Reconstitution of genetically regulated responses against random and ordered synthetic polypeptides by methylated bovine serum albumin as analyzed by isoelectric focusing.

In previous publications it was shown by avidity measurements, cross-reactivity patterns and genetic analyses, that the tetrapeptide T-T-G-G is the immuno-dominant epitope of the synthetic polypeptide (T, G)-A--L. In the present study this close immunological relationship between the random multichain copolymer (T, G)-A--L and the ordered analogue (T-T-G-G)-A--L is extended by two additional criteria. First, the immune response against (T-T-G-G)-A--L in H-2k nonresponder mouse strains can be reconstituted to high antibody levels by complexing this antigen to methylated bovine serum albumin, as was tested earlier for (T,G)-A--L. The antibodies elicited upon reconstitution in both antigenic systems are directed mainly against the same determinant, T-T-G-G. Second, isoelectric focusing analysis of specific antisera developed with radiolabeled antigen revealed restricted 7 S IgG antibody populations in high responder and reconstituted high and low responder mice. The spectra were found to be of similar complexity in the (T,G)-A--L and in the (T-T-G-G)-A--L system. From these data it was concluded that the repertoires of specific B cells to T-T-G-G are very similar in high and low responder strains, and the defect in the H-2k low responder systems should be located at the level of T-B cell cooperation.     

Genetically controlled immune responses of inbred mouse strains to conjugates of the ordered peptides (Tyr-Tyr-Glu-Glu) and (Tyr-Glu-Tyr-Glu) with multichain poly-DL-alanine as compared with the response to the random (T,G)-A--L

In previous studies we have shown that the immune response of different inbred mouse strains to the ordered synthetic antigen (Tyr-Tyr-Glu-Glu)-A--L is similar to that observed with the random polypetide (T, G)-A--L in the pattern of response and in the cross-reactivity of specific antibodies with it. By inhibition studies of the hemolytic plaque-forming cell (PFC) assay we have now further confirmed that T-T-G-G is the major determinant in the random (T, G)-A--L. C3H/HeJ mice were found to be nonresponders to (T-T-G-G)-A--L, whereas using the same techniques these mice produced low titers of antibodies to the random (T, G)-A--L. Using the hemolytic PFC technique low numbers of antibody-producing cells were detected in spleens of C3H/HeJ mice. However, the avidity of these antibodies as measured by inhibition of PFC was low as compared with that observed for antibody produced in C3H.SW mice in response to the ordered (Tyr-Tyr-Glu-Glu)-A--L. Genetic analysis experiments demonstrated a close linkage between the ability to respond to (Tyr-Tyr-Glu-Glu)-A--L and the major histocompatibility locus (H-2) of the mouse as was previously shown for (T, G)-A--L. In contrast, no linkage was observed between the immune response potential to (Tyr-Glu-Tyr-Glu)-A--L and the H-2 as indicated by the pattern of response of different inbred and congenic mouse strains and by genetic analysis. Thus, the change in the order of two amino acid residues within the sequence of the tetrapeptide caused an enormous change in the biological specificity of the antibodies produced and in the genetic control of the immune response.     

Analysis of the biological functions and fine specificity of (T,G)-A--L specific T cell clones

Two T cell lines, TPB1 and TPB2, specific for the synthetic polypeptide antigen (T,G)-A--L, were established from (T,G)-A--L primed lymph node cells of C3H.SW(H-2b) mice. Both lines proliferated in the presence of (T,G)-A--L, helped in antibody production in vitro, and secreted IL2 upon stimulation with antigen. The lines differed in the fine specificity of their responses to antigenic stimulation. The line with the broader specificity TPB2 was cloned by limiting dilution, and its derived clones were analyzed. No efficient manifestation of both proliferative activity and helper function could be detected in a single clone. Most of the clones were highly specific to (T,G)-A--L, although 2 of them cross-reacted with the closely related polypeptide (Phe,G)-A--L. Individual clones could trigger B cells for the production of antibodies of the IgM and IgG classes. All helper clones secreted (T,G)-A--L specific helper factors. No correlation was found between efficient secretion of IL2 by the clones and their other biological functions.     

Selective production of human antigen specific helper factor from normal volunteers: implications for human Ir genes.

