Gastrointestinal Stem Cells and Cancer: Bridging the Molecular Gap

Cancer is believed to be a disease involving stem cells. The digestive tract has a very high cancer prevalence partly owing to rapid epithelial cell turnover and exposure to dietary toxins. Work on the hereditary cancer syndromes including familial adenomatous polyposis (FAP) has led to significant advances, including the adenoma-carcinoma sequence. The initial mutation involved in this stepwise progression is in the "gatekeeper" tumor suppressor gene adenomatous polyposis coli (APC). In FAP somatic, second hits in this gene are nonrandom events, selected for by the position of the germ-line mutation. Extensive work in both the mouse and human has shown that crypts are clonal units and mutated stem cells may develop a selective advantage, eventually forming a clonal crypt population by a process called "niche succession." Aberrant crypt foci are then formed by the longitudinal division of crypts into two daughter units--crypt fission. The early growth of adenomas is contentious with two main theories, the "top-down" and "bottom-up" hypotheses, attempting to explain the spread of dysplastic tissue in the bowel. Initial X chromosome inactivation studies suggested that colorectal tumors were monoclonal; however, work on a rare XO/XY human patient with FAP and chimeric Min mice showed that 76% of adenomas were polyclonal. A reduction in tumor multiplicity in the chimeric mouse model has been achieved by the introduction of a homozygous tumor resistance allele. This model has been used to suggest that short-range interaction between adjacent initiated crypts, not random polyp collision, is responsible for tumor polyclonality.

Redox behaviour of Pd-based TWCs under dynamic conditions: analysis using dispersive XAS and mass spectrometry

The redox behaviour of Pd-based TWCs is studied under dynamic, cycling conditions (e.g. lambda oscillations) on a 50 millisecond scale. Pd temporal response to gas inlet mixture changes is governed by metal-promoter interface properties.

Detecting differential gene expression with a semiparametric hierarchical mixture method

Mixture modeling provides an effective approach to the differential expression problem in microarray data analysis. Methods based on fully parametric mixture models are available, but lack of fit in some examples indicates that more flexible models may be beneficial. Existing, more flexible, mixture models work at the level of one-dimensional gene-specific summary statistics, and so when there are relatively few measurements per gene these methods may not provide sensitive detectors of differential expression. We propose a hierarchical mixture model to provide methodology that is both sensitive in detecting differential expression and sufficiently flexible to account for the complex variability of normalized microarray data. EM-based algorithms are used to fit both parametric and semiparametric versions of the model. We restrict attention to the two-sample comparison problem; an experiment involving Affymetrix microarrays and yeast translation provides the motivating case study. Gene-specific posterior probabilities of differential expression form the basis of statistical inference; they define short gene lists and false discovery rates. Compared to several competing methodologies, the proposed methodology exhibits good operating characteristics in a simulation study, on the analysis of spike-in data, and in a cross-validation calculation.

On parametric empirical Bayes methods for comparing multiple groups using replicated gene expression profiles

DNA microarrays provide for unprecedented large-scale views of gene expression and, as a result, have emerged as a fundamental measurement tool in the study of diverse biological systems. Statistical questions abound, but many traditional data analytic approaches do not apply, in large part because thousands of individual genes are measured with relatively little replication. Empirical Bayes methods provide a natural approach to microarray data analysis because they can significantly reduce the dimensionality of an inference problem while compensating for relatively few replicates by using information across the array. We propose a general empirical Bayes modelling approach which allows for replicate expression profiles in multiple conditions. The hierarchical mixture model accounts for differences among genes in their average expression levels, differential expression for a given gene among cell types, and measurement fluctuations. Two distinct parameterizations are considered: a model based on Gamma distributed measurements and one based on log-normally distributed measurements. False discovery rate and related operating characteristics of the methodology are assessed in a simulation study. We also show how the posterior odds of differential expression in one version of the model is related to the ratio of the arithmetic mean to the geometric mean of the two sample means. The methodology is used in a study of mammary cancer in the rat, where four distinct patterns of expression are possible.

On differential variability of expression ratios: improving statistical inference about gene expression changes from microarray data

We consider the problem of inferring fold changes in gene expression from cDNA microarray data. Standard procedures focus on the ratio of measured fluorescent intensities at each spot on the microarray, but to do so is to ignore the fact that the variation of such ratios is not constant. Estimates of gene expression changes are derived within a simple hierarchical model that accounts for measurement error and fluctuations in absolute gene expression levels. Significant gene expression changes are identified by deriving the posterior odds of change within a similar model. The methods are tested via simulation and are applied to a panel of Escherichia coli microarrays.

In vivo mutant frequency of thioguanine-resistant T-cells in the peripheral blood and lymph nodes of melanoma patients

T-cell activation by malignant melanoma would be anticipated to stimulate T-cell proliferation, which in turn has been associated with increasing the likelihood of somatic gene mutation. The purpose of this study was to test the hypothesis that in vivo hypoxanthine guanine phosphoribosyltransferase (hprt) mutant frequencies (MFs) are increased in peripheral blood T-cells from melanoma patients compared to normal controls. Assays were made of 48 peripheral blood samples from melanoma patients with stage 3 (13 patients) and stage 4 (35 patients) disease, 38 normal controls, and of nine tumor bearing lymph nodes. The mean hprt log(10)(MF) in patient peripheral blood was -4.77 (geometric mean hprt MF=17.0x10(-6)) compared to a mean hprt log(10)(MF) of -4.87 (geometric mean hprt MF=13.5x10(-6)) in controls. Although modest, this difference is statistically significant both by t-test (P=0.049) and after adjustment for covariates of age, gender, and cigarette smoking by regression analysis (P=0.001). Among the melanoma patients, the mean log(10)(MF) for the 17 patients who had received potentially genotoxic therapies was not significantly different from the mean log(10)(MF) for the 31 patients not receiving such therapies. The hprt MFs in the nine tumor bearing nodes were compared with MFs in peripheral blood from the same patients and revealed a non-significant (P=0.07) trend for increasing MFs in blood. Furthermore, analyses of T-cell receptor gene rearrangement patterns revealed hprt mutants originating from the same in vivo clone in both peripheral blood and a tumor-bearing node. The finding of elevated hprt MFs not entirely explained by genotoxic therapies in patients compared to controls can be explained either by hypermutability or in vivo T-cell activation. The similar MFs in peripheral blood and tumor bearing lymph nodes, as well as the finding of mutant representatives of the same in vivo T-cell clone in both locations, support monitoring peripheral blood to detect events in the nodes. If in vivo proliferation accounts for the current findings, the hprt deficient (hprt-) mutant fraction in blood may be enriched for T-cells that mediate the host immune response against malignant melanoma. Further studies will characterize the functional reactivity of hprt mutant isolates against melanoma-related antigens.

