Brain Research, 171 (1979) 197-212 © Elsevier/North-Holland Biomedical press
197
R E T I N A L G A N G L I O N CELL RESPONSE TO A X O T O M Y A N D NERVE G R O W T H F A C T O R IN T H E R E G E N E R A T I N G VISUAL SYSTEM OF T H E N E W T (NOTOPHTHALMUS VIRIDESCENS) : AN U L T R A S T R U C T U R A L M O R P H O M E T R I C ANALYSIS
JAMES E. TURNER and REBECCA K. DELANEY
Department of Anatomy, Bowman Gray School of Medicine, Wake Forest University, Winston-Salem, N.C. 27103 (U.S.A.) (Accepted November 23rd, 1978)
SUMMARY
Nerve growth factor (NGF) treatment, given as a single 200 BU intraocular injection at the time of optic nerve transection, was found to significantly accelerate the retinal ganglion cell response to axotomy in the newt (Notophthalmus~iridescens). In the control series the per cent of neurons in the retinal ganglion layer demonstrating nuclear reactivity (i.e. chromatin changes) reaches a peak by 14 days post axotomy (14 DPA), plateaus through 21 DPA and falls thereafter, returning to control levels by 90 DPA. N G F treatment is shown to significantly accelerate the entrance of responding retinal ganglion cells into the reactive nuclear phase between 1 and 7 DPA, and by 7 DPA nuclear reactivity has reached a peak, in contrast to 14 DPA for control values. Consequently, N G F treatment causes retinal ganglion cells to be in the nuclear reactive state a week longer than controls but reactivity diminishes after 21 DPA as in controls. Electron microscopic morphometric analysis further substantiates these observations by demonstrating that N G F treatment can elicit certain cellular organelle changes a week earlier (i.e. at 7 DPA) than they would normally occur (i.e. at 14 DPA) in response to axotomy. In addition to eliciting cellular hypertrophy at 7 DPA, N G F treatment significantly increases Golgi field densities in the neuronal perikaryal cytoplasm as well as a doubling of the number of nucleoli per nucleus and stimulating a significant increase in nucleolar cross-sectional areas. A dose-response relationship exists between the per cent of retinal ganglion cells demonstrating nuclear reactivity at 7 DPA and various N G F concentrations which compares favorably with the dose response study involving the number of regenerating axons per nerve cross-section at 14 DPA. Studies to determine if the N G F mediated responses were a specific effect elicited by this protein molecule or whether they are also produced by other peptides
198 which share some properties in common with N G F demonstrate that only N G F is capable of eliciting these responses.
INTRODUCTION There is increasing evidence for the role of nerve growth factor (NGF) or an NGF-like substance in central nervous system (CNS) development and regeneration. Specific binding sites for [12aI]NGF have been shown to be present in the brains of new born and adult rats11,12. In addition, N G F receptor sites have been shown to appear in significant number very early in the development of the chick embryo brain, suggesting that N G F may play a role in the development of the CNS a4. The appearance of early (days 6-8) and late (days 13-18) receptors led Szutowicz et al. a2 to speculate that N G F may serve more than one type of function during the development of the chick brain. In a separate but related study, Merrel et al. 24 have shown that the temporal changes in tectal cell surface specificity, which occur in vivo between days 7-8 in the developing chick brain, can be demonstrated to be dependent on N G F or a structural analog. If N G F or an NGF-like substance does play an important role in the development and maintenance of the CNS, then an endogenous source(s) most likely exists within the CNS. Numerous studies strongly point to glial cells as a source of NGF 4,~°,21,2~,26,29. The potential importance of N G F in regenerative processes in the CNS is suggested by its stimulatory effects on the regeneration of axons in the rat dorsal catecholamine bundle~,7, 8, rabbit vestibular neurons, in vitro, and the dorsal funiculus of young kittens'5, 30. In addition, N G F has been shown to enhance recovery of the feeding response after induced hypothalamic damage in rats 5. Ebbott and Hendry 9 have reported evidence for retrograde transport of N G F in the rat central nervous system. Also, N G F has been shown to increase ornithine decarboxylase in the rat brain 2°. More recently, studies from our laboratory have demonstrated a pronounced effect of N G F and its antiserum on optic nerve regeneration in the newt 18m,42. Most importantly, we have demonstrated that the newt visual system, as a successfully regenerating model, presents an ideal opportunity for studying NGF-mediated regenerative responses in the CNS 35-40. The present paper is an attempt to further characterize our model system and deals with various morphological and ultrastructural aspects of retinal ganglion cell perikaryal response to N G F after axotomy. MATERIALS AND METHODS
