FR2789088A1 - MEANS FOR INDUCING OR INCREASING THE IMMUNOGENICITY OF MYELOID ACUTE LEUKEMIA CELLS - Google Patents

Abstract

The invention relates to a method for inducing or increasing the immunogenicity of myeloid acute leukemia cells. The inventive method consists of the following: myeloid acute leukemia cells are brought into contact with compounds by means of incubation, whereby said compounds can confer a phenotype and/or a non-tumoral cell function upon the leukemic cells and comprise at least one CD4O cell, a growth factor for said cells such as GM-CSF and a cytokine such as IL-4 in conditions which enable the cultivation of cells. The invention can be used to produce medicaments to combat myeloid acute leukemia.

Description

<Desc / Clms Page number 1>

 The present invention relates to means for inducing, or increasing, the immunogenicity of AML (acute myeloid leukemia) cells, while maintaining the leukemic characteristics of these cells, and in particular their antigenic characteristics.

 One of the technical problems posed by tumor cells is that they may present tumor antigens on their surfaces, but that they lack co-stimulatory capabilities. They are therefore not able to present these antigens appropriately, and thus do not activate T cells: the immune system is not able to respond effectively to tumor proliferation.

Among the strategies that have been developed to respond to this situation, the identification of antigenic peptides, and their vaccine administration in soluble form, or in associated form (fusion, transfection) to professional antigen presenting cells have been considered ( CPA) such as dendritic cells (DC). With the means of the prior art, however, this strategy remained technically unsatisfactory at least on two levels:
1) Since the tumor cells are not immunogenic, the identification of tumor antigens remained problematic. In the case of AML in particular, the very existence of such antigens had not yet been demonstrated,
2) the recommended solution of administering one or more identified tumor antigens could allow the body to provide an immune response directed specifically against that particular antigen (s), but it does not completely rule out the possibility that tumor cells also escape this response via other antigens not specifically identified, or not specifically taken into account.

 The strategy proposed by the inventors consists, on the contrary, in transforming the tumor cells themselves, that is to say cells that express all the possible antigens, to make them acquire

<Desc / Clms Page number 2>

 only one phenotype, but also all the functions of CPA, and of DC in particular.

 The present invention thus provides means (methods and kits) for inducing immunogenicity on AML cells (acute myeloid leukemias, subtype LAMO, LAM1, LAM2, LAM3, LAM4, LAM5, LAM6, LAM7 according to the FAB classification. French-American-British), and AML2 cells (acute myeloblastic leukemia with maturation), LAM3 (acute promyelocytic leukemia), AML4 (acute myelomonocytic leukemia), AML5 (acute monocytic leukemia) in particular, or to increase the immunogenicity of such LAM cells effectively (i.e., so that the treated LAM cells are capable of causing a significant immune response).

 Remarkably, the means according to the invention are rapidly effective, and are adapted to humans.

 The means according to the invention thus allow in particular the identification of AML antigens, the manufacture of anti-AML drugs adapted to humans and a method for predicting and / or adjusting the efficacy of these drugs.

 A first aspect of the invention therefore relates to a method for inducing or increasing the immunogenicity of LAM cells and the treated cells thus obtained as well as to their biological applications.

 This method is characterized in that it comprises contacting, in particular by incubation, AML cells with compounds capable of conferring phenotype cells and / or functions of non-tumor cells to the leukemic cells, these compounds comprising at least one ligand CD40, a growth factor of said cells, such as GM-CSF, and a cytokine such as IL-4, under conditions allowing cell culture.

 Advantageously, the contacting step is carried out by incubation under conditions adapted to the survival of the AML cells. The nature and intensities of the parameters of such culture conditions can be

<Desc / Clms Page number 3>

 selected and adjusted by those skilled in the art using his knowledge in the field of cell cultures, and leukemic cell cultures in particular, and with respect to the particular case to be treated. Examples of such conditions are given in the "Examples" section which follows. It may especially be mentioned a temperature of about 37 C, and / or an atmosphere comprising CO 2, for example about 5%, and / or a culture medium such as RPMI 1640 medium, optionally enriched in serum, for example about 5%.

 Such an incubation may optionally be accompanied by an elimination of non-adherent cells.

 The CD40 ligand (s) used may be present in soluble form or in form associated with a cell. A CD40 ligand may, for example, correspond to CD40L (also called CD 154) available in soluble form from Immunex, and in membrane form (L-CD40L cells) from Schering-Plow.

 In a free form, the CD40 ligand (s) is (are) advantageously present at a concentration of about 150 to 1150 ng / ml for an initial concentration of LAM cells of the order of 0.1. at 0.5x106 cells / ml. In associated form, the CD40 ligand (s) is (are) advantageously present) at a concentration of the order of 0.05 × 10 6 associated cells / ml for an initial concentration of LAM cells of the order of 0, 1 to 0.5 × 10 6 cells / ml.

 The growth factor (s) of AML cells, such as GM-CSF, is (are) advantageously present at a concentration of about 50 to 100 ng / ml for an initial cell concentration. AML in the range of 0.1 to 0.5 × 10 6 cells / ml.

 The cytokine (s), which IL-4, is (are) advantageously present (s) at a concentration of about 10 ng / ml for an initial concentration of LAM cells of the order of 0 , 1 to 0.5 × 10 6 cells / ml.

 The LAM cells used in the incubation step may in particular be obtained by sampling from a patient AML, and by

<Desc / Clms Page number 4>

 blood sample in particular.

 Remarkably, the method according to the invention is effective very quickly: indeed, it does not require a pre-culture step. The contacting can thus be limited to less than 6 days, preferably less than 4 days, for example to 3 days.

 According to an advantageous arrangement of the invention, the contacting step results in the acquisition by said LAM cells of a type of phenotype CPA, and type DC in particular. The nature of the elements corresponding to such a phenotype are part of the general knowledge of those skilled in the art, which also has means for determining their expression. Examples of such phenotype elements and analysis means are furthermore given in the examples which follow.

 Advantageously, the acquisition by LAM cells of a mature CPA type phenotype, and of mature DC type in particular, is observed. The treated AML cell population then generally has a CD la phenotype after treatment.

 At the level of the AML cell population, a negative regulation of the CD14 molecule can also be observed. The population of AML cells then generally exhibits a CD14- phenotype after treatment.

 In addition, one can also observe a positive regulation of the CD83 molecule. The AML cell population then generally has a CD83 + phenotype after treatment.

 The AML cells may also have at the end of the contacting step, a positive regulation of at least one adhesion molecule, such as CD54, CD58, and / or at least one MHC molecule ( Major Histocompatibility Complex), such as a class I MHC molecule, a class II MHC molecule, the DQ isoform, and / or at least one costimulatory molecule, such as CD80, CD86.

 Advantageously, the contacting step results in a positive regulation of all of these molecules. The cell population

<Desc / Clms Page number 5>

 AML then generally has a CD54 +, CD58 +, MHC class I +, MHC class II +, DQ +, CDSO +, CD86 + phenotype after treatment.

 The different populations of LAM cells mentioned above, as obtained by the method of the invention, are also, as new products, within the scope of the invention. These are cells reprogrammed to acquire a new identity card.

 According to a remarkable arrangement of the invention, the LAM cells are characterized in that they have functions of CPA, and of DC in particular. The nature of such functions is part of the general knowledge of the skilled person, who also has the means to determine their existence. Examples of such functions and means of analysis are also given in the following examples.

 Advantageously, the AML cells are capable of stimulating the proliferation of CD4 + T lymphocytes. Many means for verifying the acquisition of such a capacity are available to those skilled in the art, and examples are given in the "examples" section which follows.

 The treated LAM cells are notably capable of inducing the differentiation of CD8 + T lymphocytes into cytotoxic T lymphocytes (CTL).

The acquisition of this ability to induce CTL differentiation represents the maximum efficiency of the process, it can be obtained by combining the three components: CD40 ligand, cell growth factor such as GM-CSF, and cytokine such as IL -4 ". Numerous means making it possible to verify the acquisition of such a capacitance are at the disposal of those skilled in the art, and examples thereof are given in the "Examples" section which follows (lysis of targets, specific lysis, production of IFN).

