IS THERE CIRCADIAN VARIATION IN VILLUS HEIGHT IN RAT SMALL INTESTINE?

Transcription

1 Quarterly Journal of Experimental Physiology (1989) 74, Printed in Great Britain IS THERE CIRCADIAN VARIATION IN VILLUS HEIGHT IN RAT SMALL INTESTINE? MICHAEL L. G. GARDNER AND DEBORAH L. STEELE School of Biomedical Sciences, University of Bradford, Bradford, West Yorkshire BD7 IDP (MANUSCRIPT RECEIVED 27 SEPTEMBER 1988, ACCEPTED 22 NOVEMBER 1988) SUMMARY Measurements of villus height, crypt depth and mucosal thickness were made for each of duodenum, jejunum and ileum of rat small intestine from animals on both unrestricted and restricted feeding regimes. Contrary to a previous report, there was no evidence for consistent or synchronized diurnal variation in villus height at any region of small intestine on any of the feeding regimes. Likewise, no diurnal variation was seen in crypt depth or mucosal thickness. Variation in villus cell numbers, therefore, cannot account for previously reported diurnal variation in absorptive and digestive activities in rat small intestine. The present results support the hypothesis that the stimulus for exfoliation is probably cell recruitment and migration up the villi. INTRODUCTION Various functional activities of the small intestine are known, in rats and mice at least, to exhibit diurnal variation, e.g. absorption of amino acids, sugars and water, and hydrolase enzyme activities (Furuya & Yugari, 1971; Stevenson, Ferrigni, Parnicky, Day & Fierstein, 1975; Fisher & Gardner, 1976). The time of food intake is believed to be a major factor influencing the timing of maximum activity (Saito, Murakami & Suda, 1976; Stevenson & Fierstein, 1976; Kaufman, Korsmo & Olsen, 198). Diurnal variation in the susceptibility of the rat intestinal epithelium to damage by cytotoxic drugs has also been reported (Gardner & Plumb, 1981). However, in spite of literature on diurnal variation in mitotic indices and DNA synthetic rate (Sigdestad & Lesher, 197, 1971; Al-Dewachi, Wright, Appleton & Watson, 1976), the cellular mechanisms underlying these variations in absorptive activity and in susceptibility to toxicity cannot yet be explained. We need clearer understanding of the complex dynamics of cell proliferation, maturation and migration. Furthermore, this information may make it possible to assess whether, and how, observations on diurnal variation in susceptibility of rat intestine to toxicity can be relevant to human intestine during chemotherapy. Stevenson, Day & Sitren (1979) provided striking and potentially important observations suggesting that villi in the rat ileum vary markedly in length and cell row count throughout the day: villus height was claimed to be 5-6% greater close to the feeding time than 12 h later. This surprising finding might account for diurnal variation in absorptive and enzymic activities but, to our knowledge, it has never been independently confirmed. Another important implication that would arise from their results would be that the driving force for cell extrusion at the tips of villi cannot be immediately related to pressure from upwardly migrating cells, and hence, to cell recruitment. Another stimulus for exfoliation would need to be involved. Therefore we have set out to verify and extend the findings of Stevenson et al. (1979), and in particular to assess whether villus height measured at three- or four-hourly intervals in

