Effects of Pulmonary Hypertension on Microcirculatory Hemodynamics in Rat Skeletal Muscle

Abstract

Pulmonary hypertension (PH) has previously been characterized as a disease of the pulmonary vasculature that subsequently results in myocardial dysfunction. Heart failure compromises skeletal muscle microvascular function, which contributes to exercise intolerance. Therefore, we tested the hypothesis that such changes might be present in PH. Thus, we investigated skeletal muscle oxygen (O₂) transport in the rat model of PH to determine if O₂ delivery $\left(\dot{\mathrm{Q}}{\mathrm{O}}_2\right)$ is impaired at the level of the microcirculation as evidenced via reduced red blood cell (RBC) flux, velocity, hematocrit, and percentage of capillaries flowing in quiescent muscle. Adult male Sprague-Dawley rats were randomized into healthy (n=9) and PH groups (n=9). Progressive PH was induced via a one-time intraperitoneal injection of monocrotaline (MCT; 50 mg/kg) and rats were monitored weekly via echocardiography. Intravital microscopy in the spinotrapezius muscle was performed when echocardiograms confirmed moderate PH (preceding right ventricular (RV) failure). At 25 ± 9 days post-MCT, PH rats displayed RV hypertrophy (RV/(Left ventricle+Septum): 0.28 ± 0.05 vs. 0.44 ± 0.11), pulmonary congestion, and increased right ventricular systolic pressure (21 ± 8 vs. 55 ± 14 mmHg) compared to healthy rats (all P < 0.05). Reduced capillary RBC velocity (403 ± 140 vs. 227 ± 84 μm/s; P = 0.01), RBC flux (33 ± 12 vs. 23 ± 5 RBCs/s; P = 0.04) and % of capillaries supporting continuous RBC flux at rest (79 ± 8 vs. 56 ± 13%; P = 0.01) were evident in PH rats compared to healthy rats. When $\dot{\mathrm{Q}}{\mathrm{O}}_2$ within a given field of view was quantified (RBC flux x % of capillaries supporting continuous RBC flux), PH rats demonstrated lower overall $\dot{\mathrm{Q}}{\mathrm{O}}_2$ (↓ 50%; P = 0.002). These data support that microcirculatory hemodynamic impairments (↓$\dot{\mathrm{Q}}{\mathrm{O}}_2$ and therefore altered $\dot{\mathrm{Q}}{\mathrm{O}}_2$-to-$\dot{\mathrm{V}}{\mathrm{O}}_2$ matching) may compromise blood-myocyte O₂ transport in PH. The mechanistic bases for decreased capillary RBC flux, velocity, and percentage of capillaries supporting RBC flow remains an important topic.

Introduction

Pulmonary hypertension (PH) is a life-threatening disease characterized by exercise intolerance, dyspnea, and poor quality of life. Exercise intolerance is often defined as low maximal aerobic capacity $\left(\dot{\mathrm{V}}{\mathrm{O}}_{2\max }\right)$, which is determined by both perfusive and diffusive oxygen (O₂) transport (Wagner et al., 1997). The capillary bed represents the functional and structural framework for substrate exchange in skeletal muscle. Any aberrations in capillary hemodynamics will impact perfusive and/or diffusive O₂ transport across the blood-myocyte interface, the former of which is determined primarily by red blood cell (RBC) flux (f_RBC) and O₂ content and the latter by the number of RBCs flowing in the capillaries or capillary bed adjacent to the myocytes at any instant. Elevated pulmonary vascular pressures in PH compromise pulmonary and cardiac function, often resulting in hypoxemia and attenuated cardiac output $\left({\stackrel{.}{\text{Q}}}_{\text{TOT}}\right)$, which would effectively lower $\dot{\mathrm{V}}{\mathrm{O}}_{2\max }$ by reducing perfusive O₂ transport. Importantly, capillary density is correlated with reduced exercise capacity in PH (Mainguy et al., 2010), and may therefore impair blood-myocyte O₂ exchange at the level of the tissue, potentially highlighting a concomitant role for impaired diffusive transport to contribute to low muscle oxygenation seen in this disease state (Malenfant et al., 2015; Dimopoulos et al., 2013). Crucially, we are unaware of any direct in vivo analysis (via intravital microscopy) of skeletal muscle capillary hemodynamics in PH, which is critical for resolving the impact of this disease on capillary function and muscle O₂ delivery.