Based on previous systems for generating helper cells and factors from mouse spleen cell cultures, an in vitro system for the production and detection of human helper factors to the synthetic polypeptide antigens (T,G)-A--L and GAT10 was developed. The factors are made by human peripheral blood leucocytes and are antigen-specific, as judged both by functional criteria and specific binding and elution from antigen columns. Out of the first six volunteers studied two were high responders to (T,G)-A--L, but non-responders to GAT, two responders to GAT but not (T,G)-A--L. One subject made factors to both antigens and the sixth reacted to neither. The antigens chosen are known to be under MHC-linked immune response (Ir) gene control in all animal species tested, with some strains being responders while others are not. The selective responsiveness, different between individuals, thus suggests that the response to these antigens in man is under Ir gene control. Because of the small size of the sample initially studied and HLA typed, it was not surprising that there was no clear-cut association of response with any particular histocompatibility type at the HLA-A, B, C or D locus.     

Covalent linkage of the synthetic adjuvant MDP to the synthetic polypeptide (T,G)-A-L changes the specificity of the immune response at the T and B cell level

It is now well established that the synthetic molecule MDP (N-acetylmuramyl-L-alanyl-D-isoglutamine) can be a good adjuvant of immunity when covalently linked to antigen. The question raised in this work is whether conjugation of antigen to the immunomodulatory molecule MDP can modify the specificity of the antibodies and T cells induced following immunization. Using the well characterized synthetic polypeptide antigen, poly(L-Tyr,L-Glu)-poly(DL-Ala)--poly(L-Lys) [(T,G)-A--L], we show that immunization of C57B1/6 (H-2b) mice with MDP-(T,G)-A--L conjugate elicits at least two types of antibody directed against the poly(DL-Ala)--poly(L-Lys) (A--L) part of the antigen, and against new determinant(s) formed by MDP and a portion of the (T,G)-A--L molecule. Interestingly, the poly-(L-Try L-Glu) side chains thought to constitute the major antigenic determinants of the (T,G)-A--L molecule were not recognized. Lymph node cells from (T,G)-A--L immunized mice can be equally well stimulated in vitro by (T,G)-A--L or by MDP-(T,G)-A--L, whereas lymph node cells from MDP-(T,G)-A--L primed animals can be stimulated only when challenged by the conjugate used for immunization, and not by the free synthetic polypeptide (T,G)-A--L. The data presented here show that the coupling of a low mol. wt molecule such as MDP (mol. wt approx. 500) to an antigen can greatly modify the immune response directed against this antigen. Furthermore, (1) different antibody specificities are elicited depending upon whether the priming is done with free MDP and antigen or with MDP covalently linked to the antigen; (2) although still accessible on the conjugate, an epitope which represents the major antigenic determinant on the free polypeptide appears to be silent when presented on the conjugate; and (3) new determinant(s) formed by the chemical linkage of the polypeptide to the synthetic adjuvant are involved in the priming of T cells.     

Analysis of antigen specific T cell helper function in first degree relatives of patients with systemic lupus erythematosus (SLE).

Fourteen families with first degree relatives of patients with systemic lupus erythematosus (SLE) were studied for the ability of their members to respond to the synthetic polypeptide antigen (T,G)-A-L. The family members were also tested for their HLA determinants. All SLE patients tested responded to (T,G)-A-L as measured by the production of (T,G)-A-L specific T cell helper factors by their antigen activated T cells, confirming our previous findings that 100% of SLE donors responded to (T,G)-A-L in contrast to 50% responders in a control population of healthy donors. The general defect in the regulation of immune responses in SLE patients was further indicated by the demonstration that an SLE patient who is a daughter of non-responder parents to (T,G)-A-L, responded to this genetically regulated antigen. In contrast to our observations with SLE patients, the genetic regulation of the ability to respond to (T,G)-A-L was shown not to be impaired in healthy first degree family members of SLE patients and the segregation of the immune response potential in these families was as expected from an inherited dominant trait.     

Close association between particular I region-determined cell surface antigens and Ir gene-controlled immune responsiveness to synthetic polypeptides in wild rats.

