Modulation of mouse intestinal epithelial cell turnover in the absence of angiotensin converting enzyme

Abstract

Angiotensin converting enzyme (ACE) has been shown to be involved in regulation of apoptosis in nonintestinal tissues. This study examined the role of ACE in the modulation of intestinal adaptation utilizing ACE knockout mice (ACE^−/−). A 60% small bowel resection (SBR) was used, since this model results in a significant increase in intestinal epithelial cell (EC) apoptosis as well as proliferation. Baseline villus height, crypt depth, and intestinal EC proliferation were higher, and EC apoptosis rates were lower in ACE^−/− compared with ACE^+/+ mice. After SBR, EC apoptosis rates remained significantly lower in ACE^−/− compared with ACE^+/+ mice. Furthermore, villus height and crypt depth after SBR continued to be higher in ACE^−/− mice. The finding of a lower bax-to-bcl-2 protein ratio in ACE^−/− mice may account for reduced EC apoptotic rates after SBR in ACE^−/− compared with ACE^+/+ mice. The baseline higher rate of EC proliferation in ACE^−/− compared with ACE^+/+ mice may be due to an increase in the expression of several EC growth factor receptors. In conclusion, ACE appears to have an important role in the modulation of intestinal EC apoptosis and proliferation and suggests that the presence of ACE in the intestinal epithelium has a critical role in guiding epithelial cell adaptive response.

after massive small bowel resection (SBR), the residual intestine undergoes a series of adaptive processes resulting in a significant increase in intestinal absorptive surface area (22, 57). The exact mechanisms of postresectional intestinal adaptation remain incompletely understood, although a number of nutritive and nonnutritive factors have been identified as potential mediators of this process (35, 42, 44, 53, 65). The adaptive process includes not only increased epithelial cell (EC) proliferation, but also increased rates of enterocyte apoptosis after massive SBR in rodents (15, 18, 52). The bcl-2 family of intracellular proteins has been shown to play an important role in the regulation of intestinal EC apoptosis via the intrinsic apoptotic pathway (15, 29, 41, 50). An increase in the ratio between proapoptotic and antiapoptotic members of this family of proteins has been reported after massive SBR (49). Fas and tumor necrosis factor-α (TNF-α), both potent inducers of apoptosis by way of the death receptor pathway, have also been shown to be upregulated after SBR in mice (15, 52, 61) and may therefore also be significantly involved in the regulation of postresectional EC apoptosis in the frame of the extrinsic apoptotic pathway (12).

Our laboratory has previously identified increased gene and protein expression of intestinal EC-derived angiotensin-converting enzyme (ACE) after SBR in mice (60, 61). ACE is responsible for the cleavage of angiotensin I to angiotensin II (36). In the intestinal epithelium ACE has been shown to be a member of intestinal brush-border membrane enzymes and to play a role in the terminal digestion of various peptides, in particular prolyl peptides (10).

Because ACE has been shown to promote apoptosis in various tissues (6, 32, 46, 62), we were interested to find out whether locally derived ACE may also have similar actions in the intestinal mucosal tissue. We recently reported that after massive SBR a significant reduction of EC apoptosis and a moderate enhancement of intestinal adaptation (increased crypt depth and increased EC proliferation) were found in mice treated with the ACE-inhibitor enalaprilat (61). This marked decline in enterocyte apoptosis was associated with decreased gene expression of TNF-α, indicating that ACE may be involved in the modulation of intestinal EC apoptosis via the death receptor pathway.

In the present study we investigated whether the changes observed after the administration of an ACE inhibitor could have been a drug effect or whether ACE actually influences postresectional intestinal adaptation through changes in EC apoptosis and proliferation. To approach this question, we utilized a group of ACE knockout mice (ACE^−/−). We hypothesized that ACE^−/− mice would show decreased EC apoptosis and increased intestinal adaptation after SBR. The mechanisms involved in the process of enterocyte apoptosis and proliferation were also studied in these knockout mice.

METHODS

Animals

Studies reported here conformed to the guidelines for the care and use of laboratory animals established by the University Committee on Use and Care of Animals at the University of Michigan (Ann Arbor, MI), and protocols were approved by that committee. To investigate the role of ACE in intestinal adaptation, C57BL/6J ACE^−/− (Ace^tm1Unc) mice (homozygous mutants lacking both the somatic and germinal isoforms of the ACE gene) were bred in the breeding colony of the animal care facility of the University Michigan using founders generously provided by Dr. Oliver Smithies (30). Mice used in the experiments of this study were the fourth generation of the received colony founders. Genotyping was performed using tail DNA according to the previously described protocol (23). In each set of experiments C57BL/6J wild-type (ACE^+/+) littermates of ACE^−/− mice were used as controls. Animals were maintained in a 12-h day-night rhythm at 23°C and a relative humidity of 40–60%. A standard rodent chow (LabDiet 5001 Rodent Diet, PMI Nutrition International, Brentwood, MO) was switched to microstabilized rodent liquid diet (TestDiet, Richmond, IN) 2 days prior to surgery, and mice were maintained thereafter on liquid diet until harvest.

Experimental Design

To investigate the impact of lack of ACE on the postresectional intestinal adaptive changes in jejunum and ileum, a 60% midgut resection was performed in male ACE^−/− and ACE^+/+ mice (n = 6 in each group). Jejunum and ileum from nonoperated age-matched male ACE^−/− and ACE^+/+ mice were used as controls (n = 6 in each group).

