Differences in the effects of age on intestinal proliferation, crypt fission and apoptosis on the small intestine and the colon of the rat

Abstract

The increase in gastrointestinal epithelial tissue mass and the development of the gut can occur through three main mechanisms, namely elevated cell production from the intestinal crypts, by raised crypt number, which occurs through the process of crypt fission or by altered apoptosis. The small bowel and the colon have various rates of these, which were studied in rats of various ages. Wistar rats were fed ad libitum, and were killed at 3, 4, 6, 9, 12, 18, 26 and 48 weeks of age. Tissue was later stained and microdissected and the number of native mitoses and apoptotic figures per crypt and the percentage of crypts in fission were determined. There was an almost linear increase in body weight from 3 to 9 weeks, followed by a more gradual rise until 18 weeks. The weight of the stomach and the small intestine reached maximum values at 9 weeks, whereas the caecum and the colon approached this at 12 weeks. Mitotic activity per crypt in the small intestine increased from 3.8 ± 0.1 at 3 weeks to 7.8 ± 0.4 mitoses per crypt (P < 0.001) at 9 weeks and then decreased slightly; crypt fission increased from 4.6% ± 0.8 at 3 weeks to 8.4 ± 0.9% at 6 weeks and then decreased gradually reaching a value of 1.5 ± 0.4% at 48 weeks. Apoptosis also peaked at 6 weeks and was then very low. In the colon, the proliferation decreased from 4.2 ± 0.2 mitoses per crypt in the young (3 weeks) rat and reached a plateau by 9 weeks (2.5 ± 0.1 mitoses per crypt, P < 0.001). Crypt fission also declined rapidly in the first 9 weeks (from 67.6 ± 4.2 to 23.1 ± 4.6%, P < 0.01) and then continued to decline, although at a lower rate. The crypt fission index at 48 weeks was 9.8 ± 1.0. Apoptosis in the colon persisted throughout the duration of the study, 0.19 ± 0.06 apoptotic bodies per crypt were seen at week 48. The development of the small intestine is more dependent on cell proliferation, whereas in the colon crypt fission is far more predominant, with the colon having fission indices approximately six times greater than those of the small intestine. Proliferative activity in the colon was approximately half that of the small intestine.

While cell proliferation is essential to life, excess of cell proliferation can be regarded as a classical promoter of carcinogenesis and is generally considered as one of the early events in colon cancer (Fearon & Vogelstein 1990). Intestinal cell number is also altered by cell loss, including programmed cell death (apoptosis). Apoptosis is a form of programmed cell death or genetically controlled self-destruction that is central to tissue morphogenesis in the developing animal and tissue homeostasis in the adult (Potten et al. 1997) and can be regarded as a ‘quality assurance’ mechanism (Martin et al. 1998a). Martin et al. (1998a) found that apoptosis was increased in old mice, but only after irradiation, suggesting that some damage had accumulated in the stems cells so that the cells were near the threshold level required to trigger cell death.

Furthermore, the intestinal epithelium can be profoundly moderated by the division of intestinal crypts by crypt fission, in which bifurcations in the base of the crypts ‘unzip’ to create new crypts (Totafurno et al. 1987). Crypt fission is, thus, a mechanism for the clonal spread of mutated crypts in carcinogenesis (Park et al. 1997; Goodlad et al. 2001; Brittan & Wright 2002). Thus, while cell proliferation and crypt fission are essential components of gut renewal and defence, excess of proliferation may be associated with cancer risk (Preston-Martin et al. 1990) and both proliferation and crypt fission are elevated in crypts isolated from human adenomas and hyperplastic polyps (Wong et al. 2002).

Most changes in proliferation and fission occur during the first few weeks of life. The width of villi (Martin et al. 1998a) and the number of cells per villus increased with age and this mirrors the number of crypts associated with a villus and the high rates of fission in the first weeks of life of the mouse (Cheng & Bjerknes 1985). A similar inverse relationship between crypts in fission and crypt number was found in rats with fission rates falling off more rapidly in the colon (St. Clair & Osborne 1985). There have been several studies, suggesting that proliferation is increased in old age (Holt et al. 1988; Xiao et al. 2001); however, our earlier investigations did not find any differences in the gut of old rats (Goodlad & Wright 1990; Goodlad et al. 1992).

