Lumican exhibits anti-angiogenic activity in a context specific manner.

A series of overexpression studies have shown that lumican suppresses angiogenesis in tumors produced from pancreatic adenocarcinoma, fibrosarcoma, and melanoma tumor cells. Despite lumican's anti-angiogenic activity, a clear correlation of differential expression of lumican in various cancers and cancer malignancy has failed to emerge. Therefore, we hypothesized that either 1.) endogenously expressed lumican is not anti-angiogenic or alternatively that 2.) lumican exhibits angiostatic activity only in limited microenvironments. Previously, lumican was shown to suppress tumor growth and angiogenesis in subcutaneously injected PanO2 pancreatic adenocarcinoma cells. Therefore, to determine if endogenously expressed lumican is anti-angiogenic we subcutaneously injected PanO2 cells into wild-type and lumican knockout mice and compared tumor growth and vascular densities of the resulting tumors. We found that tumors grown in lumican knockout animals were larger and contained significantly elevated vascular densities compared to those grown in wild-type mice. Interestingly however lumican knockout animals did not exhibit enhanced angiogenesis in aortic ring assays, matrigel plugs, or healing wound biopsies raising the possibility that lumican suppresses angiogenesis only in tumor microenvironments. To test this possibility, we sought a tumor model wherein lumican did not exhibit anti-angiogenic activity. Utilizing the 4T1 breast cancer model, we found that lumican suppressed 4T1 tumor growth and lung metastasis, but not angiogenesis. In conclusion, these results show that the angiostatic activity of lumican is dependent on currently undefined microenvironmental cues and therefore helps to understand why differential expression of lumican does not consistently correlate with human tumor malignancy.