 Particularly advantageously, the CTLs induced using the means according to the invention are capable of specifically lysing cells corresponding to the AML cells before treatment.

 For such induction of CTL, a ratio (CD8 + T lymphocytes / AM T cells) of advantage -and greater than 5/1

<Desc / Clms Page number 6>

According to one provision of the invention, said AML cells retain leukemic features, such as possible chromosomal abnormalities. Advantageously, said LAM cells retain their antigenic characteristics (preservation of tumor antigens and / or histocompatibility complex they expressed before treatment).

 Also remarkably, said acquisition of phenotype CPA, and DC in particular, and / or said acquisition of functions of CPA, and DC in particular, and / or said conservation of leukemic and / or antigenic characteristics is carried out (are carried out). for a majority of LAM cells contacted, and for an average percentage greater than 75% of them approximately.

 The properties conferred on the treated LAM cells are advantageously exploited, according to the invention, in various biological applications.

 The means according to the invention may thus be for the manufacture of anti-LAM drugs adapted to humans or animals. The present invention therefore also aims at: the use of a composition capable of conferring phenotype and / or non-tumor cell functions on leukemic cells, this composition containing at least one CD40 ligand, a leukemia cell growth factor , such as GM-CSF, and a cytokine, such as IL-4, for the manufacture of an anti-LAM drug, and in particular anti-LAM2 and / or antiLAM3 and / or anti-LAM4 and / or anti-LAM5 , which can be administered to humans or animals, the use of treated LAM cells as defined above, in particular as obtained with the process according to the invention, for the manufacture of an anti-LAM drug, and in particular anti-LAM2 and / or antiLAM3 and / or anti-LAM4 and / or anti-LAM5, administrable to humans or animals, and the use of cytotoxic lymphocytes (CTL) as produced by using LAM cells as obtained with the process according to the invention for the

<Desc / Clms Page number 7>

 manufacture of an anti-AML drug, and in particular anti-LAM2 and / or antiLAM3 and / or anti-LAM4 and / or anti-LAM5, which can be administered to humans or animals.

 The present invention also provides, in a second aspect, means (method and kit) for identifying AML cell antigens.

It provides in particular a method for the identification of an AML antigen, characterized in that it comprises the use of treated AML cells as defined above, in particular as obtained with the method according to the invention. invention, so as to identify at least one molecule on the surface of these cells which participates in an immune reaction. The identified antigen may be a tumor antigen or a histocompatibility complex antigen.

 For the implementation of this method, kits comprising the reagents and compounds necessary for the identification of antigens are advantageously used, together with a user guide.

 An embodiment of the identification method according to the invention may further comprise the constitution of an RNA library of the treated LAM cells (constitution of an expression library by cloning the RNAs of the LAM cells in a eukaryotic vector ). The LAM RNA library can then be contacted with lymphocytes, and CD8 + T lymphocytes in particular.

Such CD8 + T lymphocytes advantageously come from the same patient as the AML cells before treatment. A RNA from the library is then identified as corresponding to an AML antigen when the corresponding clone in the library is capable of inducing differentiation of CD8 + T lymphocytes into CTL, and / or inducing production by CD8 + T lymphocytes. interferon gamma and / or TNFα. From the identification of an RNA corresponding to a LAM antigen, the person skilled in the art can, using his knowledge and the means of the art, sequence this RNA, obtain the cDNA, DNA and protein sequences , (poly) peptide corresponding.

 An antigen as identified by the process according to the invention is particularly advantageously used for the manufacture of an anti-LAM, and anti-LAM2, and / or anti-LAM3 drug. and / or anti-LAM4 and / or

<Desc / Clms Page number 8>

 anti-LAM5 in particular. The present application is aimed in particular at such use for the manufacture of a preventive or therapeutic AML vaccine.

 The vaccine composition also falls within the scope of the invention as such. This composition contains the vaccine active ingredient with the usual pharmaceutically acceptable adjuvants.

 In yet another aspect, the present invention also provides means (method and kit) for predicting and / or adjusting the efficacy on a given patient of a drug as manufactured using a drug. any according to the invention. Such a prediction and / or adjustment method comprises the analysis of the CD40 expression of the leukemic cell population of said patient. This prediction method makes it possible, in addition to a therapeutic decision adapted to each case of leukemia considered to adjust the reactive doses and / or the dosage of said drug to the patient for whom this drug is intended. Kits for the implementation of this method are also part of the invention.

 The present invention is illustrated by the following examples in which reference is made to the following figures: FIGS. 1A, 1B, 1C, 1D show microscopy images (May-Grünwald-Giemsa staining) of leukemic cells after different treatments according to the invention (GM-CSF + IL-4, GM-CSF + IL-4 + CD40L) of three days, - Figures 2A, 2B show graphs illustrating, as a function of the number of days of treatment, the negative regulation of CD 14 (FIG. 2A) on leukemic cells treated according to the invention by GM-CSF and IL-4, and illustrating the overall CD1α phenotype of leukemia cell populations treated according to the invention, FIGS. 3A, 3B, 3C illustrate the expression of CD83 (flow cytometry) on the leukemic cells as a function of the treatment carried out according to the invention,

<Desc / Clms Page number 9>

 FIGS. 4A, 4B, 4C, 4D illustrate the expression of DC surface markers by the leukemic cells after various treatments according to the invention: CD86 (FIG. 4A), MHC Class 1 (FIG. 4B), DQ isoform (FIG. 4C) ), analysis of parameters (CD86, CD83, CD54) in dispersion (FIG. 4D); FIG. 5 illustrates the preservation of genetic anomalies by the leukemic cells treated according to the invention (hybridization using DNA probes), FIGS. 6A and 6B illustrate the production of IL-12 (FIG. 6A) and IL-8 (FIG. 6B) by different LAM5s treated according to the invention and by the immature and mature MoDC controls, FIGS. 7A, 7B1, 7B2, 7B3, 7C illustrate the stimulation of CD4 + T lymphocyte proliferation obtained using leukemic cells undergoing different treatments (cpm versus number of stimulator cells) compared to PBMC, activated monocytes, immature MoDC and mature, - Figure 8 illustrates the correlates between ability to stimulate CD4 + proliferation and acquisition of the CD83 + phenotype by the leukemic cells treated according to the invention (incorporation of 3H thymidine into cpm depending on the number of stimulator cells used); FIGS. 9A, 9B, 9C illustrate the capacity leukemia cells treated according to the invention to induce the differentiation of CTL: lysis of P815 as a function of the ratio E / T (effector cells / stimulator cells) in FIG. 9A,% specific lysis as a function of the E / T ratio in FIG. 9B, and production of IFNγ by induced CTLs as a function of the treatment of the leukemic inducer cells (GM-CSF on the left; GM-CSF / IL-4 / CD40L on the right) in FIG. 9C, and FIGS. 10A and 10B illustrate the CTL capacity induced by a patient to destroy the own leukemic cells thereof.

<Desc / Clms Page number 10>

EXAMPLES
MATERIALS AND METHODS
Patient samples, peripheral blood mononuclear cells and cord blood cells
Leukemic cell samples were obtained, after informed consent, from patients before chemotherapy. The LAM5 blasts analyzed in this study (see Table 1 below) come from 10 patients treated in Marseille at the Paoli-Calmettes Institute (9 adults aged 18 to 71) or Timone Hospital (1 child) between 1994 and 1997. All monocyte cells observed on blood smears presented a leukemic phenotype at diagnosis, and leukemic cells accounted for 57-95% of circulating cells. Six of the ten leukemias presented blood cells with cytogenic abnormalities, of which 3 were for chromosome 8. The three oldest patients died after the first induction before an accurate assessment of their condition was possible. Five patients had complete remission after the first induction with chemotherapy, and two patients after the second induction with chemotherapy. For this study, we selected those leukemic blast populations that had more than 50% live cells after 5 days of in vitro culture. The cell morphology was analyzed after three days of culture.

