| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Journal of Physiology (2001), 535.3, pp. 825-839
© Copyright 2001 The Physiological Society
 |
ABSTRACT |
|---|
 TopAbstract IntroductionMethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Long-term depression (LTD) of excitatory synaptic transmission between parallel fibres (PFs) - the axons of granule cells - and Purkinje cells (PCs) is the most widely studied and best understood form of synaptic plasticity within the cerebellar cortex. LTD is classically induced by repetitive and coincidental activation of PFs with the other excitatory synaptic input to PCs, climbing fibres (CFs) - the axons of neurones located in the inferior olive (Ito & Kano, 1982).
In recent years, it has become apparent that PF-PC synapses can also maintain a form of long-term potentiation (LTP) of synaptic transmission following a brief period of raised frequency stimulation to PFs (Salin et al. 1996). This potentiation resembles that observed in the hippocampus between mossy fibres and CA3 pyramidal cells (Nicoll & Malenka, 1995). Unlike LTD, which is considered to be a post-synaptic process, LTP is thought to result from a pre-synaptic increase in transmitter release via a mechanism involving calcium, cyclic adenosine 3'5'-monophosphate (cAMP) and protein kinase A (PKA).
Raised frequency stimulation of the molecular layer, through which PFs run, also results in the production of the diffusible messenger NO (Shibuki & Kimura, 1997) and a NO-dependent production of cGMP in Purkinje cells (Hartell et al. 2001). Intriguingly, NO production shares a number of similarities with cerebellar LTP in that its production is calcium, cAMP and PKA dependent and its release can be facilitated by tetanic stimulation. The anatomical distribution of NOS within the cerebellar cortex suggests that the PFs themselves are a likely source of NO (Southam et al. 1992; Bredt et al. 1992).
Evidence for a participatory role of NO in cerebellar LTD in brain slices is now generally accepted (Daniel et al. 1993, 1998; Hartell, 1994, 1996a). We have recently provided evidence to suggest that under particular conditions of PF activation, a function of NO may be to spread synaptic depression to distant synapses. Raising the frequency of PF stimulation, at intensities that produce spatially restricted post-synaptic increase in calcium (Eilers et al. 1995), leads to a form of LTD that spreads over tens of micrometres (Hartell, 1996b,c, 2000). Inhibition of NOS prevents this spread of LTD. Although the origin of this depression is probably post-synaptic, the spread of depression must be transcellular since NOS is not present in Purkinje cells (Southam et al. 1992; Schmidt et al. 1992).
Since cAMP and NO have the potential to modulate each other with respect to production (Inada et al. 1998; Polte & Schroder, 1998; Dubey et al. 1998) and given that NO has been demonstrated to have pre-synaptic effects on plasticity elsewhere in the brain (Haley et al. 1992; Schuman & Madison, 1994a,b; Arancio et al. 1996; Son et al. 1996), we have examined whether NO might also play a role in the induction of cerebellar LTP. To this end, we investigated the nature of plasticity following the application of raised frequency PF activation to one of two synaptically independent PF pathways to the same cell. Under conditions designed to suppress LTD by reducing post-synaptic calcium activity, raised frequency PF activation gave rise to a heterosynaptic, NO- and PKA-dependent potentiation of pre-synaptic origin. Although NO was essential for the induction of LTP at both sites, its effect was not mediated by cGMP or protein kinase G (PKG) and NO was not essential for the maintenance of potentiation once induced. Input-specific LTP was observed, however, in the presence of a NO scavenger. These data show for the first time that NO is required for the production of cerebellar LTP and that, as with certain forms of LTD (Hartell, 1996b), NO mediates the lateral spread of synaptic plasticity.
|
METHODS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Experiments were performed on 200 µm thick sagittal slices of cerebellar vermis obtained from 14- to 21-day-old male Wistar rats, decapitated under halothane anaesthesia, using standard techniques approved by the Aston University Bioethics Committee. Slices were incubated at room temperature in artificial cerebro-spinal fluid (aCSF) of the following composition (mM): NaCl 118, KCl 4.7, CaCl2.2H2O 2.5, NaHCO3 25, KH2PO4 1.2, MgSO4.7H2O 1.2, glucose 11, equilibrated with 95 % O2 + 5 % CO2 gas (pH 7.4).
Slices were placed between two nylon nets in a chamber mounted on an upright microscope, fully submerged and perfused with aCSF containing 20 µM picrotoxin at a rate of approximately 2 ml min-1. Purkinje cells were visualized using a 
40 water immersion lens (0.75 NA) and whole cell patch recordings were made from the somata of cells close to the surface of the slice. Two aCSF-filled patch electrodes were positioned in the molecular layer, at separations of between 20 and 168 µm (median 62 µm, mean 76 µm), to activate discrete PF inputs to the same cell. The electrodes were sited equidistant from the soma, at a similar level within the molecular layer, so that they activated regions on opposite sides of the main proximal dendrite.
Patch electrodes, with resistances of between 3 and 5 M
, were filled with solutions of the following composition (mM); potassium gluconate 132, NaCl 8, MgCl2 2, Hepes 30, Na2ATP 4, BAPTA 10, GTP 0.3, adjusted to pH 7.3. Cells were held in voltage clamp mode at a holding potential of -70 mV. Shortly after entering whole cell recording mode, each of the two PF inputs to the cell were activated alternately (50-100 µs, 1-20 V, 0.2 Hz). Stimulus intensities were minimized to limit the amplitude of the PF-excitatory post-synaptic currents (EPSCs) to less than 300 pA to limit the possible entry of calcium through voltage-dependent calcium channels (Hartell, 1996b). EPSC amplitudes were matched as closely as possible between pathways and experiments so that a similar number of PFs were activated in each input pathway. After approximately 10 min, the frequency of stimulation to one input, termed P0, was raised to 8 Hz for 15 s. The second input (designated P1) was not stimulated during this period. After this phase of raised frequency stimulation (RFS), stimulation to both pathways was resumed at 0.2 Hz. In accordance with the observations of Salin et al. (1996), LTP was also reliably induced in the absence of picrotoxin or by 16 Hz RFS. Picrotoxin was included in all but a few preliminary experiments to prevent the activation of 'off-beam' inhibitory currents that were particularly prevalent at large electrode separations.
To establish whether changes in synaptic strength may have had a pre-synaptic origin, pairs of pulses were delivered at 50 ms intervals and the ratio of the second to the first was plotted over time. The extent of pathway independence of P0 and P1 was also checked at intervals during experiments using a protocol previously described (Hartell, 1996b). Briefly, pathways P0 and P1 were activated at a 50 ms interval and after a 150 ms delay, the stimulation order was reversed. If P0 and P1 shared a significant number of PFs, then activation of P0 should cause some facilitation of P1 and vice versa. In the event of a 100 % overlap, the level of facilitation of P1 when preceded by P0, and P0 when preceded by P1, should approach the paired pulse ratio (PPR) of P0-P0 or P1-P1. Conversely, complete pathway independence should mean that the amplitude of P0 when preceded by P1 should remain unchanged compared to a naive response and vice versa. The following equations were used to estimate the relative levels of pathway overlap between P0-P1 (left) and P0-P0 (right) at various time points throughout experiments.
| (1) |
Data were acquired and partially analysed on-line with 'The LTP Program' (Anderson & Collingridge, 1999). Additional off-line analysis was performed with custom procedures written with Igor Pro (Wavemetrics Inc., Lake Oswego, OR, USA). Changes in synaptic strength were monitored by measuring the EPSC amplitudes and expressing these changes as percentages of the mean baseline levels measured over 10 min prior to RFS. Mann-Whitney U or Wilcoxon signed-rank tests were performed to determine the level of statistical significance between groups of data. P values less than 0.05 were considered significant.
mEPSCs were recorded in the presence of aCSF containing a raised level of calcium (5 mM) and 1 µM TTX. mEPSCs were collected using pCLAMP 7 and analysed off-line using Igor Pro. Selection criteria were kept constant throughout analyses and mEPSCs with amplitudes and areas exceeding -10 pA and 40 fC, respectively, were accepted. Data sets were collected over 5 min epochs before and after drug application and compared statistically using the Kolmogorov-Smirnov test. Statistical significance was taken to be at the 5 % level. In all experiments, input and electrode resistances were monitored throughout and experiments were discarded from the analysis if either of these measurements changed significantly over the course of the experiment.
