What the papers say
Do viral chemokines modulate Kaposi’s sarcoma? Dirk Dittmer and Dean H. Kedes* Summary
Kaposi’s sarcoma (KS) is an angiogenic tumor of mixed cellularity most commonly found in homosexual men infected with HIV. Both molecular and epidemiologic evidence has linked a newly described herpesvirus to this disease. This virus, Kaposi’s sarcoma-associated herpesvirus (KSHV), encodes a number of cellular homologues, including two genes that share remarkable similarity to the human chemokine macrophage inhibitory factor-1a. Recently, studies have begun to shed light on the roles these viral chemokines (vMIP-I and vMIP-II) may play in the complex pathogenesis of KS.1–3 The vMIP peptides may contribute to the formation of new blood vessels (neovascularization), inhibit infection by certain strains of HIV-1 and modify the cellular immune response. BioEssays 20:367–370, 1998. r 1998 John Wiley & Sons, Inc. Introduction Kaposi’s sarcoma (KS), though well recognized as a moderately rare tumor of the skin since the late nineteenth century, was mainly confined to populations in equatorial Africa and Mediterranean countries, until its meteoric surge during the early 1980s among homosexual men with acquired immunodeficiency syndrome (AIDS). The lesions of KS present more as an unusual immune-modulated hyperplasia of mixed cellularity than as a classic neoplasm. Central to all such lesions are (1) the so-called spindle cells, which express endothelial cell markers; (2) infiltrating lymphocytes; and (3) an intense neovascularization with marked capillary permeability. The resulting erythrocyte extravasation from these rudimentary vessels gives the lesions their characteristic purple pigmentation. In 1994 Chang and Moore4 redirected the field of KS research by identifying genomic DNA of a new human gammaherpesvirus, Kaposi’s sarcoma-associated herpesvirus, KSHV (also referred to as human herpesvirus-8, HHV-8) in KS lesions. Subsequently, the same genome was
Department of Microbiology and Immunology, University of California, San Francisco, San Francisco, California. Contract grant sponsor: National Institutes of Health; Contract grant number: R01 CA73506; Contract grant sponsor: University-wide AIDS Research; Contract grant number: R96-SF-142. *Correspondence to: Dean H. Kedes, Box 0414, University of California, San Francisco, San Francisco, CA 94143–0414. E-mail:
[email protected]
BioEssays 20:367–370, r 1998 John Wiley & Sons, Inc.
found in human immunodeficiency virus (HIV)-negative KS lesions, Castleman’s disease (an angiogenic lymphocytic tumor often coincident with KS) and a rare AIDS-associated B-cell lymphoma (primary effusion lymphoma, PEL). To date, a wealth of seroepidemiologic evidence has established KSHV as a necessary cofactor for KS development.5 However, the mechanisms by which KSHV contributes to the KS-phenotype remain unknown. Recent reports by Moore et al., Kledal et al., and Boshoff et al. described investigations into the KSHV-encoded homologues of human chemokine genes (see below) and proposed a role for them in KSangiogenesis,1 potential inhibition of HIV-1 infection and modification of the cellular immune response.1–3
Why investigate virally encoded chemokines? Chemokines (chemotactic cytokines) are a superfamily of small proteins (8–10 kD) originally identified by their ability to induce leukocyte chemotaxis.6 They can be secreted by most cell types and upon binding their cognate receptors cause GTP-dependent Ca21 influx and activation of an intracellular signaling cascade, resulting in differentiation, proliferation or chemotaxis of the target cell. The vast majority of the 28 known human chemokines fall into one of two subfamilies, the CC- (or b-) or the CXC- (or a-) chemokines, depending on the absence or presence of an additional amino acid between the first 2 cysteines of a signature 4-cysteine motif. The CCand CXC- subfamily members bind with varying specificity to corresponding receptors, referred to as CC-CRs and CXCCRs, respectively.
