The glycan-binding protein galectin-1 links hypoxia and angiogenesis in Kaposi Sarcoma

DIEGO O CROCI; NATALIA RUBISTEIN; MARIANA SALATINO; JUAN M ILARREGUI; MARTA A TOSCANO; GERMAN BIANCO; LAURA PALEARI; ADRIANA ALBINI; GABRIEL A RABINOVICH

Denver, USA

Congreso; 100th Annual Meeting of the American Association for Cancer Research; 2009

AACR

body { background: #FFFFFF; margin: 0px; padding: 0px; } Angiogenesis is a critical process for tumor progression. We previously demonstrated that galectin-1 (Gal-1), a  glycan-binding protein, controls tumor growth by favoring escape from T-cell-dependent immunity.The aim of this study was to investigate, using in vitro and in vivo strategies, the role of Gal-1 in the control of tumor angiogenesis and to analyze the mechanisms underlying this effect.To examine the contribution of Gal-1 to tumor angiogenesis in vivo we inhibited Gal-1 expression by silencing-gene strategies in human Kaposi sarcoma tumor cell lines. Targeted inhibition of gal-1 gene expression inhibited the formation of new blood vessels and suppressed tumor growth in vivo (p<0,05 , p<0,01 knockdown vs wild-type). Expression of Gal-1 in Kaposi cells was found to be up-regulated under hypoxic conditions, both in vitro (as evaluated by WB, Real Time RT-PCR and promoter activity) and in vivo (IHC). This over expression was dependent  of NFêB but not HIF-1á activity, since inhibitor of IêB phosphorylation BAY 11-7082 (1.5 ìM) but not Hif-1á (2ìM) prevented hypoxia-induced expression of Gal-1 (p<0.05). Moreover, conditioned media from KS cells (CMKS) incubated 16 h under hypoxic conditions (O2 1 KPa) were more effective at inducing tubulogenesis p<0.01) and invasion (p<0.05) than normoxic cells. Interestingly, this effect was abrogated when KS hypoxic cells were infected with a retrovirus encoding shRNA to gal-1(p<0.01). This result was confirmed in vivo using a matrigel sponges model in which hypoxic MC-KS were more effective to induce hemoglobin accumulation in matrigel plugs (p<0.05) hypoxic MC-KS vs normoxic MC-KS). Remarkably, the presence of shRNA-gal-1 significantly reduced accumulation of hemoglobin in matrigel plugs either in normoxic or hypoxic conditions (p<0.05 vs empty vector).We next addressed the effect of Gal-1 on endothelial cells. Recombinant Gal-1 bound human umbilical vein endothelial cells (HUVEC) in a dose and carbohydrate-dependent manner (p<0.01). Moreover, treatment of HUVEC with recombinant Gal-1 promoted tubulogenesis, (p<0.01) proliferation (p<0.05) and invasion (p<0.05). A mechanistic analysis of the intracellular signals involved in this effects revealed a substantial increase in phosphorylated-Akt and Erk ½ following Gal-1 treatment (F.I>3). Pretreatment of HUVEC with either PI3Kinhibitor LY294.002 (2ìM) or Erk1/2-inhibitor U0126 (5ìM) but not NFêB, p38 or JNKs inhibitors, abolished the ability of Gal-1 to induce tubulogenesis (p<0.01), proliferation (p<0.05) and invasion (p<0.05). Our results suggest that hypoxia-regulated Gal-1 is a key modulator of tumor cell angiogenesis by promoting tubulogenesis, proliferation and invasion of endothelial cells. In addition, we propose a model in which Gal-1regulated by hypoxia is secreted to the extracellular milieu to regulate angiogenesis and tumor-immune escape, thus influencing tumor progression.in vitro and in vivo strategies, the role of Gal-1 in the control of tumor angiogenesis and to analyze the mechanisms underlying this effect.To examine the contribution of Gal-1 to tumor angiogenesis in vivo we inhibited Gal-1 expression by silencing-gene strategies in human Kaposi sarcoma tumor cell lines. Targeted inhibition of gal-1 gene expression inhibited the formation of new blood vessels and suppressed tumor growth in vivo (p<0,05 , p<0,01 knockdown vs wild-type). Expression of Gal-1 in Kaposi cells was found to be up-regulated under hypoxic conditions, both in vitro (as evaluated by WB, Real Time RT-PCR and promoter activity) and in vivo (IHC). This over expression was dependent  of NFêB but not HIF-1á activity, since inhibitor of IêB phosphorylation BAY 11-7082 (1.5 ìM) but not Hif-1á (2ìM) prevented hypoxia-induced expression of Gal-1 (p<0.05). Moreover, conditioned media from KS cells (CMKS) incubated 16 h under hypoxic conditions (O2 1 KPa) were more effective at inducing tubulogenesis p<0.01) and invasion (p<0.05) than normoxic cells. Interestingly, this effect was abrogated when KS hypoxic cells were infected with a retrovirus encoding shRNA to gal-1(p<0.01). This result was confirmed in vivo using a matrigel sponges model in which hypoxic MC-KS were more effective to induce hemoglobin accumulation in matrigel plugs (p<0.05) hypoxic MC-KS vs normoxic MC-KS). Remarkably, the presence of shRNA-gal-1 significantly reduced accumulation of hemoglobin in matrigel plugs either in normoxic or hypoxic conditions (p<0.05 vs empty vector).We next addressed the effect of Gal-1 on endothelial cells. Recombinant Gal-1 bound human umbilical vein endothelial cells (HUVEC) in a dose and carbohydrate-dependent manner (p<0.01). Moreover, treatment of HUVEC with recombinant Gal-1 promoted tubulogenesis, (p<0.01) proliferation (p<0.05) and invasion (p<0.05). A mechanistic analysis of the intracellular signals involved in this effects revealed a substantial increase in phosphorylated-Akt and Erk ½ following Gal-1 treatment (F.I>3). Pretreatment of HUVEC with either PI3Kinhibitor LY294.002 (2ìM) or Erk1/2-inhibitor U0126 (5ìM) but not NFêB, p38 or JNKs inhibitors, abolished the ability of Gal-1 to induce tubulogenesis (p<0.01), proliferation (p<0.05) and invasion (p<0.05). Our results suggest that hypoxia-regulated Gal-1 is a key modulator of tumor cell angiogenesis by promoting tubulogenesis, proliferation and invasion of endothelial cells. In addition, we propose a model in which Gal-1regulated by hypoxia is secreted to the extracellular milieu to regulate angiogenesis and tumor-immune escape, thus influencing tumor progression.