Calcaneal Tendon Regions Exhibit Different MMP-2 Activation After Vertical Jumping and Treadmill Running


  • Olga Cristina De Mello Malheiro,

    1. São Paulo State University (UNESP), School of Science and Technology, Department of Physical Education, Presidente Prudente, SP, Brazil
    Search for more papers by this author
  • Camila Tateishi Giacomini,

    1. São Paulo State University (UNESP), Institute of Biosciences, Department of Morphology, Botucatu, SP, Brazil
    Search for more papers by this author
  • Luis Antonio Justulin Jr.,

    1. São Paulo State University (UNESP), Institute of Biosciences, Department of Morphology, Botucatu, SP, Brazil
    Search for more papers by this author
  • Flávia Karina Delella,

    1. São Paulo State University (UNESP), Institute of Biosciences, Department of Morphology, Botucatu, SP, Brazil
    Search for more papers by this author
  • Maeli Dal-Pai-Silva,

    1. São Paulo State University (UNESP), Institute of Biosciences, Department of Morphology, Botucatu, SP, Brazil
    Search for more papers by this author
  • Sérgio Luis Felisbino

    Corresponding author
    1. São Paulo State University (UNESP), Institute of Biosciences, Department of Morphology, Botucatu, SP, Brazil
    • Department of Morphology, Institute of Biosciences, São Paulo State University (UNESP) 18.618-000, Botucatu, SP, Brazil
    Search for more papers by this author
    • Fax: +55 0XX 14 3811 6264


Increased activity of matrix metalloproteinases (MMPs) -2 and -9 was found in calcaneal tendon after physical training. However, little attention has been given to the distinct biomechanical and tissue structure of the calcaneal tendon's proximal and distal regions. Herein, we evaluated the effect of two types of physical activities on tendon morphology and matrix metalloproteinase activities in the proximal and distal regions of rat calcaneal tendon, separately. Adult male Wistar rats from control, water-adapted, vertical-jumping, and treadmill-running groups were sacrificed after 1 or 4 days of physical exercise, 6 hr after the end of that day's exercise session. Tendons were processed for histology, morphometry, and gelatin zymography. Tendons from adapted and trained animals showed active secretory cells and increased thickness, cellularity, and blood vessel volume fraction of peritendinous sheath, but without inflammatory process. In the proximal region, both pro- and active MMP-2 were increased after vertical jumping, but only pro-MMP-2 was increased after treadmill running. In contrast, in the distal region, both exercise types increased the activity of pro- and active MMP-2, especially treadmill running, which increased the active MMP-2 by about 11- and eightfold, respectively, after 1 and 4 days of training. No activity of MMP-9 was observed in either tendon region in this study. In conclusion, distal and proximal regions of calcaneal tendon exhibit differential intensities of tissue remodeling after treadmill running or vertical jumping and MMP-2, in the absence of inflammation, plays a major role in this adaptive response. Anat Rec, 2009. © 2009 Wiley-Liss, Inc.

The majority of tendons are composed mostly of parallel bundles of collagen fibers interposed between elongated fibroblasts and fibrocytes, and support great traction forces (Birch,2007). However, some tendons must resist additional forces of compression or friction against bone or cartilages, and are known as wrap-around tendons (Benjamin and Ralphs,1998). In these tendons, the areas that are subjected to the compressive forces develop a fibrocartilage-like tissue. Furthermore, tendons attached to the bone also present an enthesis or insertion fibrocartilage (Benjamin et al.,2006). The tendon's fibrocartilages possess distinct cellular and extracellular matrix composition and organization. In contrast with typical tendon structures, tendon fibrocartilages present chondrocyte-like cells surrounded by a basophilic pericellular matrix with a higher content of hyaluronan and proteoglycans, such as aggrecan, and type II collagen fibrils and type VI microfibrils (Ralphs et al.,1991; Vogel et al.,1994; Carvalho and Vidal;1995; Benjamin and Ralphs,1998; Felisbino and Carvalho,1999; Carvalho et al.,2000; Covizi et al.,2001; Carvalho et al.,2006). This specialization guarantees to tendon, the capacity to support compressive forces without compromising muscle force transmission to the bone.