Surgical Procedure

Resection of the small bowel was performed between the point 6 cm distal to the ligament of Treitz and 6 cm proximal to the ileocecal junction as previously described (16, 20). On the first postoperative day mice were given only water and thereafter had free access to water and liquid diet. No antibiotics were used. Body weights were determined preoperatively and at harvest.

Harvesting

Mice were euthanized at 7 days postoperatively by use of CO₂. Small bowel (0.5 cm) segments were preserved in 10% buffered formalin. Jejunal and ileal tissues were taken 3 cm distal to the ligament of Treitz and 3 cm proximal to the ileocecal junction. Small bowel 0.5 cm proximal and distal to anastomotic sites were discarded. The remaining small bowel was immediately processed for mucosal cell isolation.

Histochemical Detection of ACE

Paraffin sections were used to detect ACE activity along the crypt-villus axis by previously described techniques (63). Primary antibody ACE monoclonal antibody (0.5 μg/ml; BD PharMingen), or PBS (negative control) were applied overnight at 4°C, followed by secondary antibody and horseradish peroxidase.

Measurements of Mucosal and Mesenteric Blood Flow

Because ACE^−/− mice have been shown to have approximately a 15–20 mmHg lower blood pressure compared with ACE^+/+ mice (54), we evaluated intestinal blood flow in ACE^−/− and ACE^+/+ mice using laser Doppler perfusion imaging (LDPI, Perimed, North Royalton, OH) as previously reported (48). Mesenteric as well as jejunal and ileal mucosal blood flows were measured in anesthetized mice (n = 5, in each group). A LDPI 670-nm helium-neon laser beam was placed 12 cm above the mesentery and sequentially scanned the surface of the mesentery, as well as the jejunal and ileal mucosa over a 2-cm length in each segment. Maximum, minimum, and mean percent perfusion was normalized to total pixel area. At the end of the measurements the mice were euthanized.

Intestinal Morphology and Histology

Paraffin embedded tissues were (5-μm thickness) stained with hematoxylin and eosin. Image Pro Plus Software (Media Cybernetics, Silver Spring, MD) was used for measurements of villus height and crypt depth. Means of 10 replicate measurements were made per tissue section.

Measurements of Intestinal Epithelial Cell Diameters

Because intestinal adaptation after SBR is comprised of both EC hyperplasia (increase in total number of ECs) as well as cellular hypertrophy (14, 52, 59), we analyzed the postresectional changes in EC diameters compared with the nonoperated mice. EC diameters were calculated as previously reported (16). In short, the villus height, or crypt depth, was divided by the total number of ECs counted on the one half-side of the respective villus or crypt, respectively. A minimum of five well-oriented villi and crypts were counted per tissue section and the results were averaged. The comparison between percent changes in EC diameters and respective percent changes in crypt depths and villus heights allowed for better insights concerning enterocyte hyperplasia and/or hypertrophy at 1 wk postresection compared with respective control groups in ACE^−/− and ACE^+/+ mice.

Epithelial Cell Proliferation Assay

5-Bromo-2-deoxyuridine (BrdU) was injected into mice 1 h before harvest (50 mg/kg ip; Roche Diagnostic, Indianapolis, IN) and used to determine EC proliferation rates (7), via a BrdU In Situ Detection Kit II according to the manufacturer's guidelines (BD PharMingen). An index of the crypt cell proliferation rate was calculated by the ratio of number of crypt cells incorporating BrdU to total number of crypt cells. The total number of proliferating cells per crypt was defined as a mean of proliferating cells in 10 well-oriented crypts (counted at ×400).

Detection of Epithelial Cell Apoptosis

A terminal deoxynucleotidyl transferase biotin-dUTP nick-end labeling (TUNEL) staining method was used to detect apoptosis, according to manufacturer's instructions (ApopTag Plus Peroxidase InSitu Apoptosis Detection Kit, Chemicon International, Temecula, CA), with slight modification. Slides were incubated with only one-third of the recommended concentration of terminal deoxynucleotidyl transferase enzyme, to avoid overstaining.

Quantification of Apoptosis

Assessment of apoptosis consisted of separate counting of all TUNEL-positive ECs in all well-oriented crypts and villi separately and dividing the total number of counted apoptotic cells per number of analyzed crypts and villi, respectively. Apoptotic index in the region of villi is expressed as the number of TUNEL-positive cells per one villus. Apoptotic index in the crypt region is expressed in two different ways, based on a modification of previously described approaches (34, 61): first as the number of apoptotic cells per one crypt, and second (for more detailed analysis of the location of apoptosis in the crypt) cells in each half-crypt were numbered starting at the base of the crypt column. The number of apoptotic cells at each position in the half-crypt was recorded as the ratio of all counted TUNEL-positive cells at this position per 100 cells at this position.

Mucosal Cell Isolation and Purification

Mucosal cells were isolated and EC purified as previously described (26).

RT-PCR

Total RNA was isolated by a guanidine isothiocyanate/chloroform extraction method using Trizol (GIBCO BRL, Gaithersburg, MD). EC mRNA (poly-A positive) was reversed transcribed into cDNA by following a standard protocol (64). Specific primers were designed with use of proprietary software (LaserGene, DNAStar, Madison, WI). PCR and gels were run under standard conditions (64). Results were expressed as the ratio of the investigated mRNA over β-actin mRNA expression.

