Exp Brain Res (1992) 90:253-261

Experimental BrainResearch 9 Springer-Verlag1992

Induction of high frequency activity in the somatosensory thalamus of rats in vivo results in long-term potentiation of responses in SI cortex S.M. Lee and F.F. Ebner* Center for Neural Science and Section of Neurobiology, Brown University, Providence, RI 02912, USA Received October 18, 1991 /Accepted March 19, 1992

Summary. Extracellular single-unit techniques were employed to record unitary activity simultaneously from the thalamic ventral posterior medial (VPM) nucleus and the ipsilateral primary somatosensory cortex of adult rats. Cross-correlation analysis triggered by the spontaneous firing of thalamocortical relay neurons in VPM and the discharge of layer IV neurons in the corresponding ipsilateral cortical barrel indicated that the paired-units included in this study were strongly correlated in their activity. The baseline responses of highly correlated cortical/thalamic pairs to a 10 ms deflection of a vibrissa on the contralateral side were measured using poststimulus time histograms. After establishing the baseline response, high frequency activity in VPM was induced in one of two ways: i) direct electrical stimulation of thalamic neurons or ii) whisker stimulation in the presence of bicuculline methiodide (BIC) released near the thalamic neurons. Both methods resulted in a conditioning stimulus (CS) paradigm consisting of "bursts" of highfrequency activity (50-100 Hz) with an inter-burst interval of 150 ms ( ~ 7 Hz). Almost immediately following the presentation of the CS, the response of layer IV cortical neurons to vibrissa stimulation increased by 37-62 % over baseline values, which was maintained after the effects of BIC had worn off in VPM. This enhancement in the response of the cortical neurons was not accompanied by a concomitant increase in the thalamic responses. Thus, these results strongly suggest that the potentiation first occurred at the thalamocortical synapse.

Key words: Long-term potentiation (LTP) - Ventral posterior medial- Thalamus - Somatosensory cortex - Bicuculline - in vivo

* Present address: Institute for Developmental Neuroscience, Vanderbilt University, Box 152 GPC, Nashville, TN 37203, USA Correspondence to: F.F. Ebner

Introduction

A number of studies, primarily using in vitro slice preparations, have indicated that a high frequency stimulation of the white matter in adult neocortex leads to an enhancement in the evoked responses which may last for several hours (Lee 1982; Artola and Singer 1987; Bindman et al. 1988; Perkins and Teyler 1988; Wilson 1984; Berry et al. 1989; Artola et al. 1990; Lee et al. 1992a). The stimulation frequencies used to induce LTP have varied from 2 to 50 Hz, and the magnitude of the enhancement has ranged between 20-217% over pre-tetanus values. Despite the variation in the degree of potentiation, a consistent finding from these studies has been that LTP can be induced in neocortex if the level of excitability in neocortical slices is sufficiently elevated. The threshold for LTP can be achieved most readily in mature neocortex by introducing low concentrations of bicuculline, a GABAA receptor antagonist (Artola and Singer 1987, 1991 ; Bindman et al. 1988) or by reducing the extracellular concentration of M g 2+ (Lee et al. 1991). However, the effectiveness of these parameters for inducing LTP in neocortical slices has not been fully investigated in the intact animal. The pioneering studies of Bliss and his co-workers (Bliss and Lomo 1973; Bliss and Gardner-Medwin 1973) showed that a high frequency electrical stimulation leads to a long-term modification of synaptic efficacy within the adult vertebrate central nervous system. They reported that granule cells in the dentate gyrus of anesthetized rabbits displayed a 40% increase in the amplitude of the population spike when brief conditioning stimuli of 10-20 Hz for 10-15 s or 100 Hz for 3-4 s were presented to the perforant pathway. To our knowledge, a comparable demonstration of LTP in the intact sensory neocortex of adult animals has not been reported. Despite the paucity of in vivo data, the results from several in vitro studies have offered conclusive evidence that the presentation of tetanizing stimuli to the white matter leads to a long-term modification in the responsiveness of cortical neurons in layers II/III (Artola and

