JOURNAL OF MORPHOLOGY 212:269-280 (1992)
Four Forearm Flexor Muscles of the Horse, Equus caballus: Anatomy and Histochemistry JOHN W. HERMANSON AND MArTHEW A. COBB Department ofAnatomy, College of Veterrnor?/Medicine, Cornell Unruersity, Ithaca, New York 14853
ABSTRACT Two of the forearm flexors of the horse, the superficial and deep digital flexor muscles, are critical to support the digital and fetlock joints, exhibit differing insertions, and are passively supported by the proximal and distal check ligaments, respectively. These two muscles differ in histochemical composition and architecture. The differences are correlated with the different stress levels transmitted through their tendons, and the different frequencies of clinical breakdown that have been reported. Both muscles contain type I and type IIa fibers. A few type IIb fibers occurred in the deep digital flexor. The superficial digital flexor contained approximately 56%type I fibers, extremely short muscle fibers, and extensive connective tissue investment. In contrast, the deep digital flexor had three muscle heads: ulnar, radial, and “long” and “short” regions of the humeral head. The “long” and “short” regions of the humeral head contained 33%and 44% type I fibers, respectively, fiber lengths three to four times as long as those in the superficial digital flexor, and relatively less connective tissue investment. Flexor carpi radialis and flexor carpi ulnaris compared most closely with the humeral head of the deep digital flexor. These data suggest a correlation of the unique architecture of superficial digital flexor with its proposed elastic storage properties during locomotion in horses, and a n explanation for the frequent breakdown of the superficial digital flexor in athletic horses. The forelimbs of horses are subject to substantial forces during quiet stance as well as during the stance phase of locomotion. A consideration of mechanics during quiet standing is important as domestic horses stand for 90% of every day, even while sleeping (Ruckebusch, ’72). The passive stayapparatus of the horse is an elaborate system of muscles and ligaments designed to permit stance with minimal muscular effort yet without compromising the cursorial ability of this animal (Sack, ’91). Components of the passive stay-apparatusof the hindlimb have been described in detail (Sack, ’89), but less information is available regarding components of the forelimb passive stay-apparatus, specifically those involving the elbow and shoulder joints. Although several studies have examined the biomechanical properties of the digital flexor muscles of horse forelimbs (Kingsbury et al., ’78; Dimery et al., ’86; Ker et al., ’88;Biewener and Rizzo, ’89)’ the physiological properties of these muscles remain largely unknown. The present study presents a n 8
1992 WILEY-LISS. INC
analysis of muscle morphology and histochemistry in the forearm flexors of the horse, including the flexor carpi radialis, flexor carpi ulnaris, superficial digital flexor, and deep digital flexor. The flexor carpi radialis and flexor carpi ulnaris are considered relatively simple in terms of their roles a t the elbow and carpal joints. The superficial digital flexor and deep digital flexors are significantly more complex both in terms of architecture (the deep flexor has three major heads with differing origins) and function (the deep and superficial flexors traverse from four to six functional joints). In particular, this study emphasizes comparison of presumed function in the superficial and deep digital flexor muscles. Although both of these muscles function as digital flexors, their insertions differ sufficiently to effect subtly different roles in locomotion and stance (Stashak, ’87). Biomechanical modeling of these muscles suggests that the superficial flexor tendon is exposed to almost twice the amount of stress as that imposed on the deep flexor tendon
270
J.W. HERMANSON AND M.A. COBB
(Ker et al., '88). Ker et al. suggest that such factors, as well as the relatively smaller crosssectional area of the superficial flexor tendon in the metacarpal region, predispose the muscle to the higher frequency of tendon breakdowns than is observed in the deep flexor tendon (Webbon, '77). Architectural and histochemical data are correlated with these mechanical hypotheses. MATERIALS AND METHODS
Whole flexor carpi radialis, flexor carpi ulnaris, deep digital flexor, and superficial digital flexor muscles were obtained from nine horses, Equus caballus (Table l), immediately following euthanasia by overdose of sodium pentobarbital. The animals were euthanized for reasons unrelated to musculoskeletal performance. The muscles were blotted dry and 2 cm3 blocks of muscle were taken from the following regions of each muscle. One or two sections were taken from the flexor carpi radialis: one from the proximal region (N = 6) and one from the distal region (N = 3). One to three sections of each flexor carpi ulnaris were taken including superficial (N = 8) and deep midbelly (N = 7) regions of the humeral head and a midbelly section of the ulnar head (N = 5). A maximum of four sections were taken from each deep flexor: midbelly regions of both parts of the humeral head (N = 9 or 5), and of the radial (N = 7) and ulnar (N = 8) heads. Three samples were taken from superficial flexors: proximal (N = 9) and distal (N = 5) superficial regions and a deep midbelly region (N = 7). Connective tissue landmarks were used to ensure that samples were taken from similar locations in all muscles. The muscle samples were mounted on cork with 5% gum tragacanth and quick frozen in isopentane cooled to approximately - 160°C by liquid nitrogen. Samples were then stored at -60 to 80°C.
TABLE 1. Description of horses used in the histochemical studv o f forearm flexor muscles Horse
Age (yrs) 8 7 9 7 6 20 12 6 20
Sex
Breed
Weight (kg)
F
Thoroughbred Quarterhorse Thoroughbred Standardbred Standardbred Standardbred Standardbred Thoroughbred Standardbred
511 591 490 485 400 499 494 421 412
F F F F F F F F
Full sets of all muscle samples were not obtained from each animal. Transverse serial sections (10 pm thickness) of muscle samples were obtained using a cryostat and mounted on glass slides for staining. Sections of each muscle sample were stained for myofibrillar ATPase (mATPase) following acidic and basic (pH 10.3) preincubation. A range of pHs (from 4.3 to 4.55) was used following initial studies spanning a pH range of 4.2 to 4.7 that indicated the usefulness of this range for fiber type identification following acidic preincubation and mATPase staining (see also Matoba and Gollnick, '84). Serial sections were stained for nicotinamide dinucleotide tetrazolium reductase (NADHTR) to show oxidative activity, and alpha glycerophosphate dehydrogenase (alpha GPD) to show glycolytic ability. Histochemical and morphometric procedures were used as previously described in Hermanson and Hurley ('90). Differences in histochemical profiles were assessed using ANOVA and a posthoc Bonferroni's comparison of multiple means (SAS Institute, Cary, NC). Significance was accepted when P I .05. Additional 10 km serial sections of each sample were stained for fast and slow myosin isoforms using immunocytochemical techniques. Sections were obtained in the same manner as for histochemistry, mounted on glass slides, and kept at -20°C until needed. Sections were allowed to warm to room temperature and then covered with 2% normal goat serum and incubated at 4°C in a humid chamber for 30 minutes. The normal goat serum was then shaken off and the primary antibody applied and allowed to incubate for 14 hours in a humid chamber a t 4°C. Fast myosin was reacted with a commercially available anti-fast myosin antibody Sigma MY-32 (Sigma Chemical Co., St. Louis). Slow myosin was reacted with anti-slow myosin antibody S-58 (provided by Dr. F. Stockdale, Stanford Medical School). The sections were dipped and then soaked for 10 minutes in a 0.05 M phosphate buffer solution (PBS) with 0.85%NaCl. A biotinylated rabbit anti-mouse secondary antibody was then applied for 10 minutes and the PBS rinsing repeated. Next, the sections were reacted with a streptavidin enzyme conjugate for 5 minutes and the PBS rinse was repeated. Staining was performed using a substrate chromagen for 15 minutes. The secondary antibody, enzyme conjugate, and substrate chromagen make up a commercially available staining kit (Histostain Sp.,
HORSE FLEXOR MUSCLES
Zymed Inc., San Francisco). Dissections were performed on two fresh and one formalin fixed forelimb specimen to determine patterns of innervation and connective tissue in the muscles studied. Whole muscles were obtained from two adult horses and a pony at postmortem, blotted dry, and fixed at their in situ lengths in 5%formalin solution. These specimens were then transferred to an approximately 15% nitric acid solution for connective tissue digestion. When fiber bundles could be teased apart, the muscles were transferred to a 50% glycerin solution. Under a dissecting microscope, individual muscle fibers were teased apart and measured to obtain estimates of mean fiber length for each of the regions sampled for histochemistry and immunocytochemistry. This yielded a measure of mean muscle fiber length and should not be confused with any attempt to measure overall muscle length or the distance between a tendon of origin and of insertion. Two flexor carpi radialis, flexor carpi ulnaris, deep digital flexor, and superficial digital flexor muscles were obtained from two adult horses and stained for acetylcholinesterase (Ypey, '78) to demonstrate the distribution of motor endplates within each muscle. Terminology used in this study conforms with the Nomina Anatomica Veterinaria (1983).
