- Email: [email protected]
A new form of synaptic plasticity is transiently expressed in the developing rat visual cortex: a modulatory role for visual experience and brain-derived neurotrophic factor
Pergamon PII:
Neuroscience Vol. 91, No. 1, pp. 163–173, 1999 Copyright 䉷 1999 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306-4522/99 $20.00+0.00 S0306-4522(98)00598-3
A NEW FORM OF SYNAPTIC PLASTICITY IS TRANSIENTLY EXPRESSED IN THE DEVELOPING RAT VISUAL CORTEX: A MODULATORY ROLE FOR VISUAL EXPERIENCE AND BRAINDERIVED NEUROTROPHIC FACTOR E. SERMASI,* D. TROPEA* and L. DOMENICI†‡ *International School for Advanced Studies (SISSA), Neuroscience Program, Via Beirut 2-4, 34014 Trieste, Italy †Institute of Neurophysiology, CNR, Via San Zeno 51, 56127 Pisa, Italy
Abstract—Synaptic plasticity has been implicated in the mechanisms contributing to the shaping of the cortical circuits responsible for the transmission of the visual input in the rat primary visual cortex. However, the degree of plasticity of the thalamocortical synapse may change during development, perhaps reflecting the degree of stabilization of the circuitry subserving it. We have chosen the ability of this synapse to be first depressed and then potentiated as a specific indicator of its plasticity. In this study we have investigated how this parameter changes during development and the factors controlling it. Extracellular field potentials in cortical layers 2/3 were evoked by stimulation of the white matter in rat primary visual cortex slices prepared at different postnatal ages. Low-frequency stimulation (900 pulses at 1 Hz) of the white matter was used to induce long-term depression of field potential amplitude, whereas long-term potentiation was evoked by high-frequency stimulation consisting of three trains at 100 Hz. We provide evidence that while it is possible to potentiate previously depressed synapses soon after eye opening (postnatal day 17) this synaptic characteristic decreases rapidly thereafter. The decrease in this form of cortical synaptic plasticity closely matches the stabilization of the cortical circuitry towards an adult pattern of connectivity and function. Depressed cortical synapses cannot be potentiated in normal rats at postnatal 23, but they can be potentiated in rats reared in the dark from postnatal days 17 to 29. Moreover, application of brain-derived neurotrophic factor, known to be expressed in an activitydependent manner, was able to restore the ability of synapses to be potentiated after long-term depression, thus indicating its important modulatory role in brain development. 䉷 1999 IBRO. Published by Elsevier Science Ltd. Key words: synaptic plasticity, visual cortex, activity-dependent development, dark-rearing, neurotrophins, brainderived neurotrophic factor.
The mammalian sensory cortices are immature at birth, both anatomically and physiologically, and develop gradually during defined postnatal time windows. 3,13,16–18 The underlying developmental processes are activity dependent 33 and follow the Hebbian rule: the presynaptic inputs that are simultaneously active with the postsynaptic cells are reinforced while the others are lost. For many years the visual cortex has been used as a model to study activity-dependent developmental processes. In the rat visual cortex, for example, neuronal functional properties such as orientation selectivity, receptive field size and ocular dominance are largely immature soon after eye opening [postnatal day (P)14–15], reaching progressively adult properties. 15,34 In particular, orientation selectivity and ‡To whom correspondence should be addressed. Abbreviations: aCSF, artificial cerebrospinal fluid; BDNF, brain-derived neurotrophic factor; DMSO, dimethylsulfoxide; HFS, high-frequency stimulation; LFS, low-frequency stimulation; LTD, long-term depression; LTP, long-term potentiation; P, postnatal day; PCCB, percentage change from control baseline.
ocular dominance distribution develop markedly between P17 and P23, reaching adult-like properties by P26–P30. Visual experience is a necessary requirement for maturation of the cortical connections that subserve the visual function. Indeed, in the rat 15 as well as in other species 27,36,37 dark-rearing from birth slows down the maturation of the cortical circuitry. For example, functional properties of visual cortical neurons such as ocular dominance and orientation selectivity are maintained in an immature state in dark-reared rats. Activity-dependent synaptic plasticity, and in particular long-term changes in synaptic efficacy such as long-term potentiation (LTP) and longterm depression (LTD), have been implicated in the mechanisms controlling the cortical developmental process. 20,31 In the rat primary visual cortex LTP and LTD can be elicited by stimulation of the white matter containing the geniculocortical fibres. Both forms of synaptic plasticity have been reported to be N-methyl-d-aspartate dependent. 2,23 However, while LTD can be evoked throughout the lifespan, LTP expression is restricted to postnatal
163
164
E. Sermasi et al.
development. 22–24,32 In particular, it has been shown that LTP is high soon after eye opening but progressively decreases thereafter and disappears by P35. As a consequence, the degree and characteristics of cortical synaptic plasticity change during postnatal development, possibly reflecting the level of maturation of the circuitry subserving the sensory function. Up to now, work on visual cortex synaptic plasticity has dealt with either LTP or LTD but not the two together. We have chosen to investigate the ability of cortical synapses to be first depressed and then potentiated as a specific indicator of the degree of cortical synaptic plasticity at different developmental ages. This phenomenon has been studied in slices of rat brain containing the primary visual cortex taken at postnatal ages representing different stages of both the maturation of the functional properties of cortical neurons and LTP expression (P17, i.e. two to three days after eye opening, P23 and P35). Similar experiments were also performed in light-deprived animals to verify a potential visual experience dependence of this form of synaptic plasticity. In addition, we have examined the possible role of brain-derived neurotrophic factor (BDNF) on the plastic property of developing cortical synapses. Indeed, neurotrophins have been shown to play a role in the cortical synaptic plasticity of this part of the cortex. 1,11,35 EXPERIMENTAL PROCEDURES
Visual cortex slices preparation and electrophysiological recordings Albino rats (Harlan Nossan) were deeply anaesthetized with urethane and then decapitated. All efforts were made to minimize animal suffering and to reduce the number of animals used. The brain was rapidly removed and coronal slices (400 mm) containing the visual cortex were prepared using a Vibratome (Pelco, U.S.A.) and superfused in a submerged recording chamber at 33⬚C with artificial cerebrospinal fluid (aCSF) at a rate of 4 ml/min. An average of two to three slices per animal (the second and third slices from the occipital pole of each hemisphere) was used for experimental purposes. The aCSF was gassed with 95% O2 and 5% CO2 and had the following composition in mM: NaCl, 126; KCl, 3.5; NaH2PO4, 1.2; MgCl, 1.3; CaCl2, 2; NaHCO3, 25; glucose, 11. Extracellular field potentials in the inferior half of cortical layers 2/3 were recorded with an electrode filled with a 2 M NaCl solution and evoked, by stimulation of the white matter containing geniculocortical fibres, using a bipolar concentric stimulating electrode as previously described by Domenici et al. 12 LTD was induced by low-frequency stimulation (LFS, 900 pulses at 1 Hz) of the white matter. The protocol for inducing LTP consisted of three trains of high-frequency stimulation (HFS, 100 Hz, 1 s). Slices containing the primary visual cortex were prepared from either postnatal day 16–17 (P17) or P23– P24 (P23) or P35 (P35) rats. In dark-rearing experiments, slices were prepared from animals light deprived from P17 to P29, that is for a period considered sufficient to maintain P17 synaptic characteristics. BDNF (Alamone Labs, Israel) was applied by filling the recording electrode with a 100 ng/ ml in 1 M NaCl solution. K252a (Alamone Labs, Israel) was first aliquoted in dimethylsulphoxide (DMSO), then diluted in aCSF to a final concentration of 200 nM and administered to slices by superfusion. All animal experiments were
carried out in accordance with the European Community Council Directive of 24 November 1986 (86/609/EEC). Data analysis The amplitude of the maximum negative field potential in layer 2/3 was used as a measure of the evoked population excitatory current. 25 Changes in the amplitude of the maximum negative field potential mirror changes in the magnitude of a monosynaptic current sink and correlate with changes in the initial slope of the negative potentials. In order to avoid polysynaptic sinks before starting the recording we ensured that field potentials were monophasic and that the shape was not changed by frequency of stimulation higher than 1 Hz. Following this recording protocol the latency of negative field potentials ranged between 5 and 7 ms. To minimize the possibility of widely activating fibres in the white matter the stimulating electrode was placed at the border between the white matter and layer 6 in a way that according to Kenan-Vaknin and Teyler 21 and Woodward et al. 38 would facilitate the selective stimulation of vertical pathways projecting to cortical layers 2/3 and 4. Baseline responses were obtained every 15 s and averaged every four responses with a stimulation intensity that yielded a halfmaximal response. The magnitude of both LTD and potentiation after LTD (dedepression) was measured starting 20 min after the end of the respective conditioning protocol. The mean value of 20 consecutive averaged responses was calculated relative to the baseline averaged control values for each experiment. In the case of failure of potentiation after LTD at P23, baseline responses were followed for at least 20 min unless otherwise stated. Data were then pooled together for each experimental group and expressed as percentage change from control baseline (PCCB) ^ S.E.M. Statistical comparison was done by applying a Mann–Whitney rank sum test among LTD and dedepression PCCB values. Difference was considered significant with P ⬍ 0.05. RESULTS
Reversal of long-term depression at P17 and P23 At P17, having initially induced LTD by LFS (900 pulses at 1 Hz) of the white matter (Table 1, PCCB 69 ^ 4%, n 5) and followed it for 40 min, stable potentiation was reliably elicited by a subsequent HFS (100 Hz, 1 s) in all cases (Fig. 1A, n 5/5). Potentiated responses were equal or higher to the control values recorded prior to the LTD protocol (Table 1, PCCB 105 ^ 10%, n 5). The opposite phenomenon, i.e. the depression of cortical synapses after induction of LTP, was equally observed in slices from P17 rats (LTP 179 ^ 17%, depotentiation 107 ^ 4%, n 3). However, depotentiation was not further investigated at later ages, as it is known that LTP expression declines during development. 24 At P23 (Fig. 1B) the effects of LFS were similar to those evoked at P17 (Table 1, PCCB 62 ^ 4%, n 8). In fact, comparison of LTD amplitude at these two ages revealed no statistically significant difference. However, in sharp contrast to that observed at P17, at P23 it was never possible (8/8) to potentiate the evoked responses after the induction of LTD (Table 1, PCCB 56 ^ 7%, n 8). In order to verify a possible delay in the onset of potentiation after LTD at P23, in two cases the
Visual experience and BDNF modulate cortical synaptic plasticity Table 1. Long-term depression and dedepression at P17 and P23 and in dark-reared rats
LTD De-Dep
P17
P23
Rats dark-reared from P17 to P29
69 ^ 4% 105 ^ 10%
62 ^ 4% 56 ^ 7%
82 ^ 2% 111 ^ 6%
Values are expressed as mean percentage change from control baseline ^ S.E.M. De-Dep, dedepression.
