signal transduction by tumour necrosis factor and …tumour necrosis factor related ligands and...

Signal transduction by tumour necrosis factor andtumour necrosis factor related ligands and theirreceptors

Bryant G Darnay, Bharat B Aggarwal

Many biological functions are regulatedthrough interactions of extracellular moleculeswith their cognate cell surface receptors. Thetransduction of these signals by their receptorsat the plasma membrane to the intracellularmachinery results in such cellular activities asgene activation, protein phosphorylation, cellproliferation, and cell destruction. Though thekinetics of these activities may diVer, theirinteractions are coordinated by selective inter-play between the receptors’ intracellular do-mains and a select set of intracellular receptorbinding proteins. One such family of extracel-lular molecules and their cell-surface receptorsare the tumour necrosis factor (TNF) family ofrelated cytokines and receptors, which is thetopic of this review.

Rheumatoid arthritis, a disease of jointinflammation and destruction, is the result ofinappropriate activation of resident and inflam-matory cells within the synovial tissue. Theconsequence of an initiating and as yetunknown stimulus, the cascade of inflamma-tory processes are chronic and self perpetuat-ing. The inflammation in the joints characteris-tic of arthritis is believed to be atrributablelargely to misregulation of cytokine produc-tion, abnormal expression of receptors, or theabsence of counter-regulatory pathways. Twoproinflammatory cytokines, TNF and inter-leukin 1 (IL1), are believed to be the majorcytokines cooperating in the pathology of thisdisease.1 Therapeutic approaches that inhibitthe interaction of these ligands with theirreceptors has been a successful avenue in thetreatment of rheumatoid arthritis.2

One characteristic common to both TNFand IL1 is their ability to activate thetranscription factor nuclear factor-kappa B(NF-êB), which is responsible for regulationof a number of genes necessary for the inflam-mation process.3 More recently, the elucida-tion of the TNF and IL1 signalling pathwayshas provided novel candidate molecules fromwhich to develop therapeutic inhibitors thatwould block NF-êB activation. Furthermore,additional members of the TNF family havebeen discovered and are also capable ofactivating NF-êB. To date, 21 members of theTNF receptor superfamily and 17 members ofthe TNF ligand superfamily have been identi-fied. Most of these ligand/receptor pairsparticipate in modulating various physiologi-cal processes, including the immune response,anti-tumour activity, cellular proliferation anddiVerentiation, and apoptosis. Many of thesephysiological processes are controlled through

a network of these and other cytokinesproduced by various types of cells and theiraberrant regulation may result in inflamma-tory diseases. In this review, we would like tofocus on the recent developments in the char-acterisation of the TNF signalling pathwaylearned from multiple approaches includinggene disruption in mice and on reports ofrecently discovered members of the TNFligand and receptor superfamilies. It is likelythat these novel cytokines also cooperate inregulating the immune system, and thus maybe involved in inflammatory diseases.

TNF signal transductionAlthough produced primarily by activatedmacrophages, small amounts of TNF are pro-duced by several other cell types. TNF isexpressed as a 26 kDa transmembrane protein,which is processed to a soluble 17 kDa proteinreleased via specific proteolytic cleavage. Someof the well known activities ascribed to TNFinclude septic shock, cytotoxicity, inflamma-tion, and viral replication. Clearly, TNF is apleiotropic cytokine perhaps because virtuallyall cells express at least one of the two types ofTNF receptors. The signalling pathways initi-ated by TNF binding to its receptor have beenextensively investigated, clarifying the signal-ling components linking receptor activation tobiological activities.4 The advent of the yeasttwo hybrid system for identifying protein-protein interactions and the availability ofexpressed sequence tag (EST) databases haveassisted in the identification of a unique andnovel set of signalling machinery used by TNFand other related family members. These noveladaptor proteins seem to be promiscuous andthus are used by more than one TNF receptorfamily member for signal transduction. Al-though specific functions have been assigned tothese adaptors in relation to the cellularresponses activated by TNF receptor engage-ment, the physiological relevance of each adap-tor protein in the context of ligand stimulationmust await its targeted disruption in mice.Where these experiments have been done,however, some unexpected findings haveemerged.

Signalling cascades initiated by variousmembers of the TNF receptor family includethose that activate transcription factors (that is,NF-êB and AP1),3 protein kinases (that is,MAPK, JNK, p38),5 and proteases.6 7 Over thepast few years, a number of novel adaptor pro-teins have been identified that initiate thesesignalling cascades. One family, the death-

Ann Rheum Dis 1999;58:(Suppl I) I2–I13I2

Cytokine ResearchLaboratory,Department ofMolecular Oncology,The University ofTexas M D AndersonCancer Center,Houston, Texas 77030,USA

Correspondence to:Dr B Darnay.

on October 31, 2020 by guest. P

rotected by copyright.http://ard.bm

j.com/

Ann R

heum D

is: first published as 10.1136/ard.58.2008.i2 on 1 Novem

ber 1999. Dow

nloaded from

http://ard.bmj.com/

domain proteins,8 link death receptors todownstream proteases of the caspase familynecessary for activation of apoptosis. The deathdomain is a protein-protein interaction motif,which is a conserved stretch of approximately90 residues. The homophillic or heterophillicinteraction between death domain containingproteins is most probably through electrostaticinteractions, as revealed by the structure of thedeath domain of Fas,9 which consists of a seriesof antiparallel amphipathic á-helices withmany exposed charged residues. For example,ligand binding to the cell surface receptorcauses a rearrangement of the intracellulardomain to oligomerise with adaptors and inturn initiate signal transduction cascades.Consistent with this model, forced overexpres-sion of death receptors in cultured cells causesa ligand independent apoptotic eVect indistin-guishable from ligand stimulation. Thus, it

seems that oligomerisation caused by ligandbinding to the receptor initiates the signallingcascades.

A second family of adaptor proteins identi-fied as signalling components of the TNFreceptor family is the TNF receptor associatedfactor (TRAFs) family, which appears to func-tion primarily in the activation of transcriptionfactors and protein kinases.10 The TRAF familyconsists of six distinct proteins, each containinga ring and zinc finger motif in their N-terminaland C-terminal domains that appear to beresponsible for self association and proteininteraction (fig 1). All, except for TRAF4, wereidentified through yeast two-hybrid screeningusing a cytoplasmic domain of various mem-bers of the TNF receptor family. To date,TRAF4 has no known function. The interac-tion of TRAF1, TRAF2, and TRAF5 with vari-ous cytoplasmic domains of TNF receptor

Figure 1 The TRAF family of proteins. Top, each of the TRAF molecules is depicted with the indicated motifs. Bottom,members of the TNF receptor family are listed with the TRAF molecules that directly interact with the receptor. LMP-1 andIL1 receptors are not members of this family, but have been shown to bind to TRAF molecules.

TRAF6

TRAF5

TRAF4

TRAF3

TRAF1

TRAF2

TRAF6TRAF5TRAF3TRAF1Receptor

TNFR2

LTβ R

CD40

CD30

CD27

HVEM

RANK

LMP-1

IL1R

OX40

4-1BB

GITR

–

– – – –

–

–

– –

– –

–

–

–––

–

–

?

?

?

?

?

TRAF2

Ring finger Zn finger TRAF-N TRAF-C

��¡¢��stu��wx��rstuvw

TNF related ligands and their receptors I3



j.com/

Ann R

heum D


ber 1999. Dow

nloaded from

http://ard.bmj.com/

family members requires a specific motif in thereceptor (that is, PXQXT). Unlike these TRAFmolecules, TRAF6 uses a distinct motif (that is,QXPXE), which has been identified in CD40and RANK.11 12 However, of the known TRAFmolecules, only TRAF2, TRAF5, and TRAF6have been demonstrated to mediate NF-êB andJNK activation.

Over the past few years, the signallingmachinery linking the TNF receptor to threedownstream targets (that is, apoptosis, NF-êB,and JNK activation) has been elucidated (fig2). To understand the complexity of the identi-fication of the signalling components of a path-way, one must first understand how they areidentified. For example, after the adaptor isidentified, it is examined for its ability to eitheractivate or inhibit downstream signalling path-ways by transfection of its cDNA into culturedcells. Furthermore, a mutant version of thepotential adaptor molecule is introduced intocells and examined for its ability to inhibit aspecific ligand dependent end point. If themutant version blocks this pathway, then oneconcludes this adaptor molecule participates insignalling by the tested ligand. Although this isnot a foolproof scheme, it has become a verypowerful tool in the study of the signallingevents aVected by TNF and other members ofthis family. There are potential pitfalls to arriv-ing at general conclusions when performing

these types of experiments.4 13 However, thephysiological role of these adaptor molecules inTNF signalling and development has beenrecently revealed by targeted disruption of theirgenes in mice, most notably TRAF2, RIP,FADD, caspase 8, and FAN.

