A Proteomic Screen Reveals Novel Fas Ligand Interacting Proteins within Nervous System Schwann cells

Peter Thornhill Department of Physiology

McGill University June, 2007

A thesis submitted to the Faculty of Graduate Studies and Research in partial fulfillment of the requirements of the degree of Master of

Science (MSc.).

© Peter Thornhill, 2007

1+1 Libraryand Archives Canada

Bibliothèque et Archives Canada

Published Heritage Branch

Direction du Patrimoine de l'édition

395 Wellington Street Ottawa ON K1A ON4 Canada

395, rue Wellington Ottawa ON K1A ON4 Canada

NOTICE: The author has granted a non­exclusive license allowing Library and Archives Canada to reproduce, publish, archive, preserve, conserve, communicate to the public by telecommunication or on the Internet, loan, distribute and sell theses worldwide, for commercial or non­commercial purposes, in microform, paper, electronic and/or any other formats.

The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.

ln compliance with the Canadian Privacy Act some supporting forms may have been removed from this thesis.

While these forms may be included in the document page count, their removal does not represent any loss of content from the thesis.

• •• Canada

AVIS:

Your file Votre référence ISBN: 978-0-494-38437-4 Our file Notre référence ISBN: 978-0-494-38437-4

L'auteur a accordé une licence non exclusive permettant à la Bibliothèque et Archives Canada de reproduire, publier, archiver, sauvegarder, conserver, transmettre au public par télécommunication ou par l'Internet, prêter, distribuer et vendre des thèses partout dans le monde, à des fins commerciales ou autres, sur support microforme, papier, électronique et/ou autres formats.

L'auteur conserve la propriété du droit d'auteur et des droits moraux qui protège cette thèse. Ni la thèse ni des extraits substantiels de celle-ci ne doivent être imprimés ou autrement reproduits sans son autorisation.

Conformément à la loi canadienne sur la protection de la vie privée, quelques formulaires secondaires ont été enlevés de cette thèse.

Bien que ces formulaires aient inclus dans la pagination, il n'y aura aucun contenu manquant.

ACKNOWLEDGEMENTS:

Firstly, 1 would like to greatly thank my supervisor, Dr. Julie Desbarats, for guiding me well, supporting part of my tenure in her lab, and giving such a young graduate student like myself such a long leash to try many of my own experimental ideas, sorne of which worked, and unfortunately many of which didn't.

1 would like to thank Dr. Yaël Mamane of the Sonenberg lab for advice in co­immunoprecipitation.

1 would like to thank Dr. Kurt Dejgaard for his expertise and sage advice during the interpretation and preparation part of my mass spectrometry experiments.

1 would like to thank Dr. Veen a Sangwan of the Park lab for advice and furnishing me with several antibodies.

1 would like to thank Gillian Drury for many valuable discussions both lab related and otherwise and for helping to guide me through problems with the molecular biology experiments.

Thanks to Manuelle Rongy for originaIly purifying the primary Schwann cells 1 used to create the stable FasL expressing celllines.

1 would like to thank Ken McDonald for his expertise in ceIl sorting/flow cytometry and for being an all around ni ce guy.

Thanks to Dr. Peter McPherson for generously providing antibodies to alpha and gamma adaptins

Thanks to the National Science and Engineering Research Council of Canada for supporting my masters degree.

1 would like to thank the proteomic platform at the Institute for Research in Immunology and Cancer, led by Dr. Pierre Thibault, for performing the mass spectrometry.

This work was supported by grant 53337 from the Canadian Institute of Health Research (CIHR). My collaborator, Jason B. Cohn, was supported by a CIHR Doctoral Fellowship and my supervisor, Dr. Julie Desbarats, by a CIHR New Investigator award.

Thornhill, pg. 2

CONTRIBUTIONS OF AUTHORS:

This rnanuscript based thesis contains two rnanuscripts. One is in revision after subrnission to

FEBS Letters and the second will be subrnitted shortly, also to FEBS Letters. I, Peter Thornhill,

attest that I performed aIl of the experiments presented within this thesis myself, except for the

following contributions by other authors:

Gillian Drury - Helped to perform RT-PCR experirnents for Fig. IC of Chapter 2 to compare

FasL to enhanced green fluorescent protein expression at the mRNA level.

Jason B. Cohn - Developed and provided the author with antiserurn against Sorting Nexin

18/Sorting Nexin Associated Golgi protein 1 (SNXI8/SNAGI), as weIl as advice for its use.

Dr. William L. Stanford - Supervised Jason B. Cohn.

Dr. Alan Bernstein - Aiso supervised Jason B. Cohn.

Dr. Julie Desbarats- Supervised the author, as weIl as performed sorne of the bench work found

in Fig 3A, B of Chapter 3.

y oon Kow Young - Counted a large number of FasL staining puncta within Schwann cells to

generate the statistics found in Fig. 3E of Chapter 3.

Others- Other non-authors have also helped in ways that do not merit authorship and are outlined

in the acknowledgements section.

I, Dr. Julie Desbarats, attest to the accuracy of these statements:

I, Peter Thornhill, attest to the accuracy of these statements:

Thornhill, pg. 3

~ 1

ABSTRACT:

Fas Ligand (FasL) binds to the Fas receptor to induce apoptosis or activate other

signaling pathways. FasL can also transduce "reverse signais" and thus participate in

bidirectional signaling. The FasL intracellular domain contains consensus sequences for

phosphorylation and a proline rich protein interaction domain. This latter region of FasL

has previously been implicated in FasL reverse signaling and regulation of FasL surface

expression. In this report, we sought to identify novel FasL interacting proteins to help

understand signaling through and trafficking of this death factor Using mass

spectrometry, we identified sorting nexin 18, adaptin ~, Grh2, PACSIN2 and PACSIN3

as FasL interacting proteins. RNAi mediated knockdown of Grh2 significantly reduced

the surface expression of FasL and increased its expression intracellularly. Our data

show that Grb2 con troIs the suhcellular localization of FasL. AlI other proteins

identified in our screen could be classified as trafficking-associated proteins, highlighting

the complex regulation of the surface expression of this death factor.

Thornhill, pg. 4

RÉSUMÉ:

La protéine Fas Ligand (FasL) se lie au récepteur Fas afin d'induire l'apoptose ou

activer d'autres voies de signalisation. La protéine FasL est aussi impliquée dans la

signalisation rétrograde et participe à une signalisation bidirectionnelle. Le domaine

intracellulaire de FasL contient une séquence consensus de phosphorylation ainsi qu'un

domaine d'interaction protéique riche en proline. Ce domaine a précédemment été

impliqué dans la signalisation rétrograde de FasL et dans la régulation de son expression

à la surface cellulaire. Dans cette thèse, nous désirions identifier de nouvelles

interactions protéiques avec FasL afin de contribuer à la compréhension du trafic

cellulaire et de la signalisation de ce facteur apoptotique. Grâce à la spectrométrie de

masse, nous avons identifié Sorting Nexin 18, Adaptin ~,PACSIN2, Grb2 et PACSIN3

comme étant des protéines interagissant avec FasL. En utilisant la méthode de l'ARN

par interférence, nous avons diminué l'expression de Grb2 et observé que l'expression de

FasL à la surface cellulaire a été significativement réduite alors que son expression

intracellulaire a été augmentée. Nos résultats démontrent que Grb2 est impliquée dans le

contrôle de la localisation subcellulaire de FasL. Les autres protéines identifiées dans

notre criblage pourraient être classifiées comme protéines associées au trafic cellulaire,

suggérant ainsi la complexité de la régulation de ce facteur apoptotique à la surface

cellulaire.

Thornhill, pg. 5

TABLE OF CONTENTS:

Acknowledgements ........................................................................................................ 2 Contributions of Authors ................................................................................................. .3 Abstract. .................................................................................................................... .4 Résumé ................................................................................................................................ 5 Abbreviations ........................................................................................................................ 7 Chapter 1: Introduction .......................................................................................................... 10

Fas: the Molecule ...................................................................................................... 11 Fas Death Signaling .................................................................................................. .12 Non Apoptotic Fas signaling ........................................................................................ 12 FasL: the Molecule .................................................................................................... 14 FasL Expression and Function ....................................................................................... 15 Regulation of FasL .................................................................................................... 19 FasL Reverse Signaling .............................................................................................. 22 FasL Interacting Proteins ............................................................................................. 24

Objectives .......................................................................................................................... 28 Chapter 2: A Proteomic Screen Reveals Novel Fas Ligand Interacting Proteins within Nervous System Schwann cells ............................................................................................................................ 29

Abstract. ................................................................................................................ 31 Introduction ............................................................................................................. 32 Materials and Methods ............................................................................................... 34 Results and Figures ................................................................................................... .40 Discussion ............................................................................................................. .46

Transition to Chapter 3 .......................................................................................................... .49 Chapter 3: The Adaptor Protein Grb2 Interacts with Fas Ligand and Regulates its Cell Surface Expression in Schwann cells ........................................................................................................................... 50

Abstract. ................................................................................................................ 51 Introduction ............................................................................................................. 53 Materials and Methods ............................................................................................... 54 Results and Figures .................................................................................................... 57 Discussion ............................................................................................................... 64

Chapter 4: Final Conclusions .................................................................................................... 66 References ................................................................................................................. 71 Appendix ................................................................................................................... 82

Permissions ..................................................................................................... 83 Ethics Certificates .............................................................................................. 85

TABLES & FIGURES: Chapter 1

Figure 1 Autoimmune Lymphoproliferative Syndrome Subtypes ...................................... 17 Chapter 2

Figure 1 FasL deletion mutants are expressed more efficiently than wtFasL. ........................ .41 Figure 2 FasL specifically co-immunoprecipitates six bands ............................................ .42 Table 1 LC-MSMS Proteins "Hits" Co-precipitated with FasLL'133 in Schwann Cells ............... .43 Figure 3 Adaptin P co-precipitates with FasL. ............................................................ .44 Figure 4 SNX18 co-precipitates with FasL. ............................................................... .45

Chapter 3 Figure 1 FasL deletion mutants: revisited .................................................................. .57 Figure 2 Grb2 co-precipitates with amino acids 34-70 ofFasL. ......................................... 58 Figure 3 Grb2 regulates the surface expression of FasL. .................................................. 60 Figure 4 Grb2 interacts with the FasL interacting proteins Sorting Nexin 18 and adaptin p .......... 62 Figure 5 Grb2 knockdown dramatically attenuates FasL-adaptin P co-precipitation ................... 63

Chapter4 Figure 1 Possible functions of novel FasL interacting proteins .......................................... 67 Figure 2 Model of FasL trafficking based on experiments found in Chapter 3 ........................ 69

Thornhill, pg. 6

1"'"' .. ABBREVIATIONS: ,

ADAM A Disintegrin And Metalloproteinase

AICD Activation Induced Cell Death

ALPS Autoirnrnune LymphoProliferative Syndrome

AP Adaptor Protein

APAFI APoptosis Activating Factor 1

APC Antigen Presenting Cell

CAD Caspase Activated DNase

CC Coiled Coil domain

CKI Casein Kinase 1

cPLA2 cytosolic Phospholipase A2

CRD Cysteine Rich Domain

CTL Cytotoxic T Lymphocyte

DD Death Domain

DISC Death Inducing Signaling Complex

EGFP Enhanced Green Fluorescent Protein

EGFR Epidermal Growth Factor Receptor

EGR Early Growth factor Response

EH Epsin Homology domain

ERK Extracellular-signal Regulated Kinase

FADD Fas Associated Death Domain

FAP-l Fas Associated Phosphatase

FasL Fas Ligand

FKHRLI ForKHead transcriptional ReguLator 1

FLIP FLICE-like Inhibitory Protein

GDP Guanosine DiPhosphate

GFAP Glial Fibrillary Acidic Protein

Grb2 Growth Receptor Bound 2

GSK3~ Glycogen Synthase Kinase 3~

r GST Glutathione-S-Transferase

GTP Guanosine TriPhosphate

Thornhill, pg. 7

ICD

IL

IP

IRES

JNK

LC-MSMS

MAPK

MEK

MFI

MHC

MMP

mRNA

NFAT

NF-KB

NGFR

NK

PACSIN

PI3K

PKB

PLAD

PSTPIP

PTP-PEST

PX

RNAi

RT-PCR

sFasL

SH2

SH3

siRNA

SNX18

SOS

IntraCeIlular Domain

Interleukin

ImmunoPrecipitation

InternaI Ribosomal Entry Site

c-Jun N-terminal Kinase

Liquid Chromatography cou pIed tandem Mass Spectrometry

Mitogen Activated Protein Kinase

Mitogen activated ERK Kinase

Mean Fluorescence Intensity

Major Histocompatibility Complex

Matrix MetaIloProteinase

messenger RiboNuc1eic Acid

Nuc1ear Factor of Activated T cells

Nuc1ear Factor KB

Nerve Growth Factor Receptor

N aturaI Killer cell

Protein kinase c And Casein kinase Substrate ln Neurons

Phosphatidyl Inositol 3' Kinase

Protein Kinase B

Pre Ligand Assembly Domain

Proline, Serine and Threonine Phosphatase Interacting Protein

Protein Tyrosine Phosphatase PEST

Phox Homology domain

RNA interference

Reverse Transcription-Polymerase Chain Reaction

soluble Fas Ligand

Src Homology 2 domain

Src Homology 3 domain

short interfering RNA

Sorting NeXin 18

Son Of Sevenless

Thornhill, pg. 8

r TCR T Cell Receptor

TNF Tumor Necrosis Factor

TNFR Tumor Necrosis Factor Receptor

TNFSF Tumor Necrosis Factor SuperFamily

WASP Wiskott Aldrich Syndrome Protein

Thornhill, pg. 9

Chapter 1

Introduction and Objectives

Thornhill, pg. 10

LITERA TURE REVIEW:

PARTI: Fas

Fas: The mole cule

The discovery of the Fas molecule arose from the observation that a specific

antibody (termed anti-Fas) secreted from aB ceIl clone, created by immunizing mice

with human fibroblast cells, was cytolytic to human cells in culture [1,2]. Nanogram

quantities of anti-Fas antibodies arrested proliferation, and induced DNA fragmentation

and morphological changes consistent with programmed ceIl death, or apoptosis [2]. In

the early 1990s, the Fas antigen, hereafter caIled Fas, was cloned from a human T

lymphoma cDNA library [3]. Today Fas is known to be a 327 amino acid, type 1

transmembrane protein and a member of the Tumor Necrosis Factor ReceptorlNerve

Growth Factor Receptor (TNFRINGFR) superfamily of cytokine receptors. Membership

in this superfamily is defined based on the presence of cysteine rich pseudo-repeats (also

known as cysteine rich domains; CRDs) , and Fas possesses three such extracellular

CRDs [4]. The N-terminal CRD, CRD1, confers the ability of Fas to self-trimerize in the

absence of ligand and is also known as the pre-ligand assembly domain (PLAD) [5,6] AlI

three CRDs of the Fas extraceIlular domain are required for binding to FasL [7].

Mutagenesis of the Fas intracellular domain revealed the presence of another

conserved domain, termed the death domain (DD), necessary for apoptosis in response to

agonistic Fas antibodies [8]. Death domains are protein modules containing six

antiparallel, amphipathic alpha helices organized into a conserved overaIl fold, and are

present in a subset of the TNFR superfamily [9]. In addition to TNF receptors, multiple

cytosolic adaptor proteins also contain DDs and are recruited to Fas via homotypic

interactions to initiate Fas signaling.

Thornhill, pg. Il

Fas death signaling

While FasL-Fas interactions activate multiple and pleiotropic signaling cascades

[10-12], the best characterized outcome of Fas signaling is apoptosis. Vpon activation by

FasL or agonistic antibodies, Fas associates with the Fas associated death domain

(FADD), a DD containing adaptor protein, which further recruits Caspase 8 and/or

Caspase 10 to form the Death-Inducing Signaling Complex (DISC) [12]. Caspase 8 is

activated within the DISC by autocatalytic c1eavage. Activated Caspase 8 proceeds to

activate either Caspase 3 directly, as part of the extrinsic apoptosis pathway or indirectly,

via mitochondrial destabilization, as part of the intrinsic pathway [13]. In the intrinsic

pathway, c1eavage of the protein Bid by activated Caspase 8 results in destabilization of

the mitochondrial membrane potential, and release of cytochrome cinto the cytosol [14].

Release of cytochrome c stimulates formation of the apoptosome, composed of Caspase

9 and the apoptotic protease activating factor 1 (APAF1), which can c1eave and activate

Caspase 3 [15]. Both pathways converge on activation of the executioner Caspase 3,

resulting in cleavage of multiple cellular proteins, and activation of the Caspase-activated

DNase (CAD), resulting in characteristic internuc1eosomal chromatin fragmentation and

apoptosis [16].

Non-apoptotic Fas Signaling

Cellular outcomes

In addition to the better characterized apoptotic outcomes of Fas signaling, Fas

engagement can also result in multiple cellular outcomes besides apoptosis, complicating

the interpretation of many experiments. For instance, Fas signaling can induce both

proliferation and apoptosis of fibroblast cells [17 ,18]. Intriguingly, although Fas

engagement normally kills hepatocytes in the healthy liver, the outcome of Fas signaling

is reversed after partial hepatectomy, where Fas stimulation accelerates liver regeneration

[19]. Fas also induces hypertrophy of cardiomyocytes in vitro and in vivo [20]. In the

nervous system, Fas engagement induces neurite outgrowth as well as neurite branching

[21,22] and Fas deficient lpr mice exhibit increased susceptibility to Parkinson's disease

[23] and delayed peripheral nerve regeneration [21]. Even in the immune system, where

Fas mutants exhibit excessive lymphoproliferation due to apoptotic defects, Fas has been

Thornhill, pg. 12

documented to have non apoptotic effects. Fas engagement can induce secretion of pro

inflammatory cytokines, such as Interleukin 1 (IL-l), from activated immune cells [24].

Stimulation of Fas with agonistic antibodies, which stimulate Fas signaling, can cause T

and B lymphocytes to proliferate in culture [25,26]. This interesting ability of FasL,

acting through Fas, to paradoxically cause both proliferation and apoptosis parallels other

proteins of the TNF superfamily, many of which are also multifunctional [27].

Molecular mechanisms

The non apoptotic effects of Fas signaling downstream of FasL imply the ability

of Fas to activate other pathways, distinct from the classical apoptotic caspase cascade.

Surprisingly, even sorne of the non-apoptotic Fas signaling pathways require caspase

activation. Fas induced T cell proliferation is blocked by caspase inhibitors [28].

Interestingly, there is a growing body of work indicating that the caspases, key mediators

of the cell death pathway, also regulate nonapoptotic cellular functions such as

proliferation and differentiation [29].

Other non apoptotic effects of Fas require activation of signaling pathways

distinct from the caspase cascade. Fas induced neurite outgrowth requires activation of

the Extracellular-signal Regulated Kinase (ERK) Mitogen Activated Protein Kinase

(MAPK) pathway [21], and Fas activation in fibroblasts triggers activation ofERK as

weIl as nuclear factor KB (NF-KB) [30]. The ERK MAPK cascade is one of three MAPK

phosphorylation cascades activated downstream of multiple growth factors, and is

associated with growth and differentiation. NF-KB is a ubiquitous transcription factor

that controls the expression of genes involved in inflammation, apoptosis, and cell cycle.

NF-KB is activated by multiple stimuli which trigger the phosphorylation and subsequent

degradation of its cytosolic inhibitor, I-KB [31]. The cytosolic Fas- associated death

domain-Like interleukin Ibeta-converting enzyme)-Inhibitory Protein (FLIP), has been

shown to mediate activation of both of these pathways downstream of Fas [32]. FLIP

binds to the adaptor protein FADD, and acts as an inhibitor of apoptosis downstream of

death receptors [33], and can instead activate other signaling pathways. Therefore, FLIP

may be a key regulator of Fas non apoptotic versus apoptotic signaling. Rather than

proliferate, FLIP-/- fibroblasts die in response to FasL [34]. The Fas Associated

Thornhill, pg. 13

Phosphatase (FAP-l) is another Fas interacting protein that can inhibit apoptosis [35],

and alternatively activate signaling via NF-KB [36]. The outcome of Fas signaling may

also be controlled by exposure of cells to growth factors, whose downstream effects can

interfere with Fas induced DISC formation [37], and upregulate mitochondrial proteins

which act as inhibitors of apoptosis [38].

PART II: Fas Ligand

FasL: the mole cule

Fas Ligand (FasL) is a 281 amino acid single pass transmembrane protein, first

identified in a screen for cell surface proteins that bound to Fas. In 1993, it was

discovered that a certain cellline, PC60-dlOS, was capable of killing Fas positive, but

not Fas negative, target cells in co-culture experiments, implying the presence of a Fas

"ligand" protein [39]. To identify the ligand, a Fas-IgG fusion protein was produced and

used to sort high FasL expressing cells as a source for the creation of a cDNA library

[40]. Plasmid DNA was then transfected into COS cells and plasmids were re-isolated

from cells that bound to Fas-IgG in vitro. Sequences identified after multiple rounds of

this type of panning, led to the identification of the Fas Ligand. FasL is a type II

transmembrane protein, meaning the C terminus rather than the N terminus is presented

extracellularly, and a member of the TNF superfamily. Membership in this family is

based on homology within the extracellular domains of these molecules. Although there

is a relatively low level of primary sequence homology (not exceeding 35%) amongst

members of the TNF superfamily, they are predicted to have similar topology and exhibit

a homotrimeric structure [41]. Mutagenic analyses of FasL indicate that the membrane

proximal amino acids of the extracellular domain are responsible for this trimeric self

association, while the extreme C terminal amino acids are required for Fas receptor

binding [42]. The FasL extracellular domain also contains three sites ofN or asparagine

linked glycosylation, which do not appear to be required for induction of apoptosis via

Fas [42]. In addition to binding Fas, the FasL extracellular domain also binds to the

secreted decoy receptor 3 protein, which blocks its activity as an apoptosis inducer via

Fas [43,44]. Intracellularly, FasL contains a long, 80 amino acid cytoplasmic tail which

Thornhill, pg. 14

is distinct from the shorter intracellular tails of other TNF superfamily members. Within

this intracellular tail are putative consensus sequences for phosphorylation by Casein

Kinase 1 (CKl) and glycogen synthase kinase 3~ (GSK3~), as well as an extended region

of very high proline content, capable of interacting with Src Homology 3 (SH3) and WW

domain containing proteins [45,46].

