Chromosome 1 heterochromatin variants and cancer: a reassessment

Centromeric Satellite DNAs: Hidden Sequence Variation in the Human Population

The central goal of medical genomics is to understand the inherited basis of sequence variation that underlies human physiology, evolution, and disease. Functional association studies currently ignore millions of bases that span each centromeric region and acrocentric short arm. These regions are enriched in long arrays of tandem repeats, or satellite DNAs, that are known to vary extensively in copy number and repeat structure in the human population. Satellite sequence variation in the human genome is often so large that it is detected cytogenetically, yet due to the lack of a reference assembly and informatics tools to measure this variability, contemporary high-resolution disease association studies are unable to detect causal variants in these regions. Nevertheless, recently uncovered associations between satellite DNA variation and human disease support that these regions present a substantial and biologically important fraction of human sequence variation. Therefore, there is a pressing and unmet need to detect and incorporate this uncharacterized sequence variation into broad studies of human evolution and medical genomics. Here I discuss the current knowledge of satellite DNA variation in the human genome, focusing on centromeric satellites and their potential implications for disease.

Heterochromatin variants in 109 ovarian cancer patients and 192 healthy subjects

Aberrations of the C-band region of chromosome no. 1 (1qh) were studied in 109 patients with ovarian cancer and 192 healthy subjects. The groups were compared for heterochromatin size variations, intrapair size asymmetry, and inversion. No significant correlation was found between the size of 1qh and ovarian cancer. Heterochromatin size asymmetry was estimated visually and determined by objective measurement of 1qh length or area; the methods show strong correlation. The measurements were normalised by comparison with the length or area of 16p or the entire chromosome no. 1. However, since good reliability was found by simply relating the 1qh size difference to the mean 1qh size, this was considered an appropriate and simpler method of normalisation. Asymmetry indices of length and area measurements correlated well, implying that the simpler method of length measurements can be readily used. 1qh asymmetry, measured objectively or estimated visually, was significantly increased in the cancer patient group. The incidence of C-band inversion was significantly increased in the patient group. Moreover, inversion increased significantly with increasing 1qh asymmetry.

Familial occurrence of cancer and heteromorphism of the heterochromatic segment of chromosome 1

During the last decade, evidence has been forthcoming in support of the correlation between heteromorphism of human chromosome 1qh and the incidence of various malignancies in the carriers of such heteromorphism. We present data from a family with hereditary predisposition to cancer. In this family, five members in a sibship of seven developed ovary and/or colon carcinoma at comparatively young ages. A further 4 cases of malignant disease were ascertained, when a pedigree of 36 family members of 3 generations was constructed. Chromosome analysis was carried out in G- and C-banding from peripheral blood cultures of 19 family members. Distinct heteromorphism in the chromosome 1qh region was detected in 15 (79%) of them, including all 3 cancer patients investigated.

Constitutional C-band polymorphism in lymphocytes from patients with chronic myeloid leukemia

The C-band heterochromatin polymorphism of chromosomes 1, 9, and 16 was studied in lymphocytes from 53 patients with Ph1-positive chronic myeloid leukemia (CML) and 183 control persons. The patients had significantly larger heterochromatic blocks on chromosome 16 (p less than 0.01) and fewer partial inversions of chromosome 9 (p less than 0.05) than the control persons, whereas no differences were found for the symmetry/asymmetry pattern. We suggest that the increased constitutive heterochromatin regions may, via sister chromosome exchange, facilitate homo- or hemizygotization of genes which favor neoplasia development and/or progression.

Chromosomes in the genesis and progression of ependymomas

Cytogenetic analysis was performed on cultures of primary ependymal tumors with different degrees of malignancy (I-IV) obtained from four patients, none of whom had received therapy before karyotypic evaluation. The most common abnormalities were monosomy 17 and 22 in four cases and losses of sex chromosomes in three cases. Structural rearrangements of chromosome 2 were a finding for all cases and involved loss of material at 2q32-34. Other structural chromosome abnormalities detected involved chromosomes 4, 6, 10, 11, 12, and X. We also reviewed data on 22 cases previously reported.

