- Email: [email protected]
Recognition and receptors in virus transmission by arthropods
REVIEWS
Acknowledgements Support from NIH grants from the US Public Health Service is gratefully acknowledged. References 1 Deepe, G.S. and Bullock, W.E. (1992) in Inflammation: Basic Principles and Clinical Correlates (Gallin, J.I. et al., eds), pp. 943–958, Raven Press 2 Bullock, W.E. and Wright, S.D. (1987) J. Exp. Med. 165, 195–210 3 Newman, S.L. et al. (1990) J. Clin. Invest. 85, 223–230 4 Kimberlin, C.L. et al. (1981) Infect. Immun. 34, 6–10 5 Eissenberg, L.G. and Goldman, W.E. (1987) Infect. Immun. 55, 29–34 6 Wolf, J.E. et al. (1987) J. Immunol. 138, 582–586 7 Wolf, J.E. et al. (1989) Infect. Immun. 57, 513–519 8 Howard, D.H. (1965) J. Bacteriol. 89, 518–523 9 Newman, S.L., Gootee, L. and Gabay, J.E. (1993) J. Clin. Invest. 92, 624–631 10 Eissenberg, L.P., Schlesinger, P.H. and Goldman, W.E. (1988) J. Leukocyte Biol. 43, 483–491 11 Taylor, M.L. et al. (1989) Clin. Exp. Immunol. 75, 466–470 12 Newman, S.L. et al. (1997) J. Immunol. 158, 1779–1786 13 Howard, D.H. (1964) J. Bacteriol. 87, 33–38 14 Newman, S.L. et al. (1991) Infect. Immun. 59, 737–741 15 Newman, S.L. et al. (1992) J. Immunol. 149, 574–580 16 Eissenberg, L.G., Goldman, W.E. and Schlesinger, P.H. (1993) J. Exp. Med. 177, 1605–1611 17 Dautry-Varsat, A., Ciechanover, A. and Lodish, H.F. (1983) Proc. Natl. Acad. Sci. U. S. A. 80, 2258–2262 18 Lestas, A.N. (1976) Br. J. Haematol. 32, 341–350 19 Lane, T.E., Wu-Hsieh, B.A. and Howard, D.H. (1991) Infect. Immun. 59, 2274–2278 20 Newman, S.L. et al. (1994) J. Clin. Invest. 93, 1422–1429
21 Lukacs, G.L., Rotstein, O.D. and Grinstein, S. (1990) J. Biol. Chem. 265, 21009–21017 22 Lukacs, G.L., Rotstein, O.D. and Grinstein, S. (1991) J. Biol. Chem. 266, 24540–24544 23 Zhou, P. et al. (1995) J. Immunol. 155, 785–795 24 Lane, T.E., Wu-Hsieh, B.A. and Howard, D.H. (1993) Infect. Immun. 61, 1468–1473 25 Smith, J.G. et al. (1990) J. Infect. Dis. 162, 1349–1353 26 Zhou, P., Miller, G. and Seder, R.A. (1998) J. Immunol. 160, 1359–1368 27 Lane, T.E., Wu-Hsieh, B.A. and Howard, D.H. (1994) Infect. Immun. 62, 1940–1945 28 Fleischmann, J., Wu-Hsieh, B. and Howard, D.H. (1990) J. Infect. Dis. 161, 143–145 29 Newman, S.L. and Gootee, L. (1992) Infect. Immun. 11, 4593–4597 30 Sturgill-Koszycki, S. et al. (1994) Science 263, 678–681 31 Russell, D.G., Dant, J. and Sturgill-Koszycki, S. (1996) J. Immunol. 156, 4764–4773 32 Sturgill-Koszycki, S., Shiable, U.E. and Russell, D.G. (1996) EMBO J. 15, 6960–6968 33 Xu, S. et al. (1994) J. Immunol. 153, 2568–2578 34 Hackam, D.J. et al. (1997) J. Biol. Chem. 272, 29810–29820 35 Hanano, R. and Kaufmann, S.H. (1995) Immunol. Lett. 45, 23–27 36 Chan, J. et al. (1992) J. Exp. Med. 175, 1111–1122 37 Rook, G.A. et al. (1987) Immunology 67, 229–234 38 Schneemann, M. et al. (1993) J. Infect. Dis. 167, 1358–1363 39 Bonecini-Almeida, M.G. et al. (1998) J. Immunol. 160, 4490–4499 40 Huang, S. et al. (1993) Science 259, 1742–1745 41 Cooper, A.M. et al. (1993) J. Exp. Med. 178, 2243–2247 42 Flynn, J. et al. (1993) J. Exp. Med. 178, 2249–2254 43 Newport, M.J. et al. (1996) New Engl. J. Med. 335, 1941–1949 44 Jouanguy, E. et al. (1996) New Engl. J. Med. 335, 1956–1961
Recognition and receptors in virus transmission by arthropods Johannes F.J.M. van den Heuvel, Saskia A. Hogenhout and Frank van der Wilk
M
ore than 70% of Fundamental knowledge of the molecular propagative. However, the viruses infecting plants mechanisms underlying virus transmission majority of plant viruses is and .40% of viruses by arthropods is a prerequisite for the transmitted in a noncirculative infecting mammals use arthrocreation of new strategies to modulate manner; i.e. they do not replipod vectors to move from one vector competence. There have been several cate in their vectors, as they host to another. As a result, recent advances in identifying the viral and are only associated with the the health of .2.5 billion peovector determinants involved in virus cuticular linings of the mouthple, mainly from tropical and recognition, attachment and retention. parts and the anterior part of subtropical regions, and the the alimentary tract. Generally, J.F.J.M. van den Heuvel*, S.A. Hogenhout and production of agricultural comsuch plant viruses are mechaniF. van der Wilk are in the Dept of Virology, DLO modities worldwide are at Research cally transmissible as well. Institute for Plant Protection (IPO-DLO), PO stake. Not only do arthropodMechanical transmission by Box 9060, 6700 GW Wageningen, The Netherlands. borne viruses have to cross epiarthropods has been reported *tel: 131 317 476141, fax: 131 317 410113, thelial cell linings of the gut for a few mammalian viruses, e-mail: [email protected] and salivary glands but they also including members of the have to resist the potentially Retroviridae1. hostile environment within their vectors. Almost all Phloem-feeding insects, such as aphids, whiteflies, mammalian viruses and some plant viruses replicate leafhoppers and planthoppers, and haematophagous in their vector during this cyclic passage; i.e. they are mosquitoes and ticks play a prominent role in the 0966-842X/99/$ - see front matter © 1999 Elsevier Science. All rights reserved. TRENDS
IN
MICROBIOLOGY
71
PII: S0966-842X(98)01434-6
VOL. 7 No. 2 February 1999
REVIEWS
Table 1. Virus families and genera59 containing species transmitted in a circulative manner by arthropods Family, genus
Hosta
Arthropod vector
Propagative
Bunyaviridae Bunyavirus Nairovirus Phlebovirus Tospovirus
M M M P
Mosquitoes, ticks, ceratopogonids, phlebotomines Ticks Phlebotomines, mosquitoes, ceratopogonids Thrips
Yes Yes Yes Yes
Circoviridae Nanovirus
P
Aphids
No
Flaviviridae Flavivirus
M
Mosquitoes, ticks
Yes
Geminiviridae Mastrevirus Curtovirus Begomovirus
P P P
Leafhoppers (Cicadellidae) Leafhoppers, treehoppers (Membracidae) Whiteflies (Bemisia tabaci)
No No No?
Luteoviridae Luteovirus Polerovirus Enamovirus
P P P
Aphids Aphids Aphids
No No No
Reoviridae Fijivirus Phytoreovirus Oryzavirus Orbivirus
P P P M
Yes Yes Yes Yes
Coltivirus
M
Planthoppers (Delphacidae) Leafhoppers Planthoppers Anopheline and culicine mosquitoes, sandflies (Phlebotomus spp.), gnats, ticks Ticks
Rhabdoviridae Cytorhabdovirus Nucleorhabdovirus Ephemerovirus Unassigned Rhabdoviruses
P P M P
Leafhoppers, aphids Leafhoppers, aphids Culicoides, mosquitoes Lacewings, mites, leafhoppers, aphids
Yes Yes Yes
Togaviridae Alphavirus
M
Mosquitoes, Cimidae
Yes
Soft ticks (Ornithodoros spp.)
