Journal of Orthopaedic Research 8:13-20 Raven Press, Ltd., New York 0 1990 Orthopaedic Research Society
Collagen Synthesis During Primate Flexor Tendon Repair In Vitro Jean E. Russell and Paul R. Manske Division of Orthopedic Surgery, Department of Surgery, Washington University School of Medicine, St. Louis, Missouri, U.S.A.
Summary: The concept that flexor tendons have the intrinsic capacity to participate actively in the repair process following laceration has been developed in recent years as the result of experimental studies from numerous laboratories. However, the role of the outerlepitenon and innerlendotenon cell populations with regard to protein synthesis is still controversial. The purpose of this study was to investigate the respective participation of these tendon fibroblast populations in the synthesis of the collagen matrix during in vitro repair of the flexor tendon from nonhuman primates, utilizing immunohistochemical techniques and a Type I procollagen antibody. Zone I1 profundus flexor tendon segments were obtained from young adult Macaca nemestrina monkeys. One centimeter segments were cultured, either with or without a transverse laceration across 90% of the midsection of the tendon segment. Frozen sections of the cultured tendon segments were reacted with the (mouse monoclonal) Type I procollagen antisera. At all times of culture of the nonlacerated tendon segment, only a few of the epitenon cells along the surface of the tendon and distant from the cut end stained positively for active collagen synthesis. In the lacerated segments, and as early as 9 days of culture and repair, the procollagen reaction product was starting to appear in those cells of the outer epitenon cell layer. These studies support the concept that the inner fibroblasts do actively participate in collagen production. However, it now also appears that a significant degree of collagen synthesis during tendon repair resides in the outerlepitenon layer of cells enveloping the tendon segment, and that the repair response of the flexor tendon in vitro is proportional to the degree of injury. Key Words: Collagen-Tendon-Epitenon-Wound-Repair.
The concept that flexor tendons have the intrinsic capacity to participate actively in the repair process following laceration has been developed in recent years as the result of experimental studies from several laboratories. The support for this concept is based in great part on histologic examination of healing tendons from various experimental animals (1,2,6-12,1&19,21). These morphological studies demonstrated proliferation and migration of the
outer epitenon layer of cells, as well as the central endotenon cells located near the cut, into the laceration site. Most previous studies did not identify which cell population was responsible for protein synthesis, and the role of the two populations is still controversial. Since both flexor tendon epitenon and endotenon cells are fibroblasts, it is entirely possible that both would contribute to the production of Type I collagen during the repair process. The purpose of this study was therefore to evaluate the cellular activity of the tendon fibroblasts, from nonhuman primate
Address correspondence and reprint requests to Dr. J. E. Russell at Box 8109, Division of Orthopedic Surgery, Washington University, 660 So. Euclid Ave., St. Louis, MO 63110, U.S.A.
13
J . E. RUSSELL A N D P . R. MANSKE
14
flexor tendons, during in vitro repair with respect to Type I collagen synthesis. Utilization of the monoclonal antibody that labels intracellular Type I procollagen (20) has permitted the identification of those cells that were most actively producing Type I collageri during the in vitro repair period. MATERIALS AND METHODS Studies utilized three young adult Macaca nemestrina monkeys, which were involved in an ongoing ligament prosthesis study. All studies were done in accordance with the animal protection protocols of Washington University, St. Louis, MO. Procedure Profundus flexor tendon segments were obtained from within the digital sheath (i.e., Zone 11), under sterile conditions, from both the hands and the feet of the monkeys at the time of euthanasia. Three 1 cm segments were obtained from each tendon. One segment was proximal to the vinculum longus, one centered on the vinculum longus, and one distal to the vinculum longus. Thus, 48 tendon segments were obtained from each monkey. The 1 cm segments were placed in Hank’s balanced salt solution and transverse lacerations made across 90% of the midsection of each tendon segment with a scalpel. The segments were positioned on triangular stainless steel grids, and the grids folded so as to immobilize the tendon and align the laceration site, thereby eliminating the need to use suture material (4,13,15). Segments were cultured in BGJb:DME medium (1:4), supplemented with 2 mM glutamine, 50 pg/ml of ascorbic acid, 5% heat-inactivated homologous serum, and 100 pg/ml of streptomycin and 100 units/ ml of penicillin. The cultures were incubated in 3 ml of media in 17 mm Costar petri dishes (Costar, Cambridge, MA, U.S.A.) at 37°C in an atmosphere of 7.5% CO, in air at 100% humidity. Media was changed every 48 h. Tendon segments remained in culture for periods of 4 and 9 days and 2, 4, 6, and 9 weeks. Twelve of the tendon segments were cultured without a midsection laceration to evaluate the nature and specificity ofthe repair response generated by the laceration iniurv. These tendon segments remained in culture for periods of 1, 2,4, an; 8 weeks. fresh tendon segments were harvested and prepared for immediate immunohistochemistry I
-
J Orthop Res, Vol. 8 , N o . I , 1990
without placement in culture. These served as control segments. Immunohistochemistry Harvested tendon segments (unfixed) were mounted in tissue blocks and frozen in Tissue-Tek OTC mounting medium (Miles Laboratory). The blocks were stored at - 70°C until sectioning with a steel knife in a cryostat at - 15°C. The 6-pm sections were picked up on glass slides coated with 1% gelatin. The sections were briefly air-dried and fixed immediately in acetone. The monoclonal antibody to the C-terminal end of anti-procollagen (anti-pC) was that of McDonald et al. (20). The sections were incubated with the procollagen antiserum (3,20) at 1:20 dilution for 40-
FIG. 1. Epitenon surface of a freshly harvested control tendon segment. The frozen section was incubated with the procollagen antibody (1 :20 dilution) and reacted with an avidin-biotin-peroxidase. conjugate as described in the Methods section. The peroxidase staining is localized to a very few outer epitenon cells (arrows) and fewer inner fibroblasts (arrows). ( x 163.)
