Homepage bandeau_genéral

  Director : DIGHIERO Guillaume (dighiero@pasteur.fr)




1) Natural autoantibodies and autoimmunity

Murine transgenic model of a human autoimmune hemolytic anemia (cold agglutinin disease).

2) Chronic lymphocytic leukemia (CLL)

a) Microarray study of gene expression in CLL.

b) Mechanisms underlying underexpression of the B cell receptor and impaired functional response in CLL B cells.

c) Anti-idiotypic vaccines in CLL.

3) AIDS immunopathology.

Mechanisms of immunological reconstitution following high anti-viral therapy in AIDS.



I) Natural autoantibodies and autoimmunity

Contributions of the laboratory in the field.

At the Pasteur Institute with S. Avrameas we have been working since 1980 in the model of natural autoantibodies (NAAB). We could demonstrate that in normal human serum, a substantial part of circulating Igs are indeed NAAB. We could also demonstrate by preparing hybridomas derived from normal unimmunized BALB/c mouse spleens, that the precursors of this autoreactive repertoire account for a substantial part of normal B cells. As these autoantibodies express recurrent idiotopes, express V genes frequently in germinal configuration and predominate early in life, they are the expression of the germinal repertoire. Since then, many groups demonstrated that these experiments were true. This repertoire has been found in species phylogenetically distant from mammals like fishs and batracians. Hence, we have evolved from Ehrlich's notion of "horror autotoxicus" to Burnett's notion of forbidden clones, to reach the modern idea that autoimmunity is a physiological event

During recent years, we could obtain evidence coming from three different models which supported the view that B cell-malignancies frequently involve this autoreactive repertoire. More recently, the same high frequency of NAA specificities was found amoung mouse plasmocytomas induced by intra-peritoneal injection of silicone. Taken together these results show that the NAAB repertoire is prone to malignant transformation. It is presently unclear whether the continuous challenge of this autoreactive repertoire by self-antigens could create propitious conditions for malignant transformation to occur.

Another important point that we could show in our earlier studies, was the frequent polyreactivity of these antibodies. Although, this notion was difficult to accept by classical immunologists, it could be extensively substantiated in numerous laboratories and the term polyreactive antibody is usually employed nowadays. However, the structural basis of polyreactivity remains elusive, though the possibility exists that molecular surface adaptability, which is a general principle underlying promiscuity in protein-protein interactions, could play a role.

As concerns the fonction of this repertoire, it could constitute an archaic immune system which may have a role as a first line of defense. Interestingly, this NAA polyreactive repertoire, has been found to be prevalent in fish and in batracians, which have a primitive immune repertoire. We have evoked the possibility that this polyreactive B cell could be induced through an antigen driven process to mutate and to become a high-affinity monoreactive antibody (template hypothesis). This hypothesis needs to be substantiated, however.

Research Project:

a) Creation of a murine transgenic model of human CA disease (Responsible: C. Pourcel).

Scientific background

Autoimmune hemolytic anemia (AIHA) induced by cold agglutinins (CA) is due to autoantibodies usually expressing the µk isotype, that recognize carbohydrate epitopes on human red blood cells (RBC) and cause hemagglutination at temperatures below 37°C and, optimally, in the cold (i.e. 4°C). A substantial proportions of cases develop in patients with B-cell lymphomas, Waldenström's macroglulinemia or chronic lymphocytic leukemia. The fact that CA disease can also evolve in two steps, the first one limited to AIHA, which subsequently can evolve to a lymphoma, makes this disease a model of the relationship between autoimmune and malignant diseases.

Present status.In a previous work, one human cold agglutinin (CAGAS) displaying the rare anti-Gd specificity (sialilated antigen found on glycolipids) was studied at the molecular level (De la Fuente et al, 1994). In contrast to classical anti-Ii CA, which do not bind to mouse red blood cells, CAGAS agglutinates murine erythrocytes (MRBC) with better affinity than it binds human red blood cells, and is able to hemolyse MRBC in the presence of complement. We have introduced CAGAS VH and VL domains into eukaryotic expression vectors and transfected them into the non-secreting mouse myeloma X63 cell line. Clones expressing complete engineered pentameric IgMk CAGAS (eCAGAS) recapitulating the characteristics of serum CA (sCAGAS) were obtained. Intraperitoneal injection of X63 transfected line or of eCAGAS into BALB/c mice induced a typical hemolytic anemia, thus proving its pathogenic role in mice (Dumas et al, 1997). Interestingly, one mouse developped an important abnormality of the pinna, characterized by bilateral external ear loss, which could be the consequence of the classical cryopatic syndrome observed in CA disease.

