Differential Biomechanical Responses of Elastic and Muscular Arteries to Angiotensin II-induced Hypertension

Abstract

Elastic and muscular arteries are distinguished by their distinct microstructures, biomechanical properties, and smooth muscle cell contractile functions. They also exhibit differential remodeling in aging and hypertension. Although regional differences in biomechanical properties have been compared, few studies have quantified biaxial differences in response to hypertension. Here, we contrast passive and active changes in large elastic and medium- and small-sized muscular arteries in adult mice in response to chronic infusion of angiotensin over 14 days. We found a significant increase in wall thickness, both medial and adventitial, in the descending thoracic aorta that associated with trends of an increased collagen:elastin ratio. There was adventitial thickening in the small-sized mesenteric artery, but also significant changes in elastic lamellar structure and contractility. An increased contractile response to phenylephrine coupled with a reduced vasodilatory response to acetylcholine in the mesenteric artery suggested an increased contractile state in response to hypertension. Overall reductions in the calculated gradients in pulse wave velocity and elastin energy storage capability from elastic-to-muscular arteries suggested a possible transfer of excessive pulsatile energy into the small-sized muscular arteries resulting in significant functional consequences in response to hypertension.

Introduction

Elastic and muscular arteries are distinguished by their distinct microstructures and unique biomechanical functions. Large elastic arteries consist of a thick elastin-rich medial layer and a relatively thin collagen-rich adventitial layer. The medial elastin is organized into multiple nearly concentric laminae that endow the wall with considerable resilience, namely, an ability to store energy elastically during systole that can be used to augment blood flow during diastole (Wagenseil and Mecham, 2009). Monolayers of smooth muscle cells (SMCs) exist within each of the intra-lamellar spaces and exhibit mainly a matrix phenotype, meaning that their primary role is to mechano-sense and mechano-regulate medial composition (Kelleher et al., 2004; Murtada et al., 2020b), which includes modest amounts of collagen (mostly type III) and glycosaminoglycans (GAGs). In contrast, muscular arteries consist of an SMC–rich medial layer and a collagen-rich adventitial layer. The medial layer in the muscular artery is typically delimited by prominent internal (IEL) and external (EEL) elastic laminae and is occupied primarily by contiguous layers of SMCs of the contractile phenotype (Yoshida and Owens, 2005) that endow the wall with an ability to vasoconstrict and vasodilate, thereby controlling regional blood flow to end organs (Lacolley et al., 2017). Muscular arteries can be categorized into medium- and small-sized before transitioning into arterioles and which feed the capillaries. Similar to small muscular arteries, arterioles consist of an SMC-rich medial layer and a collagen-rich adventitial layer but lack a defined EEL. The transition across vessel types and functions are continuous rather than distinct, including from small-sized muscular arteries to arterioles, both of which regulate vascular resistance and blood pressure. Regional differences in SMC phenotype are further characterized by different functional regulatory pathways through calcium-dependent myosin light-chain kinase activity and calcium-sensitizing myosin light-chain phosphatase activity (Murtada and Humphrey, 2018). Whether thin or thick, the collagen-rich adventitial layer appears to serve largely as a protective sheath to prevent acute over-distensions that could damage the underlying smooth muscle or elastin (Bellini et al., 2014). Arteries thus reflect well the ubiquitous structure-function relationship characteristic of connective tissues (Humphrey, 2002).

Elastic and muscular arteries exhibit differential remodeling in aging and hypertension. M.E Safar and colleagues showed that in vivo distensibility is greater in carotid (elastic) than femoral (muscular) arteries in young patients, but distensibility decreases dramatically in carotid arteries with aging (from ^~20 to 60 years of age), while femoral arteries change little (Benetos et al., 1993). Similar differential remodeling was observed when comparing carotid and radial (muscular) arteries even when adjusted for differences in blood pressure from young to older normotensive individuals (Bortolotto et al., 1999). Importantly, hypertension tends to accelerate age-induced changes in arterial properties (Benetos et al., 1993) and leads to marked decreases in the distensibility of elastic (aorta and carotid) but not muscular (brachial, femoral, and radial) arteries (Laurent and Boutouyrie, 2015). Notwithstanding the strong association of increased structural stiffening of central arteries with all-cause mortality (Laurent et al., 2006), in vivo measures of distensibility do not distinguish biaxial changes in material and geometric properties or contributions of passive versus active behaviors.

