Portal de Periódicos da CAPES

Cyclooxygenase‐1, but not ‐2, in blast cells of patients with acute leukemia

2007; Wiley; Volume: 121; Issue: 4 Linguagem: Inglês

10.1002/ijc.22786

1097-0215

Véronique Truffinet, Magali Donnard, Christelle Vincent‐Fabert, Jean Luc Faucher, Dominique Bordessoule, Pascal Turlure, Franck Trimoreau, Yves Denizot,

Estrogen and related hormone effects

Dear Sir, The ability of prostaglandin E2 (PGE2) to regulate the immune system has been widely explored.1, 2 Data reported its ability to regulate several functions in mature blood cells such as T and B lymphocytes, monocyte-macrophages and dendritic cells. PGE2 effects are mediated through interactions with 4 distinct membrane-bound G-protein coupled EP receptors: EP1, EP2, EP3 and EP4.1 EP2 and EP4 are coupled to Gs and stimulate cAMP production, which leads to gene regulation. EP3 are coupled to Gi and inhibit cAMP synthesis. EP1 are coupled to Gq/p and ligand binding induces intracellular calcium level variations. Cyclooxygenases (Cox) catalyse the production of PGE2 from arachidonic acid.3 The Cox-1 isoform is typically constitutively expressed unlike the inducible Cox-2 one. Several studies have reported that the growth-promoting properties of Cox-2 in physiological responses are diverted in malignancies.3 The effects of PGE2 on mature leukocytes have been widely explored.1, 2 PGE2 effects on immature progenitors are poorly documented. PGE2 has been reported to stimulate the growth of human erythroid progenitor cells (BFU-E) and granulocyte-macrophage progenitor cells (CFU-GM) from normal purified CD34+ cells.4 The production and role of PGE2 from leukemic blasts are currently unknown. Our laboratory previously reported the presence of functional EP2 receptors on freshly isolated blast cells from patients with immature forms of acute myeloid leukemia (AML M0–2) and acute lymphoid leukemia (ALL).5, 6 In contrast to EP2 receptors, no functional EP1, EP3 and EP4 receptors were found in freshly isolated AML M0–2 and ALL blasts.6 In view of the potentially important role of PGE2 in processes of cancer and leukocyte maturation and function, we investigated whether leukemic blasts express Cox-1 and Cox-2, produce PGE2 and whether PGE2 affect blast cell proliferation and apoptosis. Over a period of 2 years, blood samples recovered on EDTA were obtained from 22 patients at diagnosis according to the Helsinki recommendations. The population (graded according to the French–American–British classification) consisted of 2 AML M0, 12 AML M1, 4 AML M2 and 4 ALL (3 of B and 1 of T origin, respectively). Leucocytosis ranged from 9.3 to 219 G/l. Blood samples from patients with more than 82% blast cells as circulating leukocytes were used. Blood mononuclear cells were isolated by separation on a Ficoll gradient and washed 2 times with Hank's balanced salts solution (HBSS). The blast purity (>98%) was controlled by flow cytometry analysis (XL II; Beckman Coulter; Margency, France). Blast viability (>95%) was judged by Trypan blue exclusion. The presence of Cox-1 and Cox-2 transcripts was studied by quantitative polymerase chain reaction (Q-PCR) using TaqMan assay reagents (Applied Biosystems, Foster City, CA). The presence of Cox-1 and Cox-2 protein was studied using flow cytometry with fluorescein isothiocyanate (FITC)-labelled mouse anti-human Cox-1 and phycoerythrin (PE)-labelled mouse anti-human Cox-2 antibodies (Becton Dickinson, Pont de Claix, France). The ability of blast cells to produce PGE2 was tested using an EIA kit (Cayman Chemical, Ann Arbor, MI). The effect of PGE2 on blast cell growth was analysed using the CellTiter 96® One Solution Cell Proliferation assay (Promega Corporation, Madison, WI) according to the manufacturer's recommendations. The effect of PGE2 on blast apoptosis rate was determined by flow cytometry after incubation with 7-AAD and FITC-labelled Annexin V antibodies (Becton