H-89, forskolin and TTX were supplied by Sigma. H-[1,2,4]oxadiazolo[4,3-a]quinoxaline-1-one (ODQ), 7-nitro-indazole (7-NI), N G-nitro-L-arginine methyl ester (L-NAME), spermine NONOate, 2-(4-carboxyphenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide (cPTIO) and DL-2-amino-5-phosphonopentanoic acid (AP5) were obtained from Tocris Cookson. Spermine NONOate, cPTIO, AP5 and L-NAME were dissolved directly in aCSF. All other compounds were dissolved in DMSO to final concentrations of less than 0.1 % DMSO.
|
RESULTS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Pathway-specific RFS induces heterosynaptic, PKA-sensitive potentiation accompanied by a reduction in paired pulse facilitation
We initially examined the extent to which synaptic potentiation induced by raised frequency PF stimulation remained input specific. Excitatory post-synaptic currents (EPSCs) elicited by alternate, 0.2 Hz stimulation of two spatially distinct PF inputs were recorded from the soma of Purkinje cells held in voltage clamp mode at a holding potential of -70 mV. After a period of 10 min, during which responses remained stable in amplitude, the frequency of stimulation to one of the two inputs (designated P0) was raised to 8 Hz for 15 s. This protocol was termed raised frequency stimulation (RFS). The other input (P1) was not activated during this period. Following RFS, alternate stimulation to P0 and P1 was resumed at 0.2 Hz.
To reduce the likelihood of simultaneous LTD induction, which is dependent upon post-synaptic calcium, 10 mM BAPTA was included in the patch pipette. Under these conditions, RFS to P0 resulted in a significant potentiation of PF EPSCs above baseline levels at both pathways (Fig. 1A, C and D). The mean EPSC amplitudes of P0 (146.8 ± 9.7 %) and P1 responses (152.4 ± 9.5 %) measured 20 min after RFS were statistically indistinguishable from each other but significantly elevated compared to pre-RFS baseline levels (P < 0.05, Wilcoxon signed-rank test, n = 6; Fig. 1D).
 |
View larger version [in this window] [in a new window] |
|
|
Figure 1. RFS to one of two PF inputs induces PKA-dependent LTP in Purkinje cells that is associated with a decrease in the PPR in both pathways A, once baseline responses to 0.2 Hz activation of two separate PF inputs to Purkinje cells were stable, one pathway, designated P0 ( | ||
To help establish the origin of the potentiation, pairs of pulses at a 50 ms interval were applied to P0 and P1 throughout the experiments. Under this paired pulse protocol, the second response is facilitated compared to the first. Paired pulse facilitation is associated with a residual calcium transient in the PF terminal after the first action potential (Atluri & Regehr, 1996) and an increase in transmitter release (Zucker, 1989). Changes in the paired pulse ratio (PPR) of the second pulse to the first can be effected by known modulators of transmitter release and are generally taken to reflect changes in the release probability (Dittman & Regehr, 1996). RFS to P0 led not only to potentiation in both pathways but also to a decrease in the PPR in both pathways (Fig. 1B, C and E). The normalized PPRs of P0 and P1 responses 20 min after RFS were significantly reduced, compared to baseline levels, to 94.6 ± 0.9 and 94.5 ± 2.9 %, respectively (P < 0.05, n = 6, Wilcoxon signed-rank test; Fig. 1E). In contrast to the potentiation, which was sustained for the duration of the recordings (up to 60 min), the decrease in paired pulse facilitation tended to return towards baseline levels within approximately 25 min of RFS (Fig. 1B). This might indicate an early and a late phase LTP mediated by different mechanisms.
To confirm that the potentiation we observed was mechanistically similar to that previously described (Salin et al. 1996), we repeated the experiments with the PKA inhibitor H-89 in the extracellular bathing media at 0.2 µM. At this concentration, H-89 is 15-20 times more selective for PKA than protein kinase C or calcium- calmodulin-dependent protein kinase (Kawasaki et al. 1998). Inhibition of PKA prevented potentiation in P0 and P1 pathways (99.4 ± 4.8 and 103.4 ± 4.3 %, respectively, P < 0.05, n = 6, Mann-Whitney U test, Fig. 1D) and no reduction in PPR was seen (100.2 ± 2.6 and 99.9 ± 3.0 %, respectively, P < 0.05, n = 6, Fig. 1E) after 20 min. Potentiation was also observed in neurones containing 0.5 mM EGTA, provided cells were hyperpolarized to -90 mV during the period of RFS (Fig. 1D and E). After 20 min, P0 and P1 responses were potentiated to 130 ± 9.3 and 129.3 ± 14.9 %, respectively (P < 0.05, n = 6). Without either hyperpolarization or 10 mM BAPTA in the recording pipette, potentiation was small and transient and concealed by a more dominant, post-synaptic, cGMP-dependent long-term depression of synaptic responses (Jacoby & Hartell, 1999).
Heterosynaptic LTP is not due to pathway overlap
There are several possible explanations for our observation that the potentiation induced by RFS to P0 under conditions of reduced post-synaptic calcium activity does not remain pathway specific but spreads to the distant P1 site. The first is that RFS leads to a generalized increase in the post-synaptic sensitivity of AMPA receptors on the Purkinje cell. In view of the accompanying reduction in the PPR and hence the likely pre-synaptic origin of cerebellar LTP, a more feasible explanation is that RFS leads to a generalized increase in transmitter release. This could arise either through an increase in the probability of transmitter release, an increase in the number of transmitter release sites per fibre and/or an increase in the number of contributing fibres. Although the majority of PF-stimulating electrodes were placed at separations of 60-70 µm, well outside the distance where overlap might be expected at the stimulus intensities used (Reynolds & Hartell, 2000; Wang et al. 2000), we tested for the possibility that the pathways were not completely independent in case the apparent loss of input specificity resulted from a significant number of synapses being activated by both P0 and P1, either prior to or following the induction of LTP.
To this end, we used a modified paired pulse protocol that allowed us to estimate the degree of overlap between the two pathways before and after LTP induction (Hartell, 1996b). The principle is described in the experimental procedures section and illustrated in Fig. 2. The degree of overlap between P0 and P1 was estimated 5 min before RFS-induced potentiation and again 5-10, 15-20 and 30 min after using the equation described in Methods. Figure 2 provides a representative example in which this test for pathway independence was carried out 5 min prior to RFS. Measurements taken during the baseline period revealed that P0 had no influence on P1 (0 % of the P1-P1 PPR; n = 5) and vice versa (0 % of the P0-P0 PPR; n = 5). Therefore, we can be confident that P0 and P1 did not share a significant number of fibres prior to RFS-induced potentiation. The estimated percentage overlap measured 5 min after RFS did not show any significant increase (3 and 6 % respectively; n = 5) and remained constant at this level for all remaining measurements up to 30 min post induction.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 2. P0 and P1 display pathway independence A, parallel fibre pathways P0 and P1 to a single Purkinje cell were activated at 50 ms intervals and again in reverse order 150 ms later. When preceded by activation of the alternate pathway, P0 and P1 EPSC amplitudes remained similar to naïve responses, in this and in four other examples. B, the second response to P1-P0 stimulation (*) and P0-P1 stimulation (#) shown in A are superimposed on P0-P0 and P1-P1 responses, respectively. C, the ratios of the percentage level of potentiation compared to baseline levels observed in P0 compared to that in P1 are plotted against the separations between electrodes for a group of 15 cells. The dashed line represents the fitted linear regression. | ||
LTP can spread more than 150 µm from the site of RFS
We next examined the spatial extent of the spread of LTP from the site of RFS to distant synapses. In a separate group of 15 cells, the level of input specificity between P0, which received RFS, and P1, which did not, was assessed by calculating the ratio of the percentage potentiation in each pathway (LTP P0/P1) 20 min after RFS. An increase in LTP P0/P1 ratio above a value of 1 will reflect an increase in the degree of input specificity. At electrode separations ranging between 20 and 168 µm we did not find any evidence to suggest that LTP became more input specific as the electrode separation increased (Fig. 2C; r = 0.01, P > 0.05). Since the electrode separation did not influence the degree of input specificity over this range, we also examined whether the number of PFs activated during RFS affected the degree of input specificity. No correlation between P0 EPSC amplitude (range 190-300 pA) and the P0/P1 potentiation ratio was found (r = 0.29, P > 0.05).