BioEssays 20.5
367
What the papers say
MIP-1 : vMIP-I/K6: vMIP-II/K4:
MQVSTAALAVLLCTMALCNQVLSAPLAADTPTACCFSYTSRQIPQNFIADYFETSSQCSKPSVIFLTKRGRQVCADPSEEWVQKYVSDLELSA || || | || | || || ||||||| ||||| || | | MAPVHVLCCVSVLLATFYLTPTESAGSLVSYTPNSCCYGFQQHPPPVQILKEWYPTSPACPKPGVILLTKRGRQICADPSKNWVRQLMQRLPAIA | || | || | || || ||||||| ||| | || | | MDTKGILLVAVLTALLCLQSGDTLGASWHRPDKCCLGYQKRPLPQVLLSSWYPTSQLCSKPGVIFLTKRGRQVCADKSKDWVKKLMQQLPVTAR
synthesized vMIP-II: authentic secreted vMIP-II:
GDTLGASWHRPDKCCLGYQKRPLPQVLLSSWYPTSQLCSKPGVIFLTKRGRQVCADKSKDWVKKLMQQLPVTAR LGASWHRPDKCCLGYQKRPLPQVLLSSWYPTSQLCSKPGVIFLTKRGRQVCADKSKDWVKKLMQQLPVTA
MIP-1 : vMIP-I/orfK6: vMIP-II/orfK4:
g127077 g1718266 g1718264
Figure 1. Predicted amino acid sequence of the open reading frames of human MIP-1a, vMIP-I, and vMIP-II (lines 1–3, respectively) compared with that of the synthesized (line 4) and the secreted (line 5) versions of vMIP-II employed by Boshoff et al.1 and Kledal et al.,3 respectively. Amino acids common to MIP-1a and both vMIPs are indicated by vertical lines. Red lines denote the conserved disulfide bridges within the 4 cysteine motif. (Accession numbers are given below.)
The double-stranded 165-kbp DNA genome of KSHV is predicted to encode 81 open reading frames (ORFs).7,8 Two KSHV genes, vMIP-I (orf K6) and vMIP-II (orf K4), show ,40% sequence identity to the human macrophage inflammatory protein-1a (MIP-1a) and, based on their primary amino acid sequence, belong to the CC-chemokines. In the host, MIP-1a activates a wide range of target cells such as monocytes, eosinophils, and neutrophils and likely plays an important part in antiviral defense.9 Boshoff et al. worked with chemically synthesized vMIP-I (amino acids 25–95) and vMIP-II (amino acids 21–94) peptides, while Kledal et al. showed that the secreted form of vMIP-II has lost its N-terminal 23-amino acid signal sequence as well as a single C-terminal arginine (3) leaving an active peptide consisting of amino acids 24–93 (Fig. 1). N- and C-terminal processing events are common for many chemokines and crucial to their activity. In CC-chemokines, single amino acid deletions or substitutions at the N-terminus can change agonists into antagonists and vice versa.10–12 Therefore, differences in the peptides need to be considered when comparing the results of otherwise similar chemokine assays. A role for the vMIPs in angiogenesis? Although studies to date have implicated only CXC-chemokines in neovascularization, angiogenesis and angiostasis, Boshoff et al. tested whether the highly angiogenic phenotype of KS could be attributed to the CC-chemokine homologues encoded by KSHV. They placed vMIP-I and vMIP-II on methylcellulose discs, implanted these onto chorioallantoic membranes of chicken eggs13 and then scored the degree of
368
BioEssays 20.5
neovascularization. Both vMIP-I and vMIP-II induced angiogenesis in this assay while two control CC-chemokines, RANTES and human MIP-1a, did not, leading the authors to conclude that vMIP-I and vMIP-II may contribute to the angiogenic phenotype of KS. KS spindle cell lines derived from primary tumors release a multitude of endogenous angiogenic factors (including IL-1b, IL-6, oncostatin-M, VEGF, and bFGF), some of which are regulated by HIV-1 tat.14 KSHV likewise encodes a homologue of interleukin-6 (IL-6).15,16 By injecting SCID mice with a KSHV-positive B-cell line, Pecchio et al. found that the resulting tumors were marked by highly exaggerated neovascularization and vascular permeability, characteristics reminiscent of the KS phenotype.17 Similarly, Bais et al. found highly vascularized tumors in nude mice implanted with NIH 3T3 cells stably expressing KSHV’s own constitutively active chemokine-receptor-like homologue (vCC-CR/orf74).18 Here and in tumor models employing KS-derived cell lines, the angiogenic phenotype could be attributed to VEGF-1.19,20 The degree to which the vMIPs or other KSHV-encoded angiogenic factors (e.g., vIL-6) or cellular cofactors (e.g., IL-1b, VEGF-1) contribute to KSHV-related tumor angiogenesis in these animal models remains to be determined. The vMIPs may, in fact, exert their effects on KS neovascularization through indirect mechanisms such as recruiting inflammatory cells or stimulating spindle or endothelial cells to release conventional angiogenesis factors. It will be difficult to prove a direct involvement of vMIPs (or other KSHVencoded gene products) in angiogenesis in vitro, since KS-derived cell lines retain their angiogenic potential during
