DAP12 IMPACTS TRAFFICKING AND SURFACE STABILITY OF KILLER CELL

IMMUNOGLOBULIN-LIKE RECEPTORS ON NATURAL KILLER CELLS

A Dissertation

submitted to the Faculty of the

Graduate School of Arts and Sciences

of Georgetown University

in partial fulfillment of the requirements for the

degree of

Doctor of Philosophy

in Tumor Biology

By

Tiernan J. Mulrooney, B.S.

Washington, DC

September 14, 2012

ii

Copyright 2012 by Tiernan J. Mulrooney

All Rights Reserved

iii

DAP12 IMPACTS TRAFFICKING AND SURFACE STABILITY OF KILLER CELL

IMMUNOGLOBULIN-LIKE RECEPTORS ON NATURAL KILLER CELLS

Tiernan J. Mulrooney, B.S.

Thesis Advisor: Carolyn Katovich Hurley, PhD.

ABSTRACT

Killer cell immunoglobulin-like receptors (KIR) aid in the regulation of natural killer (NK) cell

activity. In this study, the effect of the interaction between the two domain stimulatory KIR

(KIR2DS) and their adapter, DAP12, was investigated beyond the previously defined signaling

function. Flow cytometry analysis showed enhanced KIR2DS surface expression on NKL cells

when co-transfected with DAP12. Conversely, KIR2DS4 surface expression on primary cells

was decreased when the cells were treated with DAP12 specific siRNA. Treatment of the

KIR2DS and DAP12 transfected cells with either cycloheximide or brefeldin A repressed

KIR2DS surface expression revealing a role for DAP12 in trafficking newly synthesized KIR to

the cell surface. Immunoprecipitation of DAP12 revealed an interaction of DAP12 with an

immature isoform of KIR2DS indicating the interaction between these proteins initiates early in

the maturation process, likely within the endoplasmic reticulum. An internalization assay

demonstrated a significant impact of DAP12 on KIR2DS surface stability. Confocal microscopy

showed internalized KIR2DS molecules are recruited to lysosomal compartments independent of

DAP12 expression. Our results suggest in vivo conditions that adversely affect DAP12

expression will indirectly reduce surface expression and stability of KIR2DS. These effects

could significantly impact ligand recognition and strength of signaling through KIR2DS

molecules.

iv

ACKNOWLEDGEMENTS

The work presented within this dissertation represents a collaborative effort stemming from an

enormous network of support. I am extremely grateful to have had the opportunity to pursue my

graduate studies in the Tumor Biology Program at Georgetown University Medical Center. I

especially thank the director, Dr. Anna Riegel, and all other faculty for their support and helpful

critiques through the years. My gratitude must also be extended towards the members of my

thesis committee: Drs. Steve Byers, John Coligan, Phillip Posch, Aykut Üren, and Anton

Wellstein for their advice and time given to aid in the fulfillment of this project.

Many thanks are directed to my mentor, Dr. Carolyn Hurley, for her support in the development

of this project as well as my personal development as a scientist. I greatly appreciate her advice

and critiques in both experimental aspects and the writing of this dissertation. The advice gained

from my interactions with Carolyn will be carried through the rest of my career.

I am also thankful for the support from all the members of the Hurley lab, both past and present.

Specifically, I thank Dr. Chris VandenBussche for his guidance and enthusiasm during my initial

time in the lab. Chris’s previous work and his confidence in my help paved the road for the

evolution of this project. I also would like to recognize the support from the other lab members,

Will Frazier, Noriko Steiner, Dr. Brian Yokley, Sandra Selby, Dr. Riddhish Shah, Christian

Tabib, and Linda Jones.

My final acknowledgements are extended toward my family. I am extremely fortunate to have

countless number of family members I can rely on. My parents have given me insurmountable

v

support and encouragement throughout my education. Although, neither are scientists, their

raising of me has been instrumental in my career path to do what I can to inspire and help others.

The support from my brothers also necessitates extreme gratitude. Being the youngest of three, I

have always had two amazing people to look up to and guide me through their actions in their

own lives. I owe thanks to the rest of our great family. My grandparents, aunts, uncles, cousins,

and nephew have always supported me and continue to help mold me into respectful, humble

young man.

vi

TABLE OF CONTENTS

CHAPTER 1: INTRODUCTION ............................................................................................... 1

1.1 NATURAL KILLER CELL OVERVIEW ....................................................................... 1

1.2 OVERVIEW OF KIR ................................................................................................ 8

1.3 OVERVIEW OF DAP12 ......................................................................................... 13

1.4 GENETIC COMPOSITION OF KIR HAPLOTYPES ...................................................... 15

1.5 KIR LIGANDS ....................................................................................................... 19

1.6 COEVOLUTION OF KIR AND HLA ........................................................................ 22

1.7 NK CELL EDUCATION ........................................................................................... 25

1.8 NK CELLS AND CANCER THERAPEUTIC STRATEGIES ............................................. 31

1.9 NK CELLS AND KIR IN HEMATOPOIETIC STEM CELL TRANSPLANTATION ............. 33

1.10 NK CELLS IN PREGNANCY .................................................................................. 36

1.11 KIR AND HIV ................................................................................................... 38

1.12 STATEMENT OF PURPOSE .................................................................................... 40

CHAPTER 2: DAP12 IMPACTS TRAFFICKING AND SURFACE STABILITY OF KILLER CELL

IMMUNOGLOBULIN-LIKE RECEPTORS ON NATURAL KILLER CELLS …...………………….43

2.1 MATERIALS AND METHODS .................................................................................. 43

2.1.1 Cell lines and culture ………………………………………………….43

2.1.2 DNA constructs ………………………………………………………..43

2.1.3 KIR expression and trafficking ………………………………………..44

2.1.4 Gene silencing and relative quantity RT-PCR…………………………45

vii

2.1.5 Immunoprecipitation and Western blot ………………………………..46

2.1.6 Internalization analysis ………………………………………………...46

2.1.7 Confocal microscopy …………………………………………………..47

2.2 RESULTS ………………………………………………………………………..48

2.3 DISCUSSION …………………………………………………………………….76

2.3.1 Potential impacts of DAP12 on KIR2DS function beyond signaling …..76

2.3.2 Potential clinical relevance of observed results………………………... 78

2.3.3 Further investigation of the biology of KIR2DS maturation……………80

2.4 RESEARCH ACKNOWLEDGEMENTS………………………………………………..86

2.5 REFERENCES …………………………………………………………………… 87

viii

LIST OF FIGURES

Figure 1.1 Classical versus contemporary models of immune responses ..................... 5

Figure 1.2 The missing self hypothesis......................................................................... 7

Figure 1.3 Depiction of KIR protein structures .......................................................... 11

Figure 1.4 KIR haplotype structures ........................................................................... 18

Figure 1.5 Three models of NK cell education ........................................................... 30

Figure 2.1 KIR2DS surface expression is enhanced by exogenous DAP12............... 52

Figure 2.2 KIR2DS4 surface expression is decreased after knockdown of DAP12

in primary cells .......................................................................................... 56

Figure 2.3 Transfected KIR2DS4 surface expression continues to increase with

exogenous DAP12 over time ..................................................................... 61

Figure 2.4 DAP12 impacts trafficking of newly synthesized KIR2DS to the

cell surface ................................................................................................. 63

Figure 2.5 DAP12 interacts with KIR2DS early in the maturation process ............... 67

Figure 2.6 DAP12 stabilizes KIR2DS at the cell surface ........................................... 69

Figure 2.7 Confocal microscopy shows internalized KIR2DS1 localizes to LAMP1-

associated lysosomal compartments .......................................................... 73

Figure 2.8 DAP12 retards degradation of KIR2DS1 .................................................. 75

Figure 2.9 Overview of results and future directions ................................................ 85

ix

ABBREVIATIONS

ADCC, antibody dependent cellular cytotoxicity

ALL, acute lymphoid leukemia

AML, acute myeloid leukemia

ANOVA, analysis of variance

APC, allophycocyanin

APCs, antigen presenting cells

BFA, brefeldin A

cDNA, complementary DNA

CHX, cycloheximide

CLP, common lymphoid progenitor

Endo H, endoglycosidase H

ER, endoplasmic reticulum

EV, empty vector

EVT, extravillous trophoblasts

Fab, antibody variable region

FACS, fluorescence-activated cell sorting

FBS, fetal bovine serum

Fc, antibody constant region

GFP, green fluorescent protein

GVHD, graft versus host disease

x

HA, hemagglutinin

HLA, human leukocyte antigen

IFN-γ, interferon gamma

Ig, immunoglobulin

IL, interleukin

ITAM, immunoreceptor tyrosine-based activation motif

ITIM, immunoreceptor tyrosine-based inhibitory motif

KIR, killer cell immunoglobulin-like receptors

KIR2D, two domain KIR

KIR2DL, two domain long tailed KIR

KIR2DS, two domain short tailed KIR

KIR3D, three domain KIR

KIR3DL, three domain long tailed KIR

KIR3DS, three domain short tailed KIR

LILR, leukocyte immunoglobulin-like receptors

LRC, leukocyte receptor complex

MCMV, mouse cytomegalovirus

MFI, mean fluorescent intensity

MHC, major histocompatibility complex

MIP1-α, macrophage inflammatory protein 1 alpha

MIP1-β, macrophage inflammatory protein 1 beta

NCR, natural cytotoxicity receptors

xi

NK, natural killer

PBMC, peripheral blood mononuclear cells

PBS, phosphate buffered saline

PE, phycoerythrin

pNK, peripheral NK cells

SEM, standard error of the mean

siRNA, small interfering RNA

TCD, T cell deplete

TNF-α, tumor necrosis factor alpha

UBM, unmanipulated bone marrow

uNK, uterine NK cells

VEGF, vascular endothelial growth factor

1

INTRODUCTION

1.1 NATURAL KILLER CELL OVERVIEW

The immune system protects the body from invasion of foreign pathogens such as viruses or

bacteria and also has a role in preventing malignant progression. The functions of the immune

system are further characterized as either innate or adaptive based on the timing and specificity

of the response. The innate immune system is primarily responsible for clearing a majority of

foreign insults quickly upon host recognition. In cases where the innate system cannot

completely clear the pathogen, the adaptive immune system is activated providing proper

clearance as well as lasting immunological memory in order to efficiently eradicate a future

infection by the same pathogen. While the immune system is comprised of a variety of different

cell types that interact through an intricate network of cellular and cytokine cross-talk, the

biology of natural killer (NK) cells will be the major focus of this dissertation.

Through a process known as hematopoiesis, all immune cells are derived in the bone marrow.

During this process, NK cells differentiate from the same common lymphoid progenitor (CLP) as

T cells and B cells [1;2]. NK cells comprise between 10-15% of peripheral blood lymphocytes.

The primary cell surface markers of human NK cells are CD56 and/or CD16 in the absence of

the T-cell marker, CD3. Based on immune-phenotyping using these cell markers, three distinct

NK cell populations with divergent functional activities have been described [3]. Expression

levels of CD56 define two populations of NK cells as either bright or dim. The CD56 dim

population comprises close to 90% of the NK cells and represents a more mature differentiation

2

state [4;5]. The CD56 dim cells also express high levels of CD16 as well as members of the

killer immunoglobulin-like receptor (KIR) family. The CD56 dim NK cells also exhibit multiple

functions as they exhibit cytotoxic capabilities along with the ability to secrete pro-inflammatory

cytokines. The CD56 bright population is a less differentiated state for NK cells and unlike the

CD56 dim population, these cells typically expression low levels of CD16 and do not express

KIR. Functionally, the CD56 bright population differs from the CD56 dim cells as these cells

only demonstrate the ability to secrete immune-regulatory cytokines. The third and least

frequent population of NK cells is described as CD56 negative, CD16 positive. This population

exhibits an intermediate cytotoxic capacity compared to the CD56 dim and bright populations

driven primarily by antibody stimulation of the CD16 receptor.

As mentioned NK cells are derived from the same CLP as T cells and B cells; however, several

key characteristics differentiate NK cells from their lymphoid counterparts. NK cells lack the

antigen specific receptors displayed by B cells and T cells resulting from genetic rearrangement.

NK cells instead exhibit stochastic expression of a variety of activating and inhibitory receptors.

Another major distinguishing functional characteristic that separates them from the other

lymphoid counterparts is the NK cell’s ability to react immediately against a foreign pathogen

without initial priming of the cells. This rapid response of NK cells was first described in their

unique ability to kill tumor cells during brief co-cultures without prior stimulation [6;7]. In

conjunction with the cytotoxic capabilities of NK cells, the secretion of cytokines also impacts

the immune system. Through secretion of cytokines such as interferon gamma (IFN-γ),

macrophage inflammatory protein 1 alpha and beta (MIP1-α and MIP1– β), and tumor necrosis

3

factor alpha (TNF-α), NK cells can recruit cells of the adaptive immune system to the site of

invasion further enhancing the immune response to pathogens [8-10].

Resulting from the ability to respond immediately without prior antigen stimulation, NK cells

have been designated members of the innate immune system. The innate immune response

classically has been thought to lack clonal expansions of the innate cells as well as to lack a

memory response to multiple infections by the same pathogen. In the classical model, NK cells

are activated and respond to the primary infection. If the infection is not efficiently cleared,

clonally expanded T cells and B cells are recruited resulting in the clearance of the pathogen.

The clonal expansion of the T cells and B cells results in a population of memory cells that are

quick to react with a more robust response to a secondary insult by the same pathogen. During a

secondary infection under the classical model, NK cells respond as they did after the first

infection with no evidence of clonal expansion or enhancement of response as seen with the T

cells and B cells. While the designation of NK cells as members of the innate immune system

still remains, new evidence has described some adaptive characteristics of “memory” NK cells

[11;12]. These studies suggest populations of NK cells that initially respond to viral infection

rapidly proliferate and are activated upon secondary infection by the same virus suggesting the

existence of a clonal population of NK cells with memory characteristics. These data contradict

the classical model of innate immune responses to primary and secondary infections as two

independent occurrences. Resulting from these studies, a contemporary model of immune system

activation has been formulated, accounting for the presence of NK cell memory (Figure 1.1).

4

5

Figure 1.1 Classical versus contemporary models of immune responses

NK cell responses are shown in blue and T cell and B cell responses are indicated in orange.

Under the classical model, NK cells respond initially to the primary infection. If the NK cells

along with the other functioning cells of the innate immune system do not clear the pathogen,

T cells and B cells become the predominant effector cells. Clonally expanded, antigen-specific

T cells and B cells efficiently eradicate the pathogen. During this process, the clonally expanded

T cells and B cells develop a memory phenotype. Upon a secondary infection by the same

pathogen, NK cells respond just as they did during the first infection. However, the memory

T cells and B cells respond rapidly and more robustly to quickly clear the infection. The

contemporary model reveals a similar pattern of response to primary infection for NK cells,

T cells, and B cells. The major distinction is during the primary infection NK cells undergo a

clonal expansion resulting in a memory phenotype of these cells for the specific pathogen. These

memory NK cells expand rapidly upon secondary infection and with a greater response

compared to the initial response similar to the mechanisms of activation described for T cells and

B cells.

6

As NK cells exhibit rapid cytotoxicity, the mechanism by which normal cells were protected

from NK cell lysis was unknown. The “missing self” hypothesis explains the basic principle of

how NK cells become activated against foreign pathogens without damage to self tissue [13;14].

The hypothesis proposed NK cell activation was regulated by expression of major

histocompatibility complex (MHC) molecules on potential target cell surfaces (e.g. human

leukocyte antigen (HLA) class I). Since HLA class I molecules are expressed on almost all cell

types within the body, inhibitory receptors expressed on NK cells recognize these HLA ligands

preventing the lysis of normal tissue. Contrary to normal cells, many malignant and infected

cells downregulate HLA expression in order to evade T cell and B cell recognition. This

decreased expression of HLA in combination with expression of stimulatory ligands makes these

cells more susceptible to NK cell lysis as inhibitory signals are no longer propagated within the

NK cell [15] (Figure 1.2).

As research has progressed, the models of activation have become much more complex, as a

multitude of receptors are known to play a role in dictating NK function. Many of these

receptors are categorized into diverse families that can consist of both inhibitory and stimulatory

receptors. Some examples of the major receptor families expressed by NK cells are KIR, natural

cytotoxicity receptors (NCR), NKG2x receptors, and leukocyte immunoglobulin-like receptors

(LILR). Within this dissertation, the role of the KIR family is examined to further understand

how NK cells function.

7

Figure 1.2 The missing self hypothesis

NK cells are depicted on the left expressing inhibitory KIR in red and activating KIR in green.

Upon interaction with a target cell expressing a self HLA antigen depicted in the top interaction,

the inhibitory KIR abrogrates activating signals through the phosphatases, SHP1 and SHP2. As

illustrated in the lower image, interaction of an activating receptor, in the absence of inhibitory

interaction, causes activation signals within the cell transmitted by an adapter protein, DAP12,

resulting in a cytotoxic response against the target cell.

8

1.2 OVERVIEW OF KIR

The highly polymorphic KIR gene family is located within the leukocyte receptor complex

(LRC) on human chromosome 19q13.4 and is comprised of 12 genes and 2 pseudogenes. The

KIR genes are transcribed and translated into type I integral membrane proteins that are

expressed by NK cells and a subset of T-cells. KIR have an integral role in regulating NK cell

detection of self versus non-self as many of the receptors have been described to recognize HLA

class I molecules. The KIR molecules contain either three (KIR3D) or two (KIR2D)

extracellular immunoglobulin (Ig)-like domains (Figure 1.3). The KIR3D (3DL1/S1, 3DL2,

3DL3) express all exons encoding the extracellular Ig-like domains, D0, D1, and D2. KIR2D

molecules are further characterized by which Ig-like domains are expressed. Type I KIR2D

(2DL1/S1, 2DL2/L3, 2DS2, 2DS3, 2DS4, 2DS5) are genetically similar to KIR3D as all exons for

the extracellular domains are present. In contrast to KIR3D, the exon encoding the D0 domain is

spliced out of the Type I KIR2D transcripts, leaving only the translation of the D1 and D2

domains. The Type II KIR2D (2DL4, 2DL5A/B) express the D0 and D2 domains as the exon

encoding the D1 domain has been deleted within these genes.

The KIR family consists of both inhibitory and stimulatory receptors. The majority of KIR that

have a long cytoplasmic (KIR2DL and KIR3DL) region are functionally inhibitory and the short

cytoplasmic tail KIR (KIR2DS and KIR3DS) function as stimulatory receptors. The inhibitory

KIR have two immunoreceptor tyrosine-based inhibitory motifs (ITIMs). Upon ligand

engagement the phosphatases, SHP1 and SHP2, are recruited to the ITIMs resulting in the

inhibition of activation cascades within the cell [16;17]. The lone exception is KIR2DL4 which

9

has only one ITIM in its cytoplasmic region, but uniquely has a positively charged arginine at

position 4 of the transmembrane region that creates an interaction site for the adapter molecule,

FcεRIγ. Through this adapter molecule, KIR2DL4 is able to stimulate activating signals upon

ligand binding resulting in cytokine secretion [18-20].

