h exchanger, nhe1: analyses of the 3d structure of nhe1
TRANSCRIPT
D E T H U M A N I S T I S K E F A K U L T E T
K Ø B E N H A V N S U N I V E R S I T E T
Begrebet tilstande
Online counter
Supplerende
informationer
U n i v e r s i t y o f C o p e n h a g e n
D e p a r t m e n t o f B i o l o g y
Structure-function studies of the plasma membrane Na+/H+
exchanger, NHE1: analyses of the 3D structure of NHE1 using
Site-Directed Spin Labeling and Electron Spin Resonance
Master Thesis by Gabriel Peder Bjerre January 2010
2
PREFACE
This thesis represents the final part of my education as a candidate in biochemistry at the Faculty of
Sciences, University of Copenhagen, and is based upon experiments conducted from September 2008 to
September 2009 under the supervision of Associate Professor Stine Falsig Pedersen. The experimental
work was initiated at the Section of Cell and Developmental Biology situated at the August Krogh
Building, Department of Biology, University of Copenhagen and finalized at the laboratory of Professor
Peter M. Cala, Department of Physiology and Membrane Biology, School of medicine, University of
California Davis. Here I also collaborated with the laboratory of Professor John C. Voss, Department of
Biochemistry and Molecular Medicine, School of Medicine, University of California Davis.
I would like to thank my supervisor Stine Falsig Pedersen for giving me the opportunity to work in the
challenging and exciting field of structure-function relations of transporter proteins, and also for her
guidance and invaluable support and help throughout this process.
I would like to thank all professors, associates and students at the laboratories, especially Peter M. Cala,
John C. Voss, and Madhu Budamagunta for making the experimental process an interesting and most
enjoyable one.
On a personal level I would like to thank my family and friends for their caring and support. In particular
I wish to thank Vicky, Martin and Annette for their love and support.
Part of the results presented in this thesis will be presented at the first annual Joint Meeting of the
Scandinavian and German Physiological Societies in March 2010 in Copenhagen as either a poster or oral
presentation. In addition, parts of my results are included in the following research article expected to
be submitted for publishing in early 2010:
Eva B. Nygaard, Jens O. Lagerstedt, Gabriel Bjerre, Kristian A. Poulsen, Stine Meinild, Robert R. Rigor, Stine F. Pedersen, John C. Voss, and Peter M. Cala; STRUCTURAL MODELING AND ELECTRON PARAMAGNETIC RESONANCE SPECTROSCOPY OF THE HUMAN Na+/H+ EXCHANGER ISOFORM 1. Expected submission for Journal of Biological Chemistry early 2010. Manuscript presented in the appendix section of this thesis.
3
ABSTRACT
The ubiquitous plasma membrane Na+/H+ exchanger, NHE1, plays pivotal roles in regulation of cellular
pH and volume homeostasis, and is implicated in the control of multiple physiological functions,
including cell migration, proliferation, and cell death. NHE1 dysfunction is implicated in cancer
development and ischemic cell injury, hence, the ability to selectively inhibit NHE1 is of substantial
clinical importance. Little is known about the three-dimensional (3D) structure of NHE1 but, recent
evidence from a comparative study indentified a two-amino acid-motif in the TMIV of the inhibitor
insensitive paralog from Pleuronectes Americanus paNHE1 that, when substituted with the motif from
the human hNHE1 mediated inhibitor sensitivity. In addition, regions in TMX and XI were found to be
important for inhibitor binding and, since inhibitors of NHE1 are known to be competitive with the
substrate Na+, also ion translocation.
Based on the recently solved crystal structure of the Escherichia coli NhaA paralog and a cysteine
accessibility assay a model structure of the transmembrane (TM) part of hNHE1 was created and is
presented in this thesis. The model structure confirms important roles for TMIV and TMXI in ion
exchange by hNHE1. The overall aim of the present study was to elucidate the structure-function
relations of the TM region of NHE1 in general and the mechanism with which Na+/H+ exchange occurs in
particular. I employed molecular biology techniques to reintroduce cysteine residues by point mutation
and expressed the recombinant NHE1 protein in a mammalian cell line, AP-1. Protein expression and
function was assayed by Western blotting and functional screening respectively. I created an optimized
protein purification protocol and the structure-function relations of resulting NHE1 protein were studied
by Electron Paramagnetic Resonance (EPR) analysis of spinlabels in TMIV and TMXI.
I present here experimental evidence, in the form of interspin distances determined by spectral analysis
of EPR data, that show residues Ala173 and Ile461, situated in TMIV and TMXI of hNHE1 respectively, are
within 20Å proximity. This is in congruence with the presented model structure. I also present evidence
of the corresponding residues, Ala164 and Ile452, of the paralog paNHE1 to be within 20Å proximity. This
is, to the best of my knowledge, the first experimental evidence cementing the proximity of TMIV and
TMXI in hNHE1, and in the lower vertebrate homolog paNHE1. Thus, this is the most solid evidence to
date that they may actually by analogy to the data for NhaA, form the central part of the ion
translocation pathway. Interestingly, these distances are altered in response to low pH (pH=5.5),
representing an activated state of the exchanger, consistent with conformational changes of TMIV and
TMXI being central to ion translocation by NHE1. In congruence with previous studies I found the TMIV-
TMXI distance of hNHE1 to be altered in response to the NHE1 inhibitor cariporide (10mM), while the
corresponding distance in the inhibitor insensitive paNHE1 was unaltered. As Together these data are
experimental validation of our model structure of hNHE1.
4
RESUMÉ
Na+/H+ exchanger isoform 1, NHE1, der findes i stort set alle celletyper i vertebrater, spiller centrale roller i regulering af cellulær pH og volumen, og er impliceret i kontrollen af mange fysiologiske funktioner inklusiv celle- migration, proliferation og celledød. NHE1 dysfunktion er involveret i udviklingen af cancer og celledød forbundet med iskæmi. Muligheden for at kunne hæmme NHE1 selektivt er derfor af klinisk vigtighed. Den tre-dimensionelle (3D) struktur af NHE1 er ukendt men gennem komparative studier af den inhibitor insensitive paralog fra Pleuronectes Americanus paNHE1 og den humane hNHE1 blev aminosyrer i TMIV samt TMX og XI fundet at være vigtige for inhibitor binding. Da inhibitorer af NHE1 fungerer delvist kompetitivt med substratet Na+ må disse regioner også være vigtige for ion-translokation.
Baseret på krystal strukturen af den bakterielle paralog Escherichia coli NhaA og et cysteine accessibility assay skabte vi en model struktur af det transmembrane (TM) domæne af hNHE1. I denne model struktur er TMIV og TMXI essentielle for ion-translokation. Det overordnede formål med dette projekt var at undersøge relationer mellem struktur of funktion af NHE1 TM domænet generelt og mekanismen bag N+/H+ transport specifikt. Jeg har gennem molekylær biologiske teknikker introduceret aminosyren cysteine på specifikke positioner i TMIV og/eller TMXI og udtrykt disse proteiner i den mammale celle linie AP-1. Protein ekspression og funktion blev undersøgt ved Western blots og funtionstest og jeg har lavet en optimeret protokol til oprensing af rekombinant NHE1. Det oprensede NHE1 protein blev spinlabeled i TMIV og/eller TMXI og analyseret ved Elektron Paramagnetisk Resonance (EPR).
Jeg præsenterer her eksperimentel bevis ,i form af spin-spin afstande analyseret fra EPR data, der viser at Ala173 i TMIV og Ile461 i TMXI af hNHE1 er indenfor 20Å. Afstanden mellem de korresponderende aminosyrer i paNHE1, Ala164 i TMIV og Ile452 i TMXI, er ligeledes 20Å. Så vidt jeg ved er dette de første eksperimentelle beviser for at TMIV og TMXI er tæt på hinanden i NHE1. Jeg viser at afstanden mellem TMIV og TMXI ændres i respons til lav pH hvilket er konsistent med at konformationelle ændringer af TMIV/TMXI er central for ion-translokation. Ydermere præsenterer jeg bevis for at TMIV-TMXI afstanden i hNHE1 ændres i reponse til inhibering ved cariporide og at den samme afstand var uændret i den inhibitor insensitive paralog paNHE1. Disse data er eksperimentel validering af vores model struktur.
5
ABBREVIATIONS
Three-dimensional 3D Transmembrane TM Electron Paramagnetic Resonance EPR Na+/H+ exchanger isoform 1 NHE1 Buffer capacity β Carbon dioxide CO2 Cation Proton Antiporter superfamily CPA Human NHE1 hNHE1 Pleuronectes Americanus NHE1 paNHE1 SoLute Carrier protein family SLC Intracellular pH pHi
Extracellular pH pHo Intracellular loop IL Extracellular loop EL Extracellular signal-Related Kinase ERK ERK effector Ribosomal S kinase p90rsk Mitogen-Activated Protein Kinase MAPK Rho-associated kinase p160rock Protein Kinase B PKB Nck interacting kinase NIK Ca/calmodulin dependent kinase II CaMKII Calmodulin CaM Calcineurin Homologous Protein CHP Carbonic anhydrase II CAII Phosphatidylinositol 4,5-bisphosphate PIP2 Ezrin, radixin and moesin ERM-proteins Stine Falsig Pedersen SFP Peter M. Cala PMC Cryo-electron microscopy cryo-EM Multiconformation continuum electrostatics MCCE Molecular dynamics MD Site-directed spin labeling SDSL Methanethiosulfonate spin label MTS Chinese Hamster Ovary CHO Fetal bovine serum FBS Phosphate buffered saline PBS Tris-HCl buffered saline TBS Geniticin G418 Polymerase chain reaction PCR Bovine Serum Albumin BSA n-Dodecyl β-D-maltoside DDM Ethylene glycol tetraacetic acid EGTA
6
TABLE OF CONTENTS
Cover 1
Preface 2
Abstract 3
Resumé 4
Abbreviations 5
1. INTRODUCTION 9
1.1 Intracellular pH regulation 9
1.2 Taxonomy of the Na+/H+ transporter family 12
1.3 Introduction to NHE1 15
1.4 Driving force and kinetics of NHE1 16
1.5 Activation and regulation of NHE1 17
1.6 Physilogy of NHE1 19
1.7 Pathophysiology of NHE1 20
1.8 Inhibitors of NHE1 21
1.9 Structure of NHE1 21
1.10 Structure of NhaA 24
1.11 Model structure of NHE1 31
1.12 Electron Paramagnetic Resonance 35
2. AIM 40
3. MATERIALS AND METHODS 41
3.1 Cells and culturing 41
3.2 Cloning and expression of NHE1 constructs 42
3.3 Site-directed mutagenesis 42
7
3.4 Transformation, DNA purification, and sequencing 44
3.5 Transfection 45
3.6 Clonal selection 45
3.7 Lysates 46
3.8 Protein determination 46
3.9 SDS-PAGE 46
3.10 Western blotting 47
3.11 Silver staining 47
3.12 Confocal Laser Scanning Microscopy 48
3.13 Functional screening 48
3.14 Slot blot 51
3.15 Protein purification 52
3.16 EPR spectroscopy 56
4. RESULTS 57
4.1 Assessment of NHE1 clones 58
- Western blotting 58
- Functional assays 60
- CLSM images 63
4.2 Protein purification 64
4.3 EPR results 69
- hNHE1 A173C/I461C 71
- hNHE1 V160C/H473C 73
- paNHE1 A164C/I452C 74
5. DISCUSSION 76
Protein environment of EPR experiments 76
Discussion and evaluation of distance determination 77
Discussion and evaluation of structure-function 78
8
Critical assessment of methods 80
6. CONCLUSION 83
REFERENCES 84
APPENDICES 89
1. Nygaard et al. manuscript
2. Site-directed mutagenesis
3. Cell cultures
4. Transformation, DNA purification and sequencing
5. Transfection
6. Protein determination
7. Western blotting
8. CLSM
9. Functional screening
10. Protein purification
11. EPR samples
12. Ramachandran plot
9
1. INTRODUCTION
The focus of my project is the structure-function relations of the mammalian plasma membrane Na+/H+
exchanger isoform 1 (NHE1). The carrier-mediated transport of Na+ ions in exchange for protons is
universally detected throughout the various phyla, from the bacterial Na+-H+ antiporter NhaA to the
mammalian Na+-H+ exchanger NHE1 (1). The ability to regulate cellular pH, volume, and ion composition
is a basic prerequisite for life as we know it, and these proteins are a fundamental part of this
mechanism. I will begin by a brief description of intracellular pH (pHi) and the mechanism of its
regulation. I will then proceed with an introduction of the Na+/H+ exchanger protein family with special
emphasis on the mammalian isoforms. Lastly I will give a thorough description of NHE1, covering aspects
such as structure, function and regulation.
1.1 Intracellular pH regulation
The ability to regulate pHi is a fundamental prerequisite for the normal function of any cell in any
organism (2). The reason is that changes in pHi, from here on defined as the pH in the aqueous cytosolic
solution, affect the ionization state of all weak acids and weak bases in the cell (3). The nearly
inconceivably large array of cellular molecules affected includes all peptides and proteins – although to a
highly variable extent, depending e.g. on the pKa values of exposed side chains, and thus changes in pHi
have the potential to alter almost any cellular process. Therefore it is no surprise to find that all animal
cells examined tightly regulate their pHi within a narrow range, generally around values 6.9-7.4
depending on the cell type (4-7).
D-[Glucose] [Pyruvate]
+ 2 [NAD]+ + 2 [ADP] + 2 [P]i 2 + 2 [NADH] ++ 2H+ + 2 [ATP] + 2 H2O
[H+]↑
H
H
oiH
i
o10H
E12.3mV60mVVm
12.3mV7.47.261.5mVE
pHpHF
RT2.3E
H
Hlog
F
RT2.3E
H+
pHi=7.2pH
o=7.4
Vm
=-60mV
Figure 1.1 - At physiological conditions cells (here pHo=7.4, pHi=7.2 and Vm=-60mV at 37˚C) are constantly exposed to acid loading: Top) Cellular metabolism, such as the process of glycolysis shown here, produces an excess of H
+.
Bottom) The electrochemical potential (EH) of H+ calculated under these hypothetical physiological conditions is less
negative than the membrane potential (Vm), hence H+ tend to passively enter the cell. Taken together with the fact
that HCO3- and OH
- tend to passively exit the cell this imposes a chronic acidification of the cytosol.
10
At physiological conditions the complex apparatus of pHi homeostasis is running smoothly to counter
the chronic acidification of the cytosol caused by the passive influx of H+ due to the net inward
electrochemical gradient of H+, and the net H+ production of the cellular metabolism (Figure 1.1). At
steady state conditions H+ is non-passively extruded from the cell. An acute perturbation in pHi, such as
it occurs in neurons during excitation (6), will accelerate the pHi homeostasis apparatus to counter this
perturbation and return the pHi to its steady state. A simplified view of the pHi homeostasis apparatus is
seen in Figure 1.2. Briefly, it can be said that the apparatus has a passive component, namely the
capacity to withstand changes in pHi - the so called buffer capacity (β), and components that actively
regulate pHi.
HBBHn1n
H+
H+
H+
H+
H+
H+
A buffer is made of a weak acid (Brøndsted definition) and its conjugate weak base and several such
buffer pairs are responsible for buffering the cytosol. For instance carbon dioxide (CO2) can passively
cross the membrane in accordance with its concentration gradient and react with H2O to form H2CO3
which is a divalent weak acid that, along with its conjugate weak base HCO3- is very important for
buffering the cytosol. Upon addition of a strong acid, H+ from the strong acid reacts with the conjugate
weak base of the buffer pair, HCO3- in our example, and thus drives the reaction equilibrium towards the
formation of the weak acid H2CO3. Upon addition of a strong base, OH- from the strong base reacts with
the weak acid, H2CO3 in our example, thus driving the reaction equilibrium towards the formation of the
weak base HCO3-. In this manner the buffering reaction consumes strong acid (or strong base) in the
production of a weak acid (or weak base) and the result is a minor change in pHi. From this it is clear that
the capacity (β) of a buffer is dependent on the total concentration of buffer and the equilibrium
constant of the buffer reaction. The most important buffer in biological systems is the carbonic acid –
Figure 1.2 – Schematic representation of the pHi homeostasis apparatus of a cell. In this simplified view the passive component is shown as the chemical equilibrium of a weak Brøndsted acid (BH
n+1) and its conjugate base (B
n) and the
active component is shown as acid extruders (left) and acid loaders (right).
11
bicarbonate system, for which the general equation describing β of a closed system is shown in Equation
1.1 along with the other important buffer systems and their pKa values in Table 1.1.
Equation 1.1
Table 1.1
The buffer systems of the cell are able to minimize any fluctuations but cannot return pHi to its steady
state. To do this the cell has several membrane proteins that can be categorized into two groups; acid
extruders and acid loaders, both of which transport either H+ or HCO3- across the plasma membrane
(Figure 1.3). Acid extruders use energy to either extrude H+ from the cell or import HCO3- and can be
separated into primary and secondary active transporters (3). The primary active transporters are the
vacuolar (V-type) H+ ATPases and the P-type H+/K+ ATPases, both of which uses the energy of ATP
hydrolysis to extrude H+ in cells (8). The first is found in a variety of tissues and cell types whereas the
latter is found primarily in secretory epithelial cells of the stomach called parietal cells where it extrudes
H+ in exchange for K+ and is responsible for the acidification of the stomach contents (9).
H+
V-type H+ pump
H+
Na+
H+
Na-H Exchanger
Cl–
HCO3–
2
Na+
Na-Driven Cl-HCO3
Exchanger
Na/HCO3 Cotransporter(1:2 stoichiometry)HCO3
–
2
Na+
HCO3–
3
Na+
Na/HCO3 Cotransporter(1:3 stoichiometry)
HCO3–
Cl–
Cl-HCO3 Exchanger
Acid Extruders Acid Loaders
pHi pHi
OH–
HCO3–
Metabolism
H+
Weak acid Conjugate base pKa H2CO3 → HCO3
- + H
+ 6.37
H2PO4-
→ HPO42-
+ H+
7.21 NH4
+ → NH3 + H
+ 9.24
Figure 1.3 – Acid loaders and extruders. Regulation of pHi depends on the balance between the depicted acid extruders and acid loaders. Leak entry (of H
+) and extrusion (of OH
- and HCO3
-) pathways, primary transport (V-
type H+ pump), and secondary transport ( Na
+/H
+ exchanger, Na
+-dependent Cl
-/HCO3
- exchanger, Cl
-/HCO3
-
exchanger, Na+/HCO3
- cotransporter) all contribute to the complex interplay that is pHi regulation. (Modified
from Boron et al., 2004 (3)
12
The secondary active transporters are all, at least in part, driven by the Na+ gradient maintained by the
Na+/K+ ATPase. These include the Na+/H+ exchangers, which extrude H+ in the exchange for Na+ and are
allosterically activated once pHi falls below a certain threshold value(10,11), and the Na+-dependent Cl-
/HCO3- exchangers which are also allosterically regulated by H+ and are activated once pHi falls below a
certain set point (2,3,12). Acid loaders include the aforementioned passive influx of H+, and passive
efflux of OH- and HCO3-. The Na+-independent Cl-/HCO3
- exchangers mediate electroneutral extrusion of
HCO3- in exchange for Cl- (13). Alkaline pH stimulates these proteins both through substrate availability
and the presence of an allosteric H+ binding site. The Na+/HCO3- cotransporters can function both as
acid- extruder and loader. They can function as electrogenic transporters with a Na+:HCO3- stoichiometry
of 1:2, in which case they operate as acid extruders, or a stoichiometry of 1:3, in which case they
operate as acid loaders. Furthermore, they can operate as electroneutral cotransporters that also
mediate acid extrusion (3).
The Na+/H+ exchanger plays a pivotal role in the regulation of pHi (11,14) and in the next section I will
discuss the taxonomy of the protein family.
1.2 Taxonomy of the Na+/H+ transporter family
The recent advances in technology have rendered the task of sequencing entire genomes an achievable
one. As a result databases, like GenBank, are booming with genomes that await functional analysis.
Phylogenetic analysis gives insight into the evolution of Na+/H+ transporter family and can form an
effective framework in which characteristics such as function, location, ion transport and specificity,
structure-function relations, and regulation can be analyzed.
The Na+/H+ transporter family is part of the monovalent cation proton antiporter superfamily (CPA).
Proteins of the CPA superfamily are characterized by 10-12 TM helices and some sequence similarity
(1).The superfamily is subdivided into 3 groups; CPA1 includes the eukaryotic NHE family, CPA2 includes
bacterial-, plant and fungal-, and animal genes, lastly the NaT-DC family includes the distantly related
mammalian sperm NHE genes (1). The evolutionary relations among NHE family members can be used in
comparative biology to provide insight into how structure relates to function in the CPA superfamily. Of
the CPA families, the eukaryotic NHE clade is the best characterized, and can be divided into two major
clades, an intracellular and a plasma membrane clade (see respectively, the top shaded region, IC, and
the bottom shaded region, PM, in Figure 1.4), based on subcellular location (1). Brett et al. subdivides
the intracellular clade into three subgroups; endosomal/TGN, plant vacuolar, and NHE8-like. NHE8 has
later been shown to reside in the apical membrane of the proximal tubule and should perhaps be placed
in the plasma membrane clade (15). The plasma membrane clade can be subdivided into two subgroups;
recycling and resident. Based on their extensive phylogenetic analysis Bret et al. propose that the NHE
gene family originated as intracellular or organellar Na+, K+, H+ exchangers as seen in fungi. The evolution
of Na+, K+, H+ exchangers with specialized functions for the specific organelle in which they locate
resulted in offshots such as the plant vacuolar clade and the NHE8-like clade, which is exclusive to
animals. The next evolutionary step was the presence of recycling NHE gene products as seen in both
endosomes and the plasma membrane of metazoan organisms.
13
The latest evolutionary event is the occurrence of resident plasma membrane NHE gene products as first
emerged in fish and later in all vertebrates. The recently cloned NHE1 from the RBCs of the winter
flounder is an example (16). It is within this last evolutionary step we find the human NHE1 (hNHE1)
which is the main focus of my project. The winter flounder (paNHE1) paralog is also found in the
resident plasma membrane clade and is also a focus of my project (16). Interestingly the nine distinct
genes of mammalian NHE family are spread across 4 of the 5 subclades, suggesting that the eukaryotic
NHE family is quite diverse (1,17). Comparing the gene lengths and the protein sequence similarities also
testifies to this diversity (Table 1.2 and (17)). NHE1 through 5 are found in the plasma membrane clade;
NHE1, 2 and 4 in the resident subclade and NHE3 and 5 in the recycling subclade (1). NHE6 through 9 are
found in the evolutionary older intracellular clade; HNE6, 7 and 9 in the Endosomal/TGN subclade and
NHE8 in the NHE8-like subclade but recent evidence suggest it should be placed in the plasma
membrane clade (15). Another interesting characteristic revealed by this phylogenetic study, by Brett et
al., 2005, is the fact that the proteins of the resident plasma membrane clade have the longest
cytoplasmic COOH tails, in correlation with the increased complexity of interactions with the
cytoskeleton and other cell components of higher eukaryotes, of the family (1). The fact that the
members of the plasma membrane clade are all sensitive, although to a varied degree, to the inhibitor
Figure 1.4 – Phylogenetic tree of the NHE family. This is the unrooted tree of 118 eukaryotic CPA1 NHE genes. The eukaryotic NHE isoform 1 belongs to the Plasma Membrane major clade (bottom) in the resident subclade (orange). See text for further details or Brett et al., 2005 (1).
14
amiloride and none of the members of the intracellular clade are speaks of the correctness of this
classification.
At present nine mammalian isoforms of NHE are cloned and characterized and NHE1-9 are all part of the
SoLute Carrier protein family (SLC) (reviewed in(17)). Located in the plasma membrane the NHE1 is
ubiquitously expressed and considered the housekeeping form of the protein. NHE1 is the most
extensively studied isoform. NHE2-5 are also located in the plasma membrane but in a more restricted
range of tissues (Table 1.2). NHE6, 7 and 9 are, in contrast, located in the membrane of intracellular
compartments. Much less is known about NHE8, which shares only around 25% sequence identity with
the other NHEs (1,17).
Gene name
Gene length NCBI GeneID Protein name Protein length Tissue distribution Subcellular location
SLC9A1
56,097 6548 NHE1 815 Ubiquitous Basolateral plasma membrane
SLC9A2
91,644 6549 NHE2 812 Stomach, intestine, muscle
Apical plasma membrane
SLC9A3 51,023 6550 NHE3 834 Intestine, stomach, kidney, gall bladder
Apical plasma membrane and recycling endosomes
SLC9A4
60,000 6552 NHE4 888 Stomach Basolateral plasma membrane
SLC9A5
23,242 6553 NHE5 896 Brain Plasma membrane and recycling synaptic vesicles
SLC9A6
61,666 10479 NHE6 669-701 Ubiquitous Organellar, recycling endosomes
SLC9A7
152,100 84679 NHE7 725 Ubiquitous Organellar
SLC9A8
81,077 23315 NHE8 666 Ubiquitous Plasma membrane and organellar
SLC9A9
583,226 285195 NHE9 645 Ubiquitous Organellar and apical plasma membrane
The major focus of my Master’s project is the hNHE1 protein. As I outlined above, NHE1 it is part of the
CPA family that is detected throughout the various phyla, hence the carrier-mediated transport of Na+
ions in exchange for protons has been found essential always. As evolution took its course members of
the CPA family specialized in transporting monovalent cations and protons across different membranes
and most recently the resident plasma membrane clade, of which NHE1 is a member, emerged. In the
next section I will turn my focus to NHE1, covering function, regulation, structure and more.
Table 1.2 – The SLC9 gene family. This table is modified from Brett et al., 2005 (1) and Orlowski & Grinstein, 2004 (17) and serves to display the human NHE gene family. It is seen that the protein lengths of the members located intracellular (NHE6, 7 and 9) are generally shorter than members located in the plasma membrane (NHE1, 2, 3, 4 and 5). In contrast the gene for NHE9 (thought to be the evolutionary oldest paralog) is by far the longest.
15
1.3 Introduction to NHE1
The mammalian Na+/H+ exchanger 1 (NHE1) is the best characterized NHE isoform and is highly
conserved throughout the vertebrate phylum, where NHE1 is expressed in nearly all cell types. Located
in the plasma membrane NHE1 has profound effects on normal cell function, as we shall see later, NHE1
dysfunction results in a number of pathological states. NHE1 is characterized as a secondary active
transport protein, i.e. NHE1 is an exchanger, and functions by extruding one intracellular H+ in exchange
for one extracellular Na+ (14,18,19). NHE1 is hence not electrogenic. NHE1 play pivotal roles in the
maintenance and regulation of pHi and cell volume (Figure 1.5). Through a series of protein-protein
interactions NHE1 also plays a role in the reorganization of the cytoskeleton and probably also in a range
of signaling processes (20-22). Many essential cellular processes are influenced, or even initiated, by
fluctuations in pHi, cell volume, and corresponding cytoskeletal reorganization. Hence NHE1 is involved
in the control of cellular processes such as; survival, proliferation, migration and adhesion (23,24). The
importance of NHE1 in these many physiological functions along with the ubiquitous expression has
resulted in the NHE1 being called the housekeeping form of the NHE family.
Figure 1.5 – The major physiological roles of NHE1. A - NHE1 plays a pivotal role in the homeostasis of pHi
along with Na+/HCO3
- cotransporters, Na
+ dependent
HCO3-/Cl
- exchangers and Cl
-/HCO3
- exchangers
among others. The importance of pHi regulation is discussed in Section 1.1.
B - NHE1 serves as a major Na+ entry pathway in
many cell types, and as such it regulates cell volume via Na
+ fluxes along with the Cl
- /HCO3
- exchangers,
and Na+-, K
+-, and Cl
- pumps among others.
Modified from Orlowski & Grinstein, 2004 (17)
)
16
The mammalian NHE1 consists of around 813-822 amino acids and has a predicted molecular mass of
87-91kD. A number of studies have demonstrated that NHE1 functions as a dimer (25-27), but each
monomer is able to perform ion translocation at acidic pHi (28). The NHE1 protein has two structurally
and functionally distinct regions; a transmembrane (TM) region consisting of the N-terminal first ~500
amino acids and a cytosolic tail consisting of the C-terminal last ~300 amino acids. A study by
Wakabayashi et al. in 1992 (29) proved the TM region to be needed and sufficient for ion translocation
and subsequent research has established the cytosolic C-terminal as the regulatory region of NHE1
(19,30). The mature form of NHE1 is localized to the plasma membrane and both N- and O-glycosylated,
of which the O-linked glycosylation is essential for correct localization and the N-linked appears to be
non-essential (31).
1.4 Driving force and kinetics of NHE1
Under physiological conditions NHE1 exchange of Na+ and H+ is, at least in part, driven by the
electrogenic Na+/K+ ATPase that maintains a steep inward electrochemical Na+ gradient (17). The force
that drives the electroneutral Na+/H+ exchange via NHE1 is a function of the gradients of Na+ and H+ as
shown in Equation 1.1, where R is the gas constant, T is the absolute temperature and ln is the natural
logarithm. The equation is solved for the hypothetical cell in which the extracellular Na+ concentration is
145mM and the intracellular 15mM and the pHo=7.4 and the pHi=7.2, at 37˚C. At physiological
conditions and steady state in this hypothetical cell there is a large inwardly directed Na+ gradient and a
smaller outward H+ gradient. Hence there is a substantial thermodynamic driving force for net Na+
uptake and H+ efflux (32).
Equation 1.1
Cations other than Na+, such as Li+ and H+, can act as substrate and NHE1 will exchange one extracellular
Li+ for one intracellular H+ (32). In addition to this Na+/Na+ exchange has been shown in some cell types
(32). In the odd event that the Na+ and H+ ion gradients changes sufficiently, i.e. non-physiological
conditions, NHE1 can act in reverse mode and will then extrude one intracellular Na+ for one
extracellular H+ (32).
The method by which NHE1 exchanges Na+ and H+ is suggested to be a transport cycle consisting of a
Na+ favored conformation and a H+ favored conformation. A single site is thought to be sufficient for the
consecutive binding of Na+ and H+. Indeed binding of extracellular Na+ to NHE1 follows a simple
Michaelis-Menten kinetics and has a Hill coefficient of around 1, in accordance with the presence of just
17
one Na+ binding site (32,33). The apparent Km for Na+ binding to its external site is about 3 times below
the physiological extracellular Na+ concentration. Hence, at normal physiological conditions NHE1 is
close to saturation with regards to extracellular Na+ (32). NHE1’s interaction with intracellular H+ (Hi+) is
somewhat more complicated and the kinetics is consistent with an allosteric model, as the Hi+ saturation
curve for NHE1 is sigmoid (32). The sigmoidal relation of Hi+ saturation on NHE1 reflects the exchanger’s
ability to tightly regulate pHi; upon a small fall in pHi NHE1 is vigorously activated and upon a small rise
in pHi NHE1 function is abruptly lowered in accordance with the steep segment of the sigmoid curve. In
agreement with the allosteric model, the Hill coefficient of Hi+ for NHE1 is above 2, indicating at least
two independent intracellular H+ binding sites (33). Since the transport stoichiometry of NHE1 is 1:1, a
H+ transport binding site and a H+ modifier binding site has been proposed. Both sites should be located
in the N-terminal TM region of NHE1, but the H+ modifier binding site has been proposed to interact
with a regulatory domain in the C-terminal cytosolic tail of NHE1 (32,34,35). This interaction introduces
a putative pHi sensing mechanism and may allow for fine tuning of the set point at which NHE1 activates
(i.e. the steep part of the sigmoid curve of Hi+ saturation of NHE1). At physiological steady state
conditions NHE1 activity is low, just enough to counter H+ leak entry and metabolic H+ production, but in
response to acidic perturbations of pHi NHE1 is vigorously activated.
1.5 Activation and regulation of NHE1
At normal physiological cellular pH the NHE1 activity is low but is rapidly activated in response to a wide
variety of stimuli such as cellular acidification, cellular shrinkage and certain growth factors (18,30). As
previously mentioned the N-terminal TM region of NHE1 is suggested to contain a H+ modifier site that
mediates the allosteric property of NHE1 (32,34,35). Studies have identified residues in intracellular loop
5 (IL5, see Figure 1.8) and other domains, that when changed, shifts the pHi set point of the exchanger
(36,37)and there is evidence of an interaction between the H+ modifier site and the cytoplasmic C-
terminal tail of NHE1(29,38-40) suggested to be of an autoinhibitory manner and deletion drastically
alters the pHi sensitivity of NHE1 (29). The pH sensing mechanism of hNHE1 is not fully elucidated but
the recently obtained crystal structure of the bacterial homologous protein NhaA suggests the cytosolic
end of TMIX to be the pH sensor which induces a conformational change in response to a change in pHi
resulting in activation (41) (see Section 1.4). Whether this is the case for NHE1 remains to be elucidated.
Though extensively studied the exact method in which NHE1 is activated as a response to cellular
shrinkage is not yet established. A study mapped the volume-sensing site to be within the first 566
amino acids of NHE1 (42). Unless otherwise stated amino acid residue numbering is of the hNHE1
homolog. NHE1 activation by osmotic stress occurs without detectable changes in the net level of
phosphorylation of NHE1 (43,44).
Exchanger activity is regulated by an array of diverse classes of cell surface receptors, including receptor
tyrosine kinases, G protein-coupled receptors, and integrin receptors (14). The signals from these well
known signal networks converge to a limited number of NHE1-interacting proteins that regulate
modifications in the NHE1 C-terminal cytoplasmic regulatory domain (14,17). These modifications
regulate transport activity by changing the affinity of the intracellular H+ transport site and include
phosphorylation, binding of regulatory proteins, and conformational changes.
18
The distal region of the cytosolic C-terminal of NHE1 contains several serine and threonine residues that
are phosphorylated in response to sustained acidification, or to hormone and growth factor stimulation.
Phosphorylation in this region can shift the activity set point of the exchanger to activate at more
alkaline pH (14,17,30). The extracellular signal-related kinase, (ERK)effector, Ribosomal S kinase (p90rsk)
directly phosphorylates hNHE1 at Ser 703 (45). It is thought that many growth factors able to activate
NHE1, at least in part, transduce their signal through this mitogen-activated protein kinase (MAPK)
pathway. The phosphorylation of Ser703 of NHE1 enables the scaffolding protein 14-3-3 to bind to this
region of NHE1 which increases activity and may then recruit additional signaling molecules ((46) and
Figure 1.7). In addition to p90rsk, the Rho-associated kinase, p160rock (47,48), that mediates signaling
from the integrin receptors and receptors coupled to G13, and several other kinases are able to
stimulate NHE1 activity through phosphorylation of the C-terminal tail, although the phosphorylation
sites have not in all cases been directly identified. These include PKB/Akt, which phosphorylates hNHE1
at Ser648 (49), the Nck interacting kinase, NIK (50), that transduces signals from receptor tyrosine kinases
to NHE1 and the Ca/calmodulin dependent kinase II, CaMKII (51). NHE1 can also be phosphorylated by
p38 MAPK which inhibits NHE1 activity in response to angiotensin II (52). In addition NHE1 activity is also
regulated by dephosphorylation by protein phosphatase 1 (53,54).
Figure 1.7 – Regulation of NHE1. In this schematic representation from Orlowski & Grinstein, 2004 (17) the TM organization of NHE1 is shown with minor detail. The C-terminal tail contains the regulatory domain of NHE1 and regulatory interactions and other protein-protein interactions are shown here.