Genetic loci controlling breast cancer susceptibility in the Wistar-Kyoto rat

In this study, the Wistar-Kyoto (WKy) rat was genetically characterized for loci that modify susceptibility to mammary carcinogenesis. We used a genetic backcross between resistant WKy and susceptible Wistar-Furth (WF) rats as a panel for linkage mapping to genetically identify mammary carcinoma susceptibility (Mcs) loci underlying the resistance of the WKy rat. Rats were phenotyped for DMBA-induced mammary carcinomas and genotyped using microsatellite markers. To detect quantitative trait loci (QTL), we analyzed the genome scan data under both parametric and nonparametric distributional assumptions and used permutation tests to calculate significance thresholds. A generalized linear model analysis was also performed to test for interactions between significant QTL. This methodology was extended to identify interactions between the significant QTL and other genome locations. Chromosomes 5, 7, 10, and 14 were found to contain significant QTL, termed Mcs5, Mcs6, Mcs7, and Mcs8, respectively. The WKy alleles of Mcs5, -6, and -8 are associated with mammary carcinoma resistance; the WKy allele of Mcs7 is associated with an increased incidence of mammary cancer. In addition, we identified an interaction between Mcs8 and a region on chromosome 6 termed Mcsm1 (modifier of Mcs), which had no significant main effect on mammary cancer susceptibility in this genetic analysis.

Inferring the location and effect of tumor suppressor genes by instability-selection modeling of allelic-loss data

Cancerous tumor growth creates cells with abnormal DNA. Allelic-loss experiments identify genomic deletions in cancer cells, but sources of variation and intrinsic dependencies complicate inference about the location and effect of suppressor genes; such genes are the target of these experiments and are thought to be involved in tumor development. We investigate properties of an instability-selection model of allelic-loss data, including likelihood-based parameter estimation and hypothesis testing. By considering a special complete-data case, we derive an approximate calibration method for hypothesis tests of sporadic deletion. Parametric bootstrap and Bayesian computations are also developed. Data from three allelic-loss studies are reanalyzed to illustrate the methods.

Genome-wide loss of heterozygosity analysis of chemically induced rat hepatocellular carcinomas reveals elevated frequency of allelic imbalances on chromosomes 1, 6, 8, 11, 15, 17, and 20

Neoplastic development is a multistep process that involves the stochastic accumulation of heritable genetic alterations in proto-oncogenes, DNA repair genes, and tumor suppressor genes. Loss of heterozygosity (LOH) analysis has been used successfully to identify the genetic determinants of neoplastic development, including tumor suppressor genes, in several species and organs but not in the rat liver. We report the results of a sensitive genome-wide LOH analysis of rat hepatocellular carcinomas (HCCs). Heterozygous rats (Wistar-Furth x Fisher 344) were subjected to an Initiation-Promotion-Progression (IPP) protocol of hepatocarcinogenesis. Two weeks after initiation (by partial hepatectomy, 10 mg/kg diethylnitrosamine), the rats were placed on a diet containing 0.05% phenobarbital (PB). After 24 wk of PB promotion, the rats received either 100 or 1 50 mg/kg ethylnitrosourea. Hepatocellular tumors were resected after a total of 76wk of PB promotion. LOH analysis was completed on 26 HCCs by using 60 microsatellite markers covering all 20 rat autosomes and chromosome X. While 85% of the HCCs had one or more allelic imbalances, the average HCC had 3.3 allelic imbalances (range 0-9). A conditional hypothesis-testing method called the Hot-Cold model was used to determine the location of statistically significant elevations in the frequency of allelic imbalances. Elevated allelic imbalances were observed on chromosomes 1q, 6, 8, 11, 15, 17, and 20p. Together, these allelic imbalances suggest that the retinoblastoma and insulin-like growth factor genes as well as the resistance to chemical carcinogenesis (rcc) locus may be involved in HCC development in the rat but that LOH of the p53 gene is not. The elevated rate of allelic imbalances on chromosomes 8,11, and 17 may indicate the location of undiscovered tumor suppressor genes important to neoplastic development in rat liver. Microdissection-based LOH analysis of HCC revealed that contamination of non-neoplastic and nonhepatocellular tissue was not masking LOH in the whole-tumor analysis. There were no statistically significant differences in the frequency of allelic imbalances between HCC of any differentiation state (histological grade). To the degree that it does not reflect differences in etiological factors, the absence of allelic imbalances in chromosomal regions containing the p53 and mamose-6-phosphate/insulin-like growth factor II receptor tumor suppressor genes and the generally low frequency of allelic imbalances in these tumors, suggests that LOH and allelic imbalances play a less significant role in the molecular pathogenesis of HCC in rats than humans.