The short cytoplasmic tail of the stimulatory KIR lacks classical immunoreceptor tyrosine-based

activation motifs (ITAMs) and causes these receptors to rely on their adapter molecule, DAP12,

for efficient transmission of signaling. The DAP12 gene is located on human chromosome

19q13.1 and encodes a disulfide-bonded homodimer containing two ITAMs within its

cytoplasmic region. Stimulatory KIR and DAP12 interact non-covalently through a lysine at

position 9 of the transmembrane region of the KIR and an aspartic residue of DAP12. Upon

ligand binding of the receptor, DAP12 recruits ZAP-70 and Syk protein tyrosine kinases to

initiate activation cascades within the cell [21;22].

10

11

Figure 1.3 Depiction of KIR protein structures

KIR express either two (KIR2D) or three (KIR3D) extracellular Ig-like domains. The KIR3D

receptors express all of the Ig-like domains, D0, D1, and D2. The KIR2D receptors are

subclassified as Type I or Type II based on the presence of the D1 and D2 domains or the D0 and

D2 domains in the proteins, respectively. KIR2D and KIR3D receptors either have a long

(KIR2DL, KIR3DL) or short cytoplasmic tail (KIR2DS, KIR3DS). The long tailed KIR, with

the exception of KIR2DL4, contain two ITIMs within the cytoplasmic tail depicted by the red

boxes. KIR2DL4 only contains a single ITIM within this region. The short tailed KIR and

KIR2DL4 express a positively charged amino acid within the transmembrane region that plays a

role in determining specific interactionswith adapter molecules. The amino acid for the short

tailed KIR is a lysine at position 9 of the transmembrane region while KIR2DL4 contains an

arginine at position 4. The differences in amino acid and position results in KIR2DL4

interacting with FcεRIγ compared to the short tailed KIR which interact with DAP12.

12

KIR are not expressed by every NK cell and the expression frequency of a single receptor can

vary amongst individuals. The precise mechanisms underlying the regulation of KIR expression

are not well understood. Epigenetic methylation of the promoter region has been shown to

regulate KIR promoter activity in vivo [23]. The promoters of KIR and their mouse functional

homolog, Ly49, contain distal and proximal promoter elements. While the distal elements are

functional, the primary driver of KIR transcription comes from the proximal promoter element

consisting of the region 300 nucleotides upstream of the translation start site. Interestingly,

forward and reverse transcripts from the proximal promoter have been identified for the majority

of KIR [24]. A theory based on bidirectional promoter activity suggests the promoter activity of

each KIR is regulated by a balance of the quantity of forward and reverse transcripts. Data have

supported this theory as expression of a specific KIR is associated with a greater number of

forward transcripts compared to reverse transcripts of the gene [25]. A possible explanation to

the bidirectional promoter hypothesis was discovered as a 28-base PIWI-like RNA resulting

from reverse transcription within the proximal promoter of KIR3DL1 has been shown to

negatively impact KIR3DL1 expression and presence of this small RNA was correlated with

methylation of the KIR3DL1 promoter [26]. The mechanisms remain unclear but this

methylation may result from chromatin modifying enzymes that interact with the PIWI-like

RNA. In addition to understanding expression patterns of single receptors, mechanisms for

multiple KIR expression by a single NK cell remain unknown resulting in the current thought

that multiple KIR expression is stochastic and can be predicted by the product rule. According to

the product rule, the probability of expression of multiple KIR can be predicted by multiplying

the frequencies of each KIR individually. However analysis of a cohort of 44 individuals

13

revealed NK cells expressing two or more KIRs occurred more frequently than what was

predicted by the product rule suggesting a sequential acquisition of KIR expression [27].

Expression patterns and promoter regulation of KIR continues to be a target of investigation to

further understand how KIR expression is regulated during NK cell maturation.

1.3 Overview of DAP12

The adapter molecules of the immune system transmit signals following ligand recognition by

their cognate receptor. Adapter molecules and their signaling motifs, ITAMs, appear to have

evolved prior to the development of adaptive immunity as many receptors of innate immunity

rely on the signaling capabilities of these proteins. Evolutionarily, the presence of adapter

proteins is beneficial as it allows for the ligand binding regions of receptors to adapt to selective

pressures from pathogens or polymorphisms within HLA class I while the signal transducing

subunits remain conserved. An example of such evolutionary impact is the rise of multiple

activating KIR with diverse extracellular domains, but all are suspected or have been shown to

interact with the same adapter molecule, DAP12. DAP12 has been evolutionarily conserved as

evidenced by sequences of the gene from a variety of mammals, amphibians and fish [28]. The

T cell and B cell receptors further emphasize this evolutionary benefit as the adapters, CD3 and

CD79, transmit signals stemming from interactions with innumerable receptors arising from

rearrangements of the TCR and BCR genes, respectively.

DAP12 is a relatively small protein of 12 kilodaltons comprised of a small extracellular domain

containing a cysteine residue that results in the expression of homodimers of DAP12. While the

14

function of DAP12 in NK cells is the focus of this dissertation, the adapter molecule is expressed

and functions in a number of cell types including dendritic cells, neutrophils, basophils,

eosinophils, monocytes, macrophages, microglials cells, osteoclasts, as well as a subset of NKT

cells and T cells. The broad expression pattern correlates with the promiscuity of the adapter as

over twenty receptors within these cell types rely on DAP12 for signaling [29]. Interestingly cell

type specific functions have also been described as inhibitory functions of DAP12 have been

demonstrated in macrophages and dendritic cells, but only activating signaling has been

observed from DAP12 in NK cells [30]. Knockout of DAP12 in mice results in reduced abilities

to clear viral or tumor insults by NK cells and also leads to osteopetrosis [31;32]. In humans,

DAP12 deletions or loss of function mutations results in Nasu-Hakola disease causing frequent

bone cysts and osteoporosis as well as severe impacts on the nervous system leading to presenile

dementia [33]. Although to date no increased susceptibility to viral infection or malignancy

development has been described for these patients, as DAP12 interacts with a number of

activating receptors on NK cells including KIR and members of the natural cytotoxicity receptor

family, it is likely the NK cells of these individuals would demonstrate a severe inability to clear

infections or malignant cells that express the ligands for these receptors. Without a clear

understanding of the role of stimulatory KIR in humans, it is uncertain what aspects of NK cell

function are disrupted by the absence of DAP12 in these patients.

In addition to the signaling functions, adapter molecules have been implicated in affecting

surface expression of their cognate receptors. In human NK cells, a decrease of DAP12

expression correlated with a decrease in surface expression of another interacting receptor,

15

NKp44 [34]. Surface expression levels of the activating receptor, Ly49H, are significantly

reduced in DAP12 knockout mice [35]. Expression of another adapter molecule, DAP10, has

also been correlated with surface expression levels of its cognate receptor, NKG2D [36]. The

potential impact of DAP12 on KIR2DS surface expression and the underlying mechanisms are

the primary focus for the research portion of this dissertation.

1.4 GENETIC COMPOSITION OF KIR HAPLOTYPES

The KIR gene locus is the most complex genetic locus of the human genome resulting from

variations including KIR gene content, allelic polymorphisms, and inter-locus polymorphisms

[37]. The KIR genes form two clusters defined by the proximity of the position to the

centromere or telomere [38]. The centromeric cluster is bordered by the framework genes,

KIR3DL3 and KIR3DP1, while KIR2DL4 and KIR3DL2 border the telomeric cluster in the

majority of KIR haplotypes. The remaining KIR genes either are restricted to a single cluster or

have been observed to have undergone a duplication event which has resulted in the presence of

the gene in either of the two clusters or even both clusters on the same chromosome [39;40].

Meiotic recombination events appear to frequently occur within the region separating the two

gene clusters [41]. Haplotypes have been described resulting from unequal crossing over

creating haplotypes with duplicated KIR genes and others with large deletions [42;43]. Common

haplotype structures have been described for the centromeric (Cen-A1, Cen-B1, Cen-B2) and

telomeric (Tel-A1, Tel-B1) clusters [44;45] (Figure 1.4). Cen-A1 and Tel-A1 are predominantly

the most frequent structures with frequencies of 67-72% and 76-82% in individuals of European

ancestry and a random Caucasian population, respectively, as described in independent reports.

16

Polymorphisms also contribute to the genetic complexity of KIR genes. Substantial

polymorphisms have been described in the inhibitory KIR while the activating KIR exhibit less

variation compared to the inhibitory genes [46]. Allelic polymorphisms have been shown to

impact KIR and ultimate NK cell function not predicted by genotyping for presence or absence

of genes. For example, allelic polymorphisms in KIR2DL2*004 and KIR3DL1*004 cause

misfolding of these allelic products leading to retention within the cell [47;48]. KIR2DS2*003

has a polymorphism at the residue important for interaction with DAP12 within the

transmembrane region. This alteration prevents cytotoxic activity to be propagated through

KIR2DS2 in those cells [49]. Other allelic polymorphisms within the exons encoding the

extracellular domains of KIR have been described to impact KIR binding of HLA [50-52]. The

evidence described in these studies illustrates the importance for understanding and researching

how combinations of KIR haplotype structures as well as allelic polymorphisms impact

individual KIR function and ultimately impact NK cell functions.

17

18

Figure 1.4 KIR haplotype structures

Depicted are the common centromeric and telomeric haplotype structures of the KIR locus on

chromosome 19q13.4. The centromeric structure is flanked by the framework genes KIR3DL3

and KIR3DP1 while the framework genes, KIR2DL4 and KIR3DL2, demarcate the telomeric

structure. The dashed lines indicate each combination of centromeric and telomeric haplotypes

have been identified in population studies.

19

1.5 KIR LIGANDS

In the early 1990’s the discovery that inhibitory KIR recognize self-HLA molecules fulfilled the

key proposition of the missing self hypothesis [53-55]. These studies demonstrated that receptors

expressed by NK cells, coined p58 for their molecular weight, inhibited lysis of tumor cells

expressing certain HLA molecules. Blocking the p58 receptors with antibodies abrogated the

inhibition. Analysis of the HLA ligands identified HLA-C as an inhibitory ligand and pointed to

the polymorphic residues at positions 77 and 80 in the HLA class I molecule as the cause for

differences observed in receptor interactions [55-57]. Similar observations were made for a p70

molecule interacting with HLA-B molecules [58;59]. The p58 and p70 molecules are now known

as the two or three domain inhibitory KIR (KIR2DL or KIR3DL), respectively.

After these early discoveries, more intricate experiments have been performed to describe KIR

binding to HLA, including determination of polymorphic and peptide repertoire effects on

binding affinities. KIR3DL2 recognizes HLA-A3 and-A11 while its fellow three domain KIR,

KIR3DL1, binds HLA-A and HLA-B antigens carrying the Bw4 epitope designated by residues

77,80-83 of HLA [59-63]. KIR2DL1 and KIR2DL2/3 recognize HLA-C based on the amino

acids present at position 80 of the HLA molecule. The HLA-C group 1 (C1) and HLA-C group

2 (C2) are defined by the presence of either an asparagine or lysine at position 80, respectively.

KIR2DL1 preferentially binds HLA-C2 molecules while KIR2DL2/L3 strongly binds HLA-C1

molecules. The binding observed for KIR2DL1 and KIR2DL2 appears slightly promiscuous as

low level binding with functional inhibition has been observed for KIR2DL1 binding to HLA-C1

and KIR2DL2 binding to HLA-C2 [50]. KIR2DL4 differs from the other long tail KIR ligands

20

as it is known to bind the non-classical class I molecule, HLA-G [64]. Interestingly, HLA-G is

only expressed by fetal trophoblasts supporting a role of NK cells during pregnancy. HLA-G

can also be secreted by malignancies suggesting a potential evasion mechanism whereby

KIR2DL4 signaling induces a pro-inflammatory, pro-angiogenic response by the NK cell thereby

providing nutrients to the cancer.

KIR2DS1 has been shown to bind the same HLA-C2 molecules as KIR2DL1, but with roughly

50% lower affinity. Site-directed mutagenesis experiments identified the amino acid change at

residue 70 of KIR2DS1 to be the cause of the lower affinity [65-67]. Similar to KIR2DS1/L1,

KIR2DS2 was expected to share the same HLA ligands as KIR2DL2/L3 resulting from the high

sequence homology of the extracellular domains of the receptors. However, it appears the

variations observed in the KIR2DS2 extracellular domains have abolished binding to HLA class

I ligands [55]. KIR3DS1 as well was expected to bind HLA-B molecules positive for the Bw4

epitope, but only a very rare allele, KIR3DS1*014, has demonstrated detectable binding to these

molecules [51]. KIR2DS4 appears to have been a product of gene conversion of the inhibitory

receptor KIR3DL2. This conversion contributes to ligand recognition as KIR2DS4 binds HLA-

A*1102 similar to KIR3DL2 [68]. Data has also shown KIR2DS4 binding to HLA-C*04 as well

as a putative non-HLA ligand demonstrated by KIR2DS4’s ability to bind to melanoma cells

lacking beta-2-microglobulin [68-70]. Along with KIR2DS2, ligands for KIR2DS3, KIR2DS5,

KIR3DL3, and KIR2DL5 remain unknown. While the lack of HLA interaction suggests these

receptors may recognize an alternate ligand, it is possible these receptors only bind HLA when a

specific antigenic peptide is presented by the HLA molecule.

21

Crystal structures are available for KIR2D and KIR3D binding of HLA and allow researchers to

visualize how polymorphisms within KIR or HLA as well as peptide repertoires may impact the

interaction of these proteins [71-73]. The D1 and D2 extracellular domains of the Type I KIR2D

interact with the HLA molecule towards the C-terminal end of the peptide. Multiple residues of

each protein impact the interaction. As an example, a minimum of 16 residues within the D1 and

D2 domains of KIR2DL2/L3 are directly involved in the interaction with HLA [74]. The

extracellular domains of these KIR form a V-shape, with the hinge angle differing among

different KIR2DL molecules. Polymorphisms shown to affect this hinge angle by molecular

modeling associate with altered binding affinities between KIR2DL2 and KIR2DL3 [50]. As the

KIR interaction domains surround the peptide binding groove, alterations of the peptide residues

can significantly impact KIR binding [75]. Similar to the binding of KIR2D to HLA, the D1

and D2 domains of KIR3DL span the C-terminal end of the peptide region of the HLA binding

pocket. The D0 domain extends alongside the conserved region of HLA toward β-2

microglobulin. Contrary to KIR2DL binding, the D1 interaction sites are altered for KIR3DL

binding accounting for the interaction with HLA-A and HLA-B. Understanding the interaction

dynamics between KIR and HLA, accounting for effects of allelic polymorphisms and peptide

repertoires, may help in correlating NK cell function with efficiency to clear certain infections or

malignancies.

22

1.6 COEVOLUTION OF KIR AND HLA

The KIR family of genes has rapidly evolved in simian primates [76]. All KIR evolved from one

of two genes, KIR3DL or KIR3DX. In cattle, the KIR3DX gene expanded whereas KIR3DL

remained a single copy gene. Contrary to cattle KIR, KIR3DL diversified in primates, leaving

KIR3DX as a single copy, nonfunctional gene [77]. Examination of complementary DNA

(cDNA) from chimpanzee and human defined three distinct KIR lineages (I-III) [78]. Lineage I

consists of the Type II two domain KIR, KIR2DL4 and KIR2DL5. Lineage II consists of the

majority of three domain KIR, KIR3DL1/S1, and KIR3DL2. The Type I two domain KIR

(KIR2DL1, KIR2DS1, KIR2DS2, KIR2DS3, KIR2DS4, KIR2DS5, KIR2DL2/L3) comprise the

lineage III KIR. A fourth lineage, lineage V, comprises only a single human KIR, the framework

KIR3DL3 gene with an unknown function [79]. The expansion of lineages I-III correlates with

the appearance of cognate HLA ligands within each species. Each lineage of KIR corresponds

with a different HLA ligand, as KIR2DL4 from lineage I interacts with HLA-G, lineage II KIR

mostly bind HLA-A and HLA-B molecules; whereas, lineage III KIR have expanded and

recognize HLA-C [50;60;63;64]. Studies comparing the KIR and HLA genetics of humans to

Old World monkeys as well as Asian and African apes have revealed expansion of the KIR genes

based on HLA diversification and provide models for selection of HLA-C as the major ligand of

KIR in humans.

The genome of Old world monkeys contains HLA-A and HLA-B loci as well as a diversity of

lineage II KIR molecules with little lineage III diversity. These characteristics lead researchers

to hypothesize the KIR of the old world monkey, rhesus macaque, would recognize HLA-A and

23

HLA-B molecules better compared to HLA-C of humans. Surprisingly, the macaque KIR bound

best to human HLA-C molecules followed by HLA-B and only exhibited binding to HLA-A

molecules carrying the Bw4 epitope. The explanation for such an interaction is that the rhesus

MHC-B locus carries an HLA-C1 epitope possibly causing the KIR to exhibit more promiscuity

for MHC ligands [80]. The HLA-B and HLA-C allelic products resulted from a duplication of a

MHC-B-like ancestor. While one gene remained similar to MHC-B, the other has undergone

natural selection in hominoids to become HLA-C. This C1 epitope has nearly been erased from

human HLA-A and HLA-B allelic products with only two HLA-B alleles, HLA-B*46 and HLA-

B*73, still carrying the C1 epitope [81]. As a result, the lineage II rhesus macaque KIR broad

recognition of HLA-C may emulate the ability of these receptors to recognize the MHC-B

antigens carrying the C1 epitope.

Similar to the rhesus macaque, the gibbon which is a member of the Asian apes, lacks an

ortholog of HLA-C and also lacks HLA-G [82] . Corresponding to the absences of these loci,

gibbons lack diversification of lineage III KIR and KIR2DL4 has been deleted or lacks ITIMs

necessary for efficient inhibitory signaling [83].

Differing from the gibbon, the HLA-C locus first becomes present in another member of the

Asian apes, the orangutan [84]. This manifestation appears to be the initiation of selecting KIR

ligands for HLA-C as lineage III KIR begin to diversify within the orangutan [85]. Binding

assays again show that the early functional homologs of KIR likely had promiscuous binding as

the orangutan KIR, Popy2DLA, recognized HLA-C1/C2, Bw4 positive allelic products and

HLA-A3/A11 which represent all the major known ligands of human KIR. Site-directed

24

mutagenesis analysis identified key residues that play a role in determining ligand specificity.