FasL Expression and Function

Immune cells

FasL expression is best documented in the immune system, where it is thought to

regulate the survival of T cells. Peripheral T cells initially undergo a clonaI expansion

phase after exposure to antigen as the y proliferate in response to interleukin-2 (IL-2)

[47]. Later however, these expanded cells are eventually eliminated by an apoptotic

process called Activation Induced Cell Death (AICD) in order to terminate the immune

response and maintain homeostasis. FasL expression is induced in T cells in vitro after

stimulation of the T cell Receptor complex (TCR), and a soluble Fas-IgG fusion protein

inhibits AICD [48-50]. Interestingly, IL-2, which initially upregulates FasL and

promotes T cells proliferation, primes cells for death via FasL later in the immune

response [51]. Due to the ability of single, isolated T cells to undergo AICD in response

to TCR engagement, Fas and FasL have been proposed to be capable of interacting in a

cell autonomous manner to indu ce cell "suicide" [49].

In contrast to the role for FasL in regulating the survival of peripheral T cens, the

role for FasL in regulating deletion of unwanted T cells during their development and

maturation in the thymus is more controversial. While sorne reports have indicated that

FasL was involved in thymic deletion (negative selection) of autoreactive thymocytes

[52], others have found that FasL was not involved [53]. These contradictory reports

may perhaps be explained by the finding that FasL is involved in thymic deletion of T

cells upon exposure to high levels but not low levels of antigen [54].

FasL is also thought to mediate the target cell killing activities of cytotoxic T

lymphocytes (CTLs) and natural killer (NK) cens [55,56]. CTLs and NK cells are potent

killers of tumorigenic and virally infected cens at least in part via the regulated secretion

of the contents of specialized secretory lysosomes, which contain lytic proteins such as

Thornhill, pg. 15

perforin and granzyme [57]. There is evidence that FasL is localized to these secretory

lysosomes in cells that contain these specialized organelles and that appearance of

surface FasL in these cells correlates with their de granulation [58].

Immune phenotype of F as/F asL mutants

Consistent with the role for Fas and FasL in the termination of immune

responses, mice and humans with inactivating mutations in the FaslFasL system exhibit a

dramatic immune phenotype. Lpr (lymphoproliferative) mice, which cannot efficiently

transcribe Fas due to the insertion of a transposable DNA element into a gene intron,

develop lymphadenopathy and are may display symptoms resembling a systemic lupus

erythematosus-like autoimmune disease [59]. Similarly gld mice (generalized

lymphoproliferative disease) also exhibit lymphadenopathy and autoimmune symptoms

due to an inactivating missense mutation (Phenylalanine to Leucine) at the extreme

carboxyl terminus of FasL [60]. In gld animaIs, FasL is expressed normally at the cell

surface and is properly oligomerized, yet was reported to be incapable of killing Fas

expressing cells [60,61]. The phenotype of FasL knockout animals, however, indicated

a residual activity of the gld allele. FasC/- animaIs exhibit splenomegaly, and a more

severe autoimmune phenotype compared to gld animaIs, with lymphocytic infiltration

into multiple organs and death of greater than 50% of mice by 4 months of age [62]. A

third mouse mutation in the Fas system, named lpl'g (lpr complementing gld), results in

lymph node hyperplasia and high titers of autoantibodies [63]. The lprcg mutation results

in local unfolding of the Fas death domain, such that the adaptor protein FADD can no

longer bind to Fas [64].

In humans, mutations in the Fas/FasL system can result in Autoimmune

LymphoProliferative Syndrome (ALPS; Canale Smite Syndrome). Children with ALPS

develop lymphoid hyperplasia and autoimmunity reminiscent of lupus, and ALPS is a

good ex ample of a disease where the animal models have been highly informative. The

disease is segregated into subtypes, based on the functional results of the various

mutations found in the disease. ALPS 0, caused by a Fas null mutation, parallels lpr

mice; ALPS la, results primarily from intracellular Fas mutations, and parallels lprcg

mice; while ALPS lb is caused by FasL mutations, similar to gld mice [65].

Thornhill, pg. 16

Interestingly, mutations discovered in caspase 10 in ALPS II patients led to the

identification of caspase 10 as a component of the Fas DISC during apoptotic signaling

[66]. A table outlining the mutations in the various proteins along the Fas death pathway

in ALPS has been reproduced from reference [65] below:

Hmmm AniJ'lllt d~ .modds

ALft!.lb #{1d1$1d (SLBt)

.. IFNi' ALft!.9rilill B.o

ALl'SlIîUI"'ffNi'ff

ALl'Slf .(~.to,J

~.~. .~

rotO>

FAPP FADUko +: ...>}>rotlferlltlon dl!feet

~.CAsP1{}

Till l'Ill', CrmA, pJ:5fbsence .". ...... Ikatlt ~l8r II10ek or

ALPS Apoptosis

Fig. 1 Various ALPS Subtypes Related to Their Respective Mouse Mutations

Within the Cell Death Pathway.

New Abbreviations: TRAIL (TNF Related Apoptosis Inducing Ligand); DR (Death

Receptor); CASP (Caspase); TRADD (TNF Receptor Associated Death Domain); TL1-A

(TNF ligand-related molecule lA); SLE (Systemic Lupus Erythematosus)

Expression of FasL in non-immune ceUs

FasL is also expressed outside the immune system, in the nervous system [67-69],

the reproductive system [70], the eye[71], and on various epithelia [72]. Many of the se

tissues are considered to be "immune privileged," meaning that it is difficult to evoke

immune responses in these anatomicallocations. More specifically, immune privileged

sites are c1assically defined as anatornicallocations that are permissive of allograft or

xenograft survival [73]. FasL has been hypothesized to be a mediator of immune

privilege in these tissues, limiting the availability of such tissues to immune surveillance

via the engagement of Fas on tissue infiltrating lymphocytes [67,68,71]. In a similar

Thornhill, pg. 17

r

fashion, the expression of FasL upon the surface of vascular endothelial cells may

negatively regulate the extravasation of Fas expressing lymphocytes to sites of

inflammation [74]. Inflammatory mediators, such as tumor necrosis factor alpha (TNF

a), which upregulate adhesion molecules for lymphocyte extravasation [75] also

downregulate FasL [74]. FasL is also expressed by various tumors [68,76,77], and their

expression of FasL may be one of the multiple mechanisms used by neoplastie cells to

escape immune rejection.

Reports published in the journals Science and Nature in 1995 indicated that FasL

may be a key regulator of immune privilege and initially piqued the interest of many

transplant biology and tumor immunology researchers. Bellgrau et al. [78] reported that

immunologieally mismatched grafts of wild type mouse testis survived, while testis

grafts from gld miee were rapidly rejected. Griffith et al. [71] similarly reported that

inflammatory cells entering the eye in response to viral infection underwent Fas

dependent apoptosis in wild type animaIs, whereas gld animaIs exhibited inflammation

and immune invasion of ocular tissue. Therefore FasL appeared to be required for

immune privilege, but was it sufficient? A single report indicated that FasL expressing

myoblasts surrounding pancreatie Langerhans cells allografts protected them from

immune rejection [79]. However, multiple reports have indicated that ectopie FasL

expression within transplanted tissues leads to premature rejection [80-82], inc1uding one

report that used a very similar methodology [83]. These latter reports also documented

massive infiltration of neutrophils to the graft site, associated with the rejection.

Neutrophils may be attracted to sites of high FasL expression via Fas induced production

of interleukin 1 (IL-l) [24] and/or IL-8 [84], two neutrophil chemoattractants. Studies of

FasL expression in tumor cells and its role in "immune counterattack" paralleled the

contradietory findings of transplant biologists. While sorne studies found that FasL

expression increased tumor survival in vivo [85], others found that FasL expression in

tu mors caused neutrophilic infiltration and destruction [86].

Thornhill, pg. 18

r

Regulation of Fas Ligand

Transcriptional Regulation of FasL

As a protein with the capability of inducing apoptosis under certain conditions,

the expression of FasL must be tightly controlled. Few cell types constitutively express

FasL messenger RNA (mRNA) and transcription of FasL is often induced by external

stimuli. In T cells, one such stimulus is activation of the TCR complex, which results in

the expression of the FasL mRNA and precedes AICD [48]. Two sites in the FasL

promoter region have been identified as nuclear factor of activated T cells (NF A T)

transcription factor binding sites that are critical for optimal expression of FasL in

activated T cells [87]. Other transcription factors, such as the early growth factor

response (EGR) factors 1 and 3, bind adjacent regions within the FasL promoter and

cooperate synergistically with NF AT in the induction of FasL transcription downstream

ofT cell activation [88]. NF-KB has also been documented to participate in FasL

induction during AICD [89].

Another stimulus for FasL expression is cellular "stress." Exposure of cultured

cells to damaging stimuli, such as UV light, gamma rays, or certain toxic drugs, causes

them to upregulate FasL expression and commit stress induced apoptosis. Induction of

FasL expression in response to at least sorne stress stimuli is mediated by activation of

the c-Jun N terminal Kinase (JNK; also known as stress activated protein kinase), another

MAPK cascade distinct from ERK that is activated by cellular stress [90]. As its name

implies, the kinase JNK phosphorylates the protein c-Jun, allowing it to translocate to the

nucleus and activate transcription as part of the AP-1 complex. Formation of the AP-1

complex upon the FasL promoter downstream of JNK activation mediates FasL

induction in response to stress [91]. The transcription factor NF-KB has also been shown

to participate in FasL expression in response to cellular stress induced by the cancer

drugs etoposide and teniposide [92]. Growth factor withdrawal, another cellular stress,

also stimulates FasL expression. Many growth factors bind receptors which facilitate the

activation of the Akt protein kinase, which promotes cellular survival by phosphorylating

and inhibiting the forkhead transcription al regulator-1 (FKHRL1) [93]. FKHRL1 binds

to two response elements in the FasL promoter and facilitates its expression [93].

Thornhill, pg. 19

Translational Regulation of FasL

Little is known of the translation of the FasL mRNA, although it has been

observed that levels of FasL mRNA do not necessarily correlate with levels of FasL

protein expression [94]. One possible explanation for this observation could be

regulation of FasL translation. In a recent paper, Xiao et al. [95] documented a dramatic

augmentation in protein expression levels of various FasL intracellular tail deletion

mutants in stable celllines compared to the wild type protein, despite equal expression of

their respective mRNAs. These results are potentially suggestive of a negative

translational regulatory element within the nucleotides encoding the FasL amino

terminuslintracellular tail. This interpretation is complicated, however, by the

subsequent finding that the se deletion mutants exhibit decreased cytotoxicity [96], and

there are alternative explanations for differing protein levels, such as altered protein half

life, which the authors did not investigate. Dramatic increases in radioactive methionine

incorporation after very short (5 minute) labelings of cells expressing the deletion

mutants compared to cells expressing the wild type protein, [95] nonetheless lend

credibility to the translation hypothesis. This thesis utilizes the same intracellular tail

deletion mutants as Xiao et al. for the analysis of protein interactions, and using a slightly

different experimental system, reproduces the augmentation of FasL protein expression

upon FasL n-terminal deletion (see Chapter 2, Fig. 1).

Post-translational Regulation of FasL

Another mechanism for the regulation of FasL expression and function is the

shedding of its extracellular domain to generate a soluble FasL (sFasL), leaving behind a

transmembrane and intracellular domain "stump." Other TNF family members, such as

TNF, are also cleaved in a similar fashion [97]. Extracellular domain shedding of FasL

is achieved by the action of metalloproteinase enzymes [98,99]. sFasL has

approximately 1000 fold less killing activity than the membrane bound form of FasL

[100], although in certain tissues sFasL can bind to the extracellular matrix protein

fibronectin to restore its cytotoxicity [101]. Little is known of the fate of the remaining

cleavage product containing the FasL intracellular domain, although there is evidence

Thornhill, pg. 20

that the analogous cleavage product of TNF enters the nucleus and influences gene

expression [102]. The A Disintegrin And Metalloproteinase subfamily member 10

(ADAMlO) has recently been identified as an enzyme capable of generating this sFasL in

vivo [103], and Matrix Metalloproteinase 7 (MMP7) has also been demonstrated to

cleave FasL in vitro [104]. Interestingly, MMP 7 expression is negatively correlated

with survival of colorectal cancer patients [105], perhaps due to its ability to cleave FasL

and downregulate apoptosis [106].

FasL has also recently been documented to be post-translationally regulated by its

entry into lipid rafts. Lipid rafts are cholesterol, sphingolipid and saturated fatty acid rich

microdomains within the celI membrane which are thought to concentrate proteins

involved in signal transduction [107]. Two recent studies provide evidence that FasL

enters these membrane microdomains and that disruption of these microdomains strongly

decreases the apoptosis inducing activity of FasL [108,109]. Although the two studies

differ somewhat in their results, both studies also indicate that regions of the FasL

intracellular domain, particularly the proline rich region, are necessary for membrane raft

localization.

The localization of FasL to the cell surface, where it is readily available to bind

Fas and exert its biological effects, is also a regulated process. Multiple reports have

documented significant intracellular retention of FasL [58,95,110-113]. In a subset of

hematopoietic cells that contain secretory lysosomes, such as cytotoxic T lymphocytes

and mast celIs, FasL is predominantly intracellular and colocalizes with lysosomal

markers, while cells lacking the se specialized organelles express cell surface FasL with

variable levels of intracellular retenti on depending on the cell type [58,95,113]. The

FasL intracellular domain (ICD) is likely sufficient for its intracellular retention, as

fusion of the FasL ICD to the cell surface expressed CD69 protein results in intracellular

retenti on of the fusion prote in [58]. More specificalIy, deletion of the proline rich region

of the FasL ICD results in surface expression, rather than intracellular retenti on of FasL

in cells with secretory lysosomes [113]. In cytotoxic T lymphocytes, stimulation of the

TCR results in regulated exocytosis from FasL containing secretory lysosomes [114].

Secreted microvesicles bearing FasL have been isolated, and are thought to derive from

these secretory lysosomes [115-117]. In contrast to cells which contain secretory

Thornhill, pg. 21

lysosomes, little is known of the factors which regulate cell surface expression of FasL in

other cell types.

FasL Reverse Signaling

There is significant evidence that in addition to initiating signaling through Fas,

FasL can also transduce "reverse signaIs," and thus participate in bidirectional signaling.

This phenomenon is documented for several members of the TNF superfamily, which

transduce many critical and diverse signaIs in normal immune function, such as cellular

proliferation, cytokine and growth factor secretion, and immunoglobin class switching

[118]. Two papers published in 1998 were the first to document the effects of FasL

reverse signaling. Desbarats et al. [119] reported that stimulation of primary CD4+ T

cells with a Fas-Irnmunoglobin fusion protein inhibited their proliferation and caused

them to enter cell cycle arrest and later cell death. By documenting the same effect in

Fas, but not FasL mutant mice, the authors demonstrated cell cycle arrest occurred via

FasL reverse signaling and could not be explained by the blocking of Fas/FasL

interactions. Interestingly, Suzuki et al. [120] found that stimulation of CD8+ T cell

clones in a similar fashion augmented their proliferation. A follow up study later

demonstrated, that although CD4+ T cells can also be positively stimulated via FasL

reverse signaling, a concomitant upregulation of surface FasL causes CD4+ T cells to

later undergo apoptosis, whereas CD8+ T cells are relatively insensitive to Fas induced

death [121]. Therefore, differences in sensitivity to Fas induced apoptosis between the

two cell types may explain the seemingly contradictory data of the two above

experiments.

Cross-linking of endogenous Sertoli cell FasL induces activation of the ERK

cascade and subsequent activation of cytosolic phospholipase A2 (cPLA2) [70].

Inhibition of either pathway results in decreased levels of sFasL in cultured cell

supernatants [70], suggesting that FasL reverse signaling in Sertoli cells may act to

regulate the production of sFasL. In T cells, cross linking of an overexpressed FasL

intracellular domain also results in ERK activation, as well as activation of the

phosphatidyl inostiol 3' kinase (PI3K) protein kinase B (PKB) pathway [122]. The Src

homology 3 (SH3) domain containing adaptor proteins Growth Receptor Bound 2 (Grb2)

Thornhill, pg. 22

and p85, important for triggering both of these pathways, are recruited to the FasL ICD,

in response to Fas stimulation [122]. SH3 domains bind to proline rich sequences [46],

and are therefore likely recruited to the proline rich region of the FasL ICD. Cross

linking of this FasL intracellular tail construct also activates reporter plasmids for the

transcription factors NF AT , AP-l, as weIl as transcription of the cytokines Interferon y

and IL-2 [122]. Inhibition ofERK activation abrogates cytokine production induced by

FasL reverse signaling [122], suggesting that this pathway may underlie FasL reverse

signaling induced T cell proliferation.

Interestingly, TNF, the most closely related member of the TNFSF, can also

trigger the ERK cascade via reverse signaling [123]. Unlike Fas L, TNF does not contain

a large proline rich region. The TNF ICD contains a highly conserved SXXS sequence,

which is a consensus target sequence for phosphorylation by Casein Kinase 1 (CKl).

This sequence is present in many members of the TNFSF and has been documented to

change phosphorylation state during TNF reverse signaling [124]. Changes in Serine

phosphorylation have also been documented during FasL reverse signaling as weIl [122].

Collectively, these data imply that the CKI consensus sequence may be sufficient for

FasL reverse signaling induced ERK activation, as TNF is capable of activating this

cascade despite the lack of a proline rich domain. To date however, a convincing

molecular mechanism linking TNFSF member serine phosphorylation to ERK activation

has not been defined. TNFSF reverse signaling has important clinical implications, as

antibodies designed to inhibit TNF activity in autoimmune diseases such as Rheumatoid

Arthritis and Crohn's disease are currently being evaluated in clinical trials. In addition

to blocking engagement of TNFR signaling, these antibodies can also sufficiently cross

link TNF to activate TNF reverse signaling [125], and may lead to unexpected clinical

outcomes by modulating cytokine production and immune function [125,126].

Thornhill, pg. 23

".---. 1

FasL Interacting Proteins

Multiple proteins capable of binding to FasL in T cells have been identified. A

group led by O. Janssen in Germany conducted a screen in 2002 for Glutathione S

Transferase (GST) fusion proteins capable of co-precipitating FasL from FasL

overexpressing KFL-9 cells [45,46]. Many of the proteins identified contain SR3 or

WW domains, which are capable of binding to proline rich sequences. Of the proteins

identified, a surprising number of proteins modulate FasL localization, and several may

participate in FasL reverse signaling. As this thesis focuses on FasL interacting proteins,

the most relevant findings published to date for each interacting protein will be outlined

below.

Grb2

Grb2 is an adaptor protein containing one Src homology 2 (SR2) and two SR3

domains. Its SR2 domain allows it to bind to phosphotyrosine residues found in receptor

tyrosine kinases, such as the Epidermal Growth Factor Receptor (EGFR) [127]. Grb2

binding recruits the guanosine triphosphate (GTP) exhange protein Son of Sevenless

(SOS) to the membrane via its SR3 domain [128]. SOS can then complex with the Ras

GTPase protein and activate it by exchanging Ras bound guanosine diphosphate (GDP)

for GTP [128]. Activated Ras can then activate the protein kinase Raf, which can further

activate a kinase cascade by phosphorylating the Mitogen activated ERK kinase (MEK),

which can phosphorylate ERK [129]. Phosphorylated ERK enters the nucleus and can

affect gene transcription by phosphorylating transcription factors.

Grb2 was originally identified as a protein co-precipitated with a GST -FasL ICD

fusion prote in in T cells [45]. Based on its ability to function as an adaptor protein for

ERK activation downstream of receptor tyrosine kinases, Grb2 has been hypothesized to

mediate FasL reverse signaling induced ERK activation. In this thesis, evidence is

presented that in addition to its potential role in FasL reverse signaling, Grb2 may also

act as a regulator of FasL localization (see Chapter 3).

P85 Subunit of PI3K

Thornhill, pg. 24

The p85 subunit of the phosphatidyl inositol3'phosphate kinase (PI3K) is an SH3

do main containing adaptor protein for the enzyrnaticaIly active 110 kilo dalton subunit of

PI3K. As its name suggests, PI3K phosphorylates the 3' hydroxyl group ofthe

phosphatidyl inositol phospholipid within the ceIl membrane. Local increases in 3'

inositol phosphates upon the inner leaflet of the cell membrane recruît Protein kinase B

(PKB) to the membrane where it becomes activated [130]. Activated PKB is associated

with cell cycle progression and cellular growth.

The p85 subunit ofPI3K was identified as a GST -SH3 do main fusion protein

capable ofco-precipitating FasL [46]. Phosphorylation ofPKB on Serine 473, a key

residue for enzyme activation, has been documented downstream of cells treated with a

soluble Fas Irnrnunoglobin fusion protein [122]. Recruitment ofp85 to FasL has

therefore been hypothesized to participate in FasL reverse signaling induced activation of

the PI3K1PKB pathway.

