The alpha v beta 6 integrin promotes proliferation of colon carcinoma cells through a unique region of the beta 6 cytoplasmic domain.

Cell-matrix interactions are assumed to be important in regulating differentiation and tumor cell growth; however, the precise roles of individual matrix receptors in producing cellular responses are still unclear. We have previously described the alpha v beta 6 integrin, an epithelial cell fibronectin receptor expressed in many carcinoma cell lines. Here we show that heterologous expression of alpha v beta 6 in a human colon carcinoma cell line (SW480) enhances the proliferative capacity of these cells, both in vitro and in vivo in nude mice. This property of alpha v beta 6 correlates with the presence of an 11-amino acid region at the COOH terminus of the beta 6 cytoplasmic domain. This 11-amino acid sequence is required for the growth stimulatory effect, but not for other functions of the beta 6 cytoplasmic domain, such as promoting cell adhesion and focal contact localization.

Colorectal cancer and the integrin family of cell adhesion receptors: current status and future directions.

Tumour progression is thought to be determined, at least in part, by the balance between available cell surface receptors and the nature of the surrounding extracellular matrix. The integrin family of transmembrane adhesion receptors involved in tumour cell-matrix interactions mediates cell adhesion, migration, and differentiation. Certain patterns of integrin receptor expression on normal and malignant colon epithelial cells are emerging, and it is now clear that integrins can also regulate such divergent processes as cell proliferation and programmed cell death in this tumour type. This implies that integrins are involved in signal transduction events within colon carcinoma cells consequent upon their adhesive interaction with matrix molecules. A better understanding of the mechanisms involved in these events may lead to useful therapeutic strategies in the management of this disease.

Surgical history of Crohn's disease in a well-defined population.

Since 1935, 42 of 103 residents of Olmsted County, Minnesota, identified as having Crohn's disease underwent one or more surgical procedures. The incidence and type of surgical intervention, distribution of disease, and risk of recurrent disease after surgery were analyzed. Follow-up extended to 32 years, with a median of 8.5 years. Thirty-six patients underwent at least one definitive resection; eight of these patients subsequently underwent a second definitive operation. The likelihood that any patient would undergo operation was greatest within the first year of diagnosis. The proportion of patients in whom recurrent disease developed after the first and second definitive resections was 50% and 37%, respectively. Approximately half the patients who experienced a recurrence underwent further surgery. The site of diseased bowel and sex were not factors specifically influencing recurrence rates. Patients 40 years of age and older who underwent surgery appeared to fare better with respect to recurrent disease than younger patients. Although our findings suggest cumulative risks of recurrent disease after definitive resection similar to those reported from larger population groups or major referral centers, the proportion of patients who underwent surgery during the natural history of their Crohn's disease was much less than generally reported. We conclude that operation for Crohn's disease is not inevitable and that evidence to the contrary may imply both a treatment and a referral bias.

Colonic transit time assessed by 67 Ga-citrate in a case-series of patients with recto-sigmoid adenomas.

One plausible mechanism by which dietary factors may influence colorectal carcinogenesis is through their effect on intestinal transit time. This study examined colonic transit by means of oral 67Ga-citrate in a case-series of patients who had developed recto-sigmoid adenoma. Adenoma patients had a significantly shorter transit time than constipated patients (P = 0.01) and our results also suggest (but do not show conclusively) that colonic transit in adenoma patients is similar to that of normal controls. Although these findings require confirmation from a larger study, they raise the hypothesis that colonic transit times are not delayed in patients who harbour recto-sigmoid adenomas.

Cell adhesion molecules and colon cancer.

There is a general consensus that cell-cell and cell-matrix interactions determine, at least in part, the behaviour of colon cancer. The biological mediators responsible for these interactions are cell adhesion molecules belonging to several major receptor families called integrins, cadherins, the immunoglobulin superfamily, hyaluronate receptors and mucins. Emerging data indicate that certain patterns of adhesion receptor expression are associated with more aggressive disease. The present review examines the role of each of the receptor families in the development and progression of large bowel cancer.

Human colon cancer and fibroblast cell lines cultured in and on collagen gels.

Collagen is the major constituent of the in vivo extracellular matrix environment and the ability of collagen substrates to support growth of cultured cells in vitro is well recognized. The aim of the present study was to examine in vitro proliferation and matrix-binding of cells obtained from a human colon fibroblast and four colon cancer cell lines cultured in a collagen matrix environment. In contrast to colon fibroblasts, colon cancer cell lines proliferated in this culture system and their proliferative capacities were dependent upon the collagen concentration and whether tumour cells were seeded on or in the collagen. Both laminin and fibronectin stimulated growth of one of the four colon cancer cell lines without an apparent increase in cell-matrix binding. The use of collagen matrices to culture tumour cells in vitro might facilitate identification of factors which regulate growth of an individual's colorectal cancer.

A collagen matrix microassay for use in tumour-stromal cell co-cultures.

A collagen matrix microassay technique is described in which separation of collagen layers permits independent assessment of the proliferative capacity of each of two discrete cell populations. The two cell types used in this study were an established colon cancer cell line and a normal colon fibroblast cell line cultured under serum-free conditions. The implications of this in vitro technique for studies of tumour-host cell interactions are discussed.

