- Email: [email protected]
Heterotopically transplanted embryonic olfactory bulb projection neurons form selective and appropriate axonal projections over considerable distances in adult host brains
347
Neuroscience Research, 5 (1988) 347-352 Elsevier Scientific Publishers Ireland Ltd. NSR 00223
Short Communications
Heterotopically transplanted embryonic olfactory bulb projection neurons form selective and appropriate axonal projections over considerable distances in adult host brains Masako Fujii First Department of Anatomy, Hamamatsu University School of Medicine, Handacho, Hamamatsu, Shizuoka (Japan) (Received 28 August 1987; Revised version received 15 October 1987; Accepted 17 November 1987)
Key words:Transplantation; Olfactory bulb; Anterior olfactory nucleus; Wheat germ agglutinin-horseradish peroxidase conjugate (WGA-HRP) J
SUMMARY Embryonic olfactory primordia were transplanted into the region of the septum, the adjacent lateral ventricle (LV) and olfactory tubercle (Tu) in adult host rats. Alter a minimum of 7 weeks, wheat germ aggiutinin-horseradish peroxidase conjugate (WGA-HRP) were injected into the host anterior olfactory nucleus (AO). The injection retrogradely labeled the neural cell bodies in the large neuron area (the mitral and tufted cells) of the olfactory bulb (OB) transplant. In addition, the anterogradely labeled fibers projected from the host AO to the transplant. These results indicated that the transplanted mitral and tulted neurons were able to grow axons selectively to an appropriate host terminal region (the AO) and receive fibers from the AO, even when the transplant itself was in an inappropriate host site, at a considerable distance from the host AO.
It has recently been thought that transplantation of brain tissue is a valuable tool for the study of neuroplasticity of the central nervous system. By applying this technique, the present study attempts to demonstrate the neural organization of the olfactory system. Wistar albino rats were used for both donors and recipients. Olfactory bulb (OB)* primordia (easily detectable by their protruding shape) were dissected from rat brains * Abbreviations in this paper are mostly taken from the atlas of Paxinos and Watson 5.
Correspondence: M. Fujii, First Department of Anatomy, Hamamatsu University School of Medicine, 3600 Handacho, Hamamatsu, Shizuoka 431-31, Japan. 0168-0102/88/$03.50 © 1988 Elsevier Scientific Publishers Ireland Ltd.
348 on the 18th day of embryonic life. The tissue was then stereotaxically inserted using a plastic tube, 0.8 mm in diameter, into the lateral ventricle (LV), the olfactory tubercle (Tu) or the septal area. Under ketamine anesthesia (40 mg/kg body weight, intramuscularly) 23 young female adult rats (100-150g body weight) received transplants bilaterally. After 50-277 days of survival (transplants developed bilaterally in 6 animals and unilaterally in 12), the animals were given bilateral injections of wheat germ agglutinin and horseradish peroxidase conjugate (WGA-HRP, Sigma; 10~o in sterile water, 0.07-0.1/11) in their OBs under the same anesthetic procedure. Forty to 48 hours after the W G A - H R P injections, under deep Nembutal anesthesia, the animals were fixed by intracardiac perfusion of 2~o paraformaldehyde and 1.25~o glutaraldehyde mixture in 0.1 M phosphate buffer, followed by 10~o sucrose in the same buffer. Brains were removed and frozen in dry-ice, and coronal serial sections were cut at thicknesses of 25-30 #m. On each of the animals, 3 or 4 series of section were processed for the following: (a)the tetramethylbenzidine method (TMB) 4 with fixation 2 usually with Neutral red counterstain for neural connection; (b)the Koelle method for acetylcholinesterase histochemistry (ACHE) ~; and (c) Cresyl violet for cytoarchitecture.
Fig. 1. A coronal section through an OB transplant (T) in rat no. 79, 231 days after transplantation. Cresylviolet stain. Note the segregationof the two neural groups. Abbreviations:lo, lateral olfactorytract; LPO and MPO, lateral and medial preoptic areas; ox, chiasma opticus. For other abbreviations,see text. Bars: A, 250/~m: B, 1 mm.