 Naive lymphocytes were obtained from cord blood samples. The mononuclear fraction of all these samples was isolated by centrifugation according to the Ficoll-Hypaque density gradient, and frozen as live cells in liquid nitrogen.

<Desc / Clms Page number 11>

Table I: Clinical and Paraclinical Parameters of 10 Acute Monoblastic Leukemias Tested

<tb> Leukemia <SEP> Classification <SEP> karyotype <SEP>% <SEP> cells <SEP> age / sex <SEP> remission <SEP> allo / auto / transplant <SEP> status <SEP> survival
<tb> FAB <SEP> leukemic <SEP> complete <SEP> after <SEP> e <SEP> current
<tb> devices <SEP> the first <SEP>
<tb> induction
<tb> LAMa <SEP> LAM5a <SEP> inv <SEP> 8 <SEP> 79 <SEP> 60/2 <SEP> yes <SEP> no <SEP> death
<tb> LAMb <SEP> LAM5a <SEP> XX <SEP> 57 <SEP> 43/2 <SEP> yes <SEP> no <SEP> CR <SEP> alive
<tb> LAMc <SEP> LAM5b <SEP> inv <SEP> 16, <SEP> +22 <SEP> 97 <SEP> 41/2 <SEP> yes <SEP> autologous <SEP> CR <SEP> alive
<tb> LAMd <SEP> LAM5 <SEP> 8+ <SEP> 95 <SEP> 71/2 <SEP> no <SEP> no <SEP> death
<tb> LAMe <SEP> LAM5a <SEP> XY <SEP> 84 <SEP> 48/1 <SEP> yes <SEP> allograft <SEP> death
<tb> and <SEP> Cosinophylia
<tb> LAMf <SEP> LAM5a <SEP> 7-, <SEP> 8+ <SEP> 95 <SEP> 70/2 <SEP> no <SEP> no <SEP> death
<tb> LAMg <SEP> LAM5a <SEP> XX <SEP> 93 <SEP> 1/2 <SEP> no <SEP> no <SEP> CR <SEP> alive
<tb> LAMh <SEP> LAM5b <SEP> t (<SEP> 1 <SEP> Op, <SEP> 11q, <SEP> 20q) <SEP> 62 <SEP> 74/2 <SEP> no <SEP> no <SEP> death
<tb> LAMi <SEP> LAM5a <SEP> XX <SEP> 79 <SEP> 18/2 <SEP> yes <SEP> autograft <SEP> CR <SEP> alive
<tb> and <SEP> Cosinophylia
<Tb>
<tb> LAM <SEP> LAM5a <SEP> t <SEP> (9p, <SEQ> 11q, <SEQ> 16q) <SEP> 91 <SEP> 36/1 <SEP> no <SEP> no <SE> PR <SEP> living
<Tb>
CR = complete remission PR = partial remission

<Desc / Clms Page number 12>

Media and reagents
Leukemic and control cells were cultured in RPMI 1640 (Biowhitaker) supplemented with 10% fetal calf serum (FCS) at 37 ° C and 5% CO 2. For lymphocyte cultures, the medium was supplemented with 1 mmol / L sodium pyruvate (Life Sciences), 1.5 mmol / L L-glutamine (Life Sciences), 5. 10-5 mmol / L P -mercaptoethanol, 50 mg / mL streptomycin (Sigma), 50 U / mL penicillin (Sigma). The following recombinant human cytokines were used: GM-CSF (Novartis), IL-2 and y-IFN (Interferon) (Roussel Uclaf), IL-4 (Schering Plow), TNFα (Roche). An anti-CD3 antibody such as monoclonal antibody 289 (Pierres A. et al., European Journal of Immunology, 1988, vol 18,685-690).

 The following antibodies were used for flow cytometry studies: CDla (BL6, IgG1), CD4 (13B8.2, IgG1), CD8 (B9.11, IgG1), CD14 (RM052, IgG2a), CD15 (80H5). , IgM), CD33 (D3HL60, IgG1), CD34 (QBEnd10, IgG1), CD40 (5C3, IgG1), CD83 (HB15a, IgG2b) and HLA-DQ (SPVL3, IgG2A) all from Coulter-Immunotech (Marseille, France). CD80 (BB1 / B7, IgG1, Becton Dickinson), CD86 (IT2.2, IgG2b, Pharmingen).

FITC or PE conjugated goat anti-mouse (GAM) antibodies (Coulter Immunotech, Marseille, France) were used as the second antibody to label the non-conjugated monoclonal antibodies. CD1 (BU15) and CD23 (9P25) are from Sem Sealand, Schering Plow, Lyon, CD58 from ATCC (Rockville, USA), CDla (OH3.9.3), CD14 (IOG3.3), CD25 (33B3. 1), CD80 (2D10.4), MHC Class 1 (YJ4), MHC Class II (25. 1) were produced according to conventional techniques.

Cell separation
Peripheral blood mononuclear cells (PBMCs) from healthy donors were isolated according to Ficoll-Hypaque gradients. The isolation of CD4 and CD8 lymphocytes was carried out by negative selection, using magnetic beads coated with anti-CD8 or anti-CD4, CD14 monoclonal antibodies,

<Desc / Clms Page number 13>

 CD56, CD45RO and CD 19 (Immunotech, Marseille, France) For the elimination of CD4 memory cells, anti-CD45RO UCHL1 1 antibodies (Immunotech, Marseille, France). The elimination of L cells from the cultures was carried out using anti-H2Kk hybridoma supernatant (81 020) (Naquet P., Marseille Luminy Immunology Center). For screening of CD83 + cells, the leukemia cells were cultured for 5 to 7 days in the presence of GM-CSF, IL-4 and L-CD40L cells, washed once in 5 mM EDTA (saline phosphate buffer) buffer. and incubated for 0.5 hours in PBS comprising 30% human serum and then labeled for 0.5 hours with FITC-conjugated anti-CD83 monoclonal antibodies (100 L / 10 6 cells, 10 6 cells / ml). The cells were then washed twice in cold 5 mM PBS EDTA before sorting using the FACS Ventage Cell Sorting System (Becton Dickinson). After selection of the leukemic cells, the CD83 +, CD83- and total populations were sorted by green fluorescence, compared to an isotype control immunoglobulin.

Cell lines
Murine L cells transfected with human CD40L (LCD40L cells) or with human CD32 (L-CD32 cells) were used after irradiation at 75 Gy (Banchereau J. Dardilly). In cytotoxicity assays, P815 murine mastocytoma cell lines and K562 leukemia cell lines were used.

Leukemic cells and dendritic cells
Acute monoblastic leukemia cells (AML type 5, according to FAB classification) were used. After thawing, the cells, whose B and T lymphocytes were removed or not, were cultured at 0.5x106 / ml for three days, or for various periods of time, in the absence or presence of L-CD32 cells. (0.05 106 / ml) or L-CD40L cells (0.05 106 / ml), GM-CSF (100 ng / ml), TNFα (5 ng / ml), IL-4 (10 ng / ml) .

<Desc / Clms Page number 14>

 The culture medium was supplemented with cytokines every three days, and the cells were fractionated when necessary. To obtain control dendritic cells, PBMCs from healthy donors were thawed, and after 2 hours in RPMI at 5% FCS at 37 ° C., the non-adherent cells were removed by gentle rinsing with PBS. These cells comprised about 90% monocytes, and were cultured at a concentration of 0.1 to 0.5 10 / ml for 7 days in the presence of GM-CSF (100 ng / ml) and IL-4. (10 ng / ml). At day seven, 0,05. 106 ml L-CD32 or -CD40L cells were added for two to three days. To obtain control monocytes, adherent cells obtained from PMBC were cultured for 24 hours in the absence or in the presence of IFNγ at 10 ng / ml.

Flow cytometry
After removal of the murine fibroblast population co-cultured with magnetic anti-H2K monoclonal antibodies, the cells were labeled at 4 C using Ca2 + / Mg2 + free PBS containing 30% human BSA and 0, 02% azide for 0.5 hours. The cells were then incubated for 0.5 hours with the monoclonal antibodies and then incubated with FITC or PE (Immunotech) conjugated goat anti-mouse F (ab ') 2 fragments when non-conjugated monoclonal antibodies were used. After washing, the labeled cells were fixed in 2% formaldehyde PBS prior to FACScan flow cytometer analysis (Becton Dickinson). Windows have been put in place. When no fixation is made, a control window was made by staining with propidium iodide.