The relationship between major histocompatibility complex (MHC) and genetic control of immune responsiveness to the synthetic polypeptides (T,G)-A--L [poly-(LTyr,LGlu)-poly(DLAla)--poly(LLys)] and (H,G)-A--L [poly(LHis,L-Glu)-poly-(DLAla)--poly(LLys)] has been studied in 26 wild rats. The major histocompatibility complex (MHC) genotype frequencies observed were not different from those expected according to the Hardy-Weinberg formula. More than half of the wild rats carried MHC-linked responder Ir-TGAL and Ir-HGAL genes. High or intermediate responsiveness to (T,G)-A--L and high responsiveness to (H,G)-A--L were always found to be associated with particular I region-determined cell surface antigens. These antigens could be identified serologically and by primary and secondary mixed lymphocyte reactions, and were similar or identical to I region products of (T,G)-A--L high responder or (H,G)-A--L intermediate responder inbred rat strains. The strong association between cell surface antigens and immune responsiveness could be due to linkage disequilibrium or to pleiotropy. Since the same I region-determined cell surface structure could be associated either with high or intermediate anti-(T,G)-A--L antibody titers, the presence of the Ia antigen(s) identified did not seem to guarantee high antibody responsiveness to the test antigen.     

Specificity of antigen binding by T cells; competition between soluble and Ia-associated antigen

Competitive antigen binding experiments were performed with purified T and B cells of C3H.SW (H-2b) mice. As antigen, (T,G)-A--L [poly-L(Tyr,Glu)-poly-DL-ALa-poly-L Lys] was used, both in an Ia-containing form, released by adherent cells (IAC-Puri and Lonai, Eur. J. Immunol. 1980. 10:273), and in regular solution. It was found that regular (T,G)-A-L did not compete with the binding of 125I-labeled-IAC-(T,G)-A--L even at a 10(4)-fold excess, whereas IAC-(T,G)-A--L inhibited binding at 10-fold excess. The specificity of (T,G)-A--L binding to high-responder T and B cells was compared by using related branched synthetic copolymers as competitors. B cells cross-reacted with (T,G)-A--L, (H,G)-A--L, (G)-A--L and (T,G)-Pro--L. In contrast, antigen binding C3H.SW T cells cross-reacted only with (T,G)-A--L and (Phe, G)-A--L to both of which they are Ir gene-controlled high responders. Evidence for the Ir gene control of IAC-binding T cells was obtained by showing that high X low responder F1 hybrid T cells preferentally bind IAC-(T,G)-A--L processed by processor cells deriving from the high-responder parental strain. These data are interpreted to suggest that T cells have high affinity for antigen plus self Ia complexes, whereas they have a much lower, if any, affinity for free antigen. It also follows from the results that the structure of the complex ligand may have a role in defining the specificity, H-2 restriction and Ir gene control of T cells.     

T-cell receptor bias in patients with allergic bronchopulmonary aspergillosis

CD4(+) Th2 helper cell mediated immune responses have been shown to play a crucial role in the pathogenesis of ABPA. HLA and TCR are the candidate genes, which can influence the specificity of these responses. We have previously established a strong association of HLA DR2/5 in ABPA susceptibility. The study was designed to determine whether allergen specific T cell express a limited usage of T cell receptor (TCR) Vbeta gene repertoire in ABPA and to find an association of susceptible HLA-DR determinants with the identified TCR gene segments. TCR Vbeta typing was performed on antigen specific T cell lines from 14 ABPA and 12 nonABPA patients. The majority of ABPA patients (86%) expressed allergen specific T cells with Vbeta13 genes indicating its role in susceptibility, whereas in nonABPA controls, Vbeta1 genes T cell repertoires were predominantly expressed. The unrestricted pattern of Vbeta gene amplification seen before antigen stimulation suggests an oligoclonal expansion of a specific T cell population in response to the allergen Asp f 1 in ABPA and nonABPA patients. The increased usage of Vbeta13 in ABPA and Vbeta1 in nonABPA indicates their importance in susceptibility and resistance, respectively.     

Quantitative separation of antigen-specific murine antibodies by anti-allotype chromatography.

The use of Sepharose-conjugated murine anti-Iga or anti-Igb allo-antisera allowed the quantitative separation of immunoglobulins of the two allotypes. After fractionation of mixtures of anti-(T,G)-A--L antisera obtained from congenic strains differing in immunoglobulin allotype, it was possible to measure the antigen-binding capacity of specific anti-(T,G)-A--L antibodies in each allotype fraction. Analysis of artificial mixtures of immune sera obtained from homozygous Iga and Igb animals showed that this method is quantitative and internally consistent. This method of affinity chromatography was used in the analysis of specific anti (T,G)-A--L antisera from tetraparental mice.     

