Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2003;554(Pt 2):559–569. doi: 10.1113/jphysiol.2003.046953

Enhancement of microvessel tortuosity in the vastus lateralis muscle of old men in response to endurance training

N Charifi 1,2, F Kadi 2, L Féasson 1, F Costes 1, A Geyssant 1, C Denis 1
PMCID: PMC1664782  PMID: 14578492

Abstract

Muscle microvascularization is usually quantified in transverse sections, in absolute terms (capillaries around fibres, CAF, or capillary-to-fibre ratio, C/F) or as CAF related to fibre area (CAF/area, CAFA). The capillary-to-fibre perimeter exchange ratio (CFPE) has been introduced in order to assess the role of the capillary-to-fibre interface in resistance to O2 diffusion. The ratio between the length of capillaries in contact with fibres and fibre perimeter (LC/PF) has also been used as an index for capillary tortuosity. The possibility of change in capillary tortuosity with endurance training was not considered in previous studies. Consequently, this study investigated the effect of 14 weeks of endurance training on muscle microvascularization, including microvessel tortuosity, in 11 elderly men (8th decade). Microvessels were analysed using the CD31 antibody. Together with the significant increase in peak oxygen exchange and citrate synthase activity, there was a significant increase in C/F. While CFPE and CAFA remained unchanged, an important finding was the clear increase in LC/PF (56%; P < 0.001) for a same sarcomere length. We also found a strong correlation between oxidative enzyme activity and LC/PF both before and after training. These results indicate that endurance training induces significant remodelling in the microvessel network in elderly men and that an increase in the degree of microvessel tortuosity would be an important mechanism of adaptation to endurance training.

The quantification of the capillary network in muscles is a useful tool for identifying changes in muscle O2 diffusing capacity and oxidative capacity in humans following an endurance training programme (Andersen & Henriksson, 1977; Ingjer, 1979; Saltin & Gollnick, 1983). It is now well established that skeletal muscles in the elderly are able to adapt to endurance training by enhancing their capillary supply (Denis et al. 1986; Coggan et al. 1992; Proctor et al. 1995; Freyssenet et al. 1996; Hepple et al. 1997). However, the results concerning the morphological changes underlying the enhancement of capillary supply are controversial. While, in one study, the enhancement of capillary supply was due to a decrease in fibre area without any changes in the number of capillaries (Denis et al. 1986), other studies have shown that the enhancement of capillary supply was due to an increase in the number of capillaries in contact with muscle fibres (Coggan et al. 1992; Freyssenet et al. 1996; Hepple et al. 1997).

In electrically stimulated rat skeletal muscle, angiogenesis starts with the proliferation of capillary endothelial cells and results in the formation of sprouts (Mathieu-Costello, 1993; Hudlicka, 1998). The new sprouts form loops and cross-connections, but not necessarily more numerous single elongated vessels running parallel to the muscle fibres, thus contributing to a more tortuous capillary network architecture (Hansen-Smith et al. 1996). At the moment, the role of enhanced capillary tortuosity in the improvement of O2 diffusing capacity is still under debate. It has been shown that there were no changes in capillary tortuosity in skeletal muscles of rats in response to treadmill running or to chronic electrical stimulation (Poole et al. 1989; Mathieu-Costello et al. 1996). In contrast, in chronically stimulated skeletal muscles of cats, an increase in the number and in the length of the capillaries was observed (Hudlicka et al. 1987; Hudlicka, 1991). In humans, only one study showed that the percentage of capillaries cut longitudinally in muscle cross-sections was higher in trained than in untrained young adult subjects (Parsons et al. 1993). The effect of the degree of capillary tortuosity on the improvement of muscle tissue oxygenation received recent support from a theoretical mathematical computational model (Goldman & Popel, 2000). However, the question of whether capillary tortuosity is altered in old subjects remains unanswered.

The morphometrical strategies used to quantify the training-induced changes in muscle capillary network are of great importance (Lexell, 1997). The capillary supply is usually assessed by counting the number of capillaries around each fibre (CAF) or by computing the ratio between the number of capillaries present in an area and the number of fibres in the same area (C/F). Other indices derived from Krogh's hypothesis (Krogh, 1919) and based on the role of the O2 diffusion distance between each capillary and the centre of the fibre have been used: capillary density (CD) and CAF related to the fibre area (CAFA). Other studies have shown that the muscle–capillary interface is the most important factor involved in the resistance to O2 diffusion (Gayeski & Honig, 1986; Honig et al. 1992). For this reason, precise stereological procedures, for instance, capillary-to-fibre perimeter ratio, i.e. capillary perimeter divided by fibre perimeter based on the analysis of perfused muscles, were used to assess the capillary-to-fibre interface, including the tortuosity of the capillary network (Mathieu-Costello et al. 1991). However, these indices cannot be used in human studies because skeletal muscles need to be perfused in order to prevent capillary collapse. To overcome this methodological issue, the capillary-to-fibre perimeter exchange index (CFPE) has been used in the study of human tissue (Hepple, 1997). CFPE represents the ratio between the capillary-to-fibre ratio calculated for each individual fibre (C/Fi) and the perimeter of the fibre (PF). In this respect, CFPE index and capillary-to-fibre perimeter ratio were found to be correlated (Hepple & Mathieu-Costello, 2001). However, CFPE index does not take into account the orientation of the capillaries in a transverse section. Therefore, in order to identify the size of the muscle–capillary interface, it would be necessary to assess the length of the capillary-to-fibre contact. In this respect, the percentage of muscle fibre perimeter in contact with the capillary wall in transverse sections (LC/PF) can be used (Sullivan & Pittman, 1987). Capillary tortuosity can thus be indirectly determined using LC/PF in repeated muscle samples from a longitudinal study (i.e. exercise training programme), with the condition that the length of sarcomeres is equal in each sample (Mathieu-Costello et al. 1989).

The aim of the present study was to investigate whether endurance training induces an increase in capillary tortuosity of the vastus lateralis muscle in old men. For this purpose the capillary supply was assessed using the CD31 antibody (Horak et al. 1992) and different capillary indices including CFPE and LC/PF were determined. In parallel, muscle oxidative capacity was quantified by the study of citrate synthase activity.

Methods

Subjects

Eleven healthy men (age, 73 ± 3 years; body mass, 80 ± 12 kg; height, 171 ± 5 cm) participated in the study. All subjects were fully informed of the nature and possible risks of the various procedures before giving their written consent. This investigation was approved by the Consultant Committee on Human Protection from Medical Research of Rhône-Alpes-Loire Region in accordance with the French Law and with the Declaration of Helsinki.