2 258 M. L. G. GARDNER AND D. L. STEELE three regions of rat small intestine shows substantial diurnal variation, or whether villus height remains in a relatively steady state. We have also measured crypt depth. Initially we used animals with free access to food but, since Stevenson et al. (1979) used rats on two different regimes of restricted feeding, we have extended our study to include further groups of animals subjected to conditions as similar as possible to those used by these previous authors. An abstract has already reported the preliminary findings of this work (Gardner & Steele, 1987). METHODS Animals. The animals were matched as closely as possible with those used by Stevenson et al. (1979), which were male Sprague-Dawley rats 48-5 days old and weight range g at the start of the experiment, with a diet of 'Purina Rat Chow'. We used male Sprague-Dawley rats weighing g, which were transferred to our animal house on their 47th day of age. For the first 7 days thereafter, to allow initial acclimatization, all our animals were allowed free access to food (CRM rat diet; Labsure, Manea, Cambridgeshire) and water; illumination was on a 14-1 h light-dark cycle with darkness from 21. to 7. h GMT. During the second week, a total of seventy-two animals (Group A) was continued on this ad libitum regime, these being housed in groups of six. Two other groups (B and C), each of twenty-one rats, were transferred to restricted feeding regimes and to illumination and housing arrangements which were as similar as possible to those used by Stevenson et al. (1979). Group B ('morning fed') had food available between 7. and 11. h, while group C ('afternoon fed') had food available between 14. and 18. h. Water was available-ad libitum for both groups. Rats in groups B and C were housed individually and, in their second week, their illumination was on a h light-dark cycle with darkness from 18.3 to 6.3 h GMT. The food intakes and body weights of animals in groups B and C were monitored daily throughout the 2 week period. Tissue sampling and processing. At the end of the second week rats from group A (fed ad libitum) were taken at three-hourly intervals starting at 1. h, eight rats being sampled at each time point. Rats from groups B ('morning fed') and C ('afternoon fed') were taken at four-hourly intervals starting 1 h before the feeding time, three rats being sampled at each time point. Animals were lightly anaesthetized with ether and 1 cm segments of duodenum (midway between pylorus and the ligament of Treitz), jejunum (approximately 15 cm distal to the ligament of Treitz) and ileum (approximately 2 cm proximal to the caecum) were excised. These segments were placed on card, opened along their length at the mesenteric border, and placed immediately into neutral buffer formol saline (Drury & Wallington, 1967, p. 41), and the rats were then killed. After fixation for at least 48 h at room temperature, the tissue was processed for routine histology and embedded in paraffin wax. Longitudinal sections (5 /,m) were cut and stained with Harris' Haematoxylin and Eosin (Drury & Wallington, 1967, pp ). Histological morphometry. Villus height and crypt depth were measured on ten villi and crypts taken from at least two sections for each intestinal segment. Villi were selected for correct longitudinal orientation according to the criteria of Altmann & Leblond (197). Slides were viewed at x 1 magnification and the image projected by a camera lucida on to a digitizer tablet equipped with a mouse-cursor (Summagraphics Bit Pad 2; Advanced Microcomputer Applications, Beeston, Nottingham). The output from the tablet was processed with a BBC B+ microcomputer using DIGIT software (Institute of Ophthalmology, London). This provides a low-cost image analysis system that is ideal for, among other functions, linear measurements. Thus, the distances between villus tip and crypt-villus junction and between crypt-villus junction and crypt base were measured, the crypt-villus junction being identified according to the criteria of Cheng & Leblond (1974). These data were stored on floppy disc for subsequent statistical analysis. Statistical analysis. For each intestinal segment from each animal a single mean value of villus height, one of crypt depth, and one of total mucosal thickness were derived, each being the mean of measurements of ten villi or crypts observed in at least two sections. The significance of variation with

3 LACK OF CIRCADIAN VARIATION IN VILLUS HEIGHT F A 1-1 bo I.- >b 4) 33 p 31F 29 F Efed Morning fed 27 F Day number r B ba 4.S 3 p 2-1p Aftemoon fed Morning fed o Day number Fig. 1. A, the mean body weight for animals in groups B and C (restricted feeding). Days 1-7 with ad libitum feeding, followed by days 8-15 with restricted feeding. During days 8-15 group B ('morning fed', -*-) had food available from 7. to 11. h, and group C ('afternoon fed', -- --) from 14. to 18. h. B, the food intake of both groups over days Each point is the mean for n = 21 rats. time of day was assessed by a one-way analysis of variance. Data relating to each region of intestine were considered separately. RESULTS Animals' growth rates and food intakes Figure 1 A and B shows the mean body weights and food intakes respectively each day for the animals in groups B and?, i.e. those transferred to the restricted feeding regimes (B, 'morning fed'; C, 'afternoon fed'), as used by Stevenson et al. (1979), after 1 week of unrestricted feeding. During the first week (food available ad libitum) the rats grew at a mean rate of 9 82 g/ day. However, on transfer to the restricted feeding regimes (feeding restricted to a 4 h