Cytogenetic findings in 111 ovarian cancer patients: therapy-related chromosome aberrations and heterochromatic variants

The chromosomes of 111 ovarian cancer patients were studied in G- and C-banded slides from peripheral blood lymphocyte (PBL) cultures for chromosome damage caused by chemotherapy and radiotherapy and for asymmetry of the constitutive heterochromatin of chromosomes 1, 9, and 16. We also monitored the survival of these patients to determine whether any secondary neoplasia induced by the therapy and report the findings of our investigations. Melphalan (MEL) was the only drug used in single-drug chemotherapy. The incidence of chromosome abnormalities in melphalan-treated cells (25%) was higher than in the control group (17%). The incidence of structural changes was also higher (10.5%) in the MEL-treated group than in controls (6%). After treatments with combinations of drugs, the incidence of structural changes remained at the same level (11%). In the patients receiving combined treatment with MEL and radiation, the rate of structural changes increased dramatically (24%). The overall rate of chromosome aberrations in this group was also higher (50%). Combination of two or more drugs and radiation produced only 14% structural chromosome changes. The overall rate of chromosome aberrations was also low (20%) in this group. Of 111 patients studied, only 33 were alive 6 years after initiation of the study. Of the surviving patients, eight had rearranged chromosomes in the first analysis. After 5 years, new blood samples were collected from these patients and chromosome analyses showed abnormal karyotypes in all eight patients. All chromosome abnormalities in the second analysis were completely unrelated to those in the first analysis, however. Whether the chromosome changes in the second analysis were due to therapy or to other unknown factors could not be determined. Data on C-banding and the distribution of inversions indicated that 91% of the patients had C-band heteromorphisms of chromosomes 1, 91% had heteromorphisms of chromosome 9, and 69% had heteromorphisms of chromosome 16. Furthermore, inversions were observed in chromosome 1 (41% of patients), chromosome 9 (28% of patients), and chromosome 16 (5% of patients).

Constitutive heterochromatin polymorphisms in children with acute lymphoblastic leukemia

The C-band heterochromatin polymorphisms of chromosomes 1, 9, and 16 were studied on peripheral lymphocytes of 67 children with acute lymphoblastic leukemia (ALL) and 50 control individuals. A statistically significant difference between patients and controls was found for large heterochromatin regions (level 3) of chromosomes 1 and 9 (P < 0.001) and for small heterochromatin regions (level 1) of chromosome 16 (P < 0.001). The patients also showed a significant increase in chromosomes 1 and 9 heteromorphism with respect to controls (P < 0.001).

Variations in centromeric heterochromatin among patients with pre-malignant and malignant oral diseases

Polymorphism of heterochromatic regions of chromosomes 1, 9 and 16 was studied in 60 oral cancer patients, in 40 patients with oral submucous fibrosis (OSMF) and in 60 normal healthy subjects. The size heteromorphism was significantly greater (p less than 0.001) in chromosome I of the patients. Localization variants were also significantly more frequent among the patients (p less than 0.05 for OSMF and less than 0.001 for oral cancer patients). The C-band heteromorphism patterns remained comparable in OSMF and in oral cancer patients, with chromosome I being the most frequently involved. On correlating the tobacco/areca-nut chewing habit with the presence of C-band heteromorphism, we observed that C-band heteromorphism was present in 89% of the habit-free oral cancer patients and 80% of the OSMF patients with relatively shorter exposure to this habit, i.e. less than 5 years. This signifies that genetic factors are important in the causation of oral precancerous and cancerous conditions and that polymorphism of the heterochromatic regions does appear to play a role in these conditions.

Constitutive heterochromatin C-band polymorphism in prostatic cancer

The size of the heterochromatin C-bands on chromosomes has been reported to be associated with some, but by no means all, human malignancies. No studies along these lines have been performed in prostatic cancer. We therefore investigated the size, incidence of inversions, and symmetry versus asymmetry of C-band heteromorphisms on chromosomes 1, 9, and 16 in peripheral blood lymphocytes from 52 prostatic cancer patients and 183 healthy individuals. There were no differences in C-band heteromorphism on chromosomes 1, 9, and 16 between the patients and the controls. Neither were there any differences when patients with early-stage disease were compared with patients with more advanced cancer. Younger (aged less than 70 years) cancer patients had significantly higher frequencies of larger C-bands on chromosomes 1 (p less than 0.01) and 16 (p less than 0.001) than did patients aged more than 70 years at diagnosis. This could indicate a possible relationship between the amount of constitutive heterochromatin on chromosomes 1 and 16 and susceptibility to early development of prostatic cancer but could also result from the age differences between the two patient groups.

Structural chromosomal abnormalities in gynecologic malignancies

Surgical specimens taken from four patients with gynecologic malignancies were cultured, and metaphase chromosomes were prepared after staining with chromamycin-A, distamycin, and DAPI. Four specially selected karyotypes and their structural aberrations are discussed in this study and compared with those (also from carcinomas) previously described in the literature.