19
In addition to direct phosporylation NHE1 is also regulated by a wide array of directly interacting
molecules. Three calcium binding proteins interact with NHE1; calmodulin (CaM), calcineurin
homologous protein (CHP), and tescalin. The NHE1 cytosolic C-terminal is able to bind CaM at two sites;
a high affinity site – at amino acid residues (aa) 636-656, Kd=20nM, and a low affinity site – at aa657-
700, Kd=350nM (38,39,55). Unless otherwise noted amino acid residue numbering corresponds to the
hNHE1 paralog. The high affinity site is thought to interact with the H+ modifier site in an autoinhibitory
manner and binding of CaM to this site (or deletion of the entire site) abolishes this effect and thereby
stimulates NHE1 activity (39). Since all growth factors results in a transient but drastic increase in
cytosolic Ca2+ CaM may transduce, together with phosphorylation of Ser703 by p90rsk, NHE1 activation by
growth factors. CHP1 and CHP2 with two Ca2+ tightly bound associates with aa510-515 of NHE1 and this
interaction have been reported to both stimulate (56)and inhibit growth stimulated NHE1 activity (57-
59). Recently tescalin’s (CHP3) binding to the final 180aa of the NHE1 C-terminal have been reported to
promote optimal activity of the exchanger (60). Carbonic anhydrase II (CAII) binds to the C-terminal of
NHE1 (aa790-802) and increases NHE activity (17,61,62). These data support the notion that regulated
binding of CAII to NHE1 may stimulate transport activity and cellular alkalinization transiently by
increasing the local production of protons, which are extruded, while simultaneously elevating cellular
bicarbonate levels. Phosphorylation within aa634-789 of hNHE1 increases the interaction between CAII
and NHE1 (62). Two putative binding motifs for phosphatidylinositol 4,5-bisphosphate (PIP2) are found
in the C-terminal region of NHE1, at aa513-520 and aa556-564, and deletion of these results in
decreased transport activity in vivo (63). PIP2 binding is required for optimal NHE1 activity (63) and this
interaction is thought to be responsible for the requirement of physiological levels of ATP for optimal
NHE1 activity.
Finally, NHE1 interacts directly with the ERM-proteins (ezrin, radixin, and moesin), and through these
with the actin cytoskeleton of the cell (20,21,64). In this manner the distribution of NHE1 in some cells is
restricted to certain membrane areas, such as lammellipodia in fibroblasts. In fact, disruption of the
NHE1-ERM interaction impairs the formation of focal adhesion points in cultured cells, so in this manner
NHE1 acts as a structural anchor and plays a pivotal role in migration (20,21,64). The regulation of NHE1
is extensive and complicated and various modes of regulation most likely affect NHE1 at all times.
1.6 Physiology of NHE1
Changes in the growth or functional state of cells are often accompanied by, or initiated by, changes in
pHi or volume. It has been shown that in addition to, or via, regulating pHi and cellular volume, NHE1 is
involved in the control of diverse cellular processes such as growth, proliferation, differentiation,
migration and more (14). NHE1 is required for normal cell growth and proliferation, and in some cell
lines transcription of NHE1 is increased during differentiation and, if inhibited or abolished, cells exhibit
a deficiency in differentiation (23,65,66). Some of these effects are dependent on NHE1 ion
translocation (47,67) other effects are dependent on protein-protein interactions such as the NHE1-ERM
interaction (20). NHE1 transport is thought to have anti-apoptotic effects based on the fact that
intracellular acidification and cellular shrinkage are hallmarks of the apoptotic pathway and NHE1
transport counters both these events. Results confirm a role for NHE1 in cell survival (68).
20
1.7 Pathophysiology of NHE1
Altered NHE1 activity has been associated with several pathological states (12), but focus has been
placed on NHE1’s involvement in cell death in the myocardium as a result of myocardial infarction, and
NHE1’s involvement in the development of cancer and malignancy.
Briefly, ischemia of the myocardium results in increased anaerobic glycolysis, which leads to a large
increase in intracellular H+. This activates NHE1, with a rapid accumulation of intracellular Na+ as a
result. Normally the activity of the Na+/K+ ATPase would be up-regulated to counter this, but during
ischemia the ATPase may be inhibited by the decrease in intracellular ATP. As a consequence, the
increase in intracellular Na+ is thought to lead to a reversal of the Na+/Ca2+ exchanger and the resulting
rise in intracellular Ca2+ triggers various apoptotic pathways (69). Moreover, the swelling resulting from
the Na+ accumulation triggers necrotic death pathways, and cells die from a combination of the two
(70). Positive results in clinical animal experiments with NHE1 inhibitors were the basis for further
clinical trials (71), none of which have, however, proven fruitful (72).
NHE1 activity is of crucial importance in the development of at least some tumors, as seen by the fact
that tumor cells deficient of NHE1 show impaired growth when implanted into immune-deprived mice
(73). A key step of oncogenic transformation is the reversal of the pH-gradient of the cell, such that the
pHi is alkaline and the extracellular pH is acidic and recently it has been shown that excessive NHE1
activity causes this ‘malignant acidosis’ phenomenon(68,74). The microenvironment of, at least some,
tumors appear to be able to modulate NHE1 activity and location. The activation of NHE1 is due to an
increase in the affinity of the intracellular allosteric H+ modifier site for H+ (presumably resulting from
e.g. increased transporter phosphorylation downstream of e.g. MAPK activation) thereby shifting the set
point at which NHE1 activates resulting in a constitutive active exchanger. An example is the abnormal
activation of NHE1 in serum deprived breast cancer cells. This activation was shown to be mediated by
effects of RhoA and Rac1 specific to tumor cells, which showed an increase in motility and invasion (75).
A recent study showed that the microenvironment of breast cancer cells was able to activate NHE1
through CD44, and this activation resulted in extracellular acidification and further oncogenic
progression (75-77). The alkaline shift in pHi facilitates the shifts to glycolytic metabolism, resulting in an
increase in both [H+]i and [lactate] which in turn is extruded by the H+/lactate co-transporter, thus
leading to further alkaline intracellular and acidic extracellular pH values. The consequence of the high
pHi is an increase in tumor cell number and density, due to increased DNA synthesis, cell cycle
progression and substrate- and serum-independent growth (68). The increased tumor cell density
causes reduced access to the circulatory system, leading to a hypoxic environment with low levels of
nutrients (68). Excessive NHE1 activity also has a major role in coordinating the invasive properties of
tumor cells: increased cell motility and disruption of cell-matrix interactions as a consequence of
protease activity and the increased acid secretion. Invasive cancer cells are characterized by an elevated
NHE1 expression and, as for normal cells (67,78), a location of the exchanger in the leading edge is
necessary for maintaining the polarity of the migration (79,80). In the leading edge, pseudopodia project
in the direction of circulatory capillaries, a process that requires dynamic remodeling of the
cytoskeleton, and invadopodia extend into and degrade the extracellular matrix, by secreting proteases,
that are activated by the acidic environment created by NHE1-mediated H+i extrusion (68,79,80).
21
Development of drugs that either indirectly target NHE1 through signal transduction or specifically
target NHE1 and thereby inhibit the formation of pseudopodia and the function of invadopodia could
thus be a possible way to control cell invasiveness.
1.8 Inhibitors of NHE1
Residues in TMIV and TMIX have been found to be implicated in NHE1s interaction with its most
common inhibitors; amiloride-, and benzoylguanidine (HOE-) type compounds. Given the above-
mentioned pathological implications of increased NHE1 activity, researchers are very interested to learn
the mechanism with which these inhibitors affect NHE1. Kinetic studies have shown that amiloride
inhibit NHE1 in a competitive manner, hence any information about the inhibitor binding site would be
indirect information of the substrate binding site. For this reason Stine Falsig Pedersen (SFP) and Peter
M. Cala (PMC) undertook the task of elucidating the inhibitor binding site in an elegant so called knock
in study utilizing the NHE1 homologs: in 1999, atNHE1 was cloned from Amphiuma tridactylum
erythrocytes (81), and in 2003, SFP cloned paNHE1 from Pleuronectes americanus erythrocytes in
collaboration with PMC (16). Both proteins show high overall sequence identity with hNHE1 and similar
topology to hNHE1 based on hydropathy analysis. Although similar in structure and function the two
proteins show a remarkable difference in inhibitor binding profile. Whereas hNHE1, and all other known
NHE1 homologs, are sensitive to amiloride and amiloride-derivatives (like ethylisopropylamiloride, EIPA)
and benzoylguanidine-like compounds (e.g. HOE694) atNHE1 is sensitive to HOE694 but insensitive to
EIPA, and paNHE1 is insensitive to both EIPA and HOE694 (82). These knock-in studies identified a two
amino acid motif in the TM4 of paNHE1 that, when replaced with the corresponding residues in hNHE1
(paHE1 LFFFY to hNHE1 VFFLF) restored sensitivity to EIPA (82). These studies confirm that TMIV plays a
central role in inhibitor binding and the study further demonstrated a role for the TMX/TMXI region
and/or ILV/ELVI in inhibitor binding. Because of the competitive manner in which amiloride and HOE694
inhibit NHE1 these results also indicate TMIV to be important for ion translocation by NHE1 along with
the TMX/TMXI and/or ILV/ELVI (82).
1.9 Structure of NHE1
NHE1 is the best characterized of the NHE isoforms but little is known about the three dimensional
structure of NHE1 because of the inherent difficulties in crystallizing membrane proteins. This is an area
in need of development, as the proteome project assigns 30% of the entire proteome to be integral
membrane proteins and around 60% of pharmaceutical drug targets in human and plants are membrane
proteins (83). Presently over 40.000 high resolution structures of soluble proteins have been solved
whereas only around 150 high resolution structures are solved for integral membrane proteins. Close to
90% of the protein structures (62388 as of 122909) in the Protein Data Base
(www.rcsb.org/pdb/home/home.do) are per December 2009 resolved by X-ray crystallography (53784)
and the majority of the rest are resolved by NMR spectroscopy (8165). The major hindrance in studying
the 3D structure of membrane proteins is to obtain the purified membrane protein retaining its native
conformation and in sufficient amounts of sufficient purity. To purify membrane proteins, one must
22
extract the protein from its natural environment – the plasma membrane – using detergents, and in
doing so one might disrupt critical structural characteristics. The techniques mentioned above records
information about the entire sample analyzed and therefore are not ideal for detergent-solubilized
membrane proteins. Therefore, researchers have been forced to utilize other approaches to analyze the
structure of NHE1. Several molecular biology techniques and bioinformatics/computational biology
methods have been employed, as we shall see next.
The first structural insights of NHE1 were gained from hydropathy analysis using the Kyte-Doolittle
algorithm. The TM region of NHE1 was predicted to span the plasma membrane twelve times with both
the N- and C-terminal located on the cytosolic side of the membrane. Molecular biology studies such as
availability to protease cleavage (84) and the elegant research by Wakabayashi et al. in 2000 using a
substituted-cysteine-accessibility assay (85)to reveal which amino acid residues are available to the
extracellular- and intracellular- solution respectively were employed. These experiments confirmed the
predicted intracellular location of N- and C-terminal and revealed the presence of two small and one
large reentrant loops that were at least partly embedded in the membrane. The small loops are located
intracellular between TMIV-V (IL2) and TMVIII-IX (IL4) and the large loop is located on the extracellular
surface between TMIX-X (EL5) (Figure 1.8).
Figure 1.8 – Topology model of NHE1. By accessibility of 83 reintroduced cysteine residues to the cysteine-directed reagents biotin maleimide and MTSET Wakabayashi et al. 2000 (85) proposed this topological arrangement. The plasma membrane is almost impermeable to biotin maleimide and reaction with a cysteine residue covalently binds biotin to the residue. The cys-biotin is then detected by streptavidin-biotin chemistry. MTSET is membrane-impermeable and used to block biotinylation.
Red circles - Accessible to external SH-directed reagents. Blue circles - Accessible to cytosolic SH-directed reagents. Violet circles - Not readily accessible to biotin maleimide from either side of the membrane. Black circles - Not analyzed because of low level expression of the functional exchanger. Rectangles - Positions for native cysteines replaced with alanine. Blue rectangle - A native cysteine (Cys8) accessible to cytosolic SH-directed reagents.
23
Amino acid Location Mutation Effect of mutation Reference Gly
148 (rat: Gly
152) EL2 Gly→Ala Increase in Ki for EIPA (86)
Pro153
/Pro154
(rat:Pro157
/Pro158
) EL2 Pro→Ser/Pro→Phe Increase in Ki for EIPA (86) Decreased Na
+/H
+ transport
Phe161
(hamster: Phe165
) TMIV Phe→Tyr Increase in Ki for amiloride (87) Decreased Na
+/H
+ transport
Phe161
TMIV Phe→Cys Line ion transport pore (88) Phe
162 TMIV Phe→Ser Increase in Ki for cariporide (89)
Leu163
(hamster: Leu167
) TMIV Leu→Phe, Ala, Arg, Trp Increase in Ki for amiloride, HOE694 (87) Leu→Tyr Eliminates Na
+/H
+ transport
Pro167
TMIV Pro→Gly, Ala, Cys Decreased Na+/H
+ transport and expression (88)
Pro168
TMIV Pro→Gly, Ala, Cys Decreased Na+/H
+ transport (88)
Gly174
TMIV Gly→Ser, Asp Increase in Ki for amiloride, HOE694 (90) Leu
163/Gly
174 TMIV Leu→Phe/Gly→Ser Increase in Ki for HOE694 (90)
Increase in Km for Na+
Arg180
IL2 Arg→Cys MTSET decreases activity (85) Gln
181 IL2 Gln→Cys MTSET decreases activity (85)
Glu262
TMVII Glu→Gln Eliminates Na+/H
+ transport (91)
Glu→Asp Restores partial Na+/H
+ transport
Asp267
TMVII Asp→Asn Eliminates Na+/H
+ transport (91)
Asp→Glu Restores Na+/H
+ transport
His349
TMIX His→Gly, Leu Increase in Ki for amiloride (92) Glu
346 (rat: Glu
350) TMIX Glu→Asp Increase in Ki for EIPA, HO694 (86,93)
Decreased Na+/H
+ transport
Increase in Km for Na+
Gly352
(rat: Gly356
) TMIX Gly, Ala, Ser, Asp Increase in Ki for EIPA (86) Decreased Na
+/H
+ transport
Glu391
EL5 Glu→Gln Decreased Na+/H
+ transport (91)
Glu→Asp Restores Na+/H
+ transport
Arg440
IL5 Arg→Cys, Lys, His, Asp, Acidic shift of pHi set point (36) Glu, Leu Tyr
454 TMXI Tyr→Cys Retained in ER (94)
Gly455
TMXI Gly→Cys, Gln, Thr, Val Alkaline shift of pHi set point (36) Gly
456 TMXI Gly→Cys Alkaline shift of pHi set point (36)
Arg458
TMXI Arg→Cys Retained in ER (94)
Several residues in the IL2 and IL4 were found to be accessible from the extracellular side, indicating the
possibility that they take part in a pore forming structure. These putative reentrant loops are highly
conserved among several NHE1 homologs consistent with the notion that they are important for NHE1
function. Mutational analyses has identified amino acid residues within the TM segment of NHE1
involved in ion transport, inhibitor binding, pH sensitivity, and residues essential for the correct folding
and targeting of NHE1. A brief summary this experimental data is shown in Table 1.3. As noted above,
the mammalian NHE1 contains consensus sites for N- and O-linked glycosylation in EL1 (31).
Several attempts by the Fliegel group to perform NMR spectroscopy on individual TM segments have
resulted in publications (95-97)and most recently the NMR structure of TMXI (95). There are several
concerns about these structures the most important being whether the isolated oligopeptides in
question really can be expected to inhabit structures representative of their structures in the native
Table 1.3 – Mutational analysis of residues in the TM region of NHE1. This summary of residues that have an effect on the ion
transport, inhibitor binding, or the expression and targeting of the exchanger is modified from Slepkov et al. from 2007 (30).
Numbering of residues corresponds to the human NHE1 and the location corresponds to the topology shown in figure 1.8.
24
NHE1. In 2005 the full length 3D structure of the bacterial Escherichia coli Na+/H+ antiporter NhaA was
resolved to 3.45Å by X-ray crystallography (41). This gave researchers some structural clues of structure-
function relation of Na+/H+ exchange, described in more detail below.
1.10 Structure of NhaA
The bacterial NHE1 homolog NhaA from E.coli is an evolutionary distant member of the CPA family. Like
the mammalian NHE1, the NhaA is essential for the crucial processes of Na+, pH and volume
homeostasis in E. coli. NhaA differs from NHE1 in a few main functional characteristics: NhaA transport
is electrogenic and serves to extrude 1 Na+ in exchange for 2 H+ (the main problem of E. coli being
salinity rather than acidity) and for this reason, NhaA is activated at alkaline pH.
In 2005 Hunte et al. succeeded in solving a 3.45 Å crystal structure of the pH locked (pH=4.0) state of
NhaA (41). This major breakthrough allowed researchers to perform more informed structural and
functional analysis of NHE1 on the basis of the crystal structure of NhaA. Despite hNHE1 and NhaA
having less than 10% sequence identity the proteins has remarkably similar structural features owing to
the fact that structure is better conserved than sequence.
Like the predicted structure of hNHE1 the crystal structure (PDB ID: 1ZCD) of the 388 amino acid residue
NhaA consists of 12 TM domains with N- and C-terminal facing the cytosolic side. The loops on the
periplasmic side are structured, with at least a double-stranded beta sheet, and close to the surface of
the membrane causing the organization of the TM region to be flat and rigid on the periplasmic side. In
contrast, the cytoplasmic side of the TM region is bulky with several helices protruding into the cytosol
and the loops are thought to be more flexible. One of the most interesting features of the crystal
structure is the presence of periplasmic and cytoplasmic funnels pointing towards each other but
separated by a hydrophobic barrier. Several lines of experimental evidence, as outlined below, strongly
suggest that this is the ion translocation funnel and calculations based on the crystal structure data
predict that the narrowest point in each funnel can harbor hydrated Na+ or Li+ but not other monovalent
cations (41).
Before I go into further details revealed by this crystal structure I will briefly evaluate the validity of said
structure. Obviously, it is of critical importance that both the crystallization data are of satisfactory
quality and for the solved structure to be of a physiologically correct conformation of NhaA. The
diffraction data and refinement data are shown in Table 1.4 and Table 1.5 respectively.
In X-ray crystallography, the highly ordered arrangement of molecules in a crystal lattice gives rise to a
diffraction pattern based on the scattering of X-rays by the electrons of the investigated sample (98). In
the diffraction pattern the shape and symmetry of the unit cell – the smallest repeating unit that can
generate the crystal with only translation operations – define the directions of the diffracted beams, and
the locations of all atoms in the unit cell define their intensities. Because the wavelength of X-rays (≈1 Å)
is similar to atomic bond lengths it is theoretically possible to obtain structures of atomic resolution.
Each reflection of the diffraction pattern is defined by its amplitude (measurable from the intensity of
25
the reflection), wavelength (defined by the source of the X-ray, i.e. synchrotron), and its phase for which
no direct information is gathered in the diffraction experiment (98,99). The phase problem can be
solved in several ways. One is to incorporate heavy atoms such as selenium in the protein, as the
methionine analog seleno-methionine, and perform an anomalous dispersion experiment which yields
the positions of the selenium atoms and the initial phases can be calculated (98,99). Hunte et al. solved
the phase by Single-Wavelength Anomalous Dispersion (SAD) of the seleno-methionine-labeled crystal
(41). The SAD phases were then applied to one of two data sets on native NhaA and the final refinement
utilized the second native data set.
Data Collection Native set 2 Native set 1 SeMet SAD Space group P212121 P212121 P212121 Cell dimensions a,b,c (Å) 108.9, 121.7, 123.6 108.8, 121.8, 124.7 109.9, 121.5, 124.1 Wavelenght 0.9686 0.9794 0.9717 Resolution (Å) 20-3.45 (3.57) 15-3.8 (3.93) 17-4.3 (4.45) Rmerge 5.0 (29.4) 6.1 (41.3) 8.3 (43.4) I/σI 30.5 (2.1) 17.8 (2.1) 23.1 (3.8) Completeness 91.7 (64.1) 98.1 (98.8) 99.7 (100) Redundancy 18.4 12.0 21.5
The quality of a crystal structure is dependent on the quality of the diffraction data and the quality of
the subsequent refinement. In assessing the quality of the diffraction data I use the level of redundancy,
the Rmerge values, the completeness, and the signal-to-noise ratio (I/σI) to characterize the data set (see
Table 1.4). In a diffraction experiment the intensities of many reflections are measured and because of
the symmetry of the crystal some of these are equivalent and the redundancy is a measure of the
average number of measurements of each unique reflection. Higher redundancy is better. Rmerge is a
measure of the accuracy of the averaged intensities of the diffraction pattern and should generally not
exceed 10% (98). The completeness is the percentage of theoretically possible unique reflections
covered by the experimental data and is important because the intensity of every possible reflection is
needed for correct calculation, by Fourier transformation, of the electron density map (98). I find the R
merge of both native data sets (6.1% and 5.0% see Table 1.4) are within acceptable range. The redundancy
is also within acceptable range. The completeness of Native set 1 is very high, also in the high resolution
shell, whereas the completeness of Native set 2 is lower, especially in the high resolution shell. As Native
set 2 is used for the final refinement this is not cause for much concern. The signal-to-noise ratio (I/σI)of
the second native data set is better than that of the first but both well within acceptable range, and in
the high resolution shell we see the I/σI reaching the diffraction limit (2.1 for both Native set 1 and 2)
(98).
Table 1.4 – Diffraction data of the crystal structure of NhaA (PBD ID: 1ZCD). Details of the data collection from Hunte et al. 2005 (41) was taken from the supplemental material available from a) and the pdb file from b). Numbers in paranthesis are from the highest resolution shell.
a) http://www.nature.com/nature/journal/v435/n7046/suppinfo/nature03692.html b) www.rcsb.org/pdb/files/1ZCD.pdb.
26
Refinement data Native set 2 Resolution (Å) 15-3.45 No. reflections 19,993 Rwork/ Rfree 30.1/31.6 No. atoms : Protein 5,618 B-factors: Protein 121 R.m.s deviations: Bond lengths (Å) 0.01 Bond angles (º) 1.57
The quality of the diffraction data (Table 1.4) of Hunte et al. is in accordance with the quality and
resolution of the final structure. The refinement data represents how well the final structure fit the
electron density map based on the diffraction data. Rwork is a measure of well the model fits
experimental data and for high resolution structures a Rwork around 20% is expected (98). Rfree is
calculated analogous to Rwork but on a set of randomly selected reflections which have not been used in
the refinement. Rwork- Rfree represents ‘over-fitting’ of the experimental data. The RMSD of bond lenghts
and angles represents how well the model structure fit the typical bond lengths and angles. The
refinement data of the NhaA crystal structure is shown in Table 1.5. The Rwork value is high but correlates
well with the low resolution, 3.45 Å, of the final structure. There is little difference between Rwork and
Rfree indicative of no ‘over-fitting’, i.e. not getting more structural information out of the experimental
data than there actually is (98). The RMSD values of both bond lengths and angles fall within acceptable
range and the Ramachandran plot (see Appendix 12) show 95% of the φ/ψ torsion angles are within the
most favored region (68.8%) and additional allowed region (26.2%). I conclude the crystal structure of
NhaA (PDB ID: 1ZCD) to be a low resolution (3.45Å) of good quality.
The solved crystal structure is a model structure of the pH-locked NhaA. NhaA is negatively regulated by
acidic pH and at pH=4 there is no activity of the antiporter. This poses the question of whether the
solved structure is physiologically relevant. In order to test the physiological relevance the researchers
superimposed electron density data collected with 2D cryo-electron microscopy (cryo-EM) onto the
solved crystal structure (100-102). The cryo-EM data were collected from 2D crystals of the membrane-
embedded E. coli NhaA and therefore thought to represent a native conformation. The 3D crystal
structure and the 2D electron density map from the cryo-EM superimpose well and this testifies to the
validity of the crystal structure. On this basis it is suggested that many of the 12 helices do not move as a
result of pH change. It should be noted that similar to what has been found for NHE1 a monomer NhaA
is fully functional but is usually found as a dimer, as evidenced by EPR studies (28).
Table 1.5 – Refinement data of the NhaA crystal structure (PDB ID: 1ZCD). Details of the refinement data from Hunte et al. 2005 (41) was taken from the supplemental material available from a). 2.9% of the data set was excluded from the refinement to calculate Rfree.
a)http://www.nature.com/nature/journal/v435/n7046/suppinfo/nature03692.html
27
The solved high resolution crystal structure of NhaA reveals many interesting structural characteristics.
Among the most exiting is the novel fold centered on the TMIV/TMXI assembly (41,103). These helices
enter the lipid bilayer of the membrane from opposite sides (TMIV from the periplasmic side and TMXI
from the cytoplamic side) and both helices are interrupted by extended chains at the center of the
membrane (Figure 1.9). This assembly is remarkable because of the delicate electrostatic balance that
governs the structural phenomenon. The partial-positive charge of the N-termini of the two opposing
interrupted helices is compensated and stabilized by D133 of the extended chain of TMIV. The partial-
negative charge of the C-termini of the opposing (one from TMIV and one from TMXI) helices is
compensated and stabilized by K300 of TMX. The phenomenon of discontinuous helices is seen in other
transport proteins and a general theme of these novel structural folds is their tight link to the function of
Figure 1.9 – Crystal structure of NhaA. The periplasmic and cytoplasmic funnels and the putative catalytic site of NhaA is analyzed from the 3D structure. a) The TMs lining the ion translocation funnel are colored in accordance with the accessibility of Na
+ (and Li
+) as
assessed by computational methods (see for further detail). In this pH-locked conformation fully hydrated Na+ (and Li
+) is able to enter
the purple area but not the red. The gray segment is the ion passage barrier. The putative Na+ binding site centered on the conserved
D164 is shown. b) The refined model superimposed onto the electron density map of the putative Na+ binding site is shown here. The
carboxyl groups of residues D164 and D163 along with the hydroxyl group of T132 is thought to bind the positive charge of Na+. c,d,e)
Electrostatic potential surface map of the antiporter colored in accordance with charge: positive=blue and negative=red. The cytoplasmic view of the pH-locked NhaA is shown in c) and the periplasmic view in d). The dotted line indicates the cross-section view shown in d). The cytoplasmic (top) and the periplasmic (bottom) funnels are shown in d) along with D164, the putative Na
+-binding
site. From Hunte et al. 2005 (41).
28
these membrane proteins (104). It is inherent that any change in this electrostatic balance, such as the
binding of Na+, will induce a conformational change and this may be the method by which NhaA
transports Na+ ions. The substrate binding site is not seen directly in the crystal structure because of the
lack of Na+ in the crystallization medium. The pH-locked substrate-free state of NhaA has a more
compact and rigid structure than the active state as confirmed by recent studies (105). Based on
previously solved Na+-binding sites the Padan group researchers propose the Na+-binding site to be
centered on the evolutionary well conserved D164, along with D163, D133 and T132 (41,101). D163 and
D164 have previously been shown to be essential for NhaA function and mutation of D133 and T132
dramatically increases the apparent Km of the antiporter (103,106).
Previous genetic, biochemical and biophysical studies have identified many amino acids essential for
both the pH response and ion transport of NhaA. When identifying these amino acids in the 3D crystal
structure it became clear that amino acids involved in pH sensing are located at and around the
cytoplasmic orifice. A cluster of negatively charged residues capable of attracting cations and
functioning as an ionic trap and, at the same time, serve as the pH sensor is centered at the cytoplasmic
end of TMIX (41,101).
In an attempt to identify amino acid residues involved in ion translocation several mutational setups
were used: site-directed mutagenesis of residues with the chemical capacity to attract, bind or repel
cations (103,106); mutagenesis of residues that in other proteins have been shown to bind Na+ (106);
and random mutagenesis of plasmidic nhaA and selected mutants that change the specificity or affinity
to the ions (103,106). Remarkably, the mutations that affect the translocation cluster at the TM IV/XI
assembly, yet again validating the proposed translocation site of Na+. The distance from the proposed
pH-sensor to the substrate translocation site at the TMIV/TMXI extended chain assembly is calculated to
15 Å on the basis of the crystal structure(101). Furthermore, the two structural features are in direct
contact through amino acid residues F267 and F344 (101).
These are all observations and calculations made upon the rigid structure of the pH-locked NhaA and a
transport protein attain several conformations during the process of ion translocation. To study the
dynamics and conformational changes brought on by ion translocation in NhaA researchers employed
several in situ and in silico methods. Olkhova et al. used the multiconformation continuum electrostatics
(MCCE) method to study, in silico, the effect of pH values on the protonation state of ionisable residues
in NhaA (107).
29
The results revealed four clusters that are characterized by strong electrostatic interaction. The clusters
are spread along the cross axis of the membrane: two clusters are found at the cytoplasmic orifice, one
centered on the N-terminal of TMIX and the other is found on the opposite side of the cytoplasmic
funnel; one cluster is formed mainly by residues within the ion binding site and is hence centered on
D164 in the middle of the membrane; the last cluster is found at the rim of the periplasmic funnel.
Residue H256 was found to electrostatically connect the first three clusters and D133 was found to
connect the first and third cluster. Olkhova et al. suggest these electrostatic connections to be
fundamental for pH signal transduction across the membrane and NhaA activation (Figure 1.10 and
(107-109)). Combining MCCE with molecular dynamics (MD) simulations has revealed that NhaA must
undergo a conformational change in order to transport ions. Interestingly, some of these predictions
have been confirmed experimentally. The pH-induced movement of TMX toward TMII was confirmed via
crosslinking an inactive double cysteine mutation (T132C of TMS IV and G338C of TMS XI) restored NhaA
function (110). This last experiment confirms the importance of the TMIV/TMXI assembly for NhaA
function.
Figure 1.10 – Clusters of strongly electrostatically interacting ionizable groups for the NhaA antiporter. Olkhova et al. utilized the multiconformation continuum electrostatics method to evaluate the electrostatic interaction of NhaA residues. Four clusters of interacting residues were found, shown here in magenta, grey, yellow and green. The residues of the clusters are responsible for the electrostatic properties of NhaA and for the structural changes upon activation by raising the pH. Two clusters are located at the cytoplasmic side of the molecule, one at the periplasmic side and one in the center. The cluster shown in magenta is proposed to be the pH sensor that upon protonation undergoes a conformational change. This structural change along with long-range electrostatic interactions between the pH sensor and the substrate binding site, the yellow cluster, is proposed to expose the substrate binding site. From Olkhova et al. 2006 (107).
30
Figure 1.11 – Proposed mechanism of NhaA. In this schematic presentation the pH-induced conformational changes (A and B) and the conformational changes brought about by substrate binding and translocation (B and C) is shown. A) The pH-locked conformation of the crystal structure where ion translocation is blocked by the periplasmic barrier (orange bar) and only D164 of the putative Na
+-binding site is exposed to the cytoplasmic funnel (black circle and dotted lines). A pH-induced
conformational change of the pH sensor at the cytoplasmic end of TMIX (orange circle) results in a reorientation of helices IVp, XIp and X as seen in B). Here the substrate binding site is fully accessible to the cytoplasmic funnel. C) Binding of Na
+
results in a conformational change exposing the active site to the periplasmic side where Na+ is released and binding of H
+
to D164 and D163 brings the antiporter back to the active conformation in B). From Padan et al. 2009 (101).
Figure 1.12 – Proposed transport mechanism of NhaA. Based on Molecular Dynamics simulations Arkin et al. proposed this mechanism of ion translocation of NhaA. It consists of 4 sequential steps: A) D163, in this study called the accessibility control site, is deprotonated resulting in the Na
+ -binding site being exposed to the cytoplasmic side. D164 then releases a H
+ and bind a Na
+ from
the cytoplasm. B) D163 is then protonated from the periplasmic side and the resulting conformational change exposes the Na+-
binding site to the periplasm. C) D164 releases the bound Na+ and binds a H
+ from the periplasm. D) D163 releases the H
+ (from the
periplasm) to the cytosol resulting in a conformational change that exposes the Na+-binding site to the cytoplasm. In total two H
+
have been translocated from the periplasm to the cytoplasm (A and D) and one Na+ have been translocated from the cytoplasm to
the periplasm in accordance with the stoichiometry reported for NhaA. From Arkin et al. 2007 (111).
31
The molecular method by which NhaA transports Na+ and H+ has been proposed to be one of alternating
access and several detailed proposed models (Figure 1.11 and Figure 1.12) have arisen on the basis of
the 3D structure of the pH-locked NhaA (41,101,111). As mentioned above the TMIV/TMXI assembly is,
not surprisingly, the center of this alternating access mechanism and it is thought that the active
antiporter alternates between two major conformations; the substrate-binding site facing either the
periplasmic side or the cytoplasmic side. The inter-conversion between the two conformations is
brought about by substrate binding and the resulting displacement of the electrostatic balance, which
will result in a conformational change (41,101,111). The proposed conformational change between the
two conformations is small enough to be in congruence with the immense catalytic speed of the NhaA
(100,000min–1) (112). Although there is some variations in the proposed models of ion translocation the
TMIV/TMXI assembly is the key feature along with the putative substrate binding site centered on the
conserved D164 residue (along with D163). The pH sensing mechanism is proposed by both Hunter et al.
(41)and later Padan et al. (101) to be located at the cytoplasmic end of TMIX where E252 has been
shown to be of critical importance in the pH sensitivity of NhaA (113).
Although a lot of valuable information is derived from the crystal structure, some critical information is
yet to be discovered. A confirmation of Na+ binding site by a crystal structure of the active NhaA over a
range of pH is needed. A very recent cryo-electron microscopy study of 2D NhaA crystals grown at pH=4
and incubated at higher pH values relevant for NhaA function suggest a pH-induced conformational
change (from the pH-locked state to a pH value where NhaA is active) precedes a substrate-binding
induced conformational change that involve displacement of TMIV (114). In absence of the gold
standard – a crystal structure resolved at near atomic resolution – researchers have utilized the
structure of NhaA to create a model structure of the TM region of hNHE1.
1.11 Model structure of hNHE1
As mentioned in the beginning of the previous section NhaA and NHE1 are evolutionarily distant
members of the same family and share around 10% sequence identity. Both proteins serves to transport
Na+ and H+ and this shared function is affirmed by several structural similarities. The fact that structural
folds tends to be better conserved than sequence made Landau and coworkers attempt to model the
TM part of NHE1 to the 3D structure of NhaA. The process was complicated by the low sequence
homology and the resulting model structure is calculated from a pairwise alignment manually pieced
together from alignments made with three different algorithms (115). But, as is seen in the resulting
model, many of the structural features thought to be important for ion translocation were recapitulated
in the model. The assembly of the helices TMIV/TMXI with their extended chains in the middle of the
membrane, as seen in the NhaA structure, was of course also found in the model structure of NHE1,
with conserved residues matching some of the hypothesized actions above. However, the assignment of
the TM domains is inconsistent with earlier experimental evidence and no experimental validation of
inter-helix distances are in the publication (115). Therefore the researchers in the SFP and PMC
collaboration undertook the task of constructing a structural model of hNHE1 threaded onto the 3D
structure of NhaA based on a pairwise alignment that was constrained to assign residues in TM domains
that were experimentally proven (Figure 1.13)(84,85).