349 In the OB transplant, the normal organization of concentric layers of periglomerular, mitral and granular neurons was disrupted. Instead, two neuron groups were present, one composed of the large mitral and tufted neurons and the other of the small, compactly packed granular neurons (Fig. 1). In restricted parts of the transplants, there were also areas which contained small neurons, scattered, but at times, clustered (Fig. 3A). These areas were always found between the granular and large neuron areas and have received massive host AChE fibers (Fig. 3B), though the granular and large neuron areas were almost free of AChE fibers. Since ChAT fibers were known to enter into the periglomerular region in the normal OB 3, it is likely that the host AChE fibers were projecting into a comparable region in the transplants. Following the WGA-HRP injections into the host AO, labeled neurons and fibers (retrograde and anterograde WGA-HRP transports) appeared in the OB transplants. Most injections also involved the ventral part of the tenia tecta (Tr), the most anterior part of the primary olfactory cortex (PO). At times, both the TT and PO were involved, but the extent of WGA-HRP diffusion in to the T r and PO was variable and did not
Fig. 2. Low-power view of a coronal section through an OB transplant (T) in rat no. 66, 206 days survival. Cresyl violet stain. Abbreviation: LS, lateral septal nucleus. For other abbreviations, see text. Bar, 1 ram.
350 show any correlation with retrograde or anterograde labeling in the transplants. In cases where the WGA-HRP injections were restricted to the host OB, the transplant OB neurons were not labeled. Fig. 2 shows a transplant in the basal region of the most anterior part of the LV, lateral to the septum. The ventral part of the transplant was encapsulated by gliotic cells, but dorsally, the transplant neuropil was fused with the host neuropil, medially and laterally. The WGA-HRP injection in the right OB of this animal caused almost a complete diffusion to the AO, to the most anterior part of the PO, and to the ventral part of the TT (Fig. 3E). Both the anterograde and retrograde transports were clearly
Fig. 3. High-power view of the T in Fig. 2 and W G A - H R P injection site. A: Cresyl violet stain. B: ACHEstained section adjacent to A. Arrowheads in A and B show a small neuron cluster in the dense AChE fibers which seem to be periglomerular neurons in the normal OB. C: section adjacent to B, showing W G A - H R P transport in large neuron area. TMB stain. Three arrowheads show the same labeled neurons as in D. D: enlarged picture of 3 neurons in C. E: W G A - H R P in and around the AO. T M B stain. For abbreviations, see text. Bars: A, 100 #m; B and C are the same magnification as A, but there has been slight shrinkage during processing; D, 50 # m ; E, 1 mm.
351 found in the large neuron area of the transplant. Numerous large neurons in that part received WGA-HRP transport (Fig. 3C,D). In this section illustrated in Fig. 3C, 24 retrogradely labeled neurons were detected in the transplant. Labeled fibers (anterograde WGA-HRP transport) in this case scattered diffusely among the large neurons and in the AChE fiber region (periglomerular neuron area). These fibers entered the transplant at its medial border after forming a dorsally running labeled fiber bundle along the medial limit of the accumbens nucleus (Acb) from the massive deposition of WGA-HRP in the host Tu. Further tracing was impossible, but these fibers may pass through the host Tu from an origin in the host AO. These fibers may also contain axons traveling in the reverse direction (retrograde WGA-HRP transport) from the labeled somata in the transplant. These labeled fibers which go in and come out of the transplant appeared to take this long way round through the dorsomedial fusion, whereas AChE fibers apparently enter through the bottom of the transplant (Fig. 3B). Mitral and tufted neurons, in two well-developed transplants which fused medially with the host lateral septum in the anterior part of the host LV and in another transplant growing within the host septum, also showed numerous reciprocal connections with the host AO. In a case of the transplant in host LV, the host septum became thinner and in the other transplants, the host Acb together with either the anterior commissura (ac), the host septum, or the nucleus of the horizontal limb of the diagonal band of Broca (HBD) were destroyed. Three other small transplants in the same location of the host LV showed numerous labeled neurons without any damage to the adjacent host tissue or with only partial engulfment into the host Acb. These labeled neurons may send their axons to the host AO through a shorter route, through the ventral part of the host lateral septum and Tu. Two transplants, which fused laterally with the host caudoputamen (CPu) and contacted medially the host septum at a densely gliotic interface, received no detectable WGA-HRP transport. However, one of these had numerous WGA-I-IRP-containing fibers of host origin assembled in the host lateral septum, along the medial margin of the transplant**. The fibers may be prevented from entering the transplant by the glial barrier. Positive results were also obtained in the transplants growing in the host olfactory tubercle. In two small transplants, developing in the Tu, retrograde labeling appeared in the large neurons with few scattered fibers showing anterograde WGA-HRP transport. The WGA-HRP injections included almost the entire host AO. The other two small transplants showed no labeling in response to the WGA-HRP injection in the host AO. As far as the present material is concerned, transplants attached to the host septum or in the host septal region seemed to connect more effectively with the host AO than those in the host Tu. From the results mentioned above, it was clearly demonstrated that the secondary ** AO projection to the lateral septum has not been reported, but a study ofthe existence in normal animals is in progress in our laboratory.