One of the first molecules to be identifiedand required for NF-êB and JNK activation byTNF was TRAF2. Initially, when it wasdiscovered, this protein was shown to activateNF-êB, and was later found to activate JNKwhen overexpressed in cultured cells. Further-more, a mutant version of TRAF2 couldinhibit TNF induced NF-êB and JNK. Thus,from the early reports it appeared that TRAF2was essential for TNF dependent NF-êB acti-vation. However, from the TRAF2 knockoutmouse model, TNF could surprisingly stillactivate NF-êB in embryonic fibroblasts, butnot JNK.14 15 Furthermore, TRAF2 -/- miceappeared normal at birth but became progres-sively runted and died prematurely. Defects inB cell precursors and atrophy of the thymuswere also observed. Moreover, these miceexhibited increased serum concentrations ofTNF, and thymocytes and haematopoietic cellswere highly sensitive to TNF induced apopto-sis. These observations suggest that TRAF2 isrequired for TNF induced JNK activation andalso important in the regulation of lymphocytefunction and growth.

Figure 2 Schematic diagram of TNF signal transduction molecules and the biological activities activated by the adaptor proteins.

PIP5K

?

TRAP1

TRAP2

p60TRAK

MADD

Sentrin

ERK, JNK,Apoptosis

Apoptosis

Apoptosis

Apoptosis

All targets

SODD NF-κB

NF-κB

NF-κB

RAIDD

RIP

TRADD

FAN SMase Ceremide

caspase 8FADD

TRAF2 GCK MEKK1

MKK7NIK

ASK1

MAPKKK

MKK6

JNK

p38

AP1

JNKIKKS

I4 Darnay, Aggarwal



j.com/

Ann R

heum D


ber 1999. Dow

nloaded from

http://ard.bmj.com/

RIP, or receptor interacting protein, whichwas initially identified as a Fas associated deathdomain kinase, seems not to play a part in Fasmediated apoptosis but rather in TNF medi-ated NF-êB activation.16–18 In vitro RIP acti-vates apoptosis, NF-êB and JNK; however, thephysiological role of RIP was determined bytargeted disruption of its gene in mice.18 RIPdeficient mice appear normal at birth but beginto deteriorate by extensive apoptosis in boththe lymphoid and adipose tissues and die at1–3 days of age. Although TNF and Fas areable to activate apoptosis in RIP -/- cells, TNFfails to activate NF-êB. Thus, it appears thatRIP, but not TRAF2, is required for TNFinduced NF-êB activation.

The Fas associated death domain, or FADD(Mort1), was originally identified by its abilityto associate with the Fas death domain.19 20

Subsequently, a mutant version of FADDinhibited TNF, Fas, and DR3 induced apopto-sis, but not activation of NF-êB,21 22 suggestingthat the activation of NF-êB and apoptosis areseparable. The physiological role of FADD waselucidated in mice lacking FADD.22–24 TheFADD-/- mice did not survive past day 11.5 ofembryogenesis because of extensive abdominalhaemorrhage and cardiac failure. Further-more, FADD -/- mice are not susceptible toTNF, Fas, and DR3 induced apoptosis, but theapoptosis pathway induced by DR4 remainsintact. Thus, not only is FADD required to ini-tiate apoptosis by some death receptors, butalso FADD appears to be required for embry-onic development.

FADD-homologous ICE/CED-3-like pro-tease, or FLICE (MACH1), was originally dis-covered in a stimulated Fas complex25 and by ayeast two-hybrid screen using FADD as thebait.26 Based on its homology to other caspases,FLICE was later designated caspase 8. Uponligation, the TNF receptor recruits the deathdomain protein TRADD (TNF receptor asso-ciated death domain),27 which interacts withFADD and engages caspase 8 to initiate theapoptotic pathway. This signalling pathway wasverified in mice lacking caspase 8.28 Similar tothe FADD -/- mice, targeted disruption of cas-pase 8 in mice was lethal because of impairedheart muscle development and congestedaccumulation of erythrocytes. Although theability of TNF to activate NF-êB and JNK wasnot impaired, caspase 8 -/- mice exhibited adefect in activation of apoptosis by TNF, Fas,and DR3. Thus, of these known deathreceptors, all appear to require caspase 8 as theinitiating caspase leading to apoptosis.

Besides its apoptotic and inflammatoryresponses, TNF also generates other signallingmolecules including ceramide, which is a lipidsecond messenger.4 Ceramide is generatedfrom the lipid sphingosine by the activation ofneutral sphingomyelinase (N-SMase). To linkTNF receptor activation to sphingmyelinaseactivity, another protein was identified by ayeast two-hybrid screen and designated FAN,or factor associated with N-SMase activation.29

TNFR1 interacts with FAN through a smallregion N-terminal to the death domain.29 Toanalyse the physiological role of FAN in TNF

activation of N-SMase, FAN deficient micewere generated.30 FAN -/- mice are bornhealthy and exhibit no overt phenotypic abnor-malities, but the ability of TNF to activateN-SMase was impaired in FAN -/- mice.Signalling through TNFR1, TNF promotesskin permeability barrier repair involvingsphingomyelinase. As this repair process of thecutaneous barrier leads to the proliferation ofthe epidermis, FAN -/- mice have a reducedability to cause this repair process. Althoughthe lack of FAN does not appear to inhibitother TNF signalling pathways, FAN doesappear to be involved in the activation ofN-SMase by TNF.

New members of the TNF receptor familyThe TNF receptor family consists of 21 knownmembers, which are characterised by two tofour homologous cysteine rich repeats in theirextracelluar domain. Members of this receptorsuperfamily contain no significant homologywithin their intracellular domains, except forthose that possess a death domain. Despite nothaving intrinsic enzymatic activity, the TNFreceptor family recruits novel adaptor proteins,primarily death domain containing proteinsand proteins of the TRAF family. Some mem-bers of the TNF ligand superfamily bind morethan one receptor, as is the case for TRAIL,which binds five distinct receptors (that is,TRAIL R1-R4 and OPG) and LIGHT, whichbinds two receptors (that is, HVEM andLTâR). However, which receptor-ligand pairsare physiologically relevant remains to bedetermined. The previously described TNFreceptor family members (TNFRI, TNFR2,LTâR, Fas, NGFR, CD27, CD30, CD40,OX40, and 41BB) have been reviewedelsewhere.31 32 In this review we will introducethe recently discovered members of this recep-tor family (table 1). The TNF receptor familycan be divided into three groups: (1) those thatcontain a death domain, (2) those that do notcontain a death domain, and (3) those that lacka transmembrane domain, and thus are se-creted, soluble forms that may in fact inhibitcytokine signalling.

DEATH RECEPTOR 3 (DR3, LARD, WSL-1, TRAMP)Death receptor 3 was identified by a search forTNF receptors using the extracellular domain,the death domain homologous regions, and anEST database.22 Others identified this receptorand named it LARD,33 WSL-1,34 or TRAMP.35

DR3 encodes a protein of 417 amino acids witha death domain contained between residues335 and 413. The mRNA expression patternwas restricted to spleen, thymus, colon, intes-tine, prostate, and PBLs. Upon T cell activa-tion, a selective change in its alternativesplicing results in predominantly the mem-brane bound form, which may have implica-tions in lymphocyte proliferation afteractivation.33 The ligand for DR3 has now beendemonstrated to be TWEAK.36 Signal trans-duction by DR3 seems to use adaptor proteinssuch as TRADD, TRAF2, FADD, andFLICE.22 When overexpressed, DR3 activatesNF-êB, apoptosis, and JNK.22




j.com/

Ann R

heum D


ber 1999. Dow

nloaded from

http://ard.bmj.com/

TRAIL RECEPTORS (TRAIL R1- R4)

The first receptor for TRAIL was identifiedthrough a search of an EST database for TNFreceptor family members and was termeddeath receptor 4 (DR4),37 also known asTRAIL-R1,38 after which various laboratoriesidentified three more receptors for TRAIL,TRAIL-R2 (DR5, TRICK2, KILLER,Apo2),38–45 TRAIL-R3 (DcR1, TRID)38 40 46–48

and TRAIL-R4 (DcR2, TRUNDD, LIT).49–51

The TRAIL receptors have been extensivelyreviewed.8 52–59 These receptors are present in awide variety of normal tissues and in normaland tumour cell lines. Unlike TRAIL-R1 and-R2, TRAIL-R3 does not contain an intracellu-lar domain and TRAIL-R4 contains an incom-plete death domain. These data suggest thatTRAIL-R3 and -R4 could serve as decoyreceptors for TRAIL on the cell surface andprotect the cells from TRAIL inducedapoptosis.53 Consistent with this idea, overex-pression of either decoy receptor in TRAILsensitive cell lines protect them from TRAILinduced apoptosis. There are reports, however,suggesting that cells expressing R3 and R4 arestill susceptible to TRAIL induced apoptosis,(GriYths and Lynch57 and unpublished data),though there appears to be a direct correlationof the expression of the cytoplasmic caspaseinhibitor (FLIP) to the protection of TRAILinduced apoptosis. However, the results ofextensive studies by various laboratories havebeen somewhat contradictory. Some of theadaptor proteins that may be used by theTRAIL receptors include TRADD, FADD,caspase 8, caspase 10, and FLIPs. Further-more, there are contradictory reports onwhether TRAIL or its receptors activateNF-êB, perhaps because certain cell types mayor may not have the adaptor proteins necessaryfor TRAIL to induce NF-êB via the TRAIL

receptors. An additional decoy receptor forTRAIL, osteoprotegerin (OPG), a solublereceptor that binds RANKL (see below), bindsTRAIL at nanomolar concentrations andinhibits TRAIL induced apoptosis.60 However,OPG’s physiological role in TRAIL inducedapoptosis is not known. Thus, more informa-tion must be obtained before any conclusionscan be proposed for the signal transduction byTRAIL and its receptors.

