- Email: [email protected]
Microvascular architecture in rat soleus and extensor digitorum longus muscles
MICROVASCULAR
43, 192-204 (1992)
RESEARCH
Microvascular
Architecture in Rat Soleus and Extensor Digitorum Longus Muscles
DONNA A. WILLIAMS’ Laboratory
for Human
Performance University Park, Received
AND STEVEN S. SEGAL~ Research, Pennsylvania Pennsylvania 16802
State University,
June 14, 1990
Microvascular architecture was investigated in the slow-twitch soleus (SOL) and fasttwitch extensor digitorum longus (EDL) muscles. Rats (n = 5) were anesthetized and papaverine was infused into a carotid artery cannula to induce vasodilation. Microfil casting compound was then infused at an inflation pressure (caudal artery) of 100 mm Hg. Bilateral SOL and EDL muscles were excised 24-72 hr postcasting, dehydrated in ethanol, and cleared in methyl salicylate. Branch frequencies (BR) and segment lengths (SL) of intramuscular arterioles and venules were quantified along primary (lo), secondary (Z’), and tertiary (3”) order microvessels using microscopy. In both muscles, BR decreased with increasing vessel order. Regional differences in network organization were observed within the EDL muscle. SL of 1” arterioles was 47% shorter in the SOL muscle indicating more compact microvascular networks compared to the EDL muscle. These findings provide a structural basis for reported differences in blood flow between the SOL and EDL muscles at rest and during exercise. 0 1992 Academtc Press, Inc.
INTRODUCTION Skeletal muscle blood flow differs during exercise between muscles and according to fiber type within individual muscles (Folkow and Halicka, 1968; Laughlin and Armstrong, 1982; Mackie and Terjung, 1983). Mild to moderate exercise is characterized by elevated blood flow to slow-twitch-oxidative (SO) fibers. As exercise intensity increases, blood flow increases to fast-twitch-oxidative-glycolytic (FOG) and, then, to fast-twitch-glycolytic (FG) fibers (Laughlin and Armstrong, 1982). These blood flow responses correlate with the order of muscle fiber recruitment during graded exercise (Burke, 1981). Soleus (SOL) and extensor digitorum longus (EDL) muscles of the rat differ in contractile properties (Close, 1964), fiber type (Armstrong and Phelps, 1984), and blood flow (Armstrong and Laughlin, 1985). The SOL muscle is a slow-twitch, postural muscle that relies primarily upon oxidative metabolism (Close, 1972). In contrast, the fast-twitch EDL muscle is recruited during rapid, high-intensity I Present address: Department Physiology, University of Missouri Medical School, Columbia, MO 65212. ’ To whom reprint requests should be addressed at 119 No11 Laboratory, The Pennsylvania State University, University Park, PA 16802. 192 0026-2X62/92 $3.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved Printed in U.S.A.
MICROVESSELS
IN
RED
AND
WHITE
MUSCLE
193
activity and relies more upon glycolytic metabolism for energy production (Close, 1972). Differences in blood flow between the SOL and EDL muscles, at rest and during exercise, have also been documented (Armstrong and Laughlin, 1985). Blood flow differences between muscles may be explained, in part, by dissimilar vasomotor tone along resistance vessels. Another explanation may relate to length and diameter variability within the respective capillary beds (Potter and Groom, 1983). Differences in microvascular architecture proximal and distal to the capillary bed may also affect blood flow (Laughlin and Armstrong, 1982; Mackie and Terjung, 1983; Stingl, 1969); however, the structure of these portions of the vascular network has not been examined in skeletal muscles involved in locomotion. The purpose of the present study was to characterize the architecture of pre- and postcapillary networks supplying rat EDL and SOL muscles. We hypothesized that the microvascular networks would differ between these two muscles.
MATERIALS
AND METHODS
All procedures performed on animals were approved by the Institutional Animal Care and Use Committee of The Pennsylvania State University. Vumdur casting. Cage-sedentary, female, Sprague-Dawley rats (16 weeks old; 234 ? 3 g; IZ = 5) were sedated with rhompun (6.5 mgekg-‘, im) and anesthetized with ketamine hydrochloride (65 mgekg-‘, im). Carotid and caudal arteries were cannulated with polyethylene tubing (PE 190 and PE 10, respectively; the tip of the PE 190 catheter was tapered to fit snugly into the carotid artery) and the animals were heparinized (1000 Uekg- ‘, ia). Systemic vasodilation was obtained by infusing papaverine (4 mgm-‘) into the carotid artery until caudal artery pressure remained below 20 mm Hg. An infusion/withdrawal pump (Model 944; Harvard Apparatus, South Natick, MA) supported a 50-ml glass syringe (Yale; Becton-Dickinson, Rutherford, NJ) containing Microfil casting compound (Canton Bio-medical Products, Inc., Boulder, CO). Microfil was infused through the carotid cannula until an inflation pressure of 100 mm Hg was achieved in the caudal artery. Inflation pressure was maintained until the Microfil polymerized. During infusion, the angle between the lower leg and foot approximated 90”. Within 24-72 hr postcasting, bilateral EDL (n = 10) and SOL (n = 8) muscles were exposed and, with the foot angle at 90”, the dimensions of each muscle were measured in situ. Next, the muscles were dissected free and the casted feed arteries and draining veins (located exterior to the muscle) were cut leaving 2- to 3-mm lengths of the casts protruding from the muscle; these were used to identify locations of vascular supply on the muscle surface and arteriolar and venular networks within the muscle. The tendons that extended beyond the length of the muscle were removed. Each muscle was weighed, pinned to a wooden splint at in situ length, and stored at 1°C in a 2-ml cryovial (Nalgene, Rochester, NY) filled with 0.9% NaCl. Muscle fiber lengths were calculated by multiplying muscle lengths from this study (Table 1) times the fiber length/muscle length ratios of 0.595 (SOL) and 0.399 (EDL) (Segal and Faulkner, 1985). Muscle cross-sectional area (mm”) was calculated as [muscle weight (mg)] x [fiber length (mm) x muscle density
194
WILLIAMS
AND
SEGAL
(mg.mm-‘)I ‘. The value used for muscle density was 1.06 mg.mmm3 (Mendez and Keys, 1960). Identifying and classifying regions and vessels. The surface of each muscle was inspected to determine the locations of vascular supply. Casts of feed arteries and draining veins were consistently found in common areas for both the EDL and SOL muscles. These vessel groupings were classified into regions according to their location on the muscle surface and relative to the proximal tendon of the muscle. To distinguish arteriolar from venular networks within each muscle, the vessel casts external to the muscle surface were identified as arteries or veins using diameter measurements. Casts of these feed arteries and draining veins were imaged with a stereo microscope u = loo-mm objective; Model SV8; Zeiss, Germany) and a closed circuit television camera (Model HV-731U; Hitachi Denshi, Japan) coupled to a video monitor (Model PVM-122; Sony, Japan). Using digital calipers (Model 722; Starrett, Athol, MA) held against the video monitor (final magnification, 267 x ), cast diameter was measured three times and averaged. The calipers were calibrated against the video image of a stage micrometer (100 x 0.01 = 1 mm Graticules Ltd., Tonbridge Kent, England); percentage error of repeat measurements was 1.5%. In vivo observations of rat EDL and SOL muscles were used to verify the locations of vascular supply and the identity of feed arteries and draining veins. Microvascular architecture. When diameter measurements of the casted feed arteries and draining veins were complete, each muscle was dehydrated progressively in 25, 50, 75, 95, and 100% ethyl alcohol (24 hr in each solution), and then cleared by immersion in methyl salicylate (Sigma Chemical Co., St. Louis, MO). Muscles were stored in the latter solution. Negligible changes in muscle dimensions were observed during dehydration and clearing. Microvascular casts, now visible within each muscle, were magnified (42 X) using a stereo microscope (Model SZ-6 Plus; Bausch & Lomb, Rochester, NY). Representative networks were selected based on visibility and reproducibility between muscles. Starting from each region where arteries and veins entered and exited the muscle, primary (lo), secondary (2”), and tertiary (3”) branch orders of arterioles and venules were identified along the casted microvascular networks within the muscle. The 1” branches originated from the feed arteries and draining veins at the surface of the muscle, 2” branches arose from 1” microvessels, and 3” branches arose from 2” microvessels (Hutchins and Darnell, 1974). Each microvascular cast was imaged by video microscopy and a transparency was placed on the monitor face. Gross features of the network were traced and greater detail was added by hand, concomitant with inspection through the microscope. Segment lengths (SL) were measured along each of the selected microvessels using an eyepiece reticle calibrated to the stage micrometer. The value for each SL was recorded onto the network tracing. SL was defined for 1” microvessels as the distance between the midpoints of successive 2” branches, for 2” microvessels as the distance between successive 3” branches, and for 3” microvessels as the distance between successive 4” branches. During measurements of SL, the microscope was refocused continuously to follow branches that angled out of the optical plane and the muscle was rotated as necessary.