In left ventricular heart failure with reduced ejection (HFrEF), attenuations in $\dot{\mathrm{V}}{\mathrm{O}}_{2\max }$ result from impaired perfusion (blood flow,$\dot{\mathrm{Q}}$), as well as a lower percentage of capillaries supporting continuous f_RBC, thereby reducing the surface area available for O₂ exchange to the myocyte (Kindig et al., 1999). PH patients often present with lower ${\stackrel{.}{\text{Q}}}_{\text{TOT}}$, however, Tolle and colleagues (2008) suggest that an impaired systemic oxygen extraction (arteriovenous O₂ difference, a-vO₂) during maximal exercise, hypothesizing that, unlike HFrEF, the lower $\dot{\mathrm{V}}{\mathrm{O}}_{2\max }$ is not primarily due to decreased $\dot{\mathrm{Q}}$, but rather reduced contact between capillaries and contracting myocytes. Based on modeling studies by Federspiel and Popel (1986), this would, in turn, compromise the diffusive component of blood-myocyte O₂ transfer. Furthermore, capillary rarefaction has been implicated in the quadriceps muscle in PH patients (Potus et al., 2014), supporting the potential for decreased surface area available for blood-myocyte O₂ flux, which would lower intramyocyte PO₂ and impair skeletal muscle O₂ utilization (Hogan et al., 1992, Wilson et al., 1977).

The presence of capillary rarefaction, endothelial dysfunction, and reduced ${\stackrel{.}{\text{Q}}}_{\text{TOT}}$ suggests that PH patients may display changes in capillary hemodynamics at rest. Therefore, this study aimed to determine if PH elicits changes in capillary hemodynamics as a putative mechanism for impaired perfusive and diffusive O₂ transport in skeletal muscle in PH. This rationale is based, in part, upon our previous observations in HFrEF (Kindig et al., 1999; rev. Poole et al., 2012). We hypothesized that rats with monocrotaline (MCT)-induced PH would display: 1) reductions in f_RBC and RBC velocity (V_RBC) in capillaries of resting skeletal muscle, and 2) a lower proportion of capillaries supporting continuous f_RBC and thus O₂ delivery to the myocyte. The results of this investigation provide fundamental insights into the currently underexplored field of skeletal muscle microcirculatory function in PH which may support a mechanistic basis for impaired skeletal muscle oxygenation and thus exercise intolerance in patient populations.

Methods

Animals

Experiments were performed on 18 adult (~3–4 months) male Sprague-Dawley rats (~300 g; Charles River Chicago, Illinois, USA). Animals were randomized into the monocrotaline (MCT)-induced PH or healthy control (HC) groups (MCT, n=9; HC, n=9). Upon arrival, animals were maintained in accredited (Association for the Assessment and Accreditation of Laboratory and Animal Care; AAALAC) animal facilities under a 12:12 h light:dark cycle with food and water provided ad libitum. All procedures and protocols were approved by the Kansas State University Institutional Animal Care and Use Committee and conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. The present investigation contains a partial subset of a separate study (Schulze et al., 2021), but all measurements reported herein are original findings.

Monocrotaline-induced Pulmonary Hypertension

Monocrotaline alkaloid (MCT) was administered via a single intraperitoneal injection at a dosage of 50 mg/kg, which has been shown to induce progressive PH (Gomez-Arroyo et al., 2011; Brown et al., 2015). MCT (Sigma-Aldrich, St. Louis, MO) was dissolved in a solution containing 50% saline and 50% 200-proof ethanol at room temperature and a total of 1 mL of fluid was injected into each rat in the experimental group while HC rats received 1 mL of vehicle (0.5 mL saline and 0.5 mL 200-proof ethanol). Injections were made immediately following pre-injection echoes under anesthesia (see Echocardiography) with a 27 G needle in the lower right quadrant of the abdomen with care taken to avoid internal organs. Importantly, characteristics of PH (i.e., elevations in pulmonary artery pressures, remodeling of small pulmonary arteries, and right ventricular (RV) hypertrophy) occur 3–4 weeks following injection. Animals were monitored weekly via echocardiography along with noting body weight and behavioral changes.

Echocardiography

Transthoracic echocardiography was performed using a commercially available system (Logiq S8; GE Health Care, Milwaukee, WI) with a 13 MHz linear transducer (L4–12t) prior to MCT or saline injection, and then weekly following injection to monitor disease progression. All comparisons were made between the pre-injection measurement and final measurement taken just prior to experimentation. Animals were initially anaesthetized by inhalation of a 5% isofluorane-O₂ mixture and maintained on ≤2.0% isofluorane-O₂ (Butler Health Supply, Dublin, OH). Core temperature was maintained at ~37°C, measured via rectal thermometer. 2-D and M-mode images of the left ventricle (LV) were obtained from the parasternal short axis window and analyzed for end-systolic/diastolic volumes, stroke volume, and ejection fraction as previously described (Craig et al., 2019). RV measurements were taken at the level of the aortic valve, just proximal to the pulmonary valve, in the short axis view. Pulsed Doppler ultrasound was used to assess pulmonary artery ejection time (ET), acceleration time (AT), and peak velocity, and the ratio of AT to ET (AT/ET) was calculated and used as a clinical parameter for the confirmation of PH (AT/ET < 0.3) (Jones et al., 2002). Upon verification of PH, the following terminal experiments were performed.