Yes
Leafhoppers Planthoppers Aphids
Yes Yes? No
Genera not yet assigned to families ‘African swine fever-like M viruses’ Marafivirus P Tenuivirus P Umbravirus P
Yes
a
Species infects either plant (P) or mammalian (M) hosts.
dissemination of arthropod-borne viruses (Table 1). The route of some of these viruses in the invertebrate vector and the transmission barriers encountered have been well described, providing the groundwork for recent studies on the molecular basis of virus transmission by arthropods. The high degree of vector specificity and differences in tissue tropism and retention sites among arthropod-borne viruses suggest an intimate association in which both virus-encoded proteins and vector components are involved. Identification of the vector determinants involved in the different phases of the transmission process represents a nascent, but rapidly developing, research area. Despite
TRENDS
IN
MICROBIOLOGY
72
the vast body of literature dealing with host receptors for viruses, only limited information is available on vector proteins involved in recognition, attachment and retention of viruses, and only recently have insect proteins with a capacity to bind plant and mammalian viruses been identified. What are the viral determinants for transmission? It has long been recognized that the viral determinants governing transmissibility and attachment of viruses are primarily located in the structural proteins of a virus particle2,3. The domains involved are usually identified by comparative studies of the encoding
VOL. 7 No. 2 February 1999
REVIEWS
genes of transmissible and nontransmissible forms4,5. The latter often arise naturally when vector pressure is lifted; i.e. a virus is transmitted or maintained in the absence of its natural vector. A functional role for the transmission domains can be confirmed by mutational analysis of the viral genome or infectious viral cDNA clones6–10. Circulative viruses Although transmission determinants of many circulative plant viruses are known, we will limit our discussion to three recent examples, which represent taxa with different structural features (Table 1). Long-term vegetative propagation of rice infected with rice dwarf virus (Phytoreovirus sp.) results in the development of a transmission-defective isolate that lacks the outer capsid protein, P2, owing to a point mutation in the genome segment encoding this protein11. P2 is involved in the adsorption of the virus on insect cells12. Insect transmissibility of African cassava mosaic virus (Begomovirus sp.) and tomato spotted wilt virus (TSWV) (Tospovirus sp.) is also lost quite rapidly after several rounds of mechanical inoculation of the virus, because of defects in the coat protein (CP)13 or lack of the lipid envelope14, respectively. The CP of a geminivirus is the basic determinant of virus acquisition, although other gene products can influence the accumulation of the virus in the whitefly13,15. Moreover, the CP is involved in the molecular recognition of the virus by its vector. Exchanging the CP gene of the African cassava mosaic virus (Begomovirus sp.) with that of beet curly top virus (Curtovirus sp.) alters the insect specificity of the former from whiteflies to leafhoppers6. Thrips transmission of the envelope-minus TSWV isolate, NL-04, which is highly infectious upon mechanical inoculation, has so far failed, suggesting that structural components in the envelope, such as the glycoproteins G1 and G2, might be involved in recognition of a receptor in the thrips midgut14. Noncirculative viruses The transmission of some noncirculative viruses not only depends on the viral capsid but also on the presence of an active virus-encoded helper factor that is not a constituent of the virus particle8. Helper factors of the potyviruses and caulimoviruses are among the best-characterized viral products. They are bifunctional molecules capable of interaction with both the virus particle and the surface of the stylet food canal of the aphid via two separate domains and must be acquired by the vector before or during virus acquisition8. Single amino acid substitutions in the carboxyterminal region of the caulimovirus aphid transmission factor (ATF) have revealed that amino acids at positions 157 (Ile) and 159 (Gly) determine the interaction between the virus and the helper factor16. However, recent work has demonstrated that a second transmission factor, a 15-kDa protein encoded by gene III of cauliflower mosaic virus (CaMV), is involved in caulimovirus transmission as well. In vitro binding assays have shown that the carboxy-terminal region of ATF interacts with the amino terminus of
TRENDS
IN
MICROBIOLOGY
73
P15. This might indicate that the CaMV ATF does not directly interact with the viral CP but with P15, which is associated with purified CaMV particles [S. Blanc (1998) 7th International Congress of Plant Pathology, Edinburgh, UK, Abstr. 1.13.2S]. For potyviruses, the key amino acids functionally involved in binding and transmission are located in the conserved DAG motif and neighbouring residues of the capsid protein and the KITC- and PTK-containing domains of the helper factor17–22. Helper factor dependency has also been reported for noncirculative picornavirus-like plant viruses8,23,24 and has been suggested to be important for several other viruses, including the circulatively transmitted Nanovirus spp.25, which lose their ability to be transmitted by their vectors after purification from plant tissue. Helper factors of potyviruses have been shown to mediate virus retention in the food canal of the vector and vector-specific transmission26. Receptor sites for helper factors have not yet been identified. However, in recent years, significant advances have been made in identifying vector constituents involved in the transmission of viruses that are carried systemically by arthropods. Putative receptors for circulative viruses Binding between a virion and a cell surface receptor provides the initial physical association required for virus entry into vector cells. By analogy to virus entry into other animal cells, viruses can enter insect cells either by receptor-mediated endocytosis, generally through clathrin-coated vesicles, as has been proposed for luteoviruses27, or by direct fusion of the viral envelope with the cell membrane28. Based on observations made using electron microscopy, the latter mechanism has been proposed for bunyaviruses. Glycoproteins present in the membrane of the enveloped bunyaviruses have been suggested to mediate attachment29–31. Virus overlay assays have been instrumental in revealing vector proteins with a virus-binding capacity. In these assays, extracts of the vector are electrophoretically separated, blotted and probed with purified virus or anti-idiotypic antibodies, which are subsequently detected by antibodies. Anti-idiotypic antibodies mimicking the glycoproteins of TSWV specifically label a 50-kDa band in thrips homogenates on western blots and on the plasma membrane of the larval thrips midgut29. As purified virus displays an affinity for a similar-sized protein in an overlay assay of thrips proteins, the 50-kDa protein might be a cellular receptor29. A second thrips protein (~94 kDa) that has a TSWV-binding capacity has also been identified30. This protein firmly binds the G2 glycoprotein in virus overlay assays. G2 contains the highly conserved amino acid sequence Arg–Gly–Asp (RGD) near the amino terminus. This sequence is an important determinant for cellular attachment of several mammalian viruses and pathogens, including foot-and-mouth disease virus (FMDV)32, human coxsackievirus A9 (CAV-9)33 and the spirochete Borrelia burgdorferi34, which is the causal agent of Lyme disease. Owing to
VOL. 7 No. 2 February 1999
REVIEWS
the protease sensitivity of TSWV G2 and the high concentration of proteolytic enzymes that probably occur in the gut lumen, G2 is not expected to be involved in cellular attachment in the midgut of thrips30. However, studies on CAV-9 have shown that, even when the RGD-containing capsid domain is cleaved off, the virus remains infectious because it can bypass the RGD-recognizing vitronectin receptor33. Likewise, TSWV might employ alternative routes to enter thrips midgut cells, depending on its phenotype, which in turn might be regulated by thrips midgut proteases. Further characterization and localization of the 94kDa protein are required to elucidate its functional role in the cyclic passage of TSWV in thrips. In mosquitoes, putative receptor proteins for membrane-bound viruses of the genus Alphavirus (Table 1), such as chikungunya virus (CHIKV) and western equine encephalitis virus (WEEV), have been located in mosquito midguts35,36. CHIKV-binding activity is noticeably higher in the brush border membrane fraction of midguts derived from susceptible mosquitoes than in those derived from refractory insects36. Five different proteins from the brush border fraction and from the membrane fraction of Aedes albopictus
(a)
(b)
asg
psg
vp s
GroEL
mg (c)
m m hg
Trends in Microbiology
Fig. 1. (a) Diagram showing the circulative nature of a luteovirus in the aphid, and (b) the putative role of Buchnera GroEL in virus persistence. Virions are ingested with phloem sap from infected plants, move through the intestine and are transcellularly transported through the midgut (mg) or hindgut (hg) into the haemolymph, in which they persist for the life span of the aphid. The presence of Buchnera GroEL and the ability of the virus to interact with this protein determine the persistent nature of the luteovirus in the haemolymph of the vector. Upon contact with the accessory salivary gland (asg), virus particles can be transported through this gland, eventually arriving in the salivary duct, from which they are secreted with the saliva when the aphid feeds. (c) Electron micrograph of Buchnera cells in a mycetocyte (m). Abbreviations: psg, principal salivary gland; s, stomach; vp, virus particle.