Sre Farnily Tyrosine Kinases

Mulitple tyrosine kinases ofthe Src family have been documented to interact with

FasL [46,131,132]. These kinases aIl contain SH3 domains, and GST-SH3 do main

fusion proteins derived from the Fyn, Lyn and Fgr SH3 do mains are capable of co­

precipîtating FasL in vitro [46,131]. Ofthese proteins, Fgr has been demonstrated to

phosphorylate tyrosine residues within the FasL intracellular tail [131]. To date, it is

unclear ifFasL tyrosine phosphorylation is necessary for FasL reverse signaling.

Tyrosine mutated FasL is found less abundantly within the multivesicular bodies of

secretory lysosomes in cell types that contain this organelle, suggesting that tyrosine

phosphorylation of the FasL ICD may modulate the sorting ofFasL [131].

PA CSIN proteins

The Protein kinase c And Casein kinase Substrate in N eurons (P AC SIN) proteins,

also known as syndapins, are a family ofthree SH3 do main containing proteins, which

bind to FasL [45,133]. PACSIN 1 is expressed primarily in neural tissues, while

PACSINs 2 & 3 are expressed more ubiquitously [134]. PACSIN proteins each contain a

coiled coil (CC) oligomerization domain, as well as several NPF motifs, which bind to

Thornhill, pg. 25

Epsin Homology (EH) domain containing proteins, involved in the regulation of

endocytosis [134]. P ACSlN proteins bind to multiple cellular proteins in addition to

FasL. PACSINs bind to the GTPase dynamin, which controls the final fission step of

vesicle formation [l35]. PACSINs also bind to the Wiskott Aldrich Syndrome Protein

(W ASP), a potent activator of actin polymerization [l36]. Based on these two protein

interactions, P ACSIN proteins have been hypothesized to couple vesicle formation with

the actin cytoskeleton, perhaps in such a way that the cytoskeleton aids in fission or

'pinching off' of vesicles from the cell surface or organellar membranes [l34]. Co­

transfection of P ACSIN proteins with FasL results in relocation of FasL from the cell

membrane to intracellular vesicular structures in non hematopoietic cells, whereas

PACSIN proteins colocalize with lysosomal FasL in hematopoietic celIs [l33].

Interestingly, PACSIN3 protein can also bind to members of the ADAM family of

metalloproteinases and upregulate extracellular domain shedding of the heparin-binding

epidermal growth factor-like growth factor [l37]. As ADAM family members are

implicated in FasL extracellular domain cleavage, it is interesting to speculate that

PACSIN3 may regulate this process by linking members of the ADAM family to the

FasL intracellular tail.

PSTPIP

The Proline, Serine and Threonine Phosphatase-Interacting Protein (PSTPIP) was

initially identified as a FasL interacting protein in a yeast two hybrid screen [l38].

PSTPIP contains an SH3 domain, which interacts with the FasL proline rich region, and

a coiled coil domain, which allows it to interact with the Protein Tyrosine Phosphatase­

PEST (PTP-PEST). PSTPIP recruits PTP-PEST to the FasL intracellular domain [l38],

and may influence the phosphorylation state of the FasL intracellular domain.

Overexpression of PSTPIP results in decreased cell surface FasL and intracellular

retention in a proline rich domain dependent manner [l38]. PSTPIP may therefore

participate in both the trafficking and reverse signaling of FasL.

Nck

Thornhill, pg. 26

Nck is an adaptor protein containing three SH3 domains and one SH2 domain. In T

ceIls, Nck binds to both the WASP protein [139] and the T Cell Receptor (TCR) complex

[140], and recruits WASP to the TCR to facilitate local changes in actin polymerization

in response to TCR engagement by peptide/major histocompatibility complexes (MHC)

on antigen presenting cells (APCs) [141]. WASP facilitated actin polymerization at sites

of TCR-MHC contact plays a critical role in formation of an immunological synapse,

allowing the directed secretion of cytokines to sites of cell-cell contact [142]. Nck binds

to the proline rich region of the FasL ICD and partially colocalizes with FasL and

lysosomal markers at sites of cell-cell contact in cytotoxic T lymphocytes (CTLs) [143].

Treatment of cells with a short interfering RNA (siRNA) targeting Nck reduces

localization of FasL to sites of cell-cell contact [143]. Nck may therefore act as a

regulator of FasL secretion from secretory lysosomes to sites of cell-cell contact, and

thereby participate in CTL target cell killing.

Thornhill, pg. 27

OBJECTIVES:

Fas Ligand has been extensively studied as an inducer of apoptosis, particularly

in the immune system. Both Fas and FasL are also expressed in the cells of the nervous

system, inc1uding glial cells, and are upregulated after in jury [69]. Recent data document

prominent non-apoptotic roles for nervous system Fas in the regulation of neurite

outgrowth and neurite branching [21,22]. Mice expressing hypofunctional Fas and FasL

exhibit delayed peripheral nerve regeneration [21], suggesting that FasL, acting either

through Fas forward or FasL reverse signaling, may play a key role in the injured

nervous system. Schwann cells, glial cells of the peripheral nervous system, are essential

for peripheral nerve regeneration, contribute to the extensive tissue remodeling that

occurs during healing, and also express FasL after peripheral nerve injury [69].

The primary objective of this thesis is to investigate the spectrum of FasL

interacting proteins within nervous system Schwann cells. By regulating FasL

localization, as weIl as potentially participating in FasL reverse signaling, FasL

interacting proteins may participate in or modulate peripheral nerve regeneration. The

secondary objective of this thesis is to elucidate to the greatest extent possible, the

consequences of novel FasL-protein interactions in order to shed insight into the

complicated cellular biology of FasL.

Thornhill, pg. 28

r-. 1

Chapter 2

A Proteomic Screen Reveals Novel Fas Ligand Interacting Proteins within

Nervous System Schwann cells

Thornhill, pg. 29

A Proteomic Screen Reveals Novel Fas Ligand Interacting Pro teins within N ervous System Schwann cells

Peter B. Thornhill1, Gillian Druryl, Jason B. Cohn2,3, William L. Stanford5, Alan

Bernstein2,3,4, and Julie Desbarats1,*

IDepartment of Physiology, McGill University, 3655 Promenade Sir William OsIer, Montréal, Québec, H3G 1 Y6 2Samuel Lunenfeld Research Institute, Mount Sinai Hospital, 600 University Avenue, Toronto, ON M5G 1X5 3Department of Molecular and Medical Genetics, University of Toronto, Toronto, ON M5S 1A8 4Canadian Institutes of Health Research, 410 Laurier Ave. W., Ottawa, ON K1A OW9 5Institute for Biomaterials and Biomedical Engineering, University of Toronto, 164 College Street, Toronto, ON M5S 3G9

Thornhill, pg. 30

r-I

ABSTRACT:

Fas Ligand (FasL) binds Fas (CD95) to induce apoptosis or activate other

signaling pathways. In addition, FasL transduces bidirectionai or 'reverse signaIs.' The

intracellular domain of FasL contains consensus sequences for phosphorylation and an

extended proline rich region, which regulate its surface expression. Here, we used a

proteomics approach to identify novel FasL interacting proteins in Schwann cells to

investigate signaling through and trafficking of this protein in the nervous system. We

identified two novel FasL interacting proteins, sorting nexin 18 and adaptin ~, as well as

two proteins previously identified as FasL interacting proteins in T cells, PACSIN2 and

PACSIN3. These proteins are all associated with endocytosis and trafficking,

highlighting the tight regulation of cell surface expression of FasL in the nervous system.

Keywords: Fas Ligand, protein interactions, Src homology 3 domain, Schwann cell

Thornhill, pg. 31

INTRODUCTION:

Fas Ligand (FasL, CD95L, CD 178), a member of the tumor necrosis factor (TNF)

superfamily, is a type II, homotrimeric protein with cell surface and secreted isoforms.

FasL is inducibly expressed by cells of the immune system upon activation and in the

nervous system after injury, and is constitutively expressed in "immune privileged"

tissues, such as the eye and reproductive systems [70,71,144]. Traditionally, FasL has

been viewed exclusively as a ligand for the Fas receptor, stimulating formation of a death

inducing signaling complex (DISC) and activation of a caspase cascade, ultimately

culminating in apoptosis [10,145]. The apoptotic effector function of FasL is particularly

weIl documented in the immune system, where FasL is an effector of activation induced

lymphocyte cell death (AICD), and oftarget cell killing by cytotoxic T lymphocytes [47-

50,146]. Consistent with this, mice and humans with mutations in the FaslFasL system

exhibit massive lymphoproliferation and are susceptible to developing autoimmunity

[147,148]. In addition to the effector function of FasL as a ligand for Fas, increasing

evidence has suggested that FasL and other membrane-bound TNF superfamily ligands

can also transduce "reverse signaIs" and thus participate in bidirectional signaling. This

phenomenon is now documented for several members of the TNF superfamily, which

transduce many critical and diverse signaIs in normal immune function, such as cellular

proliferation, cytokine and growth factor secretion, and immunoglobin class switching

[118,126].

The FasL intracellular domain (ICD) contains a modular architecture consistent

with its role as a signal transducing molecule. It contains a 26 amino acid proline rich

region which interacts with multiple Src homology 3 (SH3) and WW domain containing

proteins in T cells, including several members of the Src tyrosine kinase family [45,46].

In T cells, stimulation of FasL with agonistic antibodies or Fas fusion proteins increases

its association with select SR3 do main containing proteins and resuIts in activation of the

extracellular-signal regulated kinase (ERK) mitogen activated protein kinase (MAPK)

pathway [70,122]. In addition to its potential role in signal transduction, the FasL proline

rich region has also been implicated in sorting of the mole cule to different cellular

compartments [113], which may be important for regulating both FasL effector and

reverse signaling activities. The FasL ICD also contains a conserved, cytosolic casein

Thornhill, pg. 32

t

• 1

kinase 1 (CKI) motif (SXXS) that has been implicated in reverse signaling in other TNF

family members [118,124]. Recently Sun et al. have documented changes in FasL serine

phosphorylation fonowing treatment of cultured cens with a Fas fusion protein [122].

We and others have previously shown that FasL can influence the proliferation of

T cens in the immune system via reverse signaling [70,119,120,122]. However, reverse

signaling in TNF superfamily members has never been studied in the nervous system.

FasL expression is induced on nervous system glial cens, inc1uding Schwann cens,

following in jury [69]. Fas is also expressed in the nervous system and is upregulated

after in jury [69], suggesting that Fas/FasL interactions, inc1uding FasL reverse signaling,

may play a role in the injured nervous system. We have found that expression of Fas and

FasL influenced peripheral nerve regeneration after crush injury [21] (and unpublished

data).

Here we used a proteomic approach to identify novel FasL interacting proteins in

Schwann cells, the glial cells of the peripheral nervous system, to gain insight into

mechanisms ofreverse signaling and trafficking of FasL in the nervous system. Using

liquid chromatography-tandem mass spectrometry (LC-MSMS) for identification of

FasL co-immunoprecipitating proteins, we identified Sorting Nexin 18 (SNXI8), adaptin

13, Protein Kinase C and Casein Kinase Substrate in Neurons 2 (PACSIN2), and

P ACSIN3 as FasL interacting proteins. The binding of FasL to multiple proteins

implicated in trafficking suggests tight regulation of the reverse signaling and effector

functions of FasL within the nervous system.

Thornhill, pg. 33

MATERIALS AND METHODS:

Cells:

Mouse Schwann cells were isolated from sciatic nerves of neonatal (P 1 to P3)

C3HIHeJ mice. Mouse nerves were digested in collagenase (Sigma-Aldrich, Oakville,

ON) then in trypsin / DNase (Worthington Biochemicals, Lakewood, NJ). The resulting

cell suspensions were grown ovemight in DMEM supplemented with 10% fetal calf

serum (FCS, Hyclone, Mississauga, ON), 100U/ml Penicillin and 100""glml

Streptomycin (Gibco, Burlington, ON), and O.lmM non-essential amino acids (Gibco,

Burlington, ON), then with 10 ""M cytosine-B-D-arabinofuranoside (Ara-C, Sigma­

Aldrich, Oakville, ON) for 48 hours. Confluent cultures of Schwann cells were treated

with anti-mouse Thy1.2 (CD90.2) antibody containing supematant derived from the HO-

13-4 hybridoma (obtained from the American Type Culture Collection) and rabbit

LowTox™ complement (Cedarlane, Burlington, ON) to remove residual fibroblasts. To

verify the purity of the cultures, cells were examined using flow cytometry as follows:

Cells were trypsinized from cell culture plates, then pelleted, and resuspended in PBS +

2% fetal calf serum (FCS). Cells were then either permeabilized and stained using the

Cytofix/Cytoperm™ fixation/permeabilization solution kit (BD Biosciences,

Mississauga, ON) according to the manufacturer' s recommendations or stained in PBS +

2% FCS for surface markers. Stainings were performed on ice for 25 minutes with the

indicated primary antibodies or isotype control antibodies. After two wash cycles in PBS

+ 2% FCS, or the provided PermlWash buffer, all samples were then stained with

phycoerythrin conjugated anti mou se secondary antibodies (Groovy Blue Genes Biotech

Ltd., Vineland, ON). Cells were washed twice more in PBS + 2% FCS or Perm/Wash

buffer and analyzed using a FACSAria™ cell analyzer (BD Biosciences, Mississauga,

ON).

Construction of FasL deletion mutants:

FasL deletion mutants, HuFasL~2-33-FLAG and HuFasL~2-70-FLAG, were

created using wildtype, human FasL cDNA (Open Biosystems, Huntsville, AL) as a

template to produce PCR products missing base pairs corresponding to N-terminal amino

Thomhill, pg. 34

1

• 1

acids 2-33 and 2-70, respectively, but still containing the initiator methionine (AUG)

base pairs. The huFasL and huFasL62-33 coding sequences were PCR amplified using

forward primers 5' CACCTGCAGCCATGCAGC-3' and 5'­

GGGCTGCAGCCATGACCTCTGTGCCC AGAAGGC-3' respectively, paired with the

reverse 5'-CTCTTAAAGCTTATATAAGC CG-3' primer and PFUltra™ DNA

polymerase (Stratagene, La Jolla, CA). The reactions were carried out for 20 cycles in

lxPCR buffer, 0.8""M each primer, 250 ""M each dideoxynucleotide triphosphate

(dNTP), and 2.5rnM MgCh, usingO.5 ng/""l template DNA as starting material. The

resulting PCR products, as well as the vector pCMV-Tag4A, (Stratagene, La Jolla, CA)

were then digested with Pst! (lnvitrogen, Burlington, ON) and HindIII (New England

Biolabs, Ipswich, MA) according to the manufacturer's recommendations to generate

complementary overhanging base pairs for cloning. The two digestion products were

then ligated together using T4 DNA Ligase (Invitrogen, Budington, ON), using a 10 fold

molar ratio of insert to vector, and transformed into chemically competent DH5a

Escherichia coli. Ligation into pCMV -Tag4A facilitated the creation of C-terminal

FLAG tagged proteins, due to the presence of this epitope in the vector, before the

translational stop site. HuFasL-FLAG and huFasL62-33-FLAG were then excised with

Pst! and Acc651 (Fermentas, Budington, ON), and the overhanging base pairs were

"filled in" using the Klenow fragment (Invitrogen, Burlington, ON) of DNA polymerase

1 according to the manufacturer' s recommendations. The resulting products were then

ligated into the pIRES2-EGFP (Clontech, Mountainview, CA) Smal (New England

Biolabs, Ipswich, MA) site using a 5 fold molar excess of insert to vector.

HuFasL62-70-FLAG was created by PCR amplification with primers 5'­

GGGCTGCAGCCATGCTGAAGAAG AGAGGGAACCACAGC-3' (forward) and 5'­

CTACTTATCGTCGTCATCCTTG-3' (reverse) using huFasLFLAG as a template and

the reaction conditions outlined above. As the PCR process does not leave terminal

phosphates necessary for ligation, the resulting PCR product was phosphorylated with T4

Polynucleotide Kinase (Gibco, Burlington, ON) according to the manufacturer's

recommendations before being ligated into the pIRES2-EGFP Smal site. AlI vectors

were full y sequenced before transfecting into cells to ensure the absence of

mutations/frameshifts created during PCR amplification.

Thornhill, pg. 35

Creation of Stably transfected celllines:

Mouse Schwann cells were transfected with huFasL-FLAG, huFasLLU-33-FLAG,

huFasLf12-70-FLAG, or empty pIRES2-EGFP vector as a control using Lipofectamine

2000™ (lnvitrogen, Burlington, ON) according to the manufacturer's recommendations.

Fort Y eight hours after transfection, cells were trypsinized and GFP positive cells were

single cell sorted into 96 well plates using a FACSAria™ fluorescence activated cell

sorter (BD Biosciences, Mississauga, ON). Cells were expanded in DMEM with 10%

FCS, O.lmM non essential amino acids (Gibco, Burlington, ON), lmM sodium pyruvate

(Gibco, Burlington, ON) and 400 [lg/ml G4l8 (Wisent Bioproducts, St. Bruno, QC).

Clones were then screened for expression of FasL constructs and GFP by

immunoblotting cellular lysat~s with FasL and GFP antibodies.

RNA isolation and RT -PCR:

RNA extraction was performed on 2 x 105 cells using the ChargeSwitch total

RNA kit (Invitrogen, Burlington, ON) according to the manufacturers instructions. RNA

was quantified using the Quant-iT Ribogreen RNA assay kit (Invitrogen, Burlington,

ON) according to the manufacturers protocol. Prior to reverse transcription, 40j.1g total

RNA was denatured at 65°C for 5 min., then placed on ice. Reverse transcription (RT)

reactions used a final volume of 20j.1L and contained 5J.1M random decamers (Ambion,

Austin, TX), 0.5mM each dNTP, lOU RNA guard (GE Healthcare, Piscataway, NJ), IX

Sensicript RT buffer (Qiagen, Mississauga, ON) and 4U Sensicript reverse transcriptase

(Qiagen, Mississauga, ON). RT reactions were incubated 60 min., at 37°C. Primers for

PCR amplification were as follows: l8S rRNA sense:

5'TGAGAAACGGCT ACCACATCC3', l8S rRNA antisense:

5'TCGCTCTGGTCCGTCTTGC3' , FasL sense:

5 'CTGGTTGCCTTGGTAGGATTGGG3 , ,FasL antisense:

5' GGACCTTGAGTTGGACTTGCCTG3 , , eGFP sense:

5' GCCACAAGTTCAGCGTGCTCG3 , , eGFP antisense:

5'TCTTCTGCTTGTCGGCCATGATAT3' (Invitrogen, Burlington, ON). PCR

reactions were performed in a 25j.1L volume and contained 2j.1L cDNA, O.4j.1M each

Thornhill, pg. 36

~. 1

primer, 0.2mM each dNTP (GE healthcare, Piscatatway, NJ), IX Hotstartaq PCR buffer

(Qiagen, Mississauga, ON), and lU Hotstartaq (Qiagen, Mississauga, ON). Cyc1ing was

performed by Techne Touchgene Gradient (Burlington, NJ) thermocyc1er as follows:

18S rRNA: 15 min. at 95°C, cyc1ed 13X 60 s. at 94°C, 40 s. at 66.2°C, 30s. at 72°C, then

final elongation 10 min. at 72°C. eGFP and FasL: 15 min. at 95°C, cyc1ed 25X 60 s. at

94°C, 20 s. at 65.2°C, 40s. at 72°C, then final elongation 10 min. at 72°C. PCR products

were visualized on 1.2% agarose gel (Invitrogen, Burlington, ON).

Immunoprecipitation (IP) and Immunoblotting:

For immunoprecipitation, cells were washed once in phosphate-buffered saline

(PBS) and lysed in buffer containing 0.5% Nonidet P-40 (Bioshop Canada, Burlington,

ON), 25mM HEPES pH 7.4,115 mM potassium acetate, 5mM magne sium chloride,

1mM sodium orthovanadate (All from Sigma-Aldrich, Oak:ville, ON), and Complete

Mini™ protease inhibitor tablet (Roche Diagnostics, Laval, QC). Lysates remained on

ice for 15 minutes before centrifugation at 15,000 rpm for 10 minutes at 4°C.

Supernatants were then prec1eared against unconjugated Sepharose 4B (Sigma-Aldrich,

Oak:ville, ON) for 1.5 hours at 4°C. Sepharose beads were then pelleted, and prec1eared

supernatants were incubated for at least 2 hours at 4°C with anti-FLAG or isotype control

antibodies (Sigma-Aldrich, Oakville, ON) covalently cross-linked to Protein G­

Sepharose (GE Healthcare, Piscataway, NJ) using dimethyl pimelimidate (Sigma­

Aldrich, Oak:ville, ON). The beads were then pelleted, washed thrice in cold lysis buffer,

boiled in Laemli buffer containing j3-mercaptoethanol, and electrophoresed on SDS

polyacrylamide gels. SDS gels were then either transferred to polyvinylidene difluoride

(PVDF) membranes (BioRad Canada, Mississauga, ON) and subjected to

immunoblotting or stained with colloidal Coomassie blue (Invitrogen, Burlington, ON)

followed by tandem mass spectrometry of unique gel bands. Gels transferred to PVDF

membranes were blocked overnight at 4°C in 5% non-fat dry milk powder (w/v) in Tris­

buffered saline/O. 1 % Tween 20 (TBST, Fisher Scientific, Ottawa, ON) and blotted with

the indicated primary and horseradish peroxidase (HRP) conjugated secondary antibodies

diluted in blocking buffer and TBST respectively. Bands were visualized using ECL

Thornhill, pg. 37

plus reagents (GE Healthcare, Piscataway, NJ) and film from Molecular Technologies

Inc. (Phillipsburg, NJ).