Although arteries have been biomechanically characterized in various mouse models of aging and hypertension, there have been few direct comparisons of differential remodeling of elastin and collagen or changes in SMC vasoactive response between elastic and medium- and small-sized muscular arteries. In this study, we contrast, for the first time, active and passive biaxial mechanical properties of elastic (aorta), medium-sized (superior mesenteric), and small-sized (first-order branch of mesenteric) muscular arteries to characterize how the differential regional functions along the vascular tree are affected in a common mouse model of hypertension. We find significant changes in regional microstructure, material properties, active responses and, most importantly, vascular function in response to hypertension.

Methods

Vessels.

Animal procedures were approved by the Institutional Animal Care and Use Committee of Yale University. Proximal descending thoracic aortas (DTA), superior mesenteric arteries (SMA), and first-order branch of mesenteric arteries (MA) were excised following euthanasia by intraperitoneal injection of Beuthanasia-D (150 mg/kg) and exsanguination. Biomechanical data (n=4–8 per region per group) were collected ex vivo and analyzed for n=14 adult (20 weeks of age) male C57BL/6J mice distributed over two groups: non-infused controls (Cntl) and angiotensin II (AngII)–infused at 1000 ng/kg/min for 14 days using an osmotic mini-pump (Alzet model 2004, DURECT Corporation).

Biomechanical Testing.

Vessel segments were mounted in a custom computer-controlled biaxial testing device (Ferruzzi et al., 2013) and submerged in 37°C Krebs-Ringer solution while oxygenated with 95% O₂ and 5% CO₂ to maintain pH at 7.4. Active biaxial mechanical testing and data analysis followed previous methods (Murtada et al., 2016a). Vessels were first preconditioned via two 15-minute contractions at the in vivo axial length stimulated by 100 mM KCl followed by a 10-minute washout with Krebs-Ringer solution. Next, they were subjected to a series of isobaric (DTA, SMA: 90 mmHg, MA: 50 mmHg) – axially isometric (in vivo axial stretch) contractions with KCl, AngII, the selective α1A-adrenergic receptor (AR) agonist A61603 (N-[5-(4,5-dihydro-1H-imidazol-2yl)-2-hydroxy-5,6,7,8-tetrahydro naphthalen-1-yl] methanesulfonamide hydrobromide), or the non-selective α1-AR agonist phenylephrine (PE). To assess endothelial function, a separate group of vessels was dilated with 10 μM acetylcholine (Ach), an endothelial cell–dependent stimulant of nitric oxide (NO) synthesis, after pre-constriction with 1 μM PE. Vessels were monitored in relaxed and contracted states using optical coherence tomography having an axial (depth) resolution <7 microns and lateral resolution of 8 microns (Callisto Model, Thorlabs).

Following active testing, vessels were washed three times with and placed in calcium-free Krebs-Ringer solution. All passive biaxial mechanical testing and data analysis followed previous methods (Ferruzzi et al., 2013), with cyclic pressurization prescribed by region (DTA, SMA: 10–140 mmHg; MA: 10–90 mmHg). Briefly, vessels were mechanically preconditioned, then exposed to seven pressure-diameter (P-d) and axial force-length (f-l) protocols. The data were fit with a validated four-fiber family constitutive model (Ferruzzi et al., 2013) via nonlinear regression of data consisting of the final cycle of unloading during all seven protocols, which yields that part of the stored energy that is available to augment blood flow. This constitutive relation includes eight model parameters within a stored energy function W, namely

$$W\left(\mathbf{C}\mathbf{,}{\mathbf{M}}^i\right)=\frac{c}{2}\left(I_C-3\right)+{\Sigma }_{i=1}^4\frac{c_1^i}{4c_2^i}\left\{\exp \left[c_2^i{\left(I{V}_c^i-1\right)}^2\right]-1\right\},$$ (1)