254 Singer 1987; B i n d m a n et al. 1988; A r t o l a et al. 1990; Lee et al. 1991), layer IV (Lee 1982) and layer V (Perkins and Teyler 1988; Berry et al. 1989). H o w e v e r , the question that remains u n a n s w e r e d is h o w the types o f n e u r o n a l activity which lead to the i n d u c t i o n o f L T P can be generated by the sensory relay n e u r o n s in an intact animal. Several speculations have been offered to explain the i n d u c t i o n o f n e u r o n a l activity at LTP-specific frequencies. T h e m o s t p r o m i n e n t o f these is related to the theta frequency (4 to 7 Hz) generated by h i p p o c a m p a l n e u r o n s o f rats during e x p l o r a t o r y behavior. T h e theta frequency in rats is a c c o m p a n i e d by a short 100 to 150 H z " b u r s t " o f action potentials in phase with the theta r h y t h m (Bland et al. 1980). W h e n the Schaffer collateral projections to the CA1 field are stimulated briefly at high frequencies (,-~ 100 Hz) in phase with the theta r h y t h m , the reliability o f inducing L T P is reported to be m a r k e d l y e n h a n c e d ( L a r s o n et al. 1986). The theta r h y t h m , in addition, closely adheres to the frequency o f whisker movement ("whisking") in rats (Welker 1964). Therefore, in o u r attempts to potentiate cortical responses to the stimulation o f the whiskers, we have patterned o u r c o n d i t i o n i n g stimulus to mimic the theta/ whisking frequency. O u r previous results in the thalamic V P M nucleus o f rats indicated t h a t high frequency n e u r o n a l responses can be generated in thalamic relay n e u r o n s if G A B A mediated inhibition is partially blocked using B I C (Lee et al. 1992b) or r e m o v e d by selectively destroying the ipsilateral thalamic reticular nucleus with kainic acid (Lee et al. 1992a). In this in vivo study, we a t t e m p t e d to measure the effectiveness o f the B I C - i n d u c e d high freq u e n c y discharge o f thalamic relay n e u r o n s in eliciting a long term e n h a n c e m e n t o f responses in barrel field neurons. To m a k e the necessary measurements, we assumed that the induction o f L T P in the t h a l a m o c o r t i c a l pathw a y following the c o n d i t i o n i n g p r o c e d u r e w o u l d be m a n ifest as an increase in the n u m b e r o f spikes generated by layer IV n e u r o n s in response to standardized test stimuli.

Methods

Preparation and recording Twelve adult Long-Evans rats between 3 and 4 months of age (250-275 g) were anesthetized with urethane (1.5 g/kg body weight, i.p.) and maintained at a constant anesthetic depth (Stage III-3) with supplemental doses by monitoring the electrocorticogram (ECoG). Previous results from our laboratory indicated that animals can be maintained at Stage III-3 throughout an experiment by monitoring the peak ECoG frequency (Friedberg et al. 1991). The dominant ECoG frequency at Stage III-3 was shown to be 3 to 4 Hz. Other physiological indicators at Stage III-3 include respiratory rate at 88 to 104 per rain, heart rate at 232 to 350 per rain and the presence of corneal and eyelid reflexes, but no withdrawal to pinch or movement of the vibrissae. Animals used in the present study were carefully monitored during the presentation of the conditioning stimuli to detect any subtle changes in the anesthetic state. This rigorous monitoring of the anesthetic state was essential because our previous studies in this system demonstrated that some of the results presented here, such as changes in the response mag-

