regulation of metalloproteinase-dependent ectodomain ......regulation of metalloproteinase-dependent...
TRANSCRIPT
Regulation of metalloproteinase-dependent ectodomain shedding in
cytokine biology and inflammation
by
Aditya K. Murthy
A thesis submitted in conformity with the requirements for the degree of
Doctor of Philosophy
Graduate Department of Medical Biophysics
University of Toronto
© Copyright by Aditya K. Murthy 2011
ii
Abstract
Regulation of metalloproteinase-dependent ectodomain shedding in cytokine biology
and inflammation.
Aditya Murthy, Doctor of Philosophy 2011. Department of Medical Biophysics,
University of Toronto.
In 1962, Gross and Lapiere described collagenolytic activity in the degradation of
tadpole tails during amphibian metamorphosis. This activity was later attributed to a
collagenase enzyme belonging to the matrix metalloproteinase family. Over the past 49
years, steady growth in the field of metalloproteinase biology has uncovered that
degradation of extracellular matrix components represents only a fraction of the functions
performed by these enzymes. The regulatory roles of these enzymes in numerous aspects
of mammalian biology remains poorly understood.
This thesis investigates the metalloproteinase ADAM17 and its natural inhibitor
TIMP3 in acute and chronic inflammation. My work describes the generation of new
murine experimental systems of compartmentalized ADAM17 or TIMP3 deficiency and
their applications in acute liver inflammation (i.e. fulminant hepatitis and T-cell mediated
autoimmune hepatitis) and atopic dermatitis. Loss of Timp3 protected mice against
fulminant hepatic failure caused by activation of the death receptor Fas. We determined
that TIMP3 simultaneously promotes pro-apoptotic signaling through TNFR1 while
suppressing anti-apoptotic EGFR activation in the liver. Mechanistically, we identified
iii
that ADAM17 is critical in shedding TNFR1 and EGFR ligands (e.g. Amphiregulin, HB-
EGF, TGF) and extended this finding to clinically relevant drug-induced hepatitis.
Adult TIMP3 deficient mice also exhibited spontaneous accumulation of CD4+ T
cells in the liver. Consequently, polyclonal T cell activation with the lectin Concanavalin
A (con A) in a model of autoimmune hepatitis resulted in accelerated liver injury. We
identified that this immunopathology relied on TNF bioavailability as mice lacking both
Timp3 and Tnf were resistant to con A. Using bone marrow chimeras we established that
non-hematopoietic tissues were the physiologically relevant source of TIMP3 in vivo,
thereby highlighting an immunosuppressive role for this stromal metalloproteinase
inhibitor in cellular immunity.
Finally, we investigated epithelial:immune crosstalk in the epidermis by
generating tissue-specific ADAM17 deficiency in basal keratinocytes. These mice
developed spontaneous inflammatory skin disease that was physiologically consistent
with atopic dermatitis. Focused investigation of keratinocyte-specific signaling
deregulated by ADAM17 deficiency revealed its requirement for tonic Notch activation,
which in turn antagonized transcriptional activity of AP-1 transcription factors on the
promoters of epithelial cytokines TSLP and G-CSF.
In summary, these works identify cellular mechanisms governing cytokine-
mediated communication between epithelial and immune cells to modulate inflammation.
The findings that TIMP3 and ADAM17 act as regulators of key inflammatory,
proliferative and developmental pathways provide impetus to expand our understanding
of this important family of enzymes in mammalian signal transduction.
iv
Acknowledgements
My most sincere thanks go to my supervisor Dr. Rama Khokha. I could not have
asked for a more ideal environment to learn and grow as a student. You have consistently
been an example of a true mentor, and your genuine curiosity, sincerity, openness and
camaraderie are qualities I hope to emulate throughout my career. Thank you. I would
like to acknowledge my committee members Dr. Razq Hakem and Dr. Rob Rottapel. Our
meetings were always challenging and rewarding. Your critiques and advice pushed my
inquiries to the next level, and for that I am grateful. A special thanks to Dr. Juan Carlos
Zúñiga-Pflücker - you sparked my budding interest in immunology, and have gone above
and beyond to introduce me to new avenues in the field. To my labmates and colleagues,
I thank you for your friendship and constant support over the last six years.
I also thank my family – Ma, Baba, Bhaskar, Farhana and the Berghoefs. You
have been examples for me in more ways than you can imagine. Your never-ending
support and interest in my work has made this thesis a success.
To my partner Naomi, I can confidently say that none of this would be possible
without your presence in my life. You encouraged me and fed my scientific curiosity
from the day I started this journey. I look forward to many more journeys to come.
v
1 Table of Contents
Abstract .............................................................................................................................. ii
Acknowledgements .......................................................................................................... iv
Table of Contents .............................................................................................................. v
List of Figures and Tables ............................................................................................... xi
List of Abbreviations ..................................................................................................... xvi
List of Publications ......................................................................................................... xx
CHAPTER 1 ...................................................................................................................... 1
1.1 Ectodomain shedding ........................................................................................... 2
1.1.1 Tissue Inhibitors of Metalloproteinases (TIMPs) ..................................... 3
1.1.1.1 The structure and evolution of TIMPs ................................................... 4
1.1.1.2 Targets of TIMPs and phenotypes of Timp deficiency ......................... 6
1.1.1.3 TIMP1 .................................................................................................. 10
1.1.1.4 TIMP2 .................................................................................................. 12
1.1.1.5 TIMP3 .................................................................................................. 12
1.1.1.6 TIMP4 .................................................................................................. 15
1.1.2 Summary ................................................................................................. 15
1.1.3 Matrix Metalloproteinases (MMPs) ........................................................ 16
1.1.4 Disintegrin and Metalloproteinases (ADAMs) ....................................... 18
1.1.4.1 ADAMs in receptor crosstalk .............................................................. 19
1.1.4.2 ADAMs in receptor signaling cascades............................................... 20
1.1.4.3 ADAM10 ............................................................................................. 21
1.1.4.4 ADAM17 ............................................................................................. 22
vi
1.2 Ectodomain shedding in physiological processes .............................................. 23
1.2.1 Cell proliferation ..................................................................................... 23
1.2.2 Apoptosis ................................................................................................. 25
1.2.3 Ectodomain shedding in immunity ......................................................... 26
1.2.3.1 Metalloproteinases facilitate lymphocyte development and
proliferation............................................................................................................ 28
1.2.3.2 Myeloid and lymphoid cell migration is dependent on
metalloproteinase activity ...................................................................................... 33
1.2.3.3 Shedding of cell surface molecules regulates humoral and cell based
immune effector function ....................................................................................... 36
1.2.4 Summary ................................................................................................. 40
1.3 Thesis outline ..................................................................................................... 43
1.3.1 Study Rationale ....................................................................................... 43
1.3.2 Thesis Objectives .................................................................................... 44
CHAPTER 2 .................................................................................................................... 45
2.1 Abstract .............................................................................................................. 46
2.2 Introduction ........................................................................................................ 47
2.3 Results ................................................................................................................ 49
2.3.1 TNF signaling sensitizes hepatocytes to Fas-mediated apoptosis ........... 49
2.3.2 Delay of Fas-induced apoptosis in Timp3−/−
livers ................................. 52
2.3.3 TNFR1 shedding dampens JNK phosphorylation and NF-B activation in
Timp3−/−
liver ............................................................................................................ 55
2.3.4 Signaling through AKT or AMPK does not contribute to
hepatoprotection in Timp3−/−
mice ............................................................................ 62
2.3.5 Compound deletions of Timp3/Tnf or Timp3/Tnfr1 completely prevent
hepatic failure and involve enhanced ERK1/2 phosphorylation ............................... 62
vii
2.3.6 TIMP3 inhibits metalloproteinase-dependent EGFR signaling .............. 66
2.3.7 Increased EGFR ligand shedding is hepatoprotective............................. 67
2.3.8 Hepatocyte-specific loss of ADAM17 or EGFR promotes Fas-induced
killing ................................................................................................................. 69
2.3.9 Inhibitors of MAPK, EGFR or ADAM17 reverse the Timp3−/−
resistance
to Fas ................................................................................................................. 70
2.3.10 Adenoviral ADAM17 prevents acute liver failure in drug-induced
toxicity ................................................................................................................. 74
2.4 Discussion .......................................................................................................... 76
2.4.1 Ectodomain shedding of TNF is a crucial step in hepatotoxicity ........... 77
2.4.2 TIMP3 regulates pro-survival and pro-apoptotic signaling in hepatocytes
................................................................................................................. 78
2.5 Methods .............................................................................................................. 83
2.5.1 Mice ......................................................................................................... 83
2.5.2 Fas-induced hepatotoxicity ..................................................................... 83
2.5.3 APAP-induced hepatotoxicity and adenoviral delivery of ADAM17 .... 84
2.5.4 Primary hepatocyte culture and apoptosis assays ................................... 84
2.5.5 LPA treatment of MEFs .......................................................................... 85
2.5.6 Immunoblotting ....................................................................................... 86
2.5.7 Caspase activity assays............................................................................ 87
2.5.8 Electrophoretic Mobility Shift Assay (EMSA) ....................................... 87
2.5.9 Enzyme-Linked ImmunoSorbent Assay (ELISA) .................................. 88
2.5.10 Histology & Immunohistochemistry ....................................................... 88
2.5.11 RNA preparation and quantitative RT-PCR............................................ 89
2.5.12 Statistical Analyses ................................................................................. 89
viii
CHAPTER 3 .................................................................................................................... 91
3.1 Abstract .............................................................................................................. 92
3.2 Introduction ........................................................................................................ 92
3.3 Results ................................................................................................................ 94
3.3.1 A basal increase in CD4+ T cell and NKT cell populations in Timp3
−/−
livers ................................................................................................................. 94
3.3.2 TIMP3 deficiency sensitizes mice to T-cell mediated hepatitis induced by
concanavalin A .......................................................................................................... 96
3.3.3 Enhanced TNF signaling and Th1 cytokine response drives liver damage
in Timp3−/−
mice ........................................................................................................ 97
3.3.4 Cell-intrinsic TIMP3 is not required for CD4+ T cell activation .......... 103
3.3.5 Stromal TIMP3 protects against con A-induced hepatitis .................... 103
3.4 Discussion ........................................................................................................ 105
3.5 Methods ............................................................................................................ 109
3.5.1 Mice ....................................................................................................... 109
3.5.2 Induction of hepatitis & generation of bone marrow chimeras ............. 109
3.5.3 CD4+ T cell culture ............................................................................... 109
3.5.4 Serum analysis....................................................................................... 110
3.5.5 Histology & Immunoblotting ................................................................ 110
3.5.6 Flow cytometry ..................................................................................... 111
3.5.7 Cell culture ............................................................................................ 111
3.5.8 Immunoblotting ..................................................................................... 112
3.5.9 RNA preparation and quantitative RT-PCR.......................................... 112
3.5.10 Statistical Analyses ............................................................................... 113
ix
CHAPTER 4 .................................................................................................................. 115
4.1 Abstract ............................................................................................................ 116
4.2 Introduction ...................................................................................................... 116
4.3 Results .............................................................................................................. 119
4.3.1 Onset of spontaneous atopic dermatitis upon deletion of epidermal
Adam17 ............................................................................................................... 119
4.3.2 Th2-polarized cellular immunity in skin-draining lymph nodes and
myeloproliferative disease in bone marrow of Adam17ep
mice ............................. 121
4.3.3 Precocious differentiation of ADAM17 deficient keratinocytes in vivo
and in vitro ............................................................................................................... 122
4.3.4 Inducible deletion of epidermal Adam17 in adult mice recapitulates
atopic dermatitis and myeloproliferative disease .................................................... 124
4.3.5 TNFR1 signaling is dispensable for atopic dermatitis and
myeloproliferative disease in Adam17ep
mice ....................................................... 130
4.3.6 Loss of Adam17 compromises tonic Notch signaling in the epidermis 133
4.3.7 Active Notch antagonizes c-Fos/AP-1 transcription of Tslp in human
keratinocytes ............................................................................................................ 136
4.3.8 Ectopic Notch rescues the enhanced AP-1 driven stress response in
ADAM17 deficient keratinocytes ............................................................................ 139
4.3.9 Adenoviral delivery of active Notch rescues local Th2 immunity and
myeloproliferation in Adam17ep
mice ................................................................... 142
4.4 Discussion ........................................................................................................ 143
4.4.1 Contexts of ADAM10 and ADAM17 function in Notch signaling ...... 146
4.5 Methods ............................................................................................................ 148
4.5.1 Mice ....................................................................................................... 148
4.5.2 Microarray Gene Expression Analysis .................................................. 148
x
4.5.3 In vivo delivery of active Notch to rescue epidermal inflammation ..... 149
4.5.4 Isolation and culture of primary keratinocytes ...................................... 149
4.5.5 AP-1 luciferase assay and PMA treatment of mouse embryonic
fibroblasts (MEFs) ................................................................................................... 150
4.5.6 Culture and treatment of HaCaT cells and chromatin
immunoprecipitation (ChIP) .................................................................................... 151
4.5.7 Flow cytometry ..................................................................................... 151
4.5.8 Histology & cytokine analysis .............................................................. 152
4.5.9 Immunoblotting and AP-1 binding assay .............................................. 152
4.5.10 RNA preparation and quantitative RT-PCR.......................................... 153
4.5.11 Statistical Analyses ............................................................................... 153
CHAPTER 5 .................................................................................................................. 156
5.1 Manipulating ADAM17 and TIMPs in models of adult skin inflammation .... 159
5.1.1 Antigen-specific atopic dermatitis ........................................................ 159
5.1.2 Mechanical cutaneous injury ................................................................. 163
5.2 Redundancy of TIMPs in hematopoietic niches .............................................. 164
5.2.1 TIMPs in the thymic stroma .................................................................. 164
5.2.1.1 Cell autonomous and stromal effects of Timps on T lymphopoiesis 164
5.2.1.2 Mechanisms underlying altered T cell development in Timp or
Adam17 null mice ................................................................................................ 170
5.3 Concluding Remarks ........................................................................................ 174
References ...................................................................................................................... 175
xi
List of Figures and Tables
Figure 1.1 – Structure and evolution of Tissue Inhibitors of Metalloproteinases (TIMP) . 7
Figure 1.2The processes of clipping, shedding and RIPping ....................................... 29
Figure 1.3Cell surface sheddases influence adaptive immunity. .................................. 32
Figure 1.4 – Localization of metalloproteinases and their substrates impacting immunity.
........................................................................................................................................... 42
Figure 2.1 Ablation of TNF signaling delays Fas-mediated hepatotoxicity. ................. 51
Figure 2.2 – Caspase 3/7 activation in TNF and TNFR1 deficient hepatocytes, hepatic
Adam and Mmp expression following Jo-2 administration, and decreased caspase
activation in Timp3
mice. ............................................................................................. 53
Figure 2.3 Histology of liver damage upon Jo-2 treatment across all genotypes. ......... 54
Figure 2.4Delay of Fas-mediated apoptosis and hepatotoxicity in Timp3 −/−
mice. .... 56
Figure 2.5 – Comparable Fas expression in WT and T3
liver....................................... 57
Figure 2.6 – PARP cleavage in primary hepatocytes, Adam17, Tnfr1 and Tnfr2
expression in liver following Jo-2 treatment. ................................................................... 58
Figure 2.7Decreased hepatocyte apoptosis correlates with enhanced TNFR1 shedding
and abrogated TNF signaling upon loss of Timp3............................................................ 60
Figure 2.8Comparable AKT and AMPKβ phosphorylation in WT and T3−/−
livers upon
induction of Fas mediated toxicity. ................................................................................... 63
Figure 2.9Compound loss of Timp3 and Tnf or Tnfr1 completely prevents Fas-
mediated hepatotoxicity and reveals accelerated ERK1/2 phosphorylation. .................... 64
xii
Figure 2.10Loss of Timp3 enhances hepatoprotective EGFR ligand shedding and
downstream ERK1/2 phosphorylation. ............................................................................. 68
Figure 2.11 – Enhanced EGFR phosphorylation in T3
MEFS, increased EGFR target
gene expression in T3
livers. ........................................................................................ 71
Figure 2.12ADAM17 and EGFR are individually hepatoprotective against Fas-
mediated hepatotoxicity. ................................................................................................... 72
Figure 2.13TIMP3 is an upstream regulator of EGFR and MAPK activity in Fas-
induced hepatotoxocity. ................................................................................................... 73
Figure 2.14 Ectopic ADAM17 delivery protects against acetaminophen-driven
fulminant hepatitis. ........................................................................................................... 75
Figure 2.15Inhibition of cell surface ADAM17 impacts MAP kinases during stress
responses and provides stromal regulation of cell survival. ............................................. 80
Figure 3.1 Timp3−/−
mice exhibit spontaneous accumulation and activation of hepatic
CD4+ T cells and NKT cells, and enhanced sensitization to con A-induced hepatitis. .... 95
Figure 3.2Enhanced necrosis, lymphocyte infiltration and macrophage activity in
Timp3−/−
livers. ................................................................................................................. 98
Figure 3.3 Loss of Timp3 enhances TNF signaling and Th1 cytokine response during
autoimmune hepatitis. ....................................................................................................... 99
Figure 3.4Timp3−/−
;Tnf −/−
compound deletion elevates serum IL-10, IL-17A and
depletes splenic and hepatic CD4+ T cells. ..................................................................... 101
xiii
Figure 3.5 – TIMP3 is dispensable for cell-autonomous activation and proliferation of
CD4+ T cells. ................................................................................................................... 102
Figure 3.6Stromal TIMP3 protects against a Th1 pro-inflammatory cyokine response
and autoimmune hepatitis. .............................................................................................. 104
Figure 4.1Epidermal ADAM17 deficiency causes atopic dermatitis (AD). ............... 120
Figure 4.2Skin-specific Th2 lymphocyte activation and myeloproliferative disease
(MPD) in Adam17ep
mice. ............................................................................................. 123
Figure 4.3Accelerated differentiation of Adam17ep
keratinocytes. ........................... 125
Figure 4.4Inducible deletion of epidermal Adam17 in adult mice recapitulates atopic
dermatitis and skin-specific Th2 lymphocyte activation. ............................................... 128
Figure 4.5 – Onset of MPD upon tamoxifen-induced deletion of epidermal Adam17 in
adult mice. ....................................................................................................................... 129
Figure 4.6 – Tnfr1 deletion does not rescue AD in Adam17ep
mice. ............................ 131
Figure 4.7 – MPD persists in Adam17ep
;Tnfr1
double deficient mice. ..................... 132
Figure 4.8ADAM17 provides ligand-independent Notch signaling and suppresses c-
Fos activity in keratinocytes. .......................................................................................... 134
Figure 4.9Stratification of atopic dermatitis, atopic eczema and psoriasis patients based
on expression analysis of genes involved in Notch signaling......................................... 135
Figure 4.10ADAM17 activates Notch to suppress c-Fos transcriptional activity on
cytokine promoters.......................................................................................................... 138
xiv
Figure 4.11Diminished Notch activation upon ADAM17 deficiency enhances c-
Fos/AP-1 driven stress signaling in keratinocytes. ......................................................... 140
Figure 4.12AdCre-deletion of Adam17 in keratinocytes enhances Csf2 gene expression
in an AP-1 dependent manner but does not impact c-Jun, JunB, JunD, Fra1 or Il33
expression. ...................................................................................................................... 141
Figure 4.13Ectopic Notch activity rescues atopic dermatitis and myeloproliferative
disease in Adam17ep
mice. ............................................................................................. 144
Figure 4.14 – An illustration of epithelial-immune crosstalk as directed by the
ADAM17/Notch/c-Fos triad. .......................................................................................... 145
Figure 5.1 – Methods of analyzing antigen-specific responses in atopic dermatitis. ..... 162
Figure 5.2 Expression of Timp genes in bone marrow stromal cell line OP9-DL1
compared to OP9-GFP. ................................................................................................... 166
Figure 5.3 Methods of analyzing hematopoietic vs. epithelial contributions of TIMPs
and ADAM17 in thymopoiesis using fetal thymus organ culture (FTOC). ................... 167
Figure 5.4 – Pharmacological inhibition of ADAM17 activity arrests T cell development
in a dose-dependent manner............................................................................................ 168
Figure 5.5 – Additive loss of Timp genes impacts T cell development with a gene-
dependent decrease in double-negative 1 (DN1) populations. ....................................... 169
Figure 5.6 – Decreased thymic cellularity in K14-Cre;Adam17fl/fl
thymus. ................... 172
xv
Table 1.1 – Profiling TIMP target MMPs and ADAMs, and their downstream substrates.9
Table 1.2 – Murine phenotypes arising from individual Timp knockouts. ...................... 11
Table 1.3TIMP effects on apoptosis and proliferation in vitro .................................... 24
Table 2.1Primer sequences used for quantitative RT-PCR analysis. ........................... 89
Table 3.1 – Primer sequences for quantitative RT-PCR analysis and genotyping. ........ 113
Table 4.1 – Probelists and significance values for differentially expressed genes in all
studies. (See Appendix A online). .................................................................................. 154
Table 4.2Primer sequences used for qRT-PCR and ChIP analysis. ........................... 154
xvi
List of Abbreviations
AD Atopic Dermatitis
ADAM A Disintegrin And Metalloproteinase
ADAM-TS A Disintegrin And Metalloproteinase with Thrombospondin motifs
AICD Activation Induced Cell Death
AKT/PI3K Phosphoinositide 3-Kinase
ALT Alanine Transaminase
AMPK Adenosine Monophosphate-activated Protein Kinase
AP-1 Activating Protein 1
APAP N-acetyl-p-aminophenol (acetaminophen)
APC Antigen Presenting Cell
AR Amphiregulin
AST Aspartate Transaminase
APP Beta Amyloid Precursor Protein
BM Bone Marrow
CBA Cytokine Bead Array
CD Cluster Domain
cDNA complementary Deoxyribonucleic Acid
c-FLIP FLICE Inhibitory Protein
CLL Chronic Lymphocytic Leukemia
Con A Concanavalin A
DAPT N-[N-(3,5-Difluorophenacetyl)-L-alanyl]-S-phenylglycine t-butyl
ester
DC Dendritic Cell
DLL Delta Like Ligand
dLN draining Lymph Node
DMSO Dimethyl Sulfoxide
DN Double Negative
DNA Deoxyribonucleic acid
DR Death Receptor
EC Epithelial Cell
ECM Extracellular Matrix
EGFR Epidermal Growth Factor Receptor
ELISA Enzyme-Linked Immunosorbent Assay
EMSA Electrophoretic Mobility Shift Assay
EPA Erythroid Potentiating Activity
ERK Extracellular signal-regulated Kinase
ERp5 Endoplasmic Reticulum Protein 5
ES cell Embryonic Stem Cell
xvii
FACS Flow Assisted Cell Sorting
FADD Fas Assodicated via Death Domain
FAK Focal Adhesion Kinase
FasL Fas Ligand
FGF Fibroblast Growth Factor
FGFR Fibroblast Growth Factor Receptor
FTOC Fetal Thymic Organ Culture
G-CSF Granulocyte Colony Stimulating Factor
GFP Green Fluorescent Protein
GM-CSF Granulocyte Macrophage Colony Stimulating Factor
GPCR G-protein Coupled Receptor
GSK Glycogen Synthase Kinase
HB-EGF Heparin Binding EGFR-like Growth Factor
HCC Hepatocellular Carcinoma
HDMEC Human Dermal Microvascular Endothelial Cells
HEMVEC Human Endometrial Microvascular Endothelial Cell
HGF Hepatocyte Growth Factor
HLA Human Leukocyte Antigen
HMVEC Human Microvascular Endothelial Cells
HSC Hepatic Stellate Cell
HUVEC Human Umbilical Vein Endothelial Cell
IAP Inhibitor of Apoptosis
ICAM Inter-Cellular Adhesion Molecule
IFN Interferon
IGF Insulin-like Growth Factor
IGFBP Insulin-like Growth Factor Binding Protein
IB nuclear factor of Kappa light polypeptide gene enhancer in B cells,
Inhibitor
IKK nuclear factor Kappa B inhibitor kinase beta
IL-(x) Interleukin - (number)
IL-(x)R Interleukin - (number) Receptor
JNK c-Jun N-terminal Kinase
KC Keratinocyte Chemoattractant
LAG Lymphocyte Activation Gene
LPA Oleoyl-L-α-lysophosphatidic acid sodium salt
LPS
LV
Lipopolysaccharide
Left Ventricle
MAML Mastermind Like
MAPK Mitogen Associated Protein Kinase
MCP Monocyte Chemoattractant Protein
xviii
M-CSF Macrophage Colony Stimulating Factor
MEF Mouse Embryonic Fibroblast
MEKK Mitogen Associated Protein Kinase Kinase
MHC Major Histocompatibility Complex
MICA MHC class I chain-related A
microCT Micro Computed Tomography
MIG Monokine Induced by Interferon Gamma
MIP Macrophage Inflammatory Protein
miRNA micro-RNA
mLN mesenteric Lymph Node
MMP Matrix Metalloproteinase
MNC Mononuclear cell
MPD Myeloproliferative Disease
mRNA Messenger Ribonucleic Acid
MT1-MMP Membrane Type 1 Matrix Metalloproteinase
MUC1 Mucin 1, cell surface associated
NECD Notch Extra-Cellular Domain
NEMO NF kappa B Essential Modulator
NF-B Nuclear Factor kappa B
NGFR Nerve Growth Factor Receptor
NICD Notch Intra-Cellular Domain
NK Natural Killer
NKG2D Natural Killer G2D
NKT Natural Killer T cell
OVA Ovalbumin
PARP Poly-ADP Ribose Polymerase
pDC Plasmacytoid Dendritic Cell
PECAM Platelet Endothelial Cell Adhesion Molecule
PI Propidium Iodide
PIP2 Phosphatidylinositol 4,5-bisphosphate
PKC Protein Kinase C
PLC Phospholipase C
PMA phorbol-12-myristate-13-acetate
PMN Polymorphonuclear cell
PTEN Phosphatase and Tensin homolog
qRT-PCR Quantitative Reverse Transcriptase Polymerase Chain Reaction
RA Rheumatoid Arthritis
RANTES Regulated on Activation Normal T cell Expressed and Secreted
RBP-j Recombination Signal-Binding Protein 1 for J-Kappa
RECK Reversion-inducing-cysteine-rich protein with Kazal motifs
xix
RIP Receptor Intramembrane Proteolysis
RNA Ribonucleic Acid
ROS Reactive Oxygen Species
RTK Receptor Tyrosine Kinase
SDF Stromal Derived Factor
SFD Sorsby's Fundus Dystrophy
SHP SH2 domain containing Phosphatase
STAT Signal Transducer and Activator of Transcription
SV40/TAg Simian Virus 40 large T antigen
SVMP Snake Venom Metalloproteinase
TACE TNF Alpha Converting Enzyme
TanIIA Tanshinone II A
TAPI TNF Alpha Protease Inhibitor
TCR T Cell Receptor
TGF Transforming Growth Factor
TH T-helper
TIMP Tissue Inhibitor of Metalloproteinases
TNF Tumor Necrosis Factor
TNFR Tumor Necrosis Factor Receptor
TPA 12-O-tetradecanoylphorbol-13-acetate
TRADD Tumor necrosis factor recetpor type 1 Associated Death Domain
TRAIL Tumor necrosis factor Related Apoptosis Inducing Ligand
TRAILR Tumor necrosis factor Related Apoptosis Inducing Ligand Receptor
TRANCE TNF Related Activation Induced Cytokine
TRE TPA Responsive Element
Treg Regulatory T cell
TSLP Thymic Stormal Lymphoprotein
TUNEL Terminal deoxynucleotidyl transferase dUTP nick end labeling
TYK Tyrosine Kinase
UV Ultraviolet
VCAM Vascular cell adhesion molecule
VE-cadherin Vascular Endothelial Cadherin
VEGF Vascular Endothelial Growth Factor
VEGFR Vascular Endothelial Growth Factor Receptor
WT Wildtype
xx
List of Publications
Murthy, A., Cruz-Munoz, W., Khokha, R. (2008). TIMPs: Extracellular Modifiers in
Cancer Development. In The Cancer Degradome. Proteases and Cancer Biology
(pp. 371-401). New York, NY: Springer Science.
Murphy, G., Murthy, A., Khokha, R. (2008). Clipping, shedding and RIPping keep
immunity on cue. Trends in Immunology. Feb;29(2):75-82. Review
Murthy, A., Defamie, V., Smookler, D.S., Di Grappa, M.A., Horiuchi, K., Federici, M.,
Sibilia, M., Blobel, C.P., Khokha, R. (2010). Ectodomain shedding of EGFR
ligands and TNFR1 dictates hepatocyte apoptosis during fulminant hepatitis in
mice. Journal of Clinical Investigation. Aug 2;120(8):2731-44.
Murthy, A., Shao, Y.W., Narala, S.R., Molyneux, S.D., Zúñiga-Pflücker, J.C., Khokha,
R. (2011). ADAM17 activation of epidermal Notch regulates atopic barrier
immunity and myeloproliferation by suppressing c-Fos-driven transcription of
epithelial cytokines. (Under review).
Murthy, A., Shao, Y.W., Wedeles, C., Smookler, D.S., Khokha, R. (2011). Stromal
TIMP3 regulates liver lymphocyte populations and provides protection against
Th1 T-cell driven autoimmune hepatitis. (Submitted).
1
CHAPTER 1
GENERAL INTRODUCTION
This chapter includes two published reviews:
Murthy, A., W. Cruz-Munoz, and R. Khokha, TIMPs: Extracellular Modifiers in Cancer
Development, in The Cancer Degradome, D. Edwards, et al., Editors. 2008, Springer
New York. p. 373-400.
Murphy, G., A. Murthy, and R. Khokha, Clipping, shedding and RIPping keep immunity
on cue. Trends Immunol, 2008. 29(2): p. 75-82.
2
1.1 Ectodomain shedding
Identifying how cells communicate, both with each other and the extracellular
microenvironment, remain central questions in biology. It is now recognized that cells
use physical attachment to the extracellular matrix to modulate numerous processes such
as differentiation of stem cells, establishing a polarized identity, promoting cell survival
and migration. Cells also produce and release soluble factors such as cytokines to interact
with each other across larger physiological spaces. Initial studies on the extracellular
matrix identified a process by which cell surface molecules like heparan sulfate
proteoglycans can regulate the binding of a cell to extracellular matrix components. This
process, termed „ectodomain shedding‟, involved enzymatic cleavage of the extracellular
component of heparin sulfate proteoglycan and suggested a mechanism of detaching
epithelial cells to allow for movement during normal biology; it also identified a potential
role for ectodomain shedding in cancer cell invasion [1, 2]. Subsequent studies added
cytokines, growth factors and their receptors as additional targets of ectodomain shedding
[3]. The current landscape of cytokine biology identifies this process as a key step in
allowing autocrine, juxtacrine or systemic signaling. A family of enzymes termed
metalloproteinases harbours the principal proteinases which perform ectodomain
shedding at the cell surface. Their function is controlled by four tissue inhibitors of
metalloproteinses (TIMPs) [4]. Together, the TIMP:metalloproteinase axis directs lines
of communication between cells and the microenvironment during health and disease.
3
1.1.1 Tissue Inhibitors of Metalloproteinases (TIMPs)
The tissue inhibitor of metalloproteinase (TIMP) family has evolved to regulate
tissue homeostasis through its ability to operate at the stromal:cellular interface. It
controls remodeling of the extracellular matrix (ECM) as well as the cell surface by
inhibiting the activity of several classes of metalloproteinases. TIMPs are ancient proteins
found in invertebrates and vertebrates including nematodes, insects, fish and mammals.
However TIMPs have not been reported in plants despite the known existence of
metalloproteinases in this kingdom [5]. The mammalian genome contains 4 distinct
TIMP proteins, each with different yet overlapping metalloproteinase inhibitory profiles
[6]. The mammalian TIMP is a two-domain protein consisting of N- and C-termini and 6
disulphide linkages, and can exist in glycosylated and unglycosylated forms. Well-known
targets of TIMPs include the enzymes from matrix metalloproteinase (MMPs), a
disintegrin and metalloproteinase (ADAMs) and ADAM with thrombospondin motif
(ADAM-TS) classes. During metalloproteinase inhibition, a single netrin domain forms a
wedge-like structure to interact with the active site of the enzyme forming a 1:1
stoichiometric complex that sterically inhibits metalloproteinase activity [7].
TIMP was independently identified as a collagenase inhibitor [8] and for its
erythroid potentiating activity [9], and additionally murine Timp1 was identified as a cell
cycle responsive gene [10]. It was then discovered that antisense RNA-mediated
downregulation of TIMP1 conferred oncogenic properties on immortal but non-
transformed murine fibroblasts showing a central role for TIMP in tumorigenesis [11].
The systems initially used to examine TIMPs included reproductive biology [12, 13],
4
wound healing (reviewed in [14]), tumorigenesis and metastasis [11, 15]. The cloning of
all four human and murine Timp genes [9, 16-20], followed by analysis of their
regulatory elements and genetic expression studies [21-25] have revealed a more
complex role in orchestrating tissue homeostasis. Cellular and matrix turnover are both
affected by TIMPs, and each TIMP distinctly influences cell function. Clinical studies
covering diverse human cancers also document the tremendous heterogeneity of TIMP
expression and their correlation with disease stage and prognosis. Furthermore, the
understanding of TIMP function has co-evolved with that of the expanding field of
metalloproteinase biology, and together they have highlighted complementary concepts
essential for the health and survival of the organism.
1.1.1.1 The structure and evolution of TIMPs
Invertebrate and vertebrate TIMPs differ significantly in both structure and
function. The two C. elegans TIMPs are single domain proteins consisting of only the N-
terminal region of mammalian TIMP, whereas the single D. melanogaster TIMP has the
2-domain structure, and functionally resembles the mammalian TIMP3. It is able to
inhibit the activity of D. melanogaster Kuzbanian/ADAM10 along with MMP inhibition
[26, 27] and associates strongly with the ECM via interactions with hyaluronic acid, both
properties unique to the mammalian TIMP3. Until recently, the N-terminal end of TIMP3
was thought to be solely responsible for the TIMP3:ECM binding. However elegant work
by Lee et al [28] has revealed that in fact both N- and C-termini are involved in ECM-
binding. Both termini utilize basic amino acids such as Lysine and Arginine in this
5
interaction, as exhaustive mutations to specific amino acids at either terminus, together
with domain swapping to generate ECM-adhering TIMP1 clearly identify the 6 amino
acids (N-terminus: Lys-26, 27, 30, and 76; C-terminus: Lys-165 and Arg-163) required
for ECM binding. Figure 1.1A represents a schematic of the prototypical 2-domain
TIMP, with the MMP-inhibitory and ECM-binding residues indicated. As TIMP1 and
TIMP3 are structurally similar, the figure also illustrates the residues of TIMP3 that are
absent in TIMP1. The crystal structure of a full length TIMP has yet to be solved, as only
the N-termini in isolation or complexed with MMP have been obtained thus far [29-31].
The significant overlap in TIMP inhibition of metalloproteinases has likely arisen due to
the events leading to the generation of each TIMP, and phylogenetic analysis shows that
the Timp genes were created by multiple duplication events [32] (Figure 1.1B).
Comparative genomic analyses of Timp evolution between invertebrate and vertebrate
genomes provide insight into these events. The vertebrate Timp family arose from an
earlier whole genome duplication before vertebrate and invertebrate divergence [33].
Furthermore, whole genome duplications are probably responsible for the formation of
the four mammalian Timp genes, before the divergence of tetrapods and teleosts. Timp1,
Timp3, and Timp4 genes are located within introns of three Synapsin genes suggesting
that the Syn-Timp locus duplicated at least three times in M. musculus and H. sapiens.
Timp2 on the other hand is not located within a Synapsin intron, and could have arisen
via duplication of Timp alone or degeneration of the Syn locus after duplication. While C.
intestinalis has a single Timp orthologous to the 4 human Timps, D. rerio has 4 Timp2
genes (Timp2a,b,c,d) orthologous to the human Timp2 [32]. This indicates that
6
duplication events occurred in the Timp2 locus in D. rerio after the tetrapod/teleost
divergence. Among the four mammalian Timp genes, Timp1 is considered to most closely
resemble the ancestral Timp, as it demonstrates the lowest rate of evolutionary change
[7]. Figure 1.1B illustrates the sequence of gene duplications leading to the creation of
each Timp gene. Timp3 is the second most ancient while Timp2 and 4 arose from the final
duplication event and are the newest members of the family.
1.1.1.2 Targets of TIMPs and phenotypes of Timp deficiency
As the number of published substrates processed by each MMP, ADAM and
ADAMTS steadily increases, the biological importance of each TIMP in regulating tissue
homeostasis comes to light. Table 1.1 lists the inhibitory capacity of individual TIMPs on
some common metalloproteinases (MMP2, MMP7, MMP9, MT1-MMP, ADAM10,
ADAM12, ADAM7, ADAM33 and ADAMTS4), as well as the substrates processed by
each of these enzymes. Given the substrate repertoire that represents ECM and cell
surface molecules, the complexity is evident in TIMP regulation of multiple signaling
pathways. At the stromal:cellular interface, TIMPs can elicit a paracrine response by
ligand processing or modulate cell autonomous function. The immune response is a
classic example where such proteolytic processing alters immune cell activation,
migration, function of clearing antigen and finally resolution of inflammation as reviewed
in [34]. Extracellular proteolytic cascades also trigger „start‟ or „stop‟ signals for
proliferation to guide cell division. The process of liver regeneration, where ~70% of the
liver is surgically removed to initiate compensatory hepatocyte proliferation, offers a
7
Figure 1.1 – Structure and evolution of Tissue Inhibitors of Metalloproteinases (TIMP)
(A) Proposed structure of the prototypical TIMP. The two domains (N-terminal and C-terminal) are shaded
in green, and major differences between TIMP1 and TIMP3 architecture are highlighted. The conserved
VIRAK sequence in TIMP is required for MMP inhibition. (B) Evolutionary pathway in the generation of
the four mammalian TIMPs. Ancestral TIMP underwent a duplication event to generate two paralogous
TIMPs thereby creating TIMP1. A second duplication event created TIMP3 and the ancestor to TIMP2 and
4. A final duplication resulted in the generation of TIMP2 and 4. TIMP1 most highly resembles the
ancestral TIMP as it has accumulated the fewest mutations across species.
8
powerful in vivo system to study the contribution of metalloproteinases and TIMPs in cell
division. Factors important for liver regeneration, as demonstrated by genetic mutant
models, are often direct or indirect target of metalloproteinases, and therefore regulated
by TIMPs as reviewed in [35].
Although there is significant overlap in the repertoire of metalloproteinases
inhibited by each TIMP, the expression and localization patterns of these inhibitors limit
the ability of an individual TIMP to comprehensively regulate MMP activity in vivo.
Murine expression analyses indicate that Timp1 is highly expressed in the muscle, lung,
and bone, Timp2 is ubiquitously expressed, Timp3 is enriched in the heart, kidney, lung,
and thymus, and Timp4 in heart, brain and muscle [19, 24, 36, 37]. Reproductive organs
are enriched in most Timps and demonstrate cell type specificity within each tissue. For
instance, Timp2 is present in stromal cells, and Timp3 and Timp4 in epithelial cells of the
developing mouse mammary gland [36, 37]. Therefore despite having common targets,
each TIMP can regulate unique cellular processes by inhibiting an ADAM, ADAM-TS or
MMP in a specific tissue compartment [38]. It is important to note that in addition to
inhibiting metalloproteinase function, TIMPs have been shown to operate via MMP-
independent mechanisms. Below we discuss phenotypes arising in TIMP deficient mice
and attempt to describe the many consequences of regulating metalloproteinase function,
connecting ECM remodeling, intracellular signaling and pathology.
9
Table 1.1 – Profiling TIMP target MMPs and ADAMs, and their downstream substrates.
TIMP1
TIMP2
TIMP3
TIMP4
Target MMP2 MMP7 MMP9 MT1-MMP ADAM10 ADAM12
ADAM17
[39] ADAM33 ADAMTS4
ECM
Collagen I[40],
IV[41], VII[42] 4 Integrin 2M 5 Integrin Aggrecan I 2 Actinin L-Selectin 9 Integrin
Decorin[43] Collagen IV Dystroglycan Collagen I, II, III Collagen IV 91 Integrin Aggrecan I
Fibrillin[44] Decorin[43] Collagen IV, V, XI Fibronectin MBP Aggrecan I Fibronectin
Galectin-3[45] E-Cadherin Elastin Gelatin
Gelatin[46] Elastin Fibrillin Laminin
Plasminogen[47] Fibronectin I MBP Progelatinase A
Laminin TGF1 ProMMP2
Nidogen Syndecan 1
Testican 1
Cell
Surface
FGF1[48] FasL CCL7 CCL7 APP peptide HB-EGF CD30 APP
peptide
MT4-MMP
FGFR1[49] IL6R CCL11 CXCL12 Delta-Like1 IGFBP3 c-Met CD23
Pro-HGF[50] CXCL12 CD23 IGFBP5 Fractalkine Kit-L1
peptide
ICAM1 CD40 GHBP Insulin
chain
IGF2 NGFR IL1R II TRANCE
IGFBP3 IL6R
IL8 IL15RA
RECK MUC1
Pro-HGF[50] MxL1
Notch
TGF
TNF
TNFR1
TNFR2
indicates inhibition by the respective TIMP
10
1.1.1.3 TIMP1
TIMP1 was identified as erythroid potentiating activity (EPA) owing to its ability
to augment red blood cell colony formation [17]. TIMP1 deficient mice display mild
phenotypes when challenged in several models including elevated cardiac ECM
breakdown [51-53], enhanced hepatocyte proliferation [54] and altered metabolic control
of obesity [55] (Table 1.2). In accord with its high expression in reproductive tissue, male
TIMP1 deficient mice have modestly lower testosterone levels. Importantly each TIMP
exhibits a unique pattern of expression during sexual maturation in both females and
males, suggesting their specific roles at puberty [56]. The well-studied MMPs and
ADAMs inhibited by TIMP1 include MMP1, MMP9, ADAM10 and ADAM-TS4 (Table
1.1); TIMP1 is a poor inhibitor of membrane type matrix metalloproteinases (MT-
MMPs). The MMP-inhibitory function of TIMP1 correlates well with the phenotypes of
heart tissue remodeling and ventricle function observed in Timp1
mice [51-53]. TIMP1
has also been shown to regulate cell proliferation by inhibiting the release of hepatocyte
growth factor (HGF) by MMP2 and MMP9. HGF is an important growth factor required
for liver regeneration [57, 58], and hepatocytes of Timp1
mice show an accelerated
entry into the cell cycle. Specifically, Timp1
mice exhibit elevated HGF signaling
culminating in accelerated hepatocyte cell division. In this model the MMP-inhibitory
function of TIMP1 is important in regulating cell proliferation [50, 54].
11
Table 1.2 – Murine phenotypes arising from individual Timp knockouts.
