Studies on endoplasmic reticulum--Golgi complex cycling pathway in herpes simplex virus-infected and brefeldin A-treated human fibroblast cells.

Brefeldin A (BFA), a fungal metabolite, significantly inhibited the release of herpes simplex virus type 1 (HSV-1) from infected human fibroblast cells. Electron micrographs of HSV-1-infected and BFA-treated human cells demonstrated the presence of enveloped particles trapped between outer and inner nuclear membranes. Analyses of viral glycoproteins B, C, and D (gB, gC, and gD) showed faster migrating, immature forms in BFA-treated cells when compared to the mature glycoproteins, as observed in the untreated control cells. The shift in mobilities of the glycoproteins in BFA-treated cells apparently was due to the disassembly of the Golgi complex when evaluated by an indirect immunofluorescence assay. The immature forms of gB, gC, and gD could not be detected on the surface of BFA-treated human fibroblast cells. Removal of BFA resulted in a reorganization of the Golgi complex and formation of fully glycosylated gB, gC, and gD. Moreover, the HSV-1 particles released from the treated cells after the removal of BFA completely restored the infectivity of the viral particles. Our results indicate that human fibroblast cells have an endoplasmic reticulum-Golgi cycling pathway.     

A rapid procedure for the enrichment of undenaturated, antigenically active herpes simplex virus glycoproteins

The usefulness of lentil lectin affinity chromatography for the rapid enrichment of HSV glycoproteins in an undenatured state for both research and clinical purposes was investigated. In order to compare the lentil lectin-binding characteristics and immunologic specificities of undenatured HSV-1 and HSV-2 glycoproteins, [35S]methionine-labelled extracts of virus-infected HEp-2 cells were subjected to lentil lectin affinity chromatography. Individual HSV-1 and HSV-2 glycoproteins in bound and unbound fractions were identified using monoclonal antibodies. With the exception of a portion of pgD and gD, all major viral glycoprotein species (gA, gB, gC, gD, gE and gF) and their glycosylated processive intermediates bound to lentil lectin indicating that all possess predominantly mannosyl and/or glucosyl carbohydrate moieties. Although the unbound pgD and gD species were glycosylated, no gD and only a portion of pgD bound to lentil lectin when reapplied to the column indicating that these subspecies possess alterations in factors required for efficient lectin binding. Immunoprecipitation of undenatured lectin-bound glycoproteins from infected cells using HSV-1 and HSV-2-specific rabbit and human antisera confirmed previous findings that the predominant type-specific glycoproteins of HSV-1 and HSV-2 are gC and gE/gF, respectively.     

Analysis of the intracellular maturation of the herpes simplex virus type 1 glycoprotein gH in infected and transfected cells.

We have expressed the HSV-1 glycoprotein, gH, in transiently transfected COS-1 cells. The expressed protein was retained intracellularly, contained unprocessed carbohydrate, and was unrecognized by the monoclonal antibody, LP11. In addition, the protein was aggregated. These properties suggest that unlike other HSV glycoproteins, gH is misfolded in transfected cells. Pulse-chase studies of HSV-1-infected cells indicate that the kinetics of processing of gH are comparable to those of gB, gC, and gD. Rescue studies suggest that gH may interact with another protein during maturation in infected cells. However, we were unable to detect any stable interaction, although analysis of gH on neutral sucrose gradients shortly after synthesis indicated a possible transient association with a high molecular weight molecule or complex. The processing and cell surface expression of gH were also analyzed in HSV-1 virus mutants lacking gB, gC, or gD. Our results indicate that the maturation and cell surface transport of gH did not require the presence of these HSV-1 glycoproteins. In addition, three truncation mutants were constructed by linker insertion mutagenesis. Each of the three truncated proteins was synthesized, but the proteins were aggregated, contained only endo H-sensitive carbohydrate, and none were secreted.     