Training procedure

The subjects trained for 14 weeks by pedalling a mechanically braked Monark bicycle ergometer (Varberg, Sweden) for 45 min daily on 4 days per week. Each session was performed with a 5 min intermittent mode, i.e. 4 min at a workload corresponding to 65–75% of peak oxygen consumption (V̇O2peak) and the subsequent 5th minute at 85–95% V̇O2peak. This sequence was repeated 8 times without interruption until the 45th minute. Heart rate was used as the criterion to maintain an adequate relative workload and to adjust the absolute workloads every week because of the training effect. All training sessions were supervised by one of the authors.

Determination of peak oxygen consumption

Before and after the training period, subjects performed an exercise test for V̇O2peak determination. Workload increments of 20 W were imposed every 2 min on the bicycle ergometer (beginning at 0 W) until voluntary exhaustion. Oxygen uptake was measured during the last 30 s of each separate exercise level. The subjects breathed through a Hans Rudolph respiratory valve 2700 and the expired gas was collected in polyethylene Douglas bags. The gas fractions were measured with an infrared CO2 analyser (Normocap, Datex, Helsinki, Finland) and a paramagnetic O2 analyser (Servomex 1440, Crowborough, England) calibrated with gas mixtures determined by Scholander's method (Scholander, 1947). The ventilatory volumes were determined with a Tissot spirometer. Heart rate was monitored and plotted from the electrocardiogram signal. Three minutes after cessation of exercise, a fingertip blood sample was taken to be analysed for lactate concentration (YSI 2300, Yellow Springs Instruments, Yellow Springs, OH, USA). Heart rate, respiratory exchange ratio and blood lactate concentration were used as criteria for acceptance of V̇O2peak.

Muscle biopsy

Muscle biopsies were taken before and after the training period from the vastus lateralis (at a level corresponding to one third of the distance from the upper margin of the patella to the anterior superior iliac spine) using Weil-Blakesley forceps. The muscle biopsies were taken from the same leg pre- and post-training. The post-training biopsy was taken 2 cm away from the pretraining biopsy site. The weight of the biopsy sample averaged 80 mg. The sample was divided into four equal parts for different analyses. A part of the sample containing well-identified fascicles was orientated under a stereo microscope, included in an embedding medium (Cryomount; Histolab, Göteborg, Sweden), frozen in isopentane cooled to its freezing point in liquid nitrogen, and stored in liquid nitrogen until further cryostat sectioning. The remainder of the sample was rapidly frozen and stored in liquid nitrogen until enzyme activity assays were performed.

Enzyme activity assays

Muscle samples were freeze dried (Lyovac GT2, Leybold-Heraeus, Köln, Germany), dissected free from connective tissue and blood, and powdered in a chamber of controlled humidity (<40% relative humidity). The muscle powder was weighed in the same chamber, homogenized at 4°C in 0.1 m phosphate buffer (pH 8.2) containing 5 mm 2-mercaptoethanol, 30 mm NaF, 5 mm MgCl2 and 0.5 mm ATP. This tissue suspension was used to measure spectrophotometrically the activity of phosphofructokinase (PFK; Essen-Gustavsson & Henriksson, 1984) as a cytosolic glycolytic marquer of metabolism. The activity of citrate synthase (CS) was determined fluorometrically (Mansour, 1966) to attest the oxidative mitochondrial capacity. All the enzyme activities were measured at 25°C and expressed in micromoles per minute per gram dry weight (UdW).

Immunocytochemical and histochemical assays

Serial transverse sections, 10 μm thick, were cut with a microtome at –20°C (Tissue Tek II, Miles Laboratories, Elkhart, IN, USA). The identification of capillaries was performed using the monoclonal antibody CD31 (Dako, Glostrup, Denmark; M0823) which recognizes PECAM-1 (platelet endothelial cell adhesion molecule), a transmembranous glycoprotein strongly expressed by vascular endothelial cells. CD31 has been used successfully to identify vascular endothelium in tumour tissue (Horak et al. 1992; Vermeulen et al. 1996) and in normal muscle (Brey et al. 2002). In a pilot study on serial sections from human muscle biopsy material (n = 5) we found no statistical difference between the number of capillaries determined using α-amylase-PAS (Andersen, 1975) and CD31. We noted that visualization of the capillaries was easier using CD31. The slides were incubated in an atmosphere saturated with water vapour, for 1 h with CD31 antibody (mouse-antihuman), for 30 min with the secondary antibody (rabbit-antimouse, Dako P0260), and for 30 min with the tertiary antibody (swine-antirabbit, Dako P0217). All the incubations were performed at room temperature, and the slides were rinsed between each incubation with a phosphate-buffered saline solution. Peroxidase labelling was performed using a DAB substrate kit (Vector, Burlingame, USA; SK-4100), that yields a brown reaction end-product at the site of the target antigen (endothelial cell, Fig. 1).

Figure 1. Muscle cross-section of the vastus lateralis stained with the monoclonal antibody CD31 for the visualization of microvessels in one subject before (A) and after training (B).

Figure 1

Open in a new tab

Microvessels are stained brown and nuclei are blue. Bar = 100 μm.

Fibre type distribution was studied on serial sections stained for myofibrillar adenosine triphosphatase (ATPase) after preincubation at different values of pH (pH 4.35, 4.55 and 10.4) according to the methodology of Brooke & Kaiser (1970). The fibres were designated as I, IIa, IIx (previously referred to as IIb by Brooke and Kaiser) and IIc (I-Iia; Smerdu et al. 1994; Kadi et al. 1998). Type I fibres were light at pH 10.4 and dark at 4.55 and 4.35. Type IIa fibres were dark at pH 10.4 and light at pH 4.55 and 4.35. Type IIx fibres were dark at pH 10.4 and 4.55 and light at pH 4.35. Type IIc fibres were intermediately stained at all pH values.

Longitudinal cryo-sections, 10 μm thick, were obtained by rotating at –20°C each block of biopsy previously cross-sectioned. They were stained using Haematoxylin, Eosin and Safran in order to measure the mean sarcomere length.

Morphological assessments

Muscle sections were viewed under a light microscope (Eclipse E400, Nikon, Badhoevedorp, the Netherlands) connected to a digital camera (Coolpix 990, Nikon). All the following measurements were made under blinding protocols. Photographs were taken at ×400 magnification and analysed using a free image section software (Scion Image). An average of eight fields were examined in each section, to give an analysis of an average of 70 fibres per individual sample.