77ssue preparation and NGF treatment Adult newts (Notophthalmus viridescens) obtained from Lee's Newt Farm, Oak Ridge, Tennessee, were anesthetized in dilute chloretone solution. Animals in the experimental groups, except those in the dose response studies, received one 2 #1 intraocular injection of 200 BU (i.e. 400 ng) NGF. Animals in the dose response series received one 2 #1 injection of either 10, 20, 200, 400 or 2000 BU of NGF. The control
199 groups received one 2 #1 injection of either the N G F diluent (i.e. phosphate buffer with NaC1), sterile saline, bovine serum albumine (Sigma) cytochrome C (Sigma), lysozyme (Sigma), insulin (Sigma) or epidermal growth factor (Collaborative Research). Injections were made into the vitreous chamber of the left eye according to the method of McEwen and Grafstein z3. The orbital portion of the left optic nerve was severed immediately after injection according to the method of Sperry 32 and Turner and Singer 38. After transection animals were kept in a moist chamber until alert and were transferred to water-filled containers and kept at a constant water temperature of 25 1 °C. Animals were kept on a constant photoperiod of a 14/10 light/dark cycle. Eyes were prepared for electron microscopy at 1, 2, 4, 7, 14, 21, 30, 45 and 90 days post axotomy (DPA). A microperfusion system was devised which allowed small quantities of fixative to be perfused directly into the vitreous chamber adjacent to the retinal ganglion cell layer. The procedure is as follows: (1) approximately 0.3 ml of cold fixative was injected into the orbital space surrounding the eye; (2) a small incision was made through the anteriolateral surface of the eye close to the ora serrata which allowed for efflux of perfused fixative; (3) a 30-gauge needle attached to a 1 ml syringe containing fixative was passed behind the cornea and lens into the vitreous chamber; (4) approximately 0.3 ml of cold fixative was perfused through the eye which was followed by fixation in situ for 15 rain. Corneas and lenses were removed and eyes were fixed over night at 4 °C. Of the various fixatives used, the best was found to be a glutaraldehyde-paraformaldehyde preparation containing acrolein and dimethyl sulfoxide buffered at pH 7.4 with 0.3 M cacodylate buffer 16. The following day eyes were halved through the optic papilla in an anteriorposterior plane, washed in cacodylate buffer, osmicated for 1 h, dehydrated in an ethanol series and embedded in Epon. Thick (1 #m) and thin (600-900 nm) sections were cut with glass and diamond knives on a Porter-Blum ultramicrotome (MT-2B). Thick sections were stained with toluidine blue in borate buffer (1 ~, pH 7.4) for examination and photography with the light microscope. Thin sections were stained with uranyl acetate and lead citrate 31. Nerve growth factor was obtained from the Wellcome Research Labolatories 7S N G F ) and from Collaborative Research (2.5S NGF). With the exception of the dose-response comparison experiment (7S vs 2.5S NGF), all results from this study concerning N G F treatment involve the use of the 7S form of the molecule. The biological activity of N G F was retested in our laboratory according to the method of Levi-Montalcini iv.
Morphometric analysis Thick and thin sections of the retinal ganglion cell layer, taken consistently from the region adjacent to the optic papilla, were quantitatively and qualitatively analyzed. For light microscopic quantitation, approximately 1000-2000 cells from each group of at least 3 retinas each were counted using a Zeiss Universal microscope at a magnification of 800 ×. The number of cells in the retinal ganglion layer at 0-90 DPA demonstrating dramatic chromatin pattern changes were expressed as a per cent of the total number of cells counted. The determination of cells demonstrating dramatic chromatin pattern changes has been reported in a previous publication 40.
