Assistant Professor of Anatomy
Department of Anatomy,
School of Medicine,
State University of New York at Buffalo,
Buffalo, New York***
Web Contact: pietsch@indiana.edu
Despite much effort and a great number of words dedicated in its behalf regeneration in the amphibian appendage remains -- insofar as primary mechanism are concerned -- an enigma. This is unfortunate, for the phenomenon exhibits many of the features encountered along the expanse of developmental events and might indeed serve to enrich understanding of all situations in which cells grow and assume new levels of organization. The overall objective in this series of studies has been to identify the principles that govern differentiation during regeneration. Muscle and cartilage in the regenerative urodele limb have served that end rather well, for both tissues are easy to recognize while the entire organ, when examined closely, does not present a clear-cut endpoint. Experimentation (see Pietsch '61a-'62b) thus far has revealed:SUMMARY
Forelimbs of Amblystoma punctatum larvae were transplanted to the orbit in place of the eye. Subsequently the musculature in the transplanted limbs underwent changes in morphology for a distance of about 1.5 mm from the point of union of host and donor tissues; beyond about 2 mm, the musclature retained its limb-like patterns. In some cases the transplanted limbs were amputated through the region where typical cross sectional anatomy of the musculature had been retained. In other experiments amputation was made through the region of altered musculature. Regeneration took place under both condition. Where the musculature of the stump had retained characteristic limb features both the muscles and skeletal elements differentiated normally. Amputation through a region with altered musculature produced regenerates with correspondingly altered muscles but with cartilaginous skeletons typical of the forelimb.These findings demonstrate that myogenesis and chondrogenesis during limb regeneration are, from the outset of the reaction, mutually independent events with different antecedents.
- some cells of the blastemal mesenchyme are chondrogenic from the outset of their histories as transplantable entities;
- cartilages produced by blastema cells isolated from stump tissues bear the anatomical features found in normally regenerated limb skeletons;
- the blastema, as such, is not myogenic;
- the differentiation of limb-like musculature seems to be a function of stump muscle: a narrow band of stump tissue remaining at the base of a transplanted blastema insures the development typically patterned limb muscles, and a blastema transplanted in place of an eye produces a limb skeleton surrounded by a musculature with morphological attributes that reflect the muscles of the orbit.
- the cells producing muscle and cartilage -- and not their environment -- determine the morphogenesis of these two tissues; and, secondly
- myogenesis and chondrogenesis are mutually independent sets of events; i. e., they are unrelated to a common antecedent.
Fig. 1. Cross section through the antibrachium of an orbitally transplanted limb of an A. opacum larva. Muscle is arranged in two demilunes on either side of the cartilaginous skeleton. This is completely atypical of the antibrachial musculature (see Pietsch, '61b, fig. 3). It is usually possible to distinguish the ulna from radius on the basis of the ulnocarpalis muscle (ventral to the former), but not in this photograph. Epithelium in the lower right corner belongs to the host site. X 70 (approx.)
The hypothesis predicted that amputation of an orbitally transplanted limb with a modified musculature would subsequently produce a regenerate with typical limb cartilages set amid muscles showing non-limb morphology.
MATERIALS AND METHODS
Experiments were performed with larvae of three species of Amblystoma that varied in size and age. However, the results followed the same trend, and in deference to simplicity this report shall focus mainly on a single group of experiments with A. punctatum 25 mm in overall length at the onset of the study. With an animal anesthetized in MS 222, and its gently but snugly head trussed in place in a crux formed by two straight pins, the eye was extirpated from the prospective host orbit and a segment of limb (level presented below) from the same animals was cut off and floated into the desired position with the host and donor wound surfaces in apposition. Autoplastic transplantations were made merely for convenience, and it is noted that essentially the same results were obtained (in ancillary experiments) with homoplastic and even heteroplastic graftings.Limbs were removed from either the neck of the humerus or from the region just below the elbow joint. The objective in the humeral level transplants was to provide controls; i. e., orbitally transplanted limbs with antibrachial regions presenting typical muscular pattern. These humeral level transplants were allowed to heal in the orbit for 15 days and at that time were subjected to amputation through the proximal third of the forearm; this left in the orbit a stump in excess of 2 mm and with a high probability that the region of amputation would contain typical muscles (see Pietsch, '61, p. 297). To be sure of this, the removed piece of transplant of each experimental was examined microscopically and the normal muscle morphology verified.
Transplants from below the elbow were amputated through the distal antibrachium after a 15-17 day healing period. This left a stump of less than 0.25 mm protruding beyond the orbital margin; amputated pieces of limb were checked microscopically to insure that the muscle patterns had changed.
Normal regeneration was studied on the contralateral limb of each animal. In each in situ limb, the amputation was made concurrent with, and through the same plane similar as in the orbitally transplanted limb.
