Allele loss, replication errors and loss of expression of E-cadherin in colorectal cancers

BACKGROUND

Loss of E-cadherin expression has been implicated in the development of invasive characteristics in colorectal carcinomas. Failure to express E-cadherin may result from mutations of the E-cadherin gene (HSECAD).

AIMS

To examine the correlation between E-cadherin expression and genetic changes at HSECAD; and to examine differences in E-cadherin expression and genetic changes at HSECAD between three different groups of colorectal cancer--replication error positive (RER+) sporadic cancers, RER--sporadic cancers and ulcerative colitis associated cancers.

SUBJECTS AND METHODS

Sixty eight colorectal cancers (22 RER+ sporadic cancers, 32 RER- sporadic cancers and 14 ulcerative colitis associated cancers) were studied using immunohistochemistry and for allele loss at the HSECAD locus. Exon 16 of HSECAD contains several mononucleotide repeat tracts which are very similar to microsatellite repeats and which may be susceptible to replication errors (manifest as new alleles). All cases were also examined for new alleles in exon 16.

RESULTS

Absent or decreased E-cadherin protein expression was found in 27 (38%) of 68 colorectal cancers and the pattern of expression did not differ significantly among the three tumour groups. Allele loss occurred at HSECAD in four (10%) of 40 informative cancers and there were no differences between the three subgroups. New alleles at exon 16 were detected in three (14%) of 22 RER+ tumours; no new alleles were detected in RER- or ulcerative colitis associated cancers. Overall, there was no correlation between allelic loss or exon 16 replication errors and immunohistochemical E-cadherin expression.

CONCLUSIONS

(1) Loss of E-cadherin expression probably does not occur as a result of mutation at the HSECAD locus in colorectal cancers. (2) There is no difference in the frequency of loss of heterozygosity at HSECAD among RER+, RER- and ulcerative colitis associated colorectal cancers.

Genetic testing is important in families with a history suggestive of hereditary non-polyposis colorectal cancer even if the Amsterdam criteria are not fulfilled

BACKGROUND

Clinical screening is still the first-line approach to identification of families with hereditary non-polyposis colorectal cancer (HNPCC). The need for uniformity of diagnosis of the syndrome, particularly in multicentre studies, led to the establishment of a set of minimum diagnostic criteria, the 'Amsterdam criteria'. It is now known that HNPCC is caused by germline defects in the human mismatch repair genes and DNA predictive testing is possible. Defects in two of the known mismatch repair genes, namely hMSH2 and hMLH1, account for over 90 per cent of mutations found in HNPCC families.

METHODS

Ten families were identified with pedigrees suggestive of HNPCC (that is with a possible dominant inheritance of HNPCC), but in which the Amsterdam criteria were not fulfilled. Using the technique of single-strand conformational polymorphism analysis, samples were screened from an affected member of each of these ten kindreds for germline mutations in the genes hMSH2 and hMLH1.

RESULTS

Mutations were identified in six families. Of these, there were three missense, one nonsense, one frameshift and one putative splice-site mutation. Three of the mutations were in hMSH2 and three in hMLH1.

CONCLUSION

This study demonstrates that all families with a pedigree suggestive of HNPCC should be referred to a geneticist even if the Amsterdam criteria are not fulfilled. A knowledge of the gene carrier status enables targeted surveillance and the possibility of early surgical intervention that could be curative.

DNA mismatch repair in lymphoblastoid cells from hereditary non-polyposis colorectal cancer (HNPCC) patients is normal under conditions of rapid cell division and increased mutational load

Patients with hereditary non-polyposis colorectal cancer (HNPCC) have germ-line mutations in one of four DNA mismatch repair genes (hMSH2, hMLH1, hPMS1 and hPMS2). It is supposed that a single functional copy of these genes is sufficient for normal mismatch repair, but it is not certain that this is the case under abnormal conditions such as rapid cell division or an increased tendency to DNA replication errors (RERs). We have analysed mismatch repair by examining replication errors in immortalised lymphoblastoid cells derived from two HNPCC patients heterozygous for mismatch repair defects (one hMSH2 mutant and one hMLH1 mutant), and from control individuals. Three conditions of cell culture have been used: (i) relatively slow cell growth and division; (ii) relatively fast growth and division; and (iii) chronic perturbation of the intracellular dNTP pool to promote a increased frequency of replication errors. No significant differences in microsatellite instability were found between HNPCC patients and controls in any of these environments. Lymphoblastoid cells from hMSH2 and hMLH1 mutant/wild-type heterozygotes appear, therefore, to have normal levels of mismatch repair, even under conditions that increase the requirement for repair. The pool bias cultures from the HNPCC patients and controls did, however, show similar, increased frequencies of RERs, suggesting that the mismatch repair capacity of the cells had been overloaded, but that the number of normal HNPCC alleles was not the limiting factor.

