J Appl Physiol 90: 164–171, 2001.

Effects of long-term exercise on the biomechanical properties of the Achilles tendon of guinea fowl CINDY I. BUCHANAN AND RICHARD L. MARSH Department of Biology, Northeastern University, Boston, Massachusetts 02115 Received 12 April 2000; accepted in final form 11 August 2000

Buchanan, Cindy I., and Richard L. Marsh. Effects of long-term exercise on the biomechanical properties of the Achilles tendon of guinea fowl. J Appl Physiol 90: 164–171, 2001.—The purpose of this study was to determine the effect of long-term exercise on tendon compliance and to ascertain whether tendons adapt differently to downhill running vs. running on a level surface. We carried out this investigation on the gastrocnemius tendon of helmeted guinea fowl (Numida meleagris) that were trained for 8–12 wk before commencing experimental procedures. We used an in situ technique to measure tendon stiffness. The animals were deeply anesthetized with isofluorane during all in situ procedures. Our results indicate that long-term exercise increased tendon stiffness. This finding held true after normalization for the cross-sectional area of the free tendon, likely reflecting a change in the material properties of the exercised tendons. Whether training consisted of level or downhill running did not appear to influence response of the tendon to exercise. We hypothesize that the increased stiffness observed in tendons after a long-term running program may be a response to repeated stress and may function as a mechanism to resist tendon damage due to mechanical fatigue. endurance training; downhill running; tendon stiffness; Numida meleagris

remodeling in response to strength or endurance training (see Kannus et al., Ref. 17, for a complete review). Some information exists on structural and biochemical changes of tendon in response to exercise, but this information is inconsistent (9, 15, 20, 36, 42). Studies that have examined mechanical changes of tendon in response to endurance training suggest that ultimate failure strength and stiffness increase with training (31, 36, 42). Viidik (36, 37) studied adaptation of rabbit tibialis posterior, peroneal, and Achilles tendons to 40 wk of training on a running machine. He reported an increase in stiffness of ⬃10% in the Achilles and tibialis posterior tendons. No differences were found in collagen content of the tendons of trained compared with those of untrained animals. He therefore reached the conclusion that tensile differences must be due to qualitative changes such as maturation or collagen type rather than quantitative changes in collagen. In studies of swine digital extensor tendons, Woo et al. (42) TENDON HAS BEEN SHOWN TO UNDERGO

Address for reprint requests and other correspondence: C. I. Buchanan, PT Dept., 6RB, Northeastern Univ., 360 Huntington Ave., Boston, MA 02115 ([email protected]). 164

found that a 12-mo training program that consisted of running at a maximal speed of 2.2 m/s increased ultimate strength by 62%. The authors also noted increased stiffness of the extensor tendons. However, the same training regime had no effect on digital flexor tendons (41). The authors noted that the digital flexors of swine work against large loads, whereas the digital extensors seldom support large loads. The results of this study are perplexing in that changes did not occur in the tendons that were thought to be loaded during running. Simonsen et al. (31) investigated adaptation response of the Achilles tendon of rats to a strength training and a swimming (endurance) training regimen. No differences were reported in force at ultimate tendon failure between 24-mo-old or 29-mo-old untrained and strength-trained groups. However, 29-mo-old swim-trained rats showed tendons with an average failure strength of 56.8 N, which was significantly higher than that (45.0 N) of untrained rats of the same age group. The authors did not measure tendon stiffness. These results are intriguing because they suggest that tendon properties may respond to the number of cycles of loading rather than the load per se. Kubo et al. (21) recently studied elastic properties of the tendon of the vastus lateralis in long-distance runners. They used ultrasonography to measure stiffness of the combined free tendon and aponeurosis in vivo. The authors report a significant difference in stiffness of the vastus lateralis muscle of runners compared with control subjects. The muscle-tendon complex of the runners was ⬃20% stiffer. The purpose of the present study was to determine the effect of long-term exercise on tendon stiffness and to ascertain whether tendons adapt differently to downhill running vs. running on a level surface. On the basis of previous studies, we hypothesized that tendon stiffness would increase in response to endurance training. We carried out this investigation on the gastrocnemius tendon of the helmeted guinea fowl (Numida meleagris). This species was chosen because of its size and running ability. In their native habitat, guinea fowl locomote primarily by running and seldom The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