Residue 44 of KIR was one of these key amino acids. Orangutans either have a lysine or

glutamate at this position while chimpanzees carry either a methionine or glutamate. The

orangutan KIRs with glutamate at position 44 are capable of binding both C1 and C2. This

broad specificity is abrogated in the chimpanzee KIR carrying glutamate as a result of an amino

acid change from phenylalanine to cysteine at position 45 allowing for only recognition of C1 by

these KIR. The presence of methionine at position 44 in the other chimpanzee KIR allow for

specific recognition of HLA-C2. The orangutan KIR with glutamate at position 44 acts as an

important intermediate to the evolution of C2-specific KIR. Human KIR either have a lysine or

methionine at position 44 along with phenylalanine at position 45 correlating with HLA-C1 or

C2 ligand specificity, respectively [86]. Variations at positions 44 and 45 of KIR have lead to

the evolution of independent C1 and C2 specific receptors in chimpanzees and humans in

conjunction with evolution of HLA-C2 molecules in humans [87].

As inhibitory KIR have evolved to develop specificity for HLA, it appears stimulatory KIR have

accumulated changes in their extracellular domains that disrupt HLA binding. Orangutans and

gorillas have activating receptors that exhibit equal affinity for their ligands as their inhibitory

counterparts [86]. Chimpanzees also have an activating receptor which binds with similar

affinity to C1 and C2 as its inhibitory counterpart but also have an activating KIR with reduced

affinity relative to its inhibitory counterpart [88]. In humans KIR2DS4 is the only activating

receptor directly inherited from the chimpanzee. KIR2DS1 and KIR2DS2 have evolved from

their corresponding inhibitory receptor genes. In humans, no activating KIR exhibits binding to

25

HLA-C1, but KIR2DS1 does display binding to HLA-C2 albeit at a lower avidity compared to

KIR2DL1. KIR2DS2, which evolved from the C1-specific inhibitory receptors, carries a

tyrosine at position 45 which abolishes the ability to bind HLA [55]. While the trend for

functionally deleting HLA specific activating receptors is amplified in humans, diminished or

complete loss of binding is also seen in chimpanzee and orangutan KIR. The loss of HLA

ligands may prevent autoimmunity with KIR binding only occurring with specific antigenic

peptides or activating KIR in humans have evolved to recognize stress-related MHC-like

proteins that are typically upregulated upon infection or in cancers.

1.7 NK CELL EDUCATION

Similar to T cells and B cells, NK cells appear to be educated in order to prevent NK cell

associated autoimmunity, but the mechanisms remain unclear. An early report suggested all

human NK cell clones expressed at least a single inhibitory receptor capable of binding HLA

[89]. These results presented an obvious mechanism of how NK cells are tolerant towards

normal self cells. However, through further investigation, it is now understood and accepted that

a significant number of mature NK cells lack expression of any HLA-associated inhibitory

receptors [90]. Such results have lead to three distinct models by which NK cells can be

educated to become tolerant to self in the absence of HLA-mediated inhibition (Figure 1.5).

The licensing model postulates NK cells are inherently anergic and rely on an interaction of an

inhibitory receptor with a self HLA molecule in order to “arm” the NK cell. The theory suggests

NK cells lacking the appropriate inhibitory receptors are hypo-responsive. Support for this

26

theory comes from both mouse and human studies. NK cells from mice deficient in MHC

expression are found to be incapable of killing tumor cell lines devoid of MHC class I and of

rejecting MHC class I deficient bone marrow [91]. In humans, NK cells lacking HLA specific

inhibitory receptors fail to become activated by class I deficient normal cells, tumor cells or by

antibody stimulation of various activating receptors [92;93]. These studies demonstrated,

without prior engagement of an inhibitory receptor, the activating signaling cascades are not

functional. Further analysis showed engagement of MHC by an inhibitory receptor alone is not

sufficient to invoke competent activation. Mutation of the ITIM located within the cytoplasmic

region of the inhibitory receptor resulted in the same hypo-responsive phenotype observed in NK

cells lacking MHC inhibitory receptors [94]. These data suggest signaling from the ITIM after

ligand engagement is necessary for efficient structuring of the activating pathways which

suggests that this initial ligand interaction may induce alternate signaling pathways from the

inhibitory receptors.

Opposing the licensing model is the “disarming” model. This theory suggests NK cells are

inherently active, but chronic engagement of activating receptors in the absence of binding of

MHC-specific inhibitory receptors renders the NK cell hypo-responsive. Consistent with the

arming model, this theory stems from observations in both murine and human models. In the

murine model, analysis of the stimulatory interactions between Rae-1 and NKG2D as well as

m157 and Ly49H support the disarming model. Rae-1 is expressed in the mouse embryo but is

then silenced before birth. In adults, Rae-1 expression serves as a marker of stress and can also

be expressed by malignant cells making these cells targets of NK cell lysis through activation of

27

NKG2D. Immunocompetent mice efficiently reject transplantation of hematopoietic cells from

Rae-1+ transgenic mice. However, NK cells from the Rae-1

+ transgenic mouse are tolerant to

Rae-1 positive tissues and do not react against targets expressing NKG2D ligands [95;96].

Similar results were observed in m157+ transgenic mice. The mouse cytomegalovirus (MCMV)

protein, m157, is a ligand for the activating receptor, Ly49H. Ly49H is unresponsive in NK cells

of m157+ transgenic mice [97]. In humans, the activating KIR2DS1 has been demonstrated to

play a role in NK cell education. As mentioned, KIR2DS1 recognizes a group of potential self

HLA molecules, HLA-C2. An increasing trend of responsiveness of NK cells expressing

KIR2DS1 was observed in donors homozygous for HLA-C2, heterozygous (C1/C2) or

homozygous (C1/C1). The education effect imparted by KIR2DS1 in HLA-C2 homozygous

donors was sufficient to overcome potential educational effects through other inhibitory

receptors [98]. These results support constant recognition of an activating ligand results in

tolerance of the particular ligand in order to prevent autoimmunity.

The third model, the rheostat model, is a combination of the aforementioned theories. Rather

than thinking of NK cell education as a binary on or off state, the rheostat model considers the

signals from both inhibitory and activating receptors and combines the strengths of binding into a

more quantitative analysis to determine NK cell education. The education of NK cells is

therefore influenced by polymorphisms found in both MHC allelic products as well as inhibitory

receptors. In the mouse, Ly49A is an inhibitory receptor with high affinity for H2Dd and low

affinity for H2Db. Based on the difference in binding affinity, NK cells positive for Ly49A from

mice that only carried H2Dd were more responsive compared to Ly49A-positive NK cells from

28

mice positive only for H2Db. The responses observed also showed the NK cells from the H2D

d

mice were more likely to be polyfunctional as evidenced by secretion of cytokines as well as by

release of cytotoxic granules [99]. Observations of the impact of different inhibitory receptors

with altered affinity for a single MHC ligand revealed similar results [100]. The quantity of

inhibitory signaling also affects activation capability as NK cells expressing two inhibitory

receptors were more responsive compared to cells expressing either receptor alone [101]. The

rheostat model also accounts for observed complexities where NK cell activity is modulated by

altered cytokine environments caused by infection or change in MHC expression occurring

during haploidentical hematopoietic stem cell transplantation [102;103].

Even though NK cells were discovered nearly 30 years ago, the mechanisms of function for these

cells are not well understood. Unlike T-cells and B-cells, the sites and mechanisms for

maturation and education of NK cells remain unclear. The data obtained to support each of the

three education models described above entice more controversy and questions with regards to

how NK cells develop tolerance to self. As research continues, it is important to fully understand

these mechanisms as they may be applied to clinical settings including, but certainly not limited

to, bone marrow transplantation.

29

30

Figure 1.5 Three models of NK cell education

A, The arming model describes NK cells as incapable of activation until recognition of HLA by

an inhibitory receptor. In the absence of an inhibitory signal, the cells remain hypo-responsive in

order to prevent NK cells from causing autoimmunity. B, Contrary to the arming model, the

disarming model dictates NK cells are inherently active and are disarmed by the lack of

interaction between HLA and an inhibitory receptor. C, The rheostat model introduces a

quantitative analysis of inhibitory and activating signals to determine NK cell education and

activation capabilities. In this model, differences of receptor affinity for HLA affect the

education process.

31

1.8 NK CELLS AND CANCER THERAPEUTIC STRATEGIES

Resulting from their innate ability to kill tumor cells, NK cell responses against tumor cells

continue to be investigated in order to develop efficient immunological therapies for cancer.

In vivo evidence in both mouse and humans implicate the importance of NK cells in elimination

of tumors [104;105]. As previously discussed, NK cells recognize HLA molecules presented on

the surface of normal cells as an inhibitory mechanism, preventing autoimmunity. However,

many malignancies reduce HLA expression through multiple mechanisms in order to evade

recognition by members of the adaptive immune system [106-108]. With the reduction of HLA

surface expression along with increased expression of stress proteins, the tumor cells become

more susceptible to NK cell lysis. A number of activation receptors including NKp30, NKG2D,

DNAM-1, and KIR2DS1 have been described for their role in tumor cell recognition.

The B7 family member, B7-H6, is the ligand for NKp30. Expression of B7-H6 has not been

observed in normal tissues but has been detected under stress induced conditions as well as in a

variety of tumor types [109] . Expression levels of the NKG2D ligands, MICA/B, Rae-1 and

ULBP proteins, have also been correlated with NK cell responses against malignancies [110].

The ligands for DNAM-1, CD155 and Nectin-2, are also expressed by an array of tumor types

[111]. KIR2DS1-positive NK cells have shown to be effective in attacking HLA-C2 positive

leukemic cell blasts [112]. The increase in expression levels of such activating ligands or

possibly specific tumor peptide residues presented by HLA-C molecules shift the functional

balance of NK cells towards activation.

32

Therapeutic strategies focused on enhancing NK cell activity in order to eradicate tumors have

been on-going since the 1980’s. In these early studies, interleukin-2 (IL-2) was administered

subcutaneously to activate NK cells in vivo or NK cells were stimulated ex vivo with IL-2 and

transferred back into patients. This treatment had modest effects in patients with advanced renal

cell carcinoma or melanoma [113]. More recent studies have combined treatment with high dose

cyclophosphamide and fludarabine with constant IL-2 administration and have resulted in a more

stable expansion of donor NK cells within the recipient [114]. This expansion was also

correlated with increased serum levels of IL-15 in these patients, consistent with the previously

defined role for IL-15 in NK cell development. Even with this stable expansion, the clinical

effectiveness for NK cell transfer is far from optimal. More strategies are being employed

including using antibodies to block inhibitory receptors on NK cells or administering different

cytokines such as IL-21or IL-15 to enhance NK cell activity [115;116]. Several phase I and II

clinical trials are underway investigating the safety and efficacy of IPH2101, a novel humanized

antibody that enhances NK cell activity by blocking inhibitory KIR, for use in treating multiple

myeloma and acute myeloid leukemia patients [117]. It is possible this antibody could be used

in addition to autologous or allogeneic NK cell transfer regimens in order to break self-tolerance

allowing for enhanced efficiency in the clearance of the malignancy.

Another mechanism by which NK cells target tumors is known as antibody dependent cellular

cytotoxicity (ADCC). This mechanism allows for specific lysis of tumor cells as the variable

(Fab) region of an antibody targets a tumor specific antigen while the constant (Fc) region binds

to Fc receptors (CD16) expressed by NK cells. Two approved clinical antibodies used for solid

33

and hematological malignancies are Trastuzumab and Rituximab, respectively. In a mouse

study, variation of the Fc regions of Trastuzumab or Rituximab resulting in altered Fc-receptor

binding caused a correlating effect on tumor cell growth implying ADCC as a major mechanism

influencing tumor control through these antibodies [118]. Similar evidence has been derived in

human studies as a polymorphism at position 158 within the Fc-receptor causes a disparity of

affinity for antibodies. Homozygosity for valine at this position allows for increased binding of

antibody compared to being heterozygous or homozygous with a phenylalanine at this position.

These in vitro data have directly correlated to clinical outcome as patients homozygous for

158V/V respond more favorably to antibody treatment compared to patients who are

heterozygous (158V/F) or homozygous 158F/F reinforcing that ADCC plays an important role

in the efficacy of Rituximab and Trastuzumab [119;120]. As these antibodies have shown to be

successful in the clinic, it is important to understand the complete mechanisms that drive their

successes and failures in order to develop more effective therapies.

1.9 NK CELLS AND KIR IN HEMATOPOIETIC STEM CELL TRANSPLANTATION

Allogeneic hematopoietic stem cell transplantation has proven to be a successful treatment for

hematological malignancies. In order to aid reconstitution of the grafted cells and prevent

treatment-related toxicity and graft versus host disease (GVHD), donors are identified based on

HLA-matching. Unfortunately, the probability of identifying a sibling donor with the identical

HLA genotype is only 25%. Even with the advent of global marrow registries, identifying a

perfectly matched HLA donor is difficult especially for ethnic minorities [121]. This low

probability has lead to investigations of alternate donors such as mismatched unrelated or

34

umbilical cord blood or haploidentical donors. The adoption of haploidentical hematopoietic

transplants has significantly increased the probabilities of identifying favorable donor-recipient

pairs. However, the early investigations highlighted the complexity of effectively treating

patients with haploidentical transplantations [122]. As investigations have continued,

conditioning regimens have been identified which result in optimal results of lower GVHD while

maintaining effective leukemia clearance. These regimens result in efficient engraftment and

rapid reconstitution of the NK cell population [105].

In a seminal study, Ruggeri et al. examined the potential effect of KIR ligand incompatibility in

the graft versus host direction on disease clearance and overall survival in patients who received

T-cell depleted HLA mismatched hematopoietic transplants from related donors. In this study,

KIR ligand incompatibility was significantly correlated with greater relapse free survival and

decreased risk of GVHD in patients with acute myeloid leukemia (AML). Interestingly, no

effect was observed in the study in patients with acute lymphoid leukemia (ALL). The decrease

in GVHD was attributed to NK cell killing of host antigen presenting cells (APCs) which would

interact with engrafting T-cells causing an immune response. These results stimulated robust

interest and investigation resulting in controversial findings as to whether KIR ligand

incompatibility has a role in relapse free survival in related and unrelated hematopoietic

transplantations [123-125].

Although the findings were controversial, the conditioning regimens differed study to study with

altered presence of T-cells within the transplant. Analysis of KIR expression in unmanipulated

bone marrow (UBM) transplantation versus T-cell depleted (TCD) bone marrow transplantation

35

revealed KIR expression was significantly decreased in the UBM transplantations. Combining

results from UBM and TCD transplantation correlated KIR expression with survival [126].

These results infer that the presence of T-cells in the graft negatively impacts NK cell

development and maturation. The mechanisms of how NK cells in a KIR ligand mismatch

setting appear to be beneficial are complex and not well understood. The NK cells in a T-cell

deplete unrelated HLA mismatch transplantation develop and are educated within the recipient

after engraftment. Only NK cells expressing KIR capable of interacting with HLA of the

recipient are responsive in ex vivo stimulation assays through evaluation of IFN-γ production. In

contrast education through inhibitory KIR interactions does not appear to be necessary for

degranulation, rather only expression of NKG2A is required for the NK cell to develop cytotoxic

function [102]. Stimulation of the NK cells with IL-15 enhanced production of IFN- γ, as well

as the capability for degranulation independent of transplant setting. The enhanced activity

following IL-15 treatment warrants further investigation of cytokine treatment post

transplantation in the effort to enhance the graft versus leukemia response by NK cells.

Analysis of KIR genotypes independent of HLA genotypes has shown that donors with KIR B/x

haplotypes have significantly enhanced relapse free survival after unrelated hematopoietic

transplantation. The KIR haplotype of the recipient had no impact on this finding [127].

Further analysis of the KIR B haplotype identified donors homozygous for KIR cen-B

haplotypes exhibited the best decrease in relapse compared with either cen-A/A or cen-A/B. The

exact genes and the underlying mechanisms by which these beneficial observations occur are

unknown. There has not been a clear direct correlation between a single activating KIR and

36

effectiveness of transplantation [128]. Presence of stimulatory KIR in the cen-B haplotype

structure such as KIR2DS2 or KIR2DS5 may recognize a ligand expressed by the AML blasts or

the presence of inhibitory KIR may affect education rates after engraftment creating an efficient

environment for NK cell lysis of the leukemia. Even though the mechanisms remain unclear, the

data support the identification of KIR B donors as ideal candidates if the choice is present.

1.10 NK CELLS IN PREGNANCY

Pregnancy introduces an immunologic paradox in which maternal immune cells interact with

invading, haploidentical fetal cells. Although originally thought to be immunologically inert,

ensuing research has defined immunological interactions at the maternal, placental interface.

These interactions primarily occur between invading fetal trophoblasts and resident uterine NK

cells.

The cytokine secreting CD56 bright NK cells are the predominant immune cell at the site of

implantation, accounting for up to 70% of lymphocytes during the first trimester of placental

development [129]. General immune-phenotyping of these uterine NK (uNK) cells revealed

unique receptor combinations compared to those expressed on peripheral NK (pNK) cells.

Contrary to the CD56 bright pNK cells, CD56 bright uNK cells express KIR and

the ability of the uNK cells to secrete cytokines such as vascular endothelial growth factor

(VEGF), granulocyte macrophage stimulating factor, IFN-γ, and macrophage inflammatory

factor 1 alpha (MIP-1α) is also enhanced compared to pNK cells [130;131].

37

The extravillous trophoblast (EVT) cells are invading fetal cells that interact with maternal

immune cells. The EVT cells invade deep through the uterine wall during decidualization, where

they replace maternal endothelial cells along the spiral arteries, resulting in increased blood flow

to the intervillous space [132] . Failure of proper invasion by these trophoblast cells leads to

inadequate transport of nutrition to the fetus possibly resulting in clinical complications such as

pre-enclampsia and still birth [133].

EVT cells express several HLA molecules, including HLA-E, HLA-G, and HLA-C that serve as

ligands to multiple NK cell receptors [134]. HLA-E serves as a ligand for the inhibitory

complex CD94/NKG2A that may function to prevent NK cell lysis of the invading EVT cells.

HLA-G is only expressed by fetal EVT cells and, while the mechanisms are unclear, it is

hypothesized that upon binding of HLA-G, KIR2DL4 initiates a pro-inflammatory, pro-

angiogenic response furthering the development of the spiral arteries. HLA-C receptors on NK

cells are expressed at higher levels on uterine NK cells during the first trimester compared to

expression levels on peripheral NK cells during the same time points. Epidemiological research

focusing on the HLA-C alleles and KIR genotypes has resulted in associations of genotype

combinations with clinical outcome of pregnancy. These studies have shown importance in

combining genotyping data from the mother, father and fetus. The combination of maternal KIR

A/A haplotype with a fetus carrying a C2 group is associated with increased risk of pre-

eclampsia. This risk is increased if the mother is homozygous C1 and the C2 antigen is of

paternal origin. A protective effect is seen with the presence of the Tel-B haplotype including

the activating receptor KIR2DS1 [135]. The mechanism underlying this effect is unclear but

38

may suggest a similar role as KIR2DL4 may serve in inducing a pro-inflammatory/pro-

angiogenic effect through activation of KIR2DS1. The data to date show multiple NK receptor

ligands on the EVT cells suggest an important role for NK cells in proper development of fetal

blood supply, but the mechanisms remain poorly understood.