Tumour recurrence is associated with Jass grouping but not with differences in E-cadherin expression in moderately differentiated Dukes' B colorectal cancers

AIMS

To assess whether immunohistochemical expression of the putative invasion suppressor, E-cadherin, is associated with tumour recurrence in colorectal cancer, independent of Dukes' stage and degree of differentiation, and thus to determine whether measurement of E-cadherin is clinically useful.

METHODS

90 moderately differentiated Dukes' B colorectal carcinomas from patients who had been followed up for at least six years were examined. All were from curative resections performed at St Mark's Hospital and the tumours were shown, on histology, to be clear of all resection margins. Tumours were reviewed and additionally classified in accordance with the Jass grouping system. Immunohistochemical analysis of E-cadherin expression was performed for each tumour using the monoclonal antibody HECD 1.

RESULTS

The Jass group ranged between I and III and there was a significant association between Jass group and tumour recurrence (p < 0.05). Positive E-cadherin expression was seen in 40/69 (58%) of non-recurrent and 13/21 (62%) recurrent cases respectively. There was no significant difference in E-cadherin expression between the two groups of tumours.

CONCLUSIONS

Even when controlled for Dukes' stage and differentiation, the Jass grouping system yields useful prognostic information; E-cadherin, however, does not predict outcome in the important group of moderately differentiated Dukes' B colon cancers, and may be of little independent prognostic value in other colon cancers.

Use of SSCP analysis to identify germline mutations in HNPCC families fulfilling the Amsterdam criteria

Hereditary non-polyposis colorectal cancer (HNPCC) is a clinical syndrome characterised by an inherited predisposition to early onset colorectal and uterine cancers and an increased incidence of other cancers. It is caused by germline defects in the human mismatch repair genes. Defects in two of the known mismatch repair genes (namely hMSH2 and hMLH1) account for over 90% of mutations found in HNPCC families. In this study we have identified 14 families that fulfilled the clinical criteria for HNPCC and screened the hMSH2 and hMLH1 genes for germline mutations using single-strand conformational polymorphism (SSCP) analysis and DNA sequencing. Seven mutations were identified. Of these, there were five frameshifts, one missense mutation and a further novel mutation that involved separate transition and transversion changes in successive amino acid residues. Three of the mutations were in hMSH2 and four in hMLH1. The identification of germ-line mutations in an HNPCC family enables targeted surveillance and the possibility of early curative intervention. SSCP is a simple and effective method for identifying most mutations in the human mismatch repair genes using DNA from fresh, frozen or archival material.

Clinical and molecular features of the hereditary mixed polyposis syndrome

BACKGROUND & AIMS

Various inherited syndromes predispose to the development of colonic juvenile polyps and colorectal cancer, with potential importance for sporadic tumorigenesis. This study describes features of a possibly new syndrome of atypical juvenile polyps and other colonic tumors and compares these features with those of known gastrointestinal tumor syndromes.

METHODS

A large family, St. Mark's family 96, with a tendency to develop colonic polyps of mixed histological types is described. Genetic linkage to known polyposis syndromes has been tested.

RESULTS

Adenomatous and hyperplastic polyps occur in affected members of the family, although the characteristic lesion is an atypical juvenile polyp. Some affected individuals have developed polyps of more than one type, and individual polyps may contain features of more than one histological type. Polyps can undergo malignant change. Typically, fewer than 15 polyps are found at colonoscopy and there is no extracolonic disease associated with the development of polyps. The family's polyps seem to be inherited in an autosomal-dominant fashion, but the disease is probably unlinked to candidate loci with importance in colorectal tumorigenesis, such as APC, hMSH2, and hMLH1.

CONCLUSIONS

We term this family's disease hereditary mixed polyposis syndrome (HMPS). Although mutations in the putative HMPS gene may be responsible for syndromes such as juvenile and Peutz-Jeghers polyposes, HMPS may also be a distinct disease.