where c (kPa), $c_1^i$ (kPa), and $c_2^i$ are parameters (i = 1,2,3,4 denote axial, circumferential, and two symmetric-diagonal fiber family directions), $I_C=tr\left(\mathbf{C}\right)$ and $I{V}_{\mathbf{C}}^i={\mathbf{M}}^i\cdot \mathbf{C}{\mathbf{M}}^i$ are coordinate invariant measures of the finite deformation, with $\mathbf{C}={\mathbf{F}}^T\mathbf{F}$ the right Cauchy-Green tensor computed from the deformation gradient tensor $F=diag\left[{\lambda }_r,{\lambda }_{\theta },{\lambda }_z\right],$ with detF = 1 assuming incompressibility. ${\mathbf{M}}^i=\left(0,\sin {\alpha }_0^i,\cos {\alpha }_0^i\right)$ captures each fiber-family direction, with ${\alpha }_0^i$ defined relative to the axial direction in the intact traction-free reference configuration: ${\alpha }_0^1=0,{\alpha }_0^2=\raisebox{1ex}{$\pi $}\!\left/ \!\raisebox{-1ex}{$2$}\right.$, and ${\alpha }_0^{3,4}=\pm {\alpha }_0.$ The Cauchy stress tensor t is

$$\mathbf{t}=-p\mathbf{I}+2\mathbf{F}\frac{\partial W}{\partial \mathbf{C}}{\mathbf{F}}^T,$$ (2)

where p is a Lagrange multiplier that enforces incompressibility. The values of the mean in-plane Cauchy wall stresses were calculated experimentally, while neglecting the radial component, with

$$\begin{array}{cc}{\sigma }_{\theta }\left({\lambda }_{\theta },{\lambda }_z\right)=\frac{pa}{h},& {\sigma }_z\end{array}\left({\lambda }_{\theta },{\lambda }_z\right)=\frac{f_T+\pi a^{2}p}{\pi h\left(2a+h\right)},$$ (3)

where P is the transmural pressure measured with standard transducers, f_T the axial load measured by a load cell, a the deformed inner radius, and h the deformed thickness. Both a and h were calculated from incompressibility using the unloaded volume and online measurements of outer diameter and axial length, which allowed calculation of mean biaxial stretches (λ_θ,λ_z) using standard formulae (i.e., ratios of current to unloaded mean radius and length, respectively). Components of the material stiffness ${\mathcal{C}}_{ijkl}$, linearized about a configuration defined by a set pressure and in vivo axial stretch, were computed as

$${{\mathcal{C}}_{ijkl}=2{\delta }_{ik}{F}_{lA}^0{F}_{jB}^0\frac{\partial W}{\partial C_{AB}}+2{\delta }_{jk}{F}_{iA}^0{F}_{lB}^0\frac{\partial W}{\partial C_{AB}}+4F_{iA}^0{F}_{jB}^0{F}_{kP}^0{F}_{lQ}^0\frac{{\partial }^{2}W}{\partial C_{AB}\partial C_{PQ}}|}_{C^0},$$ (4)

where δ_ij are components of I, F⁰ is the deformation gradient tensor between the reference configuration and a finitely deformed in vivo configuration, and C⁰ is the corresponding right Cauchy-Green tensor.

Histology.

Following biomechanical testing, specimens were unloaded and fixed overnight in 10% neutral buffered formalin, then stored in 70% ethanol at 4°C for histological examination. Fixed samples were dehydrated, embedded in paraffin, sectioned serially, and stained with Verhoeff–Van Gieson (VVG), Movat pentachrome (MOV), picro-sirius red (PSR) and smooth muscle α-actin (SMαA) anti-body counterstained with DAPI (4′,6-diamidino-2-phenylindole). Detailed analyses were performed on three biomechanically representative vessels per group. Custom MATLAB scripts extracted layer-specific cross-sectional areas and counted positively stained pixels and area fractions.

Statistics.

Analysis of variance (ANOVA) was used to compare results across regions and treatments, with Bonferroni post-hoc testing and p < 0.05 considered significant. All data are presented as mean±standard error of the mean (SEM).

