Regional adaptations in three rattendons

D. Z. Covizi, S. L. Felisbino, L. Gomes, E. R. Pimentel, H. F. Carvalho

Abstract. Although detailed histological and immunocytochemical studies have been published for

the rat calcanear tendon (CT), little is known of the structure, composition and biomechanics of the

deep (DFT) and superficial (SFT) flexor tendons. In this study, we examined the structural

specialization of these three tendons in 90-day-old rats by applying histochemical and biochemical

assays to different tendon regions (proximal, intermediate and distal regions of the DFT and SFT, and

proximal and distal regions of the CT). There were regional differences in tissue structure,

glycosaminoglycan type and content, swelling properties and in the amount and distribution of elastic

fibers. Dermatan sulfate occurred in all regions, but chondroitin sulfate predominated in the

intermediate region of the DFT and in the distal region of the CT. These two chondroitin sulfate-bearing

regions showed swelling in water, while all other regions lost fluid in water. Fibrocartilaginous sites

were observed on the CT, one at the insertion to the bone and another distally at the innermost area of

the tendon. The intermediate region of the DFT showed round cells disposed in lacunae, while the

proximal and distal regions were typically fibrous. The intermediate region of the SFT showed a wavy

array of collagen bundles but neither toluidine blue staining in the matrix nor round cells. Elastic fibers

were present in each region of the three tendons, but were more prominent in the intermediate zone of

the SFT. These results demonstrate regional variation in the three tendons. Tendon differentiation may

occur by an increase in the number of elastic fibers and by variations in the arrangement of collagen

fibers, without fibrocartilage formation. ß 2001 Harcourt Publishers Ltd

Keywords: elastic fibers, glycosaminoglycan, swelling, tendon, tendon fibrocartilage

Tissue & Cell, 2001 33 (5) 483±490

ß 2001 Harcourt Publishers Ltd

DOI: 10.1054/tice.2001.0202, available online at http://www.idealibrary.com

Tissue&Cell

Introduction

The usual histological concept views tendons as very

simple structures consisting of a parallel array of collagen

®bers and elongated ®broblasts or ®brocytes.

Physiologically, tendons are considerd to be very inactive

or quiescent structures. Only recently has the distinction

between tendons and ligaments been made on the basis of

structural and biochemical evidence (Amiel et al., 1984,

1986).

Department of Cell Biology, Institute of Biology, State University ofCampinas (UNICAMP), P. O. Box 6109, 13083±970, Campinas, SP, Brazil

Received 19 September 2000Accepted 22 May 2001

Correspondence to: Hernandes F. Carvalho, Tel./Fax: �55 19 3788 6111;E-mail: hern@unicamp.br

Marked regional variation in tendon morphology and

composition occurs when a tendon is subjected to com-

pressive and frictional forces in addition to the normal

tension exerted by the muscle. Such forces lead to the

development of a ®brocartilage-like structure (Vogel &

Koob, 1989; Benjamin & Evans, 1990; Benjamin &

Ralphs, 1998). This type of mechanical loading and phy-

siological adaptation has been described in many species

(Gillard et al., 1979; Vogel & HeinegaÊrd, 1985; Vogel &

Koob, 1989; Okuda et al., 1987; Ralphs et al., 1991;

Carvalho & Vidal, 1994a), and its differentiation and

maintenance are partly dependent on mechanical stimu-

lation (Gillard et al., 1979; Vogel & Koob, 1989).

In some instances, however, the differentiation of ten-

don regions is less pronounced. Rufai et al. (1992, 1996)

described the ®brocartilages of the calcanear tendon of

the rat and noted variations in the organization. These

483

SFT

CT

DFT

d d

i

p

i

p d

p

1B DFT SFT CT1A

Fig. 1A: Anatomical location of the calcanear tendon (CT), superficial flexor tendon (SFT) and deep flexor tendon (DFT) of the rat hind limb, as seen

from the medial aspect. B: Shows the dissected tendons and indicates the position corresponding to each of their regions. The DFT and SFT were divided

into proximal, intermediate and distal regions, while the CT was divided into proximal and distal regions. The intermediate region of the SFT is flattened

and firmly attached to the calcaneus. The distal region of The SFT is also flattened.

484 COVIZI ET AL.

authors suggested that the accumulation of vimentin in

the cytoplasm and of chondroitin sulfate in the extracel-

lular matrix are the primary responses of a tendon to

compressive forces, and that ®brocartilage may not ap-

pear in some cases of slight compression.