1.11 KIR AND HIV

NK cells function to eradicate infections through both direct lysis of infected cells, as well as

activating and recruiting an adaptive immune response through release of cytokines. The first

association of KIR and viral infection was made in a cohort of HIV patients [136]. At the time,

previous data suggested patients homozygous for HLA-Bw4 allelic products experienced slower

loss of CD4+ T-cell count [137]. Given this association, a genetic association was sought for

KIR3DL1 or KIR3DS1 since KIR3DL1 was known to bind HLA-Bw4 and KIR3DS1 was

expected to recognize the same ligand. This initial study showed delayed progression to AIDS in

HIV patients positive for KIR3DS1 and for HLA-Bw4 antigens with isoleucine at position 80

(Bw4-80Ile). These results appear on the surface to be clear as one would expect a stimulatory

receptor combined with its cognate ligand would lead to enhanced capability for activation.

However, this conclusion is complicated since only a very rare allele, KIR3DS1*014, has

exhibited binding to HLA-Bw4 [51]. As peptide repertoires presented by HLA affect KIR

binding, it is possible KIR3DS1 binding is enhanced by HIV specific peptides yet to be

identified.

39

A later study showed a similar association of KIR3DL1 with HLA-Bw4 with delayed

progression to AIDS. The observed protective effect was greater compared to the effect

associated with KIR3DS1. The most protective effect was observed in patients carrying

KIR3DL1*004 and HLA-B*57. This result was surprising as this allele of KIR3DL1 exhibits

very minimal cell surface expression and is retained within the cell [47]. KIR3DL1*004 may be

able to act intracellularly similar to KIR2DL4 in response to soluble HLA-G. The other allelic

products of KIR3DL1 have been characterized as “high” or “low” determined by their level of

surface expression as well as their inhibitory capability upon recognition of ligand. The

KIR3DL1 high alleles in conjunction with HLA-Bw4 were associated with greater protection

compared to the KIR3DL1 low allelic products. This disparity is attributed to greater

polyfunctional activity demonstrated by KIR3DL1 “high” versus “low” positive NK cells in

patients carrying HLA-Bw4 [138].

The functional explanation for the presence of an inhibitory receptor associated with decreased

viral progression is unclear. One explanation is the increase in viral antagonism is due to

enhanced NK cell activity attributed to a more educated NK cell in the presence of the KIR3DL1

“high” alleles. The stronger recognition of the HLA-Bw4 antigen, according to the rheostat

model, results in enhanced activation capabilities by these cells. Another potential mechanism is

KIR3DL1 being responsible for surveillance of HLA expression. If HLA expression is

decreased during HIV infection, then the KIR3DL1 positive cells would be more susceptible to

attacking these cells. KIR3DL1 recognition could also be lost or altered due to peptide

presentation by the HLA, causing the NK cell to be skewed towards activation. The effect of

40

peptide presentation affecting inhibitory recognition of HIV infected cells was demonstrated to

be a plausible explanation with regards to KIR2DL2. The results demonstrated HIV sequence

polymorphisms can enhance binding of inhibitory KIR to HLA presented by infected CD4+ T-

cells creating an evasion mechanism to prevent NK cell lysis of infected cells [139]. This study

is the first to show viral adaptation in response to NK cell mediated pressure. The continued

investigation into the association of KIR genetics and AIDS progression is likely to further

elucidate mechanisms by which peptide repertoires presented by HLA influence binding by KIR.

1.12 STATEMENT OF PURPOSE

Epidemiological data have identified associations of stimulatory KIR with a variety of clinical

effects including roles in effective graft versus leukemia responses in bone marrow

transplantation, slower progression to AIDS, lower risk of pre-eclampsia during pregnancy, as

well as predisposition to autoimmune diseases such as psoriatic arthritis [105;135;136]. With

clinical benefits resulting from either the presence or absence of a particular stimulatory KIR

gene, it is important to understand the biology of how these receptors mature and are trafficked

in order to develop and understand therapeutic strategies in the clinic. For instance, in the

clinical setting for bone marrow transplantation or adoptive NK cell therapy, enhanced levels of

KIR2DS1 or KIR2DS4 on the cell surface as well as increased percent of NK cells positive for

these receptors may aid in effectively clearing the leukemia or melanoma, respectively.

However, for patients suffering from autoimmunity associated with a KIR2DS molecule,

decreasing the surface expression of the receptor may help prevent further damage to normal

41

tissue. Therefore strategies focused on understanding how to manipulate KIR2DS surface

expression may aid in producing a favorable clinical outcome.

Multiple studies have shown an impact of adapter molecules on the surface expression of their

cognate receptors [34;36]. However, while many studies have analyzed the signaling function of

the DAP12, KIR2DS interaction, none to date have examined the potential impacts of the

interaction on KIR2DS surface expression. Many of the studies analyzing the clinical impact of

NK cell function have focused on utilizing NK cells to combat malignant progression in the

clinic in conjunction with interleukin treatments. The protocols for enhancing NK cell activity

call for either in vivo or ex vivo stimulation and proliferation of the NK cells with different

interleukin treatments involving IL-2, IL-12, IL-15, and IL-21 which, in certain combination

such as IL-15 and IL-21, are known to decrease DAP12 expression. It is important to understand

what functions DAP12 plays in stimulatory KIR biological development and how DAP12

expression can indirectly impact KIR function. Investigating the impact of this interaction is

also important to understand how KIR2DS properties on NK cells of Nasu Hakola patients who

lack functional DAP12 genes will be affected. This study aims to understand how DAP12

impacts KIR2DS surface expression and seeks to determine the mechanisms by which this is

accomplished in order to provide biological mechanisms that can help explain clinical outcomes

and cellular functions observed in conditions where DAP12 expression is either lost or

compromised.

42

This research was predicated on the hypothesis that DAP12 played a significant role in KIR2DS

surface expression through impacts on trafficking the receptors to the cell surface and stabilizing

KIR2DS molecules at the cell surface.

The specific aims of the project were:

1. To identify DAP12’s impact on KIR2DS surface expression.

2. To determine if DAP12 affected KIR2DS maturation and trafficking to the cell surface.

3. To analyze a potential stabilizing effect of the interaction between DAP12 and KIR2DS

molecules.

4. To assess whether DAP12 played a role in trafficking KIR2DS molecules from the cell

surface to specific internal compartments.

43

DAP12 IMPACTS TRAFFICKING AND SURFACE STABILITY OF KILLER CELL

IMMUNOGLOBULIN-LIKE RECEPTORS ON NATURAL KILLER CELLS

2.1 MATERIALS AND METHODS

2.1.1 Cell lines and culture

The NKL cell line was a gift of Dr. Francisco Borrego (NIAID, Rockville, MD, USA) and was

maintained in RPMI 1640 containing 10% FBS, 1 mM L-glutamine, 10 mM HEPES, 1 mM

sodium pyruvate and 100 U/mL IL-2 (BD Biosciences, Franklin Lakes, NJ, USA). Peripheral

blood mononuclear cells (PBMCs) were obtained from SeraCare Life Sciences (Milford, MA,

USA) and genotyped for KIR to identify a KIR2DS4*001 positive donor as previously described

[140]. PBMCs were cultured in RPMI 1640 containing 10% FBS, 1 mM L-glutamine, 10 mM

HEPES, 1 mM sodium pyruvate and 2% spent culture media from the myeloma cell line, J558L.

HEK293T cells were a gift of Dr. Todd Waldman (Georgetown Medical Center, Washington,

DC, USA) and were maintained in DMEM with 10% FBS, 1 mM L-glutamine, 10 mM HEPES,

and 1 mM sodium pyruvate.

2.1.2 DNA constructs

The cDNA encoding KIR2DS1*002 and KIR2DS4*001were cloned into the expression vector,

pEF-DEST51 (Invitrogen Life Technologies Carlsbad, CA, USA) as previously described

[141;142]. The cDNA encoding KIR2DS2*002 and FcεRIγ was obtained from Origene

Technologies Inc. (Rockville, MD, USA) and the DAP12 cDNA was obtained from Invitrogen.

These cDNA were amplified using the following primers: 2DS2-F

44

(CACCATGTCGCTCATGGTC) and 2DS2-R (TCCTGCGTATGACACCTCCTG) for

KIR2DS2 with a glycine in place of the stop codon, DAP12-F

(CACCATGGGGGGACTTGAAC) and DAP12-R (TCATTTGTAATACGGCCTC), and

GAMMA-F (CACCATGATTCCAGCAGTGGTC) and GAMMA-R

(TCACTGTGGTGGTTTCTCATG) for FcεRIγ. These amplicons were cloned into the

expression vector, pEF-DEST51, through Gateway Technology using the entry vector,

pCR8/GW/TOPO following manufacturer’s protocol (Invitrogen). Following manufacturer’s

protocol for Quikchange II (Stratagene, La Jolla, CA), site-directed mutagenesis was performed

to insert nucleotides encoding the HA-tag, YPYDVPDYA, immediately following the leader

sequence of DAP12 and FcεRIγ. The LAMP1-GFP and Rab5-GFP expression vectors (pCMV6-

AC-GFP) were obtained from Origene Technologies Inc. All constructs were prepared as per

manufacturer’s instructions using the HiSpeed Plasmid Maxi Kit (Qiagen, Valencia, CA, USA).

2.1.3 KIR expression and trafficking

For analysis of KIR surface expression on transfected NKL cells, NKL cells (107) were co-

transfected with 5 μg of a KIR2DS encoding construct and either 5 µg of empty vector (pEF-

DEST51) or 5 μg of the DAP12 encoding construct using a Nucleofector II (Lonza, Cologne AG,

Cologne, Germany) with the protocol, O-017. KIR surface expression was determined by flow

cytometry using phycoerythrin (PE) conjugated antibodies specific for CD158a/h (KIR2DS1),

CD158b/j (KIR2DS2) and CD158i (KIR2DS4) (Beckman Coulter, Fullerton, CA, USA).

Unless noted, cells were collected 16 hours post-transfection and externally stained with the

appropriate antibody for 30 minutes at 4°C. Total KIR expression was also determined by flow

45

cytometry using a FITC conjugated V5 antibody (Invitrogen) against the internal, C-terminal V5

tag of KIR as previously described [141]. Variation in total KIR expression was controlled for

by representing KIR surface expression as a ratio to total expression. To compare results

between experiments the ratios were normalized to the ratios obtained for KIR2DS co-

transfected with DAP12 or FcεRIγ. For the trafficking experiment beginning 10 hours post-

transfection, cells were treated for 6 hours with 100 μg/mL cycloheximide (Sigma Aldrich, St.

Louis, MO, USA) or 1X brefeldin A (eBioscience, San Diego, CA, USA). Cycloheximide and

brefeldin A were replenished every 2 hours for the duration of the experiment. After the

treatment, KIR2DS surface expression was determined by flow cytometry as described above.

2.1.4 Gene silencing and relative quantitative RT-PCR

Primary PBMCs from individuals positive for KIR2DS4*001 were expanded in culture for 7

days as described above. PBMCs or KIR2DS4 positively sorted cells (3x105) were cultured in

serum free Accell siRNA delivery media (Dharmacon siRNA Technologies) with either 1 μM

non-silencing siRNA or 1 μM DAP12 specific siRNA (Dharmacon) for 72 hours. KIR2DS4

surface expression was determined by flow cytometry as described above and total RNA was

isolated following manufacturer’s protocol with a RNeasy Mini kit (Qiagen). cDNA was

generated from approximately 200 ng of total RNA using TaqMan Reverse Transcription

Reagents from Applied Biosystems (Foster City, CA, USA) as per manufacturer’s instruction.

DAP12 and β-actin mRNA levels were determined by qRT-PCR using a StepOne Plus RT-PCR

instrument using DAP12 and β-actin specific pre-designed probes and primer sets and TaqMan

gene expression master mix as per manufacturer’s protocol (Applied Biosystems). β-actin

46

mRNA levels served as the internal control. The relative quantities of DAP12 mRNA obtained

after targeted siRNA treatment were normalized to the values obtained following scrambled

siRNA treatment.

2.1.5 Immunoprecipitation and Western Blot

NKL cells were co-transfected with a KIR2DS construct and either empty vector or DAP12 as

previously described. Sixteen hours post-transfection, cells were lysed for 30 minutes at 4°C

using 10% NP-40 in PBS. Lysates were immunoprecipitated using a DAP12 specific antibody

(Santa Cruz Biotechnology, Santa Cruz, CA, USA). Immunoprecipitants or whole cell lysates

from NKL cells or PBMCs were electrophoresed on 4-15% polyacrylamide Tris-HCl ready gels,

transferred to nitrocellulose membranes, and the membranes were blocked using 5.0% nonfat dry

milk as previously described [142] (Bio-Rad, Hercules, CA, USA). The membranes were

probed with a 1:5,000 dilution of a V5-specific (Invitrogen) or HA-specific (Sigma) primary

antibody or a 1:500 dilution of a DAP12 specific antibody (Santa Cruz Biotechnology, Santa

Cruz, CA, USA) followed by a secondary HRP conjugated antibody specific for mouse heavy

and light chains (Jackson Immunoresearch, West Grove, PA, USA). Proteins bands were

detected using ECL detection (Amersham Biosciences, Piscataway, NJ, USA) following

manufacturer’s protocol.

2.1.6 Internalization analysis

The internalization of KIR2DS molecules on NKL cells co-transfected with KIR2DS plus empty

vector or DAP12 was analyzed as previously described [143]. Briefly, transfectants were probed

47

16 hours post-transfection with the corresponding PE-conjugated KIR antibody and cultured at

37°C for 0, 10, 30, and 60 minutes. Samples were collected at indicated time points and were

either untreated or treated with 200 µl of PBS containing 100 mM glycine and 100 mM NaCl

(pH 2.5) for 2 minutes to strip away externally bound antibodies. Cells were washed and the

amount of receptor internalization was analyzed by flow cytometry. The percent KIR2DS

internalized was calculated using the following equation: 100 x (MFIexp – MFIstripped)/(MFItotal –

MFIstripped) where MFIexp represents internalized KIR staining at the time point taken. MFItotal

and MFIstripped are the values representing KIR expression in samples not treated with the acidic

solution or samples stained and then immediately treated with the acid at the “zero” time point

representing background values, respectively.

2.1.7 Confocal microscopy

HEK293T cells (5x10^4) were plated on a 35-mm glass Fluorodish (World Precision

Instruments, Sarasota, FL, USA). The next day, the cells were transfected with 1 µg LAMP1-

GFP or Rab5-GFP, 1 µg KIR2DS1, and 1 µg DAP12 or empty vector (pEF-DEST51) constructs

with FuGENE 6 (Roche, Indianapolis, IN, USA). Sixteen hours post-transfection, the cells were

probed with an allophycocyanin (APC)-conjugated antibody specific for KIR2DS1 (Miltenyi

biotech, Bergisch Gladbach, Germany) for 2 hours at either 4°C or 37°C. After the antibody

incubation, cells were fixed using a 4.0% paraformaldehyde solution (Pierce Scientific,

Rockford, IL, USA). The samples were viewed with an Olympus Fluoview-FV300 laser

scanning confocal system.

48

2.2 RESULTS

Analysis of the impact of DAP12 on KIR2DS surface expression on NKL cells

To analyze the potential impact of DAP12 on two domain stimulatory KIR (KIR2DS)

expression, KIR surface expression was determined on NKL cells in the presence or absence of

exogenous DAP12. The NKL cells, which do not express endogenous KIR, were co-transfected

with a singular KIR2DS construct and either empty vector or DAP12. The KIR2DS molecules

analyzed for surface expression were KIR2DS1, KIR2DS2, and KIR2DS4. KIR2DS3 was

excluded from the study as we have previously demonstrated only minimal surface expression of

this receptor [142]. Due to the lack of a specific antibody, KIR2DS5 was also excluded based on

observations that an N-terminal tag impacted trafficking patterns of the receptor. All KIR2DS

molecules analyzed were expressed at the cell surface as detected by flow cytometry. As a

representative example, surface expression of KIR2DS4 is shown (Figure 2.1A). As exemplified

by KIR2DS4, surface expression of all the KIR2DS molecules was detected independent of

exogenous DAP12, but the level of surface expression was greatly enhanced in the presence of

exogenous DAP12. We suspect the receptors, when transfected with empty vector, are expressed

at the cell surface independent of DAP12 as the level of endogenous DAP12 protein expression

in the NKL cells used in this study was undetectable by Western blot (data not shown). The

observed difference in the level of surface expression between transfection conditions was not

due to varation in protein production as total KIR2DS4 expression was comparable independent

of exogenous DAP12 (Figure 2.1B). To account for any variation in protein production across

experiments, the data are expressed as a ratio of KIR surface expression to total KIR expression

49

as described in the methods (Figure 2.1C). The ratio of surface expressed KIR2DS4 to total

KIR2DS4 expression was increased two-fold in the presence of exogenous DAP12 compared to

KIR2DS4 co-transfected with empty vector. Similar results were observed for KIR2DS1 and

KIR2DS2 surface expression with exogenous DAP12 (Figure 2.1D, E).

To confirm that the observed changes in surface expression were dependent upon the interaction

with DAP12, a mutant of KIR2DS1 that disrupts the interaction with DAP12 was created by

changing the amino acid at position 233 from a lysine to an alanine (K233A) as this position is

critical for a successful interaction between KIR2DS molecules and DAP12 [21]. The surface

expression level of the mutant was comparable to expression levels obtained with the wild type

receptor when co-transfected with empty vector (data not shown). In contrast to the observed

increase in surface expression of wild type KIR2DS1 with exogenous DAP12, the mutant

KIR2DS1 displayed equal surface expression independent of exogenous DAP12 (Figure 2.1F).

This result supports the enhanced surface expression is dependent upon a successful interaction

between KIR and DAP12.