Genotype Phenotypes Reference
Timp1-/-
Reduced luminal obliteration/increased re-epithelialization after tracheal transplantation [59]
Enhanced acute lung injury after bleomycin exposure [60]
Increased HGF activity in regenerating livers [54]
Altered LV geometry and cardiac function [53]
Exacerbated LV remodeling after myocardial infarction [51, 52]
Enhanced estrogen-induced uterine edema [61]
Decreased serum total testosterone levels [56]
Reduced serum progesterone levels during corpus luteum development [62]
Decreased adipose tissue development during nutritionally induced obesity [55]
Timp2-/-
Increased nerve branching and acetylcholine receptor expression [63]
Weakened muscle and reduced fast-twitch muscle mass [64]
Deficits in pre-attentional sensorimotor gating [65]
Required for efficient pro-MMP-2 activation both in vivo and in vitro [66, 67]
Timp3-/-
Enhanced metastatic dissemination to multiple organs [68]
Increased susceptibility to LPS-induced mortality [69]
Enhanced tumor angiogenesis in response to FGF-2 [70]
LV dilation and dilated cardiomyopathy following aortic banding [71]
Increased pulmonary compliance following LPS challenge [72]
Increased inflammatory response to intra-articular antigen injection & TNF-alpha [73]
Spontaneous LV dilatation, cardiomyocyte hypertrophy, and contractile dysfunction [74]
Impaired bronchiole branching morphogenesis [75]
Chronic hepatic inflammation and failure of liver regeneration [76]
Spontaneous air space enlargement and impaired lung function in aged mice [77]
Accelerated apoptosis during mammary gland involution [78]
Exacerbated systolic/diastolic dysfunction following myocardial infarction [79]
Increased corneal neovascularization [80]
Timp3+/-
Acceleration of type 2 diabetes when combined with insulin receptor heterozygosity [81]
Timp4-/-
Compromised cardiac function, higher susceptibility to myocardial infarction [82]
12
1.1.1.4 TIMP2
The well-established biological function of the TIMP family is to inhibit activated
metalloproteinases. While this is role is performed by all TIMPs, TIMP2 paradoxically
has a central function in MMP2 activation at the cell surface. It acts as an adaptor for
MMP2 by allowing the formation of a trimolecular complex involving MT1-
MMP/TIMP2/Pro-MMP2, where pro-MMP-2 is activated in a two-step process [66, 67,
83]. Thus TIMPs are capable of regulating metalloproteinase activity via multiple
mechanisms [67, 84]. Despite its previously mentioned ubiquitous expression pattern,
only neurological phenotypes have been reported in Timp2 null mice [63, 65]. Lluri et al
[64] have shown that Timp2 is expressed at neuromuscular junctions and colocalizes with
ß1 integrin. Interestingly, ß1 integrin (Itgb1) gene expression is decreased in TIMP2
deficient muscle, suggesting a role for this cell adhesion molecule in maintaining muscle
fiber integrity. The molecular mechanism explaining the decrease in Itgb1 expression in
Timp2
tissue has not been investigated, but one can speculate that the resulting loss of
ECM stability and enhanced metalloproteinase activity indirectly contributes to integrin
proteolysis [85] (Table 1.2).
1.1.1.5 TIMP3
TIMP3 is the only TIMP genetically linked to a human disease. Individuals
harbouring mutations in the C-terminal of TIMP3 suffer from a macular degenerative
disease termed Sorsby‟s fundus dystrophy (SFD) [86, 87]. Interestingly the MMP-
inhibitory property of the mutant TIMP3 is maintained in SFD patients, indicating that
13
the mechanism underlying the disease is MMP-independent. In fact the elevated
production and resulting accumulation of mutant TIMP3 in the Bruch‟s membrane of
SFD patients is causal to macular degeneration [88]. Additional clinical studies show that
deregulation of epigenetic and microRNA mediated control leading to decreased of
Timp3 expression are linked to age-related macular degeneration [89, 90] and epithelial
tumorigenesis [91-94].
Our group has investigated the physiological role of TIMP3 in multiple tissues by
exposing Timp3
mice to specific stimuli (Table 1.2). Two of the earliest phenotypes
identified were that of air space enlargement in the lung and accelerated involution of the
mammary gland of Timp3 knockout mice [77, 78]. Study by Fata et al [78] shows that the
loss of Timp3 is conducive to accelerated apoptosis during mammary involution, in part
owing to greater matrix proteolysis; the molecular mechanisms contributing to cell death
still remain to be understood. TIMP3 is a potent inhibitor of many MMP, ADAM and
ADAM-TS enzymes (Table 1.2), and the observations of pulmonary air space
enlargement [77], dilated cardiomyopathy [74] and cartilage degradation [95] in
Timp3mice indicate that enhanced metalloproteinase activity leads to compromised
ECM homeostasis as a function of aging. Subsequent studies by Mohammed et al [76],
Smookler et al [69] and Mahmoodi et al [73] investigated the role of TIMP3 in regulating
inflammation and proliferation dependent on the TNF signaling pathway, establishing
TIMP3 as an important negative regulator of TNF bioavailability owing to its unique
ability to inhibit ADAM17. The lack of Timp3 results in increased circulating levels of
TNF as well as its two receptors, TNFR1 and TNFR2. Federici et al [81] identified a
14
novel role for TIMP3 in providing protection from type 2 diabetes via modulation of TNF
shedding by ADAM17. Here, elevated vascular inflammation accompanied by
insensitivity to insulin signals caused the development of glucose intolerance and
hyperglycemia in Timp3+/
;InsR+/
mice at 6 months of age (Table 1.2). TNF signaling is
also crucial during myogenesis, and TIMP3 was shown to suppress myogenesis in adult
muscle by inhibiting ADAM17 activity and consequently TNF shedding. In their study,
Liu et al [96] identified that the microRNA miR-206 suppressed Timp3 gene expression
during muscle regeneration, showing that TIMP3 acts as a switch during myogenic
differentiation. Using a heart disease model of pressure overload, Kassiri et al [97]
dissected the dysregulation of MMP and ADAM activities in Timp3-/-
mice and found
that increased TNF transcriptionally upregulated several specific MMPs (MMP2, MT1-
MMP, MMP13) while exerting no effect on others (MMP7, MMP9). Intriguingly, the
combination of Timp3 and Tnf
- deletion led to a greater neutrophil influx and production
of MMP8 in cardiac tissue. Here, TIMP3 molecularly linked ECM turnover with that of
cytokine activity when cardiac tissue homeostasis was perturbed. More recently, we
showed that TIMP3 simultaneously regulates EGFR and TNFR1 signaling in the liver
following death receptor activation, thereby balancing cell survival [98]. This reveals a
role for the host genome in responding to acute stress in systems where previously
heterogeneity in the inducing agent(s) were thought to be the primary source of variations
in survival outcomes [99].
15
1.1.1.6 TIMP4
The last member of the TIMP family, TIMP4 protein is present in cardiomyocytes
and smooth muscle cells, with lower levels in the brain and muscle [100]. Its
colocalization with inflammatory cells such as macrophages and CD3+ T cells suggests a
role in inflammatory cardiac pathologies such as atherosclerotic lesions where it localizes
within necrotic regions. As Timp4 exhibits the most restricted expression pattern of all
the four Timps in mice, its effects may be limited to the target organs despite its property
of being a secreted protein. Generation of the Timp4
mouse by Koskivirta et al [82]
highlighted the in vivo function of this inhibitor in cardiovascular physiology. The mice
were highly prone to ventricular wall rupture in a model of myocardial infarction in an
MMP2-dependent manner. Corroborating evidence supports the protective role of TIMP4
in acute cardiovascular pathology as its expression is induced rat myocardium following
ischemia-reperfusion injury [101], and patients suffering from dilated cardiomyopathy
also exhibit decreased Timp4 gene expression in myocardial tissue [102].
1.1.2 Summary
Individual knockouts of Timp genes do not exhibit in utero lethality, although D.
melanogaster Timp mutant phenocopies integrin mutants displaying inflated wings and
premature lethality [103]. Overall, we see regulation of three important systems from the
in vivo analysis of TIMP function: ECM remodeling, cytokine and growth factor
bioavailability, and inflammatory cell function, critical in maintaining tissue homeostasis.
It is evident that deficiency of a single TIMP does not result in a complete loss of
16
regulation of these processes, raising the possibility of functional compensation by other
three TIMPs. Interestingly, TIMP3 demonstrates the ability to simultaneously affect all
three systems. These systems are intricately connected in vivo as ECM cleavage releases
not only the structural constraints but also ECM-bound ligands, and TIMP regulation of
receptor shedding influences ligand:receptor kinetics. Thus, TIMPs alter the amplitude of
a signaling stimulus as well as the triggers that serve to recruit infiltrating cells. The
tissue- and stimulus-specific requirement of each TIMP makes them important in
maintaining tissue homeostasis.
1.1.3 Matrix Metalloproteinases (MMPs)
This family of extracellular proteases has long been associated with cellular
invasion and motility, as its initially identified proteolytic activity was to degrade
extracellular matrix (ECM) components such as collagen and fibronectin. The MMPs
share similar modular structures beginning with a signal peptide, a pro-domain containing
the characteristic cysteine-switch motif (PRCGXPD), followed by the catalytic domain
which harbors a zinc-binding motif (HEXGHXXGXXH), a hinge region and a
hemopexin domain. The cysteine-switch and zinc-binding motifs are signatures of
MMPs. Structure/function properties have been used to generate six groups of MMPs:
Collagenases, gelatinases, stromelysins, matrilysins, membrane-type MMPs and a final
group of unclassified MMPs. Collagenases (MMP1, MMP8, MMP13) are identified by
their ability to cleave collagens I, II and III in addition to non-ECM targets. Gelatinases
(MMP2, MMP9) have structural modifications in their catalytic domains consisting of
17
fibronectin repeats, which allow them to associate with collagens and laminin. They
degrade gelatin, and MMP2 can also cleave collagens I, II and III. Stromelysins (MMP3,
MMP10, MMP11) can regulate the activity of other MMPs, for instance MMP3 activates
pro-MMP1 to its active form. Matrilysins are identified by their lack of a hemopexin
domain. MMP7 is the prototypical matrilysin and is efficient at cleaving a number of cell
surface proteins such as FasL, syndecan and E-cadherin among others. Membrane-type
MMPs (MMP14, MMP15, MMP16, MMP17, MMP24, MMP25) are transmembrane
proteins with the capacity to activate pro-MMP2. This interesting group of MMPs is
emerging as a significant regulator of several developmental processes, as mice lacking
MMP14 exhibit severe defects in skeletal development and angiogenesis [104, 105].
Metalloproteinases comprise the largest class of proteolytic enzymes in the human
genome [106]. 23 MMPs have been identified, and significant advances have been made
regarding their function in tumor metastasis and immune cell infiltration into tissues. For
example, MMP2 cleavage of fibronectin has been associated with ovarian cancer cell
invasion in humans [107]. On the other hand MMP8 has anti-tumor and anti-
inflammatory properties in mouse models since it degrades Macrophage Inflammatory
Protein 1 (MIP1) and downregulates immune cell influx into epithelial tissues [108,
109]. Membrane type metalloproteinases such as MT1-MMP are critical promoters of
invasion in 3-D collagen gels. Comparative analysis of multiple MMPs with MT1-MMP
has shown that MT1-MMP activity is required for motility of several cell types [110-
113]. This metalloproteinase is a potent collagenase and provides the anchoring
mechanism for MMP2 at the cell surface in addition to mediating its activation. TIMPs
18
display variable efficiencies at inhibiting MT1-MMP. For instance, TIMP1 is unable to
inhibit this metalloproteinase, while TIMP2 is the critical adaptor between MT1-MMP
and MMP2 at the cell surface [114]. At higher concentrations TIMP2 is an effective
inhibitor of MT1-MMP. Comparative analysis with mouse embryonic fibroblasts
deficient in individual TIMP, in a culture system designed to study MT1-MMP activity
and pro-MMP-2 activation at the cell surface, demonstrates that TIMP2 and 3 are key
modulators of this metalloproteinase [83]. How this complex may operate in vivo, upon
individual and combined TIMP2 and TIMP3 deficiencies, to regulate cell motility is an
important question that remains to be answered.
1.1.4 Disintegrin and Metalloproteinases (ADAMs)
Like MMPs, ADAMs also consist of secreted and transmembrane proteins that
perform ectodomain shedding of numerous substrates. The structure of ADAMs is a
distinguishing property from MMPs, as they harbor disintegrin-like domains that allow
for co-localization with cell surface integrins. Disintegrins were originally identified in
viper venom (hemorrhagic snake venom metalloproteinases; SVMPs) with the capacity to
bind platelet IIb3 receptors and inhibit its interaction with fibrinogen, thereby
preventing platelet coagulation. Structure analysis has identified two groups of ADAM
proteases: Conventional ADAMs and ADAMs harboring thrombospondin repeats
(ADAM-TS). All ADAMs are type I transmembrane proteins and consist of several
domains conserved with MMPs: A signal peptide, a pro-domain followed by a catalytic
domain containing the zinc-binding motif, a disintegrin domain, a cysteine rich domain,
19
EGF-like domain, a transmembrane domain and a cytoplasmic domain. ADAM-TSs have
thrombospondin motifs but lack a transmembrane domain, are known to cleave pro-
collagen, aggrecan and versican, and are play a functional role in cartilage remodeling
[115] The human genome contains 40 ADAMs and 19 ADAM-TSs, of which 13
ADAMs and 8 ADAM-TSs have shown proteolytic activity. The most intriguing
functions of ADAMs are their capacity to cleave growth factors, cytokines and their
receptors from the cell surface in a process termed „ectodomain shedding‟. The process of
ectodomain shedding regulates cytokine and growth factor signaling by either promoting
cytokine bioavailability (a positive regulation), or generating soluble receptors to
abrogate signaling (negative regulation). This post-translational control is a rapid means
of maintaining signal homeostasis in response to stimuli. It is important to note here that
ectodomain shedding also occurs in the golgi to generate pools of „shed‟ ligand or
receptor that is then mobilized to the cell surface via exocytosis.
1.1.4.1 ADAMs in receptor crosstalk
Study of crosstalk between ADAMs and other membrane-associated pathways
shows the capacity of metalloproteinases to regulate signal transduction. Ullrich and
colleagues have established that G-protein coupled receptors (GPCRs) and ADAMs
interact to activate receptor tyrosine kinases (RTKs) in a process termed “triple-
membrane pass” signaling. Specifically, ligand-stimulation of heterotrimeric G-proteins
consisting of Gi or G11 subunits have been reported to activate numerous ADAMs,
resulting in ectodomain shedding of EGFR ligands such as amphiregulin, HB-EGF or
20
TGF culminating in EGFR stimulation. The targeted activation of G11 is relevant since
stimulating cells with phorbol esters (e.g. PMA) results in similar activation of ADAMs
and RTK stimulation. Both G11 and PMA activate protein kinase C (PKC): G11 induces
PLC-mediated conversion of phosphatydilinositol 4,5-biphosphate (PIP2) to inositol
(1,4,5)-triphosphate (IP3) and diacyl glycerol (DAG). PMA is a DAG analog, and both
act as second messengers for PKC activation [116-118]. A cellular consequence of these
events is the activation of ADAM proteases, although the precise nature of biophysical
interactions between G-protein dependent signal modulator(s) and ADAM structural
domains is not fully elucidated. ADAMs in turn cleave targets such as EGFR ligands,
which once solubilized may act in juxtacrine, paracrine or systemic manners to stimulate
EGFR and induce a functional physiological response. This complex sequence in the
triple membrane pass signal occurs swiftly and relies on ADAM activity.
1.1.4.2 ADAMs in receptor signaling cascades
ADAMs also participate in proteolytic cascades of evolutionarily conserved
signal transduction pathways in eukaryotes. Examples include Notch and beta-catenin
activation in normal homeostasis and beta-amyloid processing in neurodegenerative
pathologies such as Alzheimer‟s disease. All these pathways utilize ADAMs as a-
secretases which perform ectodomain shedding of the target receptor. The cytoplasmic
domain is subsequently cleaved by -secretase complexes (e.g. presenilin). The regulation
of ADAM activity and signaling consequences of ADAM function remain a challenge for
the field and are the two areas of focus for this thesis.
21
1.1.4.3 ADAM10
The metalloproteinase ADAM10 was initially identified as a critical contributor to
Drosophila neurogenesis [119]. In the study, Rooke et al named the gene Kuzbanian and
showed that it belonged to the ADAM family of metalloproteinase enzymes. Subsequent
investigations showed that ADAM10 was a component of the Notch signaling pathway,
and loss-of-function or dominant negative mutations in the gene perturbed Notch-
dependent development of Drosophila imaginal wing discs and neurons. In a key set of
experiments, Pan et al revealed that ADAM10 performed proteolytic processing of Notch
ectodomain [120, 121]. Thus ADAM10 became the prototypic Notch sheddase in
eukaryotic cells. Following this, studies of in vivo phenotypes of ADAM10 deficiency
primarily focused on Notch signaling. Emerging evidence also suggested its contribution
to EGFR activation by ectodomain shedding of the EGFR ligand HB-EGF, however this
was only demonstrated in vitro using overexpression or dominant-negative ADAM10
constructs [122]. The generation of conditional ADAM10 deficiency (Adam10-floxed
mice) allowed a targeted study of this gene in mammalian systems. To date, mouse
models of tissue-specific ADAM10 deficiency have shown it to be required for Notch-
mediated differentiation of vascular endothelial cells, neuronal cortex, thymopoiesis,
marginal zone B-cells and keratinocytes [123-128]. Thus in a developmental context,
ADAM10 emerges as a required Notch sheddase.
22
1.1.4.4 ADAM17
The regulation of Tumor Necrosis Factor (TNF) signaling continues to be an
intensively studied event in biology owing to its significance in development, adult tissue
homeostasis, tumorigenesis and inflammation [129, 130]. ADAM17 was identified as the
first metalloproteinase with the capacity to cleave the TNF-alpha precursor from cell
surfaces [3, 131]. Following this initial finding, systemic bioavailability of TNF in vivo
has been attributed to ADAM17 function, however additional investigation using models
of ADAM17 deficiency have suggested that ADAM10 can also shed TNF from the cell
surface . The generation of ADAM17 deficient murine systems (e.g. Adam17Zn/Zn
mice
lacking the Zn2+
binding domain; Adam17-floxed mice; Adam17ex/ex
hypomorphic mice)
has revealed that ADAM17 is a key regulator of EGFR signaling, since mice lacking
Adam17 phenocopy TGF or EGFR knockout mice [39, 132-137].
A central goal in metalloproteinase biology is understanding the physiological
context where specific metalloproteinases can process the same substrate. An example is
Notch activation, which has been introduced above to as a target of ADAM10. Recent
studies have shown that ADAM17 can indeed shed Notch in a ligand-independent
manner, whereas ADAM10 is required for ligand-dependent Notch activation [138, 139].
As a corollary, combined deficiency of ADAM10 and ADAM17 results in accelerated
embryonic lethality and reveals a redundancy between these metalloproteinases [140].
Given the increasing numbers of substrates processed by ADAM10 and ADAM17, it is
important to acknowledge that they can substitute for each other in experimental and
physiological systems.
23
1.2 Ectodomain shedding in physiological processes
1.2.1 Cell proliferation
TIMP1 and TIMP2 were initially identified as important factors in erythropoeisis
due to their erythroid potentiating activity [141, 142]. Surprisingly, an antisense RNA-
mediated downregulation of TIMP1 transformed murine 3T3 fibroblasts into tumorigenic
cells [11]. Current research investigating the roles played by TIMPs in regulating
proliferation of both normal and malignant cells demonstrate that the cellular context
within which each TIMP is expressed may dictate its effect on proliferation [36, 143-
147]. Table 1.3 shows that each TIMP can exert opposing effects on proliferation, either
enhancing or inhibiting the process depending on the cell type involved. Additionally, the
mechanisms by which these effects are propagated differ between each study, involving
pathways such as ERK [148], EGFR [146], VEGF [85], HGF [54], FGF, NFB [149],
and cyclin D1. However, most of these responses have not been studied in-depth to
identify the direct connection between these signaling effectors and the ligands under
TIMP regulation. It is also important to note that not all of the mentioned pathways are
investigated in each system, and doing so would reveal common signals influenced by
each TIMP regardless of the cell type.
24
Table 1.3 - TIMP effects on apoptosis and proliferation in vitro
Method Apoptosis Reference
TIMP-1 rTIMP1 Reduce caspase 3 activity, enhance Bcl-2 expression in hepatic stellate cells (HSC) [150]
rTIMP1 increase activity of PI-3K/AKT, JAK2 tyrosine and Bad phosphorylation, maintained Bcl-XL
expression [151]
rTIMP1 Inhibit TNF-induced apoptosis, activation of PI-3K/AKT in endothelial cells [152]
rTIMP1 Protect MCF10 cells from TRAIL-induced cell death, caspase (3,8,9) activity, FAK and PI-3K
activation [153]
TIMP-2 rTIMP2 Increase apoptosis in activated T lymphocytes and Tsup or Jurkat lymphoma cell lines [154]
TIMP2 vector Increase apoptosis in HCC tumor [155]
TIMP3 Ad-TIMP3 Induction of caspase 8/9 activation and cleavage of PPAR and FAK, mitochondrial acitvation [156]
rTIMP3, Ad-TIMP3 Stabilization of death receptors (TNF-R, FAS, TRAIL-RI), caspase 8 activation [157]
Ad-TIMP3 reverses anti-apopotic effect of TNF-α on Fas-induced apoptosis, inhibit NF-B activation [158]
AdTIMP3 Increased apoptosis of rat aortic smooth muscle cells [143]
TIMP4 rTIMP4 Decrease apoptosis in MDA-MB-435 derived tumors, Increase expression of Bcl-2 and Bcl-XL [159]
Purified TIMP4 Induce apoptosis in transformed cardiac fibroblast but not in normal fibroblastst [160]
Method Cell proliferation Reference
TIMP1 purified TIMP1 Stimulate erythroid burst-forming units [145]
rTIMP1 Decrease proliferation of mammary ductal epithelial cells [36]
rTIMP1 Increase proliferation of MDA-MB-435 and activation of ERK and p38 pathways [161]
TIMP-2 rTIMP2 Suppression of TYK growth factor induced proliferation, disrupt EGFR phosphorylation/Grb-2
association [146]
rTIMP2 Inhibit HMVEC proliferation in response to FGF/VEGF, decrease SHP1-integrin association [85]
rTIMP2 Increase proliferation of A549 lung epithelial, cyclin D1 upregulation, NF-B activation, IkB decrease [149]
TIMP-3 AdTIMP3 increase proliferation of cardiac fibroblasts [162]
AdTIMP3 Decrease proliferation of RA-synovial fibroblasts [147]
TIMP4 AdTIMP4 Increase proliferation of cardiac fibroblasts [162]
rTIMP4 Decrease proliferation of G401 Wilm‟s tumor cells [144]
25
1.2.2 Apoptosis
As shown in Table 1.3, TIMPs are able to trigger apoptosis in a variety of cell
types including fibroblasts, endothelial, epithelial and hematopoietic cells. Resistance to
apoptosis is an important early trait in cellular transformation. While the intracellular
signaling pathways involved in apoptosis take center stage, the direct link between
TIMPs and cell death are not well defined. This is partly because of their target
repertoire, many of which act as triggers of apoptosis. Hepatic stellate cells
overexpressing Timp1 exhibit an anti-apoptotic phenotype in vitro, and Murphy et al
[150] demonstrate that the anti-apoptotic effect is dependent on its MMP inhibitory
function, as a specific loss-of-function mutation at the MMP-inhibitory region enhances
stellate cells susceptibility to various apoptotic stimuli. Studies on a variety of cell lines
such as the erythroleukemia cell line UT-7, endothelial cells and breast epithelial cells by
independent groups show that TIMP1 inhibits apoptosis via a PI(3)K dependent manner
and modulation of the Bcl family of proteins [151-153]. In contrast to TIMP1, adenoviral
Timp3 overexpression leads to enhanced apoptosis in several cell types [143]. Given the
specificity of TIMP3 for inhibiting ADAM17, and ADAM17-mediated shedding of TNF
and its receptors, Timp3 overexpression in vitro results in stabilization of the death
receptors Fas and TNFR1, thereby sensitizing cells to receptor mediated apoptosis via a
caspase-dependent mechanism [156, 157, 163]. Consistent with these roles of TIMP3,
clinical studies have demonstrated that epigenetic silencing of TIMP3 via methylation
occurs in several types of human cancers such as melanoma, breast, pancreatic, prostate,
colon and cervical cancers, suggesting its role as a tumor suppressor [164-172].
26
Compared to TIMP1 and TIMP3, less is known of the effects of TIMP2 and 4 on
apoptosis. Current literature suggests that the MMP inhibitory function of TIMP2
protects macrophages from apoptosis in an in vitro overexpression model [173] while
enhancing apoptosis in T lymphocytes [154].
The in vitro studies that demonstrate elevated Timp3 expression results in
enhanced apoptosis in normal and transformed cells [156, 157, 174, 175] are in conflict
with the in vivo study where a lack of TIMP3 leads to accelerated mammary epithelial
apoptosis during mammary involution. The level of Timp overexpression achieved in
these in vitro systems is well above physiological thresholds. Regardless of the manner of
TIMP3 manipulation, the cell death response remains the same, suggesting that the
balance between metalloproteinases and their inhibitors, rather than each component per
se, may dictate cell death.
1.2.3 Ectodomain shedding in immunity
Proteolytic processing at the cell surface and in the pericellular environment by
metalloproteinases is an intriguing point of control for the immune system. Immune
function depends on a series of events that activate and amplify the reaction targeted
towards the antigen, and proteolytic systems are central to all of these events. Recent
studies have firmly established the importance of proteolytic regulation of processes
associated with both innate immunity, which provides the first line of defense against
invading pathogens, as well as with adaptive immunity which affords long-term
protection via the generation of memory against specific foreign antigens. Two important
27
families involved are the matrix metalloproteinases (MMPs) and the disintegrin
metalloproteinases (ADAMs). Metalloproteinase involvement in aspects of immune cell
migration [176, 177], membrane protein ectodomain shedding [178] and cytokine and
chemokine function [14] have been recently reviewed. This section focuses on the role of
ADAMs and MMPs as regulatory triggers of lymphocyte maturation, clonal expansion,
migration and effector functions.
Metalloproteinases located on the plasma membrane, either as transmembrane
proteins or anchored to cell surface molecules serve as activators or inhibitors of immune
mediators. As illustrated in Figure 1.2, this is achieved through several types of
proteolytic processing, including clipping of low molecular weight chemokines or by
„sheddase‟ activity that releases an intact ectodomain of the target cell surface molecule.
Sheddases can also act in combination with intracellular -secretases to perform RIPping
of the remaining membrane associated. In the context of immunity, the cell surface
localization of proteolysis is ideal for the rapid processing and delivery of already
synthesized instructive signals such as cytokines or chemokines. Proteolytic cleavage is
also essential for generating soluble stimuli to initiate a systemic immune response,
although subsequent receptor shedding from the cell surface through metalloproteinase
activity is also a mode for dampening the initiated response. Chemokine gradients are a
critical early step in the immune response [14] and extracellular proteases shed
membrane chemokines or modify (clip) soluble chemokines to alter their localization
patterns and activity in order to direct immune cell migration to the site of injury [179].
28
In addition, proteolytic processing plays an important role in activating signal
transduction pathways essential for lymphocyte development, such as T cells [180, 181].
Metalloproteinases are kept in check by tissue inhibitors of metalloproteinases, or
TIMPs in the extracellular milieu. While little is known about the immunological
consequences of altered TIMP activity, deficiency of specific murine TIMPs impacts
inflammation in several models. For instance, shedding of Tumor necrosis factor (TNF)
by ADAM17 is an early step in both systemic and local inflammation: serum TNF levels
are known to regulate components of the adaptive immune system as well as affect target
tissues. TIMP3 is the physiological inhibitor of ADAM17, and Timp3
mice exhibit
elevated soluble TNF levels accompanied by an enhanced potential for inflammation in
several tissues [69, 76, 81]. Thus, many examples are continually emerging where
metalloproteinases and their specific inhibitors actively participate in various aspects of
immunity.
1.2.3.1 Metalloproteinases facilitate lymphocyte development and proliferation
Lymphopoiesis begins in the bone marrow, where B and T lymphocyte precursors
are generated. T cell progenitors migrate to the thymus and a cascade of interactions
between immature thymocytes and the thymic stroma produces mature T lymphocytes
which then migrate to peripheral lymphoid tissues [182, 183]. In the spleen or lymph
nodes, mature T lymphocytes encounter foreign antigen-MHC complexes on the surface
of antigen presenting cells (dendritic cells, macrophages, activated B cells) and an
29
Figure 1.2The processes of clipping, shedding and RIPping
Clipping Shedding RIPping
The proteolytic activation or
inactivation of chemokines and other
chemoattractants by the removal of
short N- or C-terminal peptides.
MMPs are important players [179,
184-186], however amino and
carboxypeptidases and proteinases of
other mechanistic classes are key
activities in some cell types [184].
The proteolysis of the ecto-domain
of cell surface proteins, including
chemokines, cytokines, growth
factors and receptors, as well as
adhesion molecules [3, 187-190].
Shedding can occur through
autocrine, juxtracrine or paracrine
mechanisms via soluble or
membrane-bound protease activity
[191, 192]. The process of shedding
may initiate or dampen a signaling
response depending on the
stimulatory or inhibitory role(s) of
the target being cleaved [137, 193].
ADAM17 mediated TNF shedding
can result in enhanced TNF
signaling, however since ADAM17
also cleaves TNFR1 and 2, the
shedding of these receptors dampens
the initial enhancement of systemic
TNF activity [131, 132, 194].
Regulated Intramembrane Proteolysis
is often performed by -secretases
such as the presenilin complex which
are aspartic proteinases [195].
RIPping is widespread in many
cellular responses and is critical to
downstream signaling upon receptor
activation of many pathways. Notch
signaling requires RIPping of the
intracellular domain (ICD) of the
receptor Notch 1, which translocates
to the nucleus and induces a
transcriptional response [196]. This
process has a specific sequence of
proteolysis where ectodomain
shedding of Notch by ADAMs must
occur prior to RIPping of the ICD
[180, 197]. Interestingly the Notch
ligand Delta-like 1 undergoes a
similar sequence of shedding and
RIPping in by ADAMs and -
secretases in a signal-independent
manner [198].
30
adaptive immune response is generated. During this event, a small activated T cell
population rapidly proliferates to provide sufficient activated T cells for effector function.
The immune response is then downregulated, leaving a small population of antigen
specific memory T cells to expedite a more rapid secondary immune response.
Notch signaling plays a crucial role in early lymphocyte commitment in the bone
marrow. ADAM mediated shedding of the Notch extracellular domain (ECD) is only
possible following binding to a Notch ligand such as Delta-like 1. Of the four mammalian
Notch proteins, extracellular cleavage of Notch1 and its ligand Delta-like 1 is performed
by ADAM17 [139] and potentially by ADAM10 [181]. RIPping of the Notch 1
intracellular domain by the -secretase presenilin 1 results in its translocation to the
nucleus, triggering transcription of Notch 1 target genes (Figure 1.2). ADAM mediated
cleavage of the Notch ECD commits hematopoeitic progenitor cell differentiation to the
T cell lineage and concomitantly inhibits commitment to the B cell lineage [199, 200]. In
the thymus, the requirement for ADAM17 in the transition of double negative (CD4-CD8
-
) thymocytes to a double positive state (CD4+CD8
+) has been identified by studies on
mice lacking functional ADAM17. C57BL/6 Adam17∆Zn/∆Zn
mice live only until 1 day
postnatal [131], or 2-3 weeks under a mixed (C57B/L/6 x 129) background [201], and
exhibit an overall reduction in lymphocyte numbers in addition to impaired lymphocyte
maturation. Li et al have further shown that this protease acts through the stromal
compartment since ADAM17 deficient bone marrow transplanted into rag deficient mice
is able to support T cell development from the double negative to double positive state.
Although this indicates a non-cell autonomous requirement of ADAM17 during
31
lymphopoiesis, the bone marrow transplanted mice are not fully rescued suggesting the
requirement of additional factors, and thus the cell autonomous role of ADAM17 cannot
be completely ruled out.
Another important concept in early adaptive immunity is that of clonal expansion
[202], and shedding at the cell surface influences lymphocyte proliferation by modulating
ligand receptor interactions. LAG3 (Lymphocyte Activation Gene 3) inhibits T cell
proliferation, and its cleavage by ADAM10 and ADAM17 alters clonal expansion
(Figure 1.3A). Notably, ADAM10 cleaves LAG3 constitutively whereas ADAM17
activity requires induction by phorbol ester or activation through CD3. Li et al clearly
demonstrate that ADAM10 provides a means for basal T cell proliferation while
ADAM17 enhances this response following the activation of the T cell receptor pathway
in a PKC-dependent manner [203].
The final critical step in immune regulation is the elimination of activated T
lymphocytes after successful clearance of antigen. The death inducing ligands, such as
those belonging to the TNF superfamily, play roles in such regulation. Fas ligand (FasL)
is a classic example of a factor engaged in both cell-mediated and soluble cytotoxicity.
The former involves the membrane bound form of FasL, whereas the latter requires
shedding by ADAM10 or MMP7 to generate a soluble FasL fragment, which can act
systemically [204]. Specific inhibition of ADAM10 results in significantly reduced FasL
shedding in primary human T cell cultures [205]. The resulting elevated surface
expression of FasL on cytotoxic T cells is associated with an increase in their killing
capacity, with more target T cells undergoing activation-induced cell death (AICD) [206]
32
Figure 1.3 - Cell surface sheddases influence adaptive immunity.
(A) ADAM17 sheds TNF from activated macrophages to enhance systemic toxicity and inflammation.
However ADAM17 can also shed TNF receptors 1 and 2 and dampen TNF function. Membrane-bound
TNF activates effector T cells, and thus ADAM17 may regulate T cell cytotoxicity by altering levels of cell
surface TNF and its receptors. Shedding of the inhibitory molecule LAG3 by ADAM10 and 17 on the
surface of activated T cells promotes clonal expansion required for a successful immune response. (B)
Hosts with MMP8 deficient PMNs exhibit elevated serum IL-13, IL-4 and IgE levels owing to enhanced
Helper T (TH2) cell activity. Consequently the humoral allergic response is also enhanced via elevated
CD23 signaling and antibody production. ADAM10 and 33 shed cell surface CD23 resulting in
macrophage activation and inflammation. It is unknown whether soluble CD23 directly activates
macrophages (dashed line). (C) T lymphocytes express FasL to kill activated effector T cells, termed
activation induced cell death (AICD). Tumor cells also produce FasL as a mechanism of immune evasion.
The T cell receptor NKG2D reduces Fas expression while enhancing FasL production in a cell-autonomous
manner. Tumor cells counteract this via enhanced activity of ERp5 (a disulphide isomerase-like protease),
shedding MICA from their cell surface, causing a decrease in NKG2D expression which sensitizes effector
T cells to Fas mediated apoptosis.
33
(Figure 1.3C). Members of the TNF superfamily require trimerization for activity, and it
is possible that an increased surface level of these factors promotes clustering, which may
also contribute to the increased functional capacity observed by Schulte et al [206].
1.2.3.2 Myeloid and lymphoid cell migration is dependent on metalloproteinase
activity
One of the earliest responses upon acute and chronic inflammation is that of
leukocyte recruitment to the site of infection. Polymorphonucleocytes (PMNs) are among
the first immune cells to arrive and set the stage for innate as well as adaptive immunity.
Significant knowledge has accumulated on the importance of generating chemokine and
cytokine gradients for immune cell recruitment, and the clipping performed by
metalloproteinases is an emerging paradigm for this step. Chemokines bind to
proteoglycans such as syndecan-1, providing a “homing” signal to circulating PMNs.
MMP7 secreted by PMNs cleaves syndecan-1 bound to CXCL1 and thus triggers the
generation of a CXCL1 chemokine gradient, which attracts neutrophils to the site of
injury [14]. ADAM10 and ADAM17 also shed transmembrane chemokines such as
CX3CL1, while ADAM10 releases CXCL16 [187-189, 207]. This enhances neutrophil
extravasation through the blood vessel endothelial layer into tissue interstitium. In
concert, metalloproteinase-mediated cleavage of these substrates modulates local immune
responses, providing an additional level of control over early immunity.
The generation of a gradient is often coupled with the modification of cell
adhesion properties of immune and endothelial cells, and is performed by the same
34
subsets of ADAMs and MMPs. Leukocyte recruitment is also affected by the proteolytic
modulation of cell-cell interaction. The first step of transendothelial migration is
mediated by L-selectin, an adhesion molecule expressed on many inflammatory cells
including monocytes, neutrophils as well as B and T lymphocytes. It mediates the initial
tethering and subsequent rolling of cells along vascular endothelium. Upon leukocyte
activation, L-selectin is rapidly clipped by ADAM17 at a specific extracellular site
proximal to the plasma membrane [178, 208]. Blocking ADAM17 activity by both
synthetic and natural MMP inhibitors causes the arrest of lymphocyte migration across
high endothelial venules [209]. B cell chronic lymphocytic leukemia (B-CLL)
lymphocytes have impaired transendothelial migration compared to normal peripheral
blood lymphocytes [210]. Moreover, transendothelial migration of B-CLLs involves the
universal loss of L-selectin and CD23 from their cell surface. ADAM17 deficient
neutrophils are still able to shed L-selectin from their surface and it has recently been
shown that ADAM8 is associated with L-selectin proteolysis in activated neutrophils
[211]. Activated endothelial cells express the cell adhesion molecules ICAM-1 and
VCAM-1, which direct their interaction with leukocytes. ADAM17 has been shown to
shed these molecules from endothelial cells [212-214]. Some secreted serine proteinases,
such as elastase and cathepsin G, can also shed VCAM-1 from neutrophils. The
generation of soluble chemokines and cytokines, and the cleavage of endothelial adhesion
molecules has also been termed ectodomain shedding, and its functional relevance in the
context of inflammation has been reviewed in detail by Garton et al [176].
35
MMP8 and MMP9 direct the early neutrophil response to the site of infection
[215, 216]. Several studies have shown that MMP8 or 9 deficiency results in impaired
neutrophil responses highlighted by decreased recruitment. A number of studies have
shown that neutrophil transmigration of the vascular endothelium occurs through either
transcellular or paracellular diapedesis and does not involve proteinases [217]. However,
MMPs are probably involved in the necessary processes of both chemoattractant and
basement membrane proteolysis. Neutrophils store MMP9 in tertiary granules and
release it upon activation by proinflammatory cytokines. Using a model of abdominal
sepsis, Renckens et al demonstrate that MMP9 activity is required for an adequate
neutrophil response to deal with E.coli infection [216]. Like MMP9, MMP8 is produced
by human PMNs in both forms, as a secreted pro-MMP or as an active enzyme on their
cell surface [218]. While soluble MMP8 is inhibited by TIMP1 and synthetic inhibitors,
the cell surface form of MMP8 is not. This opens the possibility of a juxtacrine
mechanism for PMN migration in which cell surface MMP8 activity affects the adjacent
cell behavior. In addition to enhanced MMP8 release from granulocytes, neutrophil
activation leads to transcriptional upregulation of this metalloproteinase. MMP8 deficient
mice have twice as many neutrophils in their alveolar space after lipopolysaccharide
inhalation and exhibit elevated serum levels of IgG1, IgE and IL4 – all markers of a TH2
response (Figure 1.3B) [219]. Work by Lopez-Otin et al proposes an unexpected anti-
inflammatory function for MMP8 in several in vivo models of acute and chronic
inflammation all as a result of altered neutrophil recruitment, demonstrating the role of
MMPs in directing an immune response [215, 219, 220].
36
1.2.3.3 Shedding of cell surface molecules regulates humoral and cell based
immune effector function
1.2.3.3.1 Humoral immunity
Chemokine and cytokine signalling is central to initiating an adequate humoral
response. TNF is a potent pro-inflammatory cytokine and many of its paracrine and
endocrine functions require shedding of its cell membrane bound form by ADAM17
[131]. Curiously, ADAM17 also regulates the levels of the receptors TNFR1 and TNFR2
by cleaving them from the cell surface to generate soluble antagonists [3, 221] (Figure
1.3A). Bacteria have been found to induce ADAM17, either through direct activation of
the EGFR or through IL-6 induction. EGFR transactivation per se is a process that
requires shedding of the membrane bound forms of the EGFR ligands through the
mobilisation of ADAMs [133]. In fact, ADAM17 deficient mice phenocopy defective
EGF signaling as they lack soluble TGF, an important EGFR ligand. Furthermore,
ADAM17 also causes IL-6R shedding; in a model of trans signaling, epithelial IL-6 and
soluble IL-6Rα first form a complex and subsequently interact with a membrane-
associated gp130 homodimer to activate CCL-2 expression and inhibiting CXCL8
production [222]. Thus, ADAM17 is involved with multiple facets of the humoral
response, beginning with its production and subsequently by influencing cytokine
signaling.
Once cleaved, soluble TNF has significant and immediate proinflammatory
consequences if left unchecked. This is evident during septic shock, where increased
systemic levels of TNF leads to non-specific inflammation causing vasodilation,
37
hypotension and ultimately multiple organ failure. Given that ADAM17 is the primary
TNF sheddase, its inhibitor TIMP3 becomes an important negative regulator of
inflammation. This control is lost in Timp3
mice, which exhibit increased sensitivity to
septic shock owing to uncontrolled TNF activity [69]. Ablating TNFR1 in parallel to
TIMP3 deficiency rescues the mouse from septic shock, thus demonstrating the central
role played by metalloproteinase-mediated regulation of TNF bioavailability in acute
inflammation.
Allergic immune responses are primarily mediated by the antibody IgE, which is
unique in that it is mostly located in tissues as opposed to lymph nodes or other primary
immune organs. IgE associates with the high affinity receptor FcRI on mast cells to
initiate the immune response via multiple activating signals [223]. A second low affinity
IgE receptor called CD23 or FcRII has been shown to exist on several lymphoid (e.g. B
cells, activated T cells, dendritic cells) and myeloid (monocytes, eosinophils, platelets)
immune cells, and even thymic epithelial cells [224]. FcRI activity is considered a non-
specific allergic response, whereas CD23 is involved in enhancing antibody response to
specific antigen in the presence of IgE. The shedding of CD23 has been shown to
increase macrophage activation via TNF release by peripheral blood mononuclear cells
that induces IFN production and release, thereby propogating inflammation [225].
Indicative of its role in exacerbating inflammation, increased soluble CD23 can be
detected in inflamed joints of rheumatoid arthritis patients [226, 227]. It is now known
that ADAM10 is the principal CD23 sheddase while ADAM8, ADAM9, ADAM12,
ADAM15, ADAM17, and ADAM19 do not participate in this event (Figure 1.3B) [227,
38
228]. The potential for functional redundancy between ADAM10 and ADAM33 has also
been addressed by gain-of-function experiments [227]. ADAM10 emerges as a more
specific and required sheddase, as B cells expressing only ADAM10 successfully shed
CD23. Even though specific inhibition of ADAM10 decreased both constitutive and
enhanced CD23 cleavage, it does not eliminate shedding completely suggesting the
presence of background ADAM33 activity.
1.2.3.3.2 Cell-mediated immunity
Adaptive immunity is mediated by activated T cells or by the antibody response
orchestrated by CD4+ and CD8
+ helper T cells and B cells. Once again, many aspects of
these events involve proteolytic modification of the status of cell membrane proteins as
part of a rapid response. HLA (human leukocyte antigen)-A2 is an MHC Class I protein
with primary functions in T cell development and initiation of immune response by
antigen presentation. It was recently shown in cell culture systems that ADAM10
processing of HLA-A2 subsequently leads to -secretase activity to release the
cytoplasmic domain which is then subjected to proteasomal degradation [229]. The
molecular contacts at the interface between T cells and antigen presenting cells define the
nature of the response against a particular antigen. As the T cell receptor engages a
peptide MHC complex on antigen presenting cells, a specialized supra-molecular
structure termed the immunological synapse assembles. This contains co-stimulatory and
adhesion molecules, many of which are regulated by proteolytic mechanisms. In addition
to endocytosis and lysosomal degradation, shedding can also occur, producing soluble
39
forms which may act as antagonists of the immune response. Shedding could play a key
role in dictating the components in the synapse, hence defining its nature [230]. CD28,
CD40, CD80 and CD86 are all detected as soluble forms in patients with autoimmune
diseases and could be the result of excessive or aberrant proteolytic processing [231].
