- Email: [email protected]
The physiological identification of pyramidal tract neurons within transplants in the rostral cortex taken from the occipital cortex during development
Brain Research. 436 ( 1987) 136-142 Elsevier
136 BRE 22628
Short Communications
The physiological identification of pyramidal tract neurons within transplants in the rostral cortex taken from the occipital cortex during development L.L. Porter 1, J.M. Cedarbaum 1, D . D . M . O'Leary 2, B.B. Stanfield 3 and H. Asanuma I t Laboratory of Neurophysiology, The Rockefeller University, New York, NY (U.S.A.); 2Department of Neurosurgery, Washington University School of Medicine, St. Louis, MO (U.S.A.) and 3Laboratory of Clinical Science, National Institute of Mental Health, National Institutes of Health Animal Center, Poolesville, MD (U.S.A.)
(Accepted 1 September 1987) Key words: Motor cortex; Cortical development; Corticospinal; Intracortical microstimulation (ICMS); Antidromic
Axons from neurons in the occipital cortex transiently extend to the pyramidal tract (PT) during the early postnatal development of rats. Normally, these axons are eliminated by the end of the third postnatal week. However, if a portion of fetal occipital cortex is transplanted to the parietofrontal region in newborn hosts then some neurons in the transplant will extend pyramidal tract axons and maintain them. Intracortical microstimulation and electrophysiological recording techniques were used to identify the physiological characteristics of the transplanted pyramidal tract cells and to determine if motor effects could be elicited from the occipital transplant. Microstimulation of the transplant did not reliably evoke movement but the low density and disarray of PT cells within the transplant might account for this. Recording from within the transplant revealed that the overall cell activity was depressed. We were able to identify neurons within the transplant which responded antidromically to stimulation of the pyramidal tract, indicating that their axons have the capacity to conduct impulses and are therefore likely to have developed some viable connections. The functional significance of such projections remains uncertain. During early postnatal development in the rat, pyramidal tract neurons are distributed throughout the tangential extent of neocortex including the occipital cortex 22. With subsequent development, the distribution of pyramidal tract (PT) neurons is increasingly restricted and eventually confined to the rostroparietal and frontal regions 1°,22. This developmental restriction is most marked in the occipital cortex in which many PT neurons can be identified in early postnatal stages but from which all PT neurons are excluded by the end of the third postnatal week 23. Exclusion of the occipital pyramidal tract neurons occurs through the process of selective elimination of PT axon collaterals rather than by the death of their cells of origin 22. The cells which give rise to the transient PT axons survive and maintain connections to subcorticai nuclei such as the superior colliculus and
the pons 14. It seems then, that regional differences in the maintained projections of layer V neurons are not intrinsic to the neurons themselves which initially have similar projectional capacities. Rather, these regional differences are eventually established through a process of collateral elimination which is influenced by 'supragenomic factors '2°. In support of this notion is the observation that pieces of fetal occipital cortex which have developed in a rostral cortical locale contain neurons which maintain pyramidal tract axons into adulthood 15. In order to examine the viability of these PT neurons we have used intracortical microstimu~,~t;~-~ !iCMS) and e_lectrophysiological recording ~e_ hniques to examine the projection of the PT fibers arising from occipital to rostral transplants and the possible motor effects of the transplanted tissue.
Correspondence: L.L. Porter, The Rockefeller University, 1230 York Ave., New York, NY 10021, U.S.A.
i37 We used l0 adult rats to determine if movement could be elicited from the transplant. These rats had received fetal occipital to rostrai cortex transplants on the day of birth. Since the method for cortical transplantation was described in detail in a previous paper 2~ only a brief outline will be given here. Tritiated thymidine ([3H]thymidine, 6.7 Ci/mmol, NEN) was injected intraperitoneally into pregnant rats on the fifteenth day of gestation. Two days later the fetuses were removed from the mother and a small piece of cortex from the occipital region was dissected out in ice-cold H A M F-10 medium. At the same time newborn rat pups from a second litter were anesthetized by hypothermia and prepared to serve as hosts for the transplants: a skull flap was cut over the parietofrontal area and a small hole was made in the underlying cortex. The piece of donor occipital cortex was trimmed and placed into the hole. The skull flap was sealed and the host pups were returned to their mothers. Later, the transplants could be identified in histological sections processed for autoradiography by the presence of [3H]thymidine labeled neurons within the donor tissue. When the rats reached the age of 3 months to one year the physiological experiments were performed. Initially, several control animals were tested to develop the experimental parameters and to establish the topographic pattern of the body parts in normal motor areas. The rats were anesthetized with 1.01.5% halothane in 50% oxygen and 50% nitrous oxide. The skull and dura were opened to expose the rostral cortex of one hemisphere which was then covered with mineral oil and photographed. A long-lasting local anesthetic was applied to the tissue around the surgical opening so that the level of general anesthesia could be kept low to facilitate move~nents evoked with ICMS. The rats were suspended in a specially designed harness attached to the stereotaxic apparatus allowing their hindlimbs and forelimbs to hang freely TM. A tungsten-in-glass microelectri~de was lowered into the motor area of the cortex and a train of 10 pulses (0.2 ms duration at 300 Hz) was delivered at 1 s intervals at 100 F~m steps along the electrode penetration. The locations of electrode tracks from which visible and consistent movements could be elicited were marked on the photographs and the threshold current in each case was recorded. At the end of the experiment, several lesions ~vere made by