On the other hand, Curwin et al. (1994) described

biochemical modi®cations along the gastrocnemius ten-

don of the growing chicken. Differences in the accumula-

tion of glycosaminoglycans and hydroxypyridinium

cross-linkings, as well as in the diameter of collagen ®-

brils, were reported for the proximal, intermediate and

distal regions of the tendon. However, the authors did not

examine the histology of the different regions. Minor

compositional variations are also observed along equine

tendons (Jones & Bee, 1990).

In the rat, the deep (DFT) and super®cial (SFT) ¯exor

tendons associated with the calcanear tendon (CT)

change direction as they thread around the ankle. While

there are a detailed descriptions of the ®brocartilage in rat

CT (Rufai et al., 1992, 1996), little is known of the struc-

tural and compositional organization of the DFT and

SFT. In this work, we describe some morphological and

biochemical characteristics of these two tendons, and

compare them to those of the CT.

Material and methods

Animals

Forty 90-day-old male Wistar rats were used in this study.

The animals were killed by cervical dislocation after

ether anesthesia, and the calcanear (CT), deep (DFT)

and super®cial (SFT) ¯exor tendons were immediately

dissected out and ®xed for histological procedures or

stored at ÿ208C until used in biochemical analyses. The

tendons (Fig. 1A) were divided into proximal, intermedi-

ate and distal regions (for the DFT and SFT) or proximal

and distal regions (for the CT).

Biochemical analyses

Glycosaminoglycan (GAG) extraction

Tissues were minced with razor blades and digested for

24 h at 608C with papain, according to Harab and

MouraÄo (1989). Undigested material was removed by

centrifugation at 1000 g for 10 min and the soluble mater-

ial in the supernatant was then precipitated with two

volumes of absolute ethanol at ÿ208C overnight. The

precipitated GAGs were resuspended in water and quan-

ti®ed by the metachromatic reaction with dimethyl-

methylene blue (Farndale et al., 1986).

Electrophoresis

Ten micrograms of the sulfated GAGs obtained as above

were subjected to electrophoresis in 0.6% agarose gels

using a diamine propylene buffer (Dietrich & Dietrich,

1977). The standards included in the electrophoretic runs

were chondroitin sulfate from whale cartilage and derma-

tan sulfate from pig skin (5 mg each; both from Sigma

Chemical Co., St Louis, MO, USA). The gels were run

at 100 mA, ®xed with Cetavlon, air dried and stained with

toluidine blue in 50% ethanol. Densitometry was used to

determine the relative amounts of sulfated GAGs re-

solved by electrophoresis.

Swelling tests

The procedure of Koob and Vogel (1987) was used.

Brie¯y, fresh tendon fragments were equilibrated in phos-

phate buffered saline (PBS) for 1 h, then blotted on ®lter

REGIONAL ORGANIZATION OF RAT TENDONS 485

paper and weighed. The tendons were then soaked in

distilled water for another hour, and again blotted and

weighed. Finally, the fragments were soaked in 0.5 M

acetic acid for 15 min, blotted and weighed. This last

incubation time was modi®ed from the original proce-

dure, since prolonged soaking in acetic acid led to almost

complete dissolution of the tendon fragments.

Histology and histochemistry

Tendon fragments were ®xed by immersion in 4% folmal-

dehyde in PBS overnight, washed in PBS, soaked in 70%

ethanol and embedded in JB-4 resin (Polysciences,

Warrington, PA, USA). Two-micrometer sections were

obtained with disposable steel knives, and stained with

0.025% toluidine blue in McIlvaine buffer at pH 4.0 or

subjected to the Weigert fuchsin±resorcin procedure for

the identi®cation of elastic ®bers.

Results

Gross morphology

The deep ¯exor tendon (DFT, Fig. 1b) attaches the m.

¯exor digitorum profundus to the digits. The tendons

threads through a groove in the medial malleolus and

slides considerably on muscle contraction. Distally, the

DFT fans out into ®ve strands that extend to the digits.