recording was continued beyond 30 min after the tetanic stimulation. However, in accordance with that observed in the other cases, no potentiation was noted. Similar experiments performed in adult animals (P35, n 2, LTD PCCB 70%, dedepression PCCB 66%) confirmed that this form of plasticity is present at P17 but declines rapidly thereafter. Control experiments In control experiments at P23 we checked whether the inability of the synapse to be potentiated after depression was a consequence of a general inability of the visual cortex slice to undergo potentiation. To test this hypothesis, we first repeated the same experimental protocol of evoking LTP having followed an LTD for 40 min (Fig. 2A, position 1). Having verified the lack of potentiation with this electrode position, the recording electrode was moved to a new location in the same slice, within layers 2/3 of the primary visual cortex. In these conditions, after at least 30 min of control stimulation in order to exclude a running down of the ability to undergo LTP, we were able to induce LTP (Fig. 2A, position 2, 5/6 slices). Thus at P23 it was possible in the same slice to induce an LTP but not to potentiate a previously depressed evoked response. In a further set of control experiments at P23, after 40 min of LTD we increased the current used for the HFS of the white matter. In order to avoid any massive stimulation of previously unrecruited fibres, the current was only increased up to a level able to evoke responses of the same amplitude as the control pre-LTD values. HFS with increased current induced in three cases a temporary increment in the responses, which lasted for less than 25 min, but in none of these experiments was it possible to potentiate the depressed responses in the long term (Fig. 2B, n 4/4). Recordings were continued for at least 30 min after the tetanic stimulation. These results demonstrate that the synaptic inability to potentiate previously depressed responses observed at P23 is not due to the small amplitude of the test response per se.
165
Effects of dark-rearing on reversal of long-term depression The loss of the ability to reverse a depression observed at P23 could be part of a developmental process under the control of visual experience. If this hypothesis were true then it could be possible to maintain the P17 thalamocortical synaptic characteristics by blocking sensory input to it. To verify this we reared normal P17 rats in the dark until P29. Experiments performed using visual cortex slices from these animals (Fig. 3) indeed showed that it was again possible to see potentiation of the evoked responses after LTD (Table 1, PCCB 111 ^ 6%, n 5/5). LTD amplitude, although slightly reduced, was still considerable (Table 1, PCCB 82 ^ 2%, n 5). Clearly, we were able to block or delay the developmental loss of this form of synaptic plasticity, suggesting that it relies on visually evoked activity. Effects of brain-derived neurotrophic factor on reversal of long-term depression To examine the possible effect of exogenous BDNF on cortical synaptic plasticity, we filled the recording electrode with a BDNF (100 ng/ml in 1 M NaCl) solution. This method was used in an attempt to concentrate and study BDNF actions in a restricted area of the visual cortex. To ensure leaking of BDNF from the recording electrode we slightly broke the tip of the pipette immediately before the beginning of the recording session. The resistance of BDNF filled electrodes was verified not to be grossly different with respect to that of the electrodes used in control experiments. BDNF released from the pipette induced a gradual increase in the size of the responses, as already shown when BDNF was applied to the slice by perfusion. 1 This effect, not observed in control slices, reached a plateau in around 15 min when control responses started to be recorded. In these experimental conditions we found that at P23 it was again possible to induce LTP after LTD (Fig. 4). LTD size (Table 2, PCCB 73 ^ 6%, n 6) was not significantly reduced in comparison to P23 control slices. In marked contrast, HFS to the white matter was again capable of inducing a stable dedepression (n 6/6) whose magnitude (Table 2, PCCB 125 ^ 8%) was not significantly different from that found at P17. We felt that it was important to check whether applied BDNF was directly responsible for restoring at P23 the potentiation after LTD observed at P17. In order to do this we used K252a, a known inhibitor of neurotrophin receptor tyrosine kinase activity, at a concentration (200 nM) known to block BDNF action on hippocampus and visual cortex synaptic plasticity. 1,19 From the results reported in Fig. 5 it is clear that perfusion of P23 slices with K252a (200 nM) blocked the action of the BDNF contained in the recording electrode
166
E. Sermasi et al.
Fig. 1.
Visual experience and BDNF modulate cortical synaptic plasticity
(Table 2, PCCB during LTD 70 ^ 1%, PCCB after high-frequency protocol 67 ^ 5%, n 4). However, the effect of K252a on BDNF action was specific to the dedepression phase as the amplitude of LTD did not change. These results prompted us to test the possible role of endogenous tyrosine kinase activated by neurotrophins in reversing LTD at P17. To do this we perfused P17 slices with K252a (200 nM) and applied the reversing LTD protocol. Contrary to the observations in control P17 preparations, the presence of K252a blocked the potentiating effects of the tetanic stimulation (Fig. 6) while LTD amplitude did not differ significantly (Table 2, PCCB during LTD 63 ^ 3%, PCCB after high-frequency protocol 70 ^ 3%, n 5). DISCUSSION
In the present work we describe an age-dependent form of synaptic plasticity in the developing rat visual cortex. Soon after eye opening, P17, LTD induced by LFS of white matter was rescued by HFS. This form of plasticity disappears in P23 animals. Visual deprivation (dark-rearing) preserves this plasticity. We further showed that application of BDNF, a neurotrophic factor of the nerve growth factor family well expressed in the visual cortex, restores this plasticity in P23 rats. Reversal of long-term change in synaptic efficacy or maximal expression of synaptic plasticity The capability of a synapse to reverse a given long-term change in synaptic efficacy could be considered as a measure of its plastic degree or, in other words, its maximal expression of plasticity. For example, in the adult hippocampus it has been shown that long-term changes of synaptic effectiveness can be modified in a bi-directional fashion. 14 Here we provide evidence showing that a similar phenomenon is present soon after eye opening in rat visual cortex. In particular, our results demonstrate the presence of a short time window, between P17 and P23, during which the thalamocortical synapse rapidly loses its ability to reverse a given LTD. Although potentiation of depressed responses is no longer inducible at P23, we and others 24 have shown that LTP per se is still present at this age in both virgin preparations and slices already challenged. Indeed, in control experiments LTD did
167
not occlude the possibility to evoke LTP in a different location within the primary visual cortex. This observation could suggest that in spite of their similarity LTP and potentiation after LTD could rely on partially different mechanisms. Alternatively, LTD could trigger an inhibitory mechanism, not yet present or operative at P17, preventing potentiation. In any case, the fact that LTD amplitude does not change significantly between P17 and P23 indicates that potentiation does not depend primarily on the magnitude of LTD. The way in which this plasticity property changes during development points to a fundamental role in the shaping of visual cortical circuitry. Malleability of cortical synapses is present soon after eye opening when selectivity for orientation and movement direction to visual stimuli is almost absent, most cells receive binocular input and their receptive field is highly immature. 15 At this developmental stage neurons in the visual cortex receive many different inputs which they subsequently integrate. The ability to easily revert the sign of information storage coming from a given synaptic input well suits the needs of a visual neuron at this age. The dramatic decline in synaptic flexibility observed in a few days coincides with the functional maturation of the cortical circuitry. For example, the functional properties of visual cortical neurons, such as ocular dominance and orientation selectivity, show a high rate of maturation between P17 and P23, reaching adult-like properties between P26 and P30. 15,34 It has been already suggested that the decline in visual cortical LTP runs in parallel with the maturation of functional properties of rat visual cortex. 24 We propose that the capacity of a synapse to invert a given long-term change in synaptic efficacy represents an intrinsic and crucial neuronal characteristic of an early stage of visual cortex functional development as can be considered P17. As maturation proceeds, under the control of visual input, a gradual decline in the degree of synaptic plasticity occurs, firstly with the loss of the capacity to reverse an LTD. Successively, by P35–P40, when all the visual cortex functional properties are adult like, LTP also disappears, 24 while LTD is still present, albeit at lower amplitude. 32 Thus, although there is no direct experimental evidence for establishing a causal relationship between the main features of cortical development and the different forms of cortical synaptic plasticity, the temporal relationship described here is very intriguing.
