Phototaxic behavior a nd the retinotectal transport of horseradish peroxidase (HRP) in surgically created cyclopean salamander larvae (Ambystoma)

School of Optometry, Indiana University, Bloomington, IN 47405 (U.S.A.) and
Department of Psychology, Indiana University of Pennsylvania, Indiana, PA 15701 (U.S.A)

Adapted from an article originally published in Neuroscience Research 18:35-43 (1993)

SUMMARY

Heavy deposits of HRP were found in the tecta of some animals that lacked NP. To find out if an optic tectum is actually required for NP, a series of ablation experiments were performed, using A. punctatum larvae. Tectectomy had the same effect on NP as bilaterally extirpating the eyes or intracranially severing both optic nerves; i.e. removing the tectum abolished NP.

The results:

1. confirm the efficacy of the ectopic eye in the cyclops preparation;
2. show that the ectopic eye can regenerate retinotectal pathways;
3. indicate that retinotectal contact is a necessary but insufficient condition for NP.

INTRODUCTION

We also inquired into whether or not the cyclopean eye can, by anterograde transport, deliver horseradish peroxidase (HRP) to the optic tectum. In addition, we present supplemental data from ablation experiments showing that phototaxic behavior requires the tectum.

MATERIALS AND METHODS

Animals

Animals were kept individually in 5 percent Holtfreter's solution, changed daily, and were fed newly hatched brine shrimp or enchytreas worms.

Operations

Two types of eye operations were performed, orthoclops and cyclops; the former involved the removal and orthotopic reimplantation of the eye, the latter the translocation of the eye (with antero-posterior axis maintained) to the top of the head, above the pretectal area. The contralateral eye was removed and discarded in both types of operations. Surgery was performed on a Petri dish lined with Vermont marble clay. The anesthetized subject was secured first by inserting a pair of insect pins into the sides of the mouth at a low angle, through the floor of the oropharynx and into the underlying clay; secondly, by gently trussing the body at the mid-thorax between two decussating straight pins. In the cyclops operation the skin over the dorsal neurocranium was reflected as a flap from the right to the left side of the head and down over the left orbit; the flap was used to draw the eye slowly from its obrit, the optic nerve and other attachments to the globe being severed as encountered; the ventral periorbital skin was temporarily left attached so as to preserve orientation. The eye was loosely packed back into the orbit while the host site was prepared. Next the right eye was removed and discarded; the still-unossified investing neurocranial cap was excised above the entire midbrain and the dorsal diencephalon. Then the left eye was cut free from its remaining attachments and moved up the head to the hole in the skull, care being taken to maintain the same longitudinal and transverse orientations the eye had had while in the orbit. The stump of the optic nerve was aimed at the pretectal area. The periorbital skin on the eye was then smoothed down, closing the wound surface on the head. A Stultz Brucke, or straddle bridge (see Rugh, 1962 ), made of transparent Tygon was then placed over the eye and slowly and gently eased down against the cornea. Smooth and transparent when wet, the Tygon permitted fine adjustments of the Brucke to align the eye's visual axis on the perpendicular. The operation complete, the MS 222 was diluted to approximately 1:7000; the Brucke and restraining pins were loosened 1-2 hours postoperatively but the animal was kept under light anesthesia for an additional 5-6 hours to ensure healing of the graft.

With orthoclopes, the right eye was removed and discarded; the left eye was then subtotally retracted from the orbit in the manner used in the cyclops operation with the ventral periorbital skin attachment intact. Its nerves and adnexa cut, the eye of the prospective orthoclops was replaced in the orbit, and the stump of the optic nerve was aimed at the optic foramen. A Brucke was used to secure the orthoclopean eye, and the subject received the postoperative care already described for the cyclops operation.

The supplemental experiments with A. punctatum larvae involved the following: tectectomy (n=7); bilateral intracranial optotomy (n=4); bilateral eye extirpation (n=4); craniotomy (n=6); unoperated (n=5). The tectectomy was performed in a manner described elsewhere (Pietsch and Schneider, 1985; 1991). To guarantee the removal of the tectum, the entire cerebral aqueduct was exposed. The reasons for chosing A. punctatum versus axolotl larvae will be presented with the results.

Behavioral Apparatus

HRP

Because our animals are aquatic and of relatively small size, we could not simply inject HRP solutions into the eye. Axolotls can survive in damp air for a few hours, but after prolonged periods out of water develop gravity-induced hemostasis; nor can their ocular tissues be sutured or sealed with surgical glues. Thus we had to devise methods for bringing very high concentrations of HRP into contact with the retinal ganglion cells while maintaining the viability of the tissues. Considering the mechanisms of anterograde transport of HRP (Mesulam and Mufson, 1980), and after having conducted a large number of preliminary experiments, we adopted the following as standard operating procedures.

