点击显示 收起
【摘要】
Guanylyl cyclase C (GC-C), the receptor for diarrheagenic enterotoxins and the paracrine ligands guanylin and uroguanylin, regulates intestinal secretion. Beyond volume homeostasis, its importance in modulating cancer cell proliferation and its uniform dysregulation early in colon carcinogenesis, reflecting loss of ligand expression, suggests a role for GC-C in organizing the crypt-villus axis. Here, eliminating GC-C expression in mice increased crypt length along a decreasing rostral-caudal gradient by disrupting component homeostatic processes. Crypt expansion reflected hyperplasia of the proliferating compartment with reciprocal increases in rapidly cycling progenitor cells and reductions in differentiated cells of the secretory lineage, including Paneth and goblet cells, but not enteroendocrine cells. GC-C signaling regulated proliferation by restricting the cell cycle at the G1/S transition. Moreover, crypt expansion in GC-CC/C mice was associated with adaptive increases in cell migration and apoptosis. Reciprocal alterations in proliferation and differentiation resulting in expansion associated with adaptive responses in migration and apoptosis suggest that GC-C coordinates component processes maintaining homeostasis of the crypt progenitor compartment. In the context of uniform loss of GC-C signaling during tumorigenesis, dysregulation of those homeostatic processes may contribute to mechanisms underlying colon cancer.
--------------------------------------------------------------------------------
The intestinal epithelium undergoes homeostatic cycles of proliferation, migration, differentiation, and apoptosis entrained by the regenerative activity of multipotent stem cells.1,2 Although mechanisms coordinating homeostasis are incompletely defined, they maintain the intestinal mucosa integrity despite constant exposure to environmental insults. In that context, specialized functions such as digestion and absorption of nutrients are accomplished by the action of a heterogeneous monolayer of columnar epithelial cells along the crypt-villus (small intestine) and crypt-surface (colon) axes.1 Slowly-cycling stem cells located near crypt bottoms produce rapidly-cycling progenitor cells in mid-crypts, which initiate programs of lineage commitment through complex regulatory signaling,2 forming mature absorptive cells or enterocytes and secretory cells, including mucus-producing goblet cells and peptide-producing enteroendocrine and Paneth cells.1,3 Mature epithelial cells are replaced as they migrate toward the lumen side of the mucosa where they undergo apoptosis and shedding.3
An imbalance in component processes underlying homeostasis may be one fundamental mechanism contributing to the development of colon cancer,4 the second leading cause of cancer-related mortality in developed countries.5 Thus, targeted inactivation of p21,6 which regulates cell cycle progression and differentiation, promotes APC mutation-dependent tumor formation by increasing proliferation and decreasing apoptosis and differentiation in intestine. Deletion of Muc2 in differentiated cells induces colorectal cancer by increasing proliferation and migration and decreasing apoptosis.7 Conversely, crypt hyperplasia associated with disrupted differentiation, without changes in proliferation or apoptosis, promotes intestinal tumorigenesis in mice exhibiting loss of imprinting of the insulin-like growth factor II gene.8
Guanylyl cyclase C (GC-C), the receptor for bacterial diarrheagenic heat-stable enterotoxins (STs) and the endogenous ligands guanylin and uroguanylin, is the primary endogenous source of cyclic guanosine 3',5' monophosphate (cGMP) in intestinal epithelial cells.9,10 Heretofore, GC-C and its downstream effector cGMP were considered principal regulators of intestinal fluid homeostasis. Intriguingly, GC-C has recently emerged as a key regulator of colon cancer cell dynamics. GC-C signaling through cyclic nucleotide gated channel disrupts tumor cell cycle progression and proliferation.11-13 In close agreement, induction of cGMP-dependent mechanisms produces enduring colon cancer cytostasis.13 Further, treatment of ApcMin/+ mice with oral uroguanylin suppresses intestinal tumor formation.14 However, although cGMP coordinates proliferation, differentiation, and apoptosis in many cell systems,15,16 a role for GC-C in regulating component processes underlying crypt-villus homeostasis remains elusive. That GC-C might regulate crypt-villus homeostasis is supported by the near-uniform loss of guanylin and uroguanylin expression early in intestinal tumorigenesis.17-19 Indeed, eliminating guanylin expression in mice induces expansion of the proliferating crypt compartment in colon,20 although this effect is ostensibly not mimicked by elimination of GC-C expression.21
These observations suggest that GC-C and its ligands may function as a paracrine system regulating component processes underlying intestinal mucosa homeostasis. Moreover, GC-C signaling may represent a novel homeostatic mechanism whose dysregulation through loss of ligand expression contributes to colon carcinogenesis. The present studies explored the role of GC-C in coordinating processes organizing the crypt-villus axis, including proliferation, cell cycle, migration, differentiation, and apoptosis.
【关键词】 homeostatic crypt-villus bacterial enterotoxin receptor guanylyl restricts proliferating compartment intestine
Materials and Methods
Cell Culture, Mice, and Reagents
NCM460 cells,22 which express GC-C mRNA (Figure 1A) and functional GC-C protein (Figure 1, BCE) , and their proprietary M3:10 medium were from In Cell (San Antonio, TX). T84 human colon carcinoma cells, used as a positive control for GC-C expression,23 were from American Type Culture Collection (Manassas, VA).
Figure 1. GC-C expression in NCM460 cells and GC-CC/C mice. A: GC-C mRNA is present in normal NCM460 colonocytes and T84 colon carcinoma cells (positive control) as assessed by RT-PCR (1 µg RNA of sample/reaction). B: Dose response of 125I-ST binding to NCM460 membranes (cold competitor, 1 µmol/L ST). kDa, ST binding affinity, Bmax, maximum number of ST binding sites. C: Total binding of 125I-ST to NCM460 membranes in the presence of the vehicle (CTR, control) or the inactive ST analog TJU 1-103 (TJU, 1 µmol/L), ST (1 µmol/L), or guanylin (1 µmol/L) as cold competitors. In NCM460 cells, ST induces cGMP accumulation in a dose-dependent manner (D), an effect mimicked by guanylin (1 µmol/L) but not the vehicle (CTR, control) or the inactive analog TJU 1-103 (TJU, 1 µmol/L) (E; ST, 1 µmol/L). F: PCR on DNA extracts yielded the expected 195-bp (GC-C+/+) and 146-bp (GC-CC/C) amplicons. G: Specific 125I-ST binding (cold competitor, 1 µmol/L ST) to intestinal mucosa membranes isolated from GC-C+/+ (hatched columns) and GC-CC/C (black columns) mice. H: H&E staining of representative GC-C+/+ and GC-CC/C mouse intestinal sections. Du, duodenum; Je/Ile, jejunum and ileum; PC, proximal colon; and DC, distal colon. *P < 0.05, two-tailed t-test. Original magnifications, x100.
