Specific Genetic Change During EAC Development

The next step, then, will be to characterize specific genetic changes occurring in the chromosomal regions identified. Since each region is quite large, this work is a substantial undertaking that is open to several different approaches. Here, we will discuss some of our results from two ofthe chromosome regions, situated on RNO4 and RNO10, respectively. Aberrations in the proximal part of RNO4 were the most common among those detectable with CGH, and gains in RNO4 occurred in 34 out of 56 tumors (61%). In 12 cases these increases were designated as gene amplification, since the number of copies gained was in excess of five. For the analysis of the proximal part of RNO4 we started by constructing a detailed physical map ofthe region (9). Subsequently, a subset of 12 tumors was subjected to analysis with FISH and Southern blotting (10). Fifteen gene probes covering the proximal 75 Mb of RNO4 (Table 3; the total length of RNO4 is presently estimated at 193 Mb) were tested in Southern blot analysis and it was shown that in 6 ofthe 12 tumors there was amplification of subsets ofthe genes (Figure 1). Notably, both the Met oncogene and the cell cycle regulating Cdk6 gene were amplified in 5 tumors. These two genes are located only about 20 Mb apart (Table 3). Still, it was found that the amplified copies of the two genes were situated in different HSR chromosomes in some ofthe tumors, making it obvious that, at least in these cases, each gene had been subjected to an independent amplification event. Furthermore, two genes located between Cdk6 and Met in native rat DNA (Tacl and Asns) were never amplified, in fact, in some cases there was actually a reduction in copy number of these genes. Our conclusion from these findings was that increased levels of Cdk6 and Met gene products somehow interacted to stimulate tumor development. Clearly, stimulation by such interaction is just one of several possibilities open in BDII EAC development, since coamplification of Cdk6 and Met was seen only in a subset of the tumors (5 out of 12 tumors analyzed). In addition, these results strongly suggest that an important tumor suppressor gene (TSG) must be situated somewhere in the interval between the two genes, since this region was never included in the amplicons.

Copy number dam

DICDipaitN UCDI'tHIJ ««tCfFOIl J tlCïEFSRI J llEbllSHlJ ApCfftfÇHIJ

Copy number dam

DICDipaitN UCDI'tHIJ ««tCfFOIl J tlCïEFSRI J llEbllSHlJ ApCfftfÇHIJ

RN04

Figure 1. On the left, the ideogram of RNO4 is shown, including higher magnification of the proximal part. The cytogenetic localization of the 15 genes analyzed is indicated. The notches on the thin line immediately to the right of the ideogram shows the approximate physical positions ofthe genes (not completely to scale, for details see Table 4). The diagrams in the right part ofthe figure show the copy number level for each studied locus in the six tumors (circles connected by dashed lines). Copy number classes refer to the number of copies per diploid chromosome set equivalent (DCSE), thus, when there is no copy number change this value will be 2, when there is amplification the value should be >5. A value <1.5 is considered to be indicative of copy number reduction (deletion). Division into classes: A = <1; B = 1-1.5; C = 1.6-4; D = 5-10; E = 11-25; F = 26-40; G = 4155; H = 56-70; I = 71-90; K = >90 copies/DCSE, ofwhich class C represents the normal range. The diagrams generated resemble the patterns seen in the CGH analysis, but give much greater detail. Note that there are some regions, which actually show reduction in copy numbers (shaded) suggestive of chromosomal deletion, perhaps occurring during the actual amplification process. The data strongly suggest the involvement of two distinct targets for gene amplification on RNO4 in these tumors.. [Reprinted by permission of Cancer Res (10)].

Even though all available indications were that Cdk6 and Met were the targets of amplification, obviously we cannot be sure of this, since in the entire 75 Mb region 461 genes have already been predicted in Ensembl based on structural criteria, and if one extrapolates available data to an estimated total of about 33000 genes in the rat, this region should harbor about 715 genes. Still, when this entire

DNA becomes fully annotated, it seems that the number ofpossible candidate genes in the region will be quite limited, and that it will be a reasonable task to determine which ones among them are involved in EAC development, using available standard molecular methodology.

Table 3. Cytogenetic and Draft DNA Sequence Position of 15 Cancer-related Genes in the Proximal Part of RN04 (see Figure 1). The present estimate is that this region contains about 715 genes (extrapolated from the Ensembl database).

Gene Symbol

Gene Description

Cytogenetic Position

DNA Sequence Position (Mb)

Cdk5

Cyclin-dependent kinase 5

4qlL2

5.0

Hgf

Hepatocytc growth factor

4ql2,l

14.6

(scatter factor)

Dtmfl

Cyclin D binding myb-like

4ql2-13

21.6

transcription factor 1

Abcbl

ATP-binding cassette, sub-family B

4ql2-13

22.0

(MDR/PGY), member 1

CypSl

Cytochrome P450, 51

4ql3

26.7

(lanosterol 14-alpha-demethylase)

Cdk6

Cyclin-dependent kinase 6

4ql3

27.5

Tad

Tachykinin, precursor 1 (substance P)

4q 13-21

37.3

Asm

Asparagine synthetase

4ql3-21

37.4

Cavl

Caveolin 1

4q21,1-21.2

47.7

Met

MET proto-oncogene

4q21.2

47.8

(hepatocyte growth factor receptor)