Dickinson). As reported in Figure 1 (upper panel), Q-PCR analysis revealed Cox-1 and Cox-2 transcripts in blast cells of AML M0–2 and ALL patients. Levels of Cox-1 or Cox-2 transcripts were independent of French–American–British classification status and immunophenotype (such as the presence of CD34, CD33 and CD117 antigens) (data not shown). We, thus, detect Cox-1 and Cox-2 transcripts in immature forms of leukemic blasts. However, it could be argued that the high sensitivity of the PCR procedure might reveal levels of mRNA expression that could be physiologically irrelevant and that the presence of mRNA does not necessarily indicate the presence of the corresponding protein. We, thus, investigated the presence of Cox-1 and Cox-2 protein using flow cytometry analysis. The Cox-1 protein, but not Cox-2, was detected in freshly isolated AML and ALL blasts (Fig. 1, medium panel). Confirming the validity of flow cytometry experiments, and as previously reported,7 Cox-1 and Cox-2 were detected in primary human chronic lymphocytic leukemia (B-CLL) B cells (Fig. 1, medium panel, right part). The present results are concordant with data reporting Cox-1 and Cox-2 transcripts but only the Cox-1 protein in HL-60 cells (a cell line with an AML M2–3 subtype).8 They are also in agreement with the presence of Cox-2 mRNA but not protein in primary bone marrow blasts from acute promyelocytic leukemia,9 and with Cox-1 protein, but not Cox-2, in freshly isolated B cells.7 We next investigated whether the blast Cox-1 protein was functional by testing the ability of leukemic blasts to secrete PGE2. As shown in Figure 1 (lower panel), leukemic blasts spontaneously secreted PGE2 (393.7 ± 169.1 pg, mean ± SEM of 4 experiments) and significantly (p < 0.05, t test for paired data) reduced by 80% their PGE2 synthesis (72.7 ± 46.6 pg) in presence of 1 μM of the Cox inhibitor indomethacin. These results are in agreement with data reporting the Cox-1-derived PGE2 synthesis of CD34+ hemopoietic progenitor cells.10 Finally, we investigated if PGE2 was able to affect leukemic blast physiology by testing PGE2 effect on the growth of leukemic blasts and on their apoptose rate. As shown in Figure 2 (upper panel), PGE2 (from 10 μM to 0.1 μM) significantly (p < 0.05, t test for paired samples) stimulated the spontaneous growth of AML leukemic blasts. PGE2 (1 μM) was also efficient to increase the LPS-induced growth of leukemic blasts. In contrast, PGE2 had no significant effect on spontaneous apoptosis from cultured AML and ALL blasts (Fig. 2, lower panel). Thus, PGE2 stimulated leukemic blast growth in vitro. However, it might be argued that at microM PGE2 concentrations, heterologous activation of the prostacyclin (PGI2) receptor (IP) is possible.11 No data are currently available concerning the presence (and functionality) of IP receptors on freshly isolated AML and ALL leukemic blasts and about the role (if any) of prostacyclin on blast physiology. However, if IP transcripts were not detected in HL-60 cells,12 Q-PCR experiments revealed the presence of IP transcripts in blast cells of AML and ALL patients (Denizot and coworkers, unpublished results) suggesting that the role of PGI2 on AML and ALL blast physiology deserve to be investigated. Detection of Cox-1 and Cox-2 transcripts and protein in AML and ALL blasts. Upper panel: Q-PCR analysis of Cox-1 and Cox-2 transcripts in blast cells of AML M0 (□), AML M1 (○), AML M2 (•), and ALL (▴) patients. Total RNA was extracted with Tripure (Roche GmbH, Mannheim, Germany). Reverse transcription was performed with 2 μg total RNA with the SuperScript reverse transcriptase (Invitrogen, Cergy Pontoise, France). Q-PCR was performed in duplicate by using TaqMan assay reagents (Applied Biosystems, website: http://www.appliedbiosystem.com) (product reference for Cox-1: Hs00277289-s1; product reference for Cox-2: Hs00153133-m1) and analysed on an ABI Prism 7000 system (Applied Biosystems). PCR were performed following the recommendations of the manufacturer. Gene expression levels were normalized to 18S RNA (product reference: Hs99999901-s1). Amounts of Cox-1 transcripts were compared to sample with the lowest level of Cox-1 (an AML M2 patient who was arbitrary quoted 1). Amounts of Cox-2 transcripts were compared to sample with the lowest level of Cox-2 (an ALL patient who was arbitrary quoted 1). Medium panel: Freshly isolated AML and ALL blasts were analysed for intracellular Cox-1 and Cox-2 expression by flow cytometry. Intracellular staining of blast cells was performed immediately after cells isolation. Cells were fixated and permeabilized with the IntraPrep™ permeabilization reagent (Immunotech, Marseille, France) according to the manufacturer's recommendations prior incubation with FITC-labelled mouse anti-human Cox-1 and PE-labelled mouse anti human Cox-2 antibodies (Becton Dickinson) or irrelevant FITC- and PE-labelled mouse antibodies as isotype controls. Cell suspensions were then submitted to flow cytometry analysis (XL II, Coulter Beckman). A representative experiment from blast cells from three AML patients (left panel) and three ALL patients (medium panel) is shown. Primary human chronic lymphocytic leukemia (B-CLL) B cells (right panel) were used as positive controls for Cox-1 and Cox-2 expression. Lower panel: Blasts (5 × 106 cells/ml) were incubated in HBSS in the presence or absence of 1 μM indomethacin (a well known Cox inhibitor) (Sigma) for 2 hr at 37°C. Cell supernatants were harvested by centrifugation and PGE2 levels were measured using a commercially available EIA kit (Cayman Chemical). The sensitivity of the assay permits detection of 15 pg/ml PGE2. Mean ± SEM of 4 independent experiments in duplicate. P < 0.05, t test for paired data. Effect of PGE2 on the growth of leukemic blasts and apoptosis. Upper panel: Left part, AML blast cells (2 × 105 cells) were cultured (in sixplicate) in 96- well plates in 100 μl of growth medium (RPMI 1640 with 10% of fetal calf serum, penicillin (100 U/ ml) and streptomycin (100 μg/ml)) in the presence of PGE2 (from 10 μM to 0.1 μM) or the appropriate vehicle (control). Right part, AML blast cells were cultured in the presence of PGE2 (1 μM) with or without 10 μg lipopolysaccharide (LPS) from Salmonella typhimurium (Sigma) or the appropriate vehicle (control). After 72 hr of growth, the number of viable cells in proliferation was assessed using the CellTiter 96 One Solution Cell Proliferation assay (Promega Corporation) according to the manufacturer's recommendations. Data represent the mean ± SEM of 5 (left part) or 4 (right part) separate experiments in sixplicate. Statistical analysis was made with the student's t test for paired data. Lower panel: Blast cells were cultured in growth medium in 24-well plate for 24 hr alone or in the presence of 1 μM PGE2. To measure the apoptosis rate, cells were washed in HBSS and incubated for 15 min at 4°C prior analysis with 7-AAD and FITC-labelled Annexin V antibodies (Becton Dickinson). Cell viability was determined by flow cytometry. Two representative experiments out of 4 (2 AML and 2 ALL) are shown. In summary, we have shown that freshly isolated leukemic blasts expressed the Cox-1, but not Cox-2, protein and spontaneously secreted the lipid mediator PGE2. A role for PGE2 as compound contributing to AML blast cell proliferation might be hypothesized and should be further explored in more differentiated AML phenotype. Yours sincerely, Véronique Truffinet*, Magali Donnard , Christelle Vincent*, Jean Luc Faucher , Dominique Bordessoule , Pascal Turlure , Franck Trimoreau , Yves Denizot*, * Université de Limoges, Centre National de la Recherche Scientifique, CNRS UMR 6101, Limoges, France, Laboratoire d'Hématologie, CHU Dupuytren, Limoges, France, Service d'Hématologie Clinique et de Thérapie Cellulaire, CHU Dupuytren, Limoges, France.