Evidence that ADAM17 is a CD40 sheddase has been presented by Contin et al [232].
Similarly, proteolytic shedding of B7 molecules on antigen presenting cells during
CD4+T cell interaction has been documented as a dynamic contributor to the modulation
of costimulatory signals for cytokine production [233]. Although B7 cleavage was
blocked by the synthetic metalloproteinase inhibitor GM6001, the identity of the specific
enzyme(s) is not yet known.
Shedding and clipping have also been implicated in tumor immunity, both in the
regulation of tumor cell survival and in immune evasion through the process of
„immunoediting‟. Effector T cells can arrest tumor growth by elevating FasL activity
triggering apoptosis of tumor cells via the Fas pathway [234]. However, the mechanism
by which T cells regulate their own survival and avoid Fas-mediated apoptosis in this
scenario has only recently been uncovered. The Spies group has identified NKG2D, a
costimulatory T cell receptor as an important regulator of T cell survival but promoter of
target cell apoptosis. NKG2D+ T cells exhibit enhanced FasL production and shedding -
possibly via MMP7 or ADAM10 - to generate soluble FasL which promotes apoptosis.
Expression of NKG2D simultaneously decreases production of the receptor Fas in an
autocrine manner, conferring resistance to NKG2D+ T cells and thereby prolonging their
survival [235]. Many tumors counter FasL cytotoxicity by cell surface shedding of the
40
MHC class I related ligand (MICA) via a disulphide-isomerase-like protein, ERp5. In an
example of immunoediting, proteolytic cleavage of MICA from tumor cells results in its
binding to NKG2D followed by internalization by T cells [236] (Figure 1.3C). These
studies highlight how sheddases impact immune regulation of tumors as well as tumor
editing of an immune response.
1.2.4 Summary
Proteolytic processing through clipping, shedding and RIPping affects a diversity
of substrates that are critical throughout immunity. These encompass cytokines and their
receptors, chemokines and their adaptors, extracellular components of cell-mediated and
humoral immunity, matrix proteins, cell adhesion molecules as well as those linked to
tumor immunity. Since proteases are involved in all aspects of an immune response -
from development, expansion, effector function, to termination - regulation of their
activity is important in maintaining immune homeostasis. These proteases, which
function extracellularly or on the cell surface, are themselves tightly regulated by
trafficking, activation and by natural inhibitors. Furthermore, interactions with
intracellular proteolytic systems such as the γ-secretases provide another layer of
complexity to the immune response. TIMPs, the biological inhibitors of MMP and
ADAM activity confer an essential point of control during normal immune response, and
synthetic inhibitors of these proteases may prove to be novel therapeutic agents in
treating autoimmune disorders or other inflammatory conditions. However caution must
be exercised here as complete specificity to particular ADAMs and MMPs has yet to be
41
achieved by small molecule or synthetic metalloproteinase inhibitors, and utilization of
currently available inhibitors would possibly result in unforeseen effects on immune cell
function in both innate and adaptive immunity.
Although only a few ADAMs, MMPs and TIMPs have been explored for their
function in immunology, the emerging data has opened a new awareness for this novel
system of regulation (Figure 1.4). Multiple members exist within these protease families
in humans (23 MMPs; 13 ADAMs), with each being able to process several diverse
substrates. This complexity of MMP or ADAM processing parallels the immune
response, and can be seen to operate at the tissue, cellular and molecular levels.
Additionally, mechanisms of controlling an immune response are highly specific and
depend on proteolysis for the release of antagonists, downregulation of pro-inflammatory
signaling, and even death of activated immune effectors. Despite the subtle biochemical
processing where proteases functionally modify rather than degrade their substrates, the
physiological consequences to the organism are significant.
42
Figure 1.4 – Localization of metalloproteinases and their substrates impacting immunity.
The metalloproteinases and their substrates discussed in this section are listed above, representing a small
fraction of the ADAMs and MMPs that influence the immune response. The shaded boxes indicate the cell
type(s) associated with each substrate. Note: CD23 is expressed on a variety of myeloid (monocytes,
eosinophils, platelets) and lymphoid (B cells, activated T cells, dendritic cells) lineages [224, 228].
43
1.3 Thesis outline
1.3.1 Study Rationale
We propose that regulated ectodomain shedding plays regulatory roles in
cytokine bioavailability and signaling. As TIMPs operate in the extra-cellular milieu,
they can potentially act as signaling switches at tissue interfaces, controlling the spectrum
of cytokine signaling that occurs during an immune response. In this manner, we
hypothesize that the TIMP3/ADAM17 axis can direct epithelial:immune crosstalk and
tissue homeostasis during acute and chronic inflammation. In vitro investigations of
TIMP function suggest paradoxical roles of these genes in different cellular processes
depending on the experimental system utilized (discussed in section 1.2), and establishing
the relevance of these findings in vivo is key to understanding which cellular processes
are indeed controlled by endogenous TIMPs.
Chapter 2 of this thesis identifies that TIMP3 simultaneously orchestrates TNFR1
and EGFR signaling in the liver during sterile inflammation caused by the Fas agonist Jo-
2. Chapter 3 establishes the contribution of non-hematopoietic TIMP3 as a stromal factor
that restricts undesired NKT and CD4+ T cell influx into the liver microenvironment,
thereby providing protection against T-cell mediated autoimmune hepatitis induced by
the lectin concanavalin A. Chapter 4 details the critical function of ADAM17 in barrier
immunity; we determine that in the skin of adult mice, keratinocyte-specific ADAM17 is
required for tonic Notch activation which in turn checks uncontrolled transcription of
epithelial cytokines (specifically TSLP and G-CSF) by the transcription factor c-Fos/AP-
44
1. This basal regulation of AP-1 activity is required to prevent atopic dermatitis and
myeloproliferative disease, thereby maintaining barrier homeostasis in vivo.
1.3.2 Thesis Objectives
Investigating the role of TIMP3 in orchestrating pro- and anti-apoptotic signals
during death receptor activation in the liver
Identifying the relevant tissue compartment of TIMP3 production and its
physiological relevance in hepatic lymphocyte function
Delineating the ADAM17-dependent cellular signaling mechanism(s) which
regulate barrier immunity in the epidermis
45
2 CHAPTER 2
Ectodomain shedding of EGFR ligands and TNFR1 dictates hepatocyte apoptosis
during fulminant hepatitis in mice
A version of this chapter is a published manuscript:
Murthy, A.., Defamie, V., Smookler, D.S., Di Grappa, M.A., Horiuchi, K., Federici, M.,
Sibilia, M., Blobel, C.P., Khokha, R. Ectodomain shedding of EGFR ligands and TNFR1
dictates hepatocyte apoptosis during fulminant hepatitis in mice. J Clin Invest, 2010.
120(8): p. 2731-44.
Author contributions:
Murthy, A. – designed and performed majority of the experiments, wrote manuscript
Defamie, V. – primary hepatocyte culture
Smookler, D.S & Di Grappa, M.A. – technical assistance with Jo-2 injections
Horiuchi, K. & Blobel C.P. – providing Adam17fl/fl
mice and Adam17-/-
MEFs
Federici, M. – experimental advice on acetaminophen toxicity model
Sibilia, M. – treatment of Egfrfl/fl
and Alb-Cre;Egfrfl/fl
hepatocytes with Jo-2
46
2.1 Abstract
Fulminant hepatitis by Fas recruits extrinsic apoptotic and hepatoprotective
signals, but mechanisms for their integration are unknown. Tissue inhibitor of
metalloproteinases 3 (TIMP3) controls the critical sheddase a disintegrin and
metalloproteinase 17 (ADAM17), and may dictate stress signaling. Using mice or cells
lacking Timp3, Tnf, Tnfr1, Egfr, and Adam17, we show that ectodomain shedding of TNF
receptors and EGF family of ligands controls JNK, NF-B, caspase and ERK1/2
activation in Fas-induced hepatitis. We first demonstrate that TNF signaling promotes
hepatotoxicity, while excessive TNFR1 shedding in Timp3−/−
mice is protective.
Compound deletions of Timp3/Tnf and Timp3/Tnfr1 conferred complete resistance to Fas-
induced toxicity, uncovering parallel EGFR activation as a significant survival signal.
Loss of TIMP3 enhanced metalloproteinase-dependent EGFR signaling due to increased
release of EGFR ligands TGF, amphiregulin and HB-EGF, while depletion of shed
amphiregulin re-sensitized Timp3−/−
hepatocytes to apoptosis. These findings
demonstrate that TIMP3 and ADAM17 cooperatively dictate cytokine signaling during
death receptor activation. Finally, adenoviral delivery of ADAM17 prevented
acetaminophen-induced liver failure in a clinically relevant model of Fas-dependent
fulminant hepatitis. Thus regulated metalloproteinase activity integrates survival and
death signals during acute stress.
47
2.2 Introduction
Hepatocytes are highly sensitive to death receptor-mediated apoptosis.
Engagement of this pathway elicits multiple and complex extrinsic signals, some
apoptotic while others hepatoprotective. The cumulative tissue response to these signals
determines survival of the organism; however our understanding of the mechanisms that
coordinate them remains incomplete. Mouse models of acute and chronic hepatotoxicity
have revealed processes downstream of TNF superfamily receptors such as TNF receptor
1 (TNFR1), Fas/CD95, and death receptors 4 and 5 (DR4, DR5) that direct context-
dependent responses. For example, lipopolysaccharide- (LPS) and concanavalin A-driven
hepatitis require TNF signaling. In the LPS model, soluble TNF secreted by components
of the immune system, primarily macrophages and neutrophils, drives the destruction of
hepatocytes and results in fatal hepatitis, whereas hepatitis caused by concanavalin A is
promoted by membrane-bound TNF on T lymphocytes [237]. Furthermore, membrane-
bound TNF on T cells is protective during listeria and mycobacteria infection [238, 239].
Such studies have shown that TNF Receptor 1 signals through c-Jun N-terminal Kinase
(JNK), NF-B and caspases during a stress response, each with significantly different
outcome on survival [240-242]. In addition to the fundamental role that TNF signaling
plays in the above systems, TNF may be a contributing factor in pathologies where it is
not the primary stimulus. TRAIL and its receptor DR4 are known to promote fulminant
hepatitis, but there is a lack of conclusive evidence for the function of TNF in Fas-
mediated hepatic failure [243].
48
Even less is known about the role of EGFR signaling during acute hepatic stress.
Of the four ERBB family members, EGFR/ERBB1 provides critical mitogenic signals
through ERK1/2 phosphorylation and mice lacking EGFR do not survive postnatally
[244]. Studies have also suggested a protective role for EGFR signaling in models of
hepatocyte regeneration and cellular stress induced by reactive oxygen species or DNA
damage [117, 245]. However controversy exists regarding the role of EGFR activation in
Fas-mediated hepatocyte apoptosis, as some groups have proposed that EGFR interaction
with Fas is required for death-receptor tyrosine phosphorylation and DISC formation
while others report that EGFR ligands EGF, HB-EGF and amphiregulin are protective in
this model [245-249]. Genetic manipulation of EGFR and its modulators is required to
establish its function in hepatocyte apoptosis.
TNF, its receptors, and several EGFR ligands are cleaved from the cell surface by
a disintegrin and metalloproteinase 17 (ADAM17/TNF Alpha Converting Enzyme) in a
process termed ectodomain shedding [3, 131]. ADAM17-mediated release of EGFR
ligands from the cell surface via ectodomain shedding is considered essential for EGFR
activation. Binding of soluble ligand exposes a dimerization loop via a conformational
change in the EGFR monomer, which is then followed by homodimerization and receptor
tyrosine kinase activity. Membrane bound or uncleavable ligand can therefore impede
EGFR activation by preventing dimerization [133, 137]. Mice that lack ADAM17
phenocopy Egfr−/−
mice, pointing to the requirement of this metalloproteinase in EGFR
signaling [135]. ADAM17 activity is blocked by TIMP3, a stromal inhibitor of several
matrix metalloproteinases (MMPs) and ADAMs [104, 250]. During inflammation,
49
TIMP3 provides an extracellular mode of controlling TNF-mediated stress response. We
have previously reported that TIMP3 checks TNF release and activity during liver
regeneration and endotoxin-mediated septic shock [69, 76]. Together the
ADAM17/TIMP3 axis may simultaneously affect the opposing signals that determine
survival following death receptor activation.
This study investigates stromal control of ectodomain shedding as an integrator of
TNF and EGF signaling during Fas-mediated hepatotoxicity. We provide the first in vivo
evidence that TNF sensitizes hepatocytes to Fas-mediated death and that increased
TNFR1 shedding in TIMP3 deficient mice protects from fulminant hepatitis by
dampening TNF activation of JNK, NF-B and caspases. We also show that TIMP3 is a
negative regulator of EGFR-mediated ERK1/2 phosphorylation, which provide
hepatoprotection. Further, this concept applies to a clinically relevant model of drug
overdose-induced acute liver failure requiring Fas. TIMP3-regulation of ectodomain
shedding is therefore necessary for the apoptotic response during Fas-mediated
hepatotoxicity.
2.3 Results
2.3.1 TNF signaling sensitizes hepatocytes to Fas-mediated apoptosis
The liver is highly sensitive to inflammatory and apoptotic stimuli of TNF
superfamily members such as Fas, TNF, and TRAIL [237, 243], however the exact role
of TNF signaling in mouse models of Fas-mediated hepatotoxicity has yet to be defined.
Primary murine hepatocytes were tested for their sensitivity to Fas-mediated apoptosis in
50
the presence of a low level of TNF (1ng/ml). TNF enhanced apoptosis in wildtype
hepatocyte cultures treated with the Fas agonist Jo-2, as measured by caspase 3/7 activity.
However TNF on its own was not cytotoxic at this low concentration (Figure 2.1A). This
highlights the potency of TNF to promote Fas-mediated apoptosis.
While mice lacking TNFR1 have previously demonstrated sensitivity to Fas-
mediated hepatotoxicity [251], the kinetics of their survival has not been explored. Here
we utilized a clone of Jo-2 (listed in Methods) that allowed us to dissect the kinetics of
survival after induction of hepatotoxicity. Wildtype, Tnfsf2 deficient (Tnf−/−
) and
Tnfrsf1a deficient (Tnfr1−/−
) mice received a lethal dose of Jo-2 (0.65µg/g i.p.). Both
knockouts showed a significant delay in fulminant hepatic failure with a time to
morbidity of approximately 10 hours compared to 5 hours for wildtype controls (Figure
2.1B). An in vitro timecourse of hepatocyte apoptosis over 24 hours using 1ng/ml TNF +
10ng/ml Jo-2 revealed comparable caspase 3/7 activation between Tnf−/−
and wildtype
cells as expected when exogenous TNF was added to culture, whereas Tnfr1−/−
hepatocytes exhibited decreased caspase 3/7 activation (Figure 2.2A). Histological
examination of livers 3 hours after administration of PBS (vehicle) or Jo-2 revealed
extensive tissue damage and hemorrhaging only in the wildtype mice, however all
genotyped eventually succumbed to fulminant hepatic failure with similar liver damage
(Figure 2.1C, Figure 2.3). Immunohistochemical staining revealed activation of caspase 3
only in wildtype livers after Jo-2 administration (Figure 2.1C). Additional markers of
hepatotoxicity such as serum alanine transaminase (ALT) were significantly elevated in
wildtype mice 3 hours after Jo-2 administration (Figure 2.1D). Further, processing of
51
Figure 2.1 Ablation of TNF signaling delays Fas-mediated hepatotoxicity.
(A) Apoptosis of wildtype primary hepatocytes treated for 12 hours with indicated doses of Jo-2 in
presence or absence of 1ng/ml TNF. Caspase 3/7 activity was measured by rate of cleavage of fluorgenic
peptide Ac-DEVD-AMC and depicted relative to untreated samples. *P<0.03. Data are mean ± S.D. (n=3).
(B) Survival curve of wildtype (WT), Tnf −/−
and Tnfr1−/−
mice injected intraperitoneally (i.p.) with
0.65μg/g Jo-2. P<0.005 vs. WT (log-rank test). (C-E) WT, Tnf −/−
and Tnfr1 −/−
mice were injected i.p. with
0.65μg/g Jo-2 or PBS. Mice were sacrificed and tissue obtained 3 hours after injection. (C) Liver histology
(H & E, active caspase 3 immunohistochemistry) depicting absence of tissue damage and caspase 3
activation in Tnf −/−
and Tnfr1−/−
mice after treatment with Jo-2. Sections are representative of at least 4
mice per genotype and treatment. Bar = 100μm (D) Hepatic toxicity measured by serum ALT levels.
*P≤0.04. Data are mean ± S.D. (n=5). (E) Immunoblots assaying the processing of caspase 8 and 3 in liver
lysates. All data are representative of at least 2 independent experiments.
52
caspases 8 and 3 were seen in wildtype liver lysates, depicting the onset of apoptosis in
these mice (Figure 2.1E). In contrast, Tnf−/−
and Tnfr1−/−
mice exhibited lower serum
ALT and a lack of caspase activation, revealing that TNF signaling sensitizes to
fulminant hepatitis in vivo.
2.3.2 Delay of Fas-induced apoptosis in Timp3−/−
livers
A hallmark of fulminant hepatic failure, often lethal in the clinical setting, is the
rapid induction of hepatocyte apoptosis [252]. Metalloproteinases and their inhibitors
may orchestrate the onset of acute tissue damage by regulating cytokine bioavailability in
the microenvironment. Thus hepatic expression of selected TIMP, ADAM and MMP
genes was assayed using qRT-PCR following administration of Jo-2 to wildtype mice.
Timp1 and Timp3, but not Timp2 or Timp4, were significantly up-regulated subsequent to
Jo-2 injection (Figure 2.4A). Of the Adam (Adam9, 10, 12, 17) and Mmp (Mmp 2, 9, Mt1-
mmp) genes evaluated, Adam12 and Mmp2 expression changed upon Jo-2 administration,
but the rest remained comparable between PBS and Jo-2 treated mice (Figure 2.2B).
Timp3−/−
mice have increased TNF bioactivity in other models of hepatic inflammation
[69, 76, 253], we thus hypothesized that they would also display enhanced susceptibility
to Jo-2. Surprisingly, Timp3−/−
mice exhibited a similar resistance to that observed in
Tnf−/−
and Tnfr1−/−
mice and survived significantly longer than wildtype controls after
Jo-2 administration (Figure 2B, P=0.005, log rank test). Negligent serum ALT and the
absence of tissue damage (H & E) 3 hours after Jo-2 administration confirmed
diminished hepatotoxicity in Timp3−/−
mice (Figure 1.4C, D).
53
Figure 2.2 – Caspase 3/7 activation in TNF and TNFR1 deficient hepatocytes, hepatic
Adam and Mmp expression following Jo-2 administration, and decreased caspase
activation in Timp3
mice.
(A) Caspase 3/7 activation as a measure of apoptosis in WT, Tnf
−/− and Tnfr
−/− hepatocytes treated with
1ng/ml TNF + 10ng/ml Jo-2 over 24 hours. *P<0.05 vs. WT. Data are mean ± S.D. (n=3). (B) qRT-PCR
analysis of indicated Adam and Mmp mRNA in livers of wildtype mice treated with PBS or 0.65μg/g Jo-2.
Mice were sacrificed 3 hours after treatment and liver mRNA was obtained. *P<0.05 ± S.D. (n=3). (C)
Caspase activity measured by cleavage of fluorogenic substrates specific to Caspase 3/7 (Ac-DEVD-AMC)
and Caspase 9 (Ac-LEHD-AMC). WT and T3−/−
mice were treated with PBS or 0.65μg/g Jo-2 and
sacrificed 3 hours later.
54
Figure 2.3 Histology of liver damage upon Jo-2 treatment across all genotypes.
H & E sections of livers obtained from mice injected with PBS of Jo-2 for the indicated timepoints.
Endpoint = Liver tissue from survival timecourse of all genotypes in Figure 2.9A. Images are representative
of at least 3 mice per genotype and condition. T3 −/−
/Tnf −/−
and T3 −/−
/Tnfr1 −/−
compound knockout mice
exhibit drastically less liver damage compared to all other genotypes at endpoint, as the experiment was
terminated at 24h. Bar = 100 μm.
55
Hepatocytes utilize the intrinsic pathway of apoptosis, requiring a loss of
mitochondrial membrane potential induced by pro-apoptotic members of the Bcl-2 family
prior to activation of downstream caspases 9 and 3 [237, 241]. Immunohistochemical
staining revealed widespread caspase 3 activation in wildtype but not Timp3−/−
livers at
this timepoint (Figure 1.4E). Processing of caspases 8 and 9 was also found to be absent
in Timp3−/−
livers (Figure 1.4F), and additionally the elevation of Bim supported
activation of intrinsic apoptosis only in wildtype livers (Figure 2.4F). Fluorogenic
caspase 3 and 9 activity assays confirmed their functional activation in wildtype mice
(Figure 2.2C). A potential mechanism explaining the dampened pro-apoptotic response in
Timp3−/−
livers could be altered expression and shedding of Fas itself. Immunoblots of
liver lysates from wildtype and Timp3−/−
mice treated with PBS or Jo-2 revealed no
differences in Fas protein expression (Figure 2.5A). Additionally, analysis of cell surface
expression on untreated hepatocytes using flow cytometry showed comparable levels of
Fas in wildtype and Timp3−/−
mice (Figure 2.5B). These findings indicate that the
hepatoprotection observed in Timp3−/−
mice is not likely due to changes in Fas levels.
Overall, diminished caspase activation and a lack of Bim EL induction demonstrated
delayed hepatocyte apoptosis in Jo-2 treated mice lacking TIMP3.
2.3.3 TNFR1 shedding dampens JNK phosphorylation and NF-B
activation in Timp3−/−
liver
We next asked whether the kinetics of apoptosis and TNF signal transduction
56
Figure 2.4Delay of Fas-mediated apoptosis and hepatotoxicity in Timp3 −/−
mice.
(A) Quantitative reverse transcriptase polymerase chain reaction (qRT-PCR) analysis of Timp1, Timp2,
Timp3 and Timp4 mRNA shows elevated induction of Timp1 and Timp3 in livers of wildtype mice treated
with 0.65μg/g Jo-2. Liver mRNA was obtained 3 hours after injection *P<0.05. Data are mean ± S.E.M.
(n=3). (B) Survival curve of WT and Timp3 −/−
(T3 −/−
) mice treated with 0.65μg/g Jo-2. P<0.005 vs. WT
(log rank test). (C-F) WT and T3 −/−
mice were treated with PBS or 0.65μg/g Jo-2. Three hours after
injection, mice were sacrificed to obtain liver and serum. (C) Hepatotoxicity measured by serum ALT
levels in WT and T3 −/−
mice. *P<0.05. Data are mean ± S.E.M. (n=4). (D) Histology (H & E) of WT (from
Figure 2.1C) and T3−/−
livers 3 hours after treatment with PBS or Jo-2 depicts absence of tissue damage,
hemorrhaging and apoptotic bodies in T3−/−
liver. Bar = 100μm. (E) Immunohistochemical staining of
active caspase 3 (arrows). T3−/−
liver exhibits lack of caspase 3 activation. Sections are representative of at
least 4 mice per genotype and treatment. Bar = 100μm. (F) Immunoblot analysis of caspase 8, 9 and 3
processing and Bim EL induction in liver lysates. All data are representative of at least three independent
experiments.
57
Figure 2.5 – Comparable Fas expression in WT and T3
liver.
(A) Immunoblot of Fas on liver lysates obtained from WT and T3
−/− mice treated with PBS, 3 hours after
Jo-2 treatment and at survival endpoints after Jo-2 treatment.
(B) Flow cytometry of cell surface Fas on WT and T3 −/−
hepatocytes perfused from untreated mice. Cells
are gated on viable (PI-negative) populations and representative of three mice per genotype.
58
Figure 2.6 – PARP cleavage in primary hepatocytes, Adam17, Tnfr1 and Tnfr2
expression in liver following Jo-2 treatment.
(A) Immunoblot of caspase 3 target PARP in WT and T3
−/− hepatocytes treated with 1ng/ml TNF + 10
ng/ml Jo-2 over 24 hours. (B) Expression of indicated mRNAs in WT and T3−/−
liver lysates 3 hours after
PBS (-) or 0.65 μg/g Jo-2 treatment (+). *P<0.008. Data are mean ± S.E.M. (n=3), expressed relative to WT
mice injected with PBS, and normalized against Actb (β-actin).
59
were altered in a TIMP3 deficient state. A 24-hour apoptosis timecourse of wildtype and
Timp3−/−
hepatocyte cultures treated with 1ng/ml TNF + 10ng/ml Jo-2 showed
diminished cleavage of Poly-ADP ribose polymerase (PARP) from 3 to 12 hours in
wildtype hepatocytes compared to Timp3−/−
cells (Figure 2.6A). PARP is a physiological
target of activated caspase 3; and we observed significantly reduced caspase 3/7 activity
in Timp3−/−
cells across all timepoints (Figure 2.7A). Of the two TNF receptors, TNFR1
associates with TRADD and FADD to initiate caspase activation, and we thus asked
whether the transient hepatoprotection observed in the absence of TIMP3 correlated with
TNF receptor shedding. ELISA specific to the extracellular domains of TNFR1 and
TNFR2 in hepatocyte culture media over the 24-hour apoptosis timecourse showed
significantly greater TNFR1 release by Timp3−/−
hepatocytes compared to wildtype, but
only modest release of TNFR2 (Figure 2.7B). Fas and Fas ligand were undetectable in
hepatocyte culture media as well as in mouse serum across all treatments in both
genotypes (data not shown). Activation of the stress-kinase JNK is an important
component of TNFR signaling during stress-response. Phosphorylated JNK promotes
intrinsic apoptosis by activating the pro-apoptotic BH3-only protein Bim and degrading
anti-apoptotic Mcl-1 [237, 243, 254-256]. In accordance with elevated TNFR1 shedding,
early JNK phosphorylation was ablated in Timp3−/−
hepatocytes 3 hours after treatment,
although it was comparable at 12 and 24 hours (Figure 2.7C). JNK phosphorylation is
known to occur in transient and prolonged phases, with each having distinct
consequences on cell survival [257-259]. Here we see that loss of Timp3 ablates the early
60
Figure 2.7Decreased hepatocyte apoptosis correlates with enhanced TNFR1 shedding
and abrogated TNF signaling upon loss of Timp3.
(A) 24-hour timecourse of apoptosis measured by caspase 3/7 activity in WT and T3
−/− primary
hepatocytes treated with 1ng/ml TNF and 10ng/ml Jo-2. *P<0.03. Data are mean ± S.D. (n=3). (B) ELISA
of TNFR1 and TNFR2 release into hepatocyte culture media from (A). *P<0.002 vs WT. Data are mean ±
S.D. (n=3). (C) Immunoblots of JNK phosphorylation in hepatocyte lysates from (A), n.s. = non-specific.
(D) ELISA of serum TNFR1 levels in WT and T3 −/−
mice before (0 hrs) and after 0.65μg/g Jo-2 treatment.
*P=0.003. Data are mean ± S.E.M. (n=4). (E) Immunoblots of JNK phosphorylation in liver lysates from
WT and T3 −/−
mice 3 hours after treatment with PBS or 0.65μg/g Jo-2. (F) NF-κB binding activity was
assayed in nuclear lysates of WT and T3 −/−
mouse livers from (E) using EMSA. (G) qRT-PCR analysis of
pro-inflammatory cytokine mRNA levels in WT and T3−/−
livers from (E). *P<0.01. Data are mean ±
S.E.M. (n=3). All data are representative of at least two independent experiments.
61
transient phase while maintaining prolonged activation. Next, in vivo analysis of TNFR1
and 2 shedding and JNK phosphorylation was performed. Constitutively higher serum
levels of TNFR1 (Figure 2.7D) but not TNFR2 (data not shown) were seen in Timp3−/−
mice, along with diminished JNK phosphorylation in Timp3−/−
livers 3 hours after Jo-2
injection (Figure 2.7E). These data indicate that loss of TIMP3 increased TNFR1
shedding and dampened early stress-kinase activation in vitro and in vivo.
In addition to the stress-kinases, NF-B activation downstream of TNFR1 is
observed in numerous models of hepatotoxicity [260, 261]. Electrophoretic mobility shift
assay (EMSA) detected active NF-B in hepatic nuclear fractions of wildtype mice
treated with Jo-2, but not in mice lacking TIMP3 (Figure 2.7F). Consistent with this,
significantly higher induction of pro-inflammatory target genes IL-1, IL-6 and TNF
was observed in wildtype livers compared to Timp3−/−
tissue (Figure 2.7G). We also
measured hepatic TNFR1 and 2 expression, which was comparable across both genotypes
before and after Jo-2 treatment (Figure 2.6B). Basal ADAM17 expression measured by
qRT-PCR was found to be higher in Timp3−/−
liver as we have previously reported [76]
(Figure 2.6B). Together, TIMP3 deficiency ablated TNF signaling in response to hepatic
death-receptor activation, as made apparent by diminished JNK phosphorylation and NF-
B activity. Further, this abrogation was not due to reduced expression of TNF receptors
but arose from increased ectodomain shedding of TNFR1.
62
2.3.4 Signaling through AKT or AMPK does not contribute to
hepatoprotection in Timp3−/−
mice
Pro-survival signaling through AKT is reported to inhibit Mcl-1 degradation by
GSK3 and JNK1/2 [256, 262]. We therefore assayed AKT phosphorylation upon
treatment with Jo-2. AKT phosphorylation was comparable between wildtype and
Timp3−/−
livers treated with and without Jo-2, implying that AKT activation during Fas-
induced hepatotoxicity does not contribute to the survival of Timp3−/−
mice (Figure
2.8A). Metabolic stress in hepatocytes is known to exacerbate tissue response to pro-
inflammatory or cytotoxic insult [253], however, AMPK phosphorylation was also
comparable between wildtype and Timp3−/−
liver lysates after Jo-2 treatment (Figure
2.8B). Altered lipid metabolism through AMPK signaling probably did not account for
the resistance observed in Timp3−/−
mice.
2.3.5 Compound deletions of Timp3/Tnf or Timp3/Tnfr1 completely prevent
hepatic failure and involve enhanced ERK1/2 phosphorylation
To investigate if decreased TNF signaling was the primary mechanism promoting the
survival of Timp3−/−
mice following Fas-mediated hepatotoxicity, we bred Timp3−/−
mice
with Tnf−/−
and Tnfr1−/−
mice. Remarkably, both strains of double knockout mice were
completely resistant to the lethal dose of Jo-2 (Figure 2.9A); histological analysis showed
that the Timp3−/−
/Tnf−/−
and Timp3−/−
/Tnfr1−/−
compound knockouts exhibited little
evidence of tissue damage even at 24 hours (Figure 2.3). Analysis of serum ALT showed
63
Figure 2.8Comparable AKT and AMPKβ phosphorylation in WT and T3−/−
livers
upon induction of Fas mediated toxicity.
(A) Immunoblot of phosphorylated AKT in liver lysates of WT and T3
−/− mice 3 hours after treatment with
PBS or Jo-2. (B) Immunoblots assaying AKT and AMPKβ phosphorylation in liver lysates 3 hours after
treatment with PBS or Jo-2. Immunoblots are representative of at least 4 mice per condition.
64
Figure 2.9Compound loss of Timp3 and Tnf or Tnfr1 completely prevents Fas-
mediated hepatotoxicity and reveals accelerated ERK1/2 phosphorylation.
(A) Survival curve of WT, T3
−/−/Tnf
−/− and T3
−/−/Tnfr1
−/− compound knockout mice (dashed lines)
treated with 0.65μg/g Jo-2, superimposed on survival curves of T3 −/−
, Tnf −/−
and Tnfr1 −/−
mice from
Figure 1B and 2C. (B) Hepatotoxicity measured by serum ALT levels in mice of indicated genotypes 3
hours after treatment with PBS or Jo-2. *P<0.05. Data are mean ± S.E.M. (n=3). (C) Immunoblots of JNK,
and ERK1/2 and phosphorylation in liver lysates of indicated genotypes 3 hours after treatment with PBS
or Jo-2. n.s. = non-specific. (D) ELISA of amphiregulin, HB-EGF and TGFα released by WT and T3 −/−
hepatocytes treated with 1ng/ml TNF + 10ng/ml Jo-2 over 24 hours. *P=0.02 vs WT. Data are mean ± S.D.
(n=3). (E) Immunoblots of ERK1/2 phosphorylation in hepatocyte lysates from (C). B, D and E are
representative of at least two independent experiments.
65
a lack of hepatotoxicity in Timp3−/−
/Tnf−/−
and Timp3−/−
/Tnfr1−/−
compound knockouts 3
hours after Jo-2 treatment when compared to wildtype mice (Figure 2.9B). This implied
the involvement of additional, parallel signals relevant to hepatocyte survival that may be
regulated by TIMP3.
Given the known activation of MAPKs during cytokine-induced stress response,
we assessed JNK, p38 and ERK1/2 phosphorylation in livers of all 6 genotypes with or
without Jo-2 injection. Loss of either TNF or TNFR1 per se eliminated JNK and p38
phosphorylation in response to Jo-2, confirming the importance of TNF signaling via
stress-activated protein kinases in Fas-mediated hepatotoxicity (Figure 2.9C). TIMP3
deficiency did not significantly affect p38 phosphorylation, ruling out a role for this
MAPK in resistance to Fas-mediated hepatotoxicity in Timp3−/−
mice (Figure 2.9C). In
contrast, constitutive ERK1/2 phosphorylation was markedly elevated in untreated livers
of Timp3−/−
mice compared to other genotypes (Figure 2.9C). After Jo-2 treatment,
ERK1/2 phosphorylation was greater in the double knockout livers compared to other
genotypes. Collectively, these data indicate that TIMP3 deficiency activates pathways
upstream of ERK1/2, potentially providing hepatoprotection.
In addition to TNF and its receptors, EGFR ligands (HB-EGF, TGF,
amphiregulin) are critical substrates of ADAM17, which is inhibited by TIMP3 [137,
263, 264]. Adam17−/−
mice phenocopy Egfr−/−
mice, and hepatic deletion of EGFR
results in failed liver regeneration indicating its requirement as a mitogenic stimulus for
hepatocyte division [265]. We therefore analyzed EGFR signaling in wildtype and
Timp3−/−
primary hepatocytes. ELISA was used to measure the shedding of the EGFR
66
ligands amphiregulin, HB-EGF, and TGF into media over a 24-hour timecourse of TNF
+ Jo-2 treatment. Significantly elevated concentrations of all 3 ligands were found in
Timp3−/−
hepatocyte culture media (Figure 2.9D). Consistent with higher EGFR ligand
release, Timp3−/−
hepatocytes showed accelerated ERK1/2 phosphorylation (Figure
2.9E). These data suggest that enhanced release of multiple EGFR ligands was
responsible for increased EGFR phosphorylation, which in turn activated survival signals
through ERK1/2, contributing to the observed resistance of Timp3−/−
mice to Fas-
mediated hepatotoxicity.
2.3.6 TIMP3 inhibits metalloproteinase-dependent EGFR signaling
Of the 4 Timp genes, only TIMP3 physiologically inhibits ADAM17 activity [76,
264], and we reasoned that EGFR signaling would be checked by TIMP3. We utilized a
previously established assay to measure GPCR-mediated EGFR signaling in a
metalloproteinase-dependent manner [116, 117, 133, 134] as shown in the schematic
(Figure 2.10A). Mouse embryonic fibroblasts (MEFs) were stimulated with Oleoyl-L-α-
lysophosphatidic acid sodium salt (LPA; a GPCR agonist) and the response was assayed
by immunoblotting for phosphorylated EGFR and ERK1/2. Wildtype MEFs showed
increasing ERK1/2 phosphorylation in a dose-dependent manner, while MEFs lacking
ADAM17 showed diminished ERK1/2 phosphorylation under the same treatment
conditions. On the other hand, loss of Timp3 resulted in markedly enhanced ERK1/2
phosphorylation (Figure 2.10B). Phosphorylated ERK1/2 was also seen in the absence of
LPA in TIMP3 deficient MEFs. As controls in this series of experiments, pre-treatment
67
of all MEFs with the small molecule ADAM inhibitor TAPI-1 abolished ERK1/2
phosphorylation at the highest dose of LPA, while soluble EGF bypassed the requirement
of metalloproteinase activity. Sustained EGFR phosphorylation in Timp3−/−
MEFs
confirmed enhanced signaling owing to elevated metalloproteinase activity (Figure
2.11A). We next tested whether shed EGFR ligands are relevant to receptor activation
through antibody-mediated depletion in culture media. Given that commercial availability
of a murine-specific neutralizing antibody was limited to amphiregulin, Timp3−/−
MEFs
were stimulated with LPA in the presence of neutralizing amphiregulin antibody (AR).
Indeed, pre-treatment with this antibody significantly diminished ERK1/2
phosphorylation indicating that soluble amphiregulin contributed to EGFR signaling
(Figure 2.10C). Thus, TIMP3 is a specific negative regulator of metalloproteinase-
dependent EGFR signaling.
2.3.7 Increased EGFR ligand shedding is hepatoprotective
We next asked whether enhanced hepatic EGFR signaling due to increased ligand
bioavailability was directly protective against Fas-induced cell death. Conditioned media
was collected from wildtype, Adam17−/−
and Timp3−/−
MEFs stimulated with LPA and
subsequently added to wildtype hepatocytes treated with Jo-2 + TNF (Figure 2.10D). As
independent controls, hepatocytes were pretreated with EGF for protection or Erlotinib
(an EGFR receptor tyrosine kinase inhibitor) for sensitization to apoptosis (Figure 2.10D,
white bars). Conditioned media from Timp3−/−
MEFs provided hepatoprotection, whereas
other genotypes had no effect. Since we had observed that shed amphiregulin plays a
68
Figure 2.10Loss of Timp3 enhances hepatoprotective EGFR ligand shedding and
downstream ERK1/2 phosphorylation.
(A) Schematic of lysophosphatidic acid (LPA)-induced EGFR signaling via the mechanism of “Triple
Membrane Pass Signal”. (B) Immunoblots assaying ERK1/2 phosphorylation in MEFs of the indicated
genotypes treated with listed doses of LPA. TAPI-1 is an ADAM inhibitor; soluble EGF was used as a
positive control. WT and T3 −/−
immunoblots are representative of at least two independently derived
primary MEF lines from different embryos. (C) T3 −/−
MEFs were pre-treated with 5.0μg/ml IgG or
amphiregulin neutralizing antibodies (αAR) 2 hours prior to treatment with 1.0μM LPA for 5 min, and
ERK1/2 phosphorylation was subsequently assayed. (D) Wildtype hepatocytes were pre-treated for 3 hours
with conditioned media from LPA-treated MEFs of the indicated genotypes (colored bars), then treated
with 1ng/ml TNF + 10ng/ml Jo-2 for an additional 3 hours. As controls, hepatocytes were left untreated,
pre-treated with EGF, or with Erlotinib (white bars). *P<0.01 vs. TNF + Jo-2 treated hepatocytes that were
not pre-treated, **P=0.004. Data are mean ± S.D. (n=3). (E) T3 −/−
hepatocytes were pre-treated with
5.0μg/ml IgG or αAR for 3 hours prior to treatment with TNF + Jo-2 as in (D). *P<0.001 vs. no pre-
treatment. Data are mean ± S.D. (n=3). All data are representative of at least two independent experiments.
69
functional role in EGFR signaling, we tested whether neutralizing soluble amphiregulin
would re-sensitize Timp3−/−
hepatocytes to apoptosis. We observed far greater caspase3/7
activation upon addition of AR during TNF + Jo-2 treatment of Timp3−/−
hepatocytes
(Figure 2.10E). A physiological consequence of EGFR ligand availability during acute
stress is induction of EGFR signaling and subsequent transcription of anti-apoptotic
genes or suppression of pro-apoptotic signals. Bcl family members Mcl-1 and Bcl-xL,
along with xIAP and cIAP1/2 are putative EGFR targets [266-268], and we observe that
loss of TIMP3 results in elevated transcription of these pro-survival genes upon Jo-2
treatment (Figure 2.11B). Additionally, MAP kinase activation downstream of EGFR is
known to antagonize Bim transcription. Consistently, we observe that Timp3−/−
livers
exhibit decreased BimEL levels upon Jo-2 treatment (Figure 2.4F).Therefore, TIMP3
deficiency enhances ectodomain shedding of EGFR ligands, among which amphiregulin
is a significant contributor to hepatoprotection against death receptor activation.
2.3.8 Hepatocyte-specific loss of ADAM17 or EGFR promotes Fas-induced
killing
In order to directly determine the role of ADAM17 and EGFR in Fas-mediated
hepatocyte apoptosis, we generated mice that lack ADAM17 or EGFR in hepatocytes by
crossing Adam17fl/fl
or Egfrfl/f
mice with those expressing Cre recombinase under control
of the Albumin promoter. Primary hepatocyte cultures were generated from control
(Adam17fl/fl
, Egfrfl/f
) and hepatocyte-specific knockout mice (Adam17hep
, Egfrhep
respectively). Significantly elevated caspase 3/7 activation was observed in Adam17hep
70
hepatocytes at 1 and 3 hours after TNF + Jo-2 treatment (Figure 2.12A). Additionally
TNFR1, amphiregulin, HB-EGF or TGF levels were undetectable over the timecourse
in Adam17hep
culture media (data not shown). As with Adam17hep
cells, Egfrhep
hepatocytes exhibited increased caspase 3/7 activity at early timepoints compared to their
Egfrfl/fl
controls (Figure 2.12B). To confirm the protective role of ADAM17 in vivo,
Adam17hep
and Adam17fl/fl
control mice were injected with a lower dose of Jo-2
(0.33g/g) and monitored for survival over 3 days after treatment. A greater fraction of
Adam17hep
mice died within 3 days compared to Adam17fl/fl
mice (5/7 versus 2/5). Cre-
mediated deletion of ADAM17 or EGFR in hepatocytes was validated via
immunoblotting (Figure 2.12C, D respectively). We conclude that ADAM17 and EGFR
individually provide hepatoprotection against Fas-induced killing.
2.3.9 Inhibitors of MAPK, EGFR or ADAM17 reverse the Timp3−/−
resistance to Fas
To confirm that TIMP3 is functionally upstream of JNK and ERK1/2 signal
transduction pathways, we first examined the in vivo consequence of MAPK inhibition
on hepatotoxicity. Pretreatment of wildtype mice with a JNK inhibitor (SP600125) led to
decreased serum ALT and hepatic caspase 3 activation versus those treated with vehicle
(Figure 2.13A, B). On the other hand, Timp3−/−
mice pretreated with an EGFR inhibitor
(Erlotinib) exhibited elevated serum ALT and hepatic caspase 3 activation upon Jo-2
injection compared to those pretreated with vehicle (Figure 2.13C, D). Next, small
molecule inhibitors of ADAM17 (TAPI-1), JNK (SP600125), ERK1/2 kinase MEKK
71
Figure 2.11 – Enhanced EGFR phosphorylation in T3
MEFS, increased EGFR target
gene expression in T3
livers.
(A) LPA enhances metalloproteinase-dependent EGFR phosphorylation in mouse embryonic fibroblasts
(MEFs) lacking TIMP3. Immunoblot of phosphorylated EGFR (p-EGFR, Tyr1173) in WT and T3 −/−
MEF
lysates from Figure 5B. (B) qRT-PCR analysis of putative EGFR target genes involved in regulation of
apoptosis. T3−/−
livers exhibit elevated expression of antiapoptotic genes Mcl1, Birc3 (Ciap1/2), Bcl2l1
(Bcl-xl) and Xiap upon Jo-2 treatment. All results are relative to untreated WT liver cDNA, normalized to
Rn18s (18S) levels. *P<0.05, S.E.M. (n=3).