TRENDS
IN
MICROBIOLOGY
74
C6/36 cells are known to bind CHIKV (Ref. 36). Viral attachment studies using this cell line further indicate that several surface proteins can serve as receptors for dengue type 4 virus37 (genus Flavivirus) and Venezuelan equine encephalitis virus (VEEV; Alphavirus sp.)38 (Table 1). There is strong evidence for the involvement of a 32-kDa laminin-binding protein in C6/36–VEEV interactions38. Laminin receptor proteins of different origins are highly conserved39, which might explain the broad range of vectors (several genera of mosquitoes and ticks) transmitting VEEV. Receptors accounting for the high degree of vector specificity that is commonly observed among invertebrates transmitting viruses, bacteria or parasites of plants and mammals still remain to be identified. It will be of great importance to reveal the true nature of these vector determinants because they offer new opportunities for disease control, including the genetic manipulation of the vector and transmission neutralization by means of vaccines or recombinant proteins expressed in transgenic plants. Towards identification of luteovirus receptors Species of the family Luteoviridae (Table 1) are solely transmitted by aphids in a circulative and non-propagative manner (Fig. 1). Based on ultrastructural studies of luteoviruses in vector and non-vector aphids, it has been postulated that virions are transcellularly transported through epithelial cell linings in the gut and salivary glands by receptor-mediated endocytosis–exocytosis27. The haemolymph acts as a reservoir in which acquired virus particles are retained in an infective form without replication for the life span of the aphid. Upon contact with the basal lamina of the accessory salivary gland, virus particles can be transported through this gland40, eventually arriving in the salivary duct, from which they are excreted with the saliva when the aphid feeds41. Vector specificity seems to be determined at the level of the accessory salivary gland40,41. Several aphid proteins, with relative molecular masses ranging from 31 to 85 kDa, have been shown to immobilize purified luteoviruses in vitro42 [X. Wang and G. Zhou (1998) 7th International Congress of Plant Pathology, Edinburgh, UK, Abstr. 1.13.16]. For example, a barley yellow dwarf virus-MAV-like isolate (BYDV-MAV; genus Luteovirus) [X. Wang and G. Zhou (1998) 7th International Congress of Plant Pathology, Edinburgh, UK, Abstr. 1.13.16] from China displays a strong affinity for two proteins of 31 and 44 kDa (P31 and P44, respectively) from the vector aphids Sitobion avenae and Schizaphis graminum but not from Rhopalosiphum padi. R. padi is unable to transmit BYDV-MAV. Antisera raised against P31 and P44 react specifically with extracts of the accessory salivary glands of vector aphids, suggesting that these proteins might be involved in luteovirus-specific recognition at the accessory salivary gland. The best-characterized protein that binds species from all three genera of the family Luteoviridae (Table 1) is GroEL (also known as symbionin),
VOL. 7 No. 2 FEBRUARY 1999
REVIEWS
which is abundantly produced by Buchnera spp., the primary endosymbiotic bacterium of aphids. The Gram-negative Buchnera spp. (Fig. 1c) are present in specialized polyploid host cells, called mycetocytes43, in the haemocoel of the aphid (Fig. 1a). The endosymbionts are distantly related to Escherichia coli44 and are maternally inherited, and the association, which has a nutritional basis, is obligatory for both the aphid and the bacterium45. Buchnera GroEL homologues are found in most aphid taxa, are immunologically closely related and share .80% sequence identity with the E. coli heat-shock protein GroEL, a member of the Hsp60 family of molecular chaperones42,46–48. Chaperonins are essential for cell viability, as they bind and stabilize newly translated or translocated aggregation-prone polypeptides and mediate their functional folding and assembly in an ATP-dependent manner49. Structural and functional characteristics of Buchnera GroEL are very similar to those of E. coli GroEL (Refs 37–48); however, unlike E. coli GroEL, Buchnera GroEL is secreted and is readily detected in the aphid haemolymph47. Native Buchnera GroEL, like E. coli GroEL, is an oligomer of ~840 kDa that consists of 14 identical subunits of 60 kDa arranged in two stacked heptameric rings47,48,50,51 (Fig. 1b). Both the subunits and the native 14-meric protein have been shown to bind luteoviruses in different ligand assays42,47,50 irrespective of whether GroEL is derived from vector or nonvector aphid species or from E. coli47,50. Pea enation mosaic virus (PEMV; genus Enamovirus) also has a high affinity for native GroEL homologues. Mutational analysis of the gene coding for Buchnera GroEL of Myzus persicae, the major vector of potato leafroll virus (PLRV; genus Luteovirus), shows that the virus-binding site is located in the equatorial domain of the subunit48. The minor capsid protein of a luteovirus, the readthrough domain (RTD), determines the interaction with GroEL (Refs 47,50). The RTD is exposed on the surface of a luteovirus particle7 and contains determinants necessary for virus transmission by aphids7,52–54. Sequence comparison of the RTDs of different luteoviruses and PEMV has identified highly conserved amino acid residues in the amino-terminal region of the RTD, which are potentially important in the interaction with Buchnera GroEL (Ref. 50). Direct injection of beet western yellows virus (BWYV; Polerovirus sp.) mutants into M. persicae has shown that virions devoid of RTD, which are unable to interact with Buchnera GroEL, are significantly less persistent in the aphid haemolymph than wild-type virions50. Moreover, treatment of M. persicae larvae with antibiotics that interfere with prokaryotic protein synthesis significantly reduces Buchnera GroEL levels in the haemolymph, inhibits transmission and results in the loss of capsid integrity in the haemolymph42. Collectively, these observations indicate that the luteovirus–GroEL interaction retards proteolytic breakdown and is essential for virus retention in the haemolymph of the aphid42,50.
TRENDS
IN
MICROBIOLOGY
75
Questions for future research • What are the cellular factors determining vector specificity of a virus? • Can mimics of viral attachment proteins be exploited to develop efficient strategies to modulate vector competence? • How do circulative viruses escape from the host’s immune response?
