Characterization of novel tumor suppressor genes, DLC-1 and DLC-2, in hepatocellular carcinoma

by Wong, Chun-ming

(Uncorrected OCR) Abstract of thesis entitled

Characterization of Novel Tumor Suppressor Genes, DLC-l and

DLC-2, in Hepatocellular Carcinoma

Submitted by

Wong Chun-Ming

For the degree of Doctor of Philosophy at the University of Hong Kong

in December 2003

Hepatocellular carcinoma (lICC) is one of the most common malignancies in the world. Although the risk factors are well known, the underlying molecular mechanisms contributing to hepatocarcinogenesis are far from clear. Deletions of chromosomal materials are the most frequent genetic alteration found in HCC. Allelic losses in HCC show a non-random pattern, and chromosomes 8p and 13q are two of the most susceptible chromosomal arms. However, critical regions of loss and putative suppressor genes on 8p and 13q have not been firmly identified.

Our experimental data indicated that allelic losses on chromosomes 8p and 13q were frequent in HCC and were associated with aggressive tumor phenotypes. In addition, our results showed recurrent losses on the sub-chromosomal regions 8p21.3-22 and 13q12.3. Such recurrent losses suggest the presence of putative tumor suppressor genes, loss of which may be important to HCC development and progression.

In search of candidate tumor suppressor genes in these regions, we identified a novel gene DLC-2 (deleted in liver cancer 2) at 13q12.3. DLC-2 shared a high sequence

homology with putative tumor suppressor gene DLC-l at 8p21.3-22. Both DLC-l and DLC-2 are GTPase activating proteins that activate the intrinsic GTPase activity of RhoA and Cdc42 and switch them offby converting the active GTP-boundstate to the inactive GDP-bound state. Several lines of evidence suggest that aberrant activation of Rho proteins is oncogenic. Thus, it is not surprising that DLC-l and DLC-2 may function as a tumor suppressor.

Although we rarely found somatic mutations in the RhoGAP domain of DLC-l or DLC-2 mRNA, deletion of DLC-l or DLC-2 locus as revealed by loss of heterozygosity analysis was frequent in human HCC. In addition, DLC-l and DLC-2 were significantly underexpressed at mRNA levels in primary HCC, suggesting that loss of DLC-l and DLC-2 functions as a result of downregulated mRNA expression might contribute to hepatocarcinogenesis. Furthermore, DLC-l promoter methylation was found in 6 (24%) of 25 primary HCCs and in hepatoma cell lines showing loss of DLC-l mRNA expression. Our results indicated that both genetic and epigenetic alterations play an important role in inactivation of the DLC-l gene in hepatocarcinogenesis.

We also characterized the tumor suppression function of DLC-l and DLC-2. We noticed that colony formation was significantly inhibited by transient transfection of DLC-l cDNA into hepatoma cell lines with no DLC-l expression. Furthermore, stable expression of DLC-l mRNA successfully suppressed the proliferation, mitogenindependent growth, and anchorage-independent growth capabilities of SMMC-7721 cells. On the other hand, expression of the GAP domain of DLC-2 in mouse fibroblasts sufficiently repressed Ras signaling and Ras-induced cellular

transformation. Overall, our findings suggest that DLC-l and DLC-2 play an important role in hepatocarcinogenesis.

An abstract of 439 words