Immunoblot Analysis

Protocols were similar to those previously described (63). Primary antibodies included: monoclonal mouse anti-bcl-2 antibody (1:400, in blocking solution; BD PharMingen, San Diego, CA), polyclonal rabbit anti mouse bax antibody (1:1,000, in blocking solution, BD PharMingen), or monoclonal Armenian hamster anti-mouse Fas antibody (1:500, in blocking solution, BD PharMingen). Detection of β-actin was performed by reprobing membranes (after striping) with anti-mouse β-actin (1:8,000, Sigma-Aldrich). The results are expressed as the ratio of target protein over β-actin protein expression.

ELISA

A commercially available ELISA kit (DuoSet, R&D Systems, Minneapolis, MN) was used for detection of TNF-α protein expression. The assay was performed in duplicate as specified by the manufacturer. Optical density was assessed by using an automated plate reader set at a wavelength of 450 nm with a correction reading of 540 nm (Synergy HT-1 automated fluorescent plate reader, Bio-Tek Instruments, Winooski, VT), and cytokine concentration was determined from the standard curve. Results are expressed as nanograms of target cytokine per microgram of total protein.

Electron Microscopy

Because of the high expression of ACE in the region of microvilli, we were interested to see whether there was a difference in the length and/or thickness of microvilli between the ACE^−/− and ACE^+/+ mice. For this purpose 3-mm full-thickness circular segments were taken from jejunum and ileum of the nonoperated ACE^−/− and ACE^+/+ mice, cut in 1-mm pieces, fixed in 2% buffered glutaraldehyde for 1 h, postfixed in buffered 1% osmium tetroxide for 1 h, dehydrated, and infiltrated with Epon epoxy resin. Blocks were sectioned and grids containing “ultrathin” sections were double stained with lead citrate and uranyl acetate. The Philips CM-100 transmission electron microscope was operated at 60 kV. The measurements of microvilli were performed at a magnitude of ×25,000.

Statistical Analysis

All data are expressed as means ± SD, unless indicated otherwise. For data in which we used repeated-measures analysis, data were expressed as means ± SE. The comparisons among groups were done using either Student t-tests or one-way ANOVA followed by a Bonferroni t-test for post hoc analysis of significance using Graph Pad Prism, Version 4.0 software (GraphPad, San Francisco, CA). A value of P < 0.05 was considered significant.

RESULTS

General Description

ACE knockout mice characteristically have lower blood pressure and poorer survival characteristics compared with wild-type mice (30). Therefore, it was not surprising that postoperative mortality rates were higher in the ACE^−/− group (2 of 8 ACE^−/− mice died in the first 48 h after SBR) compared with ACE^+/+ group (no deaths). As shown by other investigators (14, 59) intestinal resection led to a significant body weight decline at 7 days postsurgery in both SBR groups; however, there was a significantly higher body weight loss in the ACE^−/− group compared with ACE^+/+ group (percent change from weight at surgery: −19.7 ± 3.9 vs. −8.9 ± 3.3%, respectively: P < 0.001). Of further note is that body weight at 8 wk of life was significantly lower in ACE^−/− compared with ACE^+/+ group (19.4 ± 0.8 vs. 22.4 ± 0.7 g, respectively: P < 0.001). Despite these differences, surviving mice in both groups showed macroscopically strong adaptive intestinal growth changes and were without signs of intestinal obstruction. All mice had normal stooling patterns.

Small Intestinal Expression of ACE

We first wanted to confirm the expression of ACE in the intestinal epithelium. Immunohistochemical studies showed a very intense expression of ACE in the region of microvilli and a strong ACE staining in the small bowel epithelium of the ACE^+/+ mice (Fig. 1). Furthermore, positive ACE staining was found in intraepithelial lymphocytes and a small number of cells in the lamina propria. There was no positive ACE staining in ACE^−/− mice (Fig. 1).

Fig. 1.

Immunohistochemical examination of wild-type (ACE^+/+) mouse jejunum showed that the strongest immunoreactivity for angiotensin converting enzyme (ACE) appeared to be localized along the mucosal epithelium. A: negative control (staining without anti-ACE antibody). B: positive stain (with anti-ACE antibody) in ACE^+/+ mouse. C: positive stain (with anti-ACE antibody) in ACE^−/− mouse.

Intestinal Blood Flow Measurements

We next wanted to determine whether blood flow to the mesentery and mucosa of the small bowel was comparable in ACE^−/− and ACE^+/+ groups. Figure 2 shows representative two-dimensional high-resolution laser Doppler images of the mesenteric blood flow, which was found, in fact, to be significantly elevated in ACE^−/− compared with ACE^+/+ mice (P < 0.05). Furthermore, laser Doppler measurements of intestinal blood flow at the mucosal level did not show any significant differences in the degree of perfusion between ACE^−/− and ACE^+/+ mice; this was true for both jejunum and ileum. This suggested that any intestinal adaptive changes would not be due to differences in intestinal blood flow between ACE^−/− and ACE^+/+ mice.

Fig. 2.

Mesenteric and mucosal ileal and jejunal intestinal blood flow in nonoperated ACE^+/+ and ACE knockout (ACE^−/−) mice as measured by laser Doppler perfusion imager. Note that at mucosa level there was no significant difference in the intestinal blood flow between the 2 strains of mice, whereas at the level of mesentery ACE^−/− mice showed a significantly higher value of blood flow compared with ACE^+/+ mice. Representative 2-dimensional laser Doppler images of the mesenteric blood flow are shown for ACE^+/+ (A) and ACE^−/− (B) mice. Values are means ± SD. N = 5 in each group. *P < 0.05.

