Thierry Bonnefoix

Fitting limiting dilution experiments with generalized linear models results in a test of the single-hit Poisson assumption

Limiting dilution analysis is a common technique that is used in immunology to estimate accurately the frequency of cells possessing a wide variety of functional activities such as growth, cytotoxicity and production of lymphokines. The reliability of the estimated frequency is usually checked by a standard chi-square (x2) test validating the goodness-of-fit to the single-hit Poisson model (SHPM). We present evidence that modelling limiting dilution data according to a generalized linear model offers an alternative to the standard x2 test for detecting departures from the SHPM, with a considerable increase in power compared to the x2 test.

Resistance to CD95-mediated apoptosis through constitutive c-FLIP expression in a non-Hodgkin's lymphoma B cell line.

CD95 (Fas/Apo-1) is a transmembrane molecule that induces apoptosis and plays a central role in the regulation of the immune response. The present study describes two new B lymphoid cell lines, B593 and BR97, derived from non-Hodgkins lymphoma, which differ in susceptibility to CD95-mediated apoptosis. While B593 cells are sensitive to CD95-mediated apoptosis, BR97 cells are completely resistant. Activation of caspase-8 and caspase-3 proteases plays an important role in the CD95 signalling pathway. CD95 stimulation induced caspase-8 and caspase-3 activation in B593, but not in BR97 cells. However, activation of both caspase-8 and caspase-3 was achieved in BR97 cells treated with staurosporine. Furthermore, protein synthesis inhibition by cycloheximide restored sensitivity to CD95-mediated apoptosis and allowed activation of both caspase-8 and caspase-3 in BR97 cells. These results indicate that, in BR97 cells, both caspases are functional and suggest that CD95-apoptosis resistance may result from the presence of inhibitory factor(s). Constitutive high level expression of the apoptotic inhibitor c-FLIP was observed in the CD95-resistant BR97 cell line compared to B593. Moreover, downregulation of c-FLIP expression level by protein synthesis inhibition strictly correlated with restored sensitivity to CD95-mediated apoptosis in BR97 cells. Furthermore, we demonstrate that c-FLIP is recruited to the CD95 DISC in BR97 cells together with caspase-8 and FADD. The data presented in this study strongly suggests that, in a B-NHL-derived cell line, resistance to CD95-mediated apoptosis results from endogenous high level expression of apoptotic inhibitor c-FLIP.

Graphical Representation of a Generalized Linear Model-Based Statistical Test Estimating the Fit of the Single-Hit Poisson Model to Limiting Dilution Assays

Standardized statistical and graphical methods for analysis of limiting dilution assays are highly desirable to enable investigators to compare and interpret results and conclusions with greater accuracy and precision. According to these requirements, we present in this work a powerful statistical slope test that estimates the fit of the single-hit Poisson model to limiting dilution experiments. This method is readily amenable to a graphical representation. This slope test is obtained by modeling limiting dilution data according to a linear log-log regression model, which is a generalized linear model specially designed for modeling binary data. The result of the statistical slope test can then be graphed to visualize whether the data are compatible or not with the single-hit Poisson model. We demonstrate this statistical test and its graphical representation by using two examples: a real limiting dilution experiment evaluating the growth frequency of IL-2-responsive tumor-infiltrating T cells in a malignant lymph node involved by a B cell non-Hodgkin’s lymphoma, and a simulation of a limiting dilution assay corresponding to a theoretical non-single-hit Poisson model, suppressor two-target Poisson model.

1q12 chromosome translocations form aberrant heterochromatic foci associated with changes in nuclear architecture and gene expression in B cell lymphoma.

Epigenetic perturbations are increasingly described in cancer cells where they are thought to contribute to deregulated gene expression and genome instability. Here, we report the first evidence that a distinct category of chromosomal translocations observed in human tumours—those targeting 1q12 satellite DNA—can directly mediate such perturbations by promoting the formation of aberrant heterochromatic foci (aHCF). By detailed investigations of a 1q12 translocation to chromosome 2p, in a case of human B cell lymphoma, aberrant aHCF were shown to be localized to the nuclear periphery and to arise as a consequence of long range ‘pairing’ between the translocated 1q12 and chromosome 2 centromeric regions. Remarkably, adjacent 2p sequences showed increased levels of repressive histone modifications, including H4K20me3 and H3K9me3, and were bound by HP1. aHCF were associated to aberrant spatial localization and deregulated expression of a novel 2p gene (GMCL1) that was found to have prognostic impact in diffuse large B cell lymphoma. Thus constitutive heterochromatin rearrangements can contribute to tumourigenesis by perturbing gene expression via long range epigenetic mechanisms.