200 Electron microscopic montages of randomly selected retinal ganglion cell layer areas, consisting of 150-200 cells from each group of at least 3 retinas, were prepared for analysis by photographing thin sections supported on one-hole grids covered by 0.75 ~ parlodian films. For quantitative analysis, tissues were photographed with a Zeiss EM 9S-2 electron microscope at a primary magnification of 1600 x and montages constructed at a final magnification of 4960 x . In addition, light microscopic montages of randomly selected areas of the central retina consisting of from 900-1200 cells were prepared for quantitation at a magnification of 1120 ×. The light microscopic montages were utilized to measure nuclear areas of all cells within the selected portion of the retinal ganglion cell layer. This information was used to construct histograms of nuclear distributions according to their size (cross-sectional areas). Electron microscopic montages of neurons in the retinal ganglion cell layer were used to measure nuclear areas (sq./~m), perikaryal cytoplasmic areas (sq. #m), cell/nuclear ratios, mitochondrial/sq. #m of perikaryal cytoplasm, Golgi fields/sq. #m of perikaryal cytoplasm, nucleoli/nucleus and average nucleolar area/cell. All neurons in the electron microscopic montages demonstrating nucleoli were analyzed according to these criteria. Morphometric analyses were performed on light and electron microscopic montages at 0 (intact controls), 7 and 14 DPA. Quantitation of the neuronal response to axotomy was accomplished by means of a Ladd Graphic Digitizer interfaced with a Monroe 1830 calculator. This analytical system allows one to follow specific structures with a cursor with resulting x-y coordinates being continuously fed into the programmed calculator by the digitizer. The high accuracy allows for good resolution of the structures under investigation. Before processing with the digitizer, structures were first outlined to facilitate quantitation. The data were grouped for statistical analysis according to the experimental treatment and the significance of difference was tested by either the Student's t-test or an analysis of variance. Since no variations in the studied parameters were observed between central and peripheral ganglion cells, values were pooled for both populations. For all analyses, roughly equivalent numbers of cells were analyzed in each animal within each group of at least 3 retinas. RESULTS
Light microscopic analysis In a previous study 40 we reported that retinal ganglion cell nuclei undergo dramatic changes in chromatin patterns in response to axotomy and are easily observed at the light microscopic level. Fig. 1 contrasts the retinal ganglion cell nuclear response to axotomy at 7 DPA with or without N G F treatment. These light micrographs demonstrate that there are many more neurons with 'reactive nuclei' demonstrating chromatin changes (i.e. heterochromatic to more euchromatic state) in the 7 DPA + N G F group than in the 7 DPA series without N G F treatment. Also, cells in the 7 DPA + N G F series appear to have larger and often multiple nucleoli when compared with the 7 DPA control group. The number or per cent of cells demonstrating nuclear chromatin changes in response to axotomy can be plotted temporally as
VC
V¢
Fig. 1. Light micrographs of retinal ganglion cell layer 7 days after axotomy showing contrasts in the retinal ganglion cell nuclear response to axotomy with or without N G F treatment. VC, vitreous chamber; IPL, inner plexiform layer. A: control retinal ganglion cell layer (RGL) showing only a few cells with reactive nuclei (R) and the majority showing no signs of reactivity (NR). Also, note that nucleoli (N) within these cells are not very prominent. B : NGF-treated retinal ganglion layer (RGL) showing a majority of cells with reactive nuclei (R). Note the large, prominent nucleoli (N) in some of the cells, x 400.