A block in glycoprotein processing correlates with small plaque morphology and virion targetting to cell-cell junctions for an oral and an anal strain of herpes simplex virus type-1

Summary The characteristics of two clinical isolates of HSV-1 obtained from an oral (424) and an anal (490) lesion were compared with the highly passaged KOS strain. In contrast to KOS, the clinical isolates produced small plaques, were more cell-associated and the predominant viral glycoprotein species for gC and gD in infected cell lysates was the precursor, high mannose glycoform. Total virus production in Vero cells was equivalent for the three virus strains in one-step growths. Pulse-chase studies of glycoprotein C processing showed a reduction in rate at 7.5h post infection and a significant block in processing at 10.5h post infection for 424 and 490 but not KOS. Similar results were obtained for gD. The significant reduction in glycoprotein processing for 424 and 490 suggests a block in transport of viral glycoproteins or virions to and through the Golgi apparatus. Extracellular virions and the cell surface, prior to cell lysis, contained the processed gC glycoform suggesting a competent cellular glycan processing system. Upon co-infection of 424 or 490 with KOS or a gC^– KOS strain, gC was processed to levels equivalent to KOS indicating that 424 and 490 are not inhibitory but that an activity(s) encoded by KOS facilitates maturation of gC from 424 and 490. Unlike KOS infected Vero cells, virion-containing vacuoles were observed in the cytoplasm at 12h p.i. and extracellular virions were concentrated at cell-cell junctions of 424 or 490 infected cells but not in the perinuclear region. These results suggest that intracellular transport of viral glycoproteins and virions in 424 and 490 infected cells is different from KOS infected cells. The reduced level of viral glycoprotein maturation, virus release, cell surface presence and presence of virions at cell-cell junctions are consistent with small plaque production in tissue culture cells.Portions of this work were presented in the 17th International Herpesvirus Workshop, Pittsburgh, PA, 1993.     

Polymorphism of human cytomegalovirus glycoproteins characterized by monoclonal antibodies

Monoclonal antibody panels selected in this and preceding studies were employed to begin to characterize the properties of human cytomegalovirus (CMV) glycoproteins. The results were as follows. (i) Four antigenically distinct CMV glycoproteins designated as gA, gB, gC, and gD have been identified. (ii) gA, gC, and gD each form several bands when immune precipitated from infected cell extracts by the corresponding monoclonal antibodies and electrophoretically separated in sodium dodecyl sulfate-polyacrylamide gels. In contrast, gB migrated at one broad band with an apparent molecular weight in the range of 116,000 to 123,000. Bands with different molecular weights were shown to share antigenic determinants by reactivity of monoclonal antibodies with electrophoretically separated polypeptides immobilized on nitrocellulose. (iii) A panel of 16 monoclonal antibodies to gA precipitated a family of glycoproteins 160,000-148,000, 142,000, 138,000, 123,000-107,000, 95,000, and 58,500 in apparent molecular weight designated as gA1 through gA6, respectively. (iv) To identify partially glycosylated precursors of gA, infected cells were treated with tunicamycin or deoxyglucose and reacted with different monoclonal antibodies. Tunicamycin-treated infected cells labeled for a short pulse or longer intervals contained only gA5. Whereas cells treated with deoxyglucose during a pulse contained gA4, those labeled for a longer interval contained gA6 and an additional band approximately 56,500 in apparent molecular weight designated gA7. (v) Precipitates of gA from infected cells labeled for a short pulse contained gA2 and gA3 which appear to be products of rapid glycosylation. After a chase, trace amounts of gA1 and gA6 were also precipitated suggesting that these are products of slow post-translational processing. (vi) Endo-beta-N-acetylglucosaminidase H was used to identify the forms of gA which contain high-mannose oligosaccharide chains. After treatment, the electrophoretic mobility of gA2, gA3, and gA6 increased significantly suggesting that these forms contain high-mannose chains cleaved by the enzyme. A hypothesis for processing gA is presented.     

Individual herpes simplex virus 1 glycoproteins display characteristic rates of maturation from precursor to mature form both in infected cells and in cells that constitutively express the glycoproteins

Pulse-chase experiments in conjunction with quantitative immunoprecipitation have been used to study the time-course of conversion from precursor to mature form of herpes simplex virus 1 glycoproteins C, D and B (gC, gD, and gB). The experimental systems employed were two infected cell lines and cells that constitutively express gD or gB. The relative rates of conversion among the glycoproteins did not vary in the systems used; the rate of maturation of gC was about two-fold higher than that of gD which, in turn, was about one and a half-fold higher than that of gB. Treatment with phosphonoacetate which inhibits viral DNA synthesis and hence virion morphogenesis induced a striking increase in the time course of conversion of immature gC, gD, and gB to fully glycosylated forms when measured late in the infection. The model of HSV glycoproteins maturation as integral components of the virion envelope is discussed.     

