Search for new therapeutic targets in cancerAs result of the joined effort of different laboratories, originally focused on distinct aspects of molecular and cellular biology, it was possible to identify disease models to which the present project might now contribute by identifying new therapeutic targets. Two groups of diseases were selected for further analysis: lymphoproliferative disorders and acute myelogenous leukemia.
Mantle cell lymphoma (MCL) is a distinct subtype of non-Hodgkin lymphoma that is associated with the translocation t(11;14)(q13;32) and ectopic overexpression of cyclin D1. The disease is predominantly disseminated at diagnosis and a frank leukemic phase occurs in one third of patients. The pre-germinal-center naive B-cells, which populate the mantle zone of the secondary lymphoid follicles, are thought to be the cells that give origin to MCL. However, overexpression of cyclin D1 alone is not sufficient to cause lymphoma, and a better understanding of the additional molecular lesions may provide insights toward pathogenesis and new therapeutic approaches. In this context, large-scale gene expression studies may be useful in the investigation of such molecular alterations. To address this issue, we intend to compare the gene expression profiles of mantle cell lymphoma cells and normal naive B-cells using oligonucleotide microarrays. Lymphoma cells and naive B-cells (IgD+CD38±CD27-) are isolated by magnetic activated cell sorting, from the peripheral blood of patients with mantle cell lymphoma in the leukemic phase and from tonsils of normal controls. The analyses are performed in replicates using the Amersham CodeLink Human UniSet I Bioarrays with 10,000 genes. Our preliminary results identified 106 genes differentially expressed with a fold change of at least three times, 63 induced and 43 repressed in mantle cell lymphoma in comparison to naive B-cells. Ten genes were selected (6 induced and 4 repressed in lymphoma cells) for quantification by real-time RT-PCR for confirmation of microarray results. We have already stored samples from 21 patients with mantle cell lymphoma in the leukemic phase, as well as from 14 patients with other chronic lymphoproliferative diseases and 7 normal individuals for real-time RT-PCR analysis. Amongst these candidate genes, there are key regulators of the PI3K/AKT1, WNT and TGFβ signaling pathways.
Chronic Lymphocytic Leukemia (CLL) is characterized by the proliferation of mature B-lymphocytes. Several recurrent genetic abnormalities have been described in CLL, however none of them is detected in all cases and a few are associated with the clinical outcome. Recently, abnormalities in the TGFβ , WNT and PI3K/AKT1 pathways were reported. In addition, based on the microarray analysis mentioned above that also suggest that these pathways are involved in the pathogenesis of MCL, we opted to quantify the expression of genes of the WNT and PI3K/AKT1 pathways by Real Time PCR as well as to perform SAGE analysis of CLL cells. We currently have samples from 60 CLL patients and the methods have been established in our lab.
We will also analyze the SPARC gene, which may play a role in lymphoproliferative disorders (candidate genes approach). The SPARC protein was recently associated with the TGF-beta pathway and is involved in cell adhesion, metastasis and matrix metalloproteins production (such as MMP2) in many types of neoplasias. We aim to inactivate the SPARC gene in a MCL cell line using a SPARC antisense construct pCML/SP-AS. The ability of mutants to adhere and migrate will be evaluated in vitro using Matrigel. If the inactivation of SPARC changes the MCL-cell properties, we intend to analyze downstream targets of SPARC using microarray/real time RT-PCR approach. Finally, we also intend to test the effect of halofuginone, a TGF-beta inhibitor, on MCL-cell lines (with and without SPARC inactivation).
A potential target already identified in this project is the tumor-associated antigen PRAME. We have so far demonstrated that PRAME is expressed at higher levels by Mantle Cell Lymphoma and Chronic Lymphocytic Leukemia cells compared with normal peripheral blood lymphocytes (as we have also demonstrated that it is often expressed in squamous cell carcinomas and osteosarcomas). In order to determine if anti-PRAME antibodies might be of potential clinical use, we plan now: a) to demonstrate that it is expressed on the cell membrane, by means of confocal microscopy of normal, MCL and CLL lymphoid cell using the anti-PRAME MoAb and an anti-CD20 commercial Ab; b) to determine its spectrum of expression by flow cytometry in lymphoid organs (spleen, tonsils and lymphnodes), and by immuno-histochemistry on other tissues (liver, kidney, gut, esophagus, heart, brain, testicle, skin, eye, aorta, lung). Additionally, we plan to evaluate in vitro the eventual anti-leukemic effect of the anti PRAME antibody. In order to do that, we will: a) incubate tumour cells (cell lines Granta-19 and K562) with increasing concentrations of anti-PRAME (0;¬†0.02;¬†0.2; 2 and 20¬†¬Ķg/mL) and determine the percentage of apoptotic cells after 24, 48 and 72 h of culture using annexin V and propidium iodide staining; b) quantify the phagocytosis of tumour cells opsonized with the MAb; c) evaluate the ADCC in the presence and absence of antibody, and d) determine complement mediated lysis of tumour cells incubated with the anti PRAME in presence of human serum (inactivated and not inactivated).
