IDENTIFICATION OF ELASTIC FIBRES IN THE PERIPHERAL NERVE

P. L. TASSLER, A. L. DELLON, and C. CANOUN

From Johns Hopkins University School of Medicine, Baltimore, Maryland, and Oregon Health Sciences University, Portland, Oregon, USA

Traditional histological staining techniques, as well as elastin-specific antibodies and electron microscopy, have been used to assess the distribution of elastin within the peripheral nerve. The location of the elastin identified by the VerHoeff-VanGieseu or Weigert stains has been shown to coincide with the unambiguous identilication of elastin by immunospecific stains and electron microscopy. Elastin is located in all three connective layers of the peripheral nerve. Thick elastic fibres, consisting of amorphous elastiu protein and microfibrils, are located consistently in the periueurium and, to a lesser extent, in the epineurium. The endoneurium contains small collections of elastic fibres widely distributed between the axons. Compared with collagen, the overall content of elastin, however, is small, suggesting that the visco-elastic properties of peripheral nerve may be due primarily to collagen. Journal of Hand Surgery (British and European Volume, 1994) 19B: 48-54

The peripheral nerve must be constructed to respond to the range of motion of the joints it crosses, and to have a reserve resiliency sufficient to withstand the unusual traction demands imposed by injury. Wall et al (1991) have used stress relaxation properties to show that nerves are highly visco-elastic. Sunderland (1978) has suggested that these elastic properties of the peripheral nerve are due to its connective tissue supporting elements, the epineurium, perineurium and endoneur- ium. Previous researchers have focussed upon the relationship between stretch and loss of peripheral nerve function, defined electrophysiologically (Mitchell, 1872; Liu et al, 1948; Denny-Brown and Doherty, 1945; Sunderland and Bradley, 1961; Leffert and Seddon, 1965), in terms of neural regeneration (Hoen and Bracket& 1956; Haftek, 1970), blood flow (Lundborg and Rydevik, 1973; Clark et al, 1992), or stress/strain function (Rydevik et al, 1990; Wall et al, 1991). It has been assumed that the elastic properties of the peripheral nerve are due to the elastic fibres within the nerve.

An alternative explanation for the elasticity of the peripheral nerve is the visco-elastic property of the collagen molecule, whose presence has been well- demonstrated within the peripheral nerve (Kucharz, 1992). Distinguishing the presence of elastin from col- lagen is theoretically difficult because both collagen and elastin cross-stain with the traditional histochemical techniques, such as Weigert and the Verhoeff-VanGiesen (Sheehan and Hrapchak, 1980). Theoretically, with both of these stains, which use acidic iron/iodine and fuchsin reagents, elastic fibres stain blue-black and collagen fibres appear pink-red. Morphological identification of the elastic fibre by electron microscopy has distinguished a central core of amorphous elastin which stains with phosphotungstic acid, and microfibrils which stain with both uranyl acetate and lead (Greenlee et al, 1966). The position of the microfibrils varies with respect to the amorphous protein, attaining a progressively more cen-

tral location as the elastic fibre matures in its extracellu- lar environment (Ross and Bornstein, 1971; Ross et al, 1977; Cotta-Pereira et al, 1977).

In the present study, we have used immunospecific histochemistry and electron microscopy to localize elastic fibres within the peripheral nerve and we have compared these observations with the traditional histological assessment.

MATERIALS AND METHODS

Normal human sural nerves, harvested during nerve grafting procedures, have been fixed in formaldehyde, embedded in paraffin and then stained with the classic haematoxylin and eosin, Masson trichrome, Verhoeff- VanGiesen, Weigert (resorcin-fuschin) and silver staining techniques (Sheehan and Hrapchak, 1980). Both cross- sections and longitudinal sections have been evaluated, in specimens taken from three patients. The location of the collagen, as stained with the trichrome stain, corre- lated with the apparent location of the elastic fibres demonstrated by the Verhoeff-VanGiesen and Weigert stains. The location of the axons was confirmed- with the silver stain.

Normal Sprague-Dawley adult female rat sciatic nerve was fixed in glutaraldehyde, post-fixed in phosphate buffer, embedded in Epon and ultrathin sections were cut, stained with uranyl acetate and evaluated on a JEOL 1200 EXII electron microscope. Sections were examined in the epineurium, perineurium and endoneur- ium for appearances that coincided with previously identified elastic fibres in the rabbit ligamentum flavum (Serafini-Fracassini et al, 1977) bovine ligamentum nuchae and rat flexor digital tendon (Greenlee et al, 1966), rat aorta (Cliff, 1971), human dermis (Ross, 1973), guinea pig trachea and mouse testes (Brissie et al, 1975).

