Development of new approaches for tumor suppressor genes identification

Abstract

Development of tumor is a complex process involving multiple steps. New technologies for cloning and identifying genes playing critical role in cancer development are necessary. That is why we have focused our research on development of these approaches.

The new methods include CIS, cloning of identical sequences, COP, cloning of polymorphic sequences and CODE, cloning of deleted sequences. Although these methods are based on the same combination of biochemical techniques their aims are different. These methods are fully complementary; therefore they may be applied together to analyse a given object. If one aims to clone a disease gene responsible for familial cancer syndrome, these methods may be applied as follows. CIS may be used to identify the sequences identical by descent comparing the DNA obtained from affected family members. COP may be used to find sequences that are different between affected members, and CODE would be useful to compare tumor and normal (control) samples to isolate deleted sequences (putative candidate tumor suppressor genes) and amplified sequences (putative oncogenes). COP and CODE procedures may be applied to analyse the CpG islands thus allowing direct candidate gene identification.

NotI microarrays are the microarrays giving the opportunity to detect copy number and methylation changes. NotI microarrays are based on large-scale sequencing of total human NotI linking clones, which were previously shown to be tightly associated with CpG islands and genes. We have solved the main problems for genome wide screening created by the size of human genome and numerous repeat sequences by developing a new method for labelling genomic DNA where only sequences surrounding NotI sites are labelled, Nod representation (NR). A pilot experiment using NR probes demonstrated the power of the method, and we successfully detected Chr 3 NotI clones deleted in ACC-LC5 and MCH939.2 cell lines. NotI arrays will speed up cancer research very significantly and can replace CGH, LOH and many cytogenetic studies, since the high- density grids with 50.000 NotI linking clones were constructed, 22 551 unique Notl flanking sequences were generated, covering a total of 16.2 Mb of the human genome.

The candidate tumor suppressor genes (TSGs) cloned by above new methods will be entered into gene inactivation test (GIT), which is a new functional test system for TSGs identification. GIT is based on our hypothesis that TSG must be inactivated in growing tumors in experimental conditions as it happened in nature; this inactivation of a TSG can be achieved by mutation, deletion, methylation etc. To verify our hypothesis, known suppressor genes RB and p53 were built into pETI and pETE vectors that permitted tetracycline/doxycycline regulated expression of the cloned genes in cancer cell lines growing not only in vitro, but in vivo as well. These cell lines are tTA producing cell lines. Wild type but not mutated RB and p53 genes were deleted/inactivated during tumor growth in SCID mice.

Furthermore, no inactivation/deletion was observed for 3PK, MLH1, rhoA genes even after two passages in SCID mice. The two multiple cancer regions (3p21.3T and 3p21.3C regions) were identified in lung cancer and kidney cancer. The smallest overlapping homozygous deletion (app. 100 kb) in 3p21.3C region includes 8 genes. All these genes were included in functional gene inactivation test. One gene from homozygous deletion 3p21.3T region also was included in GIT. Until now, RASSF1A, RASSF1C, Gene21, SEMA3B and CACNA2D2 were shown to have growth suppression activity in vitro and in vivo, and were inactivated in the tumors following SCID mice passage. The results suggested that these genes play important role in the lung and kidney pathogenesis.