- Email: [email protected]
Initiation and spread of α-herpesvirus infections
NEWS
AND
COMMENT
not elicit an HR in the host plant? Are harpins the only elicitors of nonhost HR in tobacco and possibly in other plants? Is the same mechanism used in tobacco to recognize both the Erwinia and the P. s. pv. syringae harpins? Is host resistance different in mechanism from nonhost resistance? Answers to this fascinating puzzle require the identification of more HR elicitors and their putative plant receptors. References 1 Mekalanos, J.J. (1992) J. Bacterial. 174, l-7 2 Long, S.R. and Staskawicz, B. (1993)
174,6878-6885 10 Van Gijsegem, F., Genin, S. and Boucher, C. (1993) Trends Microbial. 1, 175-180 11 Huang, H-C. et al. (1993) Mol. Plant-Microbe Interact. 6,515-520 12 Smith, M.J. et al. (1993) Tetrahedron Lett. 34,223-226 13 Wei, Z-M. et al. (1992) Science 257, 85-88 14 He, S.Y., Huang, H-C. and Collmer, A. (1993) Cell 73,1255-1266 1.5 Huang, H-C. et al. (1988) 1. Bacterial. 170,4748-4756 16 Beer, S.V. et al. (1991) in Advances in Molecular Genetics of Plant-Microbe Interactions (Hennecke, H. and Verma, D.P.S., eds), pp. 53-60, Kluwer Academic Publishers
Cell 73,921-935 3 Willis, D.K., Rich J.J. and Hrabak, E.M. (1991) Mol. Plant-Microbe Interact. 4, 132-138 4 Klement, Z. (1982) in Phytopathogenic Prokaryotes (Vol. 2) (Mount, M.S. and Lacy, G.H., eds), pp. 149-177, Academic Press 5 Keen, N.T. (1992) Plant Mol. Biol. 19, 109-122 6 Lindsay, W.P., Lamb, C.J. and Dixon, R.A. (1993) Trends Microbial. 1, 181-187 7 Fenselau, S., Balbo, I. and Bonas, U. (1992) Mol. Plant-Microbe Interact. 5, 390-396 8 Gough, C.L. et al. (1992) Mol. Plant-Microbe Interact. 5,384-389 9 Huang, H-C. et al. (1992) J. Bacterial.
Initiationand spread of a-herpesvirus infections Thomas C. Mettenleiter
H
erpesviruses are large animal viruses with a DNA genome varying from approximately 120 to 250kb. Based on their biological properties, the Herpesviridae have been divided ‘into three subfamilies, the a-, pand y-herpesvirinae, prototypes of which are the human pathogens herpes simplex virus (HSV), (HCMV) and cytomegalovirus Epstein-Barr virus (EBV), respectively. As enveloped viruses, they depend on two consecutive processes for infectious entry into target cells: (1) attachment of free virions to cells and (2) penetration, that is, fusion of virion envelope and cellular cytoplasmic membrane leading to release of the nucleocapsid into the cell. Virion envelope glycoproteins play important roles in both these processes (see Refs 1,2 for recent reviews). After infection of primary target cells, virus spread can occur by several different mechanisms. Infected cells may release infectious
virions that reinitiate infection from outside. In addition, direct viral cell-to-cell spread from primary infected cells to adjacent noninfected cells may occur. In the host, virus may be disseminated by circulating infected cells that adhere to noninfected tissues and transmit infectivity directly. Recent results on HSV and pseudorabies virus (PrV) shed more light on these processes in a-herpesviruses. PrV causes Aujeszky’s disease in swine, which is characterized by nervous and respiratory symptoms, and reproductive failure. Unlike HSV, PrV is not pathogenic for humans. However, the two viruses have several features in common, including a broad host range in vitro, and several species besides the natural host can be infected experimentally. In addition, all of the known PrV glycoproteins are
related to homologous teins in HSV (Ref. 1)“.
Attachment Binding of free infectious virus to target cells involves interactions between virion envelope glycoproteins and cellular virus receptors. Herpes virions contain a large number of different virus-encoded envelope glycoproteins that might participate in attachment. A wellknown example of a cellular herpesvirus receptor is the B-cell membrane protein CR2 (CD2 1 ), which binds EBV (Ref. 3). Recent studies have demonstrated that several a- (reviewed in Ref. l), p- and ‘yherpesviruses4J bind to their target cells by interaction of virion components with cell-surface glycosaminoglycans, principally heparan sulfate (HS)6.