Surgical Preparations

All surgeries were performed utilizing the same anesthetic protocol as described above. The right jugular vein and right common carotid artery were surgically isolated and cannulated with a 2-French catheter-tip pressure micromanometer (Millar Instruments, Houston, TX, USA). The micromanometer was advanced into the RV and LV, respectively, to obtain pressures within each ventricle via a PowerLab/LabChart data acquisition system (AD Instruments). Following RV and LV measurements, the micromanometer was withdrawn and the right common carotid artery was catheterized (PE-10 connected to PE-50, Intra-Medic polyethylene tubing; Clay Adams Brand; Becton Dickinson, Sparks, MD) for measurement of mean arterial pressure (MAP) and heart rate (HR). The caudal artery was surgically isolated and catheterized for administration of pentobarbital sodium. Rats were then progressively transitioned from isoflurane onto pentobarbital sodium (50 mg/kg). Depth of anesthesia was continuously monitored via toe pinch and palpebral reflexes with pentobarbital sodium supplemented as deemed necessary throughout the remaining experimental protocols.

Intravital Microscopy

Incisions were made to expose the left spinotrapezius muscle, with overlying skin and fascia reflected as previously described (Hirai et al., 2018). The spinotrapezius was chosen for its muscle fiber-type composition and citrate synthase activity, being closely analogous to the untrained human quadricep muscle (Delp & Duan, 1996; Leek et al., 2001). The spinotrapezius was superfused with warmed (~37°C) Krebs-Henseleit bicarbonate-buffered solution equilibrated with 5% CO₂-95% N₂ at pH 7.4, and exposed tissue was protected with Saran Wrap (Dow Brands, Indianapolis, IN) to minimize dehydration. The muscle was carefully dissected and sutured (6–0 silk) around the caudal periphery to a thin wire horseshoe-shaped manifold at five to six equidistant points. The rat was then placed on a water circulation-heated (38°C) Lucite platform with the spinotrapezius positioned such that a microvascular field, midway between arteriolar and venular ends within the dorsal surface, could be observed using an intravital microscope (Nikon, Eclipse E600-FN; ×40 objective; 0.8 numerical aperture) equipped with a noncontact, illuminated lens, and a high resolution color monitor (total viewing area = 270 × 210 μm; Sony Trinitron PVM-1954Q, Ichinonya, Japan). The muscle was transilluminated to ensure clear resolution of the sarcomere A-bands within one-third to two-thirds of the muscle fibers. The final magnification (×1,184) was confirmed by initial calibration of the system using a stage micrometer (MA285, Meiji Techno). This magnification is adequate for measuring all essential structural and hemodynamic variables (Poole et al., 1997; Kindig et al., 1999). The manifold was attached to a muscle stretching apparatus that permitted precise length changes along the longitudinal axis of the muscle. Sarcomere length was approximated to <2.7 μm for physiological relevance throughout the observation period, and the muscle continuously superfused with the Krebs-Henseleit solution. Once the spinotrapezius muscle was positioned on the platform, 8–10 microvascular viewing fields were each recorded for ∼60–180 seconds for every animal. Recordings were time-referenced and fields and stored for subsequent offline analysis. MAP was monitored throughout the duration of data acquisition. Arterial blood samples (0.5 mL) were taken at the end of microcirculatory measurements for arterial pH, PaCO₂, PaO₂, systemic hematocrit, and lactate concentration.

Capillary Data Analysis

Five to seven fields demonstrating clear visualization of sarcomeres, fibers, and capillaries were chosen for analysis from each rat at random. Examination of the microvascular fields was conducted in real time and by frame-by-frame analysis techniques (Dartfish Video Software, Fribourg, Switzerland). Muscle fiber width was measured along a line perpendicular to the muscle fiber axis for each muscle fiber. Sarcomere length was calculated from sets of 10 consecutive sarcomeres (i.e., distance between 11 consecutive A-bands) measured parallel to the muscle fiber longitudinal axis. This measurement was repeated 3–4 times where sarcomeres were visible to obtain a mean sarcomere length for each viewing field. Capillary lineal density was determined as the number of capillaries crossing a line drawn perpendicular to the fiber axis and expressed per mm. Capillaries supporting RBC flow were assessed in real time, and each capillary was placed into one of three categories: 1) normal flow = 30 s of continuous, 2) intermittent flow = impeded flow for <10 of 30s, or 3) stopped flow = impeded flow or stopped flow for >10 of 30 s (i.e., stationary RBCs). These criteria determined the percentage of continuously flowing capillaries in a field of view [i.e., (number of capillaries supporting RBC flow/total number of visible capillaries per area) × 100]. Capillary luminal diameter (d_c) was measured in all capillaries in which hemodynamics were assessed and where the capillary endothelium was clearly visible on both sides of the lumen (2–4 measurements/capillary) with calipers accurate to ± 0.25 mm (± 0.17 μm at ×1,184 magnification). V_RBC was calculated in all capillaries that were continuously RBC perfused and in which the RBC could be followed over several frames and RBC path length measured. f_RBC was determined by counting the number of RBCs in a capillary which passed an arbitrary point. For each capillary in which hemodynamic data were measured, capillary tube hematocrit (Hct_cap) was calculated by the following equation:

$${\text{Hct}}_{\text{cap}}=\left(\text{RBC}\text{volume}\times f_{\text{RBC}}\right)/\left[\pi \times {\left({\text{d}}_{\text{c}}/2\right)}^2\times V_{\text{RBC}}\right]$$

where RBC volume was taken to be 61 μm³ (Altman & Dittmer, 1974) and capillaries were approximated as circular in cross section (Kindig et al., 1998).

Postmortem Measurements

While under a deep plane of anesthesia, all rats were euthanized via pentobarbital sodium overdose (>100 mg kg⁻¹, i.a) followed by a bilateral pneumothorax. The RV, LV, lungs, diaphragm, and spinotrapezius muscle, were dissected and weighed. The right lower leg of each rat was exposed, and the tibial bone was isolated, removed, and the length of the bone measured. RV hypertrophy was determined via the Fulton index: RV weight (mg) / LV + septum (S) weight (mg) (Fulton et al., 1952). RV hypertrophy was also expressed as RV/body weight and RV/tibia bone length.

Data and Statistical Analyses

Student paired (within animal) and independent two-sample (between HC and MCT) t-tests were utilized to determine differences in morphometric, Millar, and hemodynamic measurements, followed by Pearson correlation tests and linear regression analyses of these data. Mann Whitney tests were used when data were nonparametric. Echocardiographic data was analyzed via two-way repeated measures Analysis of Variance (ANOVA). Data are presented as mean ± standard deviation. Significance was set at P < 0.05.

Results

Morphometric and Echocardiographic Data

Eighteen rats were analyzed for morphometric and echocardiographic data (HC, n = 9; MCT, n = 9). No differences were observed in pre-injection body weight (HC: 307 ± 31 vs. MCT: 302 ± 26 g; P > 0.05) or echocardiographic measurements (P > 0.05; Table 1). However, final body weight was lower in MCT rats (HC: 445 ± 35 vs. MCT: 379 ± 24 g; P < 0.05), therefore tibial bone length was used for relative comparisons of lung mass between groups. Left ventricular echocardiographic measurements did not differ between groups (P > 0.05 for all; Table 1). MCT rats displayed hallmarks of PH including elevated right ventricular systolic pressure (Table 1), increased lung weight normalized to tibial bone length (Table 2), RV hypertrophy (Table 2), and a reduced acceleration time to ejection time ratio in the pulmonary artery (Table 1) compared to healthy rats (all P < 0.05). In addition, the MCT rats demonstrated a significant reduction in LV fractional shortening in their final echocardiogram when compared to pre-injection measurements and a lower LV dP/dt compared to healthy (Table 1). However, these rats did not exhibit any signs of overt LV failure (i.e., significant reductions in LV ejection fraction or increases in LVEDP; Table 1). These characteristics support the development of PH, pulmonary congestion, and RV compensation preceding overt RV or LV heart failure in the MCT group.

Table 1.

Echocardiographic and Millar catheter measurements.

	Healthy		MCT
	Pre	Final	Pre	Final
Echocardiographic
PA ET (ms)	85 ± 8	82 ± 6	79 ± 5	87 ± 6
PA AT (ms)	30 ± 2	29 ± 4	28 ± 2	19 ± 3 *, #
PA Peak Velocity (cm/s)	90 ± 19	101 ± 15	97 ± 16	95 ± 10
PA AT/ET	0.37 ± 0.07	0.35 ± 0.04	0.36 ± 0.04	0.22 ± 0.05 *, #
LV Stroke Volume (mL)	0.54 ± 0.14	0.76 ± 0.18 #	0.58 ± 0.16	0.63 ± 0.18
LV Ejection Fraction (%)	84 ± 5	85 ± 7	87 ± 5	81 ± 7
LV Fractional Shortening (%)	49 ± 5	54 ± 13	58 ± 14	45 ± 7 #
Millar Catheter
RVSP (mmHg)	-	21 ± 8	-	53 ± 15 *
LVEDP (mmHg)	-	11 ± 4	-	8 ± 3
LV dP/dt (mmHg/s)	-	7184 ± 1467	-	5855 ± 870 *

Data are means ± SD. Healthy (n = 9), monocrotaline (MCT, n = 9). PA, pulmonary artery; ET, ejection time; AT, acceleration time; LV, left ventricle; RVSP, right ventricular systolic pressure; LVEDP, left ventricular end-diastolic pressure; LV dP/dt, rise in pressure over time. Reduced PA AT and AT/ET determined disease progression (criteria for terminal studies). This was accompanied by increases in RVSP.