DEATH RECEPTOR 6 (DR6)

To identify additional members of the TNFreceptor family, we searched an EST databasefor genes with homology to both the extracellu-lar domain and a consensus death domain. Weidentified a novel receptor gene and named itDR6.61 DR6 consists of 655 amino acids. Itsintracellular domain contains a death domainhomologous to other known death receptors,with maximum identity with TNFR1 (27.2%)and minimum identity with TRAIL-R2(19.7%). Unlike other death receptors, thedeath domain of DR6 is located proximal tothe transmembrane domain, but the signifi-cance of this diVerence is unclear. Curiously,following the death domain is a putativeleucine zipper sequence that overlaps a prolinerich domain, similar to an SH3 binding motif.Furthermore, two putative TRAF bindingmotifs are found near the transmembraneregion. The C-terminus contains a region pre-dicted to have alpha helical character. Whatrole these domains have in signalling by DR6remains to be determined.

The transcript for DR6 was expressed abun-dantly in brain, heart, placenta, pancreas,lymph node, thymus, and prostate and mini-mally expressed in liver and PBLs. Among celllines examined, non-lymphoid tumour cells(HeLa S3, SW480, A549, and G361) had thehighest expression of DR6; haematopoietic cell

Table 1 New members of the TNF receptor superfamily

Abbreviation Receptor name Alternative names Ligand

DR3 Death Receptor 3 LARD, WSL-1, TRAMP TWEAKLARD Lymphocyte-associated Receptor of Death DR3, WSL-1, TRAMP TWEAKWSL-1 DR3, LARD, TRAMP TWEAKTRAMP TNF Receptor-associated Apoptosis-mediating Protein DR3, LARD, WSL-1 TWEAKDR4 Death Receptor 4 TRAIL-R1 TRAILDR5 Death Receptor 5 TRAIL-R2, TRICK2, KILLER, Apo2 TRAILDcR1 Decoy Receptor 1 TRAIL-R3, TRID TRAILDcR2 Decoy Receptor 2 TRAIL-R4, TRUNDD, LIT TRAILTRAIL-R1 TRAIL Receptor 1 DR4 TRAILTRAIL-R2 TRAIL Receptor 2 DR5, TRICK2, KILLER, Apo2 TRAILTRAIL-R3 TRAIL Receptor 3 TRID, DcR1 TRAILTRAIL-R4 TRAIL Receptor 4 TRUNDD, DcR3, LIT TRAILTRICK2 TRAIL Receptor Inducer of Cell Killing DR5, TRAIL-R2, KILLER, Apo2 TRAILKILLER DR5, TRAIL-R2, TRICK2, Apo2 TRAILApo2 Apoptosis Receptor 2 DR5, TRAIL-R2, KILLER, TRICK2 TRAILTRID TRAIL Receptor Without Intracellular Domain TRAIL-R3, DcR1 TRAILTRUNDD TRAIL Receptor With a Truncated Death Domain TRAIL-R4, DcR2, LIT TRAILLIT Lymphocyte Inhibitor of TRAIL TRAIL-R4, TRUNDD, DcR2 TRAILRANK Receptor Activator of NF-kB TRANCE-R RANKLTRANCE-R TRANCE Receptor RANK RANKLOPG Osteoprotegerin FDCR-1, OCIF, TR1 TRAIL, RANKLFDCR-1 Folicular Dendritic Cell Receptor 1 OPG, OCIF, TR1 TRAIL, RANKLOCIF Osteoclast Inhibitor Factor OPG, TR1, FDCR-1 TRAIL, RANKLTR1 TNF Receptor-related Receptor 1 OPG, FDCR-1, OCIF TRAIL, RANKLDR6 Death Receptor 6 unknownDcR3 Decoy Receptor 3 TR6 FasL,LIGHTTR6 TNF receptor-related receptor 6 DcR3 FasL, LIGHTHVEM Herpesvirus Entry Mediator TR2, ATAR LIGHTTR2 TNF Receptor-related Receptor 2 ATAR, HVEM LIGHTATAR Another TRAF-associated Receptor HVEM, TR2 LIGHTGITR Glucocorticoid-induced TNF-like Receptor GITRLAITR Activation-induced TNF-like Receptor GITRL

I6 Darnay, Aggarwal



j.com/

Ann R

heum D


ber 1999. Dow

nloaded from

http://ard.bmj.com/

lines (HL-60, K562, Molt4, Raji) the lowest.As with other death receptors, overexpressionof DR6 induced apoptosis of HeLa cells butthis was not observed in MCF-7 cells, indicat-ing cell type specificity. As MCF-7 cells areknown to be quite sensitive to TNFR1, Fas,and DR4, it may be that the mechanism of cellkilling by DR6 diVers from that of other deathreceptors. The deletion of the death domainfrom DR6 abolished its ability to induce apop-tosis. In co-transfection and immunoprecipita-tion assays DR6 interacted weakly withTRADD and not at all with other deathdomain proteins, including FADD, RIP, andRAIDD. Thus, DR6 may activate apoptosis byassociating with other novel, unknown deathproteins.

Like most other death receptors, overexpres-sion of DR6 induced NF-êB activation; thiswas abolished when the death domain waseliminated, suggesting there is a commonadaptor molecule for NF-êB and apoptosis.Overexpression of DR6 also activated JNK;this response was not abolished on truncationof the death domain, indicating that JNK acti-vation is mediated by a cytoplasmic region dis-tinct from that activating NF-êB and apopto-sis. As TRAF molecules are capable ofactivating JNK and NF-êB and as DR6contains two potential TRAF binding motifs, itis possible that DR6 uses TRAF molecules toactivate NF-êB and JNK.

Unlike other death receptors (that is,TNFR1, Fas, DR3, DR4, and DR5), which areexpressed in most tissues and haematopoieticcells, DR6 is expressed only in cells ofnon-haematopoietic origin, suggesting that itsphysiological role may diVer. In addition, thedeath domain of DR6 is located proximal tothe transmembrane domain, rather than at theC-terminus of the receptor, and DR6 containsat least three more protein interaction motifsthan the other death receptors (leucine zipper,SH3, and a C-terminal helical domain) locatedC-terminal to the death domain, which sug-gests that DR6 may in fact activate othersignalling cascades. DR6 induces apoptosis in acell type specific manner and is a potent activa-tor of NF-êB and JNK. Finally, becauseTRADD interacts weakly with DR6, other,alternate signalling components may be usedby DR6.

DECOY RECEPTOR 3 (DCR3/TR6)

A search of the EST databases for other TNFreceptor related genes identified a novel mem-ber of the TNF receptor family, which wasnamed DcR362 or TR6.63 DcR3 encodes a pro-tein of 300 amino acids with a molecular massof approximately 40 kDa. Unlike other mem-bers of this receptor family, DcR3 does notcontain a transmembrane domain and thus issecreted as a soluble protein similar to OPG(see below). Its mRNA appears to be expressedin lung, brain, liver, spleen, and colon. TheDcR3 transcript was detected weakly in mosthaematopoietic cell lines and was inducedupon T cell activation. Interestingly, DcR3 wasconstitutively expressed in endothelialHUVEC cells. DcR3 binds to both LIGHT

and FasL and was able to inhibit apoptosis byboth of these cytokines. The expression of thissoluble decoy receptor may contribute toimmune system evasion by certain tumours.

RANK (RECEPTOR ACTIVATOR OF NF-êB)

A recently described TNF receptor familymember, RANK (for receptor activator ofNF-êB),64 bears high similarity in its extracel-lular domain to CD40. It consists of a616-amino acid transmembrane receptor, ofwhich 383 amino acids reside in the intracellu-lar domain, and does not appear to be homolo-gous to any other family member. RANKmRNA is ubiquitiously expressed in humantissues, but cell surface RANK is expressedonly on dendritic cells, the CD4+ T cell lineMP-1, foreskin fibroblasts, osteoclast progeni-tors, and activated B and T cells.64–66 However,its ligand RANKL (see below) appears to berestricted to activated B and T cells. RANKappears to use the TRAF family of signaltransducers to activate NF-êB and JNKpathways.11 67–70 Furthermore, a novel TRAF6interaction motif was identified and shown tobe required for activation of NF-êB.11 More-over, transgenic mice expressing a soluble formof RANK have severe osteopetrosis because ofa reduction in bone resorbing osteoclasts,66

similar to OPG transgenic mice (see below).The observations that RANK interacts withTRAF6 and that TRAF6 deficient mice exhibitan osteopetrotic phenotype because of a defectin bone resorption71 suggest a direct involve-ment of RANK and its ligand in osteoclas-togenesis. Thus, how each of these TRAF mol-ecules regulates RANK/RANKL signaltransduction pathways resulting in osteoclastdiVerentiation and B and T cell modulationremains to be determined.

