Muscle capillaries represent the final step in the vascular pathway for O2 delivery to the muscle cells; therefore, capillary abundance in a muscle reflects its structural capacity for O2 flux from blood to the muscle fibers. It is clear that the denser the capillary network in a muscle, the greater the capillary surface area available per volume of tissue, and therefore the greater the capacity for O2 delivery to the muscle fibers. Intuitively, one would also expect narrower fibers to have greater O2 supply because of the shorter diffusion distances from surrounding capillaries to the center of the muscle fibers compared with wider fibers. Indeed, small fiber size, either occurring naturally in muscles or as a result of experimentation or disease, has traditionally been viewed in terms of its impact in reducing O2 diffusion distances from capillary to muscle fiber mitochondria. However, Gayeski and Honig (2) suggested a major role of capillary surface area rather than intrafiber diffusion distance in determining maximal O2 flux in muscle, based on cryospectrophotometric measurement of myoglobin saturation in quick frozen muscles that revealed uniformly low intrafiber PO2 in muscle at maximal exercise. This indicated that a major resistance to O2 flux occurs at the capillary-fiber interface, a large PO2 gradient being necessary to generate O2 flux in the carrier-free region between red blood cells in the capillaries and the sarcoplasm of the subjacent muscle fibers, whereas myoglobin-facilitated diffusion allows transport to the cell interior with a very shallow gradient (2). Subsequently, both O2 distribution and myoglobin-facilitated transport in the muscle fibers have been controversial, but the finding of a major PO2 drop at the capillary-fiber interface remained. Voter and Gayeski (15) showed that the spatial resolution of the cryospectrophotometric technique was lower than originally estimated and did not allow mapping of PO2 distribution within the muscle fibers, but confirmed low intracellular PO2 in exercising muscles. Measurement of myoglobin desaturation in human muscle by a different technique, noninvasive proton magnetic resonance spectroscopy, revealed low intracellular PO2 at even less than maximal work (14), and other physiological and morphological studies (1,4,5,7) also supported the notion of a major site of functional resistance to O2 flux at the capillary-to-fiber interface. A recent theoretical model that incorporated current information on blood and tissue O2 transport and parameters confirmed low PO2 values and shallow PO2 gradients in muscle tissue at high O2 demand (11). It also supported conclusions by others of a major role of intravascular resistance to O2 diffusion and negligible effect of myoglobin-facilitated diffusion on O2 transport (see (11), and references therein). In this review, we examine the impact of the concept of a major site of functional resistance to O2 flux at the capillary-fiber interface on the assessment of muscle structural capacity for O2 flux from capillary to fiber mitochondria, and recent findings on its plasticity.
MUSCLE CAPILLARY-TO-FIBER STRUCTURE AND ITS ASSESSMENT
Muscle capillaries run mostly parallel to the muscle fibers, with a degree of tortuosity that depends on sarcomere length, i.e., the extent of muscle contraction (Fig. 1). It was shown that capillary tortuosity increases substantially with muscle contraction in skeletal muscles of mammals, and that it is a consequence of fiber shortening rather than an indicator of the O2 requirements of the tissue (reviewed in (7)). It is also known that alteration in capillary tortuosity does not accompany the increased capillary number with increased muscle activity in endurance training (13) or chronic electrical stimulation (9), i.e., there is no additional increase in capillary surface area via altered capillary geometry when sample sarcomere length is taken into account.
Traditionally, muscle capillarization has been assessed by counting the number of capillaries per tissue cross-sectional area (capillary density) or per fiber number (capillary-to-fiber ratio) in muscle transverse sections. A major role of the capillary-fiber interface in determining maximal O2 flux in muscles implies that capillarization in a muscle transverse section needs to be assessed in relation to fiber perimeter (Fig. 2) rather than fiber cross-sectional area (used to estimate capillary density) or fiber number (used in capillary-to-fiber ratio) to estimate muscle capacity for O2 flux from capillary to fiber mitochondria. Also, the effect of geometry, such as the increase in capillary surface area due to tortuosity and branching, needs to be incorporated into the estimate of fiber capillarization. Both tasks are met in the capillary-to-fiber perimeter ratio, which is a morphometric estimate of capillary surface per fiber surface obtained by standard intersection-counting (e.g., see (9)) in muscle transverse sections (10).
