New insights into protective humoral responses and HIV vaccines

Introduction Neutralization is the reversible loss in infectivity observed when a virus forms a complex with an antibody [1]. Currently, it is difficult to test the simplest hypothesis that inducing circulating neutralizing antibodies would be protective in a phase III trial because neutralization by vaccine-induced antibodies of HIV isolates as close as possible to those involved in natural transmission cannot be easily demonstrated in vitro. However, other mechanisms may achieve the same effect. HIV buds from its host cell, taking the membrane with it. The virions can therefore be lysed by complement. Although they also carry the host's complement control proteins, this protection may be overcome if sufficient complement-fixing antibodies bind to the envelope glycoprotein. Also, antibodies may serve to target antibody-dependent cell cytotoxicity that could lyse virus-infected cells. It is believed that a vaccine against HIV that will absolutely protect from infection each recipient that becomes exposed to HIV is an unlikely proposition [2]. However, a vaccine that reduces the number of nonimmune hosts and or reduces the number of those able to transmit the virus will most probably have a significant impact on the global spread of HIV. In the simplest terms, evolution has produced cell-mediated immunity to eliminate virus from the host and then antibodies to protect from a second infection with the same agent. Many viral vaccines are believed to work by preventing the viraemic phase of infection. The virus replicates at its site of entry and then spreads via the blood stream, producing pathological effects at more remote sites. For example, polio and hepatitis virus replicate in the gastrointestinal tract, spreading to the central nervous system or the liver, respectively. Measles replicates in the respiratory tract before producing a rash in the skin. Antibodies in the blood can prevent this spread, but they cannot be expected to prevent infection at the site of entry. Edward Jenner, the father of vaccination, went to his grave believing that exposure to cowpox gave life-long immunity to smallpox. He did not recognize that vaccinated people were being infected every time there was a smallpox epidemic, boosting their immunity. Antibodies can protect against virus pathology even if a cell-associated viraemia is possible (e.g., measles). Viraemia is not the only stage in pathogenesis that can be influenced by vaccine-induced antibodies: rabies passes along the neurones while the pathology associated with Marek's disease virus and feline leukaemia virus is controlled, although they do not appear to be eliminated from vaccinated hosts. For HIV, the most vulnerable steps may be its crossing the mucosal membranes and during its journey to the local draining lymph node. IgA antibodies are induced in the genital tract in natural infection and have been shown in vitro to prevent the transcytosis of HIV virions through an epithelial cell layer. The virus is then transported to the draining lymph node by dendritic cells (DC) that appear to have a multiplier effect: HIV is infectious for target cells at very low concentrations in the presence of DC although the mechanism involves DC-SIGN rather than the CD4 receptor. The caveat has to be added that antibodies may also have an undesired effect on HIV pathogenesis. When the antibody level falls, which is inevitable without further vaccination or repeated exposure to infection, the antibodies may enhance the infectivity of HIV [3]. Also, virus antibody complexes may enter a cell carrying a provirus via a Fc or complement receptor, thus promoting the conditions for recombination leading to escape from immune control [4]. However, with viruses that do not necessarily lyse their host cell, some mechanism is required to eliminate infected cells although, with HIV, neutralizing antibodies alone can prevent infection following simulated natural transmission. Passively transferred antibodies that neutralize HIV in vitro protect rhesus macaques following exposure by the heterosexual route [5] or at birth [6] to recombinant simian immunodeficiency virus expressing envelope glycoproteins from HIV primary isolates (SHIV). Therefore, a vaccine that elicits strong cellular anti-HIV responses as well as potent neutralizing antibodies may be effective in preventing the establishment or the spread of HIV infection. First-generation vaccine candidates Phase I/II trials in human volunteers have been completed for the first generation of HIV envelope vaccine candidates. These candidates consist of the recombinant HIV-1 external envelope glycoprotein gp120 mixed in adjuvants, alum [7] or the water in oil emulsion MF59 [8]. The clinical trials have followed the traditional vaccination strategy of attempting to induce neutralizing antibodies in the blood and demonstrated that such vaccines are immunogenic and safe for human use. Two phase I/II trials have been performed in Thailand with recombinant HIV-1 Mn gp120 by Vaxgen [7] and with Chiron's HIV-1 SF2 gp120 [8]. Both