OSTEOPROTEGERIN (OPG/OCIF/TR1/FDCR-1)

OPG was first identified by sequence homol-ogy as a possible novel TNF receptor familymember during a rat intestine cDNA sequenc-ing project.72 OPG was also identified byvarious other laboratories and named OCIF,73

TR1,74 and FDCR-1.25 OPG binds not onlyRANKL,76 but also TRAIL.60 Unlike the otherTNF receptor family members, OPG, a 401amino acid protein, does not contain atransmembrane domain and thus is secreted asa soluble receptor. Its mRNA is expressed inheart, placenta, lung, liver, bone marrow,spleen, lymph node, and kidney, and at lowerlevels in the thymus, prostate, testis, ovary, andsmall intestine. Initially expressed as a 55 kDaprotein, OPG is converted to a disulphidelinked dimer of approximately 110 kDa and issecreted into the medium.72 Others haveconfirmed that OPG is membrane associated,most likely through association with the extra-cellular matrix.75 In its carboxy terminus, OPGcontains a homologous death domain that,when expressed as a transmembrane form,activates apoptosis.77 The main physiologicalfeature of OPG appears to inhibit RANKLfrom binding to osteoclast progenitors, andthus inhibits osteoclastogenesis.72–74 Consistentwith inhibition of osteoclastic cellular function,




j.com/

Ann R

heum D


ber 1999. Dow

nloaded from

http://ard.bmj.com/

TGF-â1 upregulated OPG mRNA whilesuppressing RANKL in murine bone marrowcultures.78 Moreover, OPG deficient miceexhibit an early onset of osteoporosis.79 Theunique ability of OPG to increase bone masshas resulted in a potential treatment forosteoporosis, which is entering phase I clinicaltrials in post-menopausal women.

HERPES VIRUS ENTRY MEDIATOR (HVEM/TR2/ATAR)

A novel TNF receptor family was identified bysearching an EST database for a TNF relatedreceptor protein and was termed HVEM,80

ATAR81 and TR2.82 This receptor was alsoidentified through a screen for receptors thatwould enable entry of herpes simplex virus-1into cells.83 This receptor encodes a protein of283 amino acids, whose mRNA expression isrestricted to spleen, thymus, bone marrow,lung, small intestine, PBLs, and kidney. Theligand for this receptor was recently identifiedas LIGHT84 or HVEM-L.85 The cytoplasmicdomain is much shorter than in other membersof this family. It uses TRAF1, TRAF2,TRAF3, and TRAF580–82 to activate NF-êBand JNK signalling pathways. Furthermore, aTR2-Fc fusion protein inhibited a mixed lym-phocyte reaction mediated proliferation, sug-gesting that this receptor and its ligand mayparticipate in T cell stimulation.

GLUCOCORTICOID INDUCED TNFR FAMILY

RELATED GENE (GITR/AITR)

GITR, also known as AITR,86 was identified bysearching an EST database for homologues tothe TNF receptor family.87 Initially, a murineGITR was identified by comparing untreatedand dexamethasone treated murine T cellhybridoma by using the diVerential displaytechnique.88 GITR encodes a protein of 241amino acids with a molecular mass of approxi-mately 26 kDa. The expression of GITRmRNA was highest in lymph node, PBLs, bonemarrow, thymus, lung, and spleen, and rela-tively low in other tissues. Like most TNFreceptor members, GITR was upregulated inPBMC by antigen stimulation or lymphocyteactivation. The ligand for GITR was identifiedas GITRL.86 87 GITR was shown to interactwith TRAF1, TRAF2, and TRAF3, and GITRinduced NF-êB activation appears to requireTRAF2 and NIK. Furthermore, expression of

GITR and its ligand in Jurkat T cells inhibitedantigen receptor induced apoptosis, suggestingGITR may modulate T lymphocyte survival.Unlike the mouse homologue, human GITRwas not induced by dexamethasone in periph-eral blood T cells.86 87 The observations thatthis receptor activates NF-êB and protectsagainst activation induced cell death suggeststhat GITR and its ligand may participate in Tlymphocyte survival in peripheral tissues andperhaps during interaction with the vascularendothelium.

Novel members of the TNF familyThe TNF family consists of 17 knownmembers. All members have a similar coresequence that is predicted to contain all 10â-sheet forming sequences characteristic ofTNF. This TNF-like core domain and the ESTdatabases have led to the identification of newTNF related ligands. The previously describedTNF family members (TNF, LT, FasL, NGF,CD27L, CD30L, CD40L, OX40L, and41BBL) have been reviewed elsewhere.31 32 Inthis review we will introduce the recently iden-tified members of this family (table 2).

TRAIL (TNF RELATED APOPTOSIS INDUCING

LIGAND)

One of the first TNF related ligands that wasidentified was named TRAIL89 or Apo2L.90

TRAIL is a ubiquitous type II transmembraneprotein of 281 amino acids. It can be cleavedfrom the membrane by a protease to yield asoluble protein. TRAIL specifically interactswith four membrane bound receptors knownas TRAIL R1-R4 (see above) and the solublereceptor OPG, which can inhibit TRAILinduced apoptosis.60

TRAIL appears to cause apoptosis in a vari-ety of cell types without aVecting normal (non-transformed) cells. In T cells stimulated withPMA, ionomycin, anti-CD3, interferon á, IL2,or IL15 expression of TRAIL is upregu-lated.91 92 Furthermore, TRAIL is upregulatedupon IFN á or ã stimulation of monocytes,which then acquire the ability to kill tumourcells.93 Others have demonstrated the ability ofTRAIL to induce apoptosis in humanmelanoma cells through caspase 8 and 3,94 inmelanoma cells that were resistant to FasLinduced cell killing,95 and in phenotypically

Table 2 New members of the TNF ligand superfamily

Abbreviation Ligand Name Alternative Names Receptor

THANK TNF Homologue that Activates NF-êB and JNK TALL1, BAFF unknownTALL1 TNF and ApoL-related Leucocyte-expressed Ligand 1 BAFF, THANK unknownBAFF B cell activating factor belonging to the TNF family TALL1, THANK unknownApo2L Apoptosis 2 Ligand TRAIL TRAIL-R1-R4, OPGTRAIL TNF-related Apoptosis-Inducing Ligand Apo2L TRAIL-R1-R4, OPGTWEAK TNF Relatedness and Weak Inducer of Apoptosis Apo3L DR3Apo3L Apoptosis 3 Ligand TWEAK DR3VEGI Vascular Endothelial Growth Inhibitor unknownRANKL Receptor activator of NF-êB Ligand OPGL, TRANCE, ODF RANKOPGL Osteoprotegerin Ligand RANKL, TRANCE, ODF RANKTRANCE TNF-related Activation-induced Cytokine RANKL, OPGL, ODF RANKODF Osteoclast DiVerentiation Factor RANKL, OPGL, TRANCE RANKLIGHT TL1, HVEML HVEM, LTâRTL1 TNF-related Ligand 1 HVEML, LIGHT HVEM, LTâRHVEML Herpesvirus Entry Mediator Ligand LIGHT, TL1 HVEM, LTâRAPRIL A Proliferation-inducing Ligand TALL2 unknownTALL2 TNF and ApoL-related Leucocyte-expressed Ligand 2 APRIL unknownGITRL Glucocorticoid-induced TNF-related Receptor Ligand GITR

I8 Darnay, Aggarwal



j.com/

Ann R

heum D


ber 1999. Dow

nloaded from

http://ard.bmj.com/

immature CD161+/CD56- NK cells.96 Mostremarkably, TRAIL when administered sys-temically caused tumoricidal activity of themammary adenocarcinoma cell line MDA-231in mice without causing toxic side eVects.97

TRAIL’s ability to selectively kill transformedand not normal cells and its inability to activatethe NF-êB pathway suggest that TRAIL maybe a powerful treatment for cancer.

APRIL (A PROLIFERATION INDUCING LIGAND)

APRIL was discovered by screening a publicdatabase using a profile search based on anoptimal alignment of all the currently knownTNF ligand family members.98 An identicalmolecule was identified by a similar search andnamed TALL2.99 The cDNA clone encoded atype II transmembrane protein of 250 aminoacids, which contained 28 amino acids in thecytoplasmic domain, 21 amino acids in thetransmembrane domain, and 201 amino acidsin the extracellular domain. The sequence ofAPRIL showed highest similarity in its extra-cellular domain with FasL (21%), TNFá(20%), and LTâ (18%). Expression of itsmRNA revealed that APRIL was weaklyexpressed and restricted to a few tissues, mostnotably prostate, colon, spleen, pancreas, andPBLs. Interestingly, APRIL was expressed invarious tumour cell lines including HL60,HeLa S3, K562, Molt-4, Raji, SW-480, A549,and G361. Remarkably, APRIL mRNA wasincreased in thyroid carcinoma and in lym-phoma, but in the corresponding normaltissue the expression was either weak orabsent.