352 olfactory neurons (mitral and tufted cells) in the transplants, growing in or fused with some host brain areas (the septum and Tu), both send fibers to the host AO. Reciprocally, OB transplants received fibers from the host AO. In normal animals, the AO is attached to the OB and has reciprocal connections with the OB 7. However, in this study, the distance between the host AO and OB transplants in the host LV was quite considerable (at least 2 mm apart). Various studies had emphasized the necessity for specific denervation and for direct contact in forming transplant-host projections 6,8. In certain cases of the present experiment, the insertion of transplants damaged afferent fibers to the host AO. However, reciprocal fiber connections between the OB transplant and the host AO were also clearly evident in the cases where there was no damage to the host AO afferent and efferents. Degeneration is not, therefore, essential for innervation from the OB transplant. The second olfactory neurons in the OB transplant may have potent abilities to detect and connect with the host AO. Clearly in the present experiment the transplanted mitral and tufted cells had the ability to grow to the host AO even when they were not denervated and when the transplants were located at inappropriate host brain areas not in direct contact with the host AO. Further studies will aim to clarify the mechanism of target recognition of the OB transplant. This system offers a proper model for the study because the target and the transplant can be separately examined.
ACKNOWLEDGEMENTS
The author thanks Dr. G. Raisman for his comments and critical reading of this manuscript and also Dr. J. Lawrence and Dr. N. Lenn for their discussions and correction of the English in the manuscript. REFERENCES 1 Butcher, L.L., Acetylcholinesterasehistochemistry. In A. BjOrklund and T. HOkfelt (Eds.), Handbook of Chemical Neuroanatomy, Vol. 1, Elsevier, Amsterdam, 1978, pp. 1-50. 2 Fujii, M. and Kusama, T., Fixation of horseradish peroxidase reaction product with ammonium molybdate, Neurosci. Res., 1 (1985) 65-68. 3 Halfisz, N. and Shepherd, G.M., Neurochemistry of the vertebrate olfactory bulb, Neuroscience, 10 (1-983) 579-618. 4 Mesulam, M.M., Tetramethylbenzidine for horseradish peroxidase histochemistry: a non-carcinogenic blue reaction productwith superior sensitivityfor visualizingneural afferents and efferents,J. Histochem. Cytochem., 26 (1978) 106-117. 5 Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic, New York, 1982. 6 Raisman, G. and Ebner, F.F., Mossy fibre projections into and out of hippocampal transplants, Neuroscience, 9 (1983) 783-801. 7 Switzer,R.C., De Olmos,J. and Heimer, L., The olfactorysystem. In G. Paxinos (Ed.), The RatNervous System, Vol. 1, Academic Press, New York, 1985, pp. 1-36. 8 Zhou, C.-F., Raisman, G. and Morris, R.J., Specificpatterns of fibre outgrowthfromtransplants to host mice hippocampi, shown immunohistochemicallyby the use of allelic forms of Thy-1, Neuroscience, 16 (1985) 819-833.