Fibre area and perimeter were measured. Obliquity in fibre sectioning was assessed using the form factor that represents: (4π× fibre area)/(fibre perimeter)2. The less elliptic the cross-section, the nearer the form factor is to 1. If a given fibre has a cylindrical shape, any perpendicular section will have a circular area (r = radius) and consequently the form factor is 1[(4π×r2)/[2r]2 = 1]. An oblique section will result in an elliptic slice (r and R as minor and major radius) with a form factor Inline graphic. For example, if R = 2r, then the form factor is 0.8.

Assessment of capillary network

Global indices

In a given area, capillary density (CD) was expressed as the number of capillaries counted per square millimetre (cap mm−2).

Capillary-to-fibre ratio (C/F), the ratio between the number of capillaries present in an area and the number of fibres in the same area, was calculated. In this calculation, the number of capillaries is corrected by subtracting half the number of the capillaries in the periphery of the area (Andersen & Henriksson, 1977; Brodal et al. 1977).

Individual fibre indices

The mean number of capillaries around a single fibre (CAF) and CAF relative to the area of the fibre (CAFA) were calculated.

To assess the capillary-to-fibre interface we calculated the CFPE index (capillary-to-fibre perimeter exchange; Hepple, 1997). We first determined the capillary-to-fibre ratio for each individual fibre (C/Fi): for each fibre, capillaries in contact with that fibre were counted by taking into account their sharing factor. The sharing factor represents the inverse number of fibres in contact with the same capillary. The CFPE index was then calculated as the ratio between C/Fi and fibre perimeter (Hepple, 1997).

In transverse sections, several capillaries are cut longitudinally, owing to their tortuous arrangement around muscle fibres. Capillaries were identified by the size of their cross-section: transverse or longitudinal profiles with a diameter less than 15 μm were counted. This helped to exclude vessels larger than capillaries (arterioles or venules). In the present study, most of the vessels considered in the endomysium had a diameter <10 μm. However, even within this range (diameter <10 μm), some microvessels might well represent terminal arterioles (Hansen-Smith et al. 1998). Thus, the criterion of size is important but not sufficient for the distinction between a capillary and a microarteriole. For this reason, it seems justified to use the term ‘microvessel’ and not ‘capillary’ in the interpretation of our results. This will be done in the following text. However, indices computed in this study have been termed as described in previous studies.

The immunostaining of microvessel walls allows visualization of all portions of transversally or longitudinally running microvessels and measurement on transverse sections of the length of the contact (LC) between the microvessels and the fibres. The index of tortuosity was calculated as follows: LC/PF (perimeter of fibre). LC/PF, originally called ‘length of capillary-fibre contact’ (Sullivan & Pittman, 1987), is expressed as a percentage of muscle fibre perimeter in contact with the capillary wall. In a muscle cross-section, the profile of the microvessels will depend on their orientation in relation with the fibres. Microvessels were classified into three classes using as a criterion the length of their contact with fibres (LC): structures with LC ≤ 10 μm were considered as parallel microvessels, structures with 10 μm ≤ LC = 30 μm were considered as oblique microvessels, and structures with LC > 30 μm were considered as perpendicular microvessels. The percentages of microvessels with either of these profiles were calculated before and after training.

Mean sarcomere length

Mean sarcomere length was measured on longitudinal sections. Well orientated fibre portions with regular striations were selected on approximately 20 different zones and photographed. The length occupied by 10 consecutive sarcomeres was measured with the software previously described. Consequently, an average of 200 sarcomeres per longitudinal section were measured.

Statistical analysis

Data are presented as means and standard deviations. The statistical significance of the differences between the pre- and post-training period was determined using a non-parametric Wilcoxon signed rank test. A two-way analyse of variance (ANOVA) was performed in order to study whether there were differences in C/F and C/Fi (method and training as independent variables). Linear regression and correlation coefficients (R) were used to determine the degree of relationship between two variables. The effect of the training programme on a given relationship between two variables was tested using multiple regression analysis with the introduction of a qualitative variable (0 and 1 corresponding to pre- and post-training data, respectively) as described by Daniel (1995). P values below the 0.05 level were considered statistically significant.

Results

Data on peak exercise indicators and enzyme activities are given in Table 1. The training programme did not induce any significant alterations in fibre type distribution (60%± 18 type I, 27 ± 17 type IIa, 11 ± 7% type IIx; 2 ± 3% type IIc before training and 64 ± 17% type I, 22 ± 13% type IIa, 12 ± 8% type IIx, 2 ± 1% type IIc after training).

Table 1.

Peak exercise indicators and enzyme activities before and after training

Before After
Peak exercise indicators
 V̇O2peak (ml min−1 (kg body weight) −1 28.8 ± 5.9 32.7 ± 5.4*
 Power (W) 171 ± 38 199 ± 27**
 Respiratory exchange ratio 1.07 ± 0.05 1.06 ± 0.07
 Heart rate (bpm) 159 ± 11 154 ± 13*
 Blood lactate (mmol l−1) 7.0 ± 1.6 7.7 ± 1.7
Enzyme activities
 PKF (μmol min−1 g−1) 154 ± 41 144 ± 29
 CS (μmol min−1 g−1) 45 ± 18 60 ± 27**
Open in a new tab

Values are means ± s.d.

*

P < 0.05

**

P < 0.01

Global microvessel indices and microvessel indices for each fibre type are presented in Table 2. Global C/F and individual C/F (C/Fi) were both significantly increased (+18% and +17%, respectively, P < 0.05). C/Fi was significantly increased by 40% in type IIa fibres (P < 0.05). The two-way ANOVA demonstrated that C/F and C/Fi were not statistically different (no method effect) but both were increased by training. Endurance training induced a 19% increase in CAF. This increase was very close to the significance level (P = 0.0506). Likewise, although mean increases in CAF of type I, type IIa and type IIx fibres were observed, these were not statistically significant.

Table 2.