The super®cial ¯exor tendon (SFT, Fig. 1b) originates at

the m. ¯exor digitorum super®cialis and wraps around the

calcanear tuberosity before extending to the second, third

and fourth digits. This tendon is ®rmly attached to the

articulating surface by lateral ®brous connections that

DFT SFT

CSDS

O

d i p d i p

Fig. 2 Electrophoretic profile of the GAGs from different regions of the DFT,

The intermediate region of the DFT and the distal region of the CT showed an

was also detected in the other regions but in very low amounts. The chondroit

densitometric scans, shown in Figure 3. CS, chondroitin sulfate standard; DS, de

region. o, origin of the electrophoretic run.

greatly restrict its sliding. The SFT contact region (inter-

mediate region) is ¯attened and less ¯exible than the

corresponding proximal and distal regions. The calcanear

tendon (CT, Fig. 1b) is much shorter than the others, and

attaches the gastrocnemius/soleus to the calcaneus. This

tendon changes direction close to the enthesis, where it is

compressed against the bone. The insertion ®brocartilage

(at the enthesis) and the compressive ®brocartilage de-

scribed by Rufai et al. (1992) are located in the distal

region as de®ned in this work.

Sulfated glycosaminoglycans

The GAGs in the different regions of the three tendons

were separated using the agarose gel/DAP buffer system

(Fig. 2). Dermatan sulfate was found in all regions of each

tendon. The intermediate region of the DFT and the

distal region of the CT showed a prominent band of

chondroitin sulfate, in addition to dermatan sulfate.

Small amounts of chondroitin sulfate contributed to the

total GAGs of the other regions. Figures 3A and 3B show

the contributions of each of the two GAGs to the total

GAG content of each region of the three tendons. The

distal region of the CT not only showed a major contri-

bution in terms of condroitin sulfate but also had the

highest content of GAGs (�1.2 mg/mg of the tissue wet

weight). The proximal region of the same tendon showed

three times less GAG. The smallest amounts of GAG

were found in the proximal regions of the SFT and DFT

(less than 0.2 mg/mg of the tissue wet weight). The inter-

mediate region of the SFT had a similar GAG content to

the intermediate region of the DFT, but differed from the

latter by containing mainly dermatan sulfate.

CT

CSDS

CSDS

O O

d p

SFT and CT. All regions showed a band corresponding to dermatan sulfate.

additional band corresponding to chondroitin sulfate. Chondroitin sulfate

in sulfate bands barely appear in the photographs but are seen in the

rmatan sulfate standard; p, proximal region; i, intermediate region; d, distal

0.0

100

80

60

40

20

0

0.2

0.4

0.6

0.8

1.0

1.2

DS

CSD

FT

p

DF

T i

DF

T d

SF

T p

SF

T i

SF

T d

CT

P

CT

d

DF

T p

DF

T i

DF

T d

SF

T p

SF

T i

SF

T d

CT

P

CT

d

Sul

fate

d G

AG

con

tent

(µg/

mg

of ti

ssue

wet

wei

ght)

Rel

ativ

e co

nten

t of C

S a

nd D

Sin

the

diffe

rent

reg

ions

of t

he te

ndon

A

B

Fig. 3 Glycosaminoglycan types and content in the different regions of

rat DFT, SFT and CT. A: Shows the total amount of sulfated GAGs in

each region and the contribution of dermatan sulfate (DS) and chondroitin

sulfate (CS). The values correspond to the absolute amounts of each GAG

based on their relative contents after densitometry of the gels shown in

Figure 2 and on the total amount of sulfated GAGs as determined by the

dimethylmethylene blue procedure. B: Shows the relative content of

dermatan sulfate and chondroitin sulfate in each tendon region. The distal

region of the CT and the intermediate region of the DFT show a major

contribution of chondroitin sulfate. The intermediate region of the SFT has

as much GAG as the intermediate region of the DFT, but there is a major

contribution of dermatan sulfate.

10

5

0

−5

−10

−15

−20

1000

800

600

400

200

0

Sw

ellin

g in

crem

ent

Sw

ellin

g in

crem

ent

DF

T p

DF

T i

DF

T d

SF

T p

SF

T i

SF

T d

CT

p

CT

d

A

B

Fig. 4 The swelling properties of the different regions of rat calcanear

tendons. A: Shows the changes in wet weight (relative to the initial weight)

after soaking tendon fragments in water. The intermediate region of the

DFT and the distal region of the CT increased in weight, while the other

regions shrank to variable extents. B: Shows the increase in swelling after

soaking in 0.5 M acetic acid. The distal region of the calcanear showed the

smallest gain in wet weight. Similar behavior was observed for the

intermediate regions of the DFT and SFT. The proximal and distal regions

of the SFT swelled by as much as 900%.