Mass Spectrometry of Unique Gel Bands:

Coomassie stained gel bands, detected in FLAG IP's but not in isotype control IP's, were

excised and minced manually with a scalpel, washed with water and dehydrated in

acetonitrile (Sigma-Aldrich, Oakville, ON). The corresponding gel area in the isotype

controllane was also excised for mass spectrometric analysis to separate legitimate

interacting proteins from those highly prevalent at the same molecular weight. Gel

pieces were then spun down and dried using a vacuum centrifuge, fo llowed by reduction

in 10 mM dithiotreitol/0.1 M ammonium bicarbonate (Sigma-Aldrich, Oakville, ON) for

30 minutes at 56°C. Proteins were then alkylated in 55 mM iodoacetamide (Sigma­

Aldrich, Oakville, ON)/O.l M ammonium bicarbonate for 20 min at room temperature in

the dark, followed by extensive washing in 1: 1 (v/v) O.lM ammonium

bicarbonate/acetonitrile to remove residual Coomassie stain. After dehydration in 100%

acetonitrile and vacuum drying, as above, gel particles were rehydrated in 50 mM

ammonium bicarbonate, 5 mM calcium chloride (Sigma-Aldrich, Oakville, ON) and

12.5 ng/J.lI oftrypsin (Promega Corp., Madison, WI) and digested overnight at 37°C.

Peptides remaining in the gel were then extracted twice in 1: 1 (v/v) 25 mM ammonium

bicarbonate/acetonitrile and 1: 1 (v/v) 5% formic acidlacetonitrile, respectively.

Extracted samples were then pooled and dried using a vacuum centrifuge for mass

spectrometryat the Institute for Research in Immunology and Cancer (IRlC) proteomic

center at the University of Montréal. The dried peptides were resuspended in 5 ulofO.2

% formic acid and injected on a NanoACQUITY HPLC system (Waters Limitée,

Lachine QC). Peptides were fIfst trapped by an home-made reversed phase C18

precolumn (300 J.lm internaI diameter x 5 mm length) and subsequently separated on an

home-made C18 analytical column (150 J.lm internaI diameter x 10 cm, particle size 3

J.lm). NanoLC separation was achieved using a 56-min linear gradient from 5% to 60 %

aqueous acetonitri1e (0.2 % FA) at a flow rate of600 nl/min. Mass spectrometrie

acquisitions in the MS/MS mode were performed on a Q-TOF Premier instrument

(Waters Limitée, Lachine QC). Database searches were performed against the NCBI

Thornhill, pg. 38

non-redundant protein database using Mascot software (http://www.matrixscience.com).

We used the following criteria to define potentially valid protein hits (Table 1): protein

hits had to be supported by at least 3 peptides and have a predicted molecular weight

within ~ 10 kDa of the excised gel band. Proteins with peptide hits present in the isotype

background band were excluded from further analysis, as weIl as hits for antibody

fragments, trypsin and/or other highly prevalent cellular proteins known to non­

specifically precipitate in IP experiments.

Antibodies

For immunoprecipitation, monoclonal anti-FLAG M2, or mouse IgG 1 control (Sigma­

Aldrich, Oakville, ON) were used. Anti-FasL G247.4 or anti-adaptin 13 (both BD

Biosciences, Mississauga, ON) were used for the detection of immunoprecipitated

proteins in western blots. EGFP was detected using clone JL-8 (Clontech, Mountain

View, CA). For the detection of SNX18, anti-SNX18 antisera were generated by

immunizing rabbits with a KLH-conjugated GFKKEYQKVGQSFR peptide (500~g) in

Freund's Complete Adjuvant and boosting two times with peptide (100~g) in Freund's

Incomplete Adjuvant. For flow cytometric detection, anti-SlOO (Chemicon, Temecula,

CA), anti-GFAP, or anti-Thyl.2-PE (both BD Biosciences, Mississauga, ON) were used

and compared to isotype control antibodies mouse IgG2a (BD Biosciences, Mississauga,

ON), mou se IgG2b (Sigma-Aldrich, Oakville, ON), and rat IgG2a-PE (BD Biosciences,

Mississauga, ON), respectively. Non-Fluor chrome linked primary antibodies were

detected with phycoerythrin conjugated anti mouse secondary antibodies (Groovy Blue

Genes Biotech Ltd., Vineland, ON).

Thornhill, pg. 39

RESULTS:

Stable expression of FasL constructs in Schwann celIlines

To investigate the proteins associated with FasL within the nervous system, we

first created stable Schwann cell transfectants expressing wild type or truncated FasL

constructs. Schwann celIs isolated from mouse sciatic nerve which later transformed in

celI culture were used for the establishment of stable transfectants. These celIs expressed

the Schwann celI markers S lOO and glial fibrillary acidic protein (GFAP), but not the

fibroblast marker Thy1.2 (Fig. lA). Wild type FasL-FLAG (wtFasL), FasL-FLAG

lacking amino acids 2-33 (FasL62-33), which inc1ude the casein kinase consensus

sequence, and FasL-FLAG lacking amino acids 2-70 (FasL62-70), which encode the

proline rich domain (Fig. lB), were stably expressed in Schwann celIs in conjunction

with the enhanced green fluorescent protein (EGFP), using a bicistronic approach. The

bicistronic vector, pIRES2, contains an internaI ribosomal entry site (IRES) upstream of

the EGFP gene, resulting in co-translational expression of FasL and EGFP on the same

mRNA. As expected, the ratio of FasL to EGFP mRNA was similar for all three FasL

constructs (Fig. 1 C). However, we noticed dramatic differences in the ratios of FasL to

EGFP protein expression between the different FasL constructs (Fig. ID). While wtFasL

was expressed at very low levels compared to EGFP, the ratio of FasL62-33 to EGFP

was roughly 10 fold higher. The expression ratio of FasL62-70 to EGFP was also

increased 6 fold relative to wt FasL. Our results suggest that truncations in the coding

region of FasL do not alter the transcriptional efficiency of FasL with respect to EGFP,

but may increase the translational efficiency or stability of the protein in Schwann cells.

Previous reports have documented increased translation and decreased cytotoxicity of

FasL constructs after deletion of the intracellular region of FasL [95,96]. It is therefore

possible that the dramatically greater expression of the FasL deletion mutants compared

to the wtFasL protein may be attributed to increased translational efficiency and/or

decreased cytotoxicity.

Thornhill, pg. 40

f····,~······································:

S '1 '\

.~. J.t .... \~, .............................. :

., ;S";":"'i':;' .; .... ". '(":;; '( ':" :·X.'. n~v;:;·::,: .,:::.,.,::::',: ',:. ,.:.:' .. ,:,".,: .... : ·.:··,:·, .. :c .. ::;',t,·::.,: ..... :.,.,,: "':" ".:'.;',.,.::", x·'::,,·',: :c"" .l""".""., :c"" .:c··,' ".",. :c"" .: .. ';:-".','.' :,."'.' ,:~f:.:'y :,:::, ".:>.'.' ,·:', .. ·:··,·,T t4

c :. ;.

-.. ,.,.,.) ......... - ........ ,_. ·-·v·································-.,.·· .:.;,. ••• '0:. ···.:<':·!:;.:-:.::·i:··::·,;.;.: •.• :.;~~::·:·;···.,.~· ;:~':'.::.~. :.:"':' ;::::.. ':.:" .!;:-:-:--.:-:.;.::

:0 .. ;~ "

l' i ~".,:, ;,::: . .l..

f····· .. ··· ... ········ .. ························ .. ···· ........ · ............................................................... , ................. , ................ ..

lzoCI.1&! ~ 'ai'~ : lë~'050!

1 ~.~ " 1 l .:; a. O.2B , ! ;J;< l l li. ~ j

!. .......... ~:~~ .. \i?~~t,i·· ............................................ .

Fig. 1: FasL deletion mutants are expressed more efficiently than wtFasL

A: CharacterÏzations of Schwann cells used in our experiments were performed, as described in the materials and methods, via flow cytometry using antibodies against glial fibrillary acidic protein (GFAP), S100, and the fibroblast marker Thy1.2 B: The 80 amino acid FasL intraceIlular domain is shown in biochemical single letter code. Sites of truncation for FasL constructs (b.33, b. 70), the proline rich region, and casein kinase phosphorylation consensus are indicated. C: Stably transfected Schwann cell clones expressing bicistronic wildtype (Wt) or truncated FasL-lRES-EGFP constructs were analysed for expression ofFasL constructs. FasL and EGFP mRNA, as weIl as the 18S rRNA, were amplified by RT-PCR to examine the relative expression ofFasL and EGFP message. Densitometric values were normalized to 18S rRNA to control for loading differences. Graphs show the ratios ofFasL to EGFP mRNA for each construct. D: Lysates from Schwann cells expressing Wt and truncated FasL were subjected to immunoblotting for FasL and EGFP. Graphs show the densitometric ratios ofFasL to EGFP protein expression for each construct. Blots and ratios are representative of multiple Schwann ceIl clones expressing each construct.

Thornhill, pg. 41

,~

Screen for FasL interacting proteins in Schwann cells

We used Schwann cells expressing FasLâ2-33 for our screen due to the poor

expression ofwtFasL (Fig. ID), and because previous screens in T cells failed to detect

proteins able to interact with GST -FasL2-29 [45]. Lysates from Schwann cells

expressing FasLâ2-33 were immunoprecipitated under gentle conditions to identify FasL

interacting proteins. When IPs were run on gel and stained with colloidal Coomassie, six

bands at approximately 37,52,60, 71, 74 and 110 kDa were uniquely present or greatly

increased in anti-FLAG IPs versus control IgG IPs (Fig. 2). Using LC-MSMS, we

identified these proteins as FasL, Protein Kinase C and Casein Kinase Substrate in

Neurons 2 (PACSIN2; Syndapin2), PACSIN3/Syndapin3, Sorting Nexin Associated

Golgi protein 1 (SNAG1; Sorting Nexin 18; SNX18 - present in 2 bands at 71 and 74

kDa), and adaptin 13, respectively (Table 1). Of the proteins identified, the SH3 domain­

containing adaptor proteins P ACSIN2 and 3 have been previously reported to interact

with the proline rich region ofFasL in T cells, validating our screen and suggesting that

these two interactions are conserved in multiple cell types [45].

Fig. 2: FasL specifically co­immunoprecipitates six bands

FasL was immunoprecipitated from Schwann cells expressing FasLâ2-33, run on 4-20% gradient SDS-PAGE, and stained with colloidal Coomassie Blue. Six bands at 37, 52, 60, 71, 74 and 110 kDa uniquely present or greatly increased in anti-FLAG IPs versus control IgG IPs are indicated with arrows. These bands and the corresponding gel regions from the control IPs were excised for identification using LC-MSMS.

Thornhill, pg. 42

(" Table 1: LC-MSMS Proteins "Hits" Co-precipitated with Human FasLA33 in Murine

Schwann CeUs Band Mol. Predicted # of Sequence UniProt Accession

Protein Name(s) Wt. Mol. Wt. Peptides Coverage #

Fas Ligand 35kDa 37kDa 18 20% gil2168140 Protein Kinase C and Casein Kinase Substrate in Neurons 3(PACSIN3);Syndapin 3 52kDa 48kDa 3 4% gi111127644

Protein Kinase C and Casein Kinase Substrate in Neurons 2 (PACSIN2); Syndapin 2 60kDa 56kDa 13 30% gi119483913

Sorting Nexin Associated Golgi Protein 1; Sorting Nexin 18 71kDa 68kDa 9 18% gtlfJ?~_:l~~}l

Sorting Nexin Associated Golgi Protein 1; Sorting N exin 18 74kDa 68kDa 3 7% gi126342577

Beta Adaptin Subunit 110kDA 98kDa 4 5% gmJ~Q:H?:?_7.

Validation ofFasL interaction with adaptin 13

Encouraged by the presence ofpreviously reported FasL interacting proteins in

our screen (Table 1), we proceeded to validate novel interactions by immunoblotting

FasL IPs prepared from Schwann cells stably expressing FasLb,2-33 and FasLb,2-70.

Antibodies to adaptin 13 detected a band at approximately 100kDa co-precipitating with

both FasL deletion mutants (Fig_ 3A), indicating that the proline rich region ofFasL was

not necessary for this interaction. Unsurprisingly, adaptin 13 contains no proline binding

motifs, although it contains an appendage do main that interacts directly with c1athrin in

vitro [149]. 13 adaptins make up one ofthe two large subunits present in each of the

heterotetrameric adaptor protein (AP) complexes that bind 'cargo' proteins and facilitate

the formation of clathrin coated vesicles [150]. Each ofthese heterotetrameric AP

complexes possesses a unique subunit composition: 131 is present with 'Y adaptin in the

AP-l complex, and 132 with a adaptin in the AP-2 complex_ We sought to distinguish

which AP complex bound to FasL using antibodies to these unique subunits. Although

appropriate bands were observed in celllysates with antibodies to 'Y and a adaptins, no

corresponding bands were detected in control or FasL IPs (Fig. 3B).

Thornhill, pg. 43

A Immunoprecipitation:

Lysate A2-33 A2-70

B

Lysate

kDa: A2 -33 ~2-70 250 15 100

25 15

IgG FLAG IgG FLAG

rmmunoprecipitation: 62-33 d2-70

IgG FLAG IgG FLAG

adaptin-y

Validation ofFasL interaction with SNX18

Fig. 3: Adaptin B co-precipitates withFasL A: Lysates from Schwann cells expressing FasLb.2-33 or FasLb.2-70 were immunoprecipitated using control or FLAG antibodies and subjected to immunoblotting for FasL and adaptin p. Co­precipitation of adaptin p with both FasLb.2-33 and FasLb.2-70 indicates that the FasL proline rich region is not necessary for this interaction. B: Lysates from Schwann cells expressing FasLb.2-33 or FasLb.2-70 were immunoprecipitated as above and blotted for adaptin a or adaptin y. Bands present in the lysate lanes indicate proper recognition of the respective proteins in the immunoblot, although co­precipitation with FasL was not detected.

Sorting Nexins are a family of30 proteins involved in endocytosis and protein

trafficking. This protein family is characterized by the presence of a Phox (PX) domain,

which binds various phosphatidylinositol phosphates, targeting the proteins to cellular

membranes. SNX18 contains an amino terminal SR3 domain, a PX domain, and a

carboxyl terminal coiled co il (CC) domain. As no antisera to SNX18 were available to

validate the interaction with FasL, we immunized rabbits with SNX18-derived peptides

conjugated to Keyhole Limpet Remocyanin. The resulting antisera reproducibly

recognized a protein band of approximately 50 kDa and multiple bands of 68-74 kDa in

Schwann cells (Fig. 4) and mouse tissues (data not shown). In agreement with our mass

spectrometrie data (Table 1), an antisera-reactive band of approximately 74kDa co­

precipitated with FasLb.2-33, but not with FasLb.2-70. An additional band, not detected

in our screen, at approximately 50 kDa also co-precipitated with FasL. The inability of

the FasLb.2-70 construct to co-precipitate significant amounts of either band suggests

Thornhill, pg. 44

that the binding of SNX18 to FasL requires an intact proline rich region, and may

therefore be mediated by the SH3 domain ofSNX18.

kDa:

Lysate: .

~2-33 D.2 --70

37

Immunoprecipitation: .:\2-33 1.\2-70 Immuno

FLAG IgG FLAG blot : .. SNX18

....

FasL

Figure 4

Fig. 4: SNX18 co-precipitates with FasL

Lysates from Schwann cells expressing FasLf12-33 or FasLf12-70 were immunoprecipitated using control or FLAG antibodies and subjected to immunoblotting with antibodies to FasL and antiserum to SNX18. Bands at 74 and 50 kDa (arrows) co­precipitated with FasLf12-33, but not with FasLf12-70, indicating that the proline rich region ofFasL is likely necessary for this interaction.

Thornhill, pg_ 45

~. 1

DISCUSSION:

In this report, we present the first investigation of FasL interacting proteins in

cells of the nervous system. Our screen identified four FasL interacting proteins, two

previously reported in T cells, PACSIN 2 and 3, and two novel FasL interactors, SNX18

and adaptin J3. SNX18 represents a novel FasL interacting protein in Schwann cells. Our

antisera reacted with multiple bands present at 68-74 kDa, of which only the band at

74kDa co-precipitated with FasL (Fig. 4). Interestingly, SNX18 is phosphorylated on

serine and tyrosine residues downstream of multiple receptor tyrosine kinases (JBC,

unpublished data), and these bands may correspond to differentially phosphorylated

SNXI8. In support ofthis hypothesis, phosphorylation of SNX9, the most closely

related member of the sorting nexin family, has previously been shown to alter the

cellular target of its SH3 domain in vitro [151].

Although there are to date no published data on the cellular functions of SNXI8,

insights into the potential consequences of its association with FasL can be gleaned by

comparison with its closest homologue, SNX9. SNX9 contains a domain architecture

identical to SNX18 and is 35% homologous at the amino acid level. SNX9 was

originally identified as an interactor with two members of the A Disintegrin And

Metalloprotease (ADAM) family, ADAMs 9 and 15 [152]. Interestingly, members of

this family have recently been identified as the prote as es responsible for the shedding of

the FasL extracellular domain from the cell surface to generate soluble FasL (sFasL)

[103]. SNX18 associated with FasL may therefore recruit ADAMs to FasL, promoting

FasL shedding from the membrane. Furthermore, SNX9, as well as SNXI8, associate

with the Ras activating protein SOS [153], potentially linking FasL to the ERK signaling

pathway and providing a mechanism for FasL reverse signaling via ERK. As in T cells,

we also observe ERK activation in Schwann cells upon stimulation of FasL with

agonistic antibodies or Fas fusion proteins (PBT, unpublished data), suggesting that this

pathway is conserved across cell types. Finally, SNX9 also associates with the J32 adaptin subunit of the AP-2 complex [154], stimulates the budding step of endocytosis

via activation of the GTPase activity of dynamin [155], and participates in the activation

dependent degradation of the epidermal growth factor receptor [156]. Therefore, it is

Thornhill, pg. 46

interesting to hypothesize that SNX18 may participate in the downregulation of cell

surface FasL either by facilitating its internalization, or by promoting extracellular

domain shedding of the protein.

Our screen also identified J3 adaptin as a novel FasL interacting protein. We

detected adaptin J3 associated with both FasL~2-33 and FasL~2-70 in Schwann cells

(Fig. 3). FasL contains no YXX<t> or di leucine sorting signaIs known to bind to the

adaptor protein complexes. Our antibodies recognize both J31 and J32 adaptin which are

82% homologous at the amino acid level, and we were un able to co-precipitate the AP-l

or AP-2 specific subunits, 'Y or a adaptin, with FasL. Our data suggest that the interaction

between FasL and adaptin B is mediated via the 10 amino acids in the cytoplasmic tail of

FasL that are still present in FasL~2-70. These remaining cytoplasmic residues of FasL

contain the sequence KKRn-74, which is ubiquitinated [131]. Ubiquitination has been

associated with the internalization of many cell surface proteins [157], and may therefore

indirectly recruit B2-adaptin for the internalization of FasL. In support of this, FasL

constructs with these lysine residues mutated are expressed more readily at the cell

surface compared to wild type FasL [113]. The association of FasL with adaptin J3 may

be an important regulator of the surface expression of the protein and hence control both

FasL forward (via Fas) and reverse signaling.

Our screen further identified the PACSIN proteins 2 and 3 as FasL interacting

proteins in Schwann ceIls, confirming previous findings in T cells and suggesting that

these interactions are conserved in multiple cell types. The three known PACSIN

proteins aIl contain an SH3 domain and a coiled coil domain, and are highly implicated

in endocytosis. By binding the Wiskott Aldrich Syndrome Protein (WASP), a potent

activator of actin nuc1eation, and the OTPase dynamin, they are thought to link

endocytosis to the actin cytoskeleton [134]. Based on colocalization data, the se proteins

may aIso participate in the lysosomaIlocalization of FasL [133]. Similar to the Sorting

Nexins, PACSIN proteins also bind to members of the ADAM metaIloprotease family

and regulate ectodomain shedding of the pro-heparin binding EOF-like growth factor

[137].

Several previous studies have used OST pulldowns to examine FasL protein

interactions in T cells [45,46]. Most of the FasL interacting proteins identified appeared

Thornhill, pg. 47

to attenuate the cell surface expression of FasL and/or alter its subcellular localization

[131,133,138,143,158], rather than participate in FasL reverse signaling. This is

consistent with reports documenting increased expression of FasL at the cell surface

upon deletion ofits proline rich domain [113,131]. Clearly, the presence of multiple

FasL interacting proteins in T cells, as well as in Schwann cells, with roles in regulating

its localization highlight the need for tight regulation of both the effector and reverse

signaling activities of FasL. This tight regulation of its surface expression makes

intuitive sense, as cell surface expression of FasL would need to be tightly regulated to

ensure appropriate regulation of apoptotic signaIs. FasL reverse signaling, by contrast, is

often cited as weak or modest. Our screen may have missed potential reverse signaling

effectors due to the deletion of FasL amino acids 2-33; however, a previous report failed

to identify biding partners for GST-FasL2-29 [45].

In summary, our screen for FasL interacting proteins within Schwann cells has

identified two novel proteins, and confirmed two known FasL interactors, ail with

potential functions in regulating trafficking and subcellular localization. The recurring

theme of these interacting proteins within cells of both the nervous and immune systems

highlights the need for tight regulation of cell surface expression of FasL, and

identification of these novel FasL interactors provides insight into potential molecular

mechanisms underlying this stringent regulation.