Transgenic mice were built by introducing CAGAS VH and VL domains into eukaryotic expression vectors and transgenic mice were obtained. These mice produced significant serum amounts of CAGAS, though the agglutination titer was insufficient to induce an AIHA. To circumvent the pathogenic role of CAGAS, mice developed a tolerance strategy based on deletion at the central level (bone marrow) of a majority of B cells expressing the pathogenic transgene and expression by most B cells migrating to the peripheral compartments of CAGAS heavy chain associated to mouse light chains (editing) (Havouis et al, 2000).

Since mycoplasma has been reported to induce the disease in humans, through polyclonal stimulation of the normal B cell population producing CA, we have studied the consequences of mycoplasma infection in the CA-expressing transgenic mice. Our results indicate that infection of transgenic mice with Mycoplasma Pulmonis induces a spectacular increase of blood B cells expressing the transgene and of the CAGAS level (figure 2), with CA titers reaching in some mice 1/512.

Since polyclonal stimulation by KLH only induced a small increase of CA titers, these results indicate that Mycoplasma pulmonis is able to circumvent tolerance mechanisms through positive selection of B cells expressing the CA transgene.

Future directions.

We are presently considering to reproduce the disease through immunization of transgenic animals with membrane extracts of Mycoplasma containing the cross-reactive epitope of the red blood cell Gd. In addition, we have created transgenic bcl-2 mice expressing the GAS transgene. We are planning to immunize these mice with membrane extracts of Mycoplasma. Bcl-2 transgene, known for its anti-apoptotic effect, induces a polyclonal activation in mouse which frequently results in autoimmune diseases and can also induce the emergence of lymphomas. It is plausible that some of these mice will produce lymphoma displaying a CA specificity. In the case that we fail to induce lymphomas with CAGAS specificity, the possibility exists to create transgenic mice for the c-myc, which consistently induces in short periods of time malignant transformation.

Since CA disease remains an incurable disease, a major objective of this model once we will succeed to reproduce the disease, is to develop an anti-idiotypic vaccine by pulsing dendritic cells derived from mouse bone-marrow with the idiotype. In this field, we will take advantage of the work we are carrying out in CLL (see below). This work has received grant's support from ARC (Association pour la Recherche sur le Cancer) , la Ligue contre le Cancer and l'Agence Française du sang.


Contributions of the laboratory in the field.

a) Biological studies.

1) CLL B lymphocytes are characterized by low expression of surface membrane Ig.

In relation to phenotypic characterisics of CLL B-cells, employing quantitative immunocytochemical techniques, we could demonstrate an important abnormality of the CLL proliferating B lymphocyte, characterized by a ten-fold decrease of surface membrane immunoglobulins, when compared to normal B lymphocytes. This phenomenom is acknowledged as the most reliable phenotypic characteristic of CLL B lymphocytes.

2) CLL B cells frequently correspond to expansion of a B cell clone committed to secrete NAAB.

We have examined antibody specificities of CD5+ B-chronic lymphocytic Leukemia (CLL) cells by fusing leukemic lymphocytes from 27 different CLL patients with the non-secreting X-63 mouse myeloma. We have found that more than half of B-CLL patients were expressing NAAB activity.

3) Immunoglobulin V genes studies segregate CLL patients into two groups expressing either unmutated or mutated V genes.

Although B-CLL cells phenotypically correspond to a mantle zone naive B cell, which should express unmutated V genes, with Harry Schroeder by reviewing data on 75 VH sequences, we have shown that half of CLL patients were expressing mutated VH genes and suggested that CLL could be a heterogeneous disorder, where CLLs expressing unmutated V genes could correspond to a more undifferentiated form of the disease. More recently, Hamblin et al and Damle et al confirmed these finding and demonstrated that the mutational pattern of Ig VH genes was associated to prognosis. They demonstrated that expression of unmutated VH genes was associated to poor prognosis, suggesting that a malignant transformation of less differentiated lymphocytes was associated to increased malignant potential.

b) Prognostic and therapeutic studies.