Wnt2

Wingless-type MMTV

4q21.3-21.31

48.6

integration site family member 2

Cfir

Cystic fibrosis

4q21.3-21.31

48.9

transmembrane conductance regulator

Smoh

Smoothened (Drosophila) homolog

4q22

60.9

Braf

V-raf murine sarcoma

4q22-23

71.4

viral oncogene homolog B1

Arhgeß

Rho guanine nucleotide exchange

4q23.3

75.6

factor (GEF) 5, TIM proto-oncogene

In one of the tumors (RUT7) there was amplification only of the Met oncogene. A possible explanation was obtained from expression studies in the 12 tumors. The gene product of the Met oncogene is the hepatocyte growth factor receptor, and in RUT7 there was expression of the corresponding ligand (Hgf, hepatocyte growth factor). The Hgf gene is normally turned off in endometrial tissue - the fact that it was activated in RUT7 suggests that in this tumor a different pathway involving an autocrine loop stimulating cell growth was active. Taken together, our findings suggest that there are several genes in the proximal part of RNO4 that play important roles in BDII EAC development. These genes probably include Met, Cdk6 and Hgf and at least one still undiscovered tumor suppressor gene, which is located not far from the Tac1/Asns genes.

In analyzing genetic changes in RNOIO we took a slightly different approach. Because indications from the CGH analysis were that there was loss of chromosomal material in the proximal part of RNOIO, we started out with an analysis of allelic imbalance looking for signs of LOH (loss of heterozygosity) (11, 12). We found indications that allelic imbalance was particularly common in at least four rather well-defined regions. One region, situated in RNO10q24, was shown to be affected by LOH in virtually all informative tumors, and in this region the Tp53 gene is an obvious target candidate gene. Two other regions exhibiting recurrent allelic imbalance were located in 10q11-q12 and in 10q22, but here there were no obvious specific candidate suppressor genes. It is known that RNO10, which is mostly homologous to human chromosome (HSA) 17, contains many cancer-related genes. We are currently in the process of scrutinizing the RNO10 draft sequence in search of potential candidate genes in these regions. A fourth region showing allelic imbalance was pinpointed in 10q32.1, but the CGH study had shown that in BDII EAC tumors there was copy number increase rather than decrease in this region. Thus, we will be looking for activated oncogenes rather than suppressor genes in this particular region. One thing was very clear: ifthere is a relative copy number decrease in the proximal part of RNO10, and a copy number increase in the distal part there must, by necessity, occur chromosomal breaks within a rather specific segment between the two regions. To examine this notion we have performed FISH studies using an RNO10 chromosome paint in conjunction with gene-specific probes (13). Thirty EACs were subjected to cytogenetic analysis, which was combined with RNO10-specific painting. It was found that the chromosome numbers of EAC tumors were in the diploid or triploid regions, modal numbers ranging between 3 9 and 71. The RNO10 chromosome paint revealed that the number of chromosomes entirely made up of RNO10 material or, in a few instances, translocation chromosomes with distinct RNO10 segments, in 16 ofthe tumors was equal to that expected from ploidy (i.e., two in diploid and three in triploid cells). Fourteen EACs had a total of 20 RNO10-derived chromosome segments in excess of those expected from ploidy. It should be mentioned that in most tumors only 1-2 normal-looking RNO10 chromosomes were present, additional RNO10-derived segments were mostly smaller than ordinary RNO 10 chromosomes.

To further analyze the RNOP10 involvement, two gene probes from different parts of the chromosome were used in dual-color FISH in order to determine which chromosome segments the smaller RNO10-derived elements represented. Thus, one PAC (P1 artificial chromosome) probe corresponding to Tp53 (at 10q24) and another PAC probe for the Thral gene (at 10q32.1) were labeled with red and green fluorescence, respectively, and the two probes were simultaneously hybridized to metaphase preparations from the EAC cultures. The hybridization pattern was compared to the previously determined pattern of RNO10-derived chromosomes. Based on the combined analysis it was found that EAC tumors neatly fall into two separate groups, those with no cytogenetic involvement of RNO10 (11/30 tumors; 37%) and those exhibiting deleted and/or translocated RNO10-derived chromosomes (19/30 tumors; 63%). In the latter subgroup, the analysis showed that 35 out of 37 deletion or translocation chromosomes had arisen after a break in the region between bands 10q24 and 10q32.1, leading to a situation in which they contained either Tp53 or Thral, but never both (the remaining two rearranged chromosomes were small and contained neither Tp53 nor Thral). Furthermore, among the rearranged chromosomes those derived from the distal part and containing Thral dominated (30/35 chromosomes; 86%). Thus, this analysis showed that compared to the number of copies expected from the ploidy, there was a relative deficit of the proximal part of RNO10 and a relative overrepresentation of the distal part, just as would be expected based on the CGH-results. In what way this imbalance in copy number affects EAC development remains to be determined. Based on the findings from the allelic imbalance analysis the possibility that one or more TSG loci need to be inactivated in the proximal part seems valid. Similarly, increases in the distal part may lead to overrepresentation of an oncogene. However, an interesting additional possibility is that a cancer-related gene located at the chromosomal break point is of importance in EAC development in this EAC subgroup. Only further experimentation can lead to a distinction between the different possibilities.

Was this article helpful?

0 0

Post a comment