“At the 18th International Herpesvirus Workshop, a common nomenclature for a-herpesvirus glycoproteins was agreed on, based on designations of HSV glycoproteins. This nomenclature is used here.
T.C. Mettenleiter is in the Federal Research Centre for Virus Diseases of Animals, PO Box 1149, D-72001 Tiibingen, Germany. 0 1994 Elsevier Science Ltd 0966 842X/94/$07.00
TRENDS
IN MICROBIOLOGY
2
VOL.
2
No.
glycopro-
1
JANUARY
1994
NEWS
In the a-herpesviruses, the major HS-binding glycoproteins are glycoprotein (g) C homologs7,8. Surprisingly, gC homologs are not essential for virus infectivity, indicating either that other viral glycoproteins also recognize HS or that a completely gC- and HS-independent pathway of virus attachment exists. In HSV, the essential glycoprotein gB also binds HS under physiological conditions’. However, PrV gB does not bind HS, although it may interact with PrV gC and indirectly participate in HS binding*. Therefore, there is a gC- and HS-independent mode of virus attachment for PrV. Kinetic analyses demonstrate that the primary interaction between the virion and the target cell, which is initially sensitive to competition by exogenous heparin (an HSrelated compound), converts into a heparin-resistant attachment. Glycoproteins gC and gD participate in this secondary binding, which appears to be specifically mediated by the interaction of the PrV and HSV gD homolog with a second cellular component9J0 (see Fig. 1). Adhesion of Pi-V-infected cells in suspension to noninfected cells exhibits similar characteristics to the binding of free virions”: specific adhesion is sensitive to heparin and to antibodies against PrV gC, but not to antibodies against PrV gD. In the absence of gC, specific adhesion drops by 90%, correlating with results for gC-negative PrV virions8. Interestingly, antibodies against PrV gD inhibit adhesion in the absence of gC. Hence, attachment of free PrV virions and specific adhesion of virus-infected cells occur by similar if not identical mechanisms, mainly involving glycoproteins gC and gD. Penetration After attachment, direct fusion between the herpes virion envelope and the cytoplasmic membrane occurs at neutral pH. All the viral glycoproteins involved in this process (the gB, gD, gH and gL homologs) are essential for infectivity of free virions (see Refs 1,2 for reviews), with gH and gL forming
a complex 12. The conservation of the gB, gH and gL homologs throughout the herpesviruses indicates that membrane fusion processes might be similar for all herpesviruses. Indeed, within the a-herpesviruses, complementation of lethal mutations in several gB homologs can be achieved with gB proteins from heterologous viruses2. In HCMV, there is evidence that gH may interact with a cellular receptor13 and, since herpesvirus gH homologs are closely related, similar interactions may also occur in other herpesviruses. However, a ‘classical’ fusion protein has not yet been identified in herpesviruses, although heterologous expression of gB and gD homologs in cells induced them to fuse more readily (reviewed in Ref. 1). It is likely that several virion glycoproteins acting together are necessary for successful membrane fusion. Direct cell-to-cell spread The mechanisms leading to direct virus transfer from primary infected cells to adjacent noninfected cells are still unclear. In HSV, all the glycoproteins that are necessary for penetration are also essential for direct viral cell-to-cell spread, suggesting that these processes are similar. Results from PrV, however, show that PrV gD is required only for the infectivity of free virions and not for cell-to-cell spread in culture (reviewed in Ref. 2) or transneuronal spread in u~vo’~. This shows that penetration and cell-to-cell spread, although related, are distinct and are probably regulated differently. It also indicates that cell-to-cell spread and transneuronal transfer may occur by very similar mechanisms. Whether gD is necessary for PrV spread via infected circulating cells remains to be determined. Multiple virus-cell interactions in infection and spread Initiation of infection and spread of ol-herpesviruses appear to be complex processes involving multiple sequential (or simultaneous) interactions between different viral and cellular components, the details of which are only beginning
AND
COMMENT
HSV !3B gc CID
I
HS
?