Fig. 1. Reversal of LTD at P17 and P23. Averaged ( ^ S.E.M.) responses representing the amplitude of the maximum negative field potential recorded in layer 2/3 following stimulation of the white matter. Values were normalized to average baseline. Bar and arrow indicate the conditioning protocol for LTD (1 Hz, 900 pulses) and LTP (three trains of 1 s at 100 Hz), respectively. (A) Effects of an LTD protocol (bar) followed by an LTP protocol (arrow) in slices of P17 rats. Top traces are representative field potentials taken at the time indicated by the numbers. Horizontal bar 10 ms; vertical bar 0.4 mV. Initial downward deflections are stimulation artifacts. (B) Effects of an LTD protocol (bar) followed by an LTP protocol (arrow) in slices of P23 rats. Top traces are representative field potentials taken at the time indicated by the numbers. Scale bars as in A. Note that at this age it was no longer possible to evoke potentiation after LTD (dedepression).
168
E. Sermasi et al.
Fig. 2. Averaged ( ^ S.E.M.) responses representing the amplitude of the maximum negative field potential recorded in layer 2/3 following stimulation of the white matter. Values were normalized to average baseline. The bar and arrow indicate the conditioning protocol for LTD (1 Hz, 900 pulses) and LTP (three trains of 1 s at 100 Hz), respectively. (A) Control experiments in slices of P23 rats. (1) Effects of an LTD protocol (bar) followed by a tetanic stimulation (arrow) with no LTP as shown in Fig. 1B. The recording electrode was then placed in a new position (2) not previously stimulated (S2). (2) Effects of a tetanic stimulation (arrow) given in S2 eliciting an LTP. Thus at P23 it was possible to induce an LTP in the same slice, but not to potentiate a previously depressed evoked response. S1 and S2 represent the two stimulation sites relative to recording positions 1 and 2. (B) Control experiments in slices of P23 rats. Effects of an LTD protocol (bar) followed by a tetanic stimulation (arrow). The current used for the tetanic stimulation (LTP protocol) was increased to a level able to evoke responses of the same amplitude as the control values. In these conditions only a temporary increment in the responses was observed.
Visual experience and BDNF modulate cortical synaptic plasticity
169
Fig. 3. Effects of an LTD protocol (bar) followed by an LTP protocol (arrow) in slices of rats dark-reared from P17 to P29. In these slices it was again possible to observe potentiation after LTD. Top traces are representative field potentials taken at the time indicated by the numbers. Horizontal bar 10 ms, vertical bar 0.5 mV. Initial downward deflections are stimulation artifacts.
We think it important to relate the plastic behaviour of a synapse to its physiological role during development. In this respect, the maintenance of highly plastic characteristics in the adult hippocampus fits well with its physiological role in learning and memory. 14 Role of visual input in the developmental expression of reversal of long-term depression It is well known that maturation of visual cortical neurons is under the control of visual input. Indeed, a manipulation of binocular visual experience such as dark-rearing slows down the development of functional properties of visual cortical neurons 15 and enlarges the period of expression of cortical plasticity. 27,36,37 In this study we report that dark rearing from P17 to P29 maintains the capacity of synapses to be potentiated after LTD well beyond an
age when this property is normally lost. This loss could therefore be considered part of a developmental process and, in particular, of an activitydependent development which is known to control most of the ontogenic changes in rat visual cortex. It has to be noted that dark-rearing is able to prevent both the loss of LTP in the adult 24,25 and the LTD reversal at P23, indicating that the absence of sensory triggered synaptic activity maintains the plastic properties characterizing the early stage of development. It is interesting to note that, contrary to that observed in animals light-deprived from birth where LTD amplitude is extremely reduced, 25 in rats deprived of sensory input at P17 it was still possible to induce an important LTD. Thus, by using light-imprinted rats we were able to study synapses whose development was already lighttriggered. This allowed us to clarify the role of
170
E. Sermasi et al.
Fig. 4. Effects of an LTD protocol (bar) followed by an LTP protocol (arrow) in slices of P23 rats. In these experiments the recording electrode was filled with a BDNF solution (100 ng/ml in 1 M NaCl). In contrast to that normally observed at this age it was again possible to reverse a depression. Top traces are representative field potentials taken at the time indicated by the numbers. Horizontal bar 10 ms, vertical bar 0.4 mV. Initial downward deflections are stimulation artifacts.
Table 2. Long-term depression and dedepression values obtained at P23 with either brain-derived neurotrophic factor or brainderived neurotrophic factor plus K252a and at P17 ⫹ K252a
LTD De-Dep
P23—BDNF
P23—BDNF ⫹ K252a
P17 ⫹ K252a
73 ^ 6% 125 ^ 8%
70 ^ 1% 67 ^ 5%
63 ^ 3% 70 ^ 3%
Conventions as in Table 1. De-Dep, dedepression.
sensory input on the development of synaptic plasticity without interfering with the maturation process. Exogenously applied brain-derived neurotrophic factor restores reversal of long-term depression Having first established that this form of synaptic plasticity is developmentally regulated in an
activity-dependent manner we then attempted to identify possible factors modulating it. Different factors are known to modulate cortical development and plasticity. 11,16,35 Some of these factors are under the control of visually driven electrical activity and therefore are potentially involved in modulating synaptic cortical plasticity. BDNF represents a likely candidate as it has been reported to be expressed in an activity-dependent manner, 4,9
Visual experience and BDNF modulate cortical synaptic plasticity
171
Fig. 5. Effects of an LTD protocol (bar) followed by an LTP protocol (arrow) in slices of P23 rats. In these experiments the recording electrode was filled with a BDNF solution (100 ng/ml in 1 M NaCl) while slices were perfused with K252a (200 nM). K252a (200 nM) completely blocked any BDNF-induced potentiation after LTD. Top traces are representative field potentials taken at the time indicated by the numbers. Horizontal bar 10 ms, vertical bar 0.6 mV. Initial upward deflections are stimulation artifacts.
together with its receptor trkB 6,10 in rat visual cortex and to control its maturation. 5,28 In addition, BDNF modulates both synaptic currents 8 and LTP elicited by stimulation of cortical layer IV in visual cortex. 1 In this study BDNF was released through the recording pipette. The aim was twofold: on the one hand we wanted to apply BDNF to the portion of the slice from which we were recording, avoiding in this way any general undesired activation of the Trk B receptors expressed throughout all layers of the primary visual cortex. 10 On the other hand, this ensured that enough BDNF was present at the recording site, as it has been shown that BDNF penetrates poorly when applied to the surface of the slice. 30 We have found that an exogenous supply of BDNF is able to restore the capability of cortical synapses to be potentiated after LTD at a postnatal
age when this property is normally lost. Our data indicate an involvement of BDNF in the modulation of this form of synaptic plasticity and suggest a role for endogenous BDNF and/or its receptor trkB. The possible involvement of endogenous Trk receptors in potentiating depressed responses at P17 has been tested by a functional blockade of the tyrosine kinases associated with neurotrophin receptors. We have shown that in these conditions it was no longer possible to potentiate a previously depressed synapse. These results therefore clearly indicate that Trk activation is needed for the expression of this form of synaptic plasticity. More experiments are necessary to clarify the role of endogenous BDNF and/or its receptor trkB and in particular whether a developmental regulation in their expression might be implicated in the disappearance of
172
E. Sermasi et al.
Fig. 6. Effects of an LTD protocol (bar) followed by an LTP protocol (arrow) in slices of P17 rats. In these experiments slices were perfused with K252a (200 nM). K252a (200 nM) completely blocked potentiation after LTD. Top traces are representative field potentials taken at the time indicated by the numbers. Horizontal bar 10 ms, vertical bar 0.4 mV. Initial upward deflections are stimulation artifacts.