Does GroEL-mediated retention apply to other systems? Avidity for Buchnera GroEL appears to be a common characteristic of virus species from the family Luteoviridae. It is reasonable to assume that GroELmediated retention also applies to viruses of the genus Umbravirus (Table 1), because umbravirus RNA can only be transmitted by aphids if it is packaged in the capsid of a helper virus, commonly a luteovirus or PEMV (Ref. 25). Recent unpublished work from Morin and co-workers [S. Morin et al. (1998) Second International Workshop on Geminiviruses and their Vectors, San Juan, Puerto Rico, Abstr.] has shown that the Israeli strain of tomato yellow leafcurl virus (TYLCV; Begomovirus sp.) interacts with an endosymbiotic GroEL homologue from its whitefly vector Bemisia tabaci. Whiteflies harbour mycetome endosymbionts, as do most homopteran insects43. Whiteflies feeding on anti-Buchnera GroEL antiserum prior to virus acquisition have been shown to reduce the transmission of TYLCV by .80% relative to control insects feeding on normal serum. Active antibodies were recovered from the insect haemolymph, providing almost conclusive evidence that the antibody interferes with the interaction between TYLCV and the GroEL homologue. Endosymbiotic bacteria from leafhoppers have also been implicated in the vertical transmission of rice dwarf virus55. Similarly, tsetse flies, which are important vectors of human and animal trypanosomal diseases, have evolved symbiotic associations that have similar features to those of aphids, including the overproduction of GroEL by mycetome endosymbionts56. The possibility that endosymbiotic GroEL might have additional functions in the physiology of tsetse flies or in their competence to act as vectors of protozoan parasites requires further study. Molecular characterization of the endosymbiont–host interactions, and the genetic transformation of these bacteria with antiparasite or anti-viral constructs, could provide a significant tool for modulation of vector competence57,58. Acknowledgements We greatly appreciate the contributions made to this study by Martin Verbeek, and acknowledge the Netherlands Organization for Scientific Research (NWO) for financial support. References 1 Foil, L.D. and Issel, C.J. (1991) Annu. Rev. Entomol. 36, 355–381 2 Hull, R. (1994) Adv. Dis. Vector Res. 10, 361–386 3 Ammar, E.D. (1994) Adv. Dis. Vector Res. 10, 289–331 4 MacFarlane, S.A. and Brown, D.J.F. (1995) J. Gen. Virol. 76, 1299–1304 5 Demler, S.A. (1997) J. Gen. Virol. 78, 511–523 6 Briddon, R.W. et al. (1990) Virology 177, 85–94
VOL. 7 No. 2 FEBRUARY 1999
REVIEWS
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
Brault, V. et al. (1995) EMBO J. 14, 650–659 Pirone, T.P. and Blanc, S. (1996) Annu. Rev. Phytopathol. 34, 227–247 Perry, K.L., Zhang, L. and Palukaitis, P. (1998) Virology 242, 204–210 Schmitt, C. et al. (1998) J. Gen. Virol. 79, 1281–1288 Tomaru, M. et al. (1997) J. Virol. 71, 8019–8023 Omura, T. et al. (1998) J. Virol. 72, 9370–9373 Liu, S. et al. (1997) J. Gen. Virol. 78, 1791–1794 Goldbach, R.W. and Peters, D. (1996) in The Bunyaviridae (Elliot, R.M., ed.), pp. 129–157, Plenum Press Höfer, P. et al. (1997) Virology 236, 288–295 Schmidt, I. et al. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 8885–8889 Atreya, C.D., Raccah, B. and Pirone, T.P. (1990) Virology 178, 161–165 Atreya, C.D. and Pirone, T.P. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 11919–11923 Huet, H. et al. (1994) J. Gen. Virol. 75, 1407–1414 Blanc, S. et al. (1997) Virology 231, 141–147 Peng, Y-H. et al. (1998) J. Gen. Virol. 79, 897–904 Wang, R.Y. et al. (1996) J. Gen. Virol. 77, 861–867 Hunt, R.E., Nault, L.R. and Gingery, R.E. (1988) Phytopathology 78, 499–504 Murphy, F.A. et al., eds (1995) Virus Taxonomy, Springer-Verlag Chu, P.W.G. and Helms, K. (1988) Virology 167, 38–49 Wang, R.Y. et al. (1998) J. Gen. Virol. 79, 1519–1524 Gildow, F.E. (1987) Curr. Top. Vector Res. 4, 93–120 Marsh, M. and Helenius, A. (1989) Adv. Virus Res. 36, 107–151 Bandla, M.D. et al. (1998) Phytopathology 88, 98–104 Kikkert, M. et al. (1998) Phytopathology 88, 63–69 Ludwig, G.V. et al. (1991) Virology 181, 564–571 Berinstein, A. et al. (1995) J. Virol. 69, 2664–2666 Roivainen, M. et al. (1996) Arch. Virol. 141, 1909–1919 Coburn, J. et al. (1998) Infect. Immun. 66, 1946–1952
35 Houk, E.J. et al. (1990) Virus Res. 17, 105–118 36 Mourya, D.T. et al. (1998) Indian J. Med. Res. 107, 10–14 37 Salas-Benito, J.S. and Del Angel, R.M. (1997) J. Virol. 71, 7246–7252 38 Ludwig, G.V. et al. (1996) J. Virol. 70, 5592–5599 39 Wang, K.S. et al. (1992) J. Virol. 66, 4992–5001 40 Peiffer, M.L., Gildow, F.E. and Gray, S.M. (1997) J. Gen. Virol. 78, 495–503 41 Gildow, F.E. and Gray, S.M. (1993) Phytopathology 83, 1293–1302 42 van den Heuvel, J.F.J.M. et al. (1994) J. Gen. Virol. 75, 2559–2565 43 Buchner, P. (1953) Endosymbiose der Tiere mit pflanzlichen Mikroorganismen, Birkhäuser Verlag 44 Unterman, B.M., Baumann, P. and McLean, D.L. (1989) J. Bacteriol. 171, 2790–2794 45 Douglas, A.E. (1998) Annu. Rev. Entomol. 43, 17–37 46 Ohtaka, C., Nakamura, H. and Ishikawa, H. (1992) J. Bacteriol. 174, 1869–1874 47 Filichkin, S.A. et al. (1997) J. Virol. 71, 569–577 48 Hogenhout, S.A. et al. (1998) J. Virol. 72, 358–365 49 Hartl, F.U. (1996) Nature 381, 571–580 50 van den Heuvel, J.F.J.M. et al. (1997) J. Virol. 71, 7258–7265 51 Braig, K. et al. (1994) Nature 371, 578–586 52 Jolly, C.A. and Mayo, M.A. (1994) Virology 201, 182–185 53 Chay, C.A. et al. (1996) Virology 219, 57–65 54 Bruyère, A. et al. (1997) Virology 230, 323–334 55 Nasu, S. (1965) Jpn. J. Appl. Entomol. Zool. 9, 225–237 56 Aksoy, S. (1995) Insect Mol. Biol. 4, 23–29 57 Beard, C.B. et al. (1993) Parasitol. Today 9, 179–183 58 Durvasula, R.V. et al. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 3274–3278 59 Pringle, C.R. (1998) Arch. Virol. 143, 1449–1459
Intracellular determinants of picornavirus replication Raul Andino, Nina Böddeker, Deborah Silvera and Andrea V. Gamarnik
V
iruses travel light. Tak- Viruses replicate in a restricted number of an opportunity to elucidate ing only a small ‘suithosts and tissues. In addition to viral biochemical and cellular funccase’ of essential tools, receptors, several intracellular factors can tions of specific cell types; they are confident of finding be involved in determining tissue tropism. second, the identification of everything they need. Thus, Many proteins have recently been factors essential for virus repliviruses are dependent on the implicated in picornavirus translation and cation will shed light on biochemical processes and struc- RNA replication. Although the functional the molecular basis of viral tural characteristics of the cells role of these proteins has not been pathogenesis. that they infect. Consequently, established in vivo, it is possible that they A good example of viral no virus infects all species or determine cell-type tropism and the tissue tropism is found in polioall tissues of its host, and they pathogenic outcome of the infection. myelitis caused by poliovirus, cause damage in only a few of a member of the picornavirus R. Andino*, N. Böddeker, D. Silvera and many potential target tissues. family. One of the most strikGamarnik are in the Dept of Microbiology and The fact that viral replication A.V. ing features of this disease is Immunology, Box 0414, University of California, is cell-type specific suggests that the resulting neuronal San Francisco, CA 94143-0414, USA. that cellular factors present only damage is highly selective. Al*tel: 11 415 502 6358, fax: 11 415 476 0939, in certain tissues are involved though most of the brain stem e-mail: [email protected] in the viral cycle. These factors centers are involved, other determine tissue tropism and, areas of the brain, including therefore, the type of disease that a particular virus the cerebral cortex and the olfactory, visual and audicauses. Understanding the molecular basis of this tis- tory centers, remain entirely unaffected. The cervical sue tropism has two main purposes: first, it provides and lumbar regions of the spinal cord are highly 0966-842X/99/$ - see front matter © 1999 Elsevier Science. All rights reserved. TRENDS