What the papers say
passage, but lose their viral episomes. In vivo observations present yet another conundrum: although almost all KS spindle cells harbor the virus,21,22 the vast majority are in the latent state,22 where significant vMIP transcription is low or undetectable.2 Whether the small number of lytically infected cells observed in the lesions provide sufficient levels of the chemokine homologues to elicit the pathophysiologic response remains unclear. KSHV vMIPs block HIV-1 replication in vitro In conjunction with CD4, HIV-1 requires chemokine receptors for fusion and subsequent viral entry (ref. 23 and references therein). HIV-1 employs two general classes of chemokine receptors; macrophage-tropic (M-tropic) HIV-1 strains use CC-CR5, whereas T-cell-tropic (T-tropic) strains can additionally use CXC-CR4. In vitro, a subset of M-tropic HIV-1 strains can use a third chemokine receptor, CC-CR3. It is now clear from the work of Moore, Kledal, and Boshoff, and their co-workers, that both vMIP-I and vMIP-II peptides bind various chemokine receptors with nanomolar dissociation constants in vitro.1–3 These investigators found that with the right combination of HIV-1 strain and target cell, these two chemokines can also inhibit HIV-1 spread in tissue culture. vMIP-I blocked M-tropic HIV-1 replication when target cells expressed CC-CR5,2 whereas the most pronounced effect of vMIP-II was to block the less common CC-CR3-dependent isolates and, with much less efficiency, the CC-CR5- and CXC-CR4-dependent isolates.1,3 But do vMIPs effect HIV-1 in vivo? Could the KSHV vMIPs limit HIV-1 spread in co-infected patients? Boshoff et al. envision a scenario in which the high KSHV viral load in lymph nodes of some late-stage AIDS patients may do just that. However, vMIP-II, the more potent of the two vMIPs in affecting in vitro HIV-1 entry and replication, has its most striking effect on CC-CR3, rather than CC-CR5 or CXC-CR4, the two predominant HIV-1 co-receptors in lymph nodes. Nevertheless, examining KS lesions with in situ probes for both HIV-1 and KSHV, Staskus et al. found no evidence for dually infected cells despite demonstrating the presence of the two viruses in close proximity.22 This may simply reflect mutually exclusive target populations for each virus rather than KSHV-induced local blockade of HIV-1 infection. Boshoff et al. further speculate that since microglia express high levels of CC-CR3 on their surface,24 vMIP-II might preferentially limit HIV-1 spread to these cells in the brain—the route of infection associated with AIDS encephapathy (dementia). Though KS can on rare occassion present in the brain,25 if KSHV is at all similar to all other g-herpesviruses, it is unlikely to exhibit appreciable neurotropism. Further, the major body of epidemiologic evidence available to date suggests that a diagnosis of KS is not independently associated with a reduced risk of AIDS en-
cephalopathy.26,27 In the end, establishing a link between altered HIV-1 pathogenesis and KSHV will require the same careful epidemiological studies that connected KSHV to KS. Chemokine piracy: what’s in it for KSHV? Although the pathogenic effects of KSHV are tremendously amplified in the setting of HIV-1-induced immunosuppression, the selective advantage of incorporating vMIPs was most assuredly to promote KSHV propagation in its host—not to perturb HIV-1 infection. This is especially apparent by recalling that KSHV must have evolved to include these genes well before the onset of the HIV-1 epidemic. The roles of these chemokine homologues, therefore, are perhaps better assessed by examining their effects on potential host target cells bearing cognate receptors. In the hands of both Kledal et al. and Bishoff et al., vMIP-II but not vMIP-I induced receptor desensitization to subsequent challenge by the receptors’ native chemokines (a technique commonly employed to demonstrate competitive binding of two distinct cytokines to a common receptor).1,3 Strikingly, the vMIP-II blockade was manifest not only on the CC-chemokine receptors CC-CR1, 2, 3, and 5 but also on the CXC-chemokine receptor, CXC-CR4. By examining a T-cell-depleted monocyte population, Kledal et al. attributed an antagonist function to this blockade, finding that their version of vMIP-II (see Fig. 1, line 5) blocked signaling normally