nitude, can be mimicked by a global change in state of the animal (Friedberg et al. 1991). However, the change in the responsiveness of thalamic and cortical neurons as a result of changes in the anesthetic state of the animal can be strongly correlated with a shift in the peak ECoG frequency and heart and respiratory rates (Friedberg et al. 1991 ; Lee et al. 1992a). The anesthetic state of the animals used in this study were stable at Stage III-3 before and after the CS presentation as determined by our monitoring procedure. A detailed description of the monitoring methods and the changes seen at different anesthetic depths have been presented elsewhere (Friedberg et al. 1991; Lee et al. 1992b). In brief, the depth of anesthesia was categorized using the terminology of Guedel (1920). The characteristic signs for each stage were determined using the ECoG and a number of other physiological indicators which included pupillary size, respiratory rate, electrocardiogram, corneal reflex and pinch-withdrawal reflex. For the placement of the ECoG electrode, a pin-hole was drilled through the skull over the right parietal cortex (approximately in mm, -2.0, 2.0 from bregma) to insert a lowimpedance (< 500 Kf~ at 100 Hz) tungsten electrode 1-1.5 mm beneath the surface of cortex. The signals were band-pass filtered between 0.1 to 50 Hz and displayed as a power spectrum in a condensed spectral array (CSA) format (Modular Instruments). The changes in the responsiveness of cortical neurons following the CS were assessed using extracellular single-unit recording techniques. The single units were isolated in left barrel field cortex using carbon fiber electrodes (0.5-2 MO; Armstrong-James and Millar 1979) and amplified using conventional techniques. The baseline measurement for the pre-CS activity consisted of an average of 7 poststimulus time histograms (PSTHs) at 5 rain intervals. Each PSTH displayed the spikes generated in the first 50 ms after a l0 ms deflection (300 ~tm vertical displacement) of the center receptive field (CRF) whisker using a piezoelectric mechanical stimtdator (Lee et al. 1992a). Each histogram consisted of 30 stimulus trials presented at 1 Hz. The CRF whisker was defined as the whisker which had the highest probability of eliciting a response; all other whiskers in the RF were classified as the surround receptive field (SRF). The post-CS activity was monitored for as long as the unit could be reliably "held" (~ 1 h). The responses were compared in terms of the total number of spikes per trial between 2 and 15 ms after the onset of the stimulus. The recordings in barrel field cortex were limited to deep layer III or layer IV (400-700 lain below pia), which is the primary termination zone for VPM afferents (Killackey and Leshin 1975; Jensen and Killackey 1987). Each recording site was marked by an electrolytic microlesion (10 ~tA for 5 s, tip negative). At the termination of the experiment, the animals were deeply anesthetized with sodium pentobarbital and perfused transcardially with 200 ml of phosphate buffered saline followed by 4% paraformaldehyde solution. Tangential sections 75 ~tm thick were cut through the barrel field cortex on a sliding microtome and reacted for cytochrome oxidase (CO) according to the method of Wong-Riley (1979). The placement of the thalamic electrode was verified using coronal sections through VPM stained with 2 % cresyl violet (Fig. 1A). The electrolytic microlesions marking the recording sites were readily discernible on the CO reacted sections which appeared as a pale circular region 50-120 ~tm in diameter (Fig. 1B). All cortical units presented in this study were located within a barrel "hollow" which appeared as a circular, densely-stained region in CO-reacted tangential sections (Land and Simons 1985). The number of units studied per animal was limited to three in order to minimize the overlap of the surround receptive fields. Each electrode penetration in cortex was separated by at least two barrels (~ 1.5 mm).

Conditioning stimulus ( CS) paradigm Two methods were used to induce high frequency activity in VPM ipsilateral to the cortical recording electrode. One procedure era-

255

Fig. 1A, B. The placement of the thalamic and cortical recording electrodes was verified histologically at the end of the experiment. The typical location of the thalamic electrode was in the dorsolateral half of VPM (A) The two arrows indicate the electrolytic lesions (10 ~tA, 10 s) made to mark the recording sites. The upper arrow marks the approximate location of the VPM unit responding best to the D2 whisker. The bottom mark was made 700 gm from the upper mark. The electrolytic microlesions used to mark the location of cortical neurons in layer IV were readily apparent from cytochrome oxidase (CO) reacted sections cut tangential to the surface of cortex as a pale, circular area approximately 70 gm in diameter (B, arrow). All cortical neurons analyzed were found within a CO dense "barrel". (calibration bar = 1 mm for both A, B)

ployed electrical stimulation of the thalamus. A tungsten electrode (0.5-1 Mf~) was placed in VPM to deliver three trains consisting of 4 pulses (100 las duration at 100 Hz) at an intertrain frequency of 7 Hz. The current intensity of the stimulating electrode was adjusted to limit the spread of activity in barrel field cortex to a single barrel (usually 3 5 4 0 gA). The extent of barrel field cortex activated by the electrical stimulation of VPM was measured by placing a second recording electrode in adjacent barrels and determining the responsiveness of cells in the neighboring barrels at various current intensities. The second method for eliciting high frequency activity in VPM was to disinhibit thalamic neurons using a GABAA receptor antagonist. BIC (5 raM, pH 3.5, 40-80 nA) was released iontophoretically in VPM for 3 rain using a multi-barreled electrode (Lee et al. 1992b), after which the C R F whisker was stimulated for 10 s at 7 Hz.

