Research / Scientific departments / Immunology / Teams & Projects

Teams & Projects

Tumor immunity and preclinical development of cancer vaccines

Claude Leclerc, Gilles Dadaglio, Cheng-Ming Sun
Nada Chaoul, Catherine Fayolle, Camille Guillerey, Marine Oberkampf, Francesc Rudilla Salvador, Alexandre Tang
Human Papillomaviruses (HPV) are the major cause of human cervical cancer. We have developed a therapeutic tumor vaccine, the CyclaseA-E7 (CyaA-E7), that can induce specific responses against the HPV E7 oncoprotein and eliminate E7 expressing tumors, such as TC-1. This vaccine is currently under clinical trials (
Tumor immunosuppression represents the major cause of the failure of many cancer therapies. Most of the immunotherapies found efficient in mice are indeed inefficient in cancer patients, indicating that traditional mouse tumor models do not recapitulate the human situation. We thus developed a “failure” model in which we delayed the treatment of mice-bearing TC-1 tumor by a CyaA-based therapeutic vaccine and demonstrated that the therapeutic efficacy of the vaccine is progressively lost as the tumor grows. To understand and overcome these suppressive responses, we are analyzing the interaction between tumor cells and the immune system during tumor growth.
Our analysis shows that along with the increase of the tumor volume, several immunosuppressive cells accumulate in TC-1 tissue, especially Foxp3+ regulator T cells (Treg cells) and myeloid derived suppressor cells (MDSC). The accumulation of these cells may explain the decrease of vaccine’s efficiency.

Cellular infiltrates within the tumor microenvironment 

The accumulation of CD4+ Tregs in tumor tissue is a widely described phenomenon in mouse models and in human cancer patients. Understanding the mechanisms by which Treg cells migrate and accumulate into tumors is of primary importance for the development of successful immunotherapies. We used the immunoscope technology to analyze the TCR repertoire of tumor-infiltrating T cells. We showed that tumor-infiltrating Treg and Teff cells displayed CDR3 spectratyping profiles characteristic of biased and strongly perturbed repertoires, typical of clonal expansions, suggesting that strong T-cell responses have occurred within the tumor tissue. Comparison of the TCR sequences of tumor-infiltrating Tregs demonstrated an extremely high percentage of overlapping sequences inside a BV family revealing the existence of public sequences. These public TCR sequences were tumor specific and were not shared between Treg and Teff tumor-infiltrating T-cells, demonstrating that conversion is not an active process at the tumor site. The identification of public TCR sequences in intratumoral Tregs, also present in the draining lymph node of the same animal, demonstrates that infiltration of tumor tissue by these cells is followed by an intense proliferation in the tumor, which skews the highly diverse TCR repertoire towards a few dominant clones.
We demonstrated that the alkylating reagent cyclophosphamide (CTX) can rescue the therapeutic activity of CyaA-E7 against large tumors. CTX is a widely used chemotherapeutic reagent which does not induce anti-tumor response by itself, but when injected 24 hours before CyaA-E7, it increases dramatically the expansion of E7 specific CD8+ T cells and eliminates large tumor. It is generally believed that CTX enhances immune responses through the depletion of Treg cells. However, we have demonstrated that after CTX treatment, there is only a temporary and mild decrease of Treg cells in tumor. In contrast, we have observed a recruitment of CD11b+Ly6G+ granulocytes and CD11b+Ly6C+ monocyte to spleen and to tumor, respectively. Interestingly, granulocytes and monocytes are also the major subsets of MDSC. The increase of granulocytes and monocytes along with the expansion of CD8+ T cells indicated that these cells do not suppress immune responses after CTX treatment. Our research interest is to understand the immune responses induced by CTX, focusing on the following directions:
  1. The qualitative modification of Treg cells after CTX treatment, including phenotypic, functional and genomic analysis
  2. The qualitative modification of myeloid cells after CTX treatment. MDSCs represent a group of phenotypically heterogeneous, functionally flexible cells. They could induce inflammatory responses and at same time suppress other immune cells. Tumor may induce the suppressive property of myeloid cells whereas CTX treatment may reprogram these myeloid cells from suppressor to immunogenic cells.
Understanding the responses induced by CTX and how it reprograms immune suppressors is crucial for the design of efficient immunotherapeutic strategies in the future.

Plasmacytoid dendritic cells in tumor immunity

Antigen cross-presentation by plasmacytoid dendritic cell
To induce CD8+ T cell responses against exogenous Ags, antigen presenting cells should deliver exogenous Ag to MHC class I molecules by cross presentation, allowing processing of these Ag in the MHC class I antigen-presentation pathways. We demonstrated that pDCs were able to capture efficiently exogenous Ag at steady state, but were unable to cross present these Ags. However, after activation, pDCs acquired the capacity to cross-present soluble and particular exogenous Ag leading to the generation of efficient primary CD8+ T cell responses against these Ags in vivo (Mouriès et al, Blood, 2008). These results suggested that pDCs could play an important role in the induction of CD8+ T cells responses against virus, which did not infect hematopoeitic cells or tumoral Ags. Furthermore, it is also a first evidence of a regulation of cross presentation by pDCs. The mechanisms of induction of cross-presentation are under investigation.
 Dendritic cells internalizing exogenous protein (in green) before degradation and presentation
A new mouse model lacking pDCs
To directly assess the physiological functions of pDCs in vivo, it is crucial to characterize immune responses in the absence of pDCs, and thus to develop an animal model lacking pDCs. In collaboration with P. Kastner, we obtained a strain of mice with a hypomorphic mutation in the Ikaros locus (IKL/L) leading to the absence of pDCs in periphery. Due to abnormalities in B cell development and function as well as T cell leukemia development in IKL/L mice, we crossed these mice with syngeneic Rag2-KO mice to obtain an homozygous line of mice carrying both the Ikaros mutation and deficient for the Rag2 gene (IKL/L Rag-/- mice). This new line of mice lacked pDCs, T and B cells while other cell subsets were not affected. These mice could be reconstituted with normal T and B cells offering the possibility to analyze adaptive response in the absence of pDCs. This model will be used to analyse the role of pDCs in innate and adaptive anti-cancer immunity in various grafted tumor models.
Pivotal role of plasmacytoid dendritic cells in inflammation and NK cell responses
Using our mouse model lacking pDCs, we have established that pDCs are essential for the in vivo induction of NK cell activity in response to TLR9 triggering (Guillerey et al, Blood, 2012). Furthermore, we provided the first evidence that pDCs are critical for the systemic production of a wide variety of chemokines in response to TLR9 activation. Consequently, in the absence of pDCs, we observed a profound alteration in monocytes, macrophages, neutrophils, and NK cell recruitments at the site of inflammation in response to CpG-Dotap. Our results highlight the pivotal role played by pDCs in the induction of innate immune responses following TLR9 triggering.

