No abstract

Summary. Acute promyelocytic leukaemia (APL) has been associated with a favourable prognosis in many studies of acute myeloid leukaemia. A series of 54 patients treated at the Royal Marsden Hospital between 1979 and 1996, with APL and the t(15;17) chromosome translocation at presentation, was examined for the effect of additional chromosome abnormalities in their presentation karyotype on survival.The patients were aged between 2 and 62 years with a median age of 31 years. There were approximately equal numbers of males and females. Presentation white cell count ranged from 0 . 7 to 156 × 10 9 /l with a median of 1 . 0 × 10 9 /l. 39% of patients (21/54) had additional chromosome abnormalities at presentation. Statistical analyses were performed for factors thought to influence survival such as age, sex, white cell count, and number of courses of chemotherapy required to enter remission. These showed that the presence of additional chromosome abnormalities has an adverse effect on prognosis, independent of other prognostic indicators, reducing it to the level of patients with AML from less-favourable cytogenetic subgroups. These data indicate that additional therapeutic strategies may be required in patients with APL who demonstrate cytogenetic aberrations over and above the t(15;17) at presentation. The biological basis for the more aggressive nature of these cases remains to be determined.

Six a2-macroglobulin (a2M) cDNA clones were isolated from a human liver cDNA library by using synthetic oligonucleotides as hybridization probes. One of these, pa2M1, carries a 4.6-kilobase-pair insert, which was sequenced. The insert contains the coding sequences for the mature a2M polypeptide (1451 amino acids) and for a 23-amino acid signal peptide at the NH2 terminus of the precursor proa2M. At the 3' end of the insert a poly(A) addition signal A-A-T-A-A-A and part of the poly(A) tail of the messenger RNA were found. The protein sequence deduced from the nucleotide sequence agrees with the published a2M amino acid sequence for all except three residues. The a2M locus was assigned to human chromosome 12 by Southern blot analysis with DNA from a panel of mouse/human somatic cell hybrids, using a2M cDNA as a hybridization probe.a2-Macroglobulin (a2M) is a serum glycoprotein and a major plasma proteinase inhibitor with a wide specificity. a2M-related proteins are present in all vertebrate species (1-4). Human a2M is a tetramer of four identical 185-kDa subunits, arranged as a pair of dimers each consisting of two disulfidelinked monomers (5, 6). The a2M polypeptide has a so-called bait region and an internal thiol ester bond, which account for its properties as a proteinase inhibitor. The bait region, composed of a series of target peptide bonds for plasma proteinases (7, 8) (13,14).This suggests that the conformational change, which accompanies the complex formation between a2M and proteinases and the hydrolysis of the thiol ester bond, exposes regions of the a2M molecule that are recognized by these receptors. Receptor-mediated endocytosis of proteinase-a2M complexes by macrophages and liver cells leads to clearance of the complexes from the circulation.Internal thiol ester bonds are also found in the complement proteins C3 and C4 (15), which are derived from precursor polypeptides of size similar to a2M (180-200 kDa). Their thiol ester sites are found in positions comparable to those in a2M, and the amino acid sequences of all three thiol ester sites are conserved. These observations led to the proposal that the C3, C4, and a2M genes are derived from a common ancestral gene (16). The sequences of murine and human C3 and human C4 and partial cDNA sequences of murine C4 have recently been determined (17-21). Comparison of human a2M with murine C3 revealed a 25% overall sequence homology (17,18,22

Point mutations in codons 12, 13 or 61 of the oncogenes Ha-ras, Ki-ras or N-ras have been identified in human malignancies of many types. Using the PCR (polymerase chain reaction) technique for DNA amplification in vitro and stringent probing of the amplified DNA on dot blots with a library of specific oligonucleotides, we have screened for the presence of ras mutations in oral and para-oral malignancies and some associated lesions. The material, from UK patients, consisted of 22 oral squamous-cell carcinomas including 5 neck metastases, 1 oral mucosal dysplasia, 1 proliferative verrucous leukoplakia, 1 antral and 1 tonsillar carcinoma, 1 basal-cell carcinoma, 1 salivary adenocarcinoma, 1 salivary adenoid cystic carcinoma and 1 lung adenocarcinoma metastatic to the gingiva. Genomic DNA was extracted from tissues which were fresh or preserved in liquid nitrogen. Two DNA samples contained point mutations in codon 61 of Ki-ras. One of these mutations was in the lymphocytes infiltrating a retromolar SCC. The other mutation (CAA to CAU; substitution of glutamine by histidine) was in the lung adenocarcinoma metastasis. The absence of ras mutations in the epithelium of primary oral squamous-cell carcinomas is of considerable interest as other work in our Department on Indian cases of oral carcinomas associated with chewing tobacco (quid) revealed that 35% of these had a codon 12, 13 or 61 mutation in Ha-ras. While ras activations arising from point mutations may occur in a high proportion of oral malignancies associated with chewing tobacco (quid), this was not the case in UK oral malignancies, even where tobacco was smoked.

