Sulindac enhances cell proliferation in DMH-treated mouse colonic mucosa

In a previous study we reported that the NSAID sulindac had a marked inhibitory effect on the development of colonic tumours in mice treated with the carcinogen 1,2-dimethylhydrazine (DMH). In this study we examined the effects of sulindac in respect of cell-kinetic changes in mouse colonic mucosa as determined by flash labelling with the thymidine analogue bromodeoxyuridine (BrdUrd) at varying intervals during the process of colonic carcinogenesis. We also investigated the possibility that these changes may be modulated by misoprostol a prostaglandin E1 analogue. Four groups of 36 mice each were treated for 18 weeks with the following drug/s respectively: (1) DMH; (2) DMH and sulindac; (3) DMH, sulindac and misoprostol; and (4) DMH and misoprostol. Three animals from each group were killed each week between the sixth week and the eighteenth week after the start of the experiment. A 1-h flash label technique was employed and paraffin sections of colonic mucosa were examined. For each animal a total of 50 perfect axially cut crypts were chosen and the following parameters determined: crypt length, labelling index and labelling index distribution: the data were analysed using the computer program GLIM. For each of the four groups, crypt lengths increased significantly with the duration of treatment with no significant difference between the groups. In sulindac-treated animals the labelling index for all positions increased with duration of treatment whereas for animals not treated with sulindac there was no significant difference in labelling index with respect to duration of treatment. The administration of misoprostol did not appear to significantly alter the effects of sulindac. It is postulated that the observed increase in cell proliferation could be a compensatory phenomenon occurring secondary to loss of crypt epithelial cells by apoptosis induced by sulindac. Also the finding of an increase in labelling index mediated by a chemopreventive agent indirectly questions the rationale behind the therapeutic manipulation of crypt cell proliferation in order to reduce the risk of colon cancer.

Organ culture as a model for investigating the effects of antimetabolites and nucleoside transport inhibitors on rodent colonic mucosa

The in-vitro effects of hydroxyurea 5-FU and 5-FUdR have been extensively studied in experimental systems employing cell-line techniques. In this study we investigated the effects of these drugs on the levels of incorporation of labeled nucleosides into DNA in explants of intact rat colonic mucosa maintained in organ culture. The effects of the nucleoside transport inhibitors nitrobenzylthioinosine (NBMPR) and dipyridamole--which are modulators of antimetabolite cytotoxicity--on the incorporation of tritiated thymidine ([3H]TdR) into DNA were also studied. The incorporation of tritiated TdR into DNA was reduced by hydroxyurea but was not altered by either 5-FU or 5-FUdR. The levels of tritiated deoxyuridine were reduced by 5-FU and 5-FUdR in separate experiments; this is in keeping with thymidylate synthase inhibition. NBMPR and dipyridamole also reduced 3H-TdR incorporation into DNA. These results can be explained in terms of the known mechanisms of action of these drugs. This experimental model is therefore useful in assessing the effects of antimetabolites and nucleoside transport inhibitors in intact colonic mucosa.

Effects of gastrointestinal peptides on azoxymethane-treated colonic mucosa in vitro

An organ-culture system has been used to investigate the effect of certain gastrointestinal peptides on the morphology and cell proliferation of explants of azoxymethane (AOM)-treated colonic mucosa. Our aim was to ascertain whether such factors play a direct part in the maintenance of hyperplastic changes in the large intestine. Explants of AOM-treated colonic mucosa from 15 animals were maintained in a serum-free medium in the presence of either gastrin-17 (250 pg/ml and 250 ng/ml), peptide YY (80 pmol/l and 160 pmol/l) epidermal growth factor (EGF) (10 ng/ml and 100 ng/ml) or the C-terminal fragment of glucagon-37 (30 pmol/l) for a period of up to 7 days. Other explants (controls) received fresh medium only each day. After 1, 2, 3, 5 and 7 days of culture both experimental and control explants received vincristine (4 micrograms/ml) for 3 h prior to fixation. The proportion of vincristine-arrested metaphases within the explants was determined together with crypt length. Neither gastrin nor peptide YY was found to influence cell division at either concentration. Despite an initial inhibitory effect, both concentrations of EGF exerted a trophic effect which increased with time. The glucagon-37 fragment caused an immediate increase in proliferation which then declined as time progressed. None of these factors, however, were able to maintain the hyperplastic changes seen in the pre-culture samples of AOM-treated mucosae.