IN
MICROBIOLOGY
76
PII: S0966-842X(98)01446-2
VOL. 7 No. 2 FEBRUARY 1999
Acknowledgements Support from NIH grants from the US Public Health Service is gratefully acknowledged. References 1 Deepe, G.S. and Bullock, W.E. (1992) in Inflammation: Basic Principles and Clinical Correlates (Gallin, J.I. et al., eds), pp. 943–958, Raven Press 2 Bullock, W.E. and Wright, S.D. (1987) J. Exp. Med. 165, 195–210 3 Newman, S.L. et al. (1990) J. Clin. Invest. 85, 223–230 4 Kimberlin, C.L. et al. (1981) Infect. Immun. 34, 6–10 5 Eissenberg, L.G. and Goldman, W.E. (1987) Infect. Immun. 55, 29–34 6 Wolf, J.E. et al. (1987) J. Immunol. 138, 582–586 7 Wolf, J.E. et al. (1989) Infect. Immun. 57, 513–519 8 Howard, D.H. (1965) J. Bacteriol. 89, 518–523 9 Newman, S.L., Gootee, L. and Gabay, J.E. (1993) J. Clin. Invest. 92, 624–631 10 Eissenberg, L.P., Schlesinger, P.H. and Goldman, W.E. (1988) J. Leukocyte Biol. 43, 483–491 11 Taylor, M.L. et al. (1989) Clin. Exp. Immunol. 75, 466–470 12 Newman, S.L. et al. (1997) J. Immunol. 158, 1779–1786 13 Howard, D.H. (1964) J. Bacteriol. 87, 33–38 14 Newman, S.L. et al. (1991) Infect. Immun. 59, 737–741 15 Newman, S.L. et al. (1992) J. Immunol. 149, 574–580 16 Eissenberg, L.G., Goldman, W.E. and Schlesinger, P.H. (1993) J. Exp. Med. 177, 1605–1611 17 Dautry-Varsat, A., Ciechanover, A. and Lodish, H.F. (1983) Proc. Natl. Acad. Sci. U. S. A. 80, 2258–2262 18 Lestas, A.N. (1976) Br. J. Haematol. 32, 341–350 19 Lane, T.E., Wu-Hsieh, B.A. and Howard, D.H. (1991) Infect. Immun. 59, 2274–2278 20 Newman, S.L. et al. (1994) J. Clin. Invest. 93, 1422–1429
21 Lukacs, G.L., Rotstein, O.D. and Grinstein, S. (1990) J. Biol. Chem. 265, 21009–21017 22 Lukacs, G.L., Rotstein, O.D. and Grinstein, S. (1991) J. Biol. Chem. 266, 24540–24544 23 Zhou, P. et al. (1995) J. Immunol. 155, 785–795 24 Lane, T.E., Wu-Hsieh, B.A. and Howard, D.H. (1993) Infect. Immun. 61, 1468–1473 25 Smith, J.G. et al. (1990) J. Infect. Dis. 162, 1349–1353 26 Zhou, P., Miller, G. and Seder, R.A. (1998) J. Immunol. 160, 1359–1368 27 Lane, T.E., Wu-Hsieh, B.A. and Howard, D.H. (1994) Infect. Immun. 62, 1940–1945 28 Fleischmann, J., Wu-Hsieh, B. and Howard, D.H. (1990) J. Infect. Dis. 161, 143–145 29 Newman, S.L. and Gootee, L. (1992) Infect. Immun. 11, 4593–4597 30 Sturgill-Koszycki, S. et al. (1994) Science 263, 678–681 31 Russell, D.G., Dant, J. and Sturgill-Koszycki, S. (1996) J. Immunol. 156, 4764–4773 32 Sturgill-Koszycki, S., Shiable, U.E. and Russell, D.G. (1996) EMBO J. 15, 6960–6968 33 Xu, S. et al. (1994) J. Immunol. 153, 2568–2578 34 Hackam, D.J. et al. (1997) J. Biol. Chem. 272, 29810–29820 35 Hanano, R. and Kaufmann, S.H. (1995) Immunol. Lett. 45, 23–27 36 Chan, J. et al. (1992) J. Exp. Med. 175, 1111–1122 37 Rook, G.A. et al. (1987) Immunology 67, 229–234 38 Schneemann, M. et al. (1993) J. Infect. Dis. 167, 1358–1363 39 Bonecini-Almeida, M.G. et al. (1998) J. Immunol. 160, 4490–4499 40 Huang, S. et al. (1993) Science 259, 1742–1745 41 Cooper, A.M. et al. (1993) J. Exp. Med. 178, 2243–2247 42 Flynn, J. et al. (1993) J. Exp. Med. 178, 2249–2254 43 Newport, M.J. et al. (1996) New Engl. J. Med. 335, 1941–1949 44 Jouanguy, E. et al. (1996) New Engl. J. Med. 335, 1956–1961
Recognition and receptors in virus transmission by arthropods Johannes F.J.M. van den Heuvel, Saskia A. Hogenhout and Frank van der Wilk
M
ore than 70% of Fundamental knowledge of the molecular propagative. However, the viruses infecting plants mechanisms underlying virus transmission majority of plant viruses is and .40% of viruses by arthropods is a prerequisite for the transmitted in a noncirculative infecting mammals use arthrocreation of new strategies to modulate manner; i.e. they do not replipod vectors to move from one vector competence. There have been several cate in their vectors, as they host to another. As a result, recent advances in identifying the viral and are only associated with the the health of .2.5 billion peovector determinants involved in virus cuticular linings of the mouthple, mainly from tropical and recognition, attachment and retention. parts and the anterior part of subtropical regions, and the the alimentary tract. Generally, J.F.J.M. van den Heuvel*, S.A. Hogenhout and production of agricultural comsuch plant viruses are mechaniF. van der Wilk are in the Dept of Virology, DLO modities worldwide are at Research cally transmissible as well. Institute for Plant Protection (IPO-DLO), PO stake. Not only do arthropodMechanical transmission by Box 9060, 6700 GW Wageningen, The Netherlands. borne viruses have to cross epiarthropods has been reported *tel: 131 317 476141, fax: 131 317 410113, thelial cell linings of the gut for a few mammalian viruses, e-mail: [email protected] and salivary glands but they also including members of the have to resist the potentially Retroviridae1. hostile environment within their vectors. Almost all Phloem-feeding insects, such as aphids, whiteflies, mammalian viruses and some plant viruses replicate leafhoppers and planthoppers, and haematophagous in their vector during this cyclic passage; i.e. they are mosquitoes and ticks play a prominent role in the 0966-842X/99/$ - see front matter © 1999 Elsevier Science. All rights reserved. TRENDS
IN
MICROBIOLOGY
71
PII: S0966-842X(98)01434-6
VOL. 7 No. 2 February 1999
REVIEWS
Table 1. Virus families and genera59 containing species transmitted in a circulative manner by arthropods Family, genus
Hosta
Arthropod vector
Propagative
Bunyaviridae Bunyavirus Nairovirus Phlebovirus Tospovirus
M M M P
Mosquitoes, ticks, ceratopogonids, phlebotomines Ticks Phlebotomines, mosquitoes, ceratopogonids Thrips
Yes Yes Yes Yes
Circoviridae Nanovirus
P
Aphids
No
Flaviviridae Flavivirus
M
Mosquitoes, ticks
Yes
Geminiviridae Mastrevirus Curtovirus Begomovirus
P P P
Leafhoppers (Cicadellidae) Leafhoppers, treehoppers (Membracidae) Whiteflies (Bemisia tabaci)
No No No?
Luteoviridae Luteovirus Polerovirus Enamovirus
P P P
Aphids Aphids Aphids
No No No
Reoviridae Fijivirus Phytoreovirus Oryzavirus Orbivirus
P P P M
Yes Yes Yes Yes
Coltivirus
M
Planthoppers (Delphacidae) Leafhoppers Planthoppers Anopheline and culicine mosquitoes, sandflies (Phlebotomus spp.), gnats, ticks Ticks
Rhabdoviridae Cytorhabdovirus Nucleorhabdovirus Ephemerovirus Unassigned Rhabdoviruses
P P M P
Leafhoppers, aphids Leafhoppers, aphids Culicoides, mosquitoes Lacewings, mites, leafhoppers, aphids
Yes Yes Yes
Togaviridae Alphavirus
M
Mosquitoes, Cimidae
Yes
Soft ticks (Ornithodoros spp.)