To further assess the enhanced surface expression was fully attributed to the presence of DAP12

and not due to increased protein production generated by the two expression plasmids within the

cells, KIR2DS1 was co-transfected with FcεRIγ, an unassociated adapter molecule expressed by

NK cells that is known to interact with KIR2DL4 [144] . Similar to the results for the mutant

KIR2DS1, no change in surface expression was observed when KIR2DS1 was co-transfected

with FcεRIγ compared to empty vector (Figure 2.1G) further implicating DAP12’s role in

increasing KIR2DS surface expression. Western blot analysis demonstrates equivalent

50

expression of the adapter molecules used in these assays (Figure 2.1H). These results

demonstrate, in this model system, KIR2DS molecules are capable of expression at the cell

surface, but exogenous DAP12 expression significantly enhances the number of receptors at the

surface suggesting an effect of DAP12 expression on KIR2DS surface expression.

51

52

Figure 2.1 KIR2DS surface expression is enhanced by exogenous DAP12

A, Surface expression of KIR2DS4 on NKL cells detected by a PE conjugated antibody specific

for CD158i following co-transfection of KIR2DS4 with either DAP12 (solid line) or empty

vector (EV, dashed line). NKL cells transfected with an irrelevant KIR (KIR2DS1, grey shade)

was used as a negative control. B, Total KIR2DS4 expression in NKL cells transfected with EV

(grey), or co-transfected with KIR2DS4 and EV (dashed line) or with DAP12 (solid line).

Transfectants were permeablized and probed with a V5 specific antibody for the internal, C-

terminal V5 tag linked to each KIR molecule. C, KIR2DS4 surface expression presented as a

ratio of surface expressed KIR2DS4 (A) to total KIR2DS4 expression (B). All ratios obtained

were normalized to results for KIR2DS with exogenous DAP12. D-E, Surface expression data

presented in relative ratios for KIR2DS1 and KIR2DS2. KIR2DS1 was used as a negative

control for external staining for KIR2DS2 and KIR2DS4, while KIR2DS4 was used as the

negative control for the KIR2DS1. F, Surface expression data for KIR2DS1 mutated at amino

acid position 233 from a lysine to alanine (K233A). G, Relative surface expression data for

KIR2DS1 co-transfected with an unassociated adapter molecule, FcεRIγ. H, Western blot

analysis showing relative expression levels of DAP12 and FcεRIγ used in the assays. Assays

were performed in triplicate and the data (mean + SEM) represents the results observed in two

independent experiments. Statistical analysis was performed using an unpaired Student’s t-test.

(** P < 0.01, *** P < 0.001)

53

Examination of the impact of DAP12 knockdown on KIR2DS4 surface expression on primary

cells

Resulting from the observed changes in surface expression of KIR2DS molecules in the presence

of exogenous DAP12, we sought to determine the impact of DAP12 on KIR2DS surface

expression on primary cells. Two individuals positive for the full length allele, KIR2DS4*001,

were chosen for analysis as KIR2DS4 surface expression could be evaluated without concern for

cross reactivity of the KIR2DS4 specific antibody with another KIR molecule, unlike the

antibodies specific for KIR2DS1 and KIR2DS2. Treatment of PBMCs or of KIR2DS4

positively sorted cells with DAP12-specific siRNA consistently resulted in a significant

reduction of DAP12 transcripts compared to non-targeting siRNA (Figure 2.2A). However, the

impact on DAP12 protein levels observed in multiple experiments with comparable mRNA

knockdown was not as dramatic, but there was a consistent reduction of 15-20% of DAP12

protein following incubation with DAP12 targeting siRNA (Figure 2.2B). Despite the low

efficiency in reducing DAP12 protein, a decrease of KIR2DS4 surface expression was observed

following DAP12 siRNA treatment as represented by the histogram shown (Figure 2.2C).

Following treatment with DAP12 specific siRNA, a consistent 20% decrease of KIR2DS4

surface expression was observed for both donors analyzed across multiple independent

experiments (Figure 2.2D). The observed decrease in surface expression corresponds with the

expected loss of DAP12 protein in these experiments based on consistent mRNA knockdown

achieved across all the replicates. It is likely a more stable reduction in DAP12 would cause a

greater decrease in surface expression. The correlated decrease of KIR2DS4 surface expression

54

following knockdown of DAP12 in primary cells, however, does support the data obtained from

the transient expression system indicating an impact of DAP12 on KIR2DS surface expression.

55

56

Figure 2.2 KIR2DS4 surface expression is decreased after knockdown of DAP12 in primary

cells.

Primary cells were treated with non-targeting or DAP12 specific siRNA. A, The relative quantity

of DAP12 mRNA transcripts was determined by rqRT-PCR using β-actin as an internal loading

control from RNA isolated from PBMCs or KIR2DS4 expressing cells (*** P < 0.001). The

data represents values obtained over five independent experiments performed in duplicate for

each donor. B, Representative Western blot and densitometry analysis of PBMC whole cell

lysates following treatment with non-targeting (nt) or DAP12 targeting (t) siRNA. Results were

obtained in two independent experiments performed in duplicate (* P = 0.04). C, Representative

histogram of KIR2DS4 surface expression on sorted KIR2DS4 positive primary cells after

treatment with non-targeting siRNA (solid line) or with DAP12 specific siRNA (dashed line).

The grey histogram represents cells probed with an irrelevant PE-conjugated antibody. D,

KIR2DS4 surface expression from two donors was normalized to expression levels obtained

following non-targeting siRNA treatment (*** P < 0.001). The results were recorded from five

independent experiments performed in duplicate and is represented by the normalized mean +

SEM. Statistical analyses for these experiments were performed using unpaired Student’s t-tests.

57

Further analysis of the impact of DAP12 on KIR2DS surface expression over time

The effect on KIR2DS surface expression due to changes in DAP12 expression in the previous

experiments could be attributed to effects in trafficking as well as receptor stability at the cell

surface. To further analyze the dynamics by which DAP12 enhances surface expression of

KIR2DS molecules, a time course was assessed post-transfection analyzing differences in

surface expression over time between empty vector and exogenous DAP12 conditions in the

NKL cell line. While the level of KIR2DS4 surface expression remained similar over the first

few hours following transfection independent of exogenous DAP12, a clear divergence in

surface expression was observed after the 10 hour time point (Figure 2.3A). The expression of

the KIR2DS4 at the cell surface continued to increase at a high rate with exogenous DAP12

compared to a clear plateau effect observed for KIR2DS4 with empty vector. The maximum

surface expression observed for KIR2DS4 with empty vector was obtained 14 hours post-

transfection demarcated by the vertical dashed lines. This same surface expression level was

obtained between 10-12 hours post-transfection when KIR2DS4 was transfected with DAP12.

The total KIR2DS4 expression within these experiments was similar between empty vector and

DAP12 conditions, dismissing the possibility that these results are due to increased total

KIR2DS4 levels within the cells (Figure 2.3B). The data shows that total KIR expression

continues to increase independent of exogenous DAP12; however, continued increase in receptor

surface expression is only achieved in the presence of exogenous DAP12. Conversely, in the

absence of exogenous DAP12, total KIR expression continues to increase but the level expressed

at the surface reaches a plateau. These data suggest DAP12 may be playing a role in either

58

enhancing the maturation and post-translational modifications of KIR2DS4 or that DAP12 has a

role in impacting the recycling of internalized receptors back to the cell surface .

In order to further evaluate DAP12’s potential impact on the post-translation processing of

KIR2DS molecules, expression of two different isoforms of KIR2DS was analyzed over a time

course in the presence or absence of exogenous DAP12. We have previously described

expression of two different isoforms of KIR2DS molecules in our transfected model based on

molecular weight and susceptibility to endoglycosidase H [145]. The larger isoform represents

the mature isoform of the protein that has undergone extended glycosylation in the golgi

preventing digestion of this isoform by endoglycosidase H. This mature isoform is also the only

form that is expressed at the cell surface as previously described [145]. The lower molecular

weight isoform is an immature isoform that is susceptible to endoglycosidase H treatment

suggesting this isoform has not been processed by the golgi apparatus and is likely sequestered

within the endoplasmic reticulum (ER). NKL cells were co-transfected with KIR2DS4 and either

empty vector or DAP12, harvested and lysed at multiple time points post-transfection. Western

blot analysis of these lysates show that when KIR2DS4 was expressed in the absence of DAP12,

the level of immature isoform of the receptor is clearly the predominant isoform of the protein

expressed and that the level of expression of this isoform appears to increase over the time

course (Figure 2.3C). A minimal level of increase was also observed for the mature isoform of

KIR2DS4 in the absence of DAP12. These results are dramatically different from the pattern of

KIR2DS4 isoform expression in the presence of exogenous DAP12. When KIR2DS4 was

expressed with DAP12, the level of mature isoform constantly increased over the time course

59

with an inverse decrease in expression of the immature isoform. The results for changes

observed in mature isoform expression over the time course were quantified from these Western

blots (Figure 2.4D). The quantification shows that the mature isoform of KIR2DS4 is only

slightly increased over the time course and achieves only an increase of 10% at 22 hours post-

transfection compared to the level expressed at 10 hours post-transfection. In comparison a 10%

change in percent of the isoform expressed as mature was achieved at the 13 hour time point

when KIR2DS4 was expressed with exogenous DAP12. The change in percent of receptor

expressed as the mature isoform continued to dramatically increase over the time course,

ultimately ending with a 30% change in percentage of the receptor expressed as mature at 22

hours post-transfection compared to the 10 hour time point. The data show that not only does

DAP12 increase the percent of receptor expressed as the mature form, but also shows the rate at

which the mature isoform is expressed is significantly greater compared to when the receptor is

expressed without DAP12 (P < 0.001). These results suggest KIR2DS4 is capable of being

expressed at the cell surface independent of DAP12 up to a certain threshold. After this

threshold is reached, the data show DAP12 aids in efficiently trafficking KIR2DS4 from the ER,

through the Golgi and ultimately to the cell surface. Although the data does imply a trafficking

mechanism, it does not exclude the possibility that DAP12 may also be impacting the rate at

which internalized receptors are recycled back to the cell surface.

60

61

Figure 2.3 Transfected KIR2DS4 surface expression continues to increase with exogenous

DAP12 over time

Transfected KIR2DS4 surface expression continues to increase with exogenous DAP12 over

time. KIR2DS4 was co-transfected into NKL cells with either empty vector (EV, □) or DAP12

(■). A, KIR2DS4 surface expression was analyzed by flow cytometry over a time course. B,

The total expression of KIR2DS4 was also recorded at the same time points as monitored by the

C-terminal V5 tag. Vertical dashed lines represent when the maximum surface expression of

KIR2DS4 was achieved with EV and the respective expression value obtained with DAP12.

Assays were performed in triplicate and results were reproduced in two independent

experiments. Statistical analysis was performed using ANOVA. C, Representative Western blot

analysis of KIR2DS4 isoform expression at the indicated time points post-transfection in the

presence or absence of exogenous DAP12 using a V5-specific antibody. D, Quantification of

Western blots showing the change in percent of receptor expressed as the mature isoform

evaluated at each time point compared to the values obtained 10 hours post-transfection. The

densitometry and quantification were performed on Western blot results from four independent

transfections and time course analyses of KIR2DS4 in the presence or absence of exogenous

DAP12. The results were statistically analyzed by ANOVA.

62

Investigation of the potential effect of DAP12 on trafficking newly synthesized KIR2DS

molecules to the cell surface

To determine if the continuous increase in KIR2DS4 surface expression observed during the

time course experiment was due to DAP12’s impact on the transport of newly synthesized

receptors and not due to an effect on recycling of endocytosed receptors, NKL transfectants were

treated with either brefeldin A (BFA) to inhibit protein transport or cycloheximide (CHX) to

inhibit protein synthesis starting 10 hours post-transfection. The results were consistent for all

KIR2DS molecules analyzed (Figure 2.4A-C). KIR2DS surface expression with exogenous

DAP12 at 10 hours post-transfection was significantly greater compared to KIR2DS co-

transfected with empty vector at this time point. Consistent with the previous time course

analysis, a dramatic increase of surface expression was observed only for the 16 hour time point

in the presence of exogenous DAP12. Treatment of the KIR2DS and DAP12 co-transfectants

with either BFA or CHX prevented this dramatic increase of surface expression resulting in

surface expression levels comparable to those without exogenous DAP12. The inhibition of

increased surface expression when KIR2DS and DAP12 transfectants were treated with CHX

demonstrates that the enhanced level of receptor surface expression achieved with exogenous

DAP12 is due to expression of newly synthesized proteins at the cell surface and not due to

differences in recycling patterns of the receptor in the absence or presence of exogenous DAP12.

These results support the implications of the previous time course experiments and illustrate a

role for DAP12 in efficiently trafficking newly synthesized KIR2DS molecules to the cell

surface.

63

Figure 2.4 DAP12 impacts trafficking of newly synthesized KIR2DS to the cell surface

KIR2DS encoding constructs were transfected into NKL cells with either the empty vector or the

DAP12 encoding vector. Surface expression was analyzed at 10 hours and 16 hours post-

transfection. After the 10 hour time point, cells were left untreated or treated with brefeldin A

(BFA) or cycloheximide (CHX) for 6 hours and analyzed for KIR2DS surface expression. Flow

cytometry analysis of surface expression was recorded for KIR2DS1 (A), KIR2DS2 (B), and

KIR2DS4 (C). KIR2DS1 was used as a negative control for external staining for KIR2DS2 and

KIR2DS4, while KIR2DS4 was used as the negative control for the KIR2DS1. The assays were

performed in triplicate and reproduced in independent experiments. Results (mean + SEM) were

statistically analyzed using unpaired Student’s t-test comparing samples to untreated NKL cells

co-transfected with KIR2DS and DAP12 at either the 10 or 16 hour time point independently.

64

Identification of the putative site where the interaction between DAP12 and KIR2DS initiates

As the data support DAP12’s role in trafficking, we hypothesized the interaction between

KIR2DS and DAP12 occurred early in the post-translational development of KIR resulting in

enhanced efficiency in maturation of the KIR2DS molecules. The association of DAP12 with

the different isoforms of KIR2DS in transfected NKL cells was examined by Western blot after

immunoprecipitation with a DAP12-specific antibody. Following the immunoprecipitation, two

isoforms of KIR2DS1 were detected by Western blot (Figure 2.5A). The lower isoform is an

immature form of the receptor that is susceptible to Endoglycosidase H (Endo H) treatment and

is suspected to be sequestered within the endoplasmic reticulum (ER) [142]. The larger isoform

represents a mature isoform of the protein that has been glycosylated by the Golgi and is not

cleaved by Endo H, but can be cleaved by PNGase F. Co-immunoprecipitation of the Endo H

sensitive immature isoform of KIR2DS1 suggests that the interaction between KIR2DS and

DAP12 occurs very early in the maturation process, likely within the ER.

To further understand the impact of the early interaction of DAP12 on KIR2DS maturation, the

relative quantities of immature and mature isoforms of each exogenous KIR2DS molecule in the

absence and presence of exogenous DAP12 were determined. Single KIR2DS encoding

constructs were co-transfected into NKL cells with either DAP12 vector or the empty vector.

Whole cell lysates from these transfectants were examined by Western blot along with follow up

densitometry analysis to determine changes in expression of the different isoforms of the

receptors (Figure 2.5B-D). In the absence of exogenous DAP12, each receptor was present at

approximately equal ratios of mature to immature isoforms of the protein (Figure 2.5C, D). The

65

presence of the mature isoform of all the KIR2DS molecules analyzed was enhanced in the

presence of exogenous DAP12 (Figure 2.5C) coinciding with a decrease in the amount of the

immature isoform of the receptors (Figure 2.5D). These data support the interaction of KIR2DS

molecules with DAP12 in the ER aids in driving the receptors efficiently through the maturation

process and ultimately to the cell surface.

66

67

Figure 2.5 DAP12 interacts with KIR2DS early in the maturation process

A, Western blot of immune complexes obtained following immunoprecipitation of lysates from

NKL cells co-transfected with constructs encoding KIR2DS1 and HA-DAP12 with a DAP12

specific antibody. The membrane was probed using antibodies specific for the V5 (KIR) or HA

(DAP12) epitope. B, Western blot of NKL lysates using a V5 specific antibody following

expression of KIR2DS molecules with or without exogenous DAP12. C, Densitometry ratios

calculated from the Western blot in B showing a ratio of mature KIR expression (upper band) to

total KIR expression. D, Densitometry ratios calculated from the Western blot in B showing a

ratio of immature KIR expression (lower band) to total KIR expression. The blots represent the

results of three independent experiments.

68

Analysis of the potential impact of DAP12 on KIR2DS stability at the cell surface

Thus far, the data have illustrated an important role for DAP12 in trafficking KIR2DS molecules.

The other hypothesis to be addressed that can impact surface expression and receptor function is

the possible stabilizing effects of the interaction between DAP12 and KIR2DS at the cell surface.

The relative amount of internalized KIR2DS molecules was analyzed in the absence or presence

of exogenous DAP12 as described in the methods. While internalization of each receptor was

observed independent of exogenous DAP12, a significantly higher proportion of receptors were

internalized in the absence of exogenous DAP12 (Figure 2.6A-C). For KIR2DS1 and KIR2DS2,

DAP12 approximately doubled the stability of receptors while close to a five-fold increase in

stability was observed for KIR2DS4. For all of the receptors, the internalization occurred very

rapidly, with the majority of receptors internalized within the first 10 minutes of antibody

incubation. The data suggest DAP12 may impact KIR2DS surface expression and function by

stabilizing the receptors at the cell surface.

69

Figure 2.6 DAP12 stabilizes KIR2DS at the cell surface

NKL cells were co-transfected with specified KIR2DS constructs and either empty vector (□) or

DAP12 (■). The percent of internalized receptors were quantified for NKL cells transfected with

KIR2DS1 (A), KIR2DS2 (B) or KIR2DS4 (C) following incubation with a PE conjugated

antibody specific for the extracellular domains of each receptor and then cultured for up to 1

hour at 37°C. Cells were removed at 0, 10 , 20 and 60 minutes and placed at 4°C. At each time

point, the remaining external antibody was stripped away with an acidic reagent and the

internalized receptor was analyzed by flow cytometry. The percent internalized was calculated

as described in the methods. Assays were performed in triplicate and reproduced in two

independent experiments. Statistical analysis was performed using ANOVA.

