Selection of hprt mutant T cells as surrogates for dividing cells reveals a restricted T cell receptor BV repertoire in insulin-dependent diabetes mellitus

Molecular characterization of hypoxanthine guanine phosphoribosyltransferase mutant T cells in human blood: The concept of surrogate selection for immunologically relevant cells

Somatic cell gene mutations arise in vivo due to replication errors during DNA synthesis occurring spontaneously during normal DNA synthesis or as a result of replication on a DNA template damaged by endogenous or exogenous mutagens. In principle, changes in the frequencies of mutant cells in vivo in humans reflect changes in exposures to exogenous or endogenous DNA damaging insults, other factors being equal. It is becoming increasingly evident however, that somatic mutations in humans have a far greater range of interpretations. For example, mutations in lymphocytes provide invaluable probes for in vivo cellular and molecular processes, providing identification of clonal amplifications of these cells in autoimmune and infectious diseases, transplantation recipients, paroxysmal nocturnal hemoglobinuria (PNH), and cancer. The assay for mutations of the X-chromosomal hypoxanthine guanine phosphoribosyltransferase (HPRT) gene has gained popular acceptance for this purpose since viable mutant cells can be recovered for molecular and other analyses. Although the major application of the HPRT T cell assay remains human population monitoring, the enrichment of activated T cells in the mutant fraction in individuals with ongoing immunological processes has demonstrated the utility of surrogate selection, a method that uses somatic mutation as a surrogate marker for the in vivo T cell proliferation that underlies immunological processes to investigate clinical disorders with immunological features. Studies encompassing a wide range of clinical conditions are reviewed. Despite the historical importance of the HPRT mutation system in validating surrogate selection, there are now additional mutational and other methods for identifying immunologically active T cells. These methods are reviewed and provide insights for strategies to extend surrogate selection in future studies.

In vivo 6-thioguanine-resistant T cells from melanoma patients have public TCR and share TCR beta amino acid sequences with melanoma-reactive T cells

In vivo hypoxanthine-guanine phosphoribosyltransferase (HPRT)-deficient T cells (MT) from melanoma patients are enriched for T cells with in vivo clonal amplifications that traffic between blood and tumor tissues. Melanoma is thus a model cancer to test the hypothesis that in vivo MT from cancer patients can be used as immunological probes for immunogenic tumor antigens. MT were obtained by 6-thioguanine (TG) selection of lymphocytes from peripheral blood and tumor tissues, and wild-type T cells (WT) were obtained analogously without TG selection. cDNA sequences of the T cell receptor beta chains (TRB) were used as unambiguous biomarkers of in vivo clonality and as indicators of T cell specificity. Public TRB were identified in MT from the blood and tumor of different melanoma patients. Such public TRB were not found in normal control MT or WT. As an indicator of T cell specificity for melanoma, the >2600 MT and WT TRB, including the public TRB from melanoma patients, were compared to a literature-derived empirical database of >1270 TRB from melanoma-reactive T cells. Various degrees of similarity, ranging from 100% conservation to 3-amino acid motifs (3-mer), were found between both melanoma patient MT and WT TRBs and the empirical database. The frequency of 3-mer and 4-mer TRB matching to the empirical database was significantly higher in MT compared with WT in the tumor (p=0.0285 and p=0.006, respectively). In summary, in vivo MT from melanoma patients contain public TRB as well as T cells with specificity for characterized melanoma antigens. We conclude that in vivo MT merit study as novel probes for uncharacterized immunogenic antigens in melanoma and other malignancies.