With either procedure, the VPM electrodes (tungsten or iontophoretic) were used to isolate a single unit in VPM that was strongly activated by one or more whiskers. A corresponding cortical unit was then located whose C R F whisker matched that of the thalamic unit. As further evidence that the two units were likely to be monosynaptically connected, a "thalamic spike-triggered histogram" was computed on-line by assessing the spontaneous firing of the cortical neuron as a function of VPM activity. The frequency of cortical single-unit activity within 50 ms of a thalamic spike was plotted as a histogram (Fig. 2). This procedure allowed us to estimate the level of cross-correlation between the paired thalamic/cortical units before proceeding with the conditioning procedure. Only those units which were strongly correlated were presented with the CS and studied further.

256 80

A

----o-- Layer IV .t

60

VPM

160

40 120 20

0

80 10

20

30

40

50

CS 807 B 0

0 6O

40 r~

20

O n"

10

0

20

30

40

50

C

80

i

i

;

i

20

40

60

80

Time (min) Fig. 3. The effect of high frequency stimulation in VPM on the response of cortical layer IV units to 10 ms deflection of the CRF whisker. The CS (arrow) consisted of 3 trains (4 • 100 las shocks at 100 Hz) delivered at 7 Hz. Each time point represents the average change in the number of spikes per stimulus (• SEM) for 11 paired thalamic/cortical units. Note that the thalamic responses remained unaltered although the cortical responses increased by an average of 41.2%. Changes seen 20 rain after the C3 were highly significant (p 60 min). Following the presentation of the CS, the responses generally increased by 20-30% within the first 5 min. The responses were further enhanced during the following 20-30 rain, which eventually stabilized well-above baseline values for the duration of the recording session. The remaining 4 units which failed to show potentiated responses quickly returned to baseline values after the initial increase in the cell response. The average response latency for the 11 units following the induction o f potentiation was not significantly different than pre-CS values (8.2+0.41 ms vs. 7.68+0.51 ms, respectively). In all 15 units tested, the response magnitude o f the thalamic units to the stimulation o f the C R F whisker remained identical to pre-CS measurements (Fig. 3).

Effects of high frequency activity induced by BIC in VPM The average response frequency of V P M neurons to a 10 ms deflection of the C R F whisker is summarized in Fig. 4A as a function of the ejection current for BIC. F o r

257 250

A ~

200 ,..., o

Y.

200 7 A 4 160J

| ~

---t>-- LayerIV A'" VPM

T

150 100

80t

r

50 O/ i

i

i

20 40 60 Iontophoretic current (nA)

I

i

80

100

0

200

B

,

,

L

,

20

40

60

80

1B

+

100 t f f

-.-

120

r o

t

LayerIV

-,

-

-

-

10 ms

I

I

0/

0

,

,

,

,

20

40

60

80

Fig. 4. A Plot of frequency of the response discharge of VPM units within 2-50 ms of CRF whisker stimulation as a function of the iontophoretic current used to eject BIC (mean • SEM ; n = 20). The duration of the BIC iontophoresis was kept constant at 1 rain for all VPM neurons. The frequency of the response was determined only for those units which gave 3 or more spikes per stimulus in the presence of BIC. B An example of a high-frequency response in VPM (upper trace) to a 10 ms deflection of the CRF whisker (lower trace). Iontophoretic application of 40 nA BIC for 1 min led to an increase in the number of spikes generated (from 2 to 6 spikes). The average response frequency of this unit in the presence of BIC was 131.6 Hz

Time (rain) Fig. 5. A The change in the responsiveness of VPM neurons and layer IV cortical neurons to whisker stimulation in the presence of BIC in VPM. The iontophoretic application of BIC in VPM (duration indicated by the bar) transiently increased the responsiveness of both cortical and VPM neurons to the test stimuli before returning to baseline values within 10-30 min after terminating the iontophoretic application (mean i SEM, n = 9 pairs). B The stimulation of the CRF whisker at 7 Hz for 30 s without the iontophoretic application of BIC had no detectable influence on the cortical and VPM responses (mean • SEM, n = 11 pairs)

the range o f current intensities studied (40-80 nA, for 1 min), there was a linear relationship between the a m o u n t of BIC released and the frequency of spikes generated in response to a stimulus. The range of discharge frequencies seen during the response period following the blockade of G A B A - m e d i a t e d inhibition was 57 to 240 H z (Fig. 4A). The typical response of a V P M neuron to a 10 ms deflection of the C R F whisker in the presence of BIC can be seen in Fig. 4B. This representative neuron, on average, responded with 2.4 spikes per stimulus presentation (10 ms whisker deflection) at an average interspike time of 6.3 ms (158 Hz). Following a 1 rain period of BIC iontophoresis at 62 nA, this unit increased the n u m b e r of spikes to an average of 4.7 spikes at a response frequency of 202 H z which was maintained for as long as the BIC was present (-,~ 10 min). Thus, two characteristic effects were seen in the responses o f V P M neurons disinhibited by BIC: i) a m a r k ed increase in the n u m b e r of spikes during the response period and ii) a sharp decrease in the interspike interval which was proportional to the level of disinhibition.