The implant operation began while the animal was still submerged (in MS 222 solution); a 270-degree, circumferential incision was made in the head skin about 3 mm from the edge of the globe. An arcing, parallel incision was then made through the coats of the eye approximately 1 mm behind the surface location of the ora serrata. Using the skin flap for grasping, the anterior segment of the eye, with the lens and vitreous body attached, was then slowly teased out and reflected away from the vitreous chamber. The inner surface of the undetached retina was abraded with an iris knife. The operating dish was then lifted and slowly drained in the direction of the reflected anterior eye segment. The vitreous chamber was then packed with HRP crystals (roughly 500 micrograms); the crystals immediately began to dissolve, turning the retina brown. Grasping the apron of periorbital skin, and using the lens as a fulcrum, the operator slowly rolled the reflected segment of the eye back into place, the vitreous body becoming brown upon contact with the HRP. The edge of the periorbital skin was pressed firmly against the wound surface on the head; in the process the cut surfaces of the eye were brought into apposition, and the globe was sealed. The animal was kept out of fluid for approximately 2 hours but its body and tail were moistened periodically with drops of anesthetic fluid. Finally, the animal was transferred to full strength Holtfreter's solution; the latter was diluted to 5 percent after 2 days. Supplementary control data were obtained from previously unoperated A. mexicanum larvae which received HRP unilaterally, in the manner just described.

The principal subjects (albino cyclopes and albino orthoclopes) were sacrificed 2 weeks after the implantation of HRP; supplemental controls (unoperated axolotl larvae) survived either 1 or 2 weeks after the HRP challenge. Duration of the survival time did not influence the results with previously unoperated animals, suggesting that the endpoint reaction was reaching saturation by the time the principal subjects had been sacrificed. In preparation for sacrifice, the dorsal neurocranium was opened, the meninges over the dorsal epithalamus and dorsal mesencephalon were reflected, and, so that fixative fluid would rapidly reach the interior of the midbrain, the tectum was quickly split mid-sagittally with the tip of an iris knife, exposing the floor of the cerebral aqueduct. Freed from the lower jaw, the head was severed and immediately transferred to a 4 degree-C fixative solution consisting of 1.25 percent glutaraldehyde and 1 percent paraformaldehyde (Mesulam, 1982), brought to pH 7.4 with phosphate buffer; fixation time was 1-2 hours; tissues were rinsed and processed for the TMB precisely as described by Mesulam (1978; 1982); at the end the TMB procedure, for purposes of stabilizing the reaction product, the heads were transferred to 5 percent ammonium molybdate for 30 minutes (Fujii and Kusama, 1984). In our experience, despite the latter measures the TMB reaction product remains somewhat light-sensitive; therefore the brains were kept on ice and were examined and photographed under the stereoscopic microscope immediately after the ammonium molybdate treatment. Some specimens were imbedded in sucrose, frozen at -35 degree-C and sectioned; microscopic inspection did reveal reaction product in areas where the latter could be seen as heavy blue-black deposits under the stereoscopic microscope. Relative to the whole mount methods, sectioning presented no advantages, but many disadvantages, and, therefore, was abandoned as a primary source of data.

Statistical analyses were carried out with either SPSSPC, version 2.0 or in RS/1 on a VAX 7620 computer running on VAX/VMS version 5.5-2.

RESULTS

Recovery of negative phototaxis

The albino cyclopes became our principal experimental subjects with the albino orthoclopes as the main control. To insure that we would also have subjects of similar age and genotype, yet lacking NP, we ran concurrent series with pretested wild type axolotls.

Of albino cyclopes, 25 of 26 subjects survived through run three of NP testing, as did 9 of 10 albino orthoclopes. In order to have a rational basis for setting the NP criterion, we waited until the data for run three had been collected and then let the computer determine the interval at the 95 percent confidence level for the ideal population representing the principal controls, the orthoclopes; the calculated minimum for the latter was 78.1853. Therefore, we established 78 minutes as the criterion of NP.

At run one (one month postoperatively), 4 cyclopes (16 percent) and one orthoclops (11 percent) reached the NP criterion (see Table 1). Neither the mean nor the median of either group achieved the NP criterion at run one: for albino cyclopes the mean was 59.67+/-15.7 s.d. and the median, 61.90; for albino orthoclopes the mean at run one was 65.64+/-11.16 s.d. and the median, 68.00. We postulate that, functionally, optic nerve regeneration was still in progress in most subjects during run one.