4 26 M. L. G. GARDNER AND D. L. STEELE 8 - A 6 Fg,63 = 1 12; n. s. Jejunum xf863 = 1-6; n.s.. 4 Ileum 2 :> F8,63= 1.239; n.s. 8 B 6 Jejunum F6,14 = -245; n.s. F6,14 =-997; n.s. bo z 4 - Ileum 2 * F6,14 =-973; n.s. light period, '- 8 - C F6 14 = 2-6; n.s. 6 - Jejunum F6,4 = 2-9; P<. u4 Ileum 4em 4,6 = 843; n.s. 2 Time (h GM'I) Fig. 2. For legend see facing page.

5 LACK OF CIRCADIAN VARIATION IN VILLUS HEIGHT period) the food intake fell markedly and, while some recovery was apparent between days 9 and 15, it did not return to normal (Fig. 1 B), although the animals did appear to be healthy. Hence, even by day 15, the growth rate was much less than normal; on the last day the mean growth rate was 2 g/day, i.e. 237 % of that achieved with unrestricted feeding. The animals fed in the afternoon (group C) ate significantly more food than those fed in the morning (group B, P < -1; analysis of variance), and this was reflected in higher mean body weights in the former group (P < O1). Histological morphometry of small intestine Figure 2A, B and C respectively shows the mean villus heights for each of the three regions of small intestine from the animals on each feeding regime (groups A, B and C respectively), sampled at three- or four-hourly intervals over a 24 h period. It is clear that any variation over this period is small; analysis of variance confirmed that it was insignificant except in the case of data for jejunum of the 'afternoon fed' rats (group C, P < 5). Furthermore, inspection of Fig. 2A, B and C shows no synchrony among the measurements for duodenum, jejunum and ileum. Hence, there is no evidence suggesting diurnal variation in villus height of the nature proposed by Stevenson et al. (1979). The villus height did not depend on the feeding regime. Figure 3 A, B and C shows the corresponding data for crypt depth. Analysis of variance showed significant variation with time for the data for ileum (group A, P < 1; and group B, P < O1), jejunum (group B only, P < 5), and duodenum (group C only, P < O1). None of the remaining data showed significant variation with time. To eliminate the possibility that mis-identification of the crypt-villus junction might have concealed a diurnal variation in villus height, we have also analysed the data in terms of total mucosal thickness, i.e. villus tip to crypt base (see Fig. 4A, B and C). No significant diurnal variation was revealed by the analyses of variance except for the data for the duodenum of the 'afternoon fed' rats (group C, P < 5). 261 DISCUSSION Our findings (Fig. 2) are markedly at variance with those reported by Stevenson et al. (1979): i.e. there is no evidence for diurnal variation in the villus height (and hence cell number, but see below) in any of the three groups of animals, including those subjected to feeding regimes closely similar to those adopted by Stevenson et al. (1979). A lack of diurnal variation in total protein content of mucosal homogenates was also noted by Stevenson et al. (1975) and by Stevenson & Fierstein (1976): this too is consistent with lack of diurnal variation in villus cell number. The dashed lines on Fig. 2 B and C show the data obtained by Stevenson et al. (1979, their fig. II) for ileal (apparently mid-ileal) villus height in their animals. While there is remarkable agreement between their measurements and ours at the times of their stated maxima, there are significant differences at other times. We cannot explain these differences. The two restricted feeding regimes used by Stevenson et al. (1979) were inadequate to Fig. 2. Villus height at various times of day. A, Group A, ad libitum feeding; B, group B, 'morning fed'; C, group C, 'afternoon fed'. U duodenum; -, jejunum; --, ileum. The dashed lines (--*--) show the data reported by Stevenson et al. (1979) for a mid-ileal region of rat intestine. Light and dark periods are indicated by open and filled horizontal bars respectively, and the times of food availability by hatched bars. The variance ratio (F), together with the number of degrees of freedom, for variation due to time of day is shown. n.s., not significant. Each point is the mean + S.E.M. for n = 8 rats in group A and n = 3 rats in groups B and C (ten villi per region for each rat). Where not shown, S.E.M. falls within the symbol.

