Fibers in Monkey Posterior Articular Nerves ERNEST GARDNER AND NICHOLAS J. LENN Departments of Neurology, Orthopaedic Surgery, and Human Anatomy, and the Carnegie Laboratories of Embryology, University of California, Daois, California, 9561 6

ABSTRACT The distribution of axons in the posterior articular nerves in rhesus monkey has been studied in terms of ratio of myelinated to unmyelinated fibers, and size of myelinated fibers. Conduction velocity measurements were also made. The nerves contained 2,000 to 2,200 unmyelinated fibers of 0.2 to 1.2 pm diameter. They contained 399 to 478 myelinated fibers, varying in size from 1.4 to 12.3 pm. Thus 80 to 85% of fibers in these nerves were unmyelinated. The maximum conduc- tion velocity of 70 to 80 meters per second corresponded to an initial small de- flection, with most fibers conducting at slower rates. These results are compared to previous reports which severely underestimated the number of unmyelinated fibers because electron microscope counts were not utilized.

It is suggested that pain fibers from the knee joint of monkey make up much of this large population of unmyelinated axons. An unknown number of post-gangli- onic sympathetic fibers is also included in the unmyelinated fiber group.

Joint receptors usually comprise three main types: (1) Ruffini endings in the joint capsule, and Golgi tendon endings in the ligaments (these are sometimes collec- tively termed spray endings and both are derived from myelinated fibers). (2) Lamellated corpuscles (usually pacini- form), derived from myelinated fibers. (3) Free endings, derived from unmyelinated axons and probably also from small myelin- ated fibers. The distribution and morphol- ogy of joint receptors have been studied in several species, chiefly in cats (for re- ferences, see Gardner, '50; Skoglund, '731, and in monkey and human joints (Gard- ner, '56; Stilwell, '57a,b; Frankova, '68; Palmieri and Veggetti, '71).

There have been many clinical and ex- perimental studies of the sensory and reflex functions of joint receptors, especially those concerned with pain. The role of joint receptors in kinesthesis has also re- ceived considerable attention (Skoglund, '731, although their importance in this function is questioned (Clark, '75; Clark and Burgess, '75; Grigg, '75; Horch et al., '75). Most experimental studies of joint receptors have been carried out in cats, and a number of studies have included

ANAT. REC., 187: 99-106.

analyses of fiber diameter in articular nerves correlated with type and distribu- tion of receptor (Gardner, '44; Skoglund, '56; Freeman and Wyke, '67; Burgess and Clark, '69a,b). Relatively few experimental studies have been carried out in monkeys; among them are the classical papers of Mountcastle and Powell ('59) and Mount- castle et al. ('631, who studied neurons in the thalamus and postcentral gyrus of mon- keys which responded to joint movement or discharged during the maintenance of joints at various angles. To the best of our knowledge, in neither monkeys nor humans have articular nerves been studied with re- spect to fiber size and frequency and cor- relation with receptor. Finally, no study of articular nerves has yet employed the elec- tron microscope to determine the number of unmyelinated axons.

This paper reports gross, light micro- scopic, and ultrastructural observations on the posterior nerve of the monkey knee joint. It also includes data from hitherto unpublished experimental studies.

MATERIALS AND METHODS In the course of another study in which

Received May 5, '76. Acccptcd July 20, '76.

99

100 ERNEST GARDNEH AND NICHOLAS J. LENN

an adult rhesus monkey (Mucaca mulutta) was perfused with 10% buffered formalin, the posterior nerves to the knee joints were removed bilaterally. The dissected nerves were stored in fixative for one week, washed, post-fixed in 2% OsO,, block stained with uranyl acetate and em- bedded in Durcupan. One half pm thick sections were mounted on glass slides and stained with toluidine blue. Thin sections cut adjacent to the semi-thick sections were collected either on coated large slot grids or on 300-mesh grids. Examination of the former allowed visualization of the en- tire cross section of the nerves. On the 300- mesh grids a survey photograph at 100 x was taken with the electron microscope. Photomicrographs at 12,500 original mag- nification were taken of all unmyelinated axons in each grid window covered by a single section of each nerve. The unmyelin- ated axons were counted by two observers in the resulting photographs mounted as montages where overlapping occurred. In order to estimate the percentage of the nerve cross sectional area visualized in the grid windows, the 100 x photographs were cut into pieces containing the whole nerve, and then the portion of the nerve visible in the grid squares, and the pieces weighed. The ratio of the weights of grid windows to whole nerve was calculated. Myelinated fiber counts were made on this sample, using the 100 x photographs. A similar method has been shown to be reli- able (Coggeshall et al., '74).