271
Deep digital flexor, humeral head Superficial digital flexor Proximal check lig.
Distal check lig.
Sesamoids at fetlock joint
RESULTS
Anatomy The superficial digital flexor is composed of a single head, innervated by a single branch of the ulnar nerve that enters the muscle about 3 cm from the muscle's proximal end. The muscle fibers are short, being only about 3 mm long (measured in a pony) and pinnately arranged. Three regions of the muscle were sampled for fiber length as well as histochemistry. All regions had muscle fibers of similar length. The superficial flexor was difficult to separate from the underlying deep flexor along the proximal two-thirds of its length (Figs. 1, 2) and shares a common fleshy origin with the humeral head of that muscle. A proximal check ligament arises from the distal one-third of the radius and courses distally to intersect with the superficial flexor tendon caudal to the carpus. The insertional tendon of the superficial flexor passes distally from the carpus to its attachments on the base of the proximal and middle phalanges.
Insertion of deep digital flexor Fig. 1. Equus caballus. Lateral view schematic of left forelimb showing the arrangement of the humeral head of the deep digital flexor and the superficial digital flexor muscles. Gravity tends to cause shoulder and elbow flexion (arrow), as well as overextension (dorsiflexion) of the fetlock (metacarpophalangeal),pastern (proximal interphalangeal), and coffin (distal interphalangeal) joints. Overextensionof the fetlock causes passive stretch of the deep and superficial flexors that is resisted by the distal and proximal check ligaments, respectively.Modified after Sack ('91).
The deep digital flexor is composed of three heads. The humeral head is the largest subdivision of the deep flexor, originating in common with the superficial flexor from the medial epicondyle of the humerus. The humeral head is divided by a connective tissue partition into a long and short region. The radial head comprises a relatively small subdivision with a maximum outer diameter of about 1
272
J.W. HERMANSON AND M.A. COBB
Fig. 2. Equus caballus. An oblique, caudolateral view of left forelimb of a horse. The area in the shaded box is magnified in the view to the right. The check ligaments serve as non-tiring support structures during quiet stance. The superficial digital flexor has been resected to show its region of adherence (dark shading) to the surface of the long and short humeral heads of the deep digital flexor. The ulnar and radial heads of the deep flexor are significantlysmaller than are the humeral parts. (Modified after Sack, '91).
cm, and exhibiting the shortest fiber lengths of any region of the deep flexor. The radial head fuses with the insertional tendon of deep flexor at the carpus. Both the humeral and radial heads of the deep flexor are innervated by branches of the median nerve. The median nerve splits into two primary nerve branches before entering the long and short humeral heads, suggesting that these regions may function as muscle compartments (sensu English and Letbetter, '82). A small ulnar head of the deep flexor muscle arises from the medial surface of the olecranon process. The muscular portion of this head is only about 8 cm long (in the pony) and extremely flat, but contains pinnately arranged fibers about 17 mm long (similar in length to the
much larger humeral head). The ulnar head is innervated by a branch of the ulnar nerve. A distal check ligament originates from the palmar carpal ligament and courses 5-6 cm distally to an attachment on the insertional tendon of the deep flexor. This tendon passes distally to its insertion upon the base of the terminal phalanx. The flexor carpi radialis is pinnate and originates from the medial epicondyle of the humerus and the medial collateral ligament and joint capsule of the elbow joint. Insertion is by a broad short tendon on the base of the second metacarpal. Fiber length is similar (about 37 mm in the pony) for two regions of the muscle investigated. Thus, these fibers are at least twice as long as fibers found in
2 73
HORSE FLEXOR MUSCLES
the deep flexor. The flexor carpi radialis makes up about 11%of the total mass of the forearm flexor muscle group. Innervation is by two branches of the median nerve, both entering the deep aspect of the muscle about one-third of the distance from muscle origin to insertion. The flexor carpi ulnaris is a composed of two parts, the ulnar and humeral heads. The ulnar head is small, originating directly by fibers on the medial aspect of the olecranon process and containing relatively short fibers (about 9 mm long) that are highly invested with tendinous partitions. The humeral head of flexor carpi ulnaris comprises the largest part of this muscle and originates on the medial epicondyle and elbow joint capsule immediately caudal to the origin of the flexor carpi radialis. The muscle fibers of humeral head are slightly longer than those of the ulnar head, and the muscle belly extends along the proximal seven-eights of the radius before giving rise to a broad, flat tendon inserting on the accessory carpal bone. Two branches of the ulnar nerve provide separate innervation of the ulnar and humeral heads of the flexor carpi ulnaris, respectively.
were obtained from up to three regions, one on the proximal surface and extending approximately 1cm into the depth of the muscle, yielding 61% type I fibers (N = 9 animals). A similar surface sample from the distal one-third of the muscle yielded 55% type I fibers (N = 8 animals). A sample was obtained from the deep surface of the muscle near its midbelly in 7 animals yielding 54% type I fibers, not significantly different from the two sampled regions on the superficial surface of the superficial flexor. The similarity between regions is not surprising. Because the superficial flexor is not much more than 1 cm thick, deep and superficial samples tend to be overlapping. No type IIb fibers were observed in any of the superficial flexor samples. Although the percentage of type 1 and type I1 fibers was relatively consistent between animals, there was a distinct clumping of regions of type I or type IIa fiber type predominance (Fig. 3). Thus, it was possible to observe a single fascicle containing almost 70% type I fibers immediately adjacent to a fascicle containing only 35% type I fibers. This fiber type distribution was unique to the superficial flexor among the muscles studied. The deep flexor contained relatively more Histochemistry type IIa fibers than did the superficial flexor. Three fiber types were observed in the Of the two regions studied in the humeral forearm flexor muscles, primarily including head of deep digital flexor, the short comparttype I, IIa, and infrequently including IIb ment contained a mean of 44% type I fibers, fibers in some muscles (Table 2). Differentia- whereas the adjacent long compartment contion of type I versus type I1 fibers was con- tained a mean of 33% type I fibers (Fig. 4). firmed with immunocytochemicalstaining of Only 3 of 7 short compartments, and only 3 serial sections. The superficial flexor was com- of 6 long compartments examined contained posed of about 56% type I fibers. Samples type IIb fibers. The small radial head of the TABLE 2. Mean percentage ofthree fiber &pes expressed in forearm flexor muscles of horse" Fiber tvDe ~~~
I
IIa
IIb
N
NlIb
44A(15) 33A,C(8) 6B (6) 28c ( 6 )
51F(15) 55F (11) 78G(19) 72O ( 6 )
4H(8) 13H(9) 16H(23)
9 6 8 7
4 4 3 0
59O (8) 5SD(9) 54D3E( 6 )
3gF (8) 41F (10) 45F (5)
OH
9 5
0 2 1
38A,E(6) 34'43 (11)
41F (11) 33F( 6 )
21' (11) 33' (8)
5 3
4@ (8) 39A (7) 13B( 9 )
60G(8) 61G (7) 85G(8)
OH
0 0 1
~~
Deep digital flexor: Humeral head (short) Humeralhead (long) Ulnar head Radial head Superficial digital flexor: Superficial,proximal Superficial, distal Deep, midbelly Flexor carpi radialis: Superficial,proximal Superficial, distal Flexor carpi ulnaris: Superficial, midbelly Deep, midbelly U l n a head
OH
0.4H(0.6) 1H (1)
OH
3" ( 6 )
I
*Muscleabbreviations are identified in text. Standard deviationsare given in parentheses.The No. of samples studied (N)and the No. of those samples containing IIb fibers (NIB) are given. Means identified by different superscript letters were significantly different
(P < .05).
