Abstract C74: Activated RhoA, Rac1 and ERK1/2 participate in chemokinesis triggered by fragmented 23 kDa hyaluronan

Tumors generally possess high levels of hyaluronan (HA), but have a reduced size compared with normal tissue. Experimentally increased HA induces tumor growth and metastasis in xenograft models (review by Toole, BP, 2004). Our study sought to elucidate the possible molecular mechanism of HA‐mediated cell motility in a human epithelial cell line, HeLaS3, using a Boyden chamber system. We showed that among the different molecular masses of HA (3, 23, 230 and 940 kDa) (Seikagaku Kogyo Co.) 23 kDa HA was the most potent in chemokinesis, while 230 and 940 kDa HAs had no detectable effect on chemokinesis. With a drop of 23‐kDa HA, sparsely adhered HeLaS3 cells spread, exhibiting multiple surface extensions in various directions, and altered lamellipodia. HeLaS3 cells exhibited fine cortical F‐actin at the lamellipodia and formed many small focal complexes instead of large focal adhesions at the base of membrane protrusions at 1–20 minutes after 23 kDa HA stimulation. These morphological alterations are reminiscent of Rac1 activated fibroblasts (Rottner, 1999. To properly address the modulation of Rho GTPases in HA‐induced chemokinesis, we evaluated cells carrying dominant negative mutants of either RhoA cDNA (dnRhoA) or Rac1 cDNA (dnRac1). The transduction of either dnRhoA or dnRac1 into HeLaS3 cells caused a dramatic decrease in chemokinesis with or without HA stimulation, compared with mock‐transduced control. To biochemically confirm Rho GTPase activation, we measured GTP‐RhoA and GTP‐Rac1 levels after stimulation with 23 kDa HA using ELISA based Rho GTPases activation assay kits (Cytoskeleton, Inc.). GTP‐Rho A increased after 23kD HA ligation, had a small peak at 1 minute, decreased, and then returned to baseline by 60 min. After 23kD HA stimulation, Rac1 activation was observed in dual peaks of 1.8–2 fold increases at 3 and 15 minutes; thereafter, it became near to baseline by 60 minutes. It has been shown that HA can modulate cell migration involving hyaladherin‐mediated signaling through ERK1/2 (Vigetti, 2008, Tolg, 2006). Initially, we examined the effect of the ERK inhibitor PD98059 in our system. A treatment of 6.6 µg/ ml PD98059 suppressed 23 kDa HA‐induced chemokinesis compared with the control group. Western blotting of the cell lysates showed sustained ERK1/2 phosphorylation over 120 minutes. This study indicates that sustained ERK1/2 phosphorylation is required for 23 kDa HA‐induced chemokinesis. Here we show that fragmented 23 kDa HA elicits enhanced chemokinesis and changes the shape of epithelial cells by weak Rho A activation prior to the recurrent Rac1 activation and the sustained phosphorylation of ERK1/2. Our unpublished study also indicates that hyaluronidase‐2‐mediated catabolism could play a significant role in HA oligosaccharide generation with the association of CD44 in epithelial cells. This indicates the possibility that fragmented HA acts as a mediator in the autocrine/paracrine mechanism. Our study showed that signals initiated by fragmented HA‐hyaladherin interactions induce chemokinesis of epithelial cells, expanding our understanding of the functions of HA in cancer cell biology. Citation Information: Cancer Res 2009;69(23 Suppl):C74.