Total counts of myelinated fibers were made from photomicrographs (printed at 560 x 1 of the one-half pm thick sections, using the microscope to settle questions of identity. The magnification was checked with stage and ocular micrometers. Prints were also used to construct frequency of occurrence histograms of myelinated fiber diameter, The small size of the nerves made it feasible to measure the diameter of every fiber. In the case of oval fibers, which were assumed to have been cut obliquely, the shortest distance was con- sidered to be the true diameter (Dyck, '751.

The present paper also contains some data from experiments on three monkeys fA4maca mulatta) carried out in 1956.1 The monkeys were anesthetized with sodium pentobarbital, and the posterior nerves of the knee joint were exposed, cut distally, and laid on silver electrodes. Conduction rates of the fastest conducting fibers were determined by stimulating fist the sciatic nerve and then dorsal roots, and recording the antidromically conducted impulses. Square wave stimulation through bipolar silver electrodes was used. Antidromically conducted impulses were also recorded during stimulation of lumbar and cervical levels of the posterior funiculi of the spinal cord; bipolar surface stimulating elec- trodes were used. Finally, evoked poten- tials were recorded from the surface of the opposite postcentral gyrus during electri- cal stimulation of the posterior nerve and the tibia1 nerve, A silver ball electrode was used for the surface recording, the in- different electrode being placed on adja- cent scalp. Throughout the several studies, observations were made on the origin, course, and approximate distribution of the posterior and medial articular nerves. Finally, a number of joint nerves were re- moved, osmicated, and sectioned in paraffin at 10 pm. Four posterior nerves were still available and the myelinated fibers in each were counted and the diam- eters of the largest fibers measured.

RESULTS

Gross distribution In all instances, the posterior nerves

were similar in origin, course, and gross distribution to the posterior nerve of the human knee joint as previously described (Gardner, '48). A medial nerve arising from the saphenous, and sometimes from the ob- turator also, as in cat and human, was like- wise present in the monkey. However, the distribution of other nerves to the monkey

1 These studies were carried out hy Dr. Cardner when he was Visiting Professor of Anatomy, University of California, Los Angeles, and a Public Health Service Special Research Fellow, National Institute of Arthritis and Metabolic Diseases, 1956.

FIBERS IN MONKEY POSTE

knee joint has not yet been determined. Numbers and sizes of fibers

Unm y el inated The identification of the unmyelinated

axons was generally unambiguous (figs. 1,2). The presence of a mesaxon was con- sidered a primary criterion, together with the size and the shape of the unmyelinated axons (Ochoa, '75). The axons varied in di- ameter from approximately 0.2 to 1.2 pm. Unmyelinated axons were distributed throughout the nerves. The sample size for each nerve obtained by the above method was 25%. That is to say, 25% of the total area of each nerve cross section was visualized in the windows of the 300 mesh grids. The right posterior nerve contained 500 unmyelinated fibers in the sample, giv- ing a total number of 2,000. The left poste- rior nerve contained 550 in the sample, giving a total of 2,200 unmyelnated axons in the nerve. Myelinated

In the electron micrographs, the my- elinated axons ranged from about 1 to 11 microns in average diameter (an occasional fiber was somewhat larger; see fig. 4). Counts of myelinated axons from the elec- tron micrographs revealed 135 axons in the sample for the right nerve, for a total of 540 myelinated fibers. The left nerve con- tained 100 myelinated axons in the sample for a total of 400 myelinated fibers in the whole nerve.

Figure 3 illustrates the left posterior nerve in one of the semithick sections, the nerve being about one-quarter mm in di- ameter. Total counts of myelinated axons in four different semithick sections of the left posterior nerve were: 399, 424, 432, and 449 fibers. There was evidence of branching of myelinated fibers in some of the sections. Hence more distal sections would have a higher count. In two differ- ent sections of the right posterior nerve, the counts were 469 and 478. A histogram of frequency of occurrence of fiber diam- eter for one of the nerves is shown in figure 4. The range of diameters was from 1.4 to

XIOR AKTICULAR KERVES 101

12.3 pm. A histogram w a also constructed for the same nerve using the average of the minimum and maximum diameters for each fiber. This gave a range of 2.0 to 14.0 pm.