CLL, the most frequent form of leukemia in Western countries, has long been considered, as an unpredictable disease. In this field with Jacques-Louis Binet we described a three stage prognostic classification, which has been adopted simultaneously with the Rai's staging system, as the international reference prognostic classification. This classification allows to segregate patients into three different prognostic groups, (good, intermediate and bad prognosis), which is very important when planning therapeutical strategies.

In 1976, we have created with J.L. Binet the French Cooperative Group for CLL. This group, integrated by more than 50 French teams, has been able to include in therapeutical trials, more than 3300 patients. Three different randomized trials started in 1980, 1985 and 1990, allowed our group to demonstrate that: 1) Good prognostic patients should not receive any treatment, unless disease progression occurs. Given the long term follow-up (>11 years) and the large number of patients included in these trials (>1500 patients), these results have provided important evidence favouring a watch and wait policy for good prognosis CLL patients (stage A), who represent 65% of all CLL patients. 2) The polychemotherapy CHOP regimen and the purine analogue Fludarabine, are able to induce improvement of response in advanced forms of CLL.

Research Project:

a) What are the genes involved in CLL oncogenesis? (Responsible of the programme: G. Dighiero).

A research programme called " Carte d'identité des tumeurs " was activated by the Ligue pour la Recherche contre le Cancer. This programme aims to study by the microarray technique the transcriptional expression of the different genes during human malignancies. The French Cooperative Group on CLL, has presented on behalf of this group a research project aiming to study gene expression in CLL and to answer the following questions :

1) Which is the difference in terms of molecular profile between normal CD5+ B cells and CD5+ B cells proliferating in B-CLL?

2) Is there a different molecular profile in patients belonging to good and poor prognosis of the Binet's classification ?

3) Is there a different molecular profile in CLL cases expressing unmutated VH genes when compared to those CLLs expressing mutated VH genes ?

4) What are the genetic events underlying the bimodal evolution of some CLL patients? In CLL clinical evolution is frequently given by an indolent phase that is in a second step followed by an agressive phase, though it remains to be defined whether this last event is related to the existence of a second malignant hit. We are planning to investigate this by comparing gene expression in the same patients during the indolent phase and once transformation to an agressive course has occurred.

5) What is the relationship between the molecular profile of CLL and other CD5+ B cell malignancies like mantle zone lymphoma 

This project has been approved and received the financial support of the Ligue pour la Recherche sur le Cancer and G. Dighiero assumes its coordination.

b) What are the mechanisms underlying underexpression of the BCR and impaired functional response in CLL B cells? (Responsible: Béatrice Payelle-Brogard).

Scientific background.

The B-cell receptor (BCR) consists of two heterodimers CD79a/CD79b bound non-covalently with the surface membrane Ig (SIg). In CLL, both SIg and CD79b have been reported to be consistently under-expressed, though the mechanisms accounting for this phenomenon have not been elucidated. In addition, CLL B cells display very poor signaling upon stimulation through the BCR. Since anergic B cells have been demonstrated to express low levels of the BCR and to poorly signal upon BCR stimulation, CLL B cells may correspond to malignant transformation of an anergic B cell.

Present status and future directions.

In a previous work, we have studied the complete sequence of the B29 gene, encoding for CD79b, in 20 different familial CLL patients and demonstrated the absence of genetic abnormalities underlying the low surface expression of the CD79b molecule (B. Payelle-Brogard et al, 1999). Since genetic alterations could not account for under-expression of the BCR in CLL, we have undertaken transcriptional and post-transcriptional analysis of this gene. Our results demonstrate that: 1) there is no major defect in transcriptional expression of the BCR. 2) Analysis of intracellular expression of the BCR components showed adequate synthesis, despite consistent low membrane expression of the receptor. 3) Neither a genetic defect in the transmembrane domain of SIg which associated with CD79a/CD79b nor a genetic abnormality in the chaperone protein calnexin involved in folding and assembly of the BCR were found. 4) In contrast a defect in the processing of nascent IgM, demonstrated by the failure to become resistant to endoglycosidase H treatment was found. This defect indicates that they are retained in the endoplasmic reticulum. 5) A constant defect in assembly of Ig M and CD79b chains of the BCR occured in CLL patients. These findings demonstrate that a post-transcriptional defect located at the BCR intracellular assembly and/or trafficking levels could be involved in the low expression of the BCR in B-CLL and are reminiscent to those previously reported in a transgenic murine model of anergic B cells. This project is supported by a grant of the Fondation contre la Leucémie.

c) Is there a role for anti-idiotypic vaccines in CLL? (Responsibles: F. Vuillier in collaboration with D. Scott-Algara).