If age does increase proliferation, it could help explain why cancer risk increases with the fourth power of age (Franks & Teich 1986). It has been suggested that there may be a link between tumour suppression and ageing as mammalian tumour-suppressor pathways may have evolved from ancient mechanisms that regulated embryogenesis or to protect the germline. The shift from developmental maintenance to somatic maintenance may, thus, have evolved at a cost (Campisi 2003). Mutation in one or more of the tumour-suppressor genes in older humans suggests an age-related increase in predisposition to neoplasia (Moragoda et al. 2002).

It is possible that the older gut is less able to adapt. Starvation in older rats resulted in a smaller decrease in DNA labelling of crypt cells, and refeeding produced an abrupt broadening of the proliferative zone in older rats and these responses were considered to resemble the precancerous changes seen in the colon after the administration of the chemical carcinogen dimethylhydrazine (Holt et al. 1988). However, other studies have found that the rates of enterocyte proliferation are independent of age so that intestinal adaptation in aged mice remains largely unchanged (Ferraris & Vinnakota 1995). While age did not change the size of individual villi or crypts or the cell production rate, the number of crypts and villi was reported to increase with age, indicating a continuation of mucosal growth up to 2 years in rats (Ecknauer et al. 1982).

Another report found no changes in the proliferation of the colonic epithelium with ageing in the face of a greatly increased Bax protein levels, suggesting that the colonic epithelial cells were more sensitive to apoptotic stimuli (Lee et al. 2000).

In order to try and resolve some of these questions and to provide baseline data for determining when the gut has stabilized, we decided to compare and contrast cell proliferation crypt fission and apoptosis in the small intestine and the colon of rats of various ages.

Rats and methods

Study protocol

Seven groups of six male Wistar rats were fed ad libitum, and were killed at 3, 4, 6, 9, 12, 18, 26 and 48 weeks of age. Carnoy's fixed tissue was later stained with the Feulgen reaction and was carefully microdissected; the number of native mitoses and apoptotic bodies per crypt and branched crypts were scored.

Autopsy

At autopsy, the stomachs, the small intestines, the caeca and the colons were rinsed, blotted and weighed; tissues from 10, 50 and 90% of the length of the small intestine and the colon were fixed in Carnoy's fluid for 3 h and were stored in 70% ethanol.

Assessment of proliferation/fission/apoptosis

Proliferation, apoptosis and crypt fission were measured, using previously validated methods (Goodlad et al. 1991; Goodlad 1994). Briefly, representative samples of tissue from the proximal, mid and distal small intestine and colon were hydrated, hydrolysed and stained with the Feulgen reaction. The mucosal crypts were gently teased apart under a dissection microscope, and the number of mitoses per crypt and apoptotic bodies per crypt in 20 crypts was scored. Two hundred crypts in this ‘microdissected’ tissue were then examined in order to determine the percentage of crypts in fission (crypt fission index). All samples were counted in a blinded fashion.

Statistics

Results have been presented as the mean ± SEM. Data were tested by means of analysis of variance (anova). Where statistical significance of the anova was reached (P < 0.05), individual comparisons were performed by using Dunnet's test, a method equivalent to repeated measures analyses. In addition, the non-parametric two-sided run test was used in order to establish whether the data were non-random.

Minitab Statistical Software was used (release 10.5 Xtra Minitab Ltd, Coventry, UK).

Results

Body weight and intestinal lengths and tissue weights

The rats gained weight rapidly from 3 to 12 weeks of age (Figure 1a), but the rate of weight increase then slowed down and stabilized at week 18 with only a small increase being seen after this time point.

Figure 1.