11 HS ?
Fig. 1. Involvement of envelope glycoproteins of herpes simplex virus (HSV) and pseudorabies virus (PrV) in attachment. The figure shows binding of glycoproteins gB, gC and gD to heparan sulfate (HS) or to another unidentified cellular receptor (?), and the putative interaction between gB and gC in PrV (open arrow).
emerge1,2J5. PrV and HSV both use HS for primary attachment and may even recognize similar secondary receptors 16. What determines their different natural host ranges in viva is still unclear. As more herpesviral envelope proteins are identified, these processes may turn out to be even more complex. However, initiation of infection by free virions, virus-infected circulating cells, and direct viral cellto-cell spread seem to occur by related, but distinct, mechanisms. Interference with these mechanisms of infection might help to inhibit primary infection, block viral spread and concomitantly prevent the onset of disease. to
Acknowledgements Work in my laboratory was supported by the Deutsche Forschungsgemeinschaft. I thank PG. Spear for critical reading of the manuscript.
References 1 Spear, P.G. (1993) Semin. Viral. 4, 167-180 2 Mettenleiter, T.C. (1991) Comp. lmmunol. Microbial. Infect. Dis. 14, 151-163 3 Nemerow, G. and Cooper, N. (1992) Semin. Viral. 3,117-124 4 Kari, B. and Gehrz, R. (1992) 1. Virol. 66,1761-1764 5 Vanderplasschen, A. et al. (1993) Virology 196,232-240 6 Shieh, M-T. et al. (1992) J. Cell Biol. 116,1273-1281 7 Herold, B. et al. (1991) 1. Viral. 65, 1090-1098 8 Mettenleiter, T.C. et al. (1990) 1. Virol. 64,278-286 9 Karger, A. and Mettenleiter, T.C. (1993) Virology 194,654-664 10 Johnson, D. and Ligas, M. (1988)
JANUARY
IYY4
NEWS
AND
COMMENT
1. Viral. 62,4605-4612 11 Hanssens, F., Nauwynck, H. and Pensaert, M. (1993) J. Viral. 67, 4492-4496 12 Roop, C., Hutchinson, L. and Johnson,
D. (1993)J. Viral. 67,2285-2297 13 Keay, S., Merigan, T.C. and Rasmussen, L. (1989) Proc.Nut1Ad. Sci. USA 86,10100-10103 14 Babic, N. et al. (1993) J. Viral. 67,
4421-4426 15 Fuller, A.O. and Lee, W. (1992) 1. Viral. 66,5002-5012 16 Lee, W. and Fuller, A.O. (1993) 1. Viral. 67,5088-5097
The role of TH1and T,2 subsetsin human infectiousdiseases Gianfranco Del Prete and Sergio Romagnani
T
he immune system can evoke different mechanisms for attacking pathogens, but not all mechanisms are activated after either infection or immunization. Many bacterial, protozoa1 and viral infections trigger a cell-mediated immune response, while other pathogens or their products induce a primarily antibody (humoral) response. It is now clear that T helper (TH) cells, which are required for both cell-mediated and humoral immune responses, are composed of distinct subsets distinguished by different patterns of cytokine production. T,l cells (1) produce interleukin (IL)-2, interferon (IFN)-y and lymphotoxin, (2) promote macrophage activation, which results in delayed-type hypersensitivity (DTH), and (3) promote the production of opsonizing antibodies, which are particularly required to clear infection by intracellular organisms. T,2 cells (1) secrete IL-4, IL-5, IL-6, IL-10 and IL-1 3, (2) provide optimal help for the production of antibodies that adhere to mast cells, and (3) promote both mast cell and eosinophil activation. In the absence of polarizing conditions leading to stereotyped T,l or T,2 patterns, CD4+ T cells producing both T,l-type and T,2-type cytokines (T,O cells) usually arise, which mediate effects that depend on the ratio of cytokines produced and the nature of responding cells’J. Although in principle T,l and T,2 cells might arise from distinct
precursors, experiments with homogeneous populations of cells from mice transgenic in the T cell receptor (TCR) strongly suggest that a single precursor can differentiate into either a T,l or a T,2 phenotype3-s. The possibility of a single precursor is supported further by the recent paper of Reiner and colleague&, which shows that T cells from mice infected with Leishmania major express a restricted TCR repertoire both in progressive infection and in protective immunity, regardless of histocompatibility haplotype. However, the mechanisms responsible for the differentiation of naive T, cells into the T,l or T,2 phenotype have not yet been completely clarified. Regulation of the development of T,land T,2 subsets Attention has focused recently on the possibility that the type of T, response depends on the nature of the antigen-presenting cells (APCs) or of their products. Although some differences in the response of T,l and T,2 cells to different APCs have been reported’, the type of APC of itself does not critically influence the differentiation of T, precursors into one or another phenotype3. Cytokines released by APCs and/or other cell types during exposure
G. Del Prete and S. Romagnani are in the Division of Allergy and Clinical Immunology, 1st. Clinica Medica 3, University of Florence, Viale Morgagni 85, I-50134 Florence, Italy.