passing negative current of 10 ~uA for 10 s. These lesions, along with the photographs, were used as landmarks for reconstruction of the electrode tracks in later histological examination. Results from 3 animals were drawn as a composite map of the cortical surface which depicted a topographical organization of body parts similar to those previously reported 6 including a separate rostral forepaw representation which was reported more recently 12,~3. The animals containing occipital cortical transplants were prepared similarly with the exception that both hemispheres were exposed. Several electrode penetrations were made into the normal hemisphere before moving the stimulation electrode to the side of the cortical transplant. In this way the threshold current needed to produce movement could be determined for both the control and transplant sides in individual animals. Then the area thought to contain the transplant was identified by abnormalities of the cortical surface, noting especially changes in the vascular pattern. The area was explored with a microelectrode used for both stimulation and recording. Unit activity was monitored as the electrode was lowered through the depth of the cortex with a micromanipulator and stimulus trains were delivered at 100 ~m steps to evoke movement. When movement was elicited, the threshold current and location of the movement was determined. Then a lesion was made by passing 10~A of DC current for 10-20 s through the same electrode. Penetration sites throughout the extent of the abnormal appearing tissue and around its periphery were marked on the photographs. Then the animals were deeply anesthetized with chioropent (150 mg/kg) and perfused with formalin and the brains were post-fixed, and sectioned on a freezing microtome at a thickness of 50 ~m in the frontal plane. Series of alternating sections were processed for autoradiography or for Nissl staining. The location of the electrode tracks were reconstructed from Nissl sections using the marking lesions and photographs. Alternate sections were processed for autoradiography and analyzed to deter~ mine if any of the tracks or the lesion sites fell within the boundaries of the transplant. Several electrode tracks passed through the transplants and in most of these penetrations the number of active units and the background activity encountered by the electrode was decreased noticeably rel-
138 ative to the surrounding host cortex. The lesions which marked locations where stimulation elicited movement were most often found outside the border of the identified transplant. Adjacent penetrations into host cortex evoked movement of the same or neighboring body parts of the individual animals. However, when the penetration sites were near the transplanted cortex, threshold intensities increased irregularly and it was difficult to map the entire body representation. Furthermore, amongst those electrode penetrations which were clearly in host cortex, the threshold of current needed to evoke movement from the transplanted side (x = 27 BA) was significantly higher than that needed to produce movement from the control side (x = 18 BA) (t-test, P < Two animals had lesion sites clearly located within the transplant indicating that movement had occurred in response to stimulation of transplanted neurons. In both cases, however, the electrode tip was near the border of host and donor tissues. In one animal the marking lesion was surrounded by sparsely scattered [~H]thymidine-labeled cells and therefore was within the transplant, but it was only about 200~m from the transp!ant's edge. Stimulation at this site produced elbow flexion at a threshold intensity of 40/~A. When the electrode was lowered by 500/~m the stimulus threshold for the same response dropped sharply to 16/~A. Histological reconstruction of the
electrode's path indicated that the electrode had then entered the host tissue. Results from the second case were similar. Vibrissal movement was elicited from within the transplant at a stimulus intensity of 30 BA, but here again the stimulus site was within 100-200 /~m of the border of the area containing labeled cells. Lowering the electrode 200 ltm brought it into host cortex and in this location the threshold current for vibrissal movement decreased to 20/~A. Stimulation of an adjacent penetration site located approximately 400/zm from the lesion at nearly the same cortical depth also evoked whisker movement with a threshold intensity of 30/~A. This electrode track was clearly in host cortex. In a second series of experiments, we attempted to determine the electrophysiological properties of pyramidal tract neurons located within the occipital-torostral transplants. Five animals, aged 3 months to 1 year, which had received transplant~ as newborn pups were anesthetized and surgically prepared as previously outlined. After exposure of the cortex a stimulating microelectrode was implar, ted into the pyramidal tract with stereotactic coordinates ~7 used as a guide. To ensure that the electrode was in the correct position single pulses of 0.2 ms duration were delivered to the pyramidal tract while surface recordings were made at various locations around what appeared to be normal motor cortex. Response latenc~es and stimulus thresholds were noted and corn-
A
Fig. 1. A: a tracingof ihe section containing the lesion which marks the location of the stimulating electrode in the pyramidal tract. The electrode penetrated the pyramidal tract at the level of the pyramidal decussation. The hatched area indicates the extent of the damage. Abbreviations: CC, central canal; Cu, cuneate nocleus; cu, cuneate fasciculus; gr, gracile fasciculus; py, pyramidal tract; pyx, pyramidal decussation; Sp5, nucleus spinal tract trigeminal nerve; sp5, spinal tract trigeminal nerve; 12, hypoglossal nucleus. B: a photomicrograph through a portion of the section from which the drawing in A was made. This shows the lesion (open arrows) at higher magnification. Scale bar in B, 250~m.