Indices of the microvascular supply of the muscle fibres before and after training

Before After
Total
 Global microvessel indices
  CD 333 ± 69 357 ± 61
  C/F 1.7 ± 0.4 2.0 ± 0.4*
 Individual fibre microvessel indices
  CAF 4.8 ± 1.3 5.7 ± 1.4
  CAFA 0.9 ± 0.1 1.0 ± 0.3
  C/Fi 1.8 ± 0.6 2.1 ± 0.6*
  CFPE 5.8 ± 1.5 6.7 ± 1.5
  LC/PF 16 ± 7 25 ± 10***
Type I
  CAF 5.1 ± 1.3 5.9 ± 1.2
  CAFA 0.8 ± 0.3 1.0 ± 0.4
  C/Fi 1.9 ± 0.6 2.2 ± 0.5
  CFPE 6.1 ± 1.4 6.7 ± 1.7
  LC/PF 18 ± 7 25 ± 10**
Type IIa
  CAF 4.2 ± 1.4 5.4 ± 1.9
  CAFA 0.8 ± 0.3 0.8 ± 0.4
  C/Fi 1.5 ± 0.5 2.1 ± 1.0*
  CFPE 5.0 ± 1.5 6.2 ± 2.0
  LC/PF 13 ± 7 22 ± 9**
Type IIx
  CAF 4.0 ± 1.2 4.8 ± 2.1
  CAFA 0.8 ± 0.4 0.8 ± 0.5
  C/Fi 1.4 ± 0.4 1.9 ± 0.9
  CFPE 5.0 ± 1.2 5.3 ± 2.7
  LC/PF 13 ± 5 20 ± 9
Open in a new tab

The data are means ± s.d. CD: capillary density (cap mm−2), C/F: capillary-to-fibre ratio (cap); CAF: number of capillaries around single fibres (cap); CAFA: CAF relative to the area of the fibre (cap (1000 μm)−2); C/Fi: individual capillary-to-fibre ratio (cap); CFPE: capillary-to-fibre perimeter exchange (cap (1000 μm)−1); LC/PF: length of the contact between microvessels and the fibre/perimeter of the fibre (%).

*

P < 0.05

**

P < 0.01

***

P < 0.001.

When capillary number was expressed in relation to fibre area (CAFA), there was no significant increase in this index (+11%). This is probably due to the simultaneous increase in the size of muscle fibres (13% increase, P = 0.0505) (Table 3). As indicated by the unchanged form factor between prior and after training, there were no differences in the sectioning obliquity of the fibres. Before training, the cross-sectional area of type I fibres (6012 ± 990 μm2) was significantly higher (P < 0.05) than the cross-sectional area of type IIa fibres (4965 ± 1329 μm2) and type IIx fibres (4563 ± 1023 μm2). After training, the areas were not statistically different in relation to fibre type (type I: 6393 ± 1636 μm2; type IIa: 6121 ± 2152 μm2; type IIx: 5084 ± 1741 μm2). The increase in fibre area with training was significant for type IIa fibres (+23%, P = 0.02) but not for type I and type IIx fibres. Capillary density (CD) was not significantly increased. Taken together, these results indicate that muscle microvascularization expressed in relation to the size of muscle fibres was not statistically increased.

Table 3.

Morphological data of the whole muscle fibres before and after training (n =11)

Before After
Cross-sections
 Perimeter (μm) 299 ± 25 321 ± 44
 Area (μm2) 5463 ± 971 6197 ± 1708
 Factor of form 0.76 ± 0.03 0.75 ± 0.07
Longitudinal sections
 Sarcomere length (μm) 1.65 ± 0.05 1.67 ± 0.07
Open in a new tab

Values are means ± s.d.

The capillary-to-fibre perimeter exchange (CFPE) was increased in each fibre type but not significantly.

A main outcome in the present study was the significant increase in LC/PF index (+56%; P < 0.001; Fig. 1). LC/PF increases reached 39% in type I fibres (P < 0.05) and 69% in type IIa fibres (P < 0.05). The measurement of the mean sarcomere length showed no significant changes between muscle biopsies taken before and after training (Table 3). Thus the absence of any change either in the contraction state of the postbiopsy sample or in the form factor suggests that the training-induced increase in LC/PF was the consequence of a remodelling in microvessel tortuosity.

The percentages of microvessels classified according to their length of contact with fibres are presented in Table 4. The training programme induced a significant decrease in transverse section microvessel profile and significant increases in oblique and longitudinal microvessel profiles.

Table 4.

Distribution of the microvessels classified according to their length in contact with fibres (LC) before and after training

Before After
LC = 10 μm
 Mean ± s.d. 68.5 ± 8.2 53.7 ± 10.2**
 Median ± s.d. 5.2 ± 0.9 6.1 ± 1.0**
10 < LC = 30 μm
 Mean ± s.d. 24.8 ± 8.0 31.6 ± 8,7*
 Median ± s.d. 14.4 ± 0.7 15.2 ± 0.9**
LC > 30 μm
 Mean ± s.d. 6.7 ± 4.1 14.6 ± 7.8**
 Median ± s.d. 41.6 ± 6.9 41.1 ± 4.7
Open in a new tab
*

P < 0.05;

**

P < 0.01.

The relations between CS activity and the indices of microvascularization before and after training are presented in Fig. 2. LC/PF index was strongly correlated to CS activity both before and after training (Fig. 2A). The pre- and post-training linear relations were parallel and the shift in between was found to be significant by multiple regression analysis (P < 0.05). Thus, there was a greater microvessel length in contact with fibres for a given level of CS activity after training than before training. Figure 2B shows that the number of microvessels in contact with fibres, evaluated with the sharing factor (C/Fi), was correlated to the citrate synthase activity both before and after training. Similarly, there was a significant correlation between CAF and CS activity before and after training (R = 0.74; P < 0.01 before training, and R = 0.61; P < 0.05 after training). When CAF was expressed in relation to fibre area (CAFA), there were no significant relationships between CAFA and CS (Fig. 2D). When C/Fi was expressed in relation to fibre perimeter (CFPE) there was a relationship between CFPE and CS before but not after training (Fig. 2C).

Figure 2. Relation between citrate synthase activity and microvascular supply indices before (▴) and after (▵) training.

Figure 2

Open in a new tab

Units: LC/PF:% (A); C/Fi: cap (B); CFPE: cap(1000 μm)−1 (C); CAFA: cap(1000 μm)−2 (D); CS activity: μmol min−1 g−1.

There was a significant correlation between LC/PF and CFPE indices before training (R = 0.933; P < 0.001). After training, although these parameters were significantly correlated there was a decrease in the coefficient of correlation (R = 0.659; P = 0.02). There was also a relation between LC/PF and CAF and between LC/PF and C/Fi in muscle biopsies before and after training. (R > 0.8; P < 0.001).

Discussion

The present study showed that 14 weeks of endurance training resulted in an important increase in microvessel tortuosity in a population of old men (8th decade). This adaptation would allow an improvement of oxygen transport (Goldman & Popel, 2000). Moreover, an important result in our study was the significant correlation between microvessel tortuosity and the oxidative capacity of muscles as evaluated by CS activity both before and after training.