Preparation of T lymphocytes
Cryopreserved peripheral blood or cord blood mononuclear cells were thawed on the day of culture to prepare allogeneic CD4 or CD8 T cells using negative magnetic immunoselection as described above. Briefly, 200 l of magnetic beads for

<Desc / Clms Page number 15>

 10 million cells were incubated for 15 minutes in PBS in the presence of 5 g of each purified monoclonal antibody of mouse anti-human CD14, CD56, CD19, and CD8 or CD4 respectively. To obtain pure naive T cell populations from cord blood, anti-CD45RO monoclonal antibody was added. The cells were preincubated in PBS comprising 30% human serum, before incubation in the presence of labeled beads for 10 minutes. The non-adherent fraction was selected after incubation with a magnetic element. The purity obtained is greater than 95% as verified by FACS analysis.

Proliferation tests (MLR)
Proliferation of T cells was performed in 96-well flat-bottomed plates in a final volume of 200 l. A constant number of 2.5 or 5105 purified T cells was cocultivated with increasing numbers of leukemic cells or irradiated control cells (50 Gy), in a final volume of 200 l in triplicate. T cell proliferation was monitored by measuring the incorporation of [methyl-3H] TdR (Amersham, UK) during the last 8 hours of a 6-day culture. The thymidine sample was counted on a gas-phase counter (Matrix 9600, Packard) whose efficiency is about 1/5 that of a scintillation counter ss.

Cytotoxicity tests
After 7 days of culture, the leukemic cells were irradiated (50Gy) and then cocultivated with naive allogenic CD3 T cells in 24-well plates at a rate (effector / stimulator cells) of 2 to 5 depending on the experiments. After 6 days of mixed lymphocyte and tumor cultures (NLTC), during which the necessary lymphocytes were proliferated, the cells were counted so as to test their cytolytic activity against the targets P815, K562, and against different LAM5 cells. For CTL (cytotoxic lymphocyte) assays, the target cells were labeled with 100 Ci of

<Desc / Clms Page number 16>

 Cr (NEN, USA), for 2 hours at 37 C. 103 labeled targets and serial triplicate dilutions of effector cells were incubated in RPMI at 10% FCS in 96-well V-bottom plates. For lysis of P815 targets, incubation was performed in the absence or presence of 289 anti-CD3 hybridoma supernatants. The plates were centrifuged at 500g for 3 minutes and incubated at 37 ° C for 4 hours. Then, 50 l of supernatants were collected to measure the release of radioactivity using a counter y (Topcount Packard). Specific lysis was determined for each triplicate experiment as follows: specific lysis (% = [51 Cr release experimental 51 spontaneous 51 Cr release] / [51 Cr maximum release - spontaneous 51 Cr release] x 100. The maximal release was determined by the addition of 2% MP40 detergents (Sigma, St. Louis, MO) to the target cells.

Determination of cytokines
The supernatants of differentiated leukemic cultures were collected at day 3 and then frozen. After thawing, the cytokine concentration was measured using enzyme immunoassays purchased from Coulter Immunotech (Marseille, France) for IL-12 and IL-8.

FISH (fluorescence in situ hybridization)
FISH hybridization was performed according to standard techniques on cytospin preparations from FACS sorted CD83 + and CD83- cells (FACS Venting, Becton Dickinson) from three selected patients due to detection of a cytogenetic abnormality. by interphasic FISH. In fact, the LAMc patient presented trisomy 22 at the time of diagnosis, whereas the AML and AML patients had trisomy 8. These abnormalities were observed in all the metaphases analyzed at diagnosis (n> 20). For the detection of trisomies 22 and 8, a probe specific for the BCR gene (Vysis) and a probe specific for the centromere of chromosome 8 (Onca), were

<Desc / Clms Page number 17>

 respectively used. At least 200 nuclei were examined by fluorescence microscopy by 2 independent observers.

RESULTS
Culture conditions for the production of dendritic cells from leukaemic blasts
Leukemic cell samples were grown in RPMI supplemented with FCS in the presence of various combinations of GM-CSF, IL-4, TNFα and L-CD40L cells. Most leukemias showed morphological changes by day 3, mainly after addition of GMCSF in the presence of IL-4. The cells exhibited significant morphological changes with irregular cell membrane phenomena, cytoplasmic advances and increases in variable size. These morphological changes are illustrated by FIGS. 1A and 1B: FIGS. 1A and 1B show microscopy images (May-Grünwald-Giemsa staining cytospin images) which illustrate the dendritic differentiation of leukemic cells (LAMe) after a 3-day culture on GM-CSF and IL-4, with large clusters or adhesion to plastic (Figure lA, x 10), and with cytoplasmic advancement phenomena and irregular cell membranes (Figure 1B, x 40).

 These changes were observed in the whole leukemia population for 2 leukemias (AMF and AML), or only in a subpopulation that ranged from 20 to 50% of the total population for 6 other leukemias. These results are illustrated in Table 2 below.

<Desc / Clms Page number 18>

Table II: Morphological analysis of the leukemic cells tested

<tb> Identity <SEP> Viability <SEP> without <SEP> Adhesion <SEP> at <SEP> Heterogeneity <SEP> Clusters <SEP> Increase <SEP> Appearance
<tb> Leukemia <SEP> GM-CSF <SEP> plastic <SEP> CD40L <SEP> * <SEP> of <SEP><SEP> size <SEP> of <SEP> dendrites *
<tb> cell <SEP>#
<tb> LAMa <SEP> yes <SEP> + <SEP> yes <SEP> ++ <SEP> + <SEP> ++
<tb> LAMb <SEP> yes <SEP> ++ <SEP> yes <SEP> ++ <SEP> + <SEP> +++
<tb> LAMc <SEP> yes <SEP> + <SEP> yes <SEP> ++ <SEP> ++ <SEP> ++
<tb> LAMd <SEP> oui- <SEP> yes <SEP> ++ <SEP> + <SEP> ++
<tb> LAMe <SEP> oui- <SEP> yes <SEP> ++ <SEP> ++ <SEP> ++
<tb> LAMf <SEP> oui- <SEP> no <SEP> ++ <SEP> - <SEP> ++
<tb> LAMg <SEP> oui- <SEP> no <SEP> + <SEP> + <SEP> +
<tb> LAMh <SEP> yes <SEP> ++ <SEP> yes <SEP> ++ <SEP> + <SEP> ++
<tb> LAMi <SEP> no <SEP> no <SEP> ± <SEP> +
<tb> LAMj <SEP> yes <SEP> + <SEP> no <SEP> ++ <SEP> ++ <SEP> +++
<Tb>
* CD40L cluster and appearance of fine dendrites observed in the presence of CD40L, on all or part of the cells.

# increase in cell size seen in the presence of GM-CSF or IL-4 Not all LAM5s are identical to each other, and leukemia cells are heterogeneous, but all are sensitive to CD40L.

<Desc / Clms Page number 19>

 Cells that showed such morphological changes belonged either to the adherent cell population or to the nonadherent population according to leukemia and culture conditions (see Table 2 above). Thus, two leukemias showed a strong adhesion to plastic (LAMb and LAMh) which decreased after addition of IL-4. TNFa has rarely induced such changes, except for the case of AML where clusters were observed. Co-incubation with CD40L + cells stimulated the constant formation of larger or smaller clusters of round cells around LCD40L cells, which were sensitive to the addition of EDTA. The addition of CD40L has also been associated with the acquisition of fine dendrites by leukemic cells. The constant formation of clusters is illustrated in Figure 1C, and the appearance of fine and long dendrites in Figure 1D.