Scientific background.

Since all major chemotherapeutic treatments in CLL succeed to induce good responses but are unable to cure this disease, one of the major questions that needs to be answered is whether there is a place for immunotherapeutic approaches like anti-idiotypic vaccines. Since each B cell undergoes a unique rearrangement that is characteristic for each individual B cell and malignant B cell hemopathies are characterized by clonal expansion of a clone displaying a unique rearrangement, idiotype constitutes a privileged tumor antigen.

One of the aims of tumor immunotherapy is the induction of CD8 cytotoxic T lymphocytes (CTL). It is established that various tumor cells share tumor associated antigens, though whether such antigens can elicit CTL response remains elusive.

Present status and future directions.

The French Cooperative Group on CLL has committed our group to initiate a feasibility "in vitro" study on the possibility to generate specific cytotoxic anti-idiotypic CD8 cells, through incubation of these T cells with dendritic cells (DC) previously pulsed with the tumoral idiotype. Following the complete sequence of VH and VL genes from the CLL clone, the Ig is expressed by transfecting X-63 myeloma with the same expression vectors used to construct transgenic mice. In these conditions, DCs obtained from patient's monocytes are pulsed with the complete Ig or different peptides corresponding to CDR3 hypervariable Ig region.

In a first step, freshly purified monocytes from CLL patients and normal donors were induced to differentiate in GM-CSF and IL-4 in serum-free medium and compared for their morphological, phenotypic and functional characteristics. Our results demonstrate that: 1) functional DCs can be generated from CLL patients with similar phenotype and function to that observed with normal donors, 2) in contrast to normal controls Mo-derived DCs from CLL patients spontaneously secrete endogenous IL-10 and 3) IFN-g in combination with CD40L plays a major role in priming DC from CLL patients for IL-12 and IL-15 production. These results indicate that it is possible to derive functionally competent dendritic cells from circulating monocytes in CLL patients. In a second step, VH and VL genes from CLL B lymphocytes from different patients have been isolated and cloned. This allowed us to produce patient's respective idiotype in different forms: complete IgM by recombinant technology and CDR3 peptides displaying class I high affinity. Next, monocyte derived dendritic cells from CLL patients were pulsed with these idiotypic reagents and evaluated for their capacity to induce specific CD4 and CD8 cell lines with anti-tumoral activity. In these conditions, CD8 cells lines with a moderate cytotoxic activity have been obtained (about 20% of cytotoxic activity in a 25/1 ratio; manuscript in preparation). To improve the efficiency of this system, pulsing of CLL monocyte-derived dendritic cells with exosomes derived from tumoral B cells is attempted.


Contributions of the laboratory in the field.

An important focus of our previous work was the study of the functional impairment of CD4, CD8 and NK cells in AIDS. This functional impairment intervenes early in the course of the disease, when CD4 cell counts are not severely decreased. By studying purified subsets, we have demonstrated that all these three subsets display a severe defect in proliferative and functional responses.

Next, our work was devoted to the study of the transduction signals delivered through the tyrosine-kinase/phosphatase pathway in HIV-seropositive patients and the role of chronic oxidative stress in the functional impairment of CD4 T cells. We have also demonstrated that isolated CD4 cells from HIV patients display a decreased phosphorylation process associated to abnormalities in the expression of the p56lck and the p59fyn tyrosine-kinases, which correlated with the impaired proliferative responses. Interestingly, a similar pattern has been previously reported in anergic CD4 murine clones. We could also demonstrate that oxidative stress may play a role in the functional impairment of immune effector in HIV-infected patients, since anti-oxidant treatment is able to revert the impaired proliferative activity from CD4 cells of HIV+ patients .

Research project:

Mechanisms of immunological reconstitution following high anti-viral therapy in AIDS.