Results

Histological cross-sections (Fig 1A) confirmed the highest medial elastin content, contained within 5–6 layers, in the Cntl DTA, lower in the medium-sized SMA with 3–4 layers, and lowest in the small-sized MA with distinct IEL and EEL. Following 14 days of AngII-infusion, there were several breaks in the middle elastic laminae in the AngII SMA, with only the EEL and IEL intact, and a nearly degenerated EEL but intact IEL in the AngII MA; in contrast, all layers of elastin laminae were intact in the AngII DTA. Quantification of medial elastin normalized by vessel-specific medial circumference (Fig 1B) supported a reduction in elastin content in the AngII MA (p<0.071). The media thickened in the DTA (Fig S1G), driven largely by GAG accumulation (Fig S1H), whereas such thickening was less in the SMA and absent in the MA. A trend of increased adventitial collagen (normalized with vessel-specific adventitial circumference) emerged in the DTA, but there was no significant change in collagen content in the SMA or MA (Fig 1C). This resulted in an increased collagen:elastin ratio in the DTA and trend of increase (p<0.097) in the MA following AngII-infusion. Significant increases in both medial and adventitial thickness in the AngII DTA resulted in an unaffected adventitial thickness fraction, whereas there was a trend of increase (p<0.086) in the AngII MA (Fig 1E).

Figure 1.

A: Histological images of a representative descending thoracic aorta (DTA), superior mesenteric artery (SMA), and first-order branch of mesenteric artery (MA) stained with Verhoeff–Van Gieson (VVG) and picro-sirius red (PSR) for adult C57BL/6 mice without (Cntl) or with (AngII) 14-day infusion of angiotensin II (1000 ng/kg/min). All histological images are at same magnification; black scale bar represents 20 μm and the large white scale bar represents 2 mm. B: Quantification of medial elastin normalized by vessel-specific medial circumference (Circ) revealed highest elastin content in the DTA and gradually lower content in the SMA and MA as expected. The concentration remained constant in the DTA and SMA but decreased (p<0.071) in the MA following angiotensin infusion. In addition, elastin appeared intact in the DTA, but there were signs of elastin breaks in the SMA, and degeneration of external elastic lamina in the MA (see Fig 1A, white arrows). C: Quantification of adventitial collagen normalized by vessel-specific adventitial circumference (Circ) revealed trends of increase in the DTA while remaining unchanged in the SMA and MA. D: The collagen:elastin ratio increased in the DTA, remained unchanged in the SMA, and exhibited a trend of increase in the MA (p<0.097). E: Despite changes in medial and adventitial thickness, which was most significant in the DTA (Fig S1H), there was only a trend of increase in adventitial thickness fraction in the AngII MA (p<0.086). * p<0.05.

Biomechanical metrics were compared at representative normotensive values for Cntl arteries (DTA, SMA:100 mmHg; MA:60 mmHg) and hypertensive values for AngII arteries (DTA, SMA:140 mmHg, MA:84 mmHg), values identified previously (Bersi et al., 2017; Spronck et al., 2020), and at vessel-specific values of in vivo (i.e., energetically preferred) axial stretch. Overall pressure-inner radius (Fig 2A) and axial force-stretch (Fig 2E) relationships differed by region. Despite the reduced in vivo axial stretch (Fig S2D), the increased pressure during AngII-infusion increased circumferential stress (Fig 2B) and stiffness (Fig 2C) in the DTA and SMA, but not in the MA. Axial stress and stiffness were unaffected in the DTA and MA, but increased in the SMA (Fig 2F,G). Interestingly, while wall stresses decreased with vessel size in Cntl and AngII arteries, material stiffness exhibited a different trend. Both circumferential and axial stiffness were similar in Cntl DTA and SMA, but lower in the MA. Following AngII-infusion, circumferential stiffness increased in the DTA and SMA, while axial stiffness only increased in the SMA. This suggested a differential anisotropic adaptive response to hypertension in the DTA and SMA. In contrast, the MA efficiently maintained both stress and stiffness despite hypertensive conditions. Regional pulse wave velocity, a standard metric for ‘arterial stiffness’, was calculated at Cntl and AngII pressures using the Bramwell-Hill equation $PWV=\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$\sqrt{\rho \left(dA/dP\right)/A}$}\right.,$ where ρ (1059 kg/m³) is fluid mass density, A the luminal cross-sectional area, and P the pressure. This calculation revealed a gradual increase with smaller vessel size as expected. Additionally, a significant increase was found in the DTA alone following AngII-infusion (Fig 2D). Stored elastic energy was nearly maintained in all regions when comparing Cntl vessels at normotensive and AngII vessels at hypertensive pressures (Fig S3D). However, when correlating energy storage at common fixed pressures (DTA, SMA: 100 mmHg, MA: 60 mmHg) to medial elastin content, a decreased relationship was found for AngII vessels, suggesting an impaired function of energy storage function by elastin after AngII-infusion (Fig 2H).