 FIGS. 1C and 1D represent indeed microscope slides (May-Grünwald-Giemsa staining cytospin views) which illustrate the dendritic differentiation of leukemic cells (LAMe) after 3 days of culture on GMCSF, IL-4 and CD40L, with clusters. surrounding L cells transfected with CD40L (Figure 1C, x10), and with the appearance of fine and long dendrites (Figure 1D, x40).

 Two leukemias showed very little, if any, response to all the different culture conditions tested, and presented small round non-adherent monomorphic cells (AMLg and AMI), and only weakly adhered to L cells. -CD40L.

 The changes observed were even more striking after 10 days of culture in the presence of "monocyte-derived dendritic cell-like" populations for all the LAMs tested except LAMg and LAMi.

 Thus, the cells from eight of the ten LAM5s tested were able to differentiate in vitro so as to acquire the morphological characteristics typical of dendritic cells (DC, see Table 2 above).

<Desc / Clms Page number 20>

Phenotypic Differentiation of Leukemic Cells in Response to DC Differentiation Signals
Differentiation in dendritic cells is associated with molecular modifications such as the downregulation of CD14, and the upregulation of adhesion molecules (CD54, CD58), costimulatory molecules (CD80, CD90), MHC molecules (Class I, DR, DQ) and differentiation molecules such as CD la or CD83.

The monocytic molecule CD14 was observed expressed in the diagnosis of eight out of ten leukemias. It was negatively regulated on five of them, after three days of culture in the presence of IL-4. These results are illustrated by FIG. 2A (percentage of CD14 + cells as a function of the number of days of culture on GM-CSF / IL-4) which represents the downregulation of CD14, as measured by flow cytometry on leukemic cells. LAMb LAMb grown for 10 days in the presence of GM-CSF and IL-4, and harvested at days 3.7 and 10 (days 3, 7, 10). Monocyte derived dendritic cells (MoDC) served as positive controls (black triangles). In Figure 2A, the legend is as follows:
CD14 expression was also downregulated on control monocytes treated in the same way.

 This was not associated, as it might have been with the normal MoDC model, with a major joint increase in CD1a expression, with the exception of leukemia (LAMc). These results are illustrated in FIG. 2B (percentage of CDla + cells as a function of the number of days of culture on GM-CSF / IL-4) which represents the downregulation of CD14, as measured by flow cytometry on leukemic cells. LAMb LAMb grown for 10 days in the presence of GM-CSF and IL-4, and harvested at days 3.7 and 10 (days 3, 7, 10). Monocyte derived dendritic cells (MoDC) served as positive controls (black triangles). The legend of Figure 2B corresponds to that of Figure 2A.

<Desc / Clms Page number 21>

 Prolonged cultures in the presence of GM-CSF and IL-4 could not induce the expression of CD1a (LAMb, d, e, f, g, h, i), and the CD14- / phenotype. CDla 'persisted (see Figures 2A, 2B), or induced the expression of CD la only on a small subpopulation (LAMh and LAMj).

 Other immature DC molecules, such as the mannose receptor and E-cadherin, have not been upregulated in most of the leukemic samples treated under the different culture conditions described herein. Thus, while CD14 was downregulated, immature DC markers were only rarely observed.

 It was then investigated the expression of CD83, which is normally observed on mature peripheral dendritic cells. This phenomenon is illustrated by FIGS. 3A, 3B and 3C. These figures indeed represent flow cytometric measurements of the membrane expression of CD83 (black graph surfaces) on leukemic cells after a three-day culture in the presence of different combinations of GM-CSF / IL-4 / CD40L. these combinations being explained below the graphs. For each graph of FIGS. 3A, 3B and 3C, the percentage of positive cells and the average fluorescence intensity (% / MFI) are indicated. Isotype controls are represented by white surfaces. Figure 3A gives the results obtained for LAMi, which is representative of two leukemias (LAMg and LAMi) that never express CD83. Figure 3B gives the results obtained for LAMc which is representative of six leukemias for which a strong expression of CD83 appears on a fraction of the cells in the presence of CD40L, this expression being further stimulated by addition of IL-4. Figure 3C gives the results obtained for LAMf which is representative of two leukemias (AMLf and AML) which express CD83 on almost all cells after culture on GM-CSF / IL-4 / CD40L.

 For eight leukemias, a strong expression of CD83 was thus obtained after three days of culture in the presence of CD40L, which expression was further increased when CD40L was combined with IL-4. CD83 was observed expressed

<Desc / Clms Page number 22>

 or on the entire leukemic population (AML and AML, see Figure 3C), or a leukemic subpopulation (see Figure 3B).

 Two leukemias (LAMg and LAMi, see Figure 3A) never expressed CD83, even after 15 days of culture, and even when typical conditions of MoDC culture were used (10 days of culture in the presence of GM-CSF and IL-4, followed by CD40L stimulation).

 The expression of adhesion molecules (CD54, CD58), MHC molecules (Class I, Class II and DQ isoforms) and costimulatory molecules (CD80 and CD86) was analyzed. All these molecules were upregulated (CD54, CD58, MHC Class I, MHC Class II, DQ, CD86) or induced (CD80) on 10 of the leukemias tested, mainly after coincubation with GM-CSF, IL-4 and CD40L. These results are illustrated by Figures 4A, 4B and 4C.

 FIGS. 4A, 4B and 4C show, in fact, for ten LAM5s (LAMa to LAMj) cultured for 3 days in the absence or in the presence of the factors indicated on the abscissa, the factor for increasing the expression of CD86 (FIG. 4A ), MHC class I molecules (Figure 4B) and class II MHC DQ isoform (Figure 4C).

 The newly CD83 + cells showed strong expression of adhesion and co-stimulation molecules. These results are illustrated in FIG. 4D, which shows the analysis of LAMc after culture in the presence of GM-CSF, IL-4 and CD40L, said LAMC appearing representative of six heterogeneous leukemias. Leukemias LAMg and LAMi still do not respond to these culture conditions.

 Taken together, these data demonstrate that most LAM5 leukemic blasts (eight of the ten tested), acquire in vitro mature DC markers (including CD83) and upregulation of MHC class I, MHC class II, and particularly DQ, as well as costimulatory molecules such as CD80 and CD86. However, the stadium

<Desc / Clms Page number 23>

 Immature differentiation could only be very rarely detected with these crop conditions.

Complementary evidence of the leukemic origin of DC derived from leukemia generated in vitro
The leukemic origin of CD83 + cells obtained from leukemic cell populations was confirmed by: i. the strong expression of CD 15, a molecule characteristic of the immature stage of the monocytic line expressed on leukemic cells, and ii. the persistence of initial genetic abnormalities as demonstrated by FISH for 3 leukemias (AML, AML and AML).

 The results of the LAMC are shown in FIG. 5. FIG. 5 indeed represents the hybridization results obtained with a FITC-labeled chromosome 22 specific DNA probe (fluorescence microscope analysis) for CD83 + cells that are derived from screening of LAMC, after 5 days of culture in the presence of GM-CSF, IL-4 and CD40L, and which have undergone FISH analysis on cells at interphase. This experiment is representative of three experiments (3 LAM) carried out with the appropriate probes for each of them.

 The mature DC cytokine IL-12 is produced by leukemia cells when cultured in the presence of CD40L, and this production is further enhanced by addition of IL-4.

 IL-12 is produced by mature CD after cross-linking with CD40 or Class II molecules, which amounts to mimicking the interaction between TcR CD4 or CD40L. 11-12 stimulates NK cells, but participates in the development of C cells and stimulates the differentiation of CTL (cytotoxic lymphocytes). The leukemic cells were incubated as described above, and the IL-12 p70 heterodimer was measured by the ELISA technique in the supernatants. These ELISA results are illustrated in FIG. 6A: × 10 6 / ml leukemic cells (LAMα to AMI) were incubated for 3 days in the presence of

<Desc / Clms Page number 24>

 GM-CSF (light gray bars), GM-CSF and IL-4 (dark gray bars), GM-CSF, IL-4 and CD40L (black bars); as positive controls, averages obtained for 4 immature and mature non-leukaemic MoDCs are indicated. As controls were used immature MoDCs. As expected, unlike immature MoDCs, mature MoDCs produce a very high amount of IL-12. For seven leukemias (LAMa, c, d, e, h, i, j), a similar characteristic was seen with of significant IL-12 produced, as measured in the presence of only CD40L at concentrations ranging from 150 to 1150 ng / ml with, in most cases, a synergistic effect exerted by IL-4. IL-12 production peaked at day 3 and was only weakly detectable after 6 days. The amounts of IL-12 produced are similar to those produced by mature MoDCs. Remarkably, IL-12 is produced independently of CD40L stimulation for two of the leukemias tested (AML and AML). One of the leukemias was not able to produce significant amounts of IL-12 (LAMg), although it produced, like the other AMLs tested, IL-8 under the same culture conditions. confirming that IL-12 production was in most cases restricted to leukemias that differentiated into DC.