GC-CC/C mice, generated by neomycin-resistant gene insertion,24 were genotyped after polymerase chain reaction (PCR) of genomic DNA isolated from tails (GC-C forward primer, 5'-AGGTCATGACGTCACTGCTGGGCC-3'; GC-C reverse primer, 5'-TGTCCAGTCCTTCCTCCACAG-3'; neomycin reverse primer, 5'-GGTGGGCTCTATGGCTTC-3') (Figure 1F) and exhibited loss of GC-C ligand binding in their intestine (Figure 1G) . GC-CC/C mice (I-129 background) were backcrossed with a C57BL/6 strain for seven or more generations, and GC-CC/C and GC-C+/+ littermates (males, 3 months old) were analyzed. Animals were maintained in a controlled environment (20??C, 12-hour dark/light cycle) with free access to food and water. For cell migration, three GC-CC/C and three GC-C+/+ mice per time point (2, 24, and 48 hours) were analyzed. These animals (nine GC-CC/C and nine GC-C+/+ mice) were also used for examining other morphometric or cytodynamic parameters. Two additional cohorts of six mice per genotype were used for calculations of cell cycle times. Finally, two additional cohorts of five mice per genotype were used for immunoblot analyses. Reagents used included native ST and the inactive analog ST(5C17)Ala,9,17Cys(Acm),5,106-14 disulfide (TJU 1-103) (Bachem, San Carlos, CA); thymidine (Amersham, Piscataway, NJ), and 3-isobutyl-1-methylxanthine, guanylin, 8-br-cGMP (Sigma, St. Louis, MO).
Crypt Depth
Mouse intestines were isolated, cleaned, and divided into five segments: duodenum (the rostral, 5-cm-long small intestinal segment from the pyloric sphincter), jejunum (the rostral half of small intestine from the duodenum to the ileal-cecal junction), ileum (the caudal half of small intestine from the duodenum to the ileal-cecal junction), and proximal (the rostral half) and distal (the caudal half) colon (from the cecal-colonic junction to the anal sphincter). Intestinal samples, spanning 1 cm (small intestine) or 0.5 cm (colon) from the center of each segment, were fixed overnight (10% formalin), embedded in paraffin, sectioned (5 µm), and mounted (5 to 20 sections/slide) for hematoxylin and eosin (H&E) staining. Digital images were captured with a light microscope attached to a computer and recorded using Adobe Photoshop. Crypt length was measured (in 10 crypts/intestinal segment/mouse) as the distance from the bottom to the opening of crypt with intact columnar epithelia using the NIH Image J analysis software.
In Vivo Cell Proliferation, Cell Differentiation, and Cell Death
Intestinal sections were deparaffinized and rehydrated with sequential washes of xylene, ethanol, and water, followed by heat-induced epitope retrieval (10 mmol/L citrate buffer, ph 6.0). Sections were incubated with primary antibodies (overnight, 4??C), followed by secondary antibodies and diaminobenzidine substrate (avidin-biotin kit; Vector Laboratories, Burlingame, CA). Stained cells were quantified (in 5 to 15 crypt-villus units/intestinal segment/mouse) using NIH Image J software. Proliferating cells were stained with the proliferating cell nuclear antigen (PCNA) immunohistochemistry (IHC) kit (Zymed, South San Francisco, CA) or the rat anti-mouse Ki-67 antibody (1:50 dilution; DAKO Cytomation, Carpinteria, CA). PCNA- and Ki-67-labeling indices were calculated by normalizing the number of PCNA- or Ki-67-positive cells to the total cell number in respective crypts. Paneth cells, enterocytes, and enteroendocrine cells were stained with rabbit anti-human/mouse lysozyme antibody (ready-to-use, DAKO Cytomation), goat anti-villin antibody , and rabbit anti-chromogranin A antibody (1:200 dilution, Zymed), respectively. Goblet cells were quantified using Alcian blue staining of mucin7 with fast nuclear red counterstaining. Finally, cell death was quantified by terminal deoxynucleotidyl transferase-mediated dNTP-biotin nick-end labeling (TUNEL) using the TACSXL blue label in situ apoptosis detection kit (Trevigen, Gaithersburg, MD).
In Vivo Cell Cycle and Cell Migration
Crypt cell cycles were examined by cumulative BrdU (15 mg/kg) injections (at 4- to 6-hour intervals for 24 hours) into the peritoneal cavity.25 Thirty minutes after each injection, a cohort of mice was sacrificed, and intestinal sections were stained for BrdU (IHC BrdU kit, Zymed Inc.). The relationship between the ratio of BrdU+ cells to PCNA+ cells and time was analyzed with a nonlinear mixed effects model, and cell cycle times (T) calculated from the slope (sl) of the proximal limb of biphasic curves,25 using the equation: sl = 1/T.
For cell migration, three mice per group were sacrificed at 2, 24, and 48 hours after a single intraperitoneal BrdU injection (15 mg/kg), and resulting intestinal sections were subjected to BrdU IHC. Migration rate (µm/hour) reflects the slope of the fitted line of the distance of the farthest labeled enterocyte from the crypt-bottom throughout time. The number (for cell cycle) or the distance (for cell migration) of BrdU-labeled cell(s)/crypt was quantified using NIH Image J software.