274
J.W. HERMANSON AND M.A. COBB
Fig. 3. Equus caballus. Representative cross sections of the long (A) and short (B) heads of the deep digital flexor and the superficial digital flexor ( C ) , stained for mATPase after acidic preincubation (pH 4.40). Dark stained fibers are type I and are presumed slow-twitch.Unstained (or lightly stained) fibers are type IIa. The intermediate staining fibers shown here in the short head of the deep flexor are type IIb. Scale bar = 50 pm.
HORSE FLEXOR MUSCLES
275
Fig. 4. Equus caballus. Representative cross sections of the long (A), short (B), u l n a ( C ) , and radial (D ) heads of the deep digital flexor from one horse and stained for mATPase after acidic preincubation (pH 4.40). Among these four muscle regions, the ulnar head is unique and, in this sample, contains only type Ila fibers. No type IIb fibers are present in these samples. Scale bar = 50 pm
deep flexor contained 23% type I fibers and never exhibited type IIb fibers. In sharp contrast, the ulnar head of the deep flexor contained only 6% type I fibers, significantly different from percentages in the remainder of the deep flexor. Although the mean type IIb fiber population was 19%, five of the eight animals sampled contained no type IIb fibers. Thus, three animals sampled contained a surprisingly high number of Ilb fibers in the ulnar head. Two regions of both the flexor carpi radialis and flexor carpi ulnaris were examined
and found to be comparable to the humeral head of deep flexor. The flexor carpi radialis contained about 36%type I fibers, 38%type IIa fibers, and 27% type IIb fibers (Fig. 5). Type 1Ib fibers were observed in all flexor carpi radialis muscles studied. The humeral head of flexor carpi ulnaris contained about 40% type I fibers, not significantly different from the humeral head of the deep flexor. No type IIb fibers were observed in any flexor carpi ulnaris muscles, in contrast to flexor carpi radilais, suggesting that myosin IIb expression is limited to certain muscles.
J.W. HERMANSON AND M.A. COBB
276
Fig. 5. Equus caballus. Representative serial sedions of the flexor carpi radialis muscle stained for mATPase after preincubation at pH 4.45 (A), and preincubation at pH 10.3 (B). Type I fibers appear dark in A and light in B. These same fibers do not react with anti-fast antibody (C) but do stain darkly after incubation with an anti-slow antibody (D). Conversely, type IIa and IIb fibers (dark in B) stain darkly after incubation with anti-fast but not with anti-slow antibodies. Scale bar = 50 wm.
DISCUSSION
Functional design of the forearm flexors Both the superficial and the deep digital flexors are similar in having check ligaments that perform a significant role by transferring responsibility for carrying the animal's weight to non-tiring ligamentous systems (Fig. 1).This is where the similarity ends. The deep digital flexor has three heads (Dyce et al., '87; Sack, '91). Consideration of the distribution of primary nerve branches t o these heads suggests further that the long
and short regions of the humeral head have separate innervation by the median nerve and are properly termed separate compartments (English and Letbetter, '82).Although fiber lengths are similar, on average, in the long and short heads, the long region appears to contain more connective tissue septa. The short region, in contrast, comprises one part with connective tissue similar to that found in the superficial digital flexor and another part containing long fibers and little connective tissue. The multiple heads provide extra
HORSE FLEXOR MUSCLES
degrees of freedom for the deep flexor, although its radial and ulnar heads are relatively small. The ulnar head has a uniquely fast fiber type composition, and fairly long fiber lengths. These characteristics suggest a role in rapid response to tendon length perturbations. The superficial digital flexor is much simpler than the deep flexor, possessing only a single head and a distinct morphology throughout. The superficial flexor is characterized by short fibers and significant connective tissue partitions within the muscle. In addition, the superficial flexor has a significantly higher concentration of slow twitch muscle fibers (generally exceeding 50%)than its neighboring deep flexor, and no superficial digital flexor muscles examined contained type IIb fibers. A major difference, therefore, between the superficial flexor and the deep flexor (humeral, radial, and ulnar heads) appears to be the higher proportion of type I, presumed slow twitch fibers, and the “short-fiber” architecture of superficial flexor. This difference appears consistent with recent biomechanical analysis of the digital flexor muscles. It was proposed that the superficial flexor was prone to twice as much stress as the deep flexor (Dimery et al., ’86; Ker et al., ’88)and was also more important in providing elastic storage of kinetic energy during Iocomotion (Biewener and Rizzo, ’89) in the horse. This can be explained in two ways. First, the proximal check ligament is capable of bearing the weight of the limb without muscular effort. Thus, limb movements experienced during walking (or possibly a slow trot) could serve to stretch the superficial flexor tendon-check ligament system. Elastic loading of the superficial flexor tendon and check ligament at high speeds could lead to fetlock and pastern (synomymous with metacarpholangeal and proximal interphalangeal joints of human anatomy) flexion with minimal input from the superficial digital flexor muscle belly. In fact, the superficial flexor may serve to maintain base level tension on its tendon-check ligament system. In addition to shortening caused by elastic recoil, fetlock and pastern flexion may be effected by inertial factors resulting from action in the larger elbow extensors and shoulder flexors. Similar models have been proposed to describe the interaction of hip and knee movements during galloping in cats (Hoy and Zernicke, ’85). Second, because of the architecture of the superficial flexor and its postural histochemi-
277
cal composition, the muscle belly is unlikely to account for much of the cllgital flexion effected by its tendon at the fetlock and pastern joints. Ker et al. (’88)calculated a dimensionless value, L, that indicates the relative contribution of muscle fibers to displacement at the insertion point of the tendon. These authors found the superficial flexor to have the lowest value of L among the muscles included in their study, which encompassed nine species. Thus, a contraction of fibers of the superficial flexor is ineffective at pulling the distal insertions of the muscle proximally. Such a mechanism might be compatible, however, with the observation of Goslow et al. (’81) that several dog limb muscles exhibit activity during periods in which the musculotendinous unit undergoes little or no length change or is being stretched. Although the deep flexor was also found to have a low L value, it has significantly more potential to effect displacement at the insertional site of the tendon upon the distal phalanx of the digit (Ker et al., 1988). Our data suggest that the humeral head of the deep flexor represents a muscle capable of storing some elastic energy, but also of effecting more precise control over the “spring” as well as being capable of modulating the role of the distal check ligament during trot and gallop. The fact that the fibers of the superficial flexor are relatively short when compared to the deep flexor may represent a phylogenetic constraint. A similar comparison between the superficial and deep flexors was also noted in raccoons (Procyon lotor) and coatis (Nasua narica) by McClearn (’89, who suggested that moment arm lengths and joint-angle excursions may be prime determinants of fiber length. This pattern of fiber size distribution can be interwoven with observations on the pattern of highest density of type I fibers to pose questions about the design of the mammalian limb. Although Armstrong (1980) summarized that type I fibers were most often concentrated in muscles close to the deep axis of a limb, later analysis of the dog forelimb showed that the superficial flexor contained relatively more type I fibers than did the more deeply placed deep flexor (Armstrong et al., 1982). Our observations in the horse forelimb show a similar distribution of a high frequency of type I fibers in the superficial flexor relative to surrounding or deeper muscles (Fig. 6). Whether or not the fiber size or fiber type distribution patterns
278
J.W. HERMANSON AND M.A. COBB
Lateral
Fig. 6. Equus caballus. A transverse section of the left forelimb of a horse obtained approximately twothirds of the distance along the radius from the elbow. The humeral heads (long and short regions) constitute most of the cross-sectional area of the flexor musculature. The superficial digital flexor is closely applied over the deep digital flexor and contains prominent bands of connective tissue (cross-hatched). The short and part of the long region of the deep flexor contain fleshy regions,
largely void of tendinous inscriptions. At this level, ulnaris lateralis and flexor carpi ulnaris are broad structures made up equally of myofibers and tendon, and the ulnar head of the deep flexor is entirely tendinous. The highest proportion of type I fibers in these muscles is found in the superficial flexor, which is not deeply located and thus contrasts with other published studies on fiber type distributions.