MICROVESSELS
IN
RED
TABLE WEIGHTS,
DIMENSIONS,
AND
WHITE
195
MUSCLE
1
VESSEL DIAMETERS, AND LOCATION OF VESSEL REGIONS DIGITORUM LONGUS (EDL) AND SOLEUS (SOL) MUSCLES
FOR RAT EXTENSOR
SOL
EDL Weight (mg) Length (mm) Thickness (anterior to posterior, mm) Fiber length (mm) Cross-sectional area (mm*)
107.3 27.3 1.8 10.9 9.3
Feed artery diameter (pm) Paired vein diameter (pm)
118.2 % 7.6’(15) 254.8 -c 35.1 (4)
Distance from proximal tendon (mm) Mid region Proximal region Distal region
2 3.1 ‘- 0.6’ k 0.1’ 2 0.2d 2 0.3
11.6 ? 0.8 8.1 f 0.6 NAh
100.0 24.1 2.2 14.4 6.6
-r? ” * k
4.3 0.6 0.1 0.3 0.3
162.9 -c 6.ZR (7) 218.9 f 19.0 (7) 10.7 ? 0.2 NA 17.4 ? 0.8
Note. Values are means f SE (n vessels). Artery and vein diameters are calculated for regions pooled within each muscle. y Calculated (see Materials and Methods). h EDL > SOL, P = 0.001. ’ EDL < SOL, P = 0.025. ’ EDL < SOL, P < 0.001. ’ EDL > SOL, P < 0.001. ’ EDL, artery < vein, P = 0.032. R SOL, artery < vein, P = 0.026. h NA, not applicable.
Branch frequency (BR, the number of branches per microvessel) was counted along each l”, 2”, and 3” microvessel of the network. Because tightly organized microvessels within the SOL muscles limited vision into the muscle, 3” branches could not be measured in 37% of these networks. For interruptions in the Microfil cast (14% of SOL, 9% of EDL networks), SL was measured in the visible region and BR was not counted for that microvessel. Neither SL nor BR was measured beyond 3” microvessels due to incomplete filling of the smaller branches. Data analysis. Unpaired t tests were performed to compare means using MINITAB Statistical Software (Minitab, Inc., State College, PA). Frequency distributions of SL and BR were consistently right skewed. Therefore, to approximate normal distributions, SL data were log transformed and BR data were square root transformed. Analysis of variance was performed on the transformed data according to Statistical Analysis System’s General Linear Models Procedure (SAS Institute, Inc., Cary, NC) with values weighted by number of animals. Post hoc comparisons of mean SL and BR between muscles, vascular regions, vessel type (arterioles and venules), and vessel order were obtained by constructing linear contrasts of the least squares means. To account for the possible effect of muscle dimensions on microvessel SL, each SL value was also expressed as percentage of muscle fiber and total muscle lengths. These percentages were log transformed and analyzed as above. All values are reported as means + SE. Findings were considered significant with P d 0.05.
196
WILLIAMS
AND
SEGAL
Proximal EDL
f f
Distal
1. Microvascular patterns and muscle fiber orientation viewed from the posterior surface of the soleus (SOL) and extensor digitorum longus (EDL) muscles. Muscle fibers (f) are illustrated by vertical lines; tendons (t) are depicted by darkened areas (after Close, 1964). Vascular supply regions are mid (Mso,) and distal (Ds,,J for the SOL muscle and mid (M,,,) and proximal (PEDL) for the EDL muscle. For simplicity, only arteriolar networks and one of the two M,,, arteries have been illustrated. FIG.
RESULTS Muscle weights and dimensions, feed artery and draining vein diameters, and locations of vascular regions are reported in Table 1. After clearing with methyl salicylate, mean (rt: SE) muscle weights were 133.2 + 6.5 mg for SOL and 134.7 + 6.3 mg for EDL muscles. Both muscles were heavier (P < 0.01) following the clearing procedure. Locations and Patterns of Vascular Supply
Figure 1 illustrates the locations of vascular regions supplying and representative microvascular networks within the EDL and SOL muscles. Venular networks were typically parallel with the arteriolar networks shown in Fig. 1. For reference, muscle fiber orientation (Close, 1964) is also pictured. By gross inspection, microvascular networks within the EDL muscle appeared more spacious compared to networks within the SOL muscle. Soleus muscle. Two main regions of vascular supply were identified for the SOL muscle. One region (mid soleus, MsoL ) originated from the posterior surface at the midbelly of the muscle. Vessels supplying this region (no fewer than two arteries and two veins) entered with the motor nerve and were oriented perpendicular to the longitudinal axis of the muscle. The 1” microvessels that originated were oriented toward the proximal and distal tendons (Fig. 1). The 2” at MsoL branches coursed through the muscle at ~45” angles relative to 1” microvessels. Tertiary branches were oriented in all directions from the 2” microvessels.
MICROVESSELS
-E-
IN RED
AND
WHITE
197
MUSCLE
600
2 Jz F 4
400
E E E *
200
A
v
MEDL
A V PEDL
A
v
A V &SOL
2. Mean (k SE) segment lengths for arterioles (A) and venules (V) in extensor digitorum longus (EDL) and soleus (SOL) muscles. Vascular regions are mid (M), proximal (P), and distal (D). Vessel orders are primary (l”), secondary (2”), and tertiary (3”). (n’s are reported in Table 2). a, M EDL 1” arteriole > M,,, 1” arteriole (P = 0.0002). b, M,,, 1” arteriole > P,,, 1” arteriole (P = 0.0001). c, M,,, 2” arteriole > MEoL 2” venule (P = 0.009). d, P,,, 2” arteriole > D,,, 2” arteriole (P = 0.03). See Table 2 for additional comparisons. FIG.
The second region of microvessels (distal soleus, DSoL) originated from the lateral border at the distal end of the SOL muscle. DsoL contained one paired artery and vein and the 1” microvessels arising from DsoL were perpendicular to the muscle axis. Branching at -90” from 1” microvessels, 2” arterioles and venules were oriented parallel to the longitudinal axis of the muscle. The 3” microvessels formed -45” angles with the 2” branches. Extensor digitorum longus muscle. Artery and vein pairs entered and exited the posterior side of the EDL at both the proximal (PEDL) and midbelly (Mum) sections of the muscle. These regions were distinguished by the orientation of their respective 1” microvessels. In MEDL, the 1” microvessels extended distally to the end of the muscle (Fig. 1). The 2” and 3” microvessels emanated from the respective parent vessels at 45-90” angles. The PEDL region, located between the proximal tendon and MEDL, contained one to three pairs of arteries and veins and appeared to supply the proximal one-third of the muscle. The 1” microvessels of the PEDL region followed a course perpendicular to the long axis of the muscle. Branch angles of 2” and 3” microvessels were similar to vessels of the corresponding order in MEDr,. Segment Length SL for arterioles and venules by region and order are illustrated in Fig. 2 for the EDL and SOL muscles. For 1” arterioles, SL of the MEDL region was longer (P < 0.001) than SL of 1” arterioles in all other regions. Other statistical differences are listed in the Fig. 2 legend for comparisons between muscles, vascular regions, and vessel type. The differences among orders within muscle, vascular region, and vessel type for the data shown in Fig. 2 are summarized in Table 2.
198
WILLIAMS
AND SEGAL
TABLE
2
COMPARISONS BETWEEN VESSEL ORDERS WITHIN REGION FOR SEGMENT LENGTH AND BRANCH MEASURED ALONG ARTERIOLES AND VENULES IN RAT EXTENSOR DIGITORUM L~NGUS (EDL)
(SOL)
FREQUENCY AND SOLEUS
MUSCLES
Segment length Region, vessel EDL muscle Mid, arteriole
Order
n
1”
9 58 93 3
2 3
Mid, venule
1” 2 3"
Proximal. arteriole
1” 2 3
Proximal. venule
1” 2 3
SOL Muscle Mid, arteriole Mid, venule
1”
Distal, venule
1”
~/O.OOl Y/O.001 3”/0.001 2”/0.001 3”/0.001
1" = 2'/0.086 1" i= Y/O.054 2" = 3"/0.426 1" > T/O.016
4 19 8
2 3
> > > > >
5
1”
2 3
1” 1” 2” 1” 1”
2" = Y/O.498
16
1”
Comparison/ P value
23 9 37 39 3 8 9
2 3 2 3"
Distal, arteriole
10
Branch frequency
11
4 11 7 3
11 I
1” > Y/O.017 1” > 3”/0.001 2" > 3"/0.009 1" = T/O.723
1” = 3”/0.157 2" = 3"/0.842
Order
n
1”
8 52 89
2 3
1”
1
2 3"
9 22 8 33 39 3 8 9
1” 2 3"
1” 2 3"
1”
2" = Y/O.462
2 3" 1" 2 3"
1” > 2”/0.014
1”
1" > ~/0.002 2" = Y/O.184
2 3
1” > Y/O.030
1” = 2”/0.718 1” i= Y/O.770 2"a
Y/O.996
1” 2 3
3 15 10 4 14 8 2
10 4 3
11 7
Comparison/ P value 1” > 2”/0.001 1” > 3”/0.001 2" > 3"/0.006 1" = 2"/0.215
1” > Y/O.001 2" > 3"/0.001
1” > 2”/0.003 1” > 3”/0.001 2" > 3"/0.027
1” = 2”/0.162 1” > 3”/0.009 2" i= Y/O.094
1” > T/O.004 1” > 3”/0.001 2" > 3"/0.006
1” > 2”/0.001 1” > 3”/0.001 2" > 3"/0.003 1" > 2°/0.017
1” > 2”z 1” > 1” >
3”/0.001 3"/0.057
~/O.OOl 3”/0.001
2" = Y/O.069
Note. Vessel orders are primary (lo), secondary (2)“, and tertiary (3”). Summary data (means 2 SE) are pictured in Fig. 2 for segment length and Fig. 3 for branch frequency.