8750-7587/01 $5.00 Copyright © 2001 the American Physiological Society

http://www.jap.org

EFFECTS OF LONG-TERM EXERCISE ON TENDON

fly (3). The birds used in this study were able to maintain running speeds of up to 3.0 m/s. Furthermore, on the basis of previous studies, we were confident that the gastrocnemius muscle would be loaded during running (7, 11, 28). In contrast to previous studies of tendon stiffness, we measured strain using an in situ procedure that avoids potential errors associated with tendon-clamping techniques. Furthermore, by use of an in situ technique, during strain measurements loading of the tendon more closely simulates loading that occurs during locomotion. Rather than mechanically loading the tendon, we loaded the tendon by muscle stimulation as would occur during natural movement. Finally, we were able to measure strain of the entire free tendon and aponeurosis rather than that of an isolated tendon segment. Previous studies using isolated tendons were restricted to measuring stiffness of the free tendon; however, the mechanical properties as well as adaptability of tendon may differ in the aponeurosis vs. the free tendon (10, 22, 24, 27, 35). Strain measures of isolated tendon segments may be in error if strain is not uniform throughout the entire length of the tendon and aponeurosis. Although some studies report that the muscle aponeurosis and free tendon share similar mechanical properties (24, 27, 35), others report differing degrees of stiffness in the free tendon compared with the aponeurosis (10, 22). We included a downhill-running group in our study because downhill running has been suggested to increase active stretch (eccentric contraction) of the limb muscles. Strenuous exercise that involves active muscle lengthening results in greater muscle damage than is associated with concentric or isometric contraction. (See Stauber, Ref. 32, for a review of the response of muscle to active lengthening.) Although several theories have been proposed, the actual mechanism by which active lengthening induces injury is unknown. Muscle damage attributed to active lengthening can be substantially reduced with repeated exercise, although the mechanisms of adaptation are unclear (8, 25). Muscle fibers may adapt to withstand the effects of stretch during activity and/or the connective tissue elements may be altered to limit stretching of the active fibers. When the muscle-tendon complex is lengthened, the imposed stretch is distributed between muscle fibers and tendon according to the respective stiffness of each structure. If muscle fibers are attached to a compliant tendon, the tendon could potentially absorb a large amount of stretch and thereby limit muscle fiber strain. Alternately, if the tendon is stiff, imposed stretch may lead to excessive fiber lengthening and structural damage to sarcomeres. It is conceivable that changes in response to repeated active lengthening occur in the tendon. If the tendon were to become less stiff as a result of training, muscle damage due to active strain of the myotendinous unit may be reduced.

165

METHODS

Animals and Animal Conditioning Adult guinea fowl were obtained from a local supplier and housed in cages that measured ⬃1 m ⫻ 0.5 m ⫻ 0.5 m. Either one or two birds were housed in each cage. Food and water were provided ad libitum. The birds were arbitrarily assigned to a control group (n ⫽ 4) or one of two exercise groups. Exercise consisted of running on either a level or downwardinclined motor-driven treadmill. Previous studies indicate that training effects on tendon require 30–60 min of exercise 5 days/wk for a minimum of 6 wk (9, 34). For this study, eight birds ran 5 days/wk for a minimum of 8 wk to a maximum of 12 wk. Each exercise session lasted 30 min. Level-running birds (n ⫽ 4) initially ran at a speed of 1.5 m/s. Speed was increased over a 4-wk period to a maximum of 2.5 m/s. Birds ran a minimum of eight additional weeks at 2.5 m/s. Birds in the downhill-running group (n ⫽ 4) initially ran on a 5% downward slope at a speed of 1.5 m/s. The slope of the treadmill was increased over a 4-wk period to a maximum slope of ⫺16%. Birds ran a minimum of eight additional weeks on the maximal downhill slope. Downhill-running birds increased running speed as described for the levelrunning birds. Control animals were kept caged. The weight of the birds was similar between groups and ranged from ⬃1.2 to 1.4 kg. Muscle Architecture The gastrocnemius is a superficial muscle of the shank that originates from the femur by three heads. The lateral head originates from the lateral femoral condyle and a ligamentous loop associated with the iliofibularis muscle. The lateral gastrocnemius (LG) is a pennate muscle in which the fibers extend the length of the muscle fascicles and have an average length of ⬃16 mm. Pennation angle was measured with a protractor under a dissecting microscope. The pennation angle of muscles that had been fixed at resting length was ⬃18°. Fibers of the LG insert on an aponeurosis that forms a long free tendon that passes posterior to the ankle before inserting on the tarsometatarsus. The aponeurosis of the LG is almost twice as long as the free tendon, which measures 35–40 mm. For the purposes of this study, we use the term “tendon” to refer to the entire free tendon-aponeurosis unit. The terms “free tendon” and “aponeurosis” refer to those discrete portions of the tendon. The LG acts to extend the ankle joint and also has a flexor moment at the knee. In Situ Measurement of Tendon Compliance Tendon compliance was measured in situ. The procedure involved measuring fiber length and the amount of force generated by the LG of an immobilized hindlimb. Eleven guinea fowl were used in this study; four birds were from the downhill-trained group, three were from the level-running group, and four were control animals. The animals were deeply anesthetized with isofluorane during all surgical procedures. Core temperature was maintained by placing a heating pad under the anesthetized bird. Measurement of muscle fiber length. We measured muscle fiber length by use of a Triton model 120 sonomicrometer (Triton Technology, San Diego, CA). Leads to these transducers were routed from a multipin connector (Microtech) on the back of the animal to the site of the implants. Using sterile techniques, we surgically implanted 1-mm-diameter hemispherical sonomicrometer crystals in the proximal portion of the LG just distal to the knee joint. The crystals were at-