Effects of age on body weight, intestinal lengths and tissue weights. Panel a shows the body weight and the lengths of the small intestine and of the colon; note the different scales used. Panel b shows the weight of the various regions of the gut drawn to the same scale and panel c these weights with the y-axis adjusted to allow the superimposition of the curves. Panel d shows the relative weights of the regions of the gut and panel e these weights standardized to a percentage of the initial values.

The length of the small intestine reached a plateau value earlier at 9 weeks. The length of the colon increased rapidly in the first 9 weeks and then continued to lengthen gradually.

The weights of the major regions of the gastrointestinal tract have been showed in Figure 1(b) and also in 1(c), in which the y-axes are adjusted to allow the superimposition of the curves. The small intestine increased more rapidly and peaked at week 9. The other regions showed a more gradual increase with near-maximum values being obtained at 18 weeks.

When the tissue weights were ‘standardized’ by expressing them as a percentage of total body weight, all regions showed a decrease in their relative weights until week 18 (Figure 1d). The relative decline was less pronounced for the caecum, and was especially pronounced for the small intestine (from 4.2 ± 0.05 to 1.8 ± 0.04%).

Figure 1(e) shows the weight values expressed as percentage of the values at week 3, which graphically demonstrates that the relative contribution of the small intestine to body weight decreased to 44% of the week 3 value, whereas that of the stomach decreased to 59% and the caecum and the colon only to 65 or 66% of the week 3 values.

Cell proliferation, fission and apoptosis

Mitotic activity in the small intestine more than doubled (from 3.8 ± 0.1 mitoses per crypt at 3 weeks to 7.8 ± 0.4 at 9 weeks, P < 0.001), and then gradually decreased (Figure 2). In the colon, proliferation was maximum at the first time point studied (3 weeks), then rapidly decreased until 9 weeks (from 4.2 ± 0.2 to 2.5 ± 0.1 mitoses per crypt, P < 0.001) where it remained constant.

Figure 2.

Effects of the various ages on cell proliferation crypt fission and apoptosis in the mid small intestine and the mid colon.

Crypt fission in the small intestine increased from 4.6 ± 0.8% at 3 weeks to 8.4 ± 0.9% at 6 weeks and then decreased rapidly from week 6 to week 18 and then decreased gradually reaching a value of 1.5 ± 0.4% at 48 weeks. In the colon, crypt fission decreased rapidly from 3 weeks to 9 weeks (from 67.6 ± 4.2 to 23.1 ± 4.6%, P < 0.01) and then continued to gradually decline. The colon crypt fission index at 48 weeks was 9.8 ± 1.0%, whereas in the small intestine it was 1.46 ± 0.7%.

Apoptosis was a rare event; maximum values in the small intestine were seen at 6 weeks and were very low after 12 weeks. More apoptosis was seen in the colon with peak values at 3 weeks of 0.41 ± 0.16% followed by a decrease to 0.19 ± 0.06% apoptotic bodies per crypt at week 48; the comparable value in the small intestine was 0.02 ± 0.01%.

Discussion

The changes in all parameters measured were most pronounced in the young rats. We have previously shown that crypt cell production was highest in the young rats, which had fewer crypts per unit area and a low crypt-villus ratio (Goodlad & Wright 1990), demonstrating that a considerable amount of proliferative effort in the young rats is being devoted to cryptogenesis.

The main increase in the body weight of the rats occurred by week 18; however, the rats still gained a further 8.8% in weight between weeks 18 and 48 (P < 0.001) and the stomach, the small intestine and the colon also increased significantly over this period. However, when the results were expressed as a percentage of body weight, the curves were very flat from week 18 to 48. The curves for the various regions of the gut were generally similar, but it did appear that the small intestine reached adult values earlier than the colon. The decline in the relative weights of the gastrointestinal tract as final body weight is approached will reflect the reduced demands on the gut. It is of interest that the small intestine weight decreased the most, perhaps reflecting the importance of the small bowel in the growth phase. The energy cost of digestion is high, for periodic feeders this is equivalent to 32% of the meal's energy yield (Secor & Diamond 1995).