72
Figure 2.12ADAM17 and EGFR are individually hepatoprotective against Fas-
mediated hepatotoxicity.
(A) Apoptosis timecourse of control (fl/fl) or Adam17 deficient (Δhep) hepatocytes treated with 1ng/ml
TNF + 10ng/ml Jo-2, measured by caspase 3/7 activation. (B) Control (fl/fl) or Egfr deficient (Δhep)
hepatocytes were treated as in (A) and apoptosis measured by caspase 3/7 activation. *P<0.04. Data are
mean ± S.D. (n=3). Apoptosis timecourse of both mutants were repeated in an independent experiment with
consistent results. (C, D) Confirmation of Adam17 and Egfr deletion by immunoblot.
73
Figure 2.13TIMP3 is an upstream regulator of EGFR and MAPK activity in Fas-
induced hepatotoxocity.
(A) Serum ALT measurements of WT mice pre-treated with the JNK inhibitor SP600125 (hatched) or
vehicle (PBS + 10% DMSO, white) 45 minutes prior to Jo-2 or PBS treatment. *P<0.05. Data are mean ±
S.E.M. (n=4). (B) Immunohistochemistry depicting active caspase 3 only in WT livers of mice pretreated
with DMSO prior to Jo-2 administration. (C) Serum ALT measurements of T3 −/−
mice pre-treated with
Erlotinib (hatched) or vehicle (black) 1 hour prior to Jo-2 or PBS treatment. *P<0.05. Data are mean ±
S.E.M. (n=4). (D) Immunohistochemistry depicting active caspase 3 in T3 −/−
livers pre-treated with EGFR
inhibitor Erlotinib. (E) Hepatocytes were treated with vehicle only (DMSO), inhibitors of ADAM17
(TAPI-1), JNK (SP600125), ERK1/2 kinase MEKK (U0126), EGFR (Erlotinib) for 1 hr followed by
addition of TNF + Jo-2 for 3 hours. Apoptosis was measured as caspase 3/7 activity. *P=0.0002 and
**P<0.002 vs. DMSO. Data are mean ± S.D. (n=3). Data are representative of at least two independent
experiments.
74
(U0126), and EGFR (Erlotinib) were added to wildtype and Timp3−/−
hepatocyte cultures
treated with TNF + Jo-2 for 3 hours. The inhibition of ADAM17, ERK1/2 or EGFR
activity in Timp3−/−
hepatocytes significantly elevated caspase3/7 activity, while JNK
inhibition in wildtype hepatocytes was clearly protective (Figure 2.13E). Thus, each
small molecule inhibitor was able to reverse the sensitivity to Fas-induced apoptosis in
wildtype and Timp3−/−
genotypes. The data presented here along with Figures 4 and 5
establish TIMP3 as a regulator of MAPK activity during death receptor activation.
Specifically, the loss of TIMP3 shifts the profile of the MAPK cascade, simultaneously
inhibiting JNK activation while enhancing ERK1/2 phosphorylation to ultimately
promote survival.
2.3.10 Adenoviral ADAM17 prevents acute liver failure in drug-induced
toxicity
Hepatotoxicity driven by acetaminophen (APAP) overdose is the most common
cause of death by acute liver failure in patients upon hospitalization [269-272]. Fas
signaling plays a key role in this clinical setting [273-275]. We therefore investigated the
hepatoprotective potential of ADAM17 in wildtype mice treated with a lethal dose of
APAP (600mg/kg, i.p.). Adenoviruses encoding GFP (Ad-GFP) or ADAM17 (Ad-
ADAM17) were administered at 1x109 pfu/ml 96 hours prior to APAP treatment.
Successful expression of ADAM17 was assayed via immunoblotting (Figure 2.14E).
While a significant proportion of Ad-GFP expressing mice (6 of 9) succumbed to APAP-
75
Figure 2.14 - Ectopic ADAM17 delivery protects against acetaminophen-driven
fulminant hepatitis.
(A) Survival of wildtype mice expressing hepatic Adam17 or Gfp after treatment with a lethal dose of
APAP. Adenoviral ADAM17 (Ad-ADAM17) or GFP (Ad-GFP) was injected intravenously into wildtype
mice 96 hours prior to treatment with 600mg/kg APAP. Survival was followed for the next 72 hours.
*P=0.01 vs. Ad-GFP (log-rank test). (B) Hepatotoxicity measured by serum ALT levels from mice treated
as in (A). *P<0.01. Data are mean ± S.E.M. (n=3). (C) Histological analysis of liver damage at the survival
endpoints. Significant necrosis is observed only in Ad-GFP livers (black arrows). Images are representative
of at least 5 mice. Bars = 100μm. (D) Immunohistochemical staining of active caspase 3 (black arrows)
shows induction of apoptosis in Ad-GFP livers after APAP treatment. Images are representative of 3 mice
per group. Bars = 100μm. (E) Immunoblots of ADAM17 in liver lysates confirming delivery of adenoviral
ADAM17 to the liver. Immunoblots for cleaved caspase 8 and 3 depicts onset of APAP-induced apoptosis
in Ad-GFP mice.
76
induced hepatotoxicity, all mice expressing Ad-ADAM17 (10 of 10) survived APAP
treatment (Figure 2.14A, *P=0.013, log rank test). Measuring hepatotoxicity by serum
ALT and histological examination of the two groups of mice revealed that Ad-ADAM17
provided significant protection against APAP overdose (Figure 2.14B, C). Extensive
necrosis, cellular and structural damage was only observed in the livers of Ad-GFP
expressing mice at survival endpoint (Figure 2.14C). Next induction of apoptosis was
measured via immunohistochemical staining for active caspase3 and immunoblotting for
caspases 8 and 3. We see that delivery of adenoviral ADAM17 protects the liver against
APAP-induced caspase activation (Figure 2.14D, E.) Cumulatively, these findings
strikingly demonstrate that ADAM17 delivery prevents liver failure in clinically relevant
acute hepatitis that relies on Fas signaling.
2.4 Discussion
Induction of stress signaling by Fas causes fulminant hepatitis, where the
combined cytokine and growth factor availability in the microenvironment contributes to
promoting or preventing cell death. This challenge engages multiple pathways and cell
surface proteolysis presents an attractive interface to integrate them. We show that
individually TNF, TNFR1 and TIMP3 sensitize the liver to Fas-mediated hepatotoxicity,
and that ADAM17 and EGFR protect against this insult. TIMP3 is the sole physiological
inhibitor of ADAM17, and Timp3−/−
mice highlight how ADAM17-mediated ectodomain
shedding simultaneously regulates TNF and EGF signaling during a stress response.
Enhanced TNFR1 shedding ablates pro-apoptotic JNK phosphorylation and caspase
77
cleavage in Timp3−/−
mice. The complete protection observed in compound TIMP3 and
TNF/TNFR1 mutant mice treated with Jo-2 further reveals a significant role for ERK1/2-
mediated survival signaling in hepatoprotection. We identify a novel function of TIMP3
in negative regulation of EGFR signaling, where Timp3−/−
cells exhibit elevated release
of EGFR ligands amphiregulin, TGF and HB-EGF, constitutively enhancing EGFR and
ERK1/2 phosphorylation. Together these data are the first demonstration that ectodomain
shedding serves a powerful function in modulating MAPK cascades during acute hepatic
stress.
2.4.1 Ectodomain shedding of TNF is a crucial step in hepatotoxicity
In vivo hepatotoxicity manifests upon viral infection, chronic inflammation or
fulminant hepatic failure [252, 272]. Given the crucial function of the liver in host
defense, hepatocytes are highly sensitive to cell-extrinsic cues such as TNF release by
activated resident macrophages (Kupffer cells). Recent work by a number of groups has
revealed that activators of NF-B, specifically the IKK/NEMO regulatory subunit can
dictate the basal metabolic state of hepatocytes and consequently their capacity to convert
extracellular cues into cell-intrinsic signal transduction during chronic and acute
inflammation, death receptor activation and infection [276-278]. In addition to the shifted
MAPK activation, we observed significantly blunted NF-B activation and pro-
inflammatory target gene expression in livers of Timp3−/−
mice that were treated with Jo-
2. Consistent with our observations, Beraza et al [276] report that hepatocyte-specific
deletion of NEMO protects mice from Jo-2 induced fulminant hepatitis, corroborated
78
with decreased stress-kinase activation and pro-inflammatory gene induction.
Independent reports demonstrate that deletion of IKK, NEMO or RelA/p65 promotes
inflammation-driven hepatocarcinogenesis and sensitizes mice to TNF-induced
hepatocyte damage [279, 280]. Along with this current study, our previous findings that
Timp3−/−
mice are more sensitive to LPS-induced septic shock and fail to regenerate liver
mass upon partial hepatectomy support the hypothesis that when left unchecked,
ectodomain shedding deregulates TNF signaling in the liver [69, 76, 263].
2.4.2 TIMP3 regulates pro-survival and pro-apoptotic signaling in
hepatocytes
As modeled in Figure 9, the current study shows that in addition to impacting TNF
signaling, ADAM17 activity in hepatocytes is essential for EGFR ligand release and
promotes their paracrine and/or autocrine survival signaling. Engagement of TNFR1
recruits stress kinases and NF-B, which in turn upregulate pro-apoptotic mediators such
as Bim and caspases [281-283](Figure 9C). Meanwhile, EGFR activation acts in an
opposing manner to promote cell survival by induction of anti-apoptoic genes and
suppression of Bim [266-268, 284, 285] (Figure 9D). Therefore the protective mechanism
of ADAM17 mediated EGFR activation may be to increase EGFR ligand availability,
which in turn induces gene transcription of pro-survival factors. TIMP3 acts to regulate
these opposing pathways by inhibiting the shedding of TNFR1 and EGFR ligands (Figure
9E). The physiological role of ADAM17 in TNF-induced hepatitis was confirmed by
Horiuchi et al [286], where Mx1-Cre driven loss of ADAM17 in both myeloid cells and
79
hepatocytes or LysM-Cre driven loss in myeloid cells conferred resistance to endotoxin-
shock by abrogating the delivery of TNF. We propose that when TNF acts as a secondary
sensitizer, the availability of TNF receptors and parallel activation of other pathways
dictate the extent of hepatotoxicity. Significantly, blocking EGFR signaling by the
depletion of amphiregulin in fibroblasts and hepatocytes, or small molecule inhibition of
EGFR and ERK1/2 reversed the protection provided by greater ectodomain shedding. In
addition to genetic models, our in vivo experiments conducted with these inhibitors
confirmed that countering JNK phosphorylation alleviates hepatotoxicity, while blocking
EGFR activation re-sensitizes Timp3−/−
mice. Overall the stress response to increased
TNF superfamily ligands may depend on the repertoire of cell surface receptors available
to transmit these signals and additional factors in the tissue microenvironment that
promote or counteract death receptor activation.
TRAIL is also reported to promote Fas-mediated hepatotoxicity through JNK and
Bim in primary hepatocytes and the liver [243]. We provide physiological evidence here
that TNF itself operates in a similar manner in vivo. Since TNF and TRAIL can activate
opposing signal transduction through FADD and NF-B, the parallel induction of yet
other pathways can tip the cellular response to survival or death. The pro-survival role of
AKT has been well documented, and its capacity to antagonize JNK may contribute to
protection [256, 287]. Our probing of AKT activity led us to conclude that Timp3−/−
cells
do not utilize this pathway during hepatotoxicity since no increase in phosphorylated
AKT was observed in Timp3 deficient tissues subjected to apoptotic stress. Our results
provide insight into EGFR activation and ERK1/2 signal transduction in a TIMP3
80
Figure 2.15Inhibition of cell surface ADAM17 impacts MAP kinases during stress
responses and provides stromal regulation of cell survival.
(A) Fas-activation triggers intrinsic apoptosis. (B) Macrophage (Kupffer cell) activation results in TNF
shedding, which acts as a secondary sensitizer to Fas-mediated toxicity. (C) TNF/TNFR1 binding activates
both stress kinases (JNK) and NF-κB. ADAM17 shedding of TNFR1 dampens pro-apoptotic signaling. (D)
ADAM17 shedding of EGFR ligands provides protection from apoptosis through ERK1/2 signaling. (E)
The hepatic microenvironment harbors TIMP3, which checks ADAM17 activity.
81
deficient state, which commits hepatocytes to survival following Fas-mediated toxicity.
In support of this interpretation, the protective role of the EGFR pathway is demonstrated
in models of stress induced by UV and ROS [117, 288], and a pro-apoptotic role of
TIMP3 in Fas-mediated apoptosis has been suggested in neuronal and synovial fibroblast
models in vitro [158, 289]. Our finding that dampened TNFR1 activation coupled with
enhanced EGFR signaling is protective against acute hepatic failure is consistent with
observations that individually, ablating JNK signaling or elevating EGFR activation can
protect, at least in part, against receptor-mediated apoptosis [243, 245]. While pro-
apoptotic and pro-survival kinases can operate in isolation, it is unlikely that this is the
case during physiological stress. The reported intracellular crosstalk between JNK and
ERK1/2 is another level at which the apoptotic response can be shifted [290, 291]. Our
observations are within the context of death-receptor activation during acute stress,
however the coordination of opposing signal transduction pathways applies to numerous
aspects of hepatic homeostasis which are out of the scope of the current study. A recent
review by Michalopoulos discusses the unknowns of coordinating survival and death
signals during liver regeneration and importantly liver failure upon loss of this regulation
[292].
Fulminant hepatitis is a significant cause of drug-induced liver failure and also
contributes to the hepatotoxicity observed in steatosis and end stage liver disease. Our
findings show that ectodomain shedding promotes hepatocyte survival in this acute
setting by the dual mechanism of enhanced EGFR-ERK1/2 activation and dampened
82
TNFR1 signaling through JNK and NF-B. We further demonstrate that adenoviral
delivery of ADAM17 is capable of protecting mice from acetaminophen-induced liver
failure. Even though unchecked metalloproteinase activity is implicated in chronic
hepatic inflammation, inducing transient ectodomain shedding may be an attractive
strategy to limit toxicity during fulminant liver failure.
83
2.5 Methods
2.5.1 Mice
All mice used in this study were of C57BL/6 background. Male mice aged 9-12
weeks were used for all the experimental procedures. Timp3−/−
mice have been
previously described [76], and Tnf−/−
or Tnfr1−/−
mice were obtained from the Jackson
Laboratory. These strains were crossed with Timp3−/−
mice to generate Timp3−/−
Tnf−/−
and Timp3−/−
Tnfr1−/−
mice. Hepatocyte-specific knockouts were produced by crossing
mice expressing Cre recombinase under the Albumin promoter (Jackson Labs) with
Adam17fl/fl
or Egfrfl/fl
mice. Genotypes were confirmed via PCR for the respective genes;
primer sequences are available upon request. Mice were fed 5%-fat chow ad libitum,
housed and cared for in accordance with the guidelines approved by the Canadian
Council for Animal Care, and the Animal Care Committee of the Ontario Cancer
Institute.
2.5.2 Fas-induced hepatotoxicity
Mice were injected intraperitoneally with Fas-agonist antibody (CD95, clone Jo-
2, NA/LE) in 100µL sterile PBS to trigger death receptor activation. Since Jo-2 exhibits
lot-to-lot variation in its potency of hepatocyte killing, we tested three lots and identified
BD pharmingen 554254 at 0.65µg/g to cause morbidity within 5-7 hours in wildtype
mice. Control mice were injected with 100µL sterile PBS. For JNK inhibition
experiments, wildtype mice were injected intraperitoneally with 0.25µg/g SP600125
(Calbiochem) 45 minutes prior to Jo-2 injection. For EGFR inhibition experiments,
84
Timp3−/−
mice were injected intraperitoneally with 0.50µg/g Erlotinib (generously
provided by Dr. Ming-Sound Tsao) 1 hour prior to Jo-2 injection. PBS containing 10%
DMSO was used as a vehicle. Animals were sacrificed by CO2 asphyxiation at designated
time points for analysis. Survival curves were obtained based on the time at which
animals became pre-moribund. Hepatotoxicity was determined by measuring serum ALT
levels.
2.5.3 APAP-induced hepatotoxicity and adenoviral delivery of ADAM17
To obtain an optimal proportion of adenovirus-infected hepatocytes, a dose
titration of Ad-GFP (Vector Biolabs) was performed in wildtype mice at 1x107, 1x10
8
and 1x109 pfu/ml dissolved in 500
9 pfu/ml of Ad-
GFP showed a significantly higher proportion of GFP-positive hepatocytes compared to
1x107 or 1x10
8 pfu/ml titers. Thus 1x10
9 pfu/ml Ad-GFP and Ad-ADAM17 were
dissolved in sterile PBS and delivered intravenously in the tail veins of 12-week old male
mice. 96 hours after injection of virus, mice were starved for 18 hours prior to
acetaminophen injections (APAP; Sigma A5000). APAP was dissolved in sterile PBS
and warmed to 55oC to dissolve. 600mg/kg dosed in 200l PBS was intraperitoneally
injected into mice. Survival curves were obtained based on the pre-moribund status of
animals, at which point they were sacrificed. Liver, kidney and serum were obtained for
subsequent analyses.
2.5.4 Primary hepatocyte culture and apoptosis assays
85
Primary hepatocytes were prepared by perfusing the portal vein of the liver with
0.02mg/ml Liberase (Roche) and enriched using a 10% Percoll (Sigma) gradient.
Hepatocytes were cultured in William‟s E medium containing 10% fetal bovine serum,
2mM L-glutamine, 0.1U/ml insulin, 1.0mM dexamethasone and antibiotics on 6-well
plates coated with 2.5mg/ml type I rat-tail collagen. Hepatocytes (1x106/well, 6-well
plate) were serum-starved for 24 hours, then treated with 10ng/ml Fas-agonist antibody
(Jo-2, BD Biosciences) in the presence or absence of 1ng/ml recombinant mouse TNF (R
& D Systems) for the indicated timepoints. Media and cell lysate were collected for
analysis. Hepatocytes were co-incubated with small molecule inhibitors to ADAM17
(10µM TAPI-1, Peptides International), JNK (25µM SP600125, Calbiochem), EGFR
(100nM Erlotinib) and MEKK (10µM UO-126, Sigma) where indicated. For conditioned
media (CM) experiments, culture media was obtained from MEFs of indicated genotypes
that were stimulated with 1.0M LPA for 1 hour. Wildtype hepatocytes were incubated
with CM 3 hours prior to addition of 1ng/ml TNF + 10ng/ml Jo-2. CM was maintained in
a 1:1 dilution with hepatocyte media containing TNF + Jo-2 during induction of
apoptosis for another 3 hours, after which cell lysate was collected for analysis. Timp3−/−
hepatocytes were incubated with 5.0g/ml IgG or Amphiregulin neutralizing antibody
(AF989, R & D Systems) 3 hours prior to addition of 1ng/ml TNF + 10ng/ml Jo-2.
Antibodies were maintained in hepatocyte culture media during induction of apoptosis
for another 3 hours. Cell lysate was collected for caspase3/7 activity analysis.
2.5.5 LPA treatment of MEFs
86
Primary murine embryonic fibroblasts (MEFs) were generated as described [83]
and cultured in Dulbecco‟s Modified Eagle Medium with 10% fetal bovine serum and
antibiotics. MEFs were cultured until 80% confluent and starved for 48 hours prior to
treatment. Cells were incubated for 30 minutes with 0.03M, 0.1M or 1.0M Oleoyl-L-
α-lysophosphatidic acid sodium salt (LPA, Sigma L7260) dissolved in sterile water.
MEFs were incubated with 50ng/ml recombinant human EGF (BD Biosciences) for 15
minutes as positive controls of EGFR phosphorylation and ERK1/2 phosphorylation.
Where indicated, MEFs were pre-incubated with 10µM TNF Alpha Protease Inhibitor-1
(TAPI-1, Peptides International) or 1µM Erlotinib dissolved in DMSO for 1 hour, then
incubated with the indicated concentrations of LPA for 15 minutes in the presence of
inhibitor. Two independent fibroblast lines from wildtype embryos and four lines from
Timp3−/−
embryos were stimulated with LPA as described to measure EGFR signaling.
For amphiregulin neutralization experiments, Timp3−/−
MEFs were pre-treated with
50.0g/ml IgG or amphiregulin neutralizing antibody (R & D Systems) for 2 hours. Cells
were then treated with 1.0M LPA for an additional 5 minutes. Media and cell lysate
were collected for analysis.
2.5.6 Immunoblotting
Total liver protein was extracted by mortar and pestle homogenization of frozen
tissue. RIPA extraction buffer containing 25mM Tris-HCl pH 7.6, 1% Triton X-100,
0.1% SDS, 1% NP-40, 1% sodium deoxycholate, 5mM EDTA, 50mM NaCl, 200µM
Na3VO4, 2mM PMSF, and appropriate dilution of Complete Mini, EDTA-free protease
87
inhibitor cocktail tablets (Roche) was used to lyse all tissue and cells, and lysate was
stored at -70oC. 15-50g of protein was loaded on SDS-PAGE gels for western blotting.
The following anti-mouse antibodies were used: anti-phosphorylated JNK (p-JNK,
Thr183/Tyr185), anti-JNK , anti-phosphorylated p44/42 MAP kinase (p-ERK1/2,
Thr202/Tyr204), anti-p44/42 MAP kinase (ERK1/2), anti-p38 MAPK, anti-
phosphorylated p38 MAPK (p-p38, Thr180/Tyr182), anti-EGFR, anti-phosphorylated
EGFR (p-EGFR, Tyr1173), anti-AMPK, anti-phosphorylated AMPK (p-AMPK,
Ser108), anti-AKT, anti-phosphorylated AKT (p-AKT, Ser473), anti-Caspase 3, anti-
Caspase 9 (all from Cell Signaling), anti-Caspase 8 (R & D Systems), TACE/ADAM17
specific for the disintegrin domain (Khokha Lab), and anti--Actin (Santa Cruz).
2.5.7 Caspase activity assays
Fluorometric caspase activity assays were performed using 15g of protein and
20M fluorescent substrates (Caspase 3/7: Ac-DEVD-AMC, Caspase 9: Ac-LEHD-
AMC, AnaSpec Inc.). Protein and substrates were incubated for 60 min and fluorescence
emission measured at 352 nm every 5 minutes for 1 hour to obtain kinetic measurement
of enzymatic activity.
2.5.8 Electrophoretic Mobility Shift Assay (EMSA)
NF-B activity in liver tissue was determined by EMSA using a common NF-B
biotinylated nucleic acid probe (Panomics) on nuclear fractions of livers. Briefly, liver
tissue was homogenized in Buffer A (10mM HEPES pH 7.9, 1.5mM MgCl2, 10mM KCl,
88
protease inhibitors) and centrifuged to isolate nuclear pellet. The pellet was resuspended
in Buffer B (20mM HEPES pH 7.9, 1.5mM MgCl2, 420mM NaCl, 0.2mM EDTA, 25%
v/v glycerol, protease inhibitors). After centrifugation, the supernatant containing nuclear
extract was collected and stored at -70oC. EMSA was performed according to supplier‟s
protocol (Panomics).
2.5.9 Enzyme-Linked ImmunoSorbent Assay (ELISA)
For serum ELISA, blood was collected from mice by cardiac puncture upon
sacrifice at indicated timepoints. Serum was isolated from whole blood by centrifugation
(BD Vaccutainer CPT cell preparation tube). Serum was diluted 1:200 for TNFR1 and
TNFR2 ELISA, and 1:10 for Fas, Fas ligand, and TNF. Undiluted media was used for
assays on cell culture experiments. ELISA was performed as per manufacturers‟
instructions (TNFR1, Fas, Fas ligand, Amphiregulin, HB-EGF, TGF from R & D
Systems; TNFR2, TNF from BD Pharmingen).
2.5.10 Histology & Immunohistochemistry
Liver lobes were fixed in 5% formalin for 24 hours, transferred to PBS and
processed for embedding in paraffin. 5-µm sections were placed on Superfrost/Plus
microscope slides (Fisher Scientific) and processed as previously described [76]. The
following stains were applied for immunohistochemical analysis: Harris‟ hematoxylin
(Electron Microscopy Sciences), Eosin (Fisher Scientific). Anti-Cleaved Caspase 3
antibody (Cell Signaling) was used according to manufacturer‟s specifications.
89
2.5.11 RNA preparation and quantitative RT-PCR
RNA was prepared from frozen liver tissue using TRIzol reagent (Invitrogen)
according to manufacturer‟s instructions. cDNA was obtained using a first-strand cDNA
synthesis protocol (Promega). Gene expression was measured using SYBR Green reagent
(Quanta) or Taqman primer/probesets (Applied Biosystems) in a 7800HT Real-time PCR
system (Applied Biosciences). All gene expression levels were normalized to -actin
(SYBR Green) or 18S (Taqman) and fold-change measured relative to control wildtype
samples. qPCR of RNA was used as a negative control. Amount of each product was
calculated using the 2-CT
method. Primer sequences and Taqman primer/probeset IDs
are provided in Table 2.1
2.5.12 Statistical Analyses
Data are reported as mean ± S.D. or S.E.M. All calculations were carried out
using GraphPad Prism software (GraphPad Software). Comparisons were made by two-
tailed Student‟s t-test and analysis of variance; comparisons between Kaplan-Meier
survival curves were made by log-rank test.
Table 2.1Primer sequences used for quantitative RT-PCR analysis.
Gene Forward & Reverse Primer Sequence
Timp1 F: 5‟-CATGGAAAGCCTCTGTGGAT-3‟
R: 5‟-CTCAGAGTACGCCAGGGAAC-3‟
Timp2 F: 5‟-GTCATTGCTGCCTTCCTCTC-3‟
R: 5‟-AAAGGGGTGAAGAATGGCTT-3‟
Timp3 F: 5‟-AAACATCTGCCTGGGTTCAG-3‟
R: 5‟-CAAGCTTCCAGCCAAACTTC-3‟
90
Timp4 F: 5‟-ACCTCCGGAAGGAGTACGTT-3‟
R: 5‟-TTATCTGGCAGCAACACAGC-3‟
Adam9 F: 5‟-TTGCATCCATTGTTGCTCAT-3‟
R: 5‟-CTTCTCAAAGTCCTCCGCAC-3‟
Adam10 F: 5‟-GCAACATCTGGGGACAAACT-3‟
R: 5‟-TTGCACTTGTCACTGTAGCC-3‟
Adam12 F: 5‟-AGAGAAAGGAGGCTGCATCA-3‟
R: 5‟-GCCTGCTTGACCTCTGGTAG-3‟
Adam17 F: 5‟-AGGATGCTTGGGATGTGAAG-3‟
R: 5‟-CTGTTTGCTCTGGGAGAACC-3‟
Mmp2 F: 5‟-ACTGACACTGGTACTGGCCC-3‟
R: 5‟-GTCACGTGGTGTCACTGTCC-3‟
Mmp9 F: 5‟-CCCGCTGTATAGCTACCTCG-3‟
R: 5‟-CTGTGGTTCAGTTGTGGTGG-3‟
Mmp14 F: 5‟-CTGTCCCAGATAAGCCCAAA-3‟
R: 5‟-GCATTGGGTATCCATCCATC-3‟
Il1b F: 5‟-AAGCCTCGTGCTGTCGGACC-3‟
R: 5‟-CCAGCTGCAGGGTGGGTGTG-3‟
Il6 F: 5‟-TAAGCTGGAGTCACAGAAGGAGTGGC-3‟
R: 5‟-AGGCATAACGCACTAGGTTTGCCGA-3‟
Tnf F: 5‟-ATCCGCGACGTGGAACTGGC-3‟
R: 5‟-AGAAGAGCGTGGTGGCCCCT-3‟
Tnfrsf1a (Tnfr1) F: 5‟-AGCATGCTGGAAGCCTGGCG-3‟
R: 5‟-TGGACGAGGGGGCGGATTT-3‟
Tnfrsf1b (Tnfr2) F: 5‟-TTCCCAAGCCAGCGCCACAG-3‟
R: 5‟-GCCAGCTTGGCTGGGCTTCA-3‟
Actb (-actin) F: 5‟-AGGGAAATCGTGCGTGACAT-3‟
R: 5‟-GAACCGCTCGTTGCCAATAG-3‟
Mcl1 Taqman primer/probe set Assay ID: Mm01257351_g1
Birc3 (Ciap1/2) Taqman primer/probe set Assay ID: Mm01168413_m1
Xiap Taqman primer/probe set Assay ID: Mm00776505_m1
Bcl2l1 (Bcl-xL) Taqman primer/probe set Assay ID: Mm00437783_m1
Rn18s (18S) Taqman primer/probe set Assay ID: Hs99999901_s1
Timp3 genotyping
primers
WT: 5‟-AGTTGCAGAAGGCATCCTGGGGATGGCT-3‟
Anchor: 5‟-
CAAGAATCTTCTTCTCCCGCTTCTCCGCTT-3‟
Neo3: 5‟-CCAAATTAAGGGCCAGCTCATTCCTCCCA-
3‟
91
3 CHAPTER 3
Stromal TIMP3 regulates liver lymphocyte populations and provides protection
against Th1 T-cell driven autoimmune hepatitis
A version of this manuscript has been submitted for peer review:
Aditya Murthy*, Yang Washington Shao
*, Christopher Wedeles, David Smookler, Rama
Khokha. Stromal TIMP3 regulates liver lymphocyte populations and provides protection
in a mouse model of autoimmune hepatitis.
* These authors contributed equally to this work.
Author contributions:
Murthy, A. – designed and performed in vivo experiments, wrote manuscript
Shao, Y.W. – designed and performed flow cytometry experiments, critiqued manuscript
Wedeles, C. – Performed CD4+ T cell culture experiments and RT-PCR analysis
Smookler, D. – Assisted with bone marrow chimera experiments
92
3.1 Abstract
Lymphocyte infiltration into epithelial tissues and pro-inflammatory cytokine
release are key steps in autoimmune disease. While cell-autonomous roles of
lymphocytes are well understood in autoimmunity, much less is known about the stromal
factors in these tissues that dictate immune cell function. Tissue inhibitor of
metalloproteinases 3 (TIMP3) controls systemic cytokine bioavailability and signaling by
inhibiting the ectodomain shedding of cytokines and their receptors. Here we show that
TIMP3 produced by the hepatic stroma regulates the basal lymphocyte populations in the
liver and the hepatic response to autoimmunity. TIMP3 deficiency led to spontaneous
accumulation and activation of hepatic CD4+, CD8
+ and NK T cells. Treatment with
concanavalin A resulted in a greatly enhanced T helper 1 (Th1) cytokine response and
acute liver failure in a model of autoimmune hepatitis, which mechanistically depended
on TNF signaling. Bone marrow chimeras demonstrated that TIMP3 in the stromal rather
than hematopoietic compartment was responsible for protecting against autoimmunity.
These results uncover metalloproteinase inhibitors as critical stromal factors in regulating
cellular immunity during autoimmune hepatitis.
3.2 Introduction
As a primary site of pathogen encounter, the liver microenvironment has a unique
makeup of local immune cells. The epithelial and stromal cells of the liver such as
hepatocytes, endothelial cells, stellate cells can additionally act as non-classical APCs
and generate exquisite sensitivity to entering pathogens without arming liver dendritic
93
cells [293]. Autoimmunity involves infiltration of lymphocytes into the liver and T cell
activation, which accelerates disease progression during viral, autoimmune or toxin-
induced hepatitis [293, 294]. This process typically relies on a pro-inflammatory cytokine
milieu that culminates in hepatotoxicity [295, 296]. The lectin concanavalin A (con A)
causes polyclonal T cell activation and is used as an experimental means of inducing
autoimmune hepatitis which recapitulates several aspects of the human disease. The
pathological and protective roles played by macrophages, NK/NKT cells, and the known
T cell subsets are well investigated in this model. Currently, the stromal factors
responsible for regulating peripheral immune cell homeostasis and cytokine milieu in the
liver are largely unknown [239, 297-299].
Tissue inhibitors of metalloproteinases (TIMPs) comprise a well-known family of
soluble factors, in which the 4 Timp genes post-translationally regulate the enzymatic
activity of all metalloproteinases in the mammalian genome. Metalloproteinases perform
ectodomain shedding in which chemokines, cytokines, growth factors and their receptors
are cleaved from the cell surface. This process regulates leukocyte migration and
cytokine signaling during acute and chronic inflammation [300]. Furthermore, tissue
remodeling by the TIMP/metalloproteinase axis plays a key role in dictating immune cell
function and wound healing during liver injury [301]. We and others have previously
demonstrated that TIMP3 functions in a context-dependent manner to direct cytokine
signaling during acute and chronic hepatic stress [69, 76, 253, 302]. However, the
physiological role of TIMPs in cellular immunity remains completely unknown. In this
study, we show that TIMP3 deficiency results in spontaneous lymphocyte infiltration into
94
the liver. Con A administration to TIMP3 deficient mice results in enhanced Th1
cytokine production and TNF-dependent liver damage. Finally, utilizing bone marrow
chimeras we identify that TIMP3 derived from the stromal sources provides
hepatoprotection against con A-induced autoimmunity.
3.3 Results
3.3.1 A basal increase in CD4+ T cell and NKT cell populations in Timp3
−/−
livers
We investigated the effect of Timp3 deficiency on steady-state composition of
immune cells in the periphery by characterizing the lymphocyte subsets of liver and
spleen from wildtype (WT) and Timp3−/−
mice. Flow cytometry analysis of liver
lymphocytes showed a relative increase in CD4+ T cells and NKT cells (characterized as
CD3+NK1.1
+ cells; Figure 3.1A). Specifically, absolute cell counts of basal CD4
+ T cells
and NKT cells were over two-fold higher and this was reflected in an overall increase in
the number of mononuclear cells (MNCs; Figure 3.1B). On the other hand, splenic
numbers of the same subsets, along with total MNCs were comparable between wildtype
and Timp3−/−
mice in the basal state (Figure 3.1C, D). Given that lymphocyte activation
precedes infiltration into the periphery [297, 299, 303], we next analyzed the activation
marker CD69 on splenocytes and liver lymphocytes by flow cytometry. A higher number
of CD69 expressing CD4+ and CD8
+ T cells as well as NKT cells were observed only in
resting livers of Timp3−/−
mice, while comparable numbers existed in the spleen (PBS
treated groups, Figure. 3.1E, F). Thus loss of Timp3 results in a spontaneous basal
95
Figure 3.1 Timp3−/−
mice exhibit spontaneous accumulation and activation of hepatic
CD4+ T cells and NKT cells, and enhanced sensitization to con A-induced hepatitis.
(A, B) Flow cytometry analysis and quantification of cell surface CD4/CD8 expression gated on CD3
+
lymphocytes, and CD3/NK1.1 resting lymphocytes from livers of 12-week old wildtype (WT) and
Timp3−/−
mice. (C, D) Flow cytometry analysis and quantification of cell surface CD4/CD8 expression
gated on CD3+ splenocytes, and CD3/NK1.1 resting splenocytes of 12-week old wildtype (WT) and
Timp3−/−
mice. (E, F) Counts of activated liver lymphocyte (E) and splenocyte (D) subsets gated on CD69+
cells. (G) Survival curve of mice treated with 10mg/g con A, then monitored for 48h.(H) Serum
transaminase (ALT) levels in WT and Timp3−/−
mice treated with PBS or Con A. (I) Liver histology (H &
E) of WT and Timp3−/−
mice treated with PBS or con A for 18h. Arrows depict necrosis. *P<0.05, mean ±
S.D. for B, D-F (n=4); *P<0.01, mean ± S.E.M. for H (n=4). Bar = 100m. Flow cytometry plots and
histology are representative of at least 4 mice per condition. Survival curve is based on n=11 mice per
genotype); *P<0.01 (log-rank test).
96
increase of total liver lymphocytes that are comprised of higher numbers of activated
CD4+, CD8
+ and NK T cells.
3.3.2 TIMP3 deficiency sensitizes mice to T-cell mediated hepatitis induced
by concanavalin A
We have previously found that Timp3−/−
mice have enhanced sensitivity to LPS-
induced hepatitis caused by increased TNF bioactivity but exhibit resistance to Fas-
mediated fulminant hepatitis due to compound alterations in TNFR1 and EGFR signaling
[69, 302]. These models rely on resident macrophage (Kupffer cell) activity and
hepatocyte-intrinsic responses to apoptotic stimuli, but do not address lymphocyte
function in modulating hepatitis. Intravenous administration of con A offers a model of
autoimmune hepatitis induced by lymphocytes [304, 305]; we therefore examined the
susceptibility of Timp3−/−
mice to con A-induced hepatitis. We observed significantly
greater induction of splenomegaly accompanied by an increase in CD69 expressing CD4+
and CD8+ T cells in Timp3
−/− mice. These mice also exhibited higher numbers of
activated liver lymphocytes, specifically CD4+, CD8
+ T cells and NKT cells (con A
treated groups, Figure 3.1E, F). Significantly more Timp3−/−
mice succumbed to a low
dose (10g/g) of con A compared to controls (Figure 3.1G). A greater increase in serum
transaminase (ALT) levels at 6h and 18h indicated enhanced hepatotoxicity (Figure
3.1H). Histological examination showed severe liver damage with extensive necrosis in
97
Timp3−/−
mice 18h after treatment with con A (Figure 3.2A). Thus, TIMP3 deficiency
exacerbates autoimmune hepatitis in mice.
3.3.3 Enhanced TNF signaling and Th1 cytokine response drives liver
damage in Timp3−/−
mice
TIMP3 is a negative regulator of TNF shedding in vivo, and this cytokine is a key
contributor to con A-induced hepatitis [69, 76, 81, 238, 239, 295, 304]. Immunoblotting
showed phosphorylation of the TNF signaling effector JNK/SAPK at early (6h) and late
(18h) timepoints after con A administration in Timp3−/−
livers (Figure 3.3A). We tested
whether increased TNF bioactivity underlies accelerated induction of autoimmune
hepatitis in Timp3−/−
mice and found that loss of Tnf in wildtype and Timp3−/−
backgrounds protected against hepatotoxicity as indicated by lower serum ALT (Figure
3.3B). Histological analyses confirmed negligible liver damage and hemorrhaging
(Figure 3.3C), and TUNEL staining revealed an absence of hepatocyte apoptosis in
Timp3−/−
Tnf−/−
versus Timp3−/−
mice (Figure 3.3D).
Analysis of early serum cytokine release over 6 hours following con A
administration showed accelerated kinetics of cytokine release in Timp3−/−
mice
compared to all control groups. Tnf−/−
and compound loss of Timp3−/−
/Tnf−/−
resulted in
significant abrogation of cytokine release (Figure 3.3E). Of the Th1 cytokines, TNF
peaked at 2h remaining high until 6h; IFN exhibited a transient 5-fold greater level at
4h; IL-6 continued to rise dramatically over 6h and the Th2 cytokine IL-4 peaked at 2h.
98
Figure 3.2Enhanced necrosis, lymphocyte infiltration and macrophage activity in
Timp3−/−
livers.
(A) Histology (H & E) of lymphocyte infiltration and enhanced necrosis (arrows) in Timp3
−/− livers 18h
following con A administration. Insets depict lymphocyte infiltration. (B) Immunohistochemical staining
for F4/80 depicting increased macrophage association with areas of necrosis in Timp3−/−
livers 18h
following con A administration. Bars = 100m. Images are representative of at least 3 mice per condition.
99
Figure 3.3 Loss of Timp3 enhances TNF signaling and Th1 cytokine response during
autoimmune hepatitis.
(A) Immunoblots of JNK phosphorylation in liver lysates from WT and Timp3
−/− mice treated with PBS or
con A. (B) Serum transaminase (ALT) levels 6 hours after treatment with PBS or con A. (C) H & E of
livers treated with PBS or con A. Arrowheads depict necrosis. (D) TUNEL staining of livers 6 hours after
PBS or con A treatment. Brown nuclei depict apoptotic cells (E) Th1 serum cytokine levels over 6 hours
following con A administration. (F) Serum levels of indicated cytokines and chemokines/growth factors
measured 18 hours after PBS or con A administration. Arrowheads depict non-detectable cytokine levels.
*P≤0.04, mean ± S.E.M. (n=4). Bars = 25m. Histology is representative of at least four mice per
condition.
100
IL-6 plays paradoxical roles in acute hepatic inflammation depending on the duration of
its bioavailability [306, 307], and IL-4 dependent signaling through STAT6 is required
for promoting T-cell mediated hepatitis [305, 308]. Together, these data demonstrate that
TIMP3 deficiency enhances the Th1 cytokine response during autoimmune hepatitis.
Indeed, Timp3−/−
mice displayed sustained increase in Th1 cytokines 18h after con A
treatment. We also identified increases in factors that promote the Th1 response, namely
IL-1, IL-12 and the chemoattractant MCP-1 (Figure 3.3F). Of note, IL-12 release by
activated macrophages promotes T cell activation [299, 309], and consistent with this we
observed greater association of macrophages to areas of necrosis in Timp3−/−
livers
(Figure 3.2B).
We observed increased serum IL-17A and IL-10 in Timp3−/−
Tnf −/−
mice, which
is particularly intriguing since loss of Timp3 alone only modestly affected these cytokines
(Figure 3.4A). The role Th17 cells is still emerging in autoimmune hepatitis and is
primarily associated with hepatoprotective IL-22 production [310]. IL-10 is also
suggested to be protective through its suppression of the Th1 response and its
involvement in immunological tolerance [298]. Additionally we measured decreases in
splenic and hepatic CD4+ T cell numbers, consistent with abrogated liver damage in
Timp3−/−
Tnf −/−
mice (Figure 3.4B). Further studies are needed to elucidate how
metalloproteainse inhibitors impact these immunoregulatory cytokines.
101
Figure 3.4Timp3−/−
Tnf −/−
compound deletion elevates serum IL-10, IL-17A and
depletes splenic and hepatic CD4+ T cells.
(A) Serum levels of IL-10 and IL-17A following con A treatment in mice of indicated genotypes. (B)
Counts of CD3+CD4
+ T cells in spleen and liver of mice 6 hours after treatment with PBS or con A.
*P≤0.05 vs. WT, mean ± S.E.M. (n≥3).
102
Figure 3.5 – TIMP3 is dispensable for cell-autonomous activation and proliferation of
CD4+ T cells.
(A) Gene expression of indicated Timps, Adams and Mmps in resting CD4+ T cells isolated from spleens of
wildtype mice. Data are one of three independent analyses. (B) Proliferation of WT and Timp3−/−
CD4+ T
cells following stimulation with con A over 72 hours. (C, D) Cytokine release of (C) IL-2 and (D) TNF
into culture media of CD4+ T cells stimulated with con A. Data are mean ± S.D. (n=3) and representative of
two independent experiments.
103
3.3.4 Cell-intrinsic TIMP3 is not required for CD4+ T cell activation
Together the 4 TIMPs inhibit metalloproteinase activity of all known 24 MMPs, 9
active ADAMs and 4 active ADAM-TS enzymes, yet the regulation of lymphocyte
function by TIMPs remains uninvestigated [300]. We examined the cell-autonomous role
of TIMP3 in lymphocyte activation in vitro. First, gene expression analysis of Timps and
specific Adam and Mmp genes in resting splenic CD4+ T cells showed low levels of
Timp1, Timp3 and Timp4 but significantly higher expression of Adam10 and Adam17
(Figure 3.5A). Stimulation of cultured wildtype and Timp3−/−
CD4+ T cells (1.0g/ml
con A over 72 hours) showed comparable expansion of CD4+
T cells in both genotypes
when measured by MTS (Figure 3.5B). IL-2 is required for expansion and growth of
antigen-specific T cells subsequent to TCR ligation, and its concentration was
comparable between wildtype and Timp3−/−
CD4+ T cell culture media (Figure 3.5C).