immunogens induce neutralizing antibodies directed primarily against the homologous isolate. These antibodies increase after each immunization, both in absolute titres and relative to titres of antibody able to bind gp120 or the homologous V3 peptides. However, in both studies, the antibody titres decreased after the final injection. In the case of the Vaxgen phase II trial, sera collected from the vaccinees were reported to also neutralize intra-subtype B primary isolates in assays with 72-h absorption phases and monitoring virus growth after a total of 4 days in culture. Vaxgen started phase III trials in Thailand, North America and Europe [9]. In these trials, the antibody levels are sustained by repeated immunizations. Both the Vaxgen and Chiron immunogens have also been compared in a single trial within the United States in individuals at higher risk for infection with HIV whose immune systems may be compromised [10]. Injection drug users had markedly lower antibody titres following immunization with recombinant HIV-1 SF2 gp120, although this may reflect differences in the dose of immunogen per injection. The inability of these monomeric gp120 vaccines to elicit neutralizing antibodies against heterologous primary HIV isolates is not related to their adjuvant formulation. A phase I trial of GlaxoSmithKline's HIV-1 W61D gp120, formulated with the more potent QS21 adjuvant, induced neutralizing antibody titres against the laboratory HIV-1 Mn virus approaching those observed in natural infection, although at the cost of increased clinical adverse effects, but not against primary isolates [11]. Vaccination methodologies using viral vectors to express monomeric HIV envelope immunogens did not improve on the ability of these immunogens to elicit neutralizing antibodies against heterologous primary HIV isolates. However, neutralization of primary isolates homologous to the immunogens was reported in a phase I trial of the Aventis Pasteur live-virus prime, recombinant boost vaccination strategy [12]. Volunteers were immunized with canarypox expressing, inter alia, HIV-1 Mn gp120 and then boosted with Chiron's recombinant HIV-1 SF2 gp120. Sera from immunized volunteers gave median reductions in the number of GHOST cells infected by the primary isolates of HIV-1 SF2 and HIV-1 Mn of 83% and 38%, respectively. However, results from the placebo-injected volunteers were not reported. Our own data indicate that the primary HIV-1 Mn isolate is at least 625 times more resistant to soluble CD4 (sCD4) than its laboratory-adapted version but is still at least 25 times more sensitive than the most neutralization resistant of primary isolates. A panel of eight subtype B primary isolates with high levels of resistance to neutralization by both antibodies and sCD4 has been formed to compare vaccine-induced antibodies [13]. Sera from volunteers primed with canarypox and boosted with HIV-1 Mn/IIIB gp160 were tested against this panel. Most serum/isolate combinations (84%) gave results varying between 50% reduction and 100% increase in the release of Gag p24 from conventional mitogen-transformed peripheral blood mononuclear cell (PBMC) cultures. Minorities of 8% at both extremes gave either 52-79% reductions or 102-200% increases (enhancement of infection). While 21% (n = 10) of combinations with sera from placebo recipients gave reductions of greater than 50%, none showed equivalent enhancement of infection. The proportion of enhancing sera from vaccinated volunteers was statistically significant. Although the first generation of HIV envelope vaccine candidates has been shown to be immunogenic and safe to use in phase II human trials [9,14], their failure to induce antibodies that could neutralize primary isolates delayed phase III efficacy trials. Primary isolates can be neutralized by a small number of monoclonal antibodies isolated from seropositive humans and chimpanzees as well as by seropositive sera [15,16]. The types of antibodies elicited during vaccination with first-generation vaccine candidates and those elicited during natural infection differ. Following infection, the immune the of antibodies directed and and their which is followed by an increase in the neutralization of the Antibodies elicited by monomeric immunogens may have a low for their on the of The that these antibodies neutralize laboratory but not primary isolates may be related to differences in the of exposure on the of laboratory-adapted and primary isolates. may also antibodies to only on the envelope to that exposed during or to formed from the virus or in with host proteins, during envelope vaccines are not expected to elicit such types of vaccine candidates has been to vaccine which can natural immunogens the of the envelope on the of infected cells or infectious viral These have to trials in on producing by the at the and by the site between and gp120, a to the of gp120 from the These are as soluble primarily by between within the of recombinant proteins, when as induce a of antibodies than that induced by the soluble gp120, by low titres of antibodies to the V3 immunization of macaques with