Proliferation, crypt fission and apoptosis increased dramatically in the small intestine between weeks 3 and 6 and then declined. These parameters in the colon were already at their maximum levels at the start of the time course and were nearly halved at week 9. Thus, in both tissues, the changes in proliferation and fission preceded changes in tissue weight (or length).

The results of our study are in contrast to those of Xiao et al. (2001) who reported that colonic cell production increased and apoptosis decreased in 12- to 14-month-old rats, compared to those in 4- to 6-month-old rats.

The reasons of this disparity in results is not clear, but Xiao et al. used Fischer-344 rats and used proliferating cell nuclear antigen immunoreactivity in order to assess proliferation and this method is less than ideal because PCNA can be induced in non-cycling cell (Hall et al. 1994; Goodlad et al. 2001); thus, claims of abnormality of the proliferation pattern in elderly humans (where the entire villi were PCNA-positive) cannot be taken at face value (Corazza et al. 1998). The investigation of intestinal epithelial cell proliferation and adaptation has often been bedevilled by the use of inappropriate methods (Goodlad & Wright 1982; Wright & Alison 1984) and the three-dimensional nature of the intestine means that morphological measures, such as villus height or crypt depth, should be used with caution (Clarke 1973; Hasan & Ferguson 1981). The capricious nature of many of the available kits for the TUNEL assay, combined with a lack of true specificity (Martin et al. 1998a; Wolvekamp et al. 1998; Alison 1999; Duan et al. 2003), also cast doubt on many published measures of apoptosis.

Most studies, including this one, mainly measure proliferation in the transit-amplifying part of the crypt, and it may be that there are changes in the stem cell region. Few studies have addressed this question, but Potten et al. have showed that after high doses of irradiation, the surviving crypts in old mice were both smaller and fewer in number than those in young mice; however, the number of clonogenic cells per crypt was estimated to be greater in the older mice (Martin et al. 1998b) which may be the results of altered levels of p53 and p21 expression, suggesting an age-related defect in the capacity to recognize damage and initiate apoptosis or repair (Potten et al. 2001). If this were the case and if there are two types of apoptosis spontaneous and stress-induced (Watson et al. 2000), our study would only measure the former. The low rates of apoptosis made it difficult to determine whether the apoptotic activity was in the stem cell or in the transit-amplifying zones.

When the organ weights were expressed as percentage of bodyweight at autopsy, the results showed that these relative weights were very constant after week 18 and proliferation and apoptosis were also fairly constant. Crypt fission was still falling after this time, especially in the colon. However, it must be stressed that fission indices were approximately eight times greater in the colon than those in the small bowel throughout the duration of the investigation.

Crypt fission is, in addition, seen in normal human colonic mucosa and is common in crypts isolated from adenomas and hyperplastic polyps, leading to the suggestion that sporadic human colorectal adenomas and hyperplastic polyps grow by crypt fission (Wong et al. 2002).

Elevated fission is, moreover, seen in Min mice and FAP patients (Wasan et al. 1998) and in carcinogen-treated rodents (Park et al. 1997). The greater rates of fission in the colon may lead to the more rapid evolution of wholly mutated crypts in the colon (Park et al. 1995) and could thus provide an explanation why the colon is so much more susceptible to carcinogenesis.

In studies on the ageing process, it is very important to distinguish between the effects of age itself and those of other changes that may be associated with old age. One of the most likely of such confounding factors is the decrease in food intake often found in old age, and food intake is one of the most important determinants of intestinal proliferation and absorption (Goodlad et al. 1987). The variations in the various parameters in the adult groups in the present study demonstrate the importance of making the correct choice of the reference point for the ‘control’ adult values, determining that this reference point is stable. For rats with a stable relative gut weight, the results showed in Figure 1 would indicate that 18-week-old rats are used, some of the proliferative parameters seem to have plateaued earlier. Even so the weight of a 12-week rat is over 300 g and many experiments have used much younger rats in which all the proliferative parameters would be in a state of flux, which could mask the results of potential interventions.

The results of this study confirm the dramatic changes that occur in the developing gut, but do not show any significant changes associated with age.