generated by host chemokines but did not elicit a signal of its own.3 In this way, they concluded, one potential role of the vMIPs would be to inhibit chemokine signaling by competitive binding to as many different receptors as possible, thereby obstructing the immune response to KSHV infection. Another view was lent by Boshoff et al., who, by examining isolated eosinophils (cells activated in allergic and antiparasitic responses) that express high levels of CC-CR3, assigned an agonist role to vMIP-II(1). These workers found that their synthesized version of this chemokine homologue (see Fig. 1, line 4) bound well to this receptor and induced both rapid Ca21 influx in and chemotaxis of eosinophils. In this way, Boshoff and colleagues concluded, vMIP-II effectively mimicked eotaxin, the major eosinophil chemoattractant. However, since the peptide synthesized by Boshoff et al. includes three additional N-terminal amino acids compared with vMIP-II purified from transfected mammalian cells these assays may be misleading.10–12 Importantly, eosinophils are not routinely present in KS lesions nor would it seem advantageous for KSHV to attract such a cell population— casting further doubt on the in vivo relevance of any eotaxinlike effects of vMIP-II. Evolutionary arguments can be marshaled in support for either an agonist or antagonist role of viral chemokines. Because of the complex and overlapping receptor specificities of the human chemokines, ultimately only viral recombinants in conjunction with a suitable animal model can resolve
BioEssays 20.5
369
What the papers say
these questions. To that end the recent discovery of both murine28 and monkey29,30 g-2 herpesviruses may provide such experimental systems. vMIPs: advantage but not match? How else could KSHV contribute to the KS phenotype? Herpesviruses are notorious for pirating cellular genes31,32 (e.g., murine cytomegalovirus also encodes a human chemokine homologue33). As mentioned above, KSHV also encodes a chemokine-receptor homologue (vCC-CR/orf74). While vCC-CR cannot function as a HIV-1 co-receptor,23 it may play a role in KS angiogenesis,18 and its ectopic expression induces cell proliferation in an agonist-independent fashion.34 These KSHV-encoded homologues (vMIP-I, vMIP-II, vIL-6, and vCC-CR/orf74) are most abundantly expressed during viral replication, which eventually leads to cell lysis, suggesting that, in the context of viral infection, they might be involved in transient paracrine signaling. A second set of viral genes are transcribed during latency, including two homologues of cellular genes (v-FLICE and v-cyclin) that have been implicated in intracellular cell cycle regulation and could therefore provide a more permanent growth signal.35,36 In concert, these viral gene products could create the complex growth behavior characteristic of Kaposi’s sarcoma. Acknowledgments We thank Rolf Renne and Don Ganem for helpful comments on the manuscript.
References 1 Boshoff C, et al. (1997) Angiogenic and HIV-inhibitory functions of KSHV-encoded chemokines. Science 278:290–294. 2 Moore PS, Boshoff C, Weiss RA, and Chang Y (1996) Molecular mimicry of human cytokine and cytokine response pathway genes by KSHV. Science 274:1739–1744. 3 Kledal T, et al. (1997) A broad-spectrum chemokine antagonist encoded by Kaposi’s sarcoma-associated herpesvirus. Science 277:1656–1659. 4 Chang Y, et al. (1994) Identification of herpesvirus-like DNA sequences in AIDS-associated Kaposi’s sarcoma. Science 266:1865–1869. 5 Ganem D (1997) KSHV and Kaposi’s sarcoma: The end of the beginning? Cell 91:157–160. 6 Adams D and Lloyd A (1997) Chemokines: Leucocyte recruitment and activation cytokines. Lancet 349:490–495. 7 Neipel F, Albrecht JC, and Fleckenstein B (1997) Cell-homologous genes in the Kaposi’s sarcoma-associated rhadinovirus human herpesvirus 8: Determinants of its pathogenicity? J Virol 71:4187–4192. 8 Russo J, et al. (1996) Nucleotide sequence of the Kaposi sarcomaassociated herpesvirus (HHV8). Proc Natl Acad Sci USA 93:4862–14867. 9 Cook D, et al. (1995) Requirement of MIP-1 alpha for an inflammatory response to viral infection. Science 269:1583–1585. 10 Baggiolini M and Moser B (1997) Blocking chemokine receptors. J Exp Med 186:1189–1191. 11 Gong J and Clark-Lewis I (1995) Antagonists of monocyte chemoattractant protein 1 identified by modification of functionally critical NH2-terminal residues. J Exp Med 181:631–640.