TNF levels also remained comparable across all time points (Supplemental Fig. 4D).
These data demonstrate that TIMP3 is dispensable for cell-autonomous activation,
expansion and effector function of CD4+ T cells.
3.3.5 Stromal TIMP3 protects against con A-induced hepatitis
To evaluate the physiological relevance of hematopoietic and non-hematopoietic
TIMP3 in vivo we generated congenic bone-marrow chimeras as depicted in Figure 3.6A.
Wildtype (CD45.1+) recipients were reconstituted with Timp3
−/− (CD45.2
+) bone marrow
cells and Timp3−/−
recipients with wildtype bone marrow, achieving >90% reconstitution
at 8 weeks post-transplant (bar graphs, Figure 3.6A). We observed that only stromal
104
Figure 3.6Stromal TIMP3 protects against a Th1 pro-inflammatory cytokine response
and autoimmune hepatitis.
(A) Schematic depicting generation of radiation chimeras. Stacked graphs indicate comparable bone
marrow reconstitution in WT and Timp3−/−
recipients. (B) Counts of total MNCs and subsets of liver
lymphocytes isolated from radiation chimeras 8 weeks after reconstitution. (C) Serum transaminase (ALT)
levels in WT and Timp3−/−
recipients 6 hours after con A treatment. (D, E) Histological analysis depicting
liver damage (D, H & E) and hepatocyte death (E, TUNEL) 6 hours after con A treatment. (F) Serum levels
of Th1 cytokines over 6 hours following con A treatment. (G) Schematic illustrating the regulation of
lymphocyte infiltration and cytokine release by non-hematopoietic TIMP3. Activation and accumulation of
liver lymphocytes generates a positive-feedback loop and predisposes to autoimmune hepatitis. *P<0.01,
mean ± S.D. for A, B (n=4); *P<0.05, mean ± S.E.M. for C, F (n=3). Bars = 25m. Histology is
representative of at least three mice per condition.
105
TIMP3 deficiency reproduced the spontaneous basal increase in CD3+ liver lymphocytes
(Figure 3.6B) as described in un-irradiated Timp3−/−
mice (Figure 3.1). Treatment with
con A induced serum toxicity and liver damage at 6h in Timp3−/−
recipients harboring
wildtype bone marrow, while wildtype recipients with Timp3 null hematopoietic cells did
not exhibit hepatotoxicity (Figure 3.6C, D). TUNEL staining showed extensive
hepatocyte death only in Timp3−/−
recipients (Figure 3.6E). Further, timecourse analysis
of serum cytokines showed significant elevations in Th1 cytokines IFN, IL-2 and IL-6
over 6h after con A administration (Figure 3.6F) in these mice. Taken together, these data
show that hematopoietic TIMP3 does not impact cytokine release and activation of T
lymphocytes whereas TIMP3 in non-hematopoietic tissues provides protection against
con A-induced hepatitis. Figure 3.6G models the role of Timp genes in autoimmune
hepatitis, demonstrating the requirement of non-hematopoietic TIMP3 in maintaining
hepatic lymphocyte homeostasis and controlling Th1 cytokine response.
3.4 Discussion
Here we demonstrate that the metalloproteinase inhibitor TIMP3 regulates liver
lymphocyte infiltration and pro-inflammatory function during acute hepatitis. Loss of
Timp3 led to spontaneous expansion of liver CD4+ T and NKT cells, and this resulted in
increased hepatotoxicity following con A treatment. Examination of TNF signaling
revealed that elevated serum TNF levels and signaling were causal to enhanced morbidity
in Timp3
mice. We noted that CD4+ T cells isolated from spleen expressed comparably
low levels of Timp genes in wildtype mice; furthermore, wildtype and Timp3
CD4+ T
106
cells exhibited comparable expansion and cytokine production following in vitro
stimulation with con A. We thus hypothesized that TIMP3 generated by non-
hematopoietic tissue (such as the liver parenchyma) inhibited lymphocyte infiltration and
tested this by the generation of bone marrow chimeras. Indeed, Timp3
recipients of
wildtype bone marrow exhibited spontaneous increase of CD3+ cells and total liver
mononuclear cells. In these chimeric mice, induction of con A-mediated hepatitis
reproduced the enhanced cytokine response and hepatotoxicity. Figure 8E models the role
of Timp genes in autoimmune hepatitis, illustrating the requirement of non-hematopoietic
TIMP3 in maintaining hepatic lymphocyte homeostasis and controlling a Th1 cytokine
response.
The liver has a unique role in maintaining tolerance to food antigens and gut flora
that drain in, while generating protection against exogenous biological insults such as
hepatitis B and C virus, bacterial antigens as well as non-biological agents. Kupffer cells,
sinusoidal endothelial cells and even hepatocytes participate in local antigen presentation
and expression of innate immune receptors such as TLR4; this process is postulated to
generate tolerance to frequently encountered antigen [311]. On the other hand, circulating
dendritic cells can contribute to protective pro-inflammatory responses during infection
by recruiting lymphocytes to the liver microenvironment following encounter with a
pathogen [312, 313]. Our findings introduce the role of metalloproteinase inhibitors in
tolerance versus hepatoprotection. The use of more sophisticated systems (e.g. HBV,
HCV and CMV models of viral infection, Listeria monocytogenes and Mycobacterium
tuberculosis models of bacterial infection) will prove useful in delineating the
107
contribution of Timp genes to acute CD8+ cytotoxic T lymphocyte (CTL) or CD4
+ T cell
responses against virus [314, 315], and T-helper subset function against bacterial
granulomas [316]. Given the overabundance of CD4+ T cells and NKT cells observed in
Timp3 null mice, we may find enhanced pathogen clearance and protection in an antigen-
specific manner. Potentially, TIMP3 can regulate the sensitization threshold required for
the cellular immune response in the liver.
Several possible mechanisms may be involved in lymphocyte infiltration upon
loss of TIMP3. Disruption of chemokine gradients is a prominent candidate, as increased
hepatic levels of CC- and CXC- chemokines have shown to promote acute liver injury in
murine models and patients alike [317, 318]. Specifically, increases in liver CCL2 (MCP-
1), CCL3 (MIP-1), CCL4 (MIP-1) and CCL5 (RANTES), and CXCR3 ligands
(CXCL9-11) have been reported in human liver biopsies of acute liver pathologies such
as fulminant hepatitis and acute viral hepatitis [319, 320]. Metalloproteinases MMP1, 2, 3
and 14 can cleave CC-chemokines (CCL2, 7, 8, 13), while CXCL11 is shed by numerous
MMPs (MMP1, 2, 3, 9, 13, 14) [14]. Physiologically, chemokines are immobilized on the
extracellular matrix by binding with glycosaminoglycan-associated proteoglycans such as
syndecan. Syndecan-1 harbors „sinks‟ of CXC-chemokines, and its shedding by MMP7 is
an example of the involvement of metalloproteinases in generating CXCL1 gradients. L-
selectin shedding from lymphocyte surfaces is an important early step in their egress
from lymphoid organs and into the periphery [321]. ADAM17/TACE is the only
metalloproteinase known to shed L-selectin in vivo, and TIMP3 is the only endogenous
inhibitor of ADAM17 [300]. In the current study, loss of Timp3 may potentially generate
108
a permissive environment for L-selectin shedding and uncoordinated lymphocyte egress
into the systemic microenvironment. Finally, TIMP3 is established as a suppressor of
TNF-induced hepatocyte stress signaling during chronic inflammation following partial
hepatectomy or in models of hepatosteatosis and type 2 diabetes [76, 253]. Elevated
production of reactive oxygen species (ROS) is a feature of TNF-mediated stress
signaling and potentially contributes to the immunopathology describe in this study [322,
323].
Immune cell and cytokine composition in the periphery determines the
progression of autoimmunity [324, 325]. While the identification of precise mechanisms
warrants further investigation, this study is the first to reveal a regulatory function for
endogenous metalloproteinase inhibitors in cellular immunity. Taken together, our results
show stromal TIMP3 as a non-cell-autonomous regulator of lymphocyte entry into the
liver microenvironment, implicating its requirement in immunosuppression and
prevention of liver injury during autoimmune hepatitis. Stromal TIMP3 suppresses
undesired lymphocyte infiltration and protects against hepatitis opening avenues to better
elucidate the role of Timps in autoimmune diseases.
109
3.5 Methods
3.5.1 Mice
All mice used in this study were of C57BL/6 background. Male mice aged 12-15
weeks were used for all the experimental procedures. Timp3−/−
mice have been
previously described [76]. Tnf−/−
mice were obtained from the Jackson Laboratory and
were crossed with Timp3−/−
mice to generate Timp3−/−
Tnf−/−
mice. C57BL/6-Ly5.1
(CD45.1) mice were generously provided by Dr. Norman Iscove (Professor, University of
Toronto, Department of Medical Biophysics, MaRS centre, Toronto, Canada). Mice were
fed 5%-fat chow ad libitum, housed and cared for in accordance with the guidelines
approved by the Canadian Council for Animal Care, and the Animal Care Committee of
the Ontario Cancer Institute. Genotypes were confirmed via PCR for the respective
genes; primer sequences for genotyping Timp3−/−
mice are provided in Table 3.1.
3.5.2 Induction of hepatitis & generation of bone marrow chimeras
10µg/g concanavalin A (Sigma) was injected into the tail vein of 12-week old
male mice dissolved in 200µL sterile PBS to induce hepatitis. Congenic bone marrow
transplants were performed by exposing 12-week old recipient mice to a single dose of
9Gy (900 Rd) ionizing radiation, followed by reconstitution with 4x106 donor bone
marrow cells by injection into the tail vein. Bone marrow and peripheral reconstitution
was allowed for 8 weeks prior to induction of hepatitis.
3.5.3 CD4+ T cell culture
110
Isolation of spleen CD4+ T cells is detailed in Supplemental Methods. Cells were
cultured in RPMI 1640 medium containing 10% FBS and 50M -mercaptoethanol.
1x105
CD4+ T cells were stimulated with 1.0ng/mL concanavalin A for 24, 48 and 72
hours. Proliferation was measured using a standard MTS assay. Cell culture supernatant
was obtained at indicated timepoints to measure cytokine levels by ELISA following kit
instructions (R & D Systems).
3.5.4 Serum analysis
Serum was obtained prior to and following concanavalin A injection at 2 hour
intervals by bleeding the tail vein. Hepatotoxicity was assayed by measuring serum levels
of alanine transaminase (ALT) and aspartate aminotransferase (AST). Concentrations of
cytokines were measured using a multiplexed cytokine bead array (CBA) on a BD
FACSArray Bioanalyzer system (BD Biosciences) following manufacturer‟s instructions.
3.5.5 Histology & Immunoblotting
Immunohistochemical staining and immunoblotting of liver lobes were performed
as previously described [302]. Liver lobes were fixed in 4% paraformaldehyde for 24
hours, transferred to 70% ethanol and processed for embedding in paraffin. The following
stains were applied: Harris‟ hematoxylin (Electron Microscopy Sciences), Eosin (Fisher
Scientific), TUNEL staining to measure hepatocyte apoptosis, anti-F4/80 antibody to
visualize macrophages/Kupffer cells. All staining was performed by institutional core
facilities.
111
3.5.6 Flow cytometry
For splenocyte analysis, red blood cells were lysed prior to further processing.
Liver lymphocytes were obtained by density centrifugation using Percoll. Total
splenocytes were isolated as described in Cell Culture above. Liver lymphocytes were
obtained by density centrifugation using Percoll. Briefly, livers were resuspended in 20ml
of 40% Percoll dissolved in William‟s E medium and passed through a 100um nylon
mesh to create single cell suspensions. This was underplayed with 70% Percoll dissolved
in sterile Hank‟s Buffered Salt Solution (HBSS) and centrifuged to generate an interface
containing liver lymphocytes. The interface was removed and processed for flow
cytometry. The following antibodies were used for flow cytometry (from eBiosciences):
anti-CD3 (clone 145-2C11), anti-CD4 (clone GK1.5), anti-CD8 (clone 53-6.7), anti-B220
(clone RA3-6B2), anti-NK1.1 (clone PK136), anti-Gr-1 (clone RB6-8C5), anti-CD11b
(clone M1/70), anti-F4/80 (clone BM8), anti-CD62L (clone MEL-14), anti-CD44 (clone
IM7), and anti-CD69 (clone H1.2F3). Viability was measured by Propidium Iodide (PI)
uptake. Data was obtained on a BD FACSCalibur system and analyzed using FlowJo
software.
3.5.7 Cell culture
CD4+ T cells were isolated from spleens of mice by negative selection using
magnet-assisted cell sorting (MACS, Miltenyi Biotec). Briefly, splenocytes were
obtained by passing whole spleen through a 70m mesh, followed by red blood cell lysis
in AFCS buffer. Cells were resuspended in PBS containing 0.5% BSA and 2mM EDTA.
112
Dead cells were removed prior to sorting by using the Dead Cell Removal Kit (Miltenyi
Biotec). Splenocytes were incubated with a cocktail of biotinylated antibodies
against CD8a, CD11b, CD11c, CD19, CD45R (B220), NK1.1, DX1, CD105, CD69,
MHC-class II, and Ter-119 (eBiosciences), followed by incubation with magnetically-
labeled microbeads. Magnetic separation was performed using LS columns on a magnetic
field (Miltenyi Biotec). >95% Purity of CD4+ T cell isolation was verified by flow
cytometry.
3.5.8 Immunoblotting
Total liver protein was extracted by mortar and pestle homogenization of frozen
tissue. RIPA extraction buffer containing 25mM Tris-HCl pH 7.6, 1% Triton X-100,
0.1% SDS, 1% NP-40, 1% sodium deoxycholate, 5mM EDTA, 50mM NaCl, 200µM
Na3VO4, 2mM PMSF, and appropriate dilution of Complete Mini, EDTA-free protease
inhibitor cocktail tablets (Roche) was used to lyse all tissue and cells, and lysate was
stored at -70oC. 30g of protein was loaded on SDS-PAGE gels for western blotting. The
following anti-mouse antibodies were used: anti-phosphorylated JNK (p-JNK,
Thr183/Tyr185), anti-JNK (Cell Signaling).
3.5.9 RNA preparation and quantitative RT-PCR
RNA was prepared from sorted splenic CD4+ T cells using TRIzol reagent (Invitrogen),
and cDNA was obtained using a qScript cDNA supermix protocol (Quanta Biosciences).
Gene expression was measured using SYBR Green reagent (Quanta Biosciences) in a
113
7800HT Real-time PCR system (Applied Biosciences). All gene expression levels were
normalized to Actb (-actin). Relative amount of each product was calculated using the 2-
CT method. Primer sequences are provided in Table 3.1.
3.5.10 Statistical Analyses
Data are reported as mean ± S.D. or S.E.M. Comparisons were made by two-
tailed Student‟s t-test and analysis of variance; comparisons between Kaplan-Meier
survival curves were made by log-rank test.
Table 3.1 – Primer sequences for quantitative RT-PCR analysis and genotyping.
Gene Forward & Reverse Primer Sequence
Timp1 F: 5‟-CATGGAAAGCCTCTGTGGAT-3‟
R: 5‟-CTCAGAGTACGCCAGGGAAC-3‟
Timp2 F: 5‟-GTCATTGCTGCCTTCCTCTC-3‟
R: 5‟-AAAGGGGTGAAGAATGGCTT-3‟
Timp3 F: 5‟-AAACATCTGCCTGGGTTCAG-3‟
R: 5‟-CAAGCTTCCAGCCAAACTTC-3‟
Timp4 F: 5‟-ACCTCCGGAAGGAGTACGTT-3‟
R: 5‟-TTATCTGGCAGCAACACAGC-3‟
Adam9 F: 5‟-TTGCATCCATTGTTGCTCAT-3‟
R: 5‟-CTTCTCAAAGTCCTCCGCAC-3‟
Adam10 F: 5‟-GCAACATCTGGGGACAAACT-3‟
R: 5‟-TTGCACTTGTCACTGTAGCC-3‟
Adam12 F: 5‟-AGAGAAAGGAGGCTGCATCA-3‟
R: 5‟-GCCTGCTTGACCTCTGGTAG-3‟
Adam17 F: 5‟-AGGATGCTTGGGATGTGAAG-3‟
R: 5‟-CTGTTTGCTCTGGGAGAACC-3‟
Mmp2 F: 5‟-ACTGACACTGGTACTGGCCC-3‟
R: 5‟-GTCACGTGGTGTCACTGTCC-3‟
Mmp9 F: 5‟-CCCGCTGTATAGCTACCTCG-3‟
R: 5‟-CTGTGGTTCAGTTGTGGTGG-3‟
Mt1-mmp F: 5‟-CTGTCCCAGATAAGCCCAAA-3‟
R: 5‟-GCATTGGGTATCCATCCATC-3‟
114
Timp3 genotyping
primers
WT: 5‟-AGTTGCAGAAGGCATCCTGGGGATGGCT-3‟
Anchor: 5‟-
CAAGAATCTTCTTCTCCCGCTTCTCCGCTT-3‟
Neo3: 5‟-CCAAATTAAGGGCCAGCTCATTCCTCCCA-
3‟
115
4 CHAPTER 4
ADAM17 activation of epidermal Notch regulates atopic barrier immunity and
myeloproliferation by suppressing c-Fos-driven transcription of epithelial cytokines
A version of this chapter is currently under revision at Immunity:
Aditya Murthy, Yang Washington Shao, Swami R. Narala, Sam D. Molyneux, Juan
Carlos Zúñiga-Pflücker, Rama Khokha.
ADAM17 activation of epidermal Notch
regulates atopic barrier immunity and myeloproliferation by suppressing c-Fos-driven
transcription of epithelial cytokines.
Author contributions:
Murthy, A. – designed and performed majority of the experiments, wrote manuscript
Shao, Y.W. – performed flow cytometry experiments
Narala, S.R. – assisted with chromatin immunoprecipitation experiments
Molyneux. S.D. – analyzed human gene expression datasets
Zúñiga-Pflücker J.C. – provided expertise for ligand dependent Notch activation assays
116
4.1 Abstract
Epithelial cells of mucosal tissues provide a barrier against environmental stress,
and keratinocytes are key decision-makers for immune cell function in the skin.
Currently, epithelial signaling networks that instruct barrier immunity remain
uncharacterized. Here we show that constitutive or postnatal keratinocyte-specific
deletion of a disintegrin and metalloproteinase 17 (Adam17) triggers T helper 2 (Th2)-
driven atopic dermatitis along with myeloproliferative disease. In vivo and in vitro
deficiency of ADAM17 dampened Notch signaling, increasing production of Th2-
polarizing cytokine TSLP and myeloid growth factor G-CSF. Ligand-independent Notch
activation was identified to regulate AP-1 transcriptional activity, with Notch specifically
antagonizing c-Fos recruitment to the promoters of Tslp and Csf3 (G-Csf). Further, local
skin inflammation was rescued and myeloproliferative disease ameliorated by adenoviral
delivery of active Notch to Adam17
epidermis. Our findings uncover an essential role
of ADAM17 in the adult epidermis and show the gatekeeper function of the
ADAM17/Notch/c-Fos triad in barrier immunity.
4.2 Introduction
Inflammatory skin diseases afflict greater than 10% of people in the western
world [326]. The onset of epidermal inflammation represents a collapse of epithelial
homeostasis resulting in the generation of alarm signals by keratinocytes. This in turn
recruits the immune system to eliminate infected and/or damaged cells and restore
homeostasis. Factors produced by epithelial cells that guide immune cell functions are
117
being recognized as the „epimmunome‟ and it is important to elucidate the mechanisms
that underpin their synthesis. Stress response pathways (e.g. NF-B, AP-1) have emerged
as key junctions in the maintenance of epithelial cell homeostasis [327]. Disruption of
either developmental programming or induction of physiological stress in keratinocytes
leads to copious production of „alarmins‟ including thymic stromal lymphoprotein,
(TSLP), interleukin 33 (IL-33), as well as hematopoietic growth factors such as
granulocyte colony stimulating factor (G-CSF) or granulocyte/macrophage colony
stimulating factor (GM-CSF). These over time generate atopic dermatitis by TSLP-
mediated activation of dermal dendritic cells, which results in a Th2-polarized local
cellular immune response [328, 329]. Additionally, overt production of myeloid growth
factors by keratinocytes has shown to be causal to the development of myeloproliferative
disease (MPD) in mouse models of inflammatory skin disease [330, 331]. Therefore,
keratinocytes in the epidermis direct local immunity and hematopoietic development, yet
epithelial signaling networks which instruct the epimmunome are poorly defined.
The Notch pathway is a master regulator of cell fate decisions and has established
roles in directing differentiation of all cell layers in metazoans. Recent emerging
evidence also supports a role for this pathway in the maintenance of adult tissue
homeostasis. In epithelial tissues such as the skin, Notch signaling sustains the epidermal
barrier by promoting cell-autonomous keratinocyte terminal differentiation [332].
Importantly it also orchestrates communication between the epithelial and immune
compartments of the skin to prevent chronic inflammation. Recent studies of Notch in
adult tissue homeostasis have highlighted its requirement in preventing inflammatory
118
skin disease [330, 333]; however, the molecular mechanisms by which epithelial Notch
performs this immunoregulatory function remain elusive.
ADAM17/TACE is a transmembrane metalloproteinase that cleaves cell surface
proteins [133, 300]. It is essential for postnatal survival in mice and is implicated in
immune cell development and function [201, 203, 286]. ADAM-mediated shedding of
Notch receptors is a key step in activating Notch signaling, but to date only ADAM10 has
been shown to perform this action in vivo, as demonstrated by its role in marginal zone B
cell development [123]. ADAM10 also performs S2 cleavage of Notch in keratinocytes,
and epidermal ADAM10 deficiency leads to defective barrier formation [128]. However
the relevance of ADAM17 in epithelial:immune crosstalk is unknown. We investigated
this by inactivating Adam17 in the epidermis of adult mice. The main outcomes were
spontaneous onset of Th2-driven atopic dermatitis (AD) and myeloproliferative disease
(MPD). Keratinocyte-specific deletion of Adam17 triggered spontaneous accumulation of
Th2 cells, dermal mast cell infiltration and elevated serum IgE. An increase in Gr-
1+CD11b
+ cells and concomitant loss of B220
+ B lymphocytes was observed in the bone
marrow. Using in vivo and in vitro approaches, we found that ADAM17 is crucial for
basal Notch activation in adult epidermis, and that ectopic Notch activation was sufficient
to rescue local skin inflammation and MPD in mutant mice. We further identified
elevated activator protein-1 (AP-1) transcriptional activity as causal to TSLP and G-CSF
production. Analysis of signal crosstalk between Notch and AP-1 pathways in mouse and
human keratinocytes revealed that Notch antagonized c-Fos recruitment to epithelial
cytokine promoters to keep stress signaling in check. Collectively, our results
119
demonstrate that ADAM17 permits tonic Notch activation in the adult epidermis to
regulate epithelial cytokine production and maintain barrier immunity.
4.3 Results
4.3.1 Onset of spontaneous atopic dermatitis upon deletion of epidermal
Adam17
Keratinocytes comprise the vast cellular majority of the epidermis. Basal
keratinocytes reside at the epidermal-dermal interface, and migrate outward towards the
skin surface as they undergo terminal differentiation. We induced Adam17 deletion in
basal keratinocytes by breeding Adam17fl/fl
and K14-Cre transgenic mice (Figure 4.1A).
These mice (termed Adam17ep
) exhibited a progressive atopic-dermatitis like
inflammatory skin disease with 100% penetrance as early as 3 weeks of age (Figure
4.1B). We also measured a marked increase in thymic stromal lymphoprotein (TSLP) and
IL-33, both in the epidermis and systemically (Figure 4.1C). H & E staining showed
epidermal thickening, a loss of subcutaneous fat and regions of hyperkeratosis, while
immunohistochemical analysis with Ki67 revealed keratinocyte hyperproliferation
(Figure 4.1D, E). Histological and flow cytometry analysis revealed massive infiltration
by F4/80+ macrophages (Figure 4.1F, G; FACS quantification in scatterplot to the right)
and Toluidine-B+ mast cells in the dermis (arrows, Figure 4.1H; quantification in
scatterplot to the right). Consistent with mast cell influx, serum IgE levels were >100-
120
Figure 4.1Epidermal ADAM17 deficiency causes atopic dermatitis (AD).
(A) Immunoblot performed on tail epidermis and dermis to confirm tissue-specific Adam17 deletion. (B)
Adam17fl/fl
(fl/fl) and K14-Cre transgenic mice were bred to generate tissue-specific deletion of Adam17 in
basal keratinocytes (ep mice). (C) Serum and tail epidermis analyzed by ELISA for levels of TSLP and
IL-33. (D, E) Histological analysis performed on dorsal skin (D, H&E) and keratinocyte proliferation (E,
Ki67 immunostaining). (F, G) Immunohistochemical and flow cytometry (FACS) quantification for F4/80
to measure macrophage infiltration in dermis. (H) Toluidine-B staining performed to measure mast cell
infiltration (red arrows), quantified on the right. (I) Serum IgE measured by ELISA. (J, K) Bead array
profiling of serum for indicated cytokines and chemokines. *P<0.05 mean ±S.E.M (n≥4). Histology and
immunoblotting data are representative at least six mice. Scale bars = 100m.
121
fold higher in Adam17ep
mice (Figure 4.1I). These data indicated spontaneous onset of
AD in mice lacking epidermal ADAM17.
The production of Th2 cytokines typifies atopic diseases [334]. We observed
dramatic changes in serum levels of specific factors; most notably the elevation of Th2
(IL-4, IL-15) and reduction in Th1 (IFN, IL-12p40) cytokines in 10-week old mutant
mice (Figure 4.1J). IL-17A, a major contributor to autoimmunity [335] and a marker of
acute AD was also higher (Figure 4.1J). Serum chemokines were deregulated including a
decrease in CXCL10, a factor implicated in the recruitment of regulatory T cells (Tregs) to
counteract inflammation (Figure 4.1K) [336]. This demonstrated a Th2-dominant atopic
immune response upon epidermal Adam17 loss.
4.3.2 Th2-polarized cellular immunity in skin-draining lymph nodes and
myeloproliferative disease in bone marrow of Adam17ep
mice
In order to characterize the nature and origin of the cellular immune response, we
compared skin-draining (dLN) and mesenteric (mLN) lymph nodes. Along with obvious
lymphadenopathy, flow cytometry analysis showed a 10-fold elevation in Gr-1+CD11b
+
monocytes and the presence of CD11c+MHCII
+ antigen presenting cells (APCs) in
Adam17ep
mice (Figure 4.2A-C). Activated CD4+CD69
+ T lymphocytes were observed
only in the dLNs of Adam17ep
mice (Figure 4.2D). Lymphocyte activation follows a
shift from naïve (CD62L+CD44
) to effector (CD62L
CD44
+) T cell phenotypes [337].
Again, Adam17ep
mice had a 4-fold increase in CD4+ effector T cells specifically in dLN
122
but not in mLN (Figure 4.2E). This confirmed the skin-specific origin of cellular
immunity in mice lacking epidermal ADAM17.
Mice harboring genetic epidermal defects often exhibit myeloproliferative disease
(MPD) [330, 331]. Our investigation of bone marrow homeostasis revealed strikingly
high levels of serum G-CSF along with the presence of splenomegaly in Adam17ep
mice
(Figure 4.2F, G). Interestingly, we noted a severe drop in bone marrow B220+
lymphocyte populations in Adam17ep
and mice, which upon further characterization
were found to be predominantly naïve B lymphocytes expressing high levels of IgM
compared to controls (Figure 4.2H, I). While the total number of bone marrow cells were
comparable (data not shown), Gr-1+CD11b
+ myeloid populations were proportionally
higher in bone marrow and spleen of Adam17ep
mice (Figure 4.2J), fulfilling all of the
hallmarks of MPD [338]. Thus, epidermal ADAM17 deficiency is sufficient to induce
myeloproliferation in mice.
4.3.3 Precocious differentiation of ADAM17 deficient keratinocytes in vivo
and in vitro
ADAMs perform ectodomain shedding of several factors required for keratinocyte
development including EGFR ligands and Notch. For example ADAM17 cleaves a subset
of EGFR ligands whereas ADAM10 is the recognized metalloproteinase for Notch
processing. The epidermal loss of Adam10 was recently reported to impair keratinocyte
differentiation in mice; they failed to establish the spinous layer of the epidermis [128].
We therefore asked whether the observed epidermal inflammation arose
123
Figure 4.2Skin-specific Th2 lymphocyte activation and myeloproliferative disease
(MPD) in Adam17ep
mice.
(A) Gross images depicting lymphadenopathy of the skin-draining lymph nodes in ep mice. (B) FACS
analysis of macrophage/monocytes (Gr-1+CD11b
+) in skin-draining lymph nodes (dLN) and mesenteric
lymph nodes (mLN). (C) FACS analysis of activated (CD11c+MHCII
+) dendritic cells in dLNs. (D) FACS
analysis of CD8+ vs. CD4
+ T cell activation by CD69 staining in mLNs and dLNs. (E) Induction of CD4
+
effector T cells (CD62L-CD44
+) in dLNs of ep mice measured by FACS. (F) ELISA measurement of
serum G-CSF levels in control (fl/fl) and Adam17ep
(ep) mice aged 10 weeks. (G) Representative image
of spleen depicting splenomegaly in ep mice. (H) FACS analysis of B220+ B lymphocytes in bone
marrow (BM). Analysis of CD19+B220
+ population shows a greater fraction of IgM
+ naïve B lymphocytes
in ep mice. (J) FACS analysis of CD11b+Gr-1
+ macrophage/monocytes in BM and spleen. *P<0.05 mean
±S.E.M (n=6). Data in A-E and G-I are representative of at least twelve mice. Scale bars = 5mm.
124
due to incomplete keratinocyte differentiation. This was not the case for Adam17ep
mice,
as immunofluorescence staining of the skin showed comparable establishment of terminal
keratinocyte differentiation as measured by basal (Keratin5), early (Keratin10) and late
(Filaggrin) markers (Figure 4.3A). Further, gene expression analysis of keratinocyte
differentiation markers confirmed the presence of basal and spinous (early) lineages
along with increased expression of granular (late) lineage markers in the epidermis of
newborn Adam17ep
mice (Figure 4.3B). We next created cell-autonomous ex vivo
deletion of Adam17 using Adam17fl/fl
keratinocytes transduced with adenoviral Cre
(termed AdCre-Adam17
), and tested calcium-induced differentiation. AdCre
keratinocytes exhibited accelerated differentiation as measured by gene expression of
basal, early (spinous) and late (granular/cornified) markers (Figure 4.3C). Of note,
adenoviral deletion of Adam17 induced precocious keratinocyte differentiation even in
basal conditions (AdCre Low Ca++
group, Figure 4.3C). Consistent with the onset of
terminal differentiation, AdCre keratinocytes proliferated at a slower rate in basal
conditions when compared with control (AdGFP) keratinocytes (Figure 4.3D). This data
indicates that in contrast to ADAM10 deficiency, loss of Adam17 leads to precocious
keratinocyte differentiation in a cell autonomous manner.
4.3.4 Inducible deletion of epidermal Adam17 in adult mice recapitulates
atopic dermatitis and myeloproliferative disease
Beyond keratinocyte differentiation, we reasoned that ADAM17 would be required for
epidermal homeostasis in the adult skin. We thus bred Adam17fl/fl
mice with
125
Figure 4.3Accelerated differentiation of Adam17ep
keratinocytes.
(A) Histological (Immunofluorescence) analysis of skin at indicated postnatal ages (P) measuring
expression of basal marker Keratin5 (K5, green), early differentiation marker Keratin10 (K10, red), and
late differentiation marker Filaggrin (Fil, red). (B) qPCR analysis of basal, early and late differentiation
markers in newborn (P1). (C) qPCR analysis of keratinocyte differentiation markers in primary
keratinocytes following calcium-induced terminal differentiation. Cells were maintained in basal medium
(Low Ca++
) or treated with 2mM calcium (High Ca++
) for 48 hours. AdGFP (control): Adam17fl/fl
keratinocytes transduced with adenoviral-GFP; AdCre (Adam17 knockout): Adam17fl/fl
keratinocytes
transduced with adenoviral-Cre to induced cell-autonomous Adam17 deletion. (D) Decreased rate of
proliferation in AdCre-Adam17
keratinocytes cultured in basal conditions. *P<0.05, mean ± S.E.M for B,
C, S.D. for D (n=3). Images are representative of at least 4 mice per genotype. Scale bars = 50m.
126
transgenic mice expressing tamoxifen-inducible Cre-recombinase under control of the
Keratin14 promoter (termed Adam17epERT
mice) and induced Adam17 deletion as
outlined in Figure 4.4A. Mice were analyzed at early (3 month) and late (8 month) time
points post-deletion. Tamoxifen-induced deletion of Adam17 resulted in gross
lymphadenopathy and skin inflammation (Figure 4.4B, C). We observed epidermal
thickening (H & E, Figure 4.4D) and keratinocyte hyperproliferation as measured by
Ki67 staining as early as 3 months (Figure 4.4E), as well as dermal infiltration by
Toluidine B+ mast cells (arrows, Figure 4.4F).
Flow cytometry analysis indicated higher numbers of CD4+CD69
+ activated T
cells in dLNs of Adam17epERT
mice (Figure 4.4G). Although CD8+CD69
+ T cell numbers
were also higher, the proportion of activated CD4+ T cells was 4 times greater than CD8
+
T cells (Figure 4G, quantified in Figure 4.4H). Consistent with this, a significantly higher
proportion of CD4+CD62L
CD44
+ effector T cells was observed in dLNs of
Adam17epERT
mice at all ages (Figure 4.4I, quantified in Figure 4.4J). Finally, these mice
developed MPD as characterized by loss of B220+ B lymphocytes (Figure 4.5A, B) and
proportionally greater populations of Gr-1+CD11b
+ macrophage/monocytes in bone
marrow and spleen (Figure 4.5C, D). These experiments establish that deletion of
epidermal Adam17 in the adult skin compromises barrier immunity.
127
128
Figure 4.4Inducible deletion of epidermal Adam17 in adult mice recapitulates atopic
dermatitis and skin-specific Th2 lymphocyte activation.
(A) Schematic depicting protocol for tamoxifen treatment and mouse analysis. (B) Representative images
of dLNs in Adam17fl/fl
(fl/fl) and K14-CreERT
;Adam17fl/fl
(epERT) mice following tamoxifen treatment as
outlined in Figure 4A. (C) Control (fl/fl) and K14-CreERT transgenic mice were bred to generate
tamoxifen-inducible deletion of Adam17 in basal keratinocytes (epERT mice). Images are representative
of four mice at indicated time points after tamoxifen treatment. (D-F) Histological analysis of dorsal skin
by H & E staining (D) Ki67 immunostaining depicting basal keratinocyte proliferation (E) and mast cell
influx (F, arrows) shown by Toluidine B staining. (G, H) FACS plots and quantification of CD8+ and
CD4+
T cell activation in mesenteric lymph nodes (mLN) and skin-draining lymph nodes (dLN) measured
by cell surface CD69. (I, J) FACS plots and quantification of CD8+ and CD4
+ effector T cells identified as
CD62L-CD44
+ populations in mLNs and dLNs of Adam17
fl/fl and Adam17
epERT mice. *P<0.05, mean ±
S.E.M. (n=4).
129
Figure 4.5 – Onset of MPD upon tamoxifen-induced deletion of epidermal Adam17 in
adult mice.
(A, B) FACS plots showing a loss of B220
+ B lymphocytes and a relative increase in IgM
+ naïve B
lymphocytes in bone marrow of epERT mice 3 and 8 months after tamoxifen treatment. Total B220+ B
lymphocytes in BM are quantified in the graph on the right. (C, D) MPD depicted by FACS analysis of Gr-
1+CD11b
+ macrophage/monocyte populations in bone marrow (BM) and spleen of epERT mice following
tamoxifen treatment. Bars to the right are quantifications of Gr-1+CD11b
+ myeloid cells. Flow cytometry
plots are representative of 3 mice per condition. *P<0.05, mean ± S.E.M. (n=3).
130
4.3.5 TNFR1 signaling is dispensable for atopic dermatitis and
myeloproliferative disease in Adam17ep
mice
TNF inhibitors have had measurable success in therapy of inflammatory skin
disease in humans [339-341]. Independent genetic models show that psoriasis-like
inflammatory skin disease caused by epidermal loss of Ikkb or c-Jun/JunB is ameliorated
by the deletion of TNF receptor 1 (Tnfr1), however the importance of TNFR1 signaling
in models of atopic dermatitis has not been tested. ADAM17 is the most relevant
sheddase for the cleavage of membrane bound TNF and its receptors, and we have
previously shown that regulation of ADAM17 activity by its endogenous inhibitor
TIMP3 provides a critical axis of control over TNF in models of acute and chronic
hepatic inflammation [69, 76, 98]. We bred Adam17ep
mice into a Tnfr1 deficient
background and surprisingly found that Tnfr1 loss failed to rescue epidermal
inflammation (Figure 4.6A). Analyses of peripheral immunity demonstrated activation of
CD4+ T cells (Figure 4.6B), induction of CD4
+ effector T cells (Figure 4.6C) and
elevated MHCII expression on APCs in dLN of Adam17ep
;Tnfr1mice (Figure 4.6D).
Adam17ep
;Tnfr1
mice also exhibited MPD, with increased proportions of Gr-
1+CD11b
+ myeloid cells and a loss of B220
+ B lymphocytes in the bone marrow (Figure
4.7A, B). Thus, epidermal ADAM17 prevents atopic dermatitis and maintains bone
marrow homeostasis independent of TNFR1 signaling.
131
Figure 4.6 – Tnfr1 deletion does not rescue AD in Adam17ep
mice.
(A) Representative images depicting the presence of AD in Adam17
ep;Tnfr1
(ep;Tnfr1
) double
knockout mice. (B) FACS analysis analysis and quantification of CD8+ and CD4
+ T cell activation
measured by CD69 levels in dLNs of control (fl/fl), single knockout (ep) and double knockout
(ep;Tnfr1
) mice. Activated lymphocytes are quantified in graphs to the right. (C) FACs analysis and
quantification of CD62L-CD44
+ effector CD8
+ and CD4
+ T cell subsets in dLNs. (D) Mean fluorescence
intensity (MFI) quantification of MHCII expression measured by FACS of B220+ B lymphocytes and
CD11c+ APCs. Flow cytometry plots are representative of 6 mice per condition. *P<0.05, mean ± S.E.M.
(n=6).
132
Figure 4.7 – MPD persists in Adam17ep
;Tnfr1
double deficient mice.
(A) FACS analysis of bone marrow (BM) monocyte/macrophages shows comparable numbers of Gr-
1+CD11b
+ cells in ep and ep;Tnfr1
mice. Total cell numbers are quantified by graphs on the right. (B)
Similar decrease in B220+CD19
+ bone marrow B lymphocytes in ep and ep;Tnfr1
mice. Bars are
quantifications of total cell numbers in 6 mice per genotype. *P<0.05, mean ± S.E.M. (n=6).
133
4.3.6 Loss of Adam17 compromises tonic Notch signaling in the epidermis
Remarkably, Adam17ep
mice closely phenocopy the onset of AD and MPD
previously illustrated upon inducible and combined postnatal deletion of epidermal
Notch1 and Notch2 [330]. However, there is no physiological evidence for ADAM17-
mediated Notch activation in mammals. We tested whether ADAM17 deficient skin had
defective Notch signaling and found Notch intracellular domain (NICD) to be absent in
adult Adam17ep
epidermis (Figure 4.8A). Moreover, expression of multiple Notch target
genes was diminished in Adam17ep
epidermis (Figure 4.8B). Typically, Notch activation
follows ligand binding (Delta-like 1, 3, 4, Jagged 1, 2) and involves metalloproteinase
(ADAM)-mediated S2 cleavage of the receptor at the cell surface with subsequent NICD
release via -secretase (S3 cleavage) [342]. Notch activation can occur in a ligand-
dependent or ligand-independent manner, and Bozkulak et al (2009) have shown that
ADAM10, but not ADAM17, is required for ligand-dependent Notch activation in vitro
[123, 138]. To delineate how ADAM17 contributed to Notch activation in keratinocytes,
we examined both possible scenarios for Notch activation using AdCre-Adam17
keratinocytes. In a ligand-independent model, calcium treatment of AdCre-Adam17
keratinocytes failed to produce NICD or induce Notch target gene expression (Figure
4.7C, D). On the other hand, ADAM17 was dispensable for ligand-dependent Notch
signaling when induced by Fc-Delta-like 4 fusion protein (Figure 4.8E). Together, these
data demonstrate that ADAM17 activity is required for Notch signaling in adult skin, and
suggest that ADAM17 activates Notch in a ligand-independent manner in keratinocytes.
134
Figure 4.8ADAM17 provides ligand-independent Notch signaling and suppresses c-
Fos activity in keratinocytes. (A) Immunoblot depicting loss of NICD production in adult ep epidermis. (B) qPCR analysis of newborn
(P1) epidermis for Notch target genes Hes5, Hey1 and Hey2. (C) Immunoblot depicting a lack of S3
cleavage in AdCre-Adam17
keratinocytes following calcium (Ca++
) treatment. (D) qPCR measurement
of Notch target gene expression (Hes1, Hey2) following calcium treatment. (E) qPCR of Notch target gene
expression in AdGFP-control and AdCre-Adam17
keratinocytes following ligand-dependent Notch
activation by Delta like 4 (hFc-DLL4). (F) Analysis of published microarray gene expression datasets
indicates differential expression of genes involved in Notch signaling in atopic dermatitis and psoriasis
patients compared to normal controls. *P<0.05 mean ± S.E.M (n=4 for B, n=3 for D, E). Data are
representative of 4 mice (A, B), and three independent experiments (C-E).
135
Figure 4.9Stratification of atopic dermatitis, atopic eczema and psoriasis patients
based on expression analysis of genes involved in Notch signaling.
(A-C) Heatmaps of significantly differentially expressed genes involved in Notch signaling. (D)
Unsupervised hierarchical clustering of atopic dermatitis, psoriasis and normal skin using the Notch
signaling gene set. Gene lists are provided in Table 5.1 (Appendix A).
136
We sought evidence for compromised Notch signaling in patients with
inflammatory skin disease using published microarray human gene expression datasets
[343-346]. We consistently observed genes in the Notch pathway to be differentially
expressed in patients with AD (Figure 4.8F), atopic eczema (Figure 4.9A) and psoriasis
(Figure 4.9B, C). Specifically, a decrease in Notch2, Notch3, and Presenilin1 in AD was
observed compared to normal skin (Figure 4.8F; gene lists in Table 4.1 provided in
Appendix A). Of note, Adam10 and Adam17 displayed a heterogeneous expression
pattern in AD skin, likely due to the pleiotropic functions of these metalloproteinases. We
also performed unsupervised clustering analysis and observed that genes in the Notch
signaling module stratified AD samples away from psoriasis and normal samples,
suggesting that Notch signaling is more deregulated in AD than in psoriasis (Figure
4.9D). The above data proposes a link between ADAM17 and Notch signaling in human
atopic skin disease.