Clinical development of cancer vaccines

Claude Leclerc, Gilles Dadaglio, Richard Lo-Man, Catherine Fayolle, Daphné Laubreton

Adenylate cyclase
Based on our discovery that a bacterial protein, the adenylate cyclase (CyaA), binds to dendritic cells (DCs), we have developed a highly efficient vector capable of targeting a wide range of antigens to antigen presenting cells (APC), leading to strong immune responses. A large panel of recombinant CyaAs carrying different heterologous viral and/or tumoral T-cell epitopes were shown to stimulate strong and specific Th1 and CTL responses in animals, and to induce protective immunity against viral and tumoral challenges, as well as therapeutic efficacy against transplanted tumors. We used this new vector to develop two therapeutic vaccine candidates against cervical cancer and melanoma. Clinical trials have respectively started in July 2012 and January 2011 ( and results (tolerance and immunogenicity) are expected in the following months.

A hallmark of cancer cells is a change in glycosylation processes leading to the abnormal expression of carbohydrate chains, such as antigens of the T family (Tn, sialyl-Tn and T) on most carcinomas (breast, colon, prostate and ovary). We have developed an efficient approach based on a fully synthetic glycopeptide structure, for the induction of antibodies against the Tn glycosidic antigen expressed by tumor cells.
This molecule, namely MAG-Tn3, is a glycopeptidic dendrimer including the Tn antigen and a pan-DR T cell epitope from tetanus toxin and represents the first example of a fully synthetic glycosidic vaccine with immunotherapeutic potential against cancer. We are currently organizing a Phase I/II clinical trial of the MAG-Tn3, which is planned to start at the end of 2013.

Immunity and vaccine science in neonates (Label FRM)

Richard Lo-Man, Edith Dériaud, Sebastien Lemoine, Brian Mozeleski, Dania Zhivaki
Irrespective of host genetic factors, extremes of age are characterized by a higher susceptibility to particular infections. This remains ill understood due to gaps in the knowledge of the immune system of the newborn and the elderly. Thus, these populations constitute privileged targets and reservoirs for diseases.

Age dependent regulatory mechanisms

Early life is a period of high susceptibility to infectious diseases and a stage of immune maturation. Poor CTL or Th2-skewing responses are rendering the individual particularly susceptible to intracellular pathogens. Pediatric vaccination is poorly efficient, particularly before six months of age, requiring many boosts. For decades, newborns have been considered as immunologically immature, this period being considered as prone for tolerance induction. We challenged the hypothesis of neonatal immunological immaturity by showing that mouse DC are fully functional and can efficiently induce Th1 and CTL responses in newborn mice. Our main work is to decipher regulatory mechanisms at work in earlylife. Our objective is to provide an integrated picture at the level of both effector and regulatory cells of how the neonatal immune system accomodates pediatric vaccines and pathogens.


Neonatal Bregs

We evidenced a neonatal mouse regulatory B cell (Bregs) compartment producing IL-10 in the context of Toll like receptor (TLR) activation in vivo influencing CD4 Th cell differentiation through dendritic cell control (Immunity, 2005). This process is sustained by type I IFN provided by dendritic cells (pDCs and cDCs). This mechanisms also provide protection against TLR dependent lethal inflammation in neonatal mice (Zhang et al., JEM 2007). The human counterpart of this regulatory mechanism is under investigation.

Regulatory neutrophils

We identified a new regulatory role for neutrophils triggered by mycobacteria and Gram-negative bacteria that is based on co-activation through the MyD88 and CLR/DAP12/Syk pathways. This regulation involves the production of large amounts of IL-10. IL-10-producing neutrophils are able to temper lung inflammation upon mycobacterial infection in vivo and may play a similar role in other bacterial infections (Zhang et al., Immunity, 2009).

Innate pathways for neonatal adjuvant setup

The TLR pathways are strongly impaired in neonates leading to a deficit of innate response. The consequences are a reduced capacity to control pathogens at the level of both innate and adaptive immune responses. Translation of the knowledge of the innate TLR activation pathways to vaccine and adjuvant design has also emerged over the last ten years as a promising strategy to improve prophylactic and therapeutic vaccination. In these conditions, this strategy may fail if applied to neonatal vaccination, and new strategies to improve infant vaccination need to be explored to offer alternatives. The aim of this program is to evaluate the potential of other innate activation pathways including Nod-like receptors (NLR) as well as targeting C-type lectin receptors (CLR) as an alternative for neonatal adjuvant settings.