Among the several mammalian wrap-around tendons, the calcaneal tendon or Achilles tendon has attracted special interest due to its importance in sports medicine (Benjamin et al.,2004; Shaw and Benjamin,2007). The proximal region of the calcaneal tendon, closer to the muscle, possesses the typical tendon structure to resist tensile forces exclusively. However, the distal region, closer to the calcaneus bone, presents two fibrocartilage structures: the first is a compressive fibrocartilage that progresses to calcification in older animals and the second is the enthesis fibrocartilage (Covizi et al.,2001; Esquisatto et al.,2007). This specialized and complex region of calcaneal tendon has led Benjamim et al. (2004) to propose the concept of an “enthesis organ.” Furthermore, the entheses are sites of stress concentration and consequently, are commonly subjected to overuse injuries (enthesopathies) that are well documented in a number of sports (Benjamin et al.,2006).

Although acute exercise and physical training have been demonstrated to increase type I collagen synthesis in peritendinous connective tissue of the Achilles tendon in humans, they induce a transient increase in interstitial concentration of the of type I collagen degradation product, which also indicates a rise in type I collagen degradation with exercise (Langberg et al.2001).

Collagen degradation is initiated extracellularly by a family of neutral zinc-containing endopeptidases denominated matrix metalloproteinases (MMPs), primarily by collagenases, such as MMP-1 and MMP-8, and subsequently by gelatinases, such as MMP-2 and MMP-9, which are presented in tissues mostly as latent pro-MMPs. Generally, MMP-1 and -8 are considered to be the most effective at degrading the fibrillar type I and type III collagens, however, MMP-2 and -9 can also degrade certain native collagen types (Nagase et al.,2006). MMP-9, for example, is capable of digesting collagen types IV, XI, and XIV, while MMP-2 is active against a range of collagens, including fibrillar collagens and type IV collagen (Aimes and Quigley,1995). To control the activity of these MMPs, cells also produce another class of proteins called tissue inhibitors of matrix metalloproteinases (TIMPs) (Nagase et al.,2006).

Recently, it has been demonstrated that acute exercise by treadmill running resulted in rapid (6 hr after termination of the exercise) elevation of interstitial amounts of pro-MMP-9 and TIMP-1 and a later rise (3 days after exercise) in pro-MMP-2 in human peritendinous calcaneal tendon tissue (Koskinen et al.2004), which supports the view that MMPs (and their inhibitors) play a role in ECM adaptation to exercise in tendon tissue. Moreover, Marqueti et al. (2006) also showed increases in MMP-2 and MMP-9 in rat calcaneal tendon after training by vertical jumping. Interestingly, the activities of these MMPs were reduced when the animals received the anabolic AAS (Deca-durabolin 50 and Durateston 50, alternately) simultaneously with the training, suggesting that anabolic-androgenic steroid treatment can impair tissue remodeling in the tendons of animals undergoing physical exercise by downregulating MMP activity, thus increasing the potential for tendon injury. Subsequently, using the same experimental model and training program, Marqueti et al. (2008) evaluated the MMP-2 activity in the proximal and distal calcaneal tendon, separately, and found no differences between these regions in either control or trained animals.

However, no previous studies compared the effects of two different types of exercises on the calcaneal tendon regardless of tendon region. Thus in the present study, we investigated the effects of the physical exercise, resisted, and resistance, on the activity of MMP-2 and MMP-9 in the compression and tension areas of the calcaneal tendon, to better elucidate the participation of these enzymes in the turnover of the tendon extracellular matrix submitted to two different types of exercises.



The adult male Wistar rats (13 weeks old, ∼300 g) employed in this study were obtained from the Central Stock breeder of the State University of Campinas—UNICAMP (Campinas, SP, Brazil) and were maintained under controlled temperature and lighting conditions, with free access to rodent food and water. All animal procedures were performed in accordance with the Guiding Principles in the Care and Use of Animals and approved by the ethics committee of our Institute.