Nitric oxide mediates the spread of LTP through a cGMP-independent mechanism
It has recently been shown that NO release from the molecular layer of the cerebellar cortex can be potentiated following tetanic stimulation and that this potentiation requires PKA activation (Kimura et al. 1998). In view of the fact that NO has been implicated in the spread of synaptic depression at PF-PC synapses (Hartell, 1996b) and may be released as a result of high frequency PF activation (Shibuki & Kimura, 1997), we examined whether NO was also responsible for the spread of RFS-induced LTP. With 10 mM BAPTA inside the recording pipette and with the relatively selective neuronal NOS inhibitor 7-NI (Moore & Handy, 1997) in the bathing media at a concentration of 5 µM, RFS failed to induce synaptic potentiation in either pathway (Fig. 3A and C ). The amplitudes of P0 and P1 EPSCs measured 20 min after RFS were statistically different from those measured without NOS inhibition (Fig. 3C), reaching 100.4 ± 7.8 and 100.6 ± 6.3 % of baseline levels respectively (P < 0.05, n = 6, Mann-Whitney U test). As illustrated in Fig. 3B and D no reduction in the PPR was observed in either pathway (99.4 ± 3.3 and 101.4 ± 5.0 % of baseline respectively, P < 0.05, n = 6, Mann-Whitney U test). Since this result contradicted an earlier study in which N G-nitro-L-arginine (L-NARG) failed to block LTP (Salin et al. 1996), we examined the effects of a more membrane-permeable, non-selective NOS inhibitor (L-NAME) (Pfeiffer et al. 1996). L-NAME (100 µM) similarly blocked LTP and the associated decrease in the PPR (n = 5, Fig. 3C and D). We conclude that the failure of the previous study to block LTP with L-NARG most likely reflects poor potency due to its low membrane permeability. Application of 5 µM 7-NI to slices 5 min after RFS failed to prevent or reverse potentiation in either pathway (142.0 ± 13.5 and 138.2 ± 20.0 % of baseline levels, respectively, after 20 min, n = 4, Fig. 3C) and had no effect on the associated decrease in the PPR (91.7 ± 1.9 and 94.4 ± 2.5 % of baseline levels, respectively, after 20 min, n = 4, Fig. 3D).
 |
View larger version [in this window] [in a new window] |
|
|
Figure 3. RFS-induced LTP requires NOS activity A, inhibition of NOS with 7-NI in the extracellular perfusate prevented LTP in pathways P0 and P1. B, the associated mean changes in PPR over time are shown. C, a comparison of P0 and P1 responses recorded under standard control conditions of 10 mM intracellular BAPTA and in the presence of 7-NI and L-NAME. Application of 5 µM 7-NI to the bathing medium 5 min after RFS had no effect on the extent of LTP in either P0 or P1. D, the associated changes in PPR are shown. The means and S.E.M. of 6 (7-NI), 5 (L-NAME) and 4 (7-NI application 5 min after RFS) cells are shown. Asterisks indicate a statistical difference between test and control conditions (Mann-Whitney U test, *P < 0.05; **P < 0.01). | ||
The NO scavenger cPTIO (Yoshida et al. 1994) was next used to establish whether transcellular diffusion of NO was necessary for LTP at P0 and/or P1 pathways. With 30 µM cPTIO in the extracellular perfusate, RFS to P0 produced a clear potentiation of EPSCs and a decrease in the PPR that did not spread to P1 (Fig. 4A-E). The amplitudes of P0 (131.5 ± 8.7 %) and P1 (96.6 ± 6.1 %) EPSCs measured 20 min after RFS were statistically different from each other (P < 0.01; Wilcoxon signed-rank test). In this experimental group, electrode separations ranged between 40 and 157 µm (mean 72 µm; median 68 µm, n = 7) and input-specific potentiation was observed in all cases. These data indicate that trans-cellular diffusion of NO is required for the spread of LTP to distant synapses but not for the induction of LTP at synapses specifically activated by RFS. This strongly suggests that RFS results in the generation of NO within PFs.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 4. RFS induces input-specific LTP in the presence of the NO scavenger cPTIO With 10 mM BAPTA in the patch pipette and 30 µM cPTIO in the extracellular perfusate, RFS to pathway P0 led to an input-specific, long-lasting potentiation of P0 synaptic responses (A) and an input-specific reduction in the paired pulse ratio (B). Examples of representative P0 and P1 EPSCs taken at times a and b indicated in panel A are shown below (C). D, the means and S.E.M. of P0 ( | ||
Addition of the selective guanylate cyclase inhibitor ODQ (Garthwaite et al. 1995), at a concentration of 5 µM, to the perfusion medium did not prevent RFS-induced potentiation in either pathway (Fig. 5A). P0 (132.9 ± 7.6 %) and P1 (141.2 ± 13.3 %) responses remained significantly elevated above baseline 20 min after RFS (Fig. 5C; P < 0.05, n = 6, Wilcoxon signed-rank test). The potentiation was also accompanied by a significant reduction in PPR in both pathways (P < 0.05, n = 6; Fig. 5B and D). Intracellular infusion with 5 µM ODQ also failed to prevent potentiation and the reduction in PPR (Fig. 5C and D, n = 6). Intracellular infusion of 500 nM KT5823, which is reported to have an 8- and 10-fold selectivity for PKG over PKC and PKA, respectively, at this concentration (Nakanishi, 1989), also failed to prevent potentiation of P0 (124.5 ± 4.5 %, n = 6) and P1 (124.0 ± 6.1 %) responses or the associated decreases in PPR (90.5 ± 2.7 and 94.6 ± 0.9 %). The same concentrations of ODQ and KT5823 were, however, capable of suppressing the LTD that emerged following 8 Hz RFS when post-synaptic calcium was not chelated with 10 mM BAPTA and an underlying potentiation was unmasked (Jacoby & Hartell, 1999). In the presence of ODQ, P0 and P1 responses were significantly elevated to 131.0 ± 23.9 and 119.0 ± 8.7 % of baseline levels 20 min after RFS (n = 6, P < 0.05). P0 and P1 responses reached 120.5 ± 5.5 and 119.2 ± 4.1 % of baseline (n = 6, P < 0.05) after 20 min in the presence of KT5823. Therefore, whereas guanylate cyclase and PKG are necessary for cerebellar LTD, they do not mediate the actions of NO in cerebellar LTP.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 5. RFS-induced LTP does not require guanylate cyclase Inhibition of guanylate cyclase with 5 µM extracellular ODQ did not prevent LTP (A) or the associated changes in PPR (B) in either P0 or P1. Similarly, neither intracellular ODQ (5 µM) nor the PKG inhibitor KT5823 (500 nM) prevented LTP or the decrease in PPR (D). Potentiation (C) and the reduction in PPR (D) was prevented in both pathways when the NMDA receptor antagonist AP5 (50 µM) was present in the perfusate. In all cases, the means and S.E.M. of six cells are shown. | ||
Since NMDA receptor activation has been shown to trigger NO-dependent cGMP production in the cerebellar cortex, and given that NR1 (Petralia et al. 1994b) and NR2 (Petralia et al. 1994a) NMDA receptor subunits exist on PF terminals, we also examined whether NMDA receptor activation was necessary for cerebellar LTP. Addition of 50 µM AP5 to the bathing media did not produce any discernible effect on the amplitudes or kinetics of PF EPSCs. It did, however, prevent the induction of LTP and the associated drop in PPR in both pathways (n = 6, Fig. 5C and D).