Immunohistochemical preparation to identify microvessels on muscle cross-sections

In the present study, most of the vessels considered had a diameter <10 μm. However, as discussed in the Methods, the criterion of size is not sufficient for the distinction between a capillary and a terminal arteriole. A labelling of the vessel wall revealing the presence of smooth muscle actin would represent a reliable method to separate capillaries from arterioles (Hansen-Smith et al. 1998). For this reason, we chose to use the term ‘microvessel’ and not ‘capillary’ in the interpretation of our results. Moreover, it has been shown that oxygen exchange occurs not only across the wall of capillaries but also arterioles. Thus, the arteriolar network also participates in oxygen diffusion between blood and muscle cells (Pittman, 2000).

In the field of exercise physiology, the effects of endurance exercises on the capillary network have been mostly assessed using the α-amylase PAS reaction (Andersen, 1975). During the last decade, other approaches based on immunohistochemistry have been proposed. Antibodies against Ulex europaeus agglutinin I lectin (UEA-I) as an endothelial marker (Parsons et al. 1993; Lexell, 1997; Qu et al. 1997; Porter et al. 2002) and against laminin as a basal lamina marker (Kadi et al. 1998, 1999) were used. While frequently used in the identification of tumoral neoangiogenesis in human anatomical specimens, CD31 antibody has not been extensively used to identify capillaries and microvessels in normal human muscles. Recently, CD31 has been successfully used in three-dimensional analysis of the microvascular network in rat skeletal muscle (Brey et al. 2002). Our results showed that CD31 allows a clear and reliable identification of microvessels. As illustrated in Fig. 1, microvessels can be seen as spots (when a microvessel runs parallel to the fibre) or as segments (if a microvessel runs obliquely or perpendicularly to the length of the fibre) on transverse sections. Thus, it is possible to measure a given length of microvessel wall in contact with fibre perimeter. For a given fibre, the sum of the different segments (Fig. 3) gives a quantitative indicator of the microvessel-to-fibre interface, while the length of this interface cannot be measured using a stereological approach.

Figure 3. Effects of microvessel number and microvessel length on microvessel indices.

Figure 3

Open in a new tab

A represents the values of the indices before training. CAF, C/Fi and CFPE increase when the number of microvessels increases (B and D). When the length of the contact between microvessels and fibres increases, CAF, C/Fi and CFPE remain unchanged, while LC/PF increases (C). If both microvessel number and microvessel length increase, CAF, C/Fi and CFPE increase (D). This is due to the increase in microvessel number. Thus, the amplitude of CAF, C/Fi and CFPE enhancement is the same as in B, while the amplitude of LC/PF enhancement is more important. If an increase in fibre perimeter occurs together with an increase in the number and the length of the microvessels, the enhancement of CFPE and LC/PF is less important (E). Abbreviations: d, length of a microvessel; LC = ∑ d; PF, perimeter of the fibre.

Training-induced changes in the number of microvessels

An increase in the number of capillaries around fibres has been reported in response to endurance training in young (Saltin & Gollnick, 1983; Hudlicka et al. 1992 for reviews) and in elderly subjects (Coggan et al. 1992; Freyssenet et al. 1996; Hepple et al. 1997). By relating the changes in the number of capillaries to the size of muscle fibres it was possible to investigate whether the capillary supply of a given fibre was altered. In the elderly, previous studies have shown an increase in the number of capillaries (Coggan et al. 1992; Freyssenet et al. 1996; Hepple et al. 1997), while in one study the number of capillaries was maintained (Denis et al. 1986). However, in all these studies, the capillary number in relation to fibre area was increased because of different modulations in fibre area. In the present study, an increase in CAF occurred but did not lead to an increase in CD or in CAFA due to the simultaneous increase in fibre area.

Training-induced changes in the capillary-to-fibre interface

Sullivan & Pittman (1987) proposed an index that accounts for capillary geometry in rat skeletal muscles. As they observed a high incidence of capillaries cut longitudinally on transverse sections, they measured the percentage of muscle fibre perimeter in contact with capillary wall (capillary–fibre contact, which is presented as LC/PF in our study). This parameter was considered to give a better assessment of the vascular supply of O2 to the fibre than the number of capillaries around a fibre. Based on calculations predicting a rapid drop of O2 over the short distance between the interior of the red cell and the surface of the surrounding tissue, Gayeski & Honig (1986) showed that capillaries and their immediate surrounding represent the principal site of resistance to O2 diffusion. The capillary-to-fibre interface appeared to be an important determinant of the oxygen flux rates. In this perspective, Mathieu-Costello et al. (1991) proposed an index close to the capillary–fibre contact index, i.e. the ratio between the capillary perimeter and the fibre perimeter (capillary-to-fibre perimeter ratio index). They also showed that this index was independent of sarcomere length. In consequence, this index would allow the comparison of the size of the capillary-to-fibre interface between muscles of different fibre size or different sarcomere length.

The major inconvenience of the indices proposed by Sullivan & Pittman (1987) and Mathieu-Costello et al. (1991) is related to the fact that muscles have to be perfused and fixed. In this respect, Hepple (1997) proposed the measurement of an index that can be computed on transverse sections of non-perfused muscles (i.e. muscle samples obtained from a biopsy procedure in humans and histochemically assayed). This index was defined as the capillary-to-fibre perimeter exchange index (CFPE). Hepple et al. (1997) described a relation between the capillary supply and V̇O2peak in old subjects. It was also shown that when CFPE and CD were both significantly increased after aerobic training, CFPE would explain a greater part of the variance in V̇O2peak than would CD. By taking into account the fibre perimeter, the CFPE index introduces the role of the capillary-to-fibre interface. However, a major disadvantage of CFPE index is its insensitivity to the length of capillaries in contact with the fibre. Hepple & Mathieu-Costello (2001) assumed that CFPE index does not take into account capillary geometry. However, they considered that CFPE index remains a reliable tool to estimate the capillary-to-fibre interface in muscles when capillary tortuosity is unchanged. Indeed, by reference to Poole et al. (1989), who showed no increase in capillary tortuosity in muscles of rats after endurance training, they concluded that the limitation of CFPE index in studies of capillary supply in response to exercise should not be significant (Mathieu-Costello & Hepple, 2002).