As illustrated in Figure 2, the size of the capillary-fiber interface is determined by the relative amount of capillary surface per fiber surface. Thus, it is affected by both capillary number (i.e., capillary-to-fiber ratio) and fiber size. Similarly, both capillary-to-fiber ratio and capillary density affect intercapillary diffusion distances; however, maximal diffusion distance to the center of the muscle fibers depends on fiber size and therefore capillary density (Fig. 2). Other estimates of fiber capillarization such as capillary surface and length per fiber volume incorporate the contribution of capillary size and/or geometry into capillary density, but do not reflect the size of the capillary-fiber interface. In contrast, capillary number around a fiber is an estimate of fiber capillarization on an individual fiber basis, and thus a partial indicator of the size of the capillary-fiber interface. However, unlike capillary-to-fiber perimeter ratio, capillary number around a fiber incorporates neither fiber size nor the effects of capillary geometry and fiber capillary sharing into the estimate of fiber capillarization.
DATA IN INTENSELY AEROBIC MUSCLES
Figure 3 illustrates the extremely dense capillarization and small fiber size characteristic of intensely aerobic muscles. The study of ultimate cases of muscles with extremely high O2 demand, such as the flight muscles of small birds and bats, provided insights into structural designs for high O2 flux rates, and the size of the capillary-fiber interface in muscles in relation to their O2 demand (7). It revealed extremely high capillary length·mL−1 mitochondria in flight muscles, at values two–threefold greater than previously estimated for mammalian muscles. Interestingly, the greater capillary length per fiber mitochondrial volume in flight muscles was achieved via different capillary geometry in bird versus bat. Yet, capillary surface per fiber surface, i.e., the size of the capillary-fiber interface, per fiber mitochondrial volume was similar in bird and bat flight muscles. It was about twofold greater in flight muscle than in rat hindlimb (Fig. 4), consistent with the twofold higher O2 consumption rate per unit mitochondrial volume measured by Suarez and colleagues in flying hummingbirds compared with limb muscles of mammals running at V̇o2max (reviewed in (7)). Recent data on the diaphragm of the shrew, the smallest mammal, revealed capillary surface per fiber surface ratio per fiber mitochondrial volume as high as in flight muscles (12). The greater capillary-to-fiber surface ratio per fiber mitochondrial volume in intensely aerobic muscles supports the notion of a major role of the capillary-fiber interface in determining maximal O2 flux rates in muscles (2). It suggests that the small fiber size in intensely aerobic muscles may be important not so much to reduce diffusion distance, but instead to maximize the size of the capillary-fiber interface relative to the volume of mitochondria to be supplied in the muscle fibers (7). At the extremely high mitochondrial volume densities (28–38%) characteristic of the muscle fibers, the required high capillary-to-fiber surface ratio in these muscles can only be achieved via small fiber size, i.e., fibers with a large surface to volume ratio to accommodate the necessary capillary surface area around each fiber. For example, fiber cross-sectional area and mitochondrial volume density in hummingbird flight muscle were 200 μm2and 35%, respectively (7). Using the linear regression between capillary-to-fiber surface ratio and fiber mitochondrial volume in flight muscle (Fig. 4), one calculates that a flight muscle with a fiber size comparable with that in rat hindlimb (e.g., 2000 μm2 would require a capillary-to-fiber surface ratio of 1.08 to supply the 700 μm3 (i.e., 0.35 × 2000) mitochondrial volume·μm−1 fiber length in the muscle, which is impossible.
Capillary per fiber number is not particularly high in intensely aerobic muscles, consistent with spatial limitation around small fibers (Fig. 3). For example, capillary-to-fiber ratio and capillary number around a fiber in hummingbird flight muscle were 40 and 19%smaller, respectively, than in rat soleus muscle (7), illustrating that capillary per fiber number alone does not necessarily fully reflect the extent of fiber capillarization in muscle. In addition, mitochondrial volume·μm−1 fiber length varied 5.7-fold in rat hindlimb (Fig. 4), underlining the importance of examining capillarization in relation to fiber O2 demand, i.e., mitochondrial volume. Notably, those hindlimb muscles with greater fiber mitochondrial volume did not show reduced fiber size (i.e., did not have smaller diffusion distances to the center of the muscle fibers), but instead had one or two additional capillaries per fiber compared with hindlimb muscles with lesser mitochondrial volume per fiber (7). This is also consistent with a major role of capillary surface area rather than intrafiber diffusion distance in determining maximal O2 flux rates in muscles (2).