Detection of type-specific antibody to herpes simplex virus type 1 by radioimmunoassay with herpes simplex virus type 1 glycoprotein C purified with monoclonal antibody.

Herpes simplex virus type 1 (HSV-1) and HSV-2 specify at least four glycoproteins designated gA/gB, gC, gD, and gE. Previous studies have shown that gC produced by HSV-1 is antigenically distinct from the corresponding HSV-2 glycoprotein. With the exception of gC, the glycoproteins of both serotypes share antigenic sites. Standard serological assays fail to differentiate the antibody to the shared antigenic determinants from the type-specific antibody. In this paper, we describe a procedure for purifying gC from HSV-1-infected cell extracts with an immunoadsorbent prepared with an HCL monoclonal antibody. When used in a solid-phase radioimmunoassay, gC proved to be a type-specific antigen for quantitation of antibody to HSV-1. Among individuals who had no antibody to HSV at the onset of infection, all of those with primary HSV-1 infection developed antibody to gC. Subjects with primary HSV-2 infection failed to develop antibody reactive with gC of HSV-1 (P less than 0.01). Both immunoglobulin G and M antibodies against gC were detected in sera from subjects with either primary or recurrent HSV-1 infection. Higher antibody titers to gC were found in sera from individuals with recurrent infection than in sera from those with primary HSV-1 infection.     

Intracellular localization and transport of three different bovine herpesvirus type 1 glycoproteins involved in neutralization

Summary Monoclonal antibodies against 3 different glycoproteins of bovine herpesvirus type 1 (BHV-1) involved in virus neutralization were used in indirect immunofluorescence (IIF) tests to characterize the appearance and transport to the plasma membrane of virus antigens in the infected cells.Antibodies against gp 117 and gp 71 glycoproteins first showed pronounced ring-like nuclear fluorescence at 4 hours post-infection (PI), followed by staining of the perinuclear region, presumably the Golgi apparatus. In contrast, antibody against gp 87 produced staining in cell-to-cell junctional areas at 3 hours PI before any staining close to the nucleus.The expression of the 3 glycoproteins at the surface of the infected cells was confirmed by the use of monoclonal antibodies having neutralizing activity, but not by non-neutralizing antibodies against gp 117 and gp 71. Non-neutralizing antibody against gp 87 detected the surface fluorescence only in those cells showing marked degeneration.Inhibition of glycosylation of the viral glycoproteins with tunicamycin (TM) was followed by interference with transport of gp 117 and gp 87 to the plasma membrane. On the other hand, gp 71 was incorporated into the plasma membrane despite the lack of N-linked glycosylation.With 4 Figures     

Antibody activity to herpes simplex virus in mouse Ig classes and IgG subclasses

Summary Recent studies indicate that Ig class and IgG subclass induction varies for different proteins and further that some Ig subclasses, like IgG2a, are more efficient in important biologic processes such as antibody-dependant cell-mediated cytotoxicity (ADCC). Many proteins of herpes simplex virus type 1 (HSV-1) are immunogenic and induce immunoglobulin responses. To determine the distribution of immunoglobulins induced by HSV-1 proteins, we studied immune mouse serum using an Ig isotype specific Elisa assay for antiviral activity. We found by endpoint analysis that the antiviral titer was 1:12,903 for IgG1, 1:5141 for IgG2a, 1:2140 for IgG2b and 1:229 for IgG3. To identify which isotypes were induced by individual glycoproteins and other viral proteins, Western blots containing HSV-1 proteins were probed with immune serum and isotype specific second antibodies. gB, gC, gD and the 42/44KDa nucleocapsid complex induced strong IgG1, IgG2a, IgG2b responses. IgG3 reactivity with viral proteins appeared weaker. Among the IgG3 reactivities detected on immunoblots, gB and gC were the most intense. Other proteins which elicited IgG1, IgG2a and IgG2b responses were 170KDa, 154KDa and gE. IgA responses were induced by 154KDa, gC, gB, gE and gD. Prominent IgM responses included gB, gC, gD and the 42KDa protein. These results indicate that HSV-1 glycoproteins induce prominent responses in all IgG isotypes except IgG3. The biologic implications of the data are discussed.     

