- Email: [email protected]
Analysis of collateral projections with a double retrograde labeling technique
N e u r o s c i eLetters, n c e 5 (1977)1--5
Elsevier/North-Holland Scientific Publishers Ltd.
ANALYSIS OF COLLATERAL PROJECTIONS RETROGRADE LABELING TECHNIQUE
WITH A DOUBLE
O. STEWARD, S.A. SCOVILLE and S.L. VINSANT Departments of Neurosurgery and Physiology, University of Virginia School c.f Medicine, Charlottesville, Va. 22901 (U.S.A.) (Received March 7th, 1977) (Accepted March 21st, 1977)
SUMMARY
A method is described whereby collateralinnervation by a single cell type may be determined by double retrograde labeling with horseradish peroxidase (HRP) and tritiatedbovine serum a~bumin ([SH] BSA). The projections from the entorhinal area to the hippocampal formation were analyzed as a model system, since layer Ill pyramidal cellsproject bilaterallyto the hippocarnpus. Following H R P injections into the hippocampus of one side, and [3H] B S A injections into the opposite hippocampus, double labeled cellswere found in layer Ill of the entorhinal cortex. These doubly labeled cellscored best be viewed in plasticembedded material, whereas frozen sections were found to be inadequate for clearly distinguishingbetween silvergrains and H R P reaction product.
The development and refine~ent of the horseradish peroxidase (HRP) technique for the retrograde tracing of central nervous pathways has resulted dramatic advances in neuroanatomical knowledge in recent years, since the specific cell type which gives rise to a given neuronal pathway can be ascertained. Such information has proved to be of considerable value, since different cell types within a population (even within a single cortical layer)can project differentially [ 1]. A question which has arisen in several investigations, however, is whether a s~ng]~ cell type gives rise to collateral projections whic~ terminate in different b.~in ~gions. The present report describes a technique whereby single cells can be doubly labeled, following injections of different substances into the region of termination of two collaterals of a given neuronal population. For the present study, we chose as a model system the pathway from the entorhinal area of the rat to the hippocarnp~i ~o~nation. This system has been described in detail elsewhere [6], and is useful as a model system for the study of collateralprojections since one cellpopulation in the entorhinal area
O
(the medium sized pyramidal cells in layer III) project bilaterally to regio superior of the hippocampal formation. For the double labeling studies, the HRP histochemical method was com. bined with a technique which we have recently developed for the retrograde labeling of centralnervous pathways with tritiated bovine serum albumin ([3HI BSA) ~[ 7 ] , This imate~ial was obtained from N e w England Nuclear, and was prepared by exposing 500 mg of BSA (Sigma) to 3 C i of 3H gas for 3 weeks. This exposure ~sulted in a specific activity of approximately 0.1 mCijlmg. The [3H] BSA was stored frozen in a dry form, and for the injection was suspended in 500/~1 of distilled H20. In the initial studies, this method of preparation did not result in a complete dissolution of the [3H] BSA, since some of the material tended to remain in small particles, Injections of this materiai were, however, found to be quite effective for retrograde labeling. In later experiments, it was found that dissolution could be dramatically improved by first dissolving the [3H] BSA in 3 ml of H20, followed by lyophilization (E. Geisert, personal communication), and subsequently re-dissolving the material in the desired quantiW of distilled water. For the present study, rats with unilateral entorhinal cortical lesions were utilized, since the transport oli!HRP to cells of layer III folllowing injections into the contralateral hippocampal formation has been found more efficacious following such lesions [5]. Thus, 60 days post-lesion, 0.5/~1 of 50% HRP (Sigma) was injected into the hippocampal formation contralateral to the surviving entorhinal cortex, and 1/~1 (10Ci) of [3H] BSA was injected into a similar location ipsilaterally, while the animals were anesthetized with sodium pentobarbital. Two days following the injections, 4 animals were perfused with 10% formalin in saline. The brains were removed, post-fixed in the perfusion solution for 1--3 h, and sectioned at 40 ~m on a freezing microtome. Sections were reacted in an HRP histochemical medium with 3,3'-diaminobenzidine, as pre .viously described [6], and subsequently were mounted on microscope slides, ~iefatted, and coated with I~TB-2 emulsion for autoradiography. The automdiographic preparations were exposed for 4--8 weeks, developed in Kodak D-19, and counterstained with cresyl violet. One additional animal was doubly injected, as described above, and was perfused with 2% paraformaldehyde/2% glutaraldehyde in 0.13 M sodium cacodylate buffer (pH 7.2). The brain was removed, post-fixed in the perfusion solution for 1 h, and 250--400/~m thick sections through the entorhinal area were hand-cut with a razor blade. These slices were rinsed in the cacodylate buffer for 1 h, and were then incubated for 45 rain in 30 ml of Tris buffer, with 15 mg of 3,3'-diaminobenzidine and 0.3/~1 of 3% hydrogen peroxide. These thick sections were rinsed briefly in buffer, osmicated for 45 rain, dehydrated through a graded alcohol series, and embedded in Epon-Araldite. Semi-thin (1 ~m) sections were cut on an ultramicrotome, the plastic was removed, and t h e slices were coated with Kodak NTB-2 emulsion. After 4--8 weeks, the slides were developed in Kodak D-19, lightly stained with cresyl violet, dehydrated through xylene, and covered.