Analysis of nucleophosmin–anaplastic lymphoma kinase (NPM‐ALK)‐reactive CD8+ T cell responses in children with NPM‐ALK+ anaplastic large cell lymphoma

Cellular immune responses against the oncoantigen anaplastic lymphoma kinase (ALK) in patients with ALK‐positive anaplastic large cell lymphoma (ALCL) have been detected using peptide‐based approaches in individuals preselected for human leucocyte antigen (HLA)‐A*02:01. In this study, we aimed to evaluate nucleophosmin (NPM)‐ALK‐specific CD8⁺ T cell responses in ALCL patients ensuring endogenous peptide processing of ALK antigens and avoiding HLA preselection. We also examined the HLA class I restriction of ALK‐specific CD8⁺ T cells. Autologous dendritic cells (DCs) transfected with in‐vitro‐transcribed RNA (IVT‐RNA) encoding NPM–ALK were used as antigen‐presenting cells for T cell stimulation. Responder T lymphocytes were tested in interferon‐gamma enzyme‐linked immunospot (ELISPOT) assays with NPM–ALK‐transfected autologous DCs as well as CV‐1 in Origin with SV40 genes (COS‐7) cells co‐transfected with genes encoding the patients’ HLA class I alleles and with NPM–ALK encoding cDNA to verify responses and define the HLA restrictions of specific T cell responses. NPM–ALK‐specific CD8⁺ T cell responses were detected in three of five ALK‐positive ALCL patients tested between 1 and 13 years after diagnosis. The three patients had also maintained anti‐ALK antibody responses. No reactivity was detected in samples from five healthy donors. The NPM–ALK‐specific CD8⁺ T cell responses were restricted by HLA‐C‐alleles (C*06:02 and C*12:02) in all three cases. This approach allowed for the detection of NPM–ALK‐reactive T cells, irrespective of the individual HLA status, up to 9 years after ALCL diagnosis.     

Specificity of genes controlling immune responsiveness to (T,G)-A--L and (Phe,G)-A--L

Mice possessing the H-2b haplotype are high responders to the cross-reactive antigens (T,G)-A--L and (Phe,G)-A--L whereas mice with the H-2k haplotype respond only to (Phe,G)-A--L. On the level of cross-immunization we have demonstrated that either (Phe,G)-A--L or (T,G)-A--L primed high responder C3H.SW (H-2b) mice could be boosted with both antigens. On the other hand, low responder C3H/DiSn (H-2k) mice which were primed to (Phe,G)-A--L and thus possess (T,G)-A--L specific antibodies, could not be boosted with (T,G)-A--L to mount a secondary response. Only (Phe,G)-A--L primed and boosted H-2k mice produced high levels of (T,G)-A--L reactive antibodies. Furthermore, the binding of the anti-(Phe,G)-A--L antibodies of either C3H/DiSn or C3H.SW mice to 125I-(T,G)-A--L was better inhibited by guinea-pig anti-idiotypes than the binding of C3H.SW anti-(T,G)-A--L antibodies which are the homologous idiotypes (T,G)-A--L was found to be an equally efficient tolerogen in both high and low responder mice. Thus, when C3H.SW and C3H/DiSn mice were injected with a tolerogenic dose of (T,G)-A--L and then immunized with (Phe,G)-A--L, they were found to be tolerant to (T,G)-A--L antigenic determinants, since they produced only the unique antibodies to (Phe,G)-A--L. These results suggest that the H-2 linked Ir genes controlling antibody response to (T,G)-A--L are not involved in the induction of tolerance to (T,G)-A--L.     

Fine specificity and idiotypic expression of monoclonal antibodies directed against poly(Tyr,Glu)-poly(DLAla)--poly(Lys) and its ordered analogue (Tyr-Tyr-Glu-Glu)-poly(DLAla)--poly(Lys).