Counts in the four 20-year-old osmi- cated nerves were consistently lower than the above: 310, 322, 336, 346. The largest fibers in these nerves were about 12 pm; an occasional fiber was 13 to 14 pm.

Conduction rates and central projections The fastest conducting fibers in joint

nerves reached the spinal cord at rates of about 70 to 80 meters per second. How- ever, these fibers were responsible for only an initial small deflection in the recorded response, indicating, as would be expected from the fiber size distribution, that most of the fibers conducted at much slower rates. The fastest fibers in the tibia1 nerve had about the same conduction rates. Similar records were obtained by stimulating pos- terior funiculi at lumbar levels. Stimulation of posterior funiculi at cervical levels likewise resulted in antidromically con- ducted impulses that were recorded from the posterior nerve. The action potential they formed was small and the overall con- duction rate was 50 to 60 meters per sec- ond.

When the posterior nerve was stimu- lated, small evoked potentials were re- corded from the contralateral postcentral gyrus (in the leg area shown in fig. 1, Gard- ner and Morin, '53). However, no sys- tematic exploration of buried cortex was made.

DISCUSSION

The counts of myelinated and unmyelin- ated axons obtained in these nerves indi- cate that approximately 80 to 85% of all axons were unmyelinated. There are data suggesting that this percentage varies from one system to another (Coggeshall et al., '74; Davenport and Ranson, '31; Duncan, '32). In a number of situations, there is a rather sharp separation between the size of unmyelinated and myelinated fibers. The critical diameter for myelination in the pe- ripheral nervous system has been shown to

102 ERNEST GARDNER AIVD NICHOLAS J. LENN

be about 1 pm (Matthews, '68). The current data are in agreement with this generaliza- tion. The percentage of unmyelinated axons in the joint nerves was, however, much larger than for instance, that found in rat ventral spinal roots where only up to 26% of the total axon population was un- myelinated (Coggeshall et al., '74). The percentage does correspond to that pres- ent in human sural nerve, where there are about 11,570 myelinated fibers and 53,000 unmyelinated axons (Dyck, '75).

As expected, the numbers of unmyelin- ated fibers in the posterior nerves are far greater than would have been supposed from studies in other species using silver stains. Gardner ('44) reported that in pos- terior nerves of the cat there were 115 nonmyelinated fibers. Freeman and Wyke ('67) found 162 unmyelinated fibers. Gard- ner and Jacobs ('48) reported that in dogs, posterior nerves averaged about 300 mye- linated and 600 unmyelinated fibers. It is likely, therefore, that all published figures of unmyelinated fibers in any nerve, based on counts in silver-stained material, are low.

The range of myelinated fiber diameter, because of considerable variation from one nerve to another (Skoglund, '56, '731, would be more accurate with pooling of data from a number of nerves (Burgess and Clark, '69b). However, the largest fibers in the posterior nerves were about 12 to 14 pm, and these conduct at rates of 70 to 80 meters per second. These rates are comparable to the fastest rates in human nerves. Moreover, the paucity of fibers of this size corresponds with the small amplitude of the initial fast component of the conducted action potential.

The number of myelinated fibers in mon- key posterior nerves was 399 to 540. Clark and Burgess ('75) reported a total of 1,340 myelinated fibers in five cat posterior artic- ular nerves studied with the electron mi- croscope, for an average of 268 fibers per nerve. This number is significantly higher than the number of 171 reported by Gard- ner ('441, 224 by Freeman and Wyke ('671, and 180 by Burgess and Clark ('69a,b); all

of these studies employed paraffin embed- ded nerves. The consistently lower counts in the osmicated nerves are likewise con- sidered unreliable. Evidently many of the smallest myelinated fibers are missed in paraffin sections.