Intestinal Morphometry

The jejunal and ileal morphological adaptive responses after SBR in ACE^+/+ and ACE^−/− mice are shown in Table 1. Compared with the respective controls, villus height and crypt depth were significantly increased by 1 wk after SBR in both ACE^+/+ and ACE^−/− groups of mice (P < 0.001; Table 1).

Table 1.

Villus heights and crypt depths in jejunum and ileum of ACE^+/+ and ACE^−/− mice

Morphometry	N-ACE+/+	SBR-ACE+/+	% Change	N-ACE−/−	SBR-ACE−/−	% Change
Jejunal villus height, μm	331±25	585±31c	+77%	430±29e	663±79c,f	+54%
Jejunal crypt depth, μm	92±7	168±8c	+83%	115±5e	202±19c,d	+76%
Ileal villus height, μm	217±7	290±13c	+34%	272±17d	330±37c,e	+21%
Ileal crypt depth, μm	81±4	112±3c	+38%	103±10d	115±10	+12%

Values are means ± SE; N = 6 in each group. Absolute values as well as mean percent changes from baseline are given for nonoperated mice (N) and mice at 7 days after a 60% mid small-bowel resection (SBR). Note, first, the significant postresectional increase in villus height and crypt depth in both strains of mice, and second, a significantly higher villus height and crypt depth in angiotensin converting enzyme knockout (ACE^−/−) mice compared with wild-type (ACE^+/+) mice. Significance levels are given according to the post hoc analysis. For Tables 1–3:

^aP < 0.05,

^bP < 0.01,

^cP < 0.001 (N vs. SBR groups);

^dP < 0.001,

^eP < 0.01 and

^fP < 0.05 (N-ACE^+/+ vs. N-ACE^−/− and/or SBR-ACE^+/+ vs. SBR-ACE^−/−).

Morphometric changes between ACE^+/+ and ACE^−/− mice.

Although both strains of mice underwent adaptive changes after SBR, distinct differences were noted between the mice. At baseline (prior to surgery), villus height and crypt depth were noted to be significantly greater in nonoperated ACE^−/− mice compared with ACE^+/+ mice (Table 1). This was true for both jejunum and ileum. After SBR, ACE^−/− mice continued to have greater absolute values of villus heights and crypt depths compared with ACE^+/+ mice, except for ileal crypt depth, which did not show any postresectional difference between the two strains of mice (Table 1).

Epithelial cell hyperplasia vs. hypertrophy in ACE^+/+ and ACE^−/− mice.

To examine the impact of cellular morphometric alterations on intestinal morphometry we compared changes in EC diameter to respective morphometric changes (crypt depth and villus height). Interestingly, at baseline EC diameters in intestinal crypts were significantly greater in ACE^−/− mice compared with ACE^+/+ mice for both jejunal and ileal tissues (Table 2). This suggests that cell hypertrophy may be one mechanism by which ACE^−/− mice might have increased crypt depth in nonoperated mice. EC diameters were not significantly different, at baseline, in villi of ACE^+/+ and ACE^−/− mice. One week after SBR, EC diameters increased in crypt regions of both jejunal and ileal tissues of ACE^−/− mice, whereas in ACE^+/+ mice a significant increase in crypt EC diameters was found only for jejunal tissues. EC hypertrophy remained, however, significantly greater in intestinal tissues of ACE^−/− mice, except for the ileal villi, where EC diameters strongly increased after SBR in ACE^+/+ mice and were actually significantly higher compared with ACE^−/− mice at 1 wk after SBR (P < 0.05). Finally, in jejunal villi of ACE^+/+ mice EC diameters decreased after SBR, indicating the appearance of small villus ECs 1 wk after SBR.

Table 2.

Epithelial cell diameters as a measure of cellular hypertrophy as reflected in mean cell diameter

Epithelial Cell Size	N-ACE+/+	SBR-ACE+/+	% Change	N-ACE−/−	SBR-ACE−/−	% Change
Jejunal crypt, μm	3.3±0.1	3.9±0.3a	+18%	5.2±0.2d	5.8±0.3a,d	+11%
Jejunal villus, μm	4.8±0.2	4.1±0.3a	−15%	4.8±0.1	5.7±0.2a,d	+18%
Ileal crypt, μm	4±0.1	4.3±0.2	+8%	4.8±0.2e	5.9±0.2b,d	+23%
Ileal villus, μm	4.3±0.2	5.5±0.3b,f	+28%	4.3±0.1	4.8±0.3	+12%

Values are means ± SE; N = 6 in each group. Note significant differences in crypt cell diameters between ACE^+/+ and ACE^−/− mice in both jejunal and ileal intestinal segments, at baseline (nonoperated mice) and at 7 days after mid-small-bowel resection. Absolute values as well as mean percent changes from baseline are given separately for jejunum and ileum. Significance levels are given according to the post hoc analysis. For Tables 1–3:

^aP < 0.05,

^bP < 0.01,

^cP < 0.001 (N vs. SBR groups);

^dP < 0.001,

^eP < 0.01 and

^fP < 0.05 (N-ACE^+/+ vs. N-ACE^−/− and/or SBR-ACE^+/+ vs. SBR-ACE^−/−).

Comparing the percent changes in crypt depth and villus height (Table 1) with the respective percent changes of EC diameters (Table 2), we found that in jejunal segments at 1 wk after SBR increases in villus height and crypt depth significantly exceed the respective increases in EC diameter, i.e., cell hyperplasia dominated clearly over cell hypertrophy. In contrast, in the ileal segments postresectional changes in villus height and crypt depth were comparable to the respective changes in EC diameter, i.e., cell hypertrophy dominated over cell hyperplasia. This was true for both strains of mice.