Immunoblot Analyses
Epithelia were dissected from normal intestinal mucosa from five individual animals in each group or NCM460 cells (1,000,000/well in a six-well plate) were synchronized (48 hours) by serum starvation and restimulated with serum alone (control) or that containing 1 mmol/L 8br-cGMP for 3 hours. In each case, protein was extracted in Laemmli buffer containing protease and phosphatase inhibitor cocktails (Pierce, Rockford, IL) and stored at C20??C. Homogenized proteins were subjected to immunoblot analyses using antibodies to the following antigens where indicated: cyclin D1 and pRb from Santa Cruz Biotechnology (1:200 dilution) and cleaved caspase 3 and GAPDH from Cell Signaling (Danvers, MA; 1:1000 dilution). ITF (1:10,000) anti-serum was a gift from Dr. Daniel K. Podolsky (Massachusetts General Hospital and Harvard Medical School, Boston, MA). Bovine anti-goat and anti-rabbit secondary antibodies were from Santa Cruz Biotechnology (1:5000 dilution). Staining intensity of specific bands quantified by densitometry was normalized to that for GAPDH. Average relative intensity reflects the mean of five individual animals per genotype or three independent experiments with NCM460 cells.
In Vitro Cell Proliferation and Cell Cycle
For DNA synthesis, NCM460 cells (50,000/well in a 96-well plate) were synchronized (48 hours) by serum starvation and restimulated (24 hours) with 10% fetal bovine serum in the presence of the indicated treatments. Cells were labeled by adding, for the last 3 hours of incubations, thymidine incorporation into DNA was quantified.11,13 Also, growth curves were generated by treating NCM460 cells (50,000/well in a 24-well plate) in media and determining the DNA or protein content per well. After DNA extraction with ethanol in phenol/chloroform, DNA concentrations were measured by spectrophotometry (260 nm) and cell numbers extrapolated from linear standard curves (data not shown). Conversely, protein concentrations were determined using the BCA protein assay kit (Pierce). Cell doubling times (at the period of exponential growth) and cell cycle distributions by flow cytometry (after serum restimulation of synchronized cells) were calculated as described.11
Primary Cultures of Colon Mucosa
Colonic tissues were obtained from patients undergoing surgery under a protocol approved by the institutional review board (control no. 01.0823) as a part of the clinical trial NIH R01 CA75123 (S.A. Waldman, PI). Normal colonic mucosa dissected from the lamina propria were cut into 25-mm2 sheets and placed onto six-well plates in primary culture media.26 Colonic mucosal sheets incorporated thymidine into nuclear DNA for >21 hours (data not shown). Thus, after isolation and recovery (4 hours), treatments were performed 15 minutes before addition of thymidine (2 µCi/well) for 16 hours. Then, 3H-thymidine incorporation into DNA was quantified,11 and results were normalized to the wet weight of mucosa sheets.
GC-C and Cyclic GMP Analyses
GC-C mRNA was quantified by reverse transcriptase (RT)-PCR.27 Iodination of ST and GC-C ligand binding was assessed in membranes from NCM460 cells or mouse intestinal mucosa.23 Specific GC-C binding was calculated using Prism 3.0 software after subtraction of nonspecific binding (with unlabeled ST) from total binding (without unlabeled ST). Finally, cGMP was quantified by radioimmunoassay28 in NCM460 cells incubated (30 minutes) in serum-free media with the phosphodiesterase inhibitor 3-isobutyl-1-methylxanthine (1 mmol/L) before addition of treatments (for 15 minutes).
Statistics
Crypt depth, cell proliferation by PCNA or Ki-67 IHC, and cell migration were analyzed by fitting a linear mixed effects model.29 To estimate cell cycle times, a nonlinear mixed effects model29 was fitted to the crypt-specific proportion ratios of BrdU-labeled cells to PCNA+ cells with the fixed effects part as described.25 These models accounted for animal-to-animal variability and correlation between clustered measures from the same animal. All other comparisons used the Student??s t-test. Determinations in vitro were repeated at least three times. Where appropriate, analyses were done in triplicate and results reflect mean ?? SEM of a representative experiment. In all analyses, P < 0.05 was considered statistically significant.
Results
GC-C Regulates the Size of the Proliferating Crypt Compartment
Although the gross histological architecture of crypt and villus units were not obviously different in GC-C+/+ and GC-CC/C mice (Figure 1H) , closer examination revealed that in the absence of GC-C signaling mice exhibited longer crypts (Figure 2A , Table 1 ; measured with the NIH Image J analysis software) associated with increased numbers of Ki-67+ (Figure 2, BCD ; quantified by Ki-67+ IHC) or PCNA+ (Figure 2, E and F , and Table 1 ; quantified by PCNA+ IHC) cells compared with GC-C+/+ mice. Crypt elongation reflected hyperplasia along a decreasing rostral-caudal gradient (P < 0.001), and the number of proliferating cells was the dominant predictor of hyperplasia in all intestinal segments (P < 0.001). Moreover, the fraction of proliferating cells in crypts (Ki-67- or PCNA-labeling index) increased in GC-CC/C mice compared with GC-C+/+ mice with the greatest effect in proximal colon (Figure 2, D and F) , indicating that crypt hyperplasia principally reflects expansion of the proliferating, and not quiescent, compartment. Indeed, the colonic crypts of Lieberkuhn include the entire cell phenotypic spectrum along the crypt-villus axis, whereas shorter small intestinal crypts (Figure 1H) , apically restricted by the crypt-villus junction (Figure 8) , are predominantly populated by proliferating progenitor cells (Figure 2B) . In agreement with the suggestion that GC-C critically regulates cell proliferative dynamics, induction of GC-C signaling using ST or 8-br-cGMP, a membrane-permeant cGMP analog,30 inhibited proliferation of normal epithelial cells in mucosal sheets isolated from intestines of patients, as assessed by 3H-thymidine incorporation into nuclear DNA (Figure 3A) . Further, ST or the endogenous GC-C agonist guanylin, but not the inactive ST analog TJU 1-103, inhibited proliferation of NCM460 human intestinal crypt cells22 (Figure 3B) , and this effect was mimicked by 8-br-cGMP (Figure 3C) .