6 262 M. L. G. GARDNER AND D. L. STEELE 3 A 25 Fi,63 = 1*59; n.s. u Jejunum g ; n. s. 15 -fleum R1,63 =4-3; P < -1 3 B 25 F'6,14 = 2-348; n.s. ~2 Jejunum F6,14 = 3 997; P<O5 n 15 Ileum F6,14 =4974;P<1 Q1-5 3 C F6.l4 = 6 392; P < g22 H < Fl46 =l-ooo; n. s. 15-eu 1 F6.14 = 1-185; n.s. 1-5 Light penrod Time (h GMI) Fig. 3. Crypt depth at various times of day. For details see Fig. 2.

7 LACK OF CIRCADIAN VARIATION IN VILLUS HEIGHT A 8 F63.8 = 888; n. s. Jejunum )6 F63,8 = - 798; n. s. mleum ' 4 - l,63 =-193; n.s. 2 Time (h GM) 1 - B 8 i4,6=449;n.s... Jejunum s6 6 - F6,14 Ileum 4 F6.14 =1 =1 96; n.s. - 77; n.s c 1 ~8- k1 8=3-2;P<-5 ra x Jejunum. (6( F6, 4 = 1.*39; n. s. - Seum 4 F14,6 = -78; n.s. 2 /m/5 Fig. 4. Mucosal thickness at various times of day. For details see Fig. 2.

8 264 M. L. G. GARDNER AND D. L. STEELE sustain normal growth in our rats (see Fig. 1A). Stevenson et al. (1979) did not show corresponding data for their animals, but earlier data from their laboratory were similar to ours (see Fig. 1 of Stevenson & Fierstein, 1976). This is, however, unlikely to have any bearing on the existence or not of diurnal variation in villus height, since none of our three groups exhibited synchronized or consistent variation of the nature claimed by Stevenson et al. (1979). Despite the fact that our animals' growth rates were deficient when on the restricted feeding regimes (Fig. 1 A), villus height was not significantly different from that in the animals on ad libitum feeding. Changes in villus cell number and surface area of the villus are, therefore, unlikely to be responsible for the diurnal changes in absorptive and enzymic activities that we and other investigators have already reported. The fact that villus height is in a relatively steady state throughout the day and night implies that the rate of cell exfoliation from the villi is approximately equal to the rate of recruitment of new cells from the crypts. This is consistent with the mechanisms suggested by Clarke (1973) and Cheng & Leblond (1974), and is further supported by the observation by Potten & Allen (1977) that cell loss is maximal in the mouse during the early hours of the morning, this coinciding with the time of maximal cell production reported by Sigdestad & Lesher (1971). Hence, the driving force for cell extrusion is likely to be directly associated with upward migration although it is not necessarily a passive process (Potten & Allen, 1977). In turn, the stimulus for migration is almost certainly multifactorial and related to functional demands. Although villus height and villus cell row counts are probably imperfect indices of total villus cell population, Wright & Alison (1984, p. 63) have noted that there is a good correlation between villus height and epithelial cell population in experimental animals with normally shaped villi. The alternative method for estimation of villus cell population by microdissection of villi is not applicable in the rat, since rat villi are considerably larger than mouse villi, and their convoluted shape makes their accurate microdissection virtually impossible (Goodlad, Plumb & Wright, 1987). Further, it was necessary to use villus height as the index in our present investigation since it was variation in this index that was the basis of the striking findings by Stevenson et al. (1979) and their proposed explanation of diurnal variation in absorptive and enzymic activities. It would therefore appear likely that subcellular, rather than structural, changes are involved in diurnal variations in intestinal absorptive and enzymic activities. It is noteworthy that Kaufman et al. (198) produced evidence that circadian variation of intestinal sucrase activity in the rat was accompanied by variation in protein degradation while protein synthesis remained fairly constant. Hence, it is plausible that similar changes occur diurnally in membrane transport systems which would account for diurnal variation in absorptive activities. We are very grateful to Professor T. G. Baker and Dr Diana Wood for their valued discussions about this work. We also thank the animal house staff for their extra efforts and for facilitating the special housing arrangements. REFERENCES AL-DEWACHI, H. S., WRIGHT, N. S., APPLETON, D. R. & WATSON, A. J. (1976). Studies on the mechanism of diurnal variation of proliferative indices in the small bowel mucosa of the rat. Cell and Tissue Kinetics 9, ALTMANN, G. G. & LEBLOND, C. P. (197). Factors influencing villus size in the small intestine of adult rats as revealed by transposition of intestinal segments. American Journal of Anatomy 127,

