Microstructural Characterization of Resistance Artery Remodelling in Diabetes Mellitus

Abstract

Introduction: Microvascular remodelling is a symptom of cardiovascular disease. Despite the mechanical environment being recognized as a major contributor to the remodelling process, it is currently only understood in a rudimentary way.

Objective: A morphological and mechanical evaluation of the resistance vasculature in health and diabetes mellitus.

Methods: The cells and extracellular matrix of human subcutaneous resistance arteries from abdominal fat biopsies were imaged using two-photon fluorescence and second harmonic generation at varying transmural pressure. The results informed a two-layer mechanical model.

Results: Diabetic resistance arteries reduced in wall area as pressure was increased. This was attributed to the presence of thick, straight collagen fibre bundles that braced the outer wall. The abnormal mechanical environment caused the internal elastic lamina and endothelial and vascular smooth muscle cell arrangements to twist.

Conclusions: Our results suggest diabetic microvascular remodelling is likely to be stress-driven, comprising at least 2 stages: (1) Laying down of adventitial bracing fibres that limit outward distension, and (2) Deposition of additional collagen in the media, likely due to the significantly altered mechanical environment. This work represents a step towards elucidating the local stress environment of cells, which is crucial to build accurate models of mechanotransduction in disease.

Keywords: Collagen; Diabetes mellitus; Elastin; Multiphoton microscopy; Remodelling.

Fig. 1

Schematic of the experimental apparatus, with details of the myograph base and arms omitted for simplicity. The call out shows the approximate imaging volumes probed during the high resolution (×60) and low resolution (×20) protocols. TPF, two-photon fluorescence; SHG, second harmonic generation.

Fig. 2

Comparison of healthy and diabetic morphologies in section at 30 mm Hg transmural pressure. False colours represent elastic fibres and intracellular proteins (green), and fibrous collagen (blue). a The healthy artery exhibits a distinct IEL, media and adventitia, and the wall thickness is relatively uniform. b The diabetic artery exhibits an IEL, but the adventitia and media overlap and are of uneven thickness. Bars 50 μm. T2DM, type 2 diabetes mellitus; IEL, internal elastic layer.

Fig. 3

Comparison of adventitial collagen morphology in health and diabetes mellitus, at 30 mm Hg transmural pressure. a Healthy adventitia, comprising tortuous collagen fibre bundles of varying diameter and at varying stages of recruitment. b Diabetic adventitia, comprising straight, bracing collagen fibre bundles surrounding an interior layer of tightly wound tortuous collagen fibre bundles. c The number of bracing fibre bundles present within the imaged volume of healthy arteries and arteries from individuals with T2DM. There were significantly more in arteries from individuals with T2DM (p < 0.01, Mann-Whitney U test). Bars 50 μm. T2DM, type 2 diabetes mellitus

Fig. 4

Comparison of IEL morphology in health and diabetes mellitus. a The healthy IEL at 3-mm Hg transmural pressure comprises longitudinal elastic fibres, braced periodically by thinner fibres. b At 30-mm Hg transmural pressure the IEL distends considerably and deforms into a honeycomb structure. c The diabetic IEL at 3-mm Hg transmural pressure also comprises longitudinal fibres, but they appear kinked and uneven. d At 30-mm Hg transmural pressure, the IEL has distended, but not to the same degree as the healthy artery, and looks morphologically similar to the low-pressure case, except for in discrete positions where some fibres have moved apart. e Change in longitudinal elastic fibre spacing. The change in spacing in healthy arteries is significantly greater than in arteries from individuals with T2DM (*p < 0.05), Mann Whitney U test. Bars 50 μm. IEL, internal elastic layer; T2DM, type 2 diabetes mellitus.

Fig. 5

Two-photon fluorescence optical sections through healthy vessel walls and vessel walls from individuals with T2DM (optical plane depicted in inset), with DAPI-stained cell nuclei overlaid in false colour. a In health, endothelial cell nuclei (red) are arranged longitudinally, while VSMC nuclei (blue) are arranged circumferentially. Fibroblast nuclei (yellow) exhibit no preferential shape. b In the diabetic artery, the endothelial cell nuclei (red) appear more irregular in shape and orientation, while the VSMC nuclei (blue) retain a characteristic slender morphology but are arranged more randomly. c Distributions of endothelial cell and VSMC nuclei orientations, measured as an angle relative to the longitudinal vessel axis. Bars 50 μm. T2DM, type 2 diabetes mellitus; EC, endothelial cell; VSMC, vascular smooth muscle cell.

Fig. 6

Morphological and mechanical results. a Volumetric wall strain (analogous to change in cross-sectional area) in healthy and arteries from individuals with T2DM, following pressurization from 3 to 30 mm Hg. b Ratio of layer thickness to lumen radius at 30-mm Hg transmural pressure. c Elastic modulus of each layer, and of the entire wall assuming homogeneity, determined by elastic modelling fitted to experimentally determined deformations (*p < 0.05, **p < 0.01). T2DM, type 2 diabetes mellitus.

Fig. 7

Distribution of fibrous protein orientations in health and arteries from individuals with T2DM, at varying transmural pressure. T2DM, type 2 diabetes mellitus; Adv, adventitial.

Fig. 8

Illustrations of vessel cross-sections, showing undeformed geometry (left) and deformed geometry (right). Arrows indicate predicted direction of local stress associated with transmural pressure. a, b In the healthy resistance artery the adventitia allows outward distension whilst gradually stiffening due to collagen recruitment. The significantly more compliant media distends with the adventitia, thins as the adventitia stiffens, and resists a small portion of the circumferential stress. c, d Vessels exhibiting the first stage of remodelling (constraining fibre bundles only) are less able to distend their outer edge, and so the underlying normal adventitia does not recruit. Circumferential stress is therefore limited to the pathological fibre bundles in the periphery, and the rest of the vessel wall is squeezed up against the stiff outer edge by the transmural pressure. e, f Vessels exhibiting the second stage of remodelling (constraining fibre bundles and media fibrosis) are assumed to behave more like homogeneous pipes. The presence of collagen in the inner layers shifts the distribution of circumferential stress away from the periphery and into a more uniform state.