Specimens were preserved six weeks after amputation, in Bouin's fluid if they were to be studied histologically, in ethanol-formalin if destined for methylene blue staining (see Noback, '16) and in toto examination of the skeleton. Histological studies were made on conventionally prepared serial cross sections that had been stained with Delafield's hematoxylin and eosin Y. In toto analyses were carried out with the aid of a dissecting microscope.
RESULTS
1. Muscle and cartilage in normal limb regenerates.
Studies on normal limb regenerates were in essential agreement with description already in the literature (Blount, '35, figures in Piatt, '57). Nevertheless, this investigation depended on the reliability of the morphological criteria revealed in muscle and cartilage, and, therefore, a summary of these findings will be presented here for the reader's convenience.The endpoints of analysis were the muscles and cartilages distal to the elbow. The hand of the Amblystoma larva is built around four digits with a like number of corresponding metacarpals (fig. 2). The thumb is missing. In these studies (as is usually the case) digit V was more poorly developed than other digits and sometimes was absent entirely. When digit v was missing, metacarpal V was absent and the hand was then structured around three rather than four fingers.
Fig. 2. Normally regenerated limb of an A. punctatum larva as seen from the dorsal aspect. Initial amputation was made below the elbow. Radius is on the reader's left. The proximal carpals are named by virtue of location: radiale, intermediale and ulnare. The centrale, not always found in regenerated limbs, lies distal and somewhat to the left of the intermediale. Note a carpale at the base of each metacarpal and a digital formula of 2/2/3/1. X 30.
The number of phalanges per digit varied somewhat with the most common formula being 2/3/3/1. Ideally there is a carpal as the base of each metacarpal but in about 10 percent of the cases examined, two or more of these appeared fused into a single enlarged cartilage. The proximal row of carpals consisted of three cartilages which were regular both in structure and occurrence. These proximal carpals are named for their relationships and positions (viz. radiale, intermediale, ulnare). A centrale cartilage, so name because of its position between the proximal and distal rows of carpals, was observed in a slight majority of cases. Despite minor variations, no difficulty was encountered in analyzing normally regenerated forelimb skeletons or in identifying specific, individual cartilages.
The details of the musculature in the limb of a small larva are microscopic but are readily revealed in transverse histological section (fig. 3 in Pietsch, '61b). In the forearm and writs analysis was facilitated by first identifying the ulnocarpalis muscle. A tiny ribbon of vertically directed fibers, it has been encountered ventral and parallel to the ulna in every normal Amblsytoma forelimb regenerated examined in this laboratory. It is also distinctive in the wrist (fig. 5 in Pietsch, ibid).
The musculature in the hand does not have a single identifying landmark, such as the ulnocarpalis more proximally. However, the following features always were observed:
- the dorsum of the hand contained tendons (of the long digital extensor) but no fleshy fibers.
- Ventrally, muscles occupied three planes:
- superficial (representing the distal-most extent of fleshy fibers of the palmaris superficialis);
- ventrometacarpallar muscle (individual heads of the palmaris profundus); and intermetacarpallar muscle.
2. Regeneration from humeral level transplants.
Normal limb regeneration did not present a control situation. There was no way of knowing, a priori, that the general conditions of experimentation were compatible with the differentiation of a typical limb musculature. Humeral level transplants (with typical limb muscles) after healing in the orbit were amputated through the upper forearm region and were then allowed to regenerate for six weeks. After that time regenerates exhibited typical external features and characteristic limb skeletons. Five cases were examined in histological cross section and each exhibited essentially the same things. One shall now be described more fully. Most of the preexisting stump showed muscles in non-limb arrangements, this being true for a greater distances than previous estimates had suggested (Pietsch, '61b). In the most proximal forearm sections the radius and ulna were set amid uninterrupted, unfasciculated muscle masses, but a few sections beyond this, the ulnocarpalis made its appearance. Traversing 20 u distally (fig. 5 and fig. 6), the entire cross section was a typical representative of the forelimb save for the absence of large nerve bundles. The precise location of the plane of amputation could not be ascertained, but throughout the remainder of the organ the muscular pattern was indistinguishable from that encountered under ordinary conditions of regeneration.
3. Regeneration from the antibrachial level transplants.
Forearm level transplants having healed in place for for 17 days were amputated though the region above the wrist. Fifty-four cases were examined in detail, 45 in toto and 9 histologically. Without an exception, each individual piece of cartilage was one or another constituent of the forelimb skeleton. In about 60 percent of the cases the regenerated portion of the skeleton was perfectly formed (fig. 3).
Fig. 3 Regenerate that developed from an antibrachial level transplant.
