PHC3, a component of the hPRC-H complex, associates with E2F6 during G0 and is lost in osteosarcoma tumors.

Polyhomeotic-like 3 (PHC3) is a ubiquitously expressed member of the polycomb gene family and part of the human polycomb complex hPRC-H. We found that in normal cells PHC3 associated with both hPRC-H complex components and with the transcription factor E2F6. In differentiating and confluent cells, PHC3 and E2F6 showed nuclear colocalization in a punctate pattern that resembled the binding of polycomb bodies to heterochromatin. This punctate pattern was not seen in proliferating cells suggesting that PHC3 may be part of an E2F6-polycomb complex that has been shown to occupy and silence target promoters in G(0). Previous loss of heterozygosity (LoH) analyses had shown that the region containing PHC3 underwent frequent LoH in primary human osteosarcoma tumors. When we examined normal bone and human osteosarcoma tumors, we found loss of PHC3 expression in 36 of 56 osteosarcoma tumors. Sequence analysis revealed that PHC3 was mutated in nine of 15 primary osteosarcoma tumors. These findings suggest that loss of PHC3 may favor tumorigenesis by potentially disrupting the ability of cells to remain in G(0).

Promise and challenge: Markers of prostate cancer detection, diagnosis and prognosis.

Approximately 1 man in 6 will be diagnosed with prostate cancer during his life lifetime, and over 200,000 men in the U.S. are diagnosed with prostate cancer annually. Since the widespread adoption of PSA testing, about 60-70% of men at risk in the U.S. have had a blood test for prostate cancer. With this, prostate cancer death rates have decreased, yet only slightly. Thirty thousand men still die each year from this disease. PSA testing fails to identify a small but significant proportion of aggressive cancers, and only about 30% of men with a "positive" PSA have a positive biopsy. Additionally, of men who are treated for prostate cancer, about 25% require additional treatment, presumably due to disease recurrence. Also of concern is the growing evidence that there are some prostate cancers for which treatment may not be necessary. Very long-term studies from the U.S. and Europe, following men with prostate cancer have found that some tumors do not progress over time. In these individuals, prostate cancer treatment is unnecessary and harmful as these men do not benefit from treatment but will be at risk of treatment-related side effects and complications. They suggest a fundamental problem with prostate cancer: it is not possible, at this time, to predict the natural history of the disease. It is for these reasons that the most important challenge in prostate cancer today is the inability to predict the behavior of an individual tumor in an individual patient. Here we review issues related to performance and validation of biomarkers with a focus on "doing no harm", and bearing in mind that it is the ultimate goal of early detection to save lives. Improved diagnostic and prognostic biomarkers are needed for prostate cancer, and the use of these markers should ultimately translate into increased life span and quality of life. The ultimate goal would be to not only have accurate biomarkers suitable for early diagnosis, but also biomarkers that identify men at greatest risk of developing aggressive disease. Technology has been brought to bear on this problem, and the major approaches are genomics, expression analysis, and proteomics. Proteomics and DNA methylation assays may soon be used in sensitive and specific diagnostic testing of serum and tissues for cancer. Expression arrays may be used to establish both a more specific diagnosis and prognosis for a particular tumor. The proteome is only beginning to be understood, and alternative splicing and post-translational modifications of proteins such as glycosylation and phosphorylation are challenging areas of study. Finally, risk assessment and prognosis are being pursued through analysis of genomic polymorphisms (single nucleotide polymorphisms, SNPs). This huge task is only beginning, and requires the combined expertise of molecular epidemiologists, oncologists, surgeons, pathologists, and basic scientists.

Myotonic dystrophy type 2 caused by a CCTG expansion in intron 1 of ZNF9.

Myotonic dystrophy (DM), the most common form of muscular dystrophy in adults, can be caused by a mutation on either chromosome 19q13 (DM1) or 3q21 (DM2/PROMM). DM1 is caused by a CTG expansion in the 3' untranslated region of the dystrophia myotonica-protein kinase gene (DMPK). Several mechanisms have been invoked to explain how this mutation, which does not alter the protein-coding portion of a gene, causes the specific constellation of clinical features characteristic of DM. We now report that DM2 is caused by a CCTG expansion (mean approximately 5000 repeats) located in intron 1 of the zinc finger protein 9 (ZNF9) gene. Parallels between these mutations indicate that microsatellite expansions in RNA can be pathogenic and cause the multisystemic features of DM1 and DM2.

Localization of the Fanconi anemia complementation group D gene to a 200-kb region on chromosome 3p25.3.