Figure 2.

A, E: Passive pressure-inner radius and axial force-stretch relationships of the DTA (o), SMA (□), and MA (Δ), with particular attention to behaviors at 100 (DTA, SMA) or 60 mmHg (MA) for Cntl (open) and 140 (DTA, SMA) or 84 mmHg (MA) for AngII (closed) at near vessel-specific values of in vivo axial stretch (represented by star symbol). B, C: Circumferential stress was highest in the DTA and gradually smaller in the SMA and MA in Cntl mice, while stiffness was similar in the DTA and SMA and lower in the MA. Angiotensin infusion significantly increased stress and stiffness in the DTA but less so in the SMA and not in the MA, suggesting full remodeling of the AngII MA to preserve circumferential properties. F, G: Trends differed in the axial direction, with stiffness increasing only in the SMA in AngII arteries. D: The calculated regional pulse wave velocity (PWV B-H), using the Bramwell–Hill equation, was gradually increased in the SMA and MA compared to the DTA in Cntl mice, as expected in smaller vessels. Following AngII-infusion, the calculated PWV B-H increased in only the DTA. H: A correlation was found between medial elastin and stored energy at fixed pressures (DTA, SMA: 100 mmHg, MA: 60 mmHg) when comparing all vessels, which decreased in the AngII arteries, suggesting reduced mechanical functionality in the SMA and especially DTA. *p<0.05, **p<0.01, ***p<0.001.

The MA contracted more than twice as much to 100mM KCl (Cntl MA: ~52% diameter reduction) than did the SMA and DTA, which had similar magnitude of responses (Cntl DTA and SMA: ~21%). No differences were found following AngII-infusion (Fig 3A,B). Contractile responses to 1 μM PE were similar in Cntl DTA and MA (~25%) while a significant increase was observed in AngII MA (~38%), reaching levels similar to responses in the SMA (Cntl ~36%, AngII ~39%) (Fig 3C,D). Histological images of MA stained with a fluorescent smooth muscle α-actin (SMαA) antibody suggested an increase in SMαA in AngII MA (Fig 3E,F) which could explain, in part, the increase in contractile response to PE (Fig 3G). Dose-contractile response tests revealed a similar and unchanged sensitivity to PE after AngII-infusion in all three regions (Fig S4A). Both the SMA and MA were dramatically more sensitive (~1000-fold) to the selective α1A-AR agonist A61603 than was the DTA, with no significant difference in sensitivity after AngII-infusion (Fig S4B). However, the sensitivity to acute response to angiotensin, which was higher in Cntl MA and SMA than in the DTA, was significantly reduced (EC50) following AngII-infusion (Fig S4C,D). The vasodilatory response to 10 μM acetylcholine when pre-constricted with 1 μM PE was reduced significantly in the AngII MA (Cntl ~103%, AngII ~61%), but remained unchanged in the SMA or DTA (Fig 3H). The combined effect of increased contractility and reduced relaxation resulted in a significantly smaller active inner radius in the AngII MA, less so in the SMA, and unchanged in the DTA, suggesting a significantly increased capability to reduce luminal radius in the MA following AngII-infusion (Cntl ^~ 72.4 μm, AngII ^~43.6 μm).

Figure 3.

