Large clonal expansions of CD8+ T cells in acute infectious mononucleosis.

Primary infection with Epstein-Barr virus often results in the clinical syndrome of acute infectious mononucleosis (glandular fever). This illness is characterized by a striking lymphocytosis, the nature of which has been controversial. We show that large monoclonal or oligoclonal populations of CD8+ T cells account for a significant proportion of the lymphocytosis and provide molecular evidence that these populations have been driven by antigen. The results suggest that the selective and massive expansion of a few dominant clones of CD8+ T cells is an important feature of the primary response to this virus.

Cytotoxic T cell responses to multiple conserved HIV epitopes in HIV-resistant prostitutes in Nairobi.

Many people who remain persistently seronegative despite frequent HIV exposure have HIV-specific immune responses. The study of these may provide information about mechanisms of natural protective immunity to HIV-1. We describe the specificity of cytotoxic T lymphocyte responses to HIV in seronegative prostitutes in Nairobi who are apparently resistant to HIV infection. These women have had frequent exposure to a range of African HIV-1 variants, primarily clades A, C, and D, for up to 12 yr without becoming infected. Nearly half of them have CTL directed towards epitopes previously defined for B clade virus, which are largely conserved in the A and D clade sequences. Stronger responses are frequently elicited using the A or D clade version of an epitope to stimulate CTL, suggesting that they were originally primed by exposure to these virus strains. CTL responses have been defined to novel epitopes presented by HLA class I molecules associated with resistance to infection in the cohort, HLA-A*6802 and HLA-B18. Estimates using a modified interferon-gamma Elispot assay indicate a circulating frequency of CTL to individual epitopes of between 1:3,200 and 1:50,000. Thus, HIV-specific immune responses-particularly cross-clade CTL activity- may be responsible for protection against persistent HIV infection in these African women.

Cross-clade recognition of p55 by cytotoxic T lymphocytes in HIV-1 infection.

OBJECTIVES: To evaluate cross-clade recognition of p55 antigen by cytotoxic T lymphocytes (CTL) in persons infected with diverse clades of HIV-1; to facilitate the development of a CTL-inducing vaccine to prevent transmission of multiple clades of HIV-1. DESIGN: Experiments were designed to evaluate whether persons in Uganda and the United Kingdom, infected with diverse clades of HIV-1, have CTL capable of recognizing and killing autologous target cells infected with recombinant vaccinia viruses (rVV) expressing the Gag protein from A, B, C and D clade HIV-1. The extent of cross-reactivity within such individuals, each infected with characterized virus, might reflect the type of cross-reactive immune response inducible by a monovalent vaccine. METHODS: Asymptomatic HIV-positive individuals were fully tissue-typed by ARMS (amplification of refractory mutation system) polymerase chain reaction. rVV expressing the Gag protein from identified A, B, C and D viruses were prepared. CTL were cultured and tested for cytolytic activity on autologous rVV-infected or peptide-pulsed B cells. RESULTS: Ugandan patients had inducible CTL responses recognizing A, B, C and D clade HIV-1 Gag. The majority of UK patients had inducible CTL responses that recognized two or more clades. No patient showed any HIV-2 cross-reactivity. Cross-reactive responses were characterized in three Ugandan patients. CONCLUSIONS: Most patients tested mounted cross-reactive CTL responses that recognized Gag proteins from clades of HIV-1 other than those with which they were infected.

A microELISA assay for detection of anti-HLA activity of mouse monoclonal antibodies using an Astroscan 2100 automated plate reader.