Fanconi anemia (FA) is a rare autosomal recessive disease manifested by bone-marrow failure and an elevated incidence of cancer. Cells taken from patients exhibit spontaneous chromosomal breaks and rearrangements. These breaks and rearrangements are greatly elevated by treatment of FA cells with the use of DNA cross-linking agents. The FA complementation group D gene (FANCD) has previously been localized to chromosome 3p22-26, by use of microcell-mediated chromosome transfer. Here we describe the use of noncomplemented microcell hybrids to identify small overlapping deletions that narrow the FANCD critical region. A 1.2-Mb bacterial-artificial-chromosome (BAC)/P1 contig was constructed, bounded by the marker D3S3691 distally and by the gene ATP2B2 proximally. The contig contains at least 36 genes, including the oxytocin receptor (OXTR), hOGG1, the von Hippel-Lindau tumor-suppressor gene (VHL), and IRAK-2. Both hOGG1 and IRAK-2 were excluded as candidates for FANCD. BACs were then used as probes for FISH analyses, to map the extent of the deletions in four of the noncomplemented microcell hybrid cell lines. A narrow region of common overlapping deletions limits the FANCD critical region to approximately 200 kb. The three candidate genes in this region are TIGR-A004X28, SGC34603, and AA609512.

Identification of a putative regulatory subunit of a calcium-activated potassium channel in the dup(3q) syndrome region and a related sequence on 22q11.2.

Duplication of a segment of the long arm of human chromosome 3 (3q26.3-q27) results in a syndrome characterized by multiple congenital abnormalities and neurological anomalies in some patients. We have identified a novel gene (KCNMB3) that maps to this region. KCNMB3 has significant sequence similarity to the regulatory subunit of the large-conductance calcium-activated potassium channel. Due to the significance of potassium channels in neuronal functions, the overexpression of this gene may play a role in the abnormal neurological functions seen in some of these patients. A related sequence corresponding to the second and third exons of this gene resides in the pericentromeric region of 22q11, where a number of other unprocessed pseudogenes are known to map.

An evolutionary rearrangement of the Xp11.3-11.23 region in 3p21.3, a region frequently deleted in a variety of cancers.

In searching for a tumor suppressor gene in the 3p21.3 region, we isolated two genes, RBM5 and RBM6. Sequence analysis indicated that these genes share similarity. RBM5 and-to a lesser extent-RBM6 also have similarity to DXS8237E at Xp11.3-11.23, which maps less than 20 kb upstream of UBE1. A homologue of UBE1, UBE1L, is located at 3p21. 3. FISH analysis showed that the distance between UBE1L and RBM5 in 3p21.3 is about 265 kb. DXS8237E and UBE1 on the X chromosome have the same orientation, whereas on chromosome 3 the orientation of RBM5 and that of RBM6 are opposite to the orientation of UBE1L. Presumably, part of the Xp11.3-11.23 region has duplicated to chromosome 3. Part of this region on chromosome 3 may subsequently have duplicated again within the same chromosomal region. Inversion at some stage of the evolution of the human genome would explain the change in orientation of the genes on chromosome 3 compared with that of the genes on the X chromosome.

A comparison of genomic structures and expression patterns of two closely related flanking genes in a critical lung cancer region at 3p21.3.

In the search for a tumour suppressor gene in the 3p21.3 region we isolated two genes, RBM5 and RBM6. Gene RBM5 maps to the region which is homozygously deleted in the small cell lung cancer cell line GLC20; RBM6 crosses the telomeric breakpoint of this deletion. Sequence comparison revealed that at the amino acid level both genes show 30% identity. They contain two zinc finger motifs, a bipartite nuclear signal and two RNA binding motifs, suggesting that the proteins for which RBM5 and RBM6 are coding have a DNA/RNA binding function and are located in the nucleus. Northern and Southern analysis did not reveal any abnormalities. By SSCP analysis of 16 lung cancer cell lines we found only in RBM5 a single presumably neutral mutation. By RT-PCR we demonstrated the existence of two alternative splice variants of RBM6, one including and one excluding exon 5, in both normal lung tissue and lung cancer cell lines. Exclusion of exon 5 results in a frameshift which would cause a truncated protein of 520 amino acids instead of 1123 amino acids. In normal lung tissue, the relative amount of the shorter transcript was much greater than that in the lung tumour cell lines, which raises the question whether some tumour suppressor function may be attributed to the derived shorter protein.

Physical and functional mapping of a tumor suppressor locus for renal cell carcinoma within chromosome 3p12.