The rostral portion of the brain (which contained the portion of the hippocampal formation which had received the injection) was sectioned on a freezing microtome at 40 #m. The sections were washed for 2 h in Tris buffer and the combined histochemical/autoradiographic procedure was then carried out, as described for the other cases which were frozen sectioned. For the case prepared by plastic embedding, the injections of both the [3H]BSA and the HRP resulted in the labeling of the entorhinal terminal fields in both the fascia dentata and regio superior of the hippocampus proper. As would be expected from previous investigatiohs [ 6], cells in layer III were labeled by both the HRP and the [3H] BSA. Fig. I illustrates several cells in layer III of the entorhinal area as they appear utilizing a Zeiss 100 × objective. When viewed with bright field illumination (Fig. 1A), 4 cells are observed with granular labeling, with a few grains scattered over the field which cannot be associated with a neuronal cell body. By illuminating the preparation with an oil immersion dark-field condenser, and closing the aperture on the 100 × objective, a dark field ~mage is obtained in which all the grains (which represent both silver grains and HRP reaction product) reflect light (Fig. 1B). In either bright- or dark-field, it is possible to distinguish bet~veen silver grains and HRP reaction product by the difference in focal plane in the microscope. This difference is difficult to document photographically, however. For this purpose, the preparation was illuminated with an oil immersion dark-field condenser, and the variable aperture was opened on the 100 × objective. As the aperture is opened, there is a gradual transition from a dark-field to a bright~field image. Approximately midway between dark-field and bright.field image, particles in the plane of focus appear as they would in dark-field (i.e. they reflect light), while panicles out of the plane of focus absorb light. Thus, by focusing on the autoradiographic grains, the distinction between autoradiographic grains and HRP reaction product can be documented. As is evident in Fig. 1C, all of the cells in the field which contain the HRP reaction product (see Fig. 1A and B) are also covered with silver grains (Fig. 1C). This double labeling with HRP and [3H] BSA strongly suggests that these cells project via collaterals to both the ipsilateral and contralateral hippocampal formation. In the 4 cases prepared.by frozen sectioning, cells labeled with either HRP or [SH] BSA were quite evident. In addition, there were some cells which appeared to be doubly labeled. Distinguishing between the silver grains and the HRP reaction product was difficult, however, since the surface of the frozen sections appeared irregular, interfering with distinctions based on differences in the focal plane. In addition, some cells could also be found which were labeled with HRP, but not autoradiographically, possibly because these cells were too deep in the section and thus too far from the section-emulsion interface to expose the emulsior. There are considerable advantages of the double retrograde labeling technique described h e z ~ in comparison with the one other method which has been developed utiliz.ing [3H] HRP in combination with enzymatically active HRP [2]. ~pecifically, the double labeling technique which utilizes [SH]HRP
Ij,i ~'~ ~.~.i m,~ ,,"''~ ~,~.o ~ ~® ~
w
" ~"
,~
~
~ ~ ~ o~~~ ~.~. .~ ~
~• .. g, ~'~ ~ •~ , ~
~
0