SL ranged from 20 to 1823 pm in the SOL muscle and from 10 to 2705 pm in the EDL muscle. For both muscles, there was a consistent trend for SL to decrease as vessel order increased. Microvessel SL tended to be more uniform within the SOL as compared to the EDL muscle (Fig. 2 and Table 2). When normalized to muscle fiber length, additional SL differences between the two muscles were apparent (Table 3). For the EDL muscle, SL was longer in 2 and 3” arterioles of the PEDL region and in 1” venules of MEDL compared to the corresponding order of microvessels within the SOL muscle. These differences substantiated the visual impression described earlier of more tightly packed microvascular networks within the SOL muscle compared to the EDL muscle. Branch Frequency
Figure 3 illustrates BR by muscle, region, vessel type, and order. Comparisons were the same as for SL and findings are listed in Fig. 3 and Table 2. In both
MICROVESSELS
IN
RED
AND
TABLE
WHITE
199
MUSCLE
2
COMPARISONS BETWEEN VESSEL ORDERS WITHIN REGION FORSEGMENT LENGTH AND BRANCH FREQUENCY MEASURED ALONG ARTERIOLES AND VENULES IN RAT EXTENSOR DIGITORUM LONGUS(EDL) AND SOLEUS
(SOL) MUSCLES
Region, vessel EDL muscle Mid, arteriole Mid. venule Proximal, arteriole Proximal, venule
SOL Muscle Mid, arteriole Mid, venule Distal, arteriole Distal, venule
Order
Segment length Comparison/ n P value -
Branch frequency Order
n
Comparison/ P value
1” 2 3 1” 2 3 1” 2” 3 1” 2 3
9 58 93 3 10 23 9 37 39 3 8 9
1” 1” 2” 1” 1”
> > > > > 2” a 1” > 1” > 2” > 1” = 1” = 2” =
2”/0.001 Y/O.001 3”/0.001 2”/0.001 Y/O.001 3”/0.498 T/O.017 Y/O.001 3”/0.009 2”/0.723 3”/0.157 3”/0.842
1” 2 3 1” 2 3 1” 2” 3” 1” 2” 3
8 52 89 1 9 22 8 33 39 3 8 9
1” 1” 2” 1” 1” 2” 1” 1” 2” 1” 1” 2”
> > > = > > > > > = > =
T/O.001 3”/0.001 3”/0.006 2”/0.215 3”/0.001 3”/0.001 2”/0.003 Y/O.001 Y/O.027 2”/0.162 3”/0.009 Y/O.094
1” 2” 3 1” 2 3” 1” 2 3” 1” 2 3
5 16 11 4 19 8 4 11 7 3 11 7
1” 1” 2” 1” 1” 2” 1” 1” 2” 1” 1” 2”
= 2”/0.086 zz 3”/0.054 = Y/O.426 > 2”/0.016 > 3”/0.030 = 3”/0.462 > 2”/0.014 > 3”/0.002 = Y/O.184 = 2”/0.718 = Y/O.770 = Y/O.996
1” 2 3” 1” 2” 3” 1” 2” 3” 1” 2” 3”
3 1.5 10 4 14 8 2 10 4 3 11 7
1” 1” 2” 1” 1” 2” 1” 1” 2” 1” 1” 2”
> > > > > > > > = > > =
T/O.004 3”/0.001 3”/0.006 2°/o.001 3”/0.001 Y/O.003 T/O.017 3”/0.001 3”/0.057 T/O.001 3”/0.001 Y/O.069
Note. Vessel orders are primary (l”), secondary (2)O, and tertiary (3”). Summary data (means 2 SE) are pictured in Fig. 2 for segment length and Fig. 3 for branch frequency.
muscles, BR decreased as vessel order increased (Table 2). The ranges of BR were similar across muscles, regions, vessel type, and order. DISCUSSION This study has documented differences in microvascular architecture between EDL and SOL skeletal muscles of the rat. Branch frequency consistently decreased as vessel order increased in both muscles. Differences in segment lengths between muscles consistently showed shorter vascular segments in the SOL muscle compared to the EDL muscle, indicating more compact microvascular networks in the SOL muscle. Anatomical Considerations Differences in vascular networks between red and white skeletal muscle have been of interest since first noted by Ranvier (1874). Greater density of pre- and
200
WILLIAMS
A
v
A
AND
V
SEGAL
A
V
A
V
MEDL PEEI DSOL FIG. 3. Mean (2 SE) branch frequencies for arterioles (A) and venules (V) in extensor digitorum longus (EDL) and soleus (SOL) muscles. Vascular regions are mid (M), proximal (P), and distal (D). Vessel orders are primary (lo), secondary (2”), and tertiary (3”). (n’s are reported in Table 2). a, M EDL2” arteriole < MEoL 2” venule (P = 0.02). b, Ma,, 3” arteriole > M,,, 3” venule (P = 0.0001). c, Mm. 2” venule > MsoL 2” venule (P = 0.002). d, M,,, 3” venule < P,,, 3” venule (P = 0.01). e, MS,,. 2” arteriole > Mso, , 2” venule (P = 0.01). See Table 2 for additional comparisons.
postcapillary blood vessels in red compared to white muscle has been described (Myrhage and Eriksson, 1980); however, these microvessels have not been quantified. The present study has demonstrated longer SL along 1” microvessels in M EDL compared to PEp, and compared to both regions of the SOL muscle. Capillary/fiber ratios measured in the adult rat (1.15: 1 for EDL and 2.3 : 1 for SOL; Hudlicka, 1985) have indicated greater microvascular density in the SOL muscle compared to the EDL muscle. The results here demonstrate that greater density of the microcirculation in the SOL muscle is not limited to capillaries. Rather, components of the microvascular network (arterioles, venules, and capillaries) are organized more compactly within the SOL muscle relative to the EDL muscle. These data are consistent with greater oxygen requirements in the SOL muscle compared to the EDL muscle. The magnitude of differences between SOL and EDL muscle and muscle fiber lengths (Table 1) did not account for the observed differences in SL. For example, the EDL muscle was only 12% longer than the SOL muscle, yet SL of the 1” arteriole was 51% longer in the EDL (mid region) compared to the SOL (mid region). Normalizing SL to muscle fiber length strengthened the argument of relatively shorter microvessel segments (more compact microvascular networks) in the SOL muscle compared to the EDL muscle (Table 3). Dissimilar microvascular networks were also observed within the EDL muscle. The 1” arteriole of the P,,, region had shorter SL compared to that of the MEDL region. These microvessels also differed in orientation with respect to the muscle axis (Figs. 1 and 2). Shorter SL in one region of the EDL versus another may be explained by inhibited growth due to “sensing” anatomical boundaries such
MICROVESSELS
IN RED AND
WHITE
MUSCLE
201
as other vessels, fiber orientation, or the muscle surface. A similar phenomenon may limit the length of 2” and 3” branches, which were diagonal relative to fiber orientation and displayed shorter SL and lower BR relative to 1” microvessels (Figs. 1, 2, and 3). The mechanisms determining microvessel SL remain to be established. Little is known about adaptation of pre- and postcapillary microvessels within skeletal muscles when subjected to experimental interventions, such as exercise or deconditioning. The arteriolar and venular segments measured in this study represent basic units of the microvascular networks in EDL and SOL muscles. Studying responses of these units to endurance training, muscle hypertrophy and/or atrophy relative to the control data reported here could prove fruitful in understanding growth and regression of the microcirculation. Structural similarities and differences between networks, as quantified by BR and SL, may also provide a framework for interpreting differences in physiological properties between the EDL and SOL muscles.