139 pared to previous recordings of cortical surface potentials following PT stimulation 4"16. Then a tungsten-in-glass recording microelectrode was lowered into the area thought to contain the cortical transplant. When a single unit was encountered which responded to stimulation of the pyramidal tract we ensured that the response was antidromic by testing the latency, the ability to follow high frequency stimulation with constant latency and the ability to maintain constant latency to stimulation 2-3 times threshold current ~. Lesions were made at the site of antidromic responses and at the end of the experiment a lesion was made through the stimulating electrode in the pyramidal tract (Fig. 1). Subsequent histological analysis showed that in one animal lesions marking the location of 2 cells which exhibited antidromic action potentials were clearly within the boundaries of the transplant. In both cases numerous autoradiographically labeled cells surrounded the lesions (Fig. 2). These 2 cells responded with latencies of 2.0 and 2.5 ms which is within the expected limits of PT axon conduction ve-
locity9"~6. Furthermore, they clearly met the criteria outlined above for demonstrating antidromic responses to PT stinmlation (Fig. 3). The depth at which we encountered these cells was inconsistent with that for PT cells in normal tissue 8'24 but cortical lamination was seldom retained in the grafts. Our experiments reveal some interesting features of the physiological characteristics of the occipitai-torostral cortex transplants. As already stated, the level of unit activity recorded by electrodes which penetrated the transplanted tissue was noticeably reduced compared to penetrations through host tissue. In fact, during the course of our experiments in the second part of this study, this difference could be used fairly reliably in locating the transplant. Several factors might account for the decreased level of unit activity in the transplanted cortex. One possibility is that the core of the graft may receive sparse II or patchy 5 innervation from host afferents. The resultant paucity of synaptic input might reduce unit activity. Golgi studies of cortical implants reveal unusual variations in position and orientation of somata, and
Fig. 2. A: a low-powerbright-field photomicrograph of a Nissl-stainedsection through a transplant in which PT cells were found. The arrowheads roughly outline the transplanted tissue as determined autoradiographically. B: a higher powered dark-field photomicrograph of a part of the transplant shown in A. The open arrows in A and B indicate the site of the marking lesion. The marking lesion is well within the transplant in a region containingmany neurons overlayed by clusters of silver grains. Some of these are indicated by the filled arrows in B. Scalebar in A, 500~m; in B, 125/~m.
140 abnormalities in branching patterns of both dendrites and local axon distributions 7. These factors, along with the obvious disruption of cortical lamination might have disturbed the local interactions between cells within the transplant and resulted in depression of unit activity. Although numerous penetrations were made into the transplanted cortex, reliable motor responses
A
&
B
0
i Z e;
m~
t &
L,
A
°Smv°ltmsec Fig. 3. Tests for antidromic response to pyramidal tract stimulation. A: the responses to single stimuli of threshold intensity delivered to the pyramidal tract. This cell responded with a latency of 2.0 ms. B: the response of the same cell when two times the threshold intensity was delivered to the pyramidal tract. The latency remained nearly the same. C: the cell's response to high frequency stimulation. The cell fired in response to multiple stimuli with nearly constant latency. Arrowhead indicates PT stimulation.
could not be elicited from the core of the transplant, even though in the surrounding host cortex, cells retained their ability to produce movement. From previous anatomical work 2~ and from the abnormal depths at which we encountered PT cells, it was clear that there is a loss of proper orientation of the cortical laminae within the transplants and a disruption of the density and distribution of PT cells therein. Instead of a densely packed aggregation of PT neurons within layer V, the large pyramidal cells were scattered rather widely through the grey matter. Furthermore, while Nissl sections showed a large number of healthy neurons within the transplants, tracing studies reveal fewer PT cells in the transplant than in the neighboring host cortex 5'21. Thus, when ICM5 was applied to a limited area of transplanted cortex the number of activated PT cells may have been insufficient to produce movement. In addition, the density or distribution of the synaptic terminals formed by the pyrami~ dal tract axons arising in the transplants is not known, When the occipital transplants ~rt. first placed in the rostral cortex on the day of bii~th, s o ~ e PT neurons in the adjacent host tissue have already cx~ended axons as far as the pyramidal decussation 3"19. By the time PT axons from the transplant reach the spinal cord, they may be at such a competitive disadvantage that the number and distribution of synaptic contacts they are able to form may be relatively small. Similarly~ since PT cells appear to be somewhat scattered in the transplant their axons may not form a functional group with a common target; that is, their terminal branches may be widely dispersed and stimulation would affect only a sparse number of synaptic contacts at any one spinal level, and in insufficient number to produce movement. While the number of synaptic contacts formed by transplanted occipital pyramidal tract cells may be small, some apparently remain active because, as we have shown, their axons are maintained into adulthood. The PT cells which maintain these fibers might account for the 2 cases where movement was elicited by stimulation within the transplant. However, we cannot rule out with certainty the involvement of the host cortex in these cases. Neurons within an approximate radius of 150 pm from the electrode tip 2 are expected to be directly affected with the current used, and if synaptic spread is considered the distance is even greater. Since in both cases the lesions were
141 within approximately 200/~m of host tissue, d~'ect and synaptic spread of the stimulus to the host co~'tex could have contributed to, if not accounted fo~, the movement. As mentioned above, stimulation along many electrode penetrations limited to the host cortex surrounding the transplant resulted in movement. It is difficult to explain the finding that minimum current intensities needed to produce movement from penetrations made on the side of the transplant but clearly in the intact host cortex, were significantly higher than those for the control side of the same animal. For the most part, the normal cell packing density and the laminar organization appear to be maintained in the host tissue of the transplant side. It m a y be, that with the removal of a portion of the host cortex and with the relatively late arrival of PT axons from the transplant that the remaining host cortex develops a more widely dispersed terminal field than normal. These widely scattered synaptic contacts may be less effective in eliciting movement following ICMS. In this study, :we ~dentifiec! a small n u m b e r of neurons within occip~,~.~o~i-t~',-rostral transpiar~tg as PT
cells by antidromic stimulation of their axons. The identification of PT neurons within the heterotopic cortical transplants confirms the findings of retrograde tracing studies that fetal occipital PT neurons transplanted to a rostral cortical locale can extend and maintain projections characteristic of rostral cortex ~5"2~. F u r t h e r m o r e these occipital PT axons are maintained into adulthood. However, direct motor effects in response to microstimulation were not observed with certainty in these experiments. It seems then, that under these experimental conditions the occipital PT neurons are not able to fully express the motor cortical PT neuronal phenotype. On ,he other hand, the fact that these neurons are capable of responding to low threshold microstimulation of the pyramidal tract implies that their axons have the capacity to conduct impulses and are likely therefore, to have formed some appropriate connections with their postsynaptic targets.