Clonal expansions of 6-thioguanine resistant T lymphocytes in the blood and tumor of melanoma patients

The identification of specific lymphocyte populations that mediate tumor immune responses is required for elucidating the mechanisms underlying these responses and facilitating therapeutic interventions in humans with cancer. To this end, mutant hypoxanthine-guanine phosphoribosyltransferase (HPRT) deficient (HPRT-) T-cells were used as probes to detect T-cell clonal amplifications and trafficking in vivo in patients with advanced melanoma. Mutant T-cells from peripheral blood were obtained as clonal isolates or in mass cultures in the presence of 6-thioguanine (TG) selection and from tumor-bearing lymph nodes (LNs) or metastatic melanoma tissues by TG-selected mass cultures. Nonmutant (wild-type) cells were obtained from all sites by analogous means, but without TG selection. cDNA sequences of the T-cell receptor (TCR) beta chains (TCR-beta), determined directly (clonal isolates) or following insertion into plasmids (mass cultures), were used as unambiguous biomarkers of in vivo clonality of mature T-cell clones. Clonal amplifications, identified as repetitive TCR-beta V-region, complementarity determining region 3 (CDR3), and J-region gene sequences, were demonstrated at all sites studied, that is, peripheral blood, LNs, and metastatic tumors. Amplifications were significantly enriched among the mutant compared with the wild-type T-cell fractions. Importantly, T-cell trafficking was manifested by identical TCR-beta cDNA sequences, including the hypervariable CDR3 motifs, being found in both blood and tissues in individual patients. The findings described herein indicate that the mutant T-cell fractions from melanoma patients are enriched for proliferating T-cells that infiltrate the tumor, making them candidates for investigations of potentially protective immunological responses.

Effect of chronic azathioprine treatment on germ-line transmission of Hprt mutation in mice

Azathioprine (Aza), a prodrug of 6-mercaptopurine, is used in human medicine to prevent transplant rejection and for the treatment of autoimmune diseases. Extremely high HPRT lymphocyte mutant frequencies (MFs) are found in humans and mice chronically treated with Aza, and these elevated MFs appear to be caused by selection and amplification of pre-existing HPRT mutant lymphocytes. In the present study, we investigated if in vivo selection by Aza also promotes the germ-line transmission of Hprt mutants. Fifty-five male C57BL/6 mice were treated with 10 mg/kg Aza three times/week for 24 weeks; 10 control mice were treated with the vehicle. Each of these males then was bred to unexposed females for a total of 8 weeks. Analysis of the Aza-treated males after the breeding period indicated that 12 had highly elevated Hprt lymphocyte MFs (1 x 10(-4)-2.5 x 10(-1) vs. normal MFs of <1 x 10(-5)), indicating that the Aza treatment successfully selected somatic cell mutants. The female offspring from the breeding were sacrificed at 28 days of age and Hprt MFs were measured in spleen lymphocytes. Most of the 364 female offspring (332 from Aza-treated fathers) had Hprt MFs of 0-6 x 10(-6), but seven of the offspring had moderately elevated MFs of 16 x 10(-6)-55 x 10(-6). Since one of these mice was fathered by a control male, these relatively high MFs appear to be part of the normal variation in lymphocyte Hprt MF. The present results provide no evidence that long-term Aza treatment promotes high levels of germ-line Hprt mutation transmission in mice.

HPRT mutations, TCR gene rearrangements, and HTLV-1 integration sites define in vivo T-cell clonal lineages

HPRT mutations in vivo in human T-lymphocytes are useful probes for mechanistic investigations. Molecular analyses of isolated mutants reveal their underlying mutational changes as well as the T-cell receptor (TCR) gene rearrangements present in the cells in question. The latter provide temporal reference points for other perturbations in the in vivo clones as well as evidence of clonal relationships among mutant isolates. Immunological studies and investigations of genomic instability have benefited from such analyses. A method is presented describing a T-cell lineage analysis in a patient with HTLV-1 infection. Lineage reconstruction of an in vivo proliferating HPRT mutant clone allows timing of the integration event to a postthymic differentiated cell prior to the occurrence of HPRT mutations.