Neuroscience Research, 5 (1988) 347-352 Elsevier Scientific Publishers Ireland Ltd. NSR 00223
Short Communications
Heterotopically transplanted embryonic olfactory bulb projection neurons form selective and appropriate axonal projections over considerable distances in adult host brains Masako Fujii First Department of Anatomy, Hamamatsu University School of Medicine, Handacho, Hamamatsu, Shizuoka (Japan) (Received 28 August 1987; Revised version received 15 October 1987; Accepted 17 November 1987)
Key words:Transplantation; Olfactory bulb; Anterior olfactory nucleus; Wheat germ agglutinin-horseradish peroxidase conjugate (WGA-HRP) J
SUMMARY Embryonic olfactory primordia were transplanted into the region of the septum, the adjacent lateral ventricle (LV) and olfactory tubercle (Tu) in adult host rats. Alter a minimum of 7 weeks, wheat germ aggiutinin-horseradish peroxidase conjugate (WGA-HRP) were injected into the host anterior olfactory nucleus (AO). The injection retrogradely labeled the neural cell bodies in the large neuron area (the mitral and tufted cells) of the olfactory bulb (OB) transplant. In addition, the anterogradely labeled fibers projected from the host AO to the transplant. These results indicated that the transplanted mitral and tulted neurons were able to grow axons selectively to an appropriate host terminal region (the AO) and receive fibers from the AO, even when the transplant itself was in an inappropriate host site, at a considerable distance from the host AO.
It has recently been thought that transplantation of brain tissue is a valuable tool for the study of neuroplasticity of the central nervous system. By applying this technique, the present study attempts to demonstrate the neural organization of the olfactory system. Wistar albino rats were used for both donors and recipients. Olfactory bulb (OB)* primordia (easily detectable by their protruding shape) were dissected from rat brains * Abbreviations in this paper are mostly taken from the atlas of Paxinos and Watson 5.
Correspondence: M. Fujii, First Department of Anatomy, Hamamatsu University School of Medicine, 3600 Handacho, Hamamatsu, Shizuoka 431-31, Japan. 0168-0102/88/$03.50 © 1988 Elsevier Scientific Publishers Ireland Ltd.
348 on the 18th day of embryonic life. The tissue was then stereotaxically inserted using a plastic tube, 0.8 mm in diameter, into the lateral ventricle (LV), the olfactory tubercle (Tu) or the septal area. Under ketamine anesthesia (40 mg/kg body weight, intramuscularly) 23 young female adult rats (100-150g body weight) received transplants bilaterally. After 50-277 days of survival (transplants developed bilaterally in 6 animals and unilaterally in 12), the animals were given bilateral injections of wheat germ agglutinin and horseradish peroxidase conjugate (WGA-HRP, Sigma; 10~o in sterile water, 0.07-0.1/11) in their OBs under the same anesthetic procedure. Forty to 48 hours after the W G A - H R P injections, under deep Nembutal anesthesia, the animals were fixed by intracardiac perfusion of 2~o paraformaldehyde and 1.25~o glutaraldehyde mixture in 0.1 M phosphate buffer, followed by 10~o sucrose in the same buffer. Brains were removed and frozen in dry-ice, and coronal serial sections were cut at thicknesses of 25-30 #m. On each of the animals, 3 or 4 series of section were processed for the following: (a)the tetramethylbenzidine method (TMB) 4 with fixation 2 usually with Neutral red counterstain for neural connection; (b)the Koelle method for acetylcholinesterase histochemistry (ACHE) ~; and (c) Cresyl violet for cytoarchitecture.
Fig. 1. A coronal section through an OB transplant (T) in rat no. 79, 231 days after transplantation. Cresylviolet stain. Note the segregationof the two neural groups. Abbreviations:lo, lateral olfactorytract; LPO and MPO, lateral and medial preoptic areas; ox, chiasma opticus. For other abbreviations,see text. Bars: A, 250/~m: B, 1 mm.
