Identification of key peptide-specific CD4+ T cell responses to human cytomegalovirus: implications for tracking antiviral populations.

Human cytomegalovirus (CMV) infection is normally controlled effectively by the immune response, including CD4(+) T cells. Large numbers of these cells are present in healthy seropositive individuals but their loss in immunosuppression leads to reactivation and disease. Tracking such responses in vivo is hampered by poor definition of their peptide targets. In this study, we defined the key targets of the peptide-specific CD4(+) T cell responses to the CMV pp65 protein using functional assays and a peptide library. Despite a good deal of interindividual variation in the numbers of peptides recognized, responses to CMV pp65 were strikingly targeted at three key epitopes. A response to one or more of these three key peptides was seen in all individuals tested (P < 0.0001) and this finding was tested and reproduced in a second independent population. The most common response identified was that to a DR53 restricted epitope, aa281-295. HLA-DR1 restricted CMV pp65-specific populations, although reproducibly detected, were of low frequency ex vivo. However, it was possible to detect and phenotype these cells using an enrichment protocol and this revealed them to have 'effector memory' status although, in contrast to CD8(+) T cell responses, these were CD45RA(-). These data suggest that CD4(+) T cell responses to CMV can be identified reliably using a pool of just three peptides. This simple approach will provide a robust and reliable as well as economic method for tracking peptide specific populations in health and disease.

Longitudinal mapping of protective CD4+ T cell responses against HCV: analysis of fluctuating dominant and subdominant HLA-DR11 restricted epitopes.

Cellular immunity plays an important role in the control of persistent virus infections such as hepatitis C virus (HCV). Antiviral CD4(+) T cell responses have been shown to accompany resolution of acute disease and there is also a consistent association between HLA Class II genes, notably HLADRB1*1101 (and the closely linked HLADQB1*0301) and disease resolution. We initially mapped longitudinal CD4(+) T cell responses in an individual after spontaneous resolution of acute HCV, and identified three HLA-DR11-restricted responses which vary in immunodominance over time. Functional assays and HLA Class II tetramer staining revealed one to be a response to a commonly recognized epitope, NS3(1248-1261), although cytokine capture assays showed these specific cells to be at a very low frequency. In this patient, and in others reported, this most frequently recognized HLA-DR11 restricted epitope is not immunodominant. We analysed whether sequence variability within and between genotypes might account for differences in recognition of HLA-DR11 restricted epitopes. We found that a limited number, including NS3(1248-1261), showed extreme sequence conservation. Within NS3, the ability of peptides to accept amino acid substitutions was clearly related to the structure of the protein. Overall the data provide a deeper understanding of the relationship between protein structure and variability of HLA-DR11 restricted peptides and may explain the apparent dominance of responses to NS3(1248-1261) across studies but not within an individual immune response.

Evidence for lack of cross-genotype protection of CD4+ T cell responses during chronic hepatitis C virus infection.

CD4+ T lymphocyte responses are thought to play a major role in control of the hepatitis C virus (HCV). Few, however, have been mapped down to the level of peptide and HLA restriction. Furthermore, the ability of such T cells to respond to viruses which differ in genotype has not been addressed in detail. In most cases of persistent infection with HCV, CD4 proliferative responses are weak or absent. From a large cohort of persistently infected patients, we identified an individual with unusually robust and persistent responses in the face of chronic infection. We firstly mapped two peptide epitopes to regions of the nonstructural protein NS4 (aa1686-1705 and aa 1746-1765). However, in contrast to the genotype 1a derived antigens used for mapping, the infecting virus was identified as genotype 3a. Strikingly, the patient's CD4 response to these epitopes were specific only for the genotype 1a sequence, and did not recognize genotype 3a synthetic peptides. Serologic assays indicated that prior exposure to HCV of genotype 1 had occurred. This patient therefore maintains strong CD4 proliferative responses which are genotype specific and not cross-reactive. The apparent 'misdirection' of these nonprotective responses has important implications for the role of natural and vaccine induced CD4 responses in the face of variable viruses.

HIV-1 variation diminishes CD4 T lymphocyte recognition.