COLLAGEN SYNTHESIS DURING TENDON REPAIR
60 min. The sections were then washed with excess Tris-buffered saline (TBS) for 10 min, rinsed, and incubated for 3 5 4 0 min with a biotinylated mouse IgG secondary antibody raised in horse (Vectastain) (1:50). The secondary antibody was reacted with a biotin-avidin-horseradish peroxidase conjugate (Vectastain ABC) for 30 min. After rinsing with excess TBS, the slides were counterstained for 15 min in methyl greedalcian blue for visualization of the cells and the collagen matrix, respectively. The slides were quickly dehydrated through graded alcohol, into xylene, and mounted. Controls consisted of preimmune serum and omission of the primary antibody. Sections were examined under magnification with light microscopy. Positively staining cells showed intracellular brown material. RESULTS Control The freshly harvested control tendon segments resembled uninjured tissue histologically and showed sparse positive staining for active collagen synthesis when reacted with the anti-pC antibody
15
(Fig. 1). This is in agreement with the low level of collagen synthesis that characterizes the uninjured flexor tendon. Nonlacerated Segments At all times of culture examined with this nonlacerated tendon segment, only a few of the epitenon cells along the surface of the tendon and distant from the cut end stained positively for active collagen synthesis. Synthetic activity was primarily at the cut tendon ends. At 1 week of culture and repair, there was histological evidence of cellular proliferation and migration to the cut tendon end. Only a few cells stained positively for collagen synthesis with the procollagen antibody. By 2 to 4 weeks, the cellular migration at the cut end was essentially complete, and the staining for active collagen synthesis was distinct but still sparse (Fig. 2). An occasional endotenon fibroblast w a s positively stained with the procollagen antibody. Following 8 weeks of culture and repair, the frequency of the procollagen reaction product at the cut ends was not markedly different from that of the control tendon segments (Fig. 1).
FIG. 2. Epitenon surface of the nonlacerated tendon segment following 2 weeks of culture (A), and the cut end at 4 weeks of culture (B). The frozen section was incubated with the procollagen antibody (1:20 dilution) and reacted with an avidin-biotin-peroxidase conjugate as described in the Methods Section. The peroxidase staining is distinct but sparse along the outer epitenon surface (arrows). An occasional inner fibroblast was positively stained (arrows). (A, ~ 5 6 6, ; ~60.)
J Ortkop Res, Vol. 8, N o . I , 1990
J . E . RUSSELL AND P. R . MANSKE
16
Lacerated Segments In the lacerated tendon segments, a greater number of the inner fibroblasts were actively synthesizing collagen, and the entire epitenon layer of cells enveloping the tendon segment demonstrated active collagen synthesis. No regional differences, in the pattern or intensity of the procollagen staining, were noted between the proximal, distal, and vincular segments obtained from each flexor tendon (Zone 11), at all times in culture. Four Days At 4 days following laceration and in vitro repair, there was no histological evidence of cellular proliferation nor cellular migration to either the laceration site or the cut tendon end. At this time, only a few of the epitenon cells stained positively with the procollagen antibody (anti-pC). There was no staining with anti-pC in the central endotenon cells. Nine Days
After 9 days of culture and repair, the procollagen reaction product appeared most intensely in the surface layer of epitenon cells (Fig. 3). There was a specific but less intense anti-pC reaction product in the endotenon fibroblasts that were just beneath the surface layer of epitenon cells, and in the endotenon cells that were near the cut end (Fig. 4). There was also intense staining in those tendon fibroblasts that had migrated to the cut tendon end (Fig. 4) and to the laceration site (Fig. 5).