such soluble from the laboratory-adapted isolate to elicit antibodies of neutralizing heterologous primary HIV-1 isolates However, of rhesus macaques vaccinated with recombinant HIV-1 were against the homologous viral which is that a vaccine neutralizing antibodies will a in the spread of HIV. of the of such soluble that they as of and rather than the in envelope In the of the site may the exposure and or of neutralization the of the of the site from soluble to V3 which are on the virions and the with an site have been to the of soluble while an this were between either the or between the gp120 and In the first the of at within of and Although such are not gp120 and during they appear to an they are by a panel of In the second were between the and gp120 on the of the primary HIV-1 isolate. were in to the at a are in other A second of was within the and of gp120. it was shown that the of these are as which are gp120 and only when vectors expressing are with a panel of monoclonal antibodies that the of these soluble that of a although differences were reported The of has not been reported. A to soluble with an site of of at a close to the of recombinant from the primary HIV-1 isolate induced which neutralized not only the homologous but also the heterologous primary an that neutralization are on such to increase the exposure and of neutralization In to soluble envelope vaccines that the HIV it is also to methodologies to increase the of neutralization those in the CD4 and of the HIV are varying within the is not only to the of envelope within the of the but also to the of and the presence of on the viral of from the envelope of or HIV can increase the exposure of neutralization at of the the virus more to neutralization by and monoclonal antibodies Neutralization also on of within the of HIV and HIV virions expressing the envelope but the or the or and virions both the and are that these do not the of the envelope and therefore from such can be as immunogens to elicit higher titres of antibodies against the exposed neutralization have been and the with or without the still formed from the or or both the and could also be the of from the envelope not the generation of the of from the and was compared in rhesus macaques using the followed by boosting vaccination the the were immunized with expressing the with an while during the phase they were immunized with the soluble with an Although both the and the immunogens elicited neutralizing antibodies against both the and only the was of neutralizing antibodies against heterologous primary HIV-1 isolates. These results that of the from the envelope increases the exposure and of neutralizing antibodies that are HIV-1 isolates. Although not every primary HIV-1 isolate was to neutralization by the antibodies elicited by the immunogen and the neutralization against these heterologous viruses was lower than that against the these results that envelope will increase the of neutralization most that the antibodies elicited against such will be able to their on the of virions expressing the In an to the protective of the antibodies elicited by the the were with the homologous virus the protective of antibodies in the of cell-mediated responses elicited during immunization, the were first of their cells. the phase of infection, the viral was lower in these relative to that neutralizing antibodies elicited during vaccination with the protection against the homologous viral It is not these antibodies will also protection against heterologous viral not every envelope will in an increase of the exposure of neutralization For example, of the from the envelope only increases the of the virus to neutralization by Also, the effect of a on viral to neutralization may be envelope and the effect on viral and neutralization to elicit neutralizing antibodies of with envelope from heterologous isolates In a in it was reported that such vaccines elicit antibodies of primarily the infection of virions expressing in the but not heterologous isolates. It is possible that vaccination methodologies the with the vaccine will be effective in neutralizing is to to the same or antibody responses in exposed on the of were to that to antibodies in the sera of When rhesus macaques were vaccinated with these and with they showed a or sustained CD4 and a In control macaques showed an in CD4 cell The of A of antibody is produced following HIV infection or vaccination with HIV of this antibody has effect on viral infectivity It has been shown that monoclonal antibodies these to bind to the envelope glycoproteins on infected cells with the that they would not bind to the same on virions However, antibodies to the V3 and of gp120 and both I and II antibodies to have been shown to bind to virions of a of A as to the be on the infected cell and the may be with the Most of these antibodies bind to a of and the of is to the of the in its It is that some of the envelope on the but not on the of infected are in their the of gp120 and the on The of the envelope glycoproteins are as complex of and and a within these induce neutralizing The of the not be exposed