The immunohistochemistry was performed using

48

ELASTIC FIBRES IN PERIPHERAL NERVE 19

a

Fig 1 Low power view of transverse (a) and longitudinal (b) sections of the human sural nerve with Verhoeff-VanGesen elastic fibre stain (x 40). Note dark bands and thinner strands of elastic fibres in the epineurium and perineurium. The outlined area is enlarged for Figure 2.

fresh human nerve specimens. Representative sections were chosen containing perineural vessels which served as internal controls. These tissues were fixed overnight in Carnoy’s fixative (IQ ml glacial acetic acid, 60 ml absolute ethanol, 30 ml chloroform), dehydrated in a graded series of alcohols, cleared in xylene, and infiltrated and embedded in paraffin. Serial sections of the blocks were cut at 5 urn thickness and mounted on glass slides. The sections were deparaffinized and trypsin degradation of the specimens was performed by incubating the slides in 0.1% trypsin for 10 minutes. The sections were then washed with cold phosphate buffered saline (PBS).

Monoclonal elastin antibody (Sigma Chemical, St. Louis) was diluted 1 : 500 with PBS and applied directly to the sections. PBS alone was applied to the control sections. Sections were incubated at room temperature overnight. Detection of the elastin antibody was per- formed using secondary antibody conjugated to alkaline phosphatase (Zymed Labs, South San Francisco, USA). The presence of enzyme was revealed by addition of a substrate-chromogen solution. Positive immunostaining

for elastin was localized as a red deposit in the Zymed detection system (Elias et al, 1989).

RESULTS

Traditional histochemistry

The Masson trichrome stain, which stains collagen bright blue, demonstrated the recognized pattern of collagen: thick and thin bands throughout the epi- neurium, especially dense circular bands in the transverse sections of the perineurium, and fine blue fibres through- out the endoneurium. Both the Verhoeff-VanGiesen and the Weigert stains yielded similar pictures in transverse and longitudinal sections. These “elastin” stains demonstrated thin and thick elastic fibres within the epineurium, intermixed with the collagen. In the peri- neurium, the elastic fibres appeared most prominent as a circular band in the transverse section, but were also present to a lesser degree longitudinally over short distances. At 1,000 x magnification, the elastic fibres appeared as a thin ladder or lattice over the longitudinal sections of the axons in the endoneurium. In all three

THE JOURNAL OF HAND SURGERY VOL. 19B No. 1 FEBRUARY 1994

Fig 2 The epineurium, perineurium and endoneurium may be compared in these transverse (a,c,e) and longitudinal (b,d,f) views enlarged from Figure 1 with serial sections stained sequentially with Masson trichrome (a,b: collagen is blue), Verhoeff-VanGiesen (c,d: collagen is red, elastin is blue-black), and Weigert (e,f: collagen is pink-red, elastin is black) (x 160).

connective tissue regions, the elastic fibres were sparse compared with the collagen fibres (Figs 1 and 2).

Immunohistochemistry

the internal elastic lamina. The perineurium contained the most intense pink-red stain, which was uniform in intensity with the exception of dense red, circular, incom- plete fibres in transverse section. The endoneurium had

The anti-elastin antibodies demonstrated a dense, wavy a diffuse pink stain, midway in intensity between the red band inside the artery and vein, corresponding to perineurium and the epineurium. The only dark red

XASTIC FIBRES IN PERIPHERAL NERVE

Fig 3 Immunohistochemistry. The anti-elastin antibody appears red to pink (a,c) compared with the controi sectrons of the rat sciatic nerve (b,d). The lower power v-iew (a,c) demonstrates the wide distribution of the pink colour throughout the peripheral nerve. At higher power, note the relative density of colour in the perineurium (c). The brightest red is within the internal elastic lamina of the arteries and veins.

within the endoneurium occurred within the endoneurial microvessels. The epineurium had only a few dense red fibres and the least diffuse pink staining (Fig 3).

Electron microscopy

Elastic fibres were present in the epineurium, endoneur- ium and the perineurium. In all three regions collagen flbres were present in much greater numbers than elastic fibres. There was no recurring relationship between the elastic fibres and the axons, but elastic fibres appeared within or adjacent to the fields of collagen fibres. Elastic hbres were widely distributed throughout the endoneur- ium and were thinner than those found in the epineurium and perineurium. The location of elastic fibres in the perineurium was more consistent, with thicker bands of elastin being found almost exclusively on the innermost layer of the perineurium. Smaller bundles of elastic fibres were scattered to a lesser extent through the other layers of the perineurium. The epineurium, with its abundance of collagen, contained both thick and thin elastic fibres, always located within a field of collagen

fibres. The perineurium clearly contained the most elastin of the three connective tissue layers (Fig 4).