Two and 4 Weeks
After 2 and 4 weeks of in vitro repair, the procollagen synthetic activity was still most pronounced in the epitenon layer of cells (Fig. 6), and at the cut tendon ends (Fig. 7). The increased cellular activity seen in those fibroblasts underlying the epitenon layer at 9 days persisted through 4 weeks. Six and 9 Weeks
There was a progressive diminution of the staining for collagen synthesis from 6 weeks through 9 weeks throughout the repairing tendon segment. By the ninth week, the laceration gap had closed and the individual cellular activity at the interface was not discernable (Fig. 8). DISCUSSION
In recent years, the concept has developed that flexor tendons have the intrinsic capacity to participate in the repair process. Most studies have presented histologic evidence of tendon fibroblast proliferation and migration into the laceration site. Only a few investigations have attempted to define the specific role of the tendon fibroblasts in collagen production following injury. Lindsay and associates (6-8), studying healing of chicken flexor tendons, were the first to define the tendon’s response to injury. The initial response was noted at the surface layer of epitenon fibroblasts, which proliferated and migrated to the laceration site. The tendon cells in the central, endot-
FIG. 3. Epitenon surface of a lacerated tendon segment following 9 days of culture. The frozen section was incubated with the procollagen antibody (1:20 dilution) and reacted with an avidin-biotin-peroxidase conjugate as described in the Methods section. The reaction product appeared most intensely along the entire surface layer of epitenon cells (arrows). ( ~ 1 0 8 . )
J Orthop Res, Vol. 8, N o . I, 1990
COLLAGEN SYNTHESIS DURING TENDON REPAIR
17
FIG. 4. Cut end of the lacerated tendon segment following 9 days of culture. The frozen section was incubated with the procollagen antibody (1:20 dilution) and reacted with an avidinbiotin-peroxidase conjugate as described in the Methods section. The staining is primarily localized to the outer/epitenon layer of cells, but some of the inner fibroblasts in this region also demonstrate procollagen synthesis (arrows), and appear to be migrating to the laceration site. ( x144.)
enon core of the tissue exhibited a similar response, but the response was delayed by several days to 1 week. Several subsequent investigators, using numerous ingenious in vivo and in vitro experimental models (13-19,21), have also documented morphologically the migration of fibroblasts to the tendon laceration site. However, these studies do not define the specific functional activity of either the epitenon or endotenon cells; it has been presumed that both cell populations are involved in the restoration of continuity of the tendon bundles at the laceration site. Manske (13) and Gelberman (4), using transmission electron microscopy, studied the in vitro repair response of lacerated tendon segments from various experimental animals (including rabbit, monkey,
dog, and chicken). Collectively, the results indicated that the surface layer of epitenon cells that had migrated into the laceration site assumed a phagocytic role, engulfing cellular debris and old collagen fragments. Based on the location of the newly formed collagen fibrils, collagen synthesis was thought to have been primarily dependent on the endotenon fibroblastic cell population. However, neither of these studies could exclude the epitenon cell population from involvement in the synthesis of the collagen matrix. More recently, Garner et al. (3) have demonstrated in short-term (8-day) cultures of nonlacerated chicken flexor tendons that the collagen synthesis occurred only in the epitenon cell layer. These cells were shown to divide, migrate, and be-
FIG. 5. Laceration site of the tendon segment following 9 days of culture. The frozen section was incubated with the procollagen antibody (1:20 dilution) and reacted with an avidinbiotin-peroxidase conjugate as described in the Methods Section. The peroxidase staining is most intense at the outedepitenon layer of cells (arrows). ( x 144.)
J Orthop Res, Vol. 8, No. 1 , 1990
18
J . E. RUSSELL AND P . R . MANSKE
FIG. 6. Epitenon surface of the lacerated tendon segment following 4 weeks of culture. The frozen section was incubated with the procollagen antibody (1:20 dilution) and reacted with an avidin-biotin-peroxidase conjugate as described in the Methods section. The cellular staining was most pronounced in the outer/ epitenon layer of cells. ( ~ 1 0 8 . )
gin collagen synthesis as early as 4 days following explantation. However, the major onset of collagen and noncollagen protein synthesis occurs, in the lacerated flexor tendon, at 2 to 3 weeks after the initiation of repair both in vivo ( 5 ) and in vitro (13,15). Their study may have been too short to permit the identification of the endotenon cells as protein synthesizers. Furthermore, their study only addressed the end-capping of the tendon explant and did not address the bridging of a midsection laceration site by the tenocytes. The immunohistochemical study in the nonhuman primate reported herein supports the concept that tendon fibroblasts are active participants in the repair process. As noted in earlier studies (13), the
FIG. 7. Cut end of the lacerated tendon segment following 4 weeks of culture. The frozen section was incubated with the procollagen antibody (1:20 dilution) and reacted with an avidinbiotin-peroxidase conjugate as described in the Methods section. The outer epitenon layer of cells demonstrates the most positive reaction product (arrows). ( ~ 1 0 8 . )
J Orthop Res, Vol. 8 , N o . I , 1990
endotenon fibroblasts actively participate in collagen synthesis during tendon repair; this activity is most predominant at both the laceration site and at the cut ends of the tendon segment. Manske and Lesker (14) had previously observed that the endcapping fibroblasts, noted during the in vitro repair of primate flexor tendons, are derived in part from the migration of endotenon cells to the cut end. Additionally, the results of this study extend the observations of Garner et al. (3) and indicate that the epitenon cells also participate in collagen synthesis during the tendon repair process. It appears that the greater intensity of procollagen staining was in the surface layer of epitenon cells. Finally, the increased activity in the lacerated tendons was signif-
COLLAGEN SYNTHESIS DURING TENDON REPAIR
19
sistent with the earlier observations of Lindsay (68) that all tendon cellular elements are involved in the healing process. Acknowledgment: This investigation was supported in part by the Shriners Hospital for Crippled Children Grant #15961. The authors wish to acknowledge the technical assistance of Dennis M. Oakley a n d Annernarie Schrnoeker. The authors are also grateful to John A. McDonald, Ph.D., M.D., of Washington University, St. Louis, MO, for use of his procollagen antibody in these studies. This work was presented in part at the 35th Annual Meeting, Orthopedic Research Society, February 69, 1988, Las Vegas, Nevada.