^*P < 0.05 vs. healthy

^#P < 0.05 vs. pre-injection within group.

Table 2.

Morphometric data.

	Healthy	MCT	P-value
RV (mg)	249 ± 29	365 ± 88 *	0.005
LV (mg)	888 ± 122	830 ± 117	0.316
Lungs (mg)	1403 ± 220	2321 ± 277 *	< 0.001
Diaphragm (mg)	967 ± 134	876 ± 139	0.174
RV/(LV+S) (mg/mg)	0.28 ± 0.05	0.44 ± 0.11 *	< 0.001
RV/BW (mg/g)	0.56 ± 0.06	0.96 ± 0.23 *	0.001
RV/Tibia (mg/mm)	6.0 ± 0.6	8.5 ± 2.0 *	0.010
Lung/BW (mg/g)	3.14 ± 0.32	6.15 ± 0.83 *	< 0.001
Lung/Tibia (mg/mm)	32.7 ± 4.4	53.0 ± 8.0 *	< 0.001
Diaphragm/BW (mg/g)	2.17 ± 0.24	2.30 ± 0.28	0.306
Diaphragm/Tibia (mg/mm)	22.7 ± 2.8	20.3 ± 4.1	0.204

Data are means ± SD. Healthy (n = 9), monocrotaline (MCT, n = 9). RV, right ventricle; LV + S, left ventricle plus septum; BW, body weight; Tibia, tibial bone length.

^*P < 0.05 vs. healthy.

Intravital Microscopy

Seventeen rats (HC, n = 9; MCT, n = 8) were analyzed for capillary hemodynamics. One rat from the MCT group was excluded due to insufficient clarity to measure the desired variables (i.e., f_RBC, V_RBC and Hct_cap). The proportion of capillaries flowing and sarcomere length were still measured in this animal and included in the final data set. Capillary lineal density was measured only when there was sufficient clarity to count total number of capillaries in a given field. Therefore, lineal density was calculated in a subset of fifteen rats (HC, n = 7; MCT, n = 8).

Spinotrapezius sarcomere length was not different between groups (2.5 ± 0.2 μm for both MCT and HC). The mean capillary diameter, lineal density, and muscle fiber width were not different in MCT compared to HC (Table 3). Heart rate was elevated in MCT rats (HC: 361 ± 23 vs MCT: 403 ± 25 bpm; P = 0.010), but perfusion pressures (i.e., MAP) did not differ between groups throughout the duration of data acquisition (HC: 104 ± 15 vs. MCT: 102 ± 16 mmHg). No differences were present between groups for arterial blood pH (HC: 7.43 ± 0.04 vs. MCT: 7.38 ± 0.07; P = 0.09), PaCO₂ (HC: 32 ± 3 vs. MCT: 29 ± 4 mmHg; P = 0.13), systemic hematocrit (HC: 33 ± 3 vs. MCT: 33 ± 5%; P = 0.86), or lactate (HC: 1.01 ± 0.33 vs. MCT: 1.02 ± 0.29 mmol/L; P = 0.94); however PaO₂ (HC: 81.9 ± 9.4 vs. MCT: 68.7 ± 10.0 mmHg; P = 0.01) and SaO₂ (HC: 94.5 ± 3.0 vs. MCT: 89.0 ± 8.5%; P = 0.045) were lower in MCT rats. MCT rats demonstrated a 30% reduction in the proportion of capillaries that supported continuous f_RBC and a higher percentage of capillaries that contained temporarily (>10 s) stopped RBCs or that did not support any f_RBC over the entire window of observation (Figure 1). In the capillaries supporting continuous RBC flow, MCT rats displayed a substantially lower V_RBC and f_RBC (Figure 2) compared to HC (both P < 0.05) but Hct_cap did not differ between groups (Table 3). There was a significant linear correlation between f_RBC and V_RBC in all animals (P < 0.05; Figure 3). The product of the percentage of continuously flowing capillaries and f_RBC measures continuous O₂ delivery $\left(\stackrel{.}{\text{Q}}{\text{O}}_2\right)$ within a given field sampled. There was an average ~50% decrease in $\stackrel{.}{\text{Q}}{\text{O}}_2$ to resting skeletal muscle in MCT versus healthy animals (P < 0.05; Figure 4). Notably, f_RBC was negatively linearly correlated with structural (RV/(LV+S)) and functional (RVSP) indicators of PH severity (both P < 0.05; Figures 5A–B).

Table 3.