Genome-wide loss of heterozygosity analysis of chemically induced rat hepatocellular carcinomas reveals elevated frequency of allelic imbalances on chromosomes 1, 6, 8, 11, 15, 17, and 20

Neoplastic development is a multistep process that involves the stochastic accumulation of heritable genetic alterations in proto-oncogenes, DNA repair genes, and tumor suppressor genes. Loss of heterozygosity (LOH) analysis has been used successfully to identify the genetic determinants of neoplastic development, including tumor suppressor genes, in several species and organs but not in the rat liver. We report the results of a sensitive genome-wide LOH analysis of rat hepatocellular carcinomas (HCCs). Heterozygous rats (Wistar-Furth x Fisher 344) were subjected to an Initiation-Promotion-Progression (IPP) protocol of hepatocarcinogenesis. Two weeks after initiation (by partial hepatectomy, 10 mg/kg diethylnitrosamine), the rats were placed on a diet containing 0.05% phenobarbital (PB). After 24 wk of PB promotion, the rats received either 100 or 1 50 mg/kg ethylnitrosourea. Hepatocellular tumors were resected after a total of 76wk of PB promotion. LOH analysis was completed on 26 HCCs by using 60 microsatellite markers covering all 20 rat autosomes and chromosome X. While 85% of the HCCs had one or more allelic imbalances, the average HCC had 3.3 allelic imbalances (range 0-9). A conditional hypothesis-testing method called the Hot-Cold model was used to determine the location of statistically significant elevations in the frequency of allelic imbalances. Elevated allelic imbalances were observed on chromosomes 1q, 6, 8, 11, 15, 17, and 20p. Together, these allelic imbalances suggest that the retinoblastoma and insulin-like growth factor genes as well as the resistance to chemical carcinogenesis (rcc) locus may be involved in HCC development in the rat but that LOH of the p53 gene is not. The elevated rate of allelic imbalances on chromosomes 8,11, and 17 may indicate the location of undiscovered tumor suppressor genes important to neoplastic development in rat liver. Microdissection-based LOH analysis of HCC revealed that contamination of non-neoplastic and nonhepatocellular tissue was not masking LOH in the whole-tumor analysis. There were no statistically significant differences in the frequency of allelic imbalances between HCC of any differentiation state (histological grade). To the degree that it does not reflect differences in etiological factors, the absence of allelic imbalances in chromosomal regions containing the p53 and mamose-6-phosphate/insulin-like growth factor II receptor tumor suppressor genes and the generally low frequency of allelic imbalances in these tumors, suggests that LOH and allelic imbalances play a less significant role in the molecular pathogenesis of HCC in rats than humans.

Chemopreventive efficacy of combined piroxicam and difluoromethylornithine treatment of Apc mutant Min mouse adenomas, and selective toxicity against Apc mutant embryos

Genetic knockout or pharmacological inhibition of cyclooxygenase-2 decreases the number and size of adenomas in mouse models of familial adenomatous polyposis. Epidemiological and clinical studies in humans indicate that the entire class of nonsteroidal anti-inflammatory drugs (NSAIDs) that inhibit both COX-1 and COX-2 enzymes are promising colon cancer chemopreventive agents. We used the Apc mutant Min mouse model to test combinations of agents that might maximize preventive benefit with minimal toxicity because they act via different mechanisms. Min mice (n = 144) were exposed to low doses of the nonselective COX inhibitor piroxicam and the ornithine decarboxylase (ODC) inhibitor difluoromethylornithine (DFMO), beginning at the time they were weaned and continuing throughout the duration of the experiment. Piroxicam at 12, 25, and 50 ppm in the diet caused dose-dependent decreases in the number of tumors in the middle and distal portions of the small intestine. This decrease in tumor multiplicity was associated with a striking decrease in the size of those tumors that did grow out. In contrast, none of the doses of piroxicam alone decreased tumor multiplicity in the proximal portion of the intestine (duodenum). Exposure to DFMO (0.5 or 1.0% in water) caused a dose-dependent decrease in tumor multiplicity in the middle and distal portions of the small intestine. However, this decreased multiplicity was not associated with a striking decrease in the size of the tumors. Combined treatment of mice with piroxicam plus DFMO was much more effective than either agent alone and resulted in a significant number of mice totally free of any intestinal adenomas (P < 0.001), in contrast to the 100% incidence and high multiplicity in control Min mice. In addition to this profound effectiveness in reducing tumor number, the few residual tumors in mice treated with the combined drugs were markedly smaller in size than tumors that arose from control Min mice. These experiments suggest that selective COX-2 inhibition combined with ODC inhibition is a very promising approach for colon cancer prevention. These COX-2 and ODC inhibitor drugs were not overtly toxic at the doses used when administered to mice after weaning. However, when treatment was begun in utero, the Mendelian expected progeny ratio of 1:1 that we routinely obtained in untreated control litters was no longer observed. Apc(min)/+ progeny of pregnant dams treated with piroxicam and/or DFMO were reduced in number and their ratio to Apc+/+ progeny was decreased to approximately 0.28:1. Thus, these agents are effective against adenomas that have homozygous mutation of the APC gene and also select against fetuses bearing a heterozygous mutation in the APC gene.

p16/pRb pathway alterations are required for bypassing senescence in human prostate epithelial cells

The cell cycle regulatory genes p16/CDKN2 and RB are frequently deleted in prostate cancers. In this study, we examined the role of alterations in p16 and pRb during growth, senescence, and immortalization in vitro of human prostate epithelial cells (HPECs). HPECs are established from normal prostate tissues and cultured on collagen-coated dishes. Our results show that p16 is reproducibly elevated at senescence in HPECs. HPECs are immortalized using human papilloma virus 16 E6 and/or E7 as molecular tools to inactivate p53 and/or pRb, respectively. Immortalization occurs infrequently in this system and only after a latent period during which additional genetic/epigenetic changes are thought to occur. Notably, all of the E6-immortalized HPEC lines but none of the E7 lines show inactivation of p16/CDKN2 (by deletion, methylation, or mutation) in association with immortalization. In contrast, E7 lines, in which pRb function is abrogated by E7 binding, retain the high levels of p16 observed at senescence. Thus, all lines show either a p16 or pRb inactivation. Analysis of six independent lines from metastatic prostate cancers reveals a similar loss of either p16 or pRb. Comparative genomic hybridization of HPECs shows that gains of chromosomes 5q, 8q, and 20 are nonrandomly associated with bypassing senescence (probability = 0.95). These results suggest that high levels of the cyclin-dependent kinase inhibitor p16 mediate senescence G1 arrest in HPECs and that bypassing this block by a p16/pRb pathway alteration is required for immortalization in vitro and possibly tumorigenesis in vivo. Our results further indicate that inactivation of the p16/pRb pathway alone is not sufficient to immortalize HPECs and that additional genetic alterations are required for this process.