REFERENCES
FIG. 8. Laceration sits of the tendon segment following 9 weeks of culture. The frozen section was incubated with the procollagen antibody (1:20 dilution) and reacted with an avidin-biotin-peroxidase conjugate as described in the Methods section. The laceration site is closed (arrows) and cellular procollagen synthetic activity was markedly decreased. (x166.)
icantly greater than in the nonlacerated segments, indicating that the repair response of the flexor tendon in vitro is proportional to the degree of injury. It may be of note that the culture conditions used in the present study differed from that of other investigators studying tendon repair in vitro (1,3,16); they all had included 10% fetal serum in the culture medium, while this investigation utilized adult, human sera that had been heat-inactivated. Thus, in the absence of numerous fetal serum growth factors, the repair response of the cultured tendon segment was found to be markedly influenced by the degree of injury and laceration. In the presence of the additional midsection laceration, the epitenon cellular response was noted not only at the laceration site and at the cut tendon end, but also along the entire epitenon surface. This observation is con-
1. Becker H, Graham MF, Cohen IK, Diegelmann RF: Intrinsic tendon cell proliferation in tissue culture. J Hand Surg 6:616-619, ,1981 2. Eiken 0, Lundborg G, Rank F: The role of the digital synovial sheath in tendon grafting. An experimental and clinical study on autologous tendon grafting in the digit. Scan J Plast Reconstr Surg 9:182-189, 1975 3. Garner WL, McDonald JA, Kuhn I11 C, Weeks PM: Antonomous healing of chicken flexor tendons in vitro. J Hand Surg [Am] 13:697-700, 1988 4. Gelberman RH, Manske PR, VandeBerg JS, Lesker PA: Flexor tendon repair in vitro: a comparative histologic study of the rabbit, chicken, dog and monkey. J Orthop Res 2:3% 48, 1984 5. Kain CC, Russell JE, B u m R, Dunlap J, McCarthy J, Manske PR: The effect of vascularization on avian flexor tendon repair. A biochemical study. Clin Orthop Re1 Res 233:295303, 1988 6. Lindsay WK, McDougall EP: Digital flexor tendons: an experimental study, Part 111. The fate of autogenous digital flexor tendon grafts. Br J Plast Surg 13:293-304, 1961 7. Lindsay WK, Thomson HG: Digital flexor tendons: an experimental study, Part I. The significance of each component of the flexor mechanism in tendon healing. Br J Plast Surg 12:289-319, 1960 8. Lindsay WK, Thomson HG, Walker FG: Digital flexor tendons: an experimental study, Part 11. The significance of a gap occumng at the line of suture. Br J Plasr Surg 13:l-9, 1960 9. Lundborg G: Experimental flexor tendon healing without adhesion formation-A new concept of tendon nutrition and intrinsic healing mechanisms. A preliminary report. Hand 8:235-238, 1976 10. Lundborg G, Hansson H-A, Rank F, Rydevik B: Superficial repair of severed flexor tendons in synovial environment. An experimental ultrastructural study on cellular mechanisms. J Hand Surg 5:451461, 1980 11. Lundborg G, Rank F: Experimental intrinsic healing of flexor tendon based on synovial fluid nutrition. J Hand Surg 3:21-31, 1978 12. Lundborg G, Rank F: Experimental studies on cellular mechanisms involved in healing of animal and human flexor tendon in synovial environment. Hand 12:3-11, 1980 13. Manske PR, Gelberman RH, VandeBerg JS, Lesker PA: Intrinsic flexor-tendon repair. J Bone Joint Surg [Am]66:385396, 1984 14. Manske PR, Lesker PA: Evidence of intrinsic flexor tendon
J Orthop Res, Vol. 8, No. I, 1990
J . E. RUSSELL A N D P . R . MANSKE
20
15. 16. 17. 18.
repair in various experimental animals. Clin Orrhop Re1 Res 182:297-304, 1984 Manske PR, Lesker PA, Gelberman RH, Rucinsky TE: Intrinsic restoration of the flexor tendon surface in the nonhuman primate. J Hand Surg [Am] 10:620-637, 1985 Mass DP, Martell J, Tuel R: Human flexor tendon healing in vitro. Orthop Trans 12:9, 1988 Matthews P: The fate of isolated segments of flexor tendon within the digital sheath. A study in synovial nutrition. Br J Plast Surg 29:21&224, 1976 Matthews P, Richards H: The repair potential of digital
J Orthop Res, Vol. 8 , No.
I, 1990
flexor tendons. An experimental study. J Bone Joint Surg [Br] 56:618425, 1974 19. Matthews P, Richards H: The repair reaction of flexor tendon within the digital sheath. Hand 7:27-29, 1975 20. McDonald JA, Broekelmann TJ, Matheke ML, Crouch E, Koo M, Kuhn C 111: A monoclonal antibody to the carboxyterminal domain of procollagen type I visualizes collagen-synthesizing fibroblasts. J Clin Invest 78: 1237-1244, 1986 21. McDowell CL, Snyder DM: Tendon healing. An experimental model in the dog. J Hand Surg 2:122-126, 1977
Collagen Synthesis During Primate Flexor Tendon Repair In Vitro Jean E. Russell and Paul R. Manske Division of Orthopedic Surgery, Department of Surgery, Washington University School of Medicine, St. Louis, Missouri, U.S.A.