Autotumour reactive T-cell clones among tumour-infiltrating T lymphocytes in B-cell non-Hodgkin's lymphomas.

Summary. Seventy‐three T‐cell clones (TCC) were established from tumour‐infiltrating lymphocytes‐T (TIL‐T) derived from lymph nodes involved by B‐cell non‐Hodgkins lymphomas (B‐NHL) in nine patients with different histological subtypes and clinical stages. 40 TCC (55%) expressed the CD25 Ag and were also able to proliferate in the presence of irradiated autologous B‐NHL cells. Among them, 23 autotumour (AuTu) proliferative TCC were found not to proliferate to autologous EBV‐transformed B‐cell lines, indicating that the proliferative reactivity of these TCC was preferentially directed at autologous B‐NHL cells. Tested against autologous B‐NHL cells, only three AuTu prolifera‐ tive TCC (CD8 +) showed a significant level of cytotoxicity (specific lysis > 15%). In blocking experiments, the AuTu proliferative reactivity of three TCC from one patient was strongly inhibited by anti‐DR and anti‐DQ mAbs, whereas that of three TCC from another patient was not affected by either anti‐MHC class I or class II (DR., DP, DQ) mAbs. These findings suggest that the recognition of autologous B‐NHL cells by AuTu proliferative TCC may occur through MHC‐restricted as well as MHC‐unrestricted mechanisms.

Differences in nuclear positioning of 1q12 pericentric heterochromatin in normal and tumor B lymphocytes with 1q rearrangements

The frequent rearrangement of chromosome band 1q12 constitutive heterochromatin in hematologic malignancies suggests that this rearrangement plays an important pathogenetic role in these diseases. The oncogenic mechanisms linked to 1q12 heterochromatin are unknown. Constitutive heterochromatin can epigenetically regulate gene function through the formation of transcriptional‐silencing compartments. Thus, as a first step toward understanding whether 1q12 rearrangements might compromise such activity in tumor cells, we investigated the 3‐D organization of the 1q12 heterochromatin domain (1q12HcD) in normal and tumor B lymphocytes. Strikingly, in normal B cells, we showed that the 1q12HcD dynamically organizes to the nuclear periphery in response to B‐cell receptor engagement. Specifically, we observed an almost twofold increase in 1q12Hc domains at the extreme nuclear periphery in activated versus resting B lymphocytes. Remarkably, 1q12Hc organization was noticeably altered in tumor cells that showed structural alterations of 1q12; the 1q12Hc domains were significantly displaced from the extreme nuclear periphery compared to normal activated B lymphocytes (P > 0.0001), although overall peripheral localization was maintained. In a case in which there was a translocation of IGL enhancer to 1q, the altered nuclear positioning of the 1q12HcD was even more pronounced (5% of the 1q12Hc domains at the nuclear periphery compared to 20% in other lymphoma lines), and we were able to mimic this effect in two additional B‐cell tumor lines by treatment with trichostatin A, a histone deacetylase (HDAC) inhibitor. Taken together, these results point to the 1q12HcD having a specific, nonrandom, and regulated peripheral organization in B lymphocytes. This organization is significantly disrupted in lymphoma cells harboring 1q rearrangements.

Role of autologous CD4+ T cell clones in human B non-Hodgkin's lymphoma: aborted activation and G1 blockade induced by cell-cell contact.

This article describes the study of the functional relationship between auto‐tumor‐reactive CD4+ T cell clones (TCC) and autologous malignant B cells. Four auto‐tumor‐reactive CD4+ TCC were derived from tumor‐infiltrating T lymphocytes (TIL‐T) from a freshly isolated human follicular lymphoma by the following technique: total CD4+ TIL‐T were negatively purified by an immunomagnetic procedure, then CD4+ TCC were obtained by limiting dilution in the presence of IL‐2 and autologous non‐irradiated follicular lymphoma cells as feeders. After expansion, these CD4+ TCC were co‐cultured with non‐irradiated autologous malignant B cells. All four TCC were activated by B lymphoma cells and proliferated, as assessed by CD25 expression and cell cycle analysis. Activation and proliferation of B lymphoma cells were studied in response to activated CD4+ T cells. Although all four TCC were able to induce B lymphoma cell activation (Ki‐67 antigen induction and CD40 up‐regulation), cells were subsequently blocked in G1 phase. Activation of B‐NHL cells was mediated by TCR‐HLA class II interaction, as shown by a blocking experiment using an anti‐CD4 monoclonal antibody (mAb). Since anti‐CD40 mAb with or without IL‐4 did not induce proliferation of B lymphoma cells in contrast to normal B cells, we suggest that the blockade in G1 phase is due to the presence of abnormalities in B lymphoma cells. This is the first evidence that autologous reactive CD4+ TCC can engage follicular lymphoma B cells to enter the cell cycle and induce an aborted activation stage.