dedepression at P23. Interestingly, a recent study by our group shows that the number of neurons expressing the BDNF mRNA in layer IV declines after P18. 7 Dark-rearing from birth, however, maintains the high number of cells expressing BDNF mRNA, even at a later stage of visual cortex development (P23–P25). CONCLUSIONS
This is the first time, to our knowledge, that applied BDNF has been reported to restore the loss
of synaptic plasticity in the brain of normal rats. The data reported here are in accordance with those shown previously in the CA1 region of the hippocampus from BDNF mutant mice, where reexpression of the BDNF gene restored the impaired LTP. 26,30 Moreover, the reported ability of BDNF to orientate dendritic neuronal growth geometry and therefore the neuronal capacity to integrate synaptic inputs support the growing body of evidence on the role of this neurotrophin in developmental processes. 29
REFERENCES
1. Akaneya Y., Tsumoto T., Kinoshita T. S. and Hatanaka H. (1997) Brain-derived neurotrophic factor enhances long-term potentiation in rat visual cortex. J. Neurosci. 17, 6707–6716. 2. Artola A. and Singer W. (1987) Long term potentiation and NMDA receptors in rat visual cortex. Nature 330, 649–652. 3. Boothe R. G., Dobson M. V. and Teller D. Y. (1985) Postnatal development of vision in human and nonhuman primates. A. Rev. Neurosci. 8, 495–545. 4. Bozzi Y., Pizzorusso T., Cremisi F., Rossi F. M., Barsacchi G. and Maffei L. (1995) Monocular deprivation decreases the expression of messenger RNA for brain-derived neurotrophic factor in the rat visual cortex. Neuroscience 69, 1133–1144. 5. Cabelli R. J., Hohn A. and Shatz C. J. (1995) Inhibition of ocular dominance column formation by infusion of NT-4/5 or BDNF. Science 267, 1662–1666.
Visual experience and BDNF modulate cortical synaptic plasticity
173
6. Cabelli R. J., Allendoerfer K. L., Radeke M. J., Welcher A. A., Feinstein S. C. and Shatz C. J. (1996) Changing patterns of expression and subcellular localization of TrkB in the developing visual system. J. Neurosci. 16, 7965–7980. 7. Capsoni S., Tongiorgi E., Cattaneo A. and Domenici L. (1999) Dark rearing blocks the developmental down-regulation of brain-derived neurotrophic factor messenger RNA expression in layers IV and V of the rat visual cortex. Neuroscience 88, 393–403. 8. Carmignoto G., Pizzorusso T., Tia S. and Vicini S. (1996) Brain-derived neurotrophic factor and nerve growth factor potentiate excitatory synaptic transmission in the rat visual cortex. J. Physiol., Lond. 498, 153–164. 9. Castrens E., Zafra F., Thoenen H. and Lindon D. (1992) Light regulates expression of brain-derived neurotrophic factor mRNA in rat visual cortex. Proc. natn. Acad. Sci. U.S.A. 89, 9444–9448. 10. Cellerino A., Maffei L. and Domenici L. (1996) The distribution of brain-derived neurotrophic factor and its receptor trkB in parvalbumin-containing neurons of the rat visual cortex. Eur. J. Neurosci. 8, 1190–1197. 11. Cellerino A. and Maffei L. (1996) The action of neurotrophins in the development and plasticity of the visual cortex. Prog. Neurobiol. 49, 53–71. 12. Domenici L., Harding G. W. and Burkhalter A. (1995) Patterns of synaptic activity in forward and feedback pathways within rat visual cortex. J. Neurophysiol. 74, 2649–2664. 13. Drager U. C. (1978) Observation on monocular deprivation in mice. J. Neurophysiol. 41, (1) 28–42. 14. Dudek S. M. and Bear M. F. (1993) Bi-directional long-term modification of synaptic effectiveness in the adult and immature hippocampus. J. Neurosci. 13, 2910–2918. 15. Fagiolini M., Pizzorusso T., Berardi N., Domenici L. and Maffei L. (1994) Functional postnatal development of the rat primary visual cortex and the role of visual experience: dark rearing and monocular deprivation. Vision Res. 34, 709–720. 16. Fox K. and Zahsk (1994) Critical period control in sensory cortex. Curr. Opin. Neurobiol. 4, 112–119. 17. Fox K. (1995) The critical period for long term potentiation in primary sensory cortex. Neuron 15, 485–488. 18. Hubel D. and Wiesel T. (1963) Receptive fields, binocular interaction and functional architecture in the cat’s visual cortex. J. Neurophysiol. 26, 994–1002. 19. Kang H. and Schuman E. M. (1995) Long-lasting neurotrophin-induced enhancement of synaptic transmission in the adult hippocampus. Science 267, 1658–1662. 20. Katz L. C. and Shatz C. J. (1996) Synaptic activity and the construction of cortical circuits. Science 274, 1133–1138. 21. Kenan-Vaknin G. and Teyler T. J. (1994) Laminar pattern of synaptic activity in rat primary visual cortex: comparison of in vivo and in vitro studies employing the current source density analysis. Brain Res. 635, 37–48. 22. Kirkwood A., Dudek S. M., Gold J. T., Aizenman C. D. and Bear M. F. (1993) Common forms of synaptic plasticity in the hippocampus and neocortex in vitro. Science 260, 1518–1521. 23. Kirkwood A. and Bear M. F. (1994) Homosynaptic long-term depression in the visual cortex. J. Neurosci. 14, 3404–3412. 24. Kirkwood A., Lee H. K. and Bear M. F. (1995) Co-regulation of long-term potentiation and experience-dependent synaptic plasticity in visual cortex by age and experience. Nature 375, 328–331. 25. Kirkwood A., Rioult M. G. and Bear M. F. (1996) Experience-dependent modification of synaptic plasticity in visual cortex. Nature 381, 526–528. 26. Korte M., Griesbeck O., Gravel C., Carroll P., Staiger V., Thoenen H. and Bonhoeffer T. (1996) Virus-mediated gene transfer into hippocampal CA1 region restores long-term potentiation in brain-derived neurotrophic factor mutant mice. Proc. natn. Acad. Sci. U.S.A. 93, 12547–12552. 27. Le Vay S., Wiesel T. N. and Hubel D. H. (1980) The development of ocular dominance columns in normal and visually deprived monkeys. J. comp. Neurol. 191, (1) 1–51. 28. McAllister K. A., Katz L. C. and Lo D. C. (1996) Neurotrophin regulation of cortical dendritic growth requires activity. Neuron 17, 1057–1064. 29. McAllister K. A., Katz L. C. and Lo D. C. (1997) Opposing roles for endogenous BDNF and NT-3 in regulating cortical dendritic growth. Neuron 18, 767–778. 30. Patterson S. L., Abel T., Deuel T., Martin K. C., Rose J. C. and Kandel E. R. (1996) Recombinant BDNF rescues deficits in basal synaptic transmission and hippocampal LTP in BDNF knockout mice. Neuron 16, 1137–1145. 31. Schuman E. (1997) Growth factors sculpt the synapses. Science 275, 1277–1278. 32. Sermasi E., Tropea D. and Domenici L. (1999) LTD is expressed during postnatal development in rat visual cortex: a role for visual experience. Devl Brain Res. (in press). 33. Shatz C. J. (1990) Impulse activity and the patterning of connections during CNS development. Neuron 5, 745–756. 34. Siciliano R., Fontanesi G., Casamenti F., Berardi N., Bagnoli P. and Domenici L. (1997) Postnatal development of functional properties of visual cortical cells in cats with excitotoxic lesions of basal forebrain cholinergic neurons. Vis. Neurosci. 14, 111– 123. 35. Thoenen H. (1995) Neurotrophins and neuronal plasticity. Science 270, 593–598. 36. Timney B., Mitchell D. E. and Giffin F. (1978) The development of vision in cats after extended periods of dark rearing. Expl Brain Res. 31, (4) 547–560. 37. Wiesel T. N. (1982) Postnatal development of the visual cortex and the influence of environment. Nature 299, 583–591. 38. Woodward W. R., Chiaia N., Teyler T. J., Leong L. and Coull B. M. (1990) Organization of cortical afferent and efferent pathways in the white matter of the rat visual system. Neuroscience 36, 393–401. (Accepted 6 October 1998)
Neuroscience Vol. 91, No. 1, pp. 163–173, 1999 Copyright 䉷 1999 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306-4522/99 $20.00+0.00 S0306-4522(98)00598-3
A NEW FORM OF SYNAPTIC PLASTICITY IS TRANSIENTLY EXPRESSED IN THE DEVELOPING RAT VISUAL CORTEX: A MODULATORY ROLE FOR VISUAL EXPERIENCE AND BRAINDERIVED NEUROTROPHIC FACTOR E. SERMASI,* D. TROPEA* and L. DOMENICI†‡ *International School for Advanced Studies (SISSA), Neuroscience Program, Via Beirut 2-4, 34014 Trieste, Italy †Institute of Neurophysiology, CNR, Via San Zeno 51, 56127 Pisa, Italy
Abstract—Synaptic plasticity has been implicated in the mechanisms contributing to the shaping of the cortical circuits responsible for the transmission of the visual input in the rat primary visual cortex. However, the degree of plasticity of the thalamocortical synapse may change during development, perhaps reflecting the degree of stabilization of the circuitry subserving it. We have chosen the ability of this synapse to be first depressed and then potentiated as a specific indicator of its plasticity. In this study we have investigated how this parameter changes during development and the factors controlling it. Extracellular field potentials in cortical layers 2/3 were evoked by stimulation of the white matter in rat primary visual cortex slices prepared at different postnatal ages. Low-frequency stimulation (900 pulses at 1 Hz) of the white matter was used to induce long-term depression of field potential amplitude, whereas long-term potentiation was evoked by high-frequency stimulation consisting of three trains at 100 Hz. We provide evidence that while it is possible to potentiate previously depressed synapses soon after eye opening (postnatal day 17) this synaptic characteristic decreases rapidly thereafter. The decrease in this form of cortical synaptic plasticity closely matches the stabilization of the cortical circuitry towards an adult pattern of connectivity and function. Depressed cortical synapses cannot be potentiated in normal rats at postnatal 23, but they can be potentiated in rats reared in the dark from postnatal days 17 to 29. Moreover, application of brain-derived neurotrophic factor, known to be expressed in an activitydependent manner, was able to restore the ability of synapses to be potentiated after long-term depression, thus indicating its important modulatory role in brain development. 䉷 1999 IBRO. Published by Elsevier Science Ltd. Key words: synaptic plasticity, visual cortex, activity-dependent development, dark-rearing, neurotrophins, brainderived neurotrophic factor.