APRIL’s expression in tumour derivedtissues, but not normal tissue suggested thatAPRIL may serve in tumour growth prolifera-tion. Indeed, recombinant APRIL caused pro-liferation in Jurkat T lymphoma cells, in someB cell lymphomas (that is, Raji, mouse A20),and in some cell lines of epithelial origin suchas COS, HeLa, and some melanomas. FurtherNIH-3T3 cells engineered to express APRILincreased tumour growth rate in nude mice ascompared with NIH-3T3 cells expressing noligand. The mechanism by which APRILinduces cellular proliferation is not known, butit does not appear to activate NF-êB or JNK.The APRIL receptor has not yet beenidentified, but it does not appear to be any ofthe known members of the TNF receptorfamily. The little information we do haveabout APRIL and its expression in tumourcells (compared with normal tissue) suggeststhat APRIL may play a part in tumorigenesis.Thus, antagonistic antibodies to APRIL or itsreceptor may have a potential for therapeuticintervention.

TWEAK (TNF RELATEDNESS AND WEAK INDUCER

OF APOPTOSIS)

TWEAK was first identified as a clone thatweakly hybridised to an erythropoietin probewhose primary sequence was similar to ligandsof the TNF family.100 An identical moleculewas identified through a screen of an ESTdatabase by its homology to TNF family mem-bers and was named Apo3L.36 TWEAK is a

249 amino acid type II transmembrane proteinwhose mRNA is expressed in essentially all tis-sues examined. Soluble recombinant TWEAKcaused IL8 secretion in HT29, A375, WI-38,and A549 cells.100 Additionally, TWEAKcaused weak induction of apoptosis in HT29cells when cultured with IFNã.100 In contrast,others have shown that TWEAK activatesapoptosis strongly in MCF-7 cells, the activa-tion being dependent on FADD and caspaseactivation.36 TWEAK specifically interacts withthe death receptor, DR3.36 The activation ofNF-êB by TWEAK was also demonstrated tobe TRAF2, TRADD, RIP, and NIK depend-ent.36 TWEAK induces proliferation in avariety of normal endothelial cells and in aorticsmooth muscle cells and reduces culturerequirements of serum and growth factors.101

TWEAK induces a strong angiogenic responsewhen implanted in rat corneas, suggesting aphysiological role for TWEAK in vasculatureformation in vivo.101

VEGI (VASCULAR ENDOTHELIAL GROWTH

INHIBITOR)

To identify an autocrine inhibitor of angiogen-esis specific to endothelial cells, a cDNAlibrary was constructed from RNA derivedfrom various endothelial cells. A search forTNF homologues in this EST database showeda type II transmembrane protein of 174 aminoacids with 20–30% homology to TNF familymembers. As the new protein was subsequentlyfound to be able to inhibit endothelial cellgrowth, it was designated VEGI.102 Unlikeother members of the TNF family, VEGI isexpressed predominantly in endothelial cells.Local production of a secreted form of VEGIvia gene transfer caused complete suppressionof the growth of MC-38 murine colon cancersin syngeneic C57BL/6 mice. Histologicalexamination showed marked reduction ofvascularisation in MC-38 tumours that ex-pressed soluble but not membrane boundVEGI or were transfected with control vector.The conditioned media from soluble VEGIexpressing cells showed marked inhibitoryeVect on in vitro proliferation of adult bovineaortic endothelial cells. VEGI is a novel angio-genesis inhibitor of the TNF family andfunctions in part by directly inhibiting en-dothelial cell proliferation, suggesting thatVEGI may be highly valuable in angiogenesisbased cancer therapy.

RANKL (RECEPTOR ACTIVATOR OF NF-êB LIGAND)Human RANK ligand (also known as OPGL,TRANCE, ODF) is a type II transmembraneprotein with an approximate molecular mass of45 kDa and is expressed primarily on activatedT and B cells and osteoclast progeni-tors.64 76 73 103 A recent review is available.104

Like other ligands of the TNF superfamily,RANKL has been demonstrated to activateNF-êB64 and JNK.103 Furthermore, stimulationof dendritic cells with RANKL up regulates theexpression of the anti-apoptotic protein Bcl-XL, suggesting a potential role for RANK/RANKL in dendritic cell survival.105 RANKLwas also demonstrated to be cleaved from the




j.com/

Ann R

heum D


ber 1999. Dow

nloaded from

http://ard.bmj.com/

cell surface by the TNF converting enzyme.106

Moreover, RANKL has been demonstrated toplay an essential part in osteoclast diVerentia-tion and activation.66 76 73 107 108 Targeted disrup-tion of RANKL in mice resulted in theessential requirement for RANKL to induceosteoclastogenesis. Additionally, RANKL defi-cient mice had poor lymphocyte developmentand lymph node organogenesis.109 A similarphenotype was also observed in TRAF6deficient mice.71 Moreover, in rheumatoidarthritis (RA) patients, IL17 in synovial fluidsupregulated RANKL.110 Concentrations ofIL17 in synovial fluids were significantly higherin RA patients than in osteoarthritis patients.Anti–IL17 antibody significantly inhibited os-teoclast formation induced by conditionedmedia from RA synovial tissues. These findingssuggest that IL17 first acts on osteoblasts, pro-ducing a mediator that stimulates both COX-2dependent PGE2 synthesis and RANKL geneexpression, which in turn induce diVerentia-tion of osteoclast progenitors into matureosteoclasts. They also suggest that IL17 is acrucial cytokine for osteoclastic bone resorp-tion in RA patients.

THANK (A TNF HOMOLOGUE THAT ACTIVATES

APOPTOSIS, NF-êB, AND JNK)

By using an amino acid sequence motif of TNFand searching an EST database, a novel TNFhomologue encoding 285 amino acids wasidentified and named THANK.111 The pre-dicted extracellular domain of THANK is 15,16, 18, and 19% identical to LIGHT, FasL,TNF, and LTa, respectively. Northern blotanalysis of its mRNA indicated expression inPBLs, spleen, thymus, lung, placenta, smallintestine, and pancreas. THANK mRNAexpression was highest in HL60 followed byK562, A549, and G361, but there was noexpression in HeLa, Molt-4, Raji, and SW-480.Recombinant THANK protein activatedNF-êB and JNK in the promyeloid cell lineU937. Additionally, THANK induced activa-tion of apoptosis in U937 cells. The receptorfor THANK is at present unknown, butTHANK does not bind TNFR1 or TNFR2.Identical molecules to THANK were identifiedand named TALL199 and BAFF.112

LIGHT

An additional member of the TNF family,named LIGHT, was identified by searching anEST database for sequence similarity to TNFfamily members.84 113 An identical molecule wasidentified by its interaction with HVEM anddesignated HVEM-L.85 LIGHT mRNA ishighly expressed in splenocytes, activated PBLs,CD8+ tumour infiltrating lymphocytes, granu-locytes, and monocytes, but it is not expressed inthe thymus or in tumour cells. Additionally,LIGHT is upregulated in CD4+ and CD8+ Tcells when exposed to PMA. LIGHT encodes atype II transmembrane protein of 240 aminoacids. LIGHT binds not only to HVEM, butalso to the LTâ receptor. A soluble, secretedform of LIGHT stimulates proliferation of Tlymphocytes during allogeneic responses, inhib-

its HT-29 cell growth, and weakly stimulatesNF-êB dependent transcription.85

The MDA-MB-231 human breast carci-noma transected with LIGHT caused com-plete tumour suppression in mice. Histologicalexamination showed marked neutrophil infil-tration and necrosis.84 IFNã dramatically in-creases LIGHT mediated apoptosis, andLIGHT induces apoptosis of various tumourcells that express both LTâ and HVEM recep-tors. However, LIGHT was not cytolytic to thetumour cells that express only the LTâR orHVEM or haematopoietic cells that expressonly the HVEM, such as PBLs, Jurkat cells, orCD8+ TIL cells. In contrast, treatment of theactivated PBLs with LIGHT resulted in releaseof IFNã. Taken together, LIGHT triggers dis-tinct biological responses based on the expres-sion patterns of its receptors on the target cells.Thus, LIGHT may play a part in the immunemodulation and have a potential value incancer therapy.

GITRL (GLUCOCORTICOID INDUCED TNFR FAMILY

RELATED LIGAND)

The ligand for GITR was identified by a yeastbased signal sequence trap method from aHUVEC cDNA library.87 This ligand was alsoidentified in a EST database search for TNFrelated ligands,86 GITRL encodes a 177 aminoacid type II transmembrane protein with a cal-culated mass of 20 kDa. Analysis of its mRNArevealed highest expression in small intestine,ovary, testis, and kidney, and lower to noexpression in other tissues. Expression ofmembrane-bound GITRL was detected oncultured HUVEC.87 Expression of eitherGITRL or its receptor or both the ligand andreceptor in Jurkat cells inhibited activationinduced cell death.87 Consistent with theinhibition of apoptosis, GITRL activatedthe proapoptotic transcription factorNF-êB.86 87 Thus, GITRL may modulateperipheral T cell interaction with blood vesselsin the periphery.