Accurate hematopoietic stem cell frequency estimates by fitting multicell Poisson models substituting to the single-hit Poisson model in limiting dilution transplantation assays

Limiting dilution transplantation assay (LDTA) is considered as the gold standard method to assess hematopoietic stem cell (HSC) content. Traditionally, HSC frequency estimates are based on the single-hit Poisson model (SHPM), which posits that one donor HSC is sufficient to generate a progeny of detectable differentiated cells above a threshold value in hosts. However, there is no clear support for this statement, and it is receivable that more than one donor HSC may be necessary to provide detectable reconstitution in hosts above the threshold level for detection, usually 0.5% to 1% of donor-derived cells. To address this hypothesis, we evaluated the ability of a class of multiCell Poisson models (C(≥1)PMs) to fit to LDTAs. In 7 of the 8 reanalyzed LDTAs, C(≥1)PMs plausibly compete with the traditional SHPM. Model averaging across the set of plausible models gives 1.32- to 5.88-fold increases in HSC frequencies compared with the SHPM.

Haploinsufficiency for NR3C1 , the gene encoding the glucocorticoid receptor, in blastic plasmacytoid dendritic cell neoplasms

Blastic plasmacytoid dendritic cell neoplasm (BPDCN) is a rare and highly aggressive leukemia for which knowledge on disease mechanisms and effective therapies are currently lacking. Only a handful of recurring genetic mutations have been identified and none is specific to BPDCN. In this study, through molecular cloning in an index case that presented a balanced t(3;5)(q21;q31) and molecular cytogenetic analyses in a further 46 cases, we identify monoallelic deletion of NR3C1 (5q31), encoding the glucocorticoid receptor (GCR), in 13 of 47 (28%) BPDCN patients. Targeted deep sequencing in 36 BPDCN cases, including 10 with NR3C1 deletion, did not reveal NR3C1 point mutations or indels. Haploinsufficiency for NR3C1 defined a subset of BPDCN with lowered GCR expression and extremely poor overall survival (P = .0006). Consistent with a role for GCR in tumor suppression, functional analyses coupled with gene expression profiling identified corticoresistance and loss-of-EZH2 function as major downstream consequences of NR3C1 deletion in BPDCN. Subsequently, more detailed analyses of the t(3;5)(q21;q31) revealed fusion of NR3C1 to a long noncoding RNA (lncRNA) gene (lincRNA-3q) that encodes a novel, nuclear, noncoding RNA involved in the regulation of leukemia stem cell programs and G1/S transition, via E2F. Overexpression of lincRNA-3q was a consistent feature of malignant cells and could be abrogated by bromodomain and extraterminal domain (BET) protein inhibition. Taken together, this work points to NR3C1 as a haploinsufficient tumor suppressor in a subset of BPDCN and identifies BET inhibition, acting at least partially via lncRNA blockade, as a novel treatment option in BPDCN.

The standard χ2 test used in limiting dilution assays is insufficient for estimating the goodness-of-fit to the single-hit Poisson model

Limiting dilution analysis is a common technique that is used in immunology to estimate accurately the frequency of cells possessing a wide variety of functional activities, such as growth, cytotoxicity and production of lymphokines. In the literature, most experiments are fit well by the single-hit Poisson model (SHPM), which assumes that only one cell of one defined cell subset is necessary for a positive response. This is somewhat surprising since other models such as multi-hit or multi-target models that involve the interaction of one or more cells from one or more cell subpopulations for generating or inhibiting a positive response are conceivable. Since the validity of the SHPM is usually investigated by performing a standard chi 2 test, based on the number of observed and expected positive and negative responses, we questioned here the efficiency of this test in comparison with other validity tests for the SHPM, the log likelihood test derived by Cox, and the modified Weibull plot tests, the principles of which are entirely different from that of the standard chi 2 test. We used the following theoretical approach. First, we generated artificial data corresponding to multi-hit and multi-target models. Second, considering that these data were derived from real experiments, we calculated the frequency of the desired cell subset according to the SHPM using the maximum likelihood method. Then, the goodness-of-fit of these data with the SHPM was evaluated. The log likelihood test and the modified Weibull plot tests rejected the SHPM hypothesis, while the standard chi 2 test did not. Thus, the standard chi 2 test is unable to discriminate sensitively between the SHPM and more complicated (non-single-hit) Poisson models. We concluded that the results of limiting dilution studies published thus far must be evaluated with caution. The statistical tests presented here should be routinely applied for each limiting dilution experiment.