By run two, 10 albino cyclopes (40 percent) exhibited NP, as did 8 (89 percent) albino orthoclopes; for the cyclopes the mean at run two was 71.60+/-11.6 s.d. and the median, 71.80; for the orthoclopes the mean at run two was 83.69+/-9.55 s.d. and the median, 84.70. (Note: the values in Table 1 are rounded off.)

At run three, 16 of the surviving 25 albino cyclopes (64 percent) successfully passed the NP test. Among the albino orthoclopes the NP performers remained at 88 percent. The unadjusted scores for the entire albino cyclopes group at run three generated a mean (79.3+/-12.4 s.d.) that exceeded the NP criterion. But one entire operative group of 6 albino cyclopes (** in Table 1) had failed to reach the NP criterion at run three; nor had these animals shown signs of NP at either runs 1 or 2. Although the latter animals did not change the basic argument, their scores seemed to contribute to a false impression of the quantitative circumstances attending the albino cyclopes. When the values of the non-performing group were set aside, very similar mean scores obtained in run three for the albino cyclopes and the albino orthoclopes: 84.56+/-8.48 s.d. and 84.75+/-8.55 s.d., respectively. Statistical analysis (see Table 2 ) indicated that the two populations were quantitatively similar (F ratio was 0.0031 and F probability 0.9558). At the 95 percent confidence level, the theoretical population (min-max) was 80.47 to 88.65 for albino cyclopes and 81.39 to 87.86 for albino orthoclopes; i. e., the two populations would mutually vary by less than one minute during a 120 minute session (0.92 min and 0.79 max).

In sum, when transplanted to the orbit, the eye had about a 24 percent better chance of reestablishing NP than when the host site was the top of the head. But if NP recovered at all, it reached the same level in the cyclopes as in the orthoclopes.

Horseradish peroxidase studies

Of the high-performing albino cyclopes (NP+), 4 of the 6 exhibited HRP-TMB reaction product in the tectum (Fig. 1); i.e., were HRP-positive (HRP+) while the other 2 (Fig. 2) were HRP-negative (HRP-).

Fig. 1. HRP-TMB reaction product in a dissection of the tectum of an NP+ albino cyclops, as seen in the stereoscopic microscope. The tectum is divided sagitally into right and left halves. The tectal halves were slightly everted so as to place reaction product in the photographic plane. The floor of the cerebral aqueduct, somewhat out of focus, can be seen between the tectal halves. Anterior on the specimen is directed towards the top of the photograph. Non- specific reactions with TMB show in the photograph as small spots. The spots had a brownish hue, in contradistinction to the heavy blue- black representing sites of HRP deposition, and were readily distinguishable from the latter. Primary magnification X 20.

Fig. 2. Dissection of the tectum of an NP+ albino cyclops, lacking in any evidence of an HRP-TMB reaction product. The small dark spots are non-specific reactions with TMB. Primary magnification X 20.

Fig. 3. HRP-TMB reaction product in a dissection of the tectum of an NP+ albino orthoclops, as seen in the stereoscopic microscope. Compare with Figs. 1 and 2. Primary magnification X 20.

To gain a better estimate of the distribution of reaction product in the cyclopes and orthoclopes, the tecta were mapped into quadrants, and the quadrants were scored binarily, 1 if the reaction was present or 0 if absent. The latter data are collected in Table 3. With the possible exception of quadrant I in albino orthoclopes, the HRP reaction product was distributed without any obvious pattern among the various sectors of the tectum, independent of both the subject's performance on the NP test and whether the transplanted eye was in the orbit or on top of the head.

Supplementary tectectomy experiments

Incidental observation

Fig. 4. Albino A. mexicanum larvae. From reader's left to right: cyclops with scoliosis; orthoclops with scoliosis; cyclops without scoliosis. Each subject exhibited NP. The animals were immobilized in MS 222 for the photograph. primary magnification X 1.4.

DISCUSSION

The second question, as to whether the heterotopic eye can connect to the optic tectum, also can be affirmatively answered. The cyclopes reestablished retinotectal pathways as well as, if not better than, the orthoclopes.