A microenzyme-linked immunosorbent assay (microELISA) method has been developed using an Astroscan 2100 system automated plate reader which was initially designed for tissue typing by a two colour fluorescent microcytotoxicity assay. A 96-well plate ELISA used for screening mouse monoclonal antibodies raised against surface HLA antigens has been modified for use with the Astroscan plate reader and 72-well typing trays. The existing substrate 4-methylumbelliferyl-beta-D-galactoside (4MUG) has been replaced with fluorescein-di-beta-D-galactopyranoside (FDG), to provide a wavelength (530 nm) detectable by the Astroscan or other automated plate readers designed for reading microcytotoxicity assay plates. The assay volumes have also been reduced tenfold for use with Terasaki microtest plates. The assay now has the major advantage of requiring only 5 microliters of test supernatant allowing hybridomas to be screened earlier during a fusion and on a wider cell panel. The use of the large panel which includes B lymphoblastoid cell lines (B-LCL) and mouse L cell transfectants expressing HLA genes, reduces the length of time the hybridomas need to be kept in tissue culture before selection. Other advantages include the reduction in the number of target cells required, smaller volumes of reagents throughout the assay and the ability to screen cytotoxic as well as non-cytotoxic monoclonal antibodies. The sensitivity of this microELISA proved to be comparable with the original assay and so provides an efficient screening method for monoclonal antibodies.

Broadly cross-reactive HIV-specific cytotoxic T-lymphocytes in highly-exposed persistently seronegative donors.

HIV-specific cytotoxic T-lymphocytes (CTL) are believed to play a key part in the control of virus levels throughout HIV infection. An important goal of a potential prophylactic vaccine against HIV is therefore to elicit a strong CTL response which is broadly cross-reactive against a diverse range of HIV strains. We have detected HIV-specific CTL in two groups of highly-exposed but persistently seronegative female sex workers in Africa which show extensive cross-reactivity between different viral sequences. In a small group of women exposed to both HIV-1 and HIV-2 in Gambia, studied over 4 years, we have repeatedly detected HLA-B35-restricted CTL which exhibit cross-reactivity between the HIV-1 and HIV-2 sequences of the CTL epitopes. In women with particularly intense exposure to what are likely to be multiple clades of HIV-1 in Nairobi Kenya, we have detected CTL directed towards epitopes conserved between HIV-1 clades. In neither group is there any evidence that variation in CCR5 sequence or expression is responsible for their apparent resistance to HIV infection. However, in seropositive donors from Oxford infected with African strains of HIV-1, we have defined CTL responses which are specific for particular clades and have mapped some unique A clade CTL epitopes, together with others to highly-conserved regions of the virus. Further information about the extent of cross-reactive CTL immunity will be important for future vaccine design and evaluation.

Defining the allelic variants of HLA-A30 in the Sardinian population using amplification refractory mutation system--polymerase chain reaction.

HLA-A30 is present in the Sardinian population at a frequency of 23%. We have designed a system using nested ARMS-PCR to determine the relative frequencies of the HLA-A*30 allelic variants (A*3001, A*3002, and A*3003) within this population. The use of a nested PCR approach, in which the first-round reaction provides HLA-A*30 specificity and template DNA for the subsequent nested reactions, is a powerful means of discriminating between alleles of very similar sequence. Using this method, we performed subtyping of 35 serologically defined HLA-A30 Sardinian individuals, and taking into account homozygotes, identified 38 A*30 alleles. Of these, 33 typed as A*3002, four typed as A*3001, and one sample did not conform to the patterns of reactivity of any of the published A*30 alleles. Haplotype information showed strong linkage disequilibrium between A*3002 and B18. This study underlines the potential of DNA-based methods for typing HLA class I in terms of adding further levels of definition to studies of population structure and also as a means of identifying new alleles.

Novel, cross-restricted, conserved, and immunodominant cytotoxic T lymphocyte epitopes in slow progressors in HIV type 1 infection.