Little is known about the central path- ways for articular nerves in monkeys. Gardner et al. ('53) reported that hind leg articular nerves in monkeys project to the cerebral cortex, and Mountcastle et al. ('631, and Mountcastle and Powell ('59) published extensive studies of thalamic and cortical responses to stimulation of joint receptors. Nevertheless, the routes to the thalamus and cortex remain virtually unstudied. Data reported in the present paper indicate that some impulses ascend by way of the posterior funicdi. The rate of conduction by this route appears to de- crease, a finding previously reported in cats (Gardner et al., '49; Clark, '721, where the decrease occurs as collaterals are given off as fibers ascend in the posterior funiculi. An important paper by Burgess and Clark ('69a) on projection of fibers from the cat knee joint by way of posterior funiculi, includes data from studies of four young adult squirrel monkeys. They found that only 11 to 16 fibers in each posterior articular nerve ascended in posterior funiculi to cervical levels. Hence, as in the cat, most articular fibers that enter the spinal cord from the lower limb synapse in gray matter, and most impulses destined for thalamus and cerebral cortex use path- ways other than the posterior funiculus- medial lemniscus system. Similar findings were reported by Whitsel et al. ('69, '70). Moreover, Vierck ('66) found that leg posi- tion sense in macaques is mediated by way of the ipsilateral fasciculus gracilis, and also by way of crossed and uncrossed path- ways in the lateral funiculi.

The presence of such large numbers of unmyelinated fibers is significant from the standpoint of joint sensation, especially pain. It is known from studies of cat knee joints (Samuel, '52) that some unmyelin- ated fibers in joint nerves are postgangli- onic sympathetic. (Ekholm and Skoglund,

FIBERS IN MONKEY POSTE

'63, claim that some small myelinated fibers are also postganglionic sympathetic.) However, the ratio of afferent unmyelin- ated to autonomic fibers is unknown. Hence, one can only surmise that a signifi- cant number of the unmyelinated fibers are afferent.

ACKNOWLEDGMENTS

The skillful technical assistance of Mrs. Viviana Wong and Mr. Guy Clark assisted this project. Supported by program-project Research Grant HD 08658, National In- stitute of Child Health and Human De- velopment, and by Research Grant NS 12265, National Institute of Neurological and Communicative Diseases and Stroke. The electron microscope facilities were those of the California Primate Research Center, supported by NIH Grant RR 00169.

LITERATURE CITED

Burgess, P. R., and F. J. Clark 1969a Dorsal column projection of fibres from the cat knee joint. J. Physiol., 203: 301-315.

Characteristics of knee joint recep- tors in the cat. J. Physiol., 203: 317-335.

Clark, F. J. 1972 Central projection of sensory fibers from the cat knee joint. J. Neurobiol., 3: 101-110.

1975 Information signaled by sensory fi- bers in medial articular nerve. J. Neurophysiol., 38:

Clark, F. J., and P. R. Burgess 1975 Slowly adapting receptors in cat knee joint: can they signal joint angle? J. Neurophysiol., 38: 1448-1463.

Cogeshall, R. E., J. D. Coulter and W. D. Willis, Jr. 1974 Unmyelinated axons in the ventral roots of the cat lumbosacral enlargement. J. Comp. Neur.,

Davenport, H. A,, and S. W. Ranson 1931 Ratios of cells to fibers and of myelinated to unmyelinated fibers in spinal nerve roots. Am. J. Anat., 49:

Duncan, D. 1932 A determination of the number of rinmyelinated fibers in the nerve roots of the rat and the rabbit. J. Comp. Neur., 55: 459-471.

Dyck, P. J. 1975 Pathologic alterations of the pe- ripheral nervous system of man. In: Peripheral Neuropathy. P. J Dyck, P. K. Thomas and E. H. Lamhert, eds. W. B. Saunders Company, Philadel- phia, pp. 296-336.

Ekholm, J., and S. Skoglund 1963 Autonomic con- tributions to myelinated fibers in peripheral nerves. Acta morph. neer1.-scand., 6: 55-63.

1969b

1464-1472.

153: 39-58.

193-207.

XIOR ARTICULAR NERVES 103

Frankova, H. 1968 Comparison of the occurrence and variability of joint receptors in rhesus monkey and man. Folia Morph. (Praha), 16: 83-92.

Freeman, M. A. H., and B. Wyke 1967 The innerva- tion of the knee joint. An anatomical and histologi- cal study in the cat. J. Anat., 101: 505-532.

Gardner, E. 1944 The distribution and termination of nerves in the knee joint of the cat. J. Comp. Neur.,

1948 The innervation of thc kncc joint. Anat. Rec., 101: 109-130.

1950 Physiology of movable joints. Physiol. Rev., 30:127-176.

Garher , E., and J. Jacobs 1938 Joint reflexes and regulation of respiration during exercise. Am. J. Physiol., 153: 567-579.