A: Normalized outer diameter in response to 100 mM KCl of isobaric-axially isometric mounted vessels near vessel-specific physiological loading conditions revealed significantly higher contractile response in Cntl (--, open) MA (Δ) compared to the SMA (□) and DTA (○), which was not affected after angiotensin infusion (−, closed). B: Optical coherence tomography cross-sectional images of MA revealed a patent lumen despite contractions up to an ~52 % reduction in outer diameter. C: Normalized outer diameter in response to 1 μM phenylephrine (PE) revealed an increase in contractile response in MA (Δ) after angiotensin infusion (−, closed) compared to Cntl (--, open). D: An increase in steady-state contractile response was only observed in the AngII MA (~58%) while remained unaffected in the SMA or DTA. E-G: Histological images of MA stained with a fluorescent antibody and counterstained with 4′,6-diamidino-2-phenylindole (DAPI) revealed a trend of increased smooth muscle α-actin (SMαA) following AngII-infusion; scale bar represents 20 μm. H: In addition, the endothelial-driven relaxation in response to 10 μM acetylcholine (Ach), when pre-constricted with 1 μM PE, was higher in Cntl SMA and MA than in Cntl DTA. However, a loss in relaxation (~41%) was only observed in the AngII MA. I: The combined increase in contractility (PE) and loss in relaxation (Ach) significantly reduced the active loaded inner radius in the MA, less so in the SMA, while remaining unaffected in the DTA. *p<0.05, **p<0.01.

Discussion

Vascular aging and hypertension have been studied extensively in elastic arteries but less so in muscular arteries. Clinical studies suggest that (medium-sized) muscular arteries are robust and do not remodel against aging and hypertension (Bortolotto et al., 1999; Boutouyrie et al., 1999; Boutouyrie et al., 1992) but changes in microstructure and biomechanical wall properties have not been extensively investigated (Wagner and Humphrey, 2011). We found differential changes in structure, material properties, and active responses in the large elastic and medium- and small-sized muscular arteries in response to chronic infusion of AngII over 14 days. Elastic fibers, which have an important role in endowing the arterial wall with resilience, remained intact in the elastic DTA consistent with prior observations (Bersi et al., 2017; Spronck et al., 2020). Nevertheless, the ability of elastin to store elastic energy reduced in AngII DTA (Fig 2H), as revealed in part by iso-energy contour plots that also showed altered anisotropy reflective of differential biaxial remodeling (Fig S3A–C). Also in the AngII DTA, significant medial thickening driven by accumulation of hydrophilic GAGs, which are efficient in resisting compressive loads, increased structural stiffness (Table S2) and appeared to prevent the intact elastin in the DTA from storing energy.

This effect has also been observed in a murine model of premature aging with intact aortic elastic lamellae and significant GAG accumulation (Murtada et al., 2020a). Such medial thickening and GAG accumulation were absent in both medium- and small-sized muscular arteries in response to AngII. Yet, changes were found in both elastic fiber and collagen fiber content in the small-sized muscular arteries; there was an unexpected trend toward degradation of the EEL in the MA, suggesting a consequence of hypertension that has not been well studied. Interestingly, similar observations of EEL degradation have been reported in aged human superficial femoral arteries (Kamenskiy et al., 2015), with fragmentation and degradation possibly being a consequence of fatigue. Such fatigue does not exist in young arteries. They also found similar trends of maintained circumferential stress and reduced axial stretch in the aged human femoral artery, as observed herein in the MA in response to AngII-infusion. Although quantification of elastin supported its decrease in the AngII MA, its ability to store energy appeared to be less affected (Fig 2H). It is noteworthy that some MAs even resembled more of an arteriole-like vessel, with only an IEL, than a muscular artery with both EEL and IEL intact (Fig S5), noting that cerebral arteries are classified as muscular arteries though without an EEL (Hu et al., 2007). Regardless, these significant changes in elastin in the muscular arteries suggest that hypertension-induced vascular changes could be difficult to reverse.

In contrast, collagen fibers, which engage and bear load at higher pressures to prevent over-distension of the vessel and to protect intramural cells as well as elastic fibers, displayed a trend of increase in the DTA while remaining unaffected in the SMA and MA following AngII-infusion. This resulted in a significant increase in the collagen:elastin ratio in the DTA (Fig 1D), as reported in aged human upper thoracic aorta (Sawabe, 2010). The trend in increased collagen:elastin ratio and adventitial thickness fraction in the AngII MA appeared to prevent both enlargement of the loaded passive inner radius as well as increases in circumferential wall stress in response to hypertension.