DISCUSSION

The results of this study demonstrate for the first time the presence and location of elastic hbres within the peripheral nerve in the human sural nerve and the rat sciatic nerve. One report by Fullmer ( 19%) described the presence of elastic fibres associated with nerves in human periodontal membranes, but this made no attempt to evaluate the presence of elastin within the terminal branch of the alveolar nerve. Despite the poss- ible cross-reactivity of traditional haematoxylin-iron (Verhoeff-VanGiesen) and orcein-fuchsin ( Weigert) between collagen and elastin, this study has demon- strated a good correlation between the location of elastic fibres identified by these traditional stains, and by immu- nospecific and electron microscopic analysis. The immunospecific and electron microscopic views have demonstrated, however, that the elastic fibres are less abundant than would be suggested by histochemistry,

THE JOURNAL OF HAND SURGERY VOL. 19B No. 1 FEBRUARY 1994

a/result explained by the cross-reactivity of the traditional stains.

The technique of electron microscopy employed in the present study could be used to quantitate the percent- age of the cross-sectional area of the nerve fascicle that consists of elastin fibres. The technique of immunohisto- chemistry lends itself more to a qualitative than quanti- tative description.

The spiral bands that appear to encircle the nerve fascicle, observed in 1779 by Fontana (1781), have been shown to be inherent properties of the nerve fibres, not of the perineurium (Zachary et al, 1993). Removal of the perineurium does not cause this pattern to disappear, whereas disruption of the endoneurium will result in a loss of the pattern. The present study suggests that the endoneurial elastic fibres may provide sufficient force to produce the wave-like or coiled, unstretched position of the individual axons within the fascicle.

The visco-elastic properties of the peripheral nerve have been studied extensively and the studies have been summarized by Sunderland (1978), and more recently by Rydevik et al (1990). Taken together, these studies

demonstrate that peripheral nerve may stretch by about 10% of its resting length without sequelae, up to 20% with loss of function, and rupture occurs beyond 20%. This 20% value has been confirmed recently for rat sciatic nerves following nerve repair (Butler et al, 1993). There are conflicting reports regarding which component of the nerve is responsible for elasticity. The results of the present study do confirm that the greatest concen- tration of elastic fibres is within the perineurium; how- ever, the stress-strain curve for the peripheral nerve more closely resembles that for collagen than it does for elastin beyond 20% strain (Liu, et al, 1948; Serafini- Fracassini et al, 1977; Rydevik et al, 1990). The results of the present study demonstrate that a relatively small percentage of the nerve contains elastic fibres compared with collagen. It is suggested, therefore, that the visco- elastic properties of the peripheral nerve are more likely to be due to elastin in the first phase of the curve (strain less than 20%) and collagen thereafter.

Acknowledgement The authors thank the Bowles Fund for financial support for this study

ELASTIC FIBRES IN PERIPHERAL NERVE

d

Fig 4 Electron microscopy. Low power view (a) demonstrating all three connective tissue layers of the peripheral nerve (“Ep” =epineurium, “P” =perineurium, “En” =endoneurium) (x 5,000). At higher power, a thin elastic fibre (arrowhead) is accompanied by collagen (labelled “C’) in the endoneurium (b) (x 20,000). While thin elastic fibres are distributed throughout the perineurium, a thick elastic fibre (arrowheads) lies characteristically along the innermost layer (c) (x 20,000). Elastic fibres in the epineurium (arrowhead) are always surrounded by collagen (“C” = collagen in the endoneurium) (d) (x 20,000).

References BRISSIE, R. M., SPICER, S. S., and THOMPSON, N. T. (1975). The variable

fine structure of elastin visualized with Verhoeff’s iron hematoxylin. Anatomy Research, 181: 83-94.

BUTLER, J. M., BODINE-FOWLER, S. C., ABRAM& R. A.. DULBECCO, F. L.. and BOTTE, M. J. (1993). Biomechanical properties of the rat sciatic nerve after epineurial repair without tension. Transactions of the Orthopedic Research Society, 18: 383.

CLARK, W. L.. TRUMBLE, T. E.; SWIONTKOWSKI, M. F., and TENCER, A. F. (1992). Nerve tension and blood flow in a rat model of immediate and delayed repairs. Journal of Hand Surgery, 17A: 4: 677-687.

CLIFF, W. J. (1971). The ultrastructure of aortic elastica as revealed by pro- longed treatment with 0~04. Experimental Molecular Pathology, 15: 220-229.

COTTA-PEREIRA, G., RODRIGO, F. G., and DAVID-FERREIRA, J. F. The Elastic System Fibers. In: Sandberg, L. B., Gray, W. R. and Franzblau, C. (Eds.): Elm-tin and Elastic Tissue. New York, Plenum Press, 1977: 19-30.