Amputation was made leaving a stump of less than 0.25 mm. Specimen was fixed
six weeks later and was stained in toto with methylene blue. The skeleton is
complete and typical of the limb. Compare with figure 2. The single phalanx
in digit V was present in this case but cannot be seen from this photographic
angle. X 30.The others were normal except for one or another missing cartilage or the fusion of two elements into a single enlarged structure.
Histological examination of serial cross sections of nine cases randomly selected from the antibrachial level revealed in each a non-limb muscle pattern in both the preexisting stump and throughout the newly regenerated portions (fig. 7 and fig. 8) of the limb.
DISCUSSION
The results just presented (see Fig. 4) demonstrate that myogenesis and chondrogenesis in the limb regenerate are mutually exclusive of each other and that they are independent of a common antecedent.
Fig. 4 Representation of results: - A depicts regeneration in the intact limb with the stippled area signifying the new organ.
- B represents the operative control in which a limb was transplanted from the humeral level and then suffered amputation through the region where typical muscular patterns had been retained.
- C indicates the antibrachial level transplant in which the muscle had undergone change throughout.
- A and B situations led to muscle patterns as indicated in I whereas C produced the pattern shown in II.
- Both circumstances produced the normal skeletal morphology represented at the right of the illustration.
The "independence hypothesis" outlined in the introduction accurately predicted, a priori, the experimental findings of this investigation. Evidence of the developmental independence in question also extends to nerves (Pietsch and Webber, in preparation), not only of the orbit, but of the dorsal fin (Pietsch '62a). Recently (Pietsch, '62c) it was observed that a covering with head integument did not preclude normal chondrogenesis during limb regeneration. The same has been true both for muscle and cartilage differentiation in contact with dorsal fin epithelium and mesenchymal connective tissue (see Pietsch, '61, '62a). This line of evidence refutes the field concept as applied to limb regeneration (see discussion and literature in Goss, '61). According to the field theory, a regenerate would become a limb at the direction of factors of the stump at large with the cells -- to use Weiss's ('39) allegory -- representing an unorganized herd. If the field theory is to have any objective referent then skeletogenesis should reflect the circumstances associated with muscles, nerves, etc. This was anything but true in the present findings.
Given that myogenesis and chondrogenesis are causally independent events, it follows that either two categories of cells exist or that extramural agents associated with regenerated muscle are basically unlike; i. e. tissue-tissue induction. It was reasoned a priori that if the former alternative were correct the results would turn out as they did. While the hypothesis is allowable, the truth of the matter rests on better information concerning the source(s) and particular fates of the involved cells. However, tissue-for-tissue induction as represented by Goss ('56, '61) is denied not only because it rests on the field concept but because of internal inconsistencies in that view: Goss concluded that the skeleton is histogenically induced by its counterpart in the stump. The addition of an extra piece of cartilage to the stump led to a supernumerary skeletal element in the regenerated antibrachium. But the hands in Goss's figures appear perfectly constituted (see '56, figs. 2 and 3). The question immediately arises, how did the distal blastema cells escape the alleged inductive effects of the extra cartilages? Competence might possibly be an escape from the dilemma save for the fact that carpals, metacarpals and digits consistently differentiate even after the blastema is isolated from the stump and transferred to the fin at the earliest transplanted stage (Pietsch, '61, fig. 5). In addition, skeletal elements develop under many conditions -- including regeneration -- where no bone or cartilage preexists (see Goss, '56; Holtzer, '56; Bridges, '59). Goss, who contributes to this evidence himself, asserts that skeletogenesis is accomplished after the stump regulates to reinstate the lost skeleton-inducing substances. This is a prima facie example of the logical fallacy of begging the question. The alleged regulatory capacity is simultaneously a conclusion and a proof for a thesis of which it is itself a part. Finally,a chemically purified cartilage-inducing substance (an oligonucleotide) has come not from cartilage but from notochord and spinal cord (Lash et al, '61, Hommes et al, '62). Thus, if induction -- using the term rather loosely -- is at the basis of the present findings, it is not operative in the tissue-for-tissue manner described by Goss.
Circumstantial evidence, logic and the fact that it passed the pragmatic test of truth (predicted the oucome of the experiments) allow as a working hypothesis the view that the basic reactions in the differentiation of the limb regenerate are inherent functions of the particular cells involved in the production of each tissue, that the environment in which they develop -- while doubtless of importance -- serves in an ancillary and non-specific capacity. This hypothesis is consistent with and can explain:
- the present results;
- the limb-like chondrogenesis from blastema isolated from the stump at very early stages (Pietsch, '61a);
- the perfect regenerative limb musculatures that develop in the dorsal fin when the stump remains attached to the transplanted blastema (op. cit.) and
- the extraocular muscle patterns developed following transplantation of the blastema to the orbit (Pietsch, '62b).