32
From the fold recognition algorithms at MetaServer we found that NHE1 was predicted to be most
similar in structure to NhaA (116). But because of the low sequence identity between the two proteins
the final alignment had to be modified manually. First of all, the threading of NHE1 on the NhaA
template was limited to the N-terminal domain of the protein (residues 1 to 507), the portion of NHE1
that contains the membrane-spanning domains. NHE1 is larger than NhaA and this difference is mainly
constituted by hydrophilic residues that fall within the extracellular and intracellular loops of the TM
region. Therefore several of these loops are excluded in the alignment as seen in Figure 1.13, and of
course in the resulting model structure. The final alignment had been constrained to the residues that
were experimentally proven to reside in the membrane (85).
Alignments between the known NhaA TM regions and the hNHE1 TM regions (including the flanking 10
residues) suggested by Wakabayashi et al. based on cysteine accessibility analyses were then carried
out independently using the ClustalW algorithm. The resultant TM alignments were then used to match
the regions of low homology and ensure that gaps fell within the hydrophilic loops connecting the TM
segments. In the final model structure amino acid residues Pro12 to Ala507 excluding the above
mentioned extramembraneous loops are modeled (Figure 1.14 and Figure 1.15). As seen in the space-
Figure 1.13 – Alignment of NHE1 and NhaA. From MetaServer (bioinfo.pl/meta) NHE1 was predicted to be similar in structure to NhaA
and the resulting alignment/similarity was evalutated by 3D Jury and manually adjusted before threading. See text for more detail.
Bioinformatic research was done by J. Lagerstedt; Unpublished by Nygaard et al. (see Appendix 1).
33
filling representation several charged and polar residues are found near the TMIV/TMXI assembly with
possible roles in ion translocation (Figure 1.14). For instance the conserved residue R425 (corresponding
to Lys300 of NhaA) of TMX is found in close proximity to this assembly and we hypothesize that R425
functions to screen the partial positive charge of the N-terminals of the broken helices TMIV and TMXI
found in the middle of the membrane (Figure 1.15). This was confirmed experimentally by mutation of
R425A, which was found to disrupt Na+ /H+ exchange by NHE1, at least in part by reducing trafficking of
the transporter to the plasma membrane, suggesting structural disruption. Thus, these are in
accordance with the notion that the electrostatic architecture of the TMIV/TMXI is of vital importance
for NHE1 function (unpublished results, see Appendix 1).
Figure 1.14 –Model structure of hNHE1. Shown here are solid surface structure representations of TMs III, IV, V, X, XI and XII of the
hNHE1 model structure based on the crystal structure of NhaA. Blue=positively charged residues; Red=negatively charged residues;
Green=polar residues. A) In the cytoplasmic views the arrows indicate the cytoplasmic funnel and in the slightly tilted view (15˚) the
positive charge of R425 can be seen (brown arrow). Several charged and polar residues that may be involved in the ion-
translocation are located near this cavity. These include Arg458
and Arg500
(positively charged), Glu131
(negatively charged) and,
Ser132
, Thr433
, Asn437
and Tyr454
(polar). B) In the extracellular view no apparent funnel is seen but a cluster of charged residues,
Asp470
, Lys471
, Lys472
and His473
, at the end of TMXI is seen. One must remember that this is a model forced to fit the 3D structure of
a distant relative and as such it is a great vantage point for further research but nothing more. (Unpublished Nygaard et al. see
appendix 1)
34
Even though the model structure of hNHE1 looks reasonable we must remember that it is only a model,
and a model that we have forced to fit to a given 3D structure. So similarities are a matter of course and
without any experimental data confirming inter-atomic distances and their changes upon manipulations
affecting NHE1, the model is nothing more than a qualified guess. Therefore we set out to determine
inter-atomic distances that could assess the model structure. Recently the SFP and PMC collaboration
were successful in employing a powerful approach of site-directed spin labeling (SDSL) and electron
paramagnetic resonance (EPR) to evaluate intermolecular distance. In this technique, cysteine residues
introduced into the protein at relevant positions, enables introduction of cysteine-directed spin-labels
(117,118). The EPR spectra provide information on side chain dynamics, and thus on protein
topography, conformational changes, secondary and tertiary structure (see Section 1.12) (118-121).
Introduction of a second paramagnetic center allows distance measurements within the protein
(122,123). Demonstrating the feasibility of this approach for assessing NHE1 structure, a recently
published study employed EPR spectroscopic distance measurements between spin labeled side chains
on two NhaA monomers to confirm NhaA dimerization (28). This model has proved a great foundation
for structural work on the NHE1, and together with the work done at the SFP and Cala laboratory this
has been the basis for my project.
Figure 1.15 – The TMIV/TMXI assembly of the hNHE1 model structure. A) Ribbon representation of the TMIV/TMXI assembly proposed by us to be the catalytic core of Na
+/H
+ exchange by hNHE1, residues thought to be involved in ion
translocation is shown. Schematic representation of the TMIV/TMXI assembly of the hNHE1 model structure (B) and the NhaA crystal structure (C). In comparison several residues important for function in NhaA aligns with similar residues in the hNHE1 model: Asp
133 of NhaA aligns with Asp
172 of hNHE1, Asp
163 of NhaA aligns with Thr
197 of hNHE1, and Lys
300 of NhaA
aligns with that of Arg425
of hNHE1 in our model. (Unpublished Nygaard et al. see appendix 1)
35
In this comparative approach we have utilized the model structure of hNHE1 to pose scientific questions
regarding the structure-function relations of hNHE1. Most important of these questions; what is the
molecular mechanism with which NHE1 exchanges ions? Not only would new knowledge on this
question lead to pharmaceutical advances (see section 1.8) but would also be of critical importance in
the basic science field of transporter structure, which is still in its nascent stages. On the basis of the
model structure of hNHE1 we proposed that the novel fold of two helices (TMs IV and XI) partly
unwound in the center of the plasma membrane is the structural basis for NHE1 function. The complex
has an unlikely architecture, conserved from bacteria to man (1), that places dipoles and charges in the
low dielectric membrane center and is therefore crucial for function (104).
1.12 Electron Paramagnetic Resonance
Electron paramagnetic resonance (EPR), also referred to as electron spin resonance, is a powerful
analytical technique used in the determination of secondary, tertiary and quaternary protein structure
and to study protein folding, orientation, dynamics and conformational changes. EPR is based on the
absorption of electromagnetic radiation, usually in the microwave frequency region, by a paramagnetic
sample placed in a magnetic field (117).
The basis for all EPR approaches is the electron (117,124,125). Spin is an intrinsic property of all
electrons. The spin of an electron causes the electron to behave as a magnetic dipole. The magnetic
dipole moment, µ, of electron interacts with an external magnetic field, H, resulting in the so-called
Zeeman splitting. Equation 1) describes the energy (E) of a magnetic dipole (µ) in a static magnetic field
with strength H.
Equation 1)
As mentioned before the magnetic moment of an electron is generated by its spin (S) as shown in
equation 2) where β is the Bohr magneton (intrinsic unit of the electron magnetic moment) and g is the
spectroscopic splitting factor (relates contribution of spin and orbital motion of the electron to its total
angular momentum) (117).
Equation 2)
The magnetic dipole moment of an electron is usually described as a vector quantity, i.e. it has a
magnitude and direction. When placed in an external magnetic field the magnetic dipole of the unpaired
electron aligns either parallel or antiparallel to the external field. These orientations represent two
distinct energy levels, with the difference in energy being proportional to the strength of the magnetic
field. For an electron the energy levels are quantized and can only take up the lower energy parallel
orientation ( -½) with respect to the external magnetic field or the higher energy antiparallel (+½) state.
From equations 1) and 2) the energy levels are E=±(½)gβH, and we see that the energy difference
between the two states increase proportionally with the strength of the external magnetic field.
Equation 3)
36
The electron can be exited from one energy level (parallel) to another (antiparallel) by an oscillating
magnetic field. As the energy levels of the electron are quantized, the electron exists only in these two
energy states (i.e. parallel and antiparallel) and to excite an electron, the energy of the oscillating
magnetic field must match the energy difference between the two levels. This resonance condition
between the unpaired electron and the microwave field is the basis of EPR. This is called the resonance
condition and it is shown mathematically in equation 4).
Equation 4)
When the energy (E=hv) of the oscillating magnetic field is exactly equal to the energy difference
(ΔE=gβH), the paramagnetic species absorbs the energy and it is this absorption that is detected in EPR
(see Figure 1.16).
+½
-½
+10
-1
-10
+1
hv hv hv
H
E
The reason absorption can be detected is the small difference in spin populations, i.e. that the number
of electrons in the lower energy parallel orientation (ms=-½) slightly exceeds the number of electrons in
the higher energy antiparallel orientation (ms=+½). This ratio is described by the Boltzmann distribution
and is dependent on ΔE and hence the strength of the external magnetic field and the temperature.
Increasing the magnetic field strength of lowering the temperature will increase the difference in spin
populations (117).
Figure 1.16 - Schematic diagram of the energy levels, Zeeman and hyperfine interactions for a S=½, I=1 system such as that of nitroxide (
14N). The x-axis is the external magnetic field strength (H) and the y-axis is energy (E).
The vertical arrows mark the EPR transitions with the resulting first-derivative spectrum shown below. Modified from Fajer P. G., 2000 (117).
37
Equation 5)
The EPR spectrum is the first derivative of the absorption of microwave frequency radiation depicted as
a function of the magnetic field intensity. The absorption of microwaves occurs by varying the magnetic
field strength in a limited range around a central value and, in most cases, an EPR spectrum consists of
many absorption lines. When analyzing an EPR spectrum, features and parameters such as the magnetic
field strength values that cause absorption and the number, separation, and the relative intensity of the
absorption lines, along with their widths and shapes are all related to the structure of the species
responsible for the spectrum and its interaction with its surroundings. In this way EPR spectra provide
information on side chain dynamics (118,126,127), and thus on protein topography, conformational
changes, secondary and tertiary structure (see Figure 1.18).
Most proteins, besides metalloproteins, do not contain unpaired electrons so it is necessary to introduce
a so-called spin label. Spin labels are chemical compounds containing a stable unpaired electron, usually
in the form of a nitroxide derivative, and a functional group for specific attachment to the protein,
usually at cysteine residues. The spin label used in this project is the methanethiosulfonate spin label
(MTS) seen in Figure 1.17.
The relevance of the EPR technique has increased dramatically with the advances in molecular biology.
The ability to introduce cysteine residues at desired locations through site-directed mutagenesis gave
rise to the method site-directed spin labeling (SDSL) pioneered by Hubbell et al. (118,128) in 1989. In
this method native cysteines are mutated out and new cysteine residues are reintroduced at the desired
locations. In this way it is possible to attain EPR information of almost any residue (119,120)within a
protein as best illustrated in the cysteine scanning method where every residue is mutated to a cysteine
and labeled with nitroxide (129).
Introduction of a second paramagnetic center allows distance measurements within the protein (129-
131) (see Figure 1.19). Demonstrating the feasibility of this approach for assessing NHE1 structure, a
recently published study employed EPR spectroscopic distance measurements between spin labeled
side chains on two NhaA monomers to confirm NhaA dimerization (28). Distance measurements by EPR
are based on the dipolar interaction between electron spins, i.e. the force of the magnetic field of one
spin label on the magnetic dipole of the other. The energy of the interaction between spin label with
Figure 1.17 – Schematic presentation of the methanethiosulfonate spin label (MTSSL). The methyl groups on the neighboring carbon atoms stabilizes the unpaired electron in the pπ orbital of the N-O bond and the unsaturated pyrrolidine ring limit the flexibility of the spin label. The methyl thiosulfonate group reacts specifically with the thiol group of cysteine residues and covalently binds the spin label to the protein.
38
magnetic moment µ1 and another spin label with magnetic moment µ2 is inversely proportional with the
distance between them r to the power of 3 (117,124,125).
Equation 6)
Distances of 8-20Å can be determined from changes in EPR line shape due to the spin-spin interactions
mentioned above. To discern EPR line shape changes due to dipolar coupling between two spin labels
three separate EPR experiments are necessary; the EPR spectra of each of the two single-labeled
proteins and the spectrum of the double-labeled protein. If the spin-labels are within 20Å, the dipolar
interaction will cause line broadening in the spectrum of the double-labeled protein compared the
summed spectra of the two single-labeled proteins. Using SDSL and EPR I was able to assess and
determine interatomic distances between residues in hNHE1 and paNHE1, as described in the Results
section.
Figure 1.18 – Side chain dynamics of spin labeled proteins. The EPR line shape reveals information about the dynamics of the region in which the spin label resides. The faster the motion of the spin label the higher the amplitude and little broadening are seen. As the spin label itself is quite rigid the motion seen is that of the protein backbone. At τ=1nsec corresponding to random coil sharp peaks are seen and as τ increases broadening of the spectrum is seen. It could be the motion of the spin label was restricted by secondary structure (τ=3nsec) or secondary and tertiary (τ=10nsec). The introduction of a second spin label in proximity of the first would also cause broadening. Hence the EPR line shape can be used to assess the structural and/or the magnetic local environment of the spin label. Figure kindly provided
by John C. Voss, University of California Davis.
39
> 20Å
< 20Å
The advantages of EPR in biological structure determination are several; EPR detects only unpaired
electrons, EPR detects unpaired electrons in any phase and over a wide range of temperatures, the large
electron magnetic dipole results in long range effects and importantly, EPR can be used to study
proteins that cannot be crystallized for X-ray diffraction study, proteins that are too large for study by
NMR in solution, and proteins in a much more natural environment than is possible for either NMR or X-
ray crystallography, including in membranes and even in whole cells and tissues.
There are some limitations to the EPR technique involving spin labels. The first concern is whether the
reintroduction of one or more cysteine residues results in a protein in its native structure. To determine
whether the reintroduction of cysteine residue(s) had interfered with the native structure of the NHE1
protein I performed assays testing the function of the exchanger. I inferred that if the cys-reintroduced
protein exhibited similar function to the wildtype, the native structure was preserved. The next concern
is whether the covalent binding of spin label(s) perturbs the local structure and hence shifts the protein
from its native structure. However, substantial work shows the MTSSL to be well tolerated in proteins.
The method has been applied to a wide assortment of protein types, with very few examples showing a
major functional or structural consequence resulting from this modification. Direct evidence for how
the incorporated nitroxide is accommodated in protein structures has been obtained in high-resolution
crystal structures of T4 lysozyme containing spin-labeled side chains (123,132). Even at buried sites, no
significant perturbation of the backbone is evident. The non-perturbing nature of the MTS-SL can be
attributed to the relatively compact size of the modified cysteine residue (a molecular volume on the
order of Tyr) and its ambivalent chemical nature, which does not favor highly polar or nonpolar
environments.
Figure 1.19 – Spin coupling
provides a measure of
intermolecular distance. Sites
within a protein are closer than
20Å if the spectrum of the
double-labeled protein appears
broader than the composite
spectrum made from the sum of
the two single-labeled proteins.
Figure kindly provided by John C.
Voss, University of California Davis.
40
2. AIM
The overall aim of this project was to investigate the structure-function relation of the TM region of
NHE1 in general and the mechanism with which Na+/H+ exchange occurs in particular by EPR analysis, in
order to experimentally validate a model of NHE1 threaded on the crystal structure of NhaA.
Specificially, I wanted to determine the interatomic distances of several sets of amino acid residues
located in the TMIV/TMXI assembly of the hNHE1 model structure, in order to test the hypothesis that
these two TM helices are closely adjacent and actively contribute to the ion translocation process.
A second aim was to further validate the model by comparing interatomic distances and dynamics of
TMIV and TMXI in a lower vertebrate homolog, paNHE1, which is highly homologous in the TM region
yet differs from NHE1 with respect to inhibitor sensitivity.
41
3. MATERIALS AND METHODS
3.1 Cells and culturing
For this project I have used a Chinese Hamster Ovary (CHO) cell line named AP-1 that lacks a functional
NHE1 protein. The AP-1 cell line was constructed from CHO cells by the proton suicide technique. By
utilizing the ability of NHE1 to reversibly transport Na+ and H+, and the fact that Li+ can substitute for Na+
in this translocation, researchers Rotin and Grinstein (1989) were able to create the AP-1 cell line (133).
CHO cells were exposed to a mutagen, methane sulfonate, and then loaded with Li+. When exposed to
low external pH (5.5) in media without alkali cations the cells that has functional NHE1 protein will
rapidly and massively acidify. The reversed antiporter action, taking in H+ and extruding Li+, causes a
cytosolic acidification that will be lethal, and thereby cells that lack functional NHE1 are selected. The
mutation that leads to a nonfunctional NHE1 protein in the AP-1 cell line is stable for about twenty-five
passages (133). The AP-1 cells were a kind gift from S. Grinstein at Hospital for Sick Children in Toronto.
The AP-1 cell line has been the model system for my project, and I have created several stable
transfections that I will describe later in this section.
Maintenance of cell lines - AP-1 cells were grown in α-modified minimum essential Eagle’s medium (α-
MEM, Sigma cat# 086K2439) with 10% fetal bovine serum (FBS, Gibco cat# 10 106-177), 1%
penicillin/streptomycin (pen/strep, Invitrogen cat# 15140-148), and 1% L-glutamine (L-glut, Gibco cat#
25030-024). Cells were grown in T75 flasks (75cm2, Cellstar cat# 658 170) at 37˚C with 95% humidity and
5% CO2. Cell cultures were passaged every 3-4 days, when a confluency of about 90% was reached, to
ensure continuous growth. After removal of growth medium the cell monolayer was washed in 37˚C
phosphate buffered saline (PBS) solution followed by mild trypsination for 1-2 min. Proteolytic
degradation of cell surface proteins allowed for easy detachment and the cells were then suspended in
an appropriate volume of 37˚C growth medium, and a fraction were transferred to a new flask with
growth medium. Since all my constructs are based on the pcDNA3.1(+) plasmid, which carries the
neomycin resistance gene, transfected AP-1 cells are grown with 600µg/ ml geniticin (G418) in the same
growth medium as AP-1 cells to select for positive transfectants.
Freezing and thawing of cells - Cell aliquots were routinely frozen at low passage number to maintain
cell stocks. Upon reaching 90%-95% confluency growth medium was aspirated, and the cell monolayer
washed in 37˚C PBS followed by a brief incubation with trypsin (20-30 seconds) after which the trypsin
was also aspirated as to avoid trypsin in the final cell suspension to be frozen. The cells were then gently
resuspended in an appropriate volume of freezing medium (α-MEM with 20% FBS, 1% pen/strep, 1% L-
glut, and 10% DMSO) and immediately transferred to cryotubes which were then placed in a -80˚C
freezer for gradual freezing for 48 h. For long term storage, the cryotubes were transferred to a liquid
nitrogen tank. A minimum of three cryotubes of each cell line were stocked at all times.
When thawing cells, a cryotube of the cell line in question was placed in a 37˚C heating bath for rapid
thawing. Once thawed, the cells were gently transferred to a T75 flask containing 37˚C growth medium.
Figure 2.1 -
42
In order to rid the growth medium of DMSO, the medium was replaced when the cells had adhered to
the flask bottom (4 to 6 h after thawing).
3.2 Cloning and expression of NHE1 constructs
As described in the Introduction the basic principle of EPR and SDSL is to direct a nitroxide spin label to
the thiol group of a cysteine residue. So in order to examine the chosen interatomic distances, it has
been necessary to design and produce recombinant NHE1 proteins. In a cysteine-free background, i.e. a
NHE1 version with all endogeneous cysteine residues mutated to alanine or serine, I reintroduced one
or two cysteine residues.
Name Mutation hNHE1 V160C V160C hNHE1 H473C H473C hNHE1 V160C/H473C V160C/H473C hNHE1 L166C L166C hNHE1 L465C L465C hNHE1 L166C/L465C L166C/L465C hNHE1 A173C A173C hNHE1 I461C I461C hNHE1 A173C/I461C A173C/I461C hNHE1 R180C R180C hNHE1 I451C I451C hNHE1 R180C/I451C R180C/I451C
As presented in section 2, the project has two aims and constructs were made to determine the
intermolecular distances of the TMIV/TMXI assembly of hNHE1 as shown in Table 3.1, and of paNHE1 as
shown in Table 3.2 to elucidate structure-function relations of NHE1.
Name Mutation paNHE1 TM4 A164C paNHE1 TM11 I452C paNHE1 TM4/TM11 A164/I452C
3.3 Site-directed mutagenesis
The common starting point for all constructs made was either the human gene for NHE1 in which all 9
endogenous cysteine residues had been replaced with alanine, called hNHE1 cysless, or the flounder
paNHE1 cysless equivalent. The reason is, of course, that the spin label will label any and all cysteine
residues available, and so it is necessary to remove the endogenous cysteine residues by point mutation.
Both hNHE1 cysless and paNHE1 cysless were available in the SFP laboratory in vector form prior to the
start of my project.
Table 3.2 – Flounder NHE1 constructs.
Table 3.1 – Human NHE1 constructs.
43
The vector used was the pcDNA3.1 (+) from Invitrogen (Figure 3.1), which utilizes a human
cytomegalovirus immediate-early (CMV) promoter for expression in mammalian cells. The introduction
of 6 or 9 histidine residues at the C-terminal end of NHE1 allowed me to perform protein purification
based on affinity chromatography. The reason for choosing the C-terminal for the poly-His tag was the
most N-terminal part of NHE1 was predicted to function as a signal peptide, and thus could be cleaved
off after the targeting event. In order to have a poly-His tag that could be of use in protein purification,
the plasmid was modified with a primer encoding 6 (for paNHE1 constructs) or 9 (for hNHE1 constructs)
histidine residues and overlapping the C-terminal of NHE1. Three-way ligation of the digested hNHE1
cysless ORF, the digested (and linearized) pcDNA 3.1 (+) vector, and a C-terminal His-tag produced the
hNHE1 cysless plasmid that was the starting point of my project.
Important features of the pcDNA 3.1(+) vector include the CMV promoter to drive protein expression in
our mammalian cell system, multiple cloning site for easy modification, origins of replication for both
eukaryotes and prokaryotes, and neomycin and ampicillin resistance genes for easy selection (Figure
3.1).
The cysless NHE1 clone in pcDNA3.1 (+) was modified in a polymerase chain reaction (PCR –
Quickchange, Figure 3.2) with primers designed to introduce a cysteine residue at the points of interest
in an approach called site-directed mutagenesis. The kit used was the Stratagene® QuickChange® Multi
Site-Directed Mutagenesis Kit (see Appendix 3). When designing primers for such a procedure, it is
important that the primer is able to bind to the DNA stretch of interest, i.e. the energy of the correct
hydrogen bonds must exceed the repulsive energy of the mismatch that generates the site-directed
mutation. Therefore, a stretch of 10 or more nucleotides should bind on either side of the point
Figure 3.1 – Representation of the pcDNA3.1(+) with a detailed view of the multiple cloning site. CMV promoter 232-819; Multiple cloning site 895-1010; BGH polyadenylation sequence 1028-1252; f1 origin 1298-1726; SV40 early promoter and origin 1731-2074;Neomycin resistance gene 2136-2930; SV40 polyadenylation sequence 3104-3234; pUC origin 3617-4287 (complementary strand); Ampicillin resistance gene 4432-5428 (complementary strand). From the pcDNA3.1(+) manual, cat# V790-20, Invitrogen
™
44
mutation. To ensure the correct annealing, the primers were designed to have a specific melting
temperature by using such software as OligoCalc*(see Appendix 3 for primer design). In this way, the
temperature-regulated cycle of denaturing, annealing and elongation of PCR will greatly amplify the
NHE1 DNA containing the point mutation (see appendix 1 for primer design and cycling parameters). A
final treatment with an endonuclease (Dpn1) that degrades methylated and hemi-methylated DNA
ensured that the mutated species was used for transformation of E.coli cells.
3.4 Transformation, DNA purification, and sequencing
The Dpn1 treated PCR product of single-stranded mutant plasmid DNA is used to transform
ultracompetent XL10 gold E. coli cells in order to obtain large amounts of double stranded recombinant
plasmid DNA. The ampicillin resistance gene of the pcDNA 3.1 (+) vector allows us to effectively select
cl1s that have been transformed and is expressing the resistance gene, and hence the recombinant
NHE1 protein. The transformation was carried out as described by the manufacturer (see Appendix 4).
Briefly, an aliquot of the XL10-gold ultracompetent was thawed in prechilled Falcon tubes on ice. The
cells were then treated with β-mercaptoehtanol for ten min, and 1,5 µl of the Dpn1 treated single-
stranded DNA from the PCR reaction was added and incubated 30 min on ice. The cells were then heat
Figure 3.2 – Quickchange principle of PCR is to use a temperature regulated cycle of 1) denaturing 2) annealing and 3) elongation to exp1ntially amplify a DNA sequence of interest, the template. By using primers designed to flank the DNA sequence of interest, we can limit the amplification to just that sequence. I have used the process to introduce a point mutation by designing primers that differ in just 1 or two base pairs and as such will change the codon to encode a cysteine residue. In the first step the double stranded DNA template is heated until it denatures into single stranded DNA. The melting temperature (Tm) is the temperature where the thermal energy exceeds the intermolecular forces, mostly hydrogen bonds, between the two DNA strands and thereby causes denaturing. Because of this the melting temperature is different from template to template and is dependent on the length and the G/C content of the double-stranded molecule. In the second step the temperature is lowered to allow the primers, specifically designed to introduce a point mutation in the DNA sequence of interest, to anneal to their (almost) complementary sites. This step is of crucial importance. There is a risk of nonspecific binding if the temperature is too low, and if it is too high there will be little or no annealing. In the third step the temperature is raised to allow for optimal elongation by the DNA polymerase. The temperature and length of each step of a cycle is optimized for the reaction of interest taking into account the melting temperature of the primers and the optimal temperature of the DNA polymerase used. From Stratagies Tech Talk Vol.20 No.2.
45
shocked at 42˚C for exactly 30 seconds, and after incubation on ice for 2 min, NZY+ broth was added and
the cells allowed to grow at 37˚C for 1 h. The cells were then plated on ampicillin plates and allowed to
grow overnight. Four to eight clones were selected for further growth, and extraction and purification of
plasmid DNA (see Appendix 4). An aliquot of each clone was saved as glycerol stocks, i.e. bacteria mixed
with 15% glycerol and frozen. The amplified plasmid DNA is extracted and purified using Qiaprep® Spin
Mini-prep Kit which consists of a three-step protocol: preparation and clearing of bacterial lysate,
adsorption of DNA onto a silica membrane, and washing and elution of plasmid DNA. The process was
done in accordance with the manufacturer's guidelines. To ensure that the site-directed mutation was
carried out correctly, a small amount of the purified plasmid DNA was sequenced (Eurofins MWG
Operon,Sequencing Department, Anzinger Str.7a, 85560 Ebersberg, Germany) using primers designed
for this purpose. The correctly mutated and purified plasmid DNA was then used to transfect
mammalian cells in order to create a stably transfected cell line.
3.5 Transfection
When researching the structure of a protein, it is of critical importance that the protein is folded,
modified and localized correctly since data obtained from unfolded or incorrectly folded protein would
yield no useful information. To ensure a correct folding and posttranslational modification of the
recombinant NHE1, we have chosen a mammalian expression system. As noted above, the cell line I
used as expression system is a Chinese Hamster Ovary (CHO) sub line called AP-1, which is devoid of a
functional NHE1 protein, due to a selection process described in section 2.X. Hence, any and all NHE1
purified from transfected AP-1 cells will be the recombinant protein of interest.
The transfection reagent chosen was Lipofectamine™ LTX (Invitrogen, cat # 15338-100) which forms a
complex with DNA and mediates transfection in a wide variety of eukaryotic cells. Prior to transfection
with the NHE1 plasmid DNA the AP-1 cells are grown to 60%-65% confluence. The chemical transfection
agent is able to pass through the plasma membrane but not the nuclear membrane. For this reason, it is
necessary for the cells to divide and thereby fragment the nuclear membrane to allow the plasmid DNA
access to the nucleus. The cells were washed and then incubated for 1 h in Opti-MEM® (Invitrogen, cat #
31985-070), a serum- and antibody-free medium (see Appendix 5). The purified plasmid of interest and
Lipofectamine were separately incubated with Opti-MEM® for 20 min. The two solutions were mixed and
incubated for another 20 min. The Opti-MEM was aspirated from the AP-1 cells, and the cells were
incubated with the plasmid and lipofectamine in Opti-MEM mixture for 4-6 h. The transfection medium
was aspirated from the AP-1 cells and they were incubated 18-48 h in normal AP-1 medium. To ensure
that only the transfected cells survive, selection was made based on the neomycin resistance gene in the
pcDNA3.1(+) plasmid by adding 600 µg/ml geniticin sulphate (G418) (Invitrogen).
3.6 Clonal selection
Transfection of AP-1 cells yields populations of cells in which the NHE1 plasmid DNA has incorporated
differently. To ensure uniform and high expression of the recombinant protein, I performed clonal
selection by limiting dilution. This was done by placing 100 μl of a 10 ml medium suspension containing
46
approximately 70 cells into each well of a 96-well tissue culture plate, i.e. each well will on average
contain no more than one cell. I then observed the cells with an inverted light microscope over two
weeks to determine which wells contained a single colony derived from one cellular origin. These clones
were then transferred (as described in section 3.1) in succession to 48-, 24-, 12-, and 6-well tissue
culture plates and then tested for NHE expression by western blotting and functional screening based on
the recovery of pHi following acidification, as described below.
3.7 Lysates
Lysates were made from each clone to confirm NHE1 expression by western blotting. Transfected cells
were grown in 100mm cell culture dishes to 90%-95% confluency before harvest. The cell medium was
aspirated, and the cell monolayer washed briefly in cold PBS before lysis occurred by adding 250 µl 90˚C
lysis buffer (1% SDS, 10 mM Tris-HCl, 1 mM Na2VO3, protease inhibitor (cOmplete Mini, Roche cat #
11836153001 ), pH=7,5). Using a rubber policeman the cells were scraped off and transferred to an
Eppendorff tube and placed on ice. After reheating the samples (90˚C for 5 min) the lysates were
homogenized by transferring them through a gauge 27 needle 15 times. Cell debris was removed by
cooled centrifugation for 5 min at 20,000xg and the supernatant kept in -20˚C freezer until use.
3.8 Protein determination
To ensure that equal amounts of protein were loaded from each lysate the protein concentration was
determined using the BioRad DC protein assay-kit (BIO-RAD, cat # 500-0116) by following the procedure
recommended by the manufacturer. Lysate (5 µl + 20 µl H2O) or standard (25 µl) was mixed with 125 µl
working reagent, consisting of 1:50 solution of Reagent S (BioRad, cat# 500-0115) and Reagent A
(BioRAD, cat# 500-0113), followed by addition of 1 ml Reagent B (Folin-Ciocalteau phenol reagent,
BioRAD, cat# 500-0114) immediately followed by rigorous mixing. The samples were then incubated in
the dark for 15 min. The optical densities were measured with a GeneQuant spectrophotometer
(Amersham Pharma, Biotech) at 595 nm. The protein concentrations were calculated from linear
regression of the BSA standard curve (See Appendix 6).
3.9 SDS-PAGE
Lysates were diluted according to protein concentration with ddH2O to obtain the same protein
concentration in all samples, and mixed with 25% sample buffer (4× LDS sample buffer, Invitrogen cat#
NP0007, with 50mM DTT) before loading onto a 10% Bis-Tris Gel with 10 or 12 wells (NuPage, cat#
NP0301BOX and NP0302BOX, respectively). A BenchMark Protein Ladder (Invitrogen, cat# 10 747-012)
was used as molecular size marker. Proteins were separated by gel electrophoresis for 145 min at 140 V
in a Novex system electrophoresis chamber (XcellSure Lock cat# El0001), with 400 ml running buffer
(Invitrogen, cat# NP0001) in the outer chamber, while the inner chamber was filled with 200 ml running
buffer containing 500 μl antioxidant (NuPage, Invitrogen cat# NP0005).
47
3.10 Western blotting
Separated proteins were transferred from gel to nitrocellulose membrane (Nitrocellulose
Membrane Filter Paper sandwich 0.2 μm Pore Size, Invitrogen cat# LC2000) by placing gel and
membrane, sandwiched by filter paper, in a Xcell IITM Blot Module (Invitrogen cat# EI9051) filled with
transfer buffer (NuPage cat# NP0006-1) and then exposed to electrophoresis at 25 V for 2 h. Protein
bands were visualized with 0.1 % Ponceau S (Ponceau S in 5% acetate, Sigma Aldrich cat# P7170), and
the bands of the protein ladder were marked with pencil. The membrane was cut into pieces according
to expected apparent molecular weights of the proteins to be blotted and placed in blocking buffer (5%
dry milk in TBST: 10 mM Tris/HCl pH 7.4, 2.5 M NaCl, 0.1% Tween 20 (polyoxyethylenesorbiat, Sigma,
cat# P1379-500ML) for 2 h at room temperature. The membrane pieces were then placed in a humid
chamber and incubated with primary antibody for 2 h at room temperature or overnight at 4˚C. The
blots were washed 1x15 min and 3x5 min in TBST before being transferred to a humid chamber and
incubated with Alkaline Phosphatase (AP) – or Horse Radish Peroxidase (HRP)-conjugated secondary
antibody for 1 h. After washing the blots as described above, protein bands were visualized with
BCIP/NBT (KPL cat# 50-81-08) for AP-conjugated secondary antibodies. Blots with HRP-conjugated
secondary antibodies were developed using a chemiluminescent solution (PIERCE, cat# 34080) and
protein bands visualized by exposing the blot to Kodak X-ray film for an appropriate time in the dark and
subsequent development of the film.
Antigen IgG Produced by Cat # Dilution WB Dilution CLSM NHE1
Mouse (monoclonal)
Millipore/Chemicon MAB3140 1:200 -
NHE1 Rabbit (Polyclonal)
Mark Musch, University of Chicago
Xb17 1:200 1:200
β-actin Mouse (monoclonal)
Sigma Aldrich A5441 1:5000 -
Conjugated to IgG Produced by Cat # Dilution WB Dilution CLSM Alkaline Phophatase Goat-anti-Mouse Sigma Aldrich A1293 1:5000 - Alkaline Phophatase Goat-anti-Rabbit Sigma Aldrich A3937 1:5000 - Horse Radish Peroxidase Goat-anti-Mouse Zymed 51202611 1:5000 - Alexa Fluor 488 Goat-anti-Rabbit Molecular Probes A11019 1:600
3.11 Silver Staining
After SDS-PAGE the separated proteins were visualized by Silver Staining using the Bio-Rad Silver
Staining kit (BIO-RAD cat # 161-0449). The gel was soaked in a fixative (40% methanol and 10% acetic
acid) for a minimum of 30 min followed by 5 min incubation with Oxidizer (Bio-Rad, cat # 161-0444). The
Table 3.3 - Primary antibodies used for western blotting and CLSM.
Table 3.4 - Secondary antibodies used for western blotting and CLSM.
48
gel was thoroughly rinsed with ddH2O and then incubated with Silver Reagent (Bio-Rad # 161-0445) for
20 min. After a brief water rinse the gel is developed (Bio-Rad # 161-0447). When the desired protein
bands were visible the development was stopped by soaking the gel in 5% acetic acid for 15 min.