Some subjects with high NP scores showed no HRP reaction product in the tectum. HRP methods depend on the complex interaction of numerous, fortuitous and delicately balanced circumstances, any of which could escape the attention or elude the control of the experimenter. Thus technical artifacts represent the most parsimonious, and intuitively the most probable, explanation of the failure of HRP to appear in all cases where we ideally would expect it. The available facts do not warrant extended speculation about interesting alternatives, except to take note of the existence of several other optic nuclei and relay stations in the larval salamander's brain (see Herrick, 1948; Jakway and Riss, 1972; Pietsch and Schneider, 1985, 1991). Whether one or more of the latter may substituted for the tectum seems unlikely but not impossible. Examination of other optic regions of the brain in our specimens failed to reveal HRP-TMB reaction product, possibly because our methods were insufficiently discriminating.

Nor were our methods suitable for minutely tracing the pathways from the retina to the tectum (or elsewhere). Preparation for fixation, especially of the cyclopes, disturbed the eye and, doubtless, distorted the delicate optic nerve. We have, therefore, deferred judgments concerning the precise neuroanatomical details of the cyclopes.

The disappearance of NP following tectectomy in our supplementary series shows that the tectum is necessary for this behavior, at least for the natural eye.

We were surprised by the fact that a few subjects failed the NP test but, nevertheless, were HRP+. Given the stringent cellular requirements for a histochemically demonstrable HRP-TMB reaction product (see Mesulam and Mufson, 1980), we are logically forced to postulate that subjects lacking in NP, but positive for HRP, possessed physiologically competent retinotectal connections. Is a retinotectal pathway a sufficient condition for NP? We can most succinctly treat the issue implicit in the question by examining the assertion:

not -NP =not -HRP+ {1}

(not-NP implies not-HRP in the tectum), a logically and formally valid statement which our data show to be false (see definitions in Kleene, 1967). HRP was present in abundance in some orthoclopes and cyclopes that either did not recover NP (albinos) or inherently lacked the behavioral trait, altogether (wild type). Thus, while a connection between the retina and the tectum appears to be a necessary condition for NP, based upon the tectectomy series, additional conditions must be satisfied for the animal to display NP behavior. While the posited additional conditions are not evident in our data, we can infer that regenerating optic nerve fibers have a better than even chance of reestablishing them, even when the eye is transplanted to an unusual location. Optically elicited metachromasia of pigmented Ambystoma larvae (skin camouflage reaction) depends on more than one pathway (Pietsch and Schneider, 1985). The recovery of NP could conceivably require the successful reconstruction of two (or more) sets of retinotectal connections.

Although our principal research interest only remotely touches on the specificity-plasticity question, our data illustrate the need for caution in choosing between the two sides of what has become a major controversy among students of neural regeneration (see discussion and literature in Jacobson, 1993). Our data show that the termination pattern of the regenerated pathways can vary considerably from those of the naturally occurring tracts, a result that may be an extension of findings for the adult newt, namely that many regenerating retinal fibers reach the tectum by way of abnormal routes (Fujisawa, 1981). Data such as the latter, if viewed phenomenologically, would appear to offer strong support to the plasticity school. However, from equation {1}, we might alternatively argue that viable but non-participating retino-tectal terminals obscured the specificity underlying the NP behavior of our experimental subjects.

A small but significant number of cyclopes and orthoclopes developed scoliosis. While the details concerning the latter remain to be worked out, these points warrant mention. First, in that the condition afflicted cyclopes as well as orthoclopes, the observed scoliosis would appear to be related to regeneration of the optic nerve, not the ectopic location of the eye. Secondly, despite severe postural handicaps, and often locomotor dysfunctions, the scoliotics were nevertheless able to perform the NP task.

ACKNOWLEDGEMENT

REFERENCES

Fujisawa, H. (1981) Retinotopic analysis of fiber pathways in the regenerating retinotectal system of the adult newt Cynops pyrrhogaster. Brain Res., 206: 27- 37.

Fujisawa, H., K. Watanabe, N. Tani and Y. Ibata (1981) Retinotopic analysis of fiber pathways in amphibians. I. The adult newt Cynops pyrrhogaster. Brain Res., 206: 9-20.

Herrick, C. J. (1948) The Brain of the Tiger Salamander. The University of Chicago Press, Chicago.

Jakway, J. S. and W. Riss (1972) Retinal projections in the tiger salamander Ambystoma tigrinum. Brain Behav. Evol. 5:401-442.

Jacobson, M. (1993) Historical development of the concept of neuronal specificity. In Sharma, S. C. and J. W. Fawcett, Formation and regeneration of nerve connections, Chapter 1, Birkhäuser, Boston, Basel, Berlin.

Keene, S. C. (1967) Mathematical Logic. Wiley and Sons, New York, London and Sydney.

Kirkpatrick, T., C. W. Schneider and R. Pavloski (1991) A computerized infrared monitor for following movement in aquatic animals. Behav. Res. Methods, Instr. Computers. 23:16-22.