on the there is a of within this Antibodies to the are but the is both and Antibodies to these have to be by if within subtype B laboratory isolates is to be the within I of which has been associated with a on and The first to be at the of gp120, has been to be However, it also a the monoclonal antibody has been shown to recognize the of gp120 using a of which by a single on However, following of complexes of the antibody with recombinant HIV-1 SF2 gp120, a can be by from to with either of these to bind to the Antibodies to the V3 of gp120, the neutralizing for laboratory are isolate However, antibodies can be induced with the in which and are The of these to vaccine will on the with from some is to a of antibodies to within the or is directed immunogenic of HIV envelope glycoproteins in the resistance The of primary HIV-1 resistance to vaccine-induced antibodies may not with the failure to induce the The on which neutralization assays are have been by the work of The of a neutralization is to virus and antibody to complexes and then to the infectious virus by the to cells. laboratory these Antibodies bind to virions during the phase that the complexes do not bind to target cells. The V3 the neutralizing The is the monoclonal antibody complexes bind to target cells. primary neutralization may not even after the virus has to the target cell Neutralization of primary isolates by from seropositive individuals may then antibodies to that are only or during the at the site of membrane gp120, CD4 cells and are Therefore, the of neutralization with antibodies such can be expected to be of the phase and to increase with the of the absorption recombinant gp120 envelope and that the are not expected to such gp120 elicit neutralizing antibodies against other that are exposed on the of These antibodies may bind to the of primary isolates than those of laboratory-adapted most probably because these are exposed on primary than laboratory-adapted isolates. It is that a has been by the of neutralization of primary isolates by seropositive using assays for laboratory isolates. have to be to neutralization of primary isolates by vaccine-induced antibodies with the neutralization of the is a and assays with phases are an absolute Also, neutralization of cell-associated virus may be on the target with absorption phases on can antibodies after a becomes infected than assays with absorption phases in conventional mitogen-transformed assays is that the assays for primary isolate neutralization are sensitive to infection for example, it may be possible to mitogen-transformed with virus which not sufficient envelope glycoproteins to other cell types expressing lower concentrations of CD4 or the The virus with glycoproteins infectivity for will also a target for antibodies that they more neutralization resistant The neutralization resistance of primary isolates would then be an to the of mitogen-transformed for virions whose would also probably infectious during natural transmission. The same can be from the of be the virus has to in an immune host but also has to to at least in the presence of the virus can Also, to be in to Many seropositive individuals have antibodies that can neutralize virus when these have been either by in cells or by of In the when a laboratory isolate is in a human the virus neutralization resistance not when HIV is in chimpanzees and the virus sensitive to neutralization when laboratory isolates of are in rhesus macaques or feline immunodeficiency virus is in The hypothesis is during viral in the presence of neutralizing antibodies directed at in the CD4 or of the the viral envelope will to the of the its at the of When the natural is this protection is are then from the of in which the or have and the infectious are neutralization However, it is possible that a more hypothesis is required to escape from antibody neutralization antibodies appear to bind to envelope from both laboratory and primary on the of or virions on a after antibody may infection can Also, global resistance to neutralization is not the only to the A more of neutralization resistance is Following infection, escape that only neutralization by the antibodies within their own host The of viruses that that they are not neutralized by antibodies induced to within the same host is that immunity is a significant in effect. are in that neutralization escape can be demonstrated in vitro with immune the antibody is to a or the virus can escape from neutralizing antibodies Two mechanisms appear to be able to escape at in the number and of and that the of and gp120 and also virus infectivity both of at the of the of HIV-1 this virus to antibodies against the CD4 site and the V3 The is that at these can these in the within the of the V3 can the CD4 and the in the of gp120 and the of involved in the between the and the also to a exposure to antibodies the CD4 and However, of antibodies and is either not or escape may only a single viruses some isolated from a infected with could not be neutralized by sera from vaccinated and by neutralizing antibodies were directed against the in in a vaccinated