349 In the OB transplant, the normal organization of concentric layers of periglomerular, mitral and granular neurons was disrupted. Instead, two neuron groups were present, one composed of the large mitral and tufted neurons and the other of the small, compactly packed granular neurons (Fig. 1). In restricted parts of the transplants, there were also areas which contained small neurons, scattered, but at times, clustered (Fig. 3A). These areas were always found between the granular and large neuron areas and have received massive host AChE fibers (Fig. 3B), though the granular and large neuron areas were almost free of AChE fibers. Since ChAT fibers were known to enter into the periglomerular region in the normal OB 3, it is likely that the host AChE fibers were projecting into a comparable region in the transplants. Following the WGA-HRP injections into the host AO, labeled neurons and fibers (retrograde and anterograde WGA-HRP transports) appeared in the OB transplants. Most injections also involved the ventral part of the tenia tecta (Tr), the most anterior part of the primary olfactory cortex (PO). At times, both the TT and PO were involved, but the extent of WGA-HRP diffusion in to the T r and PO was variable and did not
Fig. 2. Low-power view of a coronal section through an OB transplant (T) in rat no. 66, 206 days survival. Cresyl violet stain. Abbreviation: LS, lateral septal nucleus. For other abbreviations, see text. Bar, 1 ram.
350 show any correlation with retrograde or anterograde labeling in the transplants. In cases where the WGA-HRP injections were restricted to the host OB, the transplant OB neurons were not labeled. Fig. 2 shows a transplant in the basal region of the most anterior part of the LV, lateral to the septum. The ventral part of the transplant was encapsulated by gliotic cells, but dorsally, the transplant neuropil was fused with the host neuropil, medially and laterally. The WGA-HRP injection in the right OB of this animal caused almost a complete diffusion to the AO, to the most anterior part of the PO, and to the ventral part of the TT (Fig. 3E). Both the anterograde and retrograde transports were clearly
Fig. 3. High-power view of the T in Fig. 2 and W G A - H R P injection site. A: Cresyl violet stain. B: ACHEstained section adjacent to A. Arrowheads in A and B show a small neuron cluster in the dense AChE fibers which seem to be periglomerular neurons in the normal OB. C: section adjacent to B, showing W G A - H R P transport in large neuron area. TMB stain. Three arrowheads show the same labeled neurons as in D. D: enlarged picture of 3 neurons in C. E: W G A - H R P in and around the AO. T M B stain. For abbreviations, see text. Bars: A, 100 #m; B and C are the same magnification as A, but there has been slight shrinkage during processing; D, 50 # m ; E, 1 mm.
351 found in the large neuron area of the transplant. Numerous large neurons in that part received WGA-HRP transport (Fig. 3C,D). In this section illustrated in Fig. 3C, 24 retrogradely labeled neurons were detected in the transplant. Labeled fibers (anterograde WGA-HRP transport) in this case scattered diffusely among the large neurons and in the AChE fiber region (periglomerular neuron area). These fibers entered the transplant at its medial border after forming a dorsally running labeled fiber bundle along the medial limit of the accumbens nucleus (Acb) from the massive deposition of WGA-HRP in the host Tu. Further tracing was impossible, but these fibers may pass through the host Tu from an origin in the host AO. These fibers may also contain axons traveling in the reverse direction (retrograde WGA-HRP transport) from the labeled somata in the transplant. These labeled fibers which go in and come out of the transplant appeared to take this long way round through the dorsomedial fusion, whereas AChE fibers apparently enter through the bottom of the transplant (Fig. 3B). Mitral and tufted neurons, in two well-developed transplants which fused medially with the host lateral septum in the anterior part of the host LV and in another transplant growing within the host septum, also showed numerous reciprocal connections with the host AO. In a case of the transplant in host LV, the host septum became thinner and in the other transplants, the host Acb together with either the anterior commissura (ac), the host septum, or the nucleus of the horizontal limb of the diagonal band of Broca (HBD) were destroyed. Three other small transplants in the same location of the host LV showed numerous labeled neurons without any damage to the adjacent host tissue or with only partial engulfment into the host Acb. These labeled neurons may send their axons to the host AO through a shorter route, through the ventral part of the host lateral septum and Tu. Two transplants, which fused laterally with the host caudoputamen (CPu) and contacted medially the host septum at a densely gliotic interface, received no detectable WGA-HRP transport. However, one of these had numerous WGA-I-IRP-containing fibers of host origin assembled in the host lateral septum, along the medial margin of the transplant**. The fibers may be prevented from entering the transplant by the glial barrier. Positive results were also obtained in the transplants growing in the host olfactory tubercle. In two small transplants, developing in the Tu, retrograde labeling appeared in the large neurons with few scattered fibers showing anterograde WGA-HRP transport. The WGA-HRP injections included almost the entire host AO. The other two small transplants showed no labeling in response to the WGA-HRP injection in the host AO. As far as the present material is concerned, transplants attached to the host septum or in the host septal region seemed to connect more effectively with the host AO than those in the host Tu. From the results mentioned above, it was clearly demonstrated that the secondary ** AO projection to the lateral septum has not been reported, but a study ofthe existence in normal animals is in progress in our laboratory.