Epithelial Cell Proliferation

EC proliferation (BrdU staining) is shown for representative sections of jejunum and ileum in Fig. 3. Consistent with our previous report in ACE inhibitor-treated mice (61), at baseline (nonoperated mice) EC proliferation rates were found to be consistently higher in ACE^−/− mice compared with ACE^+/+ mice (Fig. 4). This reached statistical significance for jejunal tissue only. EC proliferation rates in both ACE^+/+ and ACE^−/− mice increased significantly at 1 wk postresection compared with nonoperated animals of respective strains, and these levels of proliferation were not different between strains. The higher baseline crypt cell proliferation rates in jejunum and ileum of nonoperated ACE^−/− mice may further explain the observed higher crypt depth and villus height in nonoperated ACE^−/− mice compared with ACE^+/+ mice.

Fig. 3.

Representative photomicrographs of jejunum and ileum of nonoperated (N) and small-bowel resected (SBR) ACE^+/+ and ACE^−/− mice after immunohistochemical staining with 5-bromodeoxyuridine (BrdU) as a marker of epithelial cell (EC) proliferation. An intraperitoneal injection of BrdU at a dosage of 50 mg/kg body wt was performed 1 h before death. Note the significant increase in the number of BrdU-positive EC after SBR and that all cells incorporating BrdU remained in intestinal crypts. Bar = 100 μm. Original magnification ×200.

Fig. 4.

Jejunal and ileal crypt cell proliferation rates at 1 h after intraperitoneal injection of 5-bromodeoxyuridine (BrdU) in ACE^+/+ ACE^−/− mice. Data are given for nonoperated mice and mice at 7 days after a 60% mid-small-bowel resection. Note significant postresectional increases in crypt epithelial cell proliferation rates in both strains of mice. Bars indicate significance levels according to the post hoc analysis. Values are means ± SE. N = 6 in each group. §P < 0.001, *P < 0.05.

Expression of Cell Proliferation Related Genes

Absence of ACE leads to a marked increased expression of growth factor receptors. Interestingly, at baseline (prior to SBR) a significantly higher gene expression of KGFR₁, EGFR, and IL-7R was noted in ACE^−/− compared with ACE^+/+ mice (Fig. 5). mRNA expression of KGFR₁ and EGFR were threefold higher in ACE^−/− mice, and IL-7R was nearly twofold higher in knockouts. Following SBR, expression of EGFR and IL-7R was also significantly increased. In ACE^+/+ mice mRNA expression of KGFR₁ increased significantly after SBR, whereas SBR had no effect on EGFR or IL-7R. This may suggest that the upregulation of these receptors was maximized already prior to surgery. Furthermore, because ACE^−/− mice still underwent similar or greater degrees of morphometric adaptive changes with SBR, it suggests that other mechanisms of action besides these growth factors may help mediate the observed increases in villus height and crypt depth.

Fig. 5.

mRNA expression of KGFR₁, EGFR and IL-7R in nonoperated mice and at 7 days after a 60% mid-small-bowel resection in ACE^+/+ and ACE^−/− mice. Note the significantly higher expression of the investigated genes in the ACE^−/− group of mice. Bars indicate significance levels according to the post hoc analysis. Values are means ± SD. N = 6 in each group. §P < 0.001, #P < 0.01.

Epithelial Cell Apoptosis

Enterocyte apoptosis rates were significantly lower in ACE^−/− compared with ACE^+/+ mice.

At baseline, rates of EC apoptosis in ileal crypts and villi were twofold lower in ACE knockout compared with wild-type mice (Fig. 6). In contrast, we found no statistical differences in baseline apoptotic rates between the two strains of mice in jejunum.

Fig. 6.

Jejunal and ileal crypt and villus cell apoptosis rates in ACE^+/+ and ACE^−/− mice. Data are given for nonoperated mice and mice at 7 days after a 60% mid-small-bowel resection. Note a significantly lower apoptotic index in the crypts of ACE^−/− mice at baseline and after SBR. Bars indicate significance levels according to the post hoc analysis. Values are means ± SD. N = 6 in each group. §P < 0.001, #P < 0.01, *P < 0.05.

SBR-associated epithelial cell apoptosis was markedly depressed in ACE^−/− mice.

SBR led to significantly increased EC apoptosis rates in both crypt and villus regions of both jejunum and ileum in ACE^+/+ mice compared with the nonoperated ACE^+/+ group (Fig. 6). Interestingly, in ACE^−/− mice SBR did not result in any change in EC apoptosis rates in the crypt regions of both jejunal and ileal segments. In the villi, however, rates of apoptosis did rise significantly after SBR in ACE^−/− and ACE^+/+ mice. However, similar to the crypt region, rates of apoptosis in the ileal segments were up to threefold lower in ACE^−/− mice compared with the respective groups of ACE^+/+ mice (Fig. 6).

We then analyzed the distribution of apoptotic cells according to the cell position in intestinal crypts. In ACE^−/− mice we found a marked reduction in apoptosis rates at the majority of cell positions in both jejunal and ileal crypts compared with ACE^+/+ mice (Fig. 7).

Fig. 7.