Summary: The concept that flexor tendons have the intrinsic capacity to participate actively in the repair process following laceration has been developed in recent years as the result of experimental studies from numerous laboratories. However, the role of the outerlepitenon and innerlendotenon cell populations with regard to protein synthesis is still controversial. The purpose of this study was to investigate the respective participation of these tendon fibroblast populations in the synthesis of the collagen matrix during in vitro repair of the flexor tendon from nonhuman primates, utilizing immunohistochemical techniques and a Type I procollagen antibody. Zone I1 profundus flexor tendon segments were obtained from young adult Macaca nemestrina monkeys. One centimeter segments were cultured, either with or without a transverse laceration across 90% of the midsection of the tendon segment. Frozen sections of the cultured tendon segments were reacted with the (mouse monoclonal) Type I procollagen antisera. At all times of culture of the nonlacerated tendon segment, only a few of the epitenon cells along the surface of the tendon and distant from the cut end stained positively for active collagen synthesis. In the lacerated segments, and as early as 9 days of culture and repair, the procollagen reaction product was starting to appear in those cells of the outer epitenon cell layer. These studies support the concept that the inner fibroblasts do actively participate in collagen production. However, it now also appears that a significant degree of collagen synthesis during tendon repair resides in the outerlepitenon layer of cells enveloping the tendon segment, and that the repair response of the flexor tendon in vitro is proportional to the degree of injury. Key Words: Collagen-Tendon-Epitenon-Wound-Repair.
The concept that flexor tendons have the intrinsic capacity to participate actively in the repair process following laceration has been developed in recent years as the result of experimental studies from several laboratories. The support for this concept is based in great part on histologic examination of healing tendons from various experimental animals (1,2,6-12,1&19,21). These morphological studies demonstrated proliferation and migration of the
outer epitenon layer of cells, as well as the central endotenon cells located near the cut, into the laceration site. Most previous studies did not identify which cell population was responsible for protein synthesis, and the role of the two populations is still controversial. Since both flexor tendon epitenon and endotenon cells are fibroblasts, it is entirely possible that both would contribute to the production of Type I collagen during the repair process. The purpose of this study was therefore to evaluate the cellular activity of the tendon fibroblasts, from nonhuman primate
Address correspondence and reprint requests to Dr. J. E. Russell at Box 8109, Division of Orthopedic Surgery, Washington University, 660 So. Euclid Ave., St. Louis, MO 63110, U.S.A.
13
J . E. RUSSELL A N D P . R. MANSKE
14
flexor tendons, during in vitro repair with respect to Type I collagen synthesis. Utilization of the monoclonal antibody that labels intracellular Type I procollagen (20) has permitted the identification of those cells that were most actively producing Type I collageri during the in vitro repair period. MATERIALS AND METHODS Studies utilized three young adult Macaca nemestrina monkeys, which were involved in an ongoing ligament prosthesis study. All studies were done in accordance with the animal protection protocols of Washington University, St. Louis, MO. Procedure Profundus flexor tendon segments were obtained from within the digital sheath (i.e., Zone 11), under sterile conditions, from both the hands and the feet of the monkeys at the time of euthanasia. Three 1 cm segments were obtained from each tendon. One segment was proximal to the vinculum longus, one centered on the vinculum longus, and one distal to the vinculum longus. Thus, 48 tendon segments were obtained from each monkey. The 1 cm segments were placed in Hank’s balanced salt solution and transverse lacerations made across 90% of the midsection of each tendon segment with a scalpel. The segments were positioned on triangular stainless steel grids, and the grids folded so as to immobilize the tendon and align the laceration site, thereby eliminating the need to use suture material (4,13,15). Segments were cultured in BGJb:DME medium (1:4), supplemented with 2 mM glutamine, 50 pg/ml of ascorbic acid, 5% heat-inactivated homologous serum, and 100 pg/ml of streptomycin and 100 units/ ml of penicillin. The cultures were incubated in 3 ml of media in 17 mm Costar petri dishes (Costar, Cambridge, MA, U.S.A.) at 37°C in an atmosphere of 7.5% CO, in air at 100% humidity. Media was changed every 48 h. Tendon segments remained in culture for periods of 4 and 9 days and 2, 4, 6, and 9 weeks. Twelve of the tendon segments were cultured without a midsection laceration to evaluate the nature and specificity ofthe repair response generated by the laceration iniurv. These tendon segments remained in culture for periods of 1, 2,4, an; 8 weeks. fresh tendon segments were harvested and prepared for immediate immunohistochemistry I
-
J Orthop Res, Vol. 8 , N o . I , 1990
without placement in culture. These served as control segments. Immunohistochemistry Harvested tendon segments (unfixed) were mounted in tissue blocks and frozen in Tissue-Tek OTC mounting medium (Miles Laboratory). The blocks were stored at - 70°C until sectioning with a steel knife in a cryostat at - 15°C. The 6-pm sections were picked up on glass slides coated with 1% gelatin. The sections were briefly air-dried and fixed immediately in acetone. The monoclonal antibody to the C-terminal end of anti-procollagen (anti-pC) was that of McDonald et al. (20). The sections were incubated with the procollagen antiserum (3,20) at 1:20 dilution for 40-
FIG. 1. Epitenon surface of a freshly harvested control tendon segment. The frozen section was incubated with the procollagen antibody (1 :20 dilution) and reacted with an avidin-biotin-peroxidase. conjugate as described in the Methods section. The peroxidase staining is localized to a very few outer epitenon cells (arrows) and fewer inner fibroblasts (arrows). ( x 163.)