Thornhill, pg. 48

TRANSITION TO CHAPTER 3:

In addition to FasL interacting proteins identified using mass spectrometry in

Schwann cells (see Chapter 2), candidate SR3 and other proline binding domain

containing proteins were also tested in parallel for co-immunoprecipitation with FasL,

when antibodies were available. Ofthe many proteins tested, only one convincingly

interacted with FasL. Chapter 3 contains experiments performed to establish the adaptor

protein Grb2 as a FasL interacting protein in Schwann cells and determine the

significance ofthis interaction.

The significance ofthe Grb2-FasL interaction was researched more extensively in

comparison to the other interacting proteins identified by LC-MSMS in Chapter 2 simply

because it was the first interacting protein identified by the author and used as a positive

control during the troubleshooting phase of the LC-MSMS experiments. Although we

did not subject gel bands at a molecular weight that would contain Grb2 to LC-MSMS in

the experiments in Chapter 2 (due to potential interference from antibody light chains in

detection by mass spectrometry), we did notice an increase in the appropriate band at 25

kilo Daltons (see Figure 2 in Chapter 2).

Thornhill, pg. 49

Chapter 3

The Adaptor Protein Grb2 Interacts with Fas Ligand and Regulates its Cell Surface

Expression in Schwann cells

Thornhill, pg. 50

The Adaptor Protein Grb2 Interacts with Fas Ligand and Regulates Us Cell Surface Expression in Schwann cells

Peter B. Thomhillt, Jason B. Cohn2,3, Yoon Kow Youngt, William L. Stanford5, Alan Bemstein2,3.4, and Julie Desbarats1,*.

IDepartment of Physiology, McGill University, 3655 Promenade Sir William OsIer, Montréal, Québec, H3G 1 Y6 2Samuel Lunenfeld Research Institute, Mount Sinai Hospital, 600 University A venue, Toronto, ON M5G 1X5 3Department of Molecular and Medical Genetics, University of Toronto, Toronto, ON M5S 1A8 4Canadian Institutes of Health Research, 410 Laurier Ave. W., Ottawa, ON K1A OW9 5Institute for Biomaterials and Biomedical Engineering, University of Toronto, 164 College Street, Toronto, ON M5S 3G9

Thomhill, pg. 51

ABSTRACT:

Fas Ligand (FasL) binds Fas (CD95) to induce apoptosis or activate other

signaling pathways. The intracellular domain of FasL contains an extended proline rich

region which modulates the surface expression of the molecule via interacting with

protein complexes. Rere, we identify the SR3 domain containing adaptor protein, Grb2,

as a regulator of FasL surface expression in Schwann cells. Mutational analyses suggest

that Grb2 binds to the proline rich region of FasL. Knockdown of Grb2 results in altered

FasL distribution, with FasL decreased on the cell surface and increased intracellularly.

Our data show that Grb2 controls the subcellular localization of FasL, either by retaining

it at, or sorting it to the cell surface.

Key Words: Fas Ligand, Src homology 3 domain, Schwann cell, Grb2, sorting nexin 18,

adaptin ~

Thornhill, pg. 52

INTRODUCTION:

Fas Ligand (FasL, CD95L, CD178) is a homotrimeric protein with membrane­

bound and soluble isoforms, and belongs to the tumor necrosis factor (TNF) superfamily.

FasL is expressed by activated immune cells, and in "immune privileged" tissues such as

the eye and the nervous system [72] [70,71,144]. The extracellular domain of FasL binds

to the Fas receptor (CD95) to induce apoptosis [10,145]. In addition to the effector

function of FasL as an inducer of apoptosis, FasL and other membrane-bound TNF

superfamily ligands can also transduce "reverse signaIs" and thus participate in

bidirectional signaling [118,126].

As FasL is a transmembrane protein able to induce apoptosis, its cell surface

expression must be tightly regulated. Cell surface FasL is c1eaved via the action of

metalloproteinases [98,103], liberating a soluble FasL (sFasL) extracellular domain,

which has limited potency as an apoptosis inducer [99]. Transmembrane FasL can also

be released extracellularly via secretion of microvesic1es bearing FasL [112,117].

Multiple reports have documented significant intracellular retenti on of FasL [58,111-

113,159]. In cells with secretory lysosomes, su ch as cytotoxic T lymphocytes and mast

cells, FasL is predominantly intracellular and colocalizes with lysosomal markers, while

cells lacking these specialized organelles express cell surface FasL with variable levels of

intracellular retenti on [58,95,113]. The FasL intracellular domain (ICD) is likely

sufficient for its intracellular retention, as fusion of the FasL ICD to the cell surface

protein CD69 results in intracellular retention of the fusion protein [58]. The FasL

intracellular domain contains several putative consensus sequences for phosphorylation,

and a 26 amino acid proline rich region capable of interacting with proteins containing

Src homology 3 (SR3) and WW domains [45,46]. The proline rich region has been

implicated in regulating surface expression of FasL through via interacting with several

of these SR3 domain containing proteins [113,138,143].

We have recently found that FasL interacts with multiple trafficking-associated

proteins in Schwann cells [160]. Schwann cells are the glial cells of the peripheral

nervous system, and they upregulate FasL after peripheral nerve injury [69], suggesting

that Fas / FasL signaling may play a role in the injured nervous system. Schwann cells

are essential for peripheral nerve regeneration, and contribute to the extensive tissue

Thornhill, pg. 53

remodeling that occurs during healing. In this report, we follow up on our initial findings

by investigating mechanisms that could regulate cell surface expression of FasL on

Schwann cells. We identify the SH3 domain containing adaptor protein, Grb2, as a

regulator of FasL surface localization in the nervous system.

MATE RIALS AND METHODS:

Cells

Mouse Schwann cells were isolated from sciatic nerves of neonatal (Pl to P3)

C3H/HeJ mice, as described [160]. To verify the purity of the cultures, cells were

examined by flow cytometry for expression of the Schwann cell markers glial fibrillary

acidic protein (GFAP) and SlOO, and the fibroblast marker Thy-1.2, as described

previously [160].

Construction of FasL deletion mutants

FasL deletion mutants, HuFasL~2-33-FLAG and HuFasL~2-70-FLAG, were created

using wildtype FasL cDNA as a template to produce PCR products missing base pairs

corresponding to N-terminal amino acids 2-33 and 2-70, respectively, but still containing

the initiator methionine (AUG) base pairs. Vector construction methodology has been

previously described [160]. AIl vectors were fully sequenced before transfecting into

cells to ensure the absence of mutations/frameshifts created during PCR amplification.

Creation of stably transfected celllines

Mouse Schwann cells were transfected with huFasL-FLAG, huFasL~2-33-FLAG,

huFasL~2-70-FLAG, or empty pIRES2-EGFP vector. EGFP-positive cells were single

cell sorted into 96 weIl plates using a FACSAria™ fluorescence activated cell sorter (BD

Biosciences, Mississauga, ON), expanded in standard media supplemented with 400

!-tg/ml G418 (Wisent Bioproducts, St. Bruno, QC), and screened for expression of FasL

constructs and EGFP by immunoblotting. .

Thornhill, pg. 54

RNA interference (RNAi)

Cells were trypsinized and transfected while still in suspension with 5nM mouse Grb2

ON-TARGETplus SMARTpool or ON-TARGETplus siCONTROL Non-Targeting

siRNA (Dharmacon, Lafayette, CO) using Lipofectamine RNAimax (lnvitrogen,

Burlington, ON) according to the manufacturers' recommendations. cens were then re­

transfected 24 hours after attachment to maximize knockdown. In an experiments,

knockdown was confrrmed 48 hours after initial transfection by immunoblotting.

Immunoprecipitation (IP) and Immunoblotting

For IP, cens were washed once in phosphate-buffered saline (PBS) and lysed in buffer

containing 0.5% Nonidet P-40 (Bioshop Canada, Burlington, ON), 25mM HEPES pH

7.4, 115 mM potassium acetate, 5mM magne sium chloride, 1 mM sodium orthovanadate

(an from Sigma-Aldrich, Oakville, ON), and Complete Mini™ protease inhibitor

cocktail (Roche Diagnostics, Laval, QC). Lysates remained on ice for 15 minutes before

centrifugation at 15,000 rpm for 10 minutes at 4°C. Supematants were then precleared

against unconjugated Sepharose 4B (Sigma-Aldrich, Oakville, ON) for 1.5 hours at 4°C.

Sepharose beads were then pelleted, and prec1eared supematants were incubated for at

least 2 hours at 4°C with specific or isotype control antibodies plus Protein G-Sepharose

(GE Healthcare, Piscataway, NJ). The beads were then pelleted, washed thrice in cold

lysis buffer, boiled in Laemmli buffer containing ~-mercaptoethanol, and

electrophoresed on SDS polyacrylamide gels. SDS gels were then transferred to

polyvinylidene difluoride membranes (PVDF, BioRad Canada, Mississauga, ON).

PVDF membranes were blocked ovemight at 4°C in 5% non-fat dry milk powder (w/v)

in Tris-buffered saline/O. 1 % Tween 20 (Fisher Scientific, Ottawa, ON; TBST) and

blotted with the indicated primary and horseradish peroxidase (HRP) conjugated

secondary antibodies diluted in blocking buffer and TBST respectively. Bands were

visualized using ECL PLUSTM reagents (GE Healthcare, Piscataway, NJ) and film from

Molecular Technologies Inc. (Phillipsburg, NJ).

Immunofluorescence

Thornhill, pg. 55

Schwann cells expressing the indicated FasL constructs were grown overnight on glass

coverslips. Cells were then fixed for 30 minutes at room temperature with 4%

paraformaldehyde, and washed thrice with PBS/O.O 1 % Triton X -100 (Sigma-Aldrich,

Oakville, ON). Cells were blocked and permeabilized with PBS/2% bovine serum

albumin (Sigma-Aldrich, Oakville, ON)/O.l % Triton X-lOO for 1 hour, then stained with

20 [lg/ml protein G purified anti-Fas Ligand antibodies (NOK-l hybridoma, International

Patent Organism Depository) or control antibodies, followed by Cy3 conjugated

secondary antibodies (Jackson Immunoresearch, West Grove, PA). All slides were

counterstained with 2.5 [lg/mI4',6-diamidino-2-phenylindole (Sigma-Aldrich, Oakville,

ON; DAPI). Stained samples were mounted on slides using Aqua Polymount

(Polysciences Inc., Warrington, PA) and analyzed using a Leica DMIRE2 fluorescent

microscope (Richmond Hill, ON).

Flow Cytometry

Cells expressing the indicated FasL constructs were detached from cell culture plates by

vigorous pipetting, pelleted, and resuspended in PBS + 2% fetal calf serum (FCS). Cells

were then stained on ice for 25 minutes with affinity purified NOK-l anti-FasL

antibodies (clone obtained from International Patent Organism Depository) or control

antibodies. After two wash cycles in PBS + 2% FCS, all samples were then stained with

phycoerythrin conjugated secondary antibodies (Jackson ImmunoResearch, West Grove,

PA). Cells were washed twice more in PBS + 2% FCS and analyzed using a

FACSArrayTM bioanalyzer (BD Biosciences, Mississauga, ON).

Antibodies

For IP, flow cytometry, and immunofluorescence, monoclonal anti-FLAG M2, mouse

IgG 1 control, Rabbit IgG control (all Sigma-Aldrich, Oakville, ON), anti-Grb2 (Santa

Cruz Biotechnology, Santa Cruz, CA), or affinity purified NOK-l anti-FasL antibodies

were used. Anti-FasL G247.4 (BD Biosciences, Mississauga, ON), anti-adaptin P (BD

Biosciences, Mississauga, ON) or anti-Grb2 (Santa Cruz Biotechnology, Santa Cruz,

CA) antibodies were used for immunoblots. For the detection of sorting nexin 18, anti­

SNX18 antibodies were generated by immunizing rabbits, as previously described [160].

Thornhill, pg. 56

RESULTS:

Cell Lines

We have previously characterized the Schwann celIlines stably expressing wild type

and truncated FasL constructs that we use here. These Schwann celIlines are GFAP+,

S100+, Thy-l-, and constitutively express wildtype FasL-FLAG (wtFasL), FasL-FLAG

lacking the casein kinase consensus sequence but including the proline-rich domain

(FasLâ2-33), or FasL-FLAG lacking the proline rich domain (FasLâ2-70; Fig. 1). We

and others have previously found that intracellularly truncated forms ofFasL are

expressed much more efficiently than wtFasL due to increased translational efficiency

and/or decreased cytotoxicity oftruncated FasL constructs [95,96,160]. Due to the

extremely low expression ofwtFasL (Fig. 1), and because we have previously shown that

FasLf12-33 and wtFasL both interact with the same novel binding partners in Schwann

cells [160], we therefore chose to use FasLf12-33 to investigate potential mechanisms of

FasL trafficking in Schwann cells.

A 1~33 ,~70

N-NQQPFNYPYPQIYWVDSSASSPWAPPGTVLPC*SVPRRPGQRRPPPPPPP?PLPPPPPPPPLPPLPLP1r.KKRGNHSTGTM •.•

B

kDa!

37

~ '~--------/ CKl./GSK3iJ consensus l?roline Rich Region

l FasL

.... EGFP

0.75

-j .g ~ ~ 0.50

Il.. .2 ... 1/1

CI '" ~ ~ 0.25 .. " ... CI>

Fig. 1: FasL deletion mutants are expressed more efficiently than wtFasL A: The 80 amino acid FasL intraceUular domain is shown in biochemical single letter code. Sites oftruncation for FasL constructs (f133, f170), the proline rich region, and casein kinase phosphorylation consensus are indicated. B: Lysates of stably transfected Schwann ceU clones expressing bicistronic wild type (Wt) or truncated FasL-lRES­EGFP constructs were run on SDS-PAGE and immunoblotted to demonstrate relative expression ofFasL and EGFP. Graph at right shows the densitometric ratios ofFasL to EGFP protein expression for each construct. Blots and ratios are representative of multiple Schwann ceU clones expressing each construct.

Thornhill, pg. 57

Interaction of Grb2 with FasL

In order to identify FasL interacting proteins that may act as regulators ofFasL surface

expression, we examined co-immunoprecipitation of candidate SR3 domain-containing

proteins with FasLâ2-33 in Schwann cells. Grb2, an adaptor protein containing one SR2

and two SR3 domains, has previously been implicated in receptor downregulation

[127,161] and is a known FasL interacting protein in T cells [45]. When FasLâ2-33

immunoprecipitates were immunoblotted for Grb2, we detected efficient co-precipitation

with FasL (Fig. 2A). Grb2 co-precipitated with FasL immunoprecipitated with

antibodies to FasL (clone NOK-l), as well as to the FLAG epitope. The NOK-l

antibody was less efficient at precipitating FasLâ2-33 than the anti-FLAG antibody, and

also co-precipitated less Grb2, consistent with a specific interaction. Antibodies to Grb2

also co-precipitated FasL from FasLâ2-33 Schwann celllysates (Fig. 2B). We detected

FasL in Grb2 immunoprecipitates from cells expressing FasLâ2-33, but not FasLâ2-70,

suggesting that the interaction with Grb2 requires an intact proline rich domain.

A Immunoprecipitation:

I,ysate IgG FLAG NOK-l

37

B Immunoprecipitation:

Lysate: ~2-33 ~2-70

~2 -33 ~2 -70 IgG Gr:Q2 IgG Grb2

25

Immuno blot:

Grb2

FasL

Immuno blot:

FasL

Grb2

Fig. 2: Grb2 co­precipitates with amino acids 34-70 ofFasL A: Lysates from Schwann cells expressing FasLâ2-33 were immunoprecipitated using control, NOK-l anti-FasL, or FLAG antibodies and subjected to immunoblotting for FasL, Grb2 B: Lysates from Schwann cells expressing FasLâ2-33 or FasLâ2-70 were immunoprecipitated with control or anti-Grb2 antibodies and blotted for and immunoblotted as above. Co-precipitation

of FasLâ2-33 , but not FasLâ2-70, suggests that Grb2 interacts with the proline rich region ofFasL encoded in amino acids 34-70.

Thornhill, pg. 58

Grb2 positively regulates cell surface FasL

We next investigated the effect of Grb2 on FasL surface expression using RNAi to

Grb2. Grb2 has previously been shown to downregulate the expression of the epidermal

growth factor receptor (EGFR) [127,161]. Unexpectedly, flow cytometric staining of

Schwann cells revealed significant (p<O.01) attenuation of cell surface expression of

FasL62-33, but not FasL62-70, 48 hours after Grb2 knockdown (Fig. 3A, B).

Immunofluorescent staining of FasL on the Schwann celllines also demonstrated

reduced surface expression of FasL62-33 after Grb2 but not control non-targeting RNAi

treatment (Fig. 3C, D). In control cells, FasL staining was observed on the surface of

lamellipodial-like extensions, as weIl as extracellularly on FasL-bearing microvesicles.

After Grb2 knockdown, this staining was shifted to the submembrane space, with

punctuate FasL staining within each cell. Quantification by a blinded observer revealed a

significant increase (p<O.0001) of the number of FasL staining puncta per cell following

Grb2 knockdown in FasL62-33 but not in FasL62-70 Schwann cells, despite efficient

knockdown in both celllines (Fig. 3 C,D). Grb2 knockdown therefore decreases FasL

surface expression and increases its expression in an intracellular compartment.

Collectively, these data imply that Grb2 promotes the cell surface expression of FasL and

affects its subcellular localization in Schwann cells.

Thornhill, pg. 59

A

>=i o ......

FasL-PE:

62-33

62-70 .. -1 ':-,0:;­::s

..Q ·ri 1-1 .;..> <il

-r< o .JJ .~!

<:: r::>

B

c

25

50 37

',-:,. "1C{. 100D

FasL-PE

.!5:~.Y..:.

- FasL+Cont:.rol RNA.i.

FasL-t-Grb2 RNA.i.

Secondax·y alone + Control RNA.i.

Secondary alone + Grb2 RNAi

Mean Fluorescence Intensity

0.0

+

62-33

Grb2 siRNA.

10.0

: ::; control :.' $iRNA

+

62-70

;::.:.:' :'.: .' . ... .. :... :. ", : : :'.::. ..:.~. ~:' :;.:::: Grb2 . . . ..

Actin

E

Thornhill, pg. 60

Number of FasL staiOÎng puncta/cell 10.00 ~2-33

8.00

6.00

4.00

2.00

0.00 Grb2

62-70 n/s

Grb2 .:.~~~J

Fig. 3: Grb2 regulates the surface expression of FasL A: Schwann cells expressing FasL~2-33 or FasL~2-70 were treated with control non­targeting or Grb2 siRNA's, stained for surface FasL expression and analyzed using flow cytometry. Thick lines represent samples given control siRNA treatment, while thin lines represent groups treated with Grb2 RNAi. B: In a separate experiment, cells were treated and stained as above in triplicate wells and similarly analyzed using flow cytometry. Geometric Mean Fluorescence intensities (MFI) are plotted. Significant decreases in (MF!) corresponding to surface expression were observed in Schwann cells expressing FasL~2-33 (p=0.0092; two tailed t test), but not FasL~2-70 (p=0.4001). C: Lysates of Schwann cells expressing FasL~2-33 and FasL~2-70 were run on SDS­PAGE and immunoblotted for Grb2 and actin following control or Grb2 siRNA treatment to confirm knockdown. Typical results are shown. D: Schwann cells expressing the indicated FasL constructs were treated with control or Grb2 siRNA and analyzed using indirect immunofluorescence following staining for FasL. Red staining represents FasL, while blue represents DAPI nuc1ear staining. E: Quantification of intracellular puncta like staining of FasL before and after Grb2 RNAi treatment performed by a blinded observer. Schwann cells expressing FasL~2-33 displayed an increased number of puncta per cell (p<O.OOOI; two tailed t test) but not cells expressing FasL~2-70 (p=0.0778). Data are representative of multiple trials.

Grb2 interacts with novel FasL interacting proteins

As an adaptor protein with multiple protein binding domains, but no inherent catalytic

activity, Grb2likely mediates its effect on FasL surface expression via linking it to other

protein complexes. We have recently identified two novel FasL interacting proteins in

Schwann cells [160], Sorting Nexin 18 (SNXI8) and adaptin ~, both with recognized

roles in protein trafficking [150,162]. Both proteins contain the minimal PXXP motifs

necessary for SR3 domain binding (NCBI database accessions Q10567, Q96RFO). We

therefore tested the ability of SNX18 and adaptin ~ to interact with Grb2 in our Schwann

celllines, using co-IP. Surprisingly, both proteins were specifically co-precipitated by

antibodies to Grb2, but not by control antibodies (Fig. 4). To our knowledge, both of

these represent novel protein interactions, not previously reported in the literature. As

these two proteins both co-precipitated with FasL as well as with Grb2 in Schwann ceIls,

we next sought to determine whether their interactions with FasL were dependent on

Grb2. Knockdown of Grb2 preceding immunoprecipitation of FasL did not appreciably

attenuate the level of co-precipitated SNXI8, despite efficient knockdown of Grb2 (Fig.

5A, B). In contrast to SNXI8, co-precipitation of adaptin ~ with FasL was dramatically

Thornhill, pg. 61

reduced by Grb2 knockdown (Fig. 5A), suggesting that the interaction between FasL and

adaptin ~ is dependent on Grb2.