Genotoxicity of therapeutic intervention in children with acute lymphocytic leukemia

The survival rates of children treated for cancer have dramatically increased after the development of standardized multiple-modality treatment protocols. As a result, there is a rapidly growing population of pediatric cancer survivors in which the long-term genotoxic effects of chemotherapeutic intervention is unknown. To study the genotoxic effects of antineoplastic treatment in children, we performed a comparative analysis of the changes in the frequency of somatic mutations (Mfs) at the hypoxanthine-guanine phosphoribosyltransferase (HPRT)-reporter gene in children treated for acute lymphocytic leukemia (ALL). We measured HPRT Mfs from 130 peripheral blood samples from 45 children with ALL (13, low risk; 22, standard risk; and 10, high risk) from the time of diagnosis, as well as during and after the completion of therapy. We observed a significant increase in mean HPRT Mfs during each phase of therapy (diagnosis, 1.4 x 10(-6); consolidation, 52.1 x 10(-6); maintenance, 93.2 x 10(-6); and off-therapy, 271.7 x 10(-6)) that were independent of the risk group treatment protocol used. This 200-fold increase in mean somatic Mf remained elevated years after the completion of therapy. We did not observe a significant difference in the genotoxicity of each risk group treatment modality despite differences in the compositional and clinical toxicity associated with these treatment protocols. These findings suggest that combination chemotherapy used to treat children with ALL is quite genotoxic, resulting in an increased somatic mutational load that may result in an elevated risk for the development of multi-factorial diseases, in particular second malignancies.

Frequent HPRT mutations in paroxysmal nocturnal haemoglobinuria reflect T cell clonal expansion, not genomic instability

Paroxysmal nocturnal haemoglobinuria (PNH) results from acquired mutations in the PIG-A gene of an haematopoietic stem cell, leading to defective biosynthesis of glycosylphosphatidylinositol (GPI) anchors and deficient expression of GPI-anchored proteins on the surface of the cell's progeny. Some laboratory and clinical findings have suggested genomic instability to be intrinsic in PNH; this possibility has been supported by mutation analysis of hypoxanthine-guanine phosphoribosyltransferase (HPRT) gene abnormalities. However, the HPRT assay examines lymphocytes in peripheral blood (PB), and T cells may be related to the pathophysiology of PNH. We analysed the molecular and functional features of HPRT mutants in PB mononuclear cells from eleven PNH patients. CD8 T cells predominated in these samples; approximately half of the CD8 cells lacked GPI-anchored protein expression, while only a small proportion of CD4 cells appeared to derive from the PNH clone. The HPRT mutant frequency (Mf) in T lymphocytes from PNH patients was significantly higher than in healthy controls. The majority of the mutant T lymphocyte clones were of CD4 phenotype, and they had phenotypically normal GPI-anchored protein expression. In PNH patients, the majority of HPRT mutant clones were contained within the Vbeta2 T cell receptor (TCR) subfamily, which was oligoclonal by complementarity-determining region three (CDR3) size analysis. Our results are more consistent with detection of uniform populations of expanded T cell clones, which presumably acquired HPRT mutations during antigen-driven cell proliferation, and not due to an increased Mf in PNH. HPRT mutant analysis does not support underlying genomic instability in PNH.

Evidence of a diverse T cell receptor repertoire for acetylcholine receptor, the autoantigen of myasthenia gravis

We utilized two methods to look for T cell clonal expansions in myasthenia gravis (MG). We analyzed TCRBV CDR3 length polymorphism (spectratyping) to look for evidence of clonal expansion of CD4 or CD8 T cells directly from peripheral blood of MG patients. No statistically significant differences were found between the diversity of TCR repertoires in MG patients compared to normal control individuals when analyzed as groups. Rare oligoclonal expansions were detected in some individual MG patients but the significance of these findings is unclear. Next, we analyzed a panel of T cell hybridomas from acetylcholine receptor (AChR) immunized, MG-susceptible HLA-DR3 transgenic mice. The epitope specificity, TCRBV gene usage and CDR3 sequences of these hybridomas were highly diverse. We conclude there is only limited evidence for restricted TCR repertoire usage in human MG and suggest this may be due to the inability of HLA-DR molecules to select for restricted TCR recognition of AChR epitopes.