Modelling the consequences of interactions between tumour cells

Classical models of tumorigenesis assume that the mutations which cause tumours to grow act in a cell-autonomous fashion. This is not necessarily true. Sometimes tumour cells may adopt genetic strategies that boost their own replication and which also influence other cells in the tumour, whether directly or as a side-effect. Tumour growth as a whole might be enhanced or retarded. We have used mathematical models to study two non-autonomous strategies that tumour cells may use. First, we have considered the production by tumour cells of an angiogenesis growth factor that benefits both the cell from which it originates and neighbouring cells. Second, we have analysed a situation in which tumour cells produce autocrine-only or paracrine-only growth factors to prevent programmed cell death. In the angiogenesis model, stable genetic polymorphisms are likely to occur between cells producing and not producing the growth factor. In the programmed cell death model, cells with autocrine growth factor production can spread throughout the tumour. Production of paracrine-only growth factor is never selected because it is 'altruistic' (that is of no benefit to the cell that makes the growth factor), despite being potentially beneficial to tumour growth as a whole. No polymorphisms can occur in the programmed cell death model. Production of angiogenesis and other growth factors in tumours may be under stable genetic, rather than epigenetic, control, with implications for therapies aimed at such targets. Many of the mutations observed in tumours may have non-autonomous effects.

Germline HNPCC gene variants have little influence on the risk for sporadic colorectal cancer

Hereditary non-polyposis colorectal cancer (HNPCC) is a syndrome of inherited bowel and other cancers that has been said to account for up to 15% of all colorectal carcinomas (CRCs). HNPCC can now be diagnosed at the molecular level by detecting germline mutations in genes involved in mismatch repair. A current problem is to determine the prevalence of HNPCC mutations in colon cancer patients with limited or no family history, especially in cases of early onset. We have identified 50 cases of non-polyposis colorectal cancer without a family history of CRC or any other HNPCC cancer, who presented under the age of 45 years. Germline HNPCC variants (at the hMSH2 or hMLH1 loci) were detected in a small minority of cases (6%). The variants that we have found may be new or low penetrance mutations, or even polymorphisms. It remains possible that some of our sample have an inherited predisposition to CRC that is not caused by HNPCC mutations or by known polyposis syndromes. Our data suggest that most HNPCC mutations occur in families and have high or moderate penetrance. New or low penetrance HNPCC mutations probably do not contribute significantly to the risk of colorectal cancer in the general population and probably account for much fewer than 15% of all CRCs. Our results question whether mass population genetic screening programmes are worthwhile for diseases such as HNPCC using current technology.

Clinical features and molecular analysis of a family with multiple colon tumours and reduced plasminogen activator activity

A family is reported in which a pair of brothers has developed recurrent venous thromboses, which have been shown in one brother (the proband) to result from a reduced level of tissue plasminogen activator (t-PA) activity. Both brothers have also developed multiple synchronous and metachronous colorectal adenomas. Other pedigree members have developed colon cancers, but not multiple colonic tumours. We have shown that HNPCC and FAP/AAPC are unlikely causes of the family's phenotypes. Previous studies have found low levels of t-PA in sporadic colon tumours. In this family, mutations in the t-PA gene, at a linked locus, or at a locus controlling t-PA activity/release may modify the colon tumour phenotype to cause multiple lesions.

The mutation rate and cancer

The selection of advantageous mutations underlies tumorigenesis. The growth of a tumor is therefore a form of evolution at the somatic level, in which the population is comprised of individual cells within the tumor. Models of tumorigenesis have considered the relative importance of mutation and selection. We show that selection is more important than an increased mutation rate in the growth of a tumor. Some cancers may acquire a "mutator phenotype," probably leading to faster growth, but mutator phenotypes are not necessary for carcinogenesis.