Crypt cell apoptosis: cell positional distribution of apoptosis in jejunal and ileal crypts from ACE^+/+ and ACE^−/− mice. Data are given for nonoperated mice and mice at 7-days after a 60% mid-small-bowel resection. A: quantitative results expressing the number of apoptotic cells at the respective position per 100 crypt cells at that position. B: representative tissue sections showing apoptotic epithelial cells at different cell positions in the intestinal crypts. Apoptotic EC are stained with the terminal deoxynucleotidyl transferase biotin-dUTP nick-end labeling (TUNEL) method [representative positive (brown) cells denoted with arrows]. Because of the overall relatively low apoptotic rates, the differences between groups cannot be shown well with presentation of the histological sections. The cell position 1 is at the base of the crypt. Note the similar peak of apoptotic cells around cell position 6 in both strains of mice. However, constantly lower apoptosis rates are noted in ACE^−/− compared with ACE^+/+ mice. Note the occurrence of additional apoptotic peaks after SBR in the ACE^+/+ mice around cell positions 12 and 18. Values are means form N = 6 in each group. Presentation of SD was omitted to simplify the graph.

Expression of Apoptosis-Related Factors

To investigate potential mechanisms that account for the marked decline in EC apoptosis rates in ACE^−/− mice, several factors of both the intrinsic and extrinsic pathways of apoptotic signaling were studied.

Intrinsic Pathway

A significant difference was noted in the expression of apoptosis related genes between ACE^−/− and ACE^+/+ mice.

The mRNA expression of the antiapoptotic factor bcl-2, a member of the bcl-2 family, showed a significant increase in postresectional expression in both strains of mice (Fig. 8). Surprisingly, despite the lower levels of EC apoptosis, the degree of bcl-2 elevation was significantly lower in the ACE^−/− group of mice.

Fig. 8.

mRNA expression of bcl-2, bax, and bid in nonoperated mice and at 7 days after a 60% mid-small-bowel resection in ACE^+/+ and ACE^−/− mice. Note the significant increase in bcl-2 gene expression after SBR in both strains of mice. Bars indicate significance levels according to the post hoc analysis. Values are means ± SD. N = 6 in each group. §P < 0.001, #P < 0.01.

The expression of the proapoptotic genes bax and bid did not change significantly after SBR compared with the respective nonoperated groups of mice (P = not significant). However, bax mRNA expression was consistently higher in ACE^−/− mice compared with ACE^+/+ mice, again, despite these mice having significantly lower rates of EC apoptosis. Similarly, postresectional mRNA expression of the proapoptotic factor bid was also found to be significantly higher in ACE^−/− compared with ACE^+/+ mice (P < 0.01).

Western immunoblot analysis was performed to further analyze some of the key changes observed in the mRNA data. Similar to our mRNA findings, a significantly increased postresectional expression of the antiapoptotic bcl-2 protein was found in both strains of mice. However, in contrast to the mRNA findings, the expression of the proapoptotic protein bax increased significantly after SBR only in the ACE^+/+ group of mice (Fig. 9). We then examined the ratio between the proapoptotic bax and antiapoptotic bcl-2 protein expression and found that a significant postresectional decrease in this ratio occurred only in the ACE knockout group of mice (bax-to-bcl-2 ratio, 6.9 ± 1.2 vs. 2.2 ± 0.9, respectively; P < 0.001), and not in the wild-type mice (bax-to-bcl-2 ratio, 2.1 ± 0.8 vs. 1.9 ± 0.7, respectively; P = not significant). Based on these protein values, these findings support the concept that alterations in the ratio of pro- to antiapoptotic members of the bcl-2 family may play a crucial role for the control of apoptosis in intestinal epithelium and could also be a mechanism by which ACE may mediate its proapoptotic effects.

Fig. 9.

Protein expression of bcl-2 and bax in nonoperated mice and at 7 days after a 60% mid-small-bowel resection in ACE^+/+ and ACE^−/− mice. Top: representative respective Western blot results from isolated intestinal epithelial cells. Bottom: quantitative results of bax and bcl-2 expression normalized for β-actin expression. Note that bcl-2 protein expression increased significantly after SBR in both strains of mice, whereas bax protein expression showed a significant postresectional increase only in the ACE^+/+ mice. Bars indicate significance levels according to the post hoc analysis. Values are means ± SD. N = 6 in each group. #P < 0.01.

Extrinsic Pathway

At baseline (nonoperated groups), gene expression of the investigated members of the extrinsic apoptotic pathways (TNF-α, TNFR₁, Fas, and FasL) were not significantly different between the ACE^+/+ and ACE^−/− mice (Fig. 10). SBR led to a significant increase in all these factors in ACE^−/− mice, and a significant increase in TNF-α and FasL in the ACE^+/+ group of mice.

Fig. 10.

Alterations in mRNA expression of the investigated members of the extrinsic apoptotic pathways in nonoperated mice and at 7 days after a 60% mid-small-bowel resection in ACE^+/+ and ACE^−/− mice. Bars indicate significance levels according to the post hoc analysis. Values are means ± SD. N = 6 in each group. #P < 0.01, *P < 0.05.

To further analyze some of the observed changes in the mRNA data, protein expression of Fas and TNF-α were assessed (Fig. 11). At baseline both these proapoptotic proteins showed a higher expression in ACE^−/− mice compared with ACE^+/+ mice. However, after SBR Fas protein expression increased significantly only in ACE^+/+ mice (P < 0.01), whereas TNF-α showed a significant decrease in the group of ACE^−/− mice (P < 0.05). Again, on the basis of these protein levels, these data suggest that a number of examined proteins known to trigger apoptosis show an increased postresectional expression only in ACE^+/+ mice (Fas and Bax) and show no significant postresectional change or even a significant decrease (TNF-α) in ACE^−/− mice.