4.3.7 Active Notch antagonizes c-Fos/AP-1 transcription of Tslp in human
keratinocytes
Our next goal was to identify the mechanism by which ADAM17-driven Notch activation
keeps epithelial inflammation in check. Epithelial cytokines TSLP and IL-33 have
emerged as key „alarmins‟ in barrier immunity [328, 329, 347]. Keratinocytes also
produce significant amounts of hematopoietic growth factors Csf3 (G-CSF) and Csf2
(GM-CSF) in an AP-1 regulated manner [331, 348]. The potential for alternative Notch
signaling exists via its interaction with transcription factors NF-B, and possibly AP-
137
1[349-351]. The AP-1 family has pleiotropic functions in stress response where specific
members dimerize, bind to target sequences and regulate gene expression [348]. Notch
has been proposed to inhibit AP-1 activity in vitro [349, 351]. To interrogate members of
the AP-1 transcription factor family antagonized by Notch, we used a human keratinocyte
cell line (HaCaT cells). We began by measuring individual AP-1 proteins (c-Jun, JunB,
JunD, c-Fos, Fra1, Fra2) over 6 hours following treatment with the phorbol ester PMA
(an AP-1 agonist). Of these AP-1 members, c-Fos displayed a distinct profile with its
levels peaking at 2 hours and rapidly terminating at 3 hours (Figure 4.10A). c-Fos
typically dimerizes with the Jun proteins (c-Jun, JunB, JunD) [348] and their levels
gradually increased in HaCaT cells following PMA stimulation (Figure 4.10A). We
inhibited either ADAM17 activity or NICD production by using small molecule
inhibitors (TAPI-1 and DAPT, respectively), and both resulted in a sustained production
of c-Fos upon PMA treatment (Figure 5G). TAPI-1 or DAPT treatment also led to the
depletion of NICD as measured by immunoblotting (Figure 4.10B). Notably, c-Fos
stabilization or NICD depletion correlated with elevated Tslp and Csf3 transcription as
measured by qPCR (Figure 4.10C, D).
The promoter regions of Tslp and Csf3 contain TPA regulatory elements (TRE)
that bind AP-1 and promoter polymorphisms that enhance AP-1 binding are associated
with higher susceptibility to bronchial asthma [352]. These promoters also harbor RBP-j
binding sites that recruit the Notch transcriptional complex (schematics shown in Figure
4.10E, F). We assessed c-Fos and NICD recruitment to these promoters by
138
Figure 4.10ADAM17 activates Notch to suppress c-Fos transcriptional activity on
cytokine promoters.
(A) Immunoblots of individual AP-1 members from whole cell lysates of HaCaT cells treated with 50ng/ml
PMA for the indicated timepoints. Data are representative of two independent experiments. (B)
Immunoblotting for NICD, c-Fos and Fra1 in HaCaT cells pre-treated with ADAM17 inhibitor TAPI-1 or
-secretase inhibitor DAPT, followed by PMA treatment for the indicated time points. (C, D) qPCR
measurement of Tslp and Csf3 (G-Csf) gene expression in HaCaT cells treated as in (A). (E, F) NICD and
c-Fos conversely regulate Tslp and Csf3 transcription by recruitment to promoter regions. HaCaT cells
were pre-treated with DMSO (control) or the ADAM17 inhibitor TAPI-1, followed by PMA treatment for
the indicated times. Chromatin was precipitated using the indicated antibodies. Tslp and Csf3 promoters
were amplified from precipitated DNA. Schematics of Tslp and Csf3 promoters depict AP-1 binding sites
(TRE), Notch binding sites (RBP-j) and primer design (small arrows). *P<0.05 mean ± S.E.M (n=4).
Data are representative of two independent experiments.
139
chromatin immunoprecipitation (ChIP) assays. NICD was constitutively present on the
Tslp promoter in HaCaT cells while c-Fos was only recruited at 2 hours after PMA
treatment. Inhibition of ADAM17 activity with TAPI-1 depleted NICD production and
abolished its recruitment to the Tslp promoter while concomitantly prolonging c-Fos
binding to this region at 3 hours after PMA treatment (Figure 4.10E). Similar but not
identical antagonism was observed between NICD and c-Fos at the Csf3 promoter
(Figure 4.10F). These results suggest that loss of Notch activation allows for unchecked
c-Fos transcriptional activity.
4.3.8 Ectopic Notch rescues the enhanced AP-1 driven stress response in
ADAM17 deficient keratinocytes
We next determined whether tonic Notch activation by ADAM17 keeps AP-1
dependent cytokine production in check. Adenoviral Cre-mediated Adam17 deletion in
keratinocytes (AdCre-Adam17
) resulted in spontaneous phosphorylation of the stress
kinase JNK1/2 (Figure 4.11A). Treatment with PMA specifically enhanced the levels of
c-Fos and Fra1 in AdCre-Adam17
compared to AdGFP control keratinocytes.
Additionally, the DNA binding activity of AP-1 members was elevated in AdCre-
Adam17
keratinocytes (Figure 4.11B, C). Next, qPCR analysis showed higher Tslp,
Csf2, Csf3, and c-fos expression in AdCre-Adam17keratinocytes after PMA treatment;
this was abrogated by the pan AP-1 inhibitor Tanshinone IIA (TanIIA) (Figure 4.11D;
Figure 4.12A). We noted the specificity of this response since ADAM17 deficiency did
140
Figure 4.11Diminished Notch activation upon ADAM17 deficiency enhances c-
Fos/AP-1 driven stress signaling in keratinocytes. (A) Immunoblotting detects spontaneous SAPK/JNK phosphorylation in AdCre-Adam17
keratinocytes.
(B) Elevated protein levels of c-Fos and Fra-1 in AdCre-Adam17
keratinocytes following PMA
treatment. (C) Enhanced AP-1 transcription factor binding for specific members of AP-1 family in AdCre-
Adam17
keratinocytes treated with PMA. AdGFP-control and AdCre-Adam17
keratinocytes were
treated with 10ng/ml PMA for 3 hours, followed by measurement of AP-1 target sequence binding by
individual members of the AP-1 family. (D, F) qPCR analysis of Csf3, Tslp and c-Fos gene expression in
keratinocytes 3 hours after PMA treatment. Presence of the pan AP-1 inhibitor Tanshinone IIA (TanIIA, D)
or ectopic expression of active Notch (AdNICD, F) abrogates gene expression. (E) Measurement of AP-1
luciferase reporter activity in MEFS after 8 hours of PMA treatment. Use of a specific c-Fos inhibitor (T-
5224) or ectopic Notch (AdNICD) abolishes AP-1 luciferase activity. *P<0.05, mean ± S.E.M. (n=3). Data
in A-C are representative of two independent experiments. Data in D and E are representative of three
independent experiments.
141
Figure 4.12AdCre-deletion of Adam17 in keratinocytes enhances Csf2 gene
expression in an AP-1 dependent manner but does not impact c-Jun, JunB, JunD, Fra1 or
Il33 expression.
(A) qPCR analysis shows increased expression of Csf2 in AdCre-Adam17
-/- keratinocytes treated with
PMA. This is abrogated in the presence of AP-1 inhibitor (Tanshinone IIA). Comparable gene expression
of AP-1 members cjun, junb, jund and fra1 in AdGFP-control and AdCre-Adam17
keratinocytes. (B)
Adenoviral induction of ectopic Notch (AdNICD) abolishes Csf1 and Csf2 expression following PMA
treatment. *P<0.05, mean ± S.E.M. (n=3). All data are representative of three independent experiments.
142
not alter gene expression of either Csf1, Il33 cytokines or the other AP-1 members (c-
jun, junb, jund, fra1, fra2) (Figure 4.12A). These data show that the loss of Adam17
results in increased AP-1 driven synthesis of the above epithelial cytokines.
We then measured AP-1 activity in an independent cell type (mouse embryonic
fibroblasts, MEFs) and observed that Adam17
MEFs had higher basal AP-1 luciferase
reporter activity. This was enhanced after PMA treatment and reduced by a c-Fos/AP-1
inhibitor (T-5224) (Figure 4.11E). More importantly, transduction of active Notch
(AdNICD) in AdCre-Adam17
keratinocytes ablated the expression of Csf2, Csf3 and
Tslp (Figure 4.11F; Figure 4.12B). In addition, AdNICD abolished both basal and PMA-
induced AP-1 luciferase reporter activity in Adam17
MEFs (Figure 4.11E). We
therefore concluded that ADAM17-dependent Notch activation inhibits AP-1
transcriptional activity, with c-Fos being a relevant AP-1 candidate.
4.3.9 Adenoviral delivery of active Notch rescues local Th2 immunity and
myeloproliferation in Adam17ep
mice
Finally, to test whether Notch could protect Adam17ep
mice from AD, we utilized a
protocol to induce ectopic Notch activity in ADAM17 deficient epidermis (schematic in
Figure 4.13A). Adenoviral delivery of active Notch (AdNICD) into Adam17ep
mice
showed a measurable increase in Notch target gene expression by qPCR analysis of
epidermal scrapings (Figure 4.13B). Importantly, CD4+ T cell activation was abrogated in
dorsal dLNs and a loss of effector CD4+ T cells was observed in AdNICD treated mice
(Figure 4.13C, D). Dorsal delivery of AdNICD also partially rescued MPD resulting in a
143
decrease in bone marrow Gr-1+CD11b
+ cells (Figure 4.13E). More importantly, a
significant replenishment of B220+ lymphocytes was seen in AdNICD treated mice
(2.93% in AdGFP vs. 13.47% in AdNICD, Figure 4.13F). Thus, active Notch was
sufficient to rescue AD as well as MPD in Adam17ep
mice. Altogether these data show
that ADAM17 activates Notch and this in turn interferes with c-Fos recruitment to the
Tslp and Csf3 promoters to keep epithelial cytokine production in check (outlined in
Figure 4.13G).
4.4 Discussion
This study provides the first demonstration, to our knowledge, of the
physiological requirement of ADAM17 in Notch signaling. We show that ADAM17
provides an inherent permissive signal for basal Notch activation throughout adult life. It
triggers Notch signaling to regulate c-Fos activation, thus maintaining epidermal barrier
homeostasis. Loss of this control induces the production of cytokines or „alarmins‟ which
catalyze local and systemic immune responses even in the absence of an overt exogenous
stress, as illustrated in Figure 4.14. Independent studies have shown that the loss of
Notch, deregulation of AP-1 or overproduction of epithelial TSLP result in inflammatory
skin disease [328, 330, 331, 353]. The ADAM17/Notch/c-Fos triad uncovered in our
study reveals a mechanistic underpinning of all these observations.
144
Figure 4.13Ectopic Notch activity rescues atopic dermatitis and myeloproliferative
disease in Adam17ep
mice.
(A) Experimental protocol to re-introduce active Notch in dorsal skin of ep mice by subcutaneous
delivery of adenoviral vector harboring NICD (AdNICD). Adenoviral GFP (AdGFP) was used as a control.
(B) qPCR analysis of Notch target genes Hes1 and Herp1 in dorsal epidermis at experimental endpoint.
Bars depict mean expression (n=4). (C, D), FACS analysis of dorsal dLNs to measure CD4+ T cell
activation by cell surface CD69 levels (C), and CD4+CD62L
-CD44
+ effector T cells (D) in ep mice
treated with AdGFP or AdNICD. (E, F), Partial rescue of myeloproliferative disease in bone marrow of
ep mice as measured by FACS analysis of Gr-1+CD11b
+ cells (E) and B220
+ cells (F). (G) ADAM17 is
required for ligand-independent Notch activation, which in turn antagonizes c-Fos transcriptional activity at
promoters of epithelial cytokines, thereby checking their production. *P<0.05 mean ± S.E.M, A, B (n=4)
and C-E (n=3). Data in C-E are representative of three independent experiments. Flow cytometry plots
represent 4 mice per condition.
145
Figure 4.14 – An illustration of epithelial-immune crosstalk as directed by the
ADAM17/Notch/c-Fos triad.
Notch represses pro-inflammatory c-Fos/AP-1 transcriptional activity in keratinocytes to prevent Th2-
driven atopic dermatitis by TSLP and myeloproliferative disease by G-CSF. ADAM17 induces ligand-
independent Notch activation in keratinocytes, which inhibits c-Fos transcriptional activity on promoter
regions of epithelial cytokines to regulate their production. The loss of ADAM17 reduces Notch activation
and results in the c-Fos-driven production of TSLP and G-CSF. Consequently, a Th2-driven local immune
response is generated in the skin and myeloproliferation is induced in the bone marrow. ADAM17
therefore provides barrier immunity by promoting Notch activation in the skin.
146
4.4.1 Contexts of ADAM10 and ADAM17 function in Notch signaling
To date, studies of Adam10 knockout mice have established its importance of in
Notch activation during development [123, 124, 128, 354]. On the other hand, the large
majority of reported ADAM17-dependent phenotypes revolve around TNF or EGFR
signaling [76, 98, 133, 136, 286, 339]. Recent in vitro studies have brought to light the
requirement of ADAM17 in ligand-independent Notch activation [138, 355]. Our study
proposes that ligand-independent Notch signaling controls barrier immunity but not
keratinocyte development. The developmental context of Notch activation may be
fulfilled by ADAM10, since mice lacking epidermal Adam10 fail to establish a spinous
layer of the epidermis [128], whereas Adam17ep
mice exhibit precocious keratinocyte
differentiation. Epidermal homeostasis balances cell-autonomous keratinocyte
development and non-autonomous communication between epithelial, stromal and
immune compartments in the skin. By generating inducible deletion of epidermal
Adam17 at 8 weeks of age, we demonstrate the significance of ADAM17 in maintaining
proper epithelial:immune crosstalk in the adult skin.
It is now accepted that Notch activation keeps the epidermal stress response in
check in addition to its established role in keratinocyte development and transformation
[356, 357]. Accumulating evidence suggests that Notch promotes NF-B signaling, but
inhibits AP-1 activity; both these pathways are key components of stress signaling in
keratinocytes but the molecular mechanism by which Notch and AP-1 pathways crosstalk
have remained elusive [331, 341, 349, 353, 358]. While we did not observe defects in
NF-B signaling (data not shown) or rescue AD through Tnfr1 deletion, we elucidated
147
how ADAM17-dependent Notch activation prevents spontaneous AP-1 activation and
Th2-driven immunity. We probed Notch/AP-1 signal crosstalk with gain/loss of function
approaches, biochemical inhibitors, AP-1 reporter and chromatin immunoprecipitation
assays in independent cell types, and found that 1) c-Fos transcribes TSLP, a key
epithelial cytokine in atopic disease; 2) Active Notch suppresses c-Fos activity by
interfering with its recruitment to cytokine promoters; 3) ADAM17 allows tonic Notch
activation in the skin; and 4) Exogenous Notch ameliorates Th2 immunity caused by
ADAM17 deficiency. Further, the data mining performed in this study indicated a clear
deregulation of the Notch pathway in human atopic dermatitis. Supporting our findings,
polymorphisms have been reported in patients where enhanced AP-1 recruitment to Tslp
and Il4 promoters occurs in several human atopic diseases [352, 359, 360].
Epithelial cells are now recognized as orchestrators of the peripheral immune
response, and several recent studies have identified epithelial mutations causal to
undesired immune cell activation in barrier tissues [327, 361-363]. We demonstrate a
gatekeeper function of ADAM17 in the skin, which may apply to other mucosal tissues
including gut, lung and vaginal epithelium. The epimmunome continues to grow and
understanding the mechanisms that instruct pro- or anti-inflammatory epithelial cytokine
synthesis will be of significant value in the treatment of diseases afflicting barrier tissues.
148
4.5 Methods
4.5.1 Mice
Adam17fl/fl
mice were generated by Dr. Carl P Blobel. K14-Cre transgenic mice
(Tg(Krt14-Cre)1Amc/J) and Tnfr1-/-
mice (C57BL/6-Tnfrsf1atm1Imx
/J) were originally
obtained from Jackson laboratories. Mice were fed 5%-fat chow ad libitum, housed and
cared for in accordance with protocols approved by the Canadian Council for Animal
Care and the Animal Care Committee of the Ontario Cancer Institute.
4.5.2 Microarray Gene Expression Analysis
Gene expression datasets were obtained from the National Center for
Biotechnology Information Gene Expression Omnibus (GEO) database under the
accessions GSE16161, GSE13355, GSE6012 and GSE6710. Data was utilized in the
processed form from its original study. A gene set comprising members of the Notch
pathway was constructed using information from additional references provided in
Supplemental Data and literature cited in main text. Probes mapping to these genes were
identified on the Affymetrix HGU133A (GSE6710 and GSE6012, 84 probes – Table S1)
and HGU133 Plus2 platforms (GSE13355 and GSE16161, 142 probes –Table S1).
Differential expression for genes within the Notch set was examined using a welch t-test,
and the False Discovery Rate [364] was employed to control for multiple hypotheses,
using a cutoff of 5.0% (Table 5.1, Appendix A). In each dataset we compared: atopic
dermatitis vs. normal (GSE16161), atopic eczema vs. normal (GSE6012) or psoriasis vs
normal (GSE13355 and GSE6710). Significantly differentially expressed genes were
149
visualized using matrix2png [365], after collapsing probes to genes at their median value,
and scaling the data to N(0,1). Samples in GSE16161 were subjected to hierarchical
clustering (uncentered correlation and average linkage) for the complete Notch signaling
gene set in Cluster 3.0 [366] and the resulting dendogram was visualized using Java
TreeView [367]. The following references were cited in addition to those listed in the
main text: [196, 342, 368-374].
4.5.3 In vivo delivery of active Notch to rescue epidermal inflammation
Mice aged 14 days were treated with a dorsal subcutaneous injection of
4x109pfu/ml adenovirus encoding active Notch (AdNICD, kindly provided by Dr. Erwin
Wagner, Centro Nacional de Investigaciones Oncologicas, Madrid, Spain) or eGFP (Ad-
eGFP, Vector Biolabs) dissolved in 50l sterile PBS. This treatment was administered
every 48h at P14, P16, P18 and P20. At P21, mice were sacrificed and tissue harvested
for analysis. Notch signaling was confirmed by qPCR analysis of target genes Hes1 and
Herp1 from dorsal epidermis scrapings. Epidermis was obtained by incubating minced
dorsal skin in Liberase Blendzyme (Roche) for 1h at 37oC with periodic vortexing. Flow
cytometry analysis is detailed below.
4.5.4 Isolation and culture of primary keratinocytes
Mouse keratinocytes were obtained from Adam17fl/fl
pups following established
protocols [375]. Cell autonomous Adam17 deletion was generated by transducing
Adam17fl/fl
keratinocytes with adenoviral Cre (AdCre-IRES-GFP, Vector Biolabs) or
150
adenoviral GFP as control (Ad-eGFP, Vector Biolabs) at an MOI of 3. Notch
overexpression was induced by transduction with AdNICD. For calcium-induced
differentiation and ligand-independent Notch activation, cells were cultured with 1mM
CaCl2 in calcium-free Eagles‟ MEM supplemented with 16% calcium-free (chelated)
serum. For ligand-dependent Notch activation, cells were incubated with a fusion protein
consisting of mouse Delta Like 4 and human-Fc at 1g/ml cross-linked with 1g/ml anti-
human IgG1 (H+L, Pierce). Anti-human IgG1 alone was used as a control. For PMA
treatments (Phorbol 12-myristate 13-acetate, P8139, Sigma), media was changed to
serum-free, calcium-free eagles MEM. Pan AP-1 activity was inhibited by pre-treating
keratinocytes with 20M Tanshinone IIA (Enzo Life Sciences, TanIIA) 18 hours prior to
PMA treatment. TanIIA was maintained for the duration of PMA treatment.
4.5.5 AP-1 luciferase assay and PMA treatment of mouse embryonic
fibroblasts (MEFs)
MEFs were transduced with lentiviral AP-1 firefly luciferase and Renilla
luciferase as a control 48 hours prior to treatment (Cignal AP1 reporter, SABiosciences).
Cells were treated with 10ng/ml PMA in serum-free DMEM for 18 hours. To inhibit c-
Fos/AP-1, MEFs were pre-treated with 20M T-5224 (kindly provided by Dr. Erwin
Wagner) 3 hours prior to PMA treatment. T-5224 was maintained for the duration of
PMA treatment. AP-1 activity was calculated as a ratio of AP-1 luciferase luminescence
divided by constitutive Renilla luciferase luminescence and depicted as relative
luminescence units (RLU).
151
4.5.6 Culture and treatment of HaCaT cells and chromatin
immunoprecipitation (ChIP)
Human keratinocyte cell line (HaCaT cells) was kindly provided by Dr. Hitoshi
Okada (Ontario Cancer Institute, Toronto, Canada). Cells were maintained in DMEM +
10% FBS and serum-starved overnight prior to any treatment. To inhibit ADAM17
activity, HaCaT cells were treated with 100M TAPI-1 3 hours prior to stimulation with
PMA. To inhibit NICD production, cells were treated with 20M of the -secretase
inhibitor DAPT. Chromatin was obtained and ChIP assay performed using the EZ-ChIP
kit (Millipore). The following antibodies were used for immunoprecipitation: anti-human
IgG1 (H+L, Pierce), anti-NICD (Val1744), anti-p-c-Fos (Cell Signaling), anti-Histone H3
(Abcam). Primer sequences are provided in Table 4.2.
4.5.7 Flow cytometry
For splenocyte analysis, red blood cell lysis was performed using AFCS buffer
prior to staining. Superficial axillary and brachial lymph nodes were denoted as skin-
draining lymph nodes (dLNs). Mesenteric lymph nodes (mLNs) were isolated as controls.
Bone marrow cells were obtained from one femur per mouse. The following antibodies
were used for analysis: anti-CD3 (clone 17A2), anti-CD4 (clone RM4-5), anti-CD8
(clone 53-6.7), anti-B220 (clone RA3-6B2), anti-NK1.1 (clone PK136), anti-F4/80 (clone
BM8), anti-MHCII (clone M5/114.15), anti-CD11c (clone N418), anti-Gr-1 (clone RB6-
8C5), anti-CD44 (clone IM7), anti-CD11b (clone M1/70), anti-CD69 (clone H1.2F3),
152
anti-CD62L (clone MEL-14), anti-IgG2a (clone m2a-15F8), anti-CD19 (clone ID3), and
anti-IgM (clone II/41). 7-AAD was used to assay cell viability.
4.5.8 Histology & cytokine analysis
Dorsal skin was obtained and dissected longitudinally along dorsal midline and
processed as in [98]. The following stains were applied for immunohistochemical
analysis: Harris‟ hematoxylin and eosin, Toluidine B, anti-Ki67 (Abcam), anti-F4/80 (BD
Biosciences), anti-Keratin 5, anti-Keratin 10, and anti-Filaggrin (all from Covance). All
staining was performed by institutional core facilities. Serum was separated by
centrifugation of whole blood (BD Vaccutainer CPT Cell Preparation Tube) and stored at
-70oC. Cytokine levels were measured by following manufacturer‟s protocols for a
cytokine bead array (Millipore). TSLP (R&D Systems) and IL-33 (eBioscience) levels
were measured by standard ELISA following manufacturer‟s protocols.
4.5.9 Immunoblotting and AP-1 binding assay
Keratinocytes were treated with 10ng/ml PMA for 3 hours. Protein was obtained
from tail epidermis by mortar and pestle homogenization of frozen tissue. Protein from
keratinocytes was obtained by washing cells with ice cold PBS followed by lysis in RIPA
extraction buffer. 10 to 50 μg protein was loaded onto SDS-PAGE gels for
immunoblotting. Nuclear lysate for AP-1 binding assays was obtained using standard
protocols [98]. AP-1 binding to target sequence (TRE oligonucleotide) was measured by
using AP-1 family EZ-TFA Transcription Factor Assay (Millipore). The AP-1 antibodies
153
listed below were used instead of those provided in the kit. The following anti-mouse
antibodies were used: anti-ADAM17 (Khokha lab), anti-active Notch1 (NICD, Abcam
for 72kda fragment, Cell Signaling for 120kda fragment), anti–phosphorylated JNK (p-
JNK, Thr183/Tyr185), anti-JNK, anti– phosphorylated p44/42 MAPK (p-ERK1/2,
Thr202/Tyr204), anti–p44/42 MAPK (ERK1/2), anti--Tubulin anti-p-c-Fos (all from
Cell Signaling), anti-c-Jun, anti-JunD, anti-c-Fos, anti-Fra1, anti-Fra2 (all from Santa
Cruz), anti-c-Jun (BD Biosciences).
4.5.10 RNA preparation and quantitative RT-PCR
RNA was prepared from frozen dorsal epidermis or fresh keratinocytes using TRIzol
reagent (Invitrogen) following standard protocols. cDNA was obtained using a first-
strand cDNA synthesis protocol (Quanta). Gene expression was measured using SYBR
Green reagent (Quanta) in a 7800HT real-time PCR system (Applied Biosciences). All
gene expression levels were normalized to Actb (β-actin), and fold change was measured
relative to control wild-type samples. qPCR of RNA was used as a negative control. The
amount of each product was calculated using the 2–ΔCT
method. Primer sequences are
provided in Table 5.2.
4.5.11 Statistical Analyses
Data are reported as mean ± S.D. or S.E.M. All calculations were carried out
using GraphPad Prism software (GraphPad Software). Comparisons were made by 2-
tailed Student‟s t test and ANOVA.
154
Table 4.1 – Probelists and significance values for differentially expressed genes in all
studies. (See Appendix A online).
Table 4.2Primer sequences used for qRT-PCR and ChIP analysis.
Gene Forward & Reverse Primer Sequence
Adam17 F: 5‟- CGGAGGAAGCAGGCTCTG-3‟
R: 5‟- GTTTCTAAGTGTGTCGCAGACTG-3‟
c-Fos F: 5‟- CAGAGCTGGAGCCCCTGTGT -3‟
R: 5‟- CGGCTGCCTTGCCTTCTCTG -3‟
c-Jun F: 5‟- TTGCCCCAACAGATCCCGGT -3‟
R: 5‟- TGCGGTTCCTCATGCGCTTC -3‟
Csf1 F: 5‟- TGGCCCATGGGAGACCAGGA -3‟
R: 5‟- GGGCTGGGTGCCTTTATGCC -3‟
Csf2 F: 5‟- ACATGTGTGCAGACCCGCCT -3‟
R: 5‟- TCCGTTTCCGGAGTTGGGGG -3‟
Csf3 F: 5‟- TGACCAGGGGAACGGCCTCT -3‟
R: 5‟- GCTGCAGCCCAGATCACCCA -3‟
Filaggrin (Fil) F: 5‟- GTGGCCAACAGAGAATGAG -3‟
R: 5‟- ATGATGCCCAGAACTATGTGAC -3‟
Fra1 F: 5‟- AACGCGGACCCTACCGAACA -3‟
R: 5‟- GGCTGCTGCTGTCGATGCTT -3‟
Hes1 F: 5‟- AAAGCCTATCATGGAGAAGAGGCG -3‟
R: 5‟- GGAATGCCGGGAGCTATCTTTCTT -3‟
Hes5 F: 5‟- AAAGCCTATCATGGAGAAGAGGCG -3‟
R: 5‟- GGAATGCCGGGAGCTATCTTTCTT -3‟
Hey1 F: 5‟- ACACTGCAGGAGGGAAAGGTT -3‟
R: 5‟- CAAACTCCGATAGTCCATAGCCA -3‟
Hey2 F: 5‟- AAGCGCCCTTGTGAGGAAAC -3‟
R: 5‟- GGTAGTTGTCGGTGAATTGGAC -3‟
Human Csf3 F: 5‟- CCCAGAGCCCCATGAAGCTG -3‟
R: 5‟- GGTGGCACACTCACTCACCAG -3‟
Human Tslp F: 5‟- AGAAAGCTCTGGAGCATCAGGGA -3‟
R: 5‟- ACATACGTGGACACCCAATTCCAC -3‟
Il33 F: 5‟- TTTATGAAGCTCCGCTCTGGCC -3‟
R: 5‟- GCTGTTGACAGGCAGCGAGTA -3‟
JunB F: 5‟- ACATGCACCACCCTTTGCGG -3‟
R: 5‟- CAGCCGCTTTCGCTCCACTT -3‟
155
JunD F: 5‟- AGGTGGCAGCTAGCGAGGAG -3‟
R: 5‟- TCAGGTTGGCGTAGACCGGG -3‟
Keratin 1 (K1) F: 5‟- GACACCACAACCCGGACCCAAAACTTAGAC -
3‟
R: 5‟- ATACTGGGCCTTGACTTCCGAGATGATG -3‟
Keratin 10 (K10) F: 5‟- GCCAGAACGCCGAGTACCAACAAC -3‟
R: 5‟- GTCACCTCCTCAATAATCGTCCTG -3‟
Keratin 14 (K14) F: 5‟- TGCTGGATGTGAAGACAAGG -3‟
R: 5‟- GGATGACTGAGAGCCAGAGG -3‟
Keratin 5 (K5) F: 5‟- TCTCTTCTGGCTACGGAGGA -3‟
R: 5‟- GAAGCTCATGCCTCCTTGAC -3‟
Loricrin (Lor) F: 5‟- TTGCAACGGAGACAACAGAG -3‟
R: 5‟- GCGACTCAATGGCTTCTTCT -3‟
Transglutaminase 1
(Tgm1)
F: 5‟- GCGGAGGGCTGTGGAGAAGG -3‟
R: 5‟- GGGTGCGCAAACGG AA GGTG -3‟
Transglutaminase 3
(Tgm3)
F: 5‟- GCTTGGGGGTTCGCTCTCG -3‟
R: 5‟- CTGCCTACTGCCTTGGTGCTGAT -3‟
Tslp F: 5‟- GAGGACTGTGAGAGCAAGCCAG -3‟
R: 5‟- GGCAGTGGTCATTGAGGGCTT -3‟
-Actin (Actb) F: 5‟- ACGAGGCCCAGAGCAAGAGA -3‟
R: 5‟- GTGAGCAGCACAGGGTGCT -3‟
ChIP: Human Csf3 F: 5‟- TGGCCAGAGCTGGGAGGCAT -3‟
R: 5‟- AGGCACCTGCCTGGCCCTAA -3‟
ChIP: Human Tslp F: 5‟- ACCCTCCATTGGTGTTGCTGGA -3‟
R: 5‟- GCATCGAGATGGCCGGGCTCA -3‟
156
5 CHAPTER 5
FUTURE DIRECTIONS
157
This thesis investigated the contributions of the metalloproteinase ADAM17 and
its endogenous inhibitor TIMP3 to acute and chronic inflammation. The insights gained
from the described studies introduce the tissue-specific roles of TIMPs in the epithelial
stress response, immune cell development and function. Proper immunity involves
communication between the afflicted tissue and resident or local immune cells, which
then calls upon context-dependent innate and/or adaptive immune effectors to eliminate
the source of tissue stress. Simultaneously a cellular repair program is initiated in the
damaged tissue, and this is now thought to depend on signals generated by active immune
cells in the local microenvironment. Following eradication or containment of infection,
an immunoregulatory program is elicited to shut down the immune response and prevent
autoimmunity. Cytokines are key messengers in each of the above steps; this thesis
demonstrates that metalloproteinase-dependent regulation of cytokine production and
signaling is a critical axis in the immune response.
A second insight gained from these studies is the observation that epithelial cells
are not bystanders in the immune response. The pervading notion for several years was
that epithelial cells act in a „responsive‟ manner to exogenous insults. Specifically for
mucosal epithelia, dogma maintained that these cells primarily provided physical barriers
to the external environment, or acted as biochemical „effector‟ cells that metabolized
toxin and died off, to be cleared by local immune components or neighboring epithelial
cells. Several lines of evidence in the study of mucosal immunity is largely revising this
notion, establishing the epithelial cell as a „regulatory‟ unit directing local and systemic
immune cell function in a very context-dependent manner [327].
158
Chapters 2, 3 and 4 show how epithelial metalloproteinases directly impact
cytokine biology and cellular immunity. In chapter 2 we show how control of ADAM17-
mediated ectodomain shedding balances pro-apoptotic signaling via Fas and TNF with
pro-survival signaling by EGFR. Chapter 3 demonstrates that non-hematopoietic tissues
are the relevant source of TIMP3 in vivo, and this metalloproteinase inhibitor plays an
immunosuppressive role in preventing undesired immune cell egress into the periphery.
Focusing on ADAM17, chapter 4 develops a new mouse model of atopic dermatitis-like
skin disease by Adam17 deletion in keratinocytes. An important yet unknown role of
ADAM17 in regulating tonic Notch signaling is revealed, where epithelial Notch
activation by ADAM17 prevents epithelial cytokine transcription by c-Fos/AP-1. Thus
we show how the control of epithelial stress signaling by metalloproteinases is an
upstream regulator of mucosal immunity. Throughout each chapter, specific cytokines are
identified as key effectors in guiding cell death, survival, acute and chronic inflammation.
Armed with this knowledge of epithelial/stromal metalloproteinase function in
immunity, new avenues can be pursued to better understand the unique and redundant
roles of Timp genes in mammalian biology. In addition to Timp3, we observed that Timp1
gene expression was also elevated in wildtype mice following death receptor activation
(Figure 2.4A). Given the previous observation that TIMP1 regulates hepatocyte growth
factor (HGF) signaling during liver regeneration [54], it is relevant to pursue the role of
TIMP1 in hepatocyte apoptosis during fulminant liver failure as described in chapter 2.
To date our investigations have focused on individual TIMP deficiencies, however it
remains unknown whether TIMPs act redundantly in regulating metalloproteinase
159
activity throughout the mammalian lifespan. The notion that TIMPs may act as stromal
factors (as demonstrated in chapter 4) are readily investigated in models of immune cell
development, specifically the thymic and bone marrow niches. The introduction of
compound knockout mice harboring multiple TIMP deficiencies allows the study of
potential in vivo redundancies during T and B lymphopoiesis and myelopoiesis.
The ability temporally induce epidermal deletion of Adam17 (generated in chapter
5) has led to the development of more sophisticated murine models to test antigen-
specific barrier immunity in the skin, allowing us to investigate how epidermal ADAM17
impacts immune cell mobilization in clinically relevant models of inflammatory skin
disease. The following sections describe future directions which have emerged from this
thesis, focusing on antigen-specific atopic dermatitis, mechanical cutaneous injury and
the investigation of epithelial TIMPs and ADAM17 in thymopoiesis. An overarching
goal of these future directions is to enhance our understanding of regulated ectodomain
shedding in epithelial:immune crosstalk.
5.1 Manipulating ADAM17 and TIMPs in models of adult skin
inflammation
5.1.1 Antigen-specific atopic dermatitis
Our investigation of epidermal ADAM17 in atopic dermatitis revealed that altered Timp3
gene expression is the most significant change in involved skin from whole genome
analysis of human psoriasis patients (unpublished observations). We have also
consistently noted that aged mice lacking both Timp1 and Timp3 develop spontaneous
160
dermatitis. These observations suggest that regulation of ectodomain shedding by TIMPs
in the skin impacts local inflammation in humans and mice. In chapter 4, we generated an
inducible model of epidermal Adam17 deletion (Adam17epER
mice). These allowed us to
generate ADAM17 deficiency in adult mice and recapitulate a more clinically relevant
phenotype which was not as severe as that observed in A17ep
mice. Moving the genetic
models forward, both Timp1
Timp3
and Adam17epER
mice will be of significant
utility in studying antigen-driven epidermal inflammation. R. Geha and colleagues
illustrate a protocol to induce atopic dermatitis in an antigen-dependent fashion using
repeated topical administration of chicken ovalbumin (OVA) as a protein allergen
(outlined in Figure 5.1A) [376, 377]. This method provides an ideal system to study the
role of TIMP1, TIMP3 and ADAM17 in progressive skin inflammation when compared
to the severe spontaneous disease seen in Adam17ep
mice (discussed in chapter 4). OVA-
mediated induction of atopic dermatitis will allow us to pose the question of how
epidermal TIMP1, TIMP3 and ADAM17 direct lymphocyte infiltration and effector
function, and dendritic cell activation in the skin microenvironment.
In addition to OVA sensitization described above, the use of established T cell
receptor (TCR) transgenic mice will be used to ask whether control of epithelial
ectodomain shedding regulates effector function of “antigen-experienced” CD4+ T cells
in the periphery. In this system, Timp3
, Timp1
Timp3
and Adam17epER
mice are
crossed with OT-II transgenic mice, which are then treated with OVA peptide fragment
(amino acids 323-339). Presumably, the presence of OVA-specific CD4 co-receptor in
161
these mice will significantly accelerate atopic dermatitis compared to the sensitization
model and generate a more acute form of skin inflammation.
Antigen presentation by skin-resident dendritic cells is a key step in maintaining
tolerance to commensals or non-pathogenic antigens while generating cellular and
humoral immunity against invading pathogens. In chapter 4 we identified that resting
splenic CD4+ T cells expressed low levels of Timp genes and thus stromal tissues were
the biologically relevant source of TIMP3 in T-cell mediated autoimmune hepatitis.
However all subsets of human dendritic cells (DCs) are thought to transcribe Timps in
response to infection [378, 379]. In a physiological context, migration of epidermal DCs
(called Langerhans cells) and dermal DCs from the skin to local draining lymph nodes
(dLNs) is required for antigen presentation and generation of immunity. Previous studies
have shown that cell-autonomous MMP2 and MMP9 are necessary for DC migration
through the skin microenvironment. Furthermore DCs produce TIMP1 and 2 to
antagonize undesired MMP activity in an autoregulatory mechanism of balancing
invasion and tissue residence [380, 381]. The genetic models listed in this section allow
us to test the role played by TIMPs in DC migration to dLNs following antigen
encounter, and importantly investigate the capacity of TIMP or ADAM17 deficient DCs
to induce Th2-polarization of dLN-resident T cells. As outlined in Figure 5.1B, a typical
method would involve tape stripping of shaved dorsal skin followed by FITC painting.
Subsequently dLNs would be probed for FITC+CD11c
+ DCs that can polarize T cells.
These cells would then be isolated and co-cultured with OT-II TCR-OVA CD4+ T cells
in the presence of OVA peptide. The observations that Timp1
Timp3
mice exhibit
162
Figure 5.1 – Methods of analyzing antigen-specific responses in atopic dermatitis. (A) OVA sensitization protocol over 50 days, followed by analysis of skin inflammation. (B) Mechanical
injury followed by FITC painting is used to study dermal DC activation and polarization of target T cells.
In our model FITC+ DCs are co-cultured with naïve OT-II TCR-OVA CD4
+ T cells to compare the effect
of individual or multiple Timp or Adam17 deletion.
163
spontaneous dermatitis lead to a hypothesis that DCs lacking Timp1 and Timp3 or
Adam17 would induce significantly greater Th2-polarization of TCR-OVA CD4+ T cells
and result in enhanced cytokine production.
Independent studies have demonstrated that epidermal sensitization to allergens
leads to enhanced airway hyperresponsiveness to the same antigen and generates asthma
[328, 377]. This is consistent with the progression of the human atopic march, where
atopic dermatitis is coupled with bronchial asthma, allergic rhinitis and forms of
gastrointestinal allergic inflammation (e.g. anaphylaxis, ulcerative colitis) [327, 382,
383]. The protective role of ADAM17 in gut epithelium was recently described by two
studies where compromised EGFR activation led to defective epithelial regeneration in
models of colitis [384, 385]. Additionally the in vitro wound healing response of
keratinocytes was also shown to rely on ADAM17-mediated EGFR activation [136]. An
important question in barrier immunity is whether similar mechanisms operate in
different atopic diseases. The models described above have the potential to identify the
contributions of metalloproteinase function in several atopic diseases.
5.1.2 Mechanical cutaneous injury
Repeated tape stripping of the skin recapitulates cutaneous injury that has been
used to study plasmacytoid dendritic cell (pDC) recruitment to the dermis, specifically
Toll-like receptor function and type 1 interferon signaling in response to self-nucleic
acids released by injured keratinocytes [378, 386, 387]. The tape strip test is also used
clinically to diagnose cutaneous lupus, as patients display an exaggerated inflammatory
164
response to mild skin injury when compared to healthy controls [387-389]. Here, an
antigen-specific response (such as that induced by OVA) is not tested. Rather, the wound
healing response induced by pDC activation is a primary focus. The role of TIMPs in
epithelial migration during wound healing remains enigmatic, with conflicting reports
showing beneficial and harmful consequences of metalloproteinase inhibition in
keratinocyte migration [390, 391]. Mice lacking individual or multiple TIMP genes
would provide valuable insight in the study of cutaneous wound healing through control
of pDC activation, interferon signaling or keratinocyte migration. A typical method
would include repeated tape stripping of shaved dorsal skin of TIMP deficient mice and
controls, followed by analysis of involved skin and local draining lymph nodes for pDC
recruitment, cytokine production and the wound healing response at early (12-48h) and
late (7-21 days) timepoints. Together, these systems can potentially enhance our
knowledge of ectodomain shedding in epithelial:immune crosstalk using genetic and
inducible models of cutaneous inflammation.
5.2 Redundancy of TIMPs in hematopoietic niches
5.2.1 TIMPs in the thymic stroma
5.2.1.1 Cell autonomous and stromal effects of Timps on T lymphopoiesis
Petrie and colleagues have mapped the global gene expression patterns in compartments
of the thymic stroma (cortical, medullary, the cortico-medullary junction) as well as
developing thymocytes; TIMP genes are exclusively relegated to the stromal
compartment, while ADAM and MMP expression is more heterogeneous [392, 393].
165
Also, quantitative RT-PCR analysis shows that Timp1, 2 and 4 are expressed at higher
levels by the modified bone marrow stromal cell line OP9-DL1 compared to OP9 control
cells (Figure 5.2). OP9-DL1 cells stably express high levels of the Notch ligand Delta-
like 1 (DL1) and preferentially induce T cell differentiation of uncommitted
hematopoietic progenitors in vitro [394]. Preliminary investigations of T cell
development using the fetal thymus organ culture (FTOC, protocol outlined in Figure
5.3A) system show that: i) small-molecule inhibition of MMPs (using GM6001) and
ADAMs (using TAPI-1) delays or arrests T cell development in a dose dependent manner
(Figure 5.4); and ii) additive loss of TIMPs alters the kinetics of T lymphopoiesis (Figure
5.5). These genetic models are new tools unique to our lab, and permit more thorough
investigation of the fundamental role of Timp genes in thymopoiesis.
Focusing on epithelial: immune crosstalk, we have asked whether regulation of stromal
versus hematopoietic MMPs/ADAMs differentially impacts thymopoiesis. Here the
chimeric FTOC system has utility, where either the stromal or hematopoietic
compartment is deleted for genes of interest. Briefly, Timp mutant FTOCs will be treated
with 2-deoxyguanosine to deplete endogenous hematopoietic cells, and then re-seeded
with congenic (CD45.1) wildtype progenitors (lineage depleted cKit+Sca-1
+) obtained
from E14.5 fetal livers (protocol outlined in Figure 5.3B) [395]. This will functionally
test the role of stromal TIMP genes in development of normal T cell progenitors, and T
cell development will be followed over 21 days.
166
Figure 5.2 Expression of Timp genes in bone marrow stromal cell line OP9-DL1
compared to OP9-GFP. OP9-GFP: blue; OP9-DL1: purple. Data are mean ±S.D. (n=3).
167
Figure 5.3 Methods of analyzing hematopoietic vs. epithelial contributions of TIMPs
and ADAM17 in thymopoiesis using fetal thymus organ culture (FTOC).
(A) Protocol for ex vivo treatment of fetal thymi with small-molecule metalloproteinase inhibitors. (B)
Protocol for generating chimeric FTOC where ADAM17 or TIMP deficient thymi are treated with 2-
deoxyguanosine (2-dGuo) to eliminate hematopoietic cells. These are subsequently seeded with congenic
(CD45.1+) hematopoietic progenitors and T cell development followed via flow cytometry.