The IL-8 production results, which were obtained in a manner comparable to that obtained for the production of IL-12 in Figure 6A, are illustrated in Figure 6B.

 Thus, differentiated leukemia cells acquire, as mature DC, the ability to produce large amounts of IL-12.

 The CD4 + naive T cells proliferate in the presence of leukemia cells cultured in the presence of CD40L. The ability to stimulate a primary response of T cells is one of the key features of DC functions. The ability of the differentiated leukemic cells obtained to stimulate the proliferation of CD4 + lymphocytes of cord blood has therefore been verified. The results obtained are illustrated by FIGS. 7A, 7B1, 7B2, 7B3 and 7C.

 In FIG. 7A (cpm as a function of the number of stimulating cells), LAMα, which belongs to the group of leukemias that differentiate in DC on a cell subpopulation, was cultured for 3 days in vitro.

<Desc / Clms Page number 25>

 presence or absence of GM-CSF, TFNa, IL-4 and CD40L. CD4 + allogeneic cord blood T cells were isolated, and 105 T cells were stimulated for 6 days in the presence of a variable number of irradiated leukemia cells (50 Gy). Proliferation was measured by incorporation of 3H thymidine for the last 8 hours. Mononuclear peripheral blood cells served as controls. This experiment is representative of three experiments performed with the five other heterogeneous differentiated leukemias. The stimulation controls used were: IFNγ-activated monocytes, CD40L-stimulated and d7-mature CD40 MoDCs, and the controls shown are representative of two donors. The results correspond to an average of measurements made in triplicate, and are representative of four experiments.

 In FIGS. 7B 1, 7B2, 7B3 (cpm as a function of the number of stimulating cells), the AMLs were precultured for 3 days in the presence of GMCSF, or GM-CSF and TNFα, or GM-CSF, IL-4 and CD40L. The same protocol was applied as in FIG. 7A (same choice of controls, measurements made in triplicate), with a constant number of 0.5 × 10 5 lymphocytes per well. LAMe (FIG. 7B 1), LAMh (FIG. 7B2) and LAM1 (FIG. 7B3) are respectively representative of the three groups of leukemias that have been observed. Each graph is representative of a group of two to five experiments. PBMCs were used as controls.

 In Figure 7C (cpm versus number of stimulator cells), LAMe, which belongs to the group of leukemias that differentiate into DC on a subpopulation, was cultured for 7 days in the presence of GM-CSF, IL-4 and CD40L, and irradiated. CD45 + CD45RO-naive lymphocytes were isolated from cord blood on magnetic beads, then the same protocol as in FIG. 7A was used (same choice of controls as in FIG. 7A, representative results of four experiments and corresponding averages of measurements made in triplicate), with a constant number of 0.5 x 105 CD4 + lymphocytes per well.

<Desc / Clms Page number 26>

For all of FIGS. 7, the legend of the graphs is as follows: Leukemia culture conditions: single fat line, middle only black triangles, tips down + GM-CSF black triangles, peaks at top + GM-CSF / TNF black circles + GM-CSF / IL-4 white squares + GM-CSF / CD40L round white + GM-CSF / IL-4 / CD40L control culture conditions: single trait PBMC white diamonds activated monocytes white triangles, dotted line MoDC immature triangles white, continuous line MoDC mature
Thus, as shown in FIG. 7A for LAMα, which belongs to the group of leukemias which present a differentiation in DC on a leukemic subpopulation, a high proliferation was observed in the presence of leukemias pre-treated with GM-CSF and CD40L. The addition of IL-4 to CD40L did not significantly enhance proliferation (see Figure 7A). A 7-day culture of leukemic cells nonetheless improved the observed T-cell proliferation.

 Among the 10 leukemias studied, three groups could be determined according to their capacity to induce the proliferation of naive T cells after in vitro culture in the presence of GM-CSF, with or without the addition of CD40L. In Figures 7B1, 7B2, 7B3, leukemia representative of these three groups is shown.

<Desc / Clms Page number 27>

 Six leukemias showed exactly the same type of naive CD4 + T cell proliferation induction (see Figure 7B 1), ie, proliferation was observed when they were precultured in the absence CD40L at the highest concentration of stimulating cells. On the other hand, a high proliferation was observed when they were precultivated in the presence of CD40L, even in the presence of small amounts of leukemic stimulating cells (see FIG. 7B 1).

 Both leukemias LAMd and LAMh showed very strong stimulation cells after culture on GM-CSF in vitro, and this stimulation could not be further increased by addition of TNFα or IL-4 and CD40L (cf. Figure 7B2).

 The two leukemias LAMg and LAMi showed rather weak stimulation cells except when CD40L was added, which then allowed a small but significant proliferation of naive TCD4 + cells at the height of leukemic cells (see Figure 7B3). .

 To further emphasize the DC-stimulating properties of precultivated leukemia for 7 days, we selected CD45RO- naive CD4 + cord blood T cells by removing the small fraction of CD40RO + cells (5 to 10% likely corresponding to PBMCs). contaminant of maternal origin at the time of collection), and used them as response cells. In addition, interferon-activated normal monocytes and immature and mature MoDCs were tested as stimulators. For LAMe leukemia, in a representative experiment of the results obtained, a 7-day preculture of leukemic cells in the presence of GM-CSF, IL-4 and CD40L induces the proliferation of naive CD4 + T cells at levels almost similar to those observed in the presence of mature control MoDCs (see Figure 7C).

 In contrast, immature DCs or activated monocytes induce very low proliferation, and only at the highest stimulator cell concentrations.

<Desc / Clms Page number 28>

 Thus, the differentiation of leukemic cells to DC is associated with the acquisition of an ability to stimulate the proliferation of naive T cells.

The expression of mature CD83 + DC molecules on leukemic blasts differentiated and correlated with the ability to proliferate naive T cells
It was then investigated whether the strong ability to stimulate naive T cells observed for leukemias precultured on CD40L is associated with the acquisition of the CD83 + differentiation marker. Two leukemic populations, of which only one subpopulation has acquired the CD83 differentiation marker (AML, AMEL), were cultured for 5-7 days in the presence of GM-CSF, IL-4 and L-CD40L cells ( see Figure 7C). The CD83 + and CD83- cell fractions were then isolated by cell sorting, and their ability to stimulate proliferation of naive CD4 + was tested. The results obtained are illustrated in FIG. 8: LAMe cells were cultured for 7 days in the presence of GM-CSF, IL-4 and CD40L, then the CD83 + and CD83- subpopulations were isolated by cell sorting; the naive CD4 + CD45 + cells were isolated from cord blood on magnetic beads, and 0.5 x 10 5 lymphocytes per well were cultured for 6 days in the presence of different concentrations of irradiated leukemia cells CD83- (triangles), CD83 + ( squares) or unsorted (circles). Proliferation was measured by incorporation of 3H thymidine for the last 8 hours.

 As shown in Figure 8, for LAMe, a strong proliferation is observed only in the presence of CD83 + leukemic cells. Similar results were obtained for the other AML leukemia.

 These experiments demonstrate that the allo-stimulatory capacities are mainly restricted to CD83 + leukemic cells since the CD83- subpopulation is only very weakly effective. This also highlights the heterogeneity of tumor cell populations.

<Desc / Clms Page number 29>

 Differentiated leukemias can generate cytotoxic T lymphocytes (naive CTDs.

