Few reports have described the macroscopic anatomy of the forelimb nerves in rats (Greene, 1963; Hebel and Stromberg, 1986; Bertelli et al., 1992, 1995; Uzun et al., 2001). Microscopic evaluations providing quantitative data on the fascicles and nerve fibers are also scanty, with most information coming from control animals in experimental studies. The use of the brachial plexus nerves in experimental models of regeneration has been defended by Bertelli et al. (1995) and Bontioti et al. (2003). In rats, experimental injury to the forelimb nerves does not usually lead to contractures or autotomy, which frequently occurs when the sciatic nerve is injured. Both contractures and autotomy can interfere with the results of functional experiments. Since the distance to the target organs (muscles and skin) is short in the rat forelimb, reinnervation is quicker, and the time required for functional recovery is less than in hindlimbs (Bertelli and Mira, 1993; Bontioti et al., 2003). The complexity of the forelimb nerves and their branches offers various experimental opportunities, since selective injuries can be easily applied to motor or sensory branches compared to hind limb nerves (Bontioti et al., 2003). Further, besides locomotion, grooming and feeding movements can be evaluated (Carry et al., 1993). Most injuries to human peripheral nerves affect the upper extremities, which constitutes another reason why an experimental model of nerve injury in the forelimb is useful (Bontioti et al., 2003). The brachial plexus of the rat and its branching exhibit various similarities with the human brachial plexus (Bertelli et al., 1992, 1995).
In this study, we present morphoquantitative data on the main brachial plexus branches: ulnar, median, radial, and digital nerves. We also compare the proximal and distal segments of the same nerves in animals 4 or 7 weeks old.
MATERIALS AND METHODS
Twelve female Wistar rats were divided into two groups: animals 4 weeks old, weighing 96–137 g (n = 6), and animals 7 weeks old, weighing 214–271 g (n = 6). The animals were anesthetized with sodium pentobarbital (Nembutal, 40 mg/kg, i.p.) and were perfused through the left ventricle with a 0.05 M phosphate-buffered saline solution, pH 7.4 (1 ml/g), followed by a 3.0% glutaraldehyde solution in 0.07 M cacodylate buffer, pH 7.2 (1.5 ml/g), using a perfusion pump (Aga 2 Intel pump).
The right median nerves, from the highest level in the axilla, through their distal branching, the lateral proper digital nerve of the third finger, were carefully dissected without stretching and removed. The right ulnar nerves were dissected from their origin to the carpal articulation level, and the right radial nerves were also dissected. The median, ulnar, and radial nerve samples selected for this study were located at either the proximal level (axilla) or the distal level (forearm, distal 1/3). The lateral proper digital nerve of the third finger was studied as a whole. After removal, nerves were kept in the fixative solution for an additional 48 hr and processed for epoxy resin embedding (Poly/Bed 812; Polysciences, Warrington, PA). After careful positioning of the nerve fragments in the embedding molds, transverse sections were cut at 0.25 μm, stained with 1% toluidine blue, and observed using an Axiophot photomicroscope (Carl Zeiss, Jena, Germany).
The images were acquired via a digital camera (TK-1270; JVC, Victor Company of Japan, Tokyo, Japan) and were analyzed using an IBM PC. The transverse sectional areas of the fascicles were obtained manually, and the number of fascicles counted. The luminal area and number of capillaries were also obtained. The endoneurial area occupied by the capillaries was measured. In the study of the myelinated fibers, the endoneural space was observed using an optical set with oil immersion lens (100×), optovar (1.6×), and camera with lens of 0.5× and an 8× magnification, which provided images with excellent resolution. The images were fully scanned using an automatic motorized microscope stage (Carl Zeiss, Jena, Germany), without overlap of the microscopic fields. Thirty percent of such microscopic fields were randomly studied in proximal segments of the median, ulnar, and radial nerves, and in distal segments of the median and ulnar nerves. All microscopic fields of the distal segment of the radial nerve and of the lateral proper digital nerve of the third finger were analyzed. Morphometric parameters for the myelinated fibers of the nerve segments were obtained as described previously (Fazan et al., 1999; Jeronimo et al., 2005). Briefly, the total number of myelinated fibers present in each microscopic field was identified by visual inspection and counted. The numerical density of the myelinated fibers was calculated. The minimal diameter of the myelinated fibers was measured using an image analysis software (KS 400; Kontron 2.0, Eching Bei München, Germany). Only fibers of circular shape were measured. Both the axonal minimal diameter and the total fiber minimal diameter were measured, and the ratio between the two diameters, the g-ratio (a measure of the degree of myelination), was obtained (Rushton, 1951; Smith and Koles, 1970). The myelin sheath area was calculated for each myelinated fiber area measured. Histograms of the distribution of the myelinated fiber and axon (not shown) populations, separated into 1 μm diameter class intervals, were constructed. Vertical lines pointing down from the base line were included, from left to right, to show the first percentile of the range of diameter, the position of the median (line in bold), and the 99th percentile of the range of diameter. A regression analysis was generated to determine the relationship between the diameter of the axon and that of the myelinated fiber diameters (shown for the median nerves and lateral proper digital nerves of the third finger).