A study of deleterious gene structure in plants using Markov chain Monte Carlo

The characteristics of deleterious genes have been of great interest in both theory and practice in genetics. Because of the complex genetic mechanism of these deleterious genes, most current studies try to estimate the overall magnitude of mortality effects on a population, which is characterized classically by the number of lethal equivalents. This number is a combination of several parameters, each of which has a distinct biological effect on genetic mortality. In conservation and breeding programs, it is important to be able to distinguish among different combinations of these parameters that lead to the same number of lethal equivalents, such as a large number of mildly deleterious genes or a few lethal genes, The ability to distinguish such parameter combinations requires more than one generation of mating. We propose a model for survival data from a two-generation mating experiment on the plant species Brassica rapa, and we enable inference with Markov chain Monte Carlo. This computational strategy is effective because a vast amount of missing genotype information must be accounted for. In addition to the lethal equivalents, the two-generation data provide separate information on the average intensity of mortality and the average number of deleterious genes carried by an individual. In our Markov chain Monte Carlo algorithm, we use a vector proposal distribution to overcome inefficiency of a single-site Gibbs sampler. Information about environmental effects is obtained from an outcrossing experiment conducted in parallel with the two-generation mating experiments.

An estimation method for the semiparametric mixed effects model

A semiparametric mixed effects regression model is proposed for the analysis of clustered or longitudinal data with continuous, ordinal, or binary outcome. The common assumption of Gaussian random effects is relaxed by using a predictive recursion method (Newton and Zhang, 1999) to provide a nonparametric smooth density estimate. A new strategy is introduced to accelerate the algorithm. Parameter estimates are obtained by maximizing the marginal profile likelihood by Powell's conjugate direction search method. Monte Carlo results are presented to show that the method can improve the mean squared error of the fixed effects estimators when the random effects distribution is not Gaussian. The usefulness of visualizing the random effects density itself is illustrated in the analysis of data from the Wisconsin Sleep Survey. The proposed estimation procedure is computationally feasible for quite large data sets.

Bayesian phylogenetic inference via Markov chain Monte Carlo methods

We derive a Markov chain to sample from the posterior distribution for a phylogenetic tree given sequence information from the corresponding set of organisms, a stochastic model for these data, and a prior distribution on the space of trees. A transformation of the tree into a canonical cophenetic matrix form suggests a simple and effective proposal distribution for selecting candidate trees close to the current tree in the chain. We illustrate the algorithm with restriction site data on 9 plant species, then extend to DNA sequences from 32 species of fish. The algorithm mixes well in both examples from random starting trees, generating reproducible estimates and credible sets for the path of evolution.

A comparative analysis of allelic imbalance events in chemically induced rat mammary, colon, and bladder tumors

In this paper, patterns of allelic imbalances (Als) in chemically induced rat mammary, colon, and bladder tumors from (Wistar Furth x Fischer 344)F1 rats are described and compared. Male F1 rats were administered azoxymethane (AOM), and colon tumors were collected at 58 wk after treatment. Female F1 rats were given either N-nitroso-N-methylurea (NMU) or N-butyl-(hydroxybutyl)-nitrosoamine (BBN), and mammary and bladder tumors were collected at 15 and 52 wk after treatment, respectively. DNA was extracted from a subset of 18 of the largest tumors from each group, and a genome scan was performed by using polymerase chain reaction and 90 polymorphic microsatellite markers. Als, such as loss of heterozygosity, gene duplication, and microsatellite instability, were observed at low frequencies in all of the tumor models. Thirty random Als were observed in the AOM-induced colon tumors but only four in the NMU-induced mammary tumors. In both these models, all the tumors were classified as adenocarcinomas, and most of the Als observed were confined to single tumors with atypical histopathology. In contrast, 27 random Als were identified in the BBN-induced bladder tumors. Als were observed in both transitional-cell carcinomas and papillomas, although most were in the carcinomas. Statistical analysis of the Al data revealed no significant nonrandom Als within or among the tumor models, although several of the infrequently observed Al events identified in the rat tumors may also be observed in the corresponding human tumor type.

A randomized, placebo-controlled trial of low-dose alpha-difluoromethylornithine in individuals at risk for colorectal cancer

DFMO is an irreversible inhibitor of ornithine decarboxylase (ODC), the key enzyme in mammalian polyamine biosynthesis. The goal of this study was to determine the effects of DFMO 0.5 g/m2/day as a single oral dose on polyamine and ODC levels in rectal, rectosigmoidal, and cecal colonic mucosae of individuals at risk for colon cancer because of a personal history of adenomatous polyps of the colon or a family history of colon cancer in at least one first-degree relative. A second goal was to determine toxicity of this treatment given over 1 year. Forty-five randomized subjects had a flexible sigmoidoscopy with no preparation and a colonoscopy after lavage preparation at baseline, a sigmoidoscopy with no preparation after 3 months, and both procedures (as at baseline) after 12 months, with mucosal biopsies taken from the rectosigmoid area (sigmoidoscopy) or rectal and cecal areas (colonoscopy) for evaluations of ODC and polyamine levels. Significantly decreased levels of putrescine and spermidine were found in rectosigmoid colonic mucosae of DFMO-treated (n = 24) compared with placebo (n = 21) subjects at 3 months (P = 0.03 and 0.04) and 12 months (P = 0.005, P = 0.004). Similar trends, none reaching statistical significance, were found for individual polyamine levels in rectal and cecal mucosae. No significant differences in ODC levels were detected marginally. There was evidence of global suppression of ODC and polyamine levels in the treatment group (P = 0.035). Three DFMO recipients (12.5%) developed clinically noticeable and audiologically demonstrated hearing loss, which was reversible and attributed to DFMO after 3 months (two subjects) and 12 months (one subject). The tissue polyamine changes demonstrated in this study are consistent with findings in other studies in colon and other tissues. The ototoxicity findings here suggest that investigation of other DFMO schedules, such as ones with a drug "holiday," will be a necessary step before Phase III chemoprevention studies can be pursued.