Bcl-2 expression in colorectal tumors: evidence of different pathways in sporadic and ulcerative-colitis-associated carcinomas

Colorectal cancers (CRCs) differ in their age at presentation, distribution, histological features, and prognosis. If tumor biology reflects genetic events, these tumors might be expected to show differences in their genetic pathways. In this study, we investigated the role of Bcl-2 in the development of three different tumor groups. Using markers at eight different microsatellite locl, we characterized one group of 34 left-sided sporadic CRCs as replication error negative (RER-) and another group of 18 left-sided sporadic CRCs as replication error positive (RER+). These tumors, together with a third group of 22 left-sided ulcerative-colitis-associated CRCs (UCACRCs), were then examined by immunohistochemistry for Bcl-2 overexpression. Of 34 of the RER- tumors, 21 (62%) and 10 of 18 (56%) of the RER+ tumors were positive for Bcl-2 overexpression. In contrast, only 5 of 22 (23%) of the UCACRCs showed similar overexpression. Our results show a significantly lower frequency of Bcl-2 overexpression in UCACRCs as compared with sporadic CRCs (P < 0.005) but no difference between sporadic left-sided RER+ and RER- CRCs. These data provide additional evidence that UCACRCs may develop along a pathway that is different from that of sporadic CRCs.

Allele loss occurs frequently at hMLH1, but rarely at hMSH2, in sporadic colorectal cancers with microsatellite instability

Mutations at the hMSH2 and hMLH1 mismatch repair loci have been implicated in the pathogenesis of colorectal cancer. Tumours with two allelic mutations at a mismatch repair locus develop replication errors (RERs). In the hereditary non-polyposis colorectal cancer (HNPCC) syndrome, one mutation is inherited and the other acquired somatically: in RER+ sporadic colorectal cancers, both mutations are somatic. RER+ tumours tend to have a low frequency of allele loss, presumably because they acquire most mutations through RERs. However, before a second mismatch repair mutation has occurred somatically, there is no reason to suppose that allele loss occurs less frequently in tumours that are to become RER+. Indeed, this second mutation might itself occur by allele loss. We have searched for allele loss at the hMSH2 and hMLH1 loci in RER+ and RER- sporadic colorectal cancers. Loss occurred at the hMLH1 locus in 7/17 (41%) RER+ tumours, compared with 6/40 (15%) RER- cancers (chi2=3.82, P approximately 0.05). At hMSH2, 2/22 RER+ sporadic cancers (9%) had lost an allele, compared with 2/40 (5%) RER- cancers (chi2=0.03, P>0.5). Taken together with previous studies which focused on colorectal cancers from HNPCC families, the data suggest that allele loss at hMLH1, but not at hMSH2, contributes to defective mismatch repair in inherited and sporadic colorectal cancer.

Allele loss on chromosome 11q and microsatellite instability in malignant melanoma

Loss of heterozygosity (allele loss, LOH) occurs frequently on the long arm of chromosome 11 in several types of cancer. We analysed 32 melanomas (almost all metastatic lesions) for allele loss at eight loci along the length of chromosome 11 (ptel-D11S922-D11S899-D11S1324-D11S1313-++ +D11S901-NCAM-D11S29-D11S968-qtel). The highest frequency of loss (38%) was at D11S29 (11q23.3). Of 13 melanomas which had lost an allele at one or more loci, all but one showed LOH at either D11S29 or NCAM (11q22). The region between these two loci is the most likely location of any tumour suppressor gene. Low frequencies of LOH occurred on 11p and there was little evidence for tumour suppressor loci outside the 11cen-q23.3 region. Unusually for melanomas, widespread microsatellite instability, with slippage of several repeat units, was observed in two of 32 tumours studied (and four other tumours showed new microsatellite alleles that differed by just one repeat unit from their normal counterparts). However, no mutations of the mismatch repair genes hMSH2 and hMLH1 were detected in these two tumours, and the observed replication errors may result from mutations in other genes involved in mismatch repair or DNA replication. LOH on 11q and replication errors appear to comprise part of the genetic pathways of several tumour types, including melanomas.

The frequency and mechanism of loss of heterozygosity on chromosome 11q in breast cancer

Loss of heterozygosity (LOH, allele loss) occurs frequently on the long arm of chromosome 11 in breast cancer. Seventy-one paired tumour/normal DNA samples from breast cancer patients under 50 years old were studied for allele loss at four microsatellite loci on 11q: D11S29 (11q23.3), NCAM (11q22-q23), D11S968 (11qtel), and D11S1313 (11qcen). The maximum frequency of LOH (approximately 35 per cent) was found at the D11S29 and NCAM loci. This result is consistent with previous studies and the frequency of allele loss is moderate to high compared with the usual baseline of 0-20 per cent. In most of the cases studied, LOH on chromosome 11q could be accounted for by one of two mechanisms. Either chromosomal non-disjunction had occurred, or sequences stretching from the telomere at least as far as NCAM had undergone deletion or mitotic recombination. These results suggest that a putative tumour suppressor gene is most likely to exist near 11q22-q23. There was a very low frequency of microsatellite instability in the tumours. An association was found between lack of progesterone receptor (PgR) expression and LOH at NCAM, suggesting that deletion of sequences on 11q may prevent high levels of PgR expression in some cases.