To examine further the mechanism of heterosynaptic potentiation, we used the adenylyl cyclase activator forskolin (Seamon & Daly, 1986), which induces a pharmacological form of LTP that resembles RFS-induced LTP in that it is PKA sensitive and it occludes further potentiation by tetanic stimulation of parallel fibres (Salin et al. 1996). As shown in Fig. 6A, bath application of 10 µM forskolin for 10 min to P0 and P1 responses activated at a constant rate of 0.2 Hz led to a gradual increase in EPSC amplitudes that persisted after washout. P0 and P1 responses rose significantly to 149.1 ± 17.0 and 152.3 ± 16.6 % of baseline levels respectively after 20 min (P < 0.05, n = 6, Wilcoxon signed-rank test; Fig. 6C). As with RFS-induced LTP, this potentiation was accompanied by a significant reduction in PPR in both pathways (Fig. 6D).
 |
View larger version [in this window] [in a new window] |
|
|
Figure 6. Forskolin induces LTP A, 10 min bath application of 10 µM forskolin led to a sustained potentiation of P0 and P1 responses. B, inhibition of NOS with extracellular 7-NI prevented forskolin-induced LTP. C, a comparison of P0 and P1 responses recorded in the presence of forskolin and additionally with 7-NI. Asterisks indicate statistical difference between test and control conditions (Mann-Whitney U test): **P < 0.01, *P < 0.05. D, the associated changes in normalized PPR are illustrated. The means and S.E.M. of six cells are shown. | ||
In view of the fact that both NO and cAMP/PKA are required for LTP, we next undertook a series of experiments designed to establish whether NO activated cAMP/PKA or vice versa. In the presence of 5 µM 7-NI, 10 µM forskolin failed to induce potentiation. P0 and P1 responses after 20 min were 104.4 ± 3.0 and 103.0 ± 6.3 % of baseline - significantly different from responses recorded in the absence of NOS inhibition (P < 0.01, n = 6, Mann-Whitney U test; Fig. 6B and C). No significant decrease in the PPR was observed compared to control data (97.4 ± 4.2 and 96.1 ± 3.2 %, P < 0.05, n = 6, Mann-Whitney U test, Fig. 6D). This result might indicate that cAMP/PKA activates NOS but does not rule out the alternative possibility that both systems are required for a long-lasting potentiation. We therefore investigated if NO application alone was capable of inducing potentiation. Bath application for 10 min of the NO donor spermine NONOate (Maragos et al. 1991) resulted in an increase in PF responses (150.8 ± 13.7 and 152.4 ± 9.1 %, n = 6) measured 20 min after NO application (Fig. 7A and C). Spermine NONOate-induced potentiation was also accompanied by a reduction in PPR (94.1 ± 2.9 and 94.3 ± 4.0 %, n = 6). This NO donor-induced potentiation was resistant to PKA inhibition. In the presence of H-89, although RFS failed to induce potentiation, subsequent application of spermine NONOate resulted in a significant potentiation of PF EPSCs above baseline (130.6 ± 19.2 and 130.5 ± 15.8 %, P < 0.05, n = 6, Wilcoxon signed-rank test, Fig. 7C, n = 6) and a decrease in PPR (92.7 ± 4.5 and 96.3 ± 3.4 %, Fig. 7C, n = 6). A representative example is shown in Fig. 8. These data support the conclusion that the cAMP/PKA cascade is required for NOS activation and that the two systems most likely work in series rather than in concert.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 7. Application of NO results in LTP A, 10 min bath application of the NO donor spermine NONOate induced an increase of responses in both pathways. B, illustration of the associated change in PPR over time. C, a comparison of P0 and P1 responses measured 20 min after RFS under control conditions of 10 mM BAPTA and 20 min after NONOate application in the absence and the presence of H-89. | ||
 |
View larger version [in this window] [in a new window] |
|
|
Figure 8. NO donor-induced potentiation does not require PKA In the presence of 0.2 µM H-89, 15 s RFS (arrow) failed to induce potentiation. Subsequent application of 100 µM spermine NONOate (horizontal bar) produced a potentiation of P0 and P1 responses. | ||
Having confirmed that forskolin is capable of inducing potentiation and that this potentiation is NOS sensitive, we next attempted to establish the site of potentiation by measuring the frequency and amplitudes of mEPSCs before and after forskolin application in the presence of 1 µM TTX. Figure 9A, B and C illustrates that application of forskolin led to a significant increase in the frequency of mEPSCs over 4 min epochs measured 15 min after application compared to similar baseline periods prior to application (P < 0.01, Kolmogorov-Smirnov test). Control experiments revealed no change in mEPSC frequency or amplitude over the same time course (Fig. 9A).
 |
View larger version [in this window] [in a new window] |
|
|
Figure 9. Forskolin increases mEPSC frequency via NOS and NO activities A, changes in mEPSC frequency over time expressed as a percentage of baseline levels under control conditions of 10 mM intracellular BAPTA. The effects of application of 10 µM forskolin and 100 µM NONOate are shown. B, the frequency distributions of mEPSCs over two 4 min periods, prior to ( | ||
We next tested whether extracellular application of 7-NI was capable of preventing forskolin-mediated increases in the frequency of mEPSCs. Figure 9B and C reveals that forskolin failed to significantly increase mEPSC frequency in the presence of 7-NI. Moreover, application of spermine NONOate mimicked the effect of forskolin alone (Fig. 9A-C). Although forskolin and NONOate produced small increases in the amplitudes of mEPSCs, perhaps indicating some post-synaptic actions, these increases were small compared to the substantial increases in mEPSC frequency. On average, forskolin and NONOate increased mEPSC amplitudes to 108 and 107 % of pre-drug baseline levels whereas synaptic responses increased to approximately 150 and 130 % of baseline, respectively. Therefore, these data further suggest that forskolin-induced LTP is mediated primarily through a pre-synaptic process that requires NOS activation.
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Our results concur with previous data obtained in cerebellar slices (Salin et al. 1996; Chen & Regehr, 1997) and in culture (Linden, 1997, 1998) in three main ways. First, raised frequency stimulation to PFs induced a long-lasting potentiation of synaptic transmission at the PF-PC synapse provided post-synaptic calcium activity was buffered. Second, LTP was associated with a reduction in paired pulse facilitation, suggesting a pre-synaptic locus. Third, LTP was blocked by inhibition of PKA and mimicked by forskolin, which promotes cAMP production by activating adenylate cyclase.
Our data reveal a number of additional, important properties of cerebellar LTP. Most notably synaptic potentiation, induced even under these conservative stimulus conditions, was not restricted to the discrete set of synapses that received raised frequency activation. Without exception, increases in synaptic strength were also observed at synapses up to the maximum range tested of 168 µm from the point of RFS. Potentiation at both P0 and P1 sites was accompanied by a reduction in paired pulse facilitation suggesting a pre-synaptic locus in each case. Long-term synaptic potentiation was prevented by NOS inhibitors and mimicked by applications of forskolin or the NO donor spermine NONOate. That these pharmacological forms of LTP were also associated with a NOS-sensitive increase in mEPSC frequency further supports a pre-synaptic site of origin for cerebellar LTP. In contrast to heterosynaptic LTD (Hartell, 2000), neither cGMP nor PKG were required for heterosynaptic LTP. Input-specific LTP was, however, clearly evident in the presence of the NO scavenger cPTIO. These data show for the first time that cerebellar LTP, like LTD (Hartell, 1996b, 2000; Reynolds & Hartell, 2000; Wang et al. 2000), does not remain input specific and that nitric oxide plays a crucial role in the induction and lateral spread of potentiation to distant PF synapses impinging on the same cell.