Physiological
Considerations
Blood flow is not distributed uniformly between muscles, among regions of muscles, or within the microcirculation either at rest or during exercise (Laughlin and Armstrong, 1982; Duling and Damon, 1987). Our present findings indicate an anatomical basis for blood flow heterogeneity. In the EDL muscle, for example, variations in blood flow could result from more compact vascular architecture in one region (PEDL) compared to the second region (MEDL), attributable to shorter SL of 1” arterioles in PEDL. Further study is required to relate microvascular network heterogeneity and blood flow distribution within a muscle. In the rat, blood flow is distributed in accord with muscle fiber type (Laughlin and Armstrong, 1982; Mackie and Terjung, 1983). Studies of mixed muscles of rat, rabbit, and human have found more capillaries associated with oxidative compared to glycolytic muscle fibers (Romanul, 1965; Gray and Renkin, 1978). However, maximal muscle blood flow was independent of capillarity (Maxwell et al., 1980). Thus, variations in the microvascular network proximal and distal to capillaries may explain the relationships between muscle blood flow and fiber type (Laughlin and Armstrong, 1982; Mackie and Terjung, 1983). Muscle fibers in the rat SOL muscle are predominantly SO (87%) with the remaining fibers classified as FOG (13%; Armstrong and Phelps, 1984). In contrast, the EDL muscle is composed of approximately equal portions of FOG and FG fibers, with only 2% SO fibers (Armstrong and Phelps, 1984). The different structural dimensions of pre- and postcapillary microvessels reported here provide an explanation for differences in blood flow between the EDL and SOL muscles of the rat (Armstrong and Laughlin, 1985) that is independent of capillarity. More compact vascular networks within the SOL muscle are consistent with greater blood flow conductance in the SOL muscle as compared to the EDL muscle. Thus varied flow between muscles could be due to the differing vascular network structures rather than to variations in vasomotor tone, resulting in a simple form of control with minimal energy requirement. For example, passive (structural) control of flow via network architecture would selectively allow flow
202
WILLIAMS
AND
SEGAL
to muscle fibers having high oxygen requirements. Measurements of the vascular structures between and within other muscles differing in oxidative metabolism would further test this hypothesis. The coordination of a hyperemic response within microvascular networks can match oxygen delivery with local demand in striated muscle (Segal, 1991). The distance over which vasomotor responses are conducted along arterioles (Segal, 1991) may, in part, explain the greater flow magnitude developed in red versus white muscle during muscle contractions (Folkow and Halicka, 1968; Hilton et al., 1970; Reis and Wooten, 1976; Laughlin and Armstrong, 1982; Mackie and Terjung, 1983). For example, with a length constant of =2 mm (Segal, 1991), a conducted response initiated at the origin of a distal 2” arteriole and spreading upstream along its parent 1” arteriole could affect flow into almost twice the number of 2” branches in the SOL muscle (0.37 mm, mean SL) compared to similar microvessels in the EDL muscle (0.69 mm, mean SL; Fig. 2). Differences in SL would not explain dissimilar hyperemic responses among 2” and 3” microvessels in SOL and EDL muscles since SL of these microvessels were similar among the respective regions of both muscles (Fig. 2). However, such similarities in architecture in 2” and 3” microvessels may not be critical if 1” arterioles are the major sites of blood flow control in the EDL and SOL muscles. Further work is needed to clarify these relationships. Limitations
and Nomenclature
In the present study, casting material was dispersed uniformly throughout the muscles. No empty regions were observed in the muscles examined despite incomplete filling of capillaries (documented previously by Gogh, 1918). In practice, capillaries devoid of casting material allowed us to visualize networks through the muscle and to measure casts of the larger microvessels. By starting at each vascular region on the muscle surface and tracing inward, we identified intact networks located similarly across muscles. We have assumed that the networks sampled were representative of the microvessel population within the SOL and EDL muscles. With the above visual assurances, we were confident in our sampling procedure. Nevertheless, these casts may not represent entirely the in vivo microvascular networks of the SOL and EDL muscles. Some secondary and tertiary branches may not have filled completely due to severe branch angles, small vessel diameters, surface tension, different filling patterns between the two muscles, or other structural constraints. Recognizing this limitation, we did not use the Horton-Strahler ordering system, which, by definition, starts with the smallest and works to the largest branches (Fenton and Zweifach, 1981). In the present study, ordering the microvascular network from large to small microvessels (Hutchins and Darnell, 1974) was consistent with methods used for lung (Weibel, 1984), brain (Coleman and Riesen, 1968), bat wing (Wiedeman, 1968), tenuissimus muscle (Eriksson and Myrhage, 1972), intestine (Bohlen, 1983), and biceps femoris muscle (Myrhage and Eriksson, 1980). In conclusion, we have documented differences in microvascular architecture between SOL and EDL muscles. When different, segment lengths were shorter in microvessels of the SOL compared to the EDL muscles, indicating more compact pre- and postcapillary networks in the SOL muscle. In both muscles, branch frequency decreased from primary to tertiary microvessels and was similar within
MICROVESSELS
IN
RED
AND
WHITE
MUSCLE
203
each order between muscles. The proximal EDL region contained more compact microvessels compared to the microvascular network that supplied the distal twothirds of the muscle. Differences in microvascular architecture between the SOL and EDL muscles reported here provide one explanation for the dissimilar values for blood flow between these muscles both at rest and during exercise.
ACKNOWLEDGMENTS The authors thank H. B. Chew for vascular casting, Joseph L. Loomis and Douglas A. Johnson for expert technical assistance, and Janice Derr and Elvessa Aragon for statistical assistance. This research was supported by NASA Grant NAGW 1196, a Grant-In-Aid from the Pennsylvania Affiliate of the American Heart Association, and NIH Grant HL41026. D. A. Williams was the recipient of the Mrs. A. Robert No11 Fellowship in Physiology.
REFERENCES ARMSTRONG, R. B., AND PHELPS, R. 0. (1984). Muscle fiber type composition of the rat hindlimb. Am. J. Anat. 171,259-272. ARMSTRONG, R. B., AND LAUGHLIN, M. H. (1985). Rat muscle blood flows during high-speed locomotion. J. Appl. Physiol. 59, 1322-1328. BOHLEN, H. G. (1983). Intestinal microvascular adaptation during maturation of spontaneously hypertensive rats. Hypertension 5, 139-745. BURKE, R. E. (1981). Motor units: Anatomy, physiology, and functional organization. In “Handbook
of Physiology, the Nervous System” (V. B. Brooks, Ed.), Vol. 2, pp. 345-422. American Physiology Society, Bethesda, MD. CLOSE, R. (1964). Dynamic properties of fast and slow skeletal muscles of the rat during development. J. Physiol.
173,74-95.
CLOSE, R. (1972). Dynamic properties of mammalian skeletal muscles. Physiol. Rev. 52, 129-197. COLEMAN, P. D., AND RIESEN, A. H. (1968). Environmental effects on cortical dendritic fields. I. Rearing in the dark. J. Anat. 102,363-374. DULING, B. R., AND DAMON, D. H. (1987). An examination of the measurement of flow heterogeneity in striated muscle. Circ. Res. 60, 1-13. ERIKSSON, E., AND MYRHAGE, R. (1972). Microvascular dimensions and blood flow in skeletal muscle. Acta Physiol. Stand. 86, 211-222. FENTON, B. M., AND ZWEIFACH, B. W. (1981). Microcirculatory model relating geometrical variation to changes in pressure and flow rate. Ann. Biomed. Eng. 9, 303-321. FOLKOW, B., AND HALICKA, H. (1968). A comparison between ‘red’ and ‘white’ muscle with respect to blood supply, capillary surface area and oxygen uptake during rest and exercise. Microvusc. Res. 1, 1-14. GRAY, S. D., AND RENKIN, E. M. (1978). Microvascular supply in relation to fiber metabolic type in mixed skeletal muscles of rabbits. Microvasc. Res. 16, 406-425. HILTON, S. M., JEFFRIES,M. G., AND VRBOV~, G. (1970). Functional specializations of the vascular bed of soleus. J. Physiol. 206, 543-562. HUDLICK~, 0. (1985). Development and adaptability of microvasculature in skeletal muscle. Z. Exp. Biol. 115,215-228. HUTCHINS, P. M., AND DARNELL, A.
E. (1974). Observation of a decreased number of small arterioles in spontaneously hypertensive rats. Circ. Res. (Suppl. Z) 34-35, I-161-1-165. KROGH, A. (1918). The number and distribution of capillaries in muscles with calculations of the oxygen pressure head necessary for supplying the tissue. J. Physiol. 52, 409-415. LAUGHLIN, M. H., AND ARMSTRONG, R. B. (1982). Muscular blood flow distribution patterns as a function of running speed in rats. Am. J. Physiol. 243, H296-H306. MACKIE, B. G., AND TERJUNG, R. L. (1983). Blood flow to different skeletal muscle fiber types during contraction. Am. J. Physiol. 245, H265-H275.
204
WILLIAMS
AND
SEGAL
L. C., WHITE, T. P., AND FAULKNER, J. A. (1980). Oxidative capacity, blood flow, and capillarity of skeletal muscles. J. Appl. Physiol. 49, 627-633. MBNDEZ, J., AND KEYS, A. (1960). Density and composition of mammalian muscle. Metabolism 9, 184-188. MYRHAGE, R., AND ERIKSSON, E. (1980). Vascular arrangements in hind limb muscles of the cat. J. Anat. Wl, l-17. POTTER, R. F., AND GROOM, A. C. (1983). Capillary diameter and geometry in cardiac and skeletal muscle studied by means of corrosion casts. Microvasc. Res. 25, 68-84. RANVIER, P. L. (1874). Note sur les vaisseaux sanguins et la circulation dans les muscles rouges. Arch. MAXWELL,
Physiol. REIS, D.
Norm. Path. 6, 446-450. J., AND WOOTEN, G. F.
(1976). Blood flow in red and white muscle: Relationship to metabolism, development and behavior. Prog. Brain Rex 44, 385-401. ROMANUL, F. C. (1965). Capillary supply and metabolism of muscle fibers. Arch. Neural. 12, 497509. SEGAL,
S. S. (1991). Microvascular recruitment in hamster striated muscle: Role for conducted vasodilation. Am. J. Physiol. 261, H181-H189. SEGAL, S. S., AND FAULKNER, J. A. (1985). Temperature-dependent physiological stability of rat skeletal muscle in vitro. Am. J. Physiol. 248, C265-C270. STINGL, J. (1969). Arrangement of the vascular bed in the skeletal muscles of the rabbit. Folia Morphologica WEIBEL, E.
II,
257-264.
R. (1984). “The Pathway for Oxygen: Structure and Function in the Mammalian Respiratory System.” Harvard Univ. Press, Cambridge, MA. WIEDEMAN, M. (1968). Blood flow through terminal arterial vessels after denervation of the bat wing. Circ.