Our study showed an increase in capillary-to-fibre interface with training (LC/PF). The fact that CAF and LC/PF increased, respectively, by 19 and 56% indicates that the mean length of microvessel segments increased per se. The mean length of sarcomeres was similar between pre- and post-training muscle samples. This allows us to conclude that the degree of contraction in pre- and post biopsy muscles was similar. Altogether these data indicate that the significant increase in LC/PF is a strong indication of the enhancement of microvessel tortuosity. This conclusion is supported by the significant increase in the proportion of oblique and longitudinal microvessels observed in muscle cross-sections (Table 4). These results are also in line with those reported in trained cross-country skiers and sedentary subjects (Parson et al. 1993).

Unexpectedly, the change in CFPE index was not significant. These conflicting results between LC/PF and CFPE changes are further analysed in Fig. 3. It appears that microvessel-to-fibre interface is dependent on three main morphological factors, i.e. microvessel number, length of the microvessel-to-fibre contacts and the perimeter of the fibre (PF). Any increase in the microvessel number (the two other factors remaining constant) is linked to a corresponding increase in microvessel-to-fibre contacts. In this case, all the indices are altered with the same amplitude (Fig. 3B). Any increase in the length of the contacts (due to tortuosity, microvessel number and PF remaining constant) only affects LC/PF (Fig. 3C). When both the number and the length of microvessels increase (PF constant), the amplitude of LC/PF change is twice that of the other indices (Fig. 3D). According to the results of the present study, Fig. 3E illustrates the changes observed in the three morphological factors (increase in CAF, tortuosity and PF): the slight increase in PF induces a decrease in both CFPE and LC/PF. Owing to the amplitude of the response, only LC/PF change remains statistically significant. Consequently, when changes in microvessel geometry result from an increase in both number and tortuosity, LC/PF seems to be the most robust index, even in the event of modest muscle hypertrophy.

Interrelationship between microvascular supply and oxidative capacity

In the present study the training-induced increase in microvascularization was accompanied by an enhancement of oxidative capacity. Moreover, a strong correlation was observed between the index of microvessel tortuosity and citrate synthase activity. In animal models, a few studies showed contradictory results (Maxwell et al. 1980; Egginton & Hudlicka, 2000). In humans, while a few studies have investigated the relationship between oxidative capacity or mitochondrial content and capillary supply (Romanul, 1965; Sullivan & Pittman, 1987; Poole & Mathieu-Costello, 1996), many have shown concomitant increases in capillarization and oxidative capacity in response to endurance training, both in young (Andersen & Henriksson, 1977; Ingjer, 1979; Coggan et al. 1992; Proctor et al. 1995) and in old subjects (Coggan et al. 1992; Proctor et al. 1995). Consequently, it is generally accepted that the increase in capillary supply would allow an enhancement of oxygen delivery to muscle fibres due to the increased mitochondrial metabolic demand.

We used different indices in our study and, as shown in Fig. 2, the results concerning the relation between microvascular supply and oxidative capacity depended largely on how microvascularization was expressed. Indeed, LC/PF is the only index that exhibited a strong relationship with CS activity before and after training (Fig. 2A). We also found a correlation between the number of microvessels and CS activity (Fig. 2B). Fig. 2C and D show that fibre dimensions can affect the relationship between CS activity and the indices of microvascularization such as CFPE and CAFA. Finally, LC/PF, which also takes into account the fibre perimeter, seems to be the most sensitive index because the correlation with CS activity was not affected by the slight changes in fibre size after training. Taken together these results suggest that the entire adaptive response to increased oxidative metabolism cannot be explained solely by increased microvascularization and reduced distance diffusion. Increased surface area available for microvessel–muscle fibre exchange seems to be another important cellular strategy in the adaptive response of the microvascular bed to the increase in the oxidative demand of the fibres. These conclusions are in line with a recent demonstration proposed by Goldman & Popel (2000) using a computational model.

Training-induced increase in muscle microvessel tortuosity in the elderly: a specificity?

Most of the earlier studies on muscle transverse sections from humans were performed with α-amylase-PAS histochemical assay. Using electron microscopy, Brodal et al. (1977) and Zumstein et al. (1983) compared the capillary supply in skeletal muscle of young trained and untrained subjects. However, in these studies, neither a clear identification nor a counting of segments of longitudinally cut capillaries has been mentioned. Consequently, a comparison of the present findings observed in elderly men with results in younger adults has yet to be performed. Younger muscle may exhibit cellular mechanisms more readily facilitating the formation of new microvessels while older muscle may rely more on a compensatory increase in the length of pre-existing capillaries. Further studies would be required to verify these hypotheses. The results of the present study do, however, support such a speculation, since training-induced increase of LC/PF expressed in as a percentage was more pronounced than the increase in CS activity (see the shift of the relation between LC/PF and CS activity). In an earlier study on young adults (Andersen & Henriksson, 1977), endurance training was found to have a greater influence on oxidative capacity than on the capillary network.

In conclusion, our findings clearly demonstrate an increase in microvessel tortuosity in skeletal muscle of elderly men in response to endurance training. Moreover, there is a significant relation between microvessel tortuosity and CS activity both before and after training.

Acknowledgments

The authors thank Ms J. Castells and Ms D. Gouttefangeas for valuable technical assistance. The research was supported by grants from the Région Rhône-Alpes (A. Geyssant, R.R.A. Action Thématique 97-99), the Swedish National Center for Research in Sports and the Loo and Hans Osterman Foundation. N. Charifi was a recipient of grants from Région Rhône-Alpes.