Using a functional genetic approach, we previously identified a novel genetic locus, NRC-1 (Nonpapillary Renal Cell Carcinoma 1), that mediated tumor suppression and rapid cell death of renal cell carcinoma (RCC) cells in vivo. For these experiments, a defined subchromosomal fragment of human chromosome 3p was transferred into a sporadic RCC cell line via microcell fusion, and microcell hybrid clones were tested for tumorigenicity in vivo. The results indicated functional evidence for a novel tumor suppressor locus within the 3p14-p12 interval known to contain the most common fragile site of the human genome (FRA3B), the FHIT gene, and the breakpoint region associated with the familial form of RCC. We now report the physical mapping of the NRC-1 critical region by detailed microsatellite analyses of novel microcell hybrid clones containing transferred fragments of chromosome 3p in the RCC cell background that were phenotypically suppressed or unsuppressed for tumorigenicity in vivo. The results limit the region containing the tumor suppressor locus within chromosome 3p12. The FHIT gene, FRA3B, and the familial RCC breakpoint region were excluded from the NRC-1 critical region. Furthermore, the NRC-1 locus falls within a well-characterized homozygous deletion region of 5-7 Mb associated with a small cell lung carcinoma cell line, U2020, suggesting that a more general tumor suppressor gene may reside in this region.

Mapping of the gene encoding the regulatory subunit RII alpha of cAMP-dependent protein kinase (locus PRKAR2A) to human chromosome region 3p21.3-p21.2.

We have determined the chromosomal localization of the gene for the regulatory subunit RII alpha of cAMP-dependent protein kinase (locus PRKAR2A) to human chromosome 3 using polymerase chain reaction (PCR) and Southern blot analysis of two different somatic cell hybrid mapping panels. Furthermore, PCR analysis of a chromosome 3 mapping panel revealed the presence of a human RII alpha-specific amplification product only in cell lines containing the region 3p21.3-p21.2. The localization of PRKAR2A was confirmed by PCR mapping using the Stanford G3 Radiation Hybrid Panel as template. The results from this analysis demonstrated that PRKAR2A is most closely linked to D3S3334 (lod score 12.5) and flanked by D3S1322E and D3S1581.

Chromosomal localization of the 5-HT1F receptor gene: no evidence for involvement in response to sumatriptan in migraine patients.

The 5-HT1F receptor, which is present in both human vascular and neuronal tissue, may mediate the therapeutic effect and/or side-effects of sumatriptan. We investigated the chromosomal localization of the 5-HT1F receptor gene and the relation between eventually existing polymorphisms and the clinical response to sumatriptan in migraine patients. The 5-HT1F receptor gene was localized using a monochromosomal mapping panel, followed by a radiation-reduced hybrid mapping and fluorescent in situ hybridization. The results of these techniques show that the 5-HT1F receptor gene is localized at 3p12. We investigated the presence of polymorphisms by single strand conformation polymorphism analysis in 14 migraine patients who consistently responded well to sumatriptan, 12 patients who consistently experienced recurrence of the headache after initial relief, 12 patients with no response to sumatriptan, and in 13 patients who consistently experienced chest symptoms after use of sumatriptan. No polymorphisms were detected in any of the patients. We therefore conclude that genetic diversity of the 5-HT1F receptor gene is most probably not responsible for the variable clinical response to sumatriptan.

Genetic structure and chromosomal mapping of MyD88.

The myeloid differentiation (MyD) marker MyD88 was initially characterized as a primary response gene, upregulated in mouse M1 myeloleukemic cells in response to differentiation induced by interleukin-6. Subsequent analysis revealed that MyD88 possesses a unique modular structure, which consists of an N-terminal "death domain," similar to the intracellular segments of TNF receptor 1 and Fas, and a C-terminal region related to the cytoplasmic domains of the Drosophila morphogen Toll and vertebrate interleukin-1 receptors. In this report we describe the cloning and gene structure of mouse MyD88. The complete coding sequence of mouse MyD88 spans five exons, with the first exon encoding the complete death domain. Zooblot analysis revealed that MyD88 is an evolutionarily conserved gene. MyD88 was localized to the distal region of mouse chromosome 9 by interspecific backcross mapping. The human homolog (hMyD88) was mapped to chromosome 3p22-p21.3 by PCR analysis of a human chromosome 3 somatic cell hybrid mapping panel. Northern blot analysis revealed widespread expression of MyD88 in many adult mouse tissues, and RT-PCR studies detected MyD88 mRNA in T and B cell lines and differentiating embryonic stem cells. The broad expression pattern demonstrates that mouse MyD88 expression is not restricted to cells of myeloid lineage as was originally believed.

Normal FHIT transcripts in renal cell cancer- and lung cancer-derived cell lines, including a cell line with a homozygous deletion in the FRA3B region.