Modulation by verapamil of vincristine pharmacokinetics and sensitivity to metaphase arrest of the normal rat colon in organ culture

The in-vitro pharmacokinetics of vincristine (VCR) in normal rat colonic mucosa were studied. Two complementary approaches were adopted using an explant organ-culture system. Firstly [G-3H]vincristine (3HVCR) accumulation, retention and efflux were characterized under basal conditions and compared with measurements made either under energy-depleted conditions, or in the presence of VRP. Secondly, a histological method--the postmetaphase index (PMI)--was used to compare the sensitivity of explants to VCR in the presence or absence of verapamil (VRP). This latter technique involves the measurement, by counting, of the proportion of mitotic figures escaping from metaphase arrest. The studies yielded the following results: 3HVCR accumulation in colonic mucosa showed no evidence of saturability up to the maximum dose studied (130 nM), at a dose of 52 nM accumulation was enhanced in energy-depleted conditions by a factor of 1.8, and in the presence of VRP (6.6 microM) by a factor of 1.4. In the presence of VRP (6.6 microM) retention of 3HVCR was increased by a factor of 1.3 and efflux was reduced by a factor of 0.8 after 2 hr. VRP (6.6 microM) reduced the PMI of colonic mucosal epithelial cells exposed to 11 nM VCR from 18.8% to 11.4% (i.e. 40% reduction) indicating sensitization of the cells to this property of VCR. These results provide evidence that the sensitivity of normal colonic mucosa to vincristine is, at least in part, regulated by drug transport. Qualitatively our observations resemble those described in multidrug resistance. Given that P-glycoprotein has been demonstrated by several groups in colonic mucosal cells, the results support a normal role for this membrane transport molecule in the protection of intestinal cells from plant alkaloids and other xenobiotic agents ingested in the diet.

Epidermal growth factor receptors in colorectal carcinoma

Nineteen samples of primary colorectal carcinoma and adjacent mucosa were examined for EGFr expression using radioligand binding assays and immunohistochemical staining with the monoclonal antibody EGFR1. Radioligand binding experiments showed expression of EGFr in both tumour and mucosa in all cases. In tumour samples EGFr levels ranged between 4 and 79 fmole per mg membrane protein (Kd = 0.1-0.4 X 10(-9) M). There was no significant difference in the level of EGFr expression between tumour and mucosa overall. Immunohistochemical staining with the EGFR1 antibody was useful in localising EGFr to epithelial elements although it was less sensitive than ligand binding assays.

A protective effect of sulindac against chemically-induced primary colonic tumours in mice

Sulindac, a non-steroidal anti-inflammatory drug, has been reported to lead to tumour regression in cases of human polyposis coli. We have investigated the effects of this drug on the growth of 1,2-dimethylhydrazine (DMH)-induced mouse colonic tumours. In one experiment, DMH and oral sulindac were administered concurrently to a group of mice for a period of up to 24 weeks, while a control group of animals received DMH only for the same period. Sulindac caused a significant reduction in both the number of mice with colonic tumours and the number of tumours per mouse. In a second experiment, two groups of mice which had already been treated with DMH for 17 weeks received either sulindac or not for 78 days. In this experiment sulindac had no effect. These results demonstrate that sulindac has a protective effect against the chemical induction of colonic tumours in mice, but does not cause the regression of established tumours.

Proliferative status of colonic mucosa in organ culture: 3H-thymidine-labelling studies and computer modelling

3H-thymidine labelling studies and a computer simulation have been employed to assess proliferative status and cellular organisation in colonic explants maintained in culture for 5 to 7 days. The one-hour flash labelling index (Is) for crypts within the middle region of explants (5.2%) was considerably lower than that observed in vivo (8.8%). Crypt length and the distribution of labelled cells appeared similar for both situations. A computer simulation program for crypt-cell proliferation was devised, facilitating the modulation of a number of parameters including the cell-cycle time (Tc) and its component phases, the cut-off position, and cell loss at mitosis. This simulation was employed to model continuous labelling (72 h) data obtained in vitro and provided an estimate of various kinetic parameters. Data for the middle region of explants was fitted with a Tc of 62 h, an S phase of 8 h and a cell loss factor (20%) which was consistent with histological findings. A fit to the experimental data obtained in vitro could be achieved by a model based upon a mode of cellular organisation known to occur within crypts in vivo. Therefore in vitro, the dynamic processes of crypt-cell proliferation and migration appear to be organised in the same manner as seen in vivo.

