No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Functional recovery following alterations in nerve-muscle connections of fishes
... XWhen peripheral motor nerves are surgically crossed or are cut and allowed to regenerate, the extent to which normal co-ordinated movement is restored is variable. The recovery of apparently normal function following nerve transaction in fish (Sperry, 1950;Sperry & Deupree, 1956;Sperry & Arora, 1965;Mark, 1965) and following nerve crosses in urodeles (Grimm, 1971;Cass & Mark, 1975) contrasts with the situation in mature anuran tadpoles (Sperry, 1947) and mammals (Sperry, 1941), which fail to re-establish co-ordinated movements. Of special interest are the findings on goldfish recovering from various experimental procedures in which extraocular muscles became reinnervated by foreign nerves. ...
... Selective reinnervation? Earlier behavioural studies in lower vertebrates indicated that when the original nerve to a given muscle was replaced by a foreign nerve, the latter could make functional connexions, consequently producing abnormal movements (Sperry & Deupree, 1956;Sperry & Arora, 1965;Mark, 1965;but see Sperry, 1950). Thus, the finding that in goldfish a foreign nerve (NIII) readily innervated the s.o. ...
... muscle when surgically crossed to it (Dual-Stag group) was expected. On the other hand, the same behavioural studies suggested that in these animals mixed motor nerves regenerating after simple transaction tended to reinnervate selectively their original muscles (Sperry, 1950;Sperry & Deupree, 1956;Sperry & Arora, 1965;Mark, 1965), as did an appropriate nerve implanted into its muscle simultaneously with a foreign nerve (Sperry & Arora, 1965). Thus, the results of experiments reported here in which NIV and the i.o. ...
1. The ability of a multiply innervated muscle to become dually innervated, that is to accept a functional innervation from both its original and a foreign nerve, was investigated using the superior oblique muscle (s.o. muscle) of the goldfish.
2. Dual innervation of s.o. muscles was achieved by allowing the original nerve (cranial NIV) to regenerate to its s.o. muscle which had been previously denervated and cross‐innervated by a foreign nerve (cranial NIII), or by allowing the original and the foreign nerve to regenerate simultaneously to a denervated muscle.
3. Behavioural observations suggested that in some fish reinnervation of the s.o. muscle by its original nerve repressed the function of a previously established foreign innervation. However, physiological tests which involved the stimulation of both foreign and appropriate nerves, and the recording of mechanical and electrical activity of the s.o. muscle, demonstrated that there was no functional displacement of foreign innervation on these muscles, even on individual dually innervated fibres.
4. Dual innervation of the s.o. muscle persisted, apparently unchanged, for as long as the observations were continued (up to 7 months). The s.o. muscle contains two populations of fibres, fast and slow, and both types became and remained dually innervated.
5. When both NIII and NIV were allowed to regenerate simultaneously to a denervated s.o. muscle there was no obvious selectivity in the final pattern of innervation. On the average both nerves elicited approximately equal tension from s.o. muscles, and evoked excitatory junctional potentials (e.j.p.s) of similar mean quantal contents.
6. ‘Myotypic respecification’ was shown not to be responsible for the discrepancy between the behavioural results which sugested that repression of foreign innervation had occurred, and the physiological results which demonstrated that this was not the case. Anatomical and physiological findings indicated that the discrepancy was attributable to eye rotation produced by regenerated inferior oblique muscle fibres which contracted simultaneously with the cross‐innervated s.o. muscle. The net result was an eye movement in which the activity of the s.o. muscle was masked.
7. It is concluded that repression of established foreign neuromuscular connexions following reinnervation by the embryologically correct nerve does not occur on goldfish extraocular muscles. The s.o. muscle can become non‐selectively innervated by both foreign and appropriate axons, and remains so, at least for several months.
... Weiss's student, Roger Sperry, adapted this experimental design to show that functional axon regeneration required target specificity, even within axons and targets of the same modality. By transecting motor nerves and suturing them to antagonistic muscles in rat (Sperry, 1941), monkey (Sperry, 1947) and fish (Sperry and Deupree, 1956), Sperry demonstrated that motor axons regenerated toward and formed electrically functional connections with antagonistic muscle, but these connections impaired rather than restored function. Based upon these observations, Sperry interpreted functional recovery after oculomotor axon transections in fish as evidence of target-selective regeneration (Sperry and Arora, 1965). ...