202
,~
64-
-
6o-
4124) 1874)
a: 56-
~
:3,
52-
48[5,1785)
~44% 40qb 36-
~ ~
32~
(4,137/)
~ 28~24-
g 2o-
I
,3198)
-.4 16874)
(4,1942) * ** ***
~ 8- *.~ j4,2145) % - II'qt-,2048
I/"i4,2287>
"--4
LI~(5,,72,1 ~
0
o I
0 2 4
7
I
14 21 DAYS AFTER
o I
3O AXOTOMY
P
197
R E T I N A L G A N G L I O N CELL RESPONSE TO A X O T O M Y A N D NERVE G R O W T H F A C T O R IN T H E R E G E N E R A T I N G VISUAL SYSTEM OF T H E N E W T (NOTOPHTHALMUS VIRIDESCENS) : AN U L T R A S T R U C T U R A L M O R P H O M E T R I C ANALYSIS
JAMES E. TURNER and REBECCA K. DELANEY
Department of Anatomy, Bowman Gray School of Medicine, Wake Forest University, Winston-Salem, N.C. 27103 (U.S.A.) (Accepted November 23rd, 1978)
SUMMARY
Nerve growth factor (NGF) treatment, given as a single 200 BU intraocular injection at the time of optic nerve transection, was found to significantly accelerate the retinal ganglion cell response to axotomy in the newt (Notophthalmus~iridescens). In the control series the per cent of neurons in the retinal ganglion layer demonstrating nuclear reactivity (i.e. chromatin changes) reaches a peak by 14 days post axotomy (14 DPA), plateaus through 21 DPA and falls thereafter, returning to control levels by 90 DPA. N G F treatment is shown to significantly accelerate the entrance of responding retinal ganglion cells into the reactive nuclear phase between 1 and 7 DPA, and by 7 DPA nuclear reactivity has reached a peak, in contrast to 14 DPA for control values. Consequently, N G F treatment causes retinal ganglion cells to be in the nuclear reactive state a week longer than controls but reactivity diminishes after 21 DPA as in controls. Electron microscopic morphometric analysis further substantiates these observations by demonstrating that N G F treatment can elicit certain cellular organelle changes a week earlier (i.e. at 7 DPA) than they would normally occur (i.e. at 14 DPA) in response to axotomy. In addition to eliciting cellular hypertrophy at 7 DPA, N G F treatment significantly increases Golgi field densities in the neuronal perikaryal cytoplasm as well as a doubling of the number of nucleoli per nucleus and stimulating a significant increase in nucleolar cross-sectional areas. A dose-response relationship exists between the per cent of retinal ganglion cells demonstrating nuclear reactivity at 7 DPA and various N G F concentrations which compares favorably with the dose response study involving the number of regenerating axons per nerve cross-section at 14 DPA. Studies to determine if the N G F mediated responses were a specific effect elicited by this protein molecule or whether they are also produced by other peptides
198 which share some properties in common with N G F demonstrate that only N G F is capable of eliciting these responses.
INTRODUCTION There is increasing evidence for the role of nerve growth factor (NGF) or an NGF-like substance in central nervous system (CNS) development and regeneration. Specific binding sites for [12aI]NGF have been shown to be present in the brains of new born and adult rats11,12. In addition, N G F receptor sites have been shown to appear in significant number very early in the development of the chick embryo brain, suggesting that N G F may play a role in the development of the CNS a4. The appearance of early (days 6-8) and late (days 13-18) receptors led Szutowicz et al. a2 to speculate that N G F may serve more than one type of function during the development of the chick brain. In a separate but related study, Merrel et al. 24 have shown that the temporal changes in tectal cell surface specificity, which occur in vivo between days 7-8 in the developing chick brain, can be demonstrated to be dependent on N G F or a structural analog. If N G F or an NGF-like substance does play an important role in the development and maintenance of the CNS, then an endogenous source(s) most likely exists within the CNS. Numerous studies strongly point to glial cells as a source of NGF 4,~°,21,2~,26,29. The potential importance of N G F in regenerative processes in the CNS is suggested by its stimulatory effects on the regeneration of axons in the rat dorsal catecholamine bundle~,7, 8, rabbit vestibular neurons, in vitro, and the dorsal funiculus of young kittens'5, 30. In addition, N G F has been shown to enhance recovery of the feeding response after induced hypothalamic damage in rats 5. Ebbott and Hendry 9 have reported evidence for retrograde transport of N G F in the rat central nervous system. Also, N G F has been shown to increase ornithine decarboxylase in the rat brain 2°. More recently, studies from our laboratory have demonstrated a pronounced effect of N G F and its antiserum on optic nerve regeneration in the newt 18m,42. Most importantly, we have demonstrated that the newt visual system, as a successfully regenerating model, presents an ideal opportunity for studying NGF-mediated regenerative responses in the CNS 35-40. The present paper is an attempt to further characterize our model system and deals with various morphological and ultrastructural aspects of retinal ganglion cell perikaryal response to N G F after axotomy. MATERIALS AND METHODS