Experimental Groups

The animals were divided randomly into four experimental groups (N = 10). Control Group (CT): they had free mobility in their cages, with free access to water and feed. Adapted group (ADP): to reduce stress, the animals were adapted to water in the first 3 days (without overload). Vertical Jumps Group with 50% overload (VJ): for four consecutive days, they were submitted to a daily physical training session composed of four sets of 10 jumps into water (at 28°C ± 2°C) in a plastic tube 25 cm in diameter filled with water to a depth of approximately double the height of the rat (∼40 cm). Sets of jumps were separated temporally by an interval equivalent to three times the duration spent by the animal to complete the previous set (a mean time of 20 sec to complete each set of jumps followed by ∼1 min of interval). The overload employed equaled 50% of the body weight of each animal. Treadmill Running Group (TR): this group of rats was submitted to a daily session of 30 min of treadmill running at a speed of 13 m/min on a surface inclined 5 degrees for four consecutive days.

The animals were sacrificed by an overdose of sodium pentobarbital (100 mg/kg) after 3 days of water adaptation and after 1 or 4 days of physical training, always 6 hr after of the end of the physical training. This latter time was chosen following the studies of Koskinen et al. (2004). The tendons were removed and some samples were processed for histological analysis and the others immediately frozen in liquid nitrogen for biochemical analysis.

Histology and Morphometry

The calcaneal tendons were immersed in 4% paraformaldehyde dissolved in phosphate buffer saline for 24 hr. Fixed samples were washed in PBS for 24 hr, dehydrated in a graded ethanol series and embedded in glycol metacrylate resin (historesin embedding kit, Leica Heidelberg). Semi-seriated resin (3 μm) sections were obtained and stained with Toluidine Blue 0.025% at pH 4.0 for morphological analysis.

The sections were examined in a Leica DMLB 80 microscope connected to a Leica DC300FX camera; the resultant digitalized images were analyzed morphometrically using the image analyzer Leica Q-win software Version 3.1 for Windows™.

The mean peritendinous sheath thickness was determined from a total of 250 random interactive measures at five different points of 10 different fields from five different individual tendon sections per experimental group at 20× magnification. To determine the mean blood vessel volume fraction (VF) of the peritendinous sheath, 50 random interactive measures were taken at 20× magnification in 10 different fields from five different individual tendon sections per experimental group. The blood vessel VF was determined as the percentage area per field of peritendinous sheath. The mean cellularity of the peritendinous sheath was determined by counting the total number of cells, excluding the blood vessel-related cells, in 10 different fields from five different individual tendon sections per experimental group at 20× magnification. The cellularity was expressed as mean cell number per field of peritendinous sheath.

Gelatin Zymography

Frozen samples of the right calcaneal tendon were divided into two regions, proximal and distal, from five different animals per experimental group and were mechanically homogenized in 50 mM Tris buffer, pH 7.5, plus 0.25% triton-X 100 by Polytron for 30 sec at 4°C. The homogenized samples were centrifuged and protein extracted from supernatant was quantified by the Bradford Methods (Bradford,1976). The extraction was performed in duplicate on left tendons from the same five animals. Aliquots (25 μg protein) from each extract were subjected to electrophoresis in gelatin-containing polyacrylamide (8% acrylamide) gels in the presence of SDS under non-reducing conditions as previously described (Carvalho et al.,2006). The gelatin substrate was present at 0.1% final concentration in the gel. The gels (0.75-mm thick) were electrophoresed for 2 hr at 100 V, 4°C, in a Bio-Rad MiniProtean III system (Bio-Rad Laboratories, Richmond, CA). Following electrophoresis, the gels were washed at room temperature in 50 mM Tris-HCl (pH 8.4) containing 2.5% Triton X-100 (two changes of 15 min). The gels were then incubated overnight (18 hr) in 50 mM Tris-HCl (pH 8.4) containing 5 mM CaCl2 and 1 μM ZnCl2 at 37°C. Following incubation, gels were stained with Coomassie Blue. Areas of proteolysis appeared as clear zones against a blue background. Molecular mass determinations were made with reference to pre-stained protein standards (Bio-Rad Laboratories) coelectrophoresed in the gels. The gels were also made in duplicate.

Image gels were captured and the bands corresponding to each enzyme form were quantified by densitometry as Integrated Optical Density (IOD) in an Image Master VDS version 3.0 (Pharmacia Biotech). The values were analyzed statistically.