A resistant genetic background leading to incomplete penetrance of intestinal neoplasia and reduced loss of heterozygosity in ApcMin/+ mice

Previous studies of Min/+ (multiple intestinal neoplasia) mice on a sensitive genetic background, C57BL/6 (B6), showed that adenomas have lost heterozygosity for the germ-line ApcMin mutation in the Apc (adenomatous polyposis coli) gene. We now report that on a strongly resistant genetic background, AKR/J (AKR), Min-induced adenoma multiplicity is reduced by about two orders of magnitude compared with that observed on the B6 background. Somatic treatment with a strong mutagen increases tumor number in AKR Min/+ mice in an age-dependent manner, similar to results previously reported for B6 Min/+ mice. Immunohistochemical analyses indicate that Apc expression is suppressed in all intestinal tumors from both untreated and treated AKR Min/+ mice. However, the mechanism of Apc inactivation in AKR Min/+ mice often differs from that observed for B6 Min/+ mice. Although loss of heterozygosity is observed in some tumors, a significant percentage of tumors showed neither loss of heterozygosity nor Apc truncation mutations. These results extend our understanding of the effects of genetic background on Min-induced tumorigenesis in several ways. First, the AKR strain carries modifiers of Min in addition to Mom1. This combination of AKR modifiers can almost completely suppress spontaneous intestinal tumorigenesis associated with the Min mutation. Second, even on such a highly resistant genetic background, tumor formation continues to involve an absence of Apc function. The means by which Apc function is inactivated is affected by genetic background. Possible scenarios are discussed.

On the statistical analysis of allelic-loss data

This paper concerns the statistical analysis of certain binary data arising in molecular studies of cancer. In allelic-loss experiments, tumour cell genomes are analysed at informative molecular marker loci to identify deleted chromosomal regions. The resulting binary data are used to infer properties of putative suppressor genes, genes involved in normal cell cycling. Various factors can complicate this inference, including background loss of heterozygosity, spatial (that is, within chromosome) dependence of the binary responses, non-informativeness of markers, covariates such as protein levels or tumour histology, heterogeneity of cells within tumours, and measurement error. We focus on the first three factors, discussing methods for statistical inference that separate background loss from significant loss. We outline the extension to other inferences, such as comparison questions and the relationship to covariates. Using characteristic features of tumourigenesis, we present a framework for the stochastic modelling of allelic-loss data, and build models within this framework; in particular, we propose a simple model that has chromosome breaks at locations of a Poisson process, and preferential selection cells with inactivated suppressor genes. We illustrate these methods on allelic-loss data from induced rat mammary tumours and human bladder cancers.

The intestinal epithelium and its neoplasms: genetic, cellular and tissue interactions

The Min (multiple intestinal neoplasia) strain of the laboratory mouse and its derivatives permit the fundamental study of factors that regulate the transition between normal and neoplastic growth. A gene of central importance in mediating these alternative patterns of growth is Apc, the mouse homologue of the human adenomatous polyposis coli (APC) gene. When adenomas form in the Min mouse, both copies of the Apc gene must be inactivated. One copy is mutated by the nonsense Apc allele carried in heterozygous form in this strain. The other copy can be silenced by any of several mechanisms. These range from loss of the homologue bearing the wild-type Apc allele; to interstitial deletions surrounding the wild-type allele; to intragenic mutation, including nonsense alleles; and finally, to a reduction in expression of the locus, perhaps owing to mutation in a regulatory locus. Each of these proposed mechanisms may constitute a two-hit genetic process as initially posited by Knudson; however, apparently the two hits could involve either a single locus or two loci. The kinetic order for the transition to adenoma may be still higher than two, if polyclonal adenomas require stronger interactions than passive fusion. The severity of the intestinal neoplastic phenotype of the Min mouse is strongly dependent upon loci other than Apc. One of these, Mom1, has now been rigorously identified at the molecular level as encoding an active resistance conferred by a secretory phospholipase. Mom1 acts locally within a crypt lineage, not systemically. Within the crypt lineage, however, its action seems to be non-autonomous: when tumours arise in Mom1 heterozygotes, the active resistance allele is maintained in the tumour (MOH or maintenance of heterozygosity). Indeed, the secretory phospholipase is synthesized by post-mitotic Paneth cells, not by the proliferative cells that presumably generate the tumour. An analysis of autonomy of modifier gene action in chimeric mice deserves detailed attention both to the number of genetic factors for which an animal is chimeric and to the clonal structure of the tissue in question. Beyond Mom1, other loci can strongly modify the severity of the Min phenotype. An emergent challenge is to find ways to identify the full set of genes that interact with the intestinal cancer predisposition of the Min mouse strain. With such a set, one can then work, using contemporary mouse genetics, to identify the molecular, cellular and organismal strategies that integrate their functions. Finally, with appropriately phenotyped human families, one can investigate by a candidate approach which modifying factors influence the epidemiology of human colon cancer. Even if a candidate modifier does not explain any of the genetic epidemiology of colon cancer in human populations, modifier activities discovered by mouse genetics provide candidates for chemopreventive and/or therapeutic modalities in the human.

Genetic identification of multiple loci that control breast cancer susceptibility in the rat

We have used a rat model of induced mammary carcinomas in an effort to identify breast cancer susceptibility genes. Using genetic crosses between the carcinoma-resistant Copenhagen (COP) and carcinoma-sensitive Wistar-Furth rats, we have confirmed the identification of the Mcs1 locus that modulates tumor number. We have now also identified two additional loci, Mcs2 and Mcs3. These three loci map to chromosomes 2, 7, and 1, respectively, and interact additively to suppress mammary carcinoma development in the COP strain. They are responsible for a major portion of the tumor-resistant phenotype of the COP rat. No loss of heterozygosity was observed surrounding the three loci. A fourth COP locus, Mcs4, has also been identified on chromosome 8 and acts in contrast to increase the number of carcinomas. These results show that mammary carcinoma susceptibility in the COP rat is a polygenic trait. Interestingly, a polymorphism in the human genomic region homologous to the rat Mcs4 region is associated with an increased breast cancer risk in African-American women. The isolation of the Mcs genes may help elucidate novel mechanisms of carcinogenesis, provide information important for human breast cancer risk estimation, and also provide unique drug discovery targets for breast cancer prevention.