370
BioEssays 20.5
12 Simmons G, et al. (1997) Potent inhibition of HIV-1 infectivity in macrophages and lymphocytes by a novel CCR5 antagonist. Science 276: 276–279. 13 Jain R , Schlenger K, Hockel M, and Yuan F (1997) Quantitative angiogenesis assays: Progress and problems. Nature Med 3:1203–1208. 14 Albini A, et al. (1996) The angiogenesis induced by HIV-1 tat protein is mediated by the Flk-1/KDR receptor on vascular endothelial cells. Nature Med 2:1371–1375. 15 Molden J, Chang Y, You Y, Moore PS, and Goldsmith MA (1997) A Kaposi’s sarcoma-associated herpesvirus-encoded cytokine homolog (vIL-6) activates signaling through the shared gp130 receptor subunit. J Biol Chem 272:19625–19631. 16 Nicholas J, et al. (1997) Kaposi’s sarcoma-associated human herpesvirus-8 encodes homologues of macrophage inflammatory protein-1 and interleukin-6. Nature Med 3:287–292. 17 Picchio GR, Sabbe RE, Gulizia RJ, McGrath M, Herndier BG, and Mosier DE (1997) The KSHV/HHV8-infected BCBL-1 l lymphoma line causes tumors in SCID mice but fails to transmit virus to a human Peripheral blood mononuclear cell graft. Virology 238:22–29. 18 Bais C, et al. (1998) G-protein-coupled receptor of Kaposi’s sarcomaassociated herpesvirus is a viral oncogene and angiogenesis activator. Nature 391:86–89. 19 Masood R, Cai J, Zheng T, Smith DL, Naidu, Y and Gill, PS (1997) Vascular endothelial growth factor/vascular permeability factor is an autocrine growth factor for AIDS–Kaposi sarcoma. Proc Natl Acad Sci USA 94:979–984. 20 Weindel K, Marme D, and Weich HA (1992) AIDS-associated Kaposi’s sarcoma cells in culture express vascular endothelial growth factor. Biochem Biophys Res Commun 183:1167–1174. 21 Boshoff C, et al. (1995) Kaposi’s sarcoma-associated herpesvirus infects endothelial and spindle cells. Nature Med 1:1274–1278. 22 Staskus KA, et al. (1997) Kaposi’s sarcoma-associated herpesvirus gene expression in endothelial (spindle) tumor cells. J Virol 71:715–719. 23 Rucker, J, et al. (1997) Utilization of chemokine receptors, orphan receptors, and herpesvirus-encoded receptors by diverse human and simian immunodeficiency viruses. J Virol 71:8999–9007. 24 He J, et al. (1997) CCR3 and CCR5 are co-receptors for HIV-1 infection of microglia. Nature 385:645–649. 25 Matthiessen L, Marche C, Labrousse F, Trophilme D, Fontaine C, and Vedrenne, C (1992) Neuropathology of the brain in 174 patients who died of AIDS in a Paris hospital 1982–1988. Ann Med Interne (Paris) 143:43–49. 26 Bell JE, et al. (1996) Influence of risk group and zidovudine therapy on the development of HIV encephalitis and cognitive impairment in AIDS patients. AIDS 10:493–499. 27 McArthur JC, et al. (1997) Relationship between human immunodeficiency virus-associated dementia and viral load in cerebrospinal fluid and brain. Ann Neurol 42:689–698. 28 Virgin HW, et al. (1997) Complete sequence and genomic analysis of murine gammaherpesvirus 68. J Virol 71:5894–5904. 29 Desrosiers RC, et al. (1997) A herpesvirus of rhesus monkeys related to the human Kaposi’s sarcoma-associated herpesvirus. J Virol 71:9764– 9769. 30 Rose T, et al. (1997) Identification of two homologs of the Kaposi’s sarcoma-associated herpesvirus (human herpesvirus 8) in retroperitoneal fibromatosis of different macaque species. J Virol 71:4138–4144. 31 Murphy PM (1997) Pirated genes in Kaposi’s sarcoma [news; comment]. Nature 385:296–297, 299. 32 Meinl E, Fickenscher H, and Fleckenstein B (1996) Chemokine receptors and chemokine-inducing molecules of lymphotropic herpesviruses [letter]. Immunol Today 17:199. 33 MacDonald MR, Li XY, and Virgin HW (1997) Late expression of a beta chemokine homolog by murine cytomegalovirus. J Virol 71:1671–1678. 34 Arvanitakis L, Geras-Raaka E, Varma A, Gershengorn M, and Cesarman E (1997) Human herpesvirus KSHV encodes a constitutively active G-protein-coupled receptor linked to cell proliferation. Nature 385:347– 349. 35 Swanton C, Mann DJ, Fleckenstein B, Neipel F, Peters G, and Jones, N (1997) Herpes viral cyclin/Cdk6 complexes evade inhibition by CDK inhibitor proteins. Nature 390:184–187. 36 Thome M, et al. (1997) Viral FLICE-inhibitory proteins (FLIPs) prevent apoptosis induced by death receptors. Nature 386:517–521.