COLLAGEN SYNTHESIS DURING TENDON REPAIR
60 min. The sections were then washed with excess Tris-buffered saline (TBS) for 10 min, rinsed, and incubated for 3 5 4 0 min with a biotinylated mouse IgG secondary antibody raised in horse (Vectastain) (1:50). The secondary antibody was reacted with a biotin-avidin-horseradish peroxidase conjugate (Vectastain ABC) for 30 min. After rinsing with excess TBS, the slides were counterstained for 15 min in methyl greedalcian blue for visualization of the cells and the collagen matrix, respectively. The slides were quickly dehydrated through graded alcohol, into xylene, and mounted. Controls consisted of preimmune serum and omission of the primary antibody. Sections were examined under magnification with light microscopy. Positively staining cells showed intracellular brown material. RESULTS Control The freshly harvested control tendon segments resembled uninjured tissue histologically and showed sparse positive staining for active collagen synthesis when reacted with the anti-pC antibody
15
(Fig. 1). This is in agreement with the low level of collagen synthesis that characterizes the uninjured flexor tendon. Nonlacerated Segments At all times of culture examined with this nonlacerated tendon segment, only a few of the epitenon cells along the surface of the tendon and distant from the cut end stained positively for active collagen synthesis. Synthetic activity was primarily at the cut tendon ends. At 1 week of culture and repair, there was histological evidence of cellular proliferation and migration to the cut tendon end. Only a few cells stained positively for collagen synthesis with the procollagen antibody. By 2 to 4 weeks, the cellular migration at the cut end was essentially complete, and the staining for active collagen synthesis was distinct but still sparse (Fig. 2). An occasional endotenon fibroblast w a s positively stained with the procollagen antibody. Following 8 weeks of culture and repair, the frequency of the procollagen reaction product at the cut ends was not markedly different from that of the control tendon segments (Fig. 1).
FIG. 2. Epitenon surface of the nonlacerated tendon segment following 2 weeks of culture (A), and the cut end at 4 weeks of culture (B). The frozen section was incubated with the procollagen antibody (1:20 dilution) and reacted with an avidin-biotin-peroxidase conjugate as described in the Methods Section. The peroxidase staining is distinct but sparse along the outer epitenon surface (arrows). An occasional inner fibroblast was positively stained (arrows). (A, ~ 5 6 6, ; ~60.)
J Ortkop Res, Vol. 8, N o . I , 1990
J . E . RUSSELL AND P. R . MANSKE
16
Lacerated Segments In the lacerated tendon segments, a greater number of the inner fibroblasts were actively synthesizing collagen, and the entire epitenon layer of cells enveloping the tendon segment demonstrated active collagen synthesis. No regional differences, in the pattern or intensity of the procollagen staining, were noted between the proximal, distal, and vincular segments obtained from each flexor tendon (Zone 11), at all times in culture. Four Days At 4 days following laceration and in vitro repair, there was no histological evidence of cellular proliferation nor cellular migration to either the laceration site or the cut tendon end. At this time, only a few of the epitenon cells stained positively with the procollagen antibody (anti-pC). There was no staining with anti-pC in the central endotenon cells. Nine Days
After 9 days of culture and repair, the procollagen reaction product appeared most intensely in the surface layer of epitenon cells (Fig. 3). There was a specific but less intense anti-pC reaction product in the endotenon fibroblasts that were just beneath the surface layer of epitenon cells, and in the endotenon cells that were near the cut end (Fig. 4). There was also intense staining in those tendon fibroblasts that had migrated to the cut tendon end (Fig. 4) and to the laceration site (Fig. 5).