In order to study the repertoire of poly(Tyr,Glu)-poly(DLAla)--poly(Lys) [(T,G)-A--L] specific antibodies, monoclonal antibodies were prepared by fusing myeloma cells with spleen cells from C3H.SW mice immunized with (T,G)-A--L and boosted with (Tyr-Tyr-Glu-Glu)-poly(DLAla)--poly(Lys)](T-T-G-G)-A--L]. Eleven clones which secreted homogeneous antibodies were obtained. In general, two families of monoclonal antibodies were detected: those which bind exclusively (T-T-G-G)-A--L and those which bind both (T-T-G-G)-A--L and (T,G)-A--L. Analysis for idiotypic expression revealed that only two antibodies (clones no. 103 and 160), which were found to be similar in their fine specificity, cross-reacted with antibodies against the major idiotypes of (T,G)A--L specific antibodies. Guinea-pig antibodies against clone no. 160 reacted with the polyclonal (T,G)-A--L specific antibodies, whereas antibodies against 103 monoclonal antibodies did not react with C3H.SW anti-(T,G)-A--L antibodies, but did cross-react with four other monoclonal antibodies. It appears that the idiotypic determinants expressed on polyclonal (T,G)-A--L specific antibodies are heterogeneous, and consist of at least two serologically different idiotypes detected by clones no. 103 and 160.     

Identification of Different Antigenic Determinants Within the Synthetic Multichain Co-Polymer Poly(Tyr, Glu)-PolyAla–PolyLys As Recognized by the Chicken

The multichain co-polymer poly(Tyr,Glu)-polyAla--polyLys, designated (T,G)-A--L, is normally thought to present only the poly(Tyr,Glu) sequences, designated (T,G), as antigenic determinants. Evidence is presented indicating that at least two different determinants in the (T,G)-A--L antigen are recognized by chickens. Studies in a partly inbred high-responder chicken strain reveal two major determinant systems: the (T,G) and the A--L in the (T,G)-A--L antigen. For one serum the antigenic determinant of the A--L backbone is shown to be a poly-D-alanine.     

Glomerulonephritis: the use of grafted hybridomas to investigate the role of epitope density, antibody affinity and antibody isotype in active serum sickness.

Anti-idiotypic sera (aIds) were raised in C57BL/6 mice against monoclonal antibodies (McAbs) which bind poly(Tyr,Glu)-poly(DLAla)--poly(Lys)--abbreviated to (T,G)-A--L--and (Tyr-Tyr-Glu-Glu)-poly(DLAla)--poly(Lys)--abbreviated to (T-T-G-G)-A--L--(nos. 103 and 160) and McAbs which react only with (T-T-G-G)-A--L (nos. 100 and 114). Anti-Id antibodies against 103 McAb reacted with (T,G)-A--L specific antibodies and specifically inhibited their binding to iodinated antigen. Similarly, conventional antibodies against the major idiotypes of (T,G)-A--L-specific antibodies inhibited the binding of 103 McAb to antigen. It is therefore suggested that 103 McAb shares major idiotypes with (T,G)-A--L-specific polyclonal antibodies of C3H.SW origin. Anti-Ids against 114 McAb also inhibited the binding of (T,G)-A--L-specific antibodies to antigen, but the binding of 114 McAb could not be inhibited by conventional aIds. Therefore, idiotypes of 103 and 114 McAbs define idiotypic determinants expressed on two different subpopulations of (T,G)-A--L-specific antibodies: those that carry major idiotypes and those which express idiotypic determinants other than the major one (minor Ids). Anti-idiotypic sera against McAbs nos. 100 and 160 reacted with the homologous idiotypes and not with the major idiotypes of (T,G)-A--L-specific antibodies. In addition to the aforementioned specificities we could define cross-reactive idiotypes (private) shared by McAbs nos. 100, 103 and 114 McAbs nos. 160 and 114. The analysis of idiotypes expressed on anti-(T,G)-A--L McAbs enabled the detection of new antigen binding site related idiotypic determinants in addition to the major idiotypes which were immunodominant in the polyclonal anti-(T,G)-A--L antibodies.     