486 COVIZI ET AL.

Swelling properties

The swelling of the different regions of the three tendons

is shown in Figure 4. Only the distal region of the CT

and the intermediate region of the DFT swelled in water

(Fig. 4A), the latter showing a 5% increment while the

former increased about 1% in weight. The other tendon

regions actually lost ¯uid in water. As expected from

predominantly ®brous structures, all regions swelled in

0.5 M acetic acid (Fig. 4B), but this swelling was not

homogeneous. Swelling varied from as much as 900%

and 850% for the distal and proximal regions of the

SFT, respectively, to as little as 175% for the distal region

of the CT. Spectrophotometric analysis of the acetic acid

solution after the swelling experiment showed that, even

with the reduced incubation period used in this study,

some collagen was dissolved from the tendon fragments

(results not shown), and this may have contributed to

variations in the ®nal weight.

REGIONAL ORGANIZATION OF RAT TENDONS 487

Histological aspects

The calcanear tendon showed two sites of ®brocartilage,

as described by others. These sites contained round cells

located in lacunae and embedded in a GAG-rich inter®-

brillar matrix, as indicated by the metachromatic staining

with toluidine blue (Fig. 5h, i), and also had ®brous

proximal region (Fig. 5g). The proximal and distal re-

gions of the SFT and DFT were typically ®brous (Figs 5a,

c, d, f). The intermediate region of the DFT showed some

aspects of the ®brocartilaginous array, with some cells

disposed in lacunae and a high accumulation of GAGs in

the inter®brillar space (Fig. 5b). This organization was

restricted to the articulating surface. The intermediate

region of the SFT showed a peculiar array of wavy col-

lagen ®bers, an aspect not seen in the other tendon re-

gions (Fig. 5e). The cells involved were typically

elongated ®broblasts which adapted to the wavy path of

the collagen ®bers. No metachromasy was detected in this

region.

Elastic fiber distribution

Elastic ®bers were rare components of the predominantly

®brous regions of these tendons (i.e. the proximal and

distal regions of the SFP and DFT and the proximal

region of the CT (Figs 6a, c, d, g, h). Most of the elastic

®bers were aligned with the collagen ®bers in these typ-

ically ®brous regions. However, there was a slight in-

crease in these number of elastic components in the

extracellular matrix around the round cells of the inter-

mediate zone of the DFT (Fig. 6b). A marked accumula-

tion of elastic ®bers was also associated with the wavy

collagen ®bers of the intermediate region of the SFT (Fig.

6e). In contrast, the compressive ®brocartilage of the CT

was poor in elastic ®bers (Fig. 6i).

Discussion

The three tendons examined in this work differed in gross

morphology and overall structure. The DFT and SFT

wrapped around bony pulleys at some distance from

their insertions. In contrast, the CT had a short portion

of the calcaneus very close to the enthesis. The distal

region of the DFT was more like a fascia, fanning out

into ®ve branches that extended to the digits. All para-

meters examined in this work indicated that the proximal

regions of the three tendons and the distal regions of the

SFT and DFT were typically ®brous.

The intermediate regions of the DFT and SFT and the

distal region of the CT followed a curved path and are,

presumably, subjected to different levels of compressive

forces, and friction. However, these regions differed from

each other with respect to one or more of the parameters

studied here. The ®brocartilage arrangement described in

the literature (Koob & Vogel, 1989; Benjamin & Evans,

1990; Carvalho & Vidal 1995; Benjamin & Ralphs, 1998)

was observed in the intermediate region of the DFT and

con®rmed at the enthesis of the CT and a short distance

from it, at the site where the CT is compressed against the

calcaneus, as mentioned by Rufai et al. (1992).

Accordingly, these regions showed round-cells in lacu-

nae, metachromatic staining of the inter®brillar matrix,

an increased GAG content with a major contribution of

chondroitin sulfate, and swelling in water. Histological

analysis showed that the DFT had an extensive area of

®brocartilage (all along the intermediate region, but con-

centrated towards the articulating surface), whereas the

®brocartilage of the CT were localized. This arrangement

may re¯ect the relative mobility of the DFT with respect

to the articulating surface. The proximity of the compres-

sive ®brocartilage of the CT to the enthesis restricts the

sliding of this tendon, whereas the DFT is relatively free

to slide. In such a situation, the compressive forces are

dissipated and result in a less evident ®brocartilage mor-

phology compared to the CT.

The swelling properties of the intermediate region of

the DFT are comparable to those of some compressed

areas of the bovine ¯exor tendon (Koob & Vogel, 1987),

but the distal region of the CT did not shrink in water and

did not swell in acetic acid. This behavior differed from

the other ®brous regions, which shrank in water.