Statistical Analysis

Values are expressed as mean ± SD. One-way analysis of variance (ANOVA) was performed to determine the presence of intergroup differences (P < 0.05) whose significance was assessed by the Tukey–Kramer post hoc test. A P value of ≤0.05 was considered significant. The statistical tests were performed on Instat (version 3.0; GraphPad, San Diego, CA).


The histological analysis in Figure 1a–c shows the typical morphology of the rat calcaneal tendon of the control animals, including the peritendinous sheath with few cells and blood vessels (Fig. 1a). Inside the proximal region of the tendon, the fibrocytes displayed condensed and flattened nuclei with reduced cytoplasm (Fig. 1b). In contrast, the distal region presented the typical morphology of tendon compressive fibrocartilage, showing chondrocytes with spherical nuclei and ample cytoplasm surrounded by basophilic pericellular matrix (Fig. 1c). After 3 days of water adaptation (Fig. 1d–f), significant augmentations in the thickness, vascularity (blood vessel and also volume fraction), and cellularity of the peritendinous sheath were observed (Table 1), vertical jumping days 1 (not shown) and 4 (Fig. 1g), and treadmill running days 1 (not shown) and 4 (Fig. 1h). No inflammatory infiltrate was observed in any tendon structure or region in this study. Inside trained tendons, the cells from both proximal and distal regions presented loose chromatin and ample cytoplasm, two characteristics of active cells (Fig. 1e,f).

Figure 1.

Longitudinal sections of rat calcaneal tendon stained by toluidine blue. (a) The proximal region from control animals presented elongated fibroblasts/fibrocytes (arrows) between very well aligned longitudinal collagen fibers. The peritendinous sheath (pe) showed few cells and blood vessels (arrowheads). (b) Detail of fibrocytes from proximal region (arrow). (c) The distal region presents groups of fibrochondrocytes (arrows) surrounded by a metachromatic pericellular matrix and a less orderly distribution of collagen fibers. (d) Proximal region from 3-day water adapted animals showing an increased cellularity around blood vessels in the peritendinous sheath (pe) and activated fibroblasts inside the tendon (arrows). (e) Proximal region detail of Figure d showing activated fibroblasts with loose chromatin and voluminous cytoplasm (arrow). (f) Distal region from 3-day water adapted animals showing activated fibrochondrocytes (arrows). (g) Proximal region from day 4 vertical jumping animals showing that peritendinous sheath (pe) appeared thickened with evident blood vessels (arrowhead). (h) Proximal region from day 4 treadmill running animals showing that peritendinous sheath (pe) also appeared thickened and presented many cells around blood vessels (arrowhead). Bars scale: a, d, g, and h = 30 μm; b, c, e, and f = 10 μm.

Table 1. Morphometric analysis of calcaneal tendon peritendinous sheath from different experimental groups
ParametersControlWAVJ day 1VJ day 4TR day 1TR day 4
  1. Values represent mean ± SD. The * indicate statistically significant differences from control with P ≤ 0.05.

  2. WA, water-adapted; VJ, vertical jumping; TR, treadmill running.

Thickness (μm)27.3 ± 5.241.7 ± 4.4*45.6 ± 6.7*48.3 ± 7.1*32.9 ± 4.647.7 ± 6.5*
Cellularity (number)14.7 ± 4.732.5 ± 7.2*35.3 ± 6.3*39.6 ± 7.1*27.9 ± 6.6*36.5 ± 6.8*
Blood vessels volume fraction (%)5.3 ± 2.412.3 ± 4.5*13.7 ± 4.8*14.7 ± 5.5*9.9 ± 2.4*12.8 ± 4.3*