Res. 22, 83-89.
43, 192-204 (1992)
RESEARCH
Microvascular
Architecture in Rat Soleus and Extensor Digitorum Longus Muscles
DONNA A. WILLIAMS’ Laboratory
for Human
Performance University Park, Received
AND STEVEN S. SEGAL~ Research, Pennsylvania Pennsylvania 16802
State University,
June 14, 1990
Microvascular architecture was investigated in the slow-twitch soleus (SOL) and fasttwitch extensor digitorum longus (EDL) muscles. Rats (n = 5) were anesthetized and papaverine was infused into a carotid artery cannula to induce vasodilation. Microfil casting compound was then infused at an inflation pressure (caudal artery) of 100 mm Hg. Bilateral SOL and EDL muscles were excised 24-72 hr postcasting, dehydrated in ethanol, and cleared in methyl salicylate. Branch frequencies (BR) and segment lengths (SL) of intramuscular arterioles and venules were quantified along primary (lo), secondary (Z’), and tertiary (3”) order microvessels using microscopy. In both muscles, BR decreased with increasing vessel order. Regional differences in network organization were observed within the EDL muscle. SL of 1” arterioles was 47% shorter in the SOL muscle indicating more compact microvascular networks compared to the EDL muscle. These findings provide a structural basis for reported differences in blood flow between the SOL and EDL muscles at rest and during exercise. 0 1992 Academtc Press, Inc.
INTRODUCTION Skeletal muscle blood flow differs during exercise between muscles and according to fiber type within individual muscles (Folkow and Halicka, 1968; Laughlin and Armstrong, 1982; Mackie and Terjung, 1983). Mild to moderate exercise is characterized by elevated blood flow to slow-twitch-oxidative (SO) fibers. As exercise intensity increases, blood flow increases to fast-twitch-oxidative-glycolytic (FOG) and, then, to fast-twitch-glycolytic (FG) fibers (Laughlin and Armstrong, 1982). These blood flow responses correlate with the order of muscle fiber recruitment during graded exercise (Burke, 1981). Soleus (SOL) and extensor digitorum longus (EDL) muscles of the rat differ in contractile properties (Close, 1964), fiber type (Armstrong and Phelps, 1984), and blood flow (Armstrong and Laughlin, 1985). The SOL muscle is a slow-twitch, postural muscle that relies primarily upon oxidative metabolism (Close, 1972). In contrast, the fast-twitch EDL muscle is recruited during rapid, high-intensity I Present address: Department Physiology, University of Missouri Medical School, Columbia, MO 65212. ’ To whom reprint requests should be addressed at 119 No11 Laboratory, The Pennsylvania State University, University Park, PA 16802. 192 0026-2X62/92 $3.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved Printed in U.S.A.
MICROVESSELS
IN
RED
AND
WHITE
MUSCLE
193
activity and relies more upon glycolytic metabolism for energy production (Close, 1972). Differences in blood flow between the SOL and EDL muscles, at rest and during exercise, have also been documented (Armstrong and Laughlin, 1985). Blood flow differences between muscles may be explained, in part, by dissimilar vasomotor tone along resistance vessels. Another explanation may relate to length and diameter variability within the respective capillary beds (Potter and Groom, 1983). Differences in microvascular architecture proximal and distal to the capillary bed may also affect blood flow (Laughlin and Armstrong, 1982; Mackie and Terjung, 1983; Stingl, 1969); however, the structure of these portions of the vascular network has not been examined in skeletal muscles involved in locomotion. The purpose of the present study was to characterize the architecture of pre- and postcapillary networks supplying rat EDL and SOL muscles. We hypothesized that the microvascular networks would differ between these two muscles.
MATERIALS
AND METHODS
All procedures performed on animals were approved by the Institutional Animal Care and Use Committee of The Pennsylvania State University. Vumdur casting. Cage-sedentary, female, Sprague-Dawley rats (16 weeks old; 234 ? 3 g; IZ = 5) were sedated with rhompun (6.5 mgekg-‘, im) and anesthetized with ketamine hydrochloride (65 mgekg-‘, im). Carotid and caudal arteries were cannulated with polyethylene tubing (PE 190 and PE 10, respectively; the tip of the PE 190 catheter was tapered to fit snugly into the carotid artery) and the animals were heparinized (1000 Uekg- ‘, ia). Systemic vasodilation was obtained by infusing papaverine (4 mgm-‘) into the carotid artery until caudal artery pressure remained below 20 mm Hg. An infusion/withdrawal pump (Model 944; Harvard Apparatus, South Natick, MA) supported a 50-ml glass syringe (Yale; Becton-Dickinson, Rutherford, NJ) containing Microfil casting compound (Canton Bio-medical Products, Inc., Boulder, CO). Microfil was infused through the carotid cannula until an inflation pressure of 100 mm Hg was achieved in the caudal artery. Inflation pressure was maintained until the Microfil polymerized. During infusion, the angle between the lower leg and foot approximated 90”. Within 24-72 hr postcasting, bilateral EDL (n = 10) and SOL (n = 8) muscles were exposed and, with the foot angle at 90”, the dimensions of each muscle were measured in situ. Next, the muscles were dissected free and the casted feed arteries and draining veins (located exterior to the muscle) were cut leaving 2- to 3-mm lengths of the casts protruding from the muscle; these were used to identify locations of vascular supply on the muscle surface and arteriolar and venular networks within the muscle. The tendons that extended beyond the length of the muscle were removed. Each muscle was weighed, pinned to a wooden splint at in situ length, and stored at 1°C in a 2-ml cryovial (Nalgene, Rochester, NY) filled with 0.9% NaCl. Muscle fiber lengths were calculated by multiplying muscle lengths from this study (Table 1) times the fiber length/muscle length ratios of 0.595 (SOL) and 0.399 (EDL) (Segal and Faulkner, 1985). Muscle cross-sectional area (mm”) was calculated as [muscle weight (mg)] x [fiber length (mm) x muscle density
194
WILLIAMS
AND
SEGAL
(mg.mm-‘)I ‘. The value used for muscle density was 1.06 mg.mmm3 (Mendez and Keys, 1960). Identifying and classifying regions and vessels. The surface of each muscle was inspected to determine the locations of vascular supply. Casts of feed arteries and draining veins were consistently found in common areas for both the EDL and SOL muscles. These vessel groupings were classified into regions according to their location on the muscle surface and relative to the proximal tendon of the muscle. To distinguish arteriolar from venular networks within each muscle, the vessel casts external to the muscle surface were identified as arteries or veins using diameter measurements. Casts of these feed arteries and draining veins were imaged with a stereo microscope u = loo-mm objective; Model SV8; Zeiss, Germany) and a closed circuit television camera (Model HV-731U; Hitachi Denshi, Japan) coupled to a video monitor (Model PVM-122; Sony, Japan). Using digital calipers (Model 722; Starrett, Athol, MA) held against the video monitor (final magnification, 267 x ), cast diameter was measured three times and averaged. The calipers were calibrated against the video image of a stage micrometer (100 x 0.01 = 1 mm Graticules Ltd., Tonbridge Kent, England); percentage error of repeat measurements was 1.5%. In vivo observations of rat EDL and SOL muscles were used to verify the locations of vascular supply and the identity of feed arteries and draining veins. Microvascular architecture. When diameter measurements of the casted feed arteries and draining veins were complete, each muscle was dehydrated progressively in 25, 50, 75, 95, and 100% ethyl alcohol (24 hr in each solution), and then cleared by immersion in methyl salicylate (Sigma Chemical Co., St. Louis, MO). Muscles were stored in the latter solution. Negligible changes in muscle dimensions were observed during dehydration and clearing. Microvascular casts, now visible within each muscle, were magnified (42 X) using a stereo microscope (Model SZ-6 Plus; Bausch & Lomb, Rochester, NY). Representative networks were selected based on visibility and reproducibility between muscles. Starting from each region where arteries and veins entered and exited the muscle, primary (lo), secondary (2”), and tertiary (3”) branch orders of arterioles and venules were identified along the casted microvascular networks within the muscle. The 1” branches originated from the feed arteries and draining veins at the surface of the muscle, 2” branches arose from 1” microvessels, and 3” branches arose from 2” microvessels (Hutchins and Darnell, 1974). Each microvascular cast was imaged by video microscopy and a transparency was placed on the monitor face. Gross features of the network were traced and greater detail was added by hand, concomitant with inspection through the microscope. Segment lengths (SL) were measured along each of the selected microvessels using an eyepiece reticle calibrated to the stage micrometer. The value for each SL was recorded onto the network tracing. SL was defined for 1” microvessels as the distance between the midpoints of successive 2” branches, for 2” microvessels as the distance between successive 3” branches, and for 3” microvessels as the distance between successive 4” branches. During measurements of SL, the microscope was refocused continuously to follow branches that angled out of the optical plane and the muscle was rotated as necessary.