Conclusions and future perspectivesAs new members of the TNF ligand andreceptor superfamilies are being discovered,one interesting characteristic seems to be com-mon, that most of these ligands have the abilityto activate the transcription factor NF-êB. Asthis factor is one of the primary modulators ofthe inflammatory process, it would not be sur-prising to find more than one of these cytokinesinvolved in RA and other types of inflammatorydiseases. As most of these ligands seem to besynthesised by cells of the immune system, itwill be most important to understand how eachof these cytokines act under physiological con-ditions. For instance, RANKL and its solublereceptor OPG, whose cDNAs were just de-scribed 18 months ago, are essential for osteo-clastogenesis. Uncovering the physiologicalrole for RANKL and OPG has recently led tothe initiation of phase I clinical trials for thetreatment of osteoporosis. With the availabilityof antibodies to these new ligands and theirrecombinant proteins, we are poised to investi-gate their physiological roles in the immune

I10 Darnay, Aggarwal



j.com/

Ann R

heum D


ber 1999. Dow

nloaded from

http://ard.bmj.com/

system and in various diseases such as RA. Forexample, a recent report demonstrated that theincreased levels of IL17 in synovial fluid fromRA patients caused an increase in osteoclas-togenesis.110 This was most probably attribut-able to the increased expression of a novelmember of the TNF family, RANKL, which isrequired for osteoclastogenesis, suggesting thatIL17 present in synovial tissues and fluids fromRA patients may be involved in the jointdestruction associated with this disease. Thus,with the identification of the signalling path-ways and the physiological roles associatedwith these new ligands, it may be possible todevelop new therapeutic approaches to combatvarious inflammatory diseases and cancer.

Funding: this research was supported by the Clayton Founda-tion for Research.

1 Arend WP, Dayer JM. Inhibition of the production andeVects of interleukin-1 and tumor necrosis factor (alpha) inrheumatoid arthritis. Arthritis Rheum 1995;38:151–60.

2 Moreland LW, Baumgartner SW, SchiV MH, et al.Treatment of rheumatoid arthritis with a recombinanthuman tumor necrosis factor receptor (p75)-Fc fusionprotein. N Engl J Med 1997;337:141–7.

3 May MJ, Gosh S. Signal transduction through NF-kB.Immunol Today 1998;19:80−8.

4 Wallach D, Varfolomeev EE, Malinin NL, Goltsev YV,Kovalenko AV, Boldin MP. Tumor necrosis factor receptorand Fas signaling mechanisms. Annu Rev Immunol 1999;17:331–67.

5 SchaeVer HJ, Weber MJ. Mitogen-activated protien kinases:specific messages from ubiquitous messengers. Mol CellBiol 1999;19:2435–44.

6 Salvesen GS, Dixit VM. Caspases: intracellular signaling byproteolysis. Cell 1997;91:443–6.

7 Thornberry NA, Lazebnik Y. Caspases: enemies within.Science 1998;281:1312–16.

8 Singh A, Ni J, Aggarwal BB. Death domain receptors andtheir role in cell demise. J Interferon Cytok Res1998;18:439–50.

9 Huang B, Eberstadt M, Olejniczak ET, Meadows RP, FesikSW. NMR structure and mutagenesis of the Fas (APO-1/CD95) death domain. Nature 1996;384:638–41.

10 Arch RH, Gedrich RW, Thompson CB. Tumor necrosisfactor receptor-associated factors (TRAFs)–a family ofadaptor proteins that regulates life and death. Genes Dev1998;12:2821–30.

11 Darnay BG, Ni J, Moore PA, Aggarwal BB. Activation ofNF-kB by RANK requires TRAF6, and NF-kB-inducingkinase (NIK): identification of a novel TRAF6 interactionmotif. J Biol Chem 1999;274:7724–31.

12 Pullen SS, Dang TTA, Crute JJ, Kehry MR. CD40 signalingthrough tumor necrosis factor receptor-associated factors(TRAFs): binding site specificity and activation ofdownstream pathways by distinct TRAFs. J Biol Chem1999;274:14246–54.

13 Karin M, Delhase M. JNK or IKK, AP-1 or NF-kB, whichare the targets for MEK kinase 1 action? Proc Natl AcadSci 1998;95:9067–9.

14 Lee S Y, Reichlin A, Santana A, Sokol KA, NussenzweigMC, Choi Y. TRAF2 is essential for JNK but not NF-kBactivation and regulates lymphocyte proliferation andsurvival. Immunity 1997;7:703–13.

15 Yeh W-C, Shahinian A, Speiser D, et al. Early lethality, fuc-ntional NF-kB activation, and increased sensitivity toTNF-induced cell death in TRAF2-deficient mice. Immu-nity 1997;7:715–25.

16 Ting AT, Pimentel-Muiños FX, Seed B. RIP mediatestumor necrosis factor receptor 1 activation of NF-B but notFas/APO-1-initiated apoptosis. EMBO J 1996;15:6189–96.

17 Stanger BZ, Leder P, Lee T-H, Kim E, Seed B. RIP: A novelprotein containing a death domain that interacts with Fas/APO1 (CD95) in yeast and causes cell death. Cell1995;81:513–23.

18 Kelliher MA, Grimm S, Ishida Y, Kuo F, Stanger BZ, LederP. The death domain kinase RIP mediates the TNF-induced NF-kB signal. Immunity 1998;8:297–303.

19 Boldin MP, Varfolomeev EE, Pancer Z, Mett IL, CamonisJH, Wallach D. A novel protein that interacts with the deathdomain of Fas/APO1 contains a sequence motif related tothe death domain. J Biol Chem 1995;270:7795–8.

20 Chinnaiyan AM, O’Rourke K, Tewari M, Dixit VM. FADD,a novel death domain-containing protein, interacts with thedeath domain of Fas and initiates apoptosis. Cell 1995;81:505–12.

21 Chinnaiyan AM, Tepper CG, Seldin MF, et al. FADD/MORT1 is a common mediator of CD95 (Fas/APO-1) andtumor necrosis factor receptor-induced apoptosis. J BiolChem 1996;271:4961–5.

22 Chinnaiyan AM, O’Rourke K, Yu G-L, et al. Signaltransduction by DR3, a death domain-containing receptorrelated to TNFR-1 and CD95. Science 1996;274: 990–2.

23 Yeh W-C, de la Pompa JL, McCurrach ME, et al. FADD:essential for embryo development and signaling from some,but not all, inducers of apoptosis. Science 1998;279:1954–8.

24 Zhang J, Cado D, Chen A, Kabra NH, Winoto A.Fas-mediated apoptosis and activation-induced T-cell pro-liferation are defective in mice lacking FADD/Mort1.Nature 1998;392:296–300.

25 Muzio M, Chinnaiyan AM, Kischkel FC, et al. FLICE, anovel FADD-homologous ICE/CED-3-like protease, isrecruited to the CD95 (Fas/APO-1) death-inducing signal-ing complex. Cell 1996;85:817–27.

26 Boldin MP, Goncharov TM, Goltsev Y V, Wallach D.Involvement of MACH, a novel MORT1/FADD-interacting protease, in Fas/APO-1- and TNF receptor-induced cell death. Cell 1996;85:803–15.

27 Hsu H, Xiong J, Goeddel DV. The TNF receptor1-associated protein TRADD signals cell death and NF-kBactivation. Cell 1995;81:495–504.

28 Varfolomeev EE, Schuchmann M, Luria V, et al.Targeteddisruption of the mouse caspase 8 gene ablates cell deathinduction by the TNF receptors, Fas/Apo1, and DR3 andis lethal prenatally. Immunity 1998;9:267–76.

29 Adam-Klages S, Adam D, Weigmann K, et al. FAN, a novelWD-repeat protein, couples the p55 TNF-receptor to neu-tral sphingomyelinase. Cell 1996;86:937–47.

30 Kreder D, Krut O, Adam-Klages S, et al. Impaired neutralsphingomyelinase activation and cutaneous barrier repairin FAN-deficient mice. EMBO J 1999;18:2472–9.

31 Wallach D, Bigda J, Engelmann H. The TNF family andrelated molecules. In: Theze J, ed. The cytokine network andimmune functions. London: Oxford University Press. (inpress).

32 Aggarwal BB, Natarajan K. Tumor necrosis factors:developments during the last decade. Eur Cytokine Netw1996;7:93–124.

33 Screaton GR, Xu X-N, Olsen AL, et al. LARD: A newlymphoid-specific death domain containing receptor regu-lated by alternative pre-mRNA splicing. Proc Natl Acad SciUSA 1997;94:4615–19.