166

EFFECTS OF LONG-TERM EXERCISE ON TENDON

tached to stainless steel holders designed by Olson and Marsh (26) that were sutured to connective tissue with 6-0 gauge silk. We inserted the crystals to a depth of ⬃3–4 mm and aligned them 8–14 mm apart parallel to muscle fascicles. Before implantation, each set of sonomicrometer crystals was calibrated by affixing the transducers to dial calipers and immersing them in physiological saline. This calibration corrects for offset errors. We assumed that the sutured arms of the holders were the points that moved relative to one another during the measurements. Although the change in length is accurately recorded by the sonomicrometer, the absolute length is underestimated because of the presence of the epoxy lens and epoxy between the holder and the crystal. The total offset error for a pair of crystals in our experiment averaged 1.95 mm. Crystals were implanted along a segment of the fascicle length. Fascicle length was measured with calipers. We assumed that fiber shortening was uniform, and we converted changes in segment length to changes in fiber length by multiplying the change in segment length times the ratio of fiber length to resting segment length. We adjusted for pennation angle by multiplying fiber length changes by the cosine of 18° (0.951). After implantation of the sonomicrometer crystals, the birds were allowed to recover, and in vivo measures of muscle strain were recorded (unpublished data). Subsequently, the birds were reanesthetized for the experiments described here. Measurement of muscle force. Skin from the hindlimb was opened to expose the gastrocnemius muscle. During subsequent procedures, the muscle was kept moist with saline. The gastrocnemius was freed from surrounding muscles, and the medial head of the gastrocnemius was disconnected from the Achilles tendon. The aponeurotic connection between the two heads of the gastrocnemius was split to ensure that only fibers from the lateral head could transfer force to the tendon. The tarsometatarsus was severed ⬃2 cm from the ankle joint, and the remaining proximal portion of this bone with the attachment of the Achilles tendon was freed from the ankle joint. All other tendons were removed from the remaining bone fragment. The bone fragment was then attached to a quartz force transducer (Kistler model 9203) via a stainless steel screw (Fig. 1). Neural stimulation of muscle contraction. We located the tibial nerve in the distal thigh just proximal to the knee. We carefully coiled two silver wire stimulating electrodes around the nerve. The nerve was severed proximal to the stimulating electrodes, and the cavity in which the nerve lay was filled with mineral oil. The electrodes were connected to a Grass S48 stimulator. Limb immobilization. The hindlimb was mounted on two aluminum plates (Fig. 1). Holes were drilled in the proximal

Fig. 1. Schematic representation of experimental setup. The lateral gastrocnemius and force transducer were fixed onto the same aluminum plate. Length changes in the myotendinous unit were measured by using sonomicrometry techniques. Force was measured by a transducer attached to the tarsometatarsus.

and distal portions of the femur. The bone was then secured to an 8-mm-thick plate with two stainless steel screws. The tibia was similarly secured to a 4-mm-thick plate. The femoral plate extended several inches distal to the knee joint and had a central cleft. The tibial plate was bolted to the femoral plate via the central cleft. When the bolt between the two plates was loosened, it could be manually glided along the cleft, allowing for changes in the knee joint angle. However, when the two plates were bolted firmly together, the knee joint was immobile. The force transducer was screwed into a 2.5-cm aluminum block, which was bolted to a cleft in the distal end of the tibial plate. Thin aluminum plates placed between the block holding the force transducer and the tibial plate were used to adjust for small differences in limb size and to ensure proper alignment of the muscle-tendon unit with the force transducer. Experimental procedure. Contraction of the LG was stimulated via the tibial nerve. Supramaximal stimuli of 0.2-ms duration were delivered at a frequency of 150 Hz within a 400-ms train. A series of 9–20 contractions was performed with the muscle at different lengths. Muscle length was adjusted by manually moving the block to which the force transducer and fragment of the tarsometatarsus were attached along the cleft of the tibial plate. The muscle was rested at least 2 min between each contraction. During each contraction, the segment length between the sonomicrometer crystals and force output were recorded (Fig. 2). Output from the sonomicrometer and the force transducer was acquired by a MacAdios II analog-to-digital converter running in a Macintosh IICi microcomputer and stored for subsequent analysis. At the conclusion of the experimental procedures, the bird was killed with an overdose of Nembutal. We then dissected out the LG muscle and tendon, weighed them, and measured tendon and fascicle length using calipers. The tendons were stored in plastic bags at ⫺20°C until sectioned. The tendons were rehydrated in saline to their original weight before sections to be used for cross-sectional studies were cut. Portions of the tendon to be used for cross-sectional area measurements were embedded in Tissue-Tek OCT compound and quick frozen in liquid nitrogen. The tendon blocks were stored in glass vials at ⫺70°C until needed. Tissue specimens were obtained from two areas along the length of the free tendon. Two or three sections were cut from each area and visualized by use of a macro lens attached to a black-and-white charge-coupled device camera. Images were captured with a Data Translation Quick Capture Video board and displayed on a monitor. We measured cross-sectional area using the application NIH Image. To calculate the cross-sectional area of the muscle, we used the formula mass 䡠 density⫺1 䡠 fascicle length⫺1. We assumed that the density of muscle was 1,060 kg/m3 (23).