After injury, axons of the peripheral nervous system (PNS) regenerate, and yet functional recovery from peripheral nerve injury is rare. This is because PNS axons regrow slowly and often toward inappropriate targets. Peripheral nerves are composed of bundles of axons that exit the spinal cord via a shared path and then diverge toward different targets forming a complex meshwork of nerve branches. These branched bundles of axons are encased in layers of glia, endothelial cells, and associated extracellular matrix (ECM). After nerve injury, severed axons degenerate and are cleared away, but the encasing cells and ECM beyond the injury site remain as branched tube-like structures that lead to nerve targets. To reconnect with their pre-injury targets, regenerating axons must navigate through these nerve tubes. Importantly, at points where nerve tubes diverge into multiple branches (branch-points), regenerating axons must select the branch that leads to their pre-injury target. Despite important implications for functional recovery, the mechanisms that guide regenerating axons at nerve branch-points are poorly understood. To probe the cellular and molecular mechanisms that guide regenerating axons, we exploit the simple architecture of spinal motor nerves in larval zebrafish, which are composed of two axonal populations that initially share a common path but diverge at a stereotyped branch-point to innervate dorsal or ventral muscles. After laser nerve transection, axons regenerate along their original nerve branch >80% of the time. Using genetic mutants and in vivo time-lapse imaging, we demonstrate that the repulsive axon guidance receptor robo2 is necessary and sufficient to promote axon regeneration along the dorsal branch. During regeneration, a small subset of glia at the nerve branch-point upregulate the Robo-ligand slit1a and the ECM component col4a5. We demonstrate that robo2 functions in a common molecular pathway with col4a5 to guide regenerating axons dorsally, and that the spatiotemporal restriction of col4a5 to the nerve branch-point during regeneration is required to guide regenerating dorsal axons. Our results provide the first cell-autonomous mechanism by which regenerating axons select between nerve branches during regeneration and provide a molecular pathway by which glia at a nerve branch-point guide regenerating axons via local ECM modifications.
This chapter presents a general review of the misdirected reflex phenomenon in anurans, with reasons for rejecting selective peripheral regrowth as a class of mechanism by which cutaneous neurites could effectuate connexions with their appropriate sensory end organs. The refuted mechanisms include (1) peripheral searching, (2) multiple branching, and (3) the degeneration–regeneration of inappropriate neurites. The chapter reviews the data that argues against sensory impulse patterning as the source of cutaneous reflex localization. Behavioral observations, combined with nerve crush experiments, have shown that the peripheral nerve trunks that mediate only dorsal wiping responses in control frogs mediate misdirected, that is, ventrally directed wipes in skin-grafted animals. The cutaneous receptive fields (CRFs) for the nerve trunks are located in the same areas on the body surface in control and in skin-grafted animals, regardless of the type of skin present. Two experimental approaches might be useful to examine the alternative to the mechanism of modulation. One would be the use of early embryonic skin rotations, coupled with behavioral observations immediately after metamorphosis. The second proposed experimental approach would employ H3-thymidine autoradiography to establish the birthdates of neurons within the dorsal root ganglion (DRG).
In man and other mammals normal motor co-ordination is not restored, as a rule, after regeneration of a severed peripheral nerve-trunk (Sperry, 1945). The random misdirection of regenerating fibres into foreign muscles tends to prevent normal dissociated action within the re-innervated musculature. In contrast, larval amphibians have been found to show excellent recovery of motor function in the form of ‘homologous or myotypic response’ (Weiss, 1936, 1941) following the cutting and regeneration of limb-nerves, limb transplantation, and the crossconnecting of limb nerves to foreign muscles. Similarly, good restoration of muscle co-ordination has been observed in the pectoral fin of adult teleost fishes (Sperry, 1950, 1956).
It has been suggested (Sperry, 1941, 1951) that such recovery is most easily explained in terms of a central readjustment of synaptic connexions to suit the altered pattern of peripheral innervation. Morphological or other direct evidence for such synaptic changes, however, has not been found.
In twenty-one cats the superior cervical sympathetic ganglion was partially denervated by transecting sympathetic rami T1, T2, and T3. The procedure completely interrupts the preganglionic innervation of the pupil. The pupil became small immediately postoperatively and gradually returned to normal size after 36 weeks. This restoration of tonus was dependent upon the development of collateral sprouts from residual, nonpupillary fibers within the superior cervical ganglion because electrical stimulation of the thoracic sympathetic trunk caudad to T3 now elicited pupillary dilation (an effect never observed normally), and because the pupil became small again after severance of the thoracic sympathetic chain caudad to T3.Although these nonpupillary nerve fibers restored tonus to the pupil, they did not participate in sympathetic darkness reflexes (the dilation of the atropinized pupil in darkness and the consensual pupillary dilation elicited by occlusion of the contralateral pupil). Thus, the new peripheral terminations of the non-pupillary fibers did not induce any demonstrable change in the reflex connections of these neurons. These experiments support the view that in mammals there is no reorganization of central synaptic pathways in response to alterations of peripheral nerve terminations.