1 Abzug, C. and Peterson. B., Antidromic stimulation in the pontomeduUary reticular formation of local axon branches in con~¢,~a~c~,d ~cstibular neurons, Brain Research, 64 ( i 977.).~i~7-413, 2 As;mq~r~ta, H., Stoney, S.D., Ir. and Abzug. C., Relatiottshi!? hz,tween afferent input and motor outflow in cat motorser~g~:rycortex, J. Neurophysiol., 31 (1968) 670-681. 3 DeMeyer, W., Ontogenesis of the rat corticospinal tract, Arch. Neurol., 16 (1967) 203-211. 4 Elger, C.E., Speckman, E.-J., Caspers, H. and Janzen, R.W.C., Cortico-spinal connections in the rat. I. Monosynaptic and polysynaptic responses of cervical motoneurons to epicorticai stimulation, Exp. Brain Res., 28 (1977) 385-404. 5 Floeter, M.K. and Jones, E.J., Connections made by transplants to the cerebral cortex of rat brains damaged in utero, J. Neurosci., 4 (1984) 141-150. 6 Hall, R.D. and Lindholm, E.P., Organization of motor and somatosensory neocortex in the albino rat, Brain Research, 66 (1974) 23-38. 7 Jaeger, C.B. and Lund, R.D., Transplantation of embryonic occipital cortex to the brain of newborn rats: a Golgi study of mature and developing transplants, J. Comp. Neurol., 200 (1981) 213-230. 8 Krieg, W.J.S., Connections of the cerebral cortex. I. The albino rat. B. Structure of the cortical areas, J. Comp. Neurol., 84 (1946) 277-324. 9 Landry, P., Labeile, A. and Deschenes, M., Intracortical distribution of axonal collaterals of pyramidal tract cells in the cat motor cortex, Brain Research, 191 (1980) 327-336. 10 Leong, S.K., Localizing the corticospina! ne-rons in neona-
tal, developing and mature albino rat, Brain Research, 265 (1983) 1-9. !1 Lund, R.D, and Hauschka, S.D., Transplanted neural tissue develops cot~nect~ons with host rat brain, Science, 193 (1976) 582-584. 12 Neafsey, E.J., Bold, E.L., Haas, G., t-']urley-Gius, K.M., Quirk, G., Sicv,ert, C.F. and Terreberry, R.R., The organization of the rat motor cortex: a microstimulation mapping study, Brain Res. Rev., 11 (1986) 77-96. 13 Neafsey, E.J. and Sievert, C., A second forelimb motor area exists in rat frontal cortex, Brain Research, 232 (1982) 151-156. 14 O'Leary, D.D.M. and Stanfield, B.B., Occipital cortical neurons with transient pyramidal tract axons extend and maintain collaterals to subcortical but not intracortical targets, Brain Research, 336 (1985) 326-333. 15 O'Leary, D.D.M. and Stanfield, B.B., Maintained projections of heterotopic transplants of fetal visual and sensorimotor cortex are characteristic of new cortical locale, Soc. Neurosci. Abstr., 12 (1986) 1049. 16 Ohta, M., Cortical motor function and the antidromic cortical response to stimulation of the medullary pyramidal tract in the rat,Jap. J. Physiol., 18 (1968) 100-124. 17 Paxinos, G. and Watson, C., The Rat Brain in Stereotct~°ic Coordinates, Academic, Sydney, 1982. 18 Sapienza, S., Talbi, B., Jacquemin, J. and Albe-Fessard, D., Relationship between input and output of cells in motor and somatosensory cortices of the chronic awake rat, Exp. Brain. Res, 43 (1981) 47-56. 19 Schreyer, D.J. and Jones. E.G., Growth and target finding by axons of the corticospinai tract in prenatal and postnatal
This work was supported by N I H Grants NS18065, EY-03653 and NS-10705. L.L.P. was an N I H Postdoctoral Fellow, F32 NS-07713.
142 rats, Neuroscience, 7 (1982) 1837-1853. 20 Stanfield, B.B., Postnatal reorganization of cortical projections: The role of collateral elimination, Trends Neurosci., 7 (1984) 37-41. 21 St,,~lfield, B.B. and O'Leary, D.D.M., Fetal occipital cortical n,:,-urones transplanted to the rostral cortex can extend ar.,i maintain a pyramidal tract axon, Nature (Lond.), 313 (1985) 135-137. 22 Stanfield, B.B., O'Leary, D.D.M. and Fricks, C., Selective collateral elimination in early postnatal development re-
stricts cortical distribution of rat pyramidal tract neurones, Nature (Lond.), 298 (1982) 371-373. 23 Stanfield, B.B. and O'Leary, D.M.M., The transient corticospinal projection from the occipital cortex during the postnatal development of the rat, J. Cornp. Neurol., 238 (1985) 236-248. 24 Zilles, K., Zilles, B. and Schleicher, A., A quantitative approach to cytoarchitectonics VI. The areal pattern of the cortex of the albino rat, Anat. Embryol., 159 (1980) 335-360.