EFFECTS OF LONG-TERM EXERCISE ON TENDON

Fig. 2. Representative recordings of muscle force and fiber segment length change during a single isometric contraction. Muscle contraction was stimulated via the tibial nerve with a supramaximal stimulus delivered at a frequency of 150 Hz.

Data analysis. Because the muscle-tendon unit was isometric during contractions of the LG, we were able to equate the amount of fiber shortening to the amount of tendon lengthening. We plotted a load-deformation curve for each muscle contraction. The stiffness reported for each animal was the mean value for three to six contractions starting at or above the length that produced maximum force. On each load-deformation curve, we measured the slope on the linear portion of the curve (Fig. 3). We calculated tendon strain (⑀) as the amount of tendon lengthening divided by resting tendon length (Lt; dl/Lt ⫽ ⑀). We calculated stress (␴) by dividing the force (F) generated during contraction by the cross-sectional area of the tendon (Ao; F/Ao ⫽ ␴). The resultant stress-strain curve provided a measure of stiffness (tan ␣) that is independent of tendon size (Young’s modulus). A nonparametric one-way analysis of variance (KruskalWallis test) was used to test for differences in tendon stiffness and cross-sectional area. The level of significance was set at 5%. Whenever P ⱕ 0.05, differences between groups of animals were tested by using the Mann-Whitney test.

167

curves for the tendon (Fig. 3). Comparison of slopes (Table 1) taken from the load-deformation curves of all animals before normalization reveals that the tendons of both the level- and downhill-running birds were significantly stiffer than those of control birds (P ⬍ 0.0001). Differences in tendon stiffness between leveland downhill-trained birds were not significant. Cross-sectional areas of the free tendon were compared to assess whether hypertrophy of this portion of the tendon contributed to the increased stiffness exhibited by the tendons of exercised birds. We found that the cross-sectional areas of tendons taken from trained birds were not significantly different from those of control birds (Table 1). As would be expected given the lack of difference in cross-sectional area, comparison of elastic moduli obtained from the stress-strain curves (Fig. 4) reveals that the tendons of the exercised birds were significantly stiffer (P ⬍ 0.0001) than those of control animals regardless of size (Table 1). Once again, tendon stiffness of downhill-running birds was not significantly different from that of level-running birds. Finally, to test whether the increase in tendon stiffness of trained birds might be in response to the production of greater muscle forces due to an increased muscle size, we compared the cross-sectional area of the LG of trained birds to that of control animals (Table 1). We found no significant difference between the cross-sectional areas of muscles obtained from exercised vs. control animals. DISCUSSION

Results Compared With Previous Studies Our values for modulus of elasticity (stiffness) are slightly lower than previously reported ranges (Table

RESULTS

Data analysis was limited to data obtained from four control birds, four downhill-trained birds, and three level-running birds. In situ data from one level-trained bird were not reliable and were therefore not included in the results. Figure 2 shows a representative set of data collected for a single contraction. Because the muscle-tendon length was fixed, shortening of the muscle fibers represents lengthening of the tendon. After correction for fascicle length, the force and length data obtained from these contractions were replotted as load-deformation

Fig. 3. Representative load-deformation curves plotted from in situ data obtained from a control bird and a level-running trained bird. Recorded muscle length changes have been converted to changes in tendon length. Slope was calculated from the linear portion of the curve lying between the arrows. The slope of the curve is the stiffness. Amount of energy absorbed by tendon during elongation is represented by the area under the curve.

168

EFFECTS OF LONG-TERM EXERCISE ON TENDON

Table 1. Comparison of stiffness and cross-sectional area of the lateral gastrocnemius tendon from control and exercised animals Group

Slope, N/mm

Tan␣, GPa

At, cm2

Am, cm2

Control Level running Downhill running

50.580 ⫾ 10.624 77.808 ⫾ 21.841* 88.054 ⫾ 12.938*

0.338 ⫾ 0.011 0.629 ⫾ 0.043* 0.564 ⫾ 0.036*

0.0645 ⫾ 0.003 0.0608 ⫾ 0.003 0.0613 ⫾ 0.001

4.7873 ⫾ 0.306 4.0665 ⫾ 0.249 4.4352 ⫾ 0.183

Values are means ⫾ SE. Slope, ratio of force to deformation; tan␣ (Young’s modulus), ratio of stress to strain; At, tendon cross-sectional area; Am, muscle cross-sectional area. * Significant difference compared to the control group (P ⬍ 0.0001).