DENNY-BROWN. D. and DOHERTY, M. M. (1945). Effects of transient stretching of peripheral nerve. Archives of Neurology and Psychiatry, 54: 116-129.

ELIAS, J. M., MARGIOTTA, M., and GABORC, D. (1989). Sensitivity and detection efficiency of the peroxidase antiperoxidase (PAP), avidin-biotin peroxidase complex (ABC), and peroxidase-labeled avidin-biotin (LAB) methods. American Journal of Clinical Pathology, 92: 62-67.

FONTANA, F. Waite SW lr Vemin de ia Vcpepere SW ks Rmons Americaim 2 volumes. Florence, 178 1.

FULLMER, H. M. (195X). Differential stainmg of the connective tissue fibers in areas of stress. Science, 127: 1240.

GREENLEE, T. K., ROSS, R.: and HARTMAN, J. L. (1966). The fine struc- ture of elastic fibers. Journal of Cell Biology, 30: 59-71.

HAFTEK, J. (1970). Stretch injury of peripheral nerve: Acute effects of stretch- ing a rabbit nerve. Journal of Bone and Joint Surgery. 52B: 2: 354-365.

HOEN, T. I. and BRACKETT; C. E. (1956). Peripheral nerve lengthening. Journal of Neurosurgery, 13: 43.

KUCHARZ, E. J. The Collagens: Biochemistry and Pathophysiology. Berlin: Springer-Verlag, 1992.

LEFFERT, R. D. and SEDDON. H. (1965). Infraclavicular brachial plexus injuries. Journal of Bone and Joint Surgery, 47B: 1: 19-22.

LIU; C. T., BENDA, C. E., and LEWEY, F. H. (1948). Tensile strength of human nerves: An experimental physical and histologic study. Archives of Neurology and Psychiatry, 59: 322-336.

LUNDBORG, G. and RYDEVIK B. (1973). Effects of stretching the tibia1 nerve of the rabbit: A preliminary study of the intraneural circulation and the barrier function of the perineurium. Journal of Bone and Joint Surgery, 55B: 2: 390%401.

MITCHELL, S. W. Injuries of Nerves and Their Conseqttences. Philadelphia, Lippincott, 1872.

ROSS, R. and BORNSTEIN, P. (1971). Elastic fibers in the body. Scientific American, 224: 44.

54

ROSS, R. (1973). The elastic fiber: A review. Journal of Histochemistry and Cvtochemistrv. 21: 199-208.

THE JOURNAL OF HAND SURGERY VOL. i9B &o. I FEBRUARY 1994

ROSS;R., FIALKOW, P. J. and ALTMAN, L. K. The Morphogenesis ofElastic Fibers. In Sandberg, L. B., Gray, W. R. and Franzblau, C. (Eds): Elastin and Elastic Tissue. New York, Plenum Press, 1977: 7-17.

RYDEVIK, B. L., KWAN, M. K., MYERS, R. R., BROWN, R. A., TRIGGS, K. .I.. WOO. S. L-Y., and GARFIN, S. R. (1990). An in vitro mechanical and histological study of acule stretching on rabbit tibia1 nerve. Journal of Orthopaedic Research, 8: 694-701.

SERAFINI-FRACASSINI, A., FIELD, J. M.; SMITH, J. W.. and STEPHENS, W. G. S. The Ultrastructure and Mechanics of Elastic Ligaments. In: Sandberg, L. B.. Gray, W. R. and Franzblau, C. (Eds): Elasr~ and Elastic Tissue. New York, Plenum Press, 1977: 97-103.

SHEEHAN, D. C. and HRAPCHAK, B. B. Theory and Practice of Hisiofechnology. 2nd Edn. Columbus, Battelle Press, 1980: 194-201.

SUNDERLAND: S. and BRADLEY, K. C. (1961). Stress-strain phenomenon in human peripheral nerve trunks. Brain. 84: 102.

SUNDERLAND, S. Nerves and ,Vmuz Iqtrirr, 2nd Edn. Edinburgh, Livingstone, 1978.

WALL, E. .I., KWAN, M. K.. RYDEVIK, B. L., WOO, S. L-Y., and GARFIN, S. R. (1991). Stress relaxation of a peripheral nerve. Journal of Hand Surgery, 16A: 5: 859-863.

ZACHARY, L. S., DELLON. E. S., NICHOLAS, E. M., and DELLON. A. L. (1993). The structural basis of Felice Fontana’s spiral bands and their relationship to nerve injury. Journa! of Reconstructive Microsurgery, 9: 2: 131-13x.

Accepted: 30 June 1993 4. Lee Dellon, MD. Suite 104, 3901 Greenspring Avenue, Baltimore, Maryland, 2121 I. USA.

a 1994 The Brrtish Society for Surgery of the Hand