136 BRE 22628
Short Communications
The physiological identification of pyramidal tract neurons within transplants in the rostral cortex taken from the occipital cortex during development L.L. Porter 1, J.M. Cedarbaum 1, D . D . M . O'Leary 2, B.B. Stanfield 3 and H. Asanuma I t Laboratory of Neurophysiology, The Rockefeller University, New York, NY (U.S.A.); 2Department of Neurosurgery, Washington University School of Medicine, St. Louis, MO (U.S.A.) and 3Laboratory of Clinical Science, National Institute of Mental Health, National Institutes of Health Animal Center, Poolesville, MD (U.S.A.)
(Accepted 1 September 1987) Key words: Motor cortex; Cortical development; Corticospinal; Intracortical microstimulation (ICMS); Antidromic
Axons from neurons in the occipital cortex transiently extend to the pyramidal tract (PT) during the early postnatal development of rats. Normally, these axons are eliminated by the end of the third postnatal week. However, if a portion of fetal occipital cortex is transplanted to the parietofrontal region in newborn hosts then some neurons in the transplant will extend pyramidal tract axons and maintain them. Intracortical microstimulation and electrophysiological recording techniques were used to identify the physiological characteristics of the transplanted pyramidal tract cells and to determine if motor effects could be elicited from the occipital transplant. Microstimulation of the transplant did not reliably evoke movement but the low density and disarray of PT cells within the transplant might account for this. Recording from within the transplant revealed that the overall cell activity was depressed. We were able to identify neurons within the transplant which responded antidromically to stimulation of the pyramidal tract, indicating that their axons have the capacity to conduct impulses and are therefore likely to have developed some viable connections. The functional significance of such projections remains uncertain. During early postnatal development in the rat, pyramidal tract neurons are distributed throughout the tangential extent of neocortex including the occipital cortex 22. With subsequent development, the distribution of pyramidal tract (PT) neurons is increasingly restricted and eventually confined to the rostroparietal and frontal regions 1°,22. This developmental restriction is most marked in the occipital cortex in which many PT neurons can be identified in early postnatal stages but from which all PT neurons are excluded by the end of the third postnatal week 23. Exclusion of the occipital pyramidal tract neurons occurs through the process of selective elimination of PT axon collaterals rather than by the death of their cells of origin 22. The cells which give rise to the transient PT axons survive and maintain connections to subcorticai nuclei such as the superior colliculus and
the pons 14. It seems then, that regional differences in the maintained projections of layer V neurons are not intrinsic to the neurons themselves which initially have similar projectional capacities. Rather, these regional differences are eventually established through a process of collateral elimination which is influenced by 'supragenomic factors '2°. In support of this notion is the observation that pieces of fetal occipital cortex which have developed in a rostral cortical locale contain neurons which maintain pyramidal tract axons into adulthood 15. In order to examine the viability of these PT neurons we have used intracortical microstimu~,~t;~-~ !iCMS) and e_lectrophysiological recording ~e_ hniques to examine the projection of the PT fibers arising from occipital to rostral transplants and the possible motor effects of the transplanted tissue.
Correspondence: L.L. Porter, The Rockefeller University, 1230 York Ave., New York, NY 10021, U.S.A.
i37 We used l0 adult rats to determine if movement could be elicited from the transplant. These rats had received fetal occipital to rostrai cortex transplants on the day of birth. Since the method for cortical transplantation was described in detail in a previous paper 2~ only a brief outline will be given here. Tritiated thymidine ([3H]thymidine, 6.7 Ci/mmol, NEN) was injected intraperitoneally into pregnant rats on the fifteenth day of gestation. Two days later the fetuses were removed from the mother and a small piece of cortex from the occipital region was dissected out in ice-cold H A M F-10 medium. At the same time newborn rat pups from a second litter were anesthetized by hypothermia and prepared to serve as hosts for the transplants: a skull flap was cut over the parietofrontal area and a small hole was made in the underlying cortex. The piece of donor occipital cortex was trimmed and placed into the hole. The skull flap was sealed and the host pups were returned to their mothers. Later, the transplants could be identified in histological sections processed for autoradiography by the presence of [3H]thymidine labeled neurons within the donor tissue. When the rats reached the age of 3 months to one year the physiological experiments were performed. Initially, several control animals were tested to develop the experimental parameters and to establish the topographic pattern of the body parts in normal motor areas. The rats were anesthetized with 1.01.5% halothane in 50% oxygen and 50% nitrous oxide. The skull and dura were opened to expose the rostral cortex of one hemisphere which was then covered with mineral oil and photographed. A long-lasting local anesthetic was applied to the tissue around the surgical opening so that the level of general anesthesia could be kept low to facilitate move~nents evoked with ICMS. The rats were suspended in a specially designed harness attached to the stereotaxic apparatus allowing their hindlimbs and forelimbs to hang freely TM. A tungsten-in-glass microelectri~de was lowered into the motor area of the cortex and a train of 10 pulses (0.2 ms duration at 300 Hz) was delivered at 1 s intervals at 100 F~m steps along the electrode penetration. The locations of electrode tracks from which visible and consistent movements could be elicited were marked on the photographs and the threshold current in each case was recorded. At the end of the experiment, several lesions ~vere made by