Fig. 11.

Alterations in protein expression of Fas and TNF-α in nonoperated mice and at 7 days after a 60% mid-small-bowel resection in ACE^+/+ and ACE^−/− mice. Top: representative Western blot result of Fas and β-actin expression from isolated intestinal epithelial cells. TNF-α was detected by ELISA. Bottom: quantitative results of Fas expression normalized for β-actin expression and quantitative results of TNF-α expression expressed as nanograms TNF-α per microgram of total protein. Note that, despite the significantly higher levels of Fas protein expression in the ACE^−/− mice compared with ACE^+/+ mice, a significant postresectional increase in Fas protein expression occurred only in the ACE^+/+ group of mice. Furthermore, there was a significant decrease in TNF-α protein expression after SBR in the ACE^−/− group of mice. Bars indicate significance levels according to the post hoc analysis. Values are means ± SD. N = 6 in each group. §P < 0.001, #P < 0.01, *P < 0.05.

Transmission Electron Microscopy

Finally, because ACE is known to be an important enzyme in the terminal digestion of a series of peptides, and our immunohistochemical studies showed ACE expression to be the strongest in the region of microvilli, we were interested to see whether there was a difference in the morphology of microvilli between ACE^+/+ and ACE^−/− mice. To address this, electron microscopic imaging was performed. Transmission electron microscopy showed that the microvilli of nonoperated ACE^−/− mice were significantly longer and thinner compared with the respective group of ACE^+/+ mice (P < 0.01; Table 3). This was found to be true for both jejunum and ileum (Fig. 12).

Table 3.

Comparison of the microvillus length and thickness between the nonoperated ACE^−/− and ACE^+/+ mice

	N-ACE+/+	N-ACE−/−	% Difference
Microvilli length, μm
Jejunum	1.132±0.18	1.480±0.09f	+31%
Ileum	1.059±0.12	1.498±0.13f	+41%
Microvilli thickness, μm
Jejunum	0.147±0.06	0.104±0.09f	−29%
Ileum	0.149±0.04	0.109±0.05f	−27%

Note that ACE^−/− mice have significantly longer and thinner microvilli compared with microvilli of ACE^+/+ mice. Absolute values as well as mean percent difference between ACE^−/− and ACE^+/+ mice are given for both jejunum and ileum. Values represent means ± SE of measurements from at least 5 sections. For Tables 1–3:

^aP < 0.05, ^bP < 0.01, ^cP < 0.001 (N vs. SBR groups); ^dP < 0.001, ^eP < 0.01 and

^fP < 0.05 (N-ACE^+/+ vs. N-ACE^−/− and/or SBR-ACE^+/+ vs. SBR-ACE^−/−).

Fig. 12.

Representative transmission electron microscopy images of ileum and jejunum from nonoperated ACE^+/+ and ACE^−/− mice at ×25,000 original magnification are shown. Note the longer and thinner microvilli in both jejunum and ileum of ACE^−/− compared with ACE^+/+ mice.

DISCUSSION

Intestinal adaptation after massive SBR in rodents is structurally characterized by an increase in EC proliferation and EC apoptosis rates (15, 17, 25, 33, 42, 58). The precise mechanisms that guide these changes continue to be investigated. This study investigated for the first time the process of postresectional intestinal adaptation in mice lacking ACE and showed that in the absence of this factor a marked reduction in EC apoptosis was noted in the ileum, and a marked decline in postresectional apoptosis was noted in both jejunum and ileum. The predominant decline in apoptosis was in the crypt region. Interestingly, these reductions in enterocyte apoptosis rates were associated with a downregulation of TNF-α protein expression, as well as a lack of increased expression of the proapoptotic proteins bax and Fas. These latter two factors were observed to increase with SBR in ACE^+/+ mice, and these changes may explain the loss of apoptosis in the knockout mice. Importantly, such changes could not account for the significantly lower rates of ileal apoptosis in nonresected mice, which had high TNF-α and Fas levels and lower levels of bcl-2 protein. This suggests that ACE and its downstream angiotensin signaling pathway have a key role in the mediation of intestinal EC apoptosis that cannot be overridden by these other important factors. Additionally, ACE^−/− mice showed significantly higher EC proliferation rates at baseline. This increase in proliferation was associated with a markedly higher gene expression of the growth factor receptors KGFR₁, EGFR, and IL-7R, compared with ACE^+/+ mice (2, 5, 19, 27, 56, 63), and the upregulation of these receptors may well explain the observed increase in EC proliferation. These differences in EC apoptosis and EC proliferation rates were reflected in the intestinal morphology by means of a significantly greater crypt depth and villus height in the intestinal epithelium of ACE^−/− compared with ACE^+/+ mice. Therefore, we consider that ACE is an important modulator of intestinal homeostasis both at baseline and during intestinal adaptation.

Previously, our group demonstrated an upregulation of intestinal mucosal ACE gene expression after SBR in mice (60). Because ACE has been shown to have a significant role in promoting inflammation, proliferation, and apoptosis in a number of organs including heart (45), lung (55), kidney (47), and colon (43), we investigated the role of ACE during the process of intestinal adaptation. We previously reported that inhibition of ACE by intraperitoneal administration of the ACE inhibitor enalaprilat led to a significant reduction in intestinal EC apoptosis (61). Enalaprilat also led to an increase in crypt cell proliferation resulting in an enhanced structural adaptation after SBR in wild-type mice (61). Because ACE inhibitors have been reported to be able to exhibit physiological actions independently, i.e., using pathways other than ACE blockade (1, 37), in the present study we examined whether our previous findings in the intestinal epithelium after ACE inhibition were a drug effect or whether ACE itself has a role in EC apoptosis and proliferation. The use of ACE^−/− mice allowed us to better examine the role of ACE.