9 LACK OF CIRCADIAN VARIATION IN VILLUS HEIGHT 265 CHENG, H. & LEBLOND, C. P. (1974). Origin, differentiation and renewal of the four main epithelial cell types in the mouse small intestine. I. Columnar cell. American Journal of Anatomy 141, CLARKE, R. M. (1973). Progress in measuring epithelial turnover in the villus of the small intestine. Digestion 8, DRURY, R. A. B. & WALLINGTON, E. A. (1967). Carleton's Histological Technique, 4th edn. London: Oxford University Press. FISHER, R. B. & GARDNER, M. L. G. (1976). A diurnal rhythm in the absorption of glucose and water by isolated rat small intestine. Journal of Physiology 254, FURUYA, S. & YUGARI, Y. (1971). Daily rhythmic change in the transport of histidine by everted sacs of rat small intestine. Biochimica et biophysica acta 241, GARDNER, M. L. G. & PLUMB, J. A. (1981). Diurnal variation in the intestinal toxicity of 5- fluorouracil in the rat. Clinical Science 61, GARDNER, M. L. G. & STEELE, D. L. (1987). Lack of circadian variation in villus height in rat small intestine. Zeitschrift far Gastroenterologie 25, 627. GOODLAD, R. A., PLUMB, J. A. & WRIGHT, N. A. (1987). The relationship between intestinal crypt cell production and intestinal water absorption measured in vitro in the rat. Clinical Science 72, KAUFMAN, M. A., KORSMO, H. A. & OLSEN, W. A. (198). Circadian rhythm of intestinal sucrase activity in rats. Journal of Clinical Investigation 65, POTTEN, C. S. & ALLEN, T. D. (1977). Ultrastructure of cell loss in intestinal mucosa. Journal of Ultrastructure Research 6, SAITO, M., MURAKAMI, E. & SUDA, M. (1976). Circadian rhythms in disaccharidases of rat small intestine and its relation to food intake. Biochimica et biophysica acta 421, SIGDESTAD, C. P. & LESHER, S. (197). Further studies on the circadian rhythm in the proliferative activity of mouse intestinal epithelium. Experientia 26, SIGDESTAD, C. P. & LESHER, S. (1971). Photo-reversal of the circadian rhythm in the proliferative activity of the mouse small intestine. Journal of Cell Physiology 78, STEVENSON, N. R., DAY, S. E. & SITREN, H. (1979). Circadian rhythmicity in rat intestinal villus length and cell number. International Journal of Chronobiology 6, STEVENSON, N. R., FERRIGNI, F., PARNICKY, K., DAY, S. E. & FIERSTEIN, J. S. (1975). Effect of changes in feeding schedule on the diurnal rhythms and daily activity levels of intestinal brush border enzymes and transport systems. Biochimica et biophysica acta 46, STEVENSON, N. R. & FIERSTEIN, J. S. (1976). Circadian rhythms of intestinal sucrase and glucose transport: cued by time of feeding. American Journal of Physiology 23, WRIGHT, N. & ALISON, M. (1984). The Biology of Epithelial Cell Populations, vol. 2. Oxford: Clarendon Press. I I EPH 74