Figure 2. GC-CC/C mice exhibit expanded proliferating compartments. A: Crypt depth was increased in GC-CC/C intestinal segments (black columns) compared with their GC-C+/+ counterparts (hatched columns). B: Ki-67 IHC staining of representative GC-C+/+ and GC-CC/C mouse intestinal sections. C: Ki-67+ cells per crypt section were increased in GC-CC/C mice. D: Ki-67 labeling index (Ki-67+ cell number/total cell number/crypt section) in different intestinal sections. E: PCNA+ cells per crypt section were increased in GC-CC/C mice. F: PCNA labeling index (PCNA+ cell number/total cell number/crypt section) in different intestinal sections. Du, duodenum; Je, jejunum; Ile, ileum; PC, proximal colon; and DC, distal colon. *P < 0.05, **P < 0.01, ***P < 0.0001 by linear mixed effects model. n = 9 mice/genotype. Original magnifications, x400.
Table 1. Crypt Depth and PCNA-Labeled Cell Number
Figure 8. The transition from proliferating to differentiated absorptive cells is retained in GC-CC/C mice. IHC staining of representative small (jejunum, Je) and large (proximal colon, PC) intestinal sections from GC-C+/+ and GC-CC/C mice. Mature enterocytes and proliferating crypt cells were visualized in adjacent sections after staining for villin and Ki-67, respectively. Double-headed arrows indicate the cell position of the transition from proliferation to enterocyte differentiation (first villin+/Ki-67C cell from the crypt bottom). Original magnifications, x400.
Figure 3. GC-C inhibits proliferation of normal human colonocytes. A: DNA synthesis in human colonic mucosa sheets treated with the vehicle (CTR, control), ST (1 µmol/L), or 8-br-cGMP (1 mmol/L). B: DNA synthesis in NCM460 cells treated with the vehicle (CTR, control), 1 µmol/L of either the inactive ST analog TJU 1-103 (TJU), ST or guanylin, or 8-br-cGMP (1 mmol/L). Experiments with control, ST, TJU 1-103 or guanylin were performed in the presence of 3-isobutyl-1-methylxanthine (1 mmol/L). C: Dose-dependent inhibition of DNA synthesis by 8-br-cGMP in NCM460 cells. Data in ACC reflect 3H-thymidine incorporation into nuclear DNA expressed as the percentage of parallel control incubations exposed to the appropriate vehicle. *P < 0.05, two-tailed t-test.
GC-C Regulates the Rate of the Cell Cycle in the Proliferating Crypt Compartment
To define further the proliferative crypt kinetics in GC-CC/C mice, cell cycle duration was quantified after serial peritoneal BrdU injections, with incorporation into DNA by cycling cells during S phase transition (Figure 4A) .25 This model presumes that proliferating crypt cells reflect two populations, including stem cells with a uniform cell cycle that is twice the duration of progenitor cells.25,31 The nonlinear mixed effects model,29 used to analyze BrdU labeling curves, was fitted to the crypt-specific proportion ratios of BrdU-labeled cells to PCNA+ cells (Figure 4B) . Only cell cycles of progenitor cells were considered, since the model and time (24 hours) used permit quantification of in vivo cell cycle rates of rapidly proliferating progenitor cells rather than slowly proliferating stem cells.3,31 Elimination of GC-C signaling accelerated the cell cycle in crypt progenitor cells of GC-CC/C, compared with GC-C+/+, mice along a rostro-caudal gradient, with significantly higher effects in proximal small intestine and modest effects in colon (Figure 4, B and C ; Table 2 ). Acceleration of the cell cycle in GC-CC/C mice was associated with increases in critical intestinal cell cycle mediators by immunoblot analysis, including cyclin D1 and p-Rb (Figure 4, D and E) .
Figure 4. GC-CC/C crypt cells exhibit shorter cell cycles. A: Representative BrdU IHC staining of GC-C+/+ mouse duodenal slides, after cumulative intraperitoneal injections with BrdU. Proliferating cells are progressively recruited in the BrdU-labeled compartment as a function of their cell cycle time. B: BrdU-labeling curves in GC-CC/C (dashed lines) and GC-C+/+ (solid lines) mice were generated by calculating the percentage of BrdU-labeled cells per crypt at each time point relative to the average genotype number of PCNA+ cells per crypt in the corresponding intestinal segment. C: Cell cycle rates in GC-CC/C (black columns) and GC-C+/+ (hatched columns) intestinal segments. D: Representative immunoblots of cell cycle mediators cyclin D1 and phosphorylated Rb (p-Rb) in jejunal mucosa from GC-C+/+ and GC-CC/C mice. E: Mean densitometry of cyclin D1 and p-Rb immunoblots, normalized to GAPDH, from jejunal mucosa of five individual GC-C+/+ or GC-CC/C mice, respectively. Intestinal segments are labeled as in Figure 2 . ACC: *P < 0.05, by linear mixed effects model. E: *P < 0.05, **P < 0.01, two-tailed t-test. Original magnifications, x400.
Table 2. Cell Cycle Times
Conversely, induction of GC-C signaling prolonged the cell cycle of NCM460 human normal colonocytes in vitro (Figure 5) . Indeed, colonocytes treated with ST (Figure 5A) or 8-br-cGMP (Figure 5B) grew more slowly than control cells receiving PBS, as quantified by the DNA (top panels) or protein (bottom panels) content of cell cultures. Also, intestinal cells exposed to ST (Figure 5C) or 8-br-cGMP (Figure 5D) exhibited a longer doubling time (assessed at the period of exponential growth)11 associated with a delay in the G1/S phase transition (Figure 5E) . Accordingly, in NCM460 cells cGMP reduced the protein levels of cyclin D1 and p-Rb (Figure 5, F and G) , key promoters of the G1/S cell cycle transition that are regulated by in vivo GC-C signaling (Figure 4, D and E) .
Figure 5. GC-C induces a G1/S transition delay in normal human colonocytes. Growth curves (A, B) and cell doubling times (C, D) of NCM460 cells treated with 1 µmol/L ST (A, C) or 1 mmol/L 8-br-cGMP (B, D). E: Time course of cell cycle distribution by flow cytometry of NCM460 cells treated with PBS (solid lines) or 1 mmol/L 8-br-cGMP (dashed lines). F: Representative immunoblots for cyclin D1 and phosphorylated Rb (p-Rb) in NCM460 cells. G: Mean densitometry of cyclin D1 and p-Rb immunoblots, normalized to GAPDH, from three independent experiments. Con, control (cells treated with PBS); 8br, 8-br-cGMP. *P < 0.05, two-tailed t-test.