with and with and involves in It is also to the of neutralization in vitro that with in protection from infection. use neutralization assays target cells and of and and it becomes to compare the neutralization of an isolate or the neutralization of a monoclonal antibody or In there is still even of levels to neutralization greater than from the of 50% and neutralization have been reported by the by this with other 50% neutralization has been because this can be with the increased levels of neutralization be more and neutralization was to between and vaccine-induced antibodies [13]. However, close to 100% neutralization to be required to of macaques by Our own results indicate in assays reductions in release greater than are required for while between GHOST cell is that reductions of than 50% are The that vaccine efficacy is by the of the virus the that of virus with neutralizing antibodies a of infectious and the virus infectivity then the of the virus not be in release which is as the for neutralization cannot between decreased viral some other associated with and an or loss in virus The of GHOST cells that can infectious may to this immunity at mucosal In heterosexual HIV can the draining lymph node without a these neutralization the of the cannot be involved in A can after through the of an epithelial cell within the on the of the of further are infection via the CD4 of or and or by the DC-SIGN of DC within the DC-SIGN HIV at low These DC have they cannot be but when they are they to the draining lymph they cells and DC with HIV can virus to in Therefore, inducing antibodies that the HIV can be in preventing infection. IgA on the may be individuals are exposed to HIV but Although these individuals are to be this only to the of which can neutralize primary isolates of HIV has been from and from The antibodies target the of of IgA from and of this as well as exposed in a of in vitro transcytosis of primary HIV through a membrane of cells. The membrane is with a well is to natural at a mucosal However, antibodies to with the to with virus infection are also in The proportion of of infected individuals whose sera have antibodies to I or to CD4 is higher than although the presence of the antibodies not with neutralization of primary isolates in a cell In other of in have antibodies to the have to be IgA antibodies in of volunteers in HIV vaccine trials were using an with recombinant These results were they were not when either from infected cells or were The was against of the to the recombinant Also, the to antibodies in may have a it was not related to the of exposure to virus in antibodies to immunogens if the traditional strategy of circulating neutralizing antibodies using envelope glycoproteins is not it may be possible to the viral using the Antibodies induced by with produced reduction in viral during the phase of infection of macaques with but there was a at the of the phase was to following CD4 cell were and vaccination with had effect on this vaccination with produced lower in rhesus macaques with both and antibody responses were These macaques also had CD4 cell greater than of their level The of virus from an immune to a host is a this to neutralization isolates that within of primary a virus that has to in an immune host with more viruses when the is a vaccinated recipient for the will the In vaccines to antibodies that neutralize the most resistant are as is by the vaccine of the mechanisms of the neutralization resistance of virus that are during the viraemic phase would be in in vaccine Although a single immune of protection following vaccination may be an it would be most for control of vaccine and for monitoring efficacy trials that results from neutralization assays are of protection would of assays against a of viral neutralization assays of primary isolates are but would be very Antibodies that bind to envelope glycoproteins may a of vaccine-induced antibodies to infected cells with neutralization of the laboratory isolate equivalent to the antibodies can bind to on cells infected with primary isolates from by cell a system that is of high However, while of induced antibodies to infected cells with neutralization of the primary only time will the equivalent assays with vaccine-induced antibodies with phase III trials use immunogens that can protect against of the viruses circulating within their target However, in the trials are in of the are circulating vaccine trials that the traditional of inducing neutralizing antibodies immunogens from each circulating subtype these The is that antibodies have the to When a cell is infected with HIV, the CD4 is that it is difficult for a second virus to enter the cell and recombinant which viruses to the same cell, a high proportion of circulating The in of may be at the transmission level (e.g., in or within an In a high proportion of isolates were with recombination within the complexes can cells of the CD4 via the and can therefore a cell that is infected with HIV. In a and are a vaccine against subtype then These could be simulated in assays than with neutralization and enhancement assays