Characterization of B virus glycoprotein antibodies induced by DNA immunization

Genes encoding glycoproteins gB, gC, gD, gE, and gG of herpes B virus (species Cercopithecine herpesvirus 1) were cloned into mammalian expression vector pcDNA3.1/V5-His. Abilities of the plasmid constructs to express recombinant glycoproteins were confirmed by Western blot analysis of transfected CHO-K1 and COS-7 cells. Antibody production was induced in rabbits by intramuscular injections with the expression constructs at four-weekly intervals. Antibodies to gB were detected after the second DNA inoculation, while it took an additional plasmid injection to induce responses to gC, gD and gE. The gG plasmid failed to stimulate antibody production. Antisera ELISA titers varied greatly depending on the gene, with gB inducing highest (21,000) and gE inducing lowest (60) antibody titer. The induced antibodies were predominantly conformation-dependent. The gB, gC, and gD antisera contained HSV cross-neutralizing antibodies, but only gB antisera contained B virus neutralizing antibodies. The gB antisera cross-reacted with HSV antigens in Western blot, ELISA, dot-blot, plaque immunostaining and immunoprecipitation assays, whereas gD and gC antisera were mostly B virus-specific. Thus, polyclonal antibodies to B virus glycoproteins can be generated by DNA immunization and used as diagnostic and research reagents.     

Estimation of the B lymphocyte precursor frequencies to herpes simplex type 1 glycoproteins by a limiting dilution assay

The precursor frequency of B lymphocytes from Balb/c mice producing HSV-1 glycoprotein B (gB), glycoprotein C (gC), and glycoprotein D (gD) antibody was determined by limiting dilution analysis under conditions to detect antibody from the clonal progeny of a single B cell precursor. In spleens of naive mice the average gC frequency was 1/48,917 +/- 5,550, while gD was 1/73,330 +/- 15,898, and gB frequency was in excess of 1/100,000. Immunization with live HSV-1 (KOS) increased the B cell frequencies of all three glycoproteins to approximately 1:3,000; however, the serum gB antibody ELISA titer was fivefold higher than gC or gD.     

Overexpression of gK in gK-transformed cells collapses the Golgi apparatus into the endoplasmic reticulum inhibiting virion egress, glycoprotein transport, and virus-induced cell fusion

Intracellular transport and egress of alphaherpesviruses require the coordinate function of multiple proteins and glycoproteins. Recently, we showed that gK is expressed on infected cell surfaces and that gK cell-surface expression required the presence of the UL20 protein [J. Virol. 77 (2003), 499]. Overexpression of gK by gK-transformed cells blocked transport of enveloped virions from perinuclear spaces and inhibited virus-induced cell fusion caused by gK syncytial mutants [J. Virol. 69 (1995), 5401]. Therefore, we investigated whether altered synthesis and transport of gK was responsible for the observed gK-mediated interference phenomena. HSV-1 infection of the gK-transformed cell line Vero (gK9) caused a profound entrapment of gK in the endoplasmic reticulum and total inhibition of gK cell surface expression. In addition, gK drastically inhibited intracellular transport and maturation of gD and caused substantial defects in Golgi-dependent glycosylation of gB. Visualization of intracellular organelles via confocal microscopy revealed a profound collapse of the Golgi apparatus into the endoplasmic reticulum. These results were analogous to those observed in the presence of brefeldin A, a known Golgi disruptor. Therefore, virion entrapment within perinuclear spaces and inhibition of glycoprotein transport are due to gK-mediated collapse of the Golgi apparatus.     

Characterisation and physical mapping of an HSV-1 glycoprotein of approximately 115 X 10(3) molecular weight

A type-specific monoclonal antibody that efficiently neutralises HSV-1 immunoprecipitated a glycoprotein of slightly greater electrophoretic mobility than gB from HSV-1 infected cells. Pulse and pulse chase experiments indicate that this glycoprotein is distinct from HSV-1 glycoproteins gB, gC, gD, and gE. This was confirmed by the reactions of LP11 with a series of intertypic recombinants the results of which indicate that the LP11 target gene is located close to the HSV-1 thymidine kinase gene between map positions 0.28 and 0.31. In accordance with the presently agreed convention this glycoprotein should be designated gH-1, and it may correspond to the 110K glycoprotein described by S. D. Showalter, M. Zweig, and B. Hampar (1981), Infect. Immun. 34, 684-692. Antibody LP11 inhibits plaque formation when added to cell monolayers after infection suggesting that gH-1 may play a role in cell-to-cell spread of infectious virus.     