GC-C-Induced Crypt Hyperplasia Is Coupled with Adaptive Increases in Cell Migration and Apoptosis
Beyond proliferative kinetics, expansion of the crypt progenitor cell compartment in GC-CC/C mice could reflect reduced egress of cells from that compartment, mediated by changes in migration or apoptosis.2 Proliferating epithelial cells migrated more rapidly along the crypt-villus axis in GC-CC/C compared with GC-C+/+ mice (Figure 6, A and B , and Table 3 ; quantified by single intraperitoneal BrdU injection). Linear mixed effects modeling, which considered animal-average scores of PCNA+ cells and crypt depth, suggests that accelerated enterocyte migration was an adaptive response to, rather than the cause of, crypt hyperplasia (P < 0.001). Indeed, as observed for proliferative kinetics, GC-C-dependent regulation of cell migration exhibited a rostro-caudal gradient (Table 3) . Similarly, elimination of GC-C increased apoptosis in intestinal crypts as assessed by TUNEL (Figure 6, C and D) and immunoblot analysis for cleaved caspase 3 (Figure 6, E and F) , also reflecting an adaptive response to crypt hyperplasia.
Figure 6. Adaptive increases in migration and apoptosis along the GC-CC/C crypt-villus axis. A: BrdU IHC staining of representative intestinal sections from mice receiving a single BrdU injection to quantify cell migration. B: Quantification of intestinal epithelial cell migration. C and D: Apoptosis was quantified by TUNEL staining (C) after scoring 100 to 800 crypts/intestinal segment in five GC-C+/+ (hatched columns) and five GC-CC/C (black columns) mice (D). E: Representative immunoblots of cleaved caspase 3 in jejunal mucosa from GC-C+/+ and GC-CC/C mice. F: Mean densitometry of cleaved caspase 3 immunoblots, normalized to GAPDH, from jejunal mucosa of five individual GC-C+/+ or GC-CC/C mice. Intestinal segments are labeled as in Figure 2 . *P < 0.05; **P < 0.01, two-tailed t-test. Original magnifications: x200 (A, jejunum); x 400 (A, proximal colon; C).
Table 3. Cell Migration Rates
GC-C Regulates Lineage-Specific Cell Differentiation
Perturbations in cell maturation dynamics also could contribute to the expansion of crypt compartments in GC-CC/C mice.2 Indeed, elimination of GC-C signaling specifically altered cell-lineage differentiation, and GC-CC/C mice exhibited fewer goblet cells compared with wild-type mice, as assessed by counting intestinal cells stained for mucin with Alcian blue (Figure 7, A and B) and by immunoblot analysis for intestine trefoil factor (Figure 7, C and D) . In addition, GC-CC/C mice exhibited fewer Paneth cells than GC-C+/+ mice, as quantified by counting mucosal cells stained with anti-lysozyme antibody (Figure 7, ECG) . However, there were no differences in enteroendocrine cells (Figure 7, HCL ; quantified by IHC for anti-chromogranin A) or absorptive enterocytes (Figure 8 ; quantified by IHC for villin). Further, the ordered transition from the proliferating (Ki-67+) to the differentiated (villin+) cell compartment was maintained in GC-CC/C mice (Figure 8) . Moreover, there were no differences in villus length in GC-CC/C, compared with GC-C+/+, mice (data not shown). Taken together, these data suggest that GC-C signaling critically affects the homeostasis of intestinal crypt compartment by coordinating processes underlying reciprocal regulation of proliferation and differentiation associated with adaptive responses in migration and apoptosis.
Figure 7. Lineage-specific cell commitment is altered in GC-CC/C mice. ACD: Goblet cells were detected by Alcian blue staining (A) and enumerated in 3 to 15 villi per mouse (B). Expression of intestinal trefoil factor (ITF), a goblet cell marker, was detected by immunoblot analysis (C) and quantified by average densitometry, after normalization to GAPDH, in colons from five individual GC-C+/+ and GC-CC/C mice, respectively (D). ECG: Paneth cells (E, F) were detected by lysozyme staining and enumerated in 10 crypts per mouse (G). HCL: Enteroendocrine cells (H, I, crypts; J, villi) were quantified in 100 crypts per mouse (K) or 20 villi per mouse (L) by chromogranin A staining. Data in B, D, G, K, and L reflect analysis of jejunal (Je) sections from nine GC-C+/+ (hatched columns) or nine GC-CC/C (black columns) mice, respectively. *P < 0.05, two-tailed t-test. Original magnifications: x200 (A, E, H); x400 (F, I, J).
Discussion
The importance of defining the molecular mechanisms organizing the crypt-villus axis is underscored by the established relationship between dysregulation of key processes underlying that organization and intestinal tumorigenesis.4,6,8 Thus, Wnt/??-catenin/Tcf-4 pathway is a principal determinant of the crypt cell phenotype,32 and mice with disrupted Tcf-4 signaling exhibit depletion of intestinal stem cell compartments.33 Ca2+ inhibits proliferation and promotes differentiation in intestinal mucosa,34 and Ca2+ supplementation abrogated intestinal hyperproliferation and tumor formation in ApcMin/+ mice.35 Further, CDX-2 is a transcription factor of the murine homeobox gene family that regulates intestinal epithelial cell differentiation36 and its reduced expression was associated with colon tumorigenesis in ApcMin/+ mice.37
In that context, GC-C signaling has emerged as a key modulator of the colon cancer cell phenotype. GC-C inhibits colon cancer cell proliferation11-13 and opposes colon carcinogenesis in ApcMin/+ mice.14,38 Also, although GC-C expression is highly conserved,23 expression of guanylin and uroguanylin is uniformly lost early in neoplastic transformation, suggesting that disruption of GC-C signaling may contribute to colon carcinogenesis.14,19 GC-C and cGMP inhibit tumor cell proliferation by inducing Ca2+ influx through cyclic nucleotide gated channels,12 suggesting that the anti-tumor effects of dietary Ca2+ 34,35 may be mediated, in part, by GC-C signaling. Moreover, activation of GC-C may oppose proproliferative Wnt/??-catenin/Tcf-4 signaling by promoting PKG1??-dependent degradation of ??-catenin, inhibiting its nuclear translocation.39
Beyond colon cancer cell proliferation, the role of GC-C in mechanisms underlying normal intestinal epithelial cell dynamics remains undefined. Elimination of guanylin expression in mice produced colonic crypt hyperplasia, reflecting expansion of proliferative compartments associated with increased migration.20 Although reminiscent of observations in the present study, these effects were not replicated in GC-CC/C mice,21 suggesting that other receptors mediate the effects of guanylin on colon crypt dynamics. Indeed, receptors other than GC-C may mediate effects of GC-C ligands on fluid homeostasis.40 However, previous studies examining crypt dynamics in GC-CC/C mice did not present data and did not specify the anatomical segment along the rostro-caudal axis analyzed.21 As demonstrated herein, experimental measurements confined to the distal colon in GC-CC/C mice could have produced artifactually negative results.