The gelatinolytic activity pattern of proximal and distal tendon regions of the different experimental groups were investigated (Fig. 2). The zymography showed clear bands with relative molecular masses of 72, 64, and 57 kDa that correspond to the pro-, intermediate, and active MMP-2, respectively. No clear bands around 92 or 81 kDa from MMP-9 were observed in this study. The analysis of the integrated optical density (IOD) values (Table 2) demonstrated that pro-MMP-2 gelatinolytic activity increased significantly in the proximal and distal regions of the Achilles tendon after 6 hr and after 4 days of vertical jumping and treadmill running (P < 0.05). Water adaptation increases pro-MMP-2 activity only in the distal region (Fig. 2). No statistically significant differences in the intermediate form of MMP-2 between the different experimental groups were observed in either tendon region (Table 2). However, the active-MMP-2 gelatinolytic activity increased significantly in the proximal region only after vertical jumping (P < 0.05), while distally all physical training increased active MMP-2, especially treadmill running, in which an 11-fold increase was achieved after 6 hr of physical training (P < 0.05) compared with an eightfold increase after 6 hr of vertical jumping (Table 2). Furthermore, in the distal region, active MMP-2 maintained an eightfold increase after 4 days of treadmill running, but dropped to a fourfold increase after 4 days of vertical jumping (P < 0.01) (Table 2).

Figure 2.

Representative gelatin zymography showing that both tendon regions present clear bands at 72, 64, and 57 kDa that correspond to pro-, intermediate, and active MMP-2, respectively. CT, control; ADP, water adapted; VJ1, vertical jumping day 1; VJ4, vertical jumping day 4; TR1, treadmill running day 1; TR4, treadmill running day 4.

Table 2. Densitometric analysis of MMP-2 gelatinolytic bands from zymography of the rat calcaneal tendon regions
Tendon regions and experimental groupsPro- (72 kDa)Inter- (64 kDa)Active (57 kDa)
  1. Values represent mean ± SD × 10,000. Superscript letters indicate statistically significant differences P ≤ 0.05. Upper cases compare the same tendon region in different experimental groups. Lower cases compare different tendon regions in the same experimental group.

Proximal regionControl2.03 ± 0.35 A6.6 ± 0.950.41 ± 0.22 A
Water-adapted2.76 ± 0.41 AB6.98 ± 1.090.31 ± 0.21 Aa
Vertical jumping day 14.12 ± 0.68 C8.95 ± 1.591.48 ± 0.35 Ba
Vertical jumping day 43.1 ± 0.57 BC7.99 ± 1.371.28 ± 0.32 B
Treadmill running day 13.91 ± 0.64 C8.41 ± 1.450.65 ± 0.29 Aa
Treadmill running day 43.14 ± 0.61 BC7.57 ± 1.260.62 ± 0.31 Aa
Distal regionControl2.58 ± 0.49 A6.89 ± 1.020.29 ± 0.19 A
Water adapted3.44 ± 0.61 B7.88 ± 1.331.38 ± 0.31 Bb
Vertical jumping day 14.21 ± 0.78 B8.38 ± 1.482.45 ± 0.52 Cb
Vertical jumping day 43.86 ± 0.66 B8.92 ± 1.541.42 ± 0.41 B
Treadmill running day 13.77 ± 0.67 B8.41 ± 1.513.39 ± 0.72 Cb
Treadmill running day 43.88 ± 0.75 B8.52 ± 1.492.65± 0.55 Cb


In addition to an increased collagen synthesis, the adaptation of tendon to loading involves changes in the quality of the tissue structure, such as an increased collagen cross-linking (Heinemeier et al.,2007). Furthermore, studies have revealed that MMP-mediated collagen degradation also takes place in this process (Riley et al.,2002; Koskinen et al.2004; Marqueti et al.,2006,2008). However, the response to the physical training is more complex in some tissues due to their regional structural variation, as observed in the calcaneal tendon (Benjamin and Ralphs,1998; Covizi et al.,2001).

In the present study, we separately evaluated the two regions of rat calcaneal tendon, proximal, and distal, with regard to the gelatinolytic activity after two different types of physical training. The results presented herein suggest that—despite differences in cell types, extracellular matrix composition and mechanical loading—both tendon regions in control rats possess the same gelatinolytic profile, with similar activities of pro- and intermediate forms of MMP-2. This result agrees with the findings of Marqueti et al. (2006). However, the two tendon regions exhibit different intensity of MMP-2 activation in response to different sets of exercises. Vertical jumping induced MMP-2 activation in both tendon regions, but treadmill running, despite its apparently low intensity, was able to induce a high level of MMP-2 activation in the distal region of calcaneal tendon, but not in the proximal region, where only pro-MMP-2 activity was increased. This greater MMP-2 activation in the distal region of the calcaneal tendon from the animals submitted to treadmill running suggests that the mechanical loading produced by this physical training induces a greater adaptive response from the fibrocartilages than vertical jumping.