are primarily determined by phylogeny or by function remains to be seen. If muscle histochemistry is a good predictor of muscle function, these regional specializations suggest that the superficial and deep flexors may perform subtly different functions during locomotion, with the deep flexor perhaps taking a more active role in imple-
menting digital flexion during higher speed gaits. Alternatively, the deep flexor may be critical to powering digital flexion during low speed gaits, a role largely mitigated at higher speeds when passive loading of the tendon during the stance phase may provide enough stored energy to passively effect digital flexion during late stancephase and early swing phase.
279
HORSE FLEXOR MUSCLES
Although not central to this discussion, observations from the radial and ulnar carpal flexors show a trend toward expression of type IIb fiber types, and a different architecture than observed in the digital flexor muscles. The flexor carpi ulnaris and the flexor carpi radialis are similar muscles in terms of attachment and size, but they show significant differences in architecture and fiber type expression. In contrast to the superficial and the deep flexors, these muscles span the carpus but not the digitaljoints. The longer fiber length of the flexor carpi radialis may reflect a greater degree of shortening effected by this muscle which spans all of the intrinsic joints of the carpus. The flexor carpi ulnaris has shorter fibers and attaches to the accessory carpal bone. The consistent presence of type IIb fibers in the flexor carpi radialis and their absence in the flexor carpi ulnaris demonstrate a muscle-specific myosin isoform expression and suggest differences in the roles of these muscles. Evolutionary and clinical correlations How might these data address issues of tendon breakdown? The major differences between the superficial and deep digital flexors, described above, indicate that the latter muscle has more heads with different attachment sites as well as a greater range of motor control functions available. The deep digital flexor has heads originating on the humerus (crossing the elbow joint) as well as on the radius and ulna (not crossing the elbow joint). These latter origins might be insulated from the joint angle changes observed at the elbow during locomotion on an incline and from activity in the superficial flexor where it is adherent to the humeral head of the deep flexor (short head). However, given the diminutive size of these two heads relative to the humeral head, it is doubtful that their functions have great significance. The degree of motor control possible in the muscles is, however, correlated with the potential range of contraction velocities and individual fiber force profiles available within the muscle. We argue that the architecture and histochemical composition of the deep flexor provide a greater breadth of motor control strategies than might be available in the superficialdigital flexor because of the latter’s uniform “short-fiber” construction and emphasis on type I fibers. Shorter muscle fibers have been shown to have decreased contraction velocities (Josephson, ’75). Type I fibers have lower potential to
generate force, and to generate force rapidly (Burke, ’81).Thus, when the musculoskeletal system of a horse is stressed, such as during training or competitive events, it is possible that the superficial flexor is less capable of defending the muscle from extrinsic (irregularities in the substrate) or intrinsic (muscle fatigue) perturbations that might alter limb kinematics. If these properties are coupled with the smaller cross-sectional area of the superficial flexor tendon, these data may provide a mechanical basis for Webbon’s (’77) finding that a majority of the forelimb tendon breakdowns occur in the superficial flexor’s tendon in the region proximal to the fetlock joint. ACKNOWLEDGMENTS
We thank Dr. Hussni Mohammed for assistance with statistical analysis. Drs. Howard Evans, Beth Valentine, Gilbert Burns, and Alan Nixon provided helpful discussion of an earlier version of this manuscript. Dr. Frank Stockdale generously provided the anti-slow antibody. We appreciate assistance provided by Ms. Ruthie LoefRer in preparing the final drafts of this manuscript. This work was supported by the H.M. Zweig Memorial Fund for Equine Research. LITERATURE CITED Armstrong, R.B. (1980) Properties and distribution of the fiber types in the locomotory muscles of mammals. In K. Schmidt-Nielson, L. Bolis, and C.R. Taylor (eds): Comparative Physiology: Primitive Mammals. Cambridge: Cambridge Univ. Press, pp. 243-284. Armstrong, R.B., C.W. Saubert, IV, H.J. Seeherman, and C.R. Taylor (1982) Distribution offiber typesin locomotory muscles of dogs. Am. J. Anat. 163:87-98. Biewener, A.A., and N. Rizzo (1989) Elastic energy storage in the horse. Am. Zool.29:182a. Burke, R.E. (1981) Motor units: anatomy, physiology, and functional organization. In J.M. Brookhart and V.B. Mountcastle (eds): Handbook of Physiology. Val. 2, Motor Control, Part 2. Baltimore: American Physiological Society, pp. 345-422. Dimery, N.J., R. M. Alexander, and R.F. Ker (1986) Elastic extension of leg tendons in the locomotion of horses (Equus caballus). J . 2001. Land. (A) 210t415425. Dyce,K.M., W.O. Sack,andC.J.G. Wensing(1987) Textbook of Veterinary Anatomy. Philadelphia: Saunders co. English, A.W., and W.D. Letbetter (1982) Anatomy and innervation patterns of cat lateral gastrocnemius and plantaris muscles. Am. J . Anat. 164:67-77. Goslow, G.E., Jr., H.J. Seeherman, C.R. Taylor, M.N. McCutchin, and N.C. Heglund (1981) Electrical activity and relative length changes of dog limb muscles as a function of speed and gait. J. Exp. Biol. 94t15-42. Hermanson, J.W., and K.J. Hurley (1990) Architectural and histochemical analysis of the biceps brachii muscle ofthe horse. ActaAnat. (Basel) 137:146-156. Hoy, M.G., and R.F. Zernicke (1985) Modulation of limb
280
J.W.HERMANSON AND M.A. COBB
dynamics in the swing phase of locomotion. J. Biomech. 18~49-60. Josephson, R.K. (1975) Extensive and intensive factors determining the performance of striated muscle. J . EXP.2001.194:135-154. Ker, R.F., R.M. Alexander, and M.B. Bennett (1988)Why are mammalian tendons so thick? J. 2001. Land. (A) 216:309-324. Kingsbury, H.B., M.A. Quddus, J.R. Rooney, and J.E. Geary (1978) A laboratory system for production of flexion rates and forces in the forelimb of the horse. Am. J. Vet. Res. 39:365-369. Matoba, H., and P.D. Gollnick (1984) Influence of ionic composition, buffering agent, and pH on the histochemical demonstration of myofibrillar actomyosin ATPase. Histochemistry 80t609-614. McClearn, D. (1985) Anatomy of raccoon (Procyon lotor) and coati (Nasua narica) forearm and leg muscles:
Relations between fiber length, moment-arm length, and joint-angle excursion. J. Morphol. 183337-115. Ruckebusch, Y. (1972) The relevance of drowsiness in the circadian cycle of farm animals. Anim. Behav. 20:637643. Sack, W.O. (1989) The stay-apparatus of the horse’s hindlimb-explained. Equine Practice 11:31-35. Sack, W.O. (1991) Rooney’s Guide to the Dissection of the Horse. Ithaca: Veterinary Textbooks. Stashak, T.S. (1987) Adam’s Lameness in Horses. Philadelphia: Lea and Febiger. Webbon, P.M. (1977) A post mortem study of equine digital flexor tendons. Equine Vet. J. 9r61-67. Ypey, D.L. (1978) A topographical study of the distribution of end-plates in the cutaneus pedoris, sartorius, and gastrocnemius muscles of the frog. J. Morphol. 155:327-348.