r~quires a documented absence of enzyme activity by the tritiated protein. At a minimum, this requirement makes it necessary to carefully assay each batch of [3HI HRP for enzymatic activity prior to use in a double labeling study. This disadvantage is of course not shared by the [3H] BSA method, since BSA is enzymatically inactive in the HRP histcchemical procedure. Among the technical notes which we feel are important with respect to double labeling studies, the use of plastic embedded sections appears of prime importance. First, the diffuse background labeling which appears in frozerL sections in areas which contain labeled cells is considerably less noticeable in the plastic sections. While the grains may not be found immediately over the cell body, the greater cytological detail afforded by plastic sections frequently makes it possible to associate grains with dendrites which originate from labeled cel~,s. In addition, the microscopic manipulation with the variable aperture oil immersion objective is not possible, at least in frozen sections of the thickness we have utilized. Finally, the use of plastic embedded material provic~es a flexibility for either light microscopy or electron microscopy. It is our hope that the double retrograde labeling technique~utilizing HRP and [3H] BSA will prove useful in studies of collateral innervation in a variety of neural systems. ACKNOWLEDGEMEI-~TS
Supported by USPHS Grant I R01 NS12333 to O. Steward. We thank E.E. Geisert for his considerable technical advice. REFERENCES 1 Catsman-Berrevoets, C.E. and Kuypers, H.G.J.M., Cells of origin of cortical projections to dorsal column nuclei, spinal cord, and bulbar medial retucular formation in the rhesus monkey~ Neuroscience Letters, 3 (1976) 245--252. 2 Geisert, E.E., Jr., The use of tritiated horseradish peroxidase for defining neuronal pathways: a new application., Brain Res., 117 (1976) 130--135. 3 LaVail, J.H., Winston, K.R. and Tish, A., A method based on retrograde intraaxonal transport of protein for identification of cell bodies of origin of axons termination within the CI~S, Brain Res., 58 (1973) 470--477. 4 Nauta, H.J.W., Pritz, M.B. and Lasek, R.J., Afferents to the rat caudato-putamen studied with horseradish peroxidase. An evaluation of a retrograde neuroanatomical research method, Brain Res., 67 (1974) 219--238. 5 Steward, O., Reinnervation of dentate gyrus by homologous afferents following entorhinal cortical lesions in adult rats, Science, 194 (1976) 426--428. 6 Steward, O. and Scoville, S.A., Cells of origin of entorhinal cortic'.d afferents to the hippocampus and fascia dentata of the rat, J. comp. Neurol., 169 (1976) 347--370. 7 Steward, O. and Scoville, S.A., Retrograde labeling of central nervous pat,hways with tritiated or Evans Blue-labeled bovine serum ~lbumin, Neuroscience Letters, 3 (1976) 191--196.
Elsevier/North-Holland Scientific Publishers Ltd.
ANALYSIS OF COLLATERAL PROJECTIONS RETROGRADE LABELING TECHNIQUE
WITH A DOUBLE
O. STEWARD, S.A. SCOVILLE and S.L. VINSANT Departments of Neurosurgery and Physiology, University of Virginia School c.f Medicine, Charlottesville, Va. 22901 (U.S.A.) (Received March 7th, 1977) (Accepted March 21st, 1977)
SUMMARY
A method is described whereby collateralinnervation by a single cell type may be determined by double retrograde labeling with horseradish peroxidase (HRP) and tritiatedbovine serum a~bumin ([SH] BSA). The projections from the entorhinal area to the hippocampal formation were analyzed as a model system, since layer Ill pyramidal cellsproject bilaterallyto the hippocarnpus. Following H R P injections into the hippocampus of one side, and [3H] B S A injections into the opposite hippocampus, double labeled cellswere found in layer Ill of the entorhinal cortex. These doubly labeled cellscored best be viewed in plasticembedded material, whereas frozen sections were found to be inadequate for clearly distinguishingbetween silvergrains and H R P reaction product.