Overcoming cellular senescence in human cancer pathogenesis

Elevation of p16, the CDKN2/p16 tumor suppressor gene (TSG) product, occurs at senescence in normal human uroepithelial cells (HUC). Immortal HUCs and bladder cancer cell lines show either alteration of p16 or pRb, the product of the retinoblastoma (RB) TSG. In addition, many human cancers show p16 or pRb alteration along with other genetic alterations that we associated with immortalization, including +20q and -3p. These observations led us to hypothesize that p16 elevation plays a critical role in senescence cell cycle arrest and that overcoming this block is an important step in tumorigenesis in vivo, as well as immortalization in vitro. Using a novel approach, we tested these hypotheses in the present study by examining p16 and pRb status in primary culture (P0) and after passage in vitro of transitional cell carcinoma (TCC) biopsies that represented both superficial bladder tumors and invasive bladder cancers. We demonstrated that all superficial TCCs showed elevated p16 after limited passage in vitro and then senesced, like normal HUCs. In contrast, all muscle invasive TCCs contained either a p16 or a pRb alteration at P0 and all spontaneously bypassed senescence (P = 0.001). Comparative genomic hybridization (CGH) was used to identify regions of chromosome loss or gain in all TCC samples. The application of a statistical model to the CGH data showed a high probability of elevated alteration rates of +20q11-q12 (0.99) and +8p22-pter (0.94) in the immortal muscle invasive TCCs, and of -9q (0.99) in the superficial TCCs. Three myoinvasive TCCs lost 3p13-p14. In this study, four of six myoinvasive TCCs also showed TP53 mutation that associated well with genome instability (P = 0.001), as previously hypothesized. Notably, TP53 mutation, which has been used as a marker of tumor progression in many human cancers, was less significant in associating with progression in this study (P = 0.04) than was p16 or pRb alteration (P = 0.001). Thus, these data support a new model in which overcoming senescence plays a critical role in human cancer pathogenesis and requires at least two genetic changes that occur in several combinations that can include either p16 or pRb loss and at least one additional alteration, such as +20q11-q12, -3p13-p14, or -8p21-pter.

Analysis of aerial survey data on Florida manatee using Markov chain Monte Carlo

We assess population trends of the Atlantic coast population of Florida manatee, Trichechus manatus latirostris, by reanalyzing aerial survey data collected between 1982 and 1992. To do so, we develop an explicit biological model that accounts for the method by which the manatees are counted, the mammals' movement between surveys, and the behavior of the population total over time. Bayesian inference, enabled by Markov chain Monte Carlo, is used to combine the survey data with the biological model. We compute marginal posterior distributions for all model parameters and predictive distributions for future counts. Several conclusions, such as a decreasing population growth rate and low sighting probabilities, are consistent across different prior specifications.

Intestinal neoplasia in the ApcMin mouse: independence from the microbial and natural killer (beige locus) status

We have tested the hypothesis that enteric bacteria are necessary for formation of intestinal adenomas in C57BL/6-ApcMin/+ mouse. Germ-free mice developed 2-fold fewer adenomas than conventional controls in the medial small intestine (7.3 versus 14.9; P < 0.003), but there were no significant differences in the rest of the intestinal tract. We conclude that microbial status does not strongly alter the adenoma phenotype in this mouse model of familial adenomatous polyposis. In parallel, we have found that C57BL/6-ApcMin/+ mice mutated at the beige locus, which controls natural killer activity, are also unaltered in adenoma multiplicity.

20q gain associates with immortalization: 20q13.2 amplification correlates with genome instability in human papillomavirus 16 E7 transformed human uroepithelial cells

Breast, bladder, colon, and ovarian carcinomas show frequent low level 20q gain and less frequently high level 20q13.2 amplification, but the significance of these 20q amplifications in transformation has not been defined. Using karyotypic and comparative genomic hybridization (CGH) analyses, chromosome losses and gains were analysed in six newly immortalized human uroepithelial cell (HUC) lines transformed by Human Papillomavirus 16 (HPV16) E7. Results showed clonal chromosomes with 20q11->qter gain in all six lines. CGH revealed a peak of 20q13.2 amplification in two cell lines. FISH with whole chromosome 20 paint showed expanded chromosome regions (ECRs) and double minute chromosomes (DMs) that contained chromosome 20 material in cell lines with 20q13.2 amplification. FISH with probes from the center of the 20q13.2 human breast cancer amplicon showed as many as 24 signals in cells with 20q13.2 amplification. The acquisition of genome instability in these E7-HUCs did not correlate with TP53 mutation, as all E7-HUCs contained only wildtype TP53. These results suggest that low level 20q gain is associated with overcoming cellular senescence in E7 transformed cells (P-value=2 x 10(-7)), but does not confer genome instability, while high level 20q13.2 amplification is associated with chromosome instability. Loss of 10p (P-value = 3 x 10(-5)) was also important in immortalization of E7-transformed HUCs. Thus, these results have profound implications for interpreting the significance of high versus low level 20q gains in human cancers.