Yes
Leafhoppers Planthoppers Aphids
Yes Yes? No
Genera not yet assigned to families ‘African swine fever-like M viruses’ Marafivirus P Tenuivirus P Umbravirus P
Yes
a
Species infects either plant (P) or mammalian (M) hosts.
dissemination of arthropod-borne viruses (Table 1). The route of some of these viruses in the invertebrate vector and the transmission barriers encountered have been well described, providing the groundwork for recent studies on the molecular basis of virus transmission by arthropods. The high degree of vector specificity and differences in tissue tropism and retention sites among arthropod-borne viruses suggest an intimate association in which both virus-encoded proteins and vector components are involved. Identification of the vector determinants involved in the different phases of the transmission process represents a nascent, but rapidly developing, research area. Despite
TRENDS
IN
MICROBIOLOGY
72
the vast body of literature dealing with host receptors for viruses, only limited information is available on vector proteins involved in recognition, attachment and retention of viruses, and only recently have insect proteins with a capacity to bind plant and mammalian viruses been identified. What are the viral determinants for transmission? It has long been recognized that the viral determinants governing transmissibility and attachment of viruses are primarily located in the structural proteins of a virus particle2,3. The domains involved are usually identified by comparative studies of the encoding
VOL. 7 No. 2 February 1999
REVIEWS
genes of transmissible and nontransmissible forms4,5. The latter often arise naturally when vector pressure is lifted; i.e. a virus is transmitted or maintained in the absence of its natural vector. A functional role for the transmission domains can be confirmed by mutational analysis of the viral genome or infectious viral cDNA clones6–10. Circulative viruses Although transmission determinants of many circulative plant viruses are known, we will limit our discussion to three recent examples, which represent taxa with different structural features (Table 1). Long-term vegetative propagation of rice infected with rice dwarf virus (Phytoreovirus sp.) results in the development of a transmission-defective isolate that lacks the outer capsid protein, P2, owing to a point mutation in the genome segment encoding this protein11. P2 is involved in the adsorption of the virus on insect cells12. Insect transmissibility of African cassava mosaic virus (Begomovirus sp.) and tomato spotted wilt virus (TSWV) (Tospovirus sp.) is also lost quite rapidly after several rounds of mechanical inoculation of the virus, because of defects in the coat protein (CP)13 or lack of the lipid envelope14, respectively. The CP of a geminivirus is the basic determinant of virus acquisition, although other gene products can influence the accumulation of the virus in the whitefly13,15. Moreover, the CP is involved in the molecular recognition of the virus by its vector. Exchanging the CP gene of the African cassava mosaic virus (Begomovirus sp.) with that of beet curly top virus (Curtovirus sp.) alters the insect specificity of the former from whiteflies to leafhoppers6. Thrips transmission of the envelope-minus TSWV isolate, NL-04, which is highly infectious upon mechanical inoculation, has so far failed, suggesting that structural components in the envelope, such as the glycoproteins G1 and G2, might be involved in recognition of a receptor in the thrips midgut14. Noncirculative viruses The transmission of some noncirculative viruses not only depends on the viral capsid but also on the presence of an active virus-encoded helper factor that is not a constituent of the virus particle8. Helper factors of the potyviruses and caulimoviruses are among the best-characterized viral products. They are bifunctional molecules capable of interaction with both the virus particle and the surface of the stylet food canal of the aphid via two separate domains and must be acquired by the vector before or during virus acquisition8. Single amino acid substitutions in the carboxyterminal region of the caulimovirus aphid transmission factor (ATF) have revealed that amino acids at positions 157 (Ile) and 159 (Gly) determine the interaction between the virus and the helper factor16. However, recent work has demonstrated that a second transmission factor, a 15-kDa protein encoded by gene III of cauliflower mosaic virus (CaMV), is involved in caulimovirus transmission as well. In vitro binding assays have shown that the carboxy-terminal region of ATF interacts with the amino terminus of
TRENDS
IN
MICROBIOLOGY
73
P15. This might indicate that the CaMV ATF does not directly interact with the viral CP but with P15, which is associated with purified CaMV particles [S. Blanc (1998) 7th International Congress of Plant Pathology, Edinburgh, UK, Abstr. 1.13.2S]. For potyviruses, the key amino acids functionally involved in binding and transmission are located in the conserved DAG motif and neighbouring residues of the capsid protein and the KITC- and PTK-containing domains of the helper factor17–22. Helper factor dependency has also been reported for noncirculative picornavirus-like plant viruses8,23,24 and has been suggested to be important for several other viruses, including the circulatively transmitted Nanovirus spp.25, which lose their ability to be transmitted by their vectors after purification from plant tissue. Helper factors of potyviruses have been shown to mediate virus retention in the food canal of the vector and vector-specific transmission26. Receptor sites for helper factors have not yet been identified. However, in recent years, significant advances have been made in identifying vector constituents involved in the transmission of viruses that are carried systemically by arthropods. Putative receptors for circulative viruses Binding between a virion and a cell surface receptor provides the initial physical association required for virus entry into vector cells. By analogy to virus entry into other animal cells, viruses can enter insect cells either by receptor-mediated endocytosis, generally through clathrin-coated vesicles, as has been proposed for luteoviruses27, or by direct fusion of the viral envelope with the cell membrane28. Based on observations made using electron microscopy, the latter mechanism has been proposed for bunyaviruses. Glycoproteins present in the membrane of the enveloped bunyaviruses have been suggested to mediate attachment29–31. Virus overlay assays have been instrumental in revealing vector proteins with a virus-binding capacity. In these assays, extracts of the vector are electrophoretically separated, blotted and probed with purified virus or anti-idiotypic antibodies, which are subsequently detected by antibodies. Anti-idiotypic antibodies mimicking the glycoproteins of TSWV specifically label a 50-kDa band in thrips homogenates on western blots and on the plasma membrane of the larval thrips midgut29. As purified virus displays an affinity for a similar-sized protein in an overlay assay of thrips proteins, the 50-kDa protein might be a cellular receptor29. A second thrips protein (~94 kDa) that has a TSWV-binding capacity has also been identified30. This protein firmly binds the G2 glycoprotein in virus overlay assays. G2 contains the highly conserved amino acid sequence Arg–Gly–Asp (RGD) near the amino terminus. This sequence is an important determinant for cellular attachment of several mammalian viruses and pathogens, including foot-and-mouth disease virus (FMDV)32, human coxsackievirus A9 (CAV-9)33 and the spirochete Borrelia burgdorferi34, which is the causal agent of Lyme disease. Owing to
VOL. 7 No. 2 February 1999
REVIEWS
the protease sensitivity of TSWV G2 and the high concentration of proteolytic enzymes that probably occur in the gut lumen, G2 is not expected to be involved in cellular attachment in the midgut of thrips30. However, studies on CAV-9 have shown that, even when the RGD-containing capsid domain is cleaved off, the virus remains infectious because it can bypass the RGD-recognizing vitronectin receptor33. Likewise, TSWV might employ alternative routes to enter thrips midgut cells, depending on its phenotype, which in turn might be regulated by thrips midgut proteases. Further characterization and localization of the 94kDa protein are required to elucidate its functional role in the cyclic passage of TSWV in thrips. In mosquitoes, putative receptor proteins for membrane-bound viruses of the genus Alphavirus (Table 1), such as chikungunya virus (CHIKV) and western equine encephalitis virus (WEEV), have been located in mosquito midguts35,36. CHIKV-binding activity is noticeably higher in the brush border membrane fraction of midguts derived from susceptible mosquitoes than in those derived from refractory insects36. Five different proteins from the brush border fraction and from the membrane fraction of Aedes albopictus
(a)
(b)
asg
psg
vp s
GroEL
mg (c)
m m hg
Trends in Microbiology
Fig. 1. (a) Diagram showing the circulative nature of a luteovirus in the aphid, and (b) the putative role of Buchnera GroEL in virus persistence. Virions are ingested with phloem sap from infected plants, move through the intestine and are transcellularly transported through the midgut (mg) or hindgut (hg) into the haemolymph, in which they persist for the life span of the aphid. The presence of Buchnera GroEL and the ability of the virus to interact with this protein determine the persistent nature of the luteovirus in the haemolymph of the vector. Upon contact with the accessory salivary gland (asg), virus particles can be transported through this gland, eventually arriving in the salivary duct, from which they are secreted with the saliva when the aphid feeds. (c) Electron micrograph of Buchnera cells in a mycetocyte (m). Abbreviations: psg, principal salivary gland; s, stomach; vp, virus particle.