The recently identified FHIT gene encompasses the FRA3B region and the breakpoint of a constitutive t(3;8) occurring in a family with hereditary renal cell cancer. Occurrence of aberrant transcripts in different types of tumours has led to the suggestion that FHIT might play a critical role in the development of various types of cancer. We have analyzed the gene and its transcripts in lung cancers and renal cell cancer-derived cell lines. A lung adenocarcinoma cell line, GLC-A2, appeared to have a homozygous deletion in intron 5 of FHIT. RT-PCR analysis revealed a normal-sized PCR product in all of the cell lines: Including GLC-A2. A number of them had an additional aberrant product. Analysis of a great number of control cell lines and tissues showed that the majority of these also had aberrant PCR products in addition to a normal-sized PCR product. Different specimens of the same cell type showed variable additional RT-PCR products. Normal-sized PCR products had a sequence identical to the FHIT sequence. PCR products longer than normal had insertions of different sizes at different positions. With three exceptions, PCR products shorter than normal represented FHIT sequences missing one or more entire exons. Thus, the presence of aberrant transcripts is not cancer-specific. Conceivably, sequence responsible for the instability of the FRA3B region are being transcribed into FHIT pre-mRNA and may cause the abnormal splicing and processing of the transcripts.

Modulation of disease severity of dystrophic epidermolysis bullosa by a splice site mutation in combination with a missense mutation in the COL7A1 gene.

Dystrophic epidermolysis bullosa (EBD) is a clinically heterogeneous skin disorder, characterized by abnormal anchoring fibrils (AF) and loss of dermal-epidermal adherence. EBD has been linked to the COL7A1 gene at chromosome 3p21 which encodes collagen VII, the major component of the AF. Here we investigated two unrelated EBD families with different clinical phenotypes and novel combinations of recessive and dominant COL7A1 mutations. Both families shared the same recessive heterozygous 14 bp deletion at the exon-intron 115 boundary of the COL7A1 gene. The deletion caused in-frame skipping of exon 115 and the elimination of 29 amino acid residues from the pro-alpha1(VII) polypeptide chain. As a result, procollagen VII was not converted to collagen VII and the C-terminal NC-2 propeptide which is normally removed from the procollagen VII prior to formation of the anchoring fibrils was retained in the skin. All affected individuals also carried missense mutations in exon 73 of COL7A1 which lead to different glycine-to-arginine substitutions in the triple-helical domain of collagen VII. Combination of the deletion mutation with a G2009R substitution resulted in a mild phenotype. In contrast, combination of the deletion with a G2043R substitution led to a severe phenotype. The G2043R substitution was a de novo mutation which alone caused a mild phenotype. Thus, different combinations of dominant and recessive COL7A1 mutations can modulate disease activity of EBD and alter the clinical presentation of the patients.

Assignment of the human peroxisomal branched-chain acyl-CoA oxidase gene to chromosome 3p21.1-p14.2 by rodent/human somatic cell hybridization.

PCR and rodent/human somatic cell hybrids were used to localize the human peroxisomal branched-chain acyl-CoA oxidase gene. Oligonucleotide primers were chosen to specifically amplify human hBCox DNA. The amplified sequence contained two restriction enzyme sites which were used to verify the authenticity of the amplified DNA. Initially, the gene was localized to human chromosome 3 by screening genomic DNA from a hybrid mapping panel. Additional hybrids retaining well-characterized fragments of human chromosome 3 were screened to further localize the gene to 3p21.1.p14.2.

Deletions of the short arm of chromosome 3 in solid tumors and the search for suppressor genes.