34 Kitson J, Raven T, Jiang Y-P, et al. A death-domain-containing receptor that mediates apoptosis. Nature 1996;384:372–5.

35 Bodmer J-L, Burns K, Schneider P, et al. TRAMP, a novelapoptosis-mediating receptor with sequence homology totumor necrosis factor receptor 1 and Fas (Apo-1/CD95).Immunity 1997;6:79–88.

36 Marsters SA, Sheridan JP, Pitti RM, Brush J, Goddard A,Ashkenazi A. Identification of a ligand for the death-domain-containing receptor Apo3. Curr Biol 1998;8:525–8.

37 Pan G, O’Rourke K, Chinnaiyan AM, et al. The receptor ofthe cytotoxic ligand TRAIL. Science 1997;276:111–13.

38 Schneider P, Bodmer JL, Thome M, Hofmann K, Holler N,Tschopp J. Characterization of two receptors for TRAIL.FEBS Lett 1997;416:329–34.

39 Pitti RM, Marsters SA, Ruppert S, Donahue CJ, Moore A,Ashkenazi A Induction of apoptosis by Apo-2 ligand, a newmember of the tumor necrosis factor cytokine family. J BiolChem 1996;271:12687–90.

40 Sheridan JP, Marsters SA, Pitti RM, et al. Control ofTRAIL-induced apoptosis by a family of signaling anddecoy receptors. Science 1997;277:818–21.

41 Schneider P, Thome M, Burns K, et al. TRAIL receptors 1(DR4) and 2 (DR5) signal FADD-dependent apoptosisand activate NF-kB. Immunity 1997;7:831–6.

42 Macfarlane M, Ahmad M, Srinivasula SM, Fernandes-Alnemri T, Cohen GM, Alnemri ES. Identification andmolecular cloning of two novel receptors for the cytotoxicligand TRAIL. J Biol Chem 1997;272:25417–20.

43 Chaudhary PM, Eby M, Jasmin A, Bookwalter A, Murray J,Hood L. Death receptor 5, a new member of the TNFRfamily, and DR4 induce FADD-dependent apoptosis andactivate NF-kB pathway. Immunity 1997;7:821–30.

44 Screaton GR, Mongkolsapaya J, Xu X-N, Cowper AE,Mcmichael AJ, Bell JI. TRICK2, a new alternatively splicedreceptor that transduces the cytotoxic signal from TRAIL.Curr Biol 1997;7:693–6.

45 Wu GS, Burns TF, McDonald ER, et al. KILLER/DR5 is aDNA damage-inducible p53-regulated death receptorgene. Nat Genet 1997;17:141–3.

46 Degli-Esposti MA, Smolak PJ, Walczak H, et al. Cloning andcharacterization of TRAIL-R3, a novel member of theemerging TRAIL receptor family. J Exp Med 1997;186:1165–70.

47 Pan G, Ni J, Wei Y-F, Yu G-L, Gentz R, Dixit VM. Anantagonist decoy receptor and a death domain-containingreceptor for TRAIL. Science 1997;277:815–17.

48 Mongkolsapaya J, Cowper AE, Xu X-N, et al. Lymphocyteinhibitor of TRAIL (TNF-related apoptosis-inducingligand): a new receptor protecting lymphocytes from thedeath ligand TRAIL. J Immunol 1998;160:3–6.

49 Degli-Esposti MA, Dougall WC, Smolak PJ, Waugh JY,Smith CA, Goodwin RG. The novel receptor TRAIL-R4induces NF-kB and protects against TRAIL-mediatedapoptosis, yet retains an incomplete death domain. Immu-nity 1997;7:813–20.

50 Marsters SA, Sheridan JP, Pitti RM, et al. A novel receptorfor Apo2L/TRAIL contains a truncated death domain.Curr Biol 1997;7:1003–6.

51 Pan G, Ni J, Yu GL, Wei Y-F, Dixit VM. TRUNDD, a newmember of the TRAIL receptor family that antagonizesTRAIL signalling. FEBS Lett 1998;424:41–5.




j.com/

Ann R

heum D


ber 1999. Dow

nloaded from

http://ard.bmj.com/

52 French LE, Tschopp J. The TRAIL to selective tumordeath. Nat Med 1999;5:146–7.

53 Ashkenazi A, Dixit VM. Death receptors: Signaling andmodulation. Science 1998;281:1305–8.

54 Degli-Esposti M. To die or not to die-the quest of theTRAIL receptors. J Leuk Biol 1999;65:535–42.

55 Darnay BG, Aggarwal BB. Early events in TNF signaling: Astory of associations and dissociations. J Leuk Biol1997;61:559–66.

56 Muzio M. Signaling by proteolysis: death receptors induceapoptosis. Int J Clin Lab Res 1998;28:141–7.

57 GriYth TS, Lynch DH. TRAIL: A molecule with multiplereceptors and control mechanisms. Curr Opin Immunol1998;10:559–63.

58 Schulze-OsthoV K, Ferrari D, Los M, Wesselborg S, PeterME. Apoptosis signaling by death receptors. Eur J Biochem1998;254:439–59.

59 Goldstein P. Cell death: TRAIL and its receptors. Curr Biol1997;7:R750–3.

60 Emery JG, McDonnell P, Brigham Burke M, et al.Osteoprotegerin is a receptor for the cytotoxic ligandTRAIL J Biol Chem 1998;273:14363–7.

61 Pan G, Bauer JH, Haridas V, et al. Identification andfunctional characterization of DR6, a novel death domain-containing TNF receptor FEBS Lett 1998;434:351–6.

62 Pitti RM, Marsters SA, Lawrence DA, et al. Geneomicamplification of a decoy receptor for Fas ligand in lung andcolon cancer. Nature 1998;396:699–703.

63 Yu K-Y, Kwon B, Ni J, Zhai Y, Ebners, Kwon BS. A newlyidentified member of the tumor necrosis factor receptorsuperfamily (TR6) suppresses LIGHT-mediated apoptosis.J Biol Chem 1999;274:13733–6.

64 Anderson DM, Maraskovsky E, Billingsley WL, et al. Ahomologue of the TNF receptor and its ligand enhanceT-cell growth and dendritic-cell function. Nature 1997:390:175–9.

65 Josien R, Wong BR, Li H-L, Steinman RM, Choi, Y.TRANCE, a TNF family member, is diVerentiallyexpressed on T cell subsets and induces cytokineproduction in dendritic cells. J Immunl 1999;162:2562–8.

66 Hsu H, Lacey DL, Dunstan CR, et al. Tumor necrosis fac-tor receptor family member RANK mediates osteoclastdiVerentiation and activation induced by osteoprotegerinligand Proc Natl Acad Sci USA 1999;96:3540–5.

67 Darnay BG, Haridas V, Ni J, Moore PA, Aggarwal BB.Characterization of the intracellular domain of receptoractivator of NF-kB (RANK): interaction with tumornecrosis factor receptor-associated factors and activation ofNF-kB and c-jun N-terminal kinase. J Biol Chem1998;273:20551–5.

68 Wong BR, Josien R, Lee SY, Vologodskaia M, SteinmanRM, Choi Y. The TRAF family of signal transducers medi-ates NF-kB activation by the TRANCE receptor. J BiolChem 1998;273:28355–9.

69 Kim H-H, Lee DE, Shin JN, et al. Receptor activator ofNF-kB recruits multiple TRAF family adaptors andactivates c-Jun N-terminal kinase. FEBS Lett 1999;443:297–302.

70 Galibert L, Tometsko ME, Anderson DM, Cosman D,Dougall WC. The involvement of multiple tumor necrosisfactor receptor (TNFR)-associated factors in the signalingmechanisms of receptor activator of NF-kB, a member ofthe TNFR superfamily. J Biol Chem 1998;273:34120–7.

71 Lomaga MA, Yeh W-C, Sarosi I, et al. TRAF6 deficiencyresults in osteopetrosis and defective interleukin-1, CD40,and LPS signaling. Genes Dev 1999;13:1015–24.

72 Simonet WS, Lacey DL, Dunstan CR, et al.Osteoprotegerin: a novel secreted protein involved in theregulation of bone density. Cell 1997;89:309–19.

73 Yasuda H, Shima N, Nakagawa N, et al. Osteoclast diVeren-tiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical toTRANCE/RANKL. Proc Natl Acad Sci USA 1998;95:3597–602.

74 Kwon BS, Wang S, Udagawa N, et al. TR1, a new memberof the tumor necrosis factor receptor superfamily, inducesfibroblast proliferation and inhibits osteoclastogenesis andbone resorption. FASEB J 1998;12:845–54.

75 Yun TJ, Chaudhary PM, Shu GL, et al. OPG/FDCR-1, aTNF receptor family member, is expressed in lymphoidcells and is up-regulated by ligating CD40. J Immunol1998;161:6113–21.

76 Lacey DL, Timms E, Tan H-L, et al. Osteoprotegerin ligandis a cytokine that regulates osteoclast diVerentiation andactivation. Cell 1998;93:165–76.