to antigen appear to have a more striking role in determining the development of the specific T,l or T,2 response. Naive, ovalbuminspecific, TCR-transgenic T cells have been used to show that heatkilled Listeria monocytogenes induced T,l development in vitro via production of IL-12 from macrophages. IL-12 can promote T-cellindependent production of IFN-)I by natural killer (NK) cells*; IFN-y in turn favors the development of T,l cells9. In contrast, early IL-4 production at the time of antigen presentation seems to be critical for the maturation of naive T, cells into T,2 cells*O. Production of IL-4 is restricted to mature T,2 cells, a small population of early thymic emigrants and cells of the mast celYbasophi1 lineage. Whether one of these populations, or another undescribed cell, serves as the source of IL-4, paralleling the role of IFN-y-producing NK cells in inducing T,l cell development, is still an unsolved question. Human CD4’ T cell clones with functional properties similar to those of murine T,l and T,2 cells have been derived from the peripheral blood and other fluids of normal subjects or patients with different diseases”. Furthermore, studies investigating the mechanisms responsible for the in vitro development of human T,l- and T,2-like cells have suggested that IL-12, IFN-cx and IFN-y are important in the in vitro development of T,l clones, and IL-4 in
0 1994 Elsevier Science Lrd 0966 842X194/$07.00
TRENDS
IN MICROBIOI_OGY
4
VOL..
2
NO.
1
JANUARY
1994
AND
COMMENT
not elicit an HR in the host plant? Are harpins the only elicitors of nonhost HR in tobacco and possibly in other plants? Is the same mechanism used in tobacco to recognize both the Erwinia and the P. s. pv. syringae harpins? Is host resistance different in mechanism from nonhost resistance? Answers to this fascinating puzzle require the identification of more HR elicitors and their putative plant receptors. References 1 Mekalanos, J.J. (1992) J. Bacterial. 174, l-7 2 Long, S.R. and Staskawicz, B. (1993)
174,6878-6885 10 Van Gijsegem, F., Genin, S. and Boucher, C. (1993) Trends Microbial. 1, 175-180 11 Huang, H-C. et al. (1993) Mol. Plant-Microbe Interact. 6,515-520 12 Smith, M.J. et al. (1993) Tetrahedron Lett. 34,223-226 13 Wei, Z-M. et al. (1992) Science 257, 85-88 14 He, S.Y., Huang, H-C. and Collmer, A. (1993) Cell 73,1255-1266 1.5 Huang, H-C. et al. (1988) 1. Bacterial. 170,4748-4756 16 Beer, S.V. et al. (1991) in Advances in Molecular Genetics of Plant-Microbe Interactions (Hennecke, H. and Verma, D.P.S., eds), pp. 53-60, Kluwer Academic Publishers
Cell 73,921-935 3 Willis, D.K., Rich J.J. and Hrabak, E.M. (1991) Mol. Plant-Microbe Interact. 4, 132-138 4 Klement, Z. (1982) in Phytopathogenic Prokaryotes (Vol. 2) (Mount, M.S. and Lacy, G.H., eds), pp. 149-177, Academic Press 5 Keen, N.T. (1992) Plant Mol. Biol. 19, 109-122 6 Lindsay, W.P., Lamb, C.J. and Dixon, R.A. (1993) Trends Microbial. 1, 181-187 7 Fenselau, S., Balbo, I. and Bonas, U. (1992) Mol. Plant-Microbe Interact. 5, 390-396 8 Gough, C.L. et al. (1992) Mol. Plant-Microbe Interact. 5,384-389 9 Huang, H-C. et al. (1992) J. Bacterial.