The absence of MMP-2 activation in the proximal region after treadmill running does not imply that the tendon is not responding to the mechanical stimulus, since tenocytes and peritendinous sheath cells appeared activated. We may speculate that MMP-2 activation was still not required, but the augmented stock of pro-MMP-2 tissue could have been further activated if mechanical loading stimulus had persisted or increased.

Marqueti et al. (2006,2008) also observed a significant increase in the gelatinolytic activity of MMP-2 in rat calcaneal tendon after a 5-week vertical jumping program with overloading, but observed no differences between proximal and distal regions, as also observed in our study, suggesting that vertical jumping stimulates a similar tissue response in both tendon regions.

In a study of human calcaneal tendon submitted to intensive uphill running, using microdialysis technique to measure the peritendinous tissue gelatinolytic activity, pro-MMP-2 displayed augmentation only after 3 days of exercise (Koskinen et al.,2004). However, the authors suggested that the microdialysis technique could not accurately reflect MMP-2 tissue levels, because, at least, part of MMP-2 is attached to membrane-anchored MMPs (Nagase,1998). On the other hand, Legerlotz et al. (2007) did not observe any regulation of tendon MMP-2 mRNA expression in response to long-term strength training in rats. In our study, the pro-MMP-2 appeared increased after 6 hr of both types of physical training. Thus, the methodology, training program, or the time of sampling used in those studies, compared with the present one, could explain the different results obtained.

Although the major role of MMP-2 is to remove denatured collagen, it may also act directly on intact collagen molecules, since it has been shown to exert some activity against triple helical type I collagen (Aimes and Quigley,1995). Thus, in agreement with Riley et al. (2002) and Marqueti et al. (2006), we also suggest that MMP-2 is the main gelatinase in tendon, where it can be active after a few hours of exercise.

In our study, inflammatory infiltration in the peritendinous sheath was not observed, which may be due to the physical training program used. Herein, we chose a low intensity of treadmill running and vertical jumping without a progressive increase in overloading and with a generous rest interval between each set of jumps in order to avoid an inflammatory reaction, which would interfere significantly with gelatinase activity in tendon tissues. Thus, we believe that the absence of inflammation may explain the undetectable levels of MMP-9 activity observed in the present study. This result agrees with Heinemeier et al. (2007) but contrasts with Koskinen et al. (2004) who found an increase of pro-MMP-9 in peritendinous tissue after 6 hr of treadmill running, an augmentation that persisted for 3 days even without additional training. The authors suggested that MMP-9, besides participating in tissue remodeling, could play a role in a potential inflammatory reaction in the peritendinous connective tissue of calcaneal tendon induced by intensive exercise or by an inflammatory tissue reaction caused by mechanical lesion of microdialysis probe implantation. Marqueti et al. (2006) also observed some MMP-9 activity in rat calcaneal tendon after 5 weeks of vertical jumping under a load equivalent to 70% of body weight, however, these authors also found inflammatory infiltration at the peritendinous sheath.

Despite the lack of any aspect of inflammation, the present study found cellular and vascular changes in peritendinous tissue after exercise. Changes in tendon vascularization have also been accounted for by others as partially an adaptation to exercise, which is induced by an increased VEGF expression (Gavin and Wagner,2001; Pufe et al.,2005; Nakamura et al.,2008). As MMP-2 activity is frequently associated with the process of angiogenesis in several tissues (Lafleur et al.,2003; Rundhaug,2005), the activation of MMP-2 after exercise observed herein may also be related to changes in blood vessel volume fraction at peritendinous sheath.

In conclusion, the present findings indicate that MMP-2 plays a major role in tendon adaptation to physical training and that different types of exercises can differentially stimulate the MMP-2 activation into the calcaneal tendon regions.


This article is part of the PhD Thesis presented by O.C.M.M. to the São Paulo State University—UNESP, Brazil. Special thanks are due to Dr. R. F. Domeniconi and to Dr. H. F. Carvalho for collaboration and helpful discussions.