Allelic imbalance in mammary carcinomas induced by either 7,12-dimethylbenz[a]anthracene or ionizing radiation in rats carrying genes conferring differential susceptibilities to mammary carcinogenesis

To identify and compare the genetic lesions associated with tumorigenesis in rats carrying the mammary carcinoma suppressor (MCS) 1 gene, we induced mammary carcinomas in (Wistar Furth (WF) x Copenhagen (Cop))F1 rats by using either 7,12-dimethylbenz[a]anthracene (DMBA) or radiation. The tumors were screened for allelic imbalances by using polymerase chain reaction and 65 polymorphic microsatellite markers spanning the genome. No allelic imbalance was detected at the mapped location of MCS-1 on chromosome 2; however, a scan of the genome revealed random allelic imbalances in the radiation-induced tumors. In addition, non-random loss of heterozygosity (LOH) on chromosome 1 in the DMBA-induced tumors was documented. We then screened three other subsets of DMBA- and radiation-induced mammary carcinomas from (WF x Fischer (F344))F1, (Wistar Kyoto x F344)F1, and (F344 x Cop)F1 rats for imbalance on chromosomes 1 and 2. No allelic imbalance was detected in the MCS-1 region of chromosome 2 in any of the tumors screened. Nonrandom imbalance on chromosome 1 was detected but only in the DMBA-induced tumors from the (F344 x Cop)F1 rats. Thus, only Cop-derived F1 rats have mammary tumors with the chromosome 1 imbalance; however, the imbalance does not favor the Cop parental allele. We also analyzed the DMBA-induced tumors with LOH at chromosome 1 for Ha-ras codon 61 mutation and found no association. These results suggest that loss of the MCS-1 Cop allele is not required for tumor formation, that the genetic background of the F1 rat appears to influence the type of genetic lesion identified in the mammary tumors, and that there is no association between Ha-ras codon 61 mutation and chromosome 1 imbalance in our model system.

A bayesian approach to detect quantitative trait loci using Markov chain Monte Carlo

Markov chain Monte Carlo (MCMC) techniques are applied to simultaneously identify multiple quantitative trait loci (QTL) and the magnitude of their effects. Using a Bayesian approach a multi-locus model is fit to quantitative trait and molecular marker data, instead of fitting one locus at a time. The phenotypic trait is modeled as a linear function of the additive and dominance effects of the unknown QTL genotypes. Inference summaries for the locations of the QTL and their effects are derived from the corresponding marginal posterior densities obtained by integrating the likelihood, rather than by optimizing the joint likelihood surface. This is done using MCMC by treating the unknown QTL, genotypes, and any missing marker genotypes, as augmented data and then by including these unknowns in the Markov chain cycle alone with the unknown parameters. Parameter estimates are obtained as means of the corresponding marginal posterior densities. High posterior density regions of the marginal densities are obtained as confidence regions. We examine flowering time data from double haploid progeny of Brassica napus to illustrate the proposed method.

Chemoprevention of spontaneous intestinal adenomas in the Apc Min mouse model by the nonsteroidal anti-inflammatory drug piroxicam

C57BL/6J-Min/+mice (n = 56), heterozygous for a nonsense mutation in the Apc gene, were randomized at weaning to seven groups, including groups treated with piroxicam at 0, 50, 100, and 200 ppm in the AIN93G diet. After only 6 weeks of treatment, intestinal adenomas and aberrant crypt foci were counted, and serum levels of piroxicam and thromboxane B2 were quantitated. Tumor multiplicity was decreased in a dose-dependent manner from 17.3 +/- 2.7 in the control to 2.1 +/- 1.1 (12%) in the high-dose piroxicam group (P < 0.001). Thromboxane B2 levels in plasma also decreased monotonically in parallel to the decrease in tumor multiplicity, consistent with the prostaglandin inhibitory effect of piroxicam. The Min mouse model demonstrates that the nonsteroidal anti-inflammatory drug piroxicam has strong biological and therapeutic effects, potentially useful for prevention of the early adenoma stage of tumor development.

Hematuria home screening: repeat testing results

To determine at what interval screening should be repeated to detect bladder cancer before it becomes muscle invasive 856 men who had 14 negative daily home tests for hematuria with a chemical reagent strip 9 months previously performed repeat tests. Of these men 50 (5.8%) had at least 1 positive test during the second 14-day screening period and 38 were evaluated, 15 of whom (39.5%) had significant urological pathological conditions, including 8 with malignancies. Bladder cancer was noted in 7 men, with no tumor invading the muscularis propria. The finding of 7 bladder cancers in 856 men (0.82%) who had a negative test 9 months previously indicates that bladder cancer has a brief preclinical duration and that testing must be repeated at least annually for screening to detect bladder cancer consistently before invasion occurs.

Comparison of bladder cancer outcome in men undergoing hematuria home screening versus those with standard clinical presentations

OBJECTIVES

Because repetitive hematuria home screening with a chemical reagent strip can detect early stage bladder cancer (BC) in asymptomatic middle-aged and elderly men, the ability of this screening to effect earlier detection and reduce BC mortality was investigated.

METHODS

Grades, stages, and outcomes of BCs detected by hematuria screening in 1575 men were compared with those of all newly diagnosed BCs in men age 50 years or older reported to the Wisconsin cancer registry in 1988. BC grades and stages were assigned by review of all pathology slides/blocks, and causes of deaths were determined from cancer registry records. As an additional control group, outcomes of BC cases diagnosed in men solicited to take part in screening, who declined, were also determined.

RESULTS

The proportions of low-grade (grades 1 and 2) superficial (Stages Ta and T1) versus high-grade (grade 3) or invasive (Stage T2 or higher) cancers in cases detected by hematuria screening (screened cases) and those reported to the tumor registry (unscreened cases) were not significantly different (52.4% versus 47.7% in 21 screened and 56.8% versus 43.3% in 511 unscreened cases) (P > 0.20). Of the high-grade or invasive cases, however, the proportion of late stage (T2 or higher) tumors was significantly lower in the screening-detected BCs compared to unscreened ones (P = 0.007). No screened case has died of BC (3- to 9-year follow-up), whereas 16.4% of unscreened cases have within 2 years of diagnosis (P = 0.025). Twenty-three of 1940 (1.2%) men who were solicited but chose not to participate in screening were diagnosed with BC within 18 months after what would have been their last home screening date, compared with 1.3% of participants having BC detected by screening. Thus, screening participants and those who were solicited and declined had similar likelihoods of developing BC.

CONCLUSIONS

Hematuria home screening detects high-grade cancers before they become muscle invading and significantly reduces BC mortality.