Verapamil sensitizes normal and neoplastic rodent intestinal tissues to the stathmokinetic effect of vincristine in vivo

A morphological method has been developed allowing measurement of the effect on intestinal epithelia of vincristine. In routinely prepared tissue sections the proportion of mitotic events progressing beyond metaphase is counted by microscopy. When estimated over a range of doses of vincristine this post-metaphase index (PMI) can be used to compare the sensitivity of differing intact tissues. Intestinal tumours were induced in rats by chemical carcinogenesis. Administration of vincristine in the presence or absence of verapamil was performed in these tumour-bearing animals. Sections were prepared from colonic and small-bowel tumours and from normal mucosa. The results show that verapamil increases the sensitivity of the tissues studied to vincristine. A dose dependent effect of verapamil on vincristine sensitisation was demonstrated in colonic tissues. These findings indicate a shared pharmacological property between the resistance of primary tumour tissue and the multidrug-resistance phenotype.

A trophic effect of epidermal growth factor (EGF) on rat colonic mucosa in organ culture

The development of an organ-culture system for rat colonic mucosa has enabled a direct assessment of the effect of epidermal growth factor (EGF) on cell division. An augmented mitotic index (AIm) has been employed to identify changes in cell proliferation. Explants of colonic mucosa from four animals were maintained in a medium containing serum for five days. On the fifth day of culture, half of the explants received fresh medium containing EGF (40 ng/ml) and the remainder (controls) fresh medium only. At 6, 12, 24 and 48 hr thereafter groups of both experimental and control explants received the metaphase-arresting drug vincristine (4 micrograms/ml) for 3 hr prior to fixation. The proportions of vincristine-arrested metaphases within the explants were determined. Analysis of the data indicates that when serum is present exogenous EGF exerts a trophic effect which increases with time (P less than 0.001). In a second experiment colonic explants from four animals were maintained for five days in a serum-free medium and were then divided into groups, each of which received one of a range of concentrations of EGF. The AIm was determined for each group after 36 hr. It was found that increasing concentrations of EGF produce a small but significant increase in cell proliferation (P less than 0.01). This effect, however, was less pronounced than that seen when serum was present. These results suggest that EGF has a trophic action on the colon and interacts with additional factors found in serum.

A comparison of crypt-cell proliferation in rat colonic mucosa in vivo and in vitro

The successful development of a long-term organ culture system has made it possible to perform experiments on rat colonic mucosa in vitro. However, the effect of trauma or the withdrawal of trophic influences in culture may result in the disturbance of proliferation within the tissue. In this paper we describe an investigation designed to characterise the culture system by a comparison of the proliferative parameters in vitro with those in vivo. Stathmokinetic experiments were performed in vivo and in vitro to estimate cell birth rate. Mitotic indices were also calculated. The in vivo birth rate (7.8 +/- 0.8 cells/1000 cells/hour) and the in vitro birth rate for the whole explant (7.7 +/- 0.5 cells/1000 cells/hour) were found to be similar. A study of crypts in the centre and at the edge of the cultured explants, however, indicated that proliferation at the two sites differed markedly, the birth rate at the edge (7.5 +/- 0.9 cells/1000 cells/hour) being approximately twice that at the centre (3.2 +/- 0.9 cells/1000 cells/hour). Provided that this topological difference in proliferation within the explant is recognised the in vitro model, in particular the basal level found at the centre, may still be regarded as a valid system for studying tissue responses to carcinogens and trophic factors.

Demonstration of vincristine resistance in primary intestinal neoplasms in the rat by the 'post-metaphase index'

A method is described enabling the direct measurement of vincristine resistance in intact tissues in vivo by morphological study. Using the metaphase arresting properties of the drug, counts were made of escaping anaphase and telophase mitotic figures at a range of doses. The proportion of post-metaphase mitotic figures is called the post-metaphase index (PMI). In 95 primary intestinal tumours induced by dimethylhydrazine (DMH) in rats, an increase in resistance to vincristine was shown over normal mucosa (P less than 0.001). The data were analysed by computer modelling and a linear relationship is demonstrated between the logit of the post-metaphase index, and log dose of vincristine. To achieve a PMI of 1% the fitted lines show an enhanced vincristine dose requirement over normal mucosa of 6 times in colonic tumours, and 8 times in small intestinal tumours. Non-neoplastic mucosa from the DMH-treated animals requires an enhanced dose of vincristine of 1.5 times, compared with normal mucosa, to achieve a PMI of 1%. Given current interest in the mechanism of vincristine resistance in cell lines this new approach provides a technique for assessing the resistance of solid tumours, both in vivo and in vitro, and for subsequent experimental manipulation.