3.12 Confocal Laser Scanning Microscopy
I wished to analyze the localization of the recombinant NHE1 in the stably transfected AP-1 cells to
ensure that the mature NHE1 was correctly localized to the plasma membrane. For this purpose I grew
stably transfected AP-1 cells as described in Section 3.1 on prewashed cover slips to 60%-75%
confluency. Upon removal of growth medium, the cell monolayer was washed once with Phosphate
Buffered Saline solution (PBS, pH=7.4) and then incubated with fixation buffer (PBS with 2%
paraformaldehyde) for 15 min at room temperature (RT), and subsequently 30 min on ice. The fixated
cells were washed 3x5 min in Tris-HCl buffer saline (TBS, pH=7.4) solution before cell membranes were
permeabilized by incubation with permeabilization buffer (TBS with 0.5% Triton X-100, pH=7.4) for 10
min at RT. After removal of permeabilization buffer the cells were incubated with blocking buffer (TBS
with 5% BSA and 0.1% Triton X-100, pH=7.4) for 30 min at RT. Cells were incubated with primary
antibody against NHE1 (1:200 in TBS with 1% BSA and 0.1% Triton X-100, pH=7.4) overnight at 4˚C (see
Table 2.3). After 3x5 min washes with washing buffer (TBS with 1% BSA and 0.1% Triton X-100, pH= 7.4)
cells were incubated with flourochrome conjugated secondary antibody (1:600 in TBS with 1% BSA and
0.1% Triton X-100, pH=7.4) for 2 h at RT. The cell monolayer was then washed, as previously described,
before being briefly (15-30 seconds) incubated with DAPI (1:1000 in TBS with 1% BSA and 0.1% Triton X-
100, pH=7.4). Following a final wash the cover slips were placed cell-side down on a microscope slide
with a drop of mounting medium (PBS with 90% glycerol and 2% N-propyl gallate). Any air bobbles were
removed by gentle pressure, and the edges were sealed with nail polish and allowed to dry for a
minimum of 20 min before being stored at 4˚C in the dark.
Localization of antibodies and immunofluorescence was visualized using a Leica DM IRB/E microscope
with a Leica TSC NT confocal laser scanning unit (Leica Lasertechnik, Heidelberg, Germany), and was
carried out by SFP, and hence, will not be further described here. The only further image processing was
overlays and adjustments of brightness/contrast, carried out using Adobe Photoshop software.
3.13 Functional screening
Western blotting or immunofluorescence confirm the presence of NHE1 protein, but beyond the
localization information obtained by immunofluorescence, tells us nothing about whether or not the
recombinant protein is functional. To ensure that the structural information gathered from this project
is from a fully functional NHE1 protein, I screened the clones for NHE1 function by testing the ability of
the stably transfected cells to recover from cytosolic acidification. As previously mentioned, the AP-1
cells are not able to regulate their pHi (133). Hence, only the transfection of a functional NHE1 protein
will render them able to recover from cytosolic acidification.
The experiments were done using the proton sensitive fluorescent probe 2'-7'-biscarboxyethyl-5-6-
carboxyfluorescein (BCECF) and a fluorescence spectrophotometer. The fluorescence characteristics of
49
BCECF make it perfect for detecting pHi recovery of cells stably transfected with recombinant NHE1
protein. BCECF fluoresces most intensely in its deprotonated state, as seen from the excitation spectrum
(Figure 3.3) (134). As the H+ concentration increases, the fluorescence intensity of BCECF decreases
because of the quenching effect of the protons (135-137). This enabled me to detect changes such as
recovery of pHi. Increasing the H+ concentration also shifts the excitation spectrum towards shorter
wavelengths. BCECF has a pKa of 6.97 making it ideally suited for measuring physiological changes in pHi
(135-137). In its ester form, BCECF-AM, BCECF can cross the plasma membrane and enter the cell by
simple diffusion enabling me to easily and noninvasively load the cells. When inside the cell, BCECF-AM
is cleaved to BCECF by intracellular esterases (135-137).
The excitation maximum is around 500 nm at the pH range of this experiment and the emission
maximum is around 525 nm. At 445 nm the fluorescence of BCECF is insensitive to changes in pH, but
excitation at 460 nm to 530 nm BCECF is pH sensitive. This enables me to utilize a ratiometric approach
to my function screenings. The advantage of being able to express the fluorescence of BCECF as a ratio
between the pH insensitive wavelength 445 nm and the pH sensitive wavelength 495 nm is that in this
way the influence of factors other than pHi (variables in the loading of BCECF, BCECF loss by bleaching or
leaking, loss of cells) is eliminated.
Fluorescence measurements were made using a Ratio Fluorescence Spectrophotometer Model C-44
(PTI, Lawrenceville, NJ). The spectrophotometer uses a 75 W Xenon Arc lamp to excite cells at
wavelengths ranging from 250-650 nm. Two to three days before the experiment, cells were seeded
onto rectangular cover slips so that, at the time of the experiment, the cells reached 90% confluency.
Prior to seeding, the cover slips were washed (1h in 12 M HCl and then thoroughly washed in ddH2O)
and autoclaved, and coated with rat collagen (1h in 0,02M acetic acid with 50 µg/ml collagen Type I, BD
Biosciences cat # 354236) to prevent excess loss of cells. On the day of the experiment, the cells were
briefly washed in 37˚C PBS and incubated with 2 µM BCECF-AM in isotonic ringer solution (IR) for 30 min
at 37˚C, 0%CO2 to load the cells with BCECF. The cells were then washed twice in 37˚C IR and incubated
another 20 min with IR to ensure the complete cleavage of BCECF-AM to BCECF.
Figure 3.3 – Excitation and Emission profile of BCECF-AM. From Dissing & Gasbjerg, 1993 (134).
50
Fl
ou
resc
ence
/ A
u
Time / S
1
2
3
4
5
6
To induce NHE1 activity the cytosol was acidified using the ammonium prepulse technique (See Figure
3.4). Initial intracellular alkalinization is caused by the ability of ammonia, NH3, to diffuse through the
membrane where it reacts with intracellular protons to produce NH4+. At a slower rate NH4
+ enters the
cell, through carriers and channels, and dissociates to protons and ammonia to cause a gradual
acidification. Upon removal of extracellular NH4+ rapid acidification is caused by the diffusion of
ammonia out of the cell which drives the following reaction towards an increase in the
concentration(3).
Experimentally, the cells were exposed to IR for 400 seconds and then NH4Cl-IR for exactly 400 seconds.
It was important to be uniformly precise since the level of intracellular acidification is dependent on the
duration of the ammonium prepulse (see Appendix 9 for solution composition). The extracellular
ammonium was removed by exposing the cells to Na+ free IR (Na+ replaced by N-methyl-d-glucamine,
NMDG) called NMDG-IR for 400 seconds. During this phase rapid cytosolic acidification occurs, and since
NMDG cannot be transported by NHE1 no recovery is possible. The cells are then allowed, if able, to
recover their pHi in IR for 400 seconds or longer (Table 3.5 and Figure 3.4).
Figure 3.4 - Schematic representation of the recovery of [Hi] following cytosolic acidification monitored by the H+ sensitive
probe BCECF. 1) Cells are exposed to IR and this represents the baseline ratio. 2) Initial alkalinization caused by NH3 diffusing into the cell and reacting with cytosolic H
+ to form NH4
+. 3) Gradual acidification caused by carrier and channel mediated
entry of NH4+. 4) Rapid acidification caused by the removal of extracellular NH4
+ causing NH3 to diffuse out of the cell leaving
an increase in H+ in the cytosol. 5) The cells are trapped at the [Hi] as there is no Na
+ to allow NHE1 to transport H
+. 6) The
cells are exposed to IR and allowed to recover if able.
51
Flo
ure
scen
ce /
Au
Time / Min
A C
B
In Figure 3.5 the initial ratio represents the [Hi+] in isotonic ringer at unperturbed conditions whereas
the final ratio represents the [Hi+] after recovery of intracellular acidification. In the schematic example
the cells recover completely from the intracellular acidification but in reality this is not always the case,
even if they have functional NHE1. The initial regulation represents the Na+-dependent H+ translocation
in terms of Δratio/min, i.e. Δ*Hi+]/time – an indirect measurement of NHE1-mediated Hi
+ transport, and,
in the AP1 cells, specifically of the activity of the introduced NHE1 mutants. None of the ringer solutions
used contains bicarbonate so the contribution from the bicarbonate transporters mentioned in Section
1.1 to pHi regulation is negligible.
Solution Effect Duration Spectrophotometer IR + 2µM BCECF-AM Load cells with the fluorescent probe BCECF 30 min No IR Ensure complete cleavage of BCECF-AM to BCECF 20 min No IR Baseline 400 sec Yes NH4CL-IR Ammonium prepulse 400 sec Yes NMDG-IR Ammonium wash out without Na
+ 400 sec Yes
IR pHi recovery 400 sec Yes
3.14 Slot blot
Stably transfected AP-1 cells were grown as described in section 3.1 in five 100mm dishes to 95%-100%
confluency. After aspirating the growth medium, the cell monolayer was washed briefly in cold PBS and
placed in liquid nitrogen. The dishes were placed on ice and 4 ml cold harvest buffer (Tris-HCl buffered
saline (TBS); 50 mM Tris-HCl, 150 mM NaCl, pH=7,4) with protease inhibitor (Roche, cat# 11836153001)
and EDTA was added, and the cells scraped off and collected in a sterile, prechilled 15 ml tube. The
harvested cells were divided into 500µl aliquots and centrifuged at 20,000xg for 1 min and the
Figure 3.5 - Quantification of the ratiometric fluorescence measurements. A) The initial ratio is the average ratio
over the first 2 minutes. B) The initial regulation is the slope (ΔRatio/min) of the line fitted to the near linear part of the curve immediately after letting the cells recover their [Hi
+]. C) The final ratio is average ratio over the last 2
minutes.
Table 3.5 – Saline solutions used for the functional analysis of NHE1. The exact compositions of these solutions are given in Appendix 9.
52
supernatant discarded. The pelleted cells were resuspended in 500µl harvest buffer with protease
inhibitor, 25U/ml benzonase, and 2mM MgCl2. The lysed cells were then centrifugated at 20,000xg for 1
min and the supernatant aspirated. The pelleted cell fragments were resuspended in 250µl of
solubilization buffer with different detergent concentrations (TBS (pH=7.4) with Triton X-100 and n-
dodecyl beta-D-maltoside, DDM) with protease inhibitor with EDTA (Roche, cat# 11836170001) and
incubated at room temperature for 30 min. All the slots of a cleaned Slot Blot module (Bio-Dot SF
apparatus, cat # 170-6542) containing a sandwich of three pieces of filter paper and a nitrocellulose
membrane, all 9x12cm and presoaked in PBS, were washed with 250µl PBS by applying vacuum until the
PBS was just drained. 250µl of each solubilized sample were added to separate slots and 250µl PBS to
the remaining slots. Sample and PBS were then forced through the nitrocellulose membrane by applying
vacuum. Each slot was then washed by adding 250µl PBS and applying vacuum. The nitrocellulose
membrane was then processed as those of a western blot (see section 3.9).
3.15 Protein purification
The first, and often biggest, challenge encountered when researching membrane proteins is to obtain
the protein in sufficient amounts and sufficiently pure form. Membrane proteins are usually present at
low levels in biological membranes, and it is rare that a single protein species is a major peptidic
constituent of the membrane. As noted above, I overcame this by expressing recombinant NHE1 protein
at the control of the CMV promoter in AP-1 cells lacking endogenous NHE1. Studying membrane
proteins has the inherent challenge of extracting said membrane protein from its natural environment
of lipids without disrupting the structure or function of the protein. For this reason there is an increased
interest in the manner in which detergents interact with membrane proteins. NHE1 has 12 TM domains
which constitute roughly half the protein. Inherent in the protein are hydrophobic forces keeping the
TM domains within the membrane and hydrophilic forces affecting the cytoplasmic tail. Different
detergents have a wide array of characteristics making them suitable for very different interactions with
membrane proteins. For this study I chose a two-detergent system to extract NHE1. Triton X-100 is a
nonionic detergent that has been widely used to extract membrane proteins on account of it being a
comparatively mild detergent that preserves protein structure and function (138,139). The other
detergent used, n-Dodecyl β-D-maltoside (DDM) is also nonionic, and hence mild, and has proven to be
able to solubilize membrane proteins in fully functional states (Figure 3.6 and Table 3.6) (138,139). The
fact that the hydrophilic head of both the detergents used are nonionic causes them to disrupt lipid-lipid
and lipid-protein interactions rather than protein-protein interactions, resulting in less denaturing of the
extracted protein.
Figure 3.6 – Detergent structure. Structure of A) Triton X-100 and B) n-Dodecyl β-D-maltoside (DDM)
A) B)
53
Name Monomer Mass Critical Micellar Concentration Aggregation Number Triton X-100 625g/mol
* 2,5×10
-4M 75-165
n-Dodecyl β-D-maltoside 511g/mol 1,8×10-4
M 110-140
The purification of the extracted NHE1 was designed as a two-step affinity chromatography procedure,
which is described in detail below, and outlined in Figure 3.7. Briefly, in the first step I utilized the
histidine tag of the recombinant protein by passing the solubilized protein through a HiTrap chelating
nickel column. Proteins with a histidine tag are retained in the column by complex binding to Ni2+ that
are immobilized by chelating to imino diacetic acid that is covalently bound to the column material. In
this way I was able to bind recombinant his-tagged NHE1 and wash off any protein unable to complex
bind to Ni2+. Although six consecutive histidine residues are very uncommon in nature, some
endogenous proteins may be able to complex bind to Ni2+, and therefore I introduced a second affinity
chromatography step to the protocol. The second step of the purification utilizes NHE1’s ability to bind
calmodulin (CaM) at a high (CaM-A, Kd ~20 nM; residues 637–656) or low (CaM-B, Kd ~350 nM; residues
657– 700) affinity, positively-charged cluster in its C-terminal regulatory domain. The partly purified
NHE1 is eluted from the nickel column into agarose beads onto which CaM is immobilized. Upon
addition of Ca2+, NHE1 will bind CaM and be retained by the resin. To avoid competition of endogenous
CaMs, EDTA was added during cell harvesting, thereby depleting the cell for Ca2+ and inactivating
endogenous CaMs. The NHE1-CaM binding is more rigid than that of NHE1 to nickel, so this step allows a
more thorough wash. I introduced a low salt wash to rid the sample of contaminants with
predominantly hydrophobic interactions and a high salt wash to rid the sample of contaminants with
predominantly hydrophilic interactions. Using this strategy I have been able to extract and purify
recombinant NHE1 for EPR analysis. Samples for western blotting and silver staining were taken
throughout the purification to analyze the procedure.
Table 3.6 – Characteristics of the detergents employed. Shown are the Monomer Mass, Critical Micellar Concentration (CMC), and Aggregation Number (AN) of the detergents used. The CMC is the concentration at which the detergent starts to form micelles. The AN is the number of detergent monomers found in a single micelle.
*For Triton X-100 the monomer mass
shown is an average. Data are taken from M. le Maire et al. 2000
54
Cells harvested and lysed by sonication
5000g
Supernatant is membrane fragments and cytosolic proteins and complexes
Pellet is large sub-cellular structures such as the nucleus
340.000g
Supernatant is cytosolic proteins and complexes
Pellet is membrane fragments
NHE1 is extracted from the membrane in a two-detergent system (Triton and DDM)
12000gSupernatant is solubilized NHE1
Pellet is insoluble membrane fragments
NHE1 is purified by its histag’s affinity to a nickel column
NHE1 is purified by its affinity to Calmodulin bound to agarose beads
NHE1 is concentrated and spin-labeled for EPR measurements
1 2
3
4
5
6
7
8
Harvest and solubilization - Cells for NHE1 purification were grown, as described in section 3.1, in
twelve 150mm culture dishes to 95%-100% confluency. After aspirating the growth medium, the cell
monolayer was washed briefly in cold PBS and placed on ice. Then 1 ml cold harvest buffer (Tris-HCl
buffered saline (TBS); 50 mM Tris-HCl, 150 mM NaCl, pH=7,4) with protease inhibitor (Roche, cat#
11836153001) and EDTA was added, and the cells scraped off and collected in a sterile, prechilled 15 ml
tube and placed on ice (step 1 in Figure 3.7). Harvested cells were centrifuged at 400xg for 10 min and
the supernatant discarded. The pelleted cells were resuspended in 4 ml harvest buffer with protease
inhibitor and lysed by sonication for 3x10 s intervals while being placed on ice at least 30 seconds in
between. The lysed cells were then centrifuged at 5000xg for 10 min at 4˚C to pellet large insoluble sub-
cellular structures, such as the nucleus (step 2 in Figure 3.7). Supernatants were then transferred to
centrifuge tubes and centrifuged at 340,000xg for 20 min at 4˚C to pellet the membrane fraction (step 3
in Figure 3.7). The membrane fraction was gently resuspended in an appropriate volume of
solubilization buffer (TBS with 0.3% Triton X-100 and 0.04% n-dodecyl beta-D-maltoside, DDM) with
protease inhibitor without EDTA (Roche, cat# 11836170001) and incubated at 30˚C for 30 min with
gentle stirring (step 4 in Figure 3.7). Centrifugation at 12,000xg for 10 min at 4˚C (step 5 in Figure
3.7)pelleted insoluble membrane fragments, such as lipid rafts, and the supernatant, the solubilized
Figure 3.7 – Experimental outline of the protein purification. This is a schematic representation of the outline of the protein
purification experiment in which each step is numbered (Blue numbers). The split arrows denote centrifugation and the boxes
denote the fraction that contains NHE1.
55
membrane fraction, was transferred to a pre-equilibrated nickel-column (GE Healthcare, cat#
17040801).
Purification by HiTrap chelating nickel column - Columns (HiTrap™ Chelating HP 1ml columns, GE
Healthcare cat # 17-0408-01) were reused, and after each elution, a column was stripped with 2 ml 0.1
M HCl and washed with 12 ml ddH2O. Columns were stored at 4˚C in 20% ethanol. To equilibrate the
HiTrap chelating nickel column before use, it was washed with 6 column volumes (column volume = 1
ml) of sterile ddH2O, and 1.5 ml of sterile 0.1M NiSO4 was added to the column. After being washed with
6 ml of ddH2O to rid the column of excess Ni2+, the column was equilibrated with 6 ml of loading buffer
(TBS with 0.04% DDM and 20 mM imidazole).
The solubilized membrane fraction was loaded onto an equilibrated nickel column and then incubated
for 30 min at 4˚C (step 6 in Figure 3.7). The column was then washed with 6 ml loading buffer followed
by 6 ml washing buffer (TBS with 0.04% DDM and 80 mM imidazole). Protein was eluted from the nickel
column in 6 ml elution buffer (TBS with 0.04% DDM and 300 mM imidazole) directly into the pre-
equilibrated CaM-conjugated affinity resin (CaM beads, Sigma cat# 094K7016).
Purification by CaM-conjugated affinity resin - Like the nickel columns, the CaM beads
(Calmodulin−Agarose, Sigma cat # MFCD00165272) were reused and, after each use the beads were
stripped by incubating the resin in elution buffer (TBS with 0.04% DDM and 5 mM EGTA) for 10 min at
room temperature. The CaM beads were pelleted by centrifugation at 1000 rpm for 1 min, and the
elution buffer carefully aspirated. After two washes in TBS the resin is stored at 4˚C in 1 ml TBS with
0.05% sodium azide. Before any purification the beads were washed twice in TBS and then washed once
in binding buffer (TBS with 0.04% DDM, 2 mM CaCl2, and 1 mM MgCl2) before being equilibrated with
binding buffer for 10 min.
Purified protein was eluted from the nickel column directly into the CaM beads, and CaCl2 and MgCl2
were added to a final concentration of 2 mM respectively (2 µl/ml of 1 M CaCl2 and 2 µl/ml of 1 M
MgCl2) to ensure optimal binding of NHE1 to the CaM-conjugated affinity resin (step 7 in Figure 3.7). The
sample was incubated overnight at 4˚C to accomplish complete binding of NHE1 to the CaM beads.
Bound protein was then washed once with binding buffer, once with low salt buffer (10 mM Tris-HCl
with 0.04% DDM, 2 mM CaCl2, 1 mM Mg2Cl2, pH=7.4), and once with high salt buffer (50 mM Tris-HCl
with 0.04% DDM, 350 mM NaCl, 2 mM CaCl2, 1 mM Mg2Cl2, pH=7.4). Purified protein was eluted by
adding 5 ml elution buffer (50 mM Tris-HCl with 0.04% DDM, 5 mM EGTA, pH=7.4) and incubating the
resin with gentle mixing for 15 min at room temperature. The purified NHE1 was concentrated into 100
µl using a 50 kD MWCO centrifugal filter device (Milipore cat # UFC805024) and 5 µl MTS spin label
(Methanethiosulfonate, Toronto Research Chemicals cat # A188600) was added and incubated for 2 h at
4˚C. To remove excess MTS, 12-18 ml washing buffer (50 mM Tris-HCl with 0.04% DDM, pH=7.4) were
run through the filter device, and the purified NHE1 was concentrated to an approximately 20 µl sample
(step 8 in Figure 3.7), usually corresponding to a protein concentration of 1µg/µl.
56
Sample Purified NHE1 Buffer, pH=7.4 Buffer, pH=1.0 Cariporide Final conditions NHE1 5.1 µl 0.9 µl NHE1 at pH7.4 NHE1, pH=5.5 5.1 µl 0.9 µl NHE1 at pH=5.5 NHE1, cariporide 5.1 µl 0.9 µl NHE1 with 10 mM cariporide
For each purification EPR measurements was performed on purified spin-labeled NHE1, purified spin-
labeled NHE1 at pH=5.5, and purified spin-labeled NHE1 with the inhibitor cariporide. For each set of
experiments an EPR measurement was performed on the flow through from the last wash of the
purified spin-labeled NHE1 to act as background.
3.16 EPR spectroscopy
I assisted on all EPR measurements, unless otherwise noted, which were kindly done by Madhu
Budamagunta, University of California Davis. EPR measurements were performed using a JEOL X-band
spectrometer fitted with a loop-gap resonator (140,141). A 6 µl aliquot of the purified, spin-labeled
protein, at a final concentration of approximately 10 μM (1.0 µg/µl) in PBS buffer (pH 7.5) containing
0.01% DDM, was placed in a sealed quartz capillary contained in the resonator. Spectra (averages of
three 2-min scans) were acquired at room temperature (22oC) at a microwave power of 4 mW and with
the amplitude optimized to the natural line width of the individual spectrum. The obtained spectra were
double integrated and normalized. The identically treated Cys-less protein that was used as a negative
control showed no signal at these settings (not shown).
Interspin distance determination - The distance between the two spin-labeled side chains was
determined from the spectral broadening of the double-labeled sample, compared to the composite
spectrum from the two corresponding single-labeled samples. All distance calculations were done by
John C. Voss, University of California Davis. First, the spectra for the double-labeled TM IV/TM XI
(Ala173Cys/Ile461Cys) construct were normalized after a baseline adjustment, followed by integration
using the Basephase program (kindly provided by C. Altenbach, Dept. of Ophtalmology at the Jules Stein
Eye Institute, UCLA, LA, California). For the integration, the best possible baseline was obtained and the
spectrum was double integrated. After the double integration, the total area was equalized for the two
single-labeled constructs, TM IV (Ala173Cys) and TM XI (Ile461Cys), respectively, which thus normalized
them to the same number of spins. The spectra of the two single-labeled constructs were summed, and
compared to the spectrum for the double-labeled construct, using the same process as described above.
The normalized spectra for the double-labeled construct and the composite spectrum for the single-
labeled constructs were analyzed for dipolar broadening using Fourier deconvolution to calculate an
ensemble of Pake splittings that can best account for the degree of broadening and, the interspin
distance was calculated from the Pake splittings as described previously (117,125,131,142).
Table 3.X – Preperation of EPR samples. The same amount of purified protein (5.1µl) was used for each EPR sample and in the NHE1 sample 0.9µl buffer (50mM Tris-HCl, 0.04% DDM) at pH=7.4 was added. In the low pH sample 0.9µl buffer (50mM Tris-HCl, 0.04% DDM) at pH=1.0 was added to a final pH≈5.5. To the inhibitor sample 0.9µl cariporide dissolved in DMSO was added to a final cariporide concentration of 10mM.
57
4. RESULTS
The results section is comprised of three parts. First I will present the experimental data from the first
check point in my experimental strategy; Western Blots and Functional tests testify to the presence of a
functional recombinant NHE1 protein. For this project it was necessary to optimize an existing protein
purification protocol and the second part of the results section contains an experimental walk through
of this optimized protocol. Lastly, but most importantly, I will present the EPR data resulting from the
purified NHE1 protein.
The interatomic distances of the TMIV/TMXI assembly of hNHE1 and paNHE1 I sat out to determine is
shown Table 4.1 and Table 4.2 respectively. These residues are spread throughout the extracellular and
intracellular funnels that make up the ion translocation pathway (see Figure 5.3) and determining the
distances shown below will assess the importance of the TMIV/TMXI assembly for NHE1 function.
Residue Location Distance to Residue Location V160 TMIVe - H473 TMXIe
L166 TMIVe - L465 TMXIe
A173 TMIVi - I461 TMXIi
R180 IL2/TMIVi - I451 TMXIi
Residue Location Distance to Residue Location A164 TMIVi - I452 TMXIi
4.1 Assessment of NHE1 clones
Western blotting - In order to ensure that the cell lines used for further protein purification and EPR
measurements were indeed stably transfected with a functional recombinant cysteine reintroduced
NHE1 protein western blotting and functional screening was performed for all constructs. After
transfection of AP-1 cells with constructs designed and produced to contain the gene for the
recombinant NHE1 proteins listed in Table 3.1, ten to twelve clones were selected by limiting dilution.
To evaluate the expression of the recombinant NHE1 protein lysates of each clone were made and
analyzed by SDS-PAGE followed by Western Blotting against NHE1. Recombinant NHE1 protein was
detected by the polyclonal antibody Xb17 or the monoclonal antibody MAB3140 and visualized by HRP-
conjugated secondary antibody subjected to its substrate. As a loading control β-actin was detected by a
monoclonal antibody and visualized by HRP-conjugated secondary antibody subjected to its substrate.
At least two forms of NHE1 can be detected; a glycosylated NHE1 which runs at an apparent molecular
weight of 100-110kDa and a non-glycosylated form which runs at <90kDa. Ultimately the recombinant
NHE1 protein will be the subject of structural studies and therefore it is crucial that the cell lines express
a mature and fully functional protein.
Table 4.2 – Interatomic distances to be determined in paNHE1. This spin pair corresponds to the hNHE1 A173C – I461C.
Table 4.1 – Interatomic distances to be determined in hNHE1
58
AP
-1
paN
HE1
WT
paN
HE1
cys
less
paN
HE1
A1
64
C
paN
HE1
I45
2C
paN
HE1
A1
64
C/I
45
2C
NHE1
β-actin
From Figure 4.1 it is clear that no NHE1 is found in the non-transfected AP-1 cells confirming that any
and all NHE1 detected in the stably transfected AP-1 cell lines originated from the respective
recombinant paNHE1 proteins. Obviously the data in Figure 4.1 is from several different blots as seen by
the loading controls and although the presentation is not very elegant it serves the purpose of
answering the binary question of recombinant NHE1 expression. It is clear that the transfected AP-1 cell
lines all express the respective recombinant protein, although at different levels compared to their
controls. High levels of NHE1 protein expression are preferred as this will facilitate the purification
process and help overcome one of the two major obstacles of this project – to obtain sufficient amounts
of purified protein. From Figure 4.1 it is clear that the paNHE1 I452C clone chosen does not have as high
a level of NHE1 expression as the other transfected cell lines. The faint band does however testify to the
expression of a NHE1 protein with a molecular weight corresponding to that of a mature recombinant
paNHE1 protein, and a Na+-dependent H+-transport is seen in the function test of this clone (see Table
4.3). I therefore conclude that the stably transfected AP-1 cell lines in Figure 4.1 all express a mature
recombinant paNHE1 protein.
Figure 4.1 - Western Blot of non-transfected AP-1 cells and AP-1 cells stably transfected with recombinant paNHE1. Samples containing 20µg protein from each cell line was loaded in the respective lanes and then separated by SDS-PAGE (140V, 100min). Proteins were then blotted onto a nitrocellulose membrane (25V, 120min). NHE1 protein was detected by the monoclonal antibody MAB3140 (1:200) and visualized by the HRP-conjugated secondary anti-mouse antibody and the HRP substrate BCIP/NBT. As a loading control β-actin was detected by a monoclonal antibody and visualized as NHE1. Arrows indicate NHE1 location (≈110kDa) and β-actin location (≈45kDa). The lanes are from different Western blots. See text for more details.
59
AP
-1
hN
HE1
WT
NHE1
β-actin
hN
HE1
V1
60C
hN
HE1
H4
73
C
hN
HE1
V1
60
C/H
47
3C
AP
-1
hN
HE1
WT
NHE1
β-actin
hN
HE1
L1
66C
hN
HE1
L4
65
C
hN
HE1
L1
66
C/L
46
5C
AP
-1
hN
HE1
I46
1C
hN
HE1
A1
73
C/I
46
1C
hN
HE1
R18
0C
*
hN
HE1
I45
1C
*
hN
HE1
R18
0C
/I4
51
C*
NHE1
β-actin
Figure 4.2 - Western Blot of non-transfected AP-1 cells and AP-1 cells stably transfected with recombinant hNHE1. Samples containing 20µg protein from each cell line was loaded in the respective lanes and then separated by SDS-PAGE (140V, 100min). Proteins were then blotted onto a nitrocellulose membrane (25V, 120min). NHE1 protein was detected by the polyclonal antibody Xb17 (1:200) and visualized by the HRP-conjugated secondary anti-rabbit antibody and the HRP substrate BCIP/NBT. As a loading control β-actin was detected by a monoclonal antibody and visualized by the HRP-conjugated secondary anti-mouse antibody and the HRP substrate BCIP/NBT. Arrows indicate NHE1 location (≈100kDa) and β-actin location (≈45kDa). See text for more details. *Samples from not clonally selected, i.e. heterogeneous, cell lines.
A)
B)
C)
60
Figure 4.2 shows the NHE1 expression pattern of all the hNHE1 transfected cell lines; the hNHE1 V160C
– H473C spin pair in A; the hNHE1 L166C – L465C spin pair in B; and finally the corrected hNHE1 A173C –
I461C spin pair in C. Also shown in Figure 4.2-C is the NHE1 expression pattern of the not clonally
selected hNHE1 R180C – I451C spin pair. Because of the limited time of my project I have not gone
further with this spin pair. From Figure 4.2-A it is clear that the selected clones of the hNHE1 V160C –
H473C spin pair all express a NHE1 protein with a molecular weight corresponding to that of a mature
recombinant hNHE1 protein. In Figure 4.2-B and -C we see that the selected clones all express the
glycosylated and non-glycosylated form of hNHE1 protein. Glycosylation of NHE1 is not essential for
function but is thought to be important for correct membrane targeting (31). I therefore presume that
the non-glycosylated form of NHE1 seen in Figure 4.2-B and -C is functionally identical to the fully
mature transporter and hence also structurally identical, at least within the ion translocation funnel and
the TM4/TM11 assembly. On the basis of the Western Blot analysis of NHE1 expression I conclude that
the selected clones of transfected AP-1 cells all express a NHE1 protein with a molecular weight
corresponding to that of a mature recombinant hNHE1 protein.
Functional assays - The H+ sensitive fluorescent probe BCECF was used to detect Na+-dependent H+-
transport by the cell lines stably transfected with recombinant NHE1. Positive clones were chosen for
further analysis on a qualitative basis. In Figure 4.3-A I have depicted the fluorescence ratio of the H+-
insensitive wavelength (445 nm) and the H+-sensitive wavelength (495 nm) as a function of time for
three separate experiments: non-transfected AP-1 cells (A) and AP-1 cells transfected with hNHE1 WT
(B) and hNHE1 V160C (C) respectively. It is evident from Figure 4.3-A that non-transfected AP-1 cells are
not able to regulate their [Hi+] within the timeframe of the experiment. When exposed to intracellular
acidification (the steep rise around 12min in ratio in A), and allowed to recover in a Na+-containing
isotonic solution no change in ratio is detected corresponding to no change in [Hi+]. This is what was
expected since AP-1 cells lack a functional endogeneous NHE1 (see Figure 4.1). From this I conclude that
any and all Na+-dependent H+-transport seen in the stably transfected cell lines originates from the
recombinant NHE1 protein.
In Figure 4.3-B we clearly see the function of hNHE1 WT. When the transfected cells were allowed to
recover in Na+-containing isotonic solution, the BCECF ratio rapidly returned to the value of the initial
ratio. In this experiment we indirectly see the extrusion of excess intracellular H+ (the fall in ratio
represents a fall in [Hi+]) by one of the fastest secondary active transport proteins known. In this way I
was able to evaluate the function of the recombinant NHE1 proteins in my project. I included AP-1 cells
transfected with hNHE1 V160C in Figure 4.3-C and here it is evident that the recombinant protein show
Na+-dependent H+-transport similar to hNHE1 WT. I therefore conclude that the recombinant hNHE1
V160C protein adopts a native and fully functional structure that is correctly transported and embedded
in the plasma membrane when expressed by transfected AP-1 cells.
61
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 5 10 15 20 25 30
Ratio (
445nm
/495nm
)
Time (min)
AP-1A)IR NH4Cl-IR NMDG-IR IR
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 5 10 15 20 25 30
Ratio (
445nm
/495nm
)
Time (min)
hNHE1 WTB)
IR NH4Cl-IR NMDG-IR IR
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 5 10 15 20 25 30
Ratio (
445nm
/495nm
)
Time (min)
hNHE1 V160CC)
IR NH4Cl-IR NMDG-IR IR
Figure 4.3 - NHE1 function tested by ratiometric fluorescence measurement of the H+ sensitive probe BCECF.
AP-1 cells, non-transfected A) or transfected with hNHE1WT B) or hNHE1 V160C C) were grown on coverslips and placed in a Ratio Fluorescence Spectrophotometer. The cells were continuously perfused by the HEPES-buffered ringers; IR followed by NH4Cl-IR, Na
+-free (NMDG-IR), and finally IR again as indicated. Each data trace
is from one experiment but is representative of three separate experiments. Initial ratio, final ratio, and initial regulation values are averages of three separate experiments (see Section 3.13 and Table 4.1).
