In vivo imaging of hemodynamic redistribution and arteriogenesis across microvascular network

Abstract

Objective: Arteriogenesis is an important mechanism that contributes to restoration of oxygen supply in chronically ischemic tissues, but remains incompletely understood due to technical limitations. This study presents a novel approach for comprehensive assessment of the remodeling pattern in a complex microvascular network containing multiple collateral microvessels.

Methods: We have developed a hardware-software integrated platform for quantitative, longitudinal, and label-free imaging of network-wide hemodynamic changes and arteriogenesis at the single-vessel level. By ligating feeding arteries in the mouse ear, we induced network-wide hemodynamic redistribution and localized arteriogenesis. The utility of this technology was demonstrated by studying the influence of obesity on microvascular arteriogenesis.

Results: Simultaneously monitoring the remodeling of competing collateral arterioles revealed a new, inverse relationship between initial vascular resistance and extent of arteriogenesis. Obese mice exhibited similar remodeling responses to lean mice through the first week, including diameter increase and flow upregulation in collateral arterioles. However, these gains were subsequently lost in obese mice.

Conclusions: Capable of label-free, comprehensive, and dynamic quantification of structural and functional changes in the microvascular network in vivo, this platform opens up new opportunities to study the mechanisms of microvascular arteriogenesis, its implications in diseases, and approaches to pharmacologically rectify microvascular dysfunction.

Keywords: arterial ligation; arteriogenesis; hemodynamic redistribution; microvascular remodeling; obesity.

Figure 1.

Multi-parametric PAM in the hardware-software platform. A, System schematic. NDF, neutraldensity filter; PBS, polarizing beam splitter; BS, beam sampler; PD, photodiode; SMF, single-mode fiber. B, Blow-up of the scanning head boxed in (A). DL, doublet; CL, correction lens; UT, ring-shaped ultrasonic transducer. C, Sequentially acquired A-line signals for multi-parametric quantification. Multi-parametric PAM of (D) the mouse ear, (E) the mouse brain, and (F) the mouse hindlimb (scale bars=500 μm). The left, middle and right columns show the maps of total concentration of hemoglobin (CHb), oxygen saturation of hemoglobin (sO2), and blood flow velocity (speed plus direction), respectively. The red and blue arrow pairs in (D–F) indicate the flow directions.

Figure 2.

Longitudinal monitoring and single-vessel analysis of hemodynamic redistribution and collateral remodeling across a microvascular network. (A–C) Multi-parametric PAM of CHb, sO2 and blood flow speed in the mouse ear before, right after, and up to 21 days following the arterial ligation (scale bar = 1mm). The red cross in (C) indicates the ligated arteriolar tree, and the ligation site lies outside PAM images. The insets in (A, C) are the blowups of the white boxed regions at the lower left corners of the two panels. (D, E) Single-vessel analysis of collateral remodeling in diameter and volumetric blood flow over the yellow boxed region in (C). The extent of remodeling was quantified in absolute value (D: μm, E: nL/s), as well as the relative changes (%) between two adjacent time points in brackets. Increased and decreased values are labeled in green and red, respectively. Purple is used to label vessel segments that are missing in one of the two time points. In (E), solid and dashed lines indicate segments without and with reversed flow direction, respectively. Arrows indicate the flow direction. Mouse strain: C57BL/6BrdCrHsd-Tyrc.

Figure 3.

Multi-parametric analysis of the competition in collateral remodeling. (A) Longitudinal monitoring of the ear microvascular network before, right after, and up to 21 days following the arterial ligation (scale bar=1mm). The three collateral arterioles are labeled in red, green and blue, respectively. Red cross indicates the ligated arteriolar tree, and the ligation site lies outside PAM images. (B–G) Multifaceted collateral remodeling quantified as relative changes (%) in vessel diameter, blood flow speed, sO2, blood oxygen supply, wall shear stress, and vascular resistance, respectively. Mouse strain: SKH1-Hrhr.

Figure 4.

Relationship between initial vascular resistance and the extent of collateral remodeling. (A–D) Inverse correlation between initial resistance and changes in vessel diameter, volumetric blood flow, tortuosity and wall shear stress, respectively. The analyses are based on 11 collateral arterioles from 6 mice.

Figure 5.

Influence of obesity on the ligation-induced hemodynamic redistribution. (A, B) Multi-parametric PAM of CHb, sO2 and blood flow speed in the control and obese mouse ears before, right after, and up to 21 days following the arterial ligation, respectively. The red crosses in (A) and (B) indicate the ligated arteriolar trees, and the ligation site lies outside PAM images. (C, D) Blowups over the white boxed region in (A) and (B). White arrows: remodeling collateral vessels. (E) Tissue-level hemodynamic responses to the ligation extracted by taking the differences between the measurements acquired at each of the post-ligation points and that acquired before ligation. Black dash area: the hypoxia region (scale bar=500μm).

Figure 6.

Influence of obesity on the ligation-induced remodeling in microvascular structure and function. (A–D) Comparison of microvascular remodeling in the control and obese mice, including diameter, volumetric blood flow, vascular tortuosity, and venous sO2. Colored asterisks indicate the statistical significance over the pre-ligation values, while black asterisks indicate that between the two animal groups at the same time point. The analyses are based on 19 collateral arterioles from 6 control mice and 5 obese mice, and data were presented as mean±SD. (*: p<0.05, **: p<0.01, and ***: p<0.001)

Figure 7.

Influence of obesity on the relationship between initial vascular resistance and the extent of collateral remodeling. Inverse correlation between initial resistance and changes in (A) vessel diameter, (B) volumetric blood flow and (C) tortuosity, respectively, in the control and obese mice. The analyses are based on a total of 19 collateral arterioles from 6 control mice and 5 obese mice. Statistical significance between the linear fittings of the two groups is indicated below each panel. (ns: not significant, **: p<0.01).