75

50

25

.I~·~:~.~:~~~·:::~:·.·;·. ~~:f .. ~:H:: ... ~.

x· .. *x~~.t:::;'.:,.:v. ~~:

SNX18

He

Grb2

Fig. 4: Grb2 interacts with the FasL interacting proteins Sorting Nexin 18 and adaptin ~ Lysates from Schwann cells expressing FasLâ2-33 were immunoprecipitated using control or anti-Grb2 antibodies, run on SDS-PAGE and subjected to multiple rounds of immunoblotting for Grb2, Sorting Nexin 18, and adaptin~. Co-precipitation ofFasL interacting proteins was observed in Grb2 but not control IPs. (HC-IgG heavy chain)

Thornhill, pg. 62

A

kDa:

75

100

37

B

Immunoprecipitation:

Control Knockdown:

IgG FLAG

Contro1

25.

20

Grb2 Knockdown:

IgG FLAG

Grb2 :RNAi

Immuno­Blot:

SNX18

FasL

Immunoblot:

Grb2

Fig. 5: Grb2 knockdown attenuates FasL-adaptin B co-precipitation A: Schwann cells expressing FasLâ2-33 were pretreated with control or Grb2 siRNAs, then lysed, and immunoprecipitated with control or FLAG antibodies. Samples were run on SDS-PAGE, followed by, repeated immunoblottings for FasL, Sorting Nexin 18 and adaptin B. B: Control and Grb2 RNAi treated Schwann celllysates from the above experiment were run in parallel on SDS-PAGE and blotted for Grb2 to confirm efficient knockdown.

Thornhill, pg. 63

DISCUSSION:

In this report, we document a role for the adaptor protein Grb2 in the trafficking

and surface expression of FasL. We also provide the first evidence that Grb2 binds to

two recently described FasL interacting proteins, SNX18 and adaptin~. FinalIy, we

show that Grb2 promotes the interaction between FasL and adaptin ~, suggesting a

potential mechanism for the regulation of FasL trafficking by Grb2.

Our data indicate that amino acids 34-70 of FasL, containing its proline rich

region, are necessary for the interaction between Grb2 and FasL. In Schwann cells

expressing FasL that contains the proline rich region, we observe a redistribution of FasL

from the cell surface to an intracellular compartment following attenuation of Grb2

expression. These data suggest that Grb2 interacts with FasL via its SH3 domain, and

promotes sorting or retention of FasL at the cell surface. FasL~2-70, which does not

contain the proline rich region and does not interact with Grb2, does not undergo

redistribution when Grb2 is knocked down. However, it must be noted that the

expression of FasL~2-70 is significantly lower than FasL~2-33 in all Schwann celIlines

examined, and the low expression levels of FasL~2-70 may compromise our ability to

detect it at the cell surface.

Grb2 has previously been documented to bind FasL in T cells [45,122], where it

has been hypothesized to mediate FasL reverse signaling via recruitment of SOS, to

activate Ras and the extracellular-signal regulated kinase (ERK) MAPK pathway

[70,122]. However, Grb2 has not previously been implicated in FasL trafficking.

Interestingly, Grb2 has been shown to mediate the internalization of the epidermal

growth factor receptor (EGFR) [127,161,163], whereas we show that Grb2 promotes

accumulation of FasL at the cell surface.

We then investigated the mechanism by which Grb2 altered the subcellular

localization of FasL. We have recently described two novel protein interactions with

FasL in Schwann celIs, SNX18 and adaptin ~. We have previously reported that SNX18

binds to amino acids 34-70 of FasL [160], and we show here that its interaction with

FasL does not depend on Grb2 (Fig. 5). Adaptin~, on the other hand, appears to require

Grb2 for its interaction with FasL (Fig. 5). Adaptin ~ contains a proline rich stalk region

Thornhill, pg. 64

(AAs 576-725; QI0567) forming a putative SH3-binding domain which may allow it to

interact with Grb2. Grb2 may therefore act as an adaptor protein for the interaction of

FasL and adaptin~. It is also possible that Grb2 may stabilize a trimeric complex of

FasL, adaptin ~, and Grb2, involving several SH3 domain interactions.

~ adaptins are an essential component of the adaptor protein (AP) complexes that

bind 'cargo' proteins and c1athrin, and facilitate the formation of c1athrin coated vesic1es

for protein sorting [164]. For ex ample , the AP-l complex facilitates budding from the

trans-Golgi network (TGN) and endosomes, while the AP-2 complex mediates budding

from the cell surface. The Grb2-dependent recruitment of adaptin ~ to FasL may

therefore underlie its ability to promote accumulation of FasL at the cell surface.

Severa! previous studies have used GST pulldown approaches to examine FasL

protein interactions in T cells [45,46]. Most of the FasL interacting proteins identified

appeared to attenuate the cell surface expression of FasL and/or alter its subcellular

localization [133,138,143]. Here we document a protein-protein interaction with FasL

that appears to increase its cell surface expression in Schwann cells. Clearly, the

presence of multiple FasL interacting proteins in T cells, as well as in Schwann cells,

with roles in regulating its localization highlight the need for tight regulation of the

effector activity of FasL. FasL is upregulated in Schwann cells after injury to the

nervous system [69], suggesting that FasL apoptotic or reverse signaling activities play a

role in regeneration in the peripheral nervous system. In fact, we have previously found

that peripheral nerve regeneration is delayed in mice with hypofunctional Fas or FasL

[21] (and unpublished data), suggesting that regulation of cell surface FasL on Schwann

cells may be critical during nerve regrowth.

In summary, we have identified Grb2 as a key regulator of FasL surface

expression in Schwann cells, and we have shown that Grb2 recruits or stabilizes the

association of adaptin ~ with FasL, providing a potential mechanism for control of FasL

trafficking by Grb2. Our findings demonstrate a new molecular mechanism for

regulating the intracellular localization of FasL, a molecule implicated in apoptosis and

peripheral nerve regeneration.

Thornhill, pg. 65

Chapter 4

Final Conclusions

Thornhill, pg. 66

CONCLUSION:

In this thesis, we used a proteomic approach to identify novel FasL interacting

proteins in Schwann cells, the glial cells of the peripheral nervous system, to gain insight

into mechanisms ofreverse signaling and trafficking ofFasL in the nervous system.

Using liquid chromatography-tandem mass spectrometry (LC-MSMS) for identification

ofFasL co-immunoprecipitating proteins (Chapter 2), we identified Sorting Nexin 18

(SNX18), adaptin~, Protein Kinase C and Casein Kinase Substrate in Neurons 2

(PACSIN2), and PACSIN3 as FasL interacting proteins. Using a parallel, candidate

approach we also identified Grb2 (Chapter 3), as a FasL binding protein. The binding of

FasL to multiple proteins implicated in trafficking suggests tight regulation ofthe reverse

signaling and effector functions ofFasL within the nervous system. As mice with

hypofunctional Fas or FasL exhibit delayed peripheral nerve regeneration, and Schwann

cells are essential for proper regeneration, factors which regulate Schwann cell surface

FasL, where it is available to bind Fas, may play important roles in nervous repair.

Although in Chapter 2 we identified several novel FasL interacting proteins,

which were verified by immunoblotting, significant work remains to elucidate the

function ofthese novel interactors in Schwann cells. Figure 1 (below) summarizes

potential functions these proteins as FasL interactors based on the literature.

How FasL interacting groteins may regulate FasL:

4. Internalization

2. Coated Pit Formation

Fig. 1 Potential Function ofFasL interacting proteins identified in this thesis

Thornhill, pg. 67

r-... Via interacting with members of the metalloproteinase family, as well as FasL,

Sorting Nexin 18 and P ACSIN3 may regulate the extracellular do main shedding (1) of

FasL [137,152]. Interestingly, it has been documented that siRNA targeting PACSIN3

decreases extracellular do main shedding of the heparin-binding epidermal growth factor­

like growth factor [137]. Coiled coil domains present in these proteins, which allow

them to oligomerize [162], could allow them to function as adaptors in this context, with

each monomer using its single SR3 do main to recruit the proline rich FasL and

metalloproteinase proteins to the same complex.

The association of FasL with adaptin ~ may facilitate the entry of FasL into

c1athrin coated pits (2). ~ adaptins make up one ofthe two large subunits present in each

ofthe heterotetrameric adaptor protein (AP) complexes that bind 'cargo' proteins and

facilitate the formation of c1athrin coated vesic1es [150]. SNX9, the most c10sely related

homologue ofSNXI8, associates with the ~2 adaptin subunit of the AP-2 complex [154]

and stimulates the budding or fission step of endocytosis (3) via activation ofthe GTPase

activityofdynamin [155]. PACSIN proteins also bind to dynamin [134] and it is

therefore interesting to hypothesize that these three proteins may act together to

participate in the downregulation of cell surface FasL by facilitating its intemalization

(4).

Altematively, FasL may bind to ~1 adaptin, a subunit ofthe AP-l complex, and

may act to facilitate budding from AP-1 target membranes, such as the Golgi apparatus

and endosomes. Our attempts to identify which AP complex bound to FasL were not

successful (Chapter 2, Fig. 3B). SNX9 can also bind to ~1 adaptin of the AP-1 complex

[165], and the association ofFasL with SNX18 may similarly facilitate or stabilize an

interaction between FasL and ~1 adaptin. These speculations assume a conservation of

binding partners between Sorting Nexins 9 and 18, an assumption which has not been

tested experimentally to date.

In Chapter 3, in addition to identifying the adaptor protein Grb2 as a FasL

interacting protein, we were also able to demonstrate that this protein acts as a positive

regulator ofFasL surface expression in Schwann cells. Mutational analyses suggested

that Grb2 binds to the proline rich region ofFasL (Chapter 3, Fig. 2) and knockdown of

Grb2 resulted in altered FasL distribution, with FasL decreased on the cell surface and

Thornhill, pg. 68

increased intracellularly (Chapter 3, Fig. 3). Our data show that Grb2 controls the

subcellular localization ofFasL, either by retaining it at, or sorting it to the cell surface.

Interestingly, we found that Grb2 interacted with two of the above identified FasL

interacting proteins, SNX18 and adaptin B (Chapter 3, Fig. 4). Ofthese two proteins

which interact with both FasL and Grb2, only adaptin B required the presence of Grb2 to

co-precipitate with FasL (Chapter 3, Fig. 5). This result suggests that Grb2 either

directly mediates or stabilizes the interaction between FasL and adaptin B. A preliminary

model summarizing the results from Chapter 3 is presented below as Figure 2.

Summary: G rb2 is expressed: Grb2 is knocked down:

FasL62-33

Surface Expression FasLâ2-33 is retained intracellularly

FasLb.2-33

Fig. 2 Preliminary model summarizing the results in Chapter 3

In the above model, we propose that the presence of Grb2 facilitates the

interaction between adaptin Band FasL by linking together the se two proline rich

proteins via its two SR3 domains. Although we were unable to experimentally define

which isoform of adaptin B bound to FasL, we propose that Grb2 mediated recruitment of

Thornhill, pg. 69

adaptin ~l may positively regulate budding from the Golgi apparatus and endosomes of

FasL containing vesicles destined for the cell surface (see left half of figure). In the

absence of Grb2, this complex does not form and FasL surface expression is greatly

attenuated (see right half of figure).

In summary, the detailed analyses of nervous system FasL interacting proteins

presented in this thesis highlight the multiple levels of regulation of this protein and

provide insight into the potential mechanisms of this regulation. Factors which regulate

the surface expression of this death factor are of obvious interest, as cell death is

dysregulated in many disease states. A more thorough understanding of the cell

biological mechanisms of protein trafficking may also provide insight into diseases

where proteins are not properly trafficked, such as cystic fibrosis.

Thornhill, pg. 70

~ 1

REFERENCES:

[1] Yonehara S, Ishii A and Yonehara M (1989) A cell-killing monoclonal antibody (anti-Fas) to a cell surface antigen co-downregulated with the receptor oftumor necrosis factor. JExp Med. 169,1747-1756.

[2] Trauth BC, Klas C, Peters AM, Matzku S, Moller P, Falk W, Debatin KM and Krammer PH (1989) Monoclonal antibody-mediated tumor regression by induction of apoptosis. Science 245, 301-305.

[3] Itoh N, Yonehara S, Ishii A, Yonehara M, Mizushima S, Sameshima M, Hase A, Seto y and Nagata S (1991) The polypeptide encoded by the cDNA for human cell surface antigen Fas can mediate apoptosis. Cell66, 233-243.

[4] Smith CA, Farrah T and Goodwin RG (1994) The TNP receptor superfamily of cellular and viral proteins: activation, costimulation, and death. Cell 76, 959-962.

[5] Chan FK, Chun HJ, Zheng L, Siegel RM, Bui KL and Lenardo MJ (2000) A domain in TNP receptors that mediates ligand-independent receptor assembly and signaling. Science 288, 2351-2354.

[6] PapoffG, Hausler P, Eramo A, Pagano MG, Di Leve G, Signore A and Ruberti G (1999) Identification and characterization ofa ligand-independent oligomerization domain in the extracellular region ofthe CD95 death receptor. J Biol Chem. 274, 38241-38250.

[7] Orlinick JR, Vaishnaw A, Elkon KB and Chao MV (1997) Requirement of cysteine-rich repeats ofthe Fas receptor for binding by the Fas ligand. J Biol Chem. 272, 28889-28894.

[8] Itoh N and Nagata S (1993) A novel protein do main required for apoptosis. Mutational analysis ofhuman Fas antigen. J Biol Chem. 268, 10932-10937.

[9] Huang B, Eberstadt M, Olejniczak ET, Meadows RP and Fesik SW (1996) NMR structure and mutagenesis ofthe Fas (APO-l/CD95) death domain. Nature 384, 638-641.

[10] Lambert C, Landau AM and Desbarats J (2003) Fas-beyond death: a regenerative role for Fas in the nervous system. Apoptosis 8,551-562.

[11] Peter ME, Budd RC, Desbarats J, Hedrick SM, Hueber AO, Newell MK, Owen LB, Pope RM, Tschopp J, Wajant H, Wallach D, Wiltrout RH, Zornig M and Lynch DH (2007) The CD95 Receptor: Apoptosis Revisited. Cell129, 447-450.

[12] Wajant H, Pfizenrnaier K and Scheurich P (2003) Non-apoptotic Fas signaling. Cytokine Growth Factor Rev. 14, 53-66.

[13] Boatright KM and Salvesen GS (2003) Mechanisms ofcaspase activation. Curr Opin Cell Biol. 15, 725-731.

[14] Wei MC, Zong WX, Cheng EH, Lindsten T, Panoutsakopoulou V, Ross AJ, Roth KA, MacGregor GR, Thompson CB and Korsmeyer SJ (2001) Proapoptotic BAX and BAK: a requisite gateway to mitochondrial dysfunction and death. Science 292, 727-730.

[15] Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, Alnernri ES and Wang X (1997) Cytochrome c and dATP-dependent formation of Apaf­l/caspase-9 complex initiates an apoptotic protease cascade. Cell91, 479-489.

[16] Widlak P (2000) The DFF40/CAD endonuclease and its role in apoptosis. Acta Biochim Pol. 47, 1037-1044.

Thornhill, pg. 71

.r" [17] Freiberg RA, Spencer DM, Choate KA, Duh HJ, Schreiber SL, Crabtree GR and

Khavari PA (1997) Fas signal transduction triggers either proliferation or apoptosis in human fibroblasts. J Invest Dermatol. 108,215-219.

[18] Aggarwal BB, Singh S, LaPushin Rand Totpal K (1995) Fas antigen signaIs proliferation of normal human diploid fibroblast and its mechanism is different from tumor necrosis factor receptor. FEBS Lett. 364, 5-8.

[19] Desbarats J and Newell MK (2000) Fas engagement accelerates liver regeneration after partial hepatectomy. Nat Med. 6, 920-923.

[20] Badorff C, Ruetten H, Mueller S, Stahmer M, Gehring D, Jung F, Ihling C, Zeiher AM and Dimmeler S (2002) Fas receptor signaling inhibits glycogen synthase kinase 3 beta and induces cardiac hypertrophy following pressure overload. J Clin Invest. 109,373-381.

[21] Desbarats J, Birge RB, Mimouni-Rongy M, Weinstein DE, Palerme JS and Newell MK (2003) Fas engagement induces neurite growth through ERK activation and p35 upregulation. Nat Cell Biol. 5, 118-125.

[22] Zuliani C, Kleber S, Klussmann S, Wenger T, Kenzelmann M, Schreglmann N, Martinez A, deI Rio JA, Soriano E, Vodrazka P, Kuner R, Groene HJ, Herr l, Krammer PH and Martin-Villalba A (2006) Control of neuronal branching by the death receptor CD95 (Fas/ Apo-1). Cell Death Differ. 13, 31-40.

[23] Landau AM, Luk KC, Jones ML, Siegrist-Johnstone R, Young YK, Kouassi E, Rymar VV, Dagher A, Sadikot AF and Desbarats J (2005) Defective Fas expression exacerbates neurotoxicity in a model of Parkinson's disease. J Exp Med. 202, 575-581.

[24] Miwa K, Asano M, Rorai R, Iwakura Y, Nagata Sand Suda T (1998) Caspase 1-independent IL-1 beta release and inflammation induced by the apoptosis inducer Fas ligand. Nat Med. 4, 1287-1292.

[25] Owen-Schaub LB, Meterissian S and Ford RJ (1993) Fas/APO-1 expression and function on malignant cells of hematologic and nonhematologic origin. J Immunother Emphasis Tumor Immunol. 14,234-241.

[26] Alderson MR, Armitage RJ, Maraskovsky E, Tough TW, Roux E, Schooley K, Ramsdell F and Lynch DR (1993) Fas transduces activation signaIs in normal human T lymphocytes. J Exp Med. 178, 2231-2235.

[27] Aggarwal BB (2003) Signalling pathways of the TNF superfamily: a double-edged sword. Nat Rev Immunol. 3, 745-756.

[28] Kennedy NJ, Kataoka T, Tschopp J and Budd RC (1999) Caspase activation is required for T cell proliferation. 190, 1891-1896.

[29] Sehra S and Dent AL (2006) Caspase function and the immune system. Crit Rev Immunol. 26, 133-148.

[30] Ahn JH, Park SM, Cho HS, Lee MS, Yoon JB, Vi1cek J and Lee TH (2001) Non-apoptotic signaling pathways activated by soluble Fas ligand in serum-starved human fibroblasts. Mitogen-activated protein kinases and NF-kappaB-dependent gene expression. J Biol Chem. 276,47100-47106.

[31] Chen F, Castranova V and Shi X (2001)'New insights into the role of nuclear factor-kappaB in cell growth regulation. Am J Pathol. 159,387-397.

[32] Kataoka T, Budd RC, Holler N, Thome M, Martinon F, Irmler M, Burns K, (' Hahne M, Kennedy N, Kovacsovics M and Tschopp J (2000) The caspase-8

Thornhill, pg. 72

r---.. inhibitor FLIP promotes activation of NF-kappaB and Erk signaling pathways. CUIT Biol. 10, 640-648.

[33] Irmler M, Thome M, Hahne M, Schneider P, Hofmann K, Steiner Y, Bodmer JL, Schroter M, Burns K, Mattmann C, Rimoldi D, French LE and Tschopp J (1997) Inhibition of death receptor signaIs by cellular FLIP. Nature 388, 190-195.

[34] Yeh WC, Itie A, Elia AJ, Ng M, Shu HB, Wakeham A, Mirtsos C, Suzuki N, Bonnard M, Goeddel DY and Mak TW (2000) Requirement for Casper (c-FLIP) in regulation of death receptor-induced apoptosis and embryonic development. Immunity 12, 633-642.

[35] Sato T, Irie S, Kitada Sand Reed JC (1995) FAP-1: a protein tyrosine phosphatase that associates with Fas. Science 268, 411-415.

[36] Wieckowski E, Atarashi Y, Stanson J, Sato TA and Whiteside TL (2007) FAP-1-mediated activation of NF-kappaB induces resistance of head and neck cancer to Fas-induced apoptosis. J Cell Biochem. 100, 16-28.

[37] Suzuki A, Hayashida M, Kawano H, Sugimoto K, Nakano T and Shiraki K (2000) Hepatocyte growth factor promotes cell survival from fas-mediated cell death in hepatocellular carcinoma cells via Akt activation and Fas-death-inducing signaling complex suppression. Hepatology. 32, 796-802.

[38] Haeffner A, Deas 0, Mollereau B, Estaquier J, Mignon A, Haeffner-Cavaillon N, Charpentier B, Senik A and Hirsch F (1999) Growth hormone prevents human monocytic cells from Fas-mediated apoptosis by up-regulating Bc1-2 expression. Eur J Immunol. 29, 334-344.

[39] Rouvier E, Luciani MF and Golstein P (1993) Fas involvement in Ca(2+)­independent T cell-mediated cytotoxicity. J Exp Med. 177, 195-200.

[40] Suda T, Takahashi T, Golstein P and Nagata S (1993) Molecular c10ning and expression of the Fas ligand, a novel member of the tumor necrosis factor family. Ce1l75,1169-1178.

[41] Peitsch MC and Tschopp J (1995) Comparative molecular modelling of the Fas­ligand and other members of the TNF family. Mol Immunol. 32,761-772.

[42] Orlinick JR, Elkon KB and Chao MY (1997) Separate domains of the human fas ligand dictate self-association and receptor binding. J. Biol. Chem 272,32221-32229.

[43] Ashkenazi A and Dixit YM (1999) Apoptosis control by death and decoy receptors. CUIT Opin Cell Biol. Il,255-260.

[44] Pitti RM, Marsters SA, Lawrence DA, Roy M, Kischkel FC, Dowd P, Huang A, Donahue CJ, Sherwood SW, Baldwin DT, Godowski PJ, Wood WI, Gurney AL, Hillan KJ, Cohen RL, Goddard AD, Botstein D and Ashkenazi A (1998) Genomic amplification of a decoy receptor for Fas ligand in lung and colon cancer. Nature 396, 699-703.