Initial ratio = 0.19
Final ratio = 0.36
Initial regulation = 0.0006
Initial ratio = 0.17
Final ratio = 0.18
Initial regulation = -0.0613
Initial ratio = 0.19
Final ratio = 0.19
Initial regulation = -0.0797
62
Construct Cell line Initial ratio Final ratio Initial regulation
- AP-1 0.19 0.36 0.0006
paNHE1 WT AP-1 0.20 0.32 -0.0238
paNHE1 cysless AP-1 0.17 0.19 -0.0907
paNHE1 A164C AP-1 0.26 0.32 -0.0503
paNHE1 I452C AP-1 0.20 0.28 -0.0183
paNHE1 A164C/I452C AP-1 0.21 0.33 -0.0392
hNHE1 WT AP-1 0.17 0.18 -0.0613
hNHE1 V160C AP-1 0.19 0.19 -0.0797
hNHE1 H473C AP-1 0.15 0.15 -0.1064
hNHE1 V160V/H473C AP-1 0.16 0.12 -0.0684
hNHE1 L166C AP-1 0.21 0.26 -0.0466
hNHE1 L465C AP-1 0.24 0.34 -0.0155
hNHE1 L166C/L465C AP-1 0.23 0.30 -0.0192
To quantify the level of NHE1 function of the recombinant proteins I operated with three values that
characterize each data set. The initial ratio is the average ratio of the first minute of the experiment and
is a measurement representative of the pHi under physiological conditions. From Table 4.1 we see that
the initial ratio can vary somewhat from cell line to cell line. It is quite interesting that the introduction
of different recombinant NHE1 proteins result in slightly different resting levels of [Hi+] but I have not
pursued this phenomenon as it is beyond the scope of this project. The final ratio is the average ratio of
the last minute of the experiment and is a measurement representative of the [Hi+] after the cells have
been allowed to recover in Na+-containing isotonic ringer solution. The difference between the initial
ratio and the final ratio is descriptive of the level of [Hi+]-recovery. From Table 4.3 it is clear that while
some transfected cell lines, hNHE1 V160C for instance, recover their [Hi+] fully within the timeframe of
the experiment others, hNHE1 L465C for instance, do not. This could cause some concern as to whether
the kinetics of the transporter has been altered by the reintroduced cysteine mutation as it is known
that the mutation of many residues can alter the set pHi set point, the [Hi+] at which the NHE1 protein
activates. However within 10-12 clones prepared from each transfected cell line the number of copies
and the location of the gene for the recombinant NHE1 protein will differ. The result is a difference in
the expression level of the recombinant NHE1 protein and there is a correlation between the cell lines
that do not fully recover their [Hi+] and the NHE1 protein level as seen in the western blot analysis
(Figure 4.2-B)). From the initial regulation it is clear that AP-1 cells transfected with hNHE1 L465C have
Na+-dependent H+-transport. However, the data in Figure 4.2-B strongly suggest that the lesser velocity
than hNHE1 WT exhibited by the hNHE1 L465C transfected cells is due to a lower expression level and
not any dysfunction of the actual hNHE1 L465C protein. The initial regulation is the slope (Δratio/min) of
the line fitted to the near linear part of the data trace immediately after the cells were allowed to
recover their [Hi+]; as such the initial regulation is an indirect measure of NHE1 activity. For cells
Table 4.3 - Data from functional screenings of non-transfected AP-1 cells and AP-1 cells stably transfected with recombinant NHE1. Initial ratio is the average ratio of the first minute and the final ratio is the average ratio of the last minute of the experiment. The initial regulation is the slope (ΔRatio/min) of the line fitted to the near linear data trace immediately after letting the cells recover their [Hi
+] in IR. All
values are average of 3 separate experiments.
63
exhibiting NHE1 activity the slope is negative because the intensity of fluorescence from the H+-sensitive
wavelength 495nm rises as Hi+ concentration drops due to the diminishing quenching of H+ - thereby
lowering the value of the ratio 445nm/495nm. It is possible to convert the ratio/min measurements to
pH/min plots by an extensive calibration of each cell line (see (82)). These experiments were performed
to answer the question; does this cell line exhibit NHE1 activity? Therefore I have chosen this qualitative
approach rather than a quantitative one as a simple yes or no answer will do.
From the data in Table 4.3 and Figures 4.1 and 4.2 I conclude that the transfected cell lines all express a
mature recombinant NHE1 protein that adopts a native and fully functional structure that is correctly
transported and embedded in the plasma membrane, although at different levels of protein expression.
CLSM images – NHE1 was detected by the primary polyclonal antibody Xb17 (see table 3.X) followed by
the flourophore-conjugated Alexa Fluor 488 secondary antibody. In Figure 4.4 merged images of hNHE1
(green) and the cell nucleus (colored blue by DAPI) are shown for AP-1 cells transfected with hNHE1
V160C (A), hNHE1 H473C (B), and hNHE1 V10C/H473C (C).
A B C
From Figure 4.4 it is clear that the recombinant NHE1 localizes primarily to the plasma membrane of
transfected cells. This is further indication that the expressed recombinant NHE1 protein has undergone
correct post-transcriptional modification, is fully mature, and proves it is correctly localized. The little
NHE1 signal seen in the hNHE1 V160C transfected AP-1 cells is more likely to be caused by poor labeling
than lack of recombinant NHE1 as the same cell line is able to fully recover pHi (see Figure 4.3-C and
Table 4.3) and clearly expresses the recombinant hNHE1 V160C protein (see Figure 4.2-A). What little
NHE1 signal seen in Figure 4.4-A does localize to the plasma membrane.
Figure 4.4 – CLSM images of the hNHE1 V160C – H473C spin pair. The merged images of NHE1 (Green, Alexa Fluor 488) and the cell nucleus (Blue, DAPI) of AP-1 cells stably transfected with A) hNHE1 V160C, B) hNHE1 H473C, and C) hNHE1 V160C/H473C. The size marker in A) corresponds to 20µm.
64
4.2 Protein Purification
The two major challenges of studying the structure of membrane proteins are to be able to obtain
sufficient amounts of protein and for the protein to be of a sufficient purity. Trying to overcome this
challenge of yield and purity I optimized a protocol previously used for NHE1 purification from epithelial
mammalian cells the outline of which is shown in Figure 4.5.
Cells harvested and lysed by sonication
5000g
Supernatant is membrane fragments and cytosolic proteins and complexes
Pellet is large sub-cellular structures such as the nucleus
90000g
Supernatant is cytosolic proteins and complexes
Pellet is membrane fragments
NHE1 is extracted from the membrane in a two-detergent system (Triton and DDM)
12000gSupernatant is solubilized NHE1
Pellet is insoluble membrane fragments
NHE1 is purified by its histag’s affinity to a nickel column
NHE1 is purified by its affinity to Calmodulin bound to agarose beads
NHE1 is concentrated and spin-labeled for EPR measurements
As many as fifteen confluent 150mm cell culture dishes were harvested to obtain enough recombinant
NHE1 protein for an EPR data set. Even under the control the powerful CMV promoter it seems that
NHE1 can only be made to constitute a small part of the protein content of a transfected AP-1 cell. An
optimization of the transfection protocol might have facilitated the purification of NHE1. The first
challenge I encountered was extracting the recombinant NHE1 protein from its native environment; the
plasma membrane. For this purpose I chose a two-detergent system using Triton X-100 and n-Dodecyl β-
D-maltoside (DDM). Both detergents are nonionic and hence considered to be mild and able to disrupt
lipid-lipid and lipid-protein interactions rather than protein-protein interactions (138,139). In this way I
was able to extract NHE1 without denaturing it. Since DDM has a rather low CMC value the addition of
Figure 4.5 - Schematic overview of the purification protocol.
65
Triton will lessen the homogeneity of the detergent layer surrounding the hydrophobic areas of the
NHE1 protein and thereby lessen the probability of the creation of micelles.
To determine what concentrations of detergent I should use to extract NHE1 I performed a slot blot and
subsequent Western blot analysis of a range of detergent concentrations.
0.01% DDM0.5% Triton
1.1% Triton
0.9% Triton
0.7% Triton
0.5% Triton
0.3% Triton
0.04% DDM
0.03% DDM
0.02%DDM
0.01% DDM
0.005% DDM
O.3% Triton 0.04% DDM
0.09% DDM
0.07% DDM
0.05% DDM
0.04% DDM
0.03% DDM
0.5% Triton
0.3% Triton
0.2% Triton
0.1% Triton
0.05% Triton
1.0% DDM
0.4% DDM
0.3% DDM
0.2% DDM
O.3% Triton
The experimental data in Figure 4.6 was the basis of my decision to use 0.04% DDM and 0.3% Triton as
my two-detergent NHE1 solubilization system. To be sure that I had indeed solubilized NHE1 from the
plasma membrane I performed a slot blot of both the supernatant and the pellet followed by western
blot analysis. In Figure 3.7 it is clear that using 0.3% Triton and 0.04% DDM increases the level of
solubilization of NHE1 from the plasma membrane.
Figure 4.6 - Detergent ratio analysis by slot blot. In the first column of A the Triton concentration is set to 0.5% and the DDM concentration is varied from 0.005% to 0.04%. In the second column of A the DDM concentration is set to 0.01% and the Triton concentration varied from 0.3% to 1.1%. B is a separate experiment that combines the optimal concentrations found in A: 0.04% DDM and 0.3% Triton and then varies both the DDM concentrations (first two columns) and Triton concentration (last column). NHE1 was detected by the monoclonal antibody MAB3140 and processed as a western blot.
A) B)
66
Pellet Supernatant
0.3% Triton
1.0% Triton
2.0% Triton
300mM
160mM
80mM
40mM
20mM
Ni-Load sample
Figure 4.7 - Slot blot analysis of the solubilization of NHE1. Harvested and lysed cells were treated with 0.04% DDM and a varied concentration of Triton. Insolubilized NHE1 was pelleted by centrifugation and samples from both the supernatant and the pellet was run on a slot blot with subsequent western blot analysis. NHE1 was detected by the monoclonal antibody MAB3140 and processed as a western blot.
Figure 4.8 - Slot blot analysis of the washing and elution imidazole concentration. The Ni-load sample is the flow through from loading the solubilized NHE1 onto the Ni
2+-
column. Subsequently the column was treated with a set of imidazole concentrations from 20mM to 300mM and all samples were analyzed by slot blot and processed as a western blot. NHE1 protein was detected by the monoclonal antibody MAB3140.
67
The next challenge was to obtain a sufficiently pure sample of recombinant NHE1 protein from the
solubilized plasma membrane solution. I chose a two-step purification protocol in which I utilized the
his-tag and the CAM-binding site of the recombinant NHE1 protein. The his-tag enabled me to purify the
recombinant NHE1 protein by reversibly binding it to Ni2+-ions chelated to the matrix material of the
column. Endogenous proteins may contain several histidine residues in close proximity and thereby
resemble the his-tag so it was necessary to wash the column with imidazole that competes with
histidine’s binding to Ni2+. At low concentrations imidazole displaces weakly bound proteins and at
higher concentrations proteins more tightly bound to the column, such as NHE1, are displaced and
thereby eluted. To test the optimal imidazole concentrations for washing and elution of the column I
performed a slot blot followed by western blotting (Figure 4.8).
No detectable amount of NHE1 protein was lost when loading of the Ni2+ column and subsequent
washes with 20mM and 40mM imidazole did not result in any detectable displacement of NHE1 either
(Figure 4.8). Applying a wash with 80mM imidazole however resulted in the loss of a small amount of
NHE1 protein. The majority of NHE1 protein was eluted by an imidazole concentration of 160mM but
substantial amounts were eluted at 300mM imidazole. In congruence with challenges of purity and
amount I chose to wash the Ni2+-bound solubilized NHE1 protein with 80-100mM imidazole thereby
ridding the sample of weakly bound impurities and to elute NHE1 protein with 300mM imidazole to
obtain as much NHE1 protein as possible.
The process of protein purification was analyzed by western blot and silver staining after EPR data was
recorded. In Figure 4.9 the western blot and analogous silver stain from the purification of a
recombinant NHE1 protein is shown. Running analogous western blots and silver stains enabled me to
make a comparative analysis of the purification protocol. The Ni-load sample is taken from the flow
through of the loading of the solubilized NHE1 protein onto the nickel column. As seen in lane 1 (Ni-
load) of the western blot in Figure 4.9 not all NHE1 is bound to the column. Throughout this study I used
the HiTrap 1ml chelating columns able to bind ≈12mg pure his-tagged protein and it is likely that upon
increasing the level of solubilized membrane proteins I have reached the capacity of the column. For this
reason I loaded the first flow through onto the nickel column again trying to displace some of the
impurities bound to the column, and thereby increasing the ratio of NHE1 to impurities. The impurities
are mainly other solubilized membrane proteins. After loading NHE1 onto the nickel column I performed
two washes; one with 20mM and one with 100mM imidazole. Both the 20mM and 100mM imidazole
wash displaces impurities as seen in lanes 2 (Ni-wash (20mM)) and 3 (Ni-wash (100mM)) of the silver
stain in Figure 4.9.
68
Ni-
load
Ni-
was
h (
20
mM
)
Ni-
was
h (
10
0m
M)
CA
M-l
oad
CA
M-w
ash
Low
sal
t w
ash
Hig
h s
alt
was
h
NH
E1
Flo
w t
hro
ugh
198kDa
126kDa
85kDa
37kDa
198kDa
126kDa
85kDa
37kDa
But a substantial amount of NHE1 is also lost in the 100mM imidazole wash so the imidazole washing
concentration was corrected to 80mM. In lane 4 (CAM-load) and lane 5 (CAM-wash) samples from the
binding and first wash of the CaM-beads are shown. The faint NHE1 band in the western blot in Figure
4.9 testifies to my problems of complete binding of NHE1 to the CaM-beads. Several tests involving
several different binding buffers did not solve this problem so I decided to proceed with the modest
amount of NHE1 that was actually bound to the CaM-beads. From the silver stain it is clear that the
Figure 4.9 - Western blot and silver stain of the purification protocol. SDS-PAGE was performed on two gels loaded identically with samples from the protein purification process. One of the gels was then processed as a Western Blot (top) in which NHE1 protein was detected by the monoclonal antibody MAB3140. The other gel was processed as a silver stain (bottom) and the location of NHE1 is marked by the right side arrow. See text for details.
69
some proteins or impurities able to bind to the nickel column were not able to bind to the CaM-beads
and therefore was washed away. NHE1’s binding to the CaM-beads allowed for a more stringent wash
procedure so I introduced a low (lane 6 – Low salt wash) and high salt wash (lane 7 – High salt wash).
The low ionic strength of the low salt wash was designed to rid the sample of hydrophobic impurities
and the high ionic strength of the high salt wash was to rid the sample of hydrophilic impurities. Finally
NHE1 was eluted by adding the chelating Ca2+ scavenger EGTA thereby releasing NHE1 from the CaM-
beads. NHE1 was eluted in 5ml and concentrated to ≈30µl using a centrifugal filter device. The flow
through (lane 9 - Flowthrough in Figure 4.9) from the filter device was used for background
measurements for the EPR experiments. The purified NHE1 sample was the subject of at least three EPR
studies and was subsequently run in lane 9 (NHE1) in the gels in Figure 4.9. Protease inhibitor was
present throughout the purification process and from the point of NHE1 solubilization 0.3% Triton and
0.04% DDM was present in order not to deplete NHE1 of detergent.
Using the optimized purification protocol I was consistently able to obtain recombinant NHE1 samples of
sufficient amount and purity for further EPR studies.
4.3 EPR results
The process of optimizing the protein purification protocol meant several months without successful
EPR experiments. In the end I only obtained complete EPR data sets on the paNHE1 A164C – I452C spin
pair and the hNHE1 A173C – I461C spin pair, the latter in corporation with Eva Nygaard. Further to this I
will present the EPR results of the double-labeled hNHE1 V160C/H473C, for which the single-labeled
data is currently being finalized, but for which I do not have yet the results. Some of these results are
presented in the, as of yet unpublished, paper “Structural modeling and electron paramagnetic
resonance spectroscopy of the human Na+/H+ exchanger isoform 1” (appendix 1).
EPR measurements were performed using a JEOL X-band spectrometer fitted with a loop-gap resonator
(140,141). A 6 µl aliquot of the purified, spin-labeled protein, at a final concentration of approximately
10 μM (1.0 µg/µl) in TBS buffer (pH 7.4) containing 0.04% DDM, was placed in a sealed quartz capillary
contained in the resonator (samples were prepared as in Table 4.4). Spectra (averages of three 2-min
scans) were acquired at room temperature (22oC) at a microwave power of 4 mW and with the
amplitude optimized to the natural line width of the individual spectrum. The obtained spectra were
double integrated and normalized. The identically treated Cys-less protein that was used as a negative
control showed no signal at these settings (not shown).
Sample Purified NHE1 Buffer, pH=7,4 Buffer, pH=1,0 Cariporide Final conditions NHE1 5.1 µl 0.9 µl NHE1 at pH7.4 NHE1, pH≈6.0 5.1 µl 0.9 µl NHE1 at pH≈5.5 NHE1, cariporide 5.1 µl 0.9 µl NHE1 with 10 mM cariporide
Table 4.4 - Conditions for EPR experiments. See text for details.
70
Each NHE1 purification resulted in at least three EPR experiments as noted in Table 4.4. The magnetic
environment of the spinlabel(s) was assessed for NHE1 at physiological pH (pH=7.4) where NHE1 activity
is low and at acidic pH (pH≈6) where NHE1 activity is vigorously upregulated and finally at physiological
pH in the presence of 10mM the inhibitor cariporide. As seen in Table 4.4 the amount of protein sample
was identical in the three experiments and hence the spectra obtained comparable. Because of the
small volume (≈20µl) of purified NHE1 protein it was not possible to determine protein concentration
prior to EPR measurements.
Interspin Distance Determinations –There are several potential strategies for how to determine the
distance between spin-labels (117,124,125,131,142). A prerequisite for all strategies are EPR data of
both the two single-labeled and the double-labeled protein. We chose to analyze the spectral
broadening of the double-labeled sample in comparison to the composite spectrum from the two
corresponding single-labeled samples. In Figure 4.10 the EPR spectra obtained for hNHE1 A173C, hNHE1
I461C, and hNHE1 A173C/I461C is shown. The protein purification and spin-labeling of hNHE1 A173C
was performed by Eva Nygaard. If there were no spin-spin interaction, that is if the spin-labels were too
far apart to affect each other, the summed spectra of the two single-labeled would equal the spectrum
of the double-labeled protein. First, the spectra for the double-labeled TM IV/TM XI (Ala173Cys/Ile461Cys)
construct were normalized after a baseline adjustment, followed by integration using the Basephase
program (kindly provided by C. Altenbach, Dept. of Ophthalmology at the Jules Stein Eye Institute, UCLA,
LA, California). For the integration, the best possible baseline was obtained and the spectrum was
double integrated. After the double integration, the total area was equalized for the two single-labeled
constructs, TM IV (A173C) and TM XI (I461C), respectively, which thus normalized them to the same
number of spins. The spectra of the two single-labeled constructs were summed, and compared to the
spectrum for the double-labeled construct, using the same process as described above. The normalized
spectra for the double-labeled construct and the composite spectrum for the single-labeled constructs
were analyzed for dipolar broadening using Fourier deconvolution to calculate an ensemble of Pake
splittings that can best account for the degree of broadening and, the interspin distance was calculated
from the Pake splittings as described previously (131,143).
71
A B
C
EPR analysis of the TM4/TM11 intermolecular distance of hNHE1
Spectral analysis - In Figure 4.10-A the spectrum for the two non-interacting single-labelled hNHE1
A173C and I461C proteins are shown. Comparison of the EPR spectra of the two single-labeled
constructs reveals different spectral shapes. The spectrum for the A173C construct exhibits higher
amplitude and less broadening, compared to the single-labeled I46C construct, which displays a strongly
immobilized component in the low field peak (see arrow in Figure 4.10-A). This reflects that position 173
experiences greater motional freedom, possibly arising from a more disordered backbone.
Figure 4.10 – EPR spectral data of the hNHE1 A173C – I461C spin pair. The EPR spectra for purified and singly spinlabeled hNHE1 A173C (TM4), RED trace, and hNHE1 I461C (TM11), BLACK trace is shown in A. In B the EPR spectra for purified and double-labeled hNHE1 A173C/I461C (TM4/TM11), BLACK trace, and the summed spectra for the two single-labeled proteins, RED trace, is shown. The effect of reduced pH (pH=5.1), GREEN trace, and inhibitor (1mM amiloride), RED trace, on the purified double-labeled hNHE1 A173C/I461C (TM4/TM11) protein is seen compared to data collected at normal pH (pH=7.5), BLACK trace. Protein purification and spin-labeling done by Eva Nygaard.
72
Ala173Cys/Ile461CyspH=7.4
10mM Cariporide
pH=5.5
Distance analysis - The EPR spectra for the double-labelled A173C/I461C construct are shown in Figure
4.10-B. As mentioned previously, in the absence of spin-spin interaction the spectrum of the double-
labeled construct would approximate the sum of the two corresponding single-labeled constructs.
However, when comparing the traces for the double-labelled versus the sum of the single-labelled
constructs, the latter shows higher amplitude and less spectral broadening (Figure 4.10-B) indicating a
dipolar component in the double-labeled sample. Essentially this means that the introduction of a
second spin-label affects the electromagnetic environment of the first spin-label, i.e. the spin-labels at
A173C and I461C are in proximity of each other. The calculated distance between the spin-labels on TM
IV and TM XI was approximately 15 ± 2 Å (SD, n=3). The ± 2 Å variation primarily reflects noise in the
spectra for the single-labeled TM IV construct. Using the strategy of SDSL and EPR we have produced
experimental evidence supporting our theoretical structural model based on the NhaA crystal structure;
TMIV and TMXI are located in proximity less than 20Å of each other in a functional hNHE1 protein. Next
we wanted to analyze the effect function or inhibition had on the distance between the spin-labels.
Functional analysis - We recorded EPR data on the double-labeled hNHE1 A173C/I461C using protein at
pH=5.1 (Figure 4.10-C) to observe any positional changes of the TMIV/TMXI assembly brought about by
activation of hNHE1. Increasing the [H+] induced a significant broadening of the spectrum for the
A173C/I461C construct as seen by its decreased amplitude (Figure 4.10-C). This indicates that the
distance between TMIV (A173) and TMXI (I461) is reduced in the active conformation of hNHE1 protein
compared to the inactive conformation. The EPR data of the double-labelled hNHE1 A173C/I461C
protein in the presence of the inhibitor amiloride (1mM) showed less broadening of the spectrum and
higher amplitude indicating an increase in distance between residues 173 and 461 in response to
inhibition. However, a mutation was found in the DNA encoding the hNHE1 A173C/I461C protein giving
rise to the EPR spectra in Figure 4.10-C, a Pro mutation unlikely to affect structure or function of the
Figure 4.11 – EPR spectral data of the hNHE1 A173C – I461C spin pair. . The effect of reduced pH (pH=5.1), GREEN trace, and inhibitor (10mM cariporide), RED trace, on the purified double-labeled hNHE1 A173C/I461C (TM4/TM11) protein is seen compared to data collected at normal pH (pH=7.4), BLACK trace.
73
exchanger. Therefore it was necessary to correct this mutation and perform the experiment again. The
EPR data for the double-labeled correct hNHE1 A173C/I461C protein is seen in Figure 4.11. Similar
effects on the line shape of the double-labeled protein was seen in response to increased [H+]; a
broadening of the signal indicative of a reduced TMIV-TMXI distance. Interestingly, the EPR data of the
double-labeled correct hNHE1 A173C/I461C protein in the presence of inhibitor (10mM cariporide) also
showed a broadening of the spectrum for the A173C/I461C construct as seen by its decreased amplitude
(Figure 4.10-C), also indicating a reduced distance between TMIV and TMXI. See Discussion for possible
interpretations of these results. These data indicate that residues 173 and 461, and hence TMIV and
TMXI, are within 15±2Å proximity and that this distance within the protein is altered in response to
activation and inhibition. This is in congruence and further validates our model structure.
EPR analysis of hNHE1 V160C/H473C
In Figure 4.12 the EPR spectra for the double-labeled hNHE1 V160C/H473C construct is shown. As
previously mentioned I did not complete the EPR analysis of this spin pair because of the limited time
frame of the project. Analyzing the spectral data for the double-labeled construct reveals that reducing
pH results in higher amplitude and less broadening indicative of an increased distance between the spin-
labels at residues 160 and 473 placed at the extracellular end of the TM domains TMIV and XI
respectively. This is in contrast to the A173C/I461C spin pair where reduced pH resulted in broadening
of the signal, indicative of a decreased distance between residues 173 and 461 in response to activation.
Figure 4.12 – EPR spectral data of the double-labeled hNHE1 V160C/H473C protein. The effect of reduced pH (pH=5.5), RED trace, and inhibitor (10mM cariporide), BLUE trace, on the purified double-labeled hNHE1 A173C/I461C (TM4/TM11) protein is seen compared to data collected at normal pH (pH=7.4), BLACK trace. See text for more details.
74
The addition of 10mM of the inhibitor cariporide resulted in slightly higher amplitude in the low- and
mid-field peaks for the hNHE1 V160C/H473C, indicative of an increased distance between residues 160
and 473 in response to inhibition of NHE1. For a discussion of the possible interpretations of this result
see section 5 – Discussion.
EPR analysis of the TM4/TM11 intermolecular distance of paNHE1
Spectral analysis – The EPR spectra for the non-interacting single-labeled paNHE1 A164C and paNHE1
I452C constructs are shown in Figure 4.13. The spectral characteristics of the two single-labeled
constructs differ and the higher amplitude and lesser broadening of the TM4 sample indicates a higher
degree of motional freedom than the TM11 sample that exhibits a significantly lower amplitude and
more broadening indicative of restricted movement of the spin-label. These characteristics are similar to
the spectral data of hNHE1 TM4/TM11 assembly seen above.
Figure 4.13 – EPR spectral data of the paNHE1 A164C – I453C spin pair. The EPR spectra for purified and singly spinlabeled A) paNHE1 A164C (TM4) and B) paNHE1 I452C (TM11) and C) double-labeled paNHE1 A164C/I452C (TM4-TM11 double) is shown. For each spectra in A, B, C : NHE1 at pH=7.4 (protein alone, BLACK trace), 10mM cariporide (RED trace), and low pH (pH=5.5, BLUE trace). In D) the EPR spectra for the two separate single-labeled (TM4 single, BLACK trace and TM11 single, RED trace) and their summed spectra (sum of singles, PINK dotted trace) are shown along with the spectra for the double-labeled protein (4-11 double, BLUE trace). See text for more details.
75
Distance analysis – The EPR spectra for the double-labeled paNHE1 A164C/I452C construct are shown in
Figure 4.13-C. As seen in Figure 4.13-C there is a significant dipolar component in the spectra for the
double-labeled construct compared to the sum of the two single-labeled constructs. The trace of the
sum of the single-labeled constructs exhibits a higher amplitude and less spectral broadening than the
EPR spectra for the double-labeled construct indicating that the residues 164 and 452, and hence TM4
and TM11, are in proximity. The signal-to-noise ratio of the baseline of the EPR data for the paNHE1
I452C construct was too large to determine an exact distance but the dipolar coupling of the spin-labels
indicates that the intermolecular distance is less than 13-15Å, as evaluated by John C. Voss, University of
California Davis. The reasoning is that the energy of dipole-dipole interaction between two spin-labels
has an rdd-3 (distance from dipole to dipole) dependence which results in discernible line-shape changes
for distances up to 25Å.
Functional analysis – The EPR spectra of the double-labeled paNHE1 A164C/I452C construct at pH=6 or
in the presence of 10mM cariporide, respectively, are shown in Figure 4.13-D. As with hNHE1 we
observed lower amplitude and a significant broadening of the EPR spectra upon increasing the [Hi+]
indicative of a reduced distance between the spin-labels. Notably, however, the EPR spectra for the
paNHE1 double-labeled construct in the presence of 10mM cariporide show no significant change when
compared to the EPR spectra for the same construct at physiological pH. This, in conjunction with the
fact that paNHE1 is insensitive to inhibition by cariporide(see section 1.X)(82) and inhibitor binding have
been shown to be, at least in part, competitive strongly supports the notion that the TMIV/TMXI
assembly is involved in ion translocation by NHE1.
76
5. DISCUSSION
From the sparse structural knowledge previously obtained by us (82) and others (85) I set out to
elucidate some central structure-function relations of the hNHE1, focusing in particular on the putative
site of ion exchange. The experimental evidence mapping the amino acid residues of the TM region of
the exchanger (84,85) was used to produce the best possible alignment of hNHE1 with the distant
bacterial homolog NhaA (see Figure 1.13). The resulting alignment was threaded onto the published
crystal structure of NhaA (41) and thus we created a 3D model structure of hNHE1 (see Figures 1.14 and
1.15). As for NhaA the central feature of the model structure was the TMIV/TMXI assembly
hypothesized to be the structural anchor of ion translocation in both NhaA and NHE1.
Protein environment of EPR experiments - In order to provide a better framework for understanding of
the EPR results I wish to briefly discuss the environment of the NHE1 protein during EPR data collection.
I have extracted the exchanger from its native habitat, the plasma membrane, by solubilization using
detergents. The interaction between detergent and membrane protein has been subject to speculation
and is no trivial matter. The ability of detergents to extract integral membrane proteins from biological
membranes is dependent upon their ability to solubilize membrane lipid (Figure 5.1) (138).
A B
Incidentally to the removal of a substantial part of the lipid by detergent, the hydrophobic membrane
embedded region of the membrane proteins becomes enwrapped in a layer of protective detergent
coating. At this stage the membrane protein can be considered to be in a solubilized state and the
Figure 5.1 – Interaction of membrane proteins with solubilizing detergents. A generalized diagram of the various phases in protein solubilization by detergents is shown in A. a) detergent is non-cooperatively taken up by the lipid phase. b) detergent molecules cooperatively interact in the membrane and start to produce large membrane fragments (see inset) but no solubilization of the vesicles occur. c) solubilization of membrane sheets containing lipids and protein components start. In d) only mixed lipid-detergent micelles and detergent-solubilized membrane components are present. B) The structure of a detergent-solubilized membrane protein as a rectangle with its hydrophobic regions covered by detergent forming a prolate ring structure. From le Marie et al., 2000 (138)
77
detergent have formed a prolate ring around the hydrophobic region of the membrane protein(Figure
5.1) (138). Hence the recombinant NHE1 protein is in a solubilized state in which the hydrophobic region
is covered by a prolate ring of detergent and the extramembraneous hydrophilic regions are exposed to
the same solution, i.e. what used to be the intracellular and extracellular regions now reside in the same
solution. My concern is; while increased [H+] is known to interact with the intracellularly exposed region
and thereby activate NHE1, increased extracellular [H+] is actually known to inhibit NHE1 activity
through, at least in part, competing with Na+ for the ion translocation site (32).
Discussion and evaluation of the determined distances - I have presented experimental evidence that
shows the TMIV and TMXI of both human NHE1 (see Figure 4.10 and 4.11) and the winter flounder
homolog paNHE1 (see Figure 4.13) to be within 20Å proximity. This supports the hypothesis of close
interaction between these domains in NHE1 function. This is, to the best of my knowledge, the first
experimental evidence cementing the proximity of TMIV and TMXI in hNHE1 (and in the lower
vertebrate homolog paNHE1), and thus the most solid evidence to date that they may actually by
analogy to the data for NhaA, form the central part of the ion translocation pathway. In comparison the
intermolecular distance between residues A173 and I461 was calculated to 8.46Å in the model structure
(Unpublished see Appendix 1). The model structure was however based on the fully inactive state of
NhaA so the experimentally determined distance would not be expected to match exactly that of the
model structure. In comparison the entire NhaA and NHE1 molecules have been estimated at
approximately 50Å by 100Å in the membrane-parallel and vertical directions respectively (102). The
spectral shapes of the single-labeled proteins revealed TMIV in both hNHE1 and paNHE1 to be more
Figure 5.2 – Cytoplasmic view of the hNHE1 model structure. In
this schematic representation only transmembrane regions are
shown: Residues 15-31 (TM I), 104-123 (TM II), 130-147 (TM III),
160-179 (TM IV), 191-209 (TM V), 228-246 (TM VI), 255-272 (TM
VII), 300-317 (TM VIII), 333-353 (TM IX), 417-437 (TM X), 453-473
(TM XI), and 481-502 (TM XII). It is clear that TMXI appears to be
more embedded in the protein structure than TMIV. Unpublished
from Nygaard et al. see Appendix 1
78
flexible than TMXI (Figure 4.10-A and Figure 4.12-A, -B respectively). This is in congruence with the
hNHE1 model structure where residue 461 of TMXI seems to be more enclosed in the protein than 173
of TMIV (see Figure 5.2).
The ability of EPR to determine precise interspin distances is a great analytical tool. However, as any
method in the field of protein structure determination, EPR is limited by the quality of the sample
subjected analysis. The lack of exact TMIV/TMXI interatomic distances from hNHE1 and paNHE1 is a
result of difficulties in the purification of one or several constructs of the spin pair. The 2Å variation of
the calculated distance of 15Å between residues 173 (TM4) and 461 (TM11) of hNHE1 is primarily
caused by noise in the EPR spectra for the A173C sample (Figure 4.10-A). No distance calculation was
performed on the interspin TM4-TM11 distance in paNHE1 because the EPR data for TM11 sample was
too noisy (Figure 4.13-B). However, the dipole-dipole interaction of spinlabels in TMIV and TMXI, as seen
by the different line shapes of the double-labeled sample in comparison with the composite spectra of
the two single-labeled samples, is undeniable evidence of the proximity of TMIV and TMXI in hNHE1 and
paNHE1. On a side note, the calculation of interspin distances relies on several assumptions and
therefore I am confident that my experimentally based conclusions on the proximity of TMIV and TMXI
are more correct (124,131,142). One of the major advantages of EPR is the ability to obtain information
about protein dynamics (119,120,132). The fact that EPR data can be collected from proteins in
physiological relevant conditions, i.e. native structure and physiological pH and level of ions, enabled me
to determine the effect of activation and inhibition on the TMIV/TMXI assembly of NHE1.
Discussion and evaluation of the structure-function relation - On the basis of the model structure we
have proposed the molecular mechanism by which hNHE1 transports ions to be one of alternating
access (Figure 5.4). In an attempt to elucidate the molecular mechanism of Na+/H+ exchange I used a
comparative approach utilizing the different inhibitor binding characteristics of hNHE1 and paNHE1.
Based on our recent findings, which point to a close correlation between inhibitor sensitivity and ion
transport, and a role for TMIV and TMXI in both processes (82) and our threaded hNHE1 model
structure (Figures 1.14 and 1.15) I performed EPR experiments to determine if activation and/or
inhibition had an effect on the intermolecular distance of TMIV and TMXI in hNHE1 and the inhibitor
insensitive paNHE1. These experiments demonstrated perturbation of the distance between TMIV and
TMXI in response to activation and inhibition of the hNHE1 exchanger (Figure 4.10-C and Figure 4.11).
However, the extramembraneous regions of the detergent-solubilized NHE1 protein (for discussion see
above) are exposed to the same solution. In the low pH EPR experiments the [H+] is increased 200 fold
(pH=7.4 to 5.1) but is still several orders of measure below the [Na+] (145mM), which is three times
higher than the apparent Km for Na+ binding to the extracellular substrate site of NHE1 (32). Thus, while
the effect of increased [H+] seen in both hNHE1 and paNHE1 could indicate activation it can be safely be
concluded that protonation of one or several sites in NHE1, in response to pH=5, results in a
rearrangement of TMIV and/or TMXI. These sites could include the ion translocation sites for H+ and Na+
and the H+ modifier site and more. Also, in our proposed mechanism and in that of NhaA ion
translocation is preceded by a protonation event which is thought to activate the exchanger (114),
presumably the allosteric site identified previously in the literature (see section 2.X). Thus, the expected
effect of increasing the [H+] on all sides of the NHE1 on NHE1 activity is actually not trivial to predict. I
79
propose that the conformational change of the TMIV/TMXI assembly seen in the EPR spectra of the
double-labeled hNHE1 is a result of activation by protonation of the pH sensor leading to a
rearrangement of the TMIV/TMXI readying the exchanger for ion translocation. My EPR data for paNHE1
revealed that the distance between TMIV and TMXI was indeed perturbed when the exchanger was at
low pH, but no perturbation was seen when the exchanger was exposed to 10mM cariporide (Figure
4.13-C). This indicates that the inhibitor effect seen in hNHE1 is specific and involves a rearrangement of
the TMIV/TMXI assembly.