352 olfactory neurons (mitral and tufted cells) in the transplants, growing in or fused with some host brain areas (the septum and Tu), both send fibers to the host AO. Reciprocally, OB transplants received fibers from the host AO. In normal animals, the AO is attached to the OB and has reciprocal connections with the OB 7. However, in this study, the distance between the host AO and OB transplants in the host LV was quite considerable (at least 2 mm apart). Various studies had emphasized the necessity for specific denervation and for direct contact in forming transplant-host projections 6,8. In certain cases of the present experiment, the insertion of transplants damaged afferent fibers to the host AO. However, reciprocal fiber connections between the OB transplant and the host AO were also clearly evident in the cases where there was no damage to the host AO afferent and efferents. Degeneration is not, therefore, essential for innervation from the OB transplant. The second olfactory neurons in the OB transplant may have potent abilities to detect and connect with the host AO. Clearly in the present experiment the transplanted mitral and tufted cells had the ability to grow to the host AO even when they were not denervated and when the transplants were located at inappropriate host brain areas not in direct contact with the host AO. Further studies will aim to clarify the mechanism of target recognition of the OB transplant. This system offers a proper model for the study because the target and the transplant can be separately examined.
ACKNOWLEDGEMENTS
The author thanks Dr. G. Raisman for his comments and critical reading of this manuscript and also Dr. J. Lawrence and Dr. N. Lenn for their discussions and correction of the English in the manuscript. REFERENCES 1 Butcher, L.L., Acetylcholinesterasehistochemistry. In A. BjOrklund and T. HOkfelt (Eds.), Handbook of Chemical Neuroanatomy, Vol. 1, Elsevier, Amsterdam, 1978, pp. 1-50. 2 Fujii, M. and Kusama, T., Fixation of horseradish peroxidase reaction product with ammonium molybdate, Neurosci. Res., 1 (1985) 65-68. 3 Halfisz, N. and Shepherd, G.M., Neurochemistry of the vertebrate olfactory bulb, Neuroscience, 10 (1-983) 579-618. 4 Mesulam, M.M., Tetramethylbenzidine for horseradish peroxidase histochemistry: a non-carcinogenic blue reaction productwith superior sensitivityfor visualizingneural afferents and efferents,J. Histochem. Cytochem., 26 (1978) 106-117. 5 Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic, New York, 1982. 6 Raisman, G. and Ebner, F.F., Mossy fibre projections into and out of hippocampal transplants, Neuroscience, 9 (1983) 783-801. 7 Switzer,R.C., De Olmos,J. and Heimer, L., The olfactorysystem. In G. Paxinos (Ed.), The RatNervous System, Vol. 1, Academic Press, New York, 1985, pp. 1-36. 8 Zhou, C.-F., Raisman, G. and Morris, R.J., Specificpatterns of fibre outgrowthfromtransplants to host mice hippocampi, shown immunohistochemicallyby the use of allelic forms of Thy-1, Neuroscience, 16 (1985) 819-833.