Rebecca B. Bonner

Mapping of the distribution of significant proteins and proteoglycans in small intestinal submucosa by fluorescence microscopy

Because small intestine submucosa (SIS) is a bioscaffold for tissue regeneration, we describe a method to analyze the material for growth peptides and for structural molecules. Immunofluorescence methods are described for relative quantification of abundant structural proteins. Additionally, a quantitative technique for comparison of the content of less abundant proteins in SIS was developed using the tyramide signal amplification (TSA) system that is applicable to paraffin-preserved tissue blocks. Frozen sections generally shredded when cut thinly enough to permit entry and washout of reagents. Five micrometer sections cut from paraffin blocks were immunolabeled for collagen, heparan sulfate proteoglycans (HSPG), FGF2, TGFβ, and VEGF. Images of tissue sections were acquired by a linear image camera and quantified by densitometry after thresholding the signal to minimize nonspecific fluorescence. Immunohistochemistry was used to confirm the immunofluorescence methods. HSPG was widely distributed but concentrated in vessels. FGF2 was distributed diffusely and was associated with fibrous structures. VEGF was distributed mainly around vessels. TGFβ was barely detectable above background. Collagen fibrils were distinctly present, and with a two-color fluorescence system, the distribution of components relative to collagen can be assessed. The anatomic structure of SIS is likely to play an important role in the regeneration of tissues, and factors in remnant vessels may facilitate penetration of the matrix along these avenues.

Identification of a High Risk Subgroup of Grade 1 Transitional Cell Carcinoma Using Image Analysis Based Deoxyribonucleic Acid Ploidy Analysis of Tumor Tissue

The use of deoxyribonucleic acid (DNA) cytometry to identify a subset of patients with grade 1, stage Ta or T1 transitional cell carcinoma at high risk for death or recurrence was investigated in a retrospective study using paraffin blocks from 88 low grade transitional cell carcinomas of the bladder with an absorptiometric video-based image analysis system. Tumors were evaluated for ploidy (70 diploid, 16 aneuploid and 2 tetraploid) and the presence of cells with greater than 5C DNA. Survival analysis of 62 patients with adequate followup (15 to 20 years) showed that 43 of 62 (69%) suffered recurrences and 13 (21%) died of bladder cancer. The single most important predictors of death and recurrence were stem line aneuploidy and the presence of cells with greater than 5C DNA, respectively.

Quantitative changes in cytoskeletal and nuclear actins during cellular transformation

Actin, a highly conserved protein comprising cell stress fibers and other cellular structures, is found in both the cytoplasm and nucleus of cells and responds to both epigenetic signals and altered gene expression occurring during tumorigenesis. We have previously shown that changes in the cytoplasmic F‐ and G‐actin ratios reflect bladder cancer risk. To determine whether nuclear actin is also altered and how nuclear and cytoplasmic actin alterations are interrelated in transformation, an in vitro model of carcinogen‐induced transformation consisting of 2 human uroepithelial cell lines immortalized by infection with SV‐40 was studied. One line, HUC‐PC, is tumorigenic in nude mice after incubation with the carcinogen 4‐ABP, the other, HUC‐BC, is not. Cytoplasmic and nuclear F‐ and G‐actin were determined by QFIA on individual cells using fluorochrome‐labeled phallicidin and DNase, I, respectively. Before exposure to 4‐ABP, the PC cells had lower cytoplasmic F‐actin content, higher cytoplasmic G‐actin content, but similar levels of nuclear G‐ and F‐actin in comparison to the BC cells. After incubation with 4‐ABP, F‐actin decreased and G‐actin increased in both cytoplasm and nuclei of PC cells and cytoplasmic F‐actin fibers were lost, but only cytoplasmic actin was altered in the BC cells. Northern blot analysis showed the expression of the β‐actin gene was only approximately 20% lower in 4‐ABP‐treated PC cells than in untreated controls, indicating the cellular change in actin was attributed to a shift between F‐and G‐actin proteins rather than to net actin synthesis. Int. J. Cancer, 70:423–429, 1997.

G-ACTIN AS A RISK FACTOR AND MODULATABLE ENDPOINT FOR CANCER CHEMOPREVENTION TRIALS

Because tumorigenesis is an ongoing process, biomarkers can be used to identify individuals at risk for bladder cancer, and treatment of those at risk to prevent or slow further progression could be an effective means of cancer control given accurate individual risk assessment. Tumorigenesis proceeds through a series of defined phenotypic changes, including those in genetically altered cells destined to become cancer as well as in surrounding normal cells responding to the altered cytokine environment. A panel of biomarkers for the changes can provide a useful system for individual risk assessment in cancer patients and in individuals exposed to carcinogens. The use of such markers can increase the specificity of chemoprevention trials by targeting therapy to patients likely to respond, and thereby markedly reduce the costs of the trials.

Pan-cadherin as a high level phenotypic biomarker for prostate cancer.