Testing candidate loci on chromosomes 1 and 6 for genetic linkage to Peutz-Jeghers' disease

Peutz-Jeghers' syndrome (PJS) is a disease with autosomal dominant inheritance, which is characterised by gastrointestinal hamartomata and characteristic melanin pigmentation. Three candidate sites for a PJS locus have recently been proposed, chromosomes 1p31-p32, 6q25 and 6p11-cen. At the first of these sites, a multipoint LOD score of 4.00 had been found, strongly suggesting genetic linkage to PJS. The last two candidate sites were suggested by the chromosomal breakpoints of a patient with an inv(6) and PJS. We have analysed up to 34 families in order to test each of the three candidate sites for linkage to PJS. No evidence was found in support of a Peutz-Jeghers' syndrome locus on chromosome 1p31-p32. The candidate region on 6q25 was also excluded. The region close to the centromere of chromosome 6 has not been excluded and there is some evidence of linkage to a marker near 6cen, although genetic heterogeneity in PJS must be proposed to account for a gene at this site.

Variants at the secretory phospholipase A2 (PLA2G2A) locus: analysis of associations with familial adenomatous polyposis and sporadic colorectal tumours

The Min mouse is a model for human familial adenomatous polyposis (FAP), an autosomal dominant disease characterised by multiple adenomatous gastrointestinal polyps. The severity of the Min phenotype is modified by a locus (Mom1) on mouse chromosome 4, at a position syntenic with human chromosome 1p35-p36. The secretory phospholipase A2 (Pla2s) gene is a candidate for this modifier locus and there is evidence that a locus on human chromosome 1p35-p36 acts to modify the severity of human duodenal FAP. We have analysed the human secretory phospholipase A2 locus (PLA2G2A) for variants that could directly influence the FAP phenotype. We found no PLA2G2A variants predicted to result in functional variation in the phospholipase A2 protein. Two PLA2G2A polymorphisms were, however, discovered, one a 'silent' base change in exon 3 and another in a noncoding region. Three other variants (possible mutations) were found in non-coding regions. In 70 FAP patients from 20 families, no associations were found between the severity of duodenal polyposis and any PLA2G2A variant. One allele at the exon 3 polymorphic site did, however, occur more often then expected in patients with relatively severe colonic FAP. Although of borderline statistical significance, this association, if genuine, is likely to result from linkage disequilibrium between the PLA2G2A alleles studied and undetected genetic variation at a closely linked locus. The frequency of the alleles at both polymorphic sites has also been determined in the germ line of patients with sporadic colorectal adenomas and carcinomas and in random controls, but no differences were found among these groups. Our results suggest that PLA2G2A variants do not influence inherited or sporadic colonic tumours. A linked locus may be a modifier of human FAP, but does not influence the risk of colorectal tumours in the general population.

Clinico-pathological features and p53 expression in left-sided sporadic colorectal cancers with and without microsatellite instability

Defects in mismatch repair (MMR) can result in the development of a 'mutator phenotype', manifest as an increase in DNA replication errors (RERs). Patients with hereditary non-polyposis colorectal cancer (HNPCC) have germline mutations in MMR genes. These patients develop carcinomas of the colon and other specific sites at a significantly earlier age than patients with sporadic carcinomas. RERs are found in the cancers from patients with HNPCC and have been demonstrated in 10-20 per cent of sporadic colorectal cancers (CRCs). Loss of MMR may simply accelerate tumour development, but it is also possible that these tumours follow a different carcinogenetic pathway from tumours with intact MMR. In particular, it has been suggested that p53 mutations occur less often in RER-positive (RER+) sporadic colorectal cancers. In this study, the clinico-pathological features and frequency of p53 overexpression in 17 left-sided RER+ CRCs were compared with 35 left-sided RER- CRCs. No differences were found in the age and tumour stage at presentation, mucinous differentiation, or Jass prognostic grouping between these two types of CRC. Thirteen out of 17 (76 per cent) RER+ and 19/35 (54 per cent) RER- tumours showed overexpression of p53, a non-significant difference (chi 2 test). Although some previous studies have suggested differences in the clinico-pathological features and p53 expression of RER+ and RER- right-sided CRCs, our results show that these differences do not exist in left-sided cancers.