 The ability of pre-cultured leukemias to generate CTLs from CD45RO- T cells from cord blood was tested. Three leukemias were cultured for 7 days in the presence of GM-CSF or GM-CSF, IL-4 and L-CD40L cells, and then were used as stimulators of naive allogeneic T cells. The CTL differentiation of the response cells was analyzed in two ways: by evaluating global differentiation in CTL by redirected lysis using P815 and anti-CD3 monoclonal antibodies, and by evaluating their ability to kill leukemic blasts. These results are illustrated by FIGS. 9A, 9B and 9C, which show the results obtained for LAMe, which is representative of the two other leukemias tested: LAMe was cultured for 7 days in the presence of GM-CSF, or in the presence of GM-CSF. CSF, IL-4 and CD40L, and its ability to generate CTLs was tested.

 Figure 9A shows the CTL differentiation results, as assayed using standard 5'Cr release tests on P815 target cells in the presence of anti-CD3 monoclonal antibodies. The effector cells are CD3 + CD45RO-naive lymphocytes which were isolated from cord blood on magnetic beads, and which were grown on 24-well plates for 6 days in the presence of leukemic cells irradiated at a rate (effector cells / stimulating cells) of 2 (leukemic cells pre-cultured in the presence of GM-CSF: black circles, leukemic cells pre-cultured in the presence of GM-CSF, IL-4 and CD40L: black).

 The ability of naive lymphocytes stimulated by the same leukemic cells to lyse the corresponding stimulatory leukaemias was also tested.

These results are illustrated by FIGS. 9B and 9C, which represent the results obtained for LAMe, which are representative of the other two leukemias tested. CD45 + CD45RO-naive lymphocytes were cultured in the presence of leukemic cells irradiated at a ratio (effector cells / stimulating cells) of 5.

<Desc / Clms Page number 30>

 In FIG. 9B, the results of lysis of leukemic cells differentiated after 7 days are reported on GM-CSF / IL-4 / CD40L (black triangles) or on GMCSF alone (black circles), as analyzed by means of tests. 51Cr release standards, after six days of culture in the presence of stimulating leukemic cells.

 As a control, mature MoDC lysis exerted by differentiated CTLs was tested in the presence of mature MoDCs (white triangles). The 51 Cr release test results are representative of three leukemias, and two leukemia experiments.

 In FIG. 9C, the results of production of IFNγ measured by ELISA on the supernatants of cultures of three leukemias (culture on GM-CSF, or on GM-CSF / IL-4 / CD40L) are reported.

 As shown in Figure 9A, treatment with IL-4 and CD40L allows differentiation of naive CTLs. DC (normal mature DC) controls were highly effective in the experimental conditions tested for CTL induction.

 As shown in FIGS. 9B and 9C, for LAMe CTL differentiation, as determined by target cell lysis and IFNγ production, requires co-incubation of leukemic cells with IL-4 and CD40L in order to obtain a specific elimination. GM-CSF alone is not enough to induce CTL differentiation. Similar results were obtained with the other two leukemias tested. Remarkably, control target cells, such as other LAM5s and the NK-sensitive K562 target, are not removed by the three CTLs tested.

 In FIGS. 10A and 10B, the release percentage of Cr51 is reported as a function of the E / T ratio. The symbols used correspond to the following targets: - # -: LAMb, -A- LAMc, -1- LAMe and + LAMj.

 These results highlight the ability of autologous cells (CTL of the patient) to kill their own leukemic cells.

<Desc / Clms Page number 31>

DISCUSSION
The experiments carried out demonstrate that it is possible to induce the differentiation of acute human leukemias in a few days in the presence of IL-4 and CD40, to transform them into mature dendritic cell-type cells that express CD83, MHC molecules and costimulators, which produce IL-12 while maintaining their leukemic status.

 In addition, these cells can lead to the proliferation and, respectively, to the differentiation of the effector cells (CD4 + and CTL T cells).

 CTLs are an essential component of the immune response against tumors. CTL can indeed, by recognition of peptides associated with MHC class I molecules, treat tumors in animal models, and lyse human cancer cells in vitro. Peptides obtained from different tumor antigens are mutant proteins, fusion proteins or normal proteins expressed by tumors or differentiation antigens. Induction of CTL requires the identification of targets such as tumor antigens, and their stimulation using professional antigen presenting cells (APCs). The identification of tumor antigens, in particular melanomas, has made it possible to propose different cell therapy strategies by anti-tumor immunization, as well as by stimulation and expansion of anti-tumor CTL clones. In order to use the immune system as an adjuvant therapy, it is necessary to identify good antigenic targets, and to produce specific lymphocyte effector cells. Various experimental procedures are used to identify tumor antigens using genetic, biochemical or serological strategies. Another objective is the activation of effector T cells after administration of the tumor antigens. Various strategies are tested which correspond to peptides, proteins, heat shock proteins isolated from tumor cells, DNA vaccines. The presentation of antigens

<Desc / Clms Page number 32>

 Tumors can be in vivo essentially performed by indirect binding, via DCs that capture and exhibit tumor antigenic peptides associated with MHC class I molecules at CTL. A large number of effective costimulatory, adhesion and MHC molecules that are present on the mature DC membrane allows for the recruitment and activation of the extremely rare, specific anti-tumor CTLs that are supposed to exist, if they exist, to belong to the together naive lymphocytes. In fact, many animal experiments have confirmed and illustrated this dendritic capacity, highlighting the ability of peptide-stimulated DCs to inhibit new tumor development, and even to induce the elimination of established tumors. In order to overcome the absence of identified tumor antigens, since most of them are restricted to melanomas in humans, the tumor cells can be used as immunogens. In addition, the use of tumors as immunogens makes it possible to induce polyvalent immunization against multiple peptides, which prevents them from escaping CTLs which would be specific for only one peptide. Finally, this approach overcomes a possible hurdle that tumor antigens may not have access to DC in vivo.

 Since tumor cells are naturally devoid of CPA capacity, various strategies, using transfection of adhesion and costimulatory molecules, are used to increase their recognition by the immune system. Some strategies have used transfection of GM-CSF, a cytokine known to activate, recruit and affect monocyte differentiation, but also transfection of IL-12, one of the cytokines produced by mature DCs that are involved in the differentiation of Thl and activation of CTL and NK. Finally, others have also been able to fuse CPA-type B cells, or DCs to tumor cells. The present invention provides, for its part, particularly effective means, adapted to humans, which make it possible to differentiate between tumor cells into tumor dendritic cells: the

<Desc / Clms Page number 33>

 obtained cells possess the ability to present the APC antigen, and the ability to express tumor antigens.

 Hematopoietic tumors are the main candidates for differentiation in DC (which are leukocytes), and different studies have succeeded in their differentiation in DC CDla + (immature). These very immature leukemias express a constant translocation (9; 22) from which derived junction peptides are good candidates for tumor antigens. These leukemic DCs elicit in vitro specific cytotoxic responses to CML (chronic myeloid leukemias).

 The present invention has, in turn, concerned the particular case of AML (myeloid leukemias / acute). The work of the inventors demonstrates that the results presented here for LAM5 can be extended to other leukemias belonging to the acute myeloid lineage. As no tumor antigen has yet been demonstrated for AML, whether or not they are associated with different chromosomal abnormalities, the means according to the invention provide CTLs which can recognize these tumor cells, and thus identify the peptides. tumor. The in vitro protocol, described above, makes it possible to induce or increase the immunogenicity of the great majority of myeloid leukemias.

Therefore, this type of protocol can be used to identify antigens on these leukemias (AML), whether they correspond to minor histocompatibility antigens or to tumor antigens.

 A critical parameter is to preserve the AML antigens during differentiation in vitro in DC, so as to retain all or part of the tumor peptides and thus allow LAM cells to be lysed by the T cells. As the leukemic cells treated according to the invention retain their chromosomal abnormalities and / or markers associated with leukemic transformation, the present strategy conserves most, if not all, of the peptides found in undifferentiated tumor cells. Moreover, in most experiments that use tumors transplanted after modification by

<Desc / Clms Page number 34>

 transfer of costimulatory molecules or cytokines, CTLs generated in vivo retain the ability to recognize and eliminate unmodified tumors.