When the trochlear nerve (NIV), which innervates the superior oblique muscle (SOM), is crushed or cut at stages 48-49 in Xenopus tadpoles, fibers from the oculomotor nerve (NIII) sprout and invade the SOM. The maximal percentage of specimens having at least one oculomotor nerve fiber on the SOM on a given day increased from 9.1% following a single crushing of NIV to 84.2% following three successive severings of NIV and the average number of silver-impregnated NIII fibers per specimen increased from 0.23 +/- 0.16 (mean +/- S.E.M.) in the single-crush experiment to 7.35 +/- 1.33 in the triple-cut experiment. This increase directly reflects the delay in the return of NIV. As NIV returns to the SOM, a portion of the inappropriate innervation is lost; while another portion appears to be stable and is in evidence 90 days after a single sectioning of NIV. The more rapidly NIV returns to the SOM, the more complete is the displacement of the NIII fibers. This suggests that the association between NIII and the SOM changes with time so that easy displacement of the inappropriate innervation is likely only when the reinnervation by the appropriate nerve fibers is rapid.
Functional reinnervation of the metathoracic retractor unguis (R.u.) muscle of adult male cockroaches (Periplaneta americana) occurred 35–60 days postneurotomy in approximately 55% of experimental animals. The nerve branches supplying reinnervated R.u. muscles contain many more axons or axon branches than in control preparations. Regenerating axons contained small mitochondria, neurotubules, and frequently synaptic vesicles in regions of the axon several micrometers distant from the muscle fibers. These axons usually made contact with specialized regions of muscle fibers which lack myofilaments and were probably previous junctional sites although on occasions contacts were also established on relatively unspecialized regions of a myofiber. Reinnervation was accompanied by a reduction in adherens junctions which had developed between contiguous myofibers following neurotomy.
Reinnervation of the R.u. muscle was signaled by the reappearance of spontaneous transmitter release from the terminals of the R.u. motoneurons and the recovery of the muscle from denervation atrophy. There was no resumption of miniature activity before reinnervation occurred. Groups of up to 100 microtubules in an approximate hexagonal array were frequently observed in reinnervated myofibers. These groups diminished in size concomitant with an increase in the number of myofilaments and sarcoplasmic reticulum thereby restoring the myofibrillar substructure. Superficially located cells tentatively identified as myoblasts/myotubes have been observed following reinnervation.
It has long been considered a general rule for nerve regeneration that the reinnervation of skeletal muscle is nonselective. Regenerating nerve fibers are supposed to reconnect with one skeletal muscle as readily as another according to studies covering a wide range of vertebrates (Weiss, 1937; Weiss & Taylor, 1944; Weiss & Hoag, 1946; Bernstein & Guth, 1961; Guth, 1961, 1962, 1963). Similarly, in embryogenesis proper functional connexions between nerve centers and particular muscles are supposedly attained, not by selective nerve outgrowth but rather through a process of ‘myotypic modulation’ (Weiss, 1955) that presupposes nonselective peripheral innervation.
Doubt about the general validity of this rule and the concepts behind it has come from a series of studies on regeneration of the oculomotor nerve in teleosts, urodeles, and anurans and of spinal fin nerves in teleosts (Sperry, 1946, 1947, 1950, 1965; Sperry & Deupree, 1956; Arora & Sperry, 1957a, 1964).
Theodore Schwann, in hisstudy on the cellular nature of animal tissues (1839), was the first to describe the development of peripheral nerves. He believed, however, that the nerve fibre was formed by fusion of a row of the cells, which have since borne his name, but which are now known to form only the sheath around the nerve fibre. The latter has a wholly separate origin.
The developing peripheral nerves of the tail of amphibian larvae have been studied by a series of authors, the first of which was Schwann himself. Hensen (1864) realized that in their earliest stage the nerves of the tail fin were devoid of nuclei. In the 1870?s, several authors described the apparent origin of nuclei there in de now . In 1886, however, Kölliker followed the multiplication by mitotic division of Schwann cells, to which, as he realized, belong all the nuclei associated with the developing nerve fibre.
Balfour (1876) maintained that the whole of the peripheral nervous system in elasmobranchs was formed from cells which migrated outwards from the spinal cord. Hensen (1876) believed that nerves were formed along the tracks of protoplasmic strands which ran the whole distance between centre and periphery, and which originated in the incomplete cleavage of cells in the ectodermal layer of the early embryo. These theories were opposed to the views of Bidder & Kupffer (1857), His (1879), and Ramon y Cajal(1890) all of whom described the development of peripheral nerves in terms of the outgrowth of fibres from cells within the spinal cord and ganglia.