We have previously shown that the level and the degree of spread of NO-dependent cGMP production in Purkinje cells are dependent on the frequency and the intensity of PF stimulation (Hartell et al. 2001). Therefore, it is likely that the extent of the spread of NO-dependent cerebellar LTP is also a function of the number and frequency of PFs activated. Even though stimulus strengths were limited so as to activate fewer than an estimated 20-25 PFs per pathway (see Barbour, 1993), we did not detect any obvious decline in the spread of LTP up to the maximum inter-electrode separation tested of 168 µm. This suggests that raised frequency activation of just a few PFs may influence synaptic transmission in a substantial proportion of synapses to a given Purkinje cell. As well as spreading within a single Purkinje cell, LTP could equally spread laterally to influence neighbouring cells and, since PFs contact more than one Purkinje cell as they pass through the molecular layer, LTP may spread along a beam of fibres. Wang et al. (2000), using similar levels of PF activation, attempted to quantify the extent of the spread of cerebellar LTD. They found that the sensitivity of AMPA receptors to the release of caged glutamate declined gradually over distances up to approximately 100 µm. Although it is tempting to speculate that NO is also responsible for the spread of cerebellar LTD, direct evidence for this is, as yet, only circumstantial (Hartell, 1996b, 2000; Reynolds & Hartell, 2001). Nevertheless, if this proves to be the case, the frequency of PF stimulation is likely to influence the degree of spread of NO. Moreover, any difference in the apparent spread of LTD and LTP could reflect a differential sensitivity of the pre-synaptic machinery regulating transmitter release and post-synaptic AMPA receptors to the direct and/or indirect actions of NO.
Origin of heterosynaptic cerebellar LTP
Although there are an increasing number of reports of synaptic plasticity spreading beyond individual synapses (Vincent & Marty, 1993; Schuman & Madison, 1994a; Fitzsimonds et al. 1997; Reynolds & Hartell, 2000), it is important to first exclude other anomalous explanations that might account for the apparent loss of input specificity that accompanied potentiation in our model. The most obvious of which is that our two PF inputs, P0 and P1, were not independent but essentially comprised the same or substantially overlapping sets of PF inputs. Using a modified paired pulse facilitation technique to test for pathway independence, we found no evidence of pathway overlap in any example prior to LTP induction over the entire range of electrode separations and stimulus intensities used. Moreover, we found little or no evidence that pathway overlap increased subsequent to LTP induction. Therefore, we can be reasonably confident that the spread of LTP represents genuine, heterosynaptic potentiation. This is further confirmed by the clear appearance of input-specific LTP over a similar range of electrode separations and EPSC amplitudes in the presence of cPTIO.
Several pieces of evidence, in this and in earlier reports, support the view that potentiation at the PF-PC synapse is primarily of pre-synaptic origin. LTP induced through raised frequency PF activation is accompanied by a reduction in paired pulse facilitation in slices (Salin et al. 1996). In culture, RFS induces a form of LTP in granule cell-PC pairs that is accompanied by a reduction in the subsequent EPSC failure rate as well as a decrease in paired pulse facilitation (Linden, 1998). Potentiation of transmitter release can also be detected in neighbouring glial cells (Linden, 1997). Pharmacological activation of the cAMP/PKA cascade with forskolin (Chen & Regehr, 1997) or application of the NO donor spermine NONOate induced a form of potentiation that shares a number of properties with synaptically induced LTP and leads to an increase in mEPSC frequency.
Assuming then that LTP at this synapse is pre-synaptic, there are two ways in which transmitter release may effectively be enhanced. Either the probability (p) of transmitter release for a given pre-synaptic stimulus may increase and/or the number (n) of release sites may rise (Zucker, 1973, 1989; Manabe et al. 1993; Chen & Regehr, 1997; Atluri & Regehr, 1998). In the latter case, this could result from an increase in release sites from the same number of activated fibres or through an increase in the number of fibres activated, i.e. a reduction in firing threshold. We directly tested for this last possibility by comparing the degree of pathway overlap before and after LTP induction in both P0 and P1 pathways. Although the levels of potentiation at P0 and P1 were essentially identical, we found little evidence of any significant increase in pathway overlap after LTP induction compared to that before. This is consistent with observations that tetanic- or forskolin-induced LTP are not accompanied by increases in the size (Salin et al. 1996) or the waveform of the PF volley (Chen & Regehr, 1997). However, whether cAMP acts preferentially to increase the probability of release and/or the number of release sites at this or at other synapses in the CNS remains a matter of debate (Trudeau et al. 1996; Chavis et al. 1998). Although we were unable to detect a change in pathway overlap following LTP induction, we did detect a decrease in paired pulse facilitation at synapses distant from the site of RFS, leading us to conclude that a widespread change in the probability of release most probably provides the causative mechanism for heterosynaptic potentiation. Input specificity of LTP and PPR reduction were, however, maintained in the presence of the NO scavenger cPTIO. These observations lead us to the conclusion that RFS triggers NO production and this diffusible messenger is responsible for the spread of potentiation by facilitating transmitter release from PFs.
Cellular mechanism of heterosynaptic LTP
Salin et al. (1996) previously concluded that NMDA receptors were not involved in cerebellar LTP. Their argument was based upon the observations that Purkinje cells in animals over 14 days old do not express NMDA receptors and because potentiation emerged intact upon washout of kynurenate when RFS was applied in the presence of this non-specific glutamate antagonist. More recently, it was shown that NMDA receptors, which are located on pre-synaptic PF terminals (Petralia et al. 1994a,b), may enhance calcium influx associated with PF activity leading to NOS activation and consequently a depression of synaptic transmission through a post-synaptic, cGMP-dependent process (Casado et al. 2000). In the light of this study, we tested directly whether NMDA receptors contributed to the RFS-induced LTP reported here and we found that potentiation was indeed prevented by the NMDA receptor antagonist AP5 (Fig. 5C and D).
It has been argued that the involvement of cAMP and PKA in LTP stems from the calcium sensitivity of adenylyl cyclases. At least two calcium-calmodulin-sensitive isoforms of adenylyl cyclase, types I (AC1) and VIII (AC8), are expressed in the cerebellar cortex (Xia et al. 1991). Whilst activity of both isoforms is required for late phase LTP in the hippocampal mossy fibre pathway (Wong et al. 1999), a single knockout of AC1 appears to be sufficient to prevent LTP in granule cell-Purkinje cell pairs in culture (Storm et al. 1998). Our current observation that NOS is also critically required for the induction but not the maintenance of LTP now provides an additional mechanism that might contribute not merely to the pre-synaptic calcium sensitivity of LTP in the cerebellar cortex, but also to its lateral spread.
There are a number of ways in which calcium, NO and cAMP/PKA could interact to produce LTP. Forskolin-induced LTP is not accompanied by an increase in influx or basal levels of calcium in pre-synaptic PF terminals (Chen & Regehr, 1997). It is unlikely, then, that a sustained or prolonged calcium signal subserves the expression of LTP, either directly or indirectly through prolonged activation of NO or cAMP/PKA. Therefore, given that both neuronal NOS and AC1 are both calcium sensitive, it is more probable that the transient calcium influx following RFS triggers either a serial interaction of NO and cAMP/PKA or alternatively, these two messenger pathways might both be required to produce a potentiation of transmitter release.