IV XI
Extracellular
Intracellular
H473
L465
I461
I451
V160
L166
A173
R180
X
XI IV
IV XI
Extracellular
Intracellular
L465
I461
I451
V160
L166
A173
R180
X
XI IV
IV XI
Extracellular
Intracellular
I452 A164
X
XI IV
A B C
H473
This is consistent with the hypothesis that the TMIV/TMXI assembly is the structural anchor of ion
exchange, as seen in our proposed mechanism, and further indicates the importance this novel fold has
in NHE1 function. Since we knew from previous studies that paNHE1 is insensitive to all commonly used
hNHE1 inhibitors (see section 1.8), I exploited this species difference to further test the model.
The EPR results of both hNHE1 (A173C/I461C; Figure 4.10-C and Figure 4.11) and paNHE1 (A164C/I452C;
Figure 4.12-C) indicate that protonation of one or several sites results in a decreased distance between
residues in TMIV and TMXI situated in the middle of the membrane (see Figure 5.3). John Voss, UC
Davis, estimated the distance decrease to be in the order of 3-5Å on both hNEH1 and paNHE1 the basis
of the EPR data. In comparison a recent cryo-EM study of the pH-activated NhaA by Appel et al. showed
a ≈7Å displacement of the periplasmic TMIV helix in response to pH=8 and substrate availability (114).
Interestingly the distance of the V160 – H437 residues located at the extracellular end of TMIV and TMXI
in hNHE1 respectively is increased in response to low pH, as indicated by the EPR spectra in Figure 4.11.
Hence, several lines of evidence suggest that altered protonation of one or more sites in NHE1 results in
Figure 5.3 – Distance alterations in NHE1 in response to low pH. In this schematic representation the TMIV (RED) / TMXI (YELLOW) assembly and the TMX (BLUE) is shown. A) The A173C – I461C distance of hNHE1 was determined to decrease in response to an increase in [H
+] (Figure 4.10-C and 4.11), while the V160 – H473 distance was shown to increase (Figure
4.12). The EPR data for paNHE1 indicated that the A164 – I452 distance (the equivalent of A) decreased in response to low pH.
80
a rearrangement of the TMIV/TMXI assembly that could correspond to one or more conformations
involved in ion translocation by NHE1.
The EPR data obtained of double-labeled hNHE1 (A173C/I461C, TM4/TM11) in the presence of inhibitor
is ambiguous. In Figure 4.10-C inhibitor induces an increase in TMIV – TMXI distance while in Figure 4.11
inhibitor induces a decreased distance. However, the conditions are different in the two experiments.
The double-labeled hNHE1 A173C/I461C protein of the initial experiment (Figure 4.10-C) had a proline
point mutation in TMIII and thus I obtained EPR data on the corrected hNHE1 protein (Figure 4.11). Also,
the inhibitor used in the first experiment is amiloride (Figure 4.10-C) whereas the more NHE1-specific
cariporide was used in the second (Figure 4.11). The proline point mutation could cause a disruption of
the helical structure of TMIII and thereby perturb the TM region of NHE1. However, the hNHE1 protein
giving rise to the EPR data in Figure 4.10-C was functional (see Section 3.13) so any perturbation was not
detrimental to the proteins ability to perform ion translocation. The inhibitors amiloride and cariporide
are similar in structure, however, previous evidence from our group shows that the sites of interaction
with NHE1 are partially different (82). As the inhibitors are free to interact with both extramembraneous
regions (see above) of NHE1 in the conditions of the EPR experiment it is possible that inhibitor binding
to the intracellular region could account for the ambiguous EPR data. However, as the inhibitors are
similar in both size and structure this is unlikely. This experiment will have to be repeated in order to be
confident of the direction of the inhibitor induced TMIV – TMXI rearrangement of hNHE1.
Critical assessment of the methods used - Caution must be taken when drawing conclusions upon EPR
data and the biggest concern is whether the EPR data recorded represents a fully functional NHE1
protein in its native conformation. I have presented experimental evidence of the expression (Figure 4.1
and 4.2) and the function (Figure 4.3 and Table 4.3) and for some cell lines even the localization of
recombinant NHE1 (Figure 4.4). The attachment of the spin label MTSSL may be cause of some concern
but there is substantial evidence that the nitroxide ring is well tolerated in proteins. This tolerance may
be caused by the compact size of the MTSSL modified cysteine residue, similar to tyrosine, and its
ambivalent chemical nature which does not favor highly polar nor nonpolar environments. I am
therefore confident of the validity of the EPR results I have presented.
The task of interpreting the EPR results presented in this study was challenged by the suboptimal
spectral quality of some experiments. As mentioned above, this stems not from any inability of the EPR
method but from difficulties obtaining a sufficiently pure NHE1 sample. The protein purification has
been the most challenging process of my work and although my optimization led to reproducible EPR
data further optimization or perhaps a different approach (see discussion below) might produce EPR
data of a higher quality, thereby facilitating the calculation of precise interspin distances. Trying to purify
membrane proteins researchers are confronted with several challenges. I choose to use the mammalian
cell line AP-1 as expression system because some post-translational modification, i.e. O-linked
glycosylation (31), has been shown to be important for correct NHE1 maturation and plasma membrane
localization. However, even under the control of the CMV promoter NHE1 protein was not a major
peptidic species in the plasma membrane of transfected AP-1 cells. To increase the yield of purified
protein I increased the amount of cells (to 0.5g-1.0g of wet cell-pellet) but obviously the ratio of NHE1 to
other membrane proteins, i.e. the signal-to-noise, remained the same. In essence the variations in the
81
quality of EPR data represent variations in the NHE1 expression level of the selected clones. Hence, to
further optimize the expression of recombinant NHE1 protein would facilitate the protein purification
and better EPR data could be obtained.
Figure 5.4 – Alternating access mechanism. In this schematic representation only the TMs IX (Red), XI (Yellow), and X (Blue) is shown. The Na+ ion is transported through NHE1 in the order: A→B→C→D→E→F. The unfavorable positive-positive and negative-negative dipoles are stabilized by R425 from TMX shown here as a black line with a positive charge. The interatomic distances that I sat out to determine samples the entire ion translocation pathway and, as shown schematically here, may result in both structural and functional information. The spin pairs are: V160C/H473C, L166C/L465C, A173C/I461C, and R180C/I451C.
A. Charge compensation from R425 located in TM X stabilizes the energetically unfavorable negative/negative and positive/positive dipole-dipole pairings in the arrangement of the TM IV/TM XI helices. B. An acidic pH change results in alteration of the protonation state of the region of TM IX located at the entrance of the proposed funnel, eliciting a conformational change in TM IX, which causes a direct contact between TM IV and TM IX (not shown). This rearrangement of TM IV results in a reorientation of TM IV and TM XI such that a Na
+ binding site is exposed to
the extracellular space. C-F. Binding of Na+ causes a charge imbalance, triggering a movement of the TM IV and TM
XI helices, exposing Na+ to the cytoplasm. The release of Na
+ results in protonation of the Na
+ binding site, causing a
conformational change leading back to the original arrangement of TM IV and TM XI.
82
Optimization of NHE1 expression could include:
Optimization of the transfection protocol to attempt to incorporate a higher number of copies
of the recombinant DNA encoding NHE1. I would attempt to transfect AP-1 cells several times
with increased amounts of recombinant DNA used for each transfection.
A selection protocol based on the activity of NHE1 and not just the presence of an antibiotic
resistance gene. The protocol involves NHE1-mediated rescue from regular acid-load of the
transfected AP-1 cells.
The experimental findings of my project will be used to further refine the model structure and to sculpt
new questions. More work will be needed to fully elucidate the TMIV/TMXI assembly and the structure-
function relations of NHE1 but an exciting notion is that if, as we propose, there are charged residues in
the ion translocation pathway, we are poised to address not only NHE1 structure-function correlates but
also the basis for some of the defining differences between channels and carriers with regard to affinity
and turnover rate. Briefly, charged residues in the permeation path could explain both the relatively
high affinity of carriers and their correspondingly lower turnover rates relative to those of channels. The
charged residue R425 was proposed by us, based on our hNHE1 model structure, to be essential for
NHE1 function by screening the dipoles of the helical ends of TMIV and TMXI situated in the middle of
the membrane. This hypothesis was verified by functional analysis of R425A mutants, which exhibited
strongly reduced function, at least in part due to reduced targeting to the plasma membrane
(unpublished, see Appendix 1).
83
6. CONCLUSION
I have presented experimental evidence in the form of intermolecular distances evaluated with SDSL
and EPR that support our 3D model structure of hNHE1 threaded on the NhaA structure. The focus of
my study was the TMIV/TMXI assembly hypothesized to be the structural anchor of the ion translocation
mechanism in hNHE1.
I have:
Introduced cysteine residues in strategic positions in cys-less human and flounder NHE1
homologs for site-directed spin labeling (method described in section 3.1).
Produced stable AP-1 clones expressing these constructs and tested them for expression of
NHE1 – by Western blotting (section 4.1); localization of NHE1 – by CLSM imaging (section 4.1);
and function of NHE1 – by recovery of pHi as detected by the ratiometric method using the pH-
sensitive probe BCECF (section 4.1).
Developed a method for purification and spin labeling of NHE1 based on the solubilization of
NHE1 by detergents DDM and Triton X-100 (section 4.2) and subsequent purification by two-
step affinity chromatography utilizing the his-tag of the recombinant NHE1 protein to purify
NHE1 by Ni2+-column (section 4.1) and the CaM binding site of the recombinant NHE1 protein to
purify NHE1 by CaM-agarose beads (section 4.1).
Presented EPR spectra that prove TM IV and TM XI of hNHE1 to be in close proximity, more
specifically about 15-20 Å apart (section 4.3).
Presented EPR spectra of the homologous paNHE1 that, in congruence with the data for hNHE1,
testifies to the proximity of TM4 and TM11, in further support of the notion of the functional
and structural importance of the TM4/TM11 assembly (section 4.3).
Demonstrated that the distance between spin-labels in TM IV and TM XI is altered in response to
acidification in both hNHE1 and paNHE1. Conversely, the addition of the NHE1 inhibitor
cariporide resulted in an alteration in the TMIV-TMXI distance in the hNHE1 but, consistent with
the model, no detectable alteration was seen in the EPR data for the inhibitor insensitive
homologous protein paNHE1.
Thus, these data validates the threaded model structure, but further studies must be undertaken to
further elucidate the structure-function relations of NHE1.
84
REFERENCES
1. Brett, C. L., Donowitz, M., and Rao, R. (2005) Am. J. Physiol Cell Physiol 288, C223-C239 2. Roos, A. and Boron, W. F. (1981) Physiol Rev. 61, 296-434 3. Boron, W. F. (2004) Adv. Physiol Educ. 28, 160-179 4. Boron, W. F. (1978) Respir. Physiol 33, 59-62 5. Chesler, M. and Nicholson, C. (1985) Brain Res. 325, 313-316 6. Chesler, M. (1990) Prog. Neurobiol. 34, 401-427 7. Chesler, M. (2003) Physiol Rev. 83, 1183-1221 8. Nelson, N. and Harvey, W. R. (1999) Physiol Rev. 79, 361-385 9. Shin, J. M., Munson, K., Vagin, O., and Sachs, G. (2009) Pflugers Arch. 457, 609-622 10. Wakabayashi, S., Shigekawa, M., and Pouyssegur, J. (1997) Physiol Rev. 77, 51-74 11. Orlowski, J. and Grinstein, S. (1997) J. Biol. Chem. 272, 22373-22376 12. Pedersen, S. F., O'Donnell, M. E., Anderson, S. E., and Cala, P. M. (2006) Am. J. Physiol Regul.
Integr. Comp Physiol 291, R1-25 13. Boron, W. F. (1992) The Kidney: Physiology and Pathophysiology, 2nd Ed., Raven, New York 14. Putney, L. K., Denker, S. P., and Barber, D. L. (2002) Annu. Rev. Pharmacol. Toxicol. 42, 527-552 15. Zhang, J., Bobulescu, I. A., Goyal, S., Aronson, P. S., Baum, M. G., and Moe, O. W. (2007) Am. J.
Physiol Renal Physiol 293, F761-F766 16. Pedersen, S. F., King, S. A., Rigor, R. R., Zhuang, Z., Warren, J. M., and Cala, P. M. (2003) Am. J.
Physiol Cell Physiol 284, C1561-C1576 17. Orlowski, J. and Grinstein, S. (2004) Pflugers Arch. 447, 549-565 18. Slepkov, E. and Fliegel, L. (2002) Biochem. Cell Biol. 80, 499-508 19. Fliegel, L. (2005) Int. J. Biochem. Cell Biol. 37, 33-37 20. Denker, S. P., Huang, D. C., Orlowski, J., Furthmayr, H., and Barber, D. L. (2000) Mol. Cell 6, 1425-
1436 21. Denker, S. P. and Barber, D. L. (2002) J. Cell Biol. 159, 1087-1096 22. Nakamura, T. Y., Iwata, Y., Arai, Y., Komamura, K., and Wakabayashi, S. (2008) Circ. Res. 103,
891-899 23. Grinstein, S., Rotin, D., and Mason, M. J. (1989) Biochim. Biophys. Acta 988, 73-97 24. Pedersen, S. F. (2006) Pflugers Arch. 452, 249-259 25. Fafournoux, P., Noel, J., and Pouyssegur, J. (1994) J. Biol. Chem. 269, 2589-2596 26. Hisamitsu, T., Pang, T., Shigekawa, M., and Wakabayashi, S. (2004) Biochemistry 43, 11135-
11143 27. Hisamitsu, T., Ben, A. Y., Nakamura, T. Y., and Wakabayashi, S. (2006) Biochemistry 45, 13346-
13355 28. Hilger, D., Polyhach, Y., Padan, E., Jung, H., and Jeschke, G. (2007) Biophys. J. biophysj 29. Wakabayashi, S., Fafournoux, P., Sardet, C., and Pouyssegur, J. (1992) Proc. Natl. Acad. Sci. U. S.
A 89, 2424-2428 30. Slepkov, E. R., Rainey, J. K., Sykes, B. D., and Fliegel, L. (2007) Biochem. J. 401, 623-633 31. Counillon, L., Pouyssegur, J., and Reithmeier, R. A. (1994) Biochemistry 33, 10463-10469 32. Aronson, P. S. (1985) Annu. Rev. Physiol 47, 545-560 33. Wakabayashi, S., Hisamitsu, T., Pang, T., and Shigekawa, M. (2003) J. Biol. Chem. 278, 43580-
43585 34. Aronson, P. S., Nee, J., and Suhm, M. A. (1982) Nature 299, 161-163
85
35. Levine, S. A., Montrose, M. H., Tse, C. M., and Donowitz, M. (1993) J. Biol. Chem. 268, 25527-25535
36. Wakabayashi, S., Hisamitsu, T., Pang, T., and Shigekawa, M. (2003) J. Biol. Chem. 278, 11828-11835
37. Hisamitsu, T., Yamada, K., Nakamura, T. Y., and Wakabayashi, S. (2007) FEBS J. 274, 4326-4335 38. Wakabayashi, S., Bertrand, B., Ikeda, T., Pouyssegur, J., and Shigekawa, M. (1994) J. Biol. Chem.
269, 13710-13715 39. Wakabayashi, S., Ikeda, T., Iwamoto, T., Pouyssegur, J., and Shigekawa, M. (1997) Biochemistry
36, 12854-12861 40. Ikeda, T., Schmitt, B., Pouyssegur, J., Wakabayashi, S., and Shigekawa, M. (1997) J. Biochem. 121,
295-303 41. Hunte, C., Screpanti, E., Venturi, M., Rimon, A., Padan, E., and Michel, H. (2005) Nature 435,
1197-1202 42. Bianchini, L., Kapus, A., Lukacs, G., Wasan, S., Wakabayashi, S., Pouyssegur, J., Yu, F. H.,
Orlowski, J., and Grinstein, S. (1995) Am. J. Physiol 269, C998-1007 43. Grinstein, S., Woodside, M., Sardet, C., Pouyssegur, J., and Rotin, D. (1992) J. Biol. Chem. 267,
23823-23828 44. Holt, M. E., King, S. A., Cala, P. M., and Pedersen, S. F. (2006) Cell Biochem. Biophys. 45, 1-18 45. Takahashi, E., Abe, J., Gallis, B., Aebersold, R., Spring, D. J., Krebs, E. G., and Berk, B. C. (1999) J.
Biol. Chem. 274, 20206-20214 46. Lehoux, S., Abe, J., Florian, J. A., and Berk, B. C. (2001) J. Biol. Chem. 276, 15794-15800 47. Tominaga, T. and Barber, D. L. (1998) Mol. Biol. Cell 9, 2287-2303 48. Tominaga, T., Ishizaki, T., Narumiya, S., and Barber, D. L. (1998) EMBO J. 17, 4712-4722 49. Meima, M. E., Webb, B. A., Witkowska, H. E., and Barber, D. L. (2009) J. Biol. Chem. 284, 26666-
26675 50. Yan, W., Nehrke, K., Choi, J., and Barber, D. L. (2001) J. Biol. Chem. 276, 31349-31356 51. Fliegel, L., Walsh, M. P., Singh, D., Wong, C., and Barr, A. (1992) Biochem. J. 282 ( Pt 1), 139-145 52. Khaled, A. R., Moor, A. N., Li, A., Kim, K., Ferris, D. K., Muegge, K., Fisher, R. J., Fliegel, L., and
Durum, S. K. (2001) Mol. Cell Biol. 21, 7545-7557 53. Sardet, C., Fafournoux, P., and Pouyssegur, J. (1991) J. Biol. Chem. 266, 19166-19171 54. Misik, A. J., Perreault, K., Holmes, C. F., and Fliegel, L. (2005) Biochemistry 44, 5842-5852 55. Bertrand, B., Wakabayashi, S., Ikeda, T., Pouyssegur, J., and Shigekawa, M. (1994) J. Biol. Chem.
269, 13703-13709 56. Pang, T., Su, X., Wakabayashi, S., and Shigekawa, M. (2001) J. Biol. Chem. 276, 17367-17372 57. Lin, X. and Barber, D. L. (1996) Proc. Natl. Acad. Sci. U. S. A 93, 12631-12636 58. Pang, T., Wakabayashi, S., and Shigekawa, M. (2002) J. Biol. Chem. 277, 43771-43777 59. Pang, T., Hisamitsu, T., Mori, H., Shigekawa, M., and Wakabayashi, S. (2004) Biochemistry 43,
3628-3636 60. Zaun, H. C., Shrier, A., and Orlowski, J. (2008) J. Biol. Chem. 283, 12456-12467 61. Li, X., Alvarez, B., Casey, J. R., Reithmeier, R. A., and Fliegel, L. (2002) J. Biol. Chem. 277, 36085-
36091 62. Li, X., Liu, Y., Alvarez, B. V., Casey, J. R., and Fliegel, L. (2006) Biochemistry 45, 2414-2424 63. Aharonovitz, O., Zaun, H. C., Balla, T., York, J. D., Orlowski, J., and Grinstein, S. (2000) J. Cell Biol.
150, 213-224 64. Meima, M. E., Mackley, J. R., and Barber, D. L. (2007) Curr. Opin. Nephrol. Hypertens. 16, 365-
372 65. Putney, L. K. and Barber, D. L. (2003) J. Biol. Chem. 278, 44645-44649 66. Wang, H., Singh, D., and Fliegel, L. (1997) J. Biol. Chem. 272, 26545-26549
86
67. Vexler, Z. S., Symons, M., and Barber, D. L. (1996) J. Biol. Chem. 271, 22281-22284 68. Cardone, R. A., Casavola, V., and Reshkin, S. J. (2005) Nat. Rev. Cancer 5, 786-795 69. Allen, D. G. and Xiao, X. H. (2003) Cardiovasc. Res. 57, 934-941 70. Pedersen, S. F., Darborg, B. V., Rasmussen, M., Nylandsted, J., and Hoffmann, E. K. (2007) Cell
Physiol Biochem. 20, 735-750 71. Chen, L., Chen, C. X., Gan, X. T., Beier, N., Scholz, W., and Karmazyn, M. (2004) Am. J. Physiol
Heart Circ. Physiol 286, H381-H387 72. Murphy, E. and Allen, D. G. (2009) J. Mol. Cell Cardiol. 46, 137-141 73. Rotin, D., Steele-Norwood, D., Grinstein, S., and Tannock, I. (1989) Cancer Res. 49, 205-211 74. Reshkin, S. J., Bellizzi, A., Caldeira, S., Albarani, V., Malanchi, I., Poignee, M., Alunni-Fabbroni, M.,
Casavola, V., and Tommasino, M. (2000) FASEB J. 14, 2185-2197 75. Cardone, R. A., Bagorda, A., Bellizzi, A., Busco, G., Guerra, L., Paradiso, A., Casavola, V., Zaccolo,
M., and Reshkin, S. J. (2005) Mol. Biol. Cell 16, 3117-3127 76. Bourguignon, L. Y., Singleton, P. A., Diedrich, F., Stern, R., and Gilad, E. (2004) J. Biol. Chem. 279,
26991-27007 77. Bourguignon, L. Y. (2008) Semin. Cancer Biol. 18, 251-259 78. Grinstein, S., Woodside, M., Waddell, T. K., Downey, G. P., Orlowski, J., Pouyssegur, J., Wong, D.
C., and Foskett, J. K. (1993) EMBO J. 12, 5209-5218 79. McLean, L. A., Roscoe, J., Jorgensen, N. K., Gorin, F. A., and Cala, P. M. (2000) Am. J. Physiol Cell
Physiol 278, C676-C688 80. Lagana, A., Vadnais, J., Le, P. U., Nguyen, T. N., Laprade, R., Nabi, I. R., and Noel, J. (2000) J. Cell
Sci. 113 ( Pt 20), 3649-3662 81. McLean, L. A., Zia, S., Gorin, F. A., and Cala, P. M. (1999) Am. J. Physiol 276, C1025-C1037 82. Pedersen, S. F., King, S. A., Nygaard, E. B., Rigor, R. R., and Cala, P. M. (2007) J. Biol. Chem. 282,
19716-19727 83. Fleming, K., Kelley, L. A., Islam, S. A., MacCallum, R. M., Muller, A., Pazos, F., and Sternberg, M. J.
(2006) Philos. Trans. R. Soc. Lond B Biol. Sci. 361, 441-451 84. Shrode, L. D., Gan, B. S., D'Souza, S. J., Orlowski, J., and Grinstein, S. (1998) Am. J. Physiol 275,
C431-C439 85. Wakabayashi, S., Pang, T., Su, X., and Shigekawa, M. (2000) J. Biol. Chem. 275, 7942-7949 86. Khadilkar, A., Iannuzzi, P., and Orlowski, J. (2001) J. Biol. Chem. 276, 43792-43800 87. Counillon, L., Franchi, A., and Pouyssegur, J. (1993) Proc. Natl. Acad. Sci. U. S. A 90, 4508-4512 88. Slepkov, E. R., Chow, S., Lemieux, M. J., and Fliegel, L. (2004) Biochem. J. 379, 31-38 89. Touret, N., Poujeol, P., and Counillon, L. (2001) Biochemistry 40, 5095-5101 90. Counillon, L., Noel, J., Reithmeier, R. A., and Pouyssegur, J. (1997) Biochemistry 36, 2951-2959 91. Murtazina, R., Booth, B. J., Bullis, B. L., Singh, D. N., and Fliegel, L. (2001) Eur. J. Biochem. 268,
4674-4685 92. Wang, D., Balkovetz, D. F., and Warnock, D. G. (1995) Am. J. Physiol 269, C392-C402 93. Noel, J., Germain, D., and Vadnais, J. (2003) Biochemistry 42, 15361-15368 94. Wakabayashi, S., Pang, T., Su, X., and Shigekawa, M. (2000) FEBS Lett. 487, 257-261 95. Lee, B. L., Li, X., Liu, Y., Sykes, B. D., and Fliegel, L. (2009) J. Biol. Chem. 284, 11546-11556 96. Reddy, T., Ding, J., Li, X., Sykes, B. D., Rainey, J. K., and Fliegel, L. (2008) J. Biol. Chem. 283,
22018-22030 97. Slepkov, E. R., Rainey, J. K., Li, X., Liu, Y., Cheng, F. J., Lindhout, D. A., Sykes, B. D., and Fliegel, L.
(2005) J. Biol. Chem. 280, 17863-17872 98. Wlodawer, A., Minor, W., Dauter, Z., and Jaskolski, M. (2008) FEBS J. 275, 1-21 99. Acharya, K. R. and Lloyd, M. D. (2005) Trends Pharmacol. Sci. 26, 10-14 100. Screpanti, E., Padan, E., Rimon, A., Michel, H., and Hunte, C. (2006) J. Mol. Biol. 362, 192-202
87
101. Padan, E., Kozachkov, L., Herz, K., and Rimon, A. (2009) J. Exp. Biol. 212, 1593-1603 102. Williams, K. A. (2000) Nature 403, 112-115 103. Galili, L., Herz, K., Dym, O., and Padan, E. (2004) J. Biol. Chem. 279, 23104-23113 104. Screpanti, E. and Hunte, C. (2007) J. Struct. Biol. 159, 261-267 105. Dzafic, E., Klein, O., Screpanti, E., Hunte, C., and Mantele, W. (2009) Spectrochim. Acta A Mol.
Biomol. Spectrosc. 72, 102-109 106. Galili, L., Rothman, A., Kozachkov, L., Rimon, A., and Padan, E. (2002) Biochemistry 41, 609-617 107. Olkhova, E., Hunte, C., Screpanti, E., Padan, E., and Michel, H. (2006) Proc. Natl. Acad. Sci. U. S. A
103, 2629-2634 108. Olkhova, E., Padan, E., and Michel, H. (2007) Biophys. J. 92, 3784-3791 109. Olkhova, E., Kozachkov, L., Padan, E., and Michel, H. (2009) Proteins 76, 548-559 110. Kozachkov, L., Herz, K., and Padan, E. (2007) Biochemistry 46, 2419-2430 111. Arkin, I. T., Xu, H., Jensen, M. O., Arbely, E., Bennett, E. R., Bowers, K. J., Chow, E., Dror, R. O.,
Eastwood, M. P., Flitman-Tene, R., Gregersen, B. A., Klepeis, J. L., Kolossvary, I., Shan, Y., and Shaw, D. E. (2007) Science 317, 799-803
112. Padan, E. (2008) Trends Biochem. Sci. 33, 435-443 113. Tzubery, T., Rimon, A., and Padan, E. (2004) J. Biol. Chem. 279, 3265-3272 114. Appel, M., Hizlan, D., Vinothkumar, K. R., Ziegler, C., and Kuhlbrandt, W. (2009) J. Mol. Biol. 388,
659-672 115. Landau, M. (2007) 116. Ginalski, K., Elofsson, A., Fischer, D., and Rychlewski, L. (2003) Bioinformatics 19, 1015-1018 117. Fajer, P. G. (2000) Electron Spin Resonance Spectroscopy Labeling in Peptide and Protein
Analysis. In R.A.Meyers, editor. Encyclopedia of Analytical Chemistry, John Wiley & Sons, Ltd.,
118. Todd, A. P., Cong, J., Levinthal, F., Levinthal, C., and Hubbell, W. L. (1989) Proteins 6, 294-305 119. Voss, J., He, M. M., Hubbell, W. L., and Kaback, H. R. (1996) Biochemistry 35, 12915-12918 120. Voss, J., Hubbell, W. L., Hernandez-Borrell, J., and Kaback, H. R. (1997) Biochemistry 36, 15055-
15061 121. Langen, R., Isas, J. M., Hubbell, W. L., and Haigler, H. T. (1998) Proc. Natl. Acad. Sci. U. S. A 95,
14060-14065 122. Gareth R.Eaton and Sandra S.Eaton (2008) Measurement of interspin distances by EPR. In
Lawrence Berliner, editor. Biological Magnetic Resonance, 123. Fanucci, G. E. and Cafiso, D. S. (2006) Curr. Opin. Struct. Biol. 16, 644-653 124. Gareth R.Eaton and Sandra S.Eaton (2008) Measurement of interspin distances by EPR. In
B.C.Gilbert, editor. Electron Paramagnetic Resonance, RSC publishing, 125. Sandra Eaton and Gareth Eaton (2000) Distance measurements by CW and pulsed EPR. In
Berliner, L. J., editor. Biological Magnetic Resonance, Kluwer Academic/Plenum, 126. Hubbell, W. L., Mchaourab, H. S., Altenbach, C., and Lietzow, M. A. (1996) Structure. 4, 779-783 127. Hubbell, W. L., Cafiso, D. S., and Altenbach, C. (2000) Nat. Struct. Biol. 7, 735-739 128. Altenbach, C., Flitsch, S. L., Khorana, H. G., and Hubbell, W. L. (1989) Biochemistry 28, 7806-7812 129. Zhao, M., Zen, K. C., Hernandez-Borrell, J., Altenbach, C., Hubbell, W. L., and Kaback, H. R. (1999)
Biochemistry 38, 15970-15977 130. Hubbell, W. L., Cafiso, D. S., and Altenbach, C. (2000) Nat. Struct. Biol. 7, 735-739 131. Rabenstein, M. D. and Shin, Y. (1995) Proceedings of the National Academy of Sciences 92, 8239-
8243 132. Mchaourab, H. S., Lietzow, M. A., Hideg, K., and Hubbell, W. L. (1996) Biochemistry 35, 7692-
7704 133. Rotin, D. and Grinstein, S. (1989) Am. J. Physiol 257, C1158-C1165
88
134. Dissing and Gasbjerg (1993) Basic principles behind the application of fluorescent dyes for the study of cellular electrolyte concentrations membrane potential and microenvironment. Membrane Transport in Biology,
135. Boyer, M. J. and Hedley, D. W. (1994) Methods Cell Biol. 41, 135-148 136. Negulescu, P. A. and Machen, T. E. (1990) Methods Enzymol. 192, 38-81 137. Tsien, R. Y. (1989) Methods Cell Biol. 30, 127-156 138. le, M. M., Champeil, P., and Moller, J. V. (2000) Biochim. Biophys. Acta 1508, 86-111 139. Seddon, A. M., Curnow, P., and Booth, P. J. (2004) Biochim. Biophys. Acta 1666, 105-117 140. Froncisz, W. and Hyde, J. S. (1982) Journal of Magnetic Resonance (1969) 47, 515-521 141. Hubbell, W. L., Froncisz, W., and Hyde, J. S. (1987) Review of Scientific Instruments 58, 1879 142. Altenbach, C., Oh, K. J., Trabanino, R. J., Hideg, K., and Hubbell, W. L. (2001) Biochemistry 40,
15471-15482 143. Altenbach, C., Oh, K., Trabanino, R. J., Hideg, K., and Hubbell, W. L. (2001) Biochemistry 40,
15471-15482
89
APPENDIX 1 – Publications The following is a nearly finalized draft with expected submission to Journal of Biological Chemistry in early 2010.
STRUCTURAL MODELING AND ELECTRON PARAMAGNETIC RESONANCE SPECTROSCOPY
OF THE HUMAN Na+/H
+ EXCHANGER ISOFORM 1, NHE1
Eva B. Nygaard§, Jens O. Lagerstedt#,†, Gabriel Bjerre§, Kristian A. Poulsen§, Stine Meinild§, Bill Shi‡, Robert R. Rigor‡, Stine F.
Pedersen§, John C. Voss†, and Peter M. Cala‡,* §Department of Biology, University of Copenhagen, DK-2100 Copenhagen, Denmark, # Department of Experimental Medical
Science, Biomedical Center, Lund University, Sweden, and Departments of †Biochemistry & Molecular Medicine, and ‡Physiology & Membrane Biology, University of California, Davis, California 95616
Running head: Structural modeling and EPR spectroscopy of NHE1
Address correspondence to: Peter M. Cala, Dept. of Physiology & Membrane Biology, School of Medicine, University of
California Davis, One Shields Ave., Davis, CA 95616. 530-752-1285; Fax: 530-752-5423; E-mail: [email protected]
NHE1 plays central roles in regulation of cellular
volume and pH, and its dysfunction is implicated in
cancer development and ischemic cell injury. We
previously presented functional evidence that TM IV
and regions in TM X to TM XI are important for
inhibitor binding and ion transport by human (h)NHE1
(Pedersen, S.F. et al. (2007) J. Biol Chem. 282:19716-
19727). Here, we present a structural model of the
transmembrane part of hNHE1 that further supports this
conclusion. The hNHE1 model was based on the crystal
structure of the Escherichia coli NhaA and on previous
cysteine scanning accessibility studies of hNHE1, and
was validated by electron paramagnetic resonance
(EPR) spectroscopy of spin labels in TM IV and TM XI
as well as by functional analysis of hNHE1 mutants.
Removal of all endogenous cysteines in hNHE1,
introduction of the mutations Ala173
Cys (TM IV) and/or
Ile461
Cys (TM XI), and expression of the constructs in
mammalian cells, resulted in functional hNHE1
proteins. The distance between spin-labels in these
positions was determined at 15 ± 2 Å, confirming that
TM IV and TM XI are in close proximity in hNHE1.
This distance was decreased at pH 5.1 (NHE1
activation) and increased in the presence of the NHE1
inhibitor cariporide. A similar TM IV-TM XI distance,
and a similar change upon a shift in pH was found for
the cariporide-insensitive winter flounder NHE1
homolog paNHE1, however, in paNHE1, cariporide
had no effect on the TM IV-TM XI distance. The
central role of the TM IV-TM XI arrangement was
further confirmed by the partial loss of function upon
mutation of R425, which is predicted by the model to
stabilize this arrangement. Data are consistent with a
role for TM IV and TM XI rearrangements in ion
translocation and inhibitor binding by hNHE1.
The ubiquitous plasma membrane Na+/H
+
exchanger isoform 1 (NHE1) plays central roles in
cellular pH and volume homeostasis, cell migration,
proliferation, and survival, yet increased NHE1 activity
contributes to ischemia-reperfusion damage as well as
tumor growth and proliferation (1;2). Hence, the ability
to selectively block NHE1 is of high clinical relevance,
yet is hampered by a lack of a detailed understanding of
NHE1 structure and mechanism of ion translocation.
Hydropathy analyses suggest that NHE1 has 12
transmembrane (TM) segments, and a large C-terminal
cytoplasmic region (3). Cysteine accessibility studies
suggest the presence of two small reentrant loops
between TM IV and TM V (intracellular loop (IL) II)
and TM VIII and TM IX (IL IV), respectively and a
larger reentrant loop between TM IX and TM X
(extracellular loop (EL) V) (1;3). Portions of IL II and
IL IV are located within the membrane, suggesting that
they may form structures lining an aqueous pore
accessible from either side of the membrane, and hence
could be involved in ion translocation by NHE1 (3). EL
V is also interesting in this regard, as it resembles the P-
loops found in e.g. voltage gated ion channels (1;3).
These putative reentrant loops are highly conserved
among several NHE1 homologs, consistent with the
notion that they are critical for NHE1 function (3-5).
A number of regions within the NHE1 protein have
been implicated in inhibitor binding, e.g. TM IV and
TM IX (6-11), however, the mechanism(s) of
interaction between NHE1 and its commonly used
inhibitors, amiloride- and benzoyl guanidine (HOE-)
type compounds, remain to be fully elucidated.