PURPOSE High level phenotypic biomarkers such as cadherins are needed to identify individuals at risk for biologically active prostate cancer and determine which individuals with elevated prostate specific antigen or a prostate nodule are candidates for re-biopsy. Cadherins are a high level phenotypic biomarker associated with decreased cell adhesion, which is a cardinal event in carcinogenesis. Recently we reported that G-actin and tissue transglutaminase type II are potential biomarkers for prostate cancer. In this study we present cadherins as a potential third component of the biomarker profile. MATERIALS AND METHODS Prostate tissues from 38 patients with cancer and 33 controls with a 10-year prostate cancer-free followup were labeled for pan-cadherin by immunohistochemical testing. Immunoreactivity was quantified using a Pathology Workstation (Autocyte Inc., Elon College, North Carolina). RESULTS Visually benign glands from controls generally expressed cadherins, whereas regions of adenocarcinoma were generally negative. On quantitative immunohistochemistry 36 of 38 prostate cancer cases expressed a lower mean percent area positive for cadherin than the 19 benign prostatic hyperplasia and 14 prostatitis cases (odds ratio 978, 95% confidence interval 45 to 21,140, relative risk 18, 95% confidence interval 5 to 67, p <0.0001). Receiver operating characteristics analysis of immunohistochemical testing data showed that an optimal threshold of 7 produced 95% sensitivity and 100% specificity. CONCLUSIONS Quantitative down-regulation of cadherin expression in prostate cancer tissue sections is a strong biomarker for prostate cancer. Analysis of cadherin and other high level phenotypic biomarker expression in the premalignant field may provide additional diagnostic information to decide which patients need re-biopsy, more intensive monitoring or chemoprevention.

Neural net-based identification of cells expressing the p300 tumor-related antigen using fluorescence image analysis.

We report on preliminary investigations of the use of an image analysis system to perform preliminary algorithmic classification of images of fluorochrome-labeled cells followed by capture of gray-level images of potentially abnormal cells for analysis by a neural network. Cells were labeled with an antibody against a bladder cancer tumor-associated antigen, and the neural net was used to distinguish true-positive cells from negative cells, false-positive cells (autofluorescent or nonspecific labeling), and cell-sized artifacts. Gray-level cell images were digitized and processed for analysis by a feed-forward neural network using back-propagation. The network was trained and tested with two independent image sets. Various network configurations and activation functions were investigated, including a sinusoidal activation function. At high power, the network agreed completely with the human observers classification. At low power, a strong clustering of cells classified by the network with expert classification was seen, while the neural network showed roughly 75% concordance with the human observer. In addition, a set of four features extracted from raw cell images were investigated. The features were: shape factor, texture, area, and average pixel intensity. A network trained with these features performed better than one operating with gray-level images. We conclude that using neural networks to recognize and classify images captured by an image analysis microscope is feasible.

Complexity, Retinoid-Responsive Gene Networks, and Bladder Carcinogenesis

Carcinogenesis involves inactivation or subversion of the normal controls of proliferation, differentiation, and apoptosis. However, these controls are robust, redundant, and interlinked at the gene expression levels, regulation of mRNA lifetimes, transcription, and recycling of proteins. One of the central systems of control of proliferation, differentiation and apoptosis is retinoid signaling. The hRAR alpha nuclear receptor occupies a central position with respect to induction of gene transcription in that when bound to appropriate retinoid ligands, its homodimers and heterodimers with hRXR alpha regulate the transcription of a number of retinoid-responsive genes. These include genes in other signaling pathways, so that the whole forms a complex network. In this study we showed that simple, cause-effect interpretations in terms of hRAR alpha gene transcription being the central regulatory event would not describe the retinoid-responsive gene network. A set of cultured bladder-derived cells representing different stages of bladder tumorigenesis formed a model system. It consisted of 2 immortalized bladder cell lines (HUC-BC and HUC-PC), one squamous cell carcinoma cell line (SCaBER), one papilloma line (RT4), and 4 transitional cell carcinomas (TCC-Sup, 5637, T24, J82) of varying stages and grades. This set of cells were used to model the range of behaviors of bladder cancers. Relative gene expression before (constitutive) and after treatment with 10 microM all-trans-retinoic acid (aTRA) was measured for androgen and estrogen receptor; a set of genes involved with retinoid metabolism and action, hRAR alpha nd beta, hRXR alpha and beta CRBP, CRABP I and II; and for signaling genes that are known to be sensitive to retinoic acid, EGFR, cytokine MK, ICAM I and transglutaminase. The phenotype for inhibition of proliferation and for apoptotic response to both aTRA and the synthetic retinoid 4-HPR was determined. Transfection with a CAT-containing plasmid containing an aTRA-sensitive promoter was used to determine if the common retinoic acid responsive element (RARE)-dependent pathway for retinoid regulation of gene expression was active. Each of the genes selected is known from previous studies to react to aTRA in a certain way, either by up- or down-regulation of the message and protein. A complex data set not readily interpretable by simple cause and effect was observed. While all cell lines expressed high levels of the mRNAs for hRXR alpha and beta that were not altered by treatment with exogenous aTRA, constitutive and stimulated responses of the other genes varied widely among the cell lines. For example, CRABP I was not expressed by J82, T24, 5637 and RT4, but was expressed at low levels that did not change in SCaBER and at moderate levels that decreased, increased, or decreased sharply in HUC-BC, TCC-Sup and HUC-PC, respectively. The expression of hRAR alpha, which governs the expression of many retinoid-sensitive genes, was expressed at moderate to high levels in all cell lines, but in some it was sharply upregulated (TCC-Sup, HUC-PC and J82), remained constant (5637 and HUC-BC), or was down-regulated (SCaBER, T24 and RT4). The phenotypes for inhibition of proliferation showed no obvious relationship to the expression of any single gene, but cell lines that were inhibited by aTRA (HUC-BC and TCC-Sup) were not sensitive to 4-HPR, and vice versa. One line (RT4) was insensitive to either retinoid. Transfection showed very little retinoid-stimulated transfection of the CAT reporter gene with RT4 or HUC-PC. About 2-fold enhancement transactivation was observed with SCaBER, HUC-BC, J82 and T24 cells and 3-8 fold with 5637, TCC-Sup cells. In HUC-BC, a G to T point mutation was found at position 606 of the hRAR alpha gene. This mutation would substitute tyrosine for asparagine in a highly conserved domain. These data indicate that retinoid signaling is probably a frequent target of inactivation in bladder carcinogenesis. (ABSTRAC