2). Our values range from 0.338 ⫾ 0.011 GPa for control animals to 0.629 ⫾ 0.043 GPa for trained birds. In a review of the mechanical properties of tendon, Zajac (43) reported that the modulus of elasticity in the linear region of the stress-strain curve is generally 0.6–1.7 GPa. Several of the studies cited (5, 19, 30) measured strain of only the free tendon. In contrast, we measured strain of both the free tendon and the aponeurosis. The lower moduli of elasticity that we have reported may reflect a difference in the free tendon vs. the aponeurosis. Ito et al. (16) used ultrasonography to measure compliance of the tibialis anterior aponeurosis and free tendon in humans. They determined that the modulus was 0.53 GPa. This value is within the range of stiffness that we measured for the entire gastrocnemius aponeurosis and free tendon. Few previous studies have reported changes in tendon compliance associated with long-term training. Viidik (36, 37) reported increases of ⬃10% in stiffness of the tibialis posterior and Achilles tendons of rabbits that had been trained for 40 wk on a running machine. All of his reported values were corrected for crosssectional area of the tendons. Similarly, we observed significant increases in tendon stiffness of ⬎40% in birds that had undergone a training program. Although the animals used in Viidik’s studies were ma-

ture at the time when tendon samples were collected, the training regime took place during their growth period. Walker et al. (39) and Shadwick (30) report that, after correction for cross-sectional area, adult dog and pig tendons have stiffer stress-strain characteristics than tendons of immature animals of these species. Woo et al. (42) reported increased stiffness in the digital extensor tendons of swine that had exercised for 1 yr on a motorized treadmill; however, they found no change in the stiffness of digital flexor tendons (41). They suggest that the different responses to exercise may be attributed to differing biochemical composition. Compared with extensor tendons, the flexor tendons of control animals had lower contents of fat, water, and other noncollagenous proteins. The authors propose that during exercise these noncollagenous materials are diminished in extensor tendons, thus increasing the concentration of collagen and correspondingly changing the mechanical properties attributed to collagen. However, because the flexor tendons have already-low amounts of noncollagenous materials, the concentration is not further reduced and hence changes in collagen concentration are not stimulated. The authors further suggest that hypertrophy of the flexor tendons is not desirable because they lie within a synovial sheath and hypertrophy would constrain gliding of the tendon within the sheath. Woo et al. (41) noted that during running the digital flexors of swine work against large loads, whereas the digital extensors seldom support large loads. However, the authors do not supply clear evidence of this loading pattern. We collected EMG and video recordings from running birds that provide us with evidence that the gastrocnemius muscle of running guinea fowl is active during stance (unpublished data). Contrary to the previous authors, we did find significant increases in stiffTable 2. Young’s modulus reported in present study compared with results of previous studies

Author

Bennet et al. (4) Ker et al. (19) Fig. 4. Representative stress-strain curves plotted from in situ data (see Fig. 3) obtained from a control bird and a level-running trained bird after correction for cross-sectional area of the free tendon. Slope was calculated from the linear portion of the curve lying between the arrows.

Shadwick (30) Ito et al. (16) Buchanan and Marsh (present study)

