Review on PDGFB (platelet-derived growth factor beta polypeptide (simian sarcoma viral (v-sis) oncogene homolog)), with data on DNA, on the protein encoded, and where the gene is implicated.

Dermatofibrosarcoma protuberans (DP), an infiltrative skin tumor of intermediate malignancy, presents specific cytogenetic features such as reciprocal translocations t(17;22)(q22;q13.1) or, more often, supernumerary ring chromosomes derived from t(17;22). Different translocations, including t(2;17) and t(X;7), have also been described. We have shown previously that both r(17;22) and t(17;22) present the same molecular rearrangement fusing the COL1A1 gene on chromosome 17 and the PDGFB gene on chromosome 22. Out of our series of 16 DPs, we detected an extra ring chromosome in tumor T96-1175, which juxtaposed sequences from chromosomes 4 and 17. As shown by fluorescence in situ hybridization (FISH) using chromosome painting and alpha-satellite probes, T96-1175 apparently lacked chromosome 22 material in the ring. However, involvement of chromosome 22 through a rearrangement of PDGFB was shown by Southern blotting, reverse transcriptase-polymerase chain reaction (RT-PCR), and FISH. This study demonstrates that a cryptic molecular rearrangement between chromosomes 17 and 22 occurred in addition to the recombination of chromosomes 4 and 17 initially identified by FISH. Assessment for cryptic molecular events should be performed in other variant DP rearrangements.

We have identified a new dermatofibrosarcoma protuberans (DP) case with a t(17;22) (q22;q13) occurring in a child. The translocation was substantiated by the presence of one or two copies of the sole der(22)t(17;22). This rearrangement added to two normal chromosomes 17 and one or two chromosomes 22, resulted in trisomy 22cen-q13 and trisomy (or tetrasomy) 17q22-25. This observation confirms the specificity of the association of DP with the t(17;22) found together with extra copies of the der(22)t(17;22). It also points out a possible prevalence of translocation rather than rings in DP of the childhood disease.

A translocation, t(17;22)(q22;q13), was identified in two cases of dermatofibrosarcoma protuberans (DP). They bring to four the number of DP cases characterized by an identical t(17;22)(q22;q13), which can be considered as a new tumor-associated chromosome rearrangement. To date, this translocation has been found only in DP and its juvenile form, giant-cell fibroblastoma. This finding has two major consequences. First, it casts light on the development and significance in DP of ring chromosomes which consistently harbor sequences derived from chromosomes 17 and 22. Second, the identification of this new chromosome marker, and eventually of the underlying molecular rearrangement, should help to classify DP, a soft-tissue tumor of still uncertain cell origin. In addition, it could be used to differentiate DP from truly benign or malignant entities, in order that this tumor of intermediate malignancy could be adequately managed.

Ring chromosomes have been found with some regularity as solid tumors have come increasingly under cytogenetic study. The full genetic content and significance of these rings remain unclear, Dermatofibrosarcoma protuberans, a tumor of the deep dermis, consistently has supernumerary ring chromosomes, sometimes as the sole detectable cytogenetic change. Using a modified method for comparative genomic hybridization and fluorescent in situ hybridization with a panel of various probes, we found that these ring chromosomes consistently contain the chromosome 22 centromere along with interstitial sequences from chromosomes 17 and 22, specifically from regions 17q23-24 and 22q11-12, The ring chromosomes in dermatofibrosarcoma protuberans are vehicles for a particular pattern of relatively low-level genomic amplification of selected sequences.

We have employed two strategies to map 13 markers located at 11q13. First, we used pulsed-field gel electrophoresis of DNA fragments obtained with methylation-sensitive restriction enzymes. The markers used in this study were scattered over 8.4 Mb and, for most of them, could not be linked one to another. A second mapping strategy employed hybridization to either DNA of somatic hybrids containing various parts of the long arm of chromosome 11 or metaphase chromosomes of a B-cell line containing the t(11;14)(q13;q32) translocation. We were able to sort out the centromeric from the telomeric probes with respect to translocation breakpoints taken as reference chromosomal landmarks by this approach. BCL1, which corresponds to the region where the t(11;14)(q13;q32) translocation breakpoints are clustered, appears as a boundary between two areas of human/mouse homology present in conserved syntenic regions on mouse chromosomes 7 and 19.

Bladder cancer corresponds to a tumor type whose clinical behavior is difficult to predict. A better understanding of this pathology is expected from molecular genetics, which brings together cytogenetics and molecular biology. Therefore, we have tried to overview correlations between chromosome abnormalities and the presence, in the vicinity of the altered loci, of genes (oncogenes and others) that could be involved in bladder oncogenesis and/or tumor progression. In addition to oncogene activation by point mutations, gene amplification, or deregulation of gene expression, several cytogenetic as well as molecular evidences point to genetic deletions (existence of "tumor suppressor genes") being involved in those processes.

In an attempt to probe the significance of HST and INT-2 gene amplification in human breast carcinomas, we have surveyed the amplification status of five molecular markers located on the long arm of chromosome 11 (BCL-1, HST, INT-2 & SEA on 11q13, and ETS-1 on 11q23) in a population of 297 mammary tumors. ETS-1 was rarely amplified and always independently from the other proto-oncogenes. Concerning band q13: (i) 50 tumors (approximately 17%) were co-amplified for BCL-1, HST & INT-2; (ii) in 3 cases, amplification extended to the SEA gene; (iii) in 6 carcinomas, BCL-1 was the only amplified marker. The fact that we never observed amplification of HST & INT-2 independently of BCL-1, which in turn can be amplified solely, suggests the presence, between HST/INT-2 and BCL-1, of a genetic element which could be important in the development of a subset of mammary tumors.

Amplification of c-myc, c-erbB-2, hst and int-2 proto-oncogenes was investigated in two independently collected breast tumor series comprising 292 carcinomas. Differences in the frequencies of amplification could be observed between these two series for c-myc (9.3% vs. 20.8%) and hst/int-2 (21.5% vs. 15.6%) whereas similar values were found for c-erbB-2 (22.5% vs. 20.3%). Statistical correlations between amplification and disease parameters were also dependent on population sampling. Therefore we performed our statistical analysis on the pooled populations and focused on the 219 primary breast carcinomas from patients without therapy prior to surgery. Amplification of c-erbB-2 was strongly correlated to the absence of either estrogen (ER-, P = 0.003) or progesterone (PR-, P = 0.004) receptors. An amplified c-myc was significantly associated with PR- (P = 0.005) and was prevalent in high grade tumors. On the contrary, hst/int-2 amplification was correlated to PR+ tumors (P = 0.01) and was more frequent in ER+ and low grade tumors, and was also correlated with lymph node involvement (P = 0.04). Our data suggest that amplification of each of these proto-oncogenes could be representative of a particular subset of breast tumors. Therefore, proto-oncogene amplification may be helpful in characterizing new biological subclasses in human breast cancer.