Here, elimination of GC-C induced selective crypt hyperplasia along a decreasing rostro-caudal gradient, in the absence of changes in villus morphology. Crypt hyperplasia reflected, in part, increases in the number of proliferating cells and acceleration of their cell cycle. These changes were associated with adaptive increases in migration and apoptosis along the crypt-villus axis, which limited expansion of the proliferating compartment. Thus, GC-C signaling represents a previously unrecognized mechanism restricting the size and proliferative rate of the intestinal progenitor compartment. The rostral-caudal gradient of this effect likely reflects a recapitulation of the established gradient along that axis of proliferation, migration, and apoptosis41,42 and the quantitative and qualitative differences in downstream effector mechanisms mediating GC-C signaling.20,43-46 In the context of the present studies, crypt elongation associated with increases in proliferation and migration previously observed in guanylinC/C mice reflect elimination of GC-C signaling, rather than loss of signaling by another unidentified receptor.20
The present study reveals a previously unappreciated role for GC-C in mechanisms regulating proliferative indices of normal intestinal epithelial cells. Indeed, GC-C signaling inhibited DNA synthesis rates in normal colonocytes in human mucosa sheets and in human intestinal epithelial cells. Moreover, GC-C signaling prolonged the intestinal epithelial cell cycle by inducing a G1/S delay. Crypt hyperplasia associated with increased numbers of rapidly cycling progenitor cells in GC-CC/C mice reflects loss of physiological GC-C regulation of the cell cycle. In contrast, GC-C-induced cytostasis in colon cancer cells reflects a generalized slowing, without delay in any specific phase, of the cell cycle.11-13 In that context, GC-C-induced G1/S delay may represent a novel tumor-suppressor mechanism in normal cells whose interruption, through loss of ligand expression,17-19 contributes to colon tumorigenesis.
In addition, this study reveals a role for GC-C signaling in lineage-specific maturation of intestinal epithelial cells. In GC-CC/C mice, crypt hyperplasia reflected increases in proliferative dynamics and reciprocal restriction of differentiation along the secretory lineage, specifically of Paneth and goblet cells, without effects on enteroendocrine cells or absorptive enterocytes. The selective impairment of cell lineage commitment suggests that, rather than a passive consequence of dysregulated cell proliferation, these effects may reflect a previously unrecognized ability of GC-C to regulate discreet molecular mechanisms associated with intestinal cell programming.
Beyond fluid balance, GC-C has recently emerged as a critical signaling molecule suppressing colorectal tumorigenesis by regulating distinct, mutually reinforcing molecular mechanisms underlying crypt-villus homeostasis, including restriction of proliferation and maintenance of genomic integrity.38 Observations presented here further support the suggestion that GC-C is one of the integrated paracrine mechanisms regulating intestinal mucosa homeostasis. Thus, GC-C signaling restricts the size of crypt progenitor compartments, impeding cell cycle progression and proliferation, promoting lineage-specific maturation, limiting cell migration, and inhibiting apoptosis. The pathophysiological significance of GC-C as a paracrine mechanism restricting the size of the proliferating compartment is underscored by the centrality of this compartment in mechanisms underlying colon carcinogenesis. Indeed, overexpression of molecules like Wnt, ??-catenin, and Tcf that promote the progenitor cell compartment in the crypt is associated with colonic tumorigenesis.32,33 Conversely, disruption of genes including APC, SMAD, and IHH, which restrict the proliferating crypt compartment and promote differentiation in the villus compartment, also is associated with colonic tumorigenesis.47-51 The prevailing model suggests that tumor initiation and progression in colon reflects accumulation of genetic mutations in proliferating crypt progenitor cells.52 Importantly, expression of guanylin and uroguanylin is invariably lost early during neoplastic transformation.17-19 In the context of the present studies, early loss of GC-C ligands and the associated physiological mechanisms restricting the cell cycle, number of proliferating cells and size of the progenitor compartment, and promoting cell differentiation will induce crypt hyperplasia, amplifying accumulation of stochastic genetic mutations underlying tumor initiation.4,6,8 Moreover, dysregulated GC-C signaling reflecting loss of ligands eliminates one important cytostatic mechanism inhibiting proliferation in transformed cells, potentiating tumor promotion.11-13,30 Of significance, loss of guanylin and uroguanylin expression in human tumors is associated with near-uniform overexpression of GC-C,23,27 suggesting paracrine hormone replacement therapy as a novel chemopreventive strategy to maintain crypt-villus homeostasis and suppress tumor initiation and promotion in intestine.38
【参考文献】
Nordgaard I, Mortensen PB: Digestive processes in the human colon. Nutrition 1995, 11:37-45