APPLICATION TO HUMAN TISSUE
As mentioned earlier, physiological studies in humans also support the notion of a major site of functional resistance to O2 flux at the capillary-fiber interface (14). Structurally, the assessment of the size of the capillary-fiber interface in human muscle tissue is difficult because of the collapse of capillaries in muscle biopsies (Fig. 5), which precludes measurement of capillary perimeter by light microscopy. Hepple (3) introduced the capillary-to-fiber perimeter exchange index, i.e., the quotient of the individual, fiber-based, capillary-to-fiber ratio (Fig. 5) and fiber perimeter, as an alternative to estimate the size of the capillary-fiber interface in muscle biopsies.
A difference with capillary-to-fiber perimeter ratio is that the capillary-to-fiber perimeter exchange index measures fiber-specific capillary number (instead of perimeter) per fiber perimeter, i.e., it does not account for the effect of capillary geometry, such as the increased capillary perimeter in muscle transverse sections with vessel tortuosity (Fig. 2), in the estimate of fiber capillarization. Human and longitudinal studies of muscle structure depend on the use of needle biopsy material, which always strongly contracts upon excision. Thus, capillary-to-fiber perimeter exchange index estimates in muscle biopsies underestimate the size of the capillary-to-fiber interface, because the high degree of capillary tortuosity in contracted muscle (Fig. 1) is not incorporated in the measurement. In comparison, capillary-to-fiber perimeter ratio varies little with muscle contraction (10), because the increased capillary perimeter in transverse sections (with increased tortuosity in the muscle) is partially compensated by the increased fiber cross-sectional area, and therefore perimeter, with muscle contraction. Because the variability in degree of muscle contraction in biopsies is small (6), and the degree of capillary tortuosity is not affected by exercise training (13), the limitation of the capillary-to-fiber perimeter exchange index in studies of muscle capillary supply in response to exercise is not expected to be significant, unless comparison is made of muscles with different capillary geometry relative to fiber sarcomere length (7). Comparison of the two methods in dog gastrocnemius muscle revealed a strong correlation between capillary-to-fiber perimeter exchange index and capillary-to-fiber perimeter ratio when both measurements were made in the same sections of perfusion-fixed tissue, whereas comparison with contralateral biopsy material is obscured by differences in fiber size, sarcomere length, and muscle sampling between the two preparations (i.e., biopsy vs perfusion-fixed material (6)).
The use of the capillary-to-fiber perimeter exchange index to assess fiber capillarization in vastus lateralis muscle of older men in response to resistance and aerobic training showed proportional changes in V̇o2 peak and the size of the capillary-fiber interface but not capillary density (Fig. 6), and supported the notion that muscle structural capacity for O2 flux is related to the size of the capillary-fiber interface rather than intrafiber diffusion distances (5).
PLASTICITY OF THE CAPILLARY-FIBER INTERFACE
Several animal studies have shown that capillary surface per fiber surface ratio and fiber mitochondrial volume increase in proportion to one another in various muscles in response to different perturbations such as endurance training, chronic electrical stimulation, and adaptation to altitude (Fig. 7). Notably, the maintenance of the size of the capillary-fiber interface occurred in muscles (hindlimb, flight muscles) with largely different fiber type composition and oxidative capacity. Also, different alterations in capillary density, geometry, and fiber mitochondrial volume occurred in response to endurance training, chronic electrical stimulation, and adaptation to altitude. Yet, capillary-to-fiber surface ratio relative to fiber mitochondrial volume was maintained in each muscle and condition, supporting the notion that the size of the capillary-fiber interface may be regulated in direct proportion to fiber mitochondrial volume or maximal O2 demand in skeletal muscle, irrespective of their fiber type composition, level of aerobic capacity, degree of capillarization, and capillary geometry (reviewed in (8)). Angiogenic signals and increased capillarization precede the increase in mitochondrial enzymes in muscles with chronic electrical stimulation. Thus, whereas time courses of capillary versus mitochondrial plasticity may differ, the unchanged ratio of capillary surface per fiber surface and fiber mitochondrial volume suggests remarkable integration of O2-dependent signaling pathways, from O2 sensing to vascular growth and increased capacity for aerobic metabolism, with increased activity and muscle hypoxia (reviewed in (8)). The molecular nature of cellular O2 sensors and their interactions with angiogenic and mitochondrial O2-regulated signaling pathways are not known.
