The heterogeneous morphology of networked collagen in distal colon and rectum of mice quantified via nonlinear microscopy

Abstract

Visceral pain from the distal colon and rectum (colorectum) is a major complaint of patients with irritable bowel syndrome. Mechanotransduction of colorectal distension/stretch appears to play a critical role in visceral nociception, and further understanding requires improved knowledge of the micromechanical environments at different sub-layers of the colorectum. In this study, we conducted nonlinear imaging via second harmonic generation to quantify the thickness of each distinct through-thickness layer of the colorectum, as well as the principal orientations, corresponding dispersions in orientations, and the distributions of diameters of collagen fibers within each of these layers. From C57BL/6 mice of both sexes (8-16 weeks of age, 25-35 g), we dissected the distal 30 mm of the large bowel including the colorectum, divided these into three even segments, and harvested specimens (~8 × 8 mm²) from each segment. We stretched the specimens either by colorectal distension to 20 mmHg (reference) or 80 mmHg (deformed) or by biaxial stretch to 10 mN (reference) or 80 mN (deformed), and fixed them with 4% paraformaldehyde. We then conducted SHG imaging through the wall thickness and analyzed post-hoc using custom-built software to quantify the orientations of collagen fibers in all distinct layers. We also quantified the thickness of each layer of the colorectum, and the corresponding distributions of collagen density and diameters of fibers. We found collagen concentrated in the submucosal layer. The average diameter of collagen fibers was greatest in the submucosal layer, followed by the serosal and muscular layers. Collagen fibers aligned with muscle fibers in the two muscular layers, whereas their orientation varied greatly with location in the serosal layer. In colonic segments, thick collagen fibers in the submucosa presented two major orientations aligned approximately ±30° to the axial direction, and form a patterned network. Our results indicate the submucosa is likely the principal passive load-bearing structure of the colorectum. In addition, afferent endings in those collagen-rich regions present likely candidates of colorectal nociceptors to encode noxious distension/stretch.

Keywords: Biaxial extension test; Colorectum; Confocal microscopy; Mechanotransduction; Submucosa; Visceral pain.

Figure 1.

Overview of our analyses of the images obtained via SHG. We (a) applied a Hamming filter, (b) performed fast-Fourier transform, and (c) calculated the power spectral density. Using (c) we quantified, via wedge-filtering, the distribution of the fibers in the original image as a histogram. We repeated this process for all the images within a z-stack and (e) generated a 2-D representation of all of the histograms where red in the color scale corresponds to a maximum in the relative magnitude. We then (f) summed these histograms from each layer to obtain an overall histogram representing the fiber distribution within that layer, facilitating our fitting with a Von Mises distribution quantifying the principal orientation and dispersion in orientation (black line).

Figure 2.

(a) A representative through-thickness image and corresponding intensity plot reveals the strongest signal originated from the submucosa. (b) We confirmed the finding by reviewing representative images before contrast enhancements. Only the submucosa (1) presents visible fibers, while the remaining specimens (2-5) appear black. (c) With enhanced contrast, we see clear fibers in the serosa (2) and blurry, but distinct, fibers in the circumferential (3) and axial (4) muscle layers and the mucosa (5). (d) With an intraluminal pressure of 20 mmHg (left) the reference specimen has wavy fibers while a pressure of 80 mmHg (right) causes the collagen fibers to stretch.

Figure 3.

Layer thicknesses of specimens from (a) colonic, (b) intermediate, and (c) rectal locations show a similar pattern. The mucosa is thickest, followed in descending order by submucosa, muscle, and serosa. The * indicates a significant difference between reference and deformed configurations.

Figure 4.

Composite (overall) histograms of the orientation of fibers in the colonic segments for (a-b) the serosa, (c-d) the muscle layer, and (e-f) submucosa for both the (a,c,e) reference and (b,d,f) deformed configurations. The black line indicates the fitted Von Mises distribution.

Figure 5.

Composite (overall) histograms of the orientation of fibers in the intermediate segments for (a-b) the serosa, (c-d) the muscle layer, and (e-f) submucosa for both the (a,c,e) reference and (b,d,f) deformed configurations. The black line indicates the fitted Von Mises distribution.

Figure 6.

Composite (overall) histograms of the orientation of fibers in the rectal segments for (a-b) the serosa, (c-d) the muscle layer, and (e-f) submucosa for both the (a,c,e) reference and (b,d,f) deformed configurations. The black line indicates the fitted Von Mises distribution.

Figure 7.

Fiber diameters of specimens from the colonic, intermediate, and rectal locations for the serosa, axial and circumferential muscle layers, and the submucosa. The * indicates a significance difference between the layers.