Effective long-term antiviral immunity requires specific cytotoxic T lymphocytes and CD4(+) T lymphocyte help. Failure of these helper responses can be a principle cause of viral persistence. We sought evidence that variation in HIV-1 CD4(+) T helper epitopes might contribute to this phenomenon. To determine this, we assayed fresh peripheral blood mononuclear cells from 43 asymptomatic HIV-1(+) patients for proliferative responses to HIV-1 antigens. 12 (28%) showed a positive response, and we went on to map dominant epitopes in two individuals, to p24 Gag restricted by human histocompatibility leukocyte antigen (HLA)-DR1 and to p17 Gag restricted by HLA-DRB52c. Nine naturally occurring variants of the p24 Gag epitope were found in the proviral DNA of the individual in whom this response was detected. All variants bound to HLA-DR1, but three of these peptides failed to stimulate a CD4(+) T lymphocyte line which recognized the index sequence. Antigenic variation was also detected in the p17 Gag epitope; a dominant viral variant present in the patient was well recognized by a specific CD4(+) T lymphocyte line, whereas several natural mutants were not. Importantly, variants detected at both epitopes also failed to stimulate fresh uncultured cells while index peptide stimulated successfully. These results demonstrate that variant antigens arise in HIV-1(+) patients which fail to stimulate the T cell antigen receptor of HLA class II-restricted lymphocytes, although the peptide epitopes are capable of being presented on the cell surface. In HIV-1 infection, naturally occurring HLA class II-restricted altered peptide ligands that fail to stimulate the circulating T lymphocyte repertoire may curtail helper responses at sites where variant viruses predominate.

Antagonism of cytotoxic T lymphocyte-mediated lysis by natural HIV-1 altered peptide ligands requires simultaneous presentation of agonist and antagonist peptides.

Mutations in human immunodeficiency virus (HIV) cluster in cytotoxic T lymphocyte (CTL) epitopes (Phillips, R. E. et al., Nature 1991. 354: 453) and are subject to immune-mediated positive selection (Price, D. A. et al., Proc. Natl. Acad. Sci. USA 1997. 94: 1890). We studied the effects of naturally occurring mutations in the HIV-1 p17 Gag HLA A2 restricted epitope SLYNTVATL on recognition by anti-HIV CTL. Most of these naturally occurring mutants escaped killing by one CTL line and the majority acted as CTL antagonists. We also investigated whether CTL exposed to a strict antagonist peptide restricted by HLA A2 were unresponsive when exposed to targets presenting the wild-type sequence. The results show that antagonism of anti-HIV CTL killing requires the simultaneous presence of agonist and antagonist peptide. We found no evidence that CTL exposed to an antagonist received a functionally negative signal since these CTL retained an unimpaired capacity to lyse targets bearing wild-type peptide.

Human vascular endothelial cells process and present autoantigen to human T cell lines.

The effectiveness of cultured human umbilical vein endothelial cells as accessory cells for T cell activation has been investigated using T cell clones and lines derived from patients with myasthenia gravis which were specific for different epitopes on the alpha subunit of the human acetylcholine receptor. The endothelial cells were induced with IFN-gamma to express HLA-DR and -DQ at high and low levels respectively. They could then efficiently present specific peptides of the alpha subunit to an HLA-DR- and an HLA-DQw5-restricted T cell line. They could also process epitopes for both T cell lines from the full-length recombinant alpha subunit (r1-437) of the human acetylcholine receptor, where the known epitopes are 80 amino acid residues apart. The endothelial presentation of r1-437, but not of the peptides, was sensitive to chloroquine inhibition. Presentation appeared slightly less efficient (by 1.5- to 3.0-fold) with endothelial cells than with presenting cells from peripheral blood. This may reflect differences in accessory signalling since mAb blocking studies suggested that ligands for CD28 provided important accessory signalling by peripheral blood presenting cells while LFA-3 was used by endothelial cells.

Critical role for the Val/Gly86 HLA-DR beta dimorphism in autoantigen presentation to human T cells.