Forskolin-induced potentiation was prevented when NOS activity was blocked. Therefore, NO must either act downstream of cAMP/PKA or both messengers must be required for LTP. The latter possibility may be less likely given that the NO donor NONOate was capable of inducing potentiation, even in the presence of PKA blockade, rather suggesting that cAMP/PKA stimulates NO production. Indeed, forskolin has been shown to potentiate NO release from the molecular layer of the cerebellar cortex and tetanus-induced potentiation of NO release is sensitive to PKA inhibition (Kimura et al. 1998). How then might PKA activate NOS? Although neuronal NOS has several phosphorylation sites that are recognized by kinases (Bredt et al. 1992), including PKA and PKC (Okada, 1995), direct evidence of PKA modification of NOS activity is limited (Inada et al. 1998, 1999) and controversial (Brune & Lapetina, 1991; Bredt et al. 1992). PKA could, like PKC, modulate the calcium sensitivity of NOS (Okada, 1995) or stimulate NO production through an indirect mechanism of which several have been described (Polte & Schroder, 1998; Dubey et al. 1998). However, it is important to note that since the actions of spermine NONOate were less persistent than those of forskolin, cAMP-dependent NO production might provide a mechanism for the short-term enhancement of transmitter release but additional cAMP-dependent mechanisms might be necessary to consolidate longer-term potentiation.
Application of 7-NI after LTP induction did not reverse potentiation, at least within a 5 min window of RFS. Therefore the expression of potentiation is not due to a sustained elevation of NO production but to a relatively transient increase. Moreover, LTP was not sensitive to inhibition of either extracellular or intracellular guanylate cyclase or to inhibition of PKG, indicating that the actions of NO in mediating potentiation do not take place via cGMP or PKG. This is consistent with the proposed pre-synaptic origin of potentiation and the post-synaptic location of guanylate cyclase (Ariano et al. 1982), but different from LTP in the CA3 region of the hippocampus where the actions of NO are thought to be dependent, at least in part, on cGMP. If not via cGMP, how else might NO enhance pre-synaptic transmitter release? NO has been shown to stimulate calcium-independent vesicular release (Meffert et al. 1994) possibly via a direct interaction with proteins involved in vesicle docking/fusion (Meffert et al. 1996).
Based upon our findings, we propose the following working model. RFS causes a transient increase in calcium levels in PF terminals. This triggers an increase in adenylate cyclase activity, leading to the production of cAMP and activation of PKA. The increase in calcium may simultaneously activate NOS in PF terminals and in the presence of PKA, NOS activity is facilitated. NO potentiates transmitter release not only at the site of generation but by also diffusing over distances of tens of micrometres to produce pre-synaptic potentiation at distant synapses. Although we cannot rule out the possibility that other NOS-containing interneurones, such as basket cells, could contribute to the spread of LTP, our experiments with the NO scavenger cPTIO indicate that LTP can be induced at the site of RFS when transcellular diffusion of NO is prevented. This predicts that PFs themselves are capable of generating NO in response to RFS and that NO must work locally within the parallel fibre terminals to enhance transmitter release. Although there are a number of reports of NO acting as an anterograde or retrograde mediator of LTD and LTP (Schuman & Madison, 1994a; Arancio et al. 1996; Hartell, 1996b, 2000; Holscher, 1997; Reynolds & Hartell, 2000), this is, to the best of our knowledge, the first example of NO acting as a 'laterograde' facilitator of pre-synaptic transmission.
This hypothesis predicts that potentiation at the site of RFS and at distant sites will differ in that PKA will be activated only at those synapses that receive high frequency stimulation. One could conceive, therefore, of a mechanism by which cAMP might trigger, as well as a temporary NO-dependent potentiation, a series of events that lead to longer-term, transcriptional or phosphorylation-based changes in pre-synaptic signalling. In other models of pre-synaptic, cAMP-dependent LTP, molecules such as cAMP response element binding protein (CREB) and tissue plasminogen activator (tPA) have been shown to contribute to longer-term facilitation of transmission (Baranes et al. 1998; Casadio et al. 1999). Whilst there is, as yet, no direct evidence for their involvement in cerebellar LTP, a recent study has demonstrated a role for CREB in a late phase form of LTD expressed post-synaptically in cultured Purkinje cells (Ahn et al. 1999). It is conceivable that NO might play a role in an early phase of LTP that could be consolidated in a more input- or synapse-specific way at a later stage. NO could act to increase the spatial dimension of synaptic plasticity in the short term before longer term, more spatially discrete events requiring protein synthesis can take effect.
Based upon our earlier studies, we also predict that NO might have a dual role in mediating potentiation through pre-synaptic, cGMP-independent events and synaptic potentiation, via actions on post-synaptic guanylate cyclase. If the calcium levels in Purkinje cells rise during raised frequency stimulation, then post-synaptic depression, which is also mediated by NO, may take place. If calcium levels do not increase, then pre-synaptic potentiation will predominate. The fact that both heterosynaptic LTD (Hartell, 1996b, 2000; Reynolds & Hartell, 2000; Wang et al. 2000) and LTP have now been demonstrated following raised frequencies of PF activation suggests that the generally accepted view that cerebellar plasticity is input specific at the cellular level requires re-evaluation.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
| AHN S., GINTY, D. D. & LINDEN, D. J. (1999). A late phase of cerebellar long-term depression requires activation of CaMKIV and CREB. Neuron 23, 559-568 | [Medline] |
| ANDERSON W. W. & COLLINGRIDGE, G. L. (1999). A data acquisition program for on-line analysis of long-term potentiation and long-term depression. Society for Neuroscience Abstracts 23, 665 | |
| ARANCIO O., KIEBLER, M., LEE, C. J., LEVRAM, V., TSIEN, R. Y., KANDEL, E. R. & HAWKINS, R. D. (1996). Nitric-oxide acts directly in the presynaptic neuron to produce long-term potentiation in cultured hippocampal-neurons. Cell 87, 1025-1035 | [Medline] |
| ARIANO M. A., LEWICKI, J. A., BRANDWEIN, H. J. & MURAD, F. (1982). Immunohistochemical localization of guanylate cyclase within neurons of rat brain. Proceedings of the National Academy of Sciences of the USA 79, 1316-1320 | [Medline] |
| ATLURI P. P. & REGEHR, W. G. (1996). Determinants of the time course of facilitation at the granule cell to Purkinje cell synapse. Journal of Neuroscience 16, 5661-5671 | [Abstract/Full Text] |
| ATLURI P. P. & REGEHR, W. G. (1998). Delayed release of neurotransmitter from cerebellar granule cells. Journal of Neuroscience 18, 8214-8227 | [Abstract/Full Text] |
| BARANES D., LEDERFEIN, D., HUANG, Y. Y., CHEN, M., BAILEY, C. H. & KANDEL, E. R. (1998). Tissue plasminogen activator contributes to the late phase of LTP and to synaptic growth in the hippocampal mossy fiber pathway. Neuron 21, 813-825 | [Medline] |