MICROVESSELS
IN
RED
TABLE WEIGHTS,
DIMENSIONS,
AND
WHITE
195
MUSCLE
1
VESSEL DIAMETERS, AND LOCATION OF VESSEL REGIONS DIGITORUM LONGUS (EDL) AND SOLEUS (SOL) MUSCLES
FOR RAT EXTENSOR
SOL
EDL Weight (mg) Length (mm) Thickness (anterior to posterior, mm) Fiber length (mm) Cross-sectional area (mm*)
107.3 27.3 1.8 10.9 9.3
Feed artery diameter (pm) Paired vein diameter (pm)
118.2 % 7.6’(15) 254.8 -c 35.1 (4)
Distance from proximal tendon (mm) Mid region Proximal region Distal region
2 3.1 ‘- 0.6’ k 0.1’ 2 0.2d 2 0.3
11.6 ? 0.8 8.1 f 0.6 NAh
100.0 24.1 2.2 14.4 6.6
-r? ” * k
4.3 0.6 0.1 0.3 0.3
162.9 -c 6.ZR (7) 218.9 f 19.0 (7) 10.7 ? 0.2 NA 17.4 ? 0.8
Note. Values are means f SE (n vessels). Artery and vein diameters are calculated for regions pooled within each muscle. y Calculated (see Materials and Methods). h EDL > SOL, P = 0.001. ’ EDL < SOL, P = 0.025. ’ EDL < SOL, P < 0.001. ’ EDL > SOL, P < 0.001. ’ EDL, artery < vein, P = 0.032. R SOL, artery < vein, P = 0.026. h NA, not applicable.
Branch frequency (BR, the number of branches per microvessel) was counted along each l”, 2”, and 3” microvessel of the network. Because tightly organized microvessels within the SOL muscles limited vision into the muscle, 3” branches could not be measured in 37% of these networks. For interruptions in the Microfil cast (14% of SOL, 9% of EDL networks), SL was measured in the visible region and BR was not counted for that microvessel. Neither SL nor BR was measured beyond 3” microvessels due to incomplete filling of the smaller branches. Data analysis. Unpaired t tests were performed to compare means using MINITAB Statistical Software (Minitab, Inc., State College, PA). Frequency distributions of SL and BR were consistently right skewed. Therefore, to approximate normal distributions, SL data were log transformed and BR data were square root transformed. Analysis of variance was performed on the transformed data according to Statistical Analysis System’s General Linear Models Procedure (SAS Institute, Inc., Cary, NC) with values weighted by number of animals. Post hoc comparisons of mean SL and BR between muscles, vascular regions, vessel type (arterioles and venules), and vessel order were obtained by constructing linear contrasts of the least squares means. To account for the possible effect of muscle dimensions on microvessel SL, each SL value was also expressed as percentage of muscle fiber and total muscle lengths. These percentages were log transformed and analyzed as above. All values are reported as means + SE. Findings were considered significant with P d 0.05.
196
WILLIAMS
AND
SEGAL
Proximal EDL
f f
Distal
1. Microvascular patterns and muscle fiber orientation viewed from the posterior surface of the soleus (SOL) and extensor digitorum longus (EDL) muscles. Muscle fibers (f) are illustrated by vertical lines; tendons (t) are depicted by darkened areas (after Close, 1964). Vascular supply regions are mid (Mso,) and distal (Ds,,J for the SOL muscle and mid (M,,,) and proximal (PEDL) for the EDL muscle. For simplicity, only arteriolar networks and one of the two M,,, arteries have been illustrated. FIG.
RESULTS Muscle weights and dimensions, feed artery and draining vein diameters, and locations of vascular regions are reported in Table 1. After clearing with methyl salicylate, mean (rt: SE) muscle weights were 133.2 + 6.5 mg for SOL and 134.7 + 6.3 mg for EDL muscles. Both muscles were heavier (P < 0.01) following the clearing procedure. Locations and Patterns of Vascular Supply
Figure 1 illustrates the locations of vascular regions supplying and representative microvascular networks within the EDL and SOL muscles. Venular networks were typically parallel with the arteriolar networks shown in Fig. 1. For reference, muscle fiber orientation (Close, 1964) is also pictured. By gross inspection, microvascular networks within the EDL muscle appeared more spacious compared to networks within the SOL muscle. Soleus muscle. Two main regions of vascular supply were identified for the SOL muscle. One region (mid soleus, MsoL ) originated from the posterior surface at the midbelly of the muscle. Vessels supplying this region (no fewer than two arteries and two veins) entered with the motor nerve and were oriented perpendicular to the longitudinal axis of the muscle. The 1” microvessels that originated were oriented toward the proximal and distal tendons (Fig. 1). The 2” at MsoL branches coursed through the muscle at ~45” angles relative to 1” microvessels. Tertiary branches were oriented in all directions from the 2” microvessels.
MICROVESSELS
-E-
IN RED
AND
WHITE
197
MUSCLE
600
2 Jz F 4
400
E E E *
200
A
v
MEDL
A V PEDL
A
v
A V &SOL
2. Mean (k SE) segment lengths for arterioles (A) and venules (V) in extensor digitorum longus (EDL) and soleus (SOL) muscles. Vascular regions are mid (M), proximal (P), and distal (D). Vessel orders are primary (l”), secondary (2”), and tertiary (3”). (n’s are reported in Table 2). a, M EDL 1” arteriole > M,,, 1” arteriole (P = 0.0002). b, M,,, 1” arteriole > P,,, 1” arteriole (P = 0.0001). c, M,,, 2” arteriole > MEoL 2” venule (P = 0.009). d, P,,, 2” arteriole > D,,, 2” arteriole (P = 0.03). See Table 2 for additional comparisons. FIG.
The second region of microvessels (distal soleus, DSoL) originated from the lateral border at the distal end of the SOL muscle. DsoL contained one paired artery and vein and the 1” microvessels arising from DsoL were perpendicular to the muscle axis. Branching at -90” from 1” microvessels, 2” arterioles and venules were oriented parallel to the longitudinal axis of the muscle. The 3” microvessels formed -45” angles with the 2” branches. Extensor digitorum longus muscle. Artery and vein pairs entered and exited the posterior side of the EDL at both the proximal (PEDL) and midbelly (Mum) sections of the muscle. These regions were distinguished by the orientation of their respective 1” microvessels. In MEDL, the 1” microvessels extended distally to the end of the muscle (Fig. 1). The 2” and 3” microvessels emanated from the respective parent vessels at 45-90” angles. The PEDL region, located between the proximal tendon and MEDL, contained one to three pairs of arteries and veins and appeared to supply the proximal one-third of the muscle. The 1” microvessels of the PEDL region followed a course perpendicular to the long axis of the muscle. Branch angles of 2” and 3” microvessels were similar to vessels of the corresponding order in MEDr,. Segment Length SL for arterioles and venules by region and order are illustrated in Fig. 2 for the EDL and SOL muscles. For 1” arterioles, SL of the MEDL region was longer (P < 0.001) than SL of 1” arterioles in all other regions. Other statistical differences are listed in the Fig. 2 legend for comparisons between muscles, vascular regions, and vessel type. The differences among orders within muscle, vascular region, and vessel type for the data shown in Fig. 2 are summarized in Table 2.
198
WILLIAMS
AND SEGAL
TABLE
2
COMPARISONS BETWEEN VESSEL ORDERS WITHIN REGION FOR SEGMENT LENGTH AND BRANCH MEASURED ALONG ARTERIOLES AND VENULES IN RAT EXTENSOR DIGITORUM L~NGUS (EDL)
(SOL)
FREQUENCY AND SOLEUS
MUSCLES
Segment length Region, vessel EDL muscle Mid, arteriole
Order
n
1”
9 58 93 3
2 3
Mid, venule
1” 2 3"
Proximal. arteriole
1” 2 3
Proximal. venule
1” 2 3
SOL Muscle Mid, arteriole Mid, venule
1”
Distal, venule
1”
~/O.OOl Y/O.001 3”/0.001 2”/0.001 3”/0.001
1" = 2'/0.086 1" i= Y/O.054 2" = 3"/0.426 1" > T/O.016
4 19 8
2 3
> > > > >
5
1”
2 3
1” 1” 2” 1” 1”
2" = Y/O.498
16
1”
Comparison/ P value
23 9 37 39 3 8 9
2 3 2 3"
Distal, arteriole
10
Branch frequency
11
4 11 7 3
11 I
1” > Y/O.017 1” > 3”/0.001 2" > 3"/0.009 1" = T/O.723
1” = 3”/0.157 2" = 3"/0.842
Order
n
1”
8 52 89
2 3
1”
1
2 3"
9 22 8 33 39 3 8 9
1” 2 3"
1” 2 3"
1”
2" = Y/O.462
2 3" 1" 2 3"
1” > 2”/0.014
1”
1" > ~/0.002 2" = Y/O.184
2 3
1” > Y/O.030
1” = 2”/0.718 1” i= Y/O.770 2"a
Y/O.996
1” 2 3
3 15 10 4 14 8 2
10 4 3
11 7
Comparison/ P value 1” > 2”/0.001 1” > 3”/0.001 2" > 3"/0.006 1" = 2"/0.215
1” > Y/O.001 2" > 3"/0.001
1” > 2”/0.003 1” > 3”/0.001 2" > 3"/0.027
1” = 2”/0.162 1” > 3”/0.009 2" i= Y/O.094
1” > T/O.004 1” > 3”/0.001 2" > 3"/0.006
1” > 2”/0.001 1” > 3”/0.001 2" > 3"/0.003 1" > 2°/0.017
1” > 2”z 1” > 1” >
3”/0.001 3"/0.057
~/O.OOl 3”/0.001
2" = Y/O.069
Note. Vessel orders are primary (lo), secondary (2)“, and tertiary (3”). Summary data (means 2 SE) are pictured in Fig. 2 for segment length and Fig. 3 for branch frequency.