Assessment of the frequency of mutant (hprt-) T lymphocytes from peripheral blood of patients with Hashimoto's thyroiditis

A salient feature of Hashimoto's thyroiditis (HT) is the T-cell-mediated destruction of the thyroid gland leading to hypothyroidism. In HT, as in other autoimmune diseases, a central premise has been that autoreactive T cells must be dividing in response to autoantigens, accumulating random spontaneous mutations during the activation process. Here, we have examined this hypothesis by using as monitor of somatic cell mutation the hprt gene, encoding the salvage pathway enzyme hypoxanthine-guanine phosphoribosyl transferase. Eleven newly diagnosed patients with HT and 10 patients with chronic disease were selected for the study, whereas 10 healthy individuals were used as controls. Peripheral T cells were cultured under limiting dilution conditions in the presence of 6-thioguanine and the frequency (MF) of surviving mutant hprt(-) T cells was calculated by Poisson statistics. It was observed that the mean MF value of either patient group (6.6 +/- 5.8 per 10(6) cells for the newly diagnosed, and 8.8 +/- 4.0 per 10(6) cells for the patients with chronic disease) was not significantly different (p > 0.05) from that of the control group (6.8 +/- 6.4 per 10(6) cells). These data do not support the concept that patients with HT have an increased number of actively dividing T cells in the circulation compared to healthy controls. Autoreactive T cells may be activated mainly in situ or home readily to the thyroid in the early stages of the disease and reach a nonexpansion stage as the chronic disease is stabilized.

Accumulation of somatic mutations in proliferating T cell clones from children treated for leukemia

There is continued controversy as to the sequential steps and mechanism(s) responsible for the in vivo acquisition of multiple mutations during neoplastic transformation. We investigated the in vivo clonality and mutational spectra of hypoxanthine-guanine phosphoribosyltransferase (HPRT) mutations in T cells from children with acute lymphocytic leukemia (ALL) to gain insight into the mutagenic mechanisms associated with leukemogenesis. We observed several instances of multiple, independent HPRT mutations accumulating in vivo in T cell receptor (TCR) gene defined clones that had undergone extensive pre- and/or post-thymic expansion following chemotherapy. In addition, we also detected the accumulation of multiple unique single mutations within distinct expanding post-thymic T cell clones. This pattern of clonally restricted hypermutability is compatible with extensive cell proliferation and selection alone without postulating genomic instability. These observations provide a paradigm for a continuum of cellular events that eventually results in the clonal accumulation of mutations in selected populations of cells in vivo and may provide insight into the primary genetic events associated with leukemogenesis, as well as the development of second malignancies and drug resistance following chemotherapy.

Molecular and flow cytometric analysis of the Vbeta repertoire for clonality assessment in mature TCRalphabeta T-cell proliferations

Clonality assessment through Southern blot (SB) analysis of TCRB genes or polymerase chain reaction (PCR) analysis of TCRG genes is important for diagnosing suspect mature T-cell proliferations. Clonality assessment through reverse transcription (RT)-PCR analysis of Vbeta-Cbeta transcripts and flow cytometry with a Vbeta antibody panel covering more than 65% of Vbeta domains was validated using 28 SB-defined clonal T-cell receptor (TCR)alphabeta(+) T-ALL samples and T-cell lines. Next, the diagnostic applicability of the V(beta) RT-PCR and flow cytometric clonality assays was studied in 47 mature T-cell proliferations. Clonal Vbeta-Cbeta RT-PCR products were detected in all 47 samples, whereas single Vbeta domain usage was found in 31 (66%) of 47 patients. The suspect leukemic cell populations in the other 16 patients showed a complete lack of Vbeta monoclonal antibody reactivity that was confirmed by molecular data showing the usage of Vbeta gene segments not covered by the applied Vbeta monoclonal antibodies. Nevertheless, this could be considered indirect evidence for the "clonal" character of these cells. Remarkably, RT-PCR revealed an oligoclonal pattern in addition to dominant Vbeta-Cbeta products and single Vbeta domain expression in many T-LGL proliferations, providing further evidence for the hypothesis raised earlier that T-LGL derive from polyclonal and oligoclonal proliferations of antigen-activated cytotoxic T cells. It is concluded that molecular Vbeta analysis serves to assess clonality in suspect T-cell proliferations. However, the faster and cheaper Vbeta antibody studies can be used as a powerful screening method for the detection of single Vbeta domain expression, followed by molecular studies in patients with more than 20% single Vbeta domain expression or large suspect T-cell populations (more than 50%-60%) without Vbeta reactivity.