Tendon

Young’s Modulus, GPa

Deer Wallaby Wallaby Pig Pig Human

Gastrocnemius Gastrocnemius Plantaris Flexor Extensor Tibialis anterior

0.92–1.00 1.18–1.90 0.59–2.54 1.66 0.76 0.53

Guinea fowl

Gastrocnemius

0.33–0.63

Species

EFFECTS OF LONG-TERM EXERCISE ON TENDON

ness of these load-bearing tendons. The discrepancy between our results and those of Woo et al. (41) may reflect differences in actual loading patterns, differences in the training regime, or species differences. Structural vs. Material Changes in Tendon Increased tendon stiffness may be attributed to hypertrophy or to a change in collagen content, composition, and/or cross-linking. In comparing the tendons of exercised rabbits to those of control animals, Viidik (36) found no difference in fresh weight, dry weight, water content, or collagen concentration. Like Viidik, we found no increase in the fresh weight of the free tendons of exercised animals. Woo et al. (42) reported an increase in collagen concentration and an increase in the cross-sectional area of the extensor tendons of exercised animals. Contrary to these authors, we did not find in increase in cross-sectional area of the free portion of the tendons taken from trained animals. The lack of change in cross-sectional area suggests that in guinea fowl tendons adapt by changes in material properties. This conclusion should be tempered by the possibility that the cross-sectional area of the aponeurosis could change independently of the free tendon. Our measures of cross-sectional area are limited to the free tendon; however, our values for stiffness are based on measures of stress and strain involving both the free tendon and aponeurosis. Therefore, our calculations of stiffness are with the assumption that the cross-sectional area of the free tendon is similar to that of the aponeurosis. This assumption is supported by a single study of kangaroo and wallaby plantaris tendons, in which Alexander and Vernon (2) reported that the cross-sectional area of the free tendon was similar to that of the aponeurosis. The relationship between size of the free tendon and that of the aponeurosis needs further study. Further investigations should also examine the possibility that independent changes occur in these two distinct compartments. The mechanical properties of connective tissue represent those of collagen fibers themselves. Evidence indicates that mechanical failure in collagen is due not to breakage of tropocollagen molecules but to pulling apart of adjacent molecules (38). Thus intermolecular cross-linking is in important factor in determining the elastic modulus and tensile strength of collagen. Changes in the material properties of tendon associated with exercise may reflect a change in collagen cross-links. Increased tendon stiffness may also reflect a change in the type of collagen that is deposited. Why Alter Tendon Stiffness? Tendon is a viscoelastic tissue that responds in a predictable manner when loaded. If the tendon is elongated at a constant rate while changes in force are recorded, a load-deformation curve can be plotted. The load-deformation curve may be adjusted by dividing force by original cross-sectional area (stress) and deformation by initial length (strain). Mechanical param-

169

eters taken from the stress-strain curve include maximum stress to failure, elastic modulus (stiffness), and energy absorption. Amount of energy absorbed by the tendon during elongation is proportional to the area under the curve (Fig. 4). The mechanical properties of tendons suggest that they do more than simply transmit force to the bones. Elastic energy storage and subsequent release by tendons is hypothesized as a mechanism used to enhance economic locomotion (1, 5, 28, 33). The lateral gastrocnemius of turkeys has been found to operate nearly isometrically during the stance phase of level running, thus facilitating effective energy storage in its tendon (28). In a study of the mechanics of wallaby and kangaroo hopping, the muscle-tendon units identified as possible sites of energy storage were the gastrocnemius, plantaris, and digital flexor muscles (19, 24). Like the LG of guinea fowl, these muscles have in common short muscle fibers and long tendons that are well suited to act as storage sites for elastic energy (4, 18). However, the degree of tendon stiffness needed to maximize elastic energy storage is not known. Although under appropriate conditions increased tendon stiffness can provide the benefit of increased capacity to store elastic energy, it is doubtful that this is the impetus for tendon remodeling. By increasing stiffness (the slope of the curve), the amount of energy stored can be substantially increased, but only if the force applied also substantially increases. If the same force is applied (e.g., Fig. 3), less energy will be stored in the stiffer tendon. We consider it unlikely that muscle forces increase during training. In comparing LG muscle cross-sectional area of exercised birds to that of control birds, we found no indication that the muscles of exercised birds had hypertrophied and would be better suited to exert higher forces. Thus increases in tendon stiffness do not appear to be necessarily related to transmitting higher forces. The findings of Kubo et al. (21) support our hypothesis that increased tendon stiffness is not associated with increased elastic energy storage or muscle strength. These authors reported significantly higher stiffness of the vastus lateralis tendon of long-distance runners compared with control subjects. However, during jumping movements, the same study revealed lower elastic energy storage potential in runners. Furthermore, the authors report no differences in torque generated during maximal voluntary contraction between runners and nonrunners. We hypothesize here that increases in tendon stiffness observed after endurance training may not be associated with a requirement for increased strength, but rather might represent a mechanism to resist tendon damage due to mechanical fatigue. Fatigue due to repeated application of stress is a phenomenon that is widely documented for inorganic materials (13). Recently, Wang et al. (40) reported fatigue behavior in the tail tendon of wallabies. They found that, when subjected to cyclical loading, the tendons failed under much lower stresses than would be sufficient to rupture them in a single pull. Schechtman and Bader (29)