Helper T lymphocytes recognize fragments of foreign (or self) antigens in the peptide-binding clefts of major histocompatibility complex class II molecules; their activation is a crucial step in the induction of many immune and autoimmune responses. While studying the latter, we raised a T-cell line from the thymus of a myasthenia gravis patient against recombinant alpha subunit of the human acetylcholine receptor, the target of this autoimmune disease. The line responds to the 144-156 region of the human sequence and not to the same region of the electric fish homolog, which differs by only three residues. These CD4+ T cells recognize this epitope only in the context of HLA-DR4 class II molecules, of which the variants with Gly86 are absolutely required. Thus the naturally occurring alternatives Dw14.2 (Gly86) and Dw14.1 (Val86)--which differ only at this one position in the entire antigen-binding region--show an all-or-nothing difference in presenting activity. This dimorphism at position 86 is widespread, occurring in subtypes of DR1, DR2, DR3, DR5, and DR6 alleles as well as DR4. Since other DR4 subtypes with substitutions at positions 70-74 also fail to present this peptide, and glycine residues can be uniquely flexible, we suggest that this replacement at position 86 acts locally or at a distance by altering the conformation of the peptide-binding cleft. Such profound functional consequences for T-cell recognition as we report here may explain this example of conserved major histocompatibility complex diversity.

Acetylcholine receptor-reactive T lymphocytes from healthy subjects and myasthenia gravis patients.

Peripheral blood lymphocytes from 23 of 114 (20%) myasthenia gravis (MG) patients showed positive T-cell proliferative responses to native acetylcholine receptor (AChR) purified from the electric fish Torpedo, compared with two of 25 (8%) healthy or other neurologic disease controls. Responsiveness appeared to fluctuate seasonally. Long-term T-cell lines and clones could be selected as readily from the two healthy responders as from the MG cases and showed similar culture behavior, CD4+ phenotype, and HLA class II restrictions. One clone from a control cross-reacted with recombinant human AChR alpha chain (r37-429A) and with the synthetic peptide 125-143(S-S) from its sequence. Both these human antigens stimulated primary proliferative responses at substantially higher frequencies (26 to 59%) than native xeno-AChR--in both patients and controls--demonstrating that truly autoreactive T cells are not inevitably deleted during normal T-cell development.

Myasthenic thymus and thymoma are selectively enriched in acetylcholine receptor-reactive T cells.

We compared T-cell proliferative responses to acetylcholine receptor (AChR) and to purified protein derivative (PPD) (of tuberculin) of hyperplastic thymus, thymoma, and blood cells from patients with myasthenia gravis (MG). Hyperplastic MG thymus cells gave significantly higher and more consistent responses to AChR than parallel cultures of autologous blood cells, whereas responses to PPD showed an opposite trend. Thus there was a preferential localization of AChR-reactive T cells in the hyperplastic MG thymus. Furthermore, there was a strong correlation between blood and thymus cell responses to PPD (but not to AChR), arguing that the hyperplastic MG thymus contains a sample of sensitized peripheral T cells. By contrast, both AChR- and PPD-responsive T cells were almost undetectable in thymus from nonmyasthenic patients, which is evidently much less receptive to circulating T cells. Cells from MG thymomas showed the highest stimulations by AChR but did not consistently react to PPD. However, the uninvolved thymus adjacent to these thymomas behaved almost identically to the hyperplastic samples described above. Our interpretation is that AChR-specific T cells are initially sensitized in the MG thymoma but are selectively trapped in the hyperplastic thymus after being primed elsewhere.

A juxta-membrane epitope on the human acetylcholine receptor recognized by T cells in myasthenia gravis.