SL ranged from 20 to 1823 pm in the SOL muscle and from 10 to 2705 pm in the EDL muscle. For both muscles, there was a consistent trend for SL to decrease as vessel order increased. Microvessel SL tended to be more uniform within the SOL as compared to the EDL muscle (Fig. 2 and Table 2). When normalized to muscle fiber length, additional SL differences between the two muscles were apparent (Table 3). For the EDL muscle, SL was longer in 2 and 3” arterioles of the PEDL region and in 1” venules of MEDL compared to the corresponding order of microvessels within the SOL muscle. These differences substantiated the visual impression described earlier of more tightly packed microvascular networks within the SOL muscle compared to the EDL muscle. Branch Frequency
Figure 3 illustrates BR by muscle, region, vessel type, and order. Comparisons were the same as for SL and findings are listed in Fig. 3 and Table 2. In both
MICROVESSELS
IN
RED
AND
TABLE
WHITE
199
MUSCLE
2
COMPARISONS BETWEEN VESSEL ORDERS WITHIN REGION FORSEGMENT LENGTH AND BRANCH FREQUENCY MEASURED ALONG ARTERIOLES AND VENULES IN RAT EXTENSOR DIGITORUM LONGUS(EDL) AND SOLEUS
(SOL) MUSCLES
Region, vessel EDL muscle Mid, arteriole Mid. venule Proximal, arteriole Proximal, venule
SOL Muscle Mid, arteriole Mid, venule Distal, arteriole Distal, venule
Order
Segment length Comparison/ n P value -
Branch frequency Order
n
Comparison/ P value
1” 2 3 1” 2 3 1” 2” 3 1” 2 3
9 58 93 3 10 23 9 37 39 3 8 9
1” 1” 2” 1” 1”
> > > > > 2” a 1” > 1” > 2” > 1” = 1” = 2” =
2”/0.001 Y/O.001 3”/0.001 2”/0.001 Y/O.001 3”/0.498 T/O.017 Y/O.001 3”/0.009 2”/0.723 3”/0.157 3”/0.842
1” 2 3 1” 2 3 1” 2” 3” 1” 2” 3
8 52 89 1 9 22 8 33 39 3 8 9
1” 1” 2” 1” 1” 2” 1” 1” 2” 1” 1” 2”
> > > = > > > > > = > =
T/O.001 3”/0.001 3”/0.006 2”/0.215 3”/0.001 3”/0.001 2”/0.003 Y/O.001 Y/O.027 2”/0.162 3”/0.009 Y/O.094
1” 2” 3 1” 2 3” 1” 2 3” 1” 2 3
5 16 11 4 19 8 4 11 7 3 11 7
1” 1” 2” 1” 1” 2” 1” 1” 2” 1” 1” 2”
= 2”/0.086 zz 3”/0.054 = Y/O.426 > 2”/0.016 > 3”/0.030 = 3”/0.462 > 2”/0.014 > 3”/0.002 = Y/O.184 = 2”/0.718 = Y/O.770 = Y/O.996
1” 2 3” 1” 2” 3” 1” 2” 3” 1” 2” 3”
3 1.5 10 4 14 8 2 10 4 3 11 7
1” 1” 2” 1” 1” 2” 1” 1” 2” 1” 1” 2”
> > > > > > > > = > > =
T/O.004 3”/0.001 3”/0.006 2°/o.001 3”/0.001 Y/O.003 T/O.017 3”/0.001 3”/0.057 T/O.001 3”/0.001 Y/O.069
Note. Vessel orders are primary (l”), secondary (2)O, and tertiary (3”). Summary data (means 2 SE) are pictured in Fig. 2 for segment length and Fig. 3 for branch frequency.
muscles, BR decreased as vessel order increased (Table 2). The ranges of BR were similar across muscles, regions, vessel type, and order. DISCUSSION This study has documented differences in microvascular architecture between EDL and SOL skeletal muscles of the rat. Branch frequency consistently decreased as vessel order increased in both muscles. Differences in segment lengths between muscles consistently showed shorter vascular segments in the SOL muscle compared to the EDL muscle, indicating more compact microvascular networks in the SOL muscle. Anatomical Considerations Differences in vascular networks between red and white skeletal muscle have been of interest since first noted by Ranvier (1874). Greater density of pre- and
200
WILLIAMS
A
v
A
AND
V
SEGAL
A
V
A
V
MEDL PEEI DSOL FIG. 3. Mean (2 SE) branch frequencies for arterioles (A) and venules (V) in extensor digitorum longus (EDL) and soleus (SOL) muscles. Vascular regions are mid (M), proximal (P), and distal (D). Vessel orders are primary (lo), secondary (2”), and tertiary (3”). (n’s are reported in Table 2). a, M EDL2” arteriole < MEoL 2” venule (P = 0.02). b, Ma,, 3” arteriole > M,,, 3” venule (P = 0.0001). c, Mm. 2” venule > MsoL 2” venule (P = 0.002). d, M,,, 3” venule < P,,, 3” venule (P = 0.01). e, MS,,. 2” arteriole > Mso, , 2” venule (P = 0.01). See Table 2 for additional comparisons.
postcapillary blood vessels in red compared to white muscle has been described (Myrhage and Eriksson, 1980); however, these microvessels have not been quantified. The present study has demonstrated longer SL along 1” microvessels in M EDL compared to PEp, and compared to both regions of the SOL muscle. Capillary/fiber ratios measured in the adult rat (1.15: 1 for EDL and 2.3 : 1 for SOL; Hudlicka, 1985) have indicated greater microvascular density in the SOL muscle compared to the EDL muscle. The results here demonstrate that greater density of the microcirculation in the SOL muscle is not limited to capillaries. Rather, components of the microvascular network (arterioles, venules, and capillaries) are organized more compactly within the SOL muscle relative to the EDL muscle. These data are consistent with greater oxygen requirements in the SOL muscle compared to the EDL muscle. The magnitude of differences between SOL and EDL muscle and muscle fiber lengths (Table 1) did not account for the observed differences in SL. For example, the EDL muscle was only 12% longer than the SOL muscle, yet SL of the 1” arteriole was 51% longer in the EDL (mid region) compared to the SOL (mid region). Normalizing SL to muscle fiber length strengthened the argument of relatively shorter microvessel segments (more compact microvascular networks) in the SOL muscle compared to the EDL muscle (Table 3). Dissimilar microvascular networks were also observed within the EDL muscle. The 1” arteriole of the P,,, region had shorter SL compared to that of the MEDL region. These microvessels also differed in orientation with respect to the muscle axis (Figs. 1 and 2). Shorter SL in one region of the EDL versus another may be explained by inhibited growth due to “sensing” anatomical boundaries such
MICROVESSELS
IN RED AND
WHITE
MUSCLE
201
as other vessels, fiber orientation, or the muscle surface. A similar phenomenon may limit the length of 2” and 3” branches, which were diagonal relative to fiber orientation and displayed shorter SL and lower BR relative to 1” microvessels (Figs. 1, 2, and 3). The mechanisms determining microvessel SL remain to be established. Little is known about adaptation of pre- and postcapillary microvessels within skeletal muscles when subjected to experimental interventions, such as exercise or deconditioning. The arteriolar and venular segments measured in this study represent basic units of the microvascular networks in EDL and SOL muscles. Studying responses of these units to endurance training, muscle hypertrophy and/or atrophy relative to the control data reported here could prove fruitful in understanding growth and regression of the microcirculation. Structural similarities and differences between networks, as quantified by BR and SL, may also provide a framework for interpreting differences in physiological properties between the EDL and SOL muscles.
