Regulation of SRF/CArG-dependent gene transcription during chronic partial obstruction of murine small intestine

Abstract

Intestinal obstructions lead to a variety of motility disorders. Small intestine smooth muscles undergo dramatic phenotypic changes in response to obstruction, but the underlying molecular mechanisms are unknown. Using RT-PCR, ChIP, Re-ChIP, and Western blots, we examined the effect of small bowel mechanical obstruction on smooth muscle gene expression. Obstruction caused a transient hyperplasia, followed by a prolonged hypertrophic response of small intestine smooth muscle cells. Smooth muscle myosin heavy chain (MHC), alpha-actin, and gamma-actin expression decreased initially, and then increased as hypertrophy developed. Myocardin expression decreased initially and then increased, while kruppel-like factors (KLF)4 and KLF5 expression increased initially, and then decreased. Serum response factor (SRF) expression decreased initially, and then recovered to sham-operated levels as hypertrophy developed. SRF binding to smooth muscle MHC and alpha-actin promoters decreased initially, but then increased above sham-operated levels as hypertrophy developed. Elk-1 binding to smooth muscle myosin heavy chain and alpha-actin promoters increased initially, and then decreased to sham-operated levels as hypertrophy developed. c-fos expression increased initially, which was associated with increased SRF/Elk-1 binding to the c-fos promoter. The Elk-1 phosphorylation inhibitor U-0126 inhibited the increase in c-fos expression. These findings indicate a dynamic response of small intestine smooth muscles to bowel obstruction involving switching between differentiated, proliferative, and hypertrophic phenotypes. These results suggest that changes in the expression and interactions between SRF, myocardin, Elk-1, and c-fos play key roles in the phenotypic switching of small intestine smooth muscles in response to mechanical obstruction.

Figure 1

Morphological changes of smooth muscles following partial intestinal obstruction. (A) Representative haematoxylin and eosin-stained sections of sham-operated (left panel), 3 days (middle panel), and 14 days post-obstruction small intestine (right panel) (scale bar: 100 μm) (n = 5). Sections were obtained from sham-operated animals 7 days after sham surgery. No differences were noted in the muscle layers from sham-operated animals 1, 3, 7 and 14 days after sham-surgery (data not shown). CM, circular muscle layer; LM, longitudinal muscle layer; (B) Changes in circular muscle layer thickness and smooth muscle cell number following partial intestinal obstruction (data obtained from four different points from five sham-operated and five small intestine partial-obstruction animals). The number of smooth muscle cells was evaluated by counting the number of nuclei in a 100 μm cross-section of the circular muscle layer (mean ± SE, *P < 0.05 compared with sham).

Figure 2

Expression of smooth muscle contractile proteins following intestinal obstruction. Real-time RT-PCR analysis of (A) smooth muscle MHC, (B) α-actin and (C) γ-actin expression (n = 3 animals per time point, mean ± SE, *P < 0.05 vs corresponding sham-operated values). C, non-operated controls; (D) Representative Western blots of smooth muscle MHC (1 μg of cytosolic protein/lane), α-actin and γ-actin (10 μg of cytosolic protein/lane). p.o., partial obstruction.

Figure 3

SRF expression following intestinal obstruction. (A) SRF mRNA levels were analysed by real-time RT-PCR (n = 3 animals per time point, mean ± SE, *P < 0.05 compared with corresponding sham-operated time point). (B) Representative SRF Western blots. SRF is primarily localized to the nucleus (arrowhead). 50 μg of cytosolic (c) or nuclear (n) protein samples were loaded into each lane. (C) Whole-mount anti-SRF immuno-fluorescence of non-operated (left panel), 1 day (middle panel), and 14-day (right panel) obstructed small intestine smooth muscles (scale bar: 100 μm).

Figure 4

SRF binding to smooth muscle MHC and α-actin promoters decreases during the cell proliferation phase and increases during the cell hypertrophy phase. ChIP was carried out with anti-SRF antibodies, and the promoter regions of smooth muscle (A) MHC and (B) α-actin were amplified and quantitated by real-time PCR (n = 3, mean ± SE, *P < 0.05 compared to corresponding sham-operated time points).

Figure 5

Elk-1 binding to SRF-bound MHC and α-actin promoters increases during the cell proliferation phase. ChIP was carried out with anti-SRF antibodies, Re-ChIP was carried out with anti-Elk-1 antibodies, and the promoter regions of smooth muscle (A) MHC and (B) α-actin were amplified and quantitated by real-time PCR (n = 3, mean ± SE, *P < 0.05 compared to corresponding sham-operated time point). (C) Real-time RT-PCR and (D) Western blot analysis (25 μg protein/lane) show no significant change in Elk-1 expression following intestinal obstruction. (E) Real-time RT-PCR shows that U-0126 reverses the down-regulation of smooth muscle MHC and α-actin expression at 1 day post-obstruction (n = 3, mean ± SE, *P < 0.05 compared with obstruction/vehicle).

Figure 6

Myocardin, KLF4 and KLF5 expression following intestinal obstruction. Real-time RT-PCR analysis of (A) myocardin, (B) KLF4 and (C) KLF5 expression (n = 3, mean ± SE, *P < 0.05 compared with corresponding sham-operated time points).

Figure 7

Increased SRF/Elk-1 binding to the c-fos promoter mediates an increase in c-fos expression during the cell proliferation phase. (A) Real-time RT-PCR analysis of c-fos expression (n = 3, mean ± SE, *P < 0.05 compared with corresponding sham-operated time points). (B) ChIP was carried out with anti-SRF antibodies, and the promoter region of c-fos was amplified and quantitated by real-time PCR (n = 3, mean ± SE, *P < 0.05 compared with corresponding sham-operated time points). (C) ChIP was carried out with anti-SRF antibodies, Re-ChIP carried out with anti-Elk-1 antibodies, and the promoter region of c-fos amplified and quantitated by real-time PCR (n = 3, mean ± SE, *P < 0.05 compared with corresponding sham-operated time points). (D) Real-time RT-PCR shows that U-0126 attenuates the up-regulation of c-fos at 6 h post-obstruction (n = 3, mean ± SE, * P < 0.05 compared with obstruction/vehicle).