77 Yamaguchi K, Kinosaki M, Goto M, et al. Characterizationof structural domains of human osteoclastogenesis inhibi-tory factor. J Biol Chem 1998;273:5117–23.

78 Takai H, Kanematsu M, Yano K, et al. Transforming growthfactor -B stimulates the production of osteoprotegerin/osteoclastognesis inhibitory factor by bone marrow stromalcells. J Biol Chem 1998;273:27091–6.

79 Bucay N, Sarosi I, Dunstan CR, et al. Osteoprotegerin-deficient mice develop early onset osteoporosis and arterialcalcification. Gene Dev 1998;12:1260–8.

80 Marsters SA, Ayers TM, Skubatch M, Gray CL, Rothe M,Ashkenazi A. Herpesvirus entry mediator, a member of thetumor necrosis factor receptor (TNFR) family, interactswith members of the TNFR-associated factor family andactivates the transcription factors NF-kB and AP-1. J BiolChem 1997;272:14029–32.

81 Hsu H, Solovyev I, Colombero A, Elliott R, Kelley M, BoyleWJ. ATAR, a novel tumor necrosis factor receptor family

member, signals through TRAF2 and TRAF5. J Biol Chem1997;272:13471–4.

82 Kwon BS, Tan KB, Ni J, et al. A newly identified member ofthe tumor necrosis factor receptor superfamily with a widetissue distribution and involvement in lymphocyte activa-tion. J Biol Chem 1997;272:14272–6.

83 Montgomery RI, Warner MS, Lum BJ, Spear PG. Herpessimplex virus-1 entry into cells mediated by a novel mem-ber of the TNF/NGF receptor family. Cell 1996;87:427–36.

84 Zhai Y, Guo R, Hsu TL, et al. LIIGHT, a novel ligand forlymphotoxin receptor and TR2/HVEM induces apoptosisand suppresses in vivo tumor formation via gene transfer. JClin Invest 1998;102:1142–51.

85 Harrop JA, McDonnell PC, Brigham-Burke M, et al.Herpesvirus entry mediator ligand (HVEM-L), a novel lig-and for HVEM/TR2, stimulates proliferation of T cells andinhibits HT29 cell growth. J Biol Chem 1998;273:27548–56.

86 Kwon B, Yu K-Y, Ni J, et al. Identification of a novelactivation-inducible protein of the tumor necrosis factorreceptor superfamily and its ligand. J. Biol Chem1999;274:6056–61.

87 Gurney AL, Marsters SA, Huang A, et al. Identification of anew member of the tumor necrosis factor family and itsreceptor, a human ortholog of mouse GITR. Curr Biol1999;9:215–18.

88 Nocentini G, Giunchi L, Ronchetti S, et al. A new memberof the tumor necrosis factor/nerve growth factor receptorfamily inhibits T cell receptor-induced apoptosis. Proc NatlAcad Sci USA 1997;94:6216–21.

89 Wiley SR, Schooley K, Smolak PJ, et al. Identification andcharacterization of a new member of the TNF family thatinduces apoptosis. Immunity 1995;3:673–82.

90 Pitti RM, Marsters SA, Ruppert S, Donahue CJ, Moore A,Ashkenazi A. Induction of apoptosis by Apo-2 ligand, anew member of the tumor necrosis factor cytokine family. JBiol Chem 1996;271:12687–90.

91 Jeremias I, Herr I, Boehler T, Debatin K-M. TRAIL/Apo-2-ligand-induced apoptosis in human T cells. Eur J Immunol1998;28:143–52.

92 Kayagaki N, Yamaguchi N, Nakayama M, Eto H, OkumuraK, Yagita H. Type I interferons (IFNs) regulate tumornecrosis factor-related apoptosis-inducing ligand (TRAIL)expression on human T cells: A novel mechanism for theantitumor eVects of type I IFNs. J Exp Med 1999;189:1451–60.

93 GriYth TS, Wiley SR, Kubin MZ, Sedger LM, MaliszewskiCR, Fanger NA. Monocyte-mediated tumoricidal activityvia the tumor necrosis factor-related cytokine, TRAIL. JExp Med 1999;189:1343–53.

94 GriYth TS, Chin WA, Jackson GC, Lynch DH, Kubin MZ.Intracellular regulation of TRAIL-induced apoptosis inhuman melanoma cells. J Immunol 1998;161:2833–40.

95 Thomas WD, Hersey P. TNF-related apoptosis-inducingligand (TRAIL) induces apoptosis in Fas ligand-resistantmelanoma cells and mediates CD4 T cell killing of targetcells. J Immunol 1998;161:2195–200.

96 Zamai L. Ahmad M, Bennett IM, Azzoni L, Alnemri ES,Perussia B. Natural killer (NK) cell-mediated cytotoxicity:diVerential use of TRAIL and Fas ligand by immature andmature primary human NK cells. J Exp Med 1998;188:2375–80.

97 Walczak H, Miller RE, Ariail K, et al. Tumoricidal activity oftumor necrosis factor-related apoptosis-inducing ligand invivo. Nat Med 1999;5:157–63.

98 Hahne M, Kataoka T, Schroter M, et al. APRIL, a new lig-and of the tumor necrosis factor family, stimulates tumorcell growth. J Exp Med 1998;188:1185–90.

99 Shu H-B, Hu W-H, Johnson H. TALL-1 is a novel memberof the TNF family that is down-regulated by mitogens. JLeuk Biol 1999;65:680–3.

100 Chicheportiche Y, Bourdon PR, Xu H, et al. TWEAK, annew secreted ligand in the tumor necrosis factor family thatweakly induces apoptosis. J Biol Chem 1997;272:32401–10.

101 Lynch CN, Wang YC, Lund JK, Chen Y-W, Leal JA, Wiley,SR. TWEAK induces angiogenesis and proliferation ofendothelial cells. J Biol Chem 1999;274;8455–9.

102 Zhai, Y, Ni J, Jiang G-W, et al. VEGI, a novel cytokine ofthe tumor necrosis factor family, is an angiogenesis inhibi-tor that suppresses the growth of colon carcinomas in vivo.FASEB J 1999;13:181–9.

103 Wong BR, Rho J, Arron J, et al. TRANCE is a novel ligandof the tumor necrosis factor receptor family that activatesc-Jun N-terminal kinase in T cells. J Biol Chem 1997;272:25190–4.

104 Green EA, Flavell RA. TRANCE-RANK, a new signalpathway involved in lymphoctye development and T cellactivation. J Exp Med 1999;189:1017–20.

105 Wong BR, Josien R, Lee SW, Sauter B, Li H-L, SteinmanRM, Choi Y. TRANCE (Tumor necrosis factor [TNF]-related activation-induced cytokine), a new TNF familymember predominantly expressed in T cells, is a dendriticcell-specific survival factor. J Exp Med 1997;186:2075–80.

106 Lum L, Wong BR, Josien R, et al. Evidence for a role of atumor necrosis factor-(TNFa)-converting enzyme-like pro-tease in shedding of TRANCE, a TNF family memberinvolved in osteoclastogenesis and dendritic cell survival. JBiol Chem 1999;274:13613–18.

107 Fuller K, Wong BR, Fox S, Choi Y, Chambers TJ.TRANCE is necessary and suYcient for osteoblast-

I12 Darnay, Aggarwal



j.com/

Ann R

heum D


ber 1999. Dow

nloaded from

http://ard.bmj.com/

mediated activation of bone resporbtion in osteoclast. J ExpMed 1998;188:997–1001.

108 Burgess TL, Qian Y-X, Kaufman S, et al. The ligand forosteoprotegerin (OPGL) directly activates mature osteo-clasts. J Cell Biol 1999;145:527–38.

109 Kong Y-Y, Yoshida H, Saros I, et al. OPGL is a key regula-tor of osteoclastogenesis, lymphocyte development andlymph-node organogenesis. Nature 1999;397:315–23.

110 Kotake S, Udagawa N, Takahashi N, et al. IL-17 insynovial fluids from patients with rheumatoid arthritis is apotent stimulator of osteoclastogenesis. J Clin Invest 1999;103:1345–52.

111 Mukhopadhyay A, Ni J, Zhai Y, Yu G-L, Aggarwal BB.Identification and characterization of a novel cytokine,THANK, a TNF homologue that activates apoptosis,nuclear factor-kB, and c-Jun NH2-terminal kinase. J BiolChem 1999;274:15978–81.

112 Schneider P, MacKay F, Steiner V, et al. BAFF, a novel lig-and of the tumor necrosis factor family, stimulates B cellgrowth. J Exp Med 1999;189:1747–56.

113 Mauri DN, Ebner R, Montgomery RI, et al. LIGHT, a newmember of the TNF superfamily, and lymphoxin a are lig-ands for herpesvirus entry mediator. Immunity 1998;8:21–30.




j.com/

Ann R

heum D


ber 1999. Dow

nloaded from

http://ard.bmj.com/

signal transduction by tumour necrosis factor and …tumour necrosis factor related ligands and...

Documents