[45] Ghadimi MP, Sanzenbacher R, Thiede B, Wenzel J, Jing Q, Plomann M, Borkhardt A, Kabelitz D and Janssen 0 (2002) Identification of interaction partners of the cytosolic polyproline region of CD95 ligand (CD 178). FEBS Lett. 519,50-58.

[46] Wenzel J, Sanzenbacher R, Ghadimi M, Lewitzky M, Zhou Q, Kaplan DR, Kabelitz D, Feller SM and Janssen 0 (2001) Multiple interactions of the cytosolic polyproline region of the CD95 ligand: hints for the reverse signal transduction capacity of a death factor. FEBS Lett. 509,255-262.

Thornhill, pg. 73

~ 1

[47] Krammer PH (2000) CD95's deadly mission in the immune system. Nature 407, 789-795.

[48] Alderson MR, Tough TW, Davis-Smith T, Braddy S, Falk B, Schooley KA, Goodwin RG, Smith CA, Ramsdell F and Lynch DH (1995) Fas ligand mediates activation-induced cell death in human T lymphocytes. J Exp Med. 181,71-77.

[49] Brunner T, Mogil RJ, LaFace D, Yoo NJ, Mahboubi A, Echeverri F, Martin SJ, Force WR, Lynch DH and Ware CF (1995) Cell-autonomous Fas (CD95)/Fas­ligand interaction mediates activation-induced apoptosis in T-cell hybridomas. Nature 373, 441-444.

[50] Ju ST, Panka DJ, Cui H, Ettinger R, el-Khatib M, Sherr DH, Stanger BZ and Marshak-Rothstein A (1995) Fas(CD95)/FasL interactions required for programmed cell death after T-cell activation. Nature. 373,444-448.

[51] Van Parijs L, R.Y., Lord JD, Nelson BH, Abbas AK, Baltimore D, (1999) Uncoupling IL-2 signaIs that regulate T cell proliferation, survival, and Fas­mediated activation-induced cell death. Immunity Il, 281-288.

[52] Kurasawa K, Hashimoto Y and Iwamoto 1 (1999) Fas modulates both positive and negative selection ofthymocytes. Cell Immunol. 194, 127-135.

[53] Singer GG and Abbas AK (1994) The fas antigen is involved in peripheral but not thymie deletion of T lymphocytes in T cell receptor transgenic mice. Immunity 1, 365-37l.

[54] Kishimoto H, Surh CD and Sprent J (1998) A role for Fas in negative selection of thymocytes in vivo. J. Exp. Med. 187, 1427-1438.

[55] Eischen CM, Schilling JD, Lynch DH, Krammer PH and Leibson PJ (1996) Fc receptor-induced expression of Fas ligand on activated NK cells facilitates cell­mediated cytotoxicity and subsequent autocrine NK cell apoptosis. J Immunol. 156,2693-2699.

[56] Kashii Y, Giorda R, Herberman RB, Whiteside TL and Vujanovic NL (1999) Constitutive expression and role of the TNF family ligands in apoptotic killing of tumor cells by human NK cells. J Immunol. 163,5358-5366.

[57] Trambas CM and Griffiths GM (2003) Delivering the kiss of death. Nat Immunol. 4, 399-403.

[58] Bossi Gand Griffiths GM (1999) Degranulation plays an essential part in regulating cell surface expression of Fas ligand in T cells and natural killer cells. Nat Med. 5, 90-96.

[59] Watanabe-Fukunaga R, Brannan CI, Copeland NG, Jenkins NA and Nagata S (1992) Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates apoptosis. Nature 356, 314-317.

[60] Takahashi T, Tanaka M, Brannan CI, Jenkins NA, Copeland NG, Suda T and Nagata S (1994) Generalized lymphoproliferative disease in mice, caused by a point mutation in the Fas ligand. Cell 76, 969-976.

[61] Hahne M, Peitsch MC, Irmler M, Schroter M, Lowin B, Rousseau M, Bron C, Renno T, French Land Tschopp J (1995) Characterization of the non-functional Fas ligand of gld mice. Int Immunol. 7, 1381-1386.

[62] Karray S, Kress C, Cuvellier S, Hue-Beauvais C, Damotte D, Babinet C and Levi-Strauss M (2004) Complete loss of Fas ligand gene causes massive lymphoproliferation and early death, indicating a residual activity of gld allele. J Immunol. 172,2118-2125.

Thornhill, pg. 74

(' [63] Matsuzawa A, Moriyama T, Kaneko T, Tanaka M, Kimura M, Ikeda H and

1 Katagiri T (1990) A new allele of the lpr locus, lprcg, that complements the gld gene in induction of lymphadenopathy in the mouse. J Exp Med. 171, 519-531.

[64] Eberstadt M, Huang B, Olejniczak ET and Fesik SW (1997) The lymphoproliferation mutation in Fas locally unfolds the Fas death domain. Nat Struct Biol. 4, 983-985.

[65] Rieux-Laucat F, Le Deist F and Fischer A (2003) Autoimmune lymphoproliferative syndromes: genetic defects of apoptosis pathways. Cell Death Differ. 10, 124-133.

[66] Wang J, C.H., Wong W, Spencer DM, Lenardo MJ, (2001) Caspase-10 is an initiator caspase in death receptor signaling. Proc Natl Acad Sei USA. 98, 13884-13888.

[67] Medana l, Li Z, Flugel A, Tschopp J, Wekerle H and Neumann H (2001) Fas ligand (CD95L) protects neurons against perforin-mediated T lymphocyte cytotoxieity. J Immunol. 167,674-681.

[68] Saas P, Walker PR, Hahne M, Quiquerez AL, Schnuriger V, Perrin G, French L, Van Meir EG, de Tribolet N, Tschopp J and Dietrich PY (1997) Fas ligand expression by astrocytoma in vivo: maintaining immune privilege in the brain? J Clin Invest. 99, 1173-1178.

[69] Choi C and Benveniste E (2004) Fas ligandIFas system in the brain: regulator of immune and apoptotic responses. Brain Res Brain Res Rev. 44, 65-81.

[70] Ulisse S, Cinque B, Silvano G, Rucci N, Biordi L, Cifone MG and D'Armiento M (2000) Erk-dependent cytosolic phospholipase A2 activity is induced by CD95 ligand cross-linking in the mou se derived Sertoli cellline TM4 and is required to trigger apoptosis in CD95 bearing cells. Cell Death Differ. 7, 916-924.

[71] Griffith TS, Brunner T, Fletcher SM, Green DR and Ferguson TA (1995) Fas Ligand-induced apoptosis as a mechanisms of immune privilege Science 270, 1189-92.

[72] Xerri L, Devilard E, Hassoun J, Mawas C and Birg F (1997) Fas ligand is not only expressed in immune privileged human organs but is also coexpressed with Fas in various epithelial tissues. Mol Pathol. 50, 87-91.

[73] Ferguson TA and Griffith TS (2006) A vision of cell death: Fas ligand and immune privilege 10 years later. Immunol Rev. 213.

[74] Walsh K and Sata M (1999) Negative regulation of inflammation by Fas ligand expression on the vascular endothelium. Trends Cardiovasc Med. 9, 34-41.

[75] Barten DM and Ruddle NH (1994) Vascular cell adhesion molecule-1 modulation by tumor necrosis factor in experimental allergic encephalomyelitis. J Neuroimmunol. 51, 123-133.

[76] O'Connell J, O'Sullivan GC, Collins JK and Shanahan F (1996) The Fas counterattack: Fas-mediated T cell killing by colon cancer cells expressing Fas ligand. J Exp Med. 184, 1075-1082.

[77] Niehans GA, Brunner T, Frizelle SP, Liston JC, Salerno CT, Knapp DJ, Green DR and Kratzke RA (1997) Human lung carcinomas express Fas ligand. Cancer Res. 57, 1007-1012.

[78] Bellgrau D, Gold D, Selawry H, Moore J, Franzusoff A and Duke RC (1995) A ~ role for CD95 ligand in preventing graft rejection. Nature. 377, 630-2.

Thornhill, pg. 75

[79] Lau HT, Y.M., Fontana A, Stoeckert CJ Jr, (1996) Prevention ofislet allograft rejection with engineered myoblasts expressing FasL in mice. Science 273, 109-112.

[80] Kang SM, Lin Z, Ascher NL and Stock PG (1998) Fas ligand expression on islets as weIl as multiple ceIllines results in accelerated neutrophilic rejection. Transplant Proc. 30, 538.

[81] Kang SM, Schneider DB, Lin Z, Hanahan D, Dichek DA, Stock PG and Baekkeskov S (1997) Fas ligand expression in islets of Langerhans does not confer immune privilege and instead targets them for rapid destruction. Nat Med. 3,738-743.

[82] Allison J, Georgiou HM, Strasser A and Vaux DL (1997) Transgenic expression of CD95 ligand on islet beta cells induces a granulocytic infiltration but does not confer immune privilege upon islet allografts. Proc Natl Acad Sci USA. 94, 3943-7.

[83] Kang SM, H.A., Le D; Springer ML, Stock PG, Blau HM, (1997) Immune response and myoblasts that express Fas ligand. Science 278, 1322-1324.

[84] Choi C, Xu X, Oh JW, Lee SJ, Gillespie GY, Park H, Jo H and Benveniste EN (2001) Fas-induced expression of chemokines in human glioma ceIls: involvement of extracellular signal-regulated kinase 1/2 and p38 mitogen­activated protein kinase. Cancer Res. 61,3084-3091.

[85] Nishimatsu H, T.T., Ueki T, Kajiwara T, Moriyama N, Ishida T, Li B, Kakizoe T, Kitamura T, (1999) CD95 ligand expression enhances growth of murine renal cell carcinoma in vivo. Cancer Immunol Immunother. 48, 56-61.

[86] Seino K, Kayagaki N, Okumura K and Yagita H (1997) Antitumor effect of locally produced CD95 ligand. Nat Med. 3, 165-170.

[87] Latinis KM, Norian LA, Eliason SL and Koretzky GA (1997) Two NFAT transcription factor binding sites participate in the regulation of CD95 (Fas) ligand expression in activated human T cells. J Biol Chem. 272, 31427-31434.

[88] Li-Weber M, Laur 0 and Krammer PH (1999) Novel Egr/NF-AT composite sites mediate activation of the CD95 (APO-1/Fas) ligand promoter in response to T cell stimulation. Eur J Immunol. 29, 3017-3027.

[89] Kasibhatla S, Genestier L and Green DR (1999) Regulation of fas-ligand expression during activation-induced cell death in T lymphocytes via nuc1ear factor kappaB. J Biol Chem. 274, 987-992.

[90] Faris M, Kokot N, Latinis K, Kasibhatla S, Green DR, Koretzky GA and Nel A (1998) The c-Jun N-terminal kinase cascade plays a role in stress-induced apoptosis in Jurkat cells by up-regulating Fas ligand expression. J Immunol. 160, 134-144.

[91] Faris M, Latinis KM, Kempiak SJ, Koretzky GA and Nel A (1998) Stress­induced Fas ligand expression in T cells is mediated through a MEK kinase 1-regulated response element in the Fas ligand promoter. Mol Cell Biol. 18,5414-5424.

[92] Kasibhatla S, Brunner T, Genestier L, Echeverri F, Mahboubi A and Green DR (1998) DNA damaging agents induce expression of Fas ligand and subsequent apoptosis in T lymphocytes via the activation of NF-kappa Band AP-l. Mol Cell. 1,543-541.

Thomhill, pg. 76

[93] Brunet A, Bonni A, Zigmond MJ, Lin MZ, Juo P, Hu LS, Anderson MJ, Arden KC, Blenis J and Greenberg ME (1999) Akt promotes cell survival by phosphorylating and inhibiting a Forkhead transcription factor. Cell 96, 857-868.

[94] Ryan AE, Lane S, Shanahan F, O'connell J and Houston A (2006) Fas ligand expression in human and mou se cancer celllines; a caveat on over-reliance on mRNA data. J Carcinog. 5, 5.

[95] Xiao S, Deshmukh US, Jodo S, Koike T, Sharma R, Furusaki A, Sung SS and Ju ST (2004) Novel negative regulator of expression in Fas ligand (CD 178) cytoplasmic tail: evidence for translational regulation and against Fas ligand retenti on in secretory lysosomes. J Immunol. 173, 5095-5102.

[96] Jodo S, Pidiyar VJ, Xiao S, Furusaki A, Sharma R, Koike T and Ju ST (2005) Fas ligand (CD 178) cytoplasmic tail is a positive regulator of Fas ligand-mediated cytotoxicity. J Immunol. 174,4470-4474.

[97] Soond SM, Everson B, Riches DW and Murphy G (2005) ERK-mediated phosphorylation of Thr735 in TNFalpha-converting enzyme and its potential role in TACE protein trafficking. J Cell Sci. 118,2371-2380.

[98] Kayagaki N, Kawasaki A, Ebata T, Ohmoto H, Ikeda S, Inoue S, Yoshino K, Okumura K and Yagita H (1995) Metalloproteinase-mediated release ofhuman Fas ligand. J Exp Med. 182, 1777-1783.

[99] Tanaka M, Itai T, Adachi M and Nagata S (1998) Downregulation of Fas ligand by shedding. Nat Med. 4, 31-36.

[100] Schneider P, Holler N, Bodmer JL, Hahne M, Frei K, Fontana A and Tschopp J (1998) Conversion of membrane-bound Fas(CD95) ligand to its soluble form is associated with downregulation of its proapoptotic activity and loss of liver toxicity. J Exp Med. 187, 1205-1213.

[101] Aoki K, Kurooka M, Chen 11, Petryniak J, Nabel EG and Nabel GJ (2001) Extracellular matrix interacts with soluble CD95L: retention and enhancement of cytotoxicity. Nat Immunol. 2, 333-337.

[102] Domonkos A, Udvardy A, Laszlo L, Nagy T and Duda E (2001) Receptor-like properties of the 26 kDa transmembrane form of TNF. Eur Cytokine N etw. 12, 411-419.

[103] Schulte M, Reiss K, Lettau M, Maretzky T, Ludwig A, Hartmann D, de Strooper B, Janssen 0 and Saftig P (2007) ADAMlO regulates FasL cell surface expression and modulates FasL-induced cytotoxicity and activation-induced cell death. Cell Death Differ. 14, 1040-1049.

[104] Mitsiades N, Yu WH, Poulaki V, Tsokos M and Stamenkovic 1 (2001) Matrix metalloproteinase-7 -mediated c1eavage of Fas ligand protects tumor cells from chemotherapeutic drug cytotoxicity. Cancer Res. 61,577-581.

[105] Wang WS, Chen PM, Wang HS, Liang WY and Su Y (2006) Matrix metalloproteinase-7 increases resistance to Fas-mediated apoptosis and is a poor prognostic factor of patients with colorectal carcinoma. Carcinogenesis. 27, 1113-1120.

[106] Fingleton B, Vargo-Gogola T, Crawford HC and Matrisian LM (2001) Matrilysin [MMP-7] expression selects for cells with reduced sensitivity to apoptosis. Neoplasia. 3, 459-468.

[107] Simons K and Toornre D (2000) Lipid rafts and signal transduction. Nat Rev Mol Cell Biol. 1,31-39.

Thornhill, pg. 77

[108] Nachbur U, Kassahn D, Yousefi S, Legler DF and Brunner T (2006) Posttranscriptional regulation of Fas (CD95) ligand killing activity by lipid rafts. Blood. 107,2790-2796.

[109] Cahuzac N, Baum W, Kirkin V, Conchonaud F, Wawrezinieck L, Marguet D, Janssen 0, Zornig M and Hueber AO (2006) Fas ligand is localized to membrane rafts, where it displays increased cell death-inducing activity. Blood. 107,2384-239l.

[110] Wasem C, Frutschi C, Arnold D, Vallan C, Lin T, Green DR, Mueller C and Brunner T (2001) Accumulation and activation-induced release ofpreformed Fas (CD95) ligand during the pathogenesis of experimental graft-versus-host disease. J hnmunol. 167,2936-2941.

[111] Tan KH and Hunziker W (2003) Compartmentalization of Fas and Fas ligand may prevent auto- or paracrine apoptosis in epithelial cells. Exp Cell Res. 284, 283-290.

[112] McKechnie NM, King BC, Fletcher E and Braun G (2006) Fas-ligand is stored in secretory lysosomes of ocular barrier epithelia and released with microvesic1es. Exp Eye Res. 83, 304-314.

[113] Blott EJ, Bossi G, Clark R, Zvelebil M and Griffiths GM (2001) Fas ligand is targeted to secretory lysosomes via a proline-rich domain in its cytoplasmic tail. J Cell Sei. 114,2405-16.

[114] Lyubchenko TA, Wurth GA and Zweifach A (2001) Role of calcium influx in cytotoxic T lymphocyte lytic granule exocytosis during target cell killing. Immunity 15,847-859.

[115] Smith AJ, Pfeiffer JR, Zhang J, Martinez AM, Griffiths GM and Wilson BS (2003) Microtubule-dependent transport of secretory vesic1es in RBL-2H3 cells. Traffic. 4, 302-312.

[116] Martinez-Lorenzo MJ, Anel A, Gamen S, Monle n l, Lasierra P, Larrad L, Pineiro A, Alava MA and Naval J (1999) Activated human T cells release bioactive Fas ligand and AP02ligand in microvesic1es. J hnmunol. 163, 1274-128l.

[117] Jodo S, Hohlbaum AM, Xiao S, Chan D, Strehlow D, Sherr DH, Marshak­Rothstein A and Ju ST (2000) CD95 (Fas) ligand-expressing vesicles display antibody-mediated, FcR-dependent enhancement of cytotoxieity. J Immunol. 165, 5487-5494.

[118] Eissner G, Kolch W and Scheurich P (2004) Ligands working as receptors: reverse signaling by members of the TNF superfamily enhance the plasticity of the immune system. Cytokine Growth Factor Rev. 15,353-66.

[119] Desbarats J, Duke RC and Newell MK (1998) Newly discovered role for Fas ligand in the cell-cyc1e arrest of CD4+ T cells. Nat Med. 4, 1377-1382.

[120] Suzuki 1 and Fink PJ (1998) Maximal proliferation of cytotoxic T lymphocytes requires reverse signaling through Fas ligand. J Exp Med. 187, 123-128.

[121] Suzuki 1 and Fink PJ (2000) The dual functions offas ligand in the regulation of peripheral CD8+ and CD4+ T cells. Proc Natl Acad Sei USA. 97, 1707-1712.

[122] Sun M, Ames KT, Suzuki 1 and Fink PJ (2006) The cytoplasmic domain of Fas ligand costimulates TCR signals. J hnmunol. 177, 1481-1491.

[123] Kirchner S, Boldt S, Kolch W, Haffner S, Kazak S, Janosch P, Holler E, Andreesen Rand Eissner G (2004) LPS resistance in monocytic cells caused by

Thornhill, pg. 78

reverse signaling through transmembrane TNF (mTNF) is mediated by the MAPKIERK pathway. J Leukoc Biol. 75, 324-33l.

[124] Watts AD, Hunt NH, Wanigasekara Y, Bloomfield G, Wallach D, Roufogalis BD and Chaudhri G (1999) A casein kinase I motif present in the cytoplasmic domain of members of the tumour necrosis factor ligand family is implicated in 'reverse signalling'. EMBO J. 18,2119-2126.

[125] Kirchner S, Holler E, Haffner S, Andreesen R and Eissner G (2004) Effect of different tumor necrosis factor (TNF) reactive agents on reverse signaling of membrane integrated TNF in monocytes. Cytokine. 28, 67-74.

[126] Waetzig GH, Rosenstiel P, Arlt A, Till A, Brautigam K, Schafer H, Rose-John S, Seegert D and Schreiber S (2005) Soluble tumor necrosis factor (TNF) receptor-1 induces apoptosis via reverse TNF signaling and autocrine transforming growth factor-betal. FASEB J. 19,91-93.

[127] Jiang X, Huang F, Marusyk A and Sorkin A (2003) Grb2 regulates internalization of EGF receptors through clathrin-coated pits. Mol Biol Cell. 14, 858-870.

[128] Buday Land Downward J (1993) Epidermal growth factor regulates p21ras through the formation of a complex of receptor, Grb2 adapter protein, and Sos nucleotide exchange factor. Cell. 73, 611-620.

[129] Kolch W (2000) Meaningful relationships: the regulation of the RasIRaflMEKlERK pathway by protein interactions. Biochem J. 351,289-305.

[130] Fresno Vara JA, Casado E, de Castro J, Cejas P, B.-I.e. and Gonzalez-Baron M (2004) PI3K/Akt signalling pathway and cancer. Cancer Treat Rev. 30, 193-204.

[131] Zuccato E, Blott EJ, HoIt 0, Sigismund S, Shaw M, Bossi Gand Griffiths GM (2006) Sorting of Fas ligand to secretory lysosomes is regulated by mono­ubiquitylation and phosphorylation. J Cell Sei. 120, 191-199.

[132] Hane M, Lowin B, Peitsch M, Becker K and Tschopp J (1995) Interaction of peptides derived from the Fas ligand with the Fyn-SH3 domain. FEBS LeU. 373, 265-268.

[133] Qian J, Chen W, Lettau M, Podda G, Zornig M, Kabelitz D and Janssen 0 (2006) Regulation of FasL expression: a SH3 domain containing protein family involved in the lysosomal association of FasL. Cell Signal. 18, 1327-1337.