Intra-nuclear localization of two envelope proteins,gB and gD,of herpes simplex virus

Summary The envelopes of herpes simplex virus (HSV) particles are acquired from the inner nuclear membrane (INM) of the infected cell and virus-coded glycoproteins are present in the envelope of mature virions. Our ultrastructural study examined the process of virus envelopment and the targeting of two major viral glycoproteins, gB and gD, to the INM in HSV-infected human embryonic fibroblasts. It was shown that envelopment and transport of virus particles from the nucleus is facilitated by the formation of a dynamic tubulo-reticulum arising from the INM. Capsids were assembled in the nucleus and collected within INM tubules which protruded into the perinuclear space and thence into the cisternae of the endoplasmic reticulum (ER). Envelopment occurred by constriction and fusion of the tubular channel walls, releasing enveloped virions into the ER. Transport to the cell surface took place in membrane-bound compartments and probably followed the normal secretory pathway through the Golgi apparatus. Immunogold probes, tagged with specific monoclonal antibodies, were used to localize gB and gD during the process of virus maturation. Cytoplasmic membranes were not labelled, but probes bound inside the nucleus, mainly at sites of virus assembly. Labelling occurred on the nucleoplasmic side of the INM which surrounded capsids in the process of envelopment, but not on the outside of that membrane, although characteristic gB glycoprotein spikes were labelled on the envelopes of extracellular virus particles and on virions intrans-Golgi transport vesicles just prior to their release from the infected cell. gB was not detected on the surface of enveloped virions in the perinuclear space, or the cisternae of the ER orcis-Golgi, which suggests that the specific epitope was masked during that stage of intracellular processing. gD probes bound to virion envelopes and also to the tegument region of some particles found in both perinuclear and extracellular sites. We postulate that precursor core proteins for both gB and gD are transported first to the nucleus, and then, together with maturing capsids, are targeted to the INM, and later inserted into viral envelopes at the site of budding. Post-translational glycosylation of envelope proteins could occur as virus particles exit the nucleus and travel through the ER and Golgi compartments.     

Cellular expression of gH confers resistance to herpes simplex virus type-1 entry

Entry of herpes simplex virus-1 (HSV-1) into cells requires a concerted action of four viral glycoproteins gB, gD, and gH-gL. Previously, cell surface expression of gD had been shown to confer resistance to HSV-1 entry. To investigate any similar effects caused by other entry glycoproteins, gB and gH-gL were coexpressed with Nectin-1 in Chinese hamster ovary (CHO) cells. Interestingly, cellular expression of gB had no effect on HSV-1(KOS) entry. In contrast, entry was significantly reduced in cells expressing gH-gL. This effect was further analyzed by expressing gH and gL separately. Cells expressing gL were normally susceptible, whereas gH-expressing cells were significantly resistant. Further experiments suggested that the gH-mediated interference phenomenon was not specific to any particular gD receptor and was also observed in gH-expressing HeLa cells. Moreover, contrary to a previous report, gL-independent cell surface expression of gH was detected in stably transfected CHO cells, possibly implicating cell surface gH in the interference phenomenon. Thus, taken together these findings indicate that cellular expression of gH interferes with HSV-1 entry.     

The immune response to herpes simplex virus: comparison of the specificity and relative titers of serum antibodies directed against viral polypeptides following primary herpes simplex virus type 1 infections

Employing an immunoblotting technique, the polypeptide specificity and relative titers of anti-HSV IgG reactive with denaturation-resistant epitopes on HSV proteins were determined in patients experiencing primary HSV-1 infections at various anatomical sites. Early sera from previously seronegative patients with primary HSV-1 infections were found to have comparatively low levels of antibody directed against the major viral glycoprotein antigens (gB, gC, and gD) relative to titers present in sera of individuals with long-standing, latent orofacial HSV-1 infections. Patients with primary infections did however have high titers of antibody directed against a series of low molecular weight HSV polypeptide antigens. These antigens were found to be antigenically related to a structural component of virion nucleocapsids. At later times postinfection, titers of antibodies directed against other viral polypeptides including the major glycoproteins increased to levels more closely approximating those observed in latently infected individuals. These results indicate that the anti-HSV IgG detected by immunoblot analysis which appears earliest following primary infection is not directed against the known major infected cell or virion glycoprotein surface antigens but rather against an internal capsid protein of HSV.     