168
Figure 5.4 – Pharmacological inhibition of ADAM17 activity arrests T cell development
in a dose-dependent manner. E15.5 fetal thymi were cultured ex vivo for 4 days in the presence of TAPI-1 and T cell development
assessed by flow cytometry. 7-AAD: viability marker. TAPI-1: TNF Alpha Protease Inhibitor-1.
169
Figure 5.5 – Additive loss of Timp genes impacts T cell development with a gene-
dependent decrease in double-negative 1 (DN1) populations.
E15.5 fetal thymi from pups of indicated genotypes were culture ex vivo up to 4 days. CD3
-CD4
-CD8
- T
cells were analyzed for kinetics of T cell development by flow cytometry. Data are representative of 4 fetal
thymi per condition.
170
5.2.1.2 Mechanisms underlying altered T cell development in Timp or Adam17
null mice
Notch signaling is a key component of T cell development and its ablation
induces ectopic B lymphopoiesis in the thymus, as Notch activity is required to suppress
thymic B cell development while permitting T cell development [182, 396, 397].
Specifically, Notch1 provides survival signals of uncommitted hematopoietic progenitors
and later, signaling through Notch1 and its transcriptional co- activator Mastermind-like
(MAML) is required for pre-TCR expression and conventional T cell development
past the double negative 3 (DN3) stage [398-400]. On the other hand, Notch appears
dispensable for T cell development in the thymus. A fraction of this unconventional T
cell subset matures in epithelial tissues, and the epidermis harbors the largest population
of T cells in the periphery [401, 402].
Both ADAM10 and ADAM17 are implicated in processing Notch via the S2
cleavage, which occurs prior to the release of the intracellular signaling domain required
to activate transcription of target genes. We thus rationalize that deletion of multiple
Timps will lead to a loss of regulation over ADAM10/17 activity and disrupt Notch
signaling during T cell development. To this end, developing T cells can be sorted at
different stages of maturation based on cell surface CD25 and CD44 expression in
CD4/CD8 double negative (DN) populations: (DN1: CD44+CD25
-; DN2: CD44
+CD25
+;
DN3a: CD44+CD25
-CD27
low intracellularTCR
-; DN3b: CD44
+CD25
-CD27
high
intracellularTCR+; DN4:CD44
-CD25
-TCR
+ or TCR
+ [403]) and analyzed for Notch
target gene expression. Notch activity can also be independently verified using CBF1-
171
eGFP reporter mice generated by Dr. Gaiano‟s group [404]. Thymic progenitors derived
from these mice will be seeded into FTOC (as in Figure 5.3B) with the selected Timp
mutants mentioned above to measure the impact of Timp deficient thymic epithelium on
Notch signaling.
We have observed that K14-Cre;Adam17fl/fl
mice have altered thymic homeostasis. Since
Keratin14 is expressed in the medullary epithelium of the thymus, we asked whether
Adam17 deletion in this compartment influenced T cell number. A ten-fold decrease in
thymocyte numbers was noted in K14-Cre;Adam17fl/fl
mice (Figure 5.6A). We also
observed disruption of the thymic architecture postnatally, specifically a decrease in the
cellularity of the thymic cortex, where the majority of CD4+CD8
+ T cells reside (Figure
5.6B, C).
Importantly, Lck-Cre;Adam17fl/fl
mice mice did not display this phenotype. Thus
in parallel to the study of TIMP function in T cell development, we will test whether
ADAM17 in the thymic epithelium is required for growth factor signaling (such as
EGFR) and/or developmental cues through Notch. Since our initial observations,
independent work by Vignali and colleagues has confirmed that non-hematopoietic
ADAM17 can impact T cell development [201]; however this study does not directly
address the role of ADAM17 in the thymic epithelium in normal T cell development
since comparisons are made between ADAM17 deficient mice and chimeric genotypes
lacking hematopoietic but not epithelial ADAM17. The authors conclude that since
conventional ADAM17 deficient mice exhibit perturbed thymopoiesis compared to those
172
Figure 5.6 – Decreased thymic cellularity in K14-Cre;Adam17fl/fl
thymus.
(A) Total thymic cell counts of indicated genotypes were obtained from 6-week old mice. (B) Flow
cytometry analysis shows a proportional decrease in CD4+CD8
+ T cells in K14-Cre-Adam17
fl/fl mice. (C)
Histological analysis shows decreased cellularity of the thymic cortex (C) in K14-Cre;Adam17fl/fl
mice. C:
Cortex; M: Medulla. Flow cytometry and histology are representative of 3 mice per condition.
173
lacking hematopoietic ADAM17, the relevant compartment of ADAM17 production is
the thymic epithelium. Our use of mice lacking ADAM17 in the thymic medulla will
provide fundamental insights into the contribution of epithelial ectodomain shedding in
thymocyte development.
174
5.3 Concluding Remarks
Investigation of intracellular signal transduction has vastly contributed to our
understanding of cytokine bioactivity, however, much less is known regarding the
extracellular control of cytokine networks. Numerous investigations of intracellular
signal crosstalk have highlighted the control of heterologous receptor activation and of
intracellular signaling nodes in immunity. Such depth of understanding is lacking with
regards to the extracellular microenvironment where discrete signals simultaneously
converge on the cell surface or where multiple cell types communicate with each other.
An adequate immune response depends on a rapid response to infection, and cytokines
are the messengers that drive this response. Proper integration of cytokine networks
maintains tissue homeostasis and prevents chronic inflammation, autoimmunity and
tumorigenesis. In addition to the studies referenced throughout this thesis, work provided
herein has established the importance of proteolytic cleavage by metalloproteinases in
numerous aspects of immunity. It has provided insights on the lateral control of multiple
cytokines by TIMPs during injury and inflammation, and shown that
compartmentalization of ADAM17-dependent ectodomain shedding can regulate
crosstalk between epithelial and immune cells to guide tissue homeostasis. Nonetheless,
the role of TIMPs and metalloproteinases in immunity still remains a largely undeveloped
field. The principles governing ectodomain shedding described throughout this work
potentially apply to numerous examples of immunopathologies, tumor immunity and
normal tissue homeostasis and further our understanding of the immune response.
175
6 References
1. Jalkanen, M., Biology of cell surface heparan sulfate proteoglycans. Med Biol,
1987. 65(1): p. 41-7.
2. Koda, J.E., A. Rapraeger, and M. Bernfield, Heparan sulfate proteoglycans from
mouse mammary epithelial cells. Cell surface proteoglycan as a receptor for
interstitial collagens. J Biol Chem, 1985. 260(13): p. 8157-62.
3. Peschon, J.J., et al., An essential role for ectodomain shedding in mammalian
development. Science, 1998. 282(5392): p. 1281-4.
4. Khokha, R. and D.T. Denhardt, Matrix metalloproteinases and tissue inhibitor of
metalloproteinases: a review of their role in tumorigenesis and tissue invasion.
Invasion Metastasis, 1989. 9(6): p. 391-405.
5. Maidment, J.M., et al., Matrix metalloproteinase homologues from Arabidopsis
thaliana. Expression and activity. J Biol Chem, 1999. 274(49): p. 34706-10.
6. Murphy, G., et al., Role of TIMPs (tissue inhibitors of metalloproteinases) in
pericellular proteolysis: the specificity is in the detail. Biochem Soc Symp,
2003(70): p. 65-80.
7. Brew, K., D. Dinakarpandian, and H. Nagase, Tissue inhibitors of
metalloproteinases: evolution, structure and function. Biochim Biophys Acta,
2000. 1477(1-2): p. 267-83.
8. Cawston, T.E., et al., Purification of rabbit bone inhibitor of collagenase.
Biochem J, 1981. 195(1): p. 159-65.
9. Docherty, A.J., et al., Sequence of human tissue inhibitor of metalloproteinases
and its identity to erythroid-potentiating activity. Nature, 1985. 318(6041): p. 66-
9.
10. Edwards, D.R., et al., A growth-responsive gene (16C8) in normal mouse
fibroblasts homologous to a human collagenase inhibitor with erythroid-
potentiating activity: evidence for inducible and constitutive transcripts. Nucleic
Acids Res, 1986. 14(22): p. 8863-78.
11. Khokha, R., et al., Antisense RNA-induced reduction in murine TIMP levels
confers oncogenicity on Swiss 3T3 cells. Science, 1989. 243(4893): p. 947-50.
12. Brenner, C.A., et al., Genes for extracellular-matrix-degrading
metalloproteinases and their inhibitor, TIMP, are expressed during early
mammalian development. Genes Dev, 1989. 3(6): p. 848-59.
13. Waterhouse, P., D.T. Denhardt, and R. Khokha, Temporal expression of tissue
inhibitors of metalloproteinases in mouse reproductive tissues during gestation.
Mol Reprod Dev, 1993. 35(3): p. 219-26.
14. Parks, W.C., C.L. Wilson, and Y.S. Lopez-Boado, Matrix metalloproteinases as
modulators of inflammation and innate immunity. Nat Rev Immunol, 2004. 4(8):
p. 617-29.
176
15. Khokha, R., et al., Suppression of invasion by inducible expression of tissue
inhibitor of metalloproteinase-1 (TIMP-1) in B16-F10 melanoma cells. J Natl
Cancer Inst, 1992. 84(13): p. 1017-22.
16. Apte, S.S., M.G. Mattei, and B.R. Olsen, Cloning of the cDNA encoding human
tissue inhibitor of metalloproteinases-3 (TIMP-3) and mapping of the TIMP3
gene to chromosome 22. Genomics, 1994. 19(1): p. 86-90.
17. Gasson, J.C., et al., Molecular characterization and expression of the gene
encoding human erythroid-potentiating activity. Nature, 1985. 315(6022): p. 768-
71.
18. Greene, J., et al., Molecular cloning and characterization of human tissue
inhibitor of metalloproteinase 4. J Biol Chem, 1996. 271(48): p. 30375-80.
19. Leco, K.J., et al., Murine tissue inhibitor of metalloproteinases-4 (Timp-4): cDNA
isolation and expression in adult mouse tissues. FEBS Lett, 1997. 401(2-3): p.
213-7.
20. Stetler-Stevenson, W.G., H.C. Krutzsch, and L.A. Liotta, Tissue inhibitor of
metalloproteinase (TIMP-2). A new member of the metalloproteinase inhibitor
family. J Biol Chem, 1989. 264(29): p. 17374-8.
21. Dean, G., et al., The human tissue inhibitor of metalloproteinases (TIMP)-1 gene
contains repressive elements within the promoter and intron 1. J Biol Chem,
2000. 275(42): p. 32664-71.
22. Edwards, D.R., et al., Involvement of AP1 and PEA3 binding sites in the
regulation of murine tissue inhibitor of metalloproteinases-1 (TIMP-1)
transcription. Biochim Biophys Acta, 1992. 1171(1): p. 41-55.
23. Leco, K.J., et al., Differential regulation of TIMP-1 and TIMP-2 mRNA
expression in normal and Ha-ras-transformed murine fibroblasts. Gene, 1992.
117(2): p. 209-17.
24. Leco, K.J., et al., Tissue inhibitor of metalloproteinases-3 (TIMP-3) is an
extracellular matrix-associated protein with a distinctive pattern of expression in
mouse cells and tissues. J Biol Chem, 1994. 269(12): p. 9352-60.
25. Wick, M., et al., Structure of the human TIMP-3 gene and its cell cycle-regulated
promoter. Biochem J, 1995. 311 ( Pt 2): p. 549-54.
26. Pohar, N., T.A. Godenschwege, and E. Buchner, Invertebrate tissue inhibitor of
metalloproteinase: structure and nested gene organization within the synapsin
locus is conserved from Drosophila to human. Genomics, 1999. 57(2): p. 293-6.
27. Wei, S., et al., Drosophila TIMP is a potent inhibitor of MMPs and TACE:
similarities in structure and function to TIMP-3. Biochemistry, 2003. 42(42): p.
12200-7.
28. Lee, M.H., S. Atkinson, and G. Murphy, Identification of the extracellular matrix
(ECM) binding motifs of tissue inhibitor of metalloproteinases (TIMP)-3 and
effective transfer to TIMP-1. J Biol Chem, 2007. 282(9): p. 6887-98.
29. Fernandez-Catalan, C., et al., Crystal structure of the complex formed by the
membrane type 1-matrix metalloproteinase with the tissue inhibitor of
metalloproteinases-2, the soluble progelatinase A receptor. Embo J, 1998.
17(17): p. 5238-48.
177
30. Iyer, S., et al., Crystal structure of the catalytic domain of matrix
metalloproteinase-1 in complex with the inhibitory domain of tissue inhibitor of
metalloproteinase-1. J Biol Chem, 2007. 282(1): p. 364-71.
31. Morgunova, E., et al., Structural insight into the complex formation of latent
matrix metalloproteinase 2 with tissue inhibitor of metalloproteinase 2. Proc Natl
Acad Sci U S A, 2002. 99(11): p. 7414-9.
32. Huxley-Jones, J., et al., The evolution of the vertebrate metzincins; insights from
Ciona intestinalis and Danio rerio. BMC Evol Biol, 2007. 7: p. 63.
33. Yu, W.P., S. Brenner, and B. Venkatesh, Duplication, degeneration and
subfunctionalization of the nested synapsin-Timp genes in Fugu. Trends Genet,
2003. 19(4): p. 180-3.
34. Murphy, G., A. Murthy, and R. Khokha, Clipping, shedding and RIPping keep
immunity on cue. Trends Immunol, 2008. In press.
35. Mohammed, F.F. and R. Khokha, Thinking outside the cell: proteases regulate
hepatocyte division. Trends Cell Biol, 2005. 15(10): p. 555-63.
36. Fata, J.E., et al., Timp-1 is important for epithelial proliferation and branching
morphogenesis during mouse mammary development. Dev Biol, 1999. 211(2): p.
238-54.
37. Nuttall, R.K., et al., Expression analysis of the entire MMP and TIMP gene
families during mouse tissue development. FEBS Lett, 2004. 563(1-3): p. 129-34.
38. Chirco, R., et al., Novel functions of TIMPs in cell signaling. Cancer Metastasis
Rev, 2006. 25(1): p. 99-113.
39. Chalaris, A., et al., Critical role of the disintegrin metalloprotease ADAM17 for
intestinal inflammation and regeneration in mice. J Exp Med, 2010. 207(8): p.
1617-24.
40. Aimes, R.T. and J.P. Quigley, Matrix metalloproteinase-2 is an interstitial
collagenase. Inhibitor-free enzyme catalyzes the cleavage of collagen fibrils and
soluble native type I collagen generating the specific 3/4- and 1/4-length
fragments. J Biol Chem, 1995. 270(11): p. 5872-6.
41. Xu, J., et al., Proteolytic exposure of a cryptic site within collagen type IV is
required for angiogenesis and tumor growth in vivo. J Cell Biol, 2001. 154(5): p.
1069-79.
42. Chen, M., et al., The recombinant expression of full-length type VII collagen and
characterization of molecular mechanisms underlying dystrophic epidermolysis
bullosa. J Biol Chem, 2002. 277(3): p. 2118-24.
43. Imai, K., et al., Degradation of decorin by matrix metalloproteinases:
identification of the cleavage sites, kinetic analyses and transforming growth
factor-beta1 release. Biochem J, 1997. 322 ( Pt 3): p. 809-14.
44. Tsuruga, E., K. Irie, and T. Yajima, Fibrillin-2 degradation by matrix
metalloproteinase-2 in periodontium. J Dent Res, 2007. 86(4): p. 352-6.
45. Shekhar, M.P., et al., Alterations in galectin-3 expression and distribution
correlate with breast cancer progression: functional analysis of galectin-3 in
breast epithelial-endothelial interactions. Am J Pathol, 2004. 165(6): p. 1931-41.
178
46. Xu, X., et al., Contributions of the MMP-2 collagen binding domain to gelatin
cleavage. Substrate binding via the collagen binding domain is required for
hydrolysis of gelatin but not short peptides. Matrix Biol, 2004. 23(3): p. 171-81.
47. O'Reilly, M.S., et al., Regulation of angiostatin production by matrix
metalloproteinase-2 in a model of concomitant resistance. J Biol Chem, 1999.
274(41): p. 29568-71.
48. Levi, E., et al., Matrix metalloproteinase 2 releases active soluble ectodomain of
fibroblast growth factor receptor 1. Proc Natl Acad Sci U S A, 1996. 93(14): p.
7069-74.
49. Steinberg, Z., et al., FGFR2b signaling regulates ex vivo submandibular gland
epithelial cell proliferation and branching morphogenesis. Development, 2005.
132(6): p. 1223-34.
50. Kim, T.H., et al., Expression and activation of pro-MMP-2 and pro-MMP-9
during rat liver regeneration. Hepatology, 2000. 31(1): p. 75-82.
51. Creemers, E.E., et al., Deficiency of TIMP-1 exacerbates LV remodeling after
myocardial infarction in mice. Am J Physiol Heart Circ Physiol, 2003. 284(1): p.
H364-71.
52. Ikonomidis, J.S., et al., Accelerated LV remodeling after myocardial infarction in
TIMP-1-deficient mice: effects of exogenous MMP inhibition. Am J Physiol Heart
Circ Physiol, 2005. 288(1): p. H149-58.
53. Roten, L., et al., Effects of gene deletion of the tissue inhibitor of the matrix
metalloproteinase-type 1 (TIMP-1) on left ventricular geometry and function in
mice. J Mol Cell Cardiol, 2000. 32(1): p. 109-20.
54. Mohammed, F.F., et al., Metalloproteinase inhibitor TIMP-1 affects hepatocyte
cell cycle via HGF activation in murine liver regeneration. Hepatology, 2005.
41(4): p. 857-67.
55. Lijnen, H.R., et al., Deficiency of tissue inhibitor of matrix metalloproteinase-1
(TIMP-1) impairs nutritionally induced obesity in mice. Thromb Haemost, 2003.
89(2): p. 249-55.
56. Nothnick, W.B., P.D. Soloway, and T.E. Curry, Jr., Pattern of messenger
ribonucleic acid expression of tissue inhibitors of metalloproteinases (TIMPs)
during testicular maturation in male mice lacking a functional TIMP-1 gene. Biol
Reprod, 1998. 59(2): p. 364-70.
57. Lindroos, P.M., R. Zarnegar, and G.K. Michalopoulos, Hepatocyte growth factor
(hepatopoietin A) rapidly increases in plasma before DNA synthesis and liver
regeneration stimulated by partial hepatectomy and carbon tetrachloride
administration. Hepatology, 1991. 13(4): p. 743-50.
58. Schmidt, C., et al., Scatter factor/hepatocyte growth factor is essential for liver
development. Nature, 1995. 373(6516): p. 699-702.
59. Chen, P., et al., Tissue inhibitor of metalloproteinase-1 deficiency abrogates
obliterative airway disease after heterotopic tracheal transplantation. Am J
Respir Cell Mol Biol, 2006. 34(4): p. 464-72.
179
60. Kim, K.H., et al., Tissue inhibitor of metalloproteinase-1 deficiency amplifies
acute lung injury in bleomycin-exposed mice. Am J Respir Cell Mol Biol, 2005.
33(3): p. 271-9.
61. Nothnick, W.B., X. Zhang, and H.E. Zhou, Steroidal regulation of uterine edema
and tissue inhibitors of metalloproteinase (TIMP)-3 messenger RNA expression is
altered in TIMP-1-deficient mice. Biol Reprod, 2004. 70(2): p. 500-8.
62. Nothnick, W.B., Tissue inhibitor of metalloproteinase-1 (TIMP-1) deficient mice
display reduced serum progesterone levels during corpus luteum development.
Endocrinology, 2003. 144(1): p. 5-8.
63. Jaworski, D.M., et al., Tissue inhibitor of metalloproteinase-2(TIMP-2)-deficient
mice display motor deficits. J Neurobiol, 2006. 66(1): p. 82-94.
64. Lluri, G., et al., Tissue inhibitor of metalloproteinase-2 (TIMP-2) regulates
neuromuscular junction development via a beta1 integrin-mediated mechanism. J
Neurobiol, 2006. 66(12): p. 1365-77.
65. Jaworski, D.M., et al., Prepulse inhibition and fear-potentiated startle are altered
in tissue inhibitor of metalloproteinase-2 (TIMP-2) knockout mice. Brain Res,
2005. 1051(1-2): p. 81-9.
66. Caterina, J.J., et al., Inactivating mutation of the mouse tissue inhibitor of
metalloproteinases-2(Timp-2) gene alters proMMP-2 activation. J Biol Chem,
2000. 275(34): p. 26416-22.
67. Wang, Z., R. Juttermann, and P.D. Soloway, TIMP-2 is required for efficient
activation of proMMP-2 in vivo. J Biol Chem, 2000. 275(34): p. 26411-5.
68. Cruz-Munoz, W., et al., Enhanced metastatic dissemination to multiple organs by
melanoma and lymphoma cells in timp-3-/- mice. Oncogene, 2006. 25(49): p.
6489-96.
69. Smookler, D.S., et al., Tissue inhibitor of metalloproteinase 3 regulates TNF-
dependent systemic inflammation. J Immunol, 2006. 176(2): p. 721-5.
70. Cruz-Munoz, W., I. Kim, and R. Khokha, TIMP-3 deficiency in the host, but not
in the tumor, enhances tumor growth and angiogenesis. Oncogene, 2006. 25(4):
p. 650-5.
71. Kassiri, Z. and R. Khokha, Myocardial extra-cellular matrix and its regulation by
metalloproteinases and their inhibitors. Thromb Haemost, 2005. 93(2): p. 212-9.
72. Martin, E.L., et al., Differential response of TIMP-3 null mice to the lung insults
of sepsis, mechanical ventilation, and hyperoxia. Am J Physiol Lung Cell Mol
Physiol, 2005. 289(2): p. L244-51.
73. Mahmoodi, M., et al., Lack of tissue inhibitor of metalloproteinases-3 results in
an enhanced inflammatory response in antigen-induced arthritis. Am J Pathol,
2005. 166(6): p. 1733-40.
74. Fedak, P.W., et al., TIMP-3 deficiency leads to dilated cardiomyopathy.
Circulation, 2004. 110(16): p. 2401-9.
75. Gill, S.E., et al., A null mutation for tissue inhibitor of metalloproteinases-3
(Timp-3) impairs murine bronchiole branching morphogenesis. Dev Biol, 2003.
261(2): p. 313-23.
180
76. Mohammed, F.F., et al., Abnormal TNF activity in Timp3-/- mice leads to chronic
hepatic inflammation and failure of liver regeneration. Nat Genet, 2004. 36(9): p.
969-77.
77. Leco, K.J., et al., Spontaneous air space enlargement in the lungs of mice lacking
tissue inhibitor of metalloproteinases-3 (TIMP-3). J Clin Invest, 2001. 108(6): p.
817-29.
78. Fata, J.E., et al., Accelerated apoptosis in the Timp-3-deficient mammary gland. J
Clin Invest, 2001. 108(6): p. 831-41.
79. Kandalam, V., et al., Early activation of matrix metalloproteinases underlies the
exacerbated systolic and diastolic dysfunction in mice lacking TIMP3 following
myocardial infarction. Am J Physiol Heart Circ Physiol, 2010. 299(4): p. H1012-
23.
80. Ebrahem, Q., et al., Increased neovascularization in mice lacking tissue inhibitor
of metalloproteinases-3. Invest Ophthalmol Vis Sci, 2011.
81. Federici, M., et al., Timp3 deficiency in insulin receptor-haploinsufficient mice
promotes diabetes and vascular inflammation via increased TNF-alpha. J Clin
Invest, 2005. 115(12): p. 3494-505.
82. Koskivirta, I., et al., Mice with tissue inhibitor of metalloproteinases 4 (Timp4)
deletion succumb to induced myocardial infarction but not to cardiac pressure
overload. J Biol Chem, 2010. 285(32): p. 24487-93.
83. English, J.L., et al., Individual Timp deficiencies differentially impact pro-MMP-2
activation. J Biol Chem, 2006. 281(15): p. 10337-46.
84. Toth, M., et al., Tissue inhibitor of metalloproteinase (TIMP)-2 acts
synergistically with synthetic matrix metalloproteinase (MMP) inhibitors but not
with TIMP-4 to enhance the (Membrane type 1)-MMP-dependent activation of
pro-MMP-2. J Biol Chem, 2000. 275(52): p. 41415-23.
85. Seo, D.W., et al., TIMP-2 mediated inhibition of angiogenesis: an MMP-
independent mechanism. Cell, 2003. 114(2): p. 171-80.
86. Weber, B.H., et al., Mutations in the tissue inhibitor of metalloproteinases-3
(TIMP3) in patients with Sorsby's fundus dystrophy. Nat Genet, 1994. 8(4): p.
352-6.
87. Weber, B.H., et al., Sorsby's fundus dystrophy is genetically linked to
chromosome 22q13-qter. Nat Genet, 1994. 7(2): p. 158-61.
88. Langton, K.P., et al., A novel tissue inhibitor of metalloproteinases-3 mutation
reveals a common molecular phenotype in Sorsby's fundus dystrophy. J Biol
Chem, 2000. 275(35): p. 27027-31.
89. Kaur, I., S. Rathi, and S. Chakrabarti, Variations in TIMP3 are associated with
age-related macular degeneration. Proc Natl Acad Sci U S A, 2010. 107(28): p.
E112-3.
90. Yu, Y., et al., Common Variants near FRK/COL10A1 and VEGFA are Associated
with Advanced Age-related Macular Degeneration. Hum Mol Genet, 2011.
91. Garofalo, M., et al., miR-221&222 regulate TRAIL resistance and enhance
tumorigenicity through PTEN and TIMP3 downregulation. Cancer Cell, 2009.
16(6): p. 498-509.
181
92. Selaru, F.M., et al., MicroRNA-21 is overexpressed in human
cholangiocarcinoma and regulates programmed cell death 4 and tissue inhibitor
of metalloproteinase 3. Hepatology, 2009. 49(5): p. 1595-601.
93. Wang, B., et al., TGFbeta-mediated upregulation of hepatic miR-181b promotes
hepatocarcinogenesis by targeting TIMP3. Oncogene, 2010. 29(12): p. 1787-97.
94. Radpour, R., et al., Hypermethylation of tumor suppressor genes involved in
critical regulatory pathways for developing a blood-based test in breast cancer.
PLoS ONE, 2011. 6(1): p. e16080.
95. Sahebjam, S., R. Khokha, and J.S. Mort, Increased collagen and aggrecan
degradation with age in the joints of Timp3(-/-) mice. Arthritis Rheum, 2007.
56(3): p. 905-9.
96. Liu, H., et al., TIMP3: a physiological regulator of adult myogenesis. J Cell Sci,
2010. 123(Pt 17): p. 2914-21.
97. Kassiri, Z., et al., Combination of tumor necrosis factor-alpha ablation and
matrix metalloproteinase inhibition prevents heart failure after pressure overload
in tissue inhibitor of metalloproteinase-3 knock-out mice. Circ Res, 2005. 97(4):
p. 380-90.
98. Murthy, A., et al., Ectodomain shedding of EGFR ligands and TNFR1 dictates
hepatocyte apoptosis during fulminant hepatitis in mice. J Clin Invest, 2010.
120(8): p. 2731-44.
99. Shackel, N.A., The balancing act of hepatocyte apoptosis. Hepatology, 2010.
52(6): p. 2231-3.
100. Koskivirta, I., et al., Tissue inhibitor of metalloproteinases 4 (TIMP4) is involved
in inflammatory processes of human cardiovascular pathology. Histochem Cell
Biol, 2006. 126(3): p. 335-42.
101. Schulze, C.J., et al., Imbalance between tissue inhibitor of metalloproteinase-4
and matrix metalloproteinases during acute myocardial [correction of
myoctardial] ischemia-reperfusion injury. Circulation, 2003. 107(19): p. 2487-92.
102. Felkin, L.E., et al., Expression of extracellular matrix genes during myocardial
recovery from heart failure after left ventricular assist device support. J Heart
Lung Transplant, 2009. 28(2): p. 117-22.
103. Godenschwege, T.A., et al., Inflated wings, tissue autolysis and early death in
tissue inhibitor of metalloproteinases mutants of Drosophila. Eur J Cell Biol,
2000. 79(7): p. 495-501.
104. Nagase, H., R. Visse, and G. Murphy, Structure and function of matrix
metalloproteinases and TIMPs. Cardiovasc Res, 2006. 69(3): p. 562-73.
105. Visse, R. and H. Nagase, Matrix metalloproteinases and tissue inhibitors of
metalloproteinases: structure, function, and biochemistry. Circ Res, 2003. 92(8):
p. 827-39.
106. Lopez-Otin, C. and L.M. Matrisian, Emerging roles of proteases in tumour
suppression. Nat Rev Cancer, 2007. 7(10): p. 800-8.
107. Kenny, H.A., et al., The initial steps of ovarian cancer cell metastasis are
mediated by MMP-2 cleavage of vitronectin and fibronectin. J Clin Invest, 2008.
118(4): p. 1367-79.
182
108. Balbin, M., et al., Loss of collagenase-2 confers increased skin tumor
susceptibility to male mice. Nat Genet, 2003. 35(3): p. 252-7.
109. Quintero, P.A., et al., Matrix metalloproteinase-8 inactivates macrophage
inflammatory protein-1 alpha to reduce acute lung inflammation and injury in
mice. J Immunol, 2010. 184(3): p. 1575-88.
110. Hotary, K., et al., A cancer cell metalloprotease triad regulates the basement
membrane transmigration program. Genes Dev, 2006. 20(19): p. 2673-86.
111. Hotary, K.B., et al., Membrane type I matrix metalloproteinase usurps tumor
growth control imposed by the three-dimensional extracellular matrix. Cell, 2003.
114(1): p. 33-45.
112. Sabeh, F., et al., Tumor cell traffic through the extracellular matrix is controlled
by the membrane-anchored collagenase MT1-MMP. J Cell Biol, 2004. 167(4): p.
769-81.
113. Zhai, Y., et al., Expression of membrane type 1 matrix metalloproteinase is
associated with cervical carcinoma progression and invasion. Cancer Res, 2005.
65(15): p. 6543-50.
114. Barbolina, M.V. and M.S. Stack, Membrane type 1-matrix metalloproteinase:
Substrate diversity in pericellular proteolysis. Semin Cell Dev Biol, 2007.
115. Nagase, H. and M. Kashiwagi, Aggrecanases and cartilage matrix degradation.
Arthritis Res Ther, 2003. 5(2): p. 94-103.
116. Daub, H., et al., Signal characteristics of G protein-transactivated EGF receptor.
EMBO J, 1997. 16(23): p. 7032-44.
117. Fischer, O.M., et al., Oxidative and osmotic stress signaling in tumor cells is
mediated by ADAM proteases and heparin-binding epidermal growth factor. Mol
Cell Biol, 2004. 24(12): p. 5172-83.
118. Fischer, O.M., et al., EGFR signal transactivation in cancer cells. Biochem Soc
Trans, 2003. 31(Pt 6): p. 1203-8.
119. Rooke, J., et al., KUZ, a conserved metalloprotease-disintegrin protein with two
roles in Drosophila neurogenesis. Science, 1996. 273(5279): p. 1227-31.
120. Pan, D. and G.M. Rubin, Kuzbanian controls proteolytic processing of Notch and
mediates lateral inhibition during Drosophila and vertebrate neurogenesis. Cell,
1997. 90(2): p. 271-80.
121. Sotillos, S., F. Roch, and S. Campuzano, The metalloprotease-disintegrin
Kuzbanian participates in Notch activation during growth and patterning of
Drosophila imaginal discs. Development, 1997. 124(23): p. 4769-79.
122. Yan, Y., K. Shirakabe, and Z. Werb, The metalloprotease Kuzbanian (ADAM10)
mediates the transactivation of EGF receptor by G protein-coupled receptors. J
Cell Biol, 2002. 158(2): p. 221-6.
123. Gibb, D.R., et al., ADAM10 is essential for Notch2-dependent marginal zone B
cell development and CD23 cleavage in vivo. J Exp Med, 2010. 207(3): p. 623-35.
124. Gibb, D.R., et al., ADAM10 overexpression shifts lympho- and myelopoiesis by
dysregulating site 2/site 3 cleavage products of Notch. J Immunol, 2011. 186(7):
p. 4244-52.
183
125. Glomski, K., et al., Deletion of Adam10 in endothelial cells leads to defects in
organ-specific vascular structures. Blood, 2011. 118(4): p. 1163-74.
126. Jorissen, E., et al., The disintegrin/metalloproteinase ADAM10 is essential for the
establishment of the brain cortex. J Neurosci, 2010. 30(14): p. 4833-44.
127. Tian, L., et al., ADAM10 is essential for proteolytic activation of Notch during
thymocyte development. Int Immunol, 2008. 20(9): p. 1181-7.
128. Weber, S., et al., The disintegrin/metalloproteinase Adam10 is essential for
epidermal integrity and Notch-mediated signaling. Development, 2011. 138(3): p.
495-505.
129. Chen, G. and D.V. Goeddel, TNF-R1 signaling: a beautiful pathway. Science,
2002. 296(5573): p. 1634-5.
130. Grivennikov, S.I., et al., Intracellular signals and events activated by cytokines of
the tumor necrosis factor superfamily: From simple paradigms to complex
mechanisms. Int Rev Cytol, 2006. 252: p. 129-61.
131. Black, R.A., et al., A metalloproteinase disintegrin that releases tumour-necrosis
factor-alpha from cells. Nature, 1997. 385(6618): p. 729-33.
132. Bell, J., et al., Role of ADAM17 in the ectodomain shedding of TNF-{alpha} and
its receptors by neutrophils and macrophages. J Leukoc Biol, 2007.
133. Blobel, C.P., ADAMs: key components in EGFR signalling and development. Nat
Rev Mol Cell Biol, 2005. 6(1): p. 32-43.
134. Gschwind, A., et al., TACE cleavage of proamphiregulin regulates GPCR-
induced proliferation and motility of cancer cells. EMBO J, 2003. 22(10): p.
2411-21.
135. Lee, D.C., et al., TACE/ADAM17 processing of EGFR ligands indicates a role as
a physiological convertase. Ann N Y Acad Sci, 2003. 995: p. 22-38.
136. Maretzky, T., et al., Migration of growth factor-stimulated epithelial and
endothelial cells depends on EGFR transactivation by ADAM17. Nat Commun,
2011. 2: p. 229.
137. Sahin, U., et al., Distinct roles for ADAM10 and ADAM17 in ectodomain
shedding of six EGFR ligands. J Cell Biol, 2004. 164(5): p. 769-79.
138. Bozkulak, E.C. and G. Weinmaster, Selective use of ADAM10 and ADAM17 in
activation of Notch1 signaling. Mol Cell Biol, 2009. 29(21): p. 5679-95.
139. Brou, C., et al., A novel proteolytic cleavage involved in Notch signaling: the role
of the disintegrin-metalloprotease TACE. Mol Cell, 2000. 5(2): p. 207-16.
140. Saftig, P. and K. Reiss, The "A Disintegrin And Metalloproteases" ADAM10 and
ADAM17: novel drug targets with therapeutic potential? Eur J Cell Biol, 2011.
90(6-7): p. 527-35.
141. Murate, T., et al., Erythroid potentiating activity of tissue inhibitor of
metalloproteinases on the differentiation of erythropoietin-responsive mouse
erythroleukemia cell line, ELM-I-1-3, is closely related to its cell growth
potentiating activity. Exp Hematol, 1993. 21(1): p. 169-76.
142. Stetler-Stevenson, W.G., N. Bersch, and D.W. Golde, Tissue inhibitor of
metalloproteinase-2 (TIMP-2) has erythroid-potentiating activity. FEBS Lett,
1992. 296(2): p. 231-4.
184
143. Baker, A.H., et al., Divergent effects of tissue inhibitor of metalloproteinase-1, -2,
or -3 overexpression on rat vascular smooth muscle cell invasion, proliferation,
and death in vitro. TIMP-3 promotes apoptosis. J Clin Invest, 1998. 101(6): p.
1478-87.
144. Celiker, M.Y., et al., Inhibition of Wilms' tumor growth by intramuscular
administration of tissue inhibitor of metalloproteinases-4 plasmid DNA.
Oncogene, 2001. 20(32): p. 4337-43.
145. Hayakawa, T., et al., Tissue inhibitor of metalloproteinases from human bone
marrow stromal cell line KM 102 has erythroid-potentiating activity, suggesting
its possibly bifunctional role in the hematopoietic microenvironment. FEBS Lett,
1990. 268(1): p. 125-8.
146. Hoegy, S.E., et al., Tissue inhibitor of metalloproteinases-2 (TIMP-2) suppresses
TKR-growth factor signaling independent of metalloproteinase inhibition. J Biol
Chem, 2001. 276(5): p. 3203-14.
147. van der Laan, W.H., et al., Cartilage degradation and invasion by rheumatoid
synovial fibroblasts is inhibited by gene transfer of TIMP-1 and TIMP-3. Gene
Ther, 2003. 10(3): p. 234-42.
148. Petitfrere, E., et al., Involvement of the p38 mitogen-activated protein kinase
pathway in tissue inhibitor of metalloproteinases-1-induced erythroid
differentiation. FEBS Lett, 2000. 485(2-3): p. 117-21.
149. Lizarraga, F., V. Maldonado, and J. Melendez-Zajgla, Tissue inhibitor of
metalloproteinases-2 growth-stimulatory activity is mediated by nuclear factor-
kappa B in A549 lung epithelial cells. Int J Biochem Cell Biol, 2004. 36(8): p.
1655-63.
150. Murphy, F.R., et al., Inhibition of apoptosis of activated hepatic stellate cells by
tissue inhibitor of metalloproteinase-1 is mediated via effects on matrix
metalloproteinase inhibition: implications for reversibility of liver fibrosis. J Biol
Chem, 2002. 277(13): p. 11069-76.
151. Lambert, E., et al., Tissue inhibitor of metalloproteinases-1 signalling pathway
leading to erythroid cell survival. Biochem J, 2003. 372(Pt 3): p. 767-74.
152. Boulday, G., et al., Exogenous tissue inhibitor of metalloproteinase-1 promotes
endothelial cell survival through activation of the phosphatidylinositol 3-
kinase/Akt pathway. Ann N Y Acad Sci, 2004. 1030: p. 28-36.
153. Liu, X.W., et al., Tissue inhibitor of metalloproteinase-1 protects human breast
epithelial cells from extrinsic cell death: a potential oncogenic activity of tissue
inhibitor of metalloproteinase-1. Cancer Res, 2005. 65(3): p. 898-906.
154. Lim, M.S., et al., Tissue inhibitor of metalloproteinase-2 induces apoptosis in
human T lymphocytes. Ann N Y Acad Sci, 1999. 878: p. 522-3.
155. Tran, P.L., et al., Gene therapy for hepatocellular carcinoma using non-viral
vectors composed of bis guanidinium-tren-cholesterol and plasmids encoding the
tissue inhibitors of metalloproteinases TIMP-2 and TIMP-3. Cancer Gene Ther,
2003. 10(6): p. 435-44.
185
156. Bond, M., et al., Tissue inhibitor of metalloproteinase-3 induces a Fas-associated
death domain-dependent type II apoptotic pathway. J Biol Chem, 2002. 277(16):
p. 13787-95.
157. Ahonen, M., et al., Tissue inhibitor of metalloproteinases-3 induces apoptosis in
melanoma cells by stabilization of death receptors. Oncogene, 2003. 22(14): p.
2121-34.
158. Drynda, A., et al., Gene transfer of tissue inhibitor of metalloproteinases-3
reverses the inhibitory effects of TNF-alpha on Fas-induced apoptosis in
rheumatoid arthritis synovial fibroblasts. J Immunol, 2005. 174(10): p. 6524-31.
159. Jiang, Y., et al., Stimulation of mammary tumorigenesis by systemic tissue
inhibitor of matrix metalloproteinase 4 gene delivery. Cancer Res, 2001. 61(6): p.
2365-70.
160. Tummalapalli, C.M., B.J. Heath, and S.C. Tyagi, Tissue inhibitor of
metalloproteinase-4 instigates apoptosis in transformed cardiac fibroblasts. J
Cell Biochem, 2001. 80(4): p. 512-21.
161. Porter, J.F., S. Shen, and D.T. Denhardt, Tissue inhibitor of metalloproteinase-1
stimulates proliferation of human cancer cells by inhibiting a metalloproteinase.
Br J Cancer, 2004. 90(2): p. 463-70.
162. Lovelock, J.D., et al., Heterogeneous effects of tissue inhibitors of matrix
metalloproteinases on cardiac fibroblasts. Am J Physiol Heart Circ Physiol,
2005. 288(2): p. H461-8.
163. Smith, M.R., et al., TIMP-3 induces cell death by stabilizing TNF-alpha receptors
on the surface of human colon carcinoma cells. Cytokine, 1997. 9(10): p. 770-80.
164. Bai, Y.X., et al., Clinicopathologic significance of BAG1 and TIMP3 expression
in colon carcinoma. World J Gastroenterol, 2007. 13(28): p. 3883-5.
165. Han, X., et al., Expression of TIMP-3 gene by construction of a eukaryotic cell
expression vector and its role in reduction of metastasis in a human breast cancer
cell line. Cell Mol Immunol, 2004. 1(4): p. 308-10.
166. House, M.G., et al., Aberrant hypermethylation of tumor suppressor genes in
pancreatic endocrine neoplasms. Ann Surg, 2003. 238(3): p. 423-31; discussion
431-2.
167. Karan, D., et al., Expression of ADAMs (a disintegrin and metalloproteases) and
TIMP-3 (tissue inhibitor of metalloproteinase-3) in human prostatic
adenocarcinomas. Int J Oncol, 2003. 23(5): p. 1365-71.
168. Kim, Y.H., et al., CpG island methylation of genes accumulates during the
adenoma progression step of the multistep pathogenesis of colorectal cancer.
Genes Chromosomes Cancer, 2006. 45(8): p. 781-9.
169. Lui, E.L., et al., DNA hypermethylation of TIMP3 gene in invasive breast ductal
carcinoma. Biomed Pharmacother, 2005. 59 Suppl 2: p. S363-5.
170. Riddick, A.C., et al., Identification of degradome components associated with
prostate cancer progression by expression analysis of human prostatic tissues. Br
J Cancer, 2005. 92(12): p. 2171-80.
186
171. van der Velden, P.A., et al., Expression profiling reveals that methylation of
TIMP3 is involved in uveal melanoma development. Int J Cancer, 2003. 106(4): p.
472-9.
172. Widschwendter, A., et al., DNA methylation in serum and tumors of cervical
cancer patients. Clin Cancer Res, 2004. 10(2): p. 565-71.
173. Johnson, J.L., et al., Suppression of atherosclerotic plaque progression and
instability by tissue inhibitor of metalloproteinase-2: involvement of macrophage
migration and apoptosis. Circulation, 2006. 113(20): p. 2435-44.
174. Ahonen, M., A.H. Baker, and V.M. Kahari, Adenovirus-mediated gene delivery of
tissue inhibitor of metalloproteinases-3 inhibits invasion and induces apoptosis in
melanoma cells. Cancer Res, 1998. 58(11): p. 2310-5.
175. Baker, A.H., et al., Inhibition of invasion and induction of apoptotic cell death of
cancer cell lines by overexpression of TIMP-3. Br J Cancer, 1999. 79(9-10): p.
1347-55.
176. Garton, K.J., P.J. Gough, and E.W. Raines, Emerging roles for ectodomain
shedding in the regulation of inflammatory responses. J Leukoc Biol, 2006. 79(6):
p. 1105-16.
177. Reiss, K., A. Ludwig, and P. Saftig, Breaking up the tie: disintegrin-like
metalloproteinases as regulators of cell migration in inflammation and invasion.
Pharmacol Ther, 2006. 111(3): p. 985-1006.