LITERATURE CITED
Blount, I. W. H. 1935 The anatomy of normal and reduplicated limbs in amphibia with special reference to musculature and vascularization. J. Exp. Zool., 69: 407-458.Bridges, J. B. 1959 Experimental heterotopic ossification. Internat. Review of Cytol., 8:253-278.
Goss, R. J. 1956. The relation of bone to the histogenesis of cartilage in regenerating forelimbs and tails of adult Triturus viridescens. J. Morph., 98: 89-123.
Goss, R. J. 1961 Regeneration of vertebrate appendages. In Advances in Morphogenesis vol. I, pp. 103-152, ed. by M. Abercrombie and J. Brachet. Academic Press, New York and London.
Holtzer, S. 1956 The inductive activity of the spinal cord in urodele tail regeneration. J. Morph., 99: 1-40
Hommes, F. A. C., C. von Leeuwen and F. Zelliken 1962 Induction of cell differentiation II. The isolation of a chondrogenic factor from embryonic chick spinal cords and notochords. Biochim. Biophys. Acta, 56: 320-325.
Lash, J. W., F. A. Hommes and F. Zilliken 1962 Induction of cell differentiation I. The in vitro induction of vertebral cartilages with a low molecular-weight tissue component. Ibid., 56: 313-319.
Morgan, T. H. 1901 Regeneration. The Macmillan Co., New York.
Needham, J. 1950. Biochemistry and Morphogenesis. Cambridge University Press. Cambridge.
Nicholas, J. S. 1955 Regeneration, Vertebrates, in Analysis of Development, ed by Willier, B. H., Weiss, P. A. and Hamburger, V. W. B. Saunders. Philadelphia.
Noback, G. J. 1916 The use of the van Wijhe method for the staining of the cartilaginous skeleton. Anat. Rec. 11: 292-293.
Piatt, J. 1957 Studies on the problem of nerve pattern. III. Innervation of the regenerated forelimb in Amblystoma. J. Exp. Zool. 136: 229-247.
Pietsch, P. 1961a Differentiation in regeneration. I. The development of muscle and cartilage following deplantation of regenerating limb blastemata of Amblystoma larvae. Develop. Biol., 3: 255-264.
Pietsch, P. 1961b. Effects of heterotopic musculature on myogenesis during limb regeneration in Amblystoma larvae. Anat. Rec. 141: 295-304.
Pietsch, P. 1962a. The influence of spinal cord on differentiation of skeletal muscle in regenerating limb blastema of Amblystoma larvae. Anat. Rec. 142:169-178.
Pietsch, P. 1962b. Skeletogensis following modification of stump muscle in regenerating limbs of Amblystoma larvae. Anat. Rec. 142: 268.
Pietsch, P. 1962c. Effects of head epithelium on limb regeneration in Amblystoma opacum larvae. Anat. Rec. 142: 321
Pietsch, P. 1962d. The specificity of deplanted regenerating limb blastema cells during skeletogenesis as studies in Amblystoma larvae. Amer. Zool., 2: 438.
Weiss, P. 1939 Principles of Development. Henry Holt and Company, New York.
FIGURES 5-8
Fig. 5 Cross section though a hand regenerated from an orbitally transplanted
limb that possessed typical muscular cross section through the region of
amputation (as judged by examining to amputated piece). The muscle pattern
that developed is typical of the normally regenerated Amblystoma hand.
The arrows indicate the three planes of fibers: one between the metacarpals;
one made up of individual heads of each lying ventral to the cartilages; and
third, a superficial sheet forming the dorsal boundary of a large volar
vascular channel. Compare with figures in Blount ('35). X 70.
Fig. 6 Cross section through the lower antibrachium of a specimen similar to
that shown in figure 5. The ulnocarpalis is indicated by an arrow, and the
basis of it other structures in the section may be identified. This is
representative of the cross sectional anatomy of normal limb regenerates, save
for the absence of nerve bundles. (Compare with figures in Piatt, '57;
Pietsch, '61b; see also illustration in figure 4.) X 70.
Fig. 7 Cross section through the hand of a transplant regenerate that
developed from a stump with modified musculature. Unlike either the situation
in figure 5 or in the normal, this hand does not exhibit individual muscles and
dorsally where there are usually no fibers one can observe a massive fillet of
muscle. X 70.
Fig. 8 Cross section through the lower antibrachium of a transplant regenerate
developed from a stump with modified musculature. The features seen in figure
6 are lacking despite the abundance of skeletal muscle. In general, the
specimen is morphologically similar to that shown in figure 1 Compare with figure 5 in Pietsch '61b. X 70.
* Supported by N.I.H. grant RG=9217
**A preliminary report of this work was presented before the 75th Session of the American Association of Anatomists, March 22, 1962.
***At Indiana University, Bloomington, Indiana, USA since 1970
return to Shufflebrain menu