The concept that cells can become malignant upon the elimination of parts of chromosomes inhibiting cell division dates back to Boveri in 1914. Deletions occurring in tumor cells are therefore considered a first indication of possible locations of tumor suppressor gene. Approaches used to localize and identify the paradigm of tumor suppressors, RB1, have also been applied to localize tumor suppressor genes on 3p, the short arm of chromosome 3. This review discusses the methodological advantages and limitations of the various approaches. From a review of the literature on losses of 3p in different types of solid tumors it appears that some tumor types show involvement of the same region, while between others the regions involved clearly differ. Also discussed are results of functional assays of tumor suppression by transfer of part of chromosome 3 into tumor cell lines. The likelihood that a common region of deletions would contain a tumor suppressor is strongly enhanced by coincidence of that region with a chromosome fragment suppressing tumorigenicity upon introduction in tumor cells. Such a situation exists for a region in 3p21.3 as well as for one or more in 3p12-p14. The former region is considered the location of a lung cancer suppressor. The same gene or a different one in the same region may also play a role in the development of other cancers including renal cell cancer. In the latter cancer, there may be additional roles of the VHL region and/or a 3p12-p14 region. The breakpoint region of a t(3;8) originally found to be constitutively present in a family with hereditary renal cell cancer now seems to be excluded from such a role. Specific genes on 3p have been suggested to act as suppressor genes based on either their location in a common deletion region, a markedly reduced expression or presence of aberrant transcripts, their capacity to suppress tumorigenicity upon transfection in to tumor cells, the presumed function of the gene product, or a combination of several of these criteria. A number of genes are evaluated for their possible role as a tumor suppressor according to these criteria. General agreement on such a role seems to exist only for VHL. Though hMLH1 plays an obvious role in the development of specific mismatch repair-deficient cancers, it cannot revert the tumor phenotype and therefore cannot be considered a proper tumor suppressor. The involvement of VHL and MLH1 also in some specific hereditary cancers allowed to successfully apply linkage analysis for their localization. TGFBR2 might well have a tumor suppressor function. It does reduce tumorigenicity upon transfection. Other 3p genes coding for receptor proteins THRB and RARB, are unlikely candidates for tumor suppression. Present observations on a possible association of FHIT with tumor development leave a number of questions unanswered, so that provisionally it cannot be considered a tumor suppressor. Regions that have been identified as crucial in solid tumor development appear to be at the edge of synteny blocks that have been rearranged through the chromosome evolution which led to the formation of human chromosome 3. Although this may merely represent a chance occurrence, it might also reflect areas of genomic instability.

Localization of a novel tumor suppressor locus on human chromosome 3q important in osteosarcoma tumorigenesis.

Mitotic recombination org nondysjunction are common mechanism for tumor-specific loss of constitutional heterozyosity (LOH) and tumor suppressor allelic inactivation and can be useful in localizing new putative tumor suppressor genes. In osteosarcoma, the highest frequencies of LOH have been reported for chromosomes 3q, 13q, 17p, and 18q. The high incidence of LOH on chromosome 3q suggests the presence of a novel tumor suppressor gene. To localize this putative tumor suppressor gene, we have used polymorphic markers on chromosome 3q to define the minimal region in which mitotic recombination or deletion results in LOH, which should contain the tumor suppressor gene. This putative tumor suppressor has been localized to a region between 3q26.2-3q26.3 of less that 1 cM between the polymorphic loci D3S1212 and D3S1246.

Assignment of the gene for human tetranectin (TNA) to chromosome 3p22-->p21.3 by somatic cell hybrid mapping.

Tetranectin is a plasminogen-binding protein that is induced during the mineralization phase of osteogenesis. By screening a human chromosome 3 somatic cell hybrid mapping panel, we have localized the human tetranectin gene (TNA) to 3p22-->p21.3, which is distinct from the loci of two human connective tissue disorders that map to the short arm of chromosome 3, MFS2 and LRS1.

An 80 Kb P1 clone from chromosome 3p21.3 suppresses tumor growth in vivo.

High frequencies of allelic loss on the short arm of chromosome 3 in small cell lung cancer (SCLC) and a number of other tumors suggest the existence of a tumor suppressor gene(s) within the deleted regions. Two small cell lung cancer lines, NCI H740 and GLC20, have been described which have homozygous deletions in the region 3p21.3. The deleted region overlaps with a 2 Mb fragment of human DNA present in the interspecies hybrid HA(3)BB9F, that suppresses tumor formation by mouse A9 fibrosarcoma cells. Human sequences from this cell hybrid were isolated using inter Alu PCR. From this starting point, a P1 contig was developed for the region of 450 Kb that is common to the homozygous deletions seen in the SCLC lines NCI H740 and GLC20 and is also present in HA(3)BB9F, the suppressed A9 hybrid. Individual P1 clones were assayed for their ability to suppress the tumorigenicity of the mouse fibrosarcoma cell line A9 as assayed by injection of transfected A9 cells into athymic nude mice. The introduction of one of the P1 clones into A9 cells resulted in suppression of tumor growth whereas two other P1 clones from the contig failed to suppress tumor formation in athymic nude mice. These data functionally delimit a tumor suppressor locus to a region of 80 kb within a P1 clone at 3p21.3.

Linkage and cytogenetic mapping of a CAG repeat containing human cDNA to 3p24.2-p22.

A trinucleotide (CAG)n repeat containing cDNA was isolated from a human cDNA library and sequenced. The locus was mapped by linkage analysis in the CEPH families and by cytogenetic analysis to 3p24.2-p22. We have additionally excluded this gene as a candidate for small cell lung carcinoma by the analysis of cell lines carrying homozygous deletions for the 3p chromosomal region.