Radiology of sigmoid volvulus.

Sigmoid volvulus is often a surgical emergency. The diagnosis is usually reached on the basis of both clinical and radiologic assessment. A review of the radiologic findings in 20 patients diagnosed as having sigmoid volvulus is presented. Typical radiographic features are illustrated. Reasons for their absence in certain cases are discussed. [Key words: Volvulus, sigmoid; Diagnostic problems; Sigmoid, volvulus].

Natural resolution of splenic injury as assessed by scintiscanning.

Two patients with suspected splenic injury following blunt external trauma are presented. Neither underwent splenectomy, and the changes on serial radioisotope scans provided a useful guide by which to assess the resolution of the splenic injury.

The role of colon fibroblasts in malignant large bowel obstruction--an experimental in vitro model.

The mechanism of bowel obstruction in colorectal cancer is likely to involve interactions between tumour cells, host fibroblasts and the extracellular matrix. The role of fibroblast-mediated matrix reorganisation in malignant structures of the large bowel was examined in an in vitro collagen matrix model in which tumour cells and fibroblasts were cultured under serum-free conditions. Colon cancer cells secreted a factor(s) which enhanced the ability of colon fibroblasts to contrast a collagen matrix without an associated mitogenic response by the fibroblasts. Within uncontracted collagen gels marked elongation of fibroblast cell processes was observed in the presence of the tumour-derived factor(s). We propose that matrix reorganisation by host fibroblasts in the wall of the human colon is responsible, at least in part, for malignant large bowel obstruction.

Portal vein thrombosis associated with rectal bleeding: non-invasive diagnosis and follow up by ultrasound.

A 65 year old woman presented with a sudden onset of abdominal pain and bright rectal bleeding. Shortly after admission a diagnosis of partial portal vein thrombosis was made by ultrasonography. This thrombosis appeared to be the first sign of a previously unsuspected hypercoagulable state due to polycythaemia rubra vera. Ultrasound proved useful in confirming complete resolution of the portal system thrombosis in response to conservative therapy.

Matrix metalloproteinase 9 activity in urine of patients at risk for premature delivery.

A simple, noninvasive assay was used to quantitate urinary matrix metalloproteinase 9 activity among 15 patients with threatened premature labor. Both positive and negative predictive values for risk for premature delivery were 80%.

Benign symmetric lipomatosis.

We report a case of massive, benign symmetric lipomatosis associated with laryngeal compression. Current views on the aetiology and management of this rare disease are discussed.

In vitro chemotherapy sensitivity testing of primary human colorectal carcinomas.

The cells obtained from 138 tumor samples taken from 135 patients with colorectal malignancies were cultured in vitro in a soft agar colony formation assay similar to that of Salmon and colleagues [1]. Significant colony formation occurred for 63 (51%) of evaluable tumor cultures, 53 of which were also tested against chemotherapeutic agents in vitro. The median number of drugs tested per tumor was 15. Using 70% inhibition of colony formation as the criterion for significant drug-induced cytotoxicity, only 3/53 (6%) of the tumors were noted to be sensitive to any drug. When colony counts generated by initially plated small tumor cell aggregates were taken into account, 8/40 (20%) of the tumors were noted to be sensitive in vitro to one or more agents. Because of the low rate of drug sensitivity found with this in vitro assay, its current role in the prospective assignment of chemotherapeutic treatment for patients with colorectal carcinoma is somewhat limited.

Nodal involvement in poorly differentiated breast cancer.

The degree of nodal involvement in a consecutive series of 400 patients with invasive ductal breast cancer is presented. A positive correlation was observed between the number of metastatic nodes identified and the number of axillary nodes examined for poorly but not moderately differentiated tumours. The relevance of this observation to breast cancer trials is discussed.

Growth factor-induced proliferative responses of human and DMH-induced rat colorectal tumour cell lines.

The proliferative response of three human and three 1,2-dimethylhydrazine (DMH)-induced rat colon tumour cell lines to mouse epidermal growth factor (mEGF) and human urogastrone was examined in two in vitro assays (3H-thymidine uptake in liquid microculture and clonogenicity in semi-solid medium). In the 3H-thymidine uptake assay, one of three rat and one of three human colon tumour cell lines, i.e., r237 and LIM1215, were stimulated by mEGF. In the clonogenic assay, mEGF stimulated the one clonogenic rat tumour cell line r113, which did not respond in the 3H-thymidine uptake assay, and the human tumour cell line LIM1215. Urogastrone stimulated LIM1215 in both assay systems. The significance of these in vitro patterns of response to growth stimulatory factors is discussed.

In vitro chemotherapy sensitivity patterns of human colon carcinoma continuous cell lines.

The cells obtained from four established human colon carcinoma cell lines were cultured in soft agar in the presence of eight chemotherapeutic agents that have been used clinically in patients with metastatic colorectal carcinoma. 5-Fluorouracil, mitomycin C, adriamycin, and cis-platinum were highly cytotoxic in vitro at the concentrations employed. These findings were in marked contrast to in vitro chemotherapy sensitivity patterns previously reported by us for a large series of primary human colorectal carcinomas cultured in soft agar. The possible reasons behind these differences are discussed.