HIV-specific cytotoxic T lymphocytes (CTLs) play an important role in the immune response to HIV infection. Long-term nonprogressors (LTNPs) or slow progressors (SPs) in HIV infection may make qualitatively different CTL responses compared to those generated by seropositive individuals who progress to disease at a faster rate. The class I molecule HLA-B*57 has been identified as one restriction element overrepresented in SP groups studied, and, together with the closely related molecule HLA-B*58, occurs commonly in ethnic groups where HIV is most prevalent. In this study, we have identified five new HLA-B*57-restricted CTL epitopes recognized by SP donors, one of which is also HLA-B*5801 restricted. These HLA-B*57-restricted responses represent the dominant HIV-specific CTL response in each of the SP donors tested. These and other such epitopes may be an important component in future vaccine design.

HIV type 1 resistance in Kenyan sex workers is not associated with altered cellular susceptibility to HIV type 1 infection or enhanced beta-chemokine production.

A small group of women (n = 80) within the Nairobi-based Pumwani Sex Workers Cohort demonstrates epidemiologic resistance to HIV-1 infection. Chemokine receptor polymorphisms and beta-chemokine overproduction have been among the mechanisms suggested to be responsible for resistance to HIV-1 infection. This study attempts to determine if any of those mechanisms are protecting the HIV-1-resistant women. Genetic analysis of CCR5 and CCR3 from the resistant women demonstrated no polymorphisms associated with resistance. Expression levels of CCR5 among the resistant women were shown to be equivalent to that found in low-risk seronegative (negative) controls, while CXCR4 expression was greater among some of the resistant women. In vitro infection experiments showed that phytohemagglutinin (PHA)-stimulated peripheral blood mononuclear cells (PBMCs) from resistant women were as susceptible to infection to T cell- and macrophage-tropic North American and Kenyan HIV-1 isolates as were the PBMCs from negative controls. No significant difference in circulating plasma levels of MIP-1alpha and MIP-1beta were found between the resistant women and negative or HIV-1-infected controls. In vitro cultures of media and PHA-stimulated PBMCs indicated that the resistant women produced significantly less MIP-1alpha and MIP-1beta than did negative controls and no significant difference in RANTES levels were observed. In contrast to studies in Caucasian cohorts, these data indicate that CCR5 polymorphisms, altered CCR5 and CXCR4 expression levels, cellular resistance to in vitro HIV-1 infection, and increased levels of beta-chemokine production do not account for the resistance to HIV-1 infection observed among the women of the Pumwani Sex Workers Cohort.

Defining the common subtypes of HLA A9, A10, A28 and A19 by use of ARMS/PCR.

We describe sequence-specific primer (SSP) combinations for use in a one-step polymerase chain reaction (PCR) typing system to determine HLA-A locus subtypes of A9 (A23, A24), A10 (A25, A26, A43), A28 (A*6801, A*6802, A*6901) and A19 (A*2901, A*2902, A*3001, A*3002, A31, A32, A33) from genomic DNA. SSP's were designed on the basis of the amplification refractory mutation system (ARMS) in which a mismatch at the 3' residue inhibits non-specific amplification. The SSP combinations described extend our low-resolution typing system, to provide a high-definition typing of the HLA-A locus.

Identification of a new HLA-B*78 allele in a Hispanic family.

A new B*78 variant, B*7804, was detected in three members of a Hispanic family. The novel B*78 sequence differs from B*78021 by two substitutions: T at nucleotide 527 (all other B*78s have A) and T at nucleotide 583 (all other B*78s have a C). Both nucleotide substitutions encode amino acid changes at codons 152 and 171, respectively.

A preliminary analysis of the 11th International Histocompatibility Workshop monoclonal antibodies.

In the 11th International Histocompatibility Workshop prescreening study, HLA Class I and Class II monoclonal antibodies (mAb) were tested on panels of peripheral T and B cells and homozygous typing cell (HTC) lymphoblastoid lines. The 102 Class I mAb, of which 64 were new, were screened on 63 HTC lines and 20 peripheral T cells. The 144 Class II mAb were tested on 63 HTC lines, 20 peripheral B and 10 peripheral T cells. For Class I, 45 of the 102 mAbs tested proved to have clear and identifiable reaction patterns on both PBL and HTC lines and included several interesting new HLA-A and -C locus antibodies. Of the 144 Class II mAb, of which 99 were new, 18 DR, 22 DQ and 7 DP antibodies were judged excellent. The DR antibodies included some which were antigen-specific but the majority were complex antibodies, while the DQ antibodies, as previously, identified the majority of DQ antigens both singly and in combinations. The encouraging number of DP antibodies was interesting in the limited range of sites they appear to detect.