| BARBOUR B. (1993). Synaptic currents evoked in Purkinje cells by stimulating individual granule cells. Neuron 11, 759-769 | [Medline] |
| BREDT D. S., FERRIS, C. D. & SNYDER, S. H. (1992). Nitric oxide synthase regulatory sites. Phosphorylation by cyclic AMP-dependent protein kinase, protein kinase C, and calcium/ calmodulin protein kinase; identification of flavin and calmodulin binding sites. Journal of Biological Chemistry 267, 10976-10981 | [Abstract] |
| BRUNE B. & LAPETINA, E. G. (1991). Phosphorylation of nitric oxide synthase by protein kinase A. Biochemical and Biophysical Research Communications 181, 921-926 | [Medline] |
| CASADIO A., MARTIN, K. C., GIUSTETTO, M., ZHU, H. X., CHEN, M., BARTSCH, D., BAILEY, C. H. & KANDEL, E. R. (1999). A transient, neuron-wide form of CREB-mediated long-term facilitation can be stabilized at specific synapses by local protein synthesis. Cell 99, 221-237 | [Medline] |
| CASADO M., DIEUDONNE, S. & ASCHER, P. (2000). Presynaptic N-methyl-D-aspartate receptors at the parallel fiber-Purkinje cell synapse. Proceedings of the National Academy of Sciences of the USA 97, 11593-11597 | [Abstract/Full Text] |
| CHAVIS P., MOLLARD, P., BOCKAERT, J. & MANZONI, O. (1998). Visualization of cyclic AMP-regulated presynaptic activity at cerebellar granule cells. Neuron 20, 773-781 | [Medline] |
| CHEN C. F. & REGEHR, W. G. (1997). The mechanism of cAMP-mediated enhancement at a cerebellar synapse. Journal of Neuroscience 17, 8687-8694 | [Abstract/Full Text] |
| DANIEL H., HEMART, N., JAILLARD, D. & CREPEL, F. (1993). Long-term depression requires nitric oxide and guanosine 3'5'cyclic monophosphate production in rat cerebellar Purkinje cells. European Journal of Neuroscience 5, 1079-1082 | [Medline] |
| DANIEL H., LEVENES, C. & CREPEL, F. (1998). Cellular mechanisms of cerebellar LTD. Trends in Neurosciences 21, 401-407 | [Medline] |
| DITTMAN J. S. & REGEHR, W. G. (1996). Contributions of calcium-dependent and calcium-independent mechanisms to presynaptic inhibition at a cerebellar synapse. Journal of Neuroscience 16, 1623-1633 | [Abstract] |
| DUBEY R. K., GILLESPIE, D. G. & JACKSON, E. K. (1998). Cyclic AMP-adenosine pathway induces nitric oxide synthesis in aortic smooth muscle cells. Hypertension 31, 296-302 | [Medline] |
| EILERS J., AUGUSTINE, G. J. & KONNERTH, A. (1995). Subthreshold synaptic Ca2+ signalling in fine dendrites and spines of cerebellar Purkinje neurons. Nature 373, 155-158 | [Medline] |
| FITZSIMONDS R. M., SONG, H. & POO, M. (1997). Propagation of activity-dependent synaptic depression in simple neural networks. Nature 388, 439-448 | [Medline] |
| GARTHWAITE J., SOUTHAM, E., BOULTON, C. L., NIELSON, E. B., SCHMIDT, K. & MAYER, B. (1995). Potent and selective-inhibition of nitric oxide-sensitive guanylyl cyclase by 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one. Molecular Pharmacology 48, 184-188 | [Abstract] |
| HALEY J. E., WILCOX, G. L. & CHAPMAN, P. F. (1992). The role of nitric oxide in hippocampal long-term potentiation. Neuron 8, 211-216 | [Medline] |
| HARTELL N. A. (1994). cGMP acts within cerebellar Purkinje cells to produce long term depression via mechanisms involving PKC and PKG. NeuroReport 5, 833-836 | [Medline] |
| HARTELL N. A. (1996a). Inhibition of cGMP breakdown promotes the induction of cerebellar long-term depression. Journal of Neuroscience 16, 2881-2890 | [Abstract/Full Text] |
| HARTELL N. A. (1996b). Strong activation of parallel fibers produces localized calcium transients and a form of LTD that spreads to distant synapses. Neuron 16, 601-610 | [Medline] |
| HARTELL N. A. (1996c). Two separate pathways for cerebellar LTD: NO-dependent and NO-independent. Behavioral and Brain Sciences 19, 453 | |
| HARTELL N. A. (2000). Receptors, second messengers and protein kinases required for heterosynaptic cerebellar long-term depression. Neuropharmacology 40, 148-161 | |
| HARTELL N. A., FURUYA, S., JACOBY, S. & OKADA, D. (2001). Intercellular action of nitric oxide increases cGMP in cerebellar Purkinje cells. NeuroReport 12, 25-28 | [Medline] |
| HOLSCHER C. (1997). Nitric oxide, the enigmatic neuronal messenger: Its role in synaptic plasticity. Trends in Neurosciences 20, 298-303 | [Medline] |
| INADA H., SHINDO, H., TAWATA, M. & ONAYA, T. (1998). cAMP regulates nitric oxide production and ouabain sensitive Na+,K+-ATPase activity in SH-SY5Y human neuroblastoma cells. Diabetologia 41, 1451-1458 | [Medline] |
| INADA H., SHINDO, H., TAWATA, M. & ONAYA, T. (1999). Cilostazol, a cyclic AMP phosphodiesterase inhibitor, stimulates nitric oxide production and sodium potassium adenosine triphosphatase activity in SH-SY5Y human neuroblastoma cells. Life Sciences 65, 1413-1422 | [Medline] |
| ITO M. & KANO, M. (1982). Long-lasting depression of parallel fiber-Purkinje cell transmission induced by conjunctive stimulation of parallel fibres and climbing fibers in the cerebellar cortex. Neuroscience Letters 33, 253-258 | [Medline] |
| JACOBY S. & HARTELL, N. A. (1999). Raised frequency parallel fibre stimulation can produce both nitric oxide-dependent LTP and LTD in the rat cerebellar cortex. Journal of Physiology 520.P, 38P | |
| KAWASAKI H., SPRINGETT, G. M., MOCHIZUKI, N., TOKI, S., NAKAYA, M., MATSUDA, M., HOUSMAN, D. E. & GRAYBIEL, A. M. (1998). A family of cAMP-binding proteins that directly activate Rap1. Science 282, 2275-2279 | |
| KIMURA S., UCHIYAMA, S., TAKAHASHI, H. E. & SHIBUKI, K. (1998). cAMP-dependent long-term potentiation of nitric oxide release from cerebellar parallel fibers in rats. Journal of Neuroscience 18, 8551-8558 | [Abstract/Full Text] |
| LINDEN D. J. (1997). Long-term potentiation of glial synaptic currents in cerebellar culture. Neuron 18, 983-994 | [Medline] |
| LINDEN D. J. (1998). Synaptically evoked glutamate transport currents may be used to detect the expression of long-term potentiation in cerebellar culture. Journal of Neurophysiology 79, 3151-3156 | [Abstract/Full Text] |
| MANABE T., WYLLIE, D. A., PERKEL, D. J. & NICOLL, R. A. (1993). Modulation of synaptic transmission and long-term potentiation - effects on paired-pulse facilitation and EPSC variance in the CA1 region of the hippocampus. Journal of Neurophysiology 70, 1451-1459 | [Medline] |
| MARAGOS C. M., MORLEY, D., WINK, D. A., DUNAMS, T. M., SAAVEDRA, J. E., HOFFMAN, A., BOVE, A. A., ISAAC, L., HRABIE, J. A. & KEEFER, L. K. (1991). Complexes of NO with nucleophiles as agents for the controlled biological release of nitric-oxide - vasorelaxant effects. Journal of Medicinal Chemistry 34, 3242-3247 | [Medline] |
| MEFFERT M. K., CALAKOS, N. C., SCHELLER, R. H. & SCHULMAN, H. (1996). Nitric oxide modulates synaptic vesicle docking fusion reactions. Neuron 16, 1229-1236 | [Medline] |
| MEFFERT M. K., PREMACK, B. A. & SCHULMAN, H. (1994). Nitric oxide stimulates Ca(2+)-independent synaptic vesicle release. Neuron 12, 1235-1244 | [Medline] |
| MOORE P. K. & HANDY, R. C. (1997). Selective inhibitors of neuronal nitric oxide synthase - Is no NOS really good NOS for the nervous system? Trends in Pharmacological Sciences 18, 204-211 | [Medline] |
| NICOLL R. A. & MALENKA, R. C. (1995). Contrasting properties of two forms of long-term potentiation in the hippocampus. Nature 377, 115-118 | [Medline] |