70

Determining the location of internalized KIR2DS molecules in the presence or absence of

DAP12 expression

The rate of immune receptor internalization as well as the ultimate localization of the receptor

after internalization can modulate immune receptor signaling and function. For instance,

KIR2DL4 is internalized to and signals from early endosomes [146]. As we observed

internalization of the KIR2DS molecules independent of exogenous DAP12, we sought to

determine if the interaction with DAP12 affected the recruitment of the receptors to specific

internal compartments. To determine the location of KIR2DS after internalization, HEK293T

cells were transfected with KIR2DS1 and DAP12 or empty vector along with either LAMP1-GFP

or Rab5-GFP encoding vectors. Although HEK293T cells are not an immune cell, KIR

localization within these cells have been previously correlated with localization of KIR within

transfected NK cell lines as well as primary cells [146]. After incubation with a conjugated

KIR2DS1 specific antibody for two hours at either 4°C or 37°C, the cells were analyzed by

confocal microscopy. Samples probed with the antibody at 4°C exhibit clear membrane staining

of KIR2DS1 (Figure 2.7A). The images observed following incubation with the antibody at

37°C revealed the internalized KIR2DS1 localized to LAMP1 associated lysosomal

compartments independent of DAP12 expression (Figure 2.7B). In the presence or absence of

exogenous DAP12, recruitment to early endosomes, marked by Rab5, was not observed for

KIR2DS1 (Figure 2.7C). KIR2DS1 molecules were found to localize to LAMP1 compartments

independent of antibody incubation as cells that were fixed, permeabilized, and probed for

KIR2DS1 showed colocalization of the receptor with LAMP1 as well (data not shown). While

71

the data support a role for DAP12 in anterograde transport of KIR2DS molecules to the cell

surface, the confocal data suggest no impact of DAP12 on retrograde transport from the cell

surface. The lack of a role in internal recruitment is consistent with data showing the

extracellular domains of KIR2DL4, and not the transmembrane region associated with adapter

interactions, are responsible for its recruitment to early endosomes [20].

72

73

Figure 2.7 Confocal microscopy shows internalized KIR2DS1 localizes to LAMP1-associated

lysosomal compartments

KIR2DS1 was transfected with EV or DAP12 along with vectors encoding either LAMP1-GFP or

Rab5-GFP constructs into HEK293T cells. The cells were probed with an APC-conjugated

KIR2DS1 antibody for 2 hours at either 4°C (A) or 37°C (B, C) resulting in internalization of the

receptor. The merged images to the right show the overlay of the KIR2DS1 and LAMP1 (B) or

RAB5 (C) images with the yellow representing observed colocalization of KIR2DS1 with

LAMP1. The image acquired after 4°C incubation was taken with a magnification of 100X

while the remaining 37°C images were obtained with a 200X magnification. The images

represent results observed in ten fields of multiple independent experiments.

74

Evaluation of the impact of DAP12 on the impedance of KIR2DS degradation by the lysosome

To further examine DAP12’s impact on KIR2DS stability and the localization to lysosomes, the

amount of KIR2DS1 degraded in the presence or absence of exogenous DAP12 was determined.

NKL cells were co-transfected with KIR2DS1 and either empty vector or DAP12. Ten hours

post-transfection, the transfectants were treated with cycloheximide for six hours. Following

cyclohexmide treatment, a significantly higher percentage of KIR2DS1 molecules were degraded

in the absence of exogenous DAP12 (Figure 2.8). These data support the previous results

revealing a stabilizing impact of the interaction of DAP12 and KIR2DS molecules. The data

also show that KIR2DS molecules are degraded independent of DAP12 expression which

supports the confocal data showing localization to lysosomes independent of DAP12 expression.

These data suggest that while DAP12 does not affect the location after internalization, the

adapter molecule does retard the rate at which the receptors are trafficked to and degraded by the

lysosomes.

75

Figure 2.8 DAP12 retards degradation of KIR2DS1

NKL cells co-transfected with KIR2DS1 with either empty vector or DAP12 were treated with

CHX 10 hours post-transfection for 6 hours. The cells were fixed, permeablized and probed with

a FITC conjugated V5 specific antibody targeting the C-terminal V5 tag on KIR2DS1. Total

expression levels for KIR2DS1 were obtained by flow cytometry at 10 hours without treatment

and compared to expression levels following 6 hours of treatment with CHX. The data are

presented as percent degraded over the time period analyzed using the following equation: 100-

((MFI16hr/MFI10hr)*100). The results are representative of two independent experiments

performed in triplicate. The results were analyzed using an unpaired Student’s t-test (P = 0.005).

76

2.3 DISCUSSION

2.3.1 Potential impacts of DAP12 on KIR2DS function beyond signaling

KIR2DS molecules are known to modify NK cell activity, but little is known regarding the

biology of their processing and maturation. In this study we hypothesized DAP12 would have an

impact on KIR2DS function beyond signaling. Immunoprecipitation data supports DAP12

interactions with KIR2DS molecules early in the maturation process likely within the

endoplasmic reticulum. This early interaction allows for efficient transport of the receptor to the

cell surface. DAP12 was shown to significantly enhance stability of the receptors at the cell

surface, but had no impact on the location of the receptors after internalization as the receptors

localized to the lysosome independent of exogenous DAP12. Our data provide novel insights in

to the biological importance of the interaction between DAP12 and KIR2DS molecules as the

data shows DAP12 has a major impact on efficient KIR2DS maturation and stability.

The surface expression and stability differences observed for KIR2DS in the presence or absence

of DAP12 described in this study may alter the capabilities of the receptor for recognizing its

ligand and may also affect the strength of signal achieved following KIR2DS ligand binding.

Differences in NK cell responses based on surface expression have previously been described for

another activating NK cells receptor, NKp46 [147]. Surface density may prove to be even more

important for receptors like KIR2DS which bind their ligand with a relatively low affinity.

Although we observed KIR2DS surface expression in the absence of DAP12, it is likely the

increased level of expression in the presence of the adapter molecule would provide greater

77

probability of interactions occurring between receptor and ligand. The observed increase in

receptor stability at the cell surface may also extend interaction time between the NK cell and its

target.

The observed changes in surface expression may also impact the ability of the KIR2DS

molecules to dimerize which is known to strengthen signaling from immune receptors.

Multimerization of the T-cell receptor (TCR) allows for enhanced sensitivity of recognition of a

HLA/peptide ligand and subsequent propagation of strong activating signals [148;149].

Multimerization is also seen with NKG2D and may impact sensitivity of this receptor towards

low abundance of ligand [150]. KIR molecules have the capability of dimerizing which

enhances ligand binding [151;152]. It is possible the enhanced surface expression of KIR2DS in

the presence of DAP12 demonstrated in this study allows for rapid and efficient dimerization of

the activating receptors creating stronger affinities for their ligand and a more robust signaling

upon ligand recognition. In conditions where DAP12 expression is insufficient or absent, the

diminished KIR2DS surface expression may result in inefficient dimerization and reduced

strength of ligand interactions.

Another major factor in determining strength of signaling from an immune receptor is the

localization of the receptor after engagement of ligand. T-cell receptors are rapidly internalized

and degraded after being stimulated. The internalization prevents superfluous secretion of

inflammatory cytokines and over-stimulation of the cell which can lead to apoptosis [153]. B-

cell receptors (BCR) display two different behaviors. The BCR can remain at the cell surface in

order to maintain strong signaling or the receptor can be internalized and initiate signaling in

78

endocytic compartments [143]. Following ligand recognition by the NK cell receptor,

KIR2DL4, the receptor internalizes to early endosomes where inflammatory signaling is

initiated. [18-20]. It is likely the stability differences at the cell surface observed for KIR2DS in

the presence or absence of DAP12 will result in differences in signaling characteristics upon

ligand binding. Upon engagement of a ligand by KIR2DS in the absence of DAP12, the receptor

will be rapidly internalized and degraded by the lysosome. This postulation is consistent with

the lack of cytotoxic activity observed after stimulation of KIR2DS2*003 which contains an

amino acid change that disrupts the interaction with DAP12 [49]. It is possible that a partial

signal may be induced by the receptor or an alternate adapter prior to internalization that may

account for the enhanced inflammatory response observed in T-cells following co-stimulation of

CD3 and KIR2DS2 in the absence of DAP12 [154;155]. Conversely, when DAP12 is

sufficiently expressed, the KIR2DS molecules are more stable at the cell surface, likely allowing

for efficient recruitment of the receptors and signaling molecules to the immune synapse

resulting in a cytotoxic and pro-inflammatory response. The stabilization of the receptor by

DAP12 may also aid in directing cytolytic granules toward the target cell. As surface

expression levels, receptor surface stability, and receptor localization all impact immune receptor

function, the mechanisms revealed in this study may shed light on the impact of KIR2DS

function and signaling capabilities in environments where DAP12 expression is compromised.

2.3.2 Potential clinical relevance of observed results

The data presented within this study provide detail to the biological impact of the interaction

between KIR2DS and DAP12 and provide potential mechanisms whereby KIR2DS function

79

could be compromised when DAP12 expression is decreased in vivo. The mechanisms described

here may help explain functional and phenotypical observations made in the clinic when DAP12

is absent or compromised. Nasu-Hakola disease is a rare clinical syndrome where the impact of

the loss of DAP12 can be observed on KIR2DS function. These patients suffer from bone cysts

and early onset presenile dementia and usually succumb to the disease during the fifth decade of

life. Patients diagnosed with this disease often carry loss-of-function mutations or deletions of

DAP12 [156]. In accordance with the data presented within this dissertation, it is likely the

surface expression levels and functional capabilities of the stimulatory KIR of these patients are

greatly compromised. Although the functions of stimulatory KIR are not well defined, the

impact of the absence of DAP12 on KIR2DS1 may be hypothesized. For instance, if NK cells

from a Nasu-Hakola patient were compared to those of a genetically identical, healthy donor it is

likely the receptor phenotyping of the NK cells as well as overall functional capabilities would

be significantly altered. Based on our defined mechanisms, the surface expression and level of

stability at the cell surface of KIR2DS1 would be significantly lower on the NK cells of the

Nasu-Hakola patient. These phenotypical characteristics of KIR2DS1 could translate into

functional alterations between these individuals as KIR2DS1 has been shown to be important for

NK cell education. Therefore without proper expression, more NK cells in the Nasu-Hakola

patient may be hypo-responsive. Although the ligands for the other stimulatory KIR are not well

understood, these patients may be more susceptible to infection or other diseases that are

regularly cleared by NK cells through stimulation of these receptors.

80

Understanding the mechanisms described within this disseration is also important as culture

conditions and clinical treatments using cytokine regiments are developed in order to enhance

NK cell proliferation and activity. Examples of cytokines used in the clinic for these purposes

include IL-2, IL-15 and IL-21. Results have shown that IL-21 alone can stimulate NK cells ex

vivo but does not induce NK cell proliferation. However, when IL-21 is added in conjunction

with either IL-2 or IL-15, the proliferation of NK cells is increased three fold compared to

culturing the cells with IL-2 or IL-15 alone [157]. While a conditioning regiment combining IL-

2 or IL-15 with IL-21 in order to maximize the number of NK cells may appear promising, these

combinations have been shown to compromise DAP12 expression. Protocols must account for

this change in expression as it may greatly impact the function of all NK receptors that interact

with DAP12. Results such as these warrant further investigation of the transcriptional regulation

of DAP12. Currently very little is known regarding promoter characterization and associated

transcription factors that impact DAP12 expression. Further investigation of the promoter region

will better aid in predicting what cytokine regiments are ideal for maximal NK cell function.

This research focus may also reveal mechanisms by which DAP12 expression can be

compromised in cytokine rich conditions in vivo, such as a tumor microenvironment.

2.3.3 Further investigation of the biology of KIR2DS maturation

This project has demonstrated new roles for DAP12 in KIR2DS function that extend beyond

signaling. As discussed in detail above, the data have defined a role in enhancing maturation and

trafficking of these receptors to the cell surface. Similar to many scientific projects, more

questions seem to arise from these results compared to questions answered. A summary of the

81

results as well as questions raised by the results are depicted (Figure 2.9). The results and

consequential new questions can be analyzed according to specific compartments within the cell.

In order to follow the receptors as they are translated and begin to undergo post-translational

modifications, the impact of the interaction between DAP12 and KIR2DS molecules within the

ER can further be investigated. The mechanism by which the interaction between these proteins

is coordinated is unknown. One hypothesis postulates the leader sequences of DAP12 and

KIR2DS coordinate the translation of these proteins in order to enhance the probability of a

successful interaction. The null hypothesis would be that the leader peptide has no effect and the

proteins randomly interact in the membrane of the endoplasmic reticulum following independent

translation. This hypothesis can be tested by creating mutant variants of the leader sequences of

each transcript followed by transfection of NKL cells. To initially determine potential impacts

on the interaction between the molecules, changes in surface expression among the different

mutants compared to wild type proteins can be analyzed in the presence or absence of exogenous

DAP12. If the leader sequence plays a role in initiating the interaction between DAP12 and

KIR2DS, allelic polymorphisms as well as inter-locus polymorphisms amongst the different

KIR2DS genes may impact the capability of a successful interaction with DAP12. Consistent

with this hypothesis, we have previously shown a polymorphism within the leader sequence of

KIR2DS1 to have a potential impact on KIR2DS1 surface expression although it is not known if

this observation was due to an altered interaction with DAP12. This effect appeared to be cell

type specific which suggests potential impacts of cell type specific chaperone proteins in

processing of KIR2DS molecules which produces another interesting focus for future studies of

82

how interactions of KIR2DS with chaperone proteins in the ER as well as the Golgi apparatus

are impacted by the interaction with DAP12.

The hypothesis driving the above investigation is DAP12 alters the conformation of KIR2DS or

recruits chaperone proteins that enhance the efficiency by which KIR2DS molecules are

glycosylated in the ER and Golgi. The interaction with DAP12 may also disrupt interactions

with certain chaperones such as calreticulin which has been shown to mediate ER retention of

KIR3DL1*004 [158]. In order to initially examine this hypothesis, NKL transfectants of

KIR2DS with empty vector or DAP12 can be analyzed for differential interacting complexes.

Proteomic analysis following immunoprecipitation of KIR from whole cell lysates could be used

to determine changes in interacting partners of KIR in the presence or absence of DAP12. These

potential investigations will further our knowledge of the molecular mechanisms required for

efficient KIR2DS maturation and trafficking to the cell surface.

As the data have demonstrated, KIR2DS surface expression is increased in the presence of

DAP12, the impact of this enhancement may also be a target for further investigation. The

hypothesis driving these experiments is the enhanced surface expression achieved with DAP12

correlates with enhanced stimulation of activating cascades as well as enhanced cytokine

secretion and cytotoxic activity. DAP12 also has an impact on stability of the receptors at the

cell surface. A hypothesis to be tested regarding this conclusion is that the increased stability of

the receptor allows the receptor to accumulate more efficiently within the immune synapse and

also allows for proper formation of signaling scaffolds within the cell upon recognition of ligand.

The enhanced stability may also aid in directing cytolytic granules towards the target cells. The

83

hypotheses described within this section aim to further investigate how molecular interactions

impact KIR2DS biology and ultimately affect NK cell function. Through understanding these

mechanisms, one can extrapolate how the functions of NK cells may be altered when KIR2DS

processing is compromised.

84

85

Figure 2.9 Overview of results and future directions

A schematic summarizes the key observations of this study as well as questions for future

investigation arising from the described mechanisms. The data reveal several new roles for

DAP12 in KIR2DS maturation, trafficking and surface stability. The molecular impacts of the

interaction between these proteins warrant further investigation as to how the interaction initially

occurs in the ER and how this interaction drives a more efficient KIR2DS maturation and

trafficking to the cell surface. The functional effects of enhanced surface expression and

increased surface stability in the presence of DAP12 also warrant future experimentation. These

further mechanistic analyses will provide further knowledge of the molecular impact of the

interaction between DAP12 and KIR2DS.

86

2.4 RESEARCH ACKNOWLEDGMENTS

This research was supported by funding from the Office of Naval Research. The views

expressed in this article are those of the authors and do not reflect the official policy of the

Department of the Navy, the Department of Defense, or the United States government.

These experiments were performed with the aid of the Tissue Culture, Flow Cytometry and Cell

Sorting, and Microscopy and Imaging Shared Resources of the Lombardi Comprehensive Cancer

Center supported by National Cancer Institute Cancer Center Support Grant. We thank William

Frazier, Karen Creswell, and Noriko Steiner for helpful discussions and technical assistance.

87

2.5 References

1. Colucci F, Caligiuri MA, and Di Santo JP., What does it take to make a natural killer?

Nat Rev. 2003. 3(5), 413-425.

2. Kondo M, Weissman IL, and Akashi K., Identification of clonogenic common

lymphoid progenitors in mouse bone marrow. Cell. 1997. 91(5), 661-672.

3. Farag SS and Caligiuri MA., Human natural killer cell development and biology.

Blood. 2006. 20, 123-127.

4. Björkström NK, Riese P, Heuts F, Andersson S, Fauriat C, Ivarsson MA, Bjorklund

AT et al., Expression patterns of NKG2A, KIR, and CD57 define a process of CD56dim

NK-cell differentiation uncoupled from NK-cell education. Blood. 2010. 116(19), 3853-

3864.

5. Chan A, Hong DL, Atzberger A, Kollnberger S, Filer AD, Buckley CD, McMichael

A et al., CD56bright human NK cells differentiate into CD56dim cells: role of contact

with peripheral fibroblasts. J Immunol. 2007. 179(1), 89-94.

6. Kiessling R, Klein E, and Wigzell H., "Natural" killer cells in the mouse. I. Cytotoxic

cells with specificity for mouse Moloney leukemia cells. Specificity and distribution

according to genotype. Eur J Immunol. 1975. 5(2), 112-117.

7. Kiessling R, Klein E, Pross H, and Wigzell H., "Natural" killer cells in the mouse. II.

Cytotoxic cells with specificity for mouse Moloney leukemia cells. Characteristics of the

killer cell. Eur J Immunol. 1975. 5(2), 117-121.

8. Hedrick JA, Saylor V, Figueroa D, and et al., Lymphotactin is produced by NK cells

and attracts both NK cells and T cells in vivo. J Immunol. 1997. 158, 1533-1540.

9. Fehniger TA, Shah MH, Turner MJ, and et al., Differential cytokine and chemokine

gene expression by human NK cells following activation with IL-18 or IL-15 in

combination with IL-12: implications for the innate immune response. J Immunol. 1999.

162, 4511-4520.

88

10. luman EM, Bartynski KJ, Avalos BR, and Caligiuri MA., Human natural killer cells

produce abundant macrophage inflammatory protein-1 in response to monocyte-derived

cytokines. J Clin Invest. 1996. 97, 2722-2727.

11. Paust S, G. H., Gill HS, Wang BZ, Flynn MP, Moseman EA, Senman B, Szczepanik

M et al., Critical role for the chemokine receptor CXCR6 in NK cell-mediated antigen-

specific memory of haptens and viruses. Nat Immunol. 2010. 11, 1127-1135.

12. Sun JC, Beilke JN, and Lanier LL., Adaptive immune features of natural killer cells.

Nature. 2009. 457, 557-561.

13. Kärre K, Ljunggren HG, Piontek G, and Kiessling R., Selective rejection of H-2-

deficient lymphoma variants suggests alternative immune defence strategy. Nature. 1986.

319, 675-678.