[134] Kessels MM and Qualmann B (2004) The syndapin protein family: linking membrane trafficking with the cytoskeleton. J Cell Sci. 117, 3077-3086.

[135] Qualmann B, Roos J, DiGregorio PJ and Kelly RB (1999) Syndapin I, a synaptic dynarnin-binding protein that associates with the neural Wiskott-Aldrich syndrome protein. Mol Biol Cell. 10, 501-513.

[136] Kessels MM and Qualmann B (2002) Syndapins integrate N-WASP in receptor­mediated endocytosis. EMBO J. 21, 6083-6094.

[137] Mori S, Tanaka M, Nanba D, Nishiwaki E, Ishiguro H, Higashiyama Sand Matsuura N (2003) PACSIN3 binds ADAM12/meItrin alpha and up-regulates ectodomain shedding of heparin-binding epidermal growth factor-like growth factor. J Biol Chem. 278, 46029-46034.

[138] Baum W, Kirkin V, Fernandez SB, Pick R, Lettau M, Janssen 0 and Zornig M (2005) Binding of the intracellular Fas ligand (FasL) domain to the adaptor protein PSTPIP results in a cytoplasmic localization of FasL. J Biol Chem. 280, 40012-40024.

Thornhill, pg. 79

[139] Rivero-Lezcano OM, Marcilla A, Sameshima JH and Robbins KC (1995) Wiskott-Aldrieh syndrome protein physically assoeiates with Nek through Sre homology 3 domains. Mol Cell Biol. 15, 5725-573l.

[140] Gil D, Sehamel WW, Montoya M, Sânehez-Madrid F and Alarcon B (2002) Recruitment of Nck by CD3 epsilon reveals a ligand-induced conformation al change essential for T cell receptor signaling and synapse formation. Cell 109, 901-912.

[141] Barda-Saad M, Braiman A, Titerence R, Bunnell SC, Barr VA and Samelson LE (2005) Dynamic molecular interactions linking the T cell antigen receptor to the aetin eytoskeleton. Nat Immunol. 6, 80-89.

[142] Badour K, Zhang J and Siminovitch KA (2004) Involvement of the Wiskott­Aldrich syndrome protein and other actin regulatory adaptors in T cell activation. Sernin Immunol. 16, 395-407.

[143] Lettau M, Qian J, Linkermann A, Latreille M, Larose L, Kabelitz D and Janssen 0(2006) The adaptor protein Nck interacts with Fas ligand: Guiding the death factor to the eytotoxie immunological synapse. Proc Natl Acad Sei USA. 103, 5911-5916.

[144] Newell M. and Desbarats J. (1999) Fas ligand: reeeptor or ligand? Apoptosis 4, 311-5.

[145] Nagata Sand Golstein P (1995) The Fas death factor. Science 267, 1449-1456. [146] Ford MS, Young KJ, Zhang Z, Ohashi PS and Zhang L (2002) The immune

regulatory function of lymphoproliferative double negative T cells in vitro and in vivo. J Exp Med. 196,261-267.

[147] Cohen PL and Eisenberg RA (1991) Lpr and gld: single gene models of systemic autoimmunity and lymphoproliferative disease. Annu Rev Immunol. 9, 243-269.

[148] Worth A, Thrasher AJ and Gaspar HB (2006) Autoimmune lymphoproliferative syndrome: molecular basis of disease and clinical phenotype. Br J Haematol. 133, 124-140.

[149] Owen DJ, Vallis Y, Pearse BM, MeMahon HT and Evans PR (2000) The structure and function of the beta 2-adaptin appendage domain. EMBO J. 19, 4216-4227.

[150] Kirchhausen T (1999) Adaptors for clathrin-mediated traffic. Annu Rev Cell Dev Biol. 15,705-732.

[151] Worby CA, Simonson-LeffN, Clemens JC, Huddler D Jr, Muda M and Dixon JE (2002) Drosophila Ack targets its substrate, the sorting nexin DSH3PX1, to a protein complex involved in axonal guidance. J Biol Chem. 277, 9422-9428.

[152] Howard L, Nelson KK, Maciewicz RA and Blobel CP (1999) Interaction of the metalloprotease disintegrins MDC9 and MDC15 with two SH3 domain­containing proteins, endophilin 1 and SH3PXl. J Biol Chem. 274, 31693-31699.

[153] Schulze WX and Mann M (2004) A novel proteomic screen for peptide-protein interactions. J Biol Chem. 279, 10756-10764.

[154] Lundmark R and Carlsson SR (2002) The beta-appendages of the four adaptor­protein (AP) complexes: structure and binding properties, and identification of sorting nexin 9 as an accessory protein to AP-2. Biochem J. 362, 597-607.

[155] Soulet F,Yarar D, Leonard M and Schmid SL (2005) SNX9 regulates dynamin assembly and is required for efficient clathrin-mediated endocytosis. Mol Biol Cell. 16,2058-2067.

Thornhill, pg. 80

[156] Lin Q, Lo CG, Cerione RA and Yang W (2002) The Cdc42 target ACK2 interacts with sorting nexin 9 (SH3PX1) to regulate epidermal growth factor receptor degradation. J Biol Chem. 277, 10134-10138.

[157] Hicke L (2001) A new ticket for entry into budding vesicles-ubiquitin. Cell106, 527-530.

[158] Qian J, Lettau M, Podda G and Janssen 0 (2005) FasL associated factors and their potential role in the regulation of FasL expression Signal Transd. 4, 195-201.

[159] Wagelie-Steffen AL, Hartmann K, Vliagoftis H and Metcalfe DD (1998) Fas ligand (FasL, CD95L, APO-1L) expression in murine mast cells. Immunology. 94,569-574.

[160] Thornhill PB, Drury G, Cohn JB, Stanford WL, Bernstein A and Desbarats J (2007) A Proteomic Screen Reveals Novel Fas Ligand Interacting Proteins within Nervous System Schwann cells FEBS Lett. Submitted.

[161] Waterman H, Katz M, Rubin C, Shtiegman K, Lavi S, Eison A, Jovin T and Yarden Y (2002) A mutant EGF-receptor defective in ubiquitylation and endocytosis unveils a role for Grb2 in negative signaling. EMBO J. 21, 303-313.

[162] Worby CA and Dixon JE (2002) Sorting out the cellular functions of sorting nexins. Nat Rev Mol Cell Biol. 3,919-931.

[163] Yamazaki T, Zaal K, Hailey D, Presley J, Lippincott-Schwartz J and Samelson LE (2002) Role of Grb2 in EGF-stimulated EGFR internalization. J Cell Sei. 115, 1791-1802.

[164] Traub LM (2005) Common prineiples in clathrin-mediated sorting at the Golgi and the plasma membrane. Biochim Biophys Acta. 1744,415-437.

[165] Hirst J, Motley A, Harasaki K, Peak Chew SY and Robinson MS (2003) EpsinR: an ENTH domain-containing protein that interacts with AP-1. Mol Biol Cell. 14, 625-641.

Thornhill, pg. 81

Appendix

Thornhill, pg. 82

·.-__________________________________________________________________________ ~3

The following procedures have not changed:

Sciatic nerve regeneration experiments (as a model ofperipheral nerve regeneration after trauma) ".

- Tagging (by ear punch) for identification of individual mice. - ip injection of anti-Fas Ligand antibody or control antibody, in 200lli saline. - Anesthesia of adult mice by avertin injection as per SOP. - Sciatic nerve crush surgery under avertin anesthetic. (Procedure as follows: Mice are anesthetized with avertin and prepared for surgery by shaving and disinfecting the right flank and leg. A l-cm long incision was made through the skin, and the gluteal and hamstring muscles of the upper dorsal thigh seperated to reveal the sciatic nerve. The nerve is isolated and grasped in Dumont #5 forceps 5 mm distal to the sciatic notch. The nerve is crushed by application of pressure for 30 seconds. The crush is immediately repeated at the same site for an additional 30 seconds. 2 Jlg anti-Fas Ligand Ab or control Ab, in physiological saline or in fibrin or matrigel™ matrix, are injected into the crush site in sorne mice at this time. The muscles and skin are closed with absorbable Dexon sutures and the mice allowed to recover. After the frrst post-operative day, the mice stop dragging their operated limb and gradually resume weight-bearing. The animaIs have never been observed to gnaw on their operated paw to the point ofinjury following this procedure.) - Walking track analysis (behavioural study for weight bearing on injured limb, which consists of coating the hind paws of mice with India Ink, and allowing the mice to walk along an 18-inch long track lined with paper) every two days post-surgery. - Euthanasia as per SOP. - Tissue collection.

Immunization of donor mice for induction of Experimental Allergic Enc~halomyelitis (EAE,a mouse model of Multiple Sclerosis)

- Immunization with protein or peptide antigens (50-400 micrograms, depending on the antigen), emulsified in Complete Freund's Adjuvant (1:1 ratio by volume ofantigen to CFA), 100 ul subcutaneously at two sites (bilaterallynear scruffofthe neck, 50 ut per site). - Euthanasia as per SOP, 10 days after immunization. - Tissue collection.

BAB induction

- Tagging (by ear punch) for identification of individual mice. - iv injection ofT cells and pertussis preparation (pTX, 200ng in 200JlI saline) to induce BAB. Pertussis preparations are used here as an adjuvant to induce EAB, but can also potentiate insulin secretion, suppress hyperglycemia, promote leukocytosis, and sensitize to histamine, and it is considered biohazardous. Therefore, the foUowing precautions are routinely employed: the powdered pertussis (Sigma) is reconstituted in the sealed vial, then aU personal protective precautions (gloves, masks, lab coats and protective eyeware) are employed during reagent preparation and mouse immunization. The reagent is very unstable and after being injected into the mouse does not present a hazard to personnel. Pertussis has been used as an adjuvant to induce BAB since the 1960's (Levine et al., Hyperacute allergic encephalomyelitis: adjuvant effect ofpertussis vaccines and extracts. J Immunol. 1966,97:363-8), with, to our knowledge, no reported ill-effects on personnel. - ip injection of anti-Fas Ligand antibody injection in 200Jll saline (as above). - Anesthesia by avertin injection as per SOP. - Intrathecal injection of anti-Fas Ligand antibody in 0.21l1 saline using a stereotaxic apparatus on anesthetised mice. - Daily observation for BAB grading. - Buthanasia as per SOP at 14, 21 or 30 days following fmal injection, OR as soon as pain or distress is observed (lack of drinking, feeding or grooming; hunched posture; piloerection; reduced mobility). - Tissue collection.

Spinal Cord Injury and Recoverv

- Behavioral testing: walking track analysis (baseline), as described under sciatic nerve surgery, and observation ofmouse posture and gait during walking to determine a numerical score. - Anesthesia - as per SOP - Survival Surgery: dorsal hemisection - asper SOP, with thefollowing details pertaining to the surgical procedure:

. One spinal cord segment in the lower thoracic region WilfOe exposed by doing a laminectomy of the vertebra. A smaIllesl0n (0.5n:un ~'l depth) will be made in the dorsal half of the spinal cord(doiiallièmisection) using a pair of micro-scissors. 2 Jlg anti-FasL AOoi: -- .

, vontrol Ab, in physiolog~cal saline or in fibri~ or matrigel™ nia~lll:~applied to. the lesion ~ so~e ~ce at ~s time .. The overlying . patavertebral muscles wIll then be sutured WIth absorbable 6-0 sutures, and the skin closed WIth skin clips. Anunals WIll be kept

.-__ ~~~-.-__ ~-.-~ __ -.-~ __ ~ __________________________________________ ~ __ -,4 wann with a heating pad or heat lamp (monitered with a thennometer) after surgery. Skins clips will be removed 7 days later. Special care will be taken to place food in the cage for easy access. The animais' bladders will be evacuated manually every 12 hours if the animais become unable to urinate due to the spinal cord damage (niost mice will require manual evacuation for the frrst 48 hours at least), and for prophylaxis ofurinary tract infections, aIl mice will receive gentamycin (1 mg! kg) administered subcutaneously once

er day for 7 days. This prophylactie regimen was reeommended by the Reeve-Irvine Institute, University of California, where tbis model was developed, and has been discussed with the facility Veterinarian (Dr. Matsumiya). Less than 10% of the animaIs might be expeeted to beeome fully paraplegie. Mice will be weighed weekly, and those showing loss ofbody weight of 15%, wound dehiscence, pressure sores, dehydration or lethargy will be euthanized. - Analgesia (post operative), as per SOP (for 48 hours post-operatively) - Behavioral testing (as described above, every two days post-surgery) - Behavioral testing for hyperalgesia (Tail-withdrawal test: Mice are lightly restrained in a clothlcardboard "pocket." The protruding distal half of the tail is then dipped into a bath of circulating water thennostatically controlled at 49.0 ± 0.2°C. Latency to respond to the heat stimulus with tail flexion is measured to the nearest O.ls. Three separate withdrawallatency detenninations, separated by~Os, are made. Only non-paraplegic mice with good motor function in the tail will be assessed, and the water temperature is too low to cause tissue damage. Von Frey (VFl filament test of mechanical sensitivity: Mice are placed withln Plexiglas observation chambers (4 x 4 x 7 cm) on a metal mesh floor. Mechanical sensitivity is measured by determining the median 50% foot withdrawal threshold with von Frey monofilaments (bending forces: 0.3,0.7, 1.6,4.0,9.8,22.0, and 53.9 mN; applied using a single, steady >1-sec application) using the up-down method ofDixon (Chaplan et al. 1994). AU mice will be tested in their front paws (as normal motor function is preserved in their front paws even immediately post-operatively, since the spinallesion is mid-thoracic). Only non-paraplegic mice with the mechanical ability to withdraw the hind paw will be assessed on their hind paws.) - Anesthesia - as per SOP - Stereotaxic surgery: injection of tracer dye to determine extent of nerve regeneration - as per SOP, and described in detail below. After a survival time of3-6 weeks animaIs will be deeply anesthetized as per SOP, and the skull over the region of the parietal cortex will be removed with the help of a fme motor drill. The dura will be incised and the underlying cerebral cortex injected with 0.1~1 of tracer per site in 6 sites. The tracer is a neuronal marker that will be taken up by nerve cells in the cortex and transported down their processes into the spinal cord. These injections will be done using a glass micropipette (tip size 50 ~). This technique is widely used to trace neuronal connections in the brain and spinal cord. After the injection, the skin closed with 5-0 sutures, and the mice will be kept warm with a heating pad or heat lamp (monitered with a thennometer). - Analgesia (post operative), as per SOP (for 48 hours post-operatively) - Euthanasia - as per SOP - Tissue collection

The procedures are the same as the original protocol:

IF NO, supply new endpoints that are different from the original protocol:

.. ·b)Fof:'i~!ll:'~~î:f Include here ALL endpoints, incIuding the ones described in the original proto col as weil as new and changed endpoints in CAPS:

Experimental endpoints: no changes: For the sciatic nerve surgery mice, the endpoint will be 3 weeks post surgery (by which time the mice are walking normally). For the EAE donor mice, the endpoint is 10 days post immunization for collection of activated T cells. For the EAE recipient mice, mice will be euthanized at days 14 or 30 days following fmal injection. For the SCI mice, 2 days post hemisection to determine local neurotrophin production in the spinal cord, or 3 weeks post hemisection to determine functional recovery, or 3 weeks post hemisection surgery, then 1 week after injection of tracer (total = 4 weeks) to determine nerve regenera,tion. For aU mice, the endpoint will OCCur when they are between 7 and 12 weeks old.

Clinical endpoints: no changes: . ___ .. pain or distress is observed (lack of drinking, feedingor.grooming; hunched posture; piloerection; dehydration; lethargy), ode ... eight loss exceeds 20% (mice will be weighed weekly), orifwôund dehiscence or pressure sores develop, the animal will be

ellthanized. .. . .

8. Hazards (check here ifnoneareused:[gI) a) Are the hazards different from original protocol? (infections, radioactive, toxic, carcinogen, tumours)

L--~"

YESO NO[gl ifyes, supply details (material, risks, precautions):

b) Have the celllines been tested for human and animal pathogens? YEs:D

;: Jackson Labs, "Charles River

C57BL/6 (B6)

Strain 1: B6 (wild-type) mice

EAE experiments: Treatment groups: 1. EAB untreated 2. EAB plus control Ab it. 3. EAB plus Fas Ligand Ab it.

Mice

Jackson Labs Jackson Labs

C57BL/6-Faslpr (lpr) C57BL/6-Faslg1d (gld)

Females and males Females and males

6-8 weeks 6-8 weeks

117 117

117 117

4. EAB induced with T ceUs pre-treated with control Ab 5. BAB induced with T ceUs pre-treated with anti-FasL Ab

Recipient mice = 6 groups x 6 mice/group x 2 time points = 72 mice

No:D None used:[gJ

C57BL/6-Fasnu 1 (Fas-nuU)

Females and males

6- 8 weeks

117

117

We have determined in previous experiments that 6 are required per group for statistical significance, since disease course is not always consistent between mice. Donor mice (as a source ofT cells), with 1 donor mouse / 2 recipient mouse required = 36 mice

Sciatic nerve experiments: Treatment groups: 1. sham surgery 2. surgery plus control Ab 3. surgeryplus Fas Ligand Ab

3 groups x 5 mice/group x 3 experimentaLrepJiçates = 45 mice

"pinai cord injury (sCI) experiments: Treatment Cl"r1'\111"\C'

5

.-~ ________________________________________________________________________ -,6

1. sham surgery 2. surgery plus control Ab 3. surgeryplus Fas Ligand Ab

J groups x 8 mice/group x 3 time points = 72 mice Due to greater variation in spinal surgery compared with peripheral nerve crush, and occasional intra-operative death (less than 1 per group on average), the number ofmice per group is greater for SCI than for sciatic surgery. Due to the labour intensive nature ofthese experiments we only anticipate one experimental replicate per strain per year, on average.

Total Strain 1 (B6 mice): 72 + 36 + 4S + 72 = 22S mice

Strains 2,3 and 4 consist ofB6 mice bearing mutations in Fas or FasL. Bach ofthese strains will be used for sciatic nerve and SCI experiments in parallel with the wild-type mice as described above. Therefore,

Total Strain 2, 3, and 4 (each) = 4S + 72 = 117

Submit to your local Facility Animal Care Committee.·Please note that after two renewals, a full protocol needs to be submitted.

This approval does no! imply tha! space will be made available. If a major increase of space needs is anticipated, please contact the appropriate animal facility manager.

170'd ltllOl -: ~ JjJU;;IU ç,'OwI.IUI'IUlIll;;:lljlll

(' iii) the protocol for decontaminating spills

Bleach will be used to decontaminate sp01s, acoordins to the following protocol:

1- ares. of spill contained

2- spill Pbsorbed with papa towels

3· area soak.ed with bleach

4- towels disposed of in labels bags for autoclaving

5· area washed. and nnsed

An wastc will be deconmminated \Vith bleaeh and plastics will be disposed of in labelled bag.!:, which will then be autoclaved.

•

7. Does the protoc:ol present conditions (e.g. handlmg oflarge volumes or high concentrations ofpathogeDs) whieh could incr~e the hazards of the infectious agent(s)7 .

NO

8. Do the specifie procedures to be employed involving genetically engine.ered orgsnisms have a lmtory ofsafe use?

NIA

9. What precautions are being taken to reduce production ofinfectious droplets and aerosols?

1 ~ No infectious material will he used 2- AlI procedures are carricd out in laminaI' flow boods, which have been duly tested and certi.fied 3. AlI procedures will be canied out by ttained personnel

O. l 'cal List the bic ogit safety t-abinets to be \lScd.

Building RoomNo.

McInryre 1234

Mcfnryre 1234

170/170'd 9è917 86~ 1715

Manufacturer ModelNo.

ThermoFonna 1286

Nuaire NU-42S-400

3JI~~O INtI~9 HJ~tl3S3~

.~

Seriai No. Date Çenified

23023-1261 ~(~ .~

SS37S C1J/~ __ -v

TOTAL P.0:3 81:51 900è-0è-Nnr

r-\

(

< MHi"<-~d-dIcl1cl4 14; 41:::l c.N~JlRONMENTRL SRt-E: TY OFF ._---- _._---

iH) the protocol for decontaminating spills

Bleach will he used to deeontamil'l.aœ spiUs, according ta the following protocol:

1- ares of spiU eontained

2- spill absorbe<! with paper towels

3- area $oaked with bleach

4- towels disposed of in labels bags for autoelaving

5- area washed and rinsed .

514 398 8047 P.03/03

Ali v,"&Ste will oc d~ontaminated ... 'iih bleach and "l~Jt3 will be disposed of ùi labelled bags, wroich will thm be autoclaved.

7. Docs the protoc:ol present conditions (c.g. handling of large volumes or rugh concentrations of pathogens) wbich could increase the h3:l:3rds of the infectious agent(s)?

NO

8. Do the specifte proeedures 10 oe employed involving gen.etica11y engineered organisms have a hlstol)' of saie use?

NIA

9. What precautions are being t.3Iœn 10 reduce production of infeetious droplets and aeroso1s?

1- No inf~ous matcrial will be used 2 .. Ail procedur~ are carrÎet:i out in laminar flow hood~ which have been duly ttsted and certi.fied 3~ AIl prooedures wlll be camed out by ttained personnel

P. = 10. List the biololl:ica.ltafety cabinets to be used.

Building RoomNo. Manufacturer ModelNo. Seriai No.

Mdntyrt 1234 Thennoronna 1286 23023-1261

McIntyre 1234 Nuaite NU·425-400 55375

Date Ctrtified

OS/2003

0512003

** TOTAL PAGE. 84 ** e> ~.<1

TOTAL P.03 MAR 22 '04 15:07 514 398 8047 PAGE. 03