The mammalian sensory cortices are immature at birth, both anatomically and physiologically, and develop gradually during defined postnatal time windows. 3,13,16–18 The underlying developmental processes are activity dependent 33 and follow the Hebbian rule: the presynaptic inputs that are simultaneously active with the postsynaptic cells are reinforced while the others are lost. For many years the visual cortex has been used as a model to study activity-dependent developmental processes. In the rat visual cortex, for example, neuronal functional properties such as orientation selectivity, receptive field size and ocular dominance are largely immature soon after eye opening [postnatal day (P)14–15], reaching progressively adult properties. 15,34 In particular, orientation selectivity and ‡To whom correspondence should be addressed. Abbreviations: aCSF, artificial cerebrospinal fluid; BDNF, brain-derived neurotrophic factor; DMSO, dimethylsulfoxide; HFS, high-frequency stimulation; LFS, low-frequency stimulation; LTD, long-term depression; LTP, long-term potentiation; P, postnatal day; PCCB, percentage change from control baseline.
ocular dominance distribution develop markedly between P17 and P23, reaching adult-like properties by P26–P30. Visual experience is a necessary requirement for maturation of the cortical connections that subserve the visual function. Indeed, in the rat 15 as well as in other species 27,36,37 dark-rearing from birth slows down the maturation of the cortical circuitry. For example, functional properties of visual cortical neurons such as ocular dominance and orientation selectivity are maintained in an immature state in dark-reared rats. Activity-dependent synaptic plasticity, and in particular long-term changes in synaptic efficacy such as long-term potentiation (LTP) and longterm depression (LTD), have been implicated in the mechanisms controlling the cortical developmental process. 20,31 In the rat primary visual cortex LTP and LTD can be elicited by stimulation of the white matter containing the geniculocortical fibres. Both forms of synaptic plasticity have been reported to be N-methyl-d-aspartate dependent. 2,23 However, while LTD can be evoked throughout the lifespan, LTP expression is restricted to postnatal
163
164
E. Sermasi et al.
development. 22–24,32 In particular, it has been shown that LTP is high soon after eye opening but progressively decreases thereafter and disappears by P35. As a consequence, the degree and characteristics of cortical synaptic plasticity change during postnatal development, possibly reflecting the level of maturation of the circuitry subserving the sensory function. Up to now, work on visual cortex synaptic plasticity has dealt with either LTP or LTD but not the two together. We have chosen to investigate the ability of cortical synapses to be first depressed and then potentiated as a specific indicator of the degree of cortical synaptic plasticity at different developmental ages. This phenomenon has been studied in slices of rat brain containing the primary visual cortex taken at postnatal ages representing different stages of both the maturation of the functional properties of cortical neurons and LTP expression (P17, i.e. two to three days after eye opening, P23 and P35). Similar experiments were also performed in light-deprived animals to verify a potential visual experience dependence of this form of synaptic plasticity. In addition, we have examined the possible role of brain-derived neurotrophic factor (BDNF) on the plastic property of developing cortical synapses. Indeed, neurotrophins have been shown to play a role in the cortical synaptic plasticity of this part of the cortex. 1,11,35 EXPERIMENTAL PROCEDURES
Visual cortex slices preparation and electrophysiological recordings Albino rats (Harlan Nossan) were deeply anaesthetized with urethane and then decapitated. All efforts were made to minimize animal suffering and to reduce the number of animals used. The brain was rapidly removed and coronal slices (400 mm) containing the visual cortex were prepared using a Vibratome (Pelco, U.S.A.) and superfused in a submerged recording chamber at 33⬚C with artificial cerebrospinal fluid (aCSF) at a rate of 4 ml/min. An average of two to three slices per animal (the second and third slices from the occipital pole of each hemisphere) was used for experimental purposes. The aCSF was gassed with 95% O2 and 5% CO2 and had the following composition in mM: NaCl, 126; KCl, 3.5; NaH2PO4, 1.2; MgCl, 1.3; CaCl2, 2; NaHCO3, 25; glucose, 11. Extracellular field potentials in the inferior half of cortical layers 2/3 were recorded with an electrode filled with a 2 M NaCl solution and evoked, by stimulation of the white matter containing geniculocortical fibres, using a bipolar concentric stimulating electrode as previously described by Domenici et al. 12 LTD was induced by low-frequency stimulation (LFS, 900 pulses at 1 Hz) of the white matter. The protocol for inducing LTP consisted of three trains of high-frequency stimulation (HFS, 100 Hz, 1 s). Slices containing the primary visual cortex were prepared from either postnatal day 16–17 (P17) or P23– P24 (P23) or P35 (P35) rats. In dark-rearing experiments, slices were prepared from animals light deprived from P17 to P29, that is for a period considered sufficient to maintain P17 synaptic characteristics. BDNF (Alamone Labs, Israel) was applied by filling the recording electrode with a 100 ng/ ml in 1 M NaCl solution. K252a (Alamone Labs, Israel) was first aliquoted in dimethylsulphoxide (DMSO), then diluted in aCSF to a final concentration of 200 nM and administered to slices by superfusion. All animal experiments were
carried out in accordance with the European Community Council Directive of 24 November 1986 (86/609/EEC). Data analysis The amplitude of the maximum negative field potential in layer 2/3 was used as a measure of the evoked population excitatory current. 25 Changes in the amplitude of the maximum negative field potential mirror changes in the magnitude of a monosynaptic current sink and correlate with changes in the initial slope of the negative potentials. In order to avoid polysynaptic sinks before starting the recording we ensured that field potentials were monophasic and that the shape was not changed by frequency of stimulation higher than 1 Hz. Following this recording protocol the latency of negative field potentials ranged between 5 and 7 ms. To minimize the possibility of widely activating fibres in the white matter the stimulating electrode was placed at the border between the white matter and layer 6 in a way that according to Kenan-Vaknin and Teyler 21 and Woodward et al. 38 would facilitate the selective stimulation of vertical pathways projecting to cortical layers 2/3 and 4. Baseline responses were obtained every 15 s and averaged every four responses with a stimulation intensity that yielded a halfmaximal response. The magnitude of both LTD and potentiation after LTD (dedepression) was measured starting 20 min after the end of the respective conditioning protocol. The mean value of 20 consecutive averaged responses was calculated relative to the baseline averaged control values for each experiment. In the case of failure of potentiation after LTD at P23, baseline responses were followed for at least 20 min unless otherwise stated. Data were then pooled together for each experimental group and expressed as percentage change from control baseline (PCCB) ^ S.E.M. Statistical comparison was done by applying a Mann–Whitney rank sum test among LTD and dedepression PCCB values. Difference was considered significant with P ⬍ 0.05. RESULTS
Reversal of long-term depression at P17 and P23 At P17, having initially induced LTD by LFS (900 pulses at 1 Hz) of the white matter (Table 1, PCCB 69 ^ 4%, n 5) and followed it for 40 min, stable potentiation was reliably elicited by a subsequent HFS (100 Hz, 1 s) in all cases (Fig. 1A, n 5/5). Potentiated responses were equal or higher to the control values recorded prior to the LTD protocol (Table 1, PCCB 105 ^ 10%, n 5). The opposite phenomenon, i.e. the depression of cortical synapses after induction of LTP, was equally observed in slices from P17 rats (LTP 179 ^ 17%, depotentiation 107 ^ 4%, n 3). However, depotentiation was not further investigated at later ages, as it is known that LTP expression declines during development. 24 At P23 (Fig. 1B) the effects of LFS were similar to those evoked at P17 (Table 1, PCCB 62 ^ 4%, n 8). In fact, comparison of LTD amplitude at these two ages revealed no statistically significant difference. However, in sharp contrast to that observed at P17, at P23 it was never possible (8/8) to potentiate the evoked responses after the induction of LTD (Table 1, PCCB 56 ^ 7%, n 8). In order to verify a possible delay in the onset of potentiation after LTD at P23, in two cases the
Visual experience and BDNF modulate cortical synaptic plasticity Table 1. Long-term depression and dedepression at P17 and P23 and in dark-reared rats
LTD De-Dep
P17
P23
Rats dark-reared from P17 to P29
69 ^ 4% 105 ^ 10%
62 ^ 4% 56 ^ 7%
82 ^ 2% 111 ^ 6%
Values are expressed as mean percentage change from control baseline ^ S.E.M. De-Dep, dedepression.