170

EFFECTS OF LONG-TERM EXERCISE ON TENDON

reported similar fatigue behavior in the extensor digitorum longus tendons of humans. We suggest that tendon remodeling producing stiffer tendons in guinea fowl may be in response to the increased number of loading cycles. If the stiffer tendons are more resistant to fatigue, this adaptation would serve to keep in balance the damage and repair process necessary to maintain tendon integrity. We believe that this hypothesis is consistent with data showing that tendons in rats respond to increases in the number of loading cycles. Simonsen et al. (31) found that a strength training regime (high force over a few loading cycles) carried out three times daily, 4 days/wk over a 38-wk period did not stimulate increases in strength of the Achilles tendon of rats. However, 90 min of swimming 4 days/wk over a 15-wk period resulted in stronger tendons. Swim training involved several hundred thousand contractions over the entire training period, whereas strength training involved ⬃1,500 contractions (an average of 5–6 contractions per day). The training regime we designed for the guinea fowl involved an average of 5,400 contractions per exercise period or ⬃200,000–300,000 contractions over the entire training period. Simonsen et al. (31) suggest that the tendons may respond to the total number of muscle contractions that occur during training rather than the absolute tension exerted by the muscle. We further suggest that increased stiffness associated with training may be related to preventing tendon fatigue. We did not observe decreases in tendon stiffness as a result of downhill running. However, muscle damage associated with active stretch may be dependent on the amount of strain incurred and/or where on its lengthtension curve the muscle operates during active stretch. During downhill running, the fibers of the LG were found to undergo active stretch of ⬍10% and operated on the ascending limb of the length-tension curve (unpublished results). These factors may have limited muscle damage, which would negate our original hypothesis that increased tendon compliance could be a mechanism to prevent damage. In this study, as well as previous studies (36, 37, 41, 42), control animals were kept caged and were not allowed the amount of activity that would normally occur in an uncontrolled environment. Thus our training regime involved a large increase in tendon loading and the number of cycles of loading. Currently, the amount of activity required to stimulate changes in the mechanical properties of tendon is unclear, as is the time course of such changes. Hopefully, future studies will allow us to investigate the optimal parameters to induce tendon adaptation. In conclusion, our results indicate that long-term exercise increased tendon stiffness in the load-bearing lateral gastrocnemius of guinea fowl. This finding held true after adjustment for the cross-sectional area of the free tendon, likely reflecting a change in the material properties of the exercised tendons. The lack of tendon hypertrophy also suggests that a change in the material properties of the tendon took place. However, the

possibility also exists that the cross-sectional area of the aponeurotic portion of the tendon substantially increased its cross-sectional area independent of the free portion of the tendon. We hypothesize that the increased stiffness observed in tendons after an endurance-training protocol may be a response to repeated stress and function as a mechanism to resist tendon damage due to mechanical fatigue. We thank Jill Boorman, Jennifer Cann, Christopher Cesario, Jason Fogu, Cory Grant, John Leary, Kerri Marron, Heather Reynolds, Stevan Simon, and Diane Weaver for assisting in the training protocol. Support for this project was provided by National Institute of Arthritis and Musculoskeletal and Skin Diseases grant AR-39318 to R. L. Marsh and a Northeastern University Research and Development award to C. I. Buchanan and R. L. Marsh. REFERENCES 1. Alexander RM and Bennet-Clark CH. Storage of elastic strain energy in muscle and other tissues. Nature 265: 114–117, 1977. 2. Alexander RM and Vernon A. The mechanics of hopping by kangaroos (Macropodidae). J Zool Lond 177: 265–303, 1975. 3. Ayeni JSO. Home range size, breeding behavior, and activities of helmeted guinea fowl (Numida meleagris) in Nigeria. Malimbus 5: 37–43, 1983. 4. Bennett MB, Ker RF, Dimery NJ, and Alexander RM. Mechanical properties of various mammalian tendons. J Zool Lond 209: 537–548, 1986. 5. Cavagna GA, Heglund NC, and Taylor CR. Mechanical work in terrestrial locomotion: two basic mechanisms for minimizing energy expenditure. Am J Physiol Regulatory Integrative Comp Physiol 233: R243–R261, 1977. 7. Clark J, and Alexander RM. Mechanics of running quail (Coturnix). J Zool Lond 176: 87–113, 1975. 8. Clarkson PM, Nosaka K, and Braun B. Muscle function after exercise-induced muscle damage and rapid adaptation. Med Sci Sports Exerc 24: 512–520, 1992. 9. Curwin SL, Vailas AC, and Wood J. Immature tendon adaptation to strenuous exercise. J Appl Physiol 65: 2297–2301, 1988. 10. Delgado-Lezama R, Raya JG, and Munoz-Martinez EJ. Methods to find aponeurosis and tendon stiffness and the onset of muscle contraction. J Neurosci Methods 78: 125–132, 1997. 11. Gatesy SM. Guinea fowl hindlimb function. II. Electromyographic analysis and motor pattern evolution. J Morphol 240: 127–142, 1999. 13. Hearle JWS. Fatigue in fibers and plastics (a review). J Mater Sci 2: 474–488, 1967. 15. Inglemark BE. The structure of tendons at various ages and under different functional conditions. II. An electronmicroscopic investigation from white rats. Acta Anat (Basel) 6: 193–225, 1948. 16. Ito M, Kawakami Y, Ichinose Y, Fukashiro S, and Fukunaga T. Nonisometric behavior of fascicles during isometric contractions of human muscle. J Appl Physiol 85: 1230–1235, 1998. 17. Kannus P, Jozsa L, Natri A, and Jarvinen J. Effects of training, immobilization and remobilization on tendons. Scand J Med Sci Sports 7: 67–71, 1997. 18. Ker RF. Dynamic tensile properties of the plantaris tendon of sheep (Ovis aries). J Exp Biol 93: 283–302, 1981. 19. Ker RF, Dimery NJ, and Alexander RM. The role of tendon elasticity in a hopping wallaby (Macropus rufogriseus). J Zool Lond 208: 417–428, 1986. 20. Kiiskinen A. Physical training and connective tissue in young mice—physical properties of Achilles tendons and long bones. Growth 41: 123–137, 1977. 21. Kubo K, Kanehisa H, Kawakami Y, and Fukunaga T. Elastic properties of muscle-tendon complex in long-distance runners. Eur J Appl Physiol 81: 181–187, 2000.