77ssue preparation and NGF treatment Adult newts (Notophthalmus viridescens) obtained from Lee's Newt Farm, Oak Ridge, Tennessee, were anesthetized in dilute chloretone solution. Animals in the experimental groups, except those in the dose response studies, received one 2 #1 intraocular injection of 200 BU (i.e. 400 ng) NGF. Animals in the dose response series received one 2 #1 injection of either 10, 20, 200, 400 or 2000 BU of NGF. The control
199 groups received one 2 #1 injection of either the N G F diluent (i.e. phosphate buffer with NaC1), sterile saline, bovine serum albumine (Sigma) cytochrome C (Sigma), lysozyme (Sigma), insulin (Sigma) or epidermal growth factor (Collaborative Research). Injections were made into the vitreous chamber of the left eye according to the method of McEwen and Grafstein z3. The orbital portion of the left optic nerve was severed immediately after injection according to the method of Sperry 32 and Turner and Singer 38. After transection animals were kept in a moist chamber until alert and were transferred to water-filled containers and kept at a constant water temperature of 25 1 °C. Animals were kept on a constant photoperiod of a 14/10 light/dark cycle. Eyes were prepared for electron microscopy at 1, 2, 4, 7, 14, 21, 30, 45 and 90 days post axotomy (DPA). A microperfusion system was devised which allowed small quantities of fixative to be perfused directly into the vitreous chamber adjacent to the retinal ganglion cell layer. The procedure is as follows: (1) approximately 0.3 ml of cold fixative was injected into the orbital space surrounding the eye; (2) a small incision was made through the anteriolateral surface of the eye close to the ora serrata which allowed for efflux of perfused fixative; (3) a 30-gauge needle attached to a 1 ml syringe containing fixative was passed behind the cornea and lens into the vitreous chamber; (4) approximately 0.3 ml of cold fixative was perfused through the eye which was followed by fixation in situ for 15 rain. Corneas and lenses were removed and eyes were fixed over night at 4 °C. Of the various fixatives used, the best was found to be a glutaraldehyde-paraformaldehyde preparation containing acrolein and dimethyl sulfoxide buffered at pH 7.4 with 0.3 M cacodylate buffer 16. The following day eyes were halved through the optic papilla in an anteriorposterior plane, washed in cacodylate buffer, osmicated for 1 h, dehydrated in an ethanol series and embedded in Epon. Thick (1 #m) and thin (600-900 nm) sections were cut with glass and diamond knives on a Porter-Blum ultramicrotome (MT-2B). Thick sections were stained with toluidine blue in borate buffer (1 ~, pH 7.4) for examination and photography with the light microscope. Thin sections were stained with uranyl acetate and lead citrate 31. Nerve growth factor was obtained from the Wellcome Research Labolatories 7S N G F ) and from Collaborative Research (2.5S NGF). With the exception of the dose-response comparison experiment (7S vs 2.5S NGF), all results from this study concerning N G F treatment involve the use of the 7S form of the molecule. The biological activity of N G F was retested in our laboratory according to the method of Levi-Montalcini iv.
Morphometric analysis Thick and thin sections of the retinal ganglion cell layer, taken consistently from the region adjacent to the optic papilla, were quantitatively and qualitatively analyzed. For light microscopic quantitation, approximately 1000-2000 cells from each group of at least 3 retinas each were counted using a Zeiss Universal microscope at a magnification of 800 ×. The number of cells in the retinal ganglion layer at 0-90 DPA demonstrating dramatic chromatin pattern changes were expressed as a per cent of the total number of cells counted. The determination of cells demonstrating dramatic chromatin pattern changes has been reported in a previous publication 40.