| OKADA D. (1995). Protein kinase C modulates calcium sensitivity of nitric oxide synthase in cerebellar slices. Journal of Neurochemistry 64, 1298-1304 | [Abstract] |
| PETRALIA R. S., WANG, Y. X. & WENTHOLD, R. J. (1994a). The NMDA receptor subunits NR2A and NR2B show histological and ultrastructural localization patterns similar to those of NR1. Journal of Neuroscience 14, 6102-6120 | [Abstract] |
| PETRALIA R. S., YOKOTANI, N. & WENTHOLD, R. J. (1994b). Light and electron microscope distribution of the NMDA receptor subunit NMDAR1 in the rat nervous system using a selective anti-peptide antibody. Journal of Neuroscience 14, 667-696 | [Abstract] |
| PFEIFFER S., LEOPOLD, E., SCHMIDT, K., BRUNNER, F. & MAYER, B. (1996). Inhibition of nitric oxide synthesis by NG-nitro-L-arginine methyl ester (L-NAME): requirement for bioactivation to the free acid, NG-nitro-L-arginine. British Journal of Pharmacology 118, 1433-1440 | [Abstract] |
| POLTE T. & SCHRODER, H. (1998). Cyclic AMP mediates endothelial protection by nitric oxide. Biochemical and Biophysical Research Communications 251, 460-465 | [Medline] |
| REYNOLDS T. & HARTELL, N. A. (2000). An evaluation of the synapse specificity of long-term depression induced in rat cerebellar slices. Journal of Physiology 527, 563-577 | [Abstract/Full Text] |
| REYNOLDS T. & HARTELL, N. A. (2001). Roles for nitric oxide and arachidonic acid in the induction of heterosynaptic cerebellar LTD. NeuroReport 12, 133-136 | [Medline] |
| SALIN P. A., MALENKA, R. C. & NICOLL, R. A. (1996). Cyclic-AMP mediates a presynaptic form of LTP at cerebellar parallel fiber synapses. Neuron 16, 797-803 | [Medline] |
| SCHMIDT H. H. H. W., GAGNE, G. D., NAKANE, M., POLLOCK, J. S., MILLER, M. F. & MURAD, F. (1992). Mapping of neural nitric oxide synthase in the rat suggests frequent co-localization with NADPH diaphorase but not soluble guanylyl cyclase, and novel paraneural functions for nitrinergic signal transduction. Journal of Histochemistry and Cytochemistry 40, 1439-1456 | [Abstract] |
| SCHUMAN E. M. & MADISON, D. V. (1994a). Locally distributed synaptic potentiation in the hippocampus. Science 263, 532-536 | |
| SCHUMAN E. M. & MADISON, D. V. (1994b). Nitric oxide and synaptic function. Annual Review of Neuroscience 17, 153-183 | [Medline] |
| SEAMON K. B. & DALY, J. W. (1986). Forskolin: its biological and chemical properties. Advances in Cyclic Nucleotide and Protein Phosphorylation Research 20, 1-150 | [Medline] |
| SHIBUKI K. & KIMURA, S. (1997). Dynamic properties of nitric oxide release from parallel fibres in rat cerebellar slices. Journal of Physiology 498, 443-452 | [Abstract] |
| SON H., HAWKINS, R. D., MARTIN, K., KIEBLER, M., HUANG, P. L., FISHMAN, M. C. & KANDEL, E. R. (1996). Long-term potentiation is reduced in mice that are doubly mutant in endothelial and neuronal nitric oxide synthase. Cell 87, 1015-1023 | [Medline] |
| SOUTHAM E., MORRIS, R. & GARTHWAITE, J. (1992). Sources and targets of nitric oxide in rat cerebellum. Neuroscience 137, 241-244 | |
| STORM D. R., HANSEL, C., HACKER, B., PARENT, A. & LINDEN, D. J. (1998). Impaired cerebellar long-term potentiation in type I adenylyl cyclase mutant mice. Neuron 20, 1199-1210 | [Medline] |
| TRUDEAU L. E., EMERY, D. G. & HAYDON, P. G. (1996). Direct modulation of the secretory machinery underlies PKA-dependent synaptic facilitation in hippocampal neurons. Neuron 17, 789-797 | [Medline] |
| VINCENT P. & MARTY, A. (1993). Neighboring cerebellar Purkinje cells communicate via retrograde inhibition of common presynaptic interneurons. Neuron 11, 885-893 | [Medline] |
| WANG S. S., KHIROUG, L. & AUGUSTINE, G. J. (2000). Quantification of spread of cerebellar long-term depression with chemical two-photon uncaging of glutamate. Proceedings of the National Academy of Sciences of the USA 97, 8635-8640 | [Abstract/Full Text] |
| WONG S. T., ATHOS, J., FIGUEROA, X. A., PINEDA, V. V., SCHAEFER, M. L., CHAVKIN, C. C., MUGLIA, L. J. & STORM, D. R. (1999). Calcium-stimulated adenylyl cyclase activity is critical for hippocampus-dependent long-term memory and late phase LTP. Neuron 23, 787-798 | [Medline] |
| XIA Z. G., REFSDAL, C. D., MERCHANT, K. M., DORSA, D. M. & STORM, D. R. (1991). Distribution of mRNA for the calmodulin-sensitive adenylate cyclase in rat brain: expression in areas associated with learning and memory. Neuron 6, 431-443 | [Medline] |
| YOSHIDA M., AKAIKE, T., WADA, Y., SATO, K., IKEDA, K., UEDA, S. & MAEDA, H. (1994). Therapeutic effects of imidazolineoxyl N-oxide against endotoxin shock through its direct nitric oxide-scavenging activity. Biochemical and Biophysical Research Communications 202, 923-930 | [Medline] |
| ZUCKER R. S. (1973). Changes in the statistics of transmitter release during facilitation. Journal of Physiology 229, 787-810 | [Medline] |
| ZUCKER R. S. (1989). Short-term synaptic plasticity. Annual Review of Neuroscience 12, 13-31 | [Medline] |
Acknowledgements
The authors would like to thank Dr Heather Cater for critically reading the manuscript and the Biotechnology and Biological Sciences Research Council, the Medical Research Council, the Royal Society of Great Britain and the Nuffield Foundation for financial support.
Corresponding author
N. A. Hartell: The Pharmaceutical Science Research Institute, School of Life and Health Sciences, Aston University, Aston Triangle, Birmingham B4 7ET, UK.
Email: n.a.hartell{at}aston.ac.uk
This article has been cited by other articles:
|
|
 |
|
  G. C. Petzold, S. Haack, O. von Bohlen und Halbach, J. Priller, T.-N. Lehmann, U. Heinemann, U. Dirnagl, and J. P. Dreier Nitric Oxide Modulates Spreading Depolarization Threshold in the Human and Rodent Cortex * Supplemental Materials and Methods Stroke, April 1, 2008; 39(4): 1292 - 1299. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D.-l. Qiu and T. Knopfel An NMDA Receptor/Nitric Oxide Cascade in Presynaptic Parallel Fiber-Purkinje Neuron Long-Term Potentiation J. Neurosci., March 28, 2007; 27(13): 3408 - 3415. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 |
), was activated at a raised frequency of 8 Hz for 15 s at time 0 (RFS, arrow). The other pathway (P1; 
) was not activated during this period. Stimulation to both pathways was then resumed at 0.2 Hz. With 10 mM BAPTA included in the patch pipette, RFS led to a long-lasting increase in synaptic responses in both pathways. B, LTP was accompanied by a reduction in the paired-pulse ratio in both pathways. Data are expressed as the percentage change in the ratio of the second pulse to the first compared to baseline levels. C, a single, representative example of EPSCs elicited by stimulation of P0 (top) and P1 (bottom) at times a and b indicated in panel A. The amplitudes of the first pulse in P0 and P1 were normalized to illustrate the accompanying reduction in the PPR (a + b). D, the means and standard errors of P0 (
) and P1 normalized EPSCs (
) measured 20 min after RFS are shown under conditions of 10 mM intracellular BAPTA, BAPTA plus 0.2 µM extracellular H-89 and following cell hyperpolarization to -90 mV during RFS with 0.5 mM EGTA inside the recording pipette. E, the associated changes in PPR measured at the same points in time are shown. In graphs (A, B, D and E ), the means and S.E.M. of six experiments are shown. Asterisks indicate a statistical difference between test and control conditions (Mann-Whitney U test, P < 0.05).