recording was continued beyond 30 min after the tetanic stimulation. However, in accordance with that observed in the other cases, no potentiation was noted. Similar experiments performed in adult animals (P35, n 2, LTD PCCB 70%, dedepression PCCB 66%) confirmed that this form of plasticity is present at P17 but declines rapidly thereafter. Control experiments In control experiments at P23 we checked whether the inability of the synapse to be potentiated after depression was a consequence of a general inability of the visual cortex slice to undergo potentiation. To test this hypothesis, we first repeated the same experimental protocol of evoking LTP having followed an LTD for 40 min (Fig. 2A, position 1). Having verified the lack of potentiation with this electrode position, the recording electrode was moved to a new location in the same slice, within layers 2/3 of the primary visual cortex. In these conditions, after at least 30 min of control stimulation in order to exclude a running down of the ability to undergo LTP, we were able to induce LTP (Fig. 2A, position 2, 5/6 slices). Thus at P23 it was possible in the same slice to induce an LTP but not to potentiate a previously depressed evoked response. In a further set of control experiments at P23, after 40 min of LTD we increased the current used for the HFS of the white matter. In order to avoid any massive stimulation of previously unrecruited fibres, the current was only increased up to a level able to evoke responses of the same amplitude as the control pre-LTD values. HFS with increased current induced in three cases a temporary increment in the responses, which lasted for less than 25 min, but in none of these experiments was it possible to potentiate the depressed responses in the long term (Fig. 2B, n 4/4). Recordings were continued for at least 30 min after the tetanic stimulation. These results demonstrate that the synaptic inability to potentiate previously depressed responses observed at P23 is not due to the small amplitude of the test response per se.
165
Effects of dark-rearing on reversal of long-term depression The loss of the ability to reverse a depression observed at P23 could be part of a developmental process under the control of visual experience. If this hypothesis were true then it could be possible to maintain the P17 thalamocortical synaptic characteristics by blocking sensory input to it. To verify this we reared normal P17 rats in the dark until P29. Experiments performed using visual cortex slices from these animals (Fig. 3) indeed showed that it was again possible to see potentiation of the evoked responses after LTD (Table 1, PCCB 111 ^ 6%, n 5/5). LTD amplitude, although slightly reduced, was still considerable (Table 1, PCCB 82 ^ 2%, n 5). Clearly, we were able to block or delay the developmental loss of this form of synaptic plasticity, suggesting that it relies on visually evoked activity. Effects of brain-derived neurotrophic factor on reversal of long-term depression To examine the possible effect of exogenous BDNF on cortical synaptic plasticity, we filled the recording electrode with a BDNF (100 ng/ml in 1 M NaCl) solution. This method was used in an attempt to concentrate and study BDNF actions in a restricted area of the visual cortex. To ensure leaking of BDNF from the recording electrode we slightly broke the tip of the pipette immediately before the beginning of the recording session. The resistance of BDNF filled electrodes was verified not to be grossly different with respect to that of the electrodes used in control experiments. BDNF released from the pipette induced a gradual increase in the size of the responses, as already shown when BDNF was applied to the slice by perfusion. 1 This effect, not observed in control slices, reached a plateau in around 15 min when control responses started to be recorded. In these experimental conditions we found that at P23 it was again possible to induce LTP after LTD (Fig. 4). LTD size (Table 2, PCCB 73 ^ 6%, n 6) was not significantly reduced in comparison to P23 control slices. In marked contrast, HFS to the white matter was again capable of inducing a stable dedepression (n 6/6) whose magnitude (Table 2, PCCB 125 ^ 8%) was not significantly different from that found at P17. We felt that it was important to check whether applied BDNF was directly responsible for restoring at P23 the potentiation after LTD observed at P17. In order to do this we used K252a, a known inhibitor of neurotrophin receptor tyrosine kinase activity, at a concentration (200 nM) known to block BDNF action on hippocampus and visual cortex synaptic plasticity. 1,19 From the results reported in Fig. 5 it is clear that perfusion of P23 slices with K252a (200 nM) blocked the action of the BDNF contained in the recording electrode
166
E. Sermasi et al.
Fig. 1.
Visual experience and BDNF modulate cortical synaptic plasticity
(Table 2, PCCB during LTD 70 ^ 1%, PCCB after high-frequency protocol 67 ^ 5%, n 4). However, the effect of K252a on BDNF action was specific to the dedepression phase as the amplitude of LTD did not change. These results prompted us to test the possible role of endogenous tyrosine kinase activated by neurotrophins in reversing LTD at P17. To do this we perfused P17 slices with K252a (200 nM) and applied the reversing LTD protocol. Contrary to the observations in control P17 preparations, the presence of K252a blocked the potentiating effects of the tetanic stimulation (Fig. 6) while LTD amplitude did not differ significantly (Table 2, PCCB during LTD 63 ^ 3%, PCCB after high-frequency protocol 70 ^ 3%, n 5). DISCUSSION
In the present work we describe an age-dependent form of synaptic plasticity in the developing rat visual cortex. Soon after eye opening, P17, LTD induced by LFS of white matter was rescued by HFS. This form of plasticity disappears in P23 animals. Visual deprivation (dark-rearing) preserves this plasticity. We further showed that application of BDNF, a neurotrophic factor of the nerve growth factor family well expressed in the visual cortex, restores this plasticity in P23 rats. Reversal of long-term change in synaptic efficacy or maximal expression of synaptic plasticity The capability of a synapse to reverse a given long-term change in synaptic efficacy could be considered as a measure of its plastic degree or, in other words, its maximal expression of plasticity. For example, in the adult hippocampus it has been shown that long-term changes of synaptic effectiveness can be modified in a bi-directional fashion. 14 Here we provide evidence showing that a similar phenomenon is present soon after eye opening in rat visual cortex. In particular, our results demonstrate the presence of a short time window, between P17 and P23, during which the thalamocortical synapse rapidly loses its ability to reverse a given LTD. Although potentiation of depressed responses is no longer inducible at P23, we and others 24 have shown that LTP per se is still present at this age in both virgin preparations and slices already challenged. Indeed, in control experiments LTD did
167
not occlude the possibility to evoke LTP in a different location within the primary visual cortex. This observation could suggest that in spite of their similarity LTP and potentiation after LTD could rely on partially different mechanisms. Alternatively, LTD could trigger an inhibitory mechanism, not yet present or operative at P17, preventing potentiation. In any case, the fact that LTD amplitude does not change significantly between P17 and P23 indicates that potentiation does not depend primarily on the magnitude of LTD. The way in which this plasticity property changes during development points to a fundamental role in the shaping of visual cortical circuitry. Malleability of cortical synapses is present soon after eye opening when selectivity for orientation and movement direction to visual stimuli is almost absent, most cells receive binocular input and their receptive field is highly immature. 15 At this developmental stage neurons in the visual cortex receive many different inputs which they subsequently integrate. The ability to easily revert the sign of information storage coming from a given synaptic input well suits the needs of a visual neuron at this age. The dramatic decline in synaptic flexibility observed in a few days coincides with the functional maturation of the cortical circuitry. For example, the functional properties of visual cortical neurons, such as ocular dominance and orientation selectivity, show a high rate of maturation between P17 and P23, reaching adult-like properties between P26 and P30. 15,34 It has been already suggested that the decline in visual cortical LTP runs in parallel with the maturation of functional properties of rat visual cortex. 24 We propose that the capacity of a synapse to invert a given long-term change in synaptic efficacy represents an intrinsic and crucial neuronal characteristic of an early stage of visual cortex functional development as can be considered P17. As maturation proceeds, under the control of visual input, a gradual decline in the degree of synaptic plasticity occurs, firstly with the loss of the capacity to reverse an LTD. Successively, by P35–P40, when all the visual cortex functional properties are adult like, LTP also disappears, 24 while LTD is still present, albeit at lower amplitude. 32 Thus, although there is no direct experimental evidence for establishing a causal relationship between the main features of cortical development and the different forms of cortical synaptic plasticity, the temporal relationship described here is very intriguing.