Morphometric data are presented as mean ± standard deviation (SD). Data were compared between the two groups (rats aged 4 or 7 weeks) using the Mann-Whitney nonparametric test, as well as between segments within the same group (proximal and distal segments) using the Wilcoxon nonparametric test for paired samples. Differences were considered significant at P ≤ 0.05.
The morphometric data obtained for all nerves studied and the comparisons between the segments (proximal or distal) and groups (4 or 7 weeks) are provided in Tables 1 and 2.
Table 1. Morphometric data for proximal and distal segments of the ulnar, median and radial nerves, and lateral proper digital nerve of the third finger in rats aged 4 weeks old
All lateral proper digital nerves of the third finger exhibited a single fascicle, as did most of the proximal segments of the ulnar, median, and radial nerves. The distal segments of the ulnar, median, and radial nerves showed 1–4, 3–9, and 1–4 nerve fascicles, respectively (Figs. 1 and 2).
Fascicular areas and the diameters of all nerves studied, like the number of myelinated fibers, were significantly less in the distal segments of both age groups, although the difference in number of fibers in the median nerve did not reach statistical significance in the 4-week-old animals. There was a significant difference in radial nerve myelinated fiber density between the two segments (distal > proximal) for both age groups. In the 4-week-old rats, the radial nerves exhibited larger myelinated fiber and axon diameters in the proximal segments, while the median nerve showed a significant difference in myelin area (proximal > distal). In the 7-week-old rats, axon diameter was greater in the proximal segments of the median and radial nerves, but not in the ulnar nerve. Fiber diameter and myelin area were significantly larger in the proximal segments of the radial nerves but not in the other nerves studied in this group. The area and the number of capillaries were greater in the proximal segments of both age groups, except for the ulnar nerve in the 7-week-old animals.
Unlike the other nerves examined, the distal segment of the radial nerve showed no significant differences between age groups with respect to nerve area. The nerve diameters tended to be larger in the 7-week-old animals, reaching statistical significance only in the proximal segment of the radial nerve. The number and density of myelinated fibers were similar between the 4- and 7-week-old animals, except for the proximal segment of the radial nerve, which showed a statistical difference in myelinated fiber density. In the proximal segments, the fiber diameters (Fig. 3), the axon diameters (Tables 1 and 2), and the myelin area (Tables 1 and 2) were greater in the 7-week-old animals for all nerves studied. In the distal segments of this group, the myelinated fiber diameters and the myelin area were significantly greater only in the median nerve, while axon diameter was significantly greater in the median, ulnar, and lateral proper digital nerve of the third finger. The number and area of the capillaries were similar in both age groups, although the distal segment of the ulnar nerve exhibited a greater number of capillaries in the 7-week-old animals. The endoneurial area occupied by the capillaries in the proximal and distal segments did not exceed 1.9% and 0.32%, respectively.
Distributional histograms of the myelinated fiber diameters (Fig. 3) and their respective axons were similar between the proximal and distal segments of the ulnar and median nerves in the two age groups. The proximal segment of the radial nerve exhibited fibers (Fig. 3) and axons of larger diameter than the distal segment for both age groups. Regression analysis obtained by the least-squares method (Fig. 4) showed the relationship between the axon and myelinated fiber diameter. The regressions were obtained by the least-squares method. The minor regression coefficient was 0.81 for the distal segment of the radial nerve in the 4-week-old rats. All the others studied nerves led to regression coefficients larger that 9.0.
Quantitative microscopic studies of the rat brachial plexus are limited. In general, histometry of nerve components is presented as control data for experimental neuropathies (Table 3). As it can be seen in Table 3, the data are limited to measurements according to the specific interests of the authors. In Table 3, we can also see that no study presented systematized morphometric data for the principal branches of the rat brachial plexus. The differences between the approaches and data obtained in this study when compared with others reports in the literature can be found in the rest of the article. Irrespective of the intrinsic value of the present morphometry quantitation, the specific data for the digital branch of the median nerve are important because rats are frequently used to study experimental dying back polyneuropathies. The initial lesions in such neuropathies involve the more distally located nerves and nerve fibers.
Table 3. Summary of experimental studies that have provided quantitative data on the forelimb nerves in rats
Myelin sheet area, area of myelinated fibers and axons, percentage of endoneurial space, axon/ myelin ratio
In this study, the nerve area and number of myelinated fibers are greatest in the proximal segments as a consequence of progressive nerve branching along its trajectory in a distal direction. The human sural nerve area diminishes from 4% to 30% distally due to nerve branching (Behse et al., 1974). The myelinated fiber density was not significantly different between the segments, with the exception of the proximal segment of the radial nerve compared to the distal segment for both age groups: the diameter was greater in the distal segments. Such findings are similar to those obtained by Hill et al. (1977). Myelinated fibers and axons of larger caliber are found in the proximal segments of the radial nerve. This may be explained by the fact that the distal portion of the radial nerve is constituted only by sensory fibers: thinner compared to motor fibers (Tomasch and Britton, 1956; Stevens et al., 1973).