Gardner, E., F. Latimer and U. Stilwell 1949 Central connections for afferent fibers from the knee joint of the cat. Am. J. Physiol., 159: 195-198.

Gardner, E., and F. Morin 1953 Spinal pathways for projection of cutaneous and muscular afferents to the sensory and motor cortex of the monkey (Mucucu rnuluttul. Am. J. Physiol., 174: 149-154.

Gardner, E., F. Morin and J. Catalano 1953 Spinal pathways for cortical and cerebellar projections in the monkey Mucucu rnuluttu). Fed. Proc., 12: 48.

Grigg, P. 1975 Mechanical factors influencing re- sponse of joint afferent neurons from cat knee. J. Neurophysiol., 38: 1473- 1484.

1975 Awareness of knee joint angle under static condi- tions. J. Noiirophysiol., 38: 1436-1447.

Illatthews, M. A. 1968 An electron microscopic study of the relationship between axon diameter and the initiation of myelin production in the pe- ripheral nervous system. Anat. Rec., 161

Mountcastle, V. B., G. F. Poggio and G. Werner 1963 The relation of thalamic cell response to peripheral stimuli varied over an intensive continuum. J. Neu- rophysiol., 26: 807-834.

Mountcastle, V. B., and T. P. S. Powell 1959 Central nervous mechanisms subserving position sense and kinesthesis. Bull. Johns Hopk. Hosp., 105: 173-200.

Ochoa, J. 1975 Microscopic anatomy of unmyelin- ated nerve fibers. In: Peripheral Neuropathy. P. J. Dyck, P. K. Thomas and E. H. Lambert, eds. W. R. Saunders Company, Philadelphia, pp. 131-150.

Palmieri, G., and A. Veggetti 1971 Osservazioni pre- liminari sui recettori articolari di Cercopitheclcs uethiops. Bolletino della Societi italiana di biologia sperimentale, 47: 148-151.

Samuel, E. P. 1952 The autonomic and somatic in- nervation of the articular capsule. Anat. Rec., 113:

Skoglund, S. 1956 Anatomical and physiological studies of knee joint innervation in the cat. Acta physiol. scand., 36: Suppl. 124.

Joint Receptors and Kinaesthesis. In: Handbook of Sensory Physiology. Vol. 11, Somato- sensory System. A. Iggo, ed., Springer-Verlag,

80: 11-32.

Horch, K. W., F. J. Clark and P. R. Burgess

53-70.

1973

- I

Berlin; pp.' 111-136. I

Stilwell, D. L., Jr. 1957a The innervation of ten-

104 ERNEST GARDNER AND NICHOLAS J. LENN

dons and aponeuroses. Am. J. Anat., 100: 289-317. 195% The innervation of deep structures

Vierck, C. J. 1966 Spinal pathways mediating limb

Whitsel, B. L., L. M. Petrucelli and G. Sapiro 1969

Modality representation in the lumbar and cervical fasciculus gracilis of squirrel monkeys. Brain Res.,

Whitsel, B. L., L. M. Petrucelli, G. Sapiro and H. Ha Fiber sorting in the fasciculus gracilis of

of the foot. Am. J. Anat., 102: 59-73. 15: 67-78.

position sense. Anat. Rec., 154: 437. 1970 squirrel monkeys. Exp. Neurol., 29: 227-242.

PLATE 1

EXPLANATION OF FIGURES

1 Electron micrograph of unmyelinated axons from left posterior articular nerve. x 18,000.

Electron micrograph of unmyelinated axons from left posterior articular nerve. x 25,000.

Photomicrograph of semithick section of left posterior articular nerve. x 450.

2

3

FIBERS IN MONKEY POSTEWIOK ARTICULAR NERVES Ernest Gardner and Nicholas J. Lenn

PLATE 1

105

FIBERS IN MONKEY POSTERIOR AHTICITLAR NERVES Ern& Gardrier and Nichda J Lenn

> 0 z W 3 a W

LL a

80

60

40

20

0

PLATE 2

2 4 6 8 10 12 DIAMETER ( p m )

EXPLANATION OF FIGURE

4 Frequency of occurrence of myelinated fiber diameters in left posterior nerve, based on the shortest diameter. The size range was 1.4 to 12.3pm. The total number of myelinated fibers was 424.

106