Fig. 1. Reversal of LTD at P17 and P23. Averaged ( ^ S.E.M.) responses representing the amplitude of the maximum negative field potential recorded in layer 2/3 following stimulation of the white matter. Values were normalized to average baseline. Bar and arrow indicate the conditioning protocol for LTD (1 Hz, 900 pulses) and LTP (three trains of 1 s at 100 Hz), respectively. (A) Effects of an LTD protocol (bar) followed by an LTP protocol (arrow) in slices of P17 rats. Top traces are representative field potentials taken at the time indicated by the numbers. Horizontal bar 10 ms; vertical bar 0.4 mV. Initial downward deflections are stimulation artifacts. (B) Effects of an LTD protocol (bar) followed by an LTP protocol (arrow) in slices of P23 rats. Top traces are representative field potentials taken at the time indicated by the numbers. Scale bars as in A. Note that at this age it was no longer possible to evoke potentiation after LTD (dedepression).
168
E. Sermasi et al.
Fig. 2. Averaged ( ^ S.E.M.) responses representing the amplitude of the maximum negative field potential recorded in layer 2/3 following stimulation of the white matter. Values were normalized to average baseline. The bar and arrow indicate the conditioning protocol for LTD (1 Hz, 900 pulses) and LTP (three trains of 1 s at 100 Hz), respectively. (A) Control experiments in slices of P23 rats. (1) Effects of an LTD protocol (bar) followed by a tetanic stimulation (arrow) with no LTP as shown in Fig. 1B. The recording electrode was then placed in a new position (2) not previously stimulated (S2). (2) Effects of a tetanic stimulation (arrow) given in S2 eliciting an LTP. Thus at P23 it was possible to induce an LTP in the same slice, but not to potentiate a previously depressed evoked response. S1 and S2 represent the two stimulation sites relative to recording positions 1 and 2. (B) Control experiments in slices of P23 rats. Effects of an LTD protocol (bar) followed by a tetanic stimulation (arrow). The current used for the tetanic stimulation (LTP protocol) was increased to a level able to evoke responses of the same amplitude as the control values. In these conditions only a temporary increment in the responses was observed.
Visual experience and BDNF modulate cortical synaptic plasticity
169
Fig. 3. Effects of an LTD protocol (bar) followed by an LTP protocol (arrow) in slices of rats dark-reared from P17 to P29. In these slices it was again possible to observe potentiation after LTD. Top traces are representative field potentials taken at the time indicated by the numbers. Horizontal bar 10 ms, vertical bar 0.5 mV. Initial downward deflections are stimulation artifacts.
We think it important to relate the plastic behaviour of a synapse to its physiological role during development. In this respect, the maintenance of highly plastic characteristics in the adult hippocampus fits well with its physiological role in learning and memory. 14 Role of visual input in the developmental expression of reversal of long-term depression It is well known that maturation of visual cortical neurons is under the control of visual input. Indeed, a manipulation of binocular visual experience such as dark-rearing slows down the development of functional properties of visual cortical neurons 15 and enlarges the period of expression of cortical plasticity. 27,36,37 In this study we report that dark rearing from P17 to P29 maintains the capacity of synapses to be potentiated after LTD well beyond an
age when this property is normally lost. This loss could therefore be considered part of a developmental process and, in particular, of an activitydependent development which is known to control most of the ontogenic changes in rat visual cortex. It has to be noted that dark-rearing is able to prevent both the loss of LTP in the adult 24,25 and the LTD reversal at P23, indicating that the absence of sensory triggered synaptic activity maintains the plastic properties characterizing the early stage of development. It is interesting to note that, contrary to that observed in animals light-deprived from birth where LTD amplitude is extremely reduced, 25 in rats deprived of sensory input at P17 it was still possible to induce an important LTD. Thus, by using light-imprinted rats we were able to study synapses whose development was already lighttriggered. This allowed us to clarify the role of
170
E. Sermasi et al.
Fig. 4. Effects of an LTD protocol (bar) followed by an LTP protocol (arrow) in slices of P23 rats. In these experiments the recording electrode was filled with a BDNF solution (100 ng/ml in 1 M NaCl). In contrast to that normally observed at this age it was again possible to reverse a depression. Top traces are representative field potentials taken at the time indicated by the numbers. Horizontal bar 10 ms, vertical bar 0.4 mV. Initial downward deflections are stimulation artifacts.
Table 2. Long-term depression and dedepression values obtained at P23 with either brain-derived neurotrophic factor or brainderived neurotrophic factor plus K252a and at P17 ⫹ K252a
LTD De-Dep
P23—BDNF
P23—BDNF ⫹ K252a
P17 ⫹ K252a
73 ^ 6% 125 ^ 8%
70 ^ 1% 67 ^ 5%
63 ^ 3% 70 ^ 3%
Conventions as in Table 1. De-Dep, dedepression.
sensory input on the development of synaptic plasticity without interfering with the maturation process. Exogenously applied brain-derived neurotrophic factor restores reversal of long-term depression Having first established that this form of synaptic plasticity is developmentally regulated in an
activity-dependent manner we then attempted to identify possible factors modulating it. Different factors are known to modulate cortical development and plasticity. 11,16,35 Some of these factors are under the control of visually driven electrical activity and therefore are potentially involved in modulating synaptic cortical plasticity. BDNF represents a likely candidate as it has been reported to be expressed in an activity-dependent manner, 4,9
Visual experience and BDNF modulate cortical synaptic plasticity
171
Fig. 5. Effects of an LTD protocol (bar) followed by an LTP protocol (arrow) in slices of P23 rats. In these experiments the recording electrode was filled with a BDNF solution (100 ng/ml in 1 M NaCl) while slices were perfused with K252a (200 nM). K252a (200 nM) completely blocked any BDNF-induced potentiation after LTD. Top traces are representative field potentials taken at the time indicated by the numbers. Horizontal bar 10 ms, vertical bar 0.6 mV. Initial upward deflections are stimulation artifacts.
together with its receptor trkB 6,10 in rat visual cortex and to control its maturation. 5,28 In addition, BDNF modulates both synaptic currents 8 and LTP elicited by stimulation of cortical layer IV in visual cortex. 1 In this study BDNF was released through the recording pipette. The aim was twofold: on the one hand we wanted to apply BDNF to the portion of the slice from which we were recording, avoiding in this way any general undesired activation of the Trk B receptors expressed throughout all layers of the primary visual cortex. 10 On the other hand, this ensured that enough BDNF was present at the recording site, as it has been shown that BDNF penetrates poorly when applied to the surface of the slice. 30 We have found that an exogenous supply of BDNF is able to restore the capability of cortical synapses to be potentiated after LTD at a postnatal
age when this property is normally lost. Our data indicate an involvement of BDNF in the modulation of this form of synaptic plasticity and suggest a role for endogenous BDNF and/or its receptor trkB. The possible involvement of endogenous Trk receptors in potentiating depressed responses at P17 has been tested by a functional blockade of the tyrosine kinases associated with neurotrophin receptors. We have shown that in these conditions it was no longer possible to potentiate a previously depressed synapse. These results therefore clearly indicate that Trk activation is needed for the expression of this form of synaptic plasticity. More experiments are necessary to clarify the role of endogenous BDNF and/or its receptor trkB and in particular whether a developmental regulation in their expression might be implicated in the disappearance of
172
E. Sermasi et al.
Fig. 6. Effects of an LTD protocol (bar) followed by an LTP protocol (arrow) in slices of P17 rats. In these experiments slices were perfused with K252a (200 nM). K252a (200 nM) completely blocked potentiation after LTD. Top traces are representative field potentials taken at the time indicated by the numbers. Horizontal bar 10 ms, vertical bar 0.4 mV. Initial upward deflections are stimulation artifacts.