Mesulam, M.-M. (1978) Tetramethyl benzidine for horseradish peroxidase neurohistochemistry: a non-carcinogenic blue reaction product with superior sensitivity for visualizing neural afferents and efferents. J. Histochem. Cytochem. 26: 106-117.

Mesulam, M.-M. (1982) Principles of horseradish peroxidase neurohistochemistry and their applications for tracing neural pathways--Axonal transport, enzyme histochemistry and light microscope analysis. In: Mesulam, M.-M.(ed), Tracing Neural Connections. John Wiley and Sons, Singapore.

Mesulam, M.-M. and E. J. Mufson (1980) The rapid anterograde transport of horseradish peroxidase. Neuroscience 5: 1277-1286.

Pietsch, P. (1972). Scrambled salamander brains: a test of holographic theories of neural program storage. Anatomical Rec. 172:383-84.

Pietsch, P. (1981) Shufflebrain. Houghton Mifflin, Boston.

Pietsch, P. and C. W. Schneider (1985) Vision and the skin camouflage reactions of Ambystoma larvae: the effects of eye transplants and brain lesions. Brain Res. 340: 37-60.

Pietsch, P. and C. W. Schneider (1991) Tectectomy in the cyclopean salamander. Physiol. Behav. 50:305-309.

Reh, T. A. and M. Constantine-Paton (1984) Retinal ganglion cell terminals change their projection sites during larval development of Rana pipiens. J. Neurosci. 4:442-457.

Rugh, R. (1962) Experimental Embryology. Burgess, Minneapolis.

Schneider, C. W. and P. Pietsch (1968) The effects of addition and subtraction of eyes on learning in salamander larvae (Amblystoma punctatum). Brain Res. 8:271-280.

Schneider, C. W., B. W. Marquette and P. Pietsch (1991) Measures of phototaxis and movement detection in the larval salamander. Physiol. Behav. 50:645-647.

Vanegas, H. (1984) Comparative anatomy of the optic tectum. Plenum, New York.

CYCLOPES                                        ORTHOCLOPES                                  
Case No.       run 1  run 2     run 3        Case No.       run 1  run 2     run 3     
22.135.2        43.9  79.1      86.4         22.140.1        70.5  79.1      87.5      
22.135.3        58.9  78.3      97.1         22.140.2        55.3  84.7      93.2      
22.135.5        38.4  70.6      85.4         22.140.3        60.8  83.4      87.5      
22.137.1        79.5  71.6      85.2         22.140.5        47.1  65.6      68.0      
22.137.2        66.0  75.8      84.4         22.140.6        76.4  79.1      95.1      
22.137.3        65.9  97.3      97.4         22.141.1        69.2  88.8      77.4      
22.137.4        81.9  93.1      95.7         22.141.2        83.8  101.8     89.9      
22.137.5        49.3  72.4      94.1         22.141.4        68.0  85.0      79.3      
22.137.6        71.3  79.8      89.9         22.141.6        59.7  85.7      84.9      
22.138.2        67.4  74.7      78.1              means      65.6  83.7      84.8      
22.138.3        15.6  65.0      72.9          s.d.(+/-)      11.2  09.6      08.6      
22.138.4        45.9  59.1      85.1            medians      68.0  84.7      87.5      
22.138.5        64.0  57.7      78.1                                                   
**22.139.1      79.7  63.5      73.5                                                   
**22.139.2      62.6  61.6      58.8                                                   
**22.139.3      61.9  52.3      56.6                                                   
**22.139.4      50.2  57.3      59.1                                                   
**22.139.5      64.1  60.3      61.1                                                   
**22.139.6      50.6  62.4      66.4                                                   
22.142.1        68.9  71.8      73.9                                                   
22.142.2        48.3  60.5      65.3                                                   
22.142.3        49.8  78.3      85.6                                                   
22.142.4        56.9  79.6      86.7                                                   
22.142.5        61.9  82.7      84.7                                                   
22.142.6        88.8  85.1      80.7                                                   
     means      59.7  71.6      79.3                                                   
 s.d.(+/-)      15.7  11.6      12.4                                                   
   medians      61.9  71.8      84.4                                                   

*run data are in minutes

**included in overall analysis but excluded from the analysis in Table 2

• Cochrans C = max. variance/sum (variances) = 0.5038, P = 0.978
• Bartlett-Box F = 0.001, P = 0.980
• Maximum variance/minimum variance = 1.015

return to Shufflebrain menu

For a non-technical work on this subject go here

Comments or inquiries: pietsch@indiana.edu