Polyclonal origin of colonic adenomas in an XO/XY patient with FAP

It is widely accepted that tumors are monoclonal in origin, arising from a mutation or series of mutations in a single cell and its descendants. The clonal origin of colonic adenomas and uninvolved intestinal mucosa from an XO/XY mosaic individual with familial adenomatous polyposis (FAP) was examined directly by in situ hybridization with Y chromosome probes. In this patient, the crypts of the small and large intestine were clonal, but at least 76 percent of the microadenomas were polyclonal in origin.

Genetic pathways in colorectal cancer

The model of colorectal tumorigenesis put forward by Fearon and Vogelstein has had great influence on molecular oncology. They proposed that a series of mutations occur in the progression from normal cells to colorectal cancer and that these mutations are associated with the histological features of such tumours. Several postulates of the model appear to be correct, particularly its emphasis on the stepwise accumulation of genetic changes and the inclusion of mutations at the adenomatous polyposis coli (APC) and TP53 loci. Since the publication of the original model, however, mutations at other loci have been identified which may be alternatives or additions. There is also evidence to suggest that some colorectal cancers develop along a different genetic pathway. In this review, we discuss how tumour development can occur as Darwinian evolution through selection of advantageous somatic mutations. The non-random nature of mutation selection gives rise to genetic pathways of tumorigenesis. In addition, we consider the Fearon and Vogelstein model, its shortcomings and possible additions to it. The evidence suggests that not all colorectal cancers follow the same genetic pathway during carcinogenesis.

Chromosome 11q in sporadic colorectal carcinoma: patterns of allele loss and their significance for tumorigenesis

AIMS

To analyse the frequency of loss of heterozygosity (allele loss, LOH) in a large sample of colorectal carcinomas using highly informative markers along chromosome 11q.

METHODS

One hundred paired samples of colorectal cancer and normal tissue were genotyped at six microsatellite markers on chromosome 11q (cen-D11S1313-D11S901-DRD2/NCAM-D11S29- D11S968-tel). The high levels of heterozygosity at these markers allow allele loss to be determined in about 80% of cases at any one locus. The frequency of replication errors (RERs, microsatellite instability) has also been determined.

RESULTS

LOH was found at frequencies of 25% and 29% at the distal D11S968 (11qter) and D11S29 (11q23.3) loci, slightly above the accepted baseline of 0-20%. Allele loss at NCAM, DRD2, D11S901, and D11S1313 was not raised above baseline levels. The probable genetic mechanism of allele loss--chromosomal non-disjunction, mitotic recombination, deletion, or gene conversion--seemed to vary between tumours and no consistent mechanism of mutation was found. Microsatellite instability was found in 23 (23%) tumours. No associations were found between LOH and clinical data (patient sex, age at presentation, tumour site, and Duke's stage).

CONCLUSIONS

Although gene(s) on 11q may have a role in the development of a minority of colorectal carcinomas, this study provides evidence against the general importance of allele loss on chromosome 11q in the pathogenesis of colorectal cancer. The results also have implications for the importance of 11q in other cancers: it seems less likely that a single tumour supressor gene at this location promotes the growth of all types of tumour when lost. Rather, one or more genes with tissue specific effects may be involved.

A modifying locus for familial adenomatous polyposis may be present on chromosome 1p35-p36

Mutations of the APC gene cause familial adenomatous polyposis (FAP) in humans and multiple intestinal neoplasia (Min) in laboratory mouse strains. A dominant modifying gene (Mom1), which partially suppresses the min phenotype, has been mapped to mouse chromosome 4. This region is syntenic with human chromosome 1p35-p36. The phospholipase A2 (Pla2s) locus is an excellent candidate for Mom1 and the equivalent human locus PLA2G2A is found on chromosome 1p35. It does not necessarily follow, however, than any modifier of mouse polyposis also influences human disease. In order to test whether a locus on 1p modifies FAP, subjects from 28 FAP families have been typed at microsatellite loci on this chromosome arm. The severity of their duodenal polyposis has also been assessed by endoscopy. Pedigree (lod score) linkage analysis found no evidence of a simple, dominant modifying gene, comparable with the action of Mom1 in inbred mouse strains. Given the more complex genetic and environmental interactions likely to exist in outbred human populations, it is probably more appropriate to use tests which do not specify a mode of inheritance. Using these methods of analysis, the data suggest that a locus on chromosome 1p35-p36 may influence the severity of duodenal FAP.