Four Forearm Flexor Muscles of the Horse, Equus caballus: Anatomy and Histochemistry JOHN W. HERMANSON AND MArTHEW A. COBB Department ofAnatomy, College of Veterrnor?/Medicine, Cornell Unruersity, Ithaca, New York 14853
ABSTRACT Two of the forearm flexors of the horse, the superficial and deep digital flexor muscles, are critical to support the digital and fetlock joints, exhibit differing insertions, and are passively supported by the proximal and distal check ligaments, respectively. These two muscles differ in histochemical composition and architecture. The differences are correlated with the different stress levels transmitted through their tendons, and the different frequencies of clinical breakdown that have been reported. Both muscles contain type I and type IIa fibers. A few type IIb fibers occurred in the deep digital flexor. The superficial digital flexor contained approximately 56%type I fibers, extremely short muscle fibers, and extensive connective tissue investment. In contrast, the deep digital flexor had three muscle heads: ulnar, radial, and “long” and “short” regions of the humeral head. The “long” and “short” regions of the humeral head contained 33%and 44% type I fibers, respectively, fiber lengths three to four times as long as those in the superficial digital flexor, and relatively less connective tissue investment. Flexor carpi radialis and flexor carpi ulnaris compared most closely with the humeral head of the deep digital flexor. These data suggest a correlation of the unique architecture of superficial digital flexor with its proposed elastic storage properties during locomotion in horses, and a n explanation for the frequent breakdown of the superficial digital flexor in athletic horses. The forelimbs of horses are subject to substantial forces during quiet stance as well as during the stance phase of locomotion. A consideration of mechanics during quiet standing is important as domestic horses stand for 90% of every day, even while sleeping (Ruckebusch, ’72). The passive stayapparatus of the horse is an elaborate system of muscles and ligaments designed to permit stance with minimal muscular effort yet without compromising the cursorial ability of this animal (Sack, ’91). Components of the passive stay-apparatusof the hindlimb have been described in detail (Sack, ’89), but less information is available regarding components of the forelimb passive stay-apparatus, specifically those involving the elbow and shoulder joints. Although several studies have examined the biomechanical properties of the digital flexor muscles of horse forelimbs (Kingsbury et al., ’78; Dimery et al., ’86; Ker et al., ’88;Biewener and Rizzo, ’89)’ the physiological properties of these muscles remain largely unknown. The present study presents a n 8
1992 WILEY-LISS. INC
analysis of muscle morphology and histochemistry in the forearm flexors of the horse, including the flexor carpi radialis, flexor carpi ulnaris, superficial digital flexor, and deep digital flexor. The flexor carpi radialis and flexor carpi ulnaris are considered relatively simple in terms of their roles a t the elbow and carpal joints. The superficial digital flexor and deep digital flexors are significantly more complex both in terms of architecture (the deep flexor has three major heads with differing origins) and function (the deep and superficial flexors traverse from four to six functional joints). In particular, this study emphasizes comparison of presumed function in the superficial and deep digital flexor muscles. Although both of these muscles function as digital flexors, their insertions differ sufficiently to effect subtly different roles in locomotion and stance (Stashak, ’87). Biomechanical modeling of these muscles suggests that the superficial flexor tendon is exposed to almost twice the amount of stress as that imposed on the deep flexor tendon
270
J.W. HERMANSON AND M.A. COBB
(Ker et al., '88). Ker et al. suggest that such factors, as well as the relatively smaller crosssectional area of the superficial flexor tendon in the metacarpal region, predispose the muscle to the higher frequency of tendon breakdowns than is observed in the deep flexor tendon (Webbon, '77). Architectural and histochemical data are correlated with these mechanical hypotheses. MATERIALS AND METHODS
Whole flexor carpi radialis, flexor carpi ulnaris, deep digital flexor, and superficial digital flexor muscles were obtained from nine horses, Equus caballus (Table l), immediately following euthanasia by overdose of sodium pentobarbital. The animals were euthanized for reasons unrelated to musculoskeletal performance. The muscles were blotted dry and 2 cm3 blocks of muscle were taken from the following regions of each muscle. One or two sections were taken from the flexor carpi radialis: one from the proximal region (N = 6) and one from the distal region (N = 3). One to three sections of each flexor carpi ulnaris were taken including superficial (N = 8) and deep midbelly (N = 7) regions of the humeral head and a midbelly section of the ulnar head (N = 5). A maximum of four sections were taken from each deep flexor: midbelly regions of both parts of the humeral head (N = 9 or 5), and of the radial (N = 7) and ulnar (N = 8) heads. Three samples were taken from superficial flexors: proximal (N = 9) and distal (N = 5) superficial regions and a deep midbelly region (N = 7). Connective tissue landmarks were used to ensure that samples were taken from similar locations in all muscles. The muscle samples were mounted on cork with 5% gum tragacanth and quick frozen in isopentane cooled to approximately - 160°C by liquid nitrogen. Samples were then stored at -60 to 80°C.
TABLE 1. Description of horses used in the histochemical studv o f forearm flexor muscles Horse
Age (yrs) 8 7 9 7 6 20 12 6 20
Sex
Breed
Weight (kg)
F
Thoroughbred Quarterhorse Thoroughbred Standardbred Standardbred Standardbred Standardbred Thoroughbred Standardbred
511 591 490 485 400 499 494 421 412
F F F F F F F F
Full sets of all muscle samples were not obtained from each animal. Transverse serial sections (10 pm thickness) of muscle samples were obtained using a cryostat and mounted on glass slides for staining. Sections of each muscle sample were stained for myofibrillar ATPase (mATPase) following acidic and basic (pH 10.3) preincubation. A range of pHs (from 4.3 to 4.55) was used following initial studies spanning a pH range of 4.2 to 4.7 that indicated the usefulness of this range for fiber type identification following acidic preincubation and mATPase staining (see also Matoba and Gollnick, '84). Serial sections were stained for nicotinamide dinucleotide tetrazolium reductase (NADHTR) to show oxidative activity, and alpha glycerophosphate dehydrogenase (alpha GPD) to show glycolytic ability. Histochemical and morphometric procedures were used as previously described in Hermanson and Hurley ('90). Differences in histochemical profiles were assessed using ANOVA and a posthoc Bonferroni's comparison of multiple means (SAS Institute, Cary, NC). Significance was accepted when P I .05. Additional 10 km serial sections of each sample were stained for fast and slow myosin isoforms using immunocytochemical techniques. Sections were obtained in the same manner as for histochemistry, mounted on glass slides, and kept at -20°C until needed. Sections were allowed to warm to room temperature and then covered with 2% normal goat serum and incubated at 4°C in a humid chamber for 30 minutes. The normal goat serum was then shaken off and the primary antibody applied and allowed to incubate for 14 hours in a humid chamber a t 4°C. Fast myosin was reacted with a commercially available anti-fast myosin antibody Sigma MY-32 (Sigma Chemical Co., St. Louis). Slow myosin was reacted with anti-slow myosin antibody S-58 (provided by Dr. F. Stockdale, Stanford Medical School). The sections were dipped and then soaked for 10 minutes in a 0.05 M phosphate buffer solution (PBS) with 0.85%NaCl. A biotinylated rabbit anti-mouse secondary antibody was then applied for 10 minutes and the PBS rinsing repeated. Next, the sections were reacted with a streptavidin enzyme conjugate for 5 minutes and the PBS rinse was repeated. Staining was performed using a substrate chromagen for 15 minutes. The secondary antibody, enzyme conjugate, and substrate chromagen make up a commercially available staining kit (Histostain Sp.,
HORSE FLEXOR MUSCLES
Zymed Inc., San Francisco). Dissections were performed on two fresh and one formalin fixed forelimb specimen to determine patterns of innervation and connective tissue in the muscles studied. Whole muscles were obtained from two adult horses and a pony at postmortem, blotted dry, and fixed at their in situ lengths in 5%formalin solution. These specimens were then transferred to an approximately 15% nitric acid solution for connective tissue digestion. When fiber bundles could be teased apart, the muscles were transferred to a 50% glycerin solution. Under a dissecting microscope, individual muscle fibers were teased apart and measured to obtain estimates of mean fiber length for each of the regions sampled for histochemistry and immunocytochemistry. This yielded a measure of mean muscle fiber length and should not be confused with any attempt to measure overall muscle length or the distance between a tendon of origin and of insertion. Two flexor carpi radialis, flexor carpi ulnaris, deep digital flexor, and superficial digital flexor muscles were obtained from two adult horses and stained for acetylcholinesterase (Ypey, '78) to demonstrate the distribution of motor endplates within each muscle. Terminology used in this study conforms with the Nomina Anatomica Veterinaria (1983).