Flow cytometric analysis of the Vbeta repertoire in healthy controls

BACKGROUND

Analysis of the T-cell receptor (TCR)-Vbeta repertoire has been used for studying selective T-cell responses in autoimmune disease, alloreactivity in transplantation, and protective immunity against microbial and tumor antigens. For the interpretation of these studies, we need information about the Vbeta repertoire usage in healthy individuals.

METHODS

We analyzed blood T-lymphocyte (sub)populations of 36 healthy controls (age range: from neonates to 86 years) with a carefully selected most complete panel of 22 Vbeta monoclonal antibodies, which together recognized 70-75% of all blood TCRalphabeta(+) T lymphocytes. Subsequently, we developed a six-tube test kit with selected Vbeta antibody combinations for easy and rapid detection of single ("clonal") Vbeta domain usage in large T-cell expansions.

RESULTS

The mean values of the Vbeta repertoire usage were stable during aging in blood TCRalphabeta(+) T lymphocytes as well as in the CD4(+) and CD8(+) T-cell subsets, although the standard deviations increased in the elderly. The increased standard deviations were caused by the occurrence of oligoclonal T-cell expansions in the elderly, mainly consisting of CD8(+) T lymphocytes. The 15 detected T-cell expansions did not reach 40% of total TCRalphabeta(+) T lymphocytes and represented less than 0.4 x 10(9) cells per liter in our study. Vbeta usage of the CD4(+) and CD8(+) subsets was comparable for most tested Vbeta domains, but significant differences (P < 0.01) between the two subsets were found for Vbeta2, Vbeta5.1, Vbeta6.7, Vbeta9.1, and Vbeta22 (higher in CD4(+)), as well as for Vbeta1, Vbeta7.1, Vbeta14, and Vbeta23 (higher in CD8(+)). Finally, single Vbeta domain expression in large T-cell expansions can indeed be detected by the six-tube test kit.

CONCLUSIONS

The results of our study can now be used as reference values in studies on distortions of the Vbeta repertoire in disease states. The six-tube test kit can be used for detection of single Vbeta domain expression in large T-cell expansions (>2.0 x 10(9)/l), which are clinically suspicious of T-cell leukemia.

Azathioprine associated T-cell mutations in insulin-dependent diabetes mellitus

Somatic mutations arise regularly in human T lymphocytes. As these events occur at increased frequencies in several autoimmune disorders, presumably because of increased T-cell proliferation, we investigated if this is also true for insulin-dependent diabetes mellitus (IDDM). Mutations of the hypoxanthine guanine phosphoribosyltransferase (hprt) gene measured by 6-thioguanine (TG) selection were studied in 28 patients (60 determinations) enrolled in a prospective double-blinded placebo-controlled study of azathioprine immunosuppression: 17 patients (34 determinations) were receiving azathioprine and 11 (26 determinations) placebo. Mean hprt T-cell mutant frequencies (MFs) were elevated in both patient groups, but only in the azathioprine group were elevations large and statistically correlated with the duration of the therapy. These results suggest that the organ-specific antigenic stimulus of the T-cell proliferation in IDDM does increase mutant cells in the peripheral blood, but this increase is relatively small. However, azathioprine, which is converted to 6-mercaptopurine in vivo, selects and amplifies the hprt mutants that do arise. Clinical azathioprine resistance may be explained by hprt mutations arising in T cells relevant to the underlying autoimmune process. Monitoring for these mutations should allow more effective use of this immunosuppressive agent.