Two and 4 Weeks
After 2 and 4 weeks of in vitro repair, the procollagen synthetic activity was still most pronounced in the epitenon layer of cells (Fig. 6), and at the cut tendon ends (Fig. 7). The increased cellular activity seen in those fibroblasts underlying the epitenon layer at 9 days persisted through 4 weeks. Six and 9 Weeks
There was a progressive diminution of the staining for collagen synthesis from 6 weeks through 9 weeks throughout the repairing tendon segment. By the ninth week, the laceration gap had closed and the individual cellular activity at the interface was not discernable (Fig. 8). DISCUSSION
In recent years, the concept has developed that flexor tendons have the intrinsic capacity to participate in the repair process. Most studies have presented histologic evidence of tendon fibroblast proliferation and migration into the laceration site. Only a few investigations have attempted to define the specific role of the tendon fibroblasts in collagen production following injury. Lindsay and associates (6-8), studying healing of chicken flexor tendons, were the first to define the tendon’s response to injury. The initial response was noted at the surface layer of epitenon fibroblasts, which proliferated and migrated to the laceration site. The tendon cells in the central, endot-
FIG. 3. Epitenon surface of a lacerated tendon segment following 9 days of culture. The frozen section was incubated with the procollagen antibody (1:20 dilution) and reacted with an avidin-biotin-peroxidase conjugate as described in the Methods section. The reaction product appeared most intensely along the entire surface layer of epitenon cells (arrows). ( ~ 1 0 8 . )
J Orthop Res, Vol. 8, N o . I, 1990
COLLAGEN SYNTHESIS DURING TENDON REPAIR
17
FIG. 4. Cut end of the lacerated tendon segment following 9 days of culture. The frozen section was incubated with the procollagen antibody (1:20 dilution) and reacted with an avidinbiotin-peroxidase conjugate as described in the Methods section. The staining is primarily localized to the outer/epitenon layer of cells, but some of the inner fibroblasts in this region also demonstrate procollagen synthesis (arrows), and appear to be migrating to the laceration site. ( x144.)
enon core of the tissue exhibited a similar response, but the response was delayed by several days to 1 week. Several subsequent investigators, using numerous ingenious in vivo and in vitro experimental models (13-19,21), have also documented morphologically the migration of fibroblasts to the tendon laceration site. However, these studies do not define the specific functional activity of either the epitenon or endotenon cells; it has been presumed that both cell populations are involved in the restoration of continuity of the tendon bundles at the laceration site. Manske (13) and Gelberman (4), using transmission electron microscopy, studied the in vitro repair response of lacerated tendon segments from various experimental animals (including rabbit, monkey,
dog, and chicken). Collectively, the results indicated that the surface layer of epitenon cells that had migrated into the laceration site assumed a phagocytic role, engulfing cellular debris and old collagen fragments. Based on the location of the newly formed collagen fibrils, collagen synthesis was thought to have been primarily dependent on the endotenon fibroblastic cell population. However, neither of these studies could exclude the epitenon cell population from involvement in the synthesis of the collagen matrix. More recently, Garner et al. (3) have demonstrated in short-term (8-day) cultures of nonlacerated chicken flexor tendons that the collagen synthesis occurred only in the epitenon cell layer. These cells were shown to divide, migrate, and be-
FIG. 5. Laceration site of the tendon segment following 9 days of culture. The frozen section was incubated with the procollagen antibody (1:20 dilution) and reacted with an avidinbiotin-peroxidase conjugate as described in the Methods Section. The peroxidase staining is most intense at the outedepitenon layer of cells (arrows). ( x 144.)
J Orthop Res, Vol. 8, No. 1 , 1990
18
J . E. RUSSELL AND P . R . MANSKE
FIG. 6. Epitenon surface of the lacerated tendon segment following 4 weeks of culture. The frozen section was incubated with the procollagen antibody (1:20 dilution) and reacted with an avidin-biotin-peroxidase conjugate as described in the Methods section. The cellular staining was most pronounced in the outer/ epitenon layer of cells. ( ~ 1 0 8 . )
gin collagen synthesis as early as 4 days following explantation. However, the major onset of collagen and noncollagen protein synthesis occurs, in the lacerated flexor tendon, at 2 to 3 weeks after the initiation of repair both in vivo ( 5 ) and in vitro (13,15). Their study may have been too short to permit the identification of the endotenon cells as protein synthesizers. Furthermore, their study only addressed the end-capping of the tendon explant and did not address the bridging of a midsection laceration site by the tenocytes. The immunohistochemical study in the nonhuman primate reported herein supports the concept that tendon fibroblasts are active participants in the repair process. As noted in earlier studies (13), the
FIG. 7. Cut end of the lacerated tendon segment following 4 weeks of culture. The frozen section was incubated with the procollagen antibody (1:20 dilution) and reacted with an avidinbiotin-peroxidase conjugate as described in the Methods section. The outer epitenon layer of cells demonstrates the most positive reaction product (arrows). ( ~ 1 0 8 . )
J Orthop Res, Vol. 8 , N o . I , 1990
endotenon fibroblasts actively participate in collagen synthesis during tendon repair; this activity is most predominant at both the laceration site and at the cut ends of the tendon segment. Manske and Lesker (14) had previously observed that the endcapping fibroblasts, noted during the in vitro repair of primate flexor tendons, are derived in part from the migration of endotenon cells to the cut end. Additionally, the results of this study extend the observations of Garner et al. (3) and indicate that the epitenon cells also participate in collagen synthesis during the tendon repair process. It appears that the greater intensity of procollagen staining was in the surface layer of epitenon cells. Finally, the increased activity in the lacerated tendons was signif-
COLLAGEN SYNTHESIS DURING TENDON REPAIR
19
sistent with the earlier observations of Lindsay (68) that all tendon cellular elements are involved in the healing process. Acknowledgment: This investigation was supported in part by the Shriners Hospital for Crippled Children Grant #15961. The authors wish to acknowledge the technical assistance of Dennis M. Oakley a n d Annernarie Schrnoeker. The authors are also grateful to John A. McDonald, Ph.D., M.D., of Washington University, St. Louis, MO, for use of his procollagen antibody in these studies. This work was presented in part at the 35th Annual Meeting, Orthopedic Research Society, February 69, 1988, Las Vegas, Nevada.