passing negative current of 10 ~uA for 10 s. These lesions, along with the photographs, were used as landmarks for reconstruction of the electrode tracks in later histological examination. Results from 3 animals were drawn as a composite map of the cortical surface which depicted a topographical organization of body parts similar to those previously reported 6 including a separate rostral forepaw representation which was reported more recently 12,~3. The animals containing occipital cortical transplants were prepared similarly with the exception that both hemispheres were exposed. Several electrode penetrations were made into the normal hemisphere before moving the stimulation electrode to the side of the cortical transplant. In this way the threshold current needed to produce movement could be determined for both the control and transplant sides in individual animals. Then the area thought to contain the transplant was identified by abnormalities of the cortical surface, noting especially changes in the vascular pattern. The area was explored with a microelectrode used for both stimulation and recording. Unit activity was monitored as the electrode was lowered through the depth of the cortex with a micromanipulator and stimulus trains were delivered at 100 ~m steps to evoke movement. When movement was elicited, the threshold current and location of the movement was determined. Then a lesion was made by passing 10~A of DC current for 10-20 s through the same electrode. Penetration sites throughout the extent of the abnormal appearing tissue and around its periphery were marked on the photographs. Then the animals were deeply anesthetized with chioropent (150 mg/kg) and perfused with formalin and the brains were post-fixed, and sectioned on a freezing microtome at a thickness of 50 ~m in the frontal plane. Series of alternating sections were processed for autoradiography or for Nissl staining. The location of the electrode tracks were reconstructed from Nissl sections using the marking lesions and photographs. Alternate sections were processed for autoradiography and analyzed to deter~ mine if any of the tracks or the lesion sites fell within the boundaries of the transplant. Several electrode tracks passed through the transplants and in most of these penetrations the number of active units and the background activity encountered by the electrode was decreased noticeably rel-
138 ative to the surrounding host cortex. The lesions which marked locations where stimulation elicited movement were most often found outside the border of the identified transplant. Adjacent penetrations into host cortex evoked movement of the same or neighboring body parts of the individual animals. However, when the penetration sites were near the transplanted cortex, threshold intensities increased irregularly and it was difficult to map the entire body representation. Furthermore, amongst those electrode penetrations which were clearly in host cortex, the threshold of current needed to evoke movement from the transplanted side (x = 27 BA) was significantly higher than that needed to produce movement from the control side (x = 18 BA) (t-test, P < Two animals had lesion sites clearly located within the transplant indicating that movement had occurred in response to stimulation of transplanted neurons. In both cases, however, the electrode tip was near the border of host and donor tissues. In one animal the marking lesion was surrounded by sparsely scattered [~H]thymidine-labeled cells and therefore was within the transplant, but it was only about 200~m from the transp!ant's edge. Stimulation at this site produced elbow flexion at a threshold intensity of 40/~A. When the electrode was lowered by 500/~m the stimulus threshold for the same response dropped sharply to 16/~A. Histological reconstruction of the
electrode's path indicated that the electrode had then entered the host tissue. Results from the second case were similar. Vibrissal movement was elicited from within the transplant at a stimulus intensity of 30 BA, but here again the stimulus site was within 100-200 /~m of the border of the area containing labeled cells. Lowering the electrode 200 ltm brought it into host cortex and in this location the threshold current for vibrissal movement decreased to 20/~A. Stimulation of an adjacent penetration site located approximately 400/zm from the lesion at nearly the same cortical depth also evoked whisker movement with a threshold intensity of 30/~A. This electrode track was clearly in host cortex. In a second series of experiments, we attempted to determine the electrophysiological properties of pyramidal tract neurons located within the occipital-torostral transplants. Five animals, aged 3 months to 1 year, which had received transplant~ as newborn pups were anesthetized and surgically prepared as previously outlined. After exposure of the cortex a stimulating microelectrode was implar, ted into the pyramidal tract with stereotactic coordinates ~7 used as a guide. To ensure that the electrode was in the correct position single pulses of 0.2 ms duration were delivered to the pyramidal tract while surface recordings were made at various locations around what appeared to be normal motor cortex. Response latenc~es and stimulus thresholds were noted and corn-
A
Fig. 1. A: a tracingof ihe section containing the lesion which marks the location of the stimulating electrode in the pyramidal tract. The electrode penetrated the pyramidal tract at the level of the pyramidal decussation. The hatched area indicates the extent of the damage. Abbreviations: CC, central canal; Cu, cuneate nocleus; cu, cuneate fasciculus; gr, gracile fasciculus; py, pyramidal tract; pyx, pyramidal decussation; Sp5, nucleus spinal tract trigeminal nerve; sp5, spinal tract trigeminal nerve; 12, hypoglossal nucleus. B: a photomicrograph through a portion of the section from which the drawing in A was made. This shows the lesion (open arrows) at higher magnification. Scale bar in B, 250~m.