Although the expression of ACE in the intestine has been demonstrated (8, 9, 11, 21), its physiological function in the intestinal mucosa is poorly defined. Recent studies have shown that all components of the renin-angiotensin system (RAS) are found in the intestine and may exhibit nonclassic RAS actions (39). In the present study we confirmed the strong expression of ACE in the brush border region and intestinal EC of the small bowel mucosa in wild-type mice. Absence of intestinal ACE in ACE^−/− mice was associated with significant aberrations in intestinal mucosal morphology at baseline, as well as after SBR. The lack of increased apoptosis following SBR in ACE^−/− mice was noted in the region of the crypts for both the jejunum and ileum. Consistent reduction of crypt enterocyte apoptosis rates and increases in EC proliferation more than likely explain the increases in intestinal morphometry after bowel resection (3, 13, 40). These morphometric changes appear to be due to both a significant increase in EC proliferation and reduction in apoptosis in ACE^−/− mice. This increase in proliferation was noted not just after an intestinal resection, which normally is associated with such an increase in proliferation, but at baseline as well. The finding of greater EC diameters in ACE^−/− mice at baseline and after SBR compared with ACE^+/+ mice suggests that accelerated growth is due to both hyperplasia and hypertrophy. Interestingly, the greater crypt depth at baseline in ACE^−/− compared with ACE^+/+ mice appears to be due to the greater EC diameters and not to the greater number of crypt enterocytes. After SBR, ACE^−/− mice increase their villus height and crypt depth beyond these baseline dimensions by additional increases in EC diameter (i.e., cell hypertrophy). Similar to our previous work, jejunal segments in both strains of mice showed that EC hyperplasia dominated clearly over the cell hypertrophy (15, 16). Such changes may be mediated by the upregulation of the various growth factor receptors.

Mechanisms leading to increased EC apoptosis after SBR are incompletely understood. The proapoptotic protein bax and other members of the intrinsic apoptotic pathway play a crucial role (4, 15, 29, 41, 49, 50, 52); although regulatory linkage between the intrinsic and extrinsic apoptotic pathways has been also speculated (52). In a study utilizing Fas-null and TNFR₁-null mice, Knott and coworkers (24) could not find any evidence for the significant involvement of the extrinsic apoptotic pathways in the postresectional enterocyte apoptosis. In addition, we have previously reported that postresectional apoptotic mechanisms may vary along the proximal to distal axis of the gut (15), which increases even more the complexity of the potential underlying apoptotic mechanisms.

Differences in mRNA and protein results were noted in the present study, and thus we rely predominately on the protein expression results in the interpretation of mechanisms responsible for alteration in EC apoptosis. The increases in both bax and Fas protein expression accompanied the increase in EC apoptosis after SBR in ACE^+/+ mice. These findings are consistent with previous studies (4, 15, 50, 52), supporting both apoptotic pathways in the regulation of postresectional EC apoptosis. In contrast, ACE^−/− mice showed a similar increase in expression of bcl-2 but an unchanged expression of bax protein after SBR. This led to a significant reduction in bax-to-bcl-2 ratio after SBR only in ACE^−/− mice, which has been shown to be a valuable indicator for cellular environments that impede apoptosis (4, 15, 29, 49). Of further interest is that post-SBR ACE^−/− mice showed a decline in TNF-α protein expression and no change in FAS. These protein changes suggest another potential mechanism that may account for the lower rates of apoptosis in ACE^−/− mice. Recent studies have linked apoptosis mediated by the intrinsic pathway to local RAS control (28, 31, 38, 51). Blockade of angiotensin II signaling, via either an ACE-inhibitor, or selective blockade of angiotensin II receptors have been shown to effectively reduce apoptosis rates in various tissues (28, 45, 47, 61).

The fact that intestinal blood flow, both mesenteric and mucosal, did not decrease in our knockout mice suggests that observed differences between ACE^+/+ and ACE^−/− mice were not due to changes in intestinal blood flow. ACE^−/− mice are characterized by the inability to concentrate urine (30), which leads to overconsumption of water and may lead to reduced intake of liquid food compared with ACE^+/+ mice. In fact, our ACE^−/− mice took in twofold more water and 30% less liquid food per day. Although these changes could potentially contribute to the greater weight loss in ACE^−/− mice after SBR, we doubt that it would explain lower rates of EC apoptosis, or increased EC proliferation. In fact, most investigators have shown greater rates of apoptosis with a decline in energy intake.

In conclusion, our results demonstrate significantly reduced EC apoptosis rates and increased EC proliferation in the absence of ACE. This was accompanied by the reduction in the bax-to-bcl-2 ratio after SBR in ACE^−/− mice. Further work regarding specific pathways for RAS-mediated regulation of intestinal EC apoptosis will be needed to elucidate its impact in the process of postresectional intestinal adaptation.

GRANTS

This research was supported by National Institutes of Health Grant AI44076-09 and Abbott Laboratories, Hospital Division. The work was also supported in part by the University of Michigan-Comprehensive Cancer Center National Institutes of Health Grant 5 P30 CA-46592 and the University of Michigan-Multipurpose Arthritic Center National Institutes of Health Grant AR-20557.