271
Deep digital flexor, humeral head Superficial digital flexor Proximal check lig.
Distal check lig.
Sesamoids at fetlock joint
RESULTS
Anatomy The superficial digital flexor is composed of a single head, innervated by a single branch of the ulnar nerve that enters the muscle about 3 cm from the muscle's proximal end. The muscle fibers are short, being only about 3 mm long (measured in a pony) and pinnately arranged. Three regions of the muscle were sampled for fiber length as well as histochemistry. All regions had muscle fibers of similar length. The superficial flexor was difficult to separate from the underlying deep flexor along the proximal two-thirds of its length (Figs. 1, 2) and shares a common fleshy origin with the humeral head of that muscle. A proximal check ligament arises from the distal one-third of the radius and courses distally to intersect with the superficial flexor tendon caudal to the carpus. The insertional tendon of the superficial flexor passes distally from the carpus to its attachments on the base of the proximal and middle phalanges.
Insertion of deep digital flexor Fig. 1. Equus caballus. Lateral view schematic of left forelimb showing the arrangement of the humeral head of the deep digital flexor and the superficial digital flexor muscles. Gravity tends to cause shoulder and elbow flexion (arrow), as well as overextension (dorsiflexion) of the fetlock (metacarpophalangeal),pastern (proximal interphalangeal), and coffin (distal interphalangeal) joints. Overextensionof the fetlock causes passive stretch of the deep and superficial flexors that is resisted by the distal and proximal check ligaments, respectively.Modified after Sack ('91).
The deep digital flexor is composed of three heads. The humeral head is the largest subdivision of the deep flexor, originating in common with the superficial flexor from the medial epicondyle of the humerus. The humeral head is divided by a connective tissue partition into a long and short region. The radial head comprises a relatively small subdivision with a maximum outer diameter of about 1
272
J.W. HERMANSON AND M.A. COBB
Fig. 2. Equus caballus. An oblique, caudolateral view of left forelimb of a horse. The area in the shaded box is magnified in the view to the right. The check ligaments serve as non-tiring support structures during quiet stance. The superficial digital flexor has been resected to show its region of adherence (dark shading) to the surface of the long and short humeral heads of the deep digital flexor. The ulnar and radial heads of the deep flexor are significantlysmaller than are the humeral parts. (Modified after Sack, '91).
cm, and exhibiting the shortest fiber lengths of any region of the deep flexor. The radial head fuses with the insertional tendon of deep flexor at the carpus. Both the humeral and radial heads of the deep flexor are innervated by branches of the median nerve. The median nerve splits into two primary nerve branches before entering the long and short humeral heads, suggesting that these regions may function as muscle compartments (sensu English and Letbetter, '82). A small ulnar head of the deep flexor muscle arises from the medial surface of the olecranon process. The muscular portion of this head is only about 8 cm long (in the pony) and extremely flat, but contains pinnately arranged fibers about 17 mm long (similar in length to the
much larger humeral head). The ulnar head is innervated by a branch of the ulnar nerve. A distal check ligament originates from the palmar carpal ligament and courses 5-6 cm distally to an attachment on the insertional tendon of the deep flexor. This tendon passes distally to its insertion upon the base of the terminal phalanx. The flexor carpi radialis is pinnate and originates from the medial epicondyle of the humerus and the medial collateral ligament and joint capsule of the elbow joint. Insertion is by a broad short tendon on the base of the second metacarpal. Fiber length is similar (about 37 mm in the pony) for two regions of the muscle investigated. Thus, these fibers are at least twice as long as fibers found in
2 73
HORSE FLEXOR MUSCLES
the deep flexor. The flexor carpi radialis makes up about 11%of the total mass of the forearm flexor muscle group. Innervation is by two branches of the median nerve, both entering the deep aspect of the muscle about one-third of the distance from muscle origin to insertion. The flexor carpi ulnaris is a composed of two parts, the ulnar and humeral heads. The ulnar head is small, originating directly by fibers on the medial aspect of the olecranon process and containing relatively short fibers (about 9 mm long) that are highly invested with tendinous partitions. The humeral head of flexor carpi ulnaris comprises the largest part of this muscle and originates on the medial epicondyle and elbow joint capsule immediately caudal to the origin of the flexor carpi radialis. The muscle fibers of humeral head are slightly longer than those of the ulnar head, and the muscle belly extends along the proximal seven-eights of the radius before giving rise to a broad, flat tendon inserting on the accessory carpal bone. Two branches of the ulnar nerve provide separate innervation of the ulnar and humeral heads of the flexor carpi ulnaris, respectively.
were obtained from up to three regions, one on the proximal surface and extending approximately 1cm into the depth of the muscle, yielding 61% type I fibers (N = 9 animals). A similar surface sample from the distal one-third of the muscle yielded 55% type I fibers (N = 8 animals). A sample was obtained from the deep surface of the muscle near its midbelly in 7 animals yielding 54% type I fibers, not significantly different from the two sampled regions on the superficial surface of the superficial flexor. The similarity between regions is not surprising. Because the superficial flexor is not much more than 1 cm thick, deep and superficial samples tend to be overlapping. No type IIb fibers were observed in any of the superficial flexor samples. Although the percentage of type 1 and type I1 fibers was relatively consistent between animals, there was a distinct clumping of regions of type I or type IIa fiber type predominance (Fig. 3). Thus, it was possible to observe a single fascicle containing almost 70% type I fibers immediately adjacent to a fascicle containing only 35% type I fibers. This fiber type distribution was unique to the superficial flexor among the muscles studied. The deep flexor contained relatively more Histochemistry type IIa fibers than did the superficial flexor. Three fiber types were observed in the Of the two regions studied in the humeral forearm flexor muscles, primarily including head of deep digital flexor, the short comparttype I, IIa, and infrequently including IIb ment contained a mean of 44% type I fibers, fibers in some muscles (Table 2). Differentia- whereas the adjacent long compartment contion of type I versus type I1 fibers was con- tained a mean of 33% type I fibers (Fig. 4). firmed with immunocytochemicalstaining of Only 3 of 7 short compartments, and only 3 serial sections. The superficial flexor was com- of 6 long compartments examined contained posed of about 56% type I fibers. Samples type IIb fibers. The small radial head of the TABLE 2. Mean percentage ofthree fiber &pes expressed in forearm flexor muscles of horse" Fiber tvDe ~~~
I
IIa
IIb
N
NlIb
44A(15) 33A,C(8) 6B (6) 28c ( 6 )
51F(15) 55F (11) 78G(19) 72O ( 6 )
4H(8) 13H(9) 16H(23)
9 6 8 7
4 4 3 0
59O (8) 5SD(9) 54D3E( 6 )
3gF (8) 41F (10) 45F (5)
OH
9 5
0 2 1
38A,E(6) 34'43 (11)
41F (11) 33F( 6 )
21' (11) 33' (8)
5 3
4@ (8) 39A (7) 13B( 9 )
60G(8) 61G (7) 85G(8)
OH
0 0 1
~~
Deep digital flexor: Humeral head (short) Humeralhead (long) Ulnar head Radial head Superficial digital flexor: Superficial,proximal Superficial, distal Deep, midbelly Flexor carpi radialis: Superficial,proximal Superficial, distal Flexor carpi ulnaris: Superficial, midbelly Deep, midbelly U l n a head
OH
0.4H(0.6) 1H (1)
OH
3" ( 6 )
I
*Muscleabbreviations are identified in text. Standard deviationsare given in parentheses.The No. of samples studied (N)and the No. of those samples containing IIb fibers (NIB) are given. Means identified by different superscript letters were significantly different
(P < .05).