REFERENCES
FIG. 8. Laceration sits of the tendon segment following 9 weeks of culture. The frozen section was incubated with the procollagen antibody (1:20 dilution) and reacted with an avidin-biotin-peroxidase conjugate as described in the Methods section. The laceration site is closed (arrows) and cellular procollagen synthetic activity was markedly decreased. (x166.)
icantly greater than in the nonlacerated segments, indicating that the repair response of the flexor tendon in vitro is proportional to the degree of injury. It may be of note that the culture conditions used in the present study differed from that of other investigators studying tendon repair in vitro (1,3,16); they all had included 10% fetal serum in the culture medium, while this investigation utilized adult, human sera that had been heat-inactivated. Thus, in the absence of numerous fetal serum growth factors, the repair response of the cultured tendon segment was found to be markedly influenced by the degree of injury and laceration. In the presence of the additional midsection laceration, the epitenon cellular response was noted not only at the laceration site and at the cut tendon end, but also along the entire epitenon surface. This observation is con-
1. Becker H, Graham MF, Cohen IK, Diegelmann RF: Intrinsic tendon cell proliferation in tissue culture. J Hand Surg 6:616-619, ,1981 2. Eiken 0, Lundborg G, Rank F: The role of the digital synovial sheath in tendon grafting. An experimental and clinical study on autologous tendon grafting in the digit. Scan J Plast Reconstr Surg 9:182-189, 1975 3. Garner WL, McDonald JA, Kuhn I11 C, Weeks PM: Antonomous healing of chicken flexor tendons in vitro. J Hand Surg [Am] 13:697-700, 1988 4. Gelberman RH, Manske PR, VandeBerg JS, Lesker PA: Flexor tendon repair in vitro: a comparative histologic study of the rabbit, chicken, dog and monkey. J Orthop Res 2:3% 48, 1984 5. Kain CC, Russell JE, B u m R, Dunlap J, McCarthy J, Manske PR: The effect of vascularization on avian flexor tendon repair. A biochemical study. Clin Orthop Re1 Res 233:295303, 1988 6. Lindsay WK, McDougall EP: Digital flexor tendons: an experimental study, Part 111. The fate of autogenous digital flexor tendon grafts. Br J Plast Surg 13:293-304, 1961 7. Lindsay WK, Thomson HG: Digital flexor tendons: an experimental study, Part I. The significance of each component of the flexor mechanism in tendon healing. Br J Plast Surg 12:289-319, 1960 8. Lindsay WK, Thomson HG, Walker FG: Digital flexor tendons: an experimental study, Part 11. The significance of a gap occumng at the line of suture. Br J Plasr Surg 13:l-9, 1960 9. Lundborg G: Experimental flexor tendon healing without adhesion formation-A new concept of tendon nutrition and intrinsic healing mechanisms. A preliminary report. Hand 8:235-238, 1976 10. Lundborg G, Hansson H-A, Rank F, Rydevik B: Superficial repair of severed flexor tendons in synovial environment. An experimental ultrastructural study on cellular mechanisms. J Hand Surg 5:451461, 1980 11. Lundborg G, Rank F: Experimental intrinsic healing of flexor tendon based on synovial fluid nutrition. J Hand Surg 3:21-31, 1978 12. Lundborg G, Rank F: Experimental studies on cellular mechanisms involved in healing of animal and human flexor tendon in synovial environment. Hand 12:3-11, 1980 13. Manske PR, Gelberman RH, VandeBerg JS, Lesker PA: Intrinsic flexor-tendon repair. J Bone Joint Surg [Am]66:385396, 1984 14. Manske PR, Lesker PA: Evidence of intrinsic flexor tendon
J Orthop Res, Vol. 8, No. I, 1990
J . E. RUSSELL A N D P . R . MANSKE
20
15. 16. 17. 18.
repair in various experimental animals. Clin Orrhop Re1 Res 182:297-304, 1984 Manske PR, Lesker PA, Gelberman RH, Rucinsky TE: Intrinsic restoration of the flexor tendon surface in the nonhuman primate. J Hand Surg [Am] 10:620-637, 1985 Mass DP, Martell J, Tuel R: Human flexor tendon healing in vitro. Orthop Trans 12:9, 1988 Matthews P: The fate of isolated segments of flexor tendon within the digital sheath. A study in synovial nutrition. Br J Plast Surg 29:21&224, 1976 Matthews P, Richards H: The repair potential of digital
J Orthop Res, Vol. 8 , No.
I, 1990
flexor tendons. An experimental study. J Bone Joint Surg [Br] 56:618425, 1974 19. Matthews P, Richards H: The repair reaction of flexor tendon within the digital sheath. Hand 7:27-29, 1975 20. McDonald JA, Broekelmann TJ, Matheke ML, Crouch E, Koo M, Kuhn C 111: A monoclonal antibody to the carboxyterminal domain of procollagen type I visualizes collagen-synthesizing fibroblasts. J Clin Invest 78: 1237-1244, 1986 21. McDowell CL, Snyder DM: Tendon healing. An experimental model in the dog. J Hand Surg 2:122-126, 1977