Using a comparative approach based on chimeras
generated using human NHE1 (hNHE1) and two NHE1
homologs (flounder, paNHE1 and Amphiuma, atNHE1)
with high sequence homology to hNHE1, yet markedly
different inhibitor profiles (4;5), we previously obtained
novel information on the regions of NHE1 important
for inhibitor binding and ion transport (12). These
studies confirmed that TM IV plays a central role in
inhibitor binding (12) as suggested by earlier point
mutation studies (6-11). Moreover, we demonstrated for
90
the first time that regions in TM X to XI and/or IL V
and EL VI are important determinants of inhibitor
sensitivity.
The three-dimensional (3D) structure of NHE1 is
unknown, however the structure of the distantly related
bacterial (Escherichia coli) Na+/H
+ antiporter NhaA
was recently solved at 3.45 Å resolution (13). Similar to
NHE1, NhaA has 12 membrane spanning domains and
intracellular N- and C-termini (13). As with NHE1,
NhaA is important for cellular pH regulation and
electrolyte homeostasis (1). Despite the low sequence
homology (<15% similarity as calculated by ClustalW
analysis), and different transport stoichiometry,
comparison of NHE1 and NhaA is relevant because of
their similar topology and the fact that structure tends to
be far better conserved than sequence. A hNHE1
lacking N-glycosylation sites and expressed in
Saccharomyces cerevisiae, was recently used to create a
22 Å resolution structure which in its overall shape was
comparable with the crystal structure of NhaA (14).
However, as glycosylation of NHE1 has been shown to
be important for surface sorting of the transporter, a
process that may involve structural changes in the
protein (15), it is uncertain whether this structure is
fully representative of the mature NHE1.
The low sequence homology between NhaA and
NHE1 makes homology modeling highly challenging.
A structural model of hNHE1 based on threading on
NhaA has recently been published (16). This model was
constructed from multiple sequence alignments, fold
recognition, and evolutionary conservation analysis.
However, the assignment of TM regions in this model
is inconsistent with experimental evidence from earlier
cysteine scanning accessibility studies of hNHE1 (3),
and the model was not validated by experimental
measurements of inter-helix distances in hNHE1.
We therefore created a 3D structural model of the
N-terminal region of hNHE1 based on threading (17)
on the NhaA structure, in which we constrained our
alignment of TM domains to regions of NHE1 that
were experimentally determined to be in a membrane-
like environment. In the NhaA structure, and thus in our
model, TM IV and TM XI are in close proximity, in
agreement with our experimental evidence for hNHE1
(12). The next step was to test the hypothesis that these
helices are involved in ion translocation and inhibitor
binding by hNHE1. This was done by experimentally
determining the relative positions of TM IV and TM
XI, and their conformational changes during activation
and inhibition. For this purpose, site-directed spin
labeling (SDSL) was employed to create spin-labeled
hNHE1 proteins, which were subjected to electron
paramagnetic resonance (EPR) spectroscopy (18). In
this technique, cysteine residues introduced into the
protein at relevant positions, enables introduction of
cysteine-directed spin-labels (19). The EPR spectra
provide information on side chain dynamics (20), and
thus on protein topography, conformational changes,
secondary and tertiary structure (21;22). Introduction of
a second paramagnetic center allows distance
measurements within the protein (18;22;23).
Demonstrating the feasibility of this approach for
assessing NHE1 structure, a recently published study
employed EPR spectroscopic distance measurements
between spin labeled side chains on two NhaA
monomers to confirm NhaA dimerization (24).
Thus, we present here a 3D structural model of
hNHE1 threaded on the NhaA structure, in which TM
IV and TM XI are in close proximity. EPR analyses
confirmed the close proximity of TM IV and TM XI in
hNHE1 and determined the distance between TM IV
and TM XI spin-labels at about 15 Å. Furthermore, we
demonstrated that the distance between spin-labels in
TM IV and TM XI de- and increases, respectively,
during hNHE1 activation and inhibition, consistent with
a role for these regions in ion translocation and
inhibitor binding by hNHE1.
EXPERIMENTAL PROCEDURES
Materials Complete protease inhibitor was from Roche
Diagnostics. Amiloride (dissolved at 1 mM in ddH2O)
and Triton X-100 were from Sigma. Cariporide was a
kind gift from Sanofi-Aventis. 5’-(N-ethyl-N-isopropyl)
amiloride (EIPA) was obtained from Sigma-Aldrich
(St. Louis, MO). n-dodecyl ß-D-maltopyranoside
(DDM) was from Anatrace. (1-Oxyl-2,2,5,5-
tetramethylpyrroline-3-methyl) methanethiosulfonate
(MTS spin-label) was from Toronto Research
Chemicals. HiTrap Chelating Nickel column and CaM
agarose beads for affinity chromatography were from
GE Healthcare and Sigma, respectively. Unless
otherwise stated, other reagents were from Sigma or
Fisher.
Structural Modeling of the N-terminal Domain of
hNHE1
The structural homology model of the N-terminal
domain of NHE1 protein was built from its primary
sequence (residues 1-507) by use of several local
structure and fold recognition methods at the
MetaServer (bioinfo.pl/meta/) (25). The established
structure of the protein with the highest scores (NhaA;
PDB accession code 1ZCD), as evaluated by the 3D-
Jury method at the same server, was used as template
and the comparison was done with the Swiss-Model
program (26).
Threading of NHE1 on the NhaA template was
limited to the N-terminal domain of the protein
(residues 1 to 507), the portion of NHE1 that contains
the membrane-spanning domains. Structurally, the
integral membrane portion of NHE1 is distinguished
91
from that of NhaA by a much greater fraction of
hydrophilic (extra-membrane) sequence. Thus, nearly
all of the assigned structure in the model for NHE1
concerns the transmembrane regions of the protein. We
therefore constrained our alignment of NHE1 and
NhaA to sequences (and their flanking 10 residues) that
have been found experimentally (3) to reside in a
membrane-like environment. Alignments between the
known NhaA TM regions and the hNHE1 TM regions
(including the flanking 10 residues) suggested by
Wakabayashi et al. based on cysteine accessibility
analyses (3) were then carried out independently using
the ClustalW algorithm. The resultant TM alignments
were then used to match the regions of low homology
and ensure that gaps fell within the hydrophilic loops
connecting the TM segments.
Analysis of the NHE1 N-terminal domain structural
model was performed by use of DeepView/Swiss-
PdbViewer (www.expasy.org/
spdbv) and by use of Insight II software (version 2005)
on the Octane workstation by Silicon Graphics. The
final figures were produced by use of the UCSF
Chimera software (www.cgl.ucsf.edu/chimera).
Calculation of the distance between Ala173
and Ile461
based on our homology model and that of Landau et al
(16) was made using xx software. We are grateful to M.
Landau, Tel-Aviv University, Israel, for providing us
with the PDB file allowing us to calculate this value
also for their model.
Cell Culture
AP-1 cells (a mutant Chinese Hamster Ovary
(CHO)-derived cell line with no endogenous NHE
activity (27) were a kind gift from Dr. S. Grinstein,
Hospital for Sick Children, Toronto, Canada, and have
previously been shown to exhibit no recovery from an
acid load in the nominal absence of HCO3¯ (12;27). AP-
1 cells transfected with the hNHE1 constructs were
grown at 37°C, 5% CO2, 95% humidity in α-modified
Eagle’s medium supplemented with 10% fetal bovine
serum, 1% L-glutamine, 1% penicillin/streptomycin,
and 600 μg/ml G418 sulfate (Invitrogen). Every 3-4
days, cells were passaged by gentle trypsinization, and
only passages 5-30 were used for experiments.
Constructs and Stable Transfection of Mutant hNHE1
and paNHE1
The full length Cys-less hNHE1 was constructed by
replacing all native cysteine residues with alanines.
Using restriction digest, hNHE1 was cloned into the
mammalian expression vector pcDNA3.1(+)
(Invitrogen). The desired residues were altered using
site-directed mutagenesis (QuickChange XL,
Stratagene, CA). To facilitate affinity chromatography
purification of the constructs, a poly-His tag was added
to the C-terminal end by three-way ligation. This Cys-
less construct was found to be fully functional, in
agreement with earlier reports (3). From the Cys-less
construct, three different hNHE1 constructs containing
Cys replacements at position 173 and/or 461 were
prepared. All constructs were verified by DNA
sequencing prior to transfection (DBS Sequencing
Facility, University of California Davis). Positive
transfectants were selected for resistance to 600 µg/ml
G418. hNHE1 expression was verified by
immunoblotting as previously described (5). Briefly,
protein homogenates were separated on 7.5% SDS-
PAGE gels and electrotransferred to nitrocellulose
membranes. Mouse monoclonal NHE1 antibody 4e9
(Chemicon) and horseradish peroxidase-conjugated
goat anti-mouse IgG secondary antibody (Zymed) were
used to label NHE1, followed by visualization by
enhanced chemiluminescence (SuperSignal, Pierce).
Clonal selection was carried out by limiting dilution,
and stably transfected clones were used in all
experiments. The full-length Cys-less paNHE1 was
constructed in a similar manner, except that the residues
replaced with cysteines for spin labeling were G165 in
TM4 and G456 in TM11.
Functional analysis of NHE1 mutants expressed in AP1
cells
All constructs were tested for NHE1 function,
employing the fluorescent, pH-sensitive probe 2´,7´-bis-
(2-carboxyethyl)-5,6-carboxyfluorescein, tetra-
acetoxymethylester (BCECF-AM) to monitor
intracellular pH (pHi) and the ammonium pre-pulse
technique to acid-load the cells, as previously described
in detail (12).
Immunofluorescence analysis of wild type and mutant
NHE1 expressed in AP1 cells
Immunofluorescence labeling of NHE1 in AP1 cells
was carried out essentially as described previously
{Rasmussen, 2008 2660 /id}. Briefly, cells grown on
glass coverslips were washed in isotonic Ringer (in
mM: 130 NaCl, 3 KCl, 20 Hepes, 1 MgCl2, 0.5 CaCl2,
10 NaOH, 10 glucose, pH 7.4), fixed in 2%
paraformaldehyde, washed in TBS (in mM: 150 NaCl,
10 Tris-HCl, 1 MgCl2, 1 EGTA), permeabilized (0.2%
triton X-100 in TBS), blocked in 5% BSA in TBST,
incubated with primary antibody against NHE1 (Xb-17,
1:100 in TBST), washed in TBST, incubated with
Alexa488-conjugated anti-mouse secondary antibody
(1:600 in TBST, 1 h), washed in TBST and mounted.
Fluorescence was visualized using a 100 X/1.4 NA plan
apochromat objective, pinhole size 1 airy disc, and the
488 nm laser line of a Leica DM IRB/E microscope and
Leica TSC NT confocal laser scanning unit (Leica
Lasertechnik GmbH, Heidelberg, Germany). Images
shown are frame averaged and presented in RGB
92
pseudocolor. No or negligible labeling was detectable
in the absence of primary antibody.
Expression and functional analysis of R425A NHE1 in
Xenopus oocytes
Expression and functional analysis of
hNHE1R425
A was carried out essentially as in {Meinild,
2009 3160 /id}. Briefly,the human R425A NHE1 was
cloned into a vector optimized for oocyte expression
(pDEST SML). The cDNA was linearized and in vitro
transcribed with T7 RNA polymerase using the T7
mMessage mMashine Kit (cat. # AM1344, Ambion). 50
ng of cRNA was injected into defoliculated X. laevis
oocytes, which were incubated in Kulori medium (90
mM NaCl, 1 mM KCl, 1 mM CaCl2, 1 mM MgCl2, 5
mM HEPES, pH 7,4) at 19 °C for 3–7 days before
experiments were performed. The two-electrode voltage
clamp method was used to control the membrane
potential and monitor the
current in oocytes expressing R425A-hNHE1.
Generally, the membrane potential (Vm) of the oocyte
was held at -50 mV and the experimental chamber was
continuously perfused by a NaCl buffer containing 100
mM NaCl, 2 mM KCl, 1 mM CaCl2, 1 mM MgCl2,
10mM HEPES, pH 7.4. In the Li+-induced leak current
experiments equimolar Li + replaced Na
+ (LiCl buffer)
and in Na+- and Li
+-free experiments, Na
+ was replaced
by equimolar methyl-D-glucamine (NMDG)(NMDG-
Cl buffer). Two-electrode voltage clamp recordings
were performed at room temperature with a Dagan
clampator interfaced to an IBM compatible PC using a
DigiData 1320 A/D converter and pCLAMP 9.0 (Axon
Instruments). Currents were low pass-filtered at 500 Hz
and sampled at 2 kHz. Electrodes were pulled from
borosilicate glass capillaries to a resistance of 0.5–2
megaohm and were filled with 1 M KCl. Preparation of Membrane Fractions
Unless otherwise indicated, subsequent procedures
were carried out at room temperature. Membrane
fractions were prepared fresh just prior to use. At 95%
confluence, cells were washed in PBS buffer (60 mM
K2HPO4, 30 mM KH2PO4, 145 mM NaCl, pH 7.5)
containing protease inhibitor (Complete protease
inhibitor w/EDTA, Roche Diagnostics), and placed in
liquid nitrogen to induce cell lysis. Next, the lysate was
placed on ice, 4 ml of PBS buffer with EDTA-
containing protease inhibitor cocktail was added, and
the lysate spun at 21,000 × g (1 min). Pellet was
resuspended in PBS buffer containing EDTA-free
protease inhibitor cocktail, 25 U/ l Benzonase
(Novagen) and 2 mM MgCl2, and centrifuged at 21,000
× g (1 min). The pellet was then resuspended in PBS
buffer containing EDTA-free protease inhibitor cocktail
and 0.5% Triton X-100 followed by incubation on ice
for 20 min. Finally, the resuspension was centrifuged at
21,000 × g (10 min, 4 °C), and the supernatant, now
containing solubilized membrane proteins, was sterile
filtered prior to purification and labeling.
Site-directed Spin-labeling and Protein Purification
To obtain highly purified hNHE1, two sequential
purification steps were employed. First, a HiTrap
Chelating Nickel column was charged with Ni2+
according to the manufacturer’s protocol (GE
Healthcare) and equilibrated with PBS buffer (pH 7.5)
containing 5 mM imidazole. The membrane protein
extract was applied onto the column at 1.0 ml/min. For
solubilization of NHE1, all imidazole buffers contained
0.01% DDM. The column was subsequently washed
with 6 volumes of PBS buffer containing 5 mM
imidazole followed by spin-labeling by applying 2
volumes of the same buffer containing MTS spin-label
(0.1 mM final concentration) to the column. The use of
MTS-SL labels for EPR measurements has been
extensively documented; for a discussion of potential
effects of the spin labels on structure, see Discussion.
After 30-min incubation in the dark, the column was
extensively washed with PBS buffer containing 5 mM
imidazole, followed by elution of hNHE1 with 150 mM
imidazole in PBS buffer.
For the second purification step, 500 µl of CaM-
agarose affinity resin was pre-equilibrated by rocking
for 10 min in 10 volumes of CaM binding buffer (PBS
buffer, 0.01% DDM and 1 mM CaCl2). The eluted
sample was incubated with the resin in the presence of
1 mM CaCl2 for 10 min. The resin was pelleted by
gentle centrifugation (200 × g, 1 min), washed in 10
volumes of CaM binding buffer, pelleted (200 × g, 1
min), rocked in 4 volumes of CaM elution buffer (PBS
buffer, 0.01% DDM and 5 mM EDTA) for 10 min, and
pelleted (200 × g, 1 min). The supernatant was saved,
the last step repeated, and the two elutions pooled. The
eluate was filtered through a 0.45 m cellulose acetate
Spin-X centrifugal filter (COSTAR) (2300 × g), and
concentrated to < 30 l (from ~1ml), using a 20 kD
Vivaspin 2 concentrator (Vivascience) (3000 × g).
Analysis of Recombinant hNHE1 protein level
The presence of hNHE1 in each purified protein
sample was confirmed by immunoblotting as described
above, and the purity was assessed by silver staining
according to the manufacturer’s (BioRad) protocol.
Briefly, purified protein samples were separated on
7.5% SDS-PAGE gels, which were fixed for 30 min,
oxidized for 5 min, and thoroughly rinsed in ddH2O.
Subsequently, the gels were incubated in silver reagent
for 20 min, rinsed quickly in ddH2O, and covered in
developer until visible bands occurred (data not shown).
Functional analysis of reconstituted hNHE1
Reconstitution of NHE1 protein. Purified poly-histidine
tagged NHE1 protein was reconstituted into liposomes
93
formed of E. coli polar lipids (Avanti Polar Lipids, Inc.,
Alabaster, AL) essentially as in {Nimigean, 2006 3152
/id}{Accardi, 2004 3153 /id} Briefly liposomes to be
assayed for either 22
Na+ uptake or H
+ flux were mixed
in a protein:lipid ratio of 1 µg:mg or 5 µg:mg,
respectively. Lipids were prepared by nitrogen
streaming to dryness, washed in pentane, and
reconstituted in appropriate intravesicular buffers (for 22
Na+ uptake: 400 mM NaCl, 10 mM MES at pH 6.0;
for H+ flux: 300 mM KCl, 100 mM citrate, 40 mM
KH2PO4, pH 4.0). 34 mM CHAPS was added to to
uniform opacity under sonication, followed by a 1 h
preincubation before adding NHE1 protein. Vesicles
were gel filtered with sephadex beads saturated with
intravesicular buffer, and eluted with the same buffer.
Vesicle-containing fractions were snap-frozen in liquid
nitrogen and stored at -80°C in 110 µl aliquots. 22
Na+ uptake or H
+ flux in reconstituted liposomes. For
22Na
+ uptake assays, liposomes were centrifuged at
low-speed through extravesicular medium (EVM1: 10
mM MES, 800 mM sucrose/sorbitol, pH 6.0) saturated
sephadex beads as in {Nimigean, 2006 3152 /id},
suspended in 100 µl EVM, and immediately combined
with 700 µl EVM1 containing 22
Na+ (0.5 µCi/ml). 100
µl aliquots were sampled by washing through DOWEX
cation exchange resin with 15 volumes of cold EVM1.
Sample radioactivitity was quantified by Cerenkov
counting. Valinomycin permeabilization was used to
obtain isotope saturation of the liposomes, and empty
liposomes were used to obtain non-specific 22
Na+
uptake (subtracted as background). For H+ flux assays,
liposomes were washed in a similar fashion with
sephadex (EVM2: 0.3 mM HEPES, 400 mM KCl, pH
7.4) and 100 µl removed to a small glass chamber with
a micro-stir bar. A pre-calibrated pH microelectrode
(Microelectrodes, Inc., Bedford, NH) was placed into
the solution, under continuous sampling to a PowerLab
data acquisition system (ADInstruments, Inc., Colorado
Springs, CO), and the solution was adjusted to 7.4. 10
µl poorly buffered NaCl solution (4M NaCl, 0.3 mM
HEPES) was added to initiate H+ flux. H
+ flux was
determined by conversion and correction for the
measured buffer capacity of EVM2. In both assays,
liposomes treated with inhibitor were suspended in
EVM in presence of 50 µM EIPA.
EPR Spectroscopy
EPR measurements were performed using a JEOL
X-band spectrometer fitted with a loop-gap resonator
(28;29). A 6 µl aliquot of the purified, spin-labeled
protein, at a final concentration of approximately 10
μM (1.0 µg/µl) in PBS buffer (pH 7.5) containing
0.01% DDM, was placed in a sealed quartz capillary
contained in the resonator. Spectra (averages of three 2-
min scans) were acquired at room temperature (22oC) at
a microwave power of 4 mW and with the amplitude
optimized to the natural line width of the individual
spectrum. The obtained spectra were double integrated
and normalized. The identically treated Cys-less protein
that was used as a negative control showed no signal at
these settings (not shown).
Interspin Distance Determinations
The distance between the two spin-labeled side
chains was determined from the spectral broadening of
the double-labeled sample, compared to the composite
spectrum from the two corresponding single-labeled
samples. First, the spectra for the double-labeled TM
IV/TM XI (Ala173
Cys/Ile461
Cys) construct were
normalized after a baseline adjustment, followed by
integration using the Basephase program (kindly
provided by C. Altenbach, Dept. of Ophtalmology at
the Jules Stein Eye Institute, UCLA, LA, California).
For the integration, the best possible baseline was
obtained and the spectrum was double integrated. After
the double integration, the total area was equalized for
the two single-labeled constructs, TM IV (Ala173
Cys)
and TM XI (Ile461
Cys), respectively, which thus
normalized them to the same number of spins. The
spectra of the two single-labeled constructs were
summed, and compared to the spectrum for the double-
labeled construct, using the same process as described
above. The normalized spectra for the double-labeled
construct and the composite spectrum for the single-
labeled constructs were analyzed for dipolar broadening
using Fourier deconvolution to calculate an ensemble of
Pake splittings that can best account for the degree of
broadening and, the interspin distance was calculated
from the Pake splittings as described previously
(23;30).
RESULTS
Structural Model of hNHE1
In order to create the tertiary structural model of the
transmembrane N-terminal domain of hNHE1 we used
in silico homology structure building, using the known
structure of NhaA as a template. The primary structure
of the N-terminal domain (residues 1 to 507) of hNHE1
was analyzed for structural homologs at the MetaServer
(bioinfo.pl/meta/) (25). Despite low primary sequence
identity, substantial structural similarities between
hNHE1 and NhaA (PDB accession code 1zcd) were
identified.
Based on previously published experimental
findings (3) and sequence analysis, the primary
sequence alignment was manually optimized prior to
threading the hNHE1 sequence onto the NhaA tertiary
structure (Fig. 1). The resulting structural homology
model of hNHE1 (Fig. 2) is limited to the N-terminal
(largely membrane spanning) domain and encompasses
amino acid residues Pro12
to Ala507
. In accordance with
the general finding that prokaryotic membrane proteins
94
possess shorter extramembrane loops and terminal
extensions than eukaryotic members of the same
superfamily (see 31), NhaA is seen to have smaller
extra-membranous loops than NHE1 (Fig. 1).
Consequently, several of the intra- and extra-cellular
loops connecting the transmembrane regions were
excluded in the presented model of hNHE1 (see also
Experimental Procedures). In the NhaA structure, and
thus in our model, TM IV and TM XI are closely
adjacent (Fig. 2 and close-up view in Fig. 4). Closer
examination of the threaded model allowed us to
identify amino acid residues that are putatively
important elements of the catalytical core of hNHE1
(Figs. 3 and 4). From the solid surface representation of
the hNHE1 model shown in Fig. 3A-B it is clearly seen
that several charged and polar residues with possible
roles in ion translocation are located near this cavity.
Importantly, in this model, Arg425
(corresponding to
Lys300
in NhaA, which is assigned a central role in the
catalytical core of NhaA (13), is accessible from the
cytoplasmic side, despite being positioned
approximately in the central plane of the lipid bilayer.
Fig. 3C shows a side view of the model, illustrating that
primarily hydrophobic side chains are pointed into the
interior of the lipid bilayer.
Functional analysis of Cys-replaced mutants and
reconstituted NHE1 protein The residues Ala
173 and Ile
461 were chosen for
cysteine replacements for site-directed spin labeling,
because of their location in close proximity to residues
in TM IV and TM XI, respectively, which have been
assigned important roles in the ion translocation
pathway (Fig. 4). In addition, the conservation of these
residues between NHEs is low, and replacement was
therefore not likely to interfere with the function of the
protein.
To ascertain that the introduction of cysteines in the
above-mentioned positions had not compromised
NHE1 function, which could render inter-helix distance
measurements unreliable, the function of each construct
was we tested the function of each construct after
expression in AP-1 cells, by monitoring pHi recovery
after acidification induced by a NH4Cl prepulse. As
seen in Fig. 5A, the three Cys-replaced constructs
(Ala173
Cys, Ile461
Cys, and Ala173
Cys/Ile461
Cys) were all
found to be functional. It may be noted that this
contrasts with a previous study in which an Ala173
Cys
mutation was found to reduce NHE1 function
{Slepkov, 2005 3154 /id}. The reason for this
discrepancy is not clear, however, it may be noted that
essentially all Cys mutants studied by Slepkov and
coworkers exhibited strongly reduced function,
including L163 and G174, comparable mutations of
which were reported by others to be fully functional
{Counillon, 1997 183 /id}, A173 not tested). Similarly,
hNHE1 V160C/H473C and all the paNHE1 constructs
were fully functional (not shown/5A).
Another concern was whether the purification and
reconstitution of NHE1 might affect function. We
therefore next monitored NHE1 function after
reconstitution in liposomes, i.e. a treatment and
environment similar to those of the EPR experiments.
As illustrated in Fig. 5B, H+ and Na
+ transport sensitive
to the NHE1 inhibitor EIPA was clearly seen after
NHE1 reconstitution. Thus, NHE1 was functional after
purification and reconstitution, indicating that structure
was preserved. A third possible caveat would be if spin
labeling significantly altered protein folding, stability or
behavior. However, this is highly unlikely to be the
case to any detectable extent (see Discussion).
EPR Spectroscopy Analyses Supports the Structural
Model of hNHE1
EPR line-shape of single-labeled hNHE1- As noted
above, to obtain a measure of the distance between two
spin-labels, EPR analyses of the non-interacting
individual spin-labeled proteins are necessary. Hence,
hNHE1 constructs with single-labeled TM IV
(Ala173
Cys) and single-labeled TM XI (Ile461
Cys),
respectively, were purified, labelled, and analysed by
EPR. Western blots and silver stains of the purified
NHE1 mutants are shown in Fig. 6A. Unless otherwise
stated, these analyses were performed at pH 7.5, i.e.
under conditions where NHE1 is expected to be in its
inactive conformation. Comparison of the EPR spectra
of the two single-labeled constructs reveals different
spectral shapes (Fig. 6B). The spectrum for the single-
labeled TM IV (Ala173
Cys) construct exhibits higher
amplitude and less broadening, compared to the single-
labeled TM XI construct (Ile461
Cys), which displays a
strongly immobilized component in the low field peak
(see arrow, Fig. 6B). This reflects that position 173
experiences greater motional freedom, possibly arising
from a more disordered backbone.
Distances measurements between the two spin-
labels on TM IV and TM XI, respectively- The EPR
spectra for the double-labeled TM IV/TM XI construct
containing spin-labels at residues 173 and 461 are
shown in Fig. 6C. In the absence of spin-spin
interaction in the double-labeled construct, the
spectrum would approximate the spectral sum of the
two corresponding single-labeled constructs. However,
when comparing the traces for the double-labeled
versus the sum of the single-labeled constructs, the
latter shows higher amplitude and less spectral
broadening (Fig. 6D), indicating a dipolar component in
the double-labeled sample. The calculated distance
between the spin-labels on TM IV and TM XI was
approximately 15 ± 2 Å (SD, n=3). The ± 2 Å variation
primarily reflects noise in the spectra for the single-
labeled TM IV construct.
95
Effect of the functional state of hNHE1 on the
distance between TM IV and TM XI- The double-
labeled construct was next studied to observe whether
activation and inhibition of the protein involved
positional changes in transmembrane domains TM IV
and TM XI based upon altered interaction between the
spin labels. Reducing the pH of the buffer to 5.1, which
is expected to activate hNHE1, induces a significant
broadening of the spectrum for the double-labeled
construct as seen by its decreased amplitude (green
trace, Fig. 6D). This indicates that the distance between
residues 173 and 461 of hNHE1 is reduced in the active
compared to the inactive conformation of the protein.
Conversely, addition of 1 mM of the NHE1 inhibitor
amiloride to the purified protein sample moderately
increased the mobility of the spin-labels, as seen from
the increased amplitude and reduced broadening of the
spectrum (red trace, Fig. 6D). This indicates that the
distance between residues 173 and 461 is increased in
the presence of amiloride.
Effect of a neutralizing mutation in the putative screen
residue Arg425
As eluded to above, the NhaA model suggests that
Lys300
plays a central role as a “screen residue” that
allows the energetically unfavorable dipole-dipole
arrangement of the catalytical core of NhaA (13). The
corresponding residue in hNHE1 is Arg425, which in
the homology model is positioned approximately in the
central plane of the lipid bilayer, yet accessible from the
cytoplasmic side (Fig. 3). We therefore hypothesized
that a neutralizing mutation of Arg425
would destabilize
the catalytical core and strongly affect NHE1 function.
Introduction of an Arg425
Ala mutation indeed reduced
NHE1-dependent pHi recovery after an acid load (Fig.
7A). Confocal imaging verified that this at least in part
reflected a strongly reduced targeting of the mutant
NHE1 to the plasma membrane (Fig. 7B). One possible
effect of the Arg425
Ala mutation that would not have
been picked up in the pHi recovery measurements
would have been to switch the ion transport mode of
NHE1 from one of Na+/H
+ exchange to conductive Na
+
transport. To test this possibility, we expressed the
Arg425
Ala mutant in Xenopus oocytes, allowing us to
monitor for Na+ currents (Fig. 7C). However, as seen,
introduction of the mutant NHE1 did not give rise to
Na+ currents, neither under unstimulated conditions, nor
after acidification to activate NHE1. Thus, consistent
with the proposed TM4/TM11 arrangement, Arg425
appears to serve a crucial function in stabilizing NHE1
structure.
DISCUSSION
Based on the crystal structure of NhaA (13) and
experimental evidence from cysteine accessibility
studies (3), we created a threaded structural model of
hNHE1. The TM IV/TM XI complex of NhaA, by
virtue of its unlikely architecture (charges and partial
charges in the membrane interior) and its apparent
conservation in NHE1 (see Fig. 4), suggests that it may
be the catalytic core of the ion translocation pathway
(13). If this inference is correct, it is likely that NHE1
and NhaA mediate Na+ and H
+ translocation through a
very similar mechanism. The inferred structural model
of hNHE1 resembles that recently proposed by Landau
et al. (16) in many respects. The two models agree on
the location of several of the helical transmembrane
domains. In addition, the NhaA crystal structure
indicates a critical role for a basic side chain on TM X,
and both predictions identify Arg425
as the residue
occupying this position. To further compare the two
models, we calculated the distance between the ends of
the amino acid side chains of Ala173 and Ile461
obtained for each model. This value was very similar
for the two models, with a distance of 8.46 Å obtained
for our model, compared to 7.91 Å for the Landau
model. However, there are significant differences
between the TM assignments in the two models.
Because of the very low homology of NHE1 to NhaA,
we constrained our alignment of TM segments to
regions of NHE1 that were experimentally determined
(3) to be in a membrane-like environment, whereas
Landau et al. achieved their alignment solely from
homology-based predictions. For example, the first
TM segment in our model starts at residue 15. In
contrast, the N-terminal extension in the Landau model
is more than 100 residues longer and the first TM
segment instead starts at residue 128 (TM I of the
Landau model co-localizes with TM III of the structure
we propose). This arrangement, however, does not
appear to take into account that residues 126 and 127
have previously been shown to be inaccessible for
MTSET labeling even in permeabilized cells, strongly
indicating that they are embedded in the bilayer (3).
Consequently, the negatively charged, or polar, residues
facing the ion binding pocket are different in the two
models, with Asp172
and Thr197
(Fig. 4) being postulated
to be part of the critical residues for translocation in our
model, whereas Landau et al. (16) depict Asp238
and
Asp267
(on TM IV and TM V, respectively, in the
Landau model) to provide the negative charges in the
core of the translocation pathway. Furthermore, the
PX3D motif is highly conserved in all NHE proteins,
and accumulated hydropathy and experimental analyses
has placed this motif in TM segment IV (see e.g.
3;5;7;32). The NhaA structure highlights the
significance of this motif, as it facilitates a distinctive
extension and crossover of TM domains IV and XI. In
our alignment, this motif is assigned to TM IV in
agreement with previous experimental evidence (3;32)
whereas in Landau et al. (16) this motif falls within TM
II. Moreover, in the Landau model, residue 173 (located
96
in TM II in this model) is much farther apart from
residue 461 than the distance observed by EPR
(approximately 15 Å). Obviously, additional
experimental work is called for to refine these models.
Validation of the structural model and ion
translocation hypothesis by SDSL and EPR- Our
structural model of hNHE1 supports our recent
evidence that TM IV and TM XI play important roles in
inhibitor binding and ion translocation by NHE1 (12).
We further validated this notion, firstly by determining
the distance between TM IV and TM XI, and the
conformational changes in these domains during
activation and inhibition of hNHE1, by SDSL and EPR
spectroscopy. For this purpose we created two single-
labeled mutants, Ala173
Cys and Ile461
Cys, and the
double-labeled mutant, Ala173
Cys/Ile461
Cys. Both
residues face the interior of the protein (Fig. 2),
however, comparison of the spectra for the two single-
labeled constructs indicates that residue 173 is in a
more flexible environment than residue 461 (Figs. 4
and 6). This corresponds to the model structure, in
which residue 461 seems to be more enclosed in the
protein than residue 173. The distance between
positions 173 and 461 in TM IV and TM XI was
obtained by comparison of the spectra for the double-
and single-labeled constructs, and was determined to 15
± 2 Å, confirming that TM IV and TM XI are in close
proximity.
When pH was reduced to 5.1 to activate hNHE1, the
Ala173
Cys/Ile461
Cys spectrum revealed stronger
interaction of the spin-labels, indicating that in the
active conformation, the spin-labeled side-chains are
brought closer together. In contrast, in the presence of
the NHE1 inhibitor amiloride, the Ala173
Cys/Ile461
Cys
spectrum exhibited reduced spin-label interaction,
indicating that amiloride inhibition of hNHE1 causes a
conformational change resulting in increased distance
between the studied residues in TM IV and TM XI.
Again, this would be in congruence with our recent
findings, which point to a close correlation between
inhibitor sensitivity and ion transport, and a role for TM
IV and TM XI in both processes (12).
These findings correlate well with the NhaA
translocation mechanism proposed by Hunte et al. (13).
In the NhaA structure (and thus in our hNHE1 model),
the extended TM IV/TM XI helices cross over each
other and conformational changes in TM IV and TM XI
(as well as in TM V and TM IX) allow an alternating-
access mechanism of ion transport (Fig. 2 and 8). Thus,
one could very likely expect an equivalent mechanism
for ion translocation in hNHE1. For example, Asp133
of
NhaA aligns with Asp172
of hNHE1, Asp163
of NhaA
aligns with Thr197
of hNHE1, and Lys300
of NhaA aligns
with that of Arg425
of hNHE1 in our model (Fig. 4).
Asp163
of NhaA is suggested to act as a molecular
switch, such that its protonation state determines
whether the Na+ binding site (Asp
164) is accessible to
the periplasm or the cytoplasm (33). In hNHE1,
Thr197
could be expected to carry out the same function
as an accessibility-control site. In the NhaA structure,
the energetically unfavorable negative/negative and
positive/positive dipole-dipole pairings due to the
arrangement of the TM IV/TM XI helices are stabilized
due to electrostatic screening provided by the negative
Asp133
and the positive Lys300
(13) presumably by
facilitating the TM IV/TM XI crossover. These residues
are conserved among bacterial NhaA homologs and
have been shown to be essential to NhaA activity
(34;35). Thus, we hypothesized that the residues Asp172
(negative charge) and Arg425
(positive charge) in the
corresponding positions of hNHE1 would have a
similar ”screening” function, which makes this
crossover architecture of TM IV/TM XI possible (Fig.
8).