The development and refine~ent of the horseradish peroxidase (HRP) technique for the retrograde tracing of central nervous pathways has resulted dramatic advances in neuroanatomical knowledge in recent years, since the specific cell type which gives rise to a given neuronal pathway can be ascertained. Such information has proved to be of considerable value, since different cell types within a population (even within a single cortical layer)can project differentially [ 1]. A question which has arisen in several investigations, however, is whether a s~ng]~ cell type gives rise to collateral projections whic~ terminate in different b.~in ~gions. The present report describes a technique whereby single cells can be doubly labeled, following injections of different substances into the region of termination of two collaterals of a given neuronal population. For the present study, we chose as a model system the pathway from the entorhinal area of the rat to the hippocarnp~i ~o~nation. This system has been described in detail elsewhere [6], and is useful as a model system for the study of collateralprojections since one cellpopulation in the entorhinal area
O
(the medium sized pyramidal cells in layer III) project bilaterally to regio superior of the hippocampal formation. For the double labeling studies, the HRP histochemical method was com. bined with a technique which we have recently developed for the retrograde labeling of centralnervous pathways with tritiated bovine serum albumin ([3HI BSA) ~[ 7 ] , This imate~ial was obtained from N e w England Nuclear, and was prepared by exposing 500 mg of BSA (Sigma) to 3 C i of 3H gas for 3 weeks. This exposure ~sulted in a specific activity of approximately 0.1 mCijlmg. The [3H] BSA was stored frozen in a dry form, and for the injection was suspended in 500/~1 of distilled H20. In the initial studies, this method of preparation did not result in a complete dissolution of the [3H] BSA, since some of the material tended to remain in small particles, Injections of this materiai were, however, found to be quite effective for retrograde labeling. In later experiments, it was found that dissolution could be dramatically improved by first dissolving the [3H] BSA in 3 ml of H20, followed by lyophilization (E. Geisert, personal communication), and subsequently re-dissolving the material in the desired quantiW of distilled water. For the present study, rats with unilateral entorhinal cortical lesions were utilized, since the transport oli!HRP to cells of layer III folllowing injections into the contralateral hippocampal formation has been found more efficacious following such lesions [5]. Thus, 60 days post-lesion, 0.5/~1 of 50% HRP (Sigma) was injected into the hippocampal formation contralateral to the surviving entorhinal cortex, and 1/~1 (10Ci) of [3H] BSA was injected into a similar location ipsilaterally, while the animals were anesthetized with sodium pentobarbital. Two days following the injections, 4 animals were perfused with 10% formalin in saline. The brains were removed, post-fixed in the perfusion solution for 1--3 h, and sectioned at 40 ~m on a freezing microtome. Sections were reacted in an HRP histochemical medium with 3,3'-diaminobenzidine, as pre .viously described [6], and subsequently were mounted on microscope slides, ~iefatted, and coated with I~TB-2 emulsion for autoradiography. The automdiographic preparations were exposed for 4--8 weeks, developed in Kodak D-19, and counterstained with cresyl violet. One additional animal was doubly injected, as described above, and was perfused with 2% paraformaldehyde/2% glutaraldehyde in 0.13 M sodium cacodylate buffer (pH 7.2). The brain was removed, post-fixed in the perfusion solution for 1 h, and 250--400/~m thick sections through the entorhinal area were hand-cut with a razor blade. These slices were rinsed in the cacodylate buffer for 1 h, and were then incubated for 45 rain in 30 ml of Tris buffer, with 15 mg of 3,3'-diaminobenzidine and 0.3/~1 of 3% hydrogen peroxide. These thick sections were rinsed briefly in buffer, osmicated for 45 rain, dehydrated through a graded alcohol series, and embedded in Epon-Araldite. Semi-thin (1 ~m) sections were cut on an ultramicrotome, the plastic was removed, and t h e slices were coated with Kodak NTB-2 emulsion. After 4--8 weeks, the slides were developed in Kodak D-19, lightly stained with cresyl violet, dehydrated through xylene, and covered.
The rostral portion of the brain (which contained the portion of the hippocampal formation which had received the injection) was sectioned on a freezing microtome at 40 #m. The sections were washed for 2 h in Tris buffer and the combined histochemical/autoradiographic procedure was then carried out, as described for the other cases which were frozen sectioned. For the case prepared by plastic embedding, the injections of both the [3H]BSA and the HRP resulted in the labeling of the entorhinal terminal fields in both the fascia dentata and regio superior of the hippocampus proper. As would be expected from previous investigatiohs [ 6], cells in layer III were labeled by both the HRP and the [3H] BSA. Fig. I illustrates several cells in layer III of the entorhinal area as they appear utilizing a Zeiss 100 × objective. When viewed with bright field illumination (Fig. 1A), 4 cells are observed with granular labeling, with a few grains scattered