Initiationand spread of a-herpesvirus infections Thomas C. Mettenleiter
H
erpesviruses are large animal viruses with a DNA genome varying from approximately 120 to 250kb. Based on their biological properties, the Herpesviridae have been divided ‘into three subfamilies, the a-, pand y-herpesvirinae, prototypes of which are the human pathogens herpes simplex virus (HSV), (HCMV) and cytomegalovirus Epstein-Barr virus (EBV), respectively. As enveloped viruses, they depend on two consecutive processes for infectious entry into target cells: (1) attachment of free virions to cells and (2) penetration, that is, fusion of virion envelope and cellular cytoplasmic membrane leading to release of the nucleocapsid into the cell. Virion envelope glycoproteins play important roles in both these processes (see Refs 1,2 for recent reviews). After infection of primary target cells, virus spread can occur by several different mechanisms. Infected cells may release infectious
virions that reinitiate infection from outside. In addition, direct viral cell-to-cell spread from primary infected cells to adjacent noninfected cells may occur. In the host, virus may be disseminated by circulating infected cells that adhere to noninfected tissues and transmit infectivity directly. Recent results on HSV and pseudorabies virus (PrV) shed more light on these processes in a-herpesviruses. PrV causes Aujeszky’s disease in swine, which is characterized by nervous and respiratory symptoms, and reproductive failure. Unlike HSV, PrV is not pathogenic for humans. However, the two viruses have several features in common, including a broad host range in vitro, and several species besides the natural host can be infected experimentally. In addition, all of the known PrV glycoproteins are
related to homologous teins in HSV (Ref. 1)“.
Attachment Binding of free infectious virus to target cells involves interactions between virion envelope glycoproteins and cellular virus receptors. Herpes virions contain a large number of different virus-encoded envelope glycoproteins that might participate in attachment. A wellknown example of a cellular herpesvirus receptor is the B-cell membrane protein CR2 (CD2 1 ), which binds EBV (Ref. 3). Recent studies have demonstrated that several a- (reviewed in Ref. l), p- and ‘yherpesviruses4J bind to their target cells by interaction of virion components with cell-surface glycosaminoglycans, principally heparan sulfate (HS)6.
“At the 18th International Herpesvirus Workshop, a common nomenclature for a-herpesvirus glycoproteins was agreed on, based on designations of HSV glycoproteins. This nomenclature is used here.
T.C. Mettenleiter is in the Federal Research Centre for Virus Diseases of Animals, PO Box 1149, D-72001 Tiibingen, Germany. 0 1994 Elsevier Science Ltd 0966 842X/94/$07.00
TRENDS
IN MICROBIOLOGY
2
VOL.
2
No.
glycopro-
1
JANUARY
1994
NEWS
In the a-herpesviruses, the major HS-binding glycoproteins are glycoprotein (g) C homologs7,8. Surprisingly, gC homologs are not essential for virus infectivity, indicating either that other viral glycoproteins also recognize HS or that a completely gC- and HS-independent pathway of virus attachment exists. In HSV, the essential glycoprotein gB also binds HS under physiological conditions’. However, PrV gB does not bind HS, although it may interact with PrV gC and indirectly participate in HS binding*. Therefore, there is a gC- and HS-independent mode of virus attachment for PrV. Kinetic analyses demonstrate that the primary interaction between the virion and the target cell, which is initially sensitive to competition by exogenous heparin (an HSrelated compound), converts into a heparin-resistant attachment. Glycoproteins gC and gD participate in this secondary binding, which appears to be specifically mediated by the interaction of the PrV and HSV gD homolog with a second cellular component9J0 (see Fig. 1). Adhesion of Pi-V-infected cells in suspension to noninfected