Capillary Measurements

	Healthy	MCT	P-value
Hctcap (%)	19 ± 4	19 ± 4	0.975
Capillary Diameter (μm)	5.8 ± 0.7	5.7 ± 0.5	0.694
Lineal Density (capillaries/mm)	36 ± 5	33 ± 3	0.131
Muscle Fiber Width (μm)	64 ± 5	62 ± 6	0.381

Data are means ± SD. Healthy (n = 9), monocrotaline (MCT, n = 8) for Hct_cap, capillary hematocrit.

Figure 1:

Percentage of capillaries supporting continuous and intermittent red blood cell (RBC) flow and stopped capillaries in healthy (HC, n = 9) and monocrotaline (MCT, n = 9) rats. * P < 0.05 versus HC.

Figure 2:

Comparison of red blood cell (RBC) hemodynamics (velocity, V_RBC and flux, f_RBC) in muscles of healthy (HC, n = 9) and monocrotaline (MCT, n = 8) rats. * P < 0.05.

Figure 3:

Relationship between RBC velocity (V_RBC) and flux (f_RBC) in healthy controls (HC, n = 9) and monocrotaline (MCT, n = 8). Note that V_RBC and f_RBC increase or decrease in proportion to one another to maintain a given hematocrit. The slope of the solid line dictates hematocrit.

Figure 4:

Index of continuous O₂ delivery to muscle in healthy (HC, n = 9) and monocrotaline (MCT, n = 8) animals. * P < 0.05.

Figure 5:

Relationship between red blood cell (RBC) flux (f_RBC) and structural (A) and functional (B) measurements of PH severity. RV/(LV+S), right ventricular weight/weight of left ventricle + septum; RVSP, right ventricular systolic pressure.

Discussion

This investigation assessed, for the first time to our knowledge, skeletal muscle capillary hemodynamics in MCT-induced PH compared to healthy rats. The distribution of skeletal muscle blood flow within the capillary network has not previously been examined and is crucial to understanding O₂ exchange in PH. The primary original findings of this study support the notion that resting skeletal muscle blood flow is reduced in MCT-induced PH, occurring via a reduction in both V_RBC and f_RBC. Further, the proportion of non-flowing capillaries (i.e., zero f_RBC) is markedly increased in MCT animals. These findings support the hypotheses that both perfusive and diffusive mechanisms of microvascular O₂ transport are compromised in PH, offering valuable insights for skeletal muscle dysfunction in PH.

Validation of MCT-induced PH

Confirmation of moderate PH was assessed using clinically established measurements (AT/ET, RVSP; Jones et al., 2002; Kitabatake et al., 1983) as well as morphometric data consistent with previous investigations in the MCT rat model used herein (Brown et al., 2015; 2017; Schulze et al., 2021). Our laboratory recently demonstrated that the dose of MCT (50 mg/kg) used herein elicits reproducible PH development within ~3 weeks in male Sprague-Dawley rats (Schulze et al., 2021). Tables 1 and 2 demonstrate that the animals in the present investigation displayed hallmarks of PH progression (i.e., ↑ RV/(LV+S), ↑ lung weight, ↓ AT/ET, ↑ RVSP) within 25 ± 9 days, consistent with previous literature (Gomez-Arroyo et al., 2011). Importantly, this model induces moderate PH, indicating that the alterations in the skeletal muscle microcirculation presented herein occur prior to and independent of distinct RV or LV failure as indicated by lack of changes in LV ejection fraction and LVEDP.

Hemodynamics

Surgical exteriorization of the spinotrapezius muscle in vivo was performed as previously described which limits disruption of neural and vascular supply and induces minimal fascial disturbance to prevent muscle damage (Gray, 1973; Bailey et al., 2000; Poole et al., 1997; Kindig et al., 1999). This preparation has been performed in healthy rats in addition to heart failure, aging, and diabetes and facilitates collection of high resolution capillary structural and hemodynamic variables which are consistent across healthy control animals in previous studies and within the present investigation (Kindig et al., 1999; Padilla et al., 2006; Copp et al., 2009; Hirai et al., 2021). Sarcomere length was maintained at <2.7 μm to prevent stretch-induced functional or structural alterations, as previously studied (Poole et al., 1997). In addition, healthy animals in this investigation presented with a majority of capillaries (>80%) flowing (Figure 1), as is expected in healthy resting skeletal muscle (Poole et al., 2011; Poole, 2019; Mendelson et al., 2021).