over the field which cannot be associated with a neuronal cell body. By illuminating the preparation with an oil immersion dark-field condenser, and closing the aperture on the 100 × objective, a dark field ~mage is obtained in which all the grains (which represent both silver grains and HRP reaction product) reflect light (Fig. 1B). In either bright- or dark-field, it is possible to distinguish bet~veen silver grains and HRP reaction product by the difference in focal plane in the microscope. This difference is difficult to document photographically, however. For this purpose, the preparation was illuminated with an oil immersion dark-field condenser, and the variable aperture was opened on the 100 × objective. As the aperture is opened, there is a gradual transition from a dark-field to a bright~field image. Approximately midway between dark-field and bright.field image, particles in the plane of focus appear as they would in dark-field (i.e. they reflect light), while panicles out of the plane of focus absorb light. Thus, by focusing on the autoradiographic grains, the distinction between autoradiographic grains and HRP reaction product can be documented. As is evident in Fig. 1C, all of the cells in the field which contain the HRP reaction product (see Fig. 1A and B) are also covered with silver grains (Fig. 1C). This double labeling with HRP and [3H] BSA strongly suggests that these cells project via collaterals to both the ipsilateral and contralateral hippocampal formation. In the 4 cases prepared.by frozen sectioning, cells labeled with either HRP or [SH] BSA were quite evident. In addition, there were some cells which appeared to be doubly labeled. Distinguishing between the silver grains and the HRP reaction product was difficult, however, since the surface of the frozen sections appeared irregular, interfering with distinctions based on differences in the focal plane. In addition, some cells could also be found which were labeled with HRP, but not autoradiographically, possibly because these cells were too deep in the section and thus too far from the section-emulsion interface to expose the emulsior. There are considerable advantages of the double retrograde labeling technique described h e z ~ in comparison with the one other method which has been developed utiliz.ing [3H] HRP in combination with enzymatically active HRP [2]. ~pecifically, the double labeling technique which utilizes [SH]HRP
Ij,i ~'~ ~.~.i m,~ ,,"''~ ~,~.o ~ ~® ~
w
" ~"
,~
~
~ ~ ~ o~~~ ~.~. .~ ~
~• .. g, ~'~ ~ •~ , ~
~
0
r~quires a documented absence of enzyme activity by the tritiated protein. At a minimum, this requirement makes it necessary to carefully assay each batch of [3HI HRP for enzymatic activity prior to use in a double labeling study. This disadvantage is of course not shared by the [3H] BSA method, since BSA is enzymatically inactive in the HRP histcchemical procedure. Among the technical notes which we feel are important with respect to double labeling studies, the use of plastic embedded sections appears of prime importance. First, the diffuse background labeling which appears in frozerL sections in areas which contain labeled cells is considerably less noticeable in the plastic sections. While the grains may not be found immediately over the cell body, the greater cytological detail afforded by plastic sections frequently makes it possible to associate grains with dendrites which originate from labeled cel~,s. In addition, the microscopic manipulation with the variable aperture oil immersion objective is not possible, at least in frozen sections of the thickness we have utilized. Finally, the use of plastic embedded material provic~es a flexibility for either light microscopy or electron microscopy. It is our hope that the double retrograde labeling technique~utilizing HRP and [3H] BSA will prove useful in studies of collateral innervation in a variety of neural systems. ACKNOWLEDGEMEI-~TS
Supported by USPHS Grant I R01 NS12333 to O. Steward. We thank E.E. Geisert for his considerable technical advice. REFERENCES 1 Catsman-Berrevoets, C.E. and Kuypers, H.G.J.M., Cells of origin of cortical projections to dorsal column nuclei, spinal cord, and bulbar medial retucular formation in the rhesus monkey~ Neuroscience Letters, 3 (1976) 245--252. 2 Geisert, E.E., Jr., The use of tritiated horseradish peroxidase for defining neuronal pathways: a new application., Brain Res., 117 (1976) 130--135. 3 LaVail, J.H., Winston, K.R. and Tish, A., A method based on retrograde intraaxonal transport of protein for identification of cell bodies of origin of axons termination within the CI~S, Brain Res., 58 (1973) 470--477. 4 Nauta, H.J.W., Pritz, M.B. and Lasek, R.J., Afferents to the rat caudato-putamen studied with horseradish peroxidase. An evaluation of a retrograde neuroanatomical research method, Brain Res., 67 (1974) 219--238. 5 Steward, O., Reinnervation of dentate gyrus by homologous afferents following entorhinal cortical lesions in adult rats, Science, 194 (1976) 426--428. 6 Steward, O. and Scoville, S.A., Cells of origin of entorhinal cortic'.d afferents to the hippocampus and fascia dentata of the rat, J. comp. Neurol., 169 (1976) 347--370. 7 Steward, O. and Scoville, S.A., Retrograde labeling of central nervous pat,hways with tritiated or Evans Blue-labeled bovine serum ~lbumin, Neuroscience Letters, 3 (1976) 191--196.