HLA-A locus DNA typing of the 4AOH cell panel by arms PCR.

As part of the Fourth Asia-Oceania Histocompatibility (4AOH) Workshop, the authors have demonstrated a method of DNA-based tissue typing of the HLA-A locus using ARMS-designed primers in a panel of specific PCR reactions. The study was carried out blind under Workshop conditions and the results confirm the method as an accurate means of determining HLA-A locus tissue types.

Complete sequence analysis of the A*1103 allele.

Here we report the full-length sequence of a novel A*11 variant. The variant was identified by ARMS-PCR and serology, the sequence was confirmed by cloning and subsequent sequencing. This variant, A*1103, found in a family of oriental origin, resembles the A*1101 sequence in exon 2 but differs in exon 3 with regard to codons 151 and 152. The polymorphism's result in two amino acid substitutions (one conserved (His->Arg), one introducing a negative charge (Ala->Glu)) located in the alpha2 helical region. The arginine at amino acid position 151 is rare amongst Alocus alleles and is besides A*1103 only observed in A*29 variants, the glutamine at amino acid position 152 is shared with A*0301, A*25, *26, *34 variants and the A*02 subtypes A*0203, *0213 and *0226. In fact, the amino acid motif comprising codons 151 and 152 is unique to A*1103 among Alocus alleles, but is common to C-locus alleles.

The monoclonal antibody TAL16.1 recognizes the aspartic acid residue at position 70 in DRB gene products.

A polymorphic monoclonal antibody (TAL16.1), raised against a mouse L-cell transfectant expressing the human DRB5*0101 gene from the HLA-DR15(2) Dw2 DR51 haplotype was shown to have a complex pattern of reactivity to DRB gene products. The antibody bound to a transfectant expressing the DRB5*0101 allele against which it was produced but not to a transfectant expressing the DRB1*1501 allele. These alleles of the DRB1 and DRB5 genes are usually coexpressed on DR15(2) Dw2 DR51 cells. A comparison of the HLA-DRB amino acid sequences of reactive and non-reactive cells identified an aspartic acid residue at position 70, conserved in all antibody-positive cells and absent in antibody-negative cells, which was postulated as being responsible for conferring the specificity of the antibody. The aspartic acid residue at position 70 is present in DRB5*0101 and DRB5*0102 alleles but absent in DRB5*0201 and DRB5*0202 alleles, allowing the antibody to distinguish between these splits of the DR51 serological specificity. TAL16.1 also binds to the product of the DRB1*0103 allele and discriminates between cells with a DR103 specificity and the other DR1 subtypes, DRB1*0101 and DRB1*0102. In this report the value of transfectants as immunogens for use in the production of monoclonal antibodies of predetermined specificity and as tools for the fine mapping of antibody specificity is discussed.

Microcytotoxicity assay using mouse L cells transfected with human MHC class II genes. A method and its application.

Modifications of the standard microcytotoxicity assay make it possible to use this technique to screen both alloantisera and monoclonal antibodies with mouse L cells transfected with Class II genes. It is necessary to maintain a protein-rich environment in order to prevent nonspecific complement lysis. Selection of the complement itself is also an important factor, the best results being achieved using a commercially available complement that had previously been absorbed with mouse cells and used at a dilution of 1/8. Using this modified method with transfectants of DW2 origin we could show that alloantisera against DRw15 recognize the DRB1*1501 gene product, whereas broad DR2 sera react only with the DRB5*0101 product. This technique can be applied successfully to study the fine specificity of polymorphic monoclonal antibodies, as shown by the reactivity of HU-30 which binds to the LDR2b transfectant and not to the LDR2a, indicating that the antibody recognizes an epitope present on the DRB1 chain and not the DRB5 chain of DR2 cell lines.