Cloned T helper cells reverting to a resting state develop increasing sensitivity in their antigen-mediated interaction with accessory cells

A cloned murine T cell line, KIII5, specific for the polypeptide poly-L(Tyr,Glu)-poly-D,L-Ala--poly-L-Lys [(T,G)-A--L] was compared at different stages after antigenic stimulation with respect to the conditions required for the reinduction of growth by varying concentrations of antigen presented on different types of accessory cells (AC). We show that the dose of antigen necessary for inducing half maximal proliferation in the presence of splenic AC shifts to considerably lower concentrations when the T cell blasts revert to a resting state (100 micrograms/ml on day 7 to 10 micrograms/ml on day 21-35). During the same time period the expression of interleukin 2 (IL2) receptor and the reactivity to IL2 decline. However, no direct correlation between the increasing sensitivity to antigen and the decreasing reactivity to IL2 appears to exist. With peritoneal AC "early" T cells (day 7) did not respond to (T,G)-A--L at all, but in the course of "aging" responsiveness increased and finally reached the same level as in the presence of splenic AC, although at a higher antigen dose (100 micrograms/ml on day 35-45). Furthermore, the antigen-induced proliferation of "aging" T cells became more resistant to inhibition both by anti-L3T4 and anti-T cell receptor antibodies. Two alternative interpretations of these data are possible: antigen-activated T cells, while gradually reverting to a resting state, interact more avidly with antigen-presenting cells or the triggering threshold of the T cells is decreasing.     

Role of MHC class II expressing CD4+ T cells in proteolipid protein(91-110)-induced EAE in HLA-DR3 transgenic mice

MHC class II molecules play a central role in the control of adaptive immune responses through selection of the CD4(+) T cell repertoire in the thymus and antigen presentation in the periphery. Inherited susceptibility to autoimmune disorders such as multiple sclerosis, rheumatoid arthritis and IDDM are associated with particular MHC class II alleles. Advent of HLA transgenic mice has helped us in deciphering the role of particular HLA DR and DQ class II molecules in human autoimmune diseases. In mice, the expression of class II is restricted to professional antigen-presenting cells (APC). However, in humans, class II is also expressed on T cells, unlike murine T cells. We have developed new humanized HLA class II transgenic mice expressing class II molecules not only on APC but also on a subset of CD4(+) T cells. The expression of class II on CD4(+) T cells is inducible, and class II(+) CD4(+) T cells can present antigen in the absence of APC. Further, using EAE, a well-established animal model of MS, we tested the functional significance of these class II(+) CD4(+) T cells. DR3.AEo transgenic mice were susceptible to proteolipid protein(91-110)-induced EAE and showed CNS pathology accompanied by widespread inflammation and demyelination seen in human MS patients, suggesting a role for class II(+) CD4(+) T cells in the pathogenesis.     

Is there any relationship between human leucocyte antigen class II and chronic urticaria? (chronic urticaria and HLA class II)

The Human Leukocyte Antigen (HLA) typing of large groups of patients with various autoimmune diseases has demonstrated that some HLA alleles occur at higher frequencies in specific diseases than in the general population. Chronic urticaria has been shown to have an autoimmune basis by a previous study which found an association between chronic urticaria and specific HLA groups. We investigated the HLA subtypes of Turkish chronic urticaria patients. For this purpose 42 Turkish patients with chronic urticaria and 115 healthy controls were typed for HLA-DR and DQ by PCR-SSP (Polymerase Chain Reaction Sequence Specific Primers) low resolution DNA technique. We found an increased frequency of DR4 (42.9%, p=0.01) in chronic urticaria patients in comparison with that in healthy controls. This study supports the hypothesis that HLA alleles may be involved in the pathogenesis of chronic urticaria and that they appear to be directly involved in the initiation of the immune response.     

How HIV-1 lentivirus causes immune deficiency disease.

Human immunodeficiency virus (HIV-1) associated immune deficiency has the characteristics of chronic graft versus host disease (GVHD) caused by human leukocyte antigen (HLA) class 2 incompatibility. The envelope glycoprotein fragment TKAKRRVVEREKR mimics HLA class 1 C molecules serologically, and also mimics an immune regulatory T cell epitope, in the region of amino acids 67 to 71, within the HLA DR beta chain. This beta chain alloepitopic region (between amino acids 67 to 80) furnishes peptides predicted to bind optimally to HLA class 1 B alleles. The hypothesis predicts that viral parameters, such as viral load, and clinical parameters, such as rate of progress to acquired immune deficiency syndrome (AIDS) and severity of the associated immune deficient state, are linked to the HLA B and HLA DR beta chain haplotype in infected patients. Immune suppression is caused by HLA class 1 B restricted CD8+ T cells which normally regulate HLA class 2 DR restricted antigen specific responses. The hypothesis further predicts the severity of immune deficiency to be linked to those HLA DR beta chain allotypes which express the amino-acid glutamine (Q) in position 70.