274
J.W. HERMANSON AND M.A. COBB
Fig. 3. Equus caballus. Representative cross sections of the long (A) and short (B) heads of the deep digital flexor and the superficial digital flexor ( C ) , stained for mATPase after acidic preincubation (pH 4.40). Dark stained fibers are type I and are presumed slow-twitch.Unstained (or lightly stained) fibers are type IIa. The intermediate staining fibers shown here in the short head of the deep flexor are type IIb. Scale bar = 50 pm.
HORSE FLEXOR MUSCLES
275
Fig. 4. Equus caballus. Representative cross sections of the long (A), short (B), u l n a ( C ) , and radial (D ) heads of the deep digital flexor from one horse and stained for mATPase after acidic preincubation (pH 4.40). Among these four muscle regions, the ulnar head is unique and, in this sample, contains only type Ila fibers. No type IIb fibers are present in these samples. Scale bar = 50 pm
deep flexor contained 23% type I fibers and never exhibited type IIb fibers. In sharp contrast, the ulnar head of the deep flexor contained only 6% type I fibers, significantly different from percentages in the remainder of the deep flexor. Although the mean type IIb fiber population was 19%, five of the eight animals sampled contained no type IIb fibers. Thus, three animals sampled contained a surprisingly high number of Ilb fibers in the ulnar head. Two regions of both the flexor carpi radialis and flexor carpi ulnaris were examined
and found to be comparable to the humeral head of deep flexor. The flexor carpi radialis contained about 36%type I fibers, 38%type IIa fibers, and 27% type IIb fibers (Fig. 5). Type 1Ib fibers were observed in all flexor carpi radialis muscles studied. The humeral head of flexor carpi ulnaris contained about 40% type I fibers, not significantly different from the humeral head of the deep flexor. No type IIb fibers were observed in any flexor carpi ulnaris muscles, in contrast to flexor carpi radilais, suggesting that myosin IIb expression is limited to certain muscles.
J.W. HERMANSON AND M.A. COBB
276
Fig. 5. Equus caballus. Representative serial sedions of the flexor carpi radialis muscle stained for mATPase after preincubation at pH 4.45 (A), and preincubation at pH 10.3 (B). Type I fibers appear dark in A and light in B. These same fibers do not react with anti-fast antibody (C) but do stain darkly after incubation with an anti-slow antibody (D). Conversely, type IIa and IIb fibers (dark in B) stain darkly after incubation with anti-fast but not with anti-slow antibodies. Scale bar = 50 wm.
DISCUSSION
Functional design of the forearm flexors Both the superficial and the deep digital flexors are similar in having check ligaments that perform a significant role by transferring responsibility for carrying the animal's weight to non-tiring ligamentous systems (Fig. 1).This is where the similarity ends. The deep digital flexor has three heads (Dyce et al., '87; Sack, '91). Consideration of the distribution of primary nerve branches t o these heads suggests further that the long
and short regions of the humeral head have separate innervation by the median nerve and are properly termed separate compartments (English and Letbetter, '82).Although fiber lengths are similar, on average, in the long and short heads, the long region appears to contain more connective tissue septa. The short region, in contrast, comprises one part with connective tissue similar to that found in the superficial digital flexor and another part containing long fibers and little connective tissue. The multiple heads provide extra
HORSE FLEXOR MUSCLES
degrees of freedom for the deep flexor, although its radial and ulnar heads are relatively small. The ulnar head has a uniquely fast fiber type composition, and fairly long fiber lengths. These characteristics suggest a role in rapid response to tendon length perturbations. The superficial digital flexor is much simpler than the deep flexor, possessing only a single head and a distinct morphology throughout. The superficial flexor is characterized by short fibers and significant connective tissue partitions within the muscle. In addition, the superficial flexor has a significantly higher concentration of slow twitch muscle fibers (generally exceeding 50%)than its neighboring deep flexor, and no superficial digital flexor muscles examined contained type IIb fibers. A major difference, therefore, between the superficial flexor and the deep flexor (humeral, radial, and ulnar heads) appears to be the higher proportion of type I, presumed slow twitch fibers, and the “short-fiber” architecture of superficial flexor. This difference appears consistent with recent biomechanical analysis of the digital flexor muscles. It was proposed that the superficial flexor was prone to twice as much stress as the deep flexor (Dimery et al., ’86; Ker et al., ’88)and was also more important in providing elastic storage of kinetic energy during Iocomotion (Biewener and Rizzo, ’89) in the horse. This can be explained in two ways. First, the proximal check ligament is capable of bearing the weight of the limb without muscular effort. Thus, limb movements experienced during walking (or possibly a slow trot) could serve to stretch the superficial flexor tendon-check ligament system. Elastic loading of the superficial flexor tendon and check ligament at high speeds could lead to fetlock and pastern (synomymous with metacarpholangeal and proximal interphalangeal joints of human anatomy) flexion with minimal input from the superficial digital flexor muscle belly. In fact, the superficial flexor may serve to maintain base level tension on its tendon-check ligament system. In addition to shortening caused by elastic recoil, fetlock and pastern flexion may be effected by inertial factors resulting from action in the larger elbow extensors and shoulder flexors. Similar models have been proposed to describe the interaction of hip and knee movements during galloping in cats (Hoy and Zernicke, ’85). Second, because of the architecture of the superficial flexor and its postural histochemi-
277
cal composition, the muscle belly is unlikely to account for much of the cllgital flexion effected by its tendon at the fetlock and pastern joints. Ker et al. (’88)calculated a dimensionless value, L, that indicates the relative contribution of muscle fibers to displacement at the insertion point of the tendon. These authors found the superficial flexor to have the lowest value of L among the muscles included in their study, which encompassed nine species. Thus, a contraction of fibers of the superficial flexor is ineffective at pulling the distal insertions of the muscle proximally. Such a mechanism might be compatible, however, with the observation of Goslow et al. (’81) that several dog limb muscles exhibit activity during periods in which the musculotendinous unit undergoes little or no length change or is being stretched. Although the deep flexor was also found to have a low L value, it has significantly more potential to effect displacement at the insertional site of the tendon upon the distal phalanx of the digit (Ker et al., 1988). Our data suggest that the humeral head of the deep flexor represents a muscle capable of storing some elastic energy, but also of effecting more precise control over the “spring” as well as being capable of modulating the role of the distal check ligament during trot and gallop. The fact that the fibers of the superficial flexor are relatively short when compared to the deep flexor may represent a phylogenetic constraint. A similar comparison between the superficial and deep flexors was also noted in raccoons (Procyon lotor) and coatis (Nasua narica) by McClearn (’89, who suggested that moment arm lengths and joint-angle excursions may be prime determinants of fiber length. This pattern of fiber size distribution can be interwoven with observations on the pattern of highest density of type I fibers to pose questions about the design of the mammalian limb. Although Armstrong (1980) summarized that type I fibers were most often concentrated in muscles close to the deep axis of a limb, later analysis of the dog forelimb showed that the superficial flexor contained relatively more type I fibers than did the more deeply placed deep flexor (Armstrong et al., 1982). Our observations in the horse forelimb show a similar distribution of a high frequency of type I fibers in the superficial flexor relative to surrounding or deeper muscles (Fig. 6). Whether or not the fiber size or fiber type distribution patterns
278
J.W. HERMANSON AND M.A. COBB
Lateral
Fig. 6. Equus caballus. A transverse section of the left forelimb of a horse obtained approximately twothirds of the distance along the radius from the elbow. The humeral heads (long and short regions) constitute most of the cross-sectional area of the flexor musculature. The superficial digital flexor is closely applied over the deep digital flexor and contains prominent bands of connective tissue (cross-hatched). The short and part of the long region of the deep flexor contain fleshy regions,
largely void of tendinous inscriptions. At this level, ulnaris lateralis and flexor carpi ulnaris are broad structures made up equally of myofibers and tendon, and the ulnar head of the deep flexor is entirely tendinous. The highest proportion of type I fibers in these muscles is found in the superficial flexor, which is not deeply located and thus contrasts with other published studies on fiber type distributions.
are primarily determined by phylogeny or by function remains to be seen. If muscle histochemistry is a good predictor of muscle function, these regional specializations suggest that the superficial and deep flexors may perform subtly different functions during locomotion, with the deep flexor perhaps taking a more active role in imple-
menting digital flexion during higher speed gaits. Alternatively, the deep flexor may be critical to powering digital flexion during low speed gaits, a role largely mitigated at higher speeds when passive loading of the tendon during the stance phase may provide enough stored energy to passively effect digital flexion during late stancephase and early swing phase.