178. Smalley, D.M. and K. Ley, L-selectin: mechanisms and physiological significance
of ectodomain cleavage. J Cell Mol Med, 2005. 9(2): p. 255-66.
179. McQuibban, G.A., et al., Inflammation dampened by gelatinase A cleavage of
monocyte chemoattractant protein-3. Science, 2000. 289(5482): p. 1202-6.
180. Deftos, M.L., et al., Notch1 signaling promotes the maturation of CD4 and CD8
SP thymocytes. Immunity, 2000. 13(1): p. 73-84.
181. Manilay, J.O., et al., Impairment of thymocyte development by dominant-negative
Kuzbanian (ADAM-10) is rescued by the Notch ligand, delta-1. J Immunol, 2005.
174(11): p. 6732-41.
182. Hayday, A.C. and D.J. Pennington, Key factors in the organized chaos of early T
cell development. Nat Immunol, 2007. 8(2): p. 137-44.
183. Takahama, Y., Journey through the thymus: stromal guides for T-cell
development and selection. Nat Rev Immunol, 2006. 6(2): p. 127-35.
184. Proost, P., S. Struyf, and J. Van Damme, Natural post-translational modifications
of chemokines. Biochem Soc Trans, 2006. 34(Pt 6): p. 997-1001.
185. Van den Steen, P.E., et al., Carboxyterminal cleavage of the chemokines MIG and
IP-10 by gelatinase B and neutrophil collagenase. Biochem Biophys Res
Commun, 2003. 310(3): p. 889-96.
186. Van Den Steen, P.E., et al., Gelatinase B/MMP-9 and neutrophil
collagenase/MMP-8 process the chemokines human GCP-2/CXCL6, ENA-
78/CXCL5 and mouse GCP-2/LIX and modulate their physiological activities. Eur
J Biochem, 2003. 270(18): p. 3739-49.
187
187. Gough, P.J., et al., A disintegrin and metalloproteinase 10-mediated cleavage and
shedding regulates the cell surface expression of CXC chemokine ligand 16. J
Immunol, 2004. 172(6): p. 3678-85.
188. Hundhausen, C., et al., The disintegrin-like metalloproteinase ADAM10 is
involved in constitutive cleavage of CX3CL1 (fractalkine) and regulates CX3CL1-
mediated cell-cell adhesion. Blood, 2003. 102(4): p. 1186-95.
189. Tsou, C.L., C.A. Haskell, and I.F. Charo, Tumor necrosis factor-alpha-converting
enzyme mediates the inducible cleavage of fractalkine. J Biol Chem, 2001.
276(48): p. 44622-6.
190. Schlondorff, J. and C.P. Blobel, Metalloprotease-disintegrins: modular proteins
capable of promoting cell-cell interactions and triggering signals by protein-
ectodomain shedding. J Cell Sci, 1999. 112 ( Pt 21): p. 3603-17.
191. Li, X. and H. Fan, Loss of ectodomain shedding due to mutations in the
metalloprotease and cysteine-rich/disintegrin domains of the tumor necrosis
factor-alpha converting enzyme (TACE). J Biol Chem, 2004. 279(26): p. 27365-
75.
192. Xia, M., et al., Shedding of Fas ectodomain that affects apoptosis of hepatocytes
occurring in regenerative liver. J Gastroenterol, 2002. 37(12): p. 1042-7.
193. Lichtenthaler, S.F., Ectodomain shedding of the amyloid precursor protein:
cellular control mechanisms and novel modifiers. Neurodegener Dis, 2006. 3(4-
5): p. 262-9.
194. Peiretti, F., et al., Proteasome inhibition activates the transport and the
ectodomain shedding of TNF-alpha receptors in human endothelial cells. J Cell
Sci, 2005. 118(Pt 5): p. 1061-70.
195. De Strooper, B., Nicastrin: gatekeeper of the gamma-secretase complex. Cell,
2005. 122(3): p. 318-20.
196. Kramer, H., RIPping notch apart: a new role for endocytosis in signal
transduction? Sci STKE, 2000. 2000(29): p. PE1.
197. Hartmann, D., et al., The disintegrin/metalloprotease ADAM 10 is essential for
Notch signalling but not for alpha-secretase activity in fibroblasts. Hum Mol
Genet, 2002. 11(21): p. 2615-24.
198. Six, E., et al., The Notch ligand Delta1 is sequentially cleaved by an ADAM
protease and gamma-secretase. Proc Natl Acad Sci U S A, 2003. 100(13): p.
7638-43.
199. Maillard, I., T. Fang, and W.S. Pear, Regulation of lymphoid development,
differentiation, and function by the Notch pathway. Annu Rev Immunol, 2005. 23:
p. 945-74.
200. Tan, J.B., et al., Requirement for Notch1 signals at sequential early stages of
intrathymic T cell development. Nat Immunol, 2005. 6(7): p. 671-9.
201. Li, N., et al., Non-cell autonomous expression of TNF-alpha-converting enzyme
ADAM17 is required for normal lymphocyte development. J Immunol, 2007.
178(7): p. 4214-21.
202. Cohn, M., et al., Reflections on the clonal-selection theory. Nat Rev Immunol,
2007. 7(10): p. 823-30.
188
203. Li, N., et al., Metalloproteases regulate T-cell proliferation and effector function
via LAG-3. EMBO J, 2007. 26(2): p. 494-504.
204. Ethell, D.W., R. Kinloch, and D.R. Green, Metalloproteinase shedding of Fas
ligand regulates beta-amyloid neurotoxicity. Curr Biol, 2002. 12(18): p. 1595-
600.
205. Kirkin, V., et al., The Fas ligand intracellular domain is released by ADAM10
and SPPL2a cleavage in T-cells. Cell Death Differ, 2007. 14(9): p. 1678-87.
206. Schulte, M., et al., ADAM10 regulates FasL cell surface expression and
modulates FasL-induced cytotoxicity and activation-induced cell death. Cell
Death Differ, 2007. 14(5): p. 1040-9.
207. Ludwig, A. and C. Weber, Transmembrane chemokines: versatile 'special agents'
in vascular inflammation. Thromb Haemost, 2007. 97(5): p. 694-703.
208. Li, Y., et al., ADAM17 deficiency by mature neutrophils has differential effects on
L-selectin shedding. Blood, 2006. 108(7): p. 2275-9.
209. Faveeuw, C., G. Preece, and A. Ager, Transendothelial migration of lymphocytes
across high endothelial venules into lymph nodes is affected by
metalloproteinases. Blood, 2001. 98(3): p. 688-95.
210. Chen, J.R., et al., Transendothelial migration of lymphocytes in chronic
lymphocytic leukaemia is impaired and involved down-regulation of both L-
selectin and CD23. Br J Haematol, 1999. 105(1): p. 181-9.
211. Gomez-Gaviro, M., et al., Expression and regulation of the metalloproteinase
ADAM-8 during human neutrophil pathophysiological activation and its catalytic
activity on L-selectin shedding. J Immunol, 2007. 178(12): p. 8053-63.
212. Tsakadze, N.L., et al., Tumor necrosis factor-alpha-converting enzyme
(TACE/ADAM-17) mediates the ectodomain cleavage of intercellular adhesion
molecule-1 (ICAM-1). J Biol Chem, 2006. 281(6): p. 3157-64.
213. Garton, K.J., et al., Stimulated shedding of vascular cell adhesion molecule 1
(VCAM-1) is mediated by tumor necrosis factor-alpha-converting enzyme (ADAM
17). J Biol Chem, 2003. 278(39): p. 37459-64.
214. Singh, R.J., et al., Cytokine stimulated vascular cell adhesion molecule-1 (VCAM-
1) ectodomain release is regulated by TIMP-3. Cardiovasc Res, 2005. 67(1): p.
39-49.
215. Van Lint, P., et al., Resistance of collagenase-2 (matrix metalloproteinase-8)-
deficient mice to TNF-induced lethal hepatitis. J Immunol, 2005. 175(11): p.
7642-9.
216. Renckens, R., et al., Matrix metalloproteinase-9 deficiency impairs host defense
against abdominal sepsis. J Immunol, 2006. 176(6): p. 3735-41.
217. Dejana, E., The transcellular railway: insights into leukocyte diapedesis. Nat Cell
Biol, 2006. 8(2): p. 105-7.
218. Weiss, S.J., et al., Oxidative autoactivation of latent collagenase by human
neutrophils. Science, 1985. 227(4688): p. 747-9.
219. Gueders, M.M., et al., Matrix metalloproteinase-8 deficiency promotes
granulocytic allergen-induced airway inflammation. J Immunol, 2005. 175(4): p.
2589-97.
189
220. Gutierrez-Fernandez, A., et al., Increased inflammation delays wound healing in
mice deficient in collagenase-2 (MMP-8). Faseb J, 2007.
221. Reddy, P., et al., Functional analysis of the domain structure of tumor necrosis
factor-alpha converting enzyme. J Biol Chem, 2000. 275(19): p. 14608-14.
222. Gomez, M.I., et al., Bacterial induction of TNF-alpha converting enzyme
expression and IL-6 receptor alpha shedding regulates airway inflammatory
signaling. J Immunol, 2005. 175(3): p. 1930-6.
223. Kinet, J.P., The high-affinity IgE receptor (Fc epsilon RI): from physiology to
pathology. Annu Rev Immunol, 1999. 17: p. 931-72.
224. Kijimoto-Ochiai, S., CD23 (the low-affinity IgE receptor) as a C-type lectin: a
multidomain and multifunctional molecule. Cell Mol Life Sci, 2002. 59(4): p.
648-64.
225. Armant, M., et al., Regulation of cytokine production by soluble CD23:
costimulation of interferon gamma secretion and triggering of tumor necrosis
factor alpha release. J Exp Med, 1994. 180(3): p. 1005-11.
226. Ribbens, C., et al., Increased synovial fluid levels of soluble CD23 are associated
with an erosive status in rheumatoid arthritis (RA). Clin Exp Immunol, 2000.
120(1): p. 194-9.
227. Weskamp, G., et al., ADAM10 is a principal 'sheddase' of the low-affinity
immunoglobulin E receptor CD23. Nat Immunol, 2006. 7(12): p. 1293-8.
228. Lemieux, G.A., et al., The low affinity IgE receptor (CD23) is cleaved by the
metalloproteinase ADAM10. J Biol Chem, 2007. 282(20): p. 14836-44.
229. Carey, B.W., D.Y. Kim, and D.M. Kovacs, Presenilin/gamma-secretase and
alpha-secretase-like peptidases cleave human MHC Class I proteins. Biochem J,
2007. 401(1): p. 121-7.
230. Gonzalez, P.A., et al., Modulation of immunological synapse by membrane-bound
and soluble ligands. Cytokine Growth Factor Rev, 2007. 18(1-2): p. 19-31.
231. Contin, C., et al., Potential role of soluble CD40 in the humoral immune response
impairment of uraemic patients. Immunology, 2003. 110(1): p. 131-40.
232. Contin, C., et al., Membrane-anchored CD40 is processed by the tumor necrosis
factor-alpha-converting enzyme. Implications for CD40 signaling. J Biol Chem,
2003. 278(35): p. 32801-9.
233. Logue, E.C., et al., ICOS-induced B7h shedding on B cells is inhibited by TLR7/8
and TLR9. J Immunol, 2006. 177(4): p. 2356-64.
234. Ebert, E.C., Interleukin-12 up-regulates perforin- and Fas-mediated lymphokine-
activated killer activity by intestinal intraepithelial lymphocytes. Clin Exp
Immunol, 2004. 138(2): p. 259-65.
235. Groh, V., et al., Fas-ligand-mediated paracrine T cell regulation by the receptor
NKG2D in tumor immunity. Nat Immunol, 2006. 7(7): p. 755-62.
236. Kaiser, B.K., et al., Disulphide-isomerase-enabled shedding of tumour-associated
NKG2D ligands. Nature, 2007. 447(7143): p. 482-6.
237. Kaufmann, T., et al., Fatal hepatitis mediated by tumor necrosis factor TNFalpha
requires caspase-8 and involves the BH3-only proteins Bid and Bim. Immunity,
2009. 30(1): p. 56-66.
190
238. Das, M., et al., Induction of hepatitis by JNK-mediated expression of TNF-alpha.
Cell, 2009. 136(2): p. 249-60.
239. Grivennikov, S.I., et al., Distinct and nonredundant in vivo functions of TNF
produced by t cells and macrophages/neutrophils: protective and deleterious
effects. Immunity, 2005. 22(1): p. 93-104.
240. Aggarwal, B.B., Signalling pathways of the TNF superfamily: a double-edged
sword. Nat Rev Immunol, 2003. 3(9): p. 745-56.
241. Chen, X., et al., Bid-independent mitochondrial activation in tumor necrosis
factor alpha-induced apoptosis and liver injury. Mol Cell Biol, 2007. 27(2): p.
541-53.
242. Hughes, P.D., et al., Apoptosis regulators Fas and Bim cooperate in shutdown of
chronic immune responses and prevention of autoimmunity. Immunity, 2008.
28(2): p. 197-205.
243. Corazza, N., et al., TRAIL receptor-mediated JNK activation and Bim
phosphorylation critically regulate Fas-mediated liver damage and lethality. J
Clin Invest, 2006. 116(9): p. 2493-9.
244. Miettinen, P.J., et al., Epithelial immaturity and multiorgan failure in mice
lacking epidermal growth factor receptor. Nature, 1995. 376(6538): p. 337-41.
245. Berasain, C., et al., Novel role for amphiregulin in protection from liver injury. J
Biol Chem, 2005. 280(19): p. 19012-20.
246. Eberle, A., et al., Fluorescence resonance energy transfer analysis of
proapoptotic CD95-EGF receptor interactions in Huh7 cells. Hepatology, 2005.
41(2): p. 315-26.
247. Khai, N.C., et al., In vivo hepatic HB-EGF gene transduction inhibits Fas-
induced liver injury and induces liver regeneration in mice: a comparative study
to HGF. J Hepatol, 2006. 44(6): p. 1046-54.
248. Musallam, L., et al., EGF mediates protection against Fas-induced apoptosis by
depleting and oxidizing intracellular GSH stocks. J Cell Physiol, 2004. 198(1): p.
62-72.
249. Reinehr, R., F. Schliess, and D. Haussinger, Hyperosmolarity and CD95L trigger
CD95/EGF receptor association and tyrosine phosphorylation of CD95 as
prerequisites for CD95 membrane trafficking and DISC formation. FASEB J,
2003. 17(6): p. 731-3.
250. Wisniewska, M., et al., Structural determinants of the ADAM inhibition by TIMP-
3: crystal structure of the TACE-N-TIMP-3 complex. J Mol Biol, 2008. 381(5): p.
1307-19.
251. Tsuji, H., et al., Tumor necrosis factor receptor p55 is essential for intrahepatic
granuloma formation and hepatocellular apoptosis in a murine model of
bacterium-induced fulminant hepatitis. Infect Immun, 1997. 65(5): p. 1892-8.
252. O'Grady, J.G., S.W. Schalm, and R. Williams, Acute liver failure: redefining the
syndromes. Lancet, 1993. 342(8866): p. 273-5.
253. Menghini, R., et al., Tissue inhibitor of metalloproteinase 3 deficiency causes
hepatic steatosis and adipose tissue inflammation in mice. Gastroenterology,
2009. 136(2): p. 663-72 e4.
191
254. Chang, L., et al., The E3 ubiquitin ligase itch couples JNK activation to
TNFalpha-induced cell death by inducing c-FLIP(L) turnover. Cell, 2006. 124(3):
p. 601-13.
255. Kamata, H., et al., Reactive oxygen species promote TNFalpha-induced death and
sustained JNK activation by inhibiting MAP kinase phosphatases. Cell, 2005.
120(5): p. 649-61.
256. Morel, C., et al., Mcl-1 integrates the opposing actions of signaling pathways that
mediate survival and apoptosis. Mol Cell Biol, 2009. 29(14): p. 3845-52.
257. Javelaud, D. and F. Besancon, NF-kappa B activation results in rapid inactivation
of JNK in TNF alpha-treated Ewing sarcoma cells: a mechanism for the anti-
apoptotic effect of NF-kappa B. Oncogene, 2001. 20(32): p. 4365-72.
258. Liu, J. and A. Lin, Role of JNK activation in apoptosis: a double-edged sword.
Cell Res, 2005. 15(1): p. 36-42.
259. Wicovsky, A., et al., Sustained JNK activation in response to tumor necrosis
factor is mediated by caspases in a cell type-specific manner. J Biol Chem, 2007.
282(4): p. 2174-83.
260. Luo, J.L., et al., Inhibition of NF-kappaB in cancer cells converts inflammation-
induced tumor growth mediated by TNFalpha to TRAIL-mediated tumor
regression. Cancer Cell, 2004. 6(3): p. 297-305.
261. Vallabhapurapu, S. and M. Karin, Regulation and function of NF-kappaB
transcription factors in the immune system. Annu Rev Immunol, 2009. 27: p. 693-
733.
262. Maurer, U., et al., Glycogen synthase kinase-3 regulates mitochondrial outer
membrane permeabilization and apoptosis by destabilization of MCL-1. Mol Cell,
2006. 21(6): p. 749-60.
263. Black, R.A., TIMP3 checks inflammation. Nat Genet, 2004. 36(9): p. 934-5.
264. Lee, M.H., et al., Full-length and N-TIMP-3 display equal inhibitory activities
toward TNF-alpha convertase. Biochem Biophys Res Commun, 2001. 280(3): p.
945-50.
265. Natarajan, A., B. Wagner, and M. Sibilia, The EGF receptor is required for
efficient liver regeneration. Proc Natl Acad Sci U S A, 2007. 104(43): p. 17081-6.
266. Alfano, D., I. Iaccarino, and M.P. Stoppelli, Urokinase signaling through its
receptor protects against anoikis by increasing BCL-xL expression levels. J Biol
Chem, 2006. 281(26): p. 17758-67.
267. Faber, A.C., et al., Differential induction of apoptosis in HER2 and EGFR
addicted cancers following PI3K inhibition. Proc Natl Acad Sci U S A, 2009.
106(46): p. 19503-8.
268. Leu, C.M., C. Chang, and C. Hu, Epidermal growth factor (EGF) suppresses
staurosporine-induced apoptosis by inducing mcl-1 via the mitogen-activated
protein kinase pathway. Oncogene, 2000. 19(13): p. 1665-75.
269. Henderson, N.C., et al., Critical role of c-jun (NH2) terminal kinase in
paracetamol- induced acute liver failure. Gut, 2007. 56(7): p. 982-90.
270. Imaeda, A.B., et al., Acetaminophen-induced hepatotoxicity in mice is dependent
on Tlr9 and the Nalp3 inflammasome. J Clin Invest, 2009. 119(2): p. 305-14.
192
271. Lee, W.M., Acetaminophen toxicity: changing perceptions on a social/medical
issue. Hepatology, 2007. 46(4): p. 966-70.
272. Malhi, H., G.J. Gores, and J.J. Lemasters, Apoptosis and necrosis in the liver: a
tale of two deaths? Hepatology, 2006. 43(2 Suppl 1): p. S31-44.
273. Tagami, A., H. Ohnishi, and R.D. Hughes, Increased serum soluble Fas in
patients with acute liver failure due to paracetamol overdose.
Hepatogastroenterology, 2003. 50(51): p. 742-5.
274. Tinel, M., et al., Subliminal Fas stimulation increases the hepatotoxicity of
acetaminophen and bromobenzene in mice. Hepatology, 2004. 39(3): p. 655-66.
275. Zhang, H., et al., Reduction of liver Fas expression by an antisense
oligonucleotide protects mice from fulminant hepatitis. Nat Biotechnol, 2000.
18(8): p. 862-7.
276. Beraza, N., et al., Hepatocyte-specific NEMO deletion promotes NK/NKT cell-
and TRAIL-dependent liver damage. J Exp Med, 2009.
277. Luedde, T., et al., Deletion of NEMO/IKKgamma in liver parenchymal cells
causes steatohepatitis and hepatocellular carcinoma. Cancer Cell, 2007. 11(2): p.
119-32.
278. Sun, B. and M. Karin, NF-kappaB signaling, liver disease and hepatoprotective
agents. Oncogene, 2008. 27(48): p. 6228-44.
279. Geisler, F., et al., Genetic inactivation of RelA/p65 sensitizes adult mouse
hepatocytes to TNF-induced apoptosis in vivo and in vitro. Gastroenterology,
2007. 132(7): p. 2489-503.
280. Maeda, S., et al., IKKbeta couples hepatocyte death to cytokine-driven
compensatory proliferation that promotes chemical hepatocarcinogenesis. Cell,
2005. 121(7): p. 977-90.
281. Inta, I., et al., Bim and Noxa are candidates to mediate the deleterious effect of the
NF-kappa B subunit RelA in cerebral ischemia. J Neurosci, 2006. 26(50): p.
12896-903.
282. Ghosh, S., M.J. May, and E.B. Kopp, NF-kappa B and Rel proteins:
evolutionarily conserved mediators of immune responses. Annu Rev Immunol,
1998. 16: p. 225-60.
283. Song, L., et al., IKKbeta programs to turn on the GADD45alpha-MKK4-JNK
apoptotic cascade specifically via p50 NF-kappaB in arsenite response. J Cell
Biol, 2006. 175(4): p. 607-17.
284. Gong, Y., et al., Induction of BIM is essential for apoptosis triggered by EGFR
kinase inhibitors in mutant EGFR-dependent lung adenocarcinomas. PLoS
medicine, 2007. 4(10): p. e294-e294.
285. Reginato, M.J., et al., Integrins and EGFR coordinately regulate the pro-
apoptotic protein Bim to prevent anoikis. Nat Cell Biol, 2003. 5(8): p. 733-40.
286. Horiuchi, K., et al., Cutting edge: TNF-alpha-converting enzyme
(TACE/ADAM17) inactivation in mouse myeloid cells prevents lethality from
endotoxin shock. J Immunol, 2007. 179(5): p. 2686-9.
287. Kim, A.H., et al., Akt phosphorylates and negatively regulates apoptosis signal-
regulating kinase 1. Mol Cell Biol, 2001. 21(3): p. 893-901.
193
288. Berasain, C., et al., Amphiregulin: an early trigger of liver regeneration in mice.
Gastroenterology, 2005. 128(2): p. 424-32.
289. Wetzel, M., et al., Tissue inhibitor of metalloproteinases-3 facilitates Fas-
mediated neuronal cell death following mild ischemia. Cell Death Differ, 2008.
15(1): p. 143-51.
290. Junttila, M.R., S.P. Li, and J. Westermarck, Phosphatase-mediated crosstalk
between MAPK signaling pathways in the regulation of cell survival. FASEB J,
2008. 22(4): p. 954-65.
291. Xia, Z., et al., Opposing effects of ERK and JNK-p38 MAP kinases on apoptosis.
Science, 1995. 270(5240): p. 1326-31.
292. Michalopoulos, G.K., Liver regeneration after partial hepatectomy: critical
analysis of mechanistic dilemmas. Am J Pathol, 2010. 176(1): p. 2-13.
293. Crispe, I.N., The liver as a lymphoid organ. Annu Rev Immunol, 2009. 27: p.
147-63.
294. Ichiki, Y., et al., T cell immunity in autoimmune hepatitis. Autoimmun Rev, 2005.
4(5): p. 315-21.
295. Croft, M., The role of TNF superfamily members in T-cell function and diseases.
Nat Rev Immunol, 2009. 9(4): p. 271-85.
296. Medzhitov, R., Origin and physiological roles of inflammation. Nature, 2008.
454(7203): p. 428-35.
297. Bonder, C.S., et al., Rules of recruitment for Th1 and Th2 lymphocytes in
inflamed liver: a role for alpha-4 integrin and vascular adhesion protein-1.
Immunity, 2005. 23(2): p. 153-63.
298. Erhardt, A., et al., IL-10, regulatory T cells, and Kupffer cells mediate tolerance
in concanavalin A-induced liver injury in mice. Hepatology, 2007. 45(2): p. 475-
85.
299. Jiang, W., et al., TLR-9 activation aggravates concanavalin A-induced hepatitis
via promoting accumulation and activation of liver CD4+ NKT cells. J Immunol,
2009. 182(6): p. 3768-74.
300. Murphy, G., A. Murthy, and R. Khokha, Clipping, shedding and RIPping keep
immunity on cue. Trends Immunol, 2008. 29(2): p. 75-82.
301. Consolo, M., et al., Matrix metalloproteinases and their inhibitors as markers of
inflammation and fibrosis in chronic liver disease (Review). Int J Mol Med, 2009.
24(2): p. 143-52.
302. Murthy, A., et al., Ectodomain shedding of EGFR ligands and TNFR1 dictates
hepatocyte apoptosis during fulminant hepatitis in mice. J Clin Invest, 2010.
303. Boisvert, J., et al., Liver-infiltrating lymphocytes in end-stage hepatitis C virus:
subsets, activation status, and chemokine receptor phenotypes. J Hepatol, 2003.
38(1): p. 67-75.
304. Tiegs, G., J. Hentschel, and A. Wendel, A T cell-dependent experimental liver
injury in mice inducible by concanavalin A. J Clin Invest, 1992. 90(1): p. 196-
203.
305. Toyabe, S., et al., Requirement of IL-4 and liver NK1+ T cells for concanavalin
A-induced hepatic injury in mice. J Immunol, 1997. 159(3): p. 1537-42.
194
306. Jin, X., et al., Paradoxical effects of short- and long-term interleukin-6 exposure
on liver injury and repair. Hepatology, 2006. 43(3): p. 474-84.
307. Klein, C., et al., The IL-6-gp130-STAT3 pathway in hepatocytes triggers liver
protection in T cell-mediated liver injury. J Clin Invest, 2005. 115(4): p. 860-9.
308. Duran, A., et al., Crosstalk between PKCzeta and the IL4/Stat6 pathway during T-
cell-mediated hepatitis. EMBO J, 2004. 23(23): p. 4595-605.
309. Deng, Z.B., et al., Immature myeloid cells induced by a high-fat diet contribute to
liver inflammation. Hepatology, 2009. 50(5): p. 1412-20.
310. Zenewicz, L.A., et al., Interleukin-22 but not interleukin-17 provides protection to
hepatocytes during acute liver inflammation. Immunity, 2007. 27(4): p. 647-59.
311. Crispe, I.N., Hepatic T cells and liver tolerance. Nat Rev Immunol, 2003. 3(1): p.
51-62.
312. Echeverria, I., et al., Enhanced T cell responses against hepatitis C virus by ex
vivo targeting of adenoviral particles to dendritic cells. Hepatology, 2011. 54(1):
p. 28-37.
313. Palucka, K., J. Banchereau, and I. Mellman, Designing vaccines based on biology
of human dendritic cell subsets. Immunity, 2010. 33(4): p. 464-78.
314. Das, A., et al., Functional skewing of the global CD8 T cell population in chronic
hepatitis B virus infection. J Exp Med, 2008. 205(9): p. 2111-24.
315. Raziorrouh, B., et al., Inhibitory Molecules that Regulate Expansion and
Restoration of HCV-Specific CD4+ T Cells in Patients with Chronic Infection.
Gastroenterology, 2011.
316. Egen, J.G., et al., Intravital imaging reveals limited antigen presentation and T
cell effector function in mycobacterial granulomas. Immunity, 2011. 34(5): p.
807-19.
317. Heydtmann, M. and D.H. Adams, Chemokines in the immunopathogenesis of
hepatitis C infection. Hepatology, 2009. 49(2): p. 676-88.
318. Sahin, H., C. Trautwein, and H.E. Wasmuth, Functional role of chemokines in
liver disease models. Nat Rev Gastroenterol Hepatol, 2010. 7(12): p. 682-90.
319. Casrouge, A., et al., Evidence for an antagonist form of the chemokine CXCL10 in
patients chronically infected with HCV. J Clin Invest, 2011. 121(1): p. 308-17.
320. Leifeld, L., et al., Early up-regulation of chemokine expression in fulminant
hepatic failure. J Pathol, 2003. 199(3): p. 335-44.
321. Amend, B., et al., Induction of autoimmunity by expansion of autoreactive
CD4+CD62Llow cells in vivo. J Immunol, 2006. 177(7): p. 4384-90.
322. Cauwels, A., et al., Caspase inhibition causes hyperacute tumor necrosis factor-
induced shock via oxidative stress and phospholipase A2. Nat Immunol, 2003.
4(4): p. 387-93.
323. Hasselblatt, P., et al., Hepatocyte survival in acute hepatitis is due to c-Jun/AP-1-
dependent expression of inducible nitric oxide synthase. Proc Natl Acad Sci U S
A, 2007. 104(43): p. 17105-10.
324. Eksteen, B., et al., Immune-mediated liver injury. Semin Liver Dis, 2007. 27(4):
p. 351-66.
195
325. Dienes, H.P. and U. Drebber, Pathology of immune-mediated liver injury. Dig
Dis, 2010. 28(1): p. 57-62.
326. Bieber, T., Atopic dermatitis. N Engl J Med, 2008. 358(14): p. 1483-94.
327. Swamy, M., et al., Epithelial decision makers: in search of the 'epimmunome'. Nat
Immunol, 2010. 11(8): p. 656-65.
328. Demehri, S., et al., Skin-derived TSLP triggers progression from epidermal-
barrier defects to asthma. PLoS Biol, 2009. 7(5): p. e1000067.
329. Ziegler, S.F. and D. Artis, Sensing the outside world: TSLP regulates barrier
immunity. Nat Immunol, 2010. 11(4): p. 289-93.
330. Dumortier, A., et al., Atopic dermatitis-like disease and associated lethal
myeloproliferative disorder arise from loss of Notch signaling in the murine skin.
PLoS One, 2010. 5(2): p. e9258.
331. Meixner, A., et al., Epidermal JunB represses G-CSF transcription and affects
haematopoiesis and bone formation. Nat Cell Biol, 2008. 10(8): p. 1003-11.
332. Blanpain, C., et al., Canonical notch signaling functions as a commitment switch
in the epidermal lineage. Genes Dev, 2006. 20(21): p. 3022-35.
333. Demehri, S., et al., Notch-deficient skin induces a lethal systemic B-
lymphoproliferative disorder by secreting TSLP, a sentinel for epidermal
integrity. PLoS Biol, 2008. 6(5): p. e123.
334. Fort, M.M., et al., IL-25 induces IL-4, IL-5, and IL-13 and Th2-associated
pathologies in vivo. Immunity, 2001. 15(6): p. 985-95.
335. Ogura, H., et al., Interleukin-17 promotes autoimmunity by triggering a positive-
feedback loop via interleukin-6 induction. Immunity, 2008. 29(4): p. 628-36.
336. Muller, M., et al., CXCR3 signaling reduces the severity of experimental
autoimmune encephalomyelitis by controlling the parenchymal distribution of
effector and regulatory T cells in the central nervous system. J Immunol, 2007.
179(5): p. 2774-86.
337. Fang, D., et al., Dysregulation of T lymphocyte function in itchy mice: a role for
Itch in TH2 differentiation. Nat Immunol, 2002. 3(3): p. 281-7.
338. Campbell, P.J. and A.R. Green, The myeloproliferative disorders. N Engl J Med,
2006. 355(23): p. 2452-66.
339. Guinea-Viniegra, J., et al., TNFalpha shedding and epidermal inflammation are
controlled by Jun proteins. Genes Dev, 2009. 23(22): p. 2663-74.
340. Leung, D.Y., et al., New insights into atopic dermatitis. J Clin Invest, 2004.
113(5): p. 651-7.
341. Pasparakis, M., et al., TNF-mediated inflammatory skin disease in mice with
epidermis-specific deletion of IKK2. Nature, 2002. 417(6891): p. 861-6.
342. Artavanis-Tsakonas, S., M.D. Rand, and R.J. Lake, Notch signaling: cell fate
control and signal integration in development. Science, 1999. 284(5415): p. 770-
6.
343. Guttman-Yassky, E., et al., Broad defects in epidermal cornification in atopic
dermatitis identified through genomic analysis. J Allergy Clin Immunol, 2009.
124(6): p. 1235-1244 e58.
196
344. Nair, R.P., et al., Genome-wide scan reveals association of psoriasis with IL-23
and NF-kappaB pathways. Nat Genet, 2009. 41(2): p. 199-204.
345. Olsson, M., et al., Increased expression of aquaporin 3 in atopic eczema. Allergy,
2006. 61(9): p. 1132-7.
346. Reischl, J., et al., Increased expression of Wnt5a in psoriatic plaques. J Invest
Dermatol, 2007. 127(1): p. 163-9.
347. Saenz, S.A., B.C. Taylor, and D. Artis, Welcome to the neighborhood: epithelial
cell-derived cytokines license innate and adaptive immune responses at mucosal
sites. Immunol Rev, 2008. 226: p. 172-90.
348. Eferl, R. and E.F. Wagner, AP-1: a double-edged sword in tumorigenesis. Nat
Rev Cancer, 2003. 3(11): p. 859-68.
349. Chu, J., et al., Repression of activator protein-1-mediated transcriptional
activation by the Notch-1 intracellular domain. J Biol Chem, 2002. 277(9): p.
7587-97.
350. Fukushima, H., et al., The association of Notch2 and NF-kappaB accelerates
RANKL-induced osteoclastogenesis. Mol Cell Biol, 2008. 28(20): p. 6402-12.
351. Talora, C., et al., Specific down-modulation of Notch1 signaling in cervical
cancer cells is required for sustained HPV-E6/E7 expression and late steps of
malignant transformation. Genes Dev, 2002. 16(17): p. 2252-63.
352. Harada, M., et al., TSLP Promoter Polymorphisms are Associated with
Susceptibility to Bronchial Asthma. Am J Respir Cell Mol Biol, 2010.
353. Zenz, R., et al., Psoriasis-like skin disease and arthritis caused by inducible
epidermal deletion of Jun proteins. Nature, 2005. 437(7057): p. 369-75.
354. Esteve, P., et al., SFRPs act as negative modulators of ADAM10 to regulate
retinal neurogenesis. Nat Neurosci, 2011. 14(5): p. 562-9.
355. van Tetering, G., et al., Metalloprotease ADAM10 is required for Notch1 site 2
cleavage. J Biol Chem, 2009. 284(45): p. 31018-27.
356. Nicolas, M., et al., Notch1 functions as a tumor suppressor in mouse skin. Nat
Genet, 2003. 33(3): p. 416-21.
357. Williams, S.E., et al., Asymmetric cell divisions promote Notch-dependent
epidermal differentiation. Nature, 2011. 470(7334): p. 353-8.
358. Chu, J. and E.H. Bresnick, Evidence that C promoter-binding factor 1 binding is
required for Notch-1-mediated repression of activator protein-1. J Biol Chem,
2004. 279(13): p. 12337-45.
359. Pastore, S., et al., Dysregulated activation of activator protein 1 in keratinocytes
of atopic dermatitis patients with enhanced expression of
granulocyte/macrophage-colony stimulating factor. J Invest Dermatol, 2000.
115(6): p. 1134-43.
360. Song, Z., et al., Polymorphic nucleotides within the human IL-4 promoter that
mediate overexpression of the gene. J Immunol, 1996. 156(2): p. 424-9.
361. Islam, S.A., et al., Mouse CCL8, a CCR8 agonist, promotes atopic dermatitis by
recruiting IL-5+ T(H)2 cells. Nat Immunol, 2011. 12(2): p. 167-77.
362. Biton, M., et al., Epithelial microRNAs regulate gut mucosal immunity via
epithelium-T cell crosstalk. Nat Immunol, 2011. 12(3): p. 239-46.
197
363. Pasparakis, M., Regulation of tissue homeostasis by NF-kappaB signalling:
implications for inflammatory diseases. Nat Rev Immunol, 2009. 9(11): p. 778-
88.
364. Benjamini, Y.a.H., Y., Controlling the false discovery rate: a practical and
powerful approach to multiple testing. Journal of the Royal Statistical Society,
Series B (Methodological), 1995. 57(1): p. 289-300.
365. Pavlidis, P. and W.S. Noble, Matrix2png: a utility for visualizing matrix data.
Bioinformatics, 2003. 19(2): p. 295-6.
366. Eisen, M.B., et al., Cluster analysis and display of genome-wide expression
patterns. Proc Natl Acad Sci U S A, 1998. 95(25): p. 14863-8.
367. Saldanha, A.J., Java Treeview--extensible visualization of microarray data.
Bioinformatics, 2004. 20(17): p. 3246-8.
368. Ehebauer, M., P. Hayward, and A. Martinez-Arias, Notch signaling pathway. Sci
STKE, 2006. 2006(364): p. cm7.
369. Harada, H., et al., Localization of putative stem cells in dental epithelium and
their association with Notch and FGF signaling. J Cell Biol, 1999. 147(1): p.
105-20.
370. Jeffries, S. and A.J. Capobianco, Neoplastic transformation by Notch requires
nuclear localization. Mol Cell Biol, 2000. 20(11): p. 3928-41.
371. Kopan, R., Notch: a membrane-bound transcription factor. J Cell Sci, 2002.
115(Pt 6): p. 1095-7.
372. Liu, Z.J., et al., Regulation of Notch1 and Dll4 by vascular endothelial growth
factor in arterial endothelial cells: implications for modulating arteriogenesis
and angiogenesis. Mol Cell Biol, 2003. 23(1): p. 14-25.
373. Ohishi, K., et al., Monocytes express high amounts of Notch and undergo cytokine
specific apoptosis following interaction with the Notch ligand, Delta-1. Blood,
2000. 95(9): p. 2847-54.
374. Wu, L., et al., Identification of a family of mastermind-like transcriptional
coactivators for mammalian notch receptors. Mol Cell Biol, 2002. 22(21): p.
7688-700.
375. Lichti, U., J. Anders, and S.H. Yuspa, Isolation and short-term culture of primary
keratinocytes, hair follicle populations and dermal cells from newborn mice and
keratinocytes from adult mice for in vitro analysis and for grafting to
immunodeficient mice. Nat Protoc, 2008. 3(5): p. 799-810.
376. Jin, H., et al., Animal models of atopic dermatitis. J Invest Dermatol, 2009.
129(1): p. 31-40.
377. Spergel, J.M., et al., Epicutaneous sensitization with protein antigen induces
localized allergic dermatitis and hyperresponsiveness to methacholine after single
exposure to aerosolized antigen in mice. J Clin Invest, 1998. 101(8): p. 1614-22.
378. Chu, C.C., P. Di Meglio, and F.O. Nestle, Harnessing dendritic cells in
inflammatory skin diseases. Semin Immunol, 2011. 23(1): p. 28-41.
379. Johnson, T.R., et al., Primary human mDC1, mDC2, and pDC dendritic cells are
differentially infected and activated by respiratory syncytial virus. PLoS ONE,
2011. 6(1): p. e16458.
198
380. Baratelli, F.E., et al., Prostaglandin E2-dependent enhancement of tissue
inhibitors of metalloproteinases-1 production limits dendritic cell migration
through extracellular matrix. J Immunol, 2004. 173(9): p. 5458-66.
381. Ratzinger, G., et al., Matrix metalloproteinases 9 and 2 are necessary for the
migration of Langerhans cells and dermal dendritic cells from human and murine
skin. J Immunol, 2002. 168(9): p. 4361-71.
382. MacDonald, T.T. and A. Di Sabatino, The immunologic basis for gastrointestinal
food allergy. Curr Opin Gastroenterol, 2009. 25(6): p. 521-6.
383. Spergel, J.M., From atopic dermatitis to asthma: the atopic march. Ann Allergy
Asthma Immunol, 2010. 105(2): p. 99-106; quiz 107-9, 117.
384. Brandl, K., et al., MyD88 signaling in nonhematopoietic cells protects mice
against induced colitis by regulating specific EGF receptor ligands. Proc Natl
Acad Sci U S A, 2010. 107(46): p. 19967-72.
385. Chalaris, A., et al., Critical role of the disintegrin metalloprotease ADAM17 for
intestinal inflammation and regeneration in mice. J Exp Med, 2010.
386. Gregorio, J., et al., Plasmacytoid dendritic cells sense skin injury and promote
wound healing through type I interferons. J Exp Med, 2010. 207(13): p. 2921-30.
387. Guiducci, C., et al., Autoimmune skin inflammation is dependent on plasmacytoid
dendritic cell activation by nucleic acids via TLR7 and TLR9. J Exp Med, 2010.
207(13): p. 2931-42.
388. Wenzel, J. and T. Tuting, An IFN-associated cytotoxic cellular immune response
against viral, self-, or tumor antigens is a common pathogenetic feature in
"interface dermatitis". J Invest Dermatol, 2008. 128(10): p. 2392-402.
389. Wenzel, J., et al., The expression pattern of interferon-inducible proteins reflects
the characteristic histological distribution of infiltrating immune cells in different
cutaneous lupus erythematosus subsets. Br J Dermatol, 2007. 157(4): p. 752-7.
390. Gao, L., L. Chao, and J. Chao, A novel signaling pathway of tissue kallikrein in
promoting keratinocyte migration: activation of proteinase-activated receptor 1
and epidermal growth factor receptor. Exp Cell Res, 2010. 316(3): p. 376-89.
391. Miyoshi, H., et al., Beneficial effects of tissue inhibitor of metalloproteinases-2
(TIMP-2) on chronic dermatitis. J Dermatol, 2005. 32(5): p. 346-53.
392. Griffith, A.V., et al., Spatial mapping of thymic stromal microenvironments
reveals unique features influencing T lymphoid differentiation. Immunity, 2009.
31(6): p. 999-1009.
393. Petrie, H.T. and J.C. Zuniga-Pflucker, Zoned out: functional mapping of stromal
signaling microenvironments in the thymus. Annu Rev Immunol, 2007. 25: p.
649-79.
394. Schmitt, T.M. and J.C. Zuniga-Pflucker, Induction of T cell development from
hematopoietic progenitor cells by delta-like-1 in vitro. Immunity, 2002. 17(6): p.
749-56.
395. Ramsdell, F., J.C. Zuniga-Pflucker, and Y. Takahama, In vitro systems for the
study of T cell development: fetal thymus organ culture and OP9-DL1 cell
coculture. Curr Protoc Immunol, 2006. Chapter 3: p. Unit 3 18.
199
396. Anderson, G. and E.J. Jenkinson, Lymphostromal interactions in thymic
development and function. Nat Rev Immunol, 2001. 1(1): p. 31-40.
397. Yuan, J.S., et al., Functions of notch signaling in the immune system: consensus
and controversies. Annu Rev Immunol, 2010. 28: p. 343-65.
398. Ciofani, M. and J.C. Zuniga-Pflucker, Notch promotes survival of pre-T cells at
the beta-selection checkpoint by regulating cellular metabolism. Nat Immunol,
2005. 6(9): p. 881-8.
399. Guidos, C.J., Synergy between the pre-T cell receptor and Notch: cementing the
alphabeta lineage choice. J Exp Med, 2006. 203(10): p. 2233-7.
400. Maillard, I., et al., The requirement for Notch signaling at the beta-selection
checkpoint in vivo is absolute and independent of the pre-T cell receptor. J Exp
Med, 2006. 203(10): p. 2239-45.
401. Barbee, S.D., et al., Skint-1 is a highly specific, unique selecting component for
epidermal T cells. Proc Natl Acad Sci U S A, 2011. 108(8): p. 3330-5.
402. Hayday, A.C., Gammadelta T cells and the lymphoid stress-surveillance
response. Immunity, 2009. 31(2): p. 184-96.
403. Taghon, T., et al., Developmental and molecular characterization of emerging
beta- and gammadelta-selected pre-T cells in the adult mouse thymus. Immunity,
2006. 24(1): p. 53-64.
404. Duncan, A.W., et al., Integration of Notch and Wnt signaling in hematopoietic
stem cell maintenance. Nat Immunol, 2005. 6(3): p. 314-22.