Because of the major role of the capillary-fiber interface as a site of functional resistance to O2 flux, muscle structural capacity for O2 flux from capillary to fiber mitochondria needs to be assessed in terms of capillary surface per fiber surface, rather than per fiber cross-sectional area (capillary density) or per fiber number (capillary-to-fiber ratio). Capillary-to-fiber perimeter ratio is a morphometric estimate of capillary surface per fiber surface, i.e., the size of the capillary interface, in muscle transverse sections. The capillary-to-fiber perimeter exchange index is an alternative estimate in muscle biopsies, where capillary perimeter cannot be directly measured by light microscopy in muscle transverse sections. When contrasted with data on fiber size and capillary density, both estimates supported the notion of a major role of the size of the capillary-fiber interface, rather than diffusion distances, in determining maximal O2 flux in muscles. Studies of muscles with large differences in O2 demand underlined the importance of examining capillarization in relation to the volume of mitochondria to be supplied in the muscle fibers, rather than fiber capillary number alone, to assess muscle structural capacity for O2 flux. Examination of the plasticity of the capillary-fiber interface in various conditions of chronically increased muscle activity and altitude adaptation indicated that the size of the capillary-fiber interface may be regulated in direct proportion to fiber mitochondrial volume or maximal O2 demand in skeletal muscles, irrespective of their fiber type composition, level of aerobic capacity, and degree of capillarization and capillary geometry, reflecting remarkable integration of O2-dependent signaling pathways regulating capillary and muscle fiber mitochondrial growth.
This work was supported by National Institutes of Health grant POI HL-17731, the Canadian Fitness and Lifestyle Research Institute, and the Ontario Ministry of Tourism, Transport, and Recreation.
1. Bebout, D. E., Hogan, M. C. Hempleman, S. C. Wagner. P. D. Effects of training and immobilization on VO2
in dog gastrocnemius muscle in situ. J. Appl. Physiol. 74: 1697–1703, 1993.
2. Gayeski, T. E. J., Honig. C. R. O2
gradients from sarcolemma to cell interior in red muscle at maximal VO2
. Am. J. Physiol. 251: H789–H799, 1986.
3. Hepple, R. T. A new measurement of tissue capillarity: the capillary-to-fibre perimeter exchange index. Can. J. Appl. Physiol. 22: 11–22, 1997.
4. Hepple, R. T., Hogan, M. C. Stary, C. Bebout, D. E. Mathieu-Costello, O. Wagner. P. D. Structural basis of muscle O2
diffusing capacity: evidence from muscle function in situ. J. Appl. Physiol. 88: 560–566, 2000.
5. Hepple, R. T., Mackinnon, S. L. M. Goodman, J. M. Thomas, S. G. Plyley. M. J. Resistance and aerobic training in older men: effects on VO2peak
and the capillary supply to skeletal muscle. J. Appl. Physiol. 82: 1305–1310, 1997.
6. Hepple, R. T., Mathieu-Costello. O. Estimating the size of the capillary-to-fiber interface in skeletal muscle: a comparison of methods. J. Appl. Physiol. 91: 2150–2156, 2001.
7. Mathieu-Costello, O. Comparative aspects of muscle capillary supply. Annu. Rev. Physiol. 55: 503–525, 1993.
8. Mathieu-Costello, O. Muscle adaptation to altitude: tissue capillarity and capacity for aerobic metabolism. High Alt. Med. Biol. 2: 413–425, 2001.
9. Mathieu-Costello, O., Agey, P. J. Wu, L. Hang, J. Adair. T. H. Capillary-to-fiber surface ratio in rat fast-twitch hindlimb muscles after chronic electrical stimulation. J. Appl. Physiol. 80: 904–909, 1996.
10. Mathieu-Costello, O., Ellis, C. G. Potter, R. F. MacDonald, I. C. Groom. A. C. Muscle capillary-to-fiber perimeter ratio: morphometry. Am. J. Physiol. 261: H1617–H1625, 1991.
11. McGuire, B. J., Secomb. T. W. A theoretical model for oxygen transport in skeletal muscle under conditions of high oxygen demand. J. Appl. Physiol. 91: 2255–2265, 2001.
12. Morales, S., Agey, P. J. Hazelwood, L. Savolainen, J. Mathieu-Costello. O. How high is the capillary supply of mitochondria in the diaphragm of shrews of different size? FASEB J. 14: A317, 2000.
13. Poole, D. C., Mathieu-Costello, O. West. J. B. Capillary tortuosity in rat soleus muscle is not affected by endurance training. Am. J. Physiol. 256: H1110–H1116, 1989.
14. Richardson, R. S., Noyszewski, E. A. Kendrick, K. F. Leigh, J. S. Wagner. P. D. Myoglobin O2
desaturation during exercise. Evidence of limited O2
transport. J. Clin. Invest. 96: 1916–1926, 1995.
15. Voter, W. A., Gayeski. T. E. J. Determination of myoglobin saturation of frozen specimens using a reflecting cryospectrophotometer. Am. J. Physiol. 269: H1328–H1341, 1995.