200 Electron microscopic montages of randomly selected retinal ganglion cell layer areas, consisting of 150-200 cells from each group of at least 3 retinas, were prepared for analysis by photographing thin sections supported on one-hole grids covered by 0.75 ~ parlodian films. For quantitative analysis, tissues were photographed with a Zeiss EM 9S-2 electron microscope at a primary magnification of 1600 x and montages constructed at a final magnification of 4960 x . In addition, light microscopic montages of randomly selected areas of the central retina consisting of from 900-1200 cells were prepared for quantitation at a magnification of 1120 ×. The light microscopic montages were utilized to measure nuclear areas of all cells within the selected portion of the retinal ganglion cell layer. This information was used to construct histograms of nuclear distributions according to their size (cross-sectional areas). Electron microscopic montages of neurons in the retinal ganglion cell layer were used to measure nuclear areas (sq./~m), perikaryal cytoplasmic areas (sq. #m), cell/nuclear ratios, mitochondrial/sq. #m of perikaryal cytoplasm, Golgi fields/sq. #m of perikaryal cytoplasm, nucleoli/nucleus and average nucleolar area/cell. All neurons in the electron microscopic montages demonstrating nucleoli were analyzed according to these criteria. Morphometric analyses were performed on light and electron microscopic montages at 0 (intact controls), 7 and 14 DPA. Quantitation of the neuronal response to axotomy was accomplished by means of a Ladd Graphic Digitizer interfaced with a Monroe 1830 calculator. This analytical system allows one to follow specific structures with a cursor with resulting x-y coordinates being continuously fed into the programmed calculator by the digitizer. The high accuracy allows for good resolution of the structures under investigation. Before processing with the digitizer, structures were first outlined to facilitate quantitation. The data were grouped for statistical analysis according to the experimental treatment and the significance of difference was tested by either the Student's t-test or an analysis of variance. Since no variations in the studied parameters were observed between central and peripheral ganglion cells, values were pooled for both populations. For all analyses, roughly equivalent numbers of cells were analyzed in each animal within each group of at least 3 retinas. RESULTS
Light microscopic analysis In a previous study 40 we reported that retinal ganglion cell nuclei undergo dramatic changes in chromatin patterns in response to axotomy and are easily observed at the light microscopic level. Fig. 1 contrasts the retinal ganglion cell nuclear response to axotomy at 7 DPA with or without N G F treatment. These light micrographs demonstrate that there are many more neurons with 'reactive nuclei' demonstrating chromatin changes (i.e. heterochromatic to more euchromatic state) in the 7 DPA + N G F group than in the 7 DPA series without N G F treatment. Also, cells in the 7 DPA + N G F series appear to have larger and often multiple nucleoli when compared with the 7 DPA control group. The number or per cent of cells demonstrating nuclear chromatin changes in response to axotomy can be plotted temporally as
VC
V¢
Fig. 1. Light micrographs of retinal ganglion cell layer 7 days after axotomy showing contrasts in the retinal ganglion cell nuclear response to axotomy with or without N G F treatment. VC, vitreous chamber; IPL, inner plexiform layer. A: control retinal ganglion cell layer (RGL) showing only a few cells with reactive nuclei (R) and the majority showing no signs of reactivity (NR). Also, note that nucleoli (N) within these cells are not very prominent. B : NGF-treated retinal ganglion layer (RGL) showing a majority of cells with reactive nuclei (R). Note the large, prominent nucleoli (N) in some of the cells, x 400.
202
,~
64-
-
6o-
4124) 1874)
a: 56-
~
:3,
52-
48[5,1785)
~44% 40qb 36-
~ ~
32~
(4,137/)
~ 28~24-
g 2o-
I
,3198)
-.4 16874)
(4,1942) * ** ***
~ 8- *.~ j4,2145) % - II'qt-,2048
I/"i4,2287>
"--4
LI~(5,,72,1 ~
0
o I
0 2 4
7
I
14 21 DAYS AFTER
o I
3O AXOTOMY
P