The intermediate region of the SFT differed from the

others in many aspects. This region featured a unique

wavy array of collagen ®bers, showed an increased con-

tent of elastic ®bers, did not swell in water and swelled

very little in acetic acid. In addition, the content of GAG

was similar to that of the intermediate region of the DFT,

but there was no metachromatic staining. Dermatan sul-

fate was the main GAG. Structurally, the intermediate

region of the STF was distinct, in that it was ¯attened and

less ¯exible than the proximal and distal regions.

As stated by Benjamin and Ralphs (1998), not every

wrap-around tendon shows compressive ®brocartilages.

These authors stated that a ®brocartilage is absent when

the tendon is pressed against a soft cartilaginous pulley

and is aided by a synovial sheath, or when the tendon only

changes direction (and then experiences compressive

forces) in certain positions of the limb.

The sliding of the SFT is largely restricted by the

®brous extensions that attach it laterally to the calcaneus.

There is no concentration of GAGs (as shown by the lack

of metachromatic staining) and the presence of dermatan

sulfate re¯ects its predominantly ®brous structure. The

presence of an increased number of elastic ®bers may be

related to the constraining of collagen ®bers in a wavy

arrangement (Oakes & Bialkower, 1977; Carvalho, 1995;

Carvalho & Vidal, 1995). As assumed for the distinct

crimp pattern in the tension region of bullfrog tendon

(Carvalho & Vidal, 1994b), this interaction between col-

lagen and elastic ®bers may adapt the tendon to the

wrapping situation in the resting position, but does not

contribute to the ¯exibility of the intermediate region of

the SFT. Finally, the limited swelling in acetic acid may

re¯ect a unique entanglement of the collagen ®bers or an

5a 5b 5c

5d 5e 5f

5g 5h 5i

488 COVIZI ET AL.

6a 6b 6c

6d 6e 6f

6g 6h 6i

REGIONAL ORGANIZATION OF RAT TENDONS 489

Fig. 5 Toluidine blue-stained sections of the proximal (a), intermediate (b) and distal (c) regions of the DFT, the proximal (d), intermediate (e) and distal

(f) regions of the SFT, and the proximal (g) and distal (h & i) regions of the CT. The same fibrous organization was observed in the proximal regions of each

tendon, and in the distal regions of the DFT and SFT. The intermediate region of the DFT contained round cells aligned in lacunae and embedded in a

metachromatic interfibrillar matrix (b: arrowheads). The intermediate region of the SFT showed a wavy distribution of collagen bundles (not seen in the

micrograph) (e). The cell nuclei followed the undulation of the fibers and appeared curved (e: arrowheads). The distal region of the CT had two

fibrocartilages. Part h corresponds to the compressive fibrocartilage found distally in the innermost area of the CT. Part i shows the fibrocartilage found at

the enthesis. Despite minor differences in collagen distribution and in cell size and organization (those at the enthesis were more aligned between the collagen

bundles), the two regions were characterized by intense metachromatic staining of the interfibrillar matrix and by the presence of round chondrocyte-like

cells. Bars � 25 mm.

Fig. 6 Weigert's fuchsin±resorcin stained sections of the proximal (a), intermediate (b) and distal (c) regions of the DFT, the proximal (d), intermediate

(e & f) and distal (g) regions of the SFT, and the proximal (h) and distal (i) regions of the CT. The tensional regions (a, c, d, g & h) have few elastic fibers

associated with the collagen bundles. The pressure-bearing intermediate regions of the DFT (b) and SFT (e & f) and the distal region of the CT (i) differed

from the tensional areas and from each other. The elastic fibers concentrated close to the round cells in the intermediate region of the DFT (b) are randomly

distributed throughout the extracellular spaces in the intermediate region of the SFT (e & f). Elastic fibers are thin and dispersed in the CT fibrocartilage (i).

Bars � 30 mm.

490 COVIZI ET AL.

increased content of reducible cross-links (Koob &

Vogel, 1987, 1989).

The biochemical and morphological data obtained

here highlight the microdomain organization and adap-

tation of different rat tendons, and help to explain the

responsiveness of tendon cells to complex biomechanical

loading.

ACKNOWLEDGMENTS

This work was supported by CNPq and CAPES.