T cell proliferative responses to synthetic peptides taken from the human nicotinic acetylcholine receptor (AChR) alpha-chain sequence, or to whole AChR purified from electric fish (Torpedo marmorata), have been studied, using blood, thymus, and lymph node cells, from 34 patients with myasthenia gravis (MG) and 17 controls mostly with other neurological diseases. Peptides were selected because they contained amino acid motifs that recur in most defined T cell epitopes. Peptide 257-269 (from the extracellular loop of the AChR alpha-chain between the second and third trans-membrane domains) stimulated cells from six patients and no controls. Peptides from region 125-143 (from the main extracellular 1-210 stretch), which is thought to be an important T cell epitope in rats, provoked responses in 26% of patients and 41% of controls. Two patients responded both to these peptides and to peptide 257-269, thereby implying some heterogeneity of their reacting T cells. Whereas the initial blood T cell samples sometimes responded both to Torpedo AChR and to the 125-143 peptides, T cell lines selected with either antigen subsequently showed no response to the other. This observation suggests that it may be essential to use human AChR sequences for studying truly autoreactive T cells in MG. Finally, no strong association was found between any of the responses to peptides and the HLA types of the responding individuals.

Rubella-specific serum and nasopharyngeal antibodies in volunteers with naturally acquired and vaccine-induced immunity after intranasal challenge.

Thirty-nine volunteers, who were either naturally immune to rubella virus or immune as a result of vaccination with RA 27/3, Cendehill, or To-336 vaccines, were challenged intranasally with high-titered RA 27/3 virus. Before and after challenge, rubella-specific IgG and IgA in serum and nasopharyngeal washings were measured by hemagglutination inhibition, neutralization, and radioimmunoassay. The reinfection rate (at least a fourfold rise in titer of serum antibody by one or more tests) was highest among recipients of Cendehill vaccine. Significant rises in titer were most frequently detected by radioimmunoassay for rubella-specific IgG. After challenge of immune volunteers, rubella-specific IgM was detected in six of the 29 with vaccine-induced immunity. Although high levels of rubella-specific serum and nasopharyngeal IgA before challenge appeared to be associated with protection in recipients of RA 27/3 vaccine, no level of any antibody tested was invariably associated with protection. For comparison, volunteers with vaccine-induced immunity challenged intranasally with the same dose of vaccine after inactivation did not show evidence of reinfection.

Rubella immunity by four different techniques: results of challenge studies.

When 42 sera with low or inconsistent levels of haemagglutination-inhibiting (HAI) antibodies were tested by single radial haemolysis (SRH) radioimmunoassay (RIA), and enzyme-linked immunoassay (ELISA), RIA was shown to be the most reliable test for detecting low levels of antibody. SRH, however, was found to be an acceptable alternative screening test for rubella antibodies and was more reliable than HAI. Although SRH plates prepared in our own laboratory failed to detect antibodies in six sera, in five of the six, antibodies were only at a low level (RIA titre 1:20-1:80). OriVir plates (Orion Diagnostica, Finland) failed to detect low levels of antibody in only three sera. There were six (14.3%) sera which were false positives in the HAI test. These women were shown to be seronegative by radioimmunoassay and, when three of these six volunteers were vaccinated, they developed a typical primary immune response which resembled that developed by 43 seronegative women following vaccination. Fifteen of the young women with consistently low HAI titres and one woman who was seronegative by HAI but seropositive by RIA and ELISA were subsequently vaccinated. Six (37.5%) of these women showed no significant rise in titre by any of the tests employed, while ten had a significant rise in titre, detected by at least one test, with a low level of rubella-specific IgM detectable in one.

HLA antigens and responses to rubella vaccination.

Attempts were made to correlate virus excretion, joint symptoms and antibody response with human leukocyte antigens (HLA) in seronegative adult women given attenuated rubella vaccine. No association was shown between HLA antigens of the A and B loci and excretion of either high or low titres of RA27/3 vaccine among 26 volunteers. However, virus excretion was influenced by such factors as the time of day at which specimens were collected and the method of virus isolation. Our study therefore failed to confirm the hypothesis that certain persons are good 'spreaders' of rubella virus and that this capacity is associated with HLA-A1 and B8. The study of joint symptoms following vaccination with Cendehill, HPV77.DE-5, RA27/3 or To-336 vaccines showed no association between such symptoms and HLA antigens. However, joint symptoms occurred within 7 days of the onset of menstruation in 33 of 47 (70%) vaccinees (P less than 0.01) and it is therefore suggested that hormonal factors must play a role. No association between HLA antigens and haemagglutination inhibition (HAI) antibody titres, 8 weeks after vaccination with RA27/3, was found amongst 34 volunteers.