139 pared to previous recordings of cortical surface potentials following PT stimulation 4"16. Then a tungsten-in-glass recording microelectrode was lowered into the area thought to contain the cortical transplant. When a single unit was encountered which responded to stimulation of the pyramidal tract we ensured that the response was antidromic by testing the latency, the ability to follow high frequency stimulation with constant latency and the ability to maintain constant latency to stimulation 2-3 times threshold current ~. Lesions were made at the site of antidromic responses and at the end of the experiment a lesion was made through the stimulating electrode in the pyramidal tract (Fig. 1). Subsequent histological analysis showed that in one animal lesions marking the location of 2 cells which exhibited antidromic action potentials were clearly within the boundaries of the transplant. In both cases numerous autoradiographically labeled cells surrounded the lesions (Fig. 2). These 2 cells responded with latencies of 2.0 and 2.5 ms which is within the expected limits of PT axon conduction ve-
locity9"~6. Furthermore, they clearly met the criteria outlined above for demonstrating antidromic responses to PT stinmlation (Fig. 3). The depth at which we encountered these cells was inconsistent with that for PT cells in normal tissue 8'24 but cortical lamination was seldom retained in the grafts. Our experiments reveal some interesting features of the physiological characteristics of the occipitai-torostral cortex transplants. As already stated, the level of unit activity recorded by electrodes which penetrated the transplanted tissue was noticeably reduced compared to penetrations through host tissue. In fact, during the course of our experiments in the second part of this study, this difference could be used fairly reliably in locating the transplant. Several factors might account for the decreased level of unit activity in the transplanted cortex. One possibility is that the core of the graft may receive sparse II or patchy 5 innervation from host afferents. The resultant paucity of synaptic input might reduce unit activity. Golgi studies of cortical implants reveal unusual variations in position and orientation of somata, and
Fig. 2. A: a low-powerbright-field photomicrograph of a Nissl-stainedsection through a transplant in which PT cells were found. The arrowheads roughly outline the transplanted tissue as determined autoradiographically. B: a higher powered dark-field photomicrograph of a part of the transplant shown in A. The open arrows in A and B indicate the site of the marking lesion. The marking lesion is well within the transplant in a region containingmany neurons overlayed by clusters of silver grains. Some of these are indicated by the filled arrows in B. Scalebar in A, 500~m; in B, 125/~m.
140 abnormalities in branching patterns of both dendrites and local axon distributions 7. These factors, along with the obvious disruption of cortical lamination might have disturbed the local interactions between cells within the transplant and resulted in depression of unit activity. Although numerous penetrations were made into the transplanted cortex, reliable motor responses
A
&
B
0
i Z e;
m~
t &
L,
A
°Smv°ltmsec Fig. 3. Tests for antidromic response to pyramidal tract stimulation. A: the responses to single stimuli of threshold intensity delivered to the pyramidal tract. This cell responded with a latency of 2.0 ms. B: the response of the same cell when two times the threshold intensity was delivered to the pyramidal tract. The latency remained nearly the same. C: the cell's response to high frequency stimulation. The cell fired in response to multiple stimuli with nearly constant latency. Arrowhead indicates PT stimulation.
could not be elicited from the core of the transplant, even though in the surrounding host cortex, cells retained their ability to produce movement. From previous anatomical work 2~ and from the abnormal depths at which we encountered PT cells, it was clear that there is a loss of proper orientation of the cortical laminae within the transplants and a disruption of the density and distribution of PT cells therein. Instead of a densely packed aggregation of PT neurons within layer V, the large pyramidal cells were scattered rather widely through the grey matter. Furthermore, while Nissl sections showed a large number of healthy neurons within the transplants, tracing studies reveal fewer PT cells in the transplant than in the neighboring host cortex 5'21. Thus, when ICM5 was applied to a limited area of transplanted cortex the number of activated PT cells may have been insufficient to produce movement. In addition, the density or distribution of the synaptic terminals formed by the pyrami~ dal tract axons arising in the transplants is not known, When the occipital transplants ~rt. first placed in the rostral cortex on the day of bii~th, s o ~ e PT neurons in the adjacent host tissue have already cx~ended axons as far as the pyramidal decussation 3"19. By the time PT axons from the transplant reach the spinal cord, they may be at such a competitive disadvantage that the number and distribution of synaptic contacts they are able to form may be relatively small. Similarly~ since PT cells appear to be somewhat scattered in the transplant their axons may not form a functional group with a common target; that is, their terminal branches may be widely dispersed and stimulation would affect only a sparse number of synaptic contacts at any one spinal level, and in insufficient number to produce movement. While the number of synaptic contacts formed by transplanted occipital pyramidal tract cells may be small, some apparently remain active because, as we have shown, their axons are maintained into adulthood. The PT cells which maintain these fibers might account for the 2 cases where movement was elicited by stimulation within the transplant. However, we cannot rule out with certainty the involvement of the host cortex in these cases. Neurons within an approximate radius of 150 pm from the electrode tip 2 are expected to be directly affected with the current used, and if synaptic spread is considered the distance is even greater. Since in both cases the lesions were
141 within approximately 200/~m of host tissue, d~'ect and synaptic spread of the stimulus to the host co~'tex could have contributed to, if not accounted fo~, the movement. As mentioned above, stimulation along many electrode penetrations limited to the host cortex surrounding the transplant resulted in movement. It is difficult to explain the finding that minimum current intensities needed to produce movement from penetrations made on the side of the transplant but clearly in the intact host cortex, were significantly higher than those for the control side of the same animal. For the most part, the normal cell packing density and the laminar organization appear to be maintained in the host tissue of the transplant side. It m a y be, that with the removal of a portion of the host cortex and with the relatively late arrival of PT axons from the transplant that the remaining host cortex develops a more widely dispersed terminal field than normal. These widely scattered synaptic contacts may be less effective in eliciting movement following ICMS. In this study, :we ~dentifiec! a small n u m b e r of neurons within occip~,~.~o~i-t~',-rostral transpiar~tg as PT
cells by antidromic stimulation of their axons. The identification of PT neurons within the heterotopic cortical transplants confirms the findings of retrograde tracing studies that fetal occipital PT neurons transplanted to a rostral cortical locale can extend and maintain projections characteristic of rostral cortex ~5"2~. F u r t h e r m o r e these occipital PT axons are maintained into adulthood. However, direct motor effects in response to microstimulation were not observed with certainty in these experiments. It seems then, that under these experimental conditions the occipital PT neurons are not able to fully express the motor cortical PT neuronal phenotype. On ,he other hand, the fact that these neurons are capable of responding to low threshold microstimulation of the pyramidal tract implies that their axons have the capacity to conduct impulses and are likely therefore, to have formed some appropriate connections with their postsynaptic targets.