Potten CS, Loeffler M: Stem cells: attributes, cycles, spirals, pitfalls and uncertainties. Lessons for and from the crypt. Development 1990, 110:1001-1020
Bach SP, Renehan AG, Potten CS: Stem cells: the intestinal stem cell as a paradigm. Carcinogenesis 2000, 21:469-476
Montgomery RK, Mulberg AE, Grand RJ: Development of the human gastrointestinal tract: twenty years of progress. Gastroenterology 1999, 116:702-731
Jemal A, Tiwari RC, Murray T, Ghafoor A, Samuels A, Ward E, Feuer EJ, Thun MJ: Cancer statistics, 2004. CA Cancer J Clin 2004, 54:8-29
Yang WC, Mathew J, Velcich A, Edelmann W, Kucherlapati R, Lipkin M, Yang K, Augenlicht LH: Targeted inactivation of the p21(WAF1/cip1) gene enhances Apc-initiated tumor formation and the tumor-promoting activity of a Western-style high-risk diet by altering cell maturation in the intestinal mucosal. Cancer Res 2001, 61:565-569
Velcich A, Yang W, Heyer J, Fragale A, Nicholas C, Viani S, Kucherlapati R, Lipkin M, Yang K, Augenlicht L: Colorectal cancer in mice genetically deficient in the mucin Muc2. Science 2002, 295:1726-1729
Sakatani T, Kaneda A, Iacobuzio-Donahue CA, Carter MG, de Boom Witzel S, Okano H, Ko MS, Ohlsson R, Longo DL, Feinberg AP: Loss of imprinting of Igf2 alters intestinal maturation and tumorigenesis in mice. Science 2005, 307:1976-1978
Lucas KA, Pitari GM, Kazerounian S, Ruiz-Stewart I, Park J, Schulz S, Chepenik KP, Waldman SA: Guanylyl cyclases and signaling by cyclic GMP. Pharmacol Rev 2000, 52:375-414
Vaandrager AB, Bot AG, De Vente J, De Jonge HR: Atriopeptins and Escherichia coli enterotoxin STa have different sites of action in mammalian intestine. Gastroenterology 1992, 102:1161-1169
Pitari GM, Di Guglielmo MD, Park J, Schulz S, Waldman SA: Guanylyl cyclase C agonists regulate progression through the cell cycle of human colon carcinoma cells. Proc Natl Acad Sci USA 2001, 98:7846-7851
Pitari GM, Zingman LV, Hodgson DM, Alekseev AE, Kazerounian S, Bienengraeber M, Hajnoczky G, Terzic A, Waldman SA: Bacterial enterotoxins are associated with resistance to colon cancer. Proc Natl Acad Sci USA 2003, 100:2695-2699
Pitari GM, Baksh RI, Harris DM, Li P, Kazerounian S, Waldman SA: Interruption of homologous desensitization in cyclic guanosine 3',5'-monophosphate signaling restores colon cancer cytostasis by bacterial enterotoxins. Cancer Res 2005, 65:11129-11135
Shailubhai K, Yu HH, Karunanandaa K, Wang JY, Eber SL, Wang Y, Joo NS, Kim HD, Miedema BW, Abbas SZ, Boddupalli SS, Currie MG, Forte LR: Uroguanylin treatment suppresses polyp formation in the Apc(Min/+) mouse and induces apoptosis in human colon adenocarcinoma cells via cyclic GMP. Cancer Res 2000, 60:5151-5157
Pilz RB, Casteel DE: Regulation of gene expression by cyclic GMP. Circ Res 2003, 93:1034-1046
Hewitson TD, Martic M, Darby IA, Kelynack KJ, Bisucci T, Tait MG, Becker GJ: Intracellular cyclic nucleotide analogues inhibit in vitro mitogenesis and activation of fibroblasts derived from obstructed rat kidneys. Nephron Exp Nephrol 2004, 96:59-66
Zhang L, Zhou W, Velculescu VE, Kern SE, Hruban RH, Hamilton SR, Vogelstein B, Kinzler KW: Gene expression profiles in normal and cancer cells. Science 1997, 276:1268-1272
Steinbrecher KA, Tuohy TM, Heppner Goss K, Scott MC, Witte DP, Groden J, Cohen MB: Expression of guanylin is downregulated in mouse and human intestinal adenomas. Biochem Biophys Res Commun 2000, 273:225-230
Notterman DA, Alon U, Sierk AJ, Levine AJ: Transcriptional gene expression profiles of colorectal adenoma, adenocarcinoma, and normal tissue examined by oligonucleotide arrays. Cancer Res 2001, 61:3124-3130
Steinbrecher KA, Wowk SA, Rudolph JA, Witte DP, Cohen MB: Targeted inactivation of the mouse guanylin gene results in altered dynamics of colonic epithelial proliferation. Am J Pathol 2002, 161:2169-2178
Mann EA, Steinbrecher KA, Stroup C, Witte DP, Cohen MB, Giannella RA: Lack of guanylyl cyclase C, the receptor for Escherichia coli heat-stable enterotoxin, results in reduced polyp formation and increased apoptosis in the multiple intestinal neoplasia (Min) mouse model. Int J Cancer 2005, 116:500-505
Moyer MP, Manzano LA, Merriman RL, Stauffer JS, Tanzer LR: NCM460, a normal human colon mucosal epithelial cell line. In Vitro Cell Dev Biol Anim 1996, 32:315-317
Carrithers SL, Barber MT, Biswas S, Parkinson SJ, Park PK, Goldstein SD, Waldman SA: Guanylyl cyclase C is a selective marker for metastatic colorectal tumors in human extraintestinal tissues. Proc Natl Acad Sci USA 1996, 93:14827-14832
Schulz S, Lopez MJ, Kuhn M, Garbers DL: Disruption of the guanylyl cyclase-C gene leads to a paradoxical phenotype of viable but heat-stable enterotoxin-resistant mice. J Clin Invest 1997, 100:1590-1595
Nowakowski RS, Lewin SB, Miller MW: Bromodeoxyuridine immunohistochemical determination of the lengths of the cell cycle and the DNA-synthetic phase for an anatomically defined population. J Neurocytol 1989, 18:311-318
Birkner S, Weber S, Dohle A, Schmahl G, Follmann W: Growth and characterisation of primary bovine colon epithelial cells in vitro. Altern Lab Anim 2004, 32:555-571