dedepression at P23. Interestingly, a recent study by our group shows that the number of neurons expressing the BDNF mRNA in layer IV declines after P18. 7 Dark-rearing from birth, however, maintains the high number of cells expressing BDNF mRNA, even at a later stage of visual cortex development (P23–P25). CONCLUSIONS
This is the first time, to our knowledge, that applied BDNF has been reported to restore the loss
of synaptic plasticity in the brain of normal rats. The data reported here are in accordance with those shown previously in the CA1 region of the hippocampus from BDNF mutant mice, where reexpression of the BDNF gene restored the impaired LTP. 26,30 Moreover, the reported ability of BDNF to orientate dendritic neuronal growth geometry and therefore the neuronal capacity to integrate synaptic inputs support the growing body of evidence on the role of this neurotrophin in developmental processes. 29
REFERENCES
1. Akaneya Y., Tsumoto T., Kinoshita T. S. and Hatanaka H. (1997) Brain-derived neurotrophic factor enhances long-term potentiation in rat visual cortex. J. Neurosci. 17, 6707–6716. 2. Artola A. and Singer W. (1987) Long term potentiation and NMDA receptors in rat visual cortex. Nature 330, 649–652. 3. Boothe R. G., Dobson M. V. and Teller D. Y. (1985) Postnatal development of vision in human and nonhuman primates. A. Rev. Neurosci. 8, 495–545. 4. Bozzi Y., Pizzorusso T., Cremisi F., Rossi F. M., Barsacchi G. and Maffei L. (1995) Monocular deprivation decreases the expression of messenger RNA for brain-derived neurotrophic factor in the rat visual cortex. Neuroscience 69, 1133–1144. 5. Cabelli R. J., Hohn A. and Shatz C. J. (1995) Inhibition of ocular dominance column formation by infusion of NT-4/5 or BDNF. Science 267, 1662–1666.
Visual experience and BDNF modulate cortical synaptic plasticity
173
6. Cabelli R. J., Allendoerfer K. L., Radeke M. J., Welcher A. A., Feinstein S. C. and Shatz C. J. (1996) Changing patterns of expression and subcellular localization of TrkB in the developing visual system. J. Neurosci. 16, 7965–7980. 7. Capsoni S., Tongiorgi E., Cattaneo A. and Domenici L. (1999) Dark rearing blocks the developmental down-regulation of brain-derived neurotrophic factor messenger RNA expression in layers IV and V of the rat visual cortex. Neuroscience 88, 393–403. 8. Carmignoto G., Pizzorusso T., Tia S. and Vicini S. (1996) Brain-derived neurotrophic factor and nerve growth factor potentiate excitatory synaptic transmission in the rat visual cortex. J. Physiol., Lond. 498, 153–164. 9. Castrens E., Zafra F., Thoenen H. and Lindon D. (1992) Light regulates expression of brain-derived neurotrophic factor mRNA in rat visual cortex. Proc. natn. Acad. Sci. U.S.A. 89, 9444–9448. 10. Cellerino A., Maffei L. and Domenici L. (1996) The distribution of brain-derived neurotrophic factor and its receptor trkB in parvalbumin-containing neurons of the rat visual cortex. Eur. J. Neurosci. 8, 1190–1197. 11. Cellerino A. and Maffei L. (1996) The action of neurotrophins in the development and plasticity of the visual cortex. Prog. Neurobiol. 49, 53–71. 12. Domenici L., Harding G. W. and Burkhalter A. (1995) Patterns of synaptic activity in forward and feedback pathways within rat visual cortex. J. Neurophysiol. 74, 2649–2664. 13. Drager U. C. (1978) Observation on monocular deprivation in mice. J. Neurophysiol. 41, (1) 28–42. 14. Dudek S. M. and Bear M. F. (1993) Bi-directional long-term modification of synaptic effectiveness in the adult and immature hippocampus. J. Neurosci. 13, 2910–2918. 15. Fagiolini M., Pizzorusso T., Berardi N., Domenici L. and Maffei L. (1994) Functional postnatal development of the rat primary visual cortex and the role of visual experience: dark rearing and monocular deprivation. Vision Res. 34, 709–720. 16. Fox K. and Zahsk (1994) Critical period control in sensory cortex. Curr. Opin. Neurobiol. 4, 112–119. 17. Fox K. (1995) The critical period for long term potentiation in primary sensory cortex. Neuron 15, 485–488. 18. Hubel D. and Wiesel T. (1963) Receptive fields, binocular interaction and functional architecture in the cat’s visual cortex. J. Neurophysiol. 26, 994–1002. 19. Kang H. and Schuman E. M. (1995) Long-lasting neurotrophin-induced enhancement of synaptic transmission in the adult hippocampus. Science 267, 1658–1662. 20. Katz L. C. and Shatz C. J. (1996) Synaptic activity and the construction of cortical circuits. Science 274, 1133–1138. 21. Kenan-Vaknin G. and Teyler T. J. (1994) Laminar pattern of synaptic activity in rat primary visual cortex: comparison of in vivo and in vitro studies employing the current source density analysis. Brain Res. 635, 37–48. 22. Kirkwood A., Dudek S. M., Gold J. T., Aizenman C. D. and Bear M. F. (1993) Common forms of synaptic plasticity in the hippocampus and neocortex in vitro. Science 260, 1518–1521. 23. Kirkwood A. and Bear M. F. (1994) Homosynaptic long-term depression in the visual cortex. J. Neurosci. 14, 3404–3412. 24. Kirkwood A., Lee H. K. and Bear M. F. (1995) Co-regulation of long-term potentiation and experience-dependent synaptic plasticity in visual cortex by age and experience. Nature 375, 328–331. 25. Kirkwood A., Rioult M. G. and Bear M. F. (1996) Experience-dependent modification of synaptic plasticity in visual cortex. Nature 381, 526–528. 26. Korte M., Griesbeck O., Gravel C., Carroll P., Staiger V., Thoenen H. and Bonhoeffer T. (1996) Virus-mediated gene transfer into hippocampal CA1 region restores long-term potentiation in brain-derived neurotrophic factor mutant mice. Proc. natn. Acad. Sci. U.S.A. 93, 12547–12552. 27. Le Vay S., Wiesel T. N. and Hubel D. H. (1980) The development of ocular dominance columns in normal and visually deprived monkeys. J. comp. Neurol. 191, (1) 1–51. 28. McAllister K. A., Katz L. C. and Lo D. C. (1996) Neurotrophin regulation of cortical dendritic growth requires activity. Neuron 17, 1057–1064. 29. McAllister K. A., Katz L. C. and Lo D. C. (1997) Opposing roles for endogenous BDNF and NT-3 in regulating cortical dendritic growth. Neuron 18, 767–778. 30. Patterson S. L., Abel T., Deuel T., Martin K. C., Rose J. C. and Kandel E. R. (1996) Recombinant BDNF rescues deficits in basal synaptic transmission and hippocampal LTP in BDNF knockout mice. Neuron 16, 1137–1145. 31. Schuman E. (1997) Growth factors sculpt the synapses. Science 275, 1277–1278. 32. Sermasi E., Tropea D. and Domenici L. (1999) LTD is expressed during postnatal development in rat visual cortex: a role for visual experience. Devl Brain Res. (in press). 33. Shatz C. J. (1990) Impulse activity and the patterning of connections during CNS development. Neuron 5, 745–756. 34. Siciliano R., Fontanesi G., Casamenti F., Berardi N., Bagnoli P. and Domenici L. (1997) Postnatal development of functional properties of visual cortical cells in cats with excitotoxic lesions of basal forebrain cholinergic neurons. Vis. Neurosci. 14, 111– 123. 35. Thoenen H. (1995) Neurotrophins and neuronal plasticity. Science 270, 593–598. 36. Timney B., Mitchell D. E. and Giffin F. (1978) The development of vision in cats after extended periods of dark rearing. Expl Brain Res. 31, (4) 547–560. 37. Wiesel T. N. (1982) Postnatal development of the visual cortex and the influence of environment. Nature 299, 583–591. 38. Woodward W. R., Chiaia N., Teyler T. J., Leong L. and Coull B. M. (1990) Organization of cortical afferent and efferent pathways in the white matter of the rat visual system. Neuroscience 36, 393–401. (Accepted 6 October 1998)