EFFECTS OF LONG-TERM EXERCISE ON TENDON 22. Lieber RL, Leonard ME, Brown CG, and Trestik CL. Frog semitendinosis tendon load-strain and stress-strain properties during passive loading. Am J Physiol Cell Physiol 261: C86–C92, 1991. 23. Mendez J and Keyes A. Density and composition of mammalian muscle. Metabolism 9: 184–188, 1960. 24. Morgan DL, Proske U, and Warren D. Measurements of muscle stiffness and the mechanism of elastic storage of energy in hopping kangaroos. J Physiol (Lond) 282: 253–261, 1978. 25. Nosaka K, Clarkson PM, McGuiggin ME, and Byrne JM. Time course of muscle adaptation after high force eccentric exercise. Eur J Appl Physiol 63: 70–76, 1991. 26. Olson JM and Marsh RL. Activation patterns and length changes in the hindlimb muscles of the bullfrog Rana catesbeiana during jumping. J Exp Biol 201: 2763–2777, 1998. 27. Racke PMH and Westbury DR. Elastic properties of the cat soleus muscle and their functional importance. J Physiol (Lond) 347: 479–495, 1984. 28. Roberts TJ, Marsh RL, Weyand PG, and Taylor CR. Muscular force in running turkeys: the economy of minimizing work. Science 275: 1113–1115, 1997. 29. Schechtman H and Bader DL. In vitro fatigue of human tendons. J Biomech 30: 829–835, 1997. 30. Shadwick RE. Elastic energy storage in tendons: mechanical differences related to function and age. J Appl Physiol 68: 1033–1040, 1990. 31. Simonsen EB, Klitgaard H, and Bojsen-Moller F. The influence of strength training, swim training and ageing on the Achilles tendon and m. soleus of the rat. J Sports Sci 13: 291–295, 1995. 32. Stauber WT. Eccentric action of muscles: physiology, injury, and adaptation. Exerc Sport Sci Rev 17: 157–185, 1985. 33. Taylor CR. Force development during sustained locomotion: a determinant of gait, speed and metabolic power. J Exp Biol 115: 253–262, 1985.

171

34. Tipton CM, Vailas AC, and Matthes RD. Experimental studies on the influences of physical activity on ligaments, tendons and joints: a brief review. Acta Med Scand Suppl 711: 157–168, 1986. 35. Trestik CL and Lieber RL. Relationship between Achilles tendon mechanical properties and gastrocnemius muscle function. J Biomech Eng 115: 225–230, 1993. 36. Viidik S. The effect of training on the tensile strength of isolated rabbit tendons. Scand J Plast Reconstr Surg Hand Surg 1: 141–147, 1967. 37. Viidik S. Tensile strength properties of Achilles tendon systems in trained and untrained rabbits. Acta Orthop Scand 40: 261– 272, 1969. 38. Wainwright SA. Mechanical Design in Organisms. Princeton, NJ: Princeton Unversity Press, 1967, p. 81–93. 39. Walker P, Amstutz HC, and Rubenfeld M. Canine tendon studies. II. Biomechanical evaluation of normal and regrown canine tendons. J Biomed Mater Res 10: 61–76, 1976. 40. Wang XT, Ker RF, and Alexander RM. Fatigue rupture of wallaby tail tendons. J Exp Biol 198: 847–852, 1995. 41. Woo SLY, Gomez MA, Amiel D, Ritter MA, Gelberman RH, and WH. The effects of exercise on the biomechanical and biochemical properties of swine digital flexor tendons. J Biomech Eng 103: 51–56, 1981. 42. Woo SLY, Ritter MA, Amiel D, Sanders TM, Gomez MA, Kuei S, Garfin SR, and Akeson WH. The biomechanical and biochemical properties of swine tendons—long term effects of exercise on the digital extensors. Connect Tissue Res 7: 177–183, 1980. 43. Zajac FE. Muscle and tendon: properties, models, scaling and application to biomechanics and motor control. Crit Rev Biomed Eng 17: 359–411, 1989.