Schulz S, Hyslop T, Haaf J, Bonaccorso C, Nielsen K, Witek ME, Birbe R, Palazzo J, Weinberg D, Waldman SA: A validated quantitative assay to detect occult micrometastases by reverse transcriptase-polymerase chain reaction of guanylyl cyclase C in patients with colorectal cancer. Clin Cancer Res 2006, 12:4545-4552
Kazerounian S, Pitari GM, Shah FJ, Frick GS, Madesh M, Ruiz-Stewart I, Schulz S, Hajnoczky G, Waldman SA: Proliferative signaling by store-operated calcium channels opposes colon cancer cell cytostasis induced by bacterial enterotoxins. J Pharmacol Exp Ther 2005, 314:1013-1022
Vonesh EF, Chinchilli VM: Linear and Nonlinear Models for the Analysis of Repeated Measures 1997 Marcel Dekker New York
Pitari GM, Li T, Baksh RI, Waldman SA: Exisulind and guanylyl cyclase C induce distinct antineoplastic signaling mechanisms in human colon cancer cells. Mol Cancer Ther 2006, 5:1190-1196
Potten CS: Cell cycles in cell hierarchies. Int J Radiat Biol Relat Stud Phys Chem Med 1986, 49:257-278
Gregorieff A, Clevers H: Wnt signaling in the intestinal epithelium: from endoderm to cancer. Genes Dev 2005, 19:877-890
Korinek V, Barker N, Moerer P, van Donselaar E, Huls G, Peters PJ, Clevers H: Depletion of epithelial stem-cell compartments in the small intestine of mice lacking Tcf-4. Nat Genet 1998, 19:379-383
Whitfield JF, Bird RP, Chakravarthy BR, Isaacs RJ, Morley P: Calcium-cell cycle regulator, differentiator, killer, chemopreventor, and maybe, tumor promoter. J Cell Biochem Suppl 1995, 22:S74-S91
Newmark HL, Yang K, Lipkin M, Kopelovich L, Liu Y, Fan K, Shinozaki H: A Western-style diet induces benign and malignant neoplasms in the colon of normal C57Bl/6 mice. Carcinogenesis 2001, 22:1871-1875
Freund JN, Domon-Dell C, Kedinger M, Duluc I: The Cdx-1 and Cdx-2 homeobox genes in the intestine. Biochem Cell Biol 1998, 76:957-969
Aoki K, Tamai Y, Horiike S, Oshima M, Taketo MM: Colonic polyposis caused by mTOR-mediated chromosomal instability in Apc+/ Delta716 Cdx2+/C compound mutant mice. Nat Genet 2003, 35:323-330
Li P, Schulz S, Bombonati A, Palazzo JP, Hyslop TM, Barab AA, Siracusa LD, Pitari GM, Waldman SA: Guanylyl cyclase C suppresses intestinal tumorigenesis by restricting proliferation and maintaining genomic integrity. Gastroenterology 2007, 133:599-607
Thompson WJ, Piazza GA, Li H, Liu L, Fetter J, Zhu B, Sperl G, Ahnen D, Pamukcu R: Exisulind induction of apoptosis involves guanosine 3',5'-cyclic monophosphate phosphodiesterase inhibition, protein kinase G activation, and attenuated beta-catenin. Cancer Res 2000, 60:3338-3342
Sindi A, Velic A, Basoglu C, Hirsch JR, Edemir B, Kuhn M, Schlatter E: Uroguanylin and guanylin regulate transport of mouse cortical collecting duct independent of guanylate cyclase C. Kidney Int 2005, 68:1008-1017
Marshman E, Booth C, Potten CS: The intestinal epithelial stem cell. Bioessays 2002, 24:91-98
Potten CS: Keratinocyte stem cells, label-retaining cells and possible genome protection mechanisms. J Invest Dermatol Symp Proc 2004, 9:183-195
Vaandrager AB, Bot AG, De Jonge HR: Guanosine 3',5'-cyclic monophosphate-dependent protein kinase II mediates heat-stable enterotoxin-provoked chloride secretion in rat intestine. Gastroenterology 1997, 112:437-443
Forte LR, Thorne PK, Eber SL, Krause WJ, Freeman RH, Francis SH, Corbin JD: Stimulation of intestinal ClC transport by heat-stable enterotoxin: activation of cAMP-dependent protein kinase by cGMP. Am J Physiol 1992, 263:C607-C615
Chao AC, de Sauvage FJ, Dong YJ, Wagner JA, Goeddel DV, Gardner P: Activation of intestinal CFTR ClC channel by heat-stable enterotoxin and guanylin via cAMP-dependent protein kinase. EMBO J 1994, 13:1065-1072
Vaandrager AB, Bot AG, Ruth P, Pfeifer A, Hofmann F, De Jonge HR: Differential role of cyclic GMP-dependent protein kinase II in ion transport in murine small intestine and colon. Gastroenterology 2000, 118:108-114
van den Brink GR, Hardwick JC, Tytgat GN, Brink MA, Ten Kate FJ, Van Deventer SJ, Peppelenbosch MP: Sonic hedgehog regulates gastric gland morphogenesis in man and mouse. Gastroenterology 2001, 121:317-328
van den Brink GR, Bleuming SA, Hardwick JC, Schepman BL, Offerhaus GJ, Keller JJ, Nielsen C, Gaffield W, van Deventer SJ, Roberts DJ, Peppelenbosch MP: Indian Hedgehog is an antagonist of Wnt signaling in colonic epithelial cell differentiation. Nat Genet 2004, 36:277-282
Tang Y, Katuri V, Srinivasan R, Fogt F, Redman R, Anand G, Said A, Fishbein T, Zasloff M, Reddy EP, Mishra B, Mishra L: Transforming growth factor-beta suppresses nonmetastatic colon cancer through Samd4 and adaptor protein Elf at an early stage of tumorigenesis. Cancer Res 2005, 65:4228-4237
Takaku K, Oshima M, Miyoshi H, Matsui M, Seldin MF, Taketo MM: Intestinal tumorigenesis in compound mutant mice of both Dpc4 (Smad4) and Apc genes. Cell 1998, 92:645-656
Fodde R, Kuipers J, Rosenberg C, Smits R, Kielman M, Gaspar C, van Es JH, Breukel C, Wiegant J, Giles RH, Clevers H: Mutations in the APC tumour suppressor gene cause chromosomal instability. Nat Cell Biol 2001, 3:433-438
Radtke F, Clevers H: Self-renewal and cancer of the gut: two sides of a coin. Science 2005, 307:1904-1909
作者单位：From the Divisions of Clinical Pharmacology,* and Biostatistics, Department of Pharmacology and Experimental Therapeutics, Thomas Jefferson University, Philadelphia, Pennsylvania