References

  1. Andersen P. Capillary density in skeletal muscle of man. Acta Physiol Scand. 1975;95:203–205. doi: 10.1111/j.1748-1716.1975.tb10043.x. [DOI] [PubMed] [Google Scholar]
  2. Andersen P, Henriksson J. Capillary supply of the quadriceps femoris muscle of man: adaptive response to exercise. J Physiol. 1977;270:677–690. doi: 10.1113/jphysiol.1977.sp011975. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Brey EM, King TW, Johnston C, McIntire LV, Reece GP, Patrick CW., Jr A technique for quantitative three-dimensional analysis of microvascular structure. Microvasc Res. 2002;63:279–294. doi: 10.1006/mvre.2002.2395. [DOI] [PubMed] [Google Scholar]
  4. Brodal P, Ingjer F, Hermansen L. Capillary supply of skeletal muscle fibers in untrained and endurance-trained men. Am J Physiol. 1977;232:H705–H712. doi: 10.1152/ajpheart.1977.232.6.H705. [DOI] [PubMed] [Google Scholar]
  5. Brooke MH, Kaiser KK. Three ‘myosin adenosine triphosphatase’ systems: the nature of their pH lability and sulfhydryl dependence. J Histochem Cytochem. 1970;18:670–672. doi: 10.1177/18.9.670. [DOI] [PubMed] [Google Scholar]
  6. Coggan AR, Spina RJ, King DS, Rogers MA, Brown M, Nemeth PM, Holloszy JO. Skeletal muscle adaptations to endurance training in 60- to 70-yr-old men and women. J Appl Physiol. 1992;72:1780–1786. doi: 10.1152/jappl.1992.72.5.1780. [DOI] [PubMed] [Google Scholar]
  7. Daniel WW. Biostatistics: A Foundation for Analysis in the Health Sciences. 6. New York: John Wiley & Sons, Inc.; 1995. pp. 460–476. [Google Scholar]
  8. Denis C, Chatard JC, Dormois D, Linossier MT, Geyssant A, Lacour JR. Effects of endurance training on capillary supply of human skeletal muscle on two age groups (20 and 60 years) J Physiol (Paris) 1986;81:379–383. [PubMed] [Google Scholar]
  9. Egginton S, Hudlicka O. Selective long-term electrical stimulation of fast glycolytic fibres increases capillary supply but not oxidative enzyme activity in rat skeletal muscles. Exp Physiol. 2000;85:567–573. [PubMed] [Google Scholar]
  10. Essen-Gustavsson B, Henriksson J. Enzyme levels in pools of microdissected human muscle fibres of identified type. Adaptive response to exercise. Acta Physiol Scand. 1984;120:505–515. doi: 10.1111/j.1748-1716.1984.tb07414.x. [DOI] [PubMed] [Google Scholar]
  11. Freyssenet D, Berthon P, Denis C, Barthelemy JC, Guezennec CY, Chatard JC. Effect of a 6-week endurance training programme and branched-chain amino acid supplementation on histomorphometric characteristics of aged human muscle. Arch Physiol Biochem. 1996;104:157–162. doi: 10.1076/apab.104.2.157.12879. [DOI] [PubMed] [Google Scholar]
  12. Gayeski TE, Honig CR. O2 gradients from sarcolemma to cell interior in red muscle at maximal V̇O2. Am J Physiol. 1986;251:H789–H799. doi: 10.1152/ajpheart.1986.251.4.H789. [DOI] [PubMed] [Google Scholar]
  13. Goldman D, Popel AS. A computational study of the effect of capillary network anastomoses and tortuosity on oxygen transport. J Theor Biol. 2000;206:181–194. doi: 10.1006/jtbi.2000.2113. [DOI] [PubMed] [Google Scholar]
  14. Hansen-Smith FM, Egginton S, Hudlicka O. Growth of arterioles in chronically stimulated adult rat skeletal muscle. Microcirculation. 1998;5:49–59. [PubMed] [Google Scholar]
  15. Hansen-Smith FM, Hudlicka O, Egginton S. In vivo angiogenesis in adult rat skeletal muscle: early changes in capillary network architecture and ultrastructure. Cell Tissue Res. 1996;286:123–136. doi: 10.1007/s004410050681. [DOI] [PubMed] [Google Scholar]
  16. Hepple RT. A new measurement of tissue capillarity: the capillary-to-fiber perimeter exchange index. Can J Appl Physiol. 1997;22:11–22. doi: 10.1139/h97-002. [DOI] [PubMed] [Google Scholar]
  17. Hepple RT, Mackinnon SL, Goodman JM, Thomas SG, Plyley MJ. Resistance and aerobic training in older men: effects on V̇O2 peak and the capillary supply to skeletal muscle. J Appl Physiol. 1997;82:1305–1310. doi: 10.1152/jappl.1997.82.4.1305. [DOI] [PubMed] [Google Scholar]
  18. Hepple RT, Mathieu-Costello O. Estimating the size of the capillary-to-fiber interface in skeletal muscle: a comparison of methods. J Appl Physiol. 2001;91:2150–2156. doi: 10.1152/jappl.2001.91.5.2150. [DOI] [PubMed] [Google Scholar]
  19. Honig CR, Connett RJ, Gayeski TE. O2 transport and its interaction with metabolism; a systems view of aerobic capacity. Med Sci Sports Exerc. 1992;24:47–53. [PubMed] [Google Scholar]
  20. Horak ER, Leek R, Klenk N, LeJeune S, Smith K, Stuart N, Greenall M, Stepniewska K, Harris AL. Angiogenesis, assessed by platelet/endothelial cell adhesion molecule antibodies, as indicator of node metastases and survival in breast cancer. Lancet. 1992;340:1120–1124. doi: 10.1016/0140-6736(92)93150-l. [DOI] [PubMed] [Google Scholar]
  21. Hudlicka O. What makes blood vessels grow? J Physiol. 1991;444:1–24. doi: 10.1113/jphysiol.1991.sp018863. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hudlicka O. Is physiological angiogenesis in skeletal muscle regulated by changes in microcirculation? Microcirculation. 1998;5:5–23. [PubMed] [Google Scholar]
  23. Hudlicka O, Brown M, Egginton S. Angiogenesis in skeletal and cardiac muscle. Physiol Rev. 1992;72:369–417. doi: 10.1152/physrev.1992.72.2.369. [DOI] [PubMed] [Google Scholar]
  24. Hudlicka O, Hoppeler H, Uhlmann E. Relationship between the size of the capillary bed and oxidative capacity in various cat skeletal muscles. Pflugers Arch. 1987;410:369–375. doi: 10.1007/BF00586513. [DOI] [PubMed] [Google Scholar]
  25. Ingjer F. Effects of endurance training on muscle fibre ATP-ase activity, capillary supply and mitochondrial content in man. J Physiol. 1979;294:419–432. doi: 10.1113/jphysiol.1979.sp012938. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kadi F, Eriksson A, Holmner S, Butler-Browne GS, Thornell LE. Cellular adaptation of the trapezius muscle in strength-trained athletes. Histochem Cell Biol. 1999;111:189–195. doi: 10.1007/s004180050348. [DOI] [PubMed] [Google Scholar]