Chromosome changes in 43 carcinomas of the cervix uteri

A summary of the chromosome changes in 43 carcinomas of the cervix studied by a direct technique showed that the most common anomaly was a small metacentric [in 77%, often in two copies: an i(5p) or possibly an i(4p)]. Others commonly involved in structural changes were: chromosome 1 (60%; most commonly an i(1q), 1p-, or translocation of part of 1q onto another chromosome); chromosome 17 (47%; translocations onto the short arm or long-arm isochromosomes), chromosome 11 (37%; translocations onto the short arm); chromosome 3 (26%; including 3p- and 31-); and chromosomes 2, 6, and 9 (each in 19%). Considering the four most frequent categories of markers--small metacentrics and markers derived from chromosomes 1, 17, and 11, none of which is specific for cervical carcinoma--almost any combination of these four might be present in a tumor (and at least one was present in all tumors) so that they were not mutually exclusive. Estimates of the average numbers of normal chromosomes based on representative karyotypes from 35 of the tumors showed that three chromosomes in particular were underrepresented (chromosomes 4, 11, and 14; 72-73% of the expected values), while chromosomes 3, 19, and 20 were those most highly represented (99-103%).

High incidence of chromosomal lesions involving C-heterochromatin in four human melanoma lines

Cytogenetic analysis of early in vitro cultures derived from human melanomas, two primary tumors (Me 10538, Me 1402) and two metastatic lesions in the same patient (Me 665/1, Me 665/2) showed non-random involvement of C-heterochromatin in clonal chromosome rearrangements. Marker chromosomes with C- and DA-Dapi-positive bands were identified in one of the metastases, Me 665/1 (m1) and in the two primary tumors, Me 10538 (m2) and Me 1402 (m3). C-positive fragments predominated in the other metastasis, Me 665/2, which lacked C-regions intercalated in rearranged chromosomes, and were also detected with appreciable frequency in the Me 665/1 and Me 1402 cells. The frequencies of marker chromosomes and their mean number per cell allowed m2 and m1 to be considered as early markers of tumor formation and m3 as a marker of tumor progression. Dissection of chromosome structure, including the origin of the intercalated C-band, has so far been achieved only with the m2 chromosome of the primary tumor Me10538. This was the only cell line which displayed few C-fragments and a narrow chromosomal distribution with a well defined mode. A gradient of malignancy could be detected in the four cell lines, by local and disseminated tumor growth in xenotransplanted mice, with the two primary melanomas 10538 and the 1402 cells at the lowest and upper extremes. This gradient closely parallels the increase in cytogenetic heterogeneity and C-heterochromatin lesions from the 10538 to the 1402 cells.

Densitometric measurements of C bands of chromosomes 1, 9, 16, and Y in leukemic and preleukemic disorders

Fifty-six patients with blood disorders (23 with chronic myeloid leukemia, 14 with acute myeloblastic leukemia, seven with acute lymphoblastic leukemia, one with chronic lymphocytic leukemia, and 11 with preleukemia states) were studied. A quantitative and objective method of C band length analysis with well-matched controls was used. The C bands of chromosome pairs 1, 9, and 16 presented a normal distribution that was similar in patients and controls, whereas the Y chromosome presented an abnormal distribution. Smaller C bands in 1qh and higher indexes of intrapair heteromorphism in pairs 1 and 9 were detected in the CML group; the group of acute leukemias (myeloblastic and lymphoblastic) presented a smaller index only in pair 1qh. No other differences in length, heteromorphism, inversion frequency, or sex were detected.

Identification by C-banding of two human prostate tumour cell lines, 1013L and DU 145

Two human prostate carcinoma cell lines are easily distinguished by C-banding. DU 145 has one metacentric chromosome with a large centromeric band, 1 metacentric and 2 acrocentric chromosomes with interstitial C-bands. The Y chromosome was present in 48% and 42% of the cells at passages 83 and 106, respectively, whereas in 1013L only one metacentric chromosome contained distinct extra C bands--one at the centromere and one interstitially. The Y chromosome was present in 72% of the cells at passage 15 and in 20% at passage 22, but by passage 44 it had disappeared. Karyotype analysis using G-banding revealed that DU 145 had retained several of its original markers, and demonstrated multiple marker chromosomes in 1013L.

Solid tumor cytogenetics. Progress since 1979

Some of the advances in the past decade in the field of solid tumor cytogenetics are described, with particular reference to nonrandom structural chromosome changes. Although it had been known for many years that meningiomas and salivary gland tumors were associated with changes involving particular chromosomes, it has only quite recently become clear, following the application of suitable culture techniques, that other benign tumors such as lipomas and leiomyomas may also be characterized by specific changes, particularly reciprocal translocations. Reciprocal translocations may also be found in malignant soft-tissue tumors such as liposarcomas (involving 12q as in lipomas) and Ewing's sarcoma. In contrast, the common forms of carcinoma present a more variable picture, although certain chromosomes may undergo nonrandom changes of various types, including translocations, which, however, are generally nonreciprocal. Some of these chromosomes may be quite specific (e.g., chromosome 10 in prostatic and #18 in colorectal cancer), while others appear to be common to many or all types of carcinoma, such as chromosomes 1, 3, 11, and 17, and a small metacentric that may be an i(5p). In carcinoma of the bladder, different chromosome changes may characterize subsets of the tumors. In carcinoma of the cervix, however, the commonly involved chromosomes, 1, 3, ?5, 11, and 17, appear in markers in any combination and are thus not mutually exclusive. Although further study of the chromosome changes in carcinomas is essential to an understanding of their relationship to the molecular changes that are associated with malignant transformation, it can be hypothesized that, while some of the changes result in the duplication of particular genes, e.g., on chromosome 1q, a more important role may be to bring about the loss of chromosomal segments containing tumor-suppressor genes. Evidence from molecular studies that has recently been accumulating for the loss of alleles on, for instance, 3p, 11p, and 17p, which could in part be due to gross chromosomal rearrangements, also strongly suggests the importance of genic loss in malignant transformation. In carcinomas, at least, the changes probably involve a number of genes, each change representing one of the several steps necessary for tumorigenesis.