14. Ljunggren HG and Kärre K., In search of the 'missing self': MHC molecules and NK

cell recognition. Immunol Today. 1990. 11(7), 237-244.

15. Smyth MJ, Cretney E, Kelly JM, Westwood JA, Street SE, Yagita H, Takeda K et

al., Activation of NK cell cytotoxicity. Mol Immunol. 2005. 42(4), 501-510.

16. Burshtyn DN, Scharenberg AM, Wagtmann N, Rajagopalan S, Berrada K, Yi T,

Kinet JP et al., Recruitment of tyrosine phosphatase HCP by the killer cell inhibitor

receptor. Immunity. 1996. 4(1), 77-85.

17. Olcese L, Lang P, Vély F, Cambiaggi A, Marguet D, Blery M, Hippen KL et al.,

Human and mouse killer-cell inhibitory receptors recruit PTP1C and PTP1D protein

tyrosine phosphatases. J Immunol. 1996. 156(12), 4531-4534.

18. Miah SM, Hughes TL, and Campbell KS., KIR2DL4 differentially signals downstream

functions in human NK cells through distinct structural modules. J Immunol. 2008.

180(5), 2922-2932.

19. Rajagopalan S, Bryceson YT, Kuppusamy SP, Geraghty DE, van der Mer Ae,

Joosten I, and Long EO., Activation of NK cells by an endocytosed receptor for soluble

HLA-G. PLoS Biol. 2006. 4(1), e9.

89

20. Rajagopalan S, Moyle MW, Joosten I, and Long EO., DNA-PKcs controls an

endosomal signaling pathway for a proinflammoatory response by natural killer cells. Sci

Signal. 2010. 3(110), ra14.

21. Feng J, Call ME, and Wucherpfenning KW., The assembly of diverse immune

receptors is focused on a polar membrane-embedded interaction site. PLoS Biol. 2006.

4(5), e142.

22. Lanier LL, Corliss BC, Wu J, Leong C, and Phillips JH., Immunoreceptor DAP12

bearing a tyrosine-based activation motif is involved in activating NK cells. Nature.

1998. 391(6668), 703-707.

23. Chan HW, Miller JS, Moore MB, and Lutz CT., Epigenetic control of highly

homologous killer Ig-like receptor gene alleles. J Immunol. 2005. 175(9), 5966-5974.

24. Davies GE, Locke SM, Wright PW, Li H, Hanson RJ, Miller JS, and Anderson SK.,

Identification of bidirectional promoters in the human KIR genes. Genes Immun. 2007.

8(3), 245-253.

25. Cichocki F, Miller JS, and Andersson SK., Killer immunoglobulin-like receptor

transcriptional regulation: a fascinating dance of multiple promoters. J Innate Immun.

2011. 3(3), 242-248.

26. Cichocki F, Lenvik T, Sharma N, Yun G, Anderson SK, and Miller JS., Cutting

edge: KIR antisense transcripts are processed into a 28-base PIWI-like RNA in human

NK cells. J Immunol. 2010. 185(4), 2009-2012.

27. Andersson S, Fauriat C, Malmberg JA, Ljunggren HG, and Malmberg KJ., KIR

acquisition probabilities are independent of self-HLA class I ligands and increase with

cellular KIR expression. Blood. 2009. 114(1), 95-104.

28. Tomasello E and Vivier E., KARAP/DAP12/TYROBP: three names and a multiplicity

of biological functions. Eur J Immunol. 2005. 35(6), 1670-1677.

29. Lanier LL., DAP10- and DAP12-associated receptors in innate immunity. Immunol Rev.

2009. 227(1), 150-160.

90

30. Hamerman JA, Tchao NK, Lowell CA, and Lanier LL., Enhanced toll-like receptor

responses in the absence of signaling adaptor DAP12. Nat Immunol. 2005. 6, 579-586.

31. Diefenbach A, Tomasello E, Lucas M, Jamieson AM, Hsia JK, Vivier E, and Raulet

DH., Selective associations with signaling proteins determine stimulatory versus

costimulatory activity of NKG2D. Nat Immunol. 2002. 3, 1142-1149.

32. Sjolin H, Tomasello E, Mousavi-Jazi M, Bartolazzi A, Karre K, Vivier E, and

Cerboni C., Pivotal role of KARAP/DAP12 adaptor molecule in the natural killer cell-

mediated resistance to murine cytomegalovirus infection. J Exp Med. 2002. 195, 825-

834.

33. Paloneva J, Kestilä M, Wu J, Salminen A, Böhling T, Ruotsalainen V, Hakola P et

al., Loss-of-function mutations in TYROBP (DAP12) result in a presenile dementia with

bone cysts. Nat Genet. 2000. 25(3), 357-361.

34. de Rham C, Ferrari-Lacraz S, Jendly S, Schneiter G, Dayer JM, and Villard J., The

proinflammatory cytokines IL-2, IL-15 and IL-21 modulate the repertoire of mature

human natural killer cell receptors. Arthritis Res Ther. 2007. 9(6), R125.

35. Orr MT, Sun JC, Hesslein DG, Arase H, Phillips JH, Takari T, and Lanier LL.,

Ly49H signaling through DAP10 is essential for optimal natural killer cell responses to

mouse cytomegalovirus infection. J Exp Med. 2009. 206(4), 807-817.

36. Burgess SJ, Marusina AI, Pathmanathan I, Borrego F, and Coligan JE., IL-21

down-regulates NKG2D/DAP10 expression on human NK and CD8+ T cells. J Immunol.

2006. 176(3), 1490-1497.

37. Shilling HG, Guethlein LA, Cheng NW, Gardiner CM, Rodrigues R, Tyan D, and

Parham P., Allelic polymorphism synergizes with variable gene content to individualize

human KIR genotype. J Immunol. 2002. 168(5), 2307-2315.

38. Trowsdale J, Barten R, Haude A, Stewart CA, Beck S, and Wilson MJ., The genomic

context of natural killer receptor extended gene families. Immunol Rev. 2001. 181, 20-38.

91

39. Ordóñez D, Meenagh A, Gomez-Lozano N, Castano J, Middleton D, and Vilches C.,

Duplication, mutation and recombination of the human orphan gene KIR2DS3 contribute

to the diversity of KIR haplotypes. Genes Immun. 2008. 9(5), 431-437.

40. Gómez-Lozano N, Gardiner CM, Parham P, and Vilches C., Some human KIR

haplotypes contain two KIR2DL5 genes: KIR2DL5A and KIR2DL5B. Immunogenetics.

2002. 54(5), 314-319.

41. Norman PJ, Abi-Rached L, Gendzehadze K, Hammond JA, Moesta AK, and et al.,

Meiotic recombination generates rich diversity in NK cell receptor genes, alleles, and

haplotypes. Genome Res. 2009. 19(5), 757-769.

42. Martin MP, Bashirova A, Traherne J, Trowsdale J, and Carrington M., Cutting

edge: expansion of the KIR locus by unequal crossing over. J Immunol. 2003. 171(5),

2192-2195.

43. Traherne JA, Martin M, Ward R, Ohashi M, Pellet F, Gladman D, Middleton D et

al., Mechanisms of copy number variation and hybrid gene formation in the KIR immune

gene complex. Mol Genet. 2010. 19, 737-751.

44. Pyo CW, Guethlein LA, Vu Q, Wang R, Abi-Rached L, Norman PJ, Marsh SG et

al., Different patterns of evolution in the centromeric and telomeric regions of group A

and B haplotypes of the human killer cell Ig-like receptor locus. PLoS ONE. 2010. 5(12),

e15115.

45. Hou L, Chen M, Ng J, and Hurley CK., Conserved KIR allele-level haplotypes are

altered by microvariation in individuals with European ancestry. Genes Immun. 2012.

1(47), 47-58.

46. Hou L, Steiner NK, Chen M, Belle I, Kubit AL, Ng J, and Hurley CK., Limited

allelic diversity of stimulatory two-domain killer cell immunoglobulin-like receptors.

Hum Immunol. 2008. 69(3), 174-178.

47. Pando MJ, Gardiner CM, Gleimer M, McQueen KL, and Parham P., The protein

made from a common allele of KIR3DL1 (3DL1*004) is poorly expressed at cell

surfaces due to substitution at positions 86 in Ig domain 0 and 182 in Ig domain 1. J

Immunol. 2003. 171, 6640-6649.

92

48. VandenBussche CJ, Dakshanamurthy S, Posch PE, and Hurley CK., A single

polymorphism disrupts the killer Ig-like receptor 2DL2/2DL3 D1 domain. J Immunol.

2006. 177(8), 5347-5357.

49. Bottino C, Falco M, Sivori S, Morreta L, Moretta A, and Biassoni R., Identification

and molecular characterization of a natural mutant of the p50.2/KIR2DS2 activating NK

receptor that fails to mediate NK cell triggering. Eur J Immunol. 2000. 30(12), 3569-

3574.

50. Moesta AK, Norman PJ, Yawata N, Gleimer M, and Parham P., Synergistic

polymorphism at two positions distal to the ligand-binding site makes KIR2DL2 a

stronger receptor for HLA-C than KIR2DL3. J Immunol. 2008. 180(6), 3969-3979.

51. O'Connor GM, Yamada E, Rampersaud A, Thomas R, Carrington M, and McVicar

DW., Analysis of binding of KIR3DS1*014 to HLA suggests distinct evolutionary

history of KIR3DS1. J Immunol. 2011. 187(5), 2162-2172.

52. Yawata M, Yawata N, Draghi M, Little AM, Partheniou F, and Parham P., Roles for

HLA and KIR polymorphisms in natural killer cell repertoire selection and modulation of

effector function. J Exp Med. 2006. 203(3), 633-645.

53. Moretta A, Bottino C, Pende D, Tripodi G, Tambussi G, Viale O, Orengo A et al.,

Identification of four subsets of human CD3-CD16+ natural killer (NK) cells by the

expression of clonally distributed functional surface molecules: correlation between

subset assignment of NK clones and ability to mediate specific alloantigen recognition. J

Exp Med. 1990. 172(6), 1589-1598.

54. Moretta A, Vitale M, Bottino C, Orengo A, Morelli L, Augugliar R, Barbaresi M et

al., P58 molecules as putative receptors for major histocompatibility complex (MHC)

class I molecules in human natural killer (NK) cells. Anti-p58 antibodies reconstitute

lysis of MHC class I-protected cells in NK clones displaying different specificities. J Exp

Med. 1993. 178(2), 597-604.

55. Winter CC, Gumperz JE, Parham P, Long EO, and Wagtmann N., Direct binding

and functional transfer of NK cell inhibitory receptors reveal novel patterns of HLA-C

allotype recognition. J Immunol. 1998. 161(2), 571-577.

93

56. Colonna M, Borsellino G, Falco M, Ferrara GB, and Strominger JL., HLA-C is the

inhibitory ligand that determines dominant resistance to lysis by NK1- and NK2-specific

natural killer cells. Proc Natl Acad Sci USA. 1993. 90, 12000-12004.

57. Colonna M, Brooks EG, Falco M, Ferrara GB, and Strominger JL., Generation of

allospecific natural killer cells by stimulation across polymorphism of HLA-C. Science.

1993. 260, 1121-1124.

58. Litwin V, Gumperz J, Parham P, Phillips JH, and Lanier LL., NKB1: a natural killer

cell receptor involved in the recognition of polymorphic HLA-B molecules. J Exp Med.

1994. 180(2), 537-543.

59. Gumperz JE, Litwin V, Phillips JH, Lanier LL, and Parham P., The Bw4 public

epitope of HLA-B molecules confers reactivity with natural killer cell clones that express

NKB1, a putative HLA receptor. J Exp Med. 1995. 181, 1133-1144.

60. Cella M, Longo A, Ferrara GB, Strominger JL, and Colonna M., NK3-specific

natural killer cells are selectively inhibited by Bw4-positive HLA alleles with isoleucine

80. J Exp Med. 1994. 180, 1235-1242.

61. Dohring C, Scheidegger D, Samaridis J, Cella M, and Colonna M., A human killer

inhibitory receptor specific for HLA-A1,2. J Immunol. 1996. 156, 3098-3101.

62. Pende D, Biassoni R, Cantoni C, Verdiani S, Falco M, di Donato C, Accame L et al.,

The natural killer cell receptor specific for HLA-A allotypes: a novel member of the

p58/p70 family of inhibitory receptors that is characterized by three immunoglobulin-like

domains and is expressed as a 140-kD disulphide-linked dimer. J Exp Med. 1996. 184(2),

505-518.

63. Stern M, Ruggeri L, Capanni M, Mancusi A, and Velardi A., Human leukocyte

antigens A23, A24, and A32 but not A25 are ligands for KIR3DL1. Blood. 2008. 112(3),

708-710.

64. Rajagopalan S and Long EO., A human histocompatibility leukocyte antigen (HLA)-G-

specific receptor expressed on all natural killer cells. J Exp Med. 1999. 189(7), 1093-

1100.

94

65. Biassoni R, Pessino A, Malaspina A, Cantoni C, Bottino C, Sivori S, Moretta L et

al., Role of amino acid position 70 in the binding affinity of p50.1 and p58.1 receptors for

HLA-Cw4 molecules. Eur J Immunol. 1997. 27(12), 3095-3099.

66. Chewning JH, Gudme CN, Hsu KC, Selvakumar A, and Dupont B., KIR2DS1-

positive NK cells mediate alloresponse against the C2 HLA-KIR ligand group in vitro. J

Immunol. 2007. 179(2), 854-868.

67. Stewart CA, Laugier-Anfossi F, Vély F, Saulquin X, Riedmuller J, Tisserant A,

Gauthier L et al., Recognition of peptide-MHC class I complexes by activating killer

immunoglobulin-like receptors. Proc Natl Acad Sci USA. 2005. 102(37), 13224-13229.

68. Graef T, Moesta AK, Norman PJ, Abi-Rached L, Vago L, Older Aguilar AM,

Gleimer M et al., KIR2DS4 is a product of gene conversion with KIR3DL2 that

introduced specificity for HLA-A*11 while diminishing avidity for HLA-C. J Exp Med.

2009. 206(11), 2557-2572.

69. Katz G, Gazit R, Arnon TI, Gonen-Gross T, Tarcic G, Markel G, Gruda R et al.,

MHC class I-independent recognition of NK-activating receptor KIR2DS4. J Immunol.

2004. 173(3), 1819-1825.

70. Katz G, Markel G, Mizrahi S, Arnon TI, and Mandelboim O., Recognition of HLA-

Cw4 but not HLA-Cw6 by the NK cell receptor killer cell Ig-like receptor two-domain

short tail number 4. J Immunol. 2001. 166(12), 7260-7267.

71. Boyington JC, Motyka SA, Schuck P, Brooks AG, and Sun PD., Crystal structure of

an NK cell immunoglobulin-like receptor in complex with its class I MHC ligand.

Nature. 2000. 405(6786), 537-543.

72. Maenaka K, Juji T, Stuart DI, and Jones EY., Crystal structure of the human p58

killer cell inhibitory receptor (KIR2DL3) specific for HLA-Cw3-related MHC class I.

Structure. 1999. 7(4), 391-398.

73. Vivian JP, Duncan RC, Berry R, O'Connor GM, Reid HH, Beddoe T, Gras S et al.,

Killer cell immunoglobulin-like receptor 3DL1-mediated recognition of human leukocyte

antigen B. Nature. 2011. 479(7373), 401-405.

95

74. Campbell KS and Purdy AK., Structure/function of human killer cell immunoglobulin-

like receptors: lessons from polymorphisms, evolution, crystal structures and mutations.

Immunology. 2011. 132(3), 315-325.

75. Fadda L, Borhis G, Ahmed P, Cheent K, Pageoa SV, Cazaly A, Stathopoulos S et al.,

Peptide antagonism as a mechanism for NK cell activation. Proc Natl Acad Sci USA.

2010. 107(22), 10160-10165.

76. Parham P., The genetic and evolutionary balances in human NK cell receptor diversity.

Semin Immunol. 2008. 20(6), 311-316.

77. Guethlein LA, Abi-Rached L, Hammond JA, and Parham P., The expanded cattle

KIR genes are orthologous to the conserved single-copy KIR3DX1 gene of primates.

Immunogenetics. 2007. 59(6), 517-522.

78. Khakoo SI, Rajalingam R, Shum BP, Weidenbach K, Flodin L, Muir DG, Canavez

F et al., Rapid evolution of NK cell receptor systems demonstrated by comparison of

chimpanzees and humans. Immunity. 2000. 12(6), 687-698.

79. Trundley AE, Hiby SE, Chang C, Sharkey AM, Santourlidis S, Uhrberg M,

Trowsdale J et al., Molecular characterization of KIR3DL3. Immunogenetics. 2006.

57(12), 904-916.

80. Older Aguilar AM, Guethlein LA, Hermes M, Walter L, and Parham P., Rhesus

macaque KIR bind human MHC class I with broad specificity and recognize HLA-C

more effectively than HLA-A and HLA-B. Immunogenetics. 2011. 63(9), 577-585.

81. Fukami-Kobayashi K, Shiina T, Anzai T, Sano K, Yamazaki M, Inoko H, and

Tateno Y., Genomic evolution of MHC class I region in primates. Proc Natl Acad Sci

USA. 2005. 102(62), 9230-9234.

82. Chen ZW, McAdam SN, Hughes AL, Dogon AL, Letvin NL, and Watkins DI.,

Molecular cloning of orangutan and gibbon MHC class I cDNA. The HLA-A and -B loci

diverged over 30 million years ago. J Immunol. 1992. 148(8), 2547-2554.

96

83. Abi-Rached L, Kuhl H, Roos C, ten Hallers B, Zhu B, Carbone L, de Jong PJ et al.,

A small, variable, and irregular killer cell Ig-like receptor locus accompanies the absence

of MHC-C and MHC-G in gibbons. J Immunol. 2010. 184(3), 1379-1391.

84. Adams EJ, Thomson G, and Parham P., Evidence for an HLA-C-like locus in the

orangutan Pongo pygmaeus. Immunogenetics. 1999. 49(10), 865-871.

85. Guethlein LA, Older Aguilar AM, Abi-Rached L, and Parham P., Evolution of killer

cell Ig-like receptor (KIR) genes: definition of an orangutan KIR haplotype reveals

expansion of lineage III KIR associated with the emergence of MHC-C. J Immunol.

2007. 179(1), 491-504.

86. Older Aguilar AM, Guethlein LA, Adams EJ, Abi-Rached L, Moesta AK, and

Parham P., Coevolution of killer cell Ig-like receptors with HLA-C to become the major

variable regulators of human NK cells. J Immunol. 2010. 185(7), 4238-4251.

87. Older Aguilar AM, Guethlein LA, Abi-Rached L, and Parham P., Natural variation

at position 45 in the D1 domain of lineage III killer cell immunoglobulin-like receptors

(KIR) has major effects on the avidity and specificity for MHC class I. Immunogenetics.