Long-term genome stability and minimal genotypic and phenotypic alterations in HPV16 E7-, but not E6-, immortalized human uroepithelial cells

Parameters of genome instability and morphological alterations associated with cell transformation were studied in an isogeneic set of clonal human uroepithelial cell (HUC) lines immortalized by the human papilloma virus 16 (HPV16) E6 and/or E7 gene(s). HPV16 E6 binds p53, leading to rapid degradation of p53, whereas E7 binds and alters pRb and other proteins. We report that two independent E7-immortalized HUC lines showed minimal phenotypic or genotypic alterations, except that both lines contained amplification of 20q DNA sequences and a greater polyploidization at an early passage. The E7-immortalized HUC line resembled normal HUC lines, except that they failed to senesce. In contrast, the E6-immortalized HUC lines were morphologically altered, contained numerous random chromosome aberrations, and showed unstable evolving karyotypes with passage in culture. No amplified DNA sequences were detected in E6-immortalized HUC lines. Instead, clonal losses of chromosome regions (i.e., -3p, -6q, -9p), putatively containing tumor suppressor or senescence genes, accompanied the E6-HUC immortalization event. E6-immortalized HUC lines showed transformed phenotypes similar to E6/E7-HUC lines. The difference in genome stability between E6- and E7-immortalized HUC was highly significant statistically (p-value < 10(-6). Thus, the HPV16 E7 gene led to HUC immortalization by a pathway that blocked cellular senescence, but did not disrupt genome stability. These results implicate p53 loss, but not pRb alteration, in genome destabilization.

Assessing the significance of chromosome-loss data: where are suppressor genes for bladder cancer?

Cytogenetic analysis reveals alterations of chromosome structure (losses, gains, and rearrangements of genetic material) in bladder cancer cells generated using an in vitro/in vivo transformation system. To predict possible locations of bladder cancer suppressor genes, we performed a robust Bayesian analysis of the chromosome-loss data. We postulated a simple stochastic model to describe chromosome loss during tumour progression. Posterior computations are enabled by a dynamic simulation algorithm. Ordered by decreasing posterior probability of putatively harbouring a suppressor gene, we observe significant losses on chromosomes 3, 18, 13, 10, 11, and y.

Behavior of feline hematopoietic stem cells years after busulfan exposure

The kinetics of hematopoietic stem cells were investigated in glucose-6-phosphate dehydrogenase (G-6-PD) heterozygous cats treated with dimethylbusulfan. Because of X-chromosome inactivation during embryogenesis, each somatic cell from these animals contains either maternal- or paternal-type G-6-PD. Therefore, all hematopoietic progenitor cells carry the G-6-PD phenotype of the most primitive cell (stem cell) from which they originate. For up to 6.5 years after dimethylbusulfan therapy, we determined the percentages of erythroid and granulocyte/macrophage progenitor cells with each G-6-PD phenotype. Significant variations were seen in studies from five of six cats, showing that the population of stem cells contributing to hematopoiesis was neither large nor constant. With mathematical analyses, we estimated that the proliferative potential of residual stem cells was much less than that of normal stem cells reduced in number by autologous transplantation (Abkowitz et al, Proc Natl Acad Sci USA 87:9062, 1990). There was no evidence for the regeneration of a normal stem cell reserve over time; rather, damage was most pronounced years after dimethylbusulfan exposure. These data may help explain the high clinical incidence of aplastic anemia and myelodysplasia after alkylating agent therapies.

Mammary carcinoma suppressor and susceptibility genes in the Wistar-Kyoto rat

The Copenhagen (Cop) rat carries a homozygous mammary carcinoma suppressor (MCS) gene that prevents both spontaneous and induced mammary cancer. Here we identify an additional rat strain, the Wistar-Kyoto (WKy), that is resistant to the development of mammary carcinomas. This rat strain is similar to Cop in having one homozygous copy of a mammary suppressor gene. The Cop MCS gene and the gene responsible for mammary cancer resistance in the WKy were found to be very closely linked genetically and thus are likely to be the same gene. It was found that, unlike the Cop strain, the WKy strain also carries multiple copies of a susceptibility gene(s). This is analogous to the susceptible Wistar-Furth strain, which carries three copies of dominant susceptibility genes. The MCS gene is epistatic in regard to these susceptibility genes. The availability of the WKy rat strain carrying MCS on a background independent of the Cop strain will aid both mechanistic studies and the molecular cloning of MCS.

Evidence for the maintenance of hematopoiesis in a large animal by the sequential activation of stem-cell clones

To test if hematopoiesis can be maintained by the sequential activation of stem-cell clones, we performed autologous marrow transplantations with limited numbers of cells in cats heterozygous for the X chromosome-linked enzyme glucose-6-phosphate dehydrogenase (G6PD) and observed the G6PD phenotypes of erythroid and granulocyte/macrophage progenitors over time. The animals were the female offspring of Geoffroy male and domestic female cats. In repeated studies of marrow from control animals (n = 5) or experimental animals prior to transplantation (n = 3), the percent of progenitors with domestic-type G6PD did not vary. After transplantation, the peripheral blood counts, marrow morphologies, frequencies of progenitors, and progenitor cell cycle kinetics returned to normal. However, abrupt and significant fluctuations were seen in the G6PD type of progenitors from each cat during the 1-1.5 years of observation. These data cannot be explained if there were either a large or constant population of active stem cells and thus imply, in a large-animal system, that hematopoiesis was maintained through clonal succession. A stochastic model was developed to estimate the numbers of active clones and their mean lifetimes.

A stochastic model for haematopoiesis in cats

Haematopoiesis is the process by which progenitor cells differentiate into competent mature blood cells. Only those cells in the latter stages of haematopoiesis can be observed in vitro, so theories about the early stages of the process cannot be tested directly. Experimental data of bone marrow samples from Safari cats provides evidence for the clonal succession hypothesis of early haematopoiesis. In this paper, a hidden Markov model is constructed to quantify this support. Recursive updating techniques are derived and are used to calculate the likelihood and to construct fitted values for the model. Inference is based on multimodal likelihood surfaces.