Consistent with our hypotheses, MCT rats demonstrated substantial decrements in capillary hemodynamics (i.e., reduced V_RBC, f_RBC, and percentage of capillaries supporting continuous flow) at rest compared to healthy controls, suggesting arteriolar and/or microcirculatory impairments in skeletal muscle. V_RBC and f_RBC represent the perfusive component of O₂ and substrate exchange, with such decreases indicating reduced muscle $\stackrel{.}{\text{Q}}{\text{O}}_2$. It is possible that lower ${\stackrel{.}{\text{Q}}}_{\text{TOT}}$ could explain, in part, the decreased V_RBC and f_RBC; however, it is important to note that stroke volume was only marginally less in MCT animals compared to HC (P=0.131) and heart rate tended to be elevated. Consistent with the findings of Tolle et al. (2008), this suggests that the changes in ${\stackrel{.}{\text{Q}}}_{\text{TOT}}$ alone would not be sufficient to lower muscle V_RBC and f_RBC in PH. This suggests the presence of arteriolar vasoconstriction to the quiescent spinotrapezius. The present investigation could not determine whether reductions are present in V_RBC and f_RBC during contractions. However, our previous observations in HFrEF rats demonstrated that capillaries not supporting f_RBC at rest failed to resume flowing during contractions (Richardson et al., 2003). Importantly, the proportional decrements in V_RBC and f_RBC (Figure 2) maintained capillary hematocrit at ~19% irrespective of PH such that, although perfusive conductance was decreased, the O₂ diffusing capacity per flowing capillary would not be expected to change with PH.

As alluded to above, one potential explanation for the lower V_RBC and f_RBC, which were reduced in proportion to the cardiac structural (Figure 5A) and functional (Figure 5B) derangements of PH, lies within the arteriolar network, which controls blood flow distribution within the capillary beds. Arteriolar control may be disrupted due to dysregulation of vasoactive substances including prostacyclin, endothelin-1, and nitric oxide (Humbert et al., 2004), or via systemic arterial stiffening (Radchenko et al., 2020), all of which have been implicated in the pathophysiology of PH. Such disturbances in skeletal muscle arteriolar function could impair vasodilation and/or facilitate chronic increases in vascular tone. Peripheral endothelial dysfunction has also been suggested in PH and, as the endothelium is responsible for sensing changes in shear stress and responding to vasoactive mediators, this could represent an additional limitation to perfusive O₂ conductance (Peled et al., 2008; Wolff et al., 2007). Furthermore, increased sympathetic nerve activity with PH could exacerbate redistribution of blood flow away from skeletal muscle (Velez-Roa et al., 2004). Any of these mechanisms, independently or in concert, would effectively lower perfusion to this microvascular bed at rest, resulting in such reductions in V_RBC and f_RBC.

Importantly, the percentage of capillaries flowing in healthy skeletal muscle at rest has been shown to be >80% (Poole et al., 2011; Poole, 2019; Mendelson et al., 2021). Our healthy values herein corroborate these previous findings; however, there was a 30% reduction in the proportion of continuously perfused capillaries in the MCT group. Three potential mechanisms exist to explain the increased proportion of stopped capillaries in MCT rats: 1) As mentioned above, compromised arteriolar function due to impaired smooth muscle or endothelium, or alterations in levels of vasoactive substances limiting the capacity for proper arteriolar control, 2) PH increases pulmonary artery and RV pressures and consequently induces venous congestion (increased central venous pressure, CVP). Because the pressure differential (ΔP) across skeletal muscle capillaries is extremely small (Mendelson et al., 2021), elevated CVP would decrease capillary flow. 3) As a compensatory mechanism to constrain the fall in f_RBC and preserve Hct_cap. Should all capillaries remain flowing, this would necessitate a greater magnitude of the fall in V_RBC, f_RBC, and perhaps Hct_cap in order to perfuse all flowing capillaries.

The 40% decrease in V_RBC (in concert with reduced f_RBC; Figure 2) lowers perfusive O₂ conductance, but it will also serve to lengthen RBC transit time through the capillary, allowing for greater fractional O₂ extraction. Indeed, it is by this mechanism that fractional O₂ extraction is elevated in left ventricular heart failure (i.e., HFrEF; Kindig et al., 1999). If extraction is represented by the formula (Roca et al., 1992):

$$\%{\text{O}}_2\text{extraction}=1-{\text{e}}^{-{\text{DmO}}_2/\beta {\dot{\text{Q}}}_{\text{m}}}$$