cells exhibits similar characteristics to the binding of free virions”: specific adhesion is sensitive to heparin and to antibodies against PrV gC, but not to antibodies against PrV gD. In the absence of gC, specific adhesion drops by 90%, correlating with results for gC-negative PrV virions8. Interestingly, antibodies against PrV gD inhibit adhesion in the absence of gC. Hence, attachment of free PrV virions and specific adhesion of virus-infected cells occur by similar if not identical mechanisms, mainly involving glycoproteins gC and gD. Penetration After attachment, direct fusion between the herpes virion envelope and the cytoplasmic membrane occurs at neutral pH. All the viral glycoproteins involved in this process (the gB, gD, gH and gL homologs) are essential for infectivity of free virions (see Refs 1,2 for reviews), with gH and gL forming
a complex 12. The conservation of the gB, gH and gL homologs throughout the herpesviruses indicates that membrane fusion processes might be similar for all herpesviruses. Indeed, within the a-herpesviruses, complementation of lethal mutations in several gB homologs can be achieved with gB proteins from heterologous viruses2. In HCMV, there is evidence that gH may interact with a cellular receptor13 and, since herpesvirus gH homologs are closely related, similar interactions may also occur in other herpesviruses. However, a ‘classical’ fusion protein has not yet been identified in herpesviruses, although heterologous expression of gB and gD homologs in cells induced them to fuse more readily (reviewed in Ref. 1). It is likely that several virion glycoproteins acting together are necessary for successful membrane fusion. Direct cell-to-cell spread The mechanisms leading to direct virus transfer from primary infected cells to adjacent noninfected cells are still unclear. In HSV, all the glycoproteins that are necessary for penetration are also essential for direct viral cell-to-cell spread, suggesting that these processes are similar. Results from PrV, however, show that PrV gD is required only for the infectivity of free virions and not for cell-to-cell spread in culture (reviewed in Ref. 2) or transneuronal spread in u~vo’~. This shows that penetration and cell-to-cell spread, although related, are distinct and are probably regulated differently. It also indicates that cell-to-cell spread and transneuronal transfer may occur by very similar mechanisms. Whether gD is necessary for PrV spread via infected circulating cells remains to be determined. Multiple virus-cell interactions in infection and spread Initiation of infection and spread of ol-herpesviruses appear to be complex processes involving multiple sequential (or simultaneous) interactions between different viral and cellular components, the details of which are only beginning
AND
COMMENT
HSV !3B gc CID
I
HS
?
11 HS ?
Fig. 1. Involvement of envelope glycoproteins of herpes simplex virus (HSV) and pseudorabies virus (PrV) in attachment. The figure shows binding of glycoproteins gB, gC and gD to heparan sulfate (HS) or to another unidentified cellular receptor (?), and the putative interaction between gB and gC in PrV (open arrow).
emerge1,2J5. PrV and HSV both use HS for primary attachment and may even recognize similar secondary receptors 16. What determines their different natural host ranges in viva is still unclear. As more herpesviral envelope proteins are identified, these processes may turn out to be even more complex. However, initiation of infection by free virions, virus-infected circulating cells, and direct viral cellto-cell spread seem to occur by related, but distinct, mechanisms. Interference with these mechanisms of infection might help to inhibit primary infection, block viral spread and concomitantly prevent the onset of disease. to
Acknowledgements Work in my laboratory was supported by the Deutsche Forschungsgemeinschaft. I thank PG. Spear for critical reading of the manuscript.