  27. Kadi F, Hagg G, Hakansson R, Holmner S, Butler-Browne GS, Thornell LE. Structural changes in male trapezius muscle with work-related myalgia. Acta Neuropathol (Berl) 1998;95:352–360. doi: 10.1007/s004010050810. [DOI] [PubMed] [Google Scholar]
  28. Krogh A. The number and distribution of capillaries in muscle with calculation of the oxygen pressure head necessary for supplying the tissue. J Physiol. 1919;52:409–415. doi: 10.1113/jphysiol.1919.sp001839. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lexell J. Muscle capillarization: morphological and morphometrical analyses of biopsy samples. Muscle Nerve supplement. 1997;5:S110–S112. [PubMed] [Google Scholar]
  30. Mansour TE. Studies on heart phosphofructokinase: purification, inhibition, and activation. J Biol Chem. 1966;238:2285–2292. [PubMed] [Google Scholar]
  31. Mathieu-Costello O. Comparative aspects of muscle capillary supply. Annu Rev Physiol. 1993;55:503–525. doi: 10.1146/annurev.ph.55.030193.002443. [DOI] [PubMed] [Google Scholar]
  32. Mathieu-Costello O, Agey PJ, Wu L, Hang J, Adair TH. Capillary-to-fiber surface ratio in rat fast-twitch hindlimb muscles after chronic electrical stimulation. J Appl Physiol. 1996;80:904–909. doi: 10.1152/jappl.1996.80.3.904. [DOI] [PubMed] [Google Scholar]
  33. Mathieu-Costello O, Ellis CG, Potter RF, MacDonald IC, Groom AC. Muscle capillary-to-fiber perimeter ratio: morphometry. Am J Physiol. 1991;261:H1617–H1625. doi: 10.1152/ajpheart.1991.261.5.H1617. [DOI] [PubMed] [Google Scholar]
  34. Mathieu-Costello O, Hepple RT. Muscle structural capacity for oxygen flux from capillary to fiber mitochondria. Exerc Sport Sci Rev. 2002;30:80–84. doi: 10.1097/00003677-200204000-00007. [DOI] [PubMed] [Google Scholar]
  35. Mathieu-Costello O, Hoppeler H, Weibel ER. Capillary tortuosity in skeletal muscles of mammals depends on muscle contraction. J Appl Physiol. 1989;66:1436–1442. doi: 10.1152/jappl.1989.66.3.1436. [DOI] [PubMed] [Google Scholar]
  36. Maxwell LC, White TP, Faulkner JA. Oxidative capacity, blood flow, and capillarity of skeletal muscles. J Appl Physiol. 1980;49:627–633. doi: 10.1152/jappl.1980.49.4.627. [DOI] [PubMed] [Google Scholar]
  37. Parsons D, McIntyre K, Schultz W, Stray-Gundersen J. Capillarity of elite cross-country skiers: a lectin (Ulex europaeus I) marker. Scand J Med Sci Sports. 1993;2:89–98. [Google Scholar]
  38. Pittman RN. Oxygen supply to contracting skeletal muscle at the microcirculatory level: diffusion vs convection. Acta Physiol Scand. 2000;168:593–602. doi: 10.1046/j.1365-201x.2000.00710.x. [DOI] [PubMed] [Google Scholar]
  39. Poole DC, Mathieu-Costello O. Relationship between fiber capillarization and mitochondrial volume density in control and trained rat soleus and plantaris muscles. Microcirculation. 1996;3:175–186. doi: 10.3109/10739689609148286. [DOI] [PubMed] [Google Scholar]
  40. Poole DC, Mathieu-Costello O, West JB. Capillary tortuosity in rat soleus muscle is not affected by endurance training. Am J Physiol. 1989;256:H1110–H1116. doi: 10.1152/ajpheart.1989.256.4.H1110. [DOI] [PubMed] [Google Scholar]
  41. Porter MM, Stuart S, Boij M, Lexell J. Capillary supply of the tibialis anterior muscle in young, healthy, and moderately active men and women. J Appl Physiol. 2002;92:1451–1457. doi: 10.1152/japplphysiol.00744.2001. [DOI] [PubMed] [Google Scholar]
  42. Proctor DN, Sinning WE, Walro JM, Sieck GC, Lemon PW. Oxidative capacity of human muscle fiber types: effects of age and training status. J Appl Physiol. 1995;78:2033–2038. doi: 10.1152/jappl.1995.78.6.2033. [DOI] [PubMed] [Google Scholar]
  43. Qu Z, Andersen JL, Zhou S. Visualisation of capillaries in human skeletal muscle. Histochem Cell Biol. 1997;107:169–174. doi: 10.1007/s004180050101. [DOI] [PubMed] [Google Scholar]
  44. Romanul FCA. Capillary supply and metabolism of muscle fibers. Arch Neurol. 1965;12:497–509. doi: 10.1001/archneur.1965.00460290053007. [DOI] [PubMed] [Google Scholar]
  45. Saltin B, Gollnick PD. Skeletal muscle adaptability: significance for metabolism and performance. In: Peachey LD, editor. Handbook of Physiology. 4. Bethesda, MD, USA: American Physiological Society; 1983. pp. 555–631. Skeletal Muscle, section 10, chap. 19. [Google Scholar]
  46. Scholander PF. J Biol Chem. Vol. 167. 1947. Analyzer for accurate estimation of respiratory gases in one-half cubic centimeter samples; pp. 235–250. [PubMed] [Google Scholar]
  47. Smerdu V, Karsch-Mizrachi I, Campione M, Leinwand L, Schiaffino S. Type IIx myosin heavy chain transcripts are expressed in type IIb fibers of human skeletal muscle. Am J Physiol. 1994;267:C1723–C1728. doi: 10.1152/ajpcell.1994.267.6.C1723. [DOI] [PubMed] [Google Scholar]
  48. Sullivan SM, Pittman RN. Relationship between mitochondrial volume density and capillarity in hamster muscles. Am J Physiol. 1987;252:H149–H155. doi: 10.1152/ajpheart.1987.252.1.H149. [DOI] [PubMed] [Google Scholar]
  49. Vermeulen PB, Gasparini G, Fox SB, Toi M, Martin L, McCulloch P, Pezzella F, Viale G, Weidner N, Harris AL, Dirix LY. Quantification of angiogenesis in solid human tumours: an international consensus on the methodology and criteria of evaluation. Eur J Cancer. 1996;32A:2474–2484. doi: 10.1016/s0959-8049(96)00379-6. [DOI] [PubMed] [Google Scholar]
  50. Zumstein A, Mathieu O, Howald H, Hoppeler H. Morphometric analysis of the capillary supply in skeletal muscles of trained and untrained subjects, its limitations in muscle biopsies. Pflugers Arch. 1983;39:277–283. doi: 10.1007/BF00580261. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES