- Email: [email protected]
Neonatal enucleation reduces the proportion of callosal boutons forming multiple synaptic contacts in rat striate cortex
Neuroscience Letters 351 (2003) 17–20 www.elsevier.com/locate/neulet
Neonatal enucleation reduces the proportion of callosal boutons forming multiple synaptic contacts in rat striate cortex Staci A. Sorensena, Theresa A. Jonesb, Jaime F. Olavarriaa,c,* b
a Neurobiology and Behavior Program, University of Washington, Seattle, WA 98195, USA Psychology Department and Institute for Neuroscience Research, University of Texas at Austin, Austin, TX 78712, USA c Psychology Department, University of Washington, Seattle, WA 98195, USA
Received 27 January 2003; received in revised form 21 July 2003; accepted 24 July 2003
Abstract Although bilateral enucleation at birth produces marked abnormalities in the overall distribution and topography of interhemispheric callosal connections in rat visual cortex, it is not known whether it also alters the morphology of callosal synapses. Here we report on the effect of neonatal enucleation on the proportion of callosal boutons making multiple postsynaptic contacts. Synapses were analyzed in adult rats after injections of the anterograde tracer biotinylated dextran amine into the opposite striate cortex. Results show that neonatal enucleation produces a significant reduction in the proportion of callosal boutons making multiple postsynaptic contacts. q 2003 Published by Elsevier Ireland Ltd. Keywords: Synaptogenesis; Visual cortex; Corpus callosum; Deprivation; Multiple synaptic boutons; Plasticity
During development of neuronal circuits, the mechanisms that guide growing axons to their synaptic targets may also influence the morphology of the forming synapses. However, while there is some information about factors that specify neural maps, less is known about the role that these factors play in synaptogenesis. For instance, previous studies have shown that bilateral enucleation at birth produces marked abnormalities in the overall distribution and topography of callosal connections in rat visual cortex [5,17,18], but whether lack of retinal input also alters the morphology of callosal synapses is not known. We have begun to investigate this possibility by examining the effect of neonatal enucleation on the morphology of visual callosal synapses [21]. In this report we focus on callosal boutons making multiple postsynaptic contacts because previous studies have shown that the incidence of these synapses can be modified by experimental paradigms associated with neural plasticity [7,8,10,11,13]. Five Long– Evans hooded rats were anesthetized with halothane (2 – 4% in air) and binocularly enucleated within * Corresponding author. Department of Psychology, University of Washington, Box 351525, Seattle, WA 98195-1525, USA. Tel.: þ 1-206543-8675; fax: þ 1-206-685-3157. E-mail address: [email protected] (J.F. Olavarria). 0304-3940/03/$ - see front matter q 2003 Published by Elsevier Ireland Ltd. doi:10.1016/S0304-3940(03)00938-8
24 h of birth (postnatal day 0, P0). Another five, normallyreared, rats served as age-matched controls. All animal use procedures were in accordance with the University of Washington Animal Care and Use Committee. At about 3 months of age, and under halothane anesthesia (2 – 4% in air), all animals received ten to 15 intracortical injections of biotinylated dextran amine (BDA, 10% solution in 0.9% saline, total volume about 1.5 ml) distributed across the primary visual cortex (area 17) of one hemisphere (Fig. 1C). After 10 days, rats were perfused transcardially with 0.1 M phosphate buffer (PB, pH 7.4), followed by 0.25% glutaraldehyde and 4% paraformaldehyde in 0.1 M PB. Coronal sections (50 mm) were left in a 20% sucrose solution for about 30 min and rapidly frozen (2 80 8C) to increase membrane permeability. After thawing, the sections were incubated in an ABC solution (Vector Laboratories) overnight and processed to reveal anterogradely transported BDA using standard immunoperoxidase methods and 3,3-diaminobenzidine as the chromogen. Tissue samples taken from the 17/18a callosal zone opposite the injection site were rinsed in 0.05 M cacodylate buffer, placed in 2% osmium tetroxide with 0.75% potassium ferrocyanide for 1 h, stained en bloc with 2% uranyl acetate and dehydrated in a series of ascending alcohol and acetone solutions, infiltrated with Eponate 12 Resin, sandwich-
18
S.A. Sorensen et al. / Neuroscience Letters 351 (2003) 17–20
Fig. 1. (A,B) Labeled callosal synapses in layers 2/3 of rat primary visual cortex. (A) Example of labeled single-synaptic bouton (SSB) on a dendritic spine (arrow). An unlabeled synapse on a dendritic spine is also shown (arrowhead). (B) Example of labeled multisynaptic bouton (MSB) forming synapses with two dendritic spines (arrows). The postsynaptic density in the dendritic spine to the left appears segmented (two arrows), suggesting it is perforated. Scale bar, 0.5 mm. (C, top) Stippled areas correspond to cortical regions containing dense accumulations of callosal axons. The border of area 17 is indicated by arrows. An asterisk indicates the region where samples containing layer 2/3 were obtained for analysis. (C, bottom) Diagram of the rat brain indicating the locations of areas 17 and 18a in occipital cortex, the injection sites (dots) on the right hemisphere, and the coronal level from which the section at the top was taken. (D) Percentage of synapses formed by MSBs versus SSBs in layers 2/3 of lateral primary visual cortex in the rat. Enucleated rats had significantly reduced proportions of labeled callosal MSBs.
embedded and polymerized at 60 8C. Wedges of tissue containing dense callosal labeling were dissected out of the 17/18a callosal zone (marked by an asterisk in Fig. 1C) and mounted on resin blocks. About three to four semithin sections (0.8 mm) were obtained using an Ultracut R Microtome (Leica, Nussloch, Germany) and alternate sections were stained with Toluidine Blue to identify cortical layers. Blocks were further trimmed so that they only contained dense callosal labeling in layer 2/3 in the region indicated with an asterisk in Fig. 1C. Ultrathin sections (70 nm) that had a silver to gray interference color were mounted on mesh grids and stained with lead citrate. A JEOL 1200 EXII electron microscope at a magnification of £ 20,000 was used to photograph fields containing at least one labeled synapse. Digitized images of these photographs were analyzed using Photoshop 3.0 (Adobe Systems) at a final magnification of £ 90,000. All images were coded to conceal the experimental condition. Due to variability in tracer uptake, labeling density varied among animals independent of experimental condition. Fields containing
labeled synapses in layer 2/3 in the region marked with an asterisk in Fig. 1C were chosen randomly and, depending on the density of labeling, about 20 – 35 synapses were photographed per animal. A total of 315 labeled synapses were analyzed (124 in enucleated and 191 in normal rats). Labeled callosal synapses were identified by the presence of dark, electron-dense precipitate within the synaptic boutons. Boutons making a single synaptic contact were classified as single-synapse boutons (SSBs), while boutons making more than one synaptic contact in the same section were classified as multisynaptic boutons (MSBs) (Fig. 1A, B). Only synapses contacting dendritic shafts or spines were included in this study. Shafts were identified by the presence of mitochondria, regularly spaced microtubules and/or other shaft-characteristic organelles. Spines were identified by the absence of these organelles and frequently by the presence of a spine apparatus. For each experimental group, the occurrence of each synapse type was reported as a proportion of the total number of synapses analyzed. Although analysis of MSBs in single sections underestimates their occurrence compared to serial section reconstruction methods (e.g. Ref. [7]), we did not perform serial reconstructions due to the low numbers of BDA filled boutons per electron micrograph (one on average). In these conditions, serial reconstructions would have required either an unacceptably large number of electron micrographs, or a much more limited sample of analyzed synapses. Moreover, while single-section identification methods have been shown to be sufficiently sensitive for detecting group differences [10,11], such differences are less likely to be detected in a small number of reconstructed synapses. Thus, we used single-section identification methods because we were primarily interested in the sensitive detection of group differences. Variables were averaged for each individual animal, and SAS general linear models were used to perform one-way analysis of variance (ANOVA) for the effects of Group (enucleated, n ¼ 5, versus control, n ¼ 5). We found a significant decrease in the proportion of axonal boutons forming synaptic contacts with more than one dendritic process (MSBs) in enucleated rats compared to control rats. Fig. 1D shows the proportion of synapses formed by MSBs and SSBs in the enucleated and control rats. The enucleated rats had nearly a 20% decrease in the proportion of MSBs compared to controls (F½1; 8 ¼ 9:04, P , 0:02) and a corresponding increase in the proportion of SSBs (the proportion of MSBs and SSBs adds up to 100%). The differences observed between groups in the proportion of MSBs were due mostly to MSBs formed by dendritic spines (i.e. 100% of MSBs in enucleated and 88 ^ 8.00% in controls were spines synapses). Although there was no significant difference between groups for the proportion of MSBs formed with dendritic shafts versus spines, MSBs formed with shafts were found only infrequently in control rats (12 ^ 8.00% of MSBs) and not at all in enucleated rats. Our finding that neonatal enucleation reduces the
S.A. Sorensen et al. / Neuroscience Letters 351 (2003) 17–20
proportion of callosal MSBs in lateral striate cortex indicates that the eyes are needed for the development and/or maintenance of a normal complement of callosal MSBs. We do not report on other aspects of synaptic morphology, such as the density of callosal synapses or the size of pre- and postsynaptic elements, and it will be important to determine whether they are also affected by neonatal enucleation. Previous studies have shown that removal of retinal input prior to synapse formation does not change the time course of synapse development and the mean synaptic density [4]. Further insight on the effect of retinal input on synaptic morphology will come from studies determining at what stage of development enucleation exerts the effects revealed in this report. The finding [18,19] that retinal input specifies the normal overall distribution and topography of visual callosal connections during a narrow time-window (P4 to P6) raises the possibility that retinal influences on cortical synaptogenesis occur at these early stages of circuit development. This scenario leads to the prediction that the reduction of MSBs we observed in enucleates occurs shortly after P6. Consistent with this possibility, circuit formation and synaptogenesis in rat visual cortex occur to a large extent during the first two postnatal weeks, before eyeopening (around P13) [2,3,16]. Moreover, the observations that blockade of spontaneous retinal activity during the first postnatal days alters several parameters of synaptogenesis in the cortex [20] and lateral geniculate nucleus [12], together with evidence that the presence of retinal input at least until P6, but not until P4, is associated with a transient increase in the duration of postsynaptic currents mediated by NMDA receptors in visual callosal cells [14,15], suggest that activity cues may influence synaptogenesis during early development. Alternatively, activity-independent factors may be important at this early stage of development, including the expression/activation and transport of chemical labels or signaling molecules along the retino-thalamocortical pathway (reviewed in Ref. [19]). It will be important to determine the extent to which these agedependent changes influence synaptogenesis prior to the onset of visual experience. Although neither the development nor the maintenance of normal callosal maps require visual experience [5, 17 –19], visual experience may be needed for the maintenance of a normal complement of MSBs, as shown in the cat geniculocortical pathway [6]. This scenario leads to the prediction that the reduction in MSBs that we found in animals enucleated at birth would only be detectable sometime after eye-opening, and that similar abnormalities would be observed in eyed-animals deprived of visual experience. Unfortunately, available studies of the effect of visual deprivation on synaptic morphology in rat visual cortex did not analyze MSBs (e.g. Ref. [1]). Evidence that adult animals have the capacity to form new MSBs is consistent with the possibility that the differences in MSBs in the present study are at least partially due to lack of visual
19
experience. For example, increases in MSBs in adult animals have been found in the visual cortex after complex environment housing [11], in the motor cortex after complex motor skills training [10,13], as well as in other conditions [7,8]. In studies reporting experience-induced synaptogenesis in adult animals, net increases in synapse number per neuron are primarily accounted for by increases in synapses formed by MSBs [9 – 11], indicating that the reutilization of boutons may be a primary mechanism of increasing synapse number in adult animals. Finally, it remains possible that synapse morphology is influenced both by mechanisms operating during neonatal circuit formation, as well as by subsequent experience-dependent factors. In the context of the present study, an interesting possibility raised by experience-induced synaptic plasticity is that expression of this capacity in adults may depend to some extent on effects of peripheral input during synaptogenesis at early stages of development. It will be of interest to determine whether synapses that form in the absence of peripheral input remain as plastic as those that develop in normal animals.
Acknowledgements We thank Peter Lee for assistance in surgical and histological procedures, and DeAnna Adkins, Tim Monahan, Scott Bury and Ann Voorhies for helping with figures and commenting on the manuscript. This study was supported by NIH grants MH52361 (T.A. Jones) and EY09343 (J.F. Olavarria), and by PERC and RRF (J.F. Olavarria).
References [1] B.W. Bakkum, L.A. Benevento, R.S. Cohen, Effects of light/dark- and dark-rearing on synaptic morphology in the superior colliculus and visual cortex of the postnatal and adult rat, J. Neurosci. Res. 28 (1991) 65–80. [2] M.E. Blue, J.G. Parnavelas, The formation and maturation of synapses in the visual cortex of the rat. II. Quantitative analysis, J. Neurocytol. 12 (1983) 697 –712. [3] M.E. Blue, J.G. Parnavelas, The formation and maturation of synapses in the visual cortex of the rat. I. Qualitative analysis, J. Neurocytol. 12 (1983) 599–616. [4] J.P. Bourgeios, P. Rakic, Synaptogenesis in the occipital cortex of macaque monkey devoid of retinal input from early embryonic stages, Eur. J. Neurosci. 8 (1996) 942–950. [5] C.G. Cusick, R.D. Lund, Modification of visual callosal projections in rats, J. Comp. Neurol. 212 (1982) 385 –398. [6] M.J. Friedlander, K.A. Martin, D. Wassenhove-McCarthy, Effects of monocular visual deprivation on geniculocortical innervation of area 18 in cat, J. Neurosci. 11 (1991) 3268–3288. [7] Y. Geinisman, R.W. Berry, J.F. Disterhoft, J.M. Power, E.A. Van der Zee, Associative learning elicits the formation of multiple-synapse boutons, J. Neurosci. 21 (2001) 5568–5573. [8] G.I. Hatton, Function-related plasticity in hypothalamus, Annu. Rev. Neurosci. 20 (1997) 375 –397. [9] T.A. Jones, Multiple synapse formation in the motor cortex opposite
20
[10]
[11]
[12]
[13]
[14]
[15]
S.A. Sorensen et al. / Neuroscience Letters 351 (2003) 17–20 unilateral sensorimotor cortex lesions in adult rats, J. Comp. Neurol. 414 (1999) 57 –66. T.A. Jones, C.J. Chu, L.A. Grande, A.D. Gregory, Motor skills training enhances lesion-induced structural plasticity in the motor cortex of adult rats, J. Neurosci. 19 (1999) 10153–10163. T.A. Jones, A.Y. Klintsova, V.L. Kilman, A.M. Sirevaag, W.T. Greenough, Induction of multiple synapses by experience in the visual cortex of adult rats, Neurobiol. Learn. Mem. 68 (1997) 13–20. R.E. Kalil, M.W. Dubin, G. Scott, L.A. Stark, Elimination of action potentials blocks the structural development of retinogeniculate synapses, Nature 323 (1986) 156 –158. J.A. Kleim, D. McNamee, E.J. Blankstein, W.T. Greenough, Interdependent changes in synapse size and number across the course of learning, Soc. Neurosci. Abstr. 23 (1997) 221. J.F. Olavarria, J.F.M. van Brederode, W.J. Spain, NMDA receptor response during development of visual callosal connections in normal and deafferented rats, Soc. Neurosci. Abstr. 25 (1999) 1005. J.F. Olavarria, J.F.M. van Brederode, W.J. Spain, Specification of retinotopically corresponding callosal linkages coincides with changes in NMDA receptor responses during first postnatal week, Soc. Neurosci. Abstr. 26 (2000) 1614.
[16] J. Olavarria, R.C. Van Sluyters, Organization and postnatal development of callosal connections in the visual cortex of the rat, J. Comp. Neurol. 239 (1985) 1–26. [17] J.F. Olavarria, C.P. Li, Effects of neonatal enucleation on the organization of callosal linkages in striate cortex of the rat, J. Comp. Neurol. 361 (1995) 138–151. [18] J. Olavarria, R. Malach, R.C. Van Sluyters, Development of visual callosal connections in neonatally enucleated rats, J. Comp. Neurol. 260 (1987) 321–348. [19] J.F. Olavarria, R. Hiroi, Retinal influences specify cortico-cortical maps by postnatal day six in rats and mice, J. Comp. Neurol. 459 (2003) 156 –172. [20] R.V. Riccio, M.A. Matthews, The postnatal development of the rat primary visual cortex during optic nerve impulse blockade by intraocular tetrodotoxin: a quantitative electron microscopic analysis, Dev. Brain Res. 20 (1985) 55–68. [21] S.A. Sorensen, T.A. Jones, E. Covey, J.F. Olavarria, Neonatal retinal deafferentation changes the structure of callosal synapses in striate cortex, Soc. Neurosci. Abstr. 26 (2000) 855.
Neonatal enucleation reduces the proportion of callosal boutons forming multiple synaptic contacts in rat striate cortex Staci A. Sorensena, Theresa A. Jonesb, Jaime F. Olavarriaa,c,* b
a Neurobiology and Behavior Program, University of Washington, Seattle, WA 98195, USA Psychology Department and Institute for Neuroscience Research, University of Texas at Austin, Austin, TX 78712, USA c Psychology Department, University of Washington, Seattle, WA 98195, USA
Received 27 January 2003; received in revised form 21 July 2003; accepted 24 July 2003
Abstract Although bilateral enucleation at birth produces marked abnormalities in the overall distribution and topography of interhemispheric callosal connections in rat visual cortex, it is not known whether it also alters the morphology of callosal synapses. Here we report on the effect of neonatal enucleation on the proportion of callosal boutons making multiple postsynaptic contacts. Synapses were analyzed in adult rats after injections of the anterograde tracer biotinylated dextran amine into the opposite striate cortex. Results show that neonatal enucleation produces a significant reduction in the proportion of callosal boutons making multiple postsynaptic contacts. q 2003 Published by Elsevier Ireland Ltd. Keywords: Synaptogenesis; Visual cortex; Corpus callosum; Deprivation; Multiple synaptic boutons; Plasticity
During development of neuronal circuits, the mechanisms that guide growing axons to their synaptic targets may also influence the morphology of the forming synapses. However, while there is some information about factors that specify neural maps, less is known about the role that these factors play in synaptogenesis. For instance, previous studies have shown that bilateral enucleation at birth produces marked abnormalities in the overall distribution and topography of callosal connections in rat visual cortex [5,17,18], but whether lack of retinal input also alters the morphology of callosal synapses is not known. We have begun to investigate this possibility by examining the effect of neonatal enucleation on the morphology of visual callosal synapses [21]. In this report we focus on callosal boutons making multiple postsynaptic contacts because previous studies have shown that the incidence of these synapses can be modified by experimental paradigms associated with neural plasticity [7,8,10,11,13]. Five Long– Evans hooded rats were anesthetized with halothane (2 – 4% in air) and binocularly enucleated within * Corresponding author. Department of Psychology, University of Washington, Box 351525, Seattle, WA 98195-1525, USA. Tel.: þ 1-206543-8675; fax: þ 1-206-685-3157. E-mail address: [email protected] (J.F. Olavarria). 0304-3940/03/$ - see front matter q 2003 Published by Elsevier Ireland Ltd. doi:10.1016/S0304-3940(03)00938-8
24 h of birth (postnatal day 0, P0). Another five, normallyreared, rats served as age-matched controls. All animal use procedures were in accordance with the University of Washington Animal Care and Use Committee. At about 3 months of age, and under halothane anesthesia (2 – 4% in air), all animals received ten to 15 intracortical injections of biotinylated dextran amine (BDA, 10% solution in 0.9% saline, total volume about 1.5 ml) distributed across the primary visual cortex (area 17) of one hemisphere (Fig. 1C). After 10 days, rats were perfused transcardially with 0.1 M phosphate buffer (PB, pH 7.4), followed by 0.25% glutaraldehyde and 4% paraformaldehyde in 0.1 M PB. Coronal sections (50 mm) were left in a 20% sucrose solution for about 30 min and rapidly frozen (2 80 8C) to increase membrane permeability. After thawing, the sections were incubated in an ABC solution (Vector Laboratories) overnight and processed to reveal anterogradely transported BDA using standard immunoperoxidase methods and 3,3-diaminobenzidine as the chromogen. Tissue samples taken from the 17/18a callosal zone opposite the injection site were rinsed in 0.05 M cacodylate buffer, placed in 2% osmium tetroxide with 0.75% potassium ferrocyanide for 1 h, stained en bloc with 2% uranyl acetate and dehydrated in a series of ascending alcohol and acetone solutions, infiltrated with Eponate 12 Resin, sandwich-
18
S.A. Sorensen et al. / Neuroscience Letters 351 (2003) 17–20
Fig. 1. (A,B) Labeled callosal synapses in layers 2/3 of rat primary visual cortex. (A) Example of labeled single-synaptic bouton (SSB) on a dendritic spine (arrow). An unlabeled synapse on a dendritic spine is also shown (arrowhead). (B) Example of labeled multisynaptic bouton (MSB) forming synapses with two dendritic spines (arrows). The postsynaptic density in the dendritic spine to the left appears segmented (two arrows), suggesting it is perforated. Scale bar, 0.5 mm. (C, top) Stippled areas correspond to cortical regions containing dense accumulations of callosal axons. The border of area 17 is indicated by arrows. An asterisk indicates the region where samples containing layer 2/3 were obtained for analysis. (C, bottom) Diagram of the rat brain indicating the locations of areas 17 and 18a in occipital cortex, the injection sites (dots) on the right hemisphere, and the coronal level from which the section at the top was taken. (D) Percentage of synapses formed by MSBs versus SSBs in layers 2/3 of lateral primary visual cortex in the rat. Enucleated rats had significantly reduced proportions of labeled callosal MSBs.
embedded and polymerized at 60 8C. Wedges of tissue containing dense callosal labeling were dissected out of the 17/18a callosal zone (marked by an asterisk in Fig. 1C) and mounted on resin blocks. About three to four semithin sections (0.8 mm) were obtained using an Ultracut R Microtome (Leica, Nussloch, Germany) and alternate sections were stained with Toluidine Blue to identify cortical layers. Blocks were further trimmed so that they only contained dense callosal labeling in layer 2/3 in the region indicated with an asterisk in Fig. 1C. Ultrathin sections (70 nm) that had a silver to gray interference color were mounted on mesh grids and stained with lead citrate. A JEOL 1200 EXII electron microscope at a magnification of £ 20,000 was used to photograph fields containing at least one labeled synapse. Digitized images of these photographs were analyzed using Photoshop 3.0 (Adobe Systems) at a final magnification of £ 90,000. All images were coded to conceal the experimental condition. Due to variability in tracer uptake, labeling density varied among animals independent of experimental condition. Fields containing
labeled synapses in layer 2/3 in the region marked with an asterisk in Fig. 1C were chosen randomly and, depending on the density of labeling, about 20 – 35 synapses were photographed per animal. A total of 315 labeled synapses were analyzed (124 in enucleated and 191 in normal rats). Labeled callosal synapses were identified by the presence of dark, electron-dense precipitate within the synaptic boutons. Boutons making a single synaptic contact were classified as single-synapse boutons (SSBs), while boutons making more than one synaptic contact in the same section were classified as multisynaptic boutons (MSBs) (Fig. 1A, B). Only synapses contacting dendritic shafts or spines were included in this study. Shafts were identified by the presence of mitochondria, regularly spaced microtubules and/or other shaft-characteristic organelles. Spines were identified by the absence of these organelles and frequently by the presence of a spine apparatus. For each experimental group, the occurrence of each synapse type was reported as a proportion of the total number of synapses analyzed. Although analysis of MSBs in single sections underestimates their occurrence compared to serial section reconstruction methods (e.g. Ref. [7]), we did not perform serial reconstructions due to the low numbers of BDA filled boutons per electron micrograph (one on average). In these conditions, serial reconstructions would have required either an unacceptably large number of electron micrographs, or a much more limited sample of analyzed synapses. Moreover, while single-section identification methods have been shown to be sufficiently sensitive for detecting group differences [10,11], such differences are less likely to be detected in a small number of reconstructed synapses. Thus, we used single-section identification methods because we were primarily interested in the sensitive detection of group differences. Variables were averaged for each individual animal, and SAS general linear models were used to perform one-way analysis of variance (ANOVA) for the effects of Group (enucleated, n ¼ 5, versus control, n ¼ 5). We found a significant decrease in the proportion of axonal boutons forming synaptic contacts with more than one dendritic process (MSBs) in enucleated rats compared to control rats. Fig. 1D shows the proportion of synapses formed by MSBs and SSBs in the enucleated and control rats. The enucleated rats had nearly a 20% decrease in the proportion of MSBs compared to controls (F½1; 8 ¼ 9:04, P , 0:02) and a corresponding increase in the proportion of SSBs (the proportion of MSBs and SSBs adds up to 100%). The differences observed between groups in the proportion of MSBs were due mostly to MSBs formed by dendritic spines (i.e. 100% of MSBs in enucleated and 88 ^ 8.00% in controls were spines synapses). Although there was no significant difference between groups for the proportion of MSBs formed with dendritic shafts versus spines, MSBs formed with shafts were found only infrequently in control rats (12 ^ 8.00% of MSBs) and not at all in enucleated rats. Our finding that neonatal enucleation reduces the
S.A. Sorensen et al. / Neuroscience Letters 351 (2003) 17–20
proportion of callosal MSBs in lateral striate cortex indicates that the eyes are needed for the development and/or maintenance of a normal complement of callosal MSBs. We do not report on other aspects of synaptic morphology, such as the density of callosal synapses or the size of pre- and postsynaptic elements, and it will be important to determine whether they are also affected by neonatal enucleation. Previous studies have shown that removal of retinal input prior to synapse formation does not change the time course of synapse development and the mean synaptic density [4]. Further insight on the effect of retinal input on synaptic morphology will come from studies determining at what stage of development enucleation exerts the effects revealed in this report. The finding [18,19] that retinal input specifies the normal overall distribution and topography of visual callosal connections during a narrow time-window (P4 to P6) raises the possibility that retinal influences on cortical synaptogenesis occur at these early stages of circuit development. This scenario leads to the prediction that the reduction of MSBs we observed in enucleates occurs shortly after P6. Consistent with this possibility, circuit formation and synaptogenesis in rat visual cortex occur to a large extent during the first two postnatal weeks, before eyeopening (around P13) [2,3,16]. Moreover, the observations that blockade of spontaneous retinal activity during the first postnatal days alters several parameters of synaptogenesis in the cortex [20] and lateral geniculate nucleus [12], together with evidence that the presence of retinal input at least until P6, but not until P4, is associated with a transient increase in the duration of postsynaptic currents mediated by NMDA receptors in visual callosal cells [14,15], suggest that activity cues may influence synaptogenesis during early development. Alternatively, activity-independent factors may be important at this early stage of development, including the expression/activation and transport of chemical labels or signaling molecules along the retino-thalamocortical pathway (reviewed in Ref. [19]). It will be important to determine the extent to which these agedependent changes influence synaptogenesis prior to the onset of visual experience. Although neither the development nor the maintenance of normal callosal maps require visual experience [5, 17 –19], visual experience may be needed for the maintenance of a normal complement of MSBs, as shown in the cat geniculocortical pathway [6]. This scenario leads to the prediction that the reduction in MSBs that we found in animals enucleated at birth would only be detectable sometime after eye-opening, and that similar abnormalities would be observed in eyed-animals deprived of visual experience. Unfortunately, available studies of the effect of visual deprivation on synaptic morphology in rat visual cortex did not analyze MSBs (e.g. Ref. [1]). Evidence that adult animals have the capacity to form new MSBs is consistent with the possibility that the differences in MSBs in the present study are at least partially due to lack of visual
19
experience. For example, increases in MSBs in adult animals have been found in the visual cortex after complex environment housing [11], in the motor cortex after complex motor skills training [10,13], as well as in other conditions [7,8]. In studies reporting experience-induced synaptogenesis in adult animals, net increases in synapse number per neuron are primarily accounted for by increases in synapses formed by MSBs [9 – 11], indicating that the reutilization of boutons may be a primary mechanism of increasing synapse number in adult animals. Finally, it remains possible that synapse morphology is influenced both by mechanisms operating during neonatal circuit formation, as well as by subsequent experience-dependent factors. In the context of the present study, an interesting possibility raised by experience-induced synaptic plasticity is that expression of this capacity in adults may depend to some extent on effects of peripheral input during synaptogenesis at early stages of development. It will be of interest to determine whether synapses that form in the absence of peripheral input remain as plastic as those that develop in normal animals.
Acknowledgements We thank Peter Lee for assistance in surgical and histological procedures, and DeAnna Adkins, Tim Monahan, Scott Bury and Ann Voorhies for helping with figures and commenting on the manuscript. This study was supported by NIH grants MH52361 (T.A. Jones) and EY09343 (J.F. Olavarria), and by PERC and RRF (J.F. Olavarria).
References [1] B.W. Bakkum, L.A. Benevento, R.S. Cohen, Effects of light/dark- and dark-rearing on synaptic morphology in the superior colliculus and visual cortex of the postnatal and adult rat, J. Neurosci. Res. 28 (1991) 65–80. [2] M.E. Blue, J.G. Parnavelas, The formation and maturation of synapses in the visual cortex of the rat. II. Quantitative analysis, J. Neurocytol. 12 (1983) 697 –712. [3] M.E. Blue, J.G. Parnavelas, The formation and maturation of synapses in the visual cortex of the rat. I. Qualitative analysis, J. Neurocytol. 12 (1983) 599–616. [4] J.P. Bourgeios, P. Rakic, Synaptogenesis in the occipital cortex of macaque monkey devoid of retinal input from early embryonic stages, Eur. J. Neurosci. 8 (1996) 942–950. [5] C.G. Cusick, R.D. Lund, Modification of visual callosal projections in rats, J. Comp. Neurol. 212 (1982) 385 –398. [6] M.J. Friedlander, K.A. Martin, D. Wassenhove-McCarthy, Effects of monocular visual deprivation on geniculocortical innervation of area 18 in cat, J. Neurosci. 11 (1991) 3268–3288. [7] Y. Geinisman, R.W. Berry, J.F. Disterhoft, J.M. Power, E.A. Van der Zee, Associative learning elicits the formation of multiple-synapse boutons, J. Neurosci. 21 (2001) 5568–5573. [8] G.I. Hatton, Function-related plasticity in hypothalamus, Annu. Rev. Neurosci. 20 (1997) 375 –397. [9] T.A. Jones, Multiple synapse formation in the motor cortex opposite
20
[10]
[11]
[12]
[13]
[14]
[15]
S.A. Sorensen et al. / Neuroscience Letters 351 (2003) 17–20 unilateral sensorimotor cortex lesions in adult rats, J. Comp. Neurol. 414 (1999) 57 –66. T.A. Jones, C.J. Chu, L.A. Grande, A.D. Gregory, Motor skills training enhances lesion-induced structural plasticity in the motor cortex of adult rats, J. Neurosci. 19 (1999) 10153–10163. T.A. Jones, A.Y. Klintsova, V.L. Kilman, A.M. Sirevaag, W.T. Greenough, Induction of multiple synapses by experience in the visual cortex of adult rats, Neurobiol. Learn. Mem. 68 (1997) 13–20. R.E. Kalil, M.W. Dubin, G. Scott, L.A. Stark, Elimination of action potentials blocks the structural development of retinogeniculate synapses, Nature 323 (1986) 156 –158. J.A. Kleim, D. McNamee, E.J. Blankstein, W.T. Greenough, Interdependent changes in synapse size and number across the course of learning, Soc. Neurosci. Abstr. 23 (1997) 221. J.F. Olavarria, J.F.M. van Brederode, W.J. Spain, NMDA receptor response during development of visual callosal connections in normal and deafferented rats, Soc. Neurosci. Abstr. 25 (1999) 1005. J.F. Olavarria, J.F.M. van Brederode, W.J. Spain, Specification of retinotopically corresponding callosal linkages coincides with changes in NMDA receptor responses during first postnatal week, Soc. Neurosci. Abstr. 26 (2000) 1614.
[16] J. Olavarria, R.C. Van Sluyters, Organization and postnatal development of callosal connections in the visual cortex of the rat, J. Comp. Neurol. 239 (1985) 1–26. [17] J.F. Olavarria, C.P. Li, Effects of neonatal enucleation on the organization of callosal linkages in striate cortex of the rat, J. Comp. Neurol. 361 (1995) 138–151. [18] J. Olavarria, R. Malach, R.C. Van Sluyters, Development of visual callosal connections in neonatally enucleated rats, J. Comp. Neurol. 260 (1987) 321–348. [19] J.F. Olavarria, R. Hiroi, Retinal influences specify cortico-cortical maps by postnatal day six in rats and mice, J. Comp. Neurol. 459 (2003) 156 –172. [20] R.V. Riccio, M.A. Matthews, The postnatal development of the rat primary visual cortex during optic nerve impulse blockade by intraocular tetrodotoxin: a quantitative electron microscopic analysis, Dev. Brain Res. 20 (1985) 55–68. [21] S.A. Sorensen, T.A. Jones, E. Covey, J.F. Olavarria, Neonatal retinal deafferentation changes the structure of callosal synapses in striate cortex, Soc. Neurosci. Abstr. 26 (2000) 855.