1 Abzug, C. and Peterson. B., Antidromic stimulation in the pontomeduUary reticular formation of local axon branches in con~¢,~a~c~,d ~cstibular neurons, Brain Research, 64 ( i 977.).~i~7-413, 2 As;mq~r~ta, H., Stoney, S.D., Ir. and Abzug. C., Relatiottshi!? hz,tween afferent input and motor outflow in cat motorser~g~:rycortex, J. Neurophysiol., 31 (1968) 670-681. 3 DeMeyer, W., Ontogenesis of the rat corticospinal tract, Arch. Neurol., 16 (1967) 203-211. 4 Elger, C.E., Speckman, E.-J., Caspers, H. and Janzen, R.W.C., Cortico-spinal connections in the rat. I. Monosynaptic and polysynaptic responses of cervical motoneurons to epicorticai stimulation, Exp. Brain Res., 28 (1977) 385-404. 5 Floeter, M.K. and Jones, E.J., Connections made by transplants to the cerebral cortex of rat brains damaged in utero, J. Neurosci., 4 (1984) 141-150. 6 Hall, R.D. and Lindholm, E.P., Organization of motor and somatosensory neocortex in the albino rat, Brain Research, 66 (1974) 23-38. 7 Jaeger, C.B. and Lund, R.D., Transplantation of embryonic occipital cortex to the brain of newborn rats: a Golgi study of mature and developing transplants, J. Comp. Neurol., 200 (1981) 213-230. 8 Krieg, W.J.S., Connections of the cerebral cortex. I. The albino rat. B. Structure of the cortical areas, J. Comp. Neurol., 84 (1946) 277-324. 9 Landry, P., Labeile, A. and Deschenes, M., Intracortical distribution of axonal collaterals of pyramidal tract cells in the cat motor cortex, Brain Research, 191 (1980) 327-336. 10 Leong, S.K., Localizing the corticospina! ne-rons in neona-
tal, developing and mature albino rat, Brain Research, 265 (1983) 1-9. !1 Lund, R.D, and Hauschka, S.D., Transplanted neural tissue develops cot~nect~ons with host rat brain, Science, 193 (1976) 582-584. 12 Neafsey, E.J., Bold, E.L., Haas, G., t-']urley-Gius, K.M., Quirk, G., Sicv,ert, C.F. and Terreberry, R.R., The organization of the rat motor cortex: a microstimulation mapping study, Brain Res. Rev., 11 (1986) 77-96. 13 Neafsey, E.J. and Sievert, C., A second forelimb motor area exists in rat frontal cortex, Brain Research, 232 (1982) 151-156. 14 O'Leary, D.D.M. and Stanfield, B.B., Occipital cortical neurons with transient pyramidal tract axons extend and maintain collaterals to subcortical but not intracortical targets, Brain Research, 336 (1985) 326-333. 15 O'Leary, D.D.M. and Stanfield, B.B., Maintained projections of heterotopic transplants of fetal visual and sensorimotor cortex are characteristic of new cortical locale, Soc. Neurosci. Abstr., 12 (1986) 1049. 16 Ohta, M., Cortical motor function and the antidromic cortical response to stimulation of the medullary pyramidal tract in the rat,Jap. J. Physiol., 18 (1968) 100-124. 17 Paxinos, G. and Watson, C., The Rat Brain in Stereotct~°ic Coordinates, Academic, Sydney, 1982. 18 Sapienza, S., Talbi, B., Jacquemin, J. and Albe-Fessard, D., Relationship between input and output of cells in motor and somatosensory cortices of the chronic awake rat, Exp. Brain. Res, 43 (1981) 47-56. 19 Schreyer, D.J. and Jones. E.G., Growth and target finding by axons of the corticospinai tract in prenatal and postnatal
This work was supported by N I H Grants NS18065, EY-03653 and NS-10705. L.L.P. was an N I H Postdoctoral Fellow, F32 NS-07713.
142 rats, Neuroscience, 7 (1982) 1837-1853. 20 Stanfield, B.B., Postnatal reorganization of cortical projections: The role of collateral elimination, Trends Neurosci., 7 (1984) 37-41. 21 St,,~lfield, B.B. and O'Leary, D.D.M., Fetal occipital cortical n,:,-urones transplanted to the rostral cortex can extend ar.,i maintain a pyramidal tract axon, Nature (Lond.), 313 (1985) 135-137. 22 Stanfield, B.B., O'Leary, D.D.M. and Fricks, C., Selective collateral elimination in early postnatal development re-
stricts cortical distribution of rat pyramidal tract neurones, Nature (Lond.), 298 (1982) 371-373. 23 Stanfield, B.B. and O'Leary, D.M.M., The transient corticospinal projection from the occipital cortex during the postnatal development of the rat, J. Cornp. Neurol., 238 (1985) 236-248. 24 Zilles, K., Zilles, B. and Schleicher, A., A quantitative approach to cytoarchitectonics VI. The areal pattern of the cortex of the albino rat, Anat. Embryol., 159 (1980) 335-360.