where D_mO₂ is muscle O₂ diffusing capacity, β is the slope of the O₂ dissociation curve in the physiologically relevant range, and ${\stackrel{.}{\text{Q}}}_{\text{m}}$ is muscle blood flow, a reduction in ${\stackrel{.}{\text{Q}}}_{\text{m}}$ would be expected to increase fractional extraction in the absence of a concomitant lowering of D_mO₂. As D_mO₂ is determined by the surface area contact between the RBCs in the flowing capillaries and adjacent myocytes, this would be dictated in an individual capillary by Hct_cap. Akin to HFrEF (Kindig et al., 1999), Hct_cap was not different between healthy and MCT animals, but the lowered perfusive O₂ conductance would necessitate an increased fractional O₂ extraction in capillaries supporting f_RBC. Despite this, the lower proportion of flowing capillaries in MCT rats would reduce the total surface area contact between the RBC and the myocyte within the muscle at any given instant, limiting overall D_mO₂ (Wagner et al., 1997). Such heterogeneity among capillaries (i.e., some displaying increased extraction while others support no O₂ flux) may, in part, explain the unchanged level of whole muscle O₂ extraction, reflected by resting interstitial oxygen pressures in the spinotrapezius (PO₂is) (Wagner et al., 1997; Schulze et al., 2021). Notwithstanding this occurrence, to sustain a given $\dot{\mathrm{V}}{\mathrm{O}}_2$, fractional O₂ extraction must be substantially increased in the individual capillaries supporting flow. According to the Fick’s principle, estimations for resting muscle $\dot{\mathrm{V}}{\mathrm{O}}_2$ herein are ~36% lower in MCT rats, suggesting that such elevations in fractional O₂ extraction are insufficient to compensate for the lower f_RBC and increased proportion of capillaries which are not participating in O₂ exchange.

Experimental Considerations

1) Females display a higher incidence of PH and, interestingly, better survivability (Lahm et al., 2014) and the cardioprotective nature of estrogen may provide some insights into this phenomenon. As the present investigation focuses on male rats, future studies examining both female and ovariectomized female rats should assess sexual dimorphism in microcirculatory hemodynamics, and whether these potential differences may be mediated by estrogen. 2) Transillumination microscopy was chosen for its ability to visualize capillaries within the skeletal muscle in real time while minimizing tissue damage. It is possible that, using this technique, we were unable to visualize all capillaries within a given field of view. However, as image clarity and lineal density were not obviously different between groups, it is unlikely that the inability to visualize individual capillaries played any substantial role in the capillary hemodynamics results herein. 3) Although we examined only a relatively modest area of tissue Per screen (270 × 210 μm), several screens and multiple capillaries per screen were analyzed, wherein we detected no significant differences across screens within a given muscle in any of the primary measurements. Blood flow data using fluorescent microspheres in skeletal muscle have not yet been reported in the MCT model and would be a valuable future direction for linking the microvascular results in the present investigation to bulk tissue blood flow. 4) The MCT rat model allows mechanistic study of PH and elicits reliable hallmarks of disease such as pulmonary vascular remodeling and subsequent RV hypertrophy (Gomez-Arroyo et al., 2011). The data herein may provide insights into the underlying bases for skeletal muscle impairments documented in human patients. While off-target effects of MCT on skeletal muscle have not been determined, the metabolism of MCT allows it to primarily affect the pulmonary vasculature, mirroring the human pathology and, to our knowledge, MCT does not affect peripheral mitochondrial or vascular function directly. Nonetheless, translation of animal research to humans should be employed with caution. Additionally, it is important to note that the present investigation focused on moderate PH. With increased duration of pathology following MCT injection and thus greater disease severity, structural microvascular changes not seen herein may potentially develop.

Conclusions

MCT-induced PH in male rats lowers resting microvascular skeletal muscle blood flow via reduced perfusive (i.e., V_RBC and f_RBC) and diffusive (proportion of flowing capillaries) O₂ transport, possibly due to upstream arteriolar dysregulation. Despite this, the lengthening of RBC capillary transit time, concurrent with unchanged Hct_cap, are expected to elevate muscle fractional O₂ extraction in the individual capillaries supporting flow which would allow the muscle to sustain a given metabolic rate. These data support the notion that capillary hemodynamics in quiescent skeletal muscle are impaired in MCT-induced PH and provide novel insights into potential mechanisms of skeletal muscle dysfunction in PH.

Highlights.

• Skeletal muscle dysfunction in PH may be due to reduced O₂ delivery to the muscle

• The proportion of capillaries supporting O₂ flux was decreased with PH

• PH lowered skeletal muscle O₂ delivery via impaired perfusive and diffusive O₂ transport

Abbreviations

AT

Acceleration time

D_mO₂

Muscle oxygen diffusing capacity

ET

Ejection time

f _RBC

Red blood cell flux

HC

Healthy control

Hct_cap

Capillary hematocrit

HFrEF

Heart failure with reduced ejection fraction

LV

Left ventricle

LVEDP

Left ventricular end diastolic pressure

MCT

Monocrotaline

PH

Pulmonary hypertension

$\stackrel{.}{\text{Q}}$

Blood flow

$\stackrel{.}{\text{Q}}{\text{O}}_2$

Oxygen delivery

${\stackrel{.}{\text{Q}}}_{\text{TOT}}$

Cardiac output

RBC

Red blood cell

RV

Right ventricle

RVSP

Right ventricular systolic pressure

$\stackrel{.}{\text{V}}{\text{O}}_{2\max }$

Maximal aerobic capacity

V _RBC

Red blood cell velocity