Mechanisms of loss of HLA class I expression on colorectal tumor cells.

For several years this laboratory has studied the expression of HLA class I on established colorectal tumor cell lines and on fresh tumors. We review here the mechanisms by which colorectal tumor cells may lose surface expression of HLA class I molecules. Several independent mechanisms have been identified, including loss or mutations in beta 2-microglobulin genes, loss of HLA heavy chain genes, selective lack of expression of HLA alleles, and regulatory defects in HLA expression including loss of expression of the peptide transporters associated with antigen processing (TAP). The data suggest that colorectal tumor cells may evade tumor specific, HLA restricted immune attack by loss of HLA class I expression through a number of mechanisms.

Tissue typing the HLA-A locus from genomic DNA by sequence-specific PCR: comparison of HLA genotype and surface expression on colorectal tumor cell lines.

A system devised for tissue typing the HLA-A locus by PCR from genomic DNA has been used to investigate abnormalities of HLA expression in a panel of 30 colorectal tumor cell lines, by comparing the HLA-A locus genotype with surface expression of HLA. Three cell lines showed complete lack of HLA expression associated with failure to express beta 2-microglobulin. In two other cell lines, loss of expression of HLA-A2 was observed, in spite of the presence of the gene in genomic DNA. Eleven cell lines gave a single HLA-A locus specificity on PCR typing. In one of these cell lines we have demonstrated the loss of an HLA-A locus gene in the tumor cell by comparison with DNA from a lymphoblastoid B-cell line derived from the same patient. These data indicate that at least three independent mechanisms were involved in the loss of HLA expression on the colorectal tumor cell lines.

HLA-A, -B, -C, -DRB1, DRB3, DRB4, DRB5 and DQB1 polymorphism detected by PCR-SSP in a semi-urban HIV-positive Ugandan population.

PCR-SSP was used to HLA-type a cohort of Ugandan HIV-positive individuals. The results represent a more comprehensive description of HLA in an African population than previously described and are in concordance with data from a general Black population. Substantial differences exist between this population and Caucasoid populations in which immunological responses to HIV have been investigated; this emphasises that the main HLA-restrictive elements for HIV-specific cytotoxic T lymphocytes will most likely be different for each population.

Loss of HLA class-I alleles, heavy chains and beta 2-microglobulin in colorectal cancer.

Using immunohistochemical methods, we have analysed colorectal biopsies of normal mucosa, metaplastic polyps (5 cases), adenomas (15 cases) and adenocarcinomas (70 cases) with 13 monoclonal antibodies (MAbs) to allelic products of the HLA-A, B, C loci. Nine of the 70 carcinomas showed total loss of HLA Class-I molecules due to an underlying defect regarding not only the expression of beta 2-microglobulin (beta 2-m), but also the heavy chains of HLA A, B and C loci, or both. Much commoner was a loss of one or more Class-I alleles as follows: A1/Aw36 (completely lost in 4 of 29 cases and focally lost in another 2), A2 (in 1 of 37 cases), A3 (in 2 of 14 cases), A1 1/28/31/33 (in 3 of 11 cases), B7 (in 3 of 13 and focally in 1 other case), B17 (in 1 case), Bw4 (in 8 of 45 and focally in another 6), Bw6 (in 9 of 62 and focally in another 3). Focal selective loss (Bw6 and a combined A1-Bw6), was observed in 2 adenomas. Normal colonic mucosa, as well as stromal and lymphoid cells present between the neoplastic glands, were studied in each case as a control. A particular allele was only considered to be lost by the malignant cells if it was still expressed on these adjacent tissues.