TRENDS
IN
MICROBIOLOGY
74
C6/36 cells are known to bind CHIKV (Ref. 36). Viral attachment studies using this cell line further indicate that several surface proteins can serve as receptors for dengue type 4 virus37 (genus Flavivirus) and Venezuelan equine encephalitis virus (VEEV; Alphavirus sp.)38 (Table 1). There is strong evidence for the involvement of a 32-kDa laminin-binding protein in C6/36–VEEV interactions38. Laminin receptor proteins of different origins are highly conserved39, which might explain the broad range of vectors (several genera of mosquitoes and ticks) transmitting VEEV. Receptors accounting for the high degree of vector specificity that is commonly observed among invertebrates transmitting viruses, bacteria or parasites of plants and mammals still remain to be identified. It will be of great importance to reveal the true nature of these vector determinants because they offer new opportunities for disease control, including the genetic manipulation of the vector and transmission neutralization by means of vaccines or recombinant proteins expressed in transgenic plants. Towards identification of luteovirus receptors Species of the family Luteoviridae (Table 1) are solely transmitted by aphids in a circulative and non-propagative manner (Fig. 1). Based on ultrastructural studies of luteoviruses in vector and non-vector aphids, it has been postulated that virions are transcellularly transported through epithelial cell linings in the gut and salivary glands by receptor-mediated endocytosis–exocytosis27. The haemolymph acts as a reservoir in which acquired virus particles are retained in an infective form without replication for the life span of the aphid. Upon contact with the basal lamina of the accessory salivary gland, virus particles can be transported through this gland40, eventually arriving in the salivary duct, from which they are excreted with the saliva when the aphid feeds41. Vector specificity seems to be determined at the level of the accessory salivary gland40,41. Several aphid proteins, with relative molecular masses ranging from 31 to 85 kDa, have been shown to immobilize purified luteoviruses in vitro42 [X. Wang and G. Zhou (1998) 7th International Congress of Plant Pathology, Edinburgh, UK, Abstr. 1.13.16]. For example, a barley yellow dwarf virus-MAV-like isolate (BYDV-MAV; genus Luteovirus) [X. Wang and G. Zhou (1998) 7th International Congress of Plant Pathology, Edinburgh, UK, Abstr. 1.13.16] from China displays a strong affinity for two proteins of 31 and 44 kDa (P31 and P44, respectively) from the vector aphids Sitobion avenae and Schizaphis graminum but not from Rhopalosiphum padi. R. padi is unable to transmit BYDV-MAV. Antisera raised against P31 and P44 react specifically with extracts of the accessory salivary glands of vector aphids, suggesting that these proteins might be involved in luteovirus-specific recognition at the accessory salivary gland. The best-characterized protein that binds species from all three genera of the family Luteoviridae (Table 1) is GroEL (also known as symbionin),
VOL. 7 No. 2 FEBRUARY 1999
REVIEWS
which is abundantly produced by Buchnera spp., the primary endosymbiotic bacterium of aphids. The Gram-negative Buchnera spp. (Fig. 1c) are present in specialized polyploid host cells, called mycetocytes43, in the haemocoel of the aphid (Fig. 1a). The endosymbionts are distantly related to Escherichia coli44 and are maternally inherited, and the association, which has a nutritional basis, is obligatory for both the aphid and the bacterium45. Buchnera GroEL homologues are found in most aphid taxa, are immunologically closely related and share .80% sequence identity with the E. coli heat-shock protein GroEL, a member of the Hsp60 family of molecular chaperones42,46–48. Chaperonins are essential for cell viability, as they bind and stabilize newly translated or translocated aggregation-prone polypeptides and mediate their functional folding and assembly in an ATP-dependent manner49. Structural and functional characteristics of Buchnera GroEL are very similar to those of E. coli GroEL (Refs 37–48); however, unlike E. coli GroEL, Buchnera GroEL is secreted and is readily detected in the aphid haemolymph47. Native Buchnera GroEL, like E. coli GroEL, is an oligomer of ~840 kDa that consists of 14 identical subunits of 60 kDa arranged in two stacked heptameric rings47,48,50,51 (Fig. 1b). Both the subunits and the native 14-meric protein have been shown to bind luteoviruses in different ligand assays42,47,50 irrespective of whether GroEL is derived from vector or nonvector aphid species or from E. coli47,50. Pea enation mosaic virus (PEMV; genus Enamovirus) also has a high affinity for native GroEL homologues. Mutational analysis of the gene coding for Buchnera GroEL of Myzus persicae, the major vector of potato leafroll virus (PLRV; genus Luteovirus), shows that the virus-binding site is located in the equatorial domain of the subunit48. The minor capsid protein of a luteovirus, the readthrough domain (RTD), determines the interaction with GroEL (Refs 47,50). The RTD is exposed on the surface of a luteovirus particle7 and contains determinants necessary for virus transmission by aphids7,52–54. Sequence comparison of the RTDs of different luteoviruses and PEMV has identified highly conserved amino acid residues in the amino-terminal region of the RTD, which are potentially important in the interaction with Buchnera GroEL (Ref. 50). Direct injection of beet western yellows virus (BWYV; Polerovirus sp.) mutants into M. persicae has shown that virions devoid of RTD, which are unable to interact with Buchnera GroEL, are significantly less persistent in the aphid haemolymph than wild-type virions50. Moreover, treatment of M. persicae larvae with antibiotics that interfere with prokaryotic protein synthesis significantly reduces Buchnera GroEL levels in the haemolymph, inhibits transmission and results in the loss of capsid integrity in the haemolymph42. Collectively, these observations indicate that the luteovirus–GroEL interaction retards proteolytic breakdown and is essential for virus retention in the haemolymph of the aphid42,50.
TRENDS
IN
MICROBIOLOGY
75
Questions for future research • What are the cellular factors determining vector specificity of a virus? • Can mimics of viral attachment proteins be exploited to develop efficient strategies to modulate vector competence? • How do circulative viruses escape from the host’s immune response?