In a randomized trial designed by ANRS, our laboratory was in charge of the immune reconstitution study in previously untreated asymptomatic HIV-infected patients, treated with a classical tri-therapy regimen. This study enrolled 343 patients that were followed over 72 weeks and were segregated according to viral charge following initial therapy into two groups: group 1 displaying aviral charge <500 copies/ml (217 patients) and group 2 displaying >500 copies/ml (136 patients). Baseline CD4 cell count (/mm3) was higher and HIV RNA (log10 copies/ml) was lower in the group 1 as compared to group 2. Over the 72 weeks, mean CD4 T cells count increased by bz 0.31 cells/µl/day in group 1 and by 0.18 cells/µl/day in-group 2 (p<0.0001). Contribution of CD4 naive cells (CD4 + CD45RA + CD62L +) to total CD4 increase was constant over time and not different in the 2 groups (44% in group 1 vs. 60% in group 2, p=0.27). Total CD8 cells decrease was not significantly different in groups 1 and 2. However, at W72, activated CD8 T cells (DR+) were stable in-group 1 and increased in-group 2 (+4 vs. +101 cells/µl, p=0.02). In conclusion baseline immuno-virological characteristics are important determinants of treatment outcome. Our data suggest that a permanent control of HIV replication could be necessary for faster immune reconstitution .

In collaboration with Dr. Rivière, we used peptide/MHC tetrameric complexes to enumerate the frequency of HLA class I-restricted CD8+ T cells recognizing Gag and Pol epitopes in 28 positive and 13 negative HLA-A*02 HIV-infected children.. The frequencies of circulating CD8+ T cells that recognized these epitopes were determined in whole blood samples. Results indicate that HIV-specific CD8+ T cells can be easily detected in peripheral blood of HIV-infected children using HLA tetramers combined to HIV peptides. These cells are CD3+, CD28-, CD45RA-, CD45RO+, CD69- and HLA-DR+, and their frequencies are comparable to those of infected adult patient.

We are presently evaluating the restoration of the immune function against three opportunistics agents (CMV, T. Gondii and P. Carinii ) in patients following the start of tri-therapy. This work is conducted by Miguel Alfonso (PhD student) under the direction of D. Scott-Algara. Scid/beige immunodeficient mice are reconstituted with human lymphoid cells (PBL) from non-infected and HIV—infected patients. Humanized mice are then challenged with T. Gondii and the specific immune response and survival are studied. Preliminary results show that immunodeficient mice can be humanized. The survival of mice reconstitued with PBLs following 12-month tri-therapy is significantly longer than that of mice reconstituted with PBLs from the same patient before tri-therapy. In addition, specific anti- T. Gondii immune response can be detected in the case of mice reconstituted following tri-therapy. These results indicate that specific immune responses against T. Gondii can be reconstituted after 1-year tri-therapy.

Photo 1

Normal mouse / Mouse suffering of bilateral loss of external ears following injection of eCAGAS

Photo 2

Intracytoplasmic staining of dTg mice PBL before (D.O) and after (D.38) M. pulmonis infection.


puce Publications of the unit on Pasteur's references database


  Office staff Researchers Scientific trainees Other personnel

BOUYSSIE Reine, bouyssie@pasteur.fr

BROGARD Béatrice, Chargée de Recherche CNRS, bbrogard@pasteur.fr

POURCEL Christine, Chef de Laboratoire IP, cpourcel@pasteur.fr

SCOTT-ALGARA Daniel, Chargé de Recherche IP, scott@pasteur.fr

ALFONZO DIAZ Miguel, malfon@pasteur.fr

HAVOUIS Séverine, shavouis@pasteur.fr

OPPEZZO Pablo, popezzzo@pasteur.fr


BEDORA Marie, Technicienne Supérieure IP, mbedora@pasteur.fr

DUMAS Gérard, Technicien Supérieur IP, dumas@pasteur.fr

JOUANNE Colette, Ingénieur d'Etudes CNRS

MAGNAC Christian, Ingénieur IP, magnac@pasteur.fr

VUILLIER Françoise, Ingénieur IP, vuillier@pasteur.fr

Laboratoire de préparation :
LAURENT Nathalie


Page Top research Institut Pasteur homepage

If you have problems with this Web page, please write to rescom@pasteur.fr.