Constitutive heterochromatin polymorphism and chromosome damage in viral hepatitis

The frequencies of chromosomal aberrations and sister-chromatid exchanges (SCEs) were scored in relation to constitutive heterochromatin in 100 patients with viral hepatitis B, 100 patients with viral hepatitis A and 100 age- and sex-matched normal controls. 23.4%, 15% and 4% of the cells showed chromosomal aberrations in patients with hepatitis B, hepatitis A and normal controls respectively. Non-random involvement of chromosomal aberrations were also noted in chromosome 1 of patients with hepatitis B and A as compared to normal controls. The frequencies of SCEs (mean +/- S.D.) were found to be 10.40 +/- 2.83 in hepatitis B and 8.70 +/- 2.34 in hepatitis A. These values were significantly higher than the SCE frequency (mean +/- S.D.) of 5.88 +/- 2.25 observed in normal controls (P less than 0.001). The intra-chromosomal distribution of SCEs revealed a relatively increased incidence of SCEs in chromosome 1 of patients with hepatitis B and A as compared to normal controls. Analysis of constitutive heterochromatin polymorphism showed chromosome 1 qh+ to be the most frequent variant in patients with hepatitis B and A as compared to normal controls. The increased involvement of C-band variant 1 qh+ in patients with hepatitis B and A as compared to normal controls may indicate that extra heterochromatin offers additional sites for viral integration.

Heterochromatic variants and their association with neoplasias: IV. Colon adenomas and carcinomas

C-band polymorphisms in peripheral blood lymphocytes of 62 patients (33 with colon adenomas and 29 with colon carcinomas) were studied. A significant difference in the frequency of heterochromatic variants in chromosomes #1 in both colon adenoma (56%) and carcinoma (67%) with respect to controls (18%) was observed (p less than 0.001). The heterochromatic variants preferentially involved in both pathologies were inv(1), 1qh-, and inv(9), compared with controls. No differences were found between colon adenomas and carcinomas. We suggest that 1qh- and inv(1) variants are important heterochromatic changes in neoplasia.

C-band polymorphisms in lymphocytes of patients with ovarian or breast adenocarcinoma

To establish the significance of the variability of human chromosome constitutive heterochromatin areas (C-band variants) in a risk of malignancy, C-banding pattern study has been performed in 33 female patients with ovarian or breast adenocarcinoma. The control group included 180 healthy women. The following characteristics of C-bands on chromosomes #1, #9, and #16 were studied: (a) size, (b) size heteromorphisms and (c) inversions, using quantitative and semiquantitative methods of analysis. Our data show no significant difference in the presence of C-band size and location variants in chromosomes #1, #9, and #16 between the patients with adenocarcinoma of the ovary or breast and healthy women. From that we conclude that there is no causal association between the presence of C-band variants on chromosomes #1, #9, and #16 and an elevated risk of ovarian and breast adenocarcinoma.

Heteromorphism of C-band positive chromosomal regions in CML patients

The heteromorphism of constitutive heterochromatin in chromosomes #1, #9, and #16 was investigated in 44 chronic myelocytic leukemia patients and 44 controls using bone marrow and peripheral blood lymphocyte cultures. A significant increase in the length of C-band region in all the three chromosome pairs as well as a statistically significant difference in the homologs of chromosome #1 was observed in chronic myelocytic leukemia patients when compared with the controls. The frequency of inversions was also greater in the patients than in the controls. A random translocation of 22q was found on either homolog of chromosome #9.

Lack of reciprocal translocations in carcinomas

Among the varied structural chromosomal aberrations present in the common forms of carcinoma, reciprocal translocations generally appear to be lacking. Although the breakpoints may be variable, the chromosomal changes nevertheless commonly result in the loss or gain of particular chromosomal segments. The possible implications of these observations are discussed, especially in relation to the thesis that some of the chromosomal changes in cancer cells are important because they lead to the expression of recessive genes.