If the hypothesized arrangement of TM IV and TM
XI of hNHE1 is correct, a mutation in one of the
“screening” residues is expected to cause a non-
functional protein. The proposed screening residue
Arg425
is located in TM X, which we in our previous
studies implicated in ion transport and inhibitor binding
(12), and which has been assigned as central in ion
translocation by NhaA (35). Moreover, in our structural
model, Arg425
is located at the bottom of the open
cavity and is thus directly accessible from the
cytoplasmic side of the membrane (Fig 3). We therefore
hypothesized that replacing Arg425
with a neutral amino
acid would result in a non-functional NHE1. Indeed, an
Arg425
Ala mutation in hNHE1 resulted in marked
reduction in NHE1 plasma membrane targeting and pH
regulatory ion transport, indicating that NHE1 structure
was compromised. These data do not allow distinction
between whether only membrane targeting, or also the
function of the transporters in the plasma membrane,
were affected. However, they strongly indicate that
consistent with the hypothesis, Arg425
is central for the
structural stability of NHE1.
We thus propose an architecture for the TM IV – TM
XI complex in which the positive charge of Arg425
stabilizes the helices by screening the partial charges
from the dipoles on the extended helix c-terminal ends
and functions as a "check valve" permitting Na+
coordination and or gating (Fig. 8). As noted above, the
negatively charged Asp172
is also likely to contribute to
dipole charge screening.
Critical assessment of the validity of the EPR data
- Functional analysis of both Cys-replaced NHE1
mutants in AP-1 cells, and of purified and reconstituted
NHE1 protein strongly indicated that NHE1 function,
and hence structure, was not compromised in our EPR
analyses. Another obvious concern in SDSL-EPR
experiments is that engineering cysteine substitutions
and subsequent modification by the spin label may
97
significantly alter protein folding, stability or behavior.
However, substantial work has revealed that the
nitroxide ring is well tolerated in proteins and assumes
a limited number of rotamers, facilitating the modeling
of the spin label within the 3D structure {Langen, 2000
3155 /id}{Guo, 2007 3156 /id}{Fajer, 2007 3157 /id}.
The method has been applied to a wide assortment of
protein types, with very few examples showing a major
functional or structural consequence resulting from this
modification. Direct evidence for how the incorporated
nitroxide is accommodated in protein structures has
been obtained in high-resolution crystal structures of T4
lysozyme containing spin-labeled side chains. Even at
buried sites, no significant perturbation of the backbone
is evident {Langen, 2000 3155 /id}{Guo, 2007 3156
/id}. Finally, others have reported modest or negligible
effects of spin labeling on protein folding {DeWeerd,
2001 3158 /id} or backbone structure of peptides as
measured by NMR {Bolin, 1998 3159 /id}. This can be
attributed to the relatively compact size of the modified
Cys residue (a molecular volume on the order of Tyr)
and its ambivalent chemical nature, which does not
favor highly polar or nonpolar environments. Small (±
2 Å or less) distortions are possible, however, as
confidence within ± 5 Å is more than enough to map
the correct TM arrangement this should not cause
concern.
Possible mechanism of H+ sensing by NHE1- In
NhaA, TM IX is located at the entrance of the
cytoplasmic funnel, where it has been proposed to
function as a “pH-sensor” (13). This arrangement is
recapitulated in our hNHE1 model (Fig. 8). Moreover,
our preliminary SDSL and EPR findings indicate a
close proximity between TM IV and TM IX of hNHE1
(King, S. A., Voss, J., & Cala, P. M., unpublished),
which could indicate interaction between these helices.
Thus, the mechanism of pH-regulated ion translocation
proposed for NhaA could also be expected for hNHE1
(Fig. 8): an acidic pH change could result in alteration
of the protonation state of the region of TM IX located
at the entrance of the proposed funnel, which could
elicit a conformational change in TM IX, causing a
direct contact between TM IV and TM IX. This
rearrangement of TM IV could result in a reorientation
of the TM IV-TM XI arrangement such that a Na+
binding site is exposed to the extracellular space.
Binding of Na+ would cause a charge imbalance that
would trigger a movement of the TM IV- and TM XI
helices, exposing Na+ to the cytoplasm. The release of
Na+ would result in protonation of the Na
+ binding site,
causing a conformational change leading back to the
original arrangement of TM IV and TM XI (Fig. 8). At
least in NhaA, this mechanism only requires small
conformational changes of the helices (13), thus the
proposed TM IV/TM XI arrangement would be suited
for relatively high turn over rate of Na+/H
+ exchange.
In conclusion, we present here a structural model of
hNHE1, which places TM IV and TM XI in close
proximity. This architecture was confirmed by EPR
analyses, from which the distance between TM IV and
TM XI was determined at 15 ± 2 Å. This distance was
decreased and increased, respectively, under conditions
of NHE1 activation and inhibition, consistent with a
role for TM IV and TM XI rearrangements in ion
translocation and inhibitor binding by hNHE1.
Acknowledgements
This work was supported by The Danish National Research
Council (SFP).
We are grateful to M. Landau, Tel-Aviv University, Israel, for
providing us with the PDB file allowing us to calculate this
value also for their model.Cariporide was a kind gift from
Sanofi-Aventis. B. Anni Bech Nielsen is gratefully
acknowledged for technical assistance.
FOOTNOTES
This study was supported by the National Institutes of
Health Grant HL-21179 (to P.M.C.), the Danish
National Research Council (to S.F.P.), the Magnus
Bergwall Foundation (to J.O.L.) and a Biocampus
Scholarship from University of Copenhagen (to
E.B.N.). We thank S. Grinstein, Hospital for Sick
Children, Ontario, Canada, for the kind gift of AP-1
cells, C. Altenbach, Dept. of Ophtalmology at the Jules
Stein Eye Institute, UCLA, LA, California for
providing us with the Basephase program, and Madhu
S. Budamagunta and Biao Shi for technical assistance.
The abbreviations used are: PaNHE1, Pleuronectes
americanus NHE1; AtNHE1, Amphiuma tridactylum
NHE1; hNHE1, human NHE1; TM, transmembrane;
IL, intracellular loop; EL, extracellular loop; EIPA, 5’-
(N-ethyl-N-isopropyl)amiloride; HOE, Hoechst type
inhibitor; TM, transmembrane; RBC, red blood cell;
3D, 3-dimensional; IL, intracellular loop; EL,
extracellular loop; SDSL, site-directed spin labeling;
EPR, electron paramagnetic resonance DDM,
octaethylene glycol mono-n-dodecyl ether (C12E8), n-
dodecyl ß-D-maltoside; CaM, calmodulin.
98
Figure1Fig. 1.Alignment of hNHE1 and NhaA. hNHE1 was predicted to be similar to the structure of NhaAby a
number of well-established secondary structure and fold recognition methods at the MetaServer(25). The
alignment of the sequences is from the joint evaluation of the outcome by the 3D-Jury system at the same server
and by manual adjustments as detailed in Experimental Procedures. The predicted TM helices of NHE1 are
visualized in gray (note the lack of fit for some loop regions).
99
Fig. 2.Full structural model of the N-terminal region of hNHE1. The structure model is based on the known
structure of the smaller bacterial Na+/H+exchanger NhaA. The shown representations include the
transmembranedomains only [Residues 15-31 (TM I), 104-123 (TM II), 130-147 (TM III), 160-179 (TM IV), 191-209
(TM V), 228-246 (TM VI), 255-272 (TM VII), 300-317 (TM VIII), 333-353 (TM IX), 417-437 (TM X), 453-473 (TM XI),
and 481-502 (TM XII)] and exclude connecting loops and terminal extension. The color code used is light-blue for
TMs I and II, dark-blue for TMs III, IV and V, green for TMs VI, VII, VIII and IX, and yellow for TMs X, XI and XII. A.
Side-view of the hNHE1 structure in the plane of the lipid bilayer. B. Cytoplasmicview. The insert shows the
numbering of the individual helices.
100
Fig. 3.Solid surface structure representations of TMs III, IV, V, X, XI and XII (referred to as the catalyticalcore in the text) in the hNHE1 model. Positively charged residues are shown in blue, negatively charged residues in red, and polar residues in green.A. The cytoplasmicview (structure to the right in panel A is tilted 15odownwards compared to the structure to the left) and reveals a cavity (arrows) in the structure that reaches down to Arg425(brown arrow). Several charged and polar residues that may be involved in the ion-translocation are located near this cavity. These include Arg458and Arg500(positively charged), Glu131(negatively charged) and, Ser132, Thr433, Asn437and Tyr454(polar). B. The face of the protein exposed to the outside of the cell does not exhibit any obvious cavity in the structure. However, a cluster of charged residues (Asp470, Lys471, Lys472and His473) at the end of TM XI and also some scattered polar residues (e.g. Tyr209, Thr417and Thr482) are accessible on the outside surface of the transmembranedomain. C. The side-view representation shows that primarily hydrophobic side-chains are pointed into the interior of the lipid bilayer. TMs I, II, VI, VII, VIII and IX are shown as grey ribbons for orientation (compare with Figure 2).
101
Fig 4.Structure of the Na+/H+exchanger catalyticalcore. The central parts of TM domains IV, V, X and XI, the
presumed catalyticalcore for hNHE1-catalyzed Na+/H+exchange, are represented as a ribbon diagram (A) and in a
schematic depiction (B). The amino acid side-chains suggested to directly participate in ion translocation are
shown, as are the positionsof the spin-labels. Note that the characteristic crossover by the extended structures of
helices IV and XI results in energetically unfavorable dipole-dipole pairings (dipoles shown as δ+ and δ-) at the ends
of the disrupted α-helices. A comparison with the central core of NhaAis shown in C.
102
103
Fig. 56.axis labels in the fig (what is standard?). hNHE1 constructs and EPR spectra. A. Immunoblotanalysis showing the presence of hNHE1 in the purified samples of Ala173Cys, Ile461Cys, and Ala173Cys/Ile461Cys, respectively. hNHE1 was detected with 4e9 mouse monoclonal NHE1 antibody and horseradish peroxidase-conjugated goat anti-mouse IgG. The bands cannot be compared quantitative, as different protein concentrations have been applied. ADD SILVER STAIN DATA HEREB-D. EPR analyses for Ala173Cys, Ile461Cys, and Ala173Cys/Ile461Cys. The spectra were collected from samples of purified hNHE1 in PBS buffer (pH 7.5) containing 0.01% DDM, and the protein concentration (10 μM) was identical in all samples. The identically treated Cys-less protein that was used as a negative control gave no EPR signal (not shown).All the spectra were normalized as described in Experimental Procedures. B. EPR spectra for Ala173Cys and Ile461Cys. The red trace shows the spectrum for the Ala173Cys, and is an average of measurements from three independently prepared samples, and the black trace shows the spectrum for the Ile461Cys, and is an average of six measurements from two independent protein samples. The arrow indicates the location of the broad component in the spectrum for the Ile461Cys construct, which indicates that position 461 experiences reduced motional freedomcompared to position 173. C. Comparison of the EPR spectrum for the single-and double-labeled constructs.The black trace shows the spectrum for the Ala173Cys/Ile461Cys construct, and the red trace is the sum of the spectra for Ala173Cys and the Ile461Cys constructs. D. EPR spectra for the Ala173Cys/461Cys construct. The black and red traces show the spectra for the Ala173Cys/461Cys construct in the absence and presence of 1 mMamiloride, respectively. The green trace shows the spectrum acquired at pH 5.1 (in the absence of inhibitor). Each spectrum in C represents an average of three measurements.
104
105
106
REFERENCES
1. Orlowski, J. and Grinstein, S. (2004) Pflugers Arch. 447, 549-565
2. Pedersen, S. F., O'Donnell, M. E., Anderson, S. E., and Cala, P. M. (2006) Am J Physiol Regul. Integr. Comp
Physiol 291, R1-25
3. Wakabayashi, S., Pang, T., Su, X., and Shigekawa, M. (2000) J Biol. Chem. 275, 7942-7949
4. McLean, L. A., Zia, S., Gorin, F. A., and Cala, P. M. (1999) Am J Physiol 276, C1025-C1037
5. Pedersen, S. F., King, S. A., Rigor, R. R., Zhuang, Z., Warren, J. M., and Cala, P. M. (2003) Am J Physiol
Cell Physiol 284, C1561-C1576
6. Counillon, L., Franchi, A., and Pouyssegur, J. (1993) Proceedings of the National Academy of Sciences
90, 4508-4512
7. Khadilkar, A., Iannuzzi, P., and Orlowski, J. (2001) J. Biol. Chem. 276, 43792-43800
8. Noël, J., Germain, D., and Vadnais, J. (2003) Biochemistry 42, 15361-15368
9. Orlowski, J. and Kandasamy, R. A. (1996) J. Biol. Chem. 271, 19922-19927
10. Touret, N., Poujeol, P., and Counillon, L. (2001) Biochemistry 40, 5095-5101
11. Wang, D., Balkovetz, D. F., and Warnock, D. G. (1995) Am J Physiol Cell Physiol 269, C392-C402
12. Pedersen, S. F., King, S. A., Nygaard, E. B., Rigor, R. R., and Cala, P. M. (2007) J. Biol. Chem. 282, 19716-
19727
13. Hunte, C., Screpanti, E., Venturi, M., Rimon, A., Padan, E., and Michel, H. (2005) Nature 435, 1197-1202
14. Moncoq, K., Kemp, G., Li, X., Fliegel, L., and Young, H. S. (2007) J. Biol. Chem. M704844200
15. Coupaye-Gerard, B., Bookstein, C., Duncan, P., Chen, X. Y., Smith, P. R., Musch, M., Ernst, S. A., Chang,
E. B., and Kleyman, T. R. (1996) Am J Physiol Cell Physiol 271, C1639-C1645
16. Landau, M., Herz, K., Padan, E., and Ben-Tal, N. (2007) J. Biol. Chem. 282, 37854-37863
17. Lagerstedt, J. O., Voss, J. C., Wieslander, A., and Persson, B. L. (2004) FEBS Letters 578, 262-268
18. Hubbell, W. L., Cafiso, D. S., and Altenbach, C. (2000) Nat. Struct. Biol. 7, 735-739
19. Todd, A. P., Cong, J., Levinthal, F., Levinthal, C., and Hubbell, W. L. (1989) Proteins 6, 294-305
20. Langen, R., Isas, J. M., Luecke, H., Haigler, H. T., and Hubbell, W. L. (1998) J. Biol. Chem. 273, 22453-
22457
21. Voss, J., He, M. M., Hubbell, W. L., and Kaback, H. R. (1996) Biochemistry 35, 12915-12918
22. Zhao, M., Zen, K. C., Hernandez-Borrell, J., Altenbach, C., Hubbell, W. L., and Kaback, H. R. (1999)
Biochemistry 38, 15970-15977
23. Rabenstein, M. D. and Shin, Y. (1995) Proceedings of the National Academy of Sciences 92, 8239-8243
24. Hilger, D., Polyhach, Y., Padan, E., Jung, H., and Jeschke, G. (2007) Biophys. J. biophysj
25. Ginalski, K., Elofsson, A., Fischer, D., and Rychlewski, L. (2003) Bioinformatics 19, 1015-1018
26. Schwede, T., Kopp, J., Guex, N., and Peitsch, M. C. (2003) Nucl. Acids Res. 31, 3381-3385
27. Rotin, D. and Grinstein, S. (1989) Am J Physiol Cell Physiol 257, C1158-C1165
28. Froncisz, W. and Hyde, J. S. (1982) Journal of Magnetic Resonance (1969) 47, 515-521
29. Hubbell, W. L., Froncisz, W., and Hyde, J. S. (1987) Review of Scientific Instruments 58, 1879
30. Altenbach, C., Oh, K., Trabanino, R. J., Hideg, K., and Hubbell, W. L. (2001) Biochemistry 40, 15471-15482
31. Pao, S. S., Paulsen, I. T., and Saier, M. H., Jr. (1998) Microbiol. Mol. Biol. Rev. 62, 1-34
32. Shrode, L. D., Gan, B. S., D'Souza, S. J., Orlowski, J., and Grinstein, S. (1998) Am J Physiol 275, C431-C439
33. Arkin, I. T., Xu, H., Jensen, M. O., Arbely, E., Bennett, E. R., Bowers, K. J., Chow, E., Dror, R. O.,
Eastwood, M. P., Flitman-Tene, R., Gregersen, B. A., Klepeis, J. L., Kolossvary, I., Shan, Y., and Shaw, D.
E. (2007) Science 317, 799-803
34. Galili, L., Herz, K., Dym, O., and Padan, E. (2004) J. Biol. Chem. 279, 23104-23113
35. Kozachkov, L., Herz, K., and Padan, E. (2007) Biochemistry 46, 2419-2430
107
APPENDIX 2 – Cell cultures
Growth medium for AP-1 cells:
500 ml Alpha-Modified Minimum Essential Eagle’s medium (α-MEM, Invitrogen cat # 12571-048)
55 ml Foetal Bovine Serum (Invitrogen, Gibco cat# 10106-151)
5.5 ml Penicillin-Streptomycin (Invitrogen, Gibco cat# 15140-148)
5.5 ml L-Glutamine (Invitrogen, Gibco cat # 25030-081)
Growth medium for transfected AP-1 cells:
500 ml Alpha-Modified Minimum Essential Eagle’s medium (α-MEM, Invitrogen cat # 12571-048)
55 ml Foetal Bovine Serum (Invitrogen, Gibco cat# 10106-151)
5.5 ml Penicillin-Streptomycin (Invitrogen, Gibco cat# 15140-148)
5.5 ml L-Glutamine (Invitrogen, Gibco cat # 25030-081)
600 µg/ml (0.44g/550ml) Geniticine (G418, Invitrogen cat #11811-098)
Trypsin-mix (1%):
1 ml Trypsin-EDTA (10x) (Invitrogen, Gibco cat# 15400-054)
9 ml PBS
PBS (1 l):
8 g NaCl
0.2 g KCl
1.15 g Na2HPO4
0.2 g KH2PO4
pH adjusted to 7.4 and added ddH2O up to 1000 ml
Freezing of cells:
+10% Foetal Bovine Serum (Invitrogen, Gibco cat# 10106-151)
+10% Dimethylsulfoxide (DMSO, Sigma-Aldrich cat # D2438)
APPENDIX 3 – Site-directed mutagenesis
All primers were designed using Vector NTI and the oligocalc web tool (http://www.basic.northwestern.edu/biotools/oligocalc.html). All designs are optimized to Stratagene rules of primerdesign:
1) Primers should be 5’ phosphorylated 2) Primers should not have any secondary structure i.e. hairpins and not be able to self anneal 3) Primers should be between 25 and 45 nucleotides long with 10-15 at each side of the mutant mismatch 4) Primers should have a Tm above 75°C
5) Example:
6) Primers should contain a minimum of 40% GC content 7) Primers should end with minimum one G or C
108
Name V160C (potential hairpin structure) H473C Original seq TCCTGCAGTCCGACGTCTTCTTCCTCTTCCTG TCCTGGACAAGAAGCACTTCCCCATGGCTGAC Primer seq TCCTGCAGTCCGACTGCTTCTTCCTCTTCCTG TCCTGGACAAGAAGTGCTTCCCCATGGCTGAC Tm/°C 77.1 77.1 %GC 56% 56% Length 32 32 Mismatch 2 2
Name L166C L465C Original seq TCTTCTTCCTCTTCCTGCTGCCGCCCATCATCCTGGATG GCCATCGCCTTCTCTCTGGGCTACCTCCTGGACAAG Primer seq TCTTCTTCCTCTTCCTGTGCCCGCCCATCATCCTGGATG GCCATCGCCTTCTCTTGCGGCTACCTCCTGGACAAG Tm/°C 79.5 79.4 %GC 56% 61% Length 39 36 Mismatch 3 3
Name A173C (potential hairpin structure) I461C Original seq GCCCATCATCCTGGATGCGGGCTACTTCCTGCCACTGC GAGGGGCCATCGCCTTCTCTCTGGGCTACC Primer seq GCCCATCATCCTGGATTGCGGCTACTTCCTGCCACTGC GAGGGGCCTGCGCCTTCTCTCTGGGCTACC Tm/°C 80.9 81.0 %GC 61% 70% Length 38 30 Mismatch 3 2
Name R180C I451C Original seq ATGCGGGCTACTTCCTGCCACTGCGGCAGTTCACAG CCAAGGACCAGTTCATCATCGCCTATGGGGGCC Primer seq ATGCGGGCTACTTCCTGCCACTGTGCCAGTTCACAG CCAAGGACCAGTTCTGCATCGCCTATGGGGGCC Tm/°C 81.0 81.2 %GC 58% 64% Length 36 33 Mismatch 2 2
DNA templates and primers were diluted to 100ng/µl work solutions and all PCR reactions were mixed in the following way, as recommended by the manufacturer (QuikChange® Multi Site-Directed Mutagenesis Kit – Instruction manual (cat #200514).
Component Concentration Volume 10xQuickChange Multi reaction buffer 10x 2.5µl QuickSolution - 0.75µl dNTP mix - 1µl QuickChange Multi enzyme blend - 1µl ds-DNA template 100ng/µl 1µl Mutagenic primer 1 100ng/µl 1µl Mutagenic primer 1 100ng/µl 1µl ddH2O - 16.75µl
All PCR reactions were run with the following parameters, as recommended by the manufacturer (QuikChange® Multi Site-Directed Mutagenesis Kit – Instruction manual (cat #200514).
Segment Cycles Temperature Time 1 1 95˚C 1 minute 2 30 95˚C 1 minute 55˚C 1 minute 65˚C 16 minutes*
*2minutes/kb of plasmid length
109
The PCR product was treated for 1h with Dpn1, a restriction enzyme that digests methylated and hemimethylated
DNA, for digestion of parental DNA.
APPENDIX 4 – Transformation, DNA purification, and sequencing
Transformation of XL10-Gold ultracompetent cells as described by the QuikChange® Multi Site-Directed
Mutagenesis Kit – Instruction manual (cat #200514):
Gently thaw the XL10-Gold ultracompetent cells (Stratagene cat # 200314) on ice. For each mutagenesis
reaction to be transformed, aliquot 30 μl of the ultracompetent cells to a prechilled Falcon® 205
polypropylene tube (BD cat #352057).
Add 1.7 μl of the β-ME mix (in the Stratagene kit) provided with the kit to the 30 μl of cells.
Swirl the contents of the tube gently. Incubate the cells on ice for 10 minutes, swirling gently every 2
minutes.
Transfer 1.5 μl of the Dpn I-treated DNA from each mutagenesis reaction to a separate aliquot of the
ultracompetent cells.
Swirl the transformation reactions gently to mix, and then incubate the reactions on ice for 30 minutes.
Preheat NZY+ broth in a 42°C water bath for later use.
Heat-pulse the tubes in a 42°C water bath for 30 seconds. The duration of the heat pulse is critical for
obtaining the highest efficiencies. Do not exceed 42°C.
Incubate the tubes on ice for 2 minutes.
Add 1.5 ml of preheated (42°C) NZY+ broth to each tube and incubate the tubes at 37°C for 1 hour with shaking at 225–250 rpm.
Plate the appropriate volume of each transformation reaction on agar plates containing the appropriate antibiotic for the plasmid vector.
From each transformation reaction 10µl, 50µl, and 100µl was plated onto agar plates with 10mg/l ampicillin for selection of positive clones. The remaining 1.34ml from the transformation reaction was mixed with 350µl autoclaved 50% glycerol, mixed thoroughly and frozen at -80˚C as glycerolstocks.
NZY+ broth (for 1 liter): 10 g of NZ amine (casein hydrolysate) 5 g of yeast extract 5 g of NaCl Add deionized H2O to a final volume of 1 liter Adjust to pH 7.5 using NaOH Autoclave Add the following filer-sterilized supplements prior to use: 12.5 ml of 1 M MgCl2 12.5 ml of 1 M MgSO4 20 ml of 20% (w/v) glucose
LB (for 1 liter): 10 g of NaCl 10 g of tryptone 5 g of yeast extract Add deionized H2O to a final volume of 1 liter Adjust pH to 7.0 with 5 N NaOH Autoclave LB agar with ampicillin (for 1 liter): 10 g of NaCl 10 g of tryptone 5 g of yeast extract 20 g of agar Add deionized H2O to a final volume of 1 liter Adjust pH to 7.0 with 5 N NaOH Autoclave When cooled to ≈55˚C add 10 ml of 10-mg/ml filter-sterilized ampicillin and pour into petri dishes (25 ml/100-mm plate)
110
DNA purification was done using the QIAprep® Miniprep kit in accordance with the guidelines provided by the manufacturer (QIAprep® Miniprep Handbook) and all components used are a part of the kit (QIAprep Spin Miniprep kit, cat # 27104). 5ml LB medium was inoculated with a single colony from the ampicillin plates and grown overnight for DNA purification.
Resuspend pelleted bacterial cells in 250 μl Buffer P1 and transfer to a microcentrifuge tube.
Add 250 μl Buffer P2 and mix thoroughly by inverting the tube 4–6 times.
Add 350 μl Buffer N3 and mix immediately and thoroughly by inverting the tube 4–6 times.
Centrifuge for 10 min at 17,900 xg in a table-top microcentrifuge.
Apply the supernatant to the QIAprep spin column by decanting pipetting.
Centrifuge for 30–60 s. Discard the flow-through.
Wash QIAprep spin column by adding 0.75 ml Buffer PE and centrifuging for 30–60 s.
Discard the flow-through, and centrifuge for an additional 1 min to remove residual wash buffer.
To elute DNA, place the QIAprep column in a clean 1.5 ml microcentrifuge tube. Add 50 μl Buffer EB or water to the center of each QIAprep spin column, let stand for 1 min, and centrifuge for 1 min.
DNA concentration was determined by absorbance and aliquots were send to MWG (http://www.eurofinsdna.com/de/home.html) for sequencing. Sequencing results were assessed by the Vector NTI software.
APPENDIX 5 – Transfection Transfection of AP-1 cells was done by the following protocol:
Place cells in 6-well trays, 2ml media + 3 drops of cells (so 60% confluence will be reached after 2 days).
When 60% confluent: aspirate media, wash with 1ml 370C OPTI-MEM and aspirate again.
Add 1-2ml OPTI-MEM to each well and incubate at 370C, 5% CO2 for 1 hour.
Prepare DNA (calculate the amount that should be added).
Prepare two 1.5ml microcentrifuge tubes (mark one with the name of the DNA and let the other be unmarked).
Add 50 l OPTI-MEM to both tubes.
Add 10 l Lipofectamine to the unmarked tube and incubate at 220C for 5-20min.
Add X l DNA (= 1µg) to the marked tube.
Transfer solution to the marked tube.Incubate at 220C for 20min-6hrs.
Aspirate OPTI-MEM from the wells and add 0.5ml OPTI-MEM.
Add the 110 l from the marked tube to 1 well.
Incubate 4-6 hrs at 370C, 5% CO2..
Aspirate OPTI-MEM and add 2ml 370C media (no antibiotics and no G418).
Incubate 18-48 hrs at 370C, 5% CO2.
Do not passage the next days, as it takes several days before the non-transfected cells die. If you passage, you will loose to many cells, since the transfection effectivity is very poor, and only a few cells gets transfected.
Lipofectamine™
LTX (Invitrogen, cat # 15338-100) Opti-MEM
® (Invitrogen, cat # 31985-070)
APPENDIX 6 – Protein determination Standard: Albumin standard (Pierce cat# 23209) diluted with ddH2O to desired concentration (30, 25, 20, 15, 10, and 5 µg BSA/ 25 µl). The linear regression is used to determine the protein content of the cell lysate. The optical density of
111
5 µl of cell lysate is measured at 595 nm, and this OD value is divided with the linear regression coefficient of the standard curve to obtain a value for the protein quantum in the 5 µl sample. When the protein content has been determined in all of the samples, each sample is diluted with lysis buffer to match the sample of lowest protein concentration. Thus, equalamounts of protein are loaded in each well of the gel to be electrophorized.
APPENDIX 7 – Western blotting
SDS PAGE running buffer: Outer chamber: 30 ml NuPAGE MOPS SDS running-buffer (20xstock) (Invitrogen cat # NP0001) and 570 ml ddH2O Inner chamber: 500 µl NuPAGE Antioxidant (Invitrogen cat # NP0005) to 200 ml running buffer Transfer buffer: 30 ml NuPage transfer buffer (20xstock) (Invitrogen cat # NP0006-1) and 60 ml 96% EtOH, and 510 ml ddH2O TBST: 10 ml 1M Tris/HCl (pH 7.6), 75 ml 2M NaCl, 1 ml Tween20. Adjusted with ddH2O to 1000 ml Blocking buffer: 50 ml TBST, 2.5 mg non-fat dry milk, 250 µl 10% Sodium Azide (NaN3)
APPENDIX 8 – CLSM Preparation of coverslips: Glass coverslips were covered with excessive 4M HCl for at least 2h with occasional shaking. The HCl was removed and the slips were washed in 4-5 times in ddH2O for dilution and removal of remaining acid. Finally, the slips were rinsed twice with 96% EtOH, and stored in 70% EtOH. Paraformaldehyde (PFA) solution (20 %): 1 g PFA dissolved in 8 ml ddH2O + three drops of 1 M NaOH. Carried out under fume hood, while wearing gloves and mask. Fixation buffer (4 % PFA): 5 ml PFA solution, 20 ml PBS Permabilization buffer: 0.25 g BSA, 50 µl Triton X-100 (Promega cat# H5141), 25 ml PBS Blocking buffer: 0.5 g BSA, 25 ml PBS Mounting medium with “anti-fade” (5 ml): 4.5 ml glycerol, 500 µl 10xPBS (70 % final), 0.1 g N-propyl gallate (2 % final)
APPENDIX 9 – Functional screening Solutions for the functional assays were prepared using the following spreadsheet in which the measured osmolarity defines the volumes of each component.
0
0.063
0.126
0.189
0.252
0.315
0.378
0 10 20 30
OD
600 n
m
µg protein
Proteinstandart kurve DC Assay
µg protein OD
0 0
5 0.06
10 0.121
15 0.164
20 0.228
25 0.279
30 0.305
112
Osm
IR (pH=7,4) NMDG-IR (pH=7,4)
Solution c/M Part Osmolarity ml mmol. mOsm ml mmol. mOsm
NaCl 1M 2 0.908 286.34 130 260 0.00 0 0
KCl 1M 2 0.914 6.56 3 6 3.28 3 6
Hepes 1M 1 1.036 38.61 20 20 19.31 20 20
MgCl2 1M 3 0.821 2.44 1 3 1.22 1 3
CaCl2 1M 3 0.454 2.20 0.5 1.5 1.10 0.5 1.5
NaOH 2M 1 2.776 3.60 10 10 0.00 0 0
KOH 1M 1 0.978 0.00 0 0 0.00 0 0
NMDGCl 1M 2 1.024 0.00 0 0 126.95 130 260
NMDGOH 1M 1 1.852 0.00 0 0 5.40 10 10
Glucose* 1M 1 1.032 19.38 10 10 9.69 10 10
Total-theoretical 2000 174.5 310.5 500 174.5 310.5
Total-actual 2000 1000
Osm NH4Cl-IR (pH=7,4)
Solution c/M Part Osmolarity ml mmol. mOsm
NaCl 1M 2 0.908 137.67 125 250
KCl 1M 2 0.914 3.28 3 6
Hepes 1M 1 1.036 19.31 20 20
MgCl2 1M 3 0.821 1.22 1 3
CaCl2 1M 3 0.454 1.10 0.5 1.5
NaOH 2M 1 2.776 1.80 10 10
NMDGCl 1M 2 1.024 0.00 0 0
NMDGOH 1M 1 1.852 0.00 0 0
Glucose 1M 1 1.032 9.69 10 10
NH4Cl 1M 2 1.706 5.86 10 20
Total-theoretical 1000 179.5 320.5
Total-actual 1000
APPENDIX 10 – Protein purification
Tris Buffered Saline solutions were all adjusted to pH=7.4 before reaching final volume.
1000mL 10xTBS – 50mM Tris
Compund Mass Concentration (10x) Concentration (1x)
Tris (C4H11NO3) 60.75g 500mM 50mM
NaCl 80.0g 1368.9mM 136.9mM
113
KCl 2.0g 26.8mM 2.7mM
250mL 10xTBS – High salt
Compund Mass Concentration (10x) Concentration (1x)
Tris (C4H11NO3) 15.14g 500mM 50mM
NaCl 51.14g 3500mM 350mM
KCl 2.0g 26.8mM 2.7mM
250mL 10xTBS – Low salt
Compund Mass Concentration (10x) Concentration (1x)
Tris (C4H11NO3) 3.03g 100mM 10mM
APPENDIX 11 – EPR samples Buffer pH=7.4: TBS(50mM tris) (pH=7.4) + 0.04% DDM
Buffer pH=1.0: TBS(50mM tris) (pH=1.0) + 0.04% DDM Cariporide mesilate (HOE 642) from Adventis Pharma Deutschland GmbH (4-Isopropyl-3-methylsulfonyl-benzoyl)-guanidine methanesulfonate) was dissolved in DMSO to a working solution of 66.67mM.
MTS spin label (2,2,5,5-tetramethyl-2,5-dihydro-1H-pyrrol-3-yl)methyl methanesulfonothioate) was dissolved in DMSO.
Cell harvest 10mL TBS(50mM tris) + 1 protease inhibitor tablet
NHE1 extraction 3mL TBS(50mM tris) + 0.3% Triton, 0.04% DDM Add 9µL Triton and 4µL DDM to 3mL TBS(50mM tris)
Nickel-column (Bind)
12mL TBS(50mM tris) + 0.04% DDM, 20mM Imidazole Add 240µL 1M Imidazole in TBS(50mM tris) to 11.76mL Buffer A
Nickel-column (Wash)
6mL TBS(50mM tris) + 0.04% DDM, 80mM Imidazole Add 480µL 1M Imidazole in TBS(50mM tris) to 5.52mL Buffer A
Nickel-column (Elute)
6mL TBS(50mM tris) + 0.04% DDM, 300mM Imidazole Add 1.8mL 1M Imidazole in TBS(50mM tris) to 4.2mL Buffer A
CAM-bind 20mL TBS(50mM tris) + 0.04% DDM, 2mM CaCl2, 1mM MgCl2
Add 40µL 1M CaCl2 and 20µL 1M MgCl2 to 20mL Buffer A
CAM-High salt wash
5mL TBS(50mM tris, 350mM NaCl) + 0.04% DDM, 2mM CaCl2, 1mM MgCl2
Add 10µL 1M CaCl2 and 5µL 1M MgCl2 to 5mL Buffer C
CAM-Low salt wash
5mL TBS(10mM tris) + 0.04% DDM, 2mM CaCl2, 1mM MgCl2
Add 10µL 1M CaCl2 and 5µL 1M MgCl2 to 5mL Buffer D
CAM-elute 15mL TBS(50mM tris) + 0.04% DDM, 5mM EGTA Add 0.75mL 100mM EGTA to 14.25mL Buffer A
Buffer A 90mL TBS(50mM tris) + 0.04% DDM Add 120µL DDM to 90mL TBS(50mM tris) Buffer B 10mL 1M Imidazole in Buffer A Dissolve 0.68g in 10mL Buffer A and adjust pH to 7.4 Buffer C 5mL TBS(50mM tris, 350mM NaCl) + 0.04% DDM Add 6.67µL DDM to 5mL TBS(50mM tris, 350mM NaCl) Buffer D 5mL TBS(10mM tris) + 0.04% DDM Add 6.67µL DDM to 5mL TBS(10mM tris)
114
APPENDIX 12 – Ramachandran plot of NhaA
115
From the PROCHECK application at the PDBsum homepage (http://www.ebi.ac.uk/pdbsum/) accession number PDB ID: 1ZCD.