Ramon y Cajal was the first to study the developing nervous system by metallic impregnation methods. He concluded that no fusion occurs at points where the processes of neighbouring cells come into contact. The concept of cellular individuality in the nervous system was the basis of the Neurone Theory of Waldeyer (1891).
The first attack on this theory came with the work of Apathy (1889) who claimed to be able to follow stainable fibrillar constituents of the nerve fibre, the ‘neurofibrilla’, through the junction of one nerve cell with the next. Apathy stated that they were the actual conducting elements of the nervous system. The study of neurofibrillae in the developing nervous system by various workers has led to no uniform concepts. Paton (1907) believed that they can enter the nerve cell from outside; Held (1909) claimed that a pre‐existing network served as a guide from the growing nerve fibre. In this respect Held's views approximate to those of Hensen.
Fresh support to the neurone theory was gained in the pioneer experiments of Ross Harrison (1907) in the cultivation of tissues in vitro , in which he observed the outgrowth of living nerve fibres from amphibian neuroblasts into a medium of clotted fibrin. In the course of their growth such fibres received no support or any contribution from adjacent cells. Harrison described the detailed behaviour of the terminal growth cone, which had already been discovered by Ramon y Cajal.
Research on the outgrowth of nerve fibres has so far been largely confined to vertebrate embryos, though the ingrowth of axons from epidermal sensory cells in insects has been studied in recent years (Wigglesworth, 1953). In some invertebrates axons undergo an apparent secondary fusion to form giant fibres (Young, 1936), though the development of this condition has not yet been followed.
The outgrowth theory by itself leaves unexplained the forces which direct the nerve fibres during their growth. His (1887) believed that the fibre grows straight outwards until deflected by some structure lying across its path. Ramon y Cajal (1893) suggested that the outgrowing tips of nerve fibres were attracted towards their end organs by a chemotropic stimulus. Weiss, on the basis of experimental study on tissue cultures (1934), has elaborated a theory whereby a growing organ orientates intercellular material around itself, and so provides a pre‐formed pathway along which a growing fibre, with its accompanying Schwann cells, can reach its destination. Contact with surfaces is certainly an important factor in nerve outgrowth.
The mutual interactions of Schwann cells and nerve fibres in the living state have been extensively studied by Speidel(1932 onwards) in the tail fin of amphibian larvae. The progress of myelination is now being followed by means of the electron microscope (Geren, 1954).
Whatever the forces which direct growing nerve fibres, it is clear that there is no complete and detailed plan set before the unfolding peripheral nervous system from the outset, for some individual fibres choose random and aberrant paths. Ramon y Cajal (1908) was the first to point out the significance of these ‘fibres égarées’.
Although no direct experimental proof of chemotropic attraction of nerve fibres has been obtained, chemical forces are invoked in the adjustments which follow when a nerve has approached its end organ. On the motor side Weiss (1936) found that in Amblystoma grafted supernumerary limbs move in unison with an adjacent normal member. Corresponding muscles contract synchronously even though the anatomical pattern of their regenerated nerves may be aberrant. Weiss explained this ‘resonance effect’ by postulating that when a regenerating motor nerve reaches a muscle at random, a chemical stimulus is transmitted centripetally, which leads to rearrangement of central synaptic junctions. Similar readjustments are inferred in the experiments of Sperry (1951) in which areas of skin on the body surface of a tadpole are interchanged. A grafted patch continues to transmit to the centre information concerning location of the body surface which refers to its original site even though in its new position it is innervated by surrounding cutaneous nerves.
0 : suppl. Experimental changes in end-feet of Held-Auerbach in the BARR, M. L. Axon reaction in motor neurons and its effect upon the end-CAMPBELL, B. 1944 The effects of retrograde degeneration upon reflex activity Functional changes EDQEWORTH, F. H. 1935 The Cranial Muscles of Vertebrates
- R I Literature Cited Barnard
- C B Downman
- J C Eccles
- A K Mcintyre Cambridge Uni-Greqory
LITERATURE CITED BARNARD, R. I. 1938 Experimental changes in end-feet. Anat. Rec., 7 0 : suppl. Experimental changes in end-feet of Held-Auerbach in the BARR, M. L. Axon reaction in motor neurons and its effect upon the end-CAMPBELL, B. 1944 The effects of retrograde degeneration upon reflex activity DOWNMAN, C. B., J. C. ECCLES AND A. K. MCINTYRE 1953 Functional changes EDQEWORTH, F. H. 1935 The Cranial Muscles of Vertebrates. Cambridge Uni-GREQORY, W. K. 1951 Evolution Emerging. Vol. 11, Macmillan Co., N. Y. RASDOLSKY, I. 1925 Die Beniitzung einer neuen Methode der Farbung der Nervenfasernkollateralen. Virchows Arch., 257 : 356-363.
- Sperry