REFERENCES

Amiel, D., Frank, C., Harwood, F., Froner, J. and Akeson, W.H.1984. Tendons and ligaments: a morphological and biochemicalcomparison. J. Orthop. Res., 1, 257±265.

Amiel, D., Kleiner, J.B., Roux, R.D., Harwood, F.L. andAkeson, W.H. 1986. The phenomenon of `ligamentization':anterior cruciate ligament reconstruction with autogenous patellartendon. J. Orthop. Res., 4, 162±172.

Benjamin, M. and Evans, E.J. 1990. Fibrocartilage. J. Anat., 171,1±15.

Benjamin, M. and Ralphs, J.R. 1998. Fibrocatilage in tendous andligaments ± an adaptation to compressive load. J. Anat., 193,481±494.

Carvalho, H.F. 1995. Understanding the biomechanics of tendonfibrocartilages. J. Theor. Biol., 172, 293±297.

Carvalho, H.F. and Vidal, B.C. 1994a. Structure and histochemistry ofa pressure-bearing tendon of the frog. Ann. Anat., 176, 161±170.

Carvalho, H.F. and Vidal, B.C. 1994b. The unique fibrillar arrange-ment of the bullfrog pressure-bearing tendon as an indicative ofgreat functional deformability. Biol. Cell, 82, 59±65.

Carvalho H.F. and Vidal, B.C. 1995. The elastic system of a pressurebearing tendon of the bullfrog. Ann. Anat., 177, 397±404.

Curwin, S.L., Roy, R.R. and Vailas, S.C. 1994. Regional and agevariations in growing tendon J. Morphol., 221, 309±320.

Dietrich, C.P. and Dietrich, S.M.C. 1976. Electrophoretic behaviour ofacidic mucopolysaccharides in diamine buffers. Anal. Biochem.,70, 645±647.

Farndale, R.W., Buttle, D.J. and Barret, A.J. 1986. A improvedquantitation and discrimination of sulphated glycosaminoglycansby use of dimethylmethylene blue. Biochim. Biophys. Acta, 883,173±177.

Gillard, G.C., Reilly, H.C., Bell-Booth, P.G. and Flint, M.H. 1979.The influence of mechanical forces on the glycosaminoglycancontent of the rabbit flexor digitorum profundus tendon. Connect.Tiss. Res., 7, 37±46.

Harab, R.C. and MouraÄ o, P.A.S., 1989. Increase of chondroitin sulfateconcentration in the endochondral ossification cartilage of normaldogs. Biochim. Biophys. Acta, 992, 237±240.

Jones, A.J. and Bee, J.A. 1990. Age- and position-related heterogeneityof equine tendon extracellular matrix. Res. Vet. Sci., 48, 357±364.

Koob, T.J. and Vogel, K.G. 1987. Site-related variations inglycosaminoglycan content and swelling properties of bovine flexortendon. J. Orthop. Res., 5, 414±424.

Koob, T.J. and Vogel, K.G. 1989. Effects of chondroitinase-ABC onproteoglycans and swelling properties of fibrocartilage in bovineflexor tendon. J. Orthop. Res., 7, 219±227.

Oakes, B.W. and Bialkower, B. 1977. Biomechanical and ultrastruc-tural studies on the elastic wing tendon from the domestic fowl.J. Anat., 123, 369±387.

Okuda, Y. Gorski, K.N. and Amadio, P.C. 1987. Biomechanical,histological and biochemical analyses of six anatomical areas ofcanine digitorum profundus tendon. J. Orthop. Res., 5, 231±241.

Ralphs, J.R. Benjamin, M and Thornett, A. 1991. Cell and matrixbiology of the suprapatella in the rat: a structural and immuno-cytochemical study of fibrocartilage in a tendon subjected tocompression. Anat. Rec., 231, 167±177.

Rufai, A., Benjamin, M. and Ralphs, J.R. 1992. Development andageing of phenotypically distinct fibrocartilages associated with therat Achilles tendon. Anat. Embryol., 186, 611±618.

Rufai, A., Ralphs, J.R. and Benjamin, M. 1996. Ultrastructure offibrocartilages at the insertion of the rat Achilles tendon. J. Anat.,189, 185±191.

Vogel, K.G. and HeinegaÊ rd, D. 1985. Characterization of proteogly-cans from adult bovine tendon. J. Biol. Chem., 260, 9298±9306.

Vogel, K.G. and Koob, T.J. 1989. Structural specialization in tendonunder compression. Int. Rev. Cytol., 115, 267±293.