References 1 Spear, P.G. (1993) Semin. Viral. 4, 167-180 2 Mettenleiter, T.C. (1991) Comp. lmmunol. Microbial. Infect. Dis. 14, 151-163 3 Nemerow, G. and Cooper, N. (1992) Semin. Viral. 3,117-124 4 Kari, B. and Gehrz, R. (1992) 1. Virol. 66,1761-1764 5 Vanderplasschen, A. et al. (1993) Virology 196,232-240 6 Shieh, M-T. et al. (1992) J. Cell Biol. 116,1273-1281 7 Herold, B. et al. (1991) 1. Viral. 65, 1090-1098 8 Mettenleiter, T.C. et al. (1990) 1. Virol. 64,278-286 9 Karger, A. and Mettenleiter, T.C. (1993) Virology 194,654-664 10 Johnson, D. and Ligas, M. (1988)
JANUARY
IYY4
NEWS
AND
COMMENT
1. Viral. 62,4605-4612 11 Hanssens, F., Nauwynck, H. and Pensaert, M. (1993) J. Viral. 67, 4492-4496 12 Roop, C., Hutchinson, L. and Johnson,
D. (1993)J. Viral. 67,2285-2297 13 Keay, S., Merigan, T.C. and Rasmussen, L. (1989) Proc.Nut1Ad. Sci. USA 86,10100-10103 14 Babic, N. et al. (1993) J. Viral. 67,
4421-4426 15 Fuller, A.O. and Lee, W. (1992) 1. Viral. 66,5002-5012 16 Lee, W. and Fuller, A.O. (1993) 1. Viral. 67,5088-5097
The role of TH1and T,2 subsetsin human infectiousdiseases Gianfranco Del Prete and Sergio Romagnani
T
he immune system can evoke different mechanisms for attacking pathogens, but not all mechanisms are activated after either infection or immunization. Many bacterial, protozoa1 and viral infections trigger a cell-mediated immune response, while other pathogens or their products induce a primarily antibody (humoral) response. It is now clear that T helper (TH) cells, which are required for both cell-mediated and humoral immune responses, are composed of distinct subsets distinguished by different patterns of cytokine production. T,l cells (1) produce interleukin (IL)-2, interferon (IFN)-y and lymphotoxin, (2) promote macrophage activation, which results in delayed-type hypersensitivity (DTH), and (3) promote the production of opsonizing antibodies, which are particularly required to clear infection by intracellular organisms. T,2 cells (1) secrete IL-4, IL-5, IL-6, IL-10 and IL-1 3, (2) provide optimal help for the production of antibodies that adhere to mast cells, and (3) promote both mast cell and eosinophil activation. In the absence of polarizing conditions leading to stereotyped T,l or T,2 patterns, CD4+ T cells producing both T,l-type and T,2-type cytokines (T,O cells) usually arise, which mediate effects that depend on the ratio of cytokines produced and the nature of responding cells’J. Although in principle T,l and T,2 cells might arise from distinct
precursors, experiments with homogeneous populations of cells from mice transgenic in the T cell receptor (TCR) strongly suggest that a single precursor can differentiate into either a T,l or a T,2 phenotype3-s. The possibility of a single precursor is supported further by the recent paper of Reiner and colleague&, which shows that T cells from mice infected with Leishmania major express a restricted TCR repertoire both in progressive infection and in protective immunity, regardless of histocompatibility haplotype. However, the mechanisms responsible for the differentiation of naive T, cells into the T,l or T,2 phenotype have not yet been completely clarified. Regulation of the development of T,land T,2 subsets Attention has focused recently on the possibility that the type of T, response depends on the nature of the antigen-presenting cells (APCs) or of their products. Although some differences in the response of T,l and T,2 cells to different APCs have been reported’, the type of APC of itself does not critically influence the differentiation of T, precursors into one or another phenotype3. Cytokines released by APCs and/or other cell types during exposure
G. Del Prete and S. Romagnani are in the Division of Allergy and Clinical Immunology, 1st. Clinica Medica 3, University of Florence, Viale Morgagni 85, I-50134 Florence, Italy.
to antigen appear to have a more striking role in determining the development of the specific T,l or T,2 response. Naive, ovalbuminspecific, TCR-transgenic T cells have been used to show that heatkilled Listeria monocytogenes induced T,l development in vitro via production of IL-12 from macrophages. IL-12 can promote T-cellindependent production of IFN-)I by natural killer (NK) cells*; IFN-y in turn favors the development of T,l cells9. In contrast, early IL-4 production at the time of antigen presentation seems to be critical for the maturation of naive T, cells into T,2 cells*O. Production of IL-4 is restricted to mature T,2 cells, a small population of early thymic emigrants and cells of the mast celYbasophi1 lineage. Whether one of these populations, or another undescribed cell, serves as the source of IL-4, paralleling the role of IFN-y-producing NK cells in inducing T,l cell development, is still an unsolved question. Human CD4’ T cell clones with functional properties similar to those of murine T,l and T,2 cells have been derived from the peripheral blood and other fluids of normal subjects or patients with different diseases”. Furthermore, studies investigating the mechanisms responsible for the in vitro development of human T,l- and T,2-like cells have suggested that IL-12, IFN-cx and IFN-y are important in the in vitro development of T,l clones, and IL-4 in
0 1994 Elsevier Science Lrd 0966 842X194/$07.00
TRENDS
IN MICROBIOI_OGY
4
VOL..
2
NO.
1
JANUARY
1994