Prior observations suggest that the degree of aortic maladaptation in hypertension correlates strongly with decreases in smooth muscle contractile capacity (Korneva and Humphrey, 2019), which also appeared to be consistent in the muscular arteries. As expected, the MA exhibited the strongest contractile capacity in response to KCl, which depolarizes the cell membrane and increases calcium influx and was unaffected following AngII-infusion. However, the response to PE, which acts through G-protein–coupled receptors (non-selective α1-AR), increased in MA alone. Interestingly, regional dose-response relationships for both the non-selective α1-AR PE and highly selective α1A-AR agonist A61603 revealed no change in sensitivity following chronic infusion of AngII (Fig S4), suggesting that a signaling pathway downstream of the G-protein–coupled receptors and separate from increases in calcium influx could be affected in the MA. Similar findings have been reported for norepinephrine, which stimulates α1 and α2 as well as β1 and β2 receptors, though to a lesser extent, to not differ in MA in mice following 14-day infusions of AngII (Barhoumi et al., 2017). The calcium sensitizing Rho-kinase, which is more active in resistance arteries than in elastic arteries (Murtada and Humphrey, 2018), is also involved in F-actin cytoskeletal polymerization and organization. Importantly, the trend of increased SMαA in AngII MA (Fig 3E–G) suggests that Rho-kinase activity could be affected in AngII MA. In addition, G-protein–coupled receptor (α1) activation is necessary in ex vivo arterial mechano-adaptation (Murtada et al., 2016b), and increased vascular tone in elastic arteries has been reported to be protective when stress-shielding vulnerable matrix (Ferruzzi et al., 2016), which could have contributed to the adaptive response in the MA as opposed to the mal-adaptive response in the DTA. These findings call for further studies on Rho-Kinase activation during hypertensive muscular arterial remodeling.

In addition to the increased contractile properties, a reduced vasodilatory function was observed solely in the MA. Endothelial dysfunction is well-known in hypertension and aging, though with complex interrelationships with oxidative stress and inflammation. Such changes affect vascular resistance (Dharmashankar and Widlansky, 2010). When considering together the PE-induced contractility and the loss in Ach-mediated relaxation, a significantly smaller active loaded inner radius was found in the MA following AngII-infusion, potentially increasing its capability to attenuate the propagation of the pressure pulse wave into end organs, though with the effect of increasing total peripheral resistance (which is dependent on flow-rate, viscosity, and segment length, and inversely on radius to the fourth power) and thus further increasing mean arterial pressure.

Arterial stiffness is a determinant of the speed at which the pressure pulse wave travels; central-to-peripheral gradients in arterial stiffness associate with cardiovascular diseases. In younger and healthier arteries, elastic artery stiffness is lower than muscular artery stiffness and associates with a smooth, continuous flow of blood to the microvasculature of major organs. With aging, however, these regional gradients reverse and associate with increased pulsatility in the microvasculature, which can contribute to end-organ damage (Yu and McEniery, 2020). We found a reduced, not reversed, PWV gradient in response to AngII-infusion (Fig 2D), suggesting that underlying changes in microstructure and active response could be an early initiator of this difference. The increased aortic stiffness, quantified via either calculated structural stiffness or regional PWV, enhances the transfer of excessive, potentially harmful pulsatile energy into the microcirculation (Mitchell, 2008) and could contribute to the change in structure observed in the muscular MA following AngII-infusion. In addition, reports of early structural changes in elastic arteries without significant changes in vascular reactivity or compliance in muscular arteries after 7- and 10-day infusions of AngII (Flamant et al., 2007; Leloup et al., 2015) further support that changes in small peripheral vessels occur subsequent to the remodeling in the aorta. T cell–derived cytokines (interleukin-17) act on vascular SMCs and fibroblasts to increase collagen synthesis and chemokine production, leading to impaired vasodilation, increased vascular stiffness, and inflammation during hypertension (McMaster et al., 2015); such changes could be involved in the observed vascular changes in the small-sized MA. Our data suggest that MA transitions to a vessel with more arteriole-like characteristics which may enhance protection of end-organs in response to hypertension. Further molecular, microstructural, and mechanical studies of peripheral arteries are warranted to better understand vascular disease progression during hypertension.