Use of monoclonal antibodies for analysis of antibody-dependent immunity to ocular herpes simplex virus type 1 infection.

Monoclonal antibodies specific for the five major glycoproteins of herpes simplex virus type 1 (HSV-1) were tested for their capacity to mediate immunity to ocular HSV-1 infection. The specificity of the immunoglobulin made by each monoclone was determined by immunoprecipitation of [14C]glucosamine-labeled polypeptides from detergent-solubilized HSV-1-infected cells. Of the five monoclonal antibodies studied, two immunoprecipitated glycoproteins gA/B, one immunoprecipitated glycoprotein gC, one immunoprecipitated glycoprotein gD, and one immunoprecipitated glycoprotein gE. All five were effective in passively transferring immunity to mice when they were given 4 to 24 h after HSV-1 infection on an abraded cornea. Four of the monoclonal antibodies were also evaluated for their capacity to neutralize HSV-1 and to promote complement-mediated cell lysis and antibody-dependent cellular cytotoxicity. It was found that none of these in vitro assays correlated with the protective activity of the antibodies in vivo. In fact, one of the monoclonal antibodies was unreactive in all three immunological reactions, even though it was highly effective in promoting recovery from HSV-1 induced ocular disease in vivo. The results suggest that antibodies can interact in vivo with virus-specific glycoproteins gA/B, gC, gD, and gE to initiate recovery from HSV-1-induced ocular disease, and that the therapeutic effectiveness of a specific monoclonal antibody does not correlate with its immunological reactivity in vitro.     

Studies on herpes simplex virus type 1 glycoproteins using monoclonal antibodies

Monoclonal antibodies against herpes simplex virus type 1 glycoproteins were isolated and utilized to study the synthesis and processing of glycoproteins B, C, and D (gB, gC, gD, respectively). Monoclonal antibodies against both gB and gD had higher virus-neutralizing activity when compared to that of gC. Differences among these glycoproteins were observed in their time of appearance in the virus-infected cells. The presence of gD was detected at a very early stage of infection when compared to gB and gC. The localization of these glycoproteins during their synthesis and processing was studied.     

Cell fusion induced by herpes simplex virus glycoproteins gB, gD, and gH-gL requires a gD receptor but not necessarily heparan sulfate

To characterize cellular factors required for herpes simplex virus type 1 (HSV-1)-induced cell fusion, we used an efficient and quantitative assay relying on expression of HSV-1 glycoproteins in transfected cells. We showed the following: (1) Cell fusion depended not only on expression of four viral glycoproteins (gB, gD, and gH-gL), as previously shown, but also on expression of cell surface entry receptors specific for gD. (2) Cell fusion required expression of all four glycoproteins in the same cell. (3) Heparan sulfate was not required for cell fusion. (4) Coexpression of receptor with the four glycoproteins in the same cell reduced fusion activity, indicating that interaction of gD and receptor can limit polykaryocyte formation. Overall, the viral and cellular determinants of HSV-1-induced cell fusion are similar to those for viral entry, except that HSV-1 entry is significantly enhanced by binding of virus to cell surface heparan sulfate.     

The immunoglobulin response to individual HSV-1 viral polypeptides: kinetics of the response during primary and secondary experimental infection with herpes simplex virus

We studied the immunoglobulin response to individual viral polypeptides in experimental primary and secondary infection with herpes simplex virus (HSV)-1 in mice. After a single footpad inoculation with 10(5.6) pfu of HSV-1, immunoglobulin to proteins mol. wts (10(3)) 44 and 75 appeared on day 7. Antibodies to gB, gC, gD, 42 x 10(3)- and (48-52) x 10(3)-mol. wt proteins appeared on day 11 and antibody to the major capsid protein, VP154, appeared on day 15 after infection. The secondary immune response was characterised by early production of antibody to gD on day 3 followed by antibodies against the 42 x 10(3)- and 44 x 10(3)-mol. wt proteins on days 4 and 5 respectively. Antibodies to glycoprotein gC and gB were delayed until day 7 of the secondary immune response. In both primary and secondary immune responses the responses against proteins of mol. wts (10(3)) 42 and 44 were particularly intense and of high titre. We conclude that the kinetics of anti-polypeptide antibody appearance is markedly asynchronous; and that the anti-glycoprotein responses occur too late in primary infection to contribute to viral clearance.