Specific chromosomal abnormalities are increasingly recognised to be associated with particular tumour subtypes. These cytogenetic abnormalities define the sites of specific genes, the alteration of which is implicated in the neoplastic process. We used comparative genomic hybridisation (CGH) to examine DNA from different breast and ovarian cancer cell lines for variations in DNA sequence copy number compared with the same normal control. We also compared different sources of the MCF7 breast line by both CGH and cDNA expression arrays. Some of the differences between the subcultures were extensive and involved large regions of the chromosome. Differences between the four subcultures were observed for gains of 2q, 5p, 5q, 6q, 7p, 7q, 9q, 10p, 11q, 13q, 14q, 16q, 18p and 20p, and losses of 4q, 5p, 5q, 6q, 7q, 8p, 11p, 11q, 12q, 13q, 15q, 19p, 19q, 20p, 21q, 22q and Xp. However, few variations were found between two subcultures examined, 5 months apart, from the same initial source. The RNA arrays also demonstrated considerable variation between the three different subcultures, with only 43% of genes expressed at the same levels in all three. Moreover, the patterns of the expressed genes did not always reflect our observed CGH aberrations. These results demonstrate extensive genomic instability and variation in RNA expression during subculture and provide supportive data for evidence that cell lines do evolve in culture, thereby weakening the direct relevance of such cultures as models of human cancer. This work also reinforces the concern that comparisons of published analyses of cultures of the same name may be dangerous.

During the past several years , a panel of human tumor cell lines (predominantly ovarian) with acquired resistance to cisplatin , the orally bioavailable analogue JM216 , and the structurally hindered analogue AMD473 , has been established and characterized for underlying mechanisms of resistance. We have examined these resistant cell lines for gains and losses of DNA associated with the acquisition of resistance using the molecular cytogenetic technique of comparative genomic hybridization. Our comparison of three analogues has shown the most frequently observed changes to include amplification of 4q (5/7) and 6q (5/7) , followed by amplification of 5q (3/7). We have defined four minimal common overrepresented regions , two each on 4q and 6q , which are potential loci of genes associated with platinum analogue resistance. Additional consistent abnormalities appear to be associated with cell lines sharing specific resistance mechanisms. For example , amplification of 12q was observed in the CH1 lines made respectively resistant to JM216 and AMD473 in which increased DNA repair appears to be a major mechanism of resistance for both agents. Hence , these comparative genomic hybridization studies have identified distinct chromosomal aberrations which may correlate with defined mechanisms of resistance and contain hitherto unrecognized genes that may provide targets for future therapeutic intervention. (Am J Pathol 1999, 155:77-84)The major thrust in anticancer therapeutic development is the identification of selective therapies against molecular targets.1,2 The identification of molecular mechanisms of drug resistance has been expedited by the examination of cell lines with acquired resistance, using modern molecular techniques. These techniques include classical cytogenetics, differential display, fluorescent in situ hybridization (FISH), and the more modern approaches of comparative genomic hybridization (CGH) and spectral karyotyping (SKY).CGH is a new technique used to examine an entire genome for variations in DNA sequence copy number 3 (amplifications and deletions). It does not require replicating cells and therefore produces results which are representative of the tumor as a whole and not just the dividing population. In contrast to FISH, it does not require a previous knowledge of genetic aberrations. It can be employed with DNA extracted from fresh tumor material or material that has been frozen, formalin-fixed, or paraffin embedded. Finally, in contrast to differential display, CGH provides information on the chromosomal location of the amplified or deleted region. We have used DNA from corresponding pairs of resistant and sensitive cell lines labeled with fluorochromes of different colors, eg, green and red. These two DNAs are hybridized simultaneously to metaphase spreads from control (normal) cells. Comparison of the ratio of red:green signal along each chromosome axis reveals regions of gain and loss between the sensitive and resistant cell lines.CGH is currently being used to determine aberrations i...

The human type II collagen gene, COL2A1, has been assigned to chromosome 12, the type III gene, COL3AI, to chromosome 2, and one of the type IV genes, COMA), to chromosome 13. These assignments were made by using cloned genes as probes on Southern blots of DNA from a panel of mouse/human somatic cell hybrids. The two genes of type I collagen, COLIAI and COL2AI, have been mapped previously to chromosomes 17 and 7, respectively. This family of conserved genes seems therefore to be dispersed throughout the genome.In vertebrates, the family of collagen proteins consists of a minimum of nine types of collagen molecules whose constituent chains are coded for by a minimum of 17 genes (1-5). Types I, II, and III, the so-called interstitial collagens, are the best characterized in terms of the amino acid and DNA sequences of their chains, and these data suggest that these chains are highly conserved evolutionarily. The genes coding for the interstitial chains have been isolated from several species and exhibit a very unusual and characteristic structure of a large number of relatively small exons [54 and 108 base pairs (bp)] at conserved positions along the length of the triple-helical Gly-X-Y portion (see ref. 6 for review). Type I collagen is a heteropolymer of two al(I) chains and one a2(I) chain, whereas types II and III are homopolymers of three al(II) and al(III) chains, respectively. These three interstitial collagen molecules each form highly ordered fibrils and are responsible for the tensile strength of tissue such as skin, bone, tendon (types I and III), and hyaline cartilage (type II) in which they are found.

Somatic cell hybrids have been constructed between a thymidine kinase-deficient mouse cell line and blood leukocytes from a patient with acute promyelocytic leukemia showing the 15q+;17q-chromosome translocation frequently associated with this disease. One hybrid contains the 15q+ translocation chromosome and very little other human material. We have shown that the c-fes oncogene, which has been mapped to chromosome 15, is not present in this hybrid and, therefore, probably is translocated to the 17q-chromosome. Analysis of the genetic markers present in this hybrid has enabled a more precise localization of the translocation breakpoints on chromosomes 15 and 17. Our experiments also have enabled an ordering and more precise mapping of several genetic markers on chromosomes 15 and 17.