Physiological
Considerations
Blood flow is not distributed uniformly between muscles, among regions of muscles, or within the microcirculation either at rest or during exercise (Laughlin and Armstrong, 1982; Duling and Damon, 1987). Our present findings indicate an anatomical basis for blood flow heterogeneity. In the EDL muscle, for example, variations in blood flow could result from more compact vascular architecture in one region (PEDL) compared to the second region (MEDL), attributable to shorter SL of 1” arterioles in PEDL. Further study is required to relate microvascular network heterogeneity and blood flow distribution within a muscle. In the rat, blood flow is distributed in accord with muscle fiber type (Laughlin and Armstrong, 1982; Mackie and Terjung, 1983). Studies of mixed muscles of rat, rabbit, and human have found more capillaries associated with oxidative compared to glycolytic muscle fibers (Romanul, 1965; Gray and Renkin, 1978). However, maximal muscle blood flow was independent of capillarity (Maxwell et al., 1980). Thus, variations in the microvascular network proximal and distal to capillaries may explain the relationships between muscle blood flow and fiber type (Laughlin and Armstrong, 1982; Mackie and Terjung, 1983). Muscle fibers in the rat SOL muscle are predominantly SO (87%) with the remaining fibers classified as FOG (13%; Armstrong and Phelps, 1984). In contrast, the EDL muscle is composed of approximately equal portions of FOG and FG fibers, with only 2% SO fibers (Armstrong and Phelps, 1984). The different structural dimensions of pre- and postcapillary microvessels reported here provide an explanation for differences in blood flow between the EDL and SOL muscles of the rat (Armstrong and Laughlin, 1985) that is independent of capillarity. More compact vascular networks within the SOL muscle are consistent with greater blood flow conductance in the SOL muscle as compared to the EDL muscle. Thus varied flow between muscles could be due to the differing vascular network structures rather than to variations in vasomotor tone, resulting in a simple form of control with minimal energy requirement. For example, passive (structural) control of flow via network architecture would selectively allow flow
202
WILLIAMS
AND
SEGAL
to muscle fibers having high oxygen requirements. Measurements of the vascular structures between and within other muscles differing in oxidative metabolism would further test this hypothesis. The coordination of a hyperemic response within microvascular networks can match oxygen delivery with local demand in striated muscle (Segal, 1991). The distance over which vasomotor responses are conducted along arterioles (Segal, 1991) may, in part, explain the greater flow magnitude developed in red versus white muscle during muscle contractions (Folkow and Halicka, 1968; Hilton et al., 1970; Reis and Wooten, 1976; Laughlin and Armstrong, 1982; Mackie and Terjung, 1983). For example, with a length constant of =2 mm (Segal, 1991), a conducted response initiated at the origin of a distal 2” arteriole and spreading upstream along its parent 1” arteriole could affect flow into almost twice the number of 2” branches in the SOL muscle (0.37 mm, mean SL) compared to similar microvessels in the EDL muscle (0.69 mm, mean SL; Fig. 2). Differences in SL would not explain dissimilar hyperemic responses among 2” and 3” microvessels in SOL and EDL muscles since SL of these microvessels were similar among the respective regions of both muscles (Fig. 2). However, such similarities in architecture in 2” and 3” microvessels may not be critical if 1” arterioles are the major sites of blood flow control in the EDL and SOL muscles. Further work is needed to clarify these relationships. Limitations
and Nomenclature
In the present study, casting material was dispersed uniformly throughout the muscles. No empty regions were observed in the muscles examined despite incomplete filling of capillaries (documented previously by Gogh, 1918). In practice, capillaries devoid of casting material allowed us to visualize networks through the muscle and to measure casts of the larger microvessels. By starting at each vascular region on the muscle surface and tracing inward, we identified intact networks located similarly across muscles. We have assumed that the networks sampled were representative of the microvessel population within the SOL and EDL muscles. With the above visual assurances, we were confident in our sampling procedure. Nevertheless, these casts may not represent entirely the in vivo microvascular networks of the SOL and EDL muscles. Some secondary and tertiary branches may not have filled completely due to severe branch angles, small vessel diameters, surface tension, different filling patterns between the two muscles, or other structural constraints. Recognizing this limitation, we did not use the Horton-Strahler ordering system, which, by definition, starts with the smallest and works to the largest branches (Fenton and Zweifach, 1981). In the present study, ordering the microvascular network from large to small microvessels (Hutchins and Darnell, 1974) was consistent with methods used for lung (Weibel, 1984), brain (Coleman and Riesen, 1968), bat wing (Wiedeman, 1968), tenuissimus muscle (Eriksson and Myrhage, 1972), intestine (Bohlen, 1983), and biceps femoris muscle (Myrhage and Eriksson, 1980). In conclusion, we have documented differences in microvascular architecture between SOL and EDL muscles. When different, segment lengths were shorter in microvessels of the SOL compared to the EDL muscles, indicating more compact pre- and postcapillary networks in the SOL muscle. In both muscles, branch frequency decreased from primary to tertiary microvessels and was similar within
MICROVESSELS
IN
RED
AND
WHITE
MUSCLE
203
each order between muscles. The proximal EDL region contained more compact microvessels compared to the microvascular network that supplied the distal twothirds of the muscle. Differences in microvascular architecture between the SOL and EDL muscles reported here provide one explanation for the dissimilar values for blood flow between these muscles both at rest and during exercise.
ACKNOWLEDGMENTS The authors thank H. B. Chew for vascular casting, Joseph L. Loomis and Douglas A. Johnson for expert technical assistance, and Janice Derr and Elvessa Aragon for statistical assistance. This research was supported by NASA Grant NAGW 1196, a Grant-In-Aid from the Pennsylvania Affiliate of the American Heart Association, and NIH Grant HL41026. D. A. Williams was the recipient of the Mrs. A. Robert No11 Fellowship in Physiology.
REFERENCES ARMSTRONG, R. B., AND PHELPS, R. 0. (1984). Muscle fiber type composition of the rat hindlimb. Am. J. Anat. 171,259-272. ARMSTRONG, R. B., AND LAUGHLIN, M. H. (1985). Rat muscle blood flows during high-speed locomotion. J. Appl. Physiol. 59, 1322-1328. BOHLEN, H. G. (1983). Intestinal microvascular adaptation during maturation of spontaneously hypertensive rats. Hypertension 5, 139-745. BURKE, R. E. (1981). Motor units: Anatomy, physiology, and functional organization. In “Handbook
of Physiology, the Nervous System” (V. B. Brooks, Ed.), Vol. 2, pp. 345-422. American Physiology Society, Bethesda, MD. CLOSE, R. (1964). Dynamic properties of fast and slow skeletal muscles of the rat during development. J. Physiol.
173,74-95.
CLOSE, R. (1972). Dynamic properties of mammalian skeletal muscles. Physiol. Rev. 52, 129-197. COLEMAN, P. D., AND RIESEN, A. H. (1968). Environmental effects on cortical dendritic fields. I. Rearing in the dark. J. Anat. 102,363-374. DULING, B. R., AND DAMON, D. H. (1987). An examination of the measurement of flow heterogeneity in striated muscle. Circ. Res. 60, 1-13. ERIKSSON, E., AND MYRHAGE, R. (1972). Microvascular dimensions and blood flow in skeletal muscle. Acta Physiol. Stand. 86, 211-222. FENTON, B. M., AND ZWEIFACH, B. W. (1981). Microcirculatory model relating geometrical variation to changes in pressure and flow rate. Ann. Biomed. Eng. 9, 303-321. FOLKOW, B., AND HALICKA, H. (1968). A comparison between ‘red’ and ‘white’ muscle with respect to blood supply, capillary surface area and oxygen uptake during rest and exercise. Microvusc. Res. 1, 1-14. GRAY, S. D., AND RENKIN, E. M. (1978). Microvascular supply in relation to fiber metabolic type in mixed skeletal muscles of rabbits. Microvasc. Res. 16, 406-425. HILTON, S. M., JEFFRIES,M. G., AND VRBOV~, G. (1970). Functional specializations of the vascular bed of soleus. J. Physiol. 206, 543-562. HUDLICK~, 0. (1985). Development and adaptability of microvasculature in skeletal muscle. Z. Exp. Biol. 115,215-228. HUTCHINS, P. M., AND DARNELL, A.
E. (1974). Observation of a decreased number of small arterioles in spontaneously hypertensive rats. Circ. Res. (Suppl. Z) 34-35, I-161-1-165. KROGH, A. (1918). The number and distribution of capillaries in muscles with calculations of the oxygen pressure head necessary for supplying the tissue. J. Physiol. 52, 409-415. LAUGHLIN, M. H., AND ARMSTRONG, R. B. (1982). Muscular blood flow distribution patterns as a function of running speed in rats. Am. J. Physiol. 243, H296-H306. MACKIE, B. G., AND TERJUNG, R. L. (1983). Blood flow to different skeletal muscle fiber types during contraction. Am. J. Physiol. 245, H265-H275.
204
WILLIAMS
AND
SEGAL
L. C., WHITE, T. P., AND FAULKNER, J. A. (1980). Oxidative capacity, blood flow, and capillarity of skeletal muscles. J. Appl. Physiol. 49, 627-633. MBNDEZ, J., AND KEYS, A. (1960). Density and composition of mammalian muscle. Metabolism 9, 184-188. MYRHAGE, R., AND ERIKSSON, E. (1980). Vascular arrangements in hind limb muscles of the cat. J. Anat. Wl, l-17. POTTER, R. F., AND GROOM, A. C. (1983). Capillary diameter and geometry in cardiac and skeletal muscle studied by means of corrosion casts. Microvasc. Res. 25, 68-84. RANVIER, P. L. (1874). Note sur les vaisseaux sanguins et la circulation dans les muscles rouges. Arch. MAXWELL,
Physiol. REIS, D.
Norm. Path. 6, 446-450. J., AND WOOTEN, G. F.
(1976). Blood flow in red and white muscle: Relationship to metabolism, development and behavior. Prog. Brain Rex 44, 385-401. ROMANUL, F. C. (1965). Capillary supply and metabolism of muscle fibers. Arch. Neural. 12, 497509. SEGAL,
S. S. (1991). Microvascular recruitment in hamster striated muscle: Role for conducted vasodilation. Am. J. Physiol. 261, H181-H189. SEGAL, S. S., AND FAULKNER, J. A. (1985). Temperature-dependent physiological stability of rat skeletal muscle in vitro. Am. J. Physiol. 248, C265-C270. STINGL, J. (1969). Arrangement of the vascular bed in the skeletal muscles of the rabbit. Folia Morphologica WEIBEL, E.
II,
257-264.
R. (1984). “The Pathway for Oxygen: Structure and Function in the Mammalian Respiratory System.” Harvard Univ. Press, Cambridge, MA. WIEDEMAN, M. (1968). Blood flow through terminal arterial vessels after denervation of the bat wing. Circ.
Res. 22, 83-89.