Acute Myelogenous Leukemia
Acute myelogenous leukemia (AML) is characterized by the abnormal proliferation of progenitors that present a block of differentiation. Acute promyelocytic leukemia (APL) is a distinct subtype of AML associated with the translocation between chromosomes 15 and 17, involving the PML and RARÔĀ° loci. The fusion protein encoded by the chimeric PML-RARÔĀ° gene retains the main functional domains of the parental proteins and, therefore can: i) bind to retinoic acid (RA) responsive elements as either homodimers or multimeric complexes containing retinoid X receptors (RXR); ii) bind to the ligand (RA) with an affinity comparable to that of wild-type RARÔĀ° and iii) physically interact with native PML delocalizing it from discrete subnuclear structures called nuclear bodies (NB). In absence of ligand, PML-RARÔĀ° and RARÔĀ° affect transcription by forming heterodimers with RXR, which recruit nuclear corepressors (N-CoR) and histone deacetylase (HDAC) complexes, thus altering chromatin structure. Upon addition of physiological concentrations of RA, complexes containing native RARÔĀ° are released and the receptor associates with coactivators. In contrast, the PML-RARÔĀ° containing repressive complexes are stable, due the presence of the coiled-coil region of PML, which cause the formation of PML-RARÔĀ° oligimers. Therefore, PML-RARÔĀ° oncoprotein acts as a constitutive transcriptional repressor, through an epigenetic mechanism.
In collaboration with Prof. Pier Paolo Pandolfi from the Memorial Sloan Kettering Cancer Center from New York, we have been analyzing a transgenic model of APL, in which the PML/RARα cDNA is under the control of the human Cathepsin G (hCG) promoter. The PML/RARα transgenic mice (TM) develop a form of leukemia that closely resembles human APL. However, the disease is developed only by mice older than one year, suggesting that other mutagenic events have to occur prior to diagnosis. We aim to characterize cell cycle, apoptosis and cyclin dependent kinase inhibitors (CDKis) gene expression at different TM age groups in order to determine which abnormalities precede the development of leukemia. We will study the distribution of myeloid progenitors CD117 positive in the different phases of the cell cycle using flow cytometry. Proliferation in vivo of this cell subset will be assessed by measuring BuDU incorporation after i.p. injection. Apoptosis will be induced by gamma irradiation and the percentage of apoptotic myeloid progenitors will be studied by flow cytometry using Annexin V and PI staining. Expression of the p21 and p16 CDKis will be evaluated semi-quantitatively by RT-PCR.
The transgenic model is a useful tool for the development of new therapeutic strategies. We are currently interested on the potential anti-leukemic activity of a derivative of vitamin E, ÔĀ°-tocopherol, which was demonstrated to inhibit cell growth of APL cell lines. Its effects were associated with the inhibition of the NFÔĀęB pathway. We will determine the in vivo effects of ÔĀ°-tocopherol, using the hCG-PML/RARα TM. Leukemic mice will be treated with ÔĀ°-tocopherol alone or in association with the all-trans retinoic acid or arsenic trioxide, which are the currently available therapy for APL. The clearence of leukemia cells, survival and toxicity will be evaluated. Moreover, we plan to analyze the changes in the gene expression profile of leukemic cells treated with ÔĀ°-tocopherol using a macroarray technique already set up in our laboratory.
A second class of drugs under investigation are HDAC inhibitors, which are synergic with RA in the induction of differentiation and may revert RA-resistance in APL patients harboring mutations in the ligand binding domain of PML/RARα. Among the HDACis are trichostatin A (TSA), sodium phenylbutirate (NaB) and valproic acid. We are currently characterizing genes which have their expression affected by RA associated or not with TSA. We have so far analyzed by macroarray the expression of genes associated with cell cycle and apoptosis control in the APL cell line NB4. Based on these results, five genes were selected: GADD153, CDC37, NEDD5, Cyclin D2, RAD23 and their expression will be quantified by real-time PCR in primary APL cells tretaed in vitro with RA with or without TSA.
Between 5-30% of APL patients treated with retinoic acid develop the retinoid syndrome (RS), characterized by respiratory distress, fever, gain of weight, renal function impairment and pleural effusion. RS pathogenesis has been associated with changes in the expression of adhesion molecules, cytokine imbalance and release of cellular enzymes, among others. We aim to study the role of adhesion molecules (AMs) in RS pathogenesis. For that, we will evaluate CD11a, CD11b, CD18, CD29, CD54, CD62L and CD162 expression in leukemic cells from patients with APL and in NB4 cells (APL cell line) treated with retinoic acid, as well as quantify the adhesion of treated cells to Matrigel. Moreover, the effect on AMs expression of new therapeutic agents for APL such as histone deacetylase inhibitors (HDACis) or filgrastim will be evaluated.
Recently, Pandolfi’s group have demonstrated that mouse embryonic fibroblasts in which the Pml gene was inactivated by homologous recombination (Pml -/-) were resistant to TGFβ induced apoptosis and growth arrest. On the other hand, Raza et al. demonstrated up-regulation of TGFβ and prolongation of S phase in bone marrow biopsy samples of 23 APL patients. These data suggest that PML/RARα may induce resistance to apoptosis by repressing native PML and interfering with the TGFβ pathway. We aim to study the role of TGFβ pathway deregulation on APL pathogenesis. We will construct a retrovirus harboring the PML/RARα cDNA under the control of tetracycline responsive element (Tet-on system). Murine hematopoietic progenitors (Sca1+ Lineage-) will be co-transfected with the tet-PML/RARα and Tta retroviruses, in order to generate a conditional model. TGFβ induced apoptosis, growth arrest as well as the activation of downstream targets of TGFβ will be analyzed in presence and absence of PML/RARα expression (controlled by the addition of doxocycline to the media). In addition, the effect of halofunginone, a small molecule inhibitory TGFβ will be assessed.