279
HORSE FLEXOR MUSCLES
Although not central to this discussion, observations from the radial and ulnar carpal flexors show a trend toward expression of type IIb fiber types, and a different architecture than observed in the digital flexor muscles. The flexor carpi ulnaris and the flexor carpi radialis are similar muscles in terms of attachment and size, but they show significant differences in architecture and fiber type expression. In contrast to the superficial and the deep flexors, these muscles span the carpus but not the digitaljoints. The longer fiber length of the flexor carpi radialis may reflect a greater degree of shortening effected by this muscle which spans all of the intrinsic joints of the carpus. The flexor carpi ulnaris has shorter fibers and attaches to the accessory carpal bone. The consistent presence of type IIb fibers in the flexor carpi radialis and their absence in the flexor carpi ulnaris demonstrate a muscle-specific myosin isoform expression and suggest differences in the roles of these muscles. Evolutionary and clinical correlations How might these data address issues of tendon breakdown? The major differences between the superficial and deep digital flexors, described above, indicate that the latter muscle has more heads with different attachment sites as well as a greater range of motor control functions available. The deep digital flexor has heads originating on the humerus (crossing the elbow joint) as well as on the radius and ulna (not crossing the elbow joint). These latter origins might be insulated from the joint angle changes observed at the elbow during locomotion on an incline and from activity in the superficial flexor where it is adherent to the humeral head of the deep flexor (short head). However, given the diminutive size of these two heads relative to the humeral head, it is doubtful that their functions have great significance. The degree of motor control possible in the muscles is, however, correlated with the potential range of contraction velocities and individual fiber force profiles available within the muscle. We argue that the architecture and histochemical composition of the deep flexor provide a greater breadth of motor control strategies than might be available in the superficialdigital flexor because of the latter’s uniform “short-fiber” construction and emphasis on type I fibers. Shorter muscle fibers have been shown to have decreased contraction velocities (Josephson, ’75). Type I fibers have lower potential to
generate force, and to generate force rapidly (Burke, ’81).Thus, when the musculoskeletal system of a horse is stressed, such as during training or competitive events, it is possible that the superficial flexor is less capable of defending the muscle from extrinsic (irregularities in the substrate) or intrinsic (muscle fatigue) perturbations that might alter limb kinematics. If these properties are coupled with the smaller cross-sectional area of the superficial flexor tendon, these data may provide a mechanical basis for Webbon’s (’77) finding that a majority of the forelimb tendon breakdowns occur in the superficial flexor’s tendon in the region proximal to the fetlock joint. ACKNOWLEDGMENTS
We thank Dr. Hussni Mohammed for assistance with statistical analysis. Drs. Howard Evans, Beth Valentine, Gilbert Burns, and Alan Nixon provided helpful discussion of an earlier version of this manuscript. Dr. Frank Stockdale generously provided the anti-slow antibody. We appreciate assistance provided by Ms. Ruthie LoefRer in preparing the final drafts of this manuscript. This work was supported by the H.M. Zweig Memorial Fund for Equine Research. LITERATURE CITED Armstrong, R.B. (1980) Properties and distribution of the fiber types in the locomotory muscles of mammals. In K. Schmidt-Nielson, L. Bolis, and C.R. Taylor (eds): Comparative Physiology: Primitive Mammals. Cambridge: Cambridge Univ. Press, pp. 243-284. Armstrong, R.B., C.W. Saubert, IV, H.J. Seeherman, and C.R. Taylor (1982) Distribution offiber typesin locomotory muscles of dogs. Am. J. Anat. 163:87-98. Biewener, A.A., and N. Rizzo (1989) Elastic energy storage in the horse. Am. Zool.29:182a. Burke, R.E. (1981) Motor units: anatomy, physiology, and functional organization. In J.M. Brookhart and V.B. Mountcastle (eds): Handbook of Physiology. Val. 2, Motor Control, Part 2. Baltimore: American Physiological Society, pp. 345-422. Dimery, N.J., R. M. Alexander, and R.F. Ker (1986) Elastic extension of leg tendons in the locomotion of horses (Equus caballus). J . 2001. Land. (A) 210t415425. Dyce,K.M., W.O. Sack,andC.J.G. Wensing(1987) Textbook of Veterinary Anatomy. Philadelphia: Saunders co. English, A.W., and W.D. Letbetter (1982) Anatomy and innervation patterns of cat lateral gastrocnemius and plantaris muscles. Am. J . Anat. 164:67-77. Goslow, G.E., Jr., H.J. Seeherman, C.R. Taylor, M.N. McCutchin, and N.C. Heglund (1981) Electrical activity and relative length changes of dog limb muscles as a function of speed and gait. J. Exp. Biol. 94t15-42. Hermanson, J.W., and K.J. Hurley (1990) Architectural and histochemical analysis of the biceps brachii muscle ofthe horse. ActaAnat. (Basel) 137:146-156. Hoy, M.G., and R.F. Zernicke (1985) Modulation of limb
280
J.W.HERMANSON AND M.A. COBB
dynamics in the swing phase of locomotion. J. Biomech. 18~49-60. Josephson, R.K. (1975) Extensive and intensive factors determining the performance of striated muscle. J . EXP.2001.194:135-154. Ker, R.F., R.M. Alexander, and M.B. Bennett (1988)Why are mammalian tendons so thick? J. 2001. Land. (A) 216:309-324. Kingsbury, H.B., M.A. Quddus, J.R. Rooney, and J.E. Geary (1978) A laboratory system for production of flexion rates and forces in the forelimb of the horse. Am. J. Vet. Res. 39:365-369. Matoba, H., and P.D. Gollnick (1984) Influence of ionic composition, buffering agent, and pH on the histochemical demonstration of myofibrillar actomyosin ATPase. Histochemistry 80t609-614. McClearn, D. (1985) Anatomy of raccoon (Procyon lotor) and coati (Nasua narica) forearm and leg muscles:
Relations between fiber length, moment-arm length, and joint-angle excursion. J. Morphol. 183337-115. Ruckebusch, Y. (1972) The relevance of drowsiness in the circadian cycle of farm animals. Anim. Behav. 20:637643. Sack, W.O. (1989) The stay-apparatus of the horse’s hindlimb-explained. Equine Practice 11:31-35. Sack, W.O. (1991) Rooney’s Guide to the Dissection of the Horse. Ithaca: Veterinary Textbooks. Stashak, T.S. (1987) Adam’s Lameness in Horses. Philadelphia: Lea and Febiger. Webbon, P.M. (1977) A post mortem study of equine digital flexor tendons. Equine Vet. J. 9r61-67. Ypey, D.L. (1978) A topographical study of the distribution of end-plates in the cutaneus pedoris, sartorius, and gastrocnemius muscles of the frog. J. Morphol. 155:327-348.