Genetic mapping of hereditary mixed polyposis syndrome to chromosome 6q

Hereditary mixed polyposis syndrome (HMPS) is characterized by atypical juvenile polyps, colonic adenomas, and colorectal carcinomas. HMPS appears to be inherited in an autosomal dominant manner. Genetic linkage analysis has been performed on a large family with HMPS. Data did not support linkage to the APC locus or to any of the loci for hereditary nonpolyposis colorectal cancer. Evidence that the HMPS locus lies on chromosome 6q was, however, provided by significant two-point LOD scores for linkage between HMPS and the D6S283 locus. Analysis of recombinants and multipoint linkage analysis suggested that the HMPS locus lies in a 4-cM interval containing the D6S283 locus and flanked by markers D6S468 and D6S301.

The HLA system and the analysis of multifactorial genetic disease

The human leukocyte antigen (HLA) system comprises closely linked genes controlling highly polymorphic proteins involved in the presentation of peptides to the T-cell receptor. Specific alleles at HLA loci are associated with diseases, often those suspected to be of autoimmune aetiology. Many of these associations result from linkage disequilibrium between the HLA gene studied and other HLA genes or non-HLA genes close by. Owing to its high level of polymorphism and its candidate role in many diseases, HLA was the first system used in many techniques of genetic mapping, such as affected-sib-pair analysis and association (linkage disequilibrium) studies. Much remains unknown about the reasons why diseases are associated with HLA. Experience gained from HLA has, however, shown how other loci involved in complex traits can be identified by studying families with multiple affected cases or sib pairs, followed by linkage-disequilibrium mapping and then analysis of candidate genes.

Failure of programmed cell death and differentiation as causes of tumors: some simple mathematical models

Most models of tumorigenesis assume that the tumor grows by increased cell division. In these models, it is generally supposed that daughter cells behave as do their parents, and cell numbers have clear potential for exponential growth. We have constructed simple mathematical models of tumorigenesis through failure of programmed cell death (PCD) or differentiation. These models do not assume that descendant cells behave as their parents do. The models predict that exponential growth in cell numbers does sometimes occur, usually when stem cells fail to die or differentiate. At other times, exponential growth does not occur: instead, the number of cells in the population reaches a new, higher equilibrium. This behavior is predicted when fully differentiated cells fail to undergo PCD. When cells of intermediate differentiation fail to die or to differentiate further, the values of growth parameters determine whether growth is exponential or leads to a new equilibrium. The predictions of the model are sensitive to small differences in growth parameters. Failure of PCD and differentiation, leading to a new equilibrium number of cells, may explain many aspects of tumor behavior--for example, early premalignant lesions such as cervical intraepithelial neoplasia, the fact that some tumors very rarely become malignant, the observation of plateaux in the growth of some solid tumors, and, finally, long lag phases of growth until mutations arise that eventually result in exponential growth.

Nonrandom mating in the two-spot ladybird (Adalia bipunctata): the influence of weight on mating success

In some populations of the two-spot ladybird, Adalia bipunctata, there is evidence that melanics of both sexes have a general mating advantage over the typical morph. There is also evidence that some female ladybirds possess a simple mating preference for melanic males. We have determined the influence of body weight on mating success in the two-spot ladybird and assessed whether weight differences might influence the mating success of the melanic morph. We found that the "formal mating tests" used in previous studies of the two-spot ladybird did not detect any influence of weight on mating success. Using more sensitive "singleton" tests, however, a significant mating advantage to heavier males was detected, irrespective of morph. There was also evidence in favor of a similar advantage to heavy females, but this was not present in all ladybird groups that were tested. Heavy individuals neither produced more eggs in matings nor showed higher activity rates. We suggest that some form of mate competition favors large individuals: for example, the larger an individual, the greater is the chance of an encounter with a potential mate. There is no evidence in this study that either sex chooses heavier mates. Effects of weight might contribute to the general melanic mating advantage found in some populations but cannot account for all the data in favor of a female mating preference for melanic males.