Does GroEL-mediated retention apply to other systems? Avidity for Buchnera GroEL appears to be a common characteristic of virus species from the family Luteoviridae. It is reasonable to assume that GroELmediated retention also applies to viruses of the genus Umbravirus (Table 1), because umbravirus RNA can only be transmitted by aphids if it is packaged in the capsid of a helper virus, commonly a luteovirus or PEMV (Ref. 25). Recent unpublished work from Morin and co-workers [S. Morin et al. (1998) Second International Workshop on Geminiviruses and their Vectors, San Juan, Puerto Rico, Abstr.] has shown that the Israeli strain of tomato yellow leafcurl virus (TYLCV; Begomovirus sp.) interacts with an endosymbiotic GroEL homologue from its whitefly vector Bemisia tabaci. Whiteflies harbour mycetome endosymbionts, as do most homopteran insects43. Whiteflies feeding on anti-Buchnera GroEL antiserum prior to virus acquisition have been shown to reduce the transmission of TYLCV by .80% relative to control insects feeding on normal serum. Active antibodies were recovered from the insect haemolymph, providing almost conclusive evidence that the antibody interferes with the interaction between TYLCV and the GroEL homologue. Endosymbiotic bacteria from leafhoppers have also been implicated in the vertical transmission of rice dwarf virus55. Similarly, tsetse flies, which are important vectors of human and animal trypanosomal diseases, have evolved symbiotic associations that have similar features to those of aphids, including the overproduction of GroEL by mycetome endosymbionts56. The possibility that endosymbiotic GroEL might have additional functions in the physiology of tsetse flies or in their competence to act as vectors of protozoan parasites requires further study. Molecular characterization of the endosymbiont–host interactions, and the genetic transformation of these bacteria with antiparasite or anti-viral constructs, could provide a significant tool for modulation of vector competence57,58. Acknowledgements We greatly appreciate the contributions made to this study by Martin Verbeek, and acknowledge the Netherlands Organization for Scientific Research (NWO) for financial support. References 1 Foil, L.D. and Issel, C.J. (1991) Annu. Rev. Entomol. 36, 355–381 2 Hull, R. (1994) Adv. Dis. Vector Res. 10, 361–386 3 Ammar, E.D. (1994) Adv. Dis. Vector Res. 10, 289–331 4 MacFarlane, S.A. and Brown, D.J.F. (1995) J. Gen. Virol. 76, 1299–1304 5 Demler, S.A. (1997) J. Gen. Virol. 78, 511–523 6 Briddon, R.W. et al. (1990) Virology 177, 85–94
VOL. 7 No. 2 FEBRUARY 1999
REVIEWS
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
Brault, V. et al. (1995) EMBO J. 14, 650–659 Pirone, T.P. and Blanc, S. (1996) Annu. Rev. Phytopathol. 34, 227–247 Perry, K.L., Zhang, L. and Palukaitis, P. (1998) Virology 242, 204–210 Schmitt, C. et al. (1998) J. Gen. Virol. 79, 1281–1288 Tomaru, M. et al. (1997) J. Virol. 71, 8019–8023 Omura, T. et al. (1998) J. Virol. 72, 9370–9373 Liu, S. et al. (1997) J. Gen. Virol. 78, 1791–1794 Goldbach, R.W. and Peters, D. (1996) in The Bunyaviridae (Elliot, R.M., ed.), pp. 129–157, Plenum Press Höfer, P. et al. (1997) Virology 236, 288–295 Schmidt, I. et al. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 8885–8889 Atreya, C.D., Raccah, B. and Pirone, T.P. (1990) Virology 178, 161–165 Atreya, C.D. and Pirone, T.P. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 11919–11923 Huet, H. et al. (1994) J. Gen. Virol. 75, 1407–1414 Blanc, S. et al. (1997) Virology 231, 141–147 Peng, Y-H. et al. (1998) J. Gen. Virol. 79, 897–904 Wang, R.Y. et al. (1996) J. Gen. Virol. 77, 861–867 Hunt, R.E., Nault, L.R. and Gingery, R.E. (1988) Phytopathology 78, 499–504 Murphy, F.A. et al., eds (1995) Virus Taxonomy, Springer-Verlag Chu, P.W.G. and Helms, K. (1988) Virology 167, 38–49 Wang, R.Y. et al. (1998) J. Gen. Virol. 79, 1519–1524 Gildow, F.E. (1987) Curr. Top. Vector Res. 4, 93–120 Marsh, M. and Helenius, A. (1989) Adv. Virus Res. 36, 107–151 Bandla, M.D. et al. (1998) Phytopathology 88, 98–104 Kikkert, M. et al. (1998) Phytopathology 88, 63–69 Ludwig, G.V. et al. (1991) Virology 181, 564–571 Berinstein, A. et al. (1995) J. Virol. 69, 2664–2666 Roivainen, M. et al. (1996) Arch. Virol. 141, 1909–1919 Coburn, J. et al. (1998) Infect. Immun. 66, 1946–1952
35 Houk, E.J. et al. (1990) Virus Res. 17, 105–118 36 Mourya, D.T. et al. (1998) Indian J. Med. Res. 107, 10–14 37 Salas-Benito, J.S. and Del Angel, R.M. (1997) J. Virol. 71, 7246–7252 38 Ludwig, G.V. et al. (1996) J. Virol. 70, 5592–5599 39 Wang, K.S. et al. (1992) J. Virol. 66, 4992–5001 40 Peiffer, M.L., Gildow, F.E. and Gray, S.M. (1997) J. Gen. Virol. 78, 495–503 41 Gildow, F.E. and Gray, S.M. (1993) Phytopathology 83, 1293–1302 42 van den Heuvel, J.F.J.M. et al. (1994) J. Gen. Virol. 75, 2559–2565 43 Buchner, P. (1953) Endosymbiose der Tiere mit pflanzlichen Mikroorganismen, Birkhäuser Verlag 44 Unterman, B.M., Baumann, P. and McLean, D.L. (1989) J. Bacteriol. 171, 2790–2794 45 Douglas, A.E. (1998) Annu. Rev. Entomol. 43, 17–37 46 Ohtaka, C., Nakamura, H. and Ishikawa, H. (1992) J. Bacteriol. 174, 1869–1874 47 Filichkin, S.A. et al. (1997) J. Virol. 71, 569–577 48 Hogenhout, S.A. et al. (1998) J. Virol. 72, 358–365 49 Hartl, F.U. (1996) Nature 381, 571–580 50 van den Heuvel, J.F.J.M. et al. (1997) J. Virol. 71, 7258–7265 51 Braig, K. et al. (1994) Nature 371, 578–586 52 Jolly, C.A. and Mayo, M.A. (1994) Virology 201, 182–185 53 Chay, C.A. et al. (1996) Virology 219, 57–65 54 Bruyère, A. et al. (1997) Virology 230, 323–334 55 Nasu, S. (1965) Jpn. J. Appl. Entomol. Zool. 9, 225–237 56 Aksoy, S. (1995) Insect Mol. Biol. 4, 23–29 57 Beard, C.B. et al. (1993) Parasitol. Today 9, 179–183 58 Durvasula, R.V. et al. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 3274–3278 59 Pringle, C.R. (1998) Arch. Virol. 143, 1449–1459
Intracellular determinants of picornavirus replication Raul Andino, Nina Böddeker, Deborah Silvera and Andrea V. Gamarnik
V
iruses travel light. Tak- Viruses replicate in a restricted number of an opportunity to elucidate ing only a small ‘suithosts and tissues. In addition to viral biochemical and cellular funccase’ of essential tools, receptors, several intracellular factors can tions of specific cell types; they are confident of finding be involved in determining tissue tropism. second, the identification of everything they need. Thus, Many proteins have recently been factors essential for virus repliviruses are dependent on the implicated in picornavirus translation and cation will shed light on biochemical processes and struc- RNA replication. Although the functional the molecular basis of viral tural characteristics of the cells role of these proteins has not been pathogenesis. that they infect. Consequently, established in vivo, it is possible that they A good example of viral no virus infects all species or determine cell-type tropism and the tissue tropism is found in polioall tissues of its host, and they pathogenic outcome of the infection. myelitis caused by poliovirus, cause damage in only a few of a member of the picornavirus R. Andino*, N. Böddeker, D. Silvera and many potential target tissues. family. One of the most strikGamarnik are in the Dept of Microbiology and The fact that viral replication A.V. ing features of this disease is Immunology, Box 0414, University of California, is cell-type specific suggests that the resulting neuronal San Francisco, CA 94143-0414, USA. that cellular factors present only damage is highly selective. Al*tel: 11 415 502 6358, fax: 11 415 476 0939, in certain tissues are involved though most of the brain stem e-mail: [email protected] in the viral cycle. These factors centers are involved, other determine tissue tropism and, areas of the brain, including therefore, the type of disease that a particular virus the cerebral cortex and the olfactory, visual and audicauses. Understanding the molecular basis of this tis- tory centers, remain entirely unaffected. The cervical sue tropism has two main purposes: first, it provides and lumbar regions of the spinal cord are highly 0966-842X/99/$ - see front matter © 1999 Elsevier Science. All rights reserved. TRENDS
IN
MICROBIOLOGY
76
PII: S0966-842X(98)01446-2
VOL. 7 No. 2 FEBRUARY 1999