 The inventors here demonstrate by their work that it is possible for AMLs, which are leukemic cells attached to the monocytic line, to differentiate them in vitro into the best possible APCs: the cells obtained have the properties of mature DCs. The blast population consists mainly of monoblastic cells (LAM5a) and / or monocyte cells (LAM5b), which corresponds to an intermediate state between immature CD34 hematopoietic cells and completely differentiated monocytes. Some observations of LAM5 naturally expressing a dendritic phenotype, although rare, could also be made. In order to develop a protocol that can be used in clinical trials, the inventors have studied protocols that can be adapted to living cell conditions, or even in vivo conditions, but that are effective within 3 days of to keep alive a maximum of leukemic cells. The most effective protocol for obtaining in 3 days the phenotype and the functions of mature DC while preserving the leukemic characteristics corresponds to combinations of GM-CSF and CD40 ligands, IL-4 being very advantageously in the case where the we aim to produce CTL. The other protocols tested were, in most cases, less effective, and in particular the combination of GM-CSF and TFNa.

 However, 2 of the 10 leukemias tested failed to differentiate in vitro. Since leukemias may exhibit heterogeneous maturation stages, or may simultaneously express cells of different lineages, longer incubation times in vitro (up to 15 days) have been tested but have not allowed induce the desired differentiation. It can be noted that TGFss and FLT3L have similarly failed. However, these leukemic cells are not completely insensitive to cytokines, since as shown in Figure 7C, after stimulation they acquire up to a certain level the ability to stimulate proliferation of allogeneic naive cells. Is

<Desc / Clms Page number 35>

 these leukemic cells do not correspond to the LAM5a or LAM5b subtype and define a new subtype, either they show abnormalities in their reactivated to IL-4 or CD40L, which implies that they are at a stage of their differentiation where they are insensitive, or have mutations in their IL-4 and / or CD40L pathway. For example, in most cases, expression of CD40 on leukemia cells cultured without added cytokine was modal, indicating that the lack of response is not related to CD40 expression. In fact, a small percentage of CD83 + cells was sufficient to produce the desired naive proliferative response characteristic of mature DC, and this behavior of mature DC was confirmed by IL-12 production even though the leukemia cells producing this IL-12 is not specified.

 Pre-culture on CD40L appears to be one of the essential parameters to obtain a good proliferation of naive CD4 + lymphocytes, although it seems advantageous to associate CD40L with GM-CSF (and IL-4) to obtain maximal expression of the costimulatory molecules. , CD83 and IL-12. The essential role of the CD40 / CD40L interaction is therefore here demonstrated in the context of CD4 + lymphocyte participation at CTL generation, with an adequate response of DC to CD40L by overexpressing the costimulatory and presentation molecules, increasing the assists in cytotoxicity and / or cytostaticity, and in enabling CTL production. Thus, the expression of CD40 on acute myeloid leukemia, associated with the ability to differentiate towards the mature DC line, is an observation of particular interest, particularly for the implementation of vaccine therapies, which may be adjuvant .

Claims (18)

 1. A method for inducing, or increasing, the immunogenicity of AML cells, characterized in that it comprises bringing into contact, in particular by incubation of the AML cells with compounds capable of conferring on the leukemic cells phenotypes and / or non-tumor cell functions, these compounds comprising at least one CD40 ligand, a growth factor of said cells, such as GM-CSF, and a cytokine such as IL-4, under conditions allowing cell culture. 2. Method according to claim 1, characterized in that the (or) ligand (s) of CD40 is (are) present (s) in soluble form or in form associated with a cell. 3. Method according to claim 2, characterized in that the (or) ligand (s) in free form is (are) present (s) at a concentration of 150 to 1150 ng / ml for an initial concentration of LAM cells. on the order of 0.1 to 0.5 x 10 6 cells / ml. 4. Method according to claim 2, characterized in that the (or) ligand (s) CD40 in the associated form is (are) present (s) at a concentration of about 0.05 x 106 associated cells / ml for an initial concentration of LAM cells of the order of 0.1 to 0.5 x 10 6 cells / ml. 5. Method according to any one of the preceding claims, characterized in that the (or) factor (s) growth of said cells, such as GM-CSF, is (are) present (s) at a concentration of the order from about 50 to 100 ng / ml for an initial AML cell concentration in the range of 0.1 to 0.5 x 10 6 cells / ml. <Desc / Clms Page number 37>  6. Method according to any one of the preceding claims, characterized in that the (or) cytokine (s), that IL-4, is (are) present (s) a concentration of the order of 10 ng / ml approximately for an initial concentration of AML cells of the order of 0.1 to 0.5 x 10 6 cells / ml. 7. Method according to any one of the preceding claims, characterized in that said bringing into contact results in the acquisition by said LAM cells of a CPA type phenotype, in particular a mature CPA type phenotype and / or of function (s) deCPA, in particular by the acquisition of a DC type phenotype, in particular of a mature DC type phenotype. 8. AML cells characterized by downregulation of the CD14 molecule, and / or upregulation of the CD83 molecule, and / or upregulation of at least one adhesion molecule, such as CD54, CD58, and / or or at least one MHC (Major Histocompatibility Complex) molecule, such as a class I MHC molecule, a class II MHC molecule, the DQ isoform, and / or at least one molecule of co-stimulation, such as CD80, CD86. 9. LAM cells according to claim 8, characterized by their ability to stimulate the proliferation of CD4 + T lymphocytes and / or to induce the differentiation of CD8 + T lymphocytes into cytotoxic T lymphocytes (CTL), the latter being advantageously capable of specifically lysing cells corresponding to the AML cells before treatment. 10. LAM cells according to any one of claims 8 or 9, characterized in that they retain leukemic characteristics, such as possible chromosomal abnormalities and / or their antigenic characteristics. <Desc / Clms Page number 38> 11. Use of a composition capable of conferring phenotypes and / or non-tumor cell functions on leukemic cells, this composition containing at least one CD40 ligand, a leukemic cell growth factor, such as GM-CSF, and a cytokine, such as IL-4, for the manufacture of an anti-AML drug, and in particular anti-LAM2 and / or anti-LAM3 and / or anti-LAM4 and / or anti-LAM5, which can be administered to humans or the animal. 12. Use of LAM cells according to any one of claims 8 to 10, for the manufacture of an anti-LAM drug, and in particular anti-LAM2 and / or anti-LAM3 and / or anti-LAM4 and / or anti -LAM5, administrable to the man or the animal. 13. Use of cytotoxic lymphocytes (CTL) as produced using AML cells according to any one of claims 8 to 10, for the manufacture of an anti-LAM drug, and in particular anti-LAM2 and / or antiLAM3 and / or anti-LAM4 and / or anti-LAM5, administrable to humans or animals. 14. Use of the AML cells according to any one of claims 8 to 10 for identifying at least one molecule on the surface of these cells which participates in an immune reaction, in particular a tumor antigen or a histocompatibility complex antigen. 15. Use according to claim 14, characterized in that it further comprises the constitution of an RNA library of said treated LAM cells. 16. Use according to claim 15, characterized in that said LAM RNA library is brought into contact with lymphocytes, and CD8 + T lymphocytes in particular, the RNA of the library being identified as corresponding to a LAM antigen when the clone corresponding to it in the library is capable of inducing the differentiation of said CD8 + T lymphocytes into CTL, and / or inducing <Desc / Clms Page number 39>  the production by CD8 + T lymphocytes of interferon gamma and / or TNFα. 17. Use of an antigen as identified according to any one of claims 14 to 16 for the manufacture of a medicament and especially a vaccine against AML, and anti-LAM2, and / or anti-LAM3 and / or anti-LAM4 and / or anti-LAM5 in particular. 18. Therapeutic monitoring method for predicting and / or adjusting the efficacy on a given patient of a medicament as manufactured using a use according to any one of claims 11, 12, 13 or characterized in that it comprises analyzing the CD40 expression of the leukemic cell population of said patient.