2011. 63(8), 543-547.

88. Moesta AK, Graef T, Abi-Rached L, Older Aguilar AM, Guethlein LA, and Parham

P., Humans differ from other hominids in lacking an activating NK cell receptor that

recognizes the C1 epitope of MHC class I. J Immunol. 2010. 185(7), 4233-4237.

89. Valiante NM, Uhrberg M, Shilling HG, Lienert-Weidenbach K, Arnett KL,

D'Andrea A, Phillips JH et al., Functionally and structurally distinct NK cell receptor

repertoires in the peripheral blood of two human donors. Immunity. 1997. 7(6), 739-751.

90. Yawata M, Yawata N, Draghi M, Partheniou F, Little AM, and Parham P., MHC

class I-specific inhibitory receptors and their ligands structure diverse human NK-cell

repertoires toward a balance of missing self-response. Blood. 2008. 112(6), 2369-2380.

91. Liao NS, Bix M, Zijlstra M, Jaenisch R, and Raulet DH., MHC class I deficiency:

susceptibility to natural killer (NK) cells and impaired NK activity. Science. 1991.

253(5016), 199-202.

97

92. Fernandez NC, Treiner E, Vance RE, Jamieson AM, Lemiieux S, and Raulet DH., A

subset of natural killer cells achieves self-tolerance without expressing inhibitory

receptors specific for self-MHC molecules. Blood. 2005. 105(11), 4416-4423.

93. Anfossi N, Andre P, Guia S, Falk CS, Roetnyck S, Stewart CA, Breso V et al.,

Human NK cell education by inhibitory receptors for MHC class I. Immunity. 2006.

25(2), 331-342.

94. Kim S, Poursine-Laurent J, Truscott SM, Lybarger L, Song YJ, Yang L, French AR

et al., Licensing of natural killer cells by host major histocompatibility complex class I

molecules. Nature. 2005. 436(7051), 709-713.

95. Ogasawara K, Benjamin J, Takaki R, Phillips JH, and Lanier LL., Function of

NKG2D in natural killer cell-mediated rejection of mouse bone marrow grafts. Nat

Immunol. 2005. 6(9), 938-945.

96. Oppenheim DE, Roberts SJ, Clarke SL, Filler R, Lewis JM, Tigelaar RE, Girardi M

et al., Sustained localized expression of ligand for the activating NKG2D receptor

impairs natural cytotoxicity in vivo and reduces tumor immunosurveillance. Nat

Immunol. 2005. 6, 928-937.

97. Sun JC and Lanier LL., Tolerance of NK cells encountering their viral ligand during

development. J Exp Med. 2008. 205(8), 1819-1828.

98. Fauriat C, Ivarsson MA, Ljunggren HG, Malmberg KJ, and Michaelsson J.,

Education of human natural killer cells by activating killer cell immunoglobulin-like

receptors. Blood. 2010. 115, 1166-1174. 6.

99. Brodin P, Kärre K, and Höglund P., NK cell education: not an on-off switch but a

tunable rheostat. Trends Immunol. 2009. 30(4), 143-149.

100. Jonsson AH, Yang L, Kim S, Taffner SM, and Yokoyama WM., Effects of MHC

class I alleles on licensing of Ly49A+ NK cells. J Immunol. 2010. 184, 3424-3432. 7.

101. Joncker NT, Fernandez NC, Treiner E, Vivier E, and Raulet DH., NK cell

responsiveness is tuned commensurate with the number of inhibitory receptors for self-

MHC class I: the rheostat model. J Immunol. 2009. 182(8), 4572-4580.

98

102. Foley B, C. S., Cooley S, Verneris MR, Curtsinger J, Luo X, Waller EK, Weisdorf

DJ et al., NK cell education after allogeneic transplantation: dissociation between

recovery of cytokine-producing and cytotoxic functions. Blood. 2011. 118(10), 2784-

2792.

103. Sun JC and Lanier LL., Cutting edge: viral infection breaks NK cell tolerance to

"missing self". J Immunol. 2008. 181(11), 7453-7457.

104. Kim S, Iizuka K, Aguila HL, Weissman IL, and Yokoyama WM., In vivo natural

killer cell activities revealed by natural killer cell-deficient mice. Proc Natl Acad Sci

USA. 2000. 97(6), 2731-2736.

105. Ruggeri L, Capanni M, Urbani E, Perruccio K, Shlomchik WD, Tosti A, Posati S et

al., Effectiveness of donor natural killer cell alloreactivity in mismatched hematopoietic

transplants. Science. 2002. 295(5562), 2097-2100.

106. Bai A and Forman J., The effect of the proteasome inhibitor lactacystin on the

presentation of transporter associated with antigen processing (TAP)-dependent and

TAP-independent peptide epitopes by class I molecules. J Immunol. 1997. 159(5), 2139-

2146.

107. Garrido F, Ruiz-Cabello F, Cabrera T, Perez-Villar JJ, Lopez-Botet M, Duggan-

Keen M, and Stern PL., Implications for immunosurveillance of altered HLA class I

phenotypes in human tumours. Immunol Today. 1997. 18, 89-95.

108. Hicklin DJ, Wang Z, Arienti F, Rivoltini L, Parmiani G, and Ferrone S., beta2-

Microglobulin mutations, HLA class I antigen loss, and tumor progression in melanoma.

J Clin Invest. 1998. 101(12), 2720-2729.

109. Brandt CS, Baratin M, Yi EC, Kennedy J, Gao Z, Fox B, Haldeman B et al., The B7

family member B7-H6 is a tumor cell ligand for the activating natural killer cell receptor

NKp30 in humans. J Exp Med. 2009. 206(7), 1495-14503.

110. Diefenbach A, Jensen ER, Jamieson AM, and Raulet DH., Rae1 and H60 ligands of

the NKG2D receptor stimulate tumour immunity. Nature. 2001. 413(6852), 165-171.

99

111. Pende D, Bottino C, Castriconi R, Cantoni C, Marcenaro S, Rivera P, Spaggiari GM

et al., PVR (CD155) and Nectin-2 (CD112) as ligands of the human DNAM-1 (CD226)

activating receptor: involvement in tumor cell lysis. Mol Immunol. 2005. 42(4), 463-469.

112. Sivori S, Carlomagno S, Falco M, Romeo E, Moretta L, and Moretta A., Natural

killer cells expressing the KIR2DS1-activating receptor efficiently kill T-cell blasts and

dendritic cells: implications in haploidentical HSCT. Blood. 2011. 117(16), 4284-4292.

113. Rosenberg SA, Lotze MT, Yang JC, Topalian SL, Chang AE, Schwartzentruber DJ,

Aebersold P et al., Prospective randomized trial of high-dose interleukin-2 alone or in

conjunction with lymphokine-activated killer cells for the treatment of patients with

advanced cancer. J Natl Cancer Inst. 1993. 85, 622-632.

114. Miller JS, Soignier Y, Panoskaltsis-Mortari A, McNearney SA, Yun GH, Fautsch

SK, McKenna D et al., Successful adoptive transfer and in vivo expansion of human

haploidentical NK cells in patients with cancer. Blood. 2005. 105(8), 3051-3057.

115. Davis ID, Skrumsager BK, Cebon J, Nicholaou T, Barlow JW, Moller NP, Skak K

et al., An open-label, two-arm, phase I trial of recombinant human interleukin-21 in

patients with metastatic melanoma. Clin Cancer Res. 2007. 13(12), 3630-3636.

116. Yu P, S. J., Steel JC, Zhang M, Morris JC, Waitz R, Fasso M, Allison JP et al.,

Simultaneous inhibition of two regulatory T-cell subsets enhanced Interleukin-15

efficacy in a prostate tumor model. Proc Natl Acad Sci USA. 2012. 109(16), 6187-6192.

117. Benson DM Jr, Backan CE, Zhang S, Collins SM, Liang J, Srivastava S, Hofmeister

CC et al., IPH2101, a novel anti-inhibitory KIR antibody, and lenalidomide combine to

enhance the natural killer cell versus multiple myeloma effect. Blood. 2011. 118(24),

6387-6391.

118. Clynes RA, Towers TL, Presta LG, and Ravetech JV., Inhibitory Fc receptors

modulate in vivo cytotoxicity against tumor targets. Nat Med. 2000. 6(4), 443-446.

119. Cartron G, Dacheux L, Salles G, Solal-Celigny P, Bardos P, Colombat P, and

Waiter H., Therapeutic activity of humanized anti-CD20 monoclonal antibody and

polymorphism in IgG Fc receptor FcgammaRIIIa gene. Blood. 2002. 99(3), 754-758.

100

120. Musolino A, Naldi N, Bortesi B, Pezzulo D, Capelletti M, Missale G, Laccabue D et

al., Immunoglobulin G fragment C receptor polymorphisms and clinical efficacy of

trastuzumab-based therapy in patients with HER-2/neu-positive metastatic breast cancer.

J Clin Oncol. 2008. 26(11), 1789-1796.

121. Kollman C, Abella E, Baitty RL, Beatty PG, Chakraborty R, Christiansen CL,

Hartzman RJ et al., Assessment of optimal size and composition of the U.S. National

Registry of hematopoietic stem cell donors. Transplantation. 2004. 78(1), 89-95.

122. Powles RL, Morgenstern GR, Kay HE, McElwain TJ, Clink HM, Dady PJ, Barrett

A et al., Mismatched family donors for bone-marrow transplantation as treatment for

acute leukaemia. 1983. 1(8325), 612-615.

123. Giebel S, Locatelli F, Lamparelli T, Velardi A, Davies S, Frumento G, Maccario R

et al., Survival advantage with KIR ligand incompatibility in hematopoietic stem cell

transplantation from unrelated donors. Blood. 2003. 102(3), 814-819.

124. Davies SM, Ruggeri L, DeFor T, Wagner JE, Weisdorf DJ, Miller JS, Velardi A et

al., Evaluation of KIR ligand incompatibility in mismatched unrelated donor

hematopoietic transplants. Killer immunoglobulin-like receptor. Blood. 2002. 100(10),

3825-3827.

125. Bornhäuser M, Schwerdtfeger R, Martin H, Frank KH, Theuser C, and Ehninger

G., Role of KIR ligand incompatibility in hematopoietic stem cell transplantation using

unrelated donors. 7. Blood. 2004. 103, 2860-2861.

126. Cooley S, McCullar V, Wangen R, Bergemann TL, Spellman S, Weisdorf DJ, and

Miller JS., KIR reconstitution is altered by T cells in the graft and correlates with clinical

outcomes after unrelated donor transplantation. Blood. 2005. 106(13), 4370-4376.

127. Cooley S, Trachtenberg E, Bergemann TL, Saeteurn K, Klein J, Le CT, Marsh SG

et al., Donors with group B KIR haplotypes improve relapse-free survival after unrelated

hematopoietic cell transplantation for acute myelogenous leukemia. Blood. 2009. 113(3),

726-732.

128. Hsu KC, Keever-Taylor CA, Wilton A, Pinto C, Heller G, Arkun K, O'Reill RJ et

al., Improved outcome in HLA-identical sibling hematopoietic stem-cell transplantation

101

for acute myelogenous leukemia predicted by KIR and HLA genotypes. Blood. 2005.

105(12), 4878-4884.

129. King A, Balendran N, Wooding P, Carter NP, and Loke YW., CD3- leukocytes

present in the human uterus during early placentation: phenotypic and morphologic

characterization of the CD56 population. Dev Immunol. 1991. 1, 169-190.

130. Verma S, King A, and Loke YW., Expression of killer cell inhibitory receptors on

human uterine natural killer cells. Eur J Immunol. 1997. 27(4), 979-983.

131. Chazara O, Xiong S, and Moffet A., Maternal KIR and fetal HLA-C: a fine balance. J

Leukoc Biol. 2011. 90(4), 703-716.

132. Pijnenborg R, Bland JM, Robertson WB, and Brosens I., Uteroplacental arterial

changes related to interstitial trophoblast migration in early human pregnancy. Placenta.

2012. 4(4), 397-413.

133. Khong TY, De Wolf F, Roberston WB, and Brosens I., Inadequate maternal vascular

response to placentation in pregnancies complicated by pre-eclampsia and by small-for-

gestational age infants. Br J Obstet Gynaecol. 1986. 93(10), 1049-1059.

134. Apps R, Murphy SP, Fernando R, Gardner L, and Ahad T., Human leucocyte

antigen (HLA) expression of primary trophoblast cells and placental cell lines,

determined using single antigen beads to characterize allotype specificities of anti-HLA

antibodies. Immunology. 2009. 127, 26-39.

135. Hiby SE, Apps R, Sharkey A, Farrel L, Gardner L, Mulder A, Claas FH et al.,

Maternal activating KIRs protect against human reproductive failure mediated by fetal

HLA-C2. J Clin Invest. 2010. 120, 4102-4110.

136. Martin MP, Gao X, Lee JH, Nelson GW, Detels R, Goedert JJ, Buchbinder S et al.,

Epistatic interaction between KIR3DS1 and HLA-B delays the progression to AIDS. Nat

Genet. 2002. 31(4), 429-434.

137. Flores-Villanueva PO, Yunis EJ, Delgado JC, Vittinghoff E, Buchbinder S, Leung

JY, Uglialoro AM et al., Control of HIV-1 viremia and protection from AIDS are

102

associated with HLA-Bw4 homozygosity. Proc Natl Acad Sci USA. 2001. 98(9), 5140-

5145.

138. Boulet S, Song R, Kamya P, Bruneau J, Shoukry NH, Tsoukas CM, and Bernard

NF., HIV protective KIR3DL1 and HLA-B genotypes influence NK cell function

following stimulation with HLA-devoid cells. J Immunol. 2010. 184(4), 2057-2064.

139. Alter G, Heckherman D, Schneidewind A, Fadda L, Kadie CM, Carlson JM,

Oniangue-Ndza C et al., HIV-1 adaptation to NK-cell-mediated immune pressure.

Nature. 2011. 476(7358), 96-100.

140. Hou L, Chen M, Jiang B, Kariyawasam K, Ng J, and Hurley CK., In contrast to other

stimulatory natural killer cell immunoglobulin-like receptor loci, several KIR2DS5

alleles predominate in African Americans. Hum Immunol. 2009. 70(9), 733-737.

141. Steiner NK, Dakshanamurthy S, VandenBussche CJ, and Hurley CK., Extracellular

domain alterations impact surface expression of stimulatory natural killer cell receptor

KIR2DS5. Immunogenetics. 2008. 60(11), 655-667.

142. VandenBussche CJ, Mulrooney TJ, Frazier WR, Dakshanamurthy S, and Hurley

CK., Dramatically reduced surface expression of NK cell receptor KIR2DS3 is attributed

to multiple residues throughout the molecule. Genes Immun. 2009. 10(2), 162-173.

143. Hou P, Araujo E, Zhao T, Zhang M, Massenburg D, Veselits M, Doyl Ce et al., B

cell antigen receptor signaling and internalization are mutually exclusive events. PLoS

Biol. 2006. 4(7), e200.

144. Kikuchi-Maki A, Catina TL, and Campbell KS., Cutting edge: KIR2DL4 transduces

signals into human NK cells through association with the Fc receptor gamma protein. J

Immunol. 2005. 174(7), 3859-3863.

145. VandenBussche CJ, Mulrooney TJ, Frazier WR, Dakshanamurthy S, and Hurley

CK., Dramatically reduced surface expression of NK cell receptor KIR2DS3 is attributed

to multiple residues throughout the molecule. Genes Immun. 2009. 10(2), 162-173.

103

146. Rajagopalan S, Bryceson YT, Kuppusamy SP, Geraghty DE, van der Mer Ae,

Joosten I, and Long EO., Activation of NK cells by an endocytosed receptor for soluble

HLA-G. PLoS Biol. 2006. 4(1), e9.

147. Sivori S, Parolini S, Falco M, Biassoni R, Bottino C, Morreta L, and Morreta A.,

2B4 functions as a co-receptor in human NK cell activation. Eur J Immunol. 2000. 30(3),

787-793.

148. Irvine DJ, Purboo MA, Krogsgaard M, and Davis MM., Direct observation of ligand

recognition by T cells. Nature. 2002. 419(6909), 845-849.

149. Sykulev Y, Joo M, Vturina I, Tsomides TJ, and Eisen HN., Evidence that a Single

Peptide-MHC Complex on a Target Cell Can Elicit a Cytolytic T Cell Response.

Immunity. 1996. 4(6), 565-571.

150. Garrity D, Call ME, Feng J, and Wucherpfenning KW., The activating NKG2D

receptor assembles in the membrane with two signaling dimers into a hexameric

structure. Proc Natl Acad Sci USA. 2005. 102(21), 7641-7646.

151. Fan QR, Long EO, and Wiley DC., Cobalt-mediated dimerization of the human natural

killer cell inhibitory receptor. J Biol Chem. 2000. 275(31), 23700-23706.

152. Fan QR, Long EO, and iley DC., A disulfide-linked natural killer cell receptor dimer

has higher affinity for HLA-C than wild-type monomer. Eur J Immunol. 2000. 30(9),

2692-2697.

153. Geisler, C., TCR trafficking in resting and stimulated T cells. Crit Rev Immunol. 2004.

24(1), 67-86.

154. Snyder MR, Lucas M, Vivier E, Weyand CM, and Goronzy JJ., Selective Activation

of the c-Jun NH2-terminal Protein Kinase Signaling Pathway by Stimulatory KIR in the

Absence of KARAP/DAP12 in CD4+ T Cells. J Exp Med. 2003. 197(4), 437-449.

155. Snyder MR, Nakajima T, Leibson PJ, Weyand CM, and Goronzy JJ., Stimulatory

killer Ig-like receptors modulate T cell activation through DAP12-dependent and DAP12-

independent mechanisms. J Immunol. 2004. 173(6), 3725-3731.

104

156. Paloneva J, Kestilä M, Wu J, Salminen A, Böhling T, Ruotsalainen V, Hakola P et

al., Loss-of-function mutations in TYROBP (DAP12) result in a presenile dementia with

bone cysts. Nat Genet. 2000. 25(3), 357-361.

157. de Rham C, Ferrari-Lacraz S, Jendly S, Schneiter G, Dayer JM, and Villard J., The

proinflammatory cytokines IL-2, IL-15 and IL-21 modulate the repertoire of mature

human natural killer cell receptors. Arthritis Res Ther. 2007. 9(6), R125.

158. Taner SB, Pando MJ, Roberts A, Schellekens J, Marsh SG, Malmberg KJ, Parham

P et al., Interactions of NK cell receptor KIR3DL1*004 with chaperones and

conformation-specific antibody reveal a functional folded state as well as predominant

intracellular retention. J Immunol. 186(1), 62-72.