Effects of HAl-i andHAI-2 on the growth and invasion of human glioblastoma cells

肝细胞生长因子激活因子抑制剂（HAl）最初被确定为肝细胞生长因子激活因子（HGFA）的有效内源性抑制剂，负责肝细胞生长因子/分散因子的激活。目前已鉴定出两种HAI，即HAI-1和HAI-2。两者都是kunitz型丝氨酸蛋白酶抑制剂，不仅抑制HGFA，还抑制其他丝氨酸蛋白酶，如纤溶酶和胰蛋白酶。值得注意的是，每个HAI在羧基末端附近都有一个跨膜结构域，因此，HAI是一种整膜蛋白酶抑制剂，在细胞表面必须具有重要的调节作用。为了探讨HAI-1和HAI-2在肿瘤细胞侵袭性生长中的可能作用，我们分析了实验中两种HAI过表达对高侵袭性恶性脑肿瘤胶质母细胞瘤（GBM）的影响。用含人HAI-1或HAI-2 cDNA的表达载体稳定转染培养的2株人GBM细胞系U251和YKG-1。随后的体外分析表明，HAI-2的过表达显著抑制了两种GBM细胞系的纤溶活性以及对纤维蛋白凝胶的侵袭。另一方面，与HAI-2相比，HAI-1表达对GBM细胞纤溶活性的影响要小得多。然后，我们分析了HAIs对这些细胞系体内侵袭性生长的影响。将GBM细胞注射到裸鼠脑内。注射后8周，处死小鼠并进行尸检，观察肿瘤的发展情况。令人惊讶的是，与对照克隆相比，转染hai -2的克隆在体内形成肿瘤的频率更高，形成的肿瘤更大。这些结果表明，HAI，特别是HAI-2，可能在体内对肿瘤细胞的生存和/或生长具有未确定的重要功能，其依赖或独立于其蛋白酶抑制活性。

Hepatocyte growth factor activator inhibitor (HAl) was initially identified as a potent endogenous inhibitor of hepatocyte growth factor activator (HGFA) that is responsible for the activation of hepatocyte growth factor/scatter factor. To date two kinds of HAI have been identified, namely HAI-1 and HAI-2. Both are Kunitz-type serine proteinase inhibitors and inhibit not only HGFA but also other serine proteinases such as plasmin and trypsin. Notably, each HAI has a transmembrane domain near the carboxyl terminal end, and thus, HAI is an integral-membrane proteinase inhibitor that must have important regulatory role on the cellular surface. To explore the possible role of HAI-1 and HAI-2 in invasive growth of tumor cells, we analyzed the effects of experimental over-expression of each HAI on glioblastoma (GBM), a malignant brain tumor that is highly invasive. Two cultured human GBM cell lines (U251 and YKG-1) were stably transfected with an expression vector harboring human HAI-1 or HAI-2 cDNA. Subsequent in vitro analyses indicated that the over-expression of HAI-2 significantly suppressed the fibrinolytic activity, as well as the invasion into the fibrin gel, of both GBM cell lines. On the other hand, the expression of HAI-1 showed much less effect relative to HAI-2 on the fibrinolytic activity of GBM cells. Then, we analyzed the effects of HAIs on in vivo invasive growth of these cell lines. The GBM cells were injected into the nude mice brain. Eight weeks after the injection, the mice were sacrificed and necropsied in order to determine the development of tumor. Rather surprisingly, HAI-2-transfected clones showed enhanced frequency of the tumor formation in vivo and formed larger tumor than control clones. These results indicated that HAI, particularly HAI-2, might have undetermined important function on the survival and/or growth of tumor cells in vivo, which is dependent or independent of its proteinase-inhibitory activity.