The following increases in nerve parameters accompanying age are corroberated by our findings: rat nerve area (Hashizume and Kanda, 1995; Jeronimo et al., 2005), myelinated fiber diameter in rats and man (Gutrecht and Dyck, 1970; Jacobs and Love, 1985; Ouvrier et al., 1987; Schellens et al., 1993; Fazan and Lachat, 1997; Jeronimo et al., 2005), axonal diameter in rats (Hashizume and Kanda, 1995), and myelin sheath area in man (Thomas and Ochoa, 1984; Jacobs and Love, 1985; Schröder et al., 1988). The present study is the first to document such differences over the short interval of 3 weeks: between the ages of 4 and 7 weeks, which is indicative of the intense maturation of the peripheral nervous system during this life period in the rat. The similarity in number of myelinated fibers between different aged members of the same species is well documented (Carry et al., 1993; Schellens et al., 1993; Hashizume and Kanda, 1995; Jeronimo et al., 2005). In man (Stevens et al., 1973; Jacobs and Love, 1985; Ouvrier et al., 1987; Schellens et al., 1993), mice (Carry et al., 1993), and rats (Jeronimo et al., 2005), myelinated fiber density diminishes with age. This reduction is related to the increase in nonneural components of the endoneurium, particularly collagen fibers. The similarity in myelinated fiber density in both age groups in our study indicates that significant quantities of nonneural components are not added to the endoneurium between the ages of 4 and 7 weeks.
Measurements of the proximal nerve segment diameters, excluding the perineurium, were greater for the radial nerves, followed by the median and ulnar nerves. The same relative proportions in diameter measurements for the same nerves have been seen in rats (Uzun et al., 2001). Other studies (Bertelli and Mira, 1995; Bertelli et al., 1995) reported similar diameter measurements for the ulnar and radial nerves. However, such measurements were less than those for the median nerve diameter. In absolute terms, the data on nerve diameter obtained by these authors were greater than our measurements because they included the epineurium and the perineurium. In the distal segments, the diameter of the ulnar nerve was greater compared to the median nerve, and between this latter and the radial nerve. At this level, the radial nerve is constituted exclusively by sensory fibers. This difference is related to the important contingent of motor fibers that innervate the hand muscles in the ulnar and the median nerves.
The maximum endoneurial area occupied in proximal segments by the capillaries (1.9%) can be ignored in the evaluation of the myelinated fiber densities in the nerves we studied. This was more evident in the distal segments, where the endoneurial area occupied by capillaries was only 0.32%. The estimation of the number and area of endoneural capillaries in each segment is important for experimental models of nerve diseases that involve blood vessels, such as diabetic neuropathy. Our findings show that the number and area of the capillaries are greater in the proximal segments of all nerves studied.
The distributions of the fiber and axon diameters were unimodal in all nerves. Nevertheless, the distributions in the proximal segments of the radial nerves, the distal segments of the ulnar and median nerves, and the lateral proper digital nerve of the third finger tended to be bimodal in animals 7 weeks old. This observation agrees with that of Jeronimo et al. (2005), who showed, for rat sural nerves, unimodal distribution in young animals (30 days), a tendency for bimodal distribution in some segments of animals 3 months old, and a well-established bimodal distribution in animals 6 months old. Further studies on the brachial plexus nerves are needed to establish fiber and axon diameter distributions for adult rats.
The relationship between the axon and the myelin sheath has been generally observed in normal nerve fibers (Gutrecht and Dyck, 1970; Schröder, 1972; Sananpanich et al., 2002). In this study, no difference in the relationship was recognized with increasing age or between the segments. Most of the myelinated fibers in all the nerves studied here lead to a g-ratio of between 0.6 and 0.7. This ratio is useful in evaluating nerve disease (Gutrecht and Dyck, 1970) and is relevant to theories of saltatory conduction (Williams and Wendell-Smith, 1971; Thomas and Ochoa, 1984). Theoretical considerations suggest that a g-ratio close to 0.6 is optimal for the spread of current from one node to the next (Rushton, 1951). This finding suggests that, despite their young age, in the animals studied, most fibers have reached the best g-ratio value (0.6) for maximal conduction velocity (Smith and Koles, 1970).
Animals taken immediately after the suckling period and young adults were used in this study. In general, this is the age at which some experimental neuropathies are investigated in rats, e.g., experimental allergic neuritis. Investigators in this field will find in the present study a systematic reference collection of normative morphometric data for the major branches of the rat brachial plexus, which should serve as referential data for such experimental studies.
The authors thank Ms. Maria Cristina Lopes Schiavoni, Mr. Antonio Renato Meirelles e Silva, and Ms. Aracy Edwirges Vieira da Silva Dias for technical assistance, as well as Mr. Geraldo Cássio dos Reis for statistical analysis.