The ability of BIC enhanced responses in V P M to alter cortical cell responsiveness was assessed in 9 barrel field layer IV neurons (6 animals). Each of the cortical neurons tested was strongly correlated with the activity of the V P M unit near the release o f BIC (see Methods). In all units tested, a 3 rain application of BIC led to a 20-40% increase response magnitude of b o t h V P M and cortical neurons (Fig. 5A). I f the whiskers were not stimulated during the presence of BIC, the responses in cortex and in V P M returned to baseline values within 30 rain (Fig. 5B). However, in 8 out of 9 cortical units (88.9%), stimulation of the C R F whisker at a frequency of 7 Hz for 10 s led to an average increase in the response magnitude of 43.2% over baseline measurements (Fig. 6, open squares). This enhancement persisted in cortex for as long as the unit responses were recorded in layer IV (--~ 1 h). The responses in V P M always returned to baseline values after the effects of BIC had worn off despite the presentation of the CS (Fig. 6, filled triangles). To ensure that the change in the response level in layer 1V was occurring in the thalamocortical synapses and not

258 ---o-- Layer 1V 200 1

120

~D

+

VPM

-

8o CS 0

0

I

t

I

I

20

4o

60

80

Time (min) Fig. 6. The effect of combining BIC iontophoresis in VPM with 7 Hz stimulation of the C R F whisker. The 3 min release of BIC in V P M is indicated by the bar. The 7 Hz stimulation of the whisker for 10 s (CS) was delivered immediately before terminating the BIC iontophoresis. Note that the responses in V P M were transiently enhanced by the presence of BIC. W h e n the CS was delivered in the presence of BIC in VPM, the cortical responsiveness increased by an average of 43.2% over baseline measurements (n = 8 pairs) 9 The average of n u m b e r of spikes/stimulus for layer IV neurons before the CS was 0.68 vs. 0.97 approximately 45 min after the conditioning procedure

VPM

LAYER IV

,r

:. ,.

A

due to a global change in cortical excitability, the peak ECoG frequency was recorded immediately after measuring the changes in the cortical and thalamic responses. In all cases, the CS presentation (electrical or BIC + whisker stimulation) did not change the peak ECoG frequency during or after the duration of the CS delivery. Further, the 7 Hz stimulation of the whiskers was not sufficient in itself to induce LTP in cortex. In 11 units (3 cases), stimulation of the CRF whisker at 7-10 Hz for 10-60 s failed to significantly alter the responsiveness of cortical neurons to subsequent test measurements (Fig. 5, filled circles) or alter the peak ECoG frequency. Typical changes in the response of cortical and VPM neurons during BIC iontophoresis and post-CS presentation can be seen in Fig. 7. Figure 7D is a dramatic example of the increase in the number of spikes and interspike frequency during the response period in the presence of bicuculline. The response of the cortical unit, which was presumed to be monosynaptically linked to the VPM neuron, was somewhat attenuated when compared to the VPM responses, but showed a 3-fold increase in the total number of spikes. Immediately after presenting the CS to the CRF whisker (7 Hz for 10 s), the BIC application was terminated. The combination of

1.

B

30 pre-CS

pre-CS

2O 10 o

I

I

L

ii

I iliW I J

I I i

i

D

C

l ~;

%.-

I iN

..

.

~-,".:..,-.-,,.. ~.'.. "....

,4~,k,."

30

40 nA BIC 2O

r.~

ttLLk, li I

10

,,,,,

| j

0

~ ,': .:

E

9

F

9 . ".,,

.

30 post-CS

20

r~

10 !

0 0

n

post-CS

1

Rm

50

~ A nnlu

100 Time (msec)

i

~

150

i L~!

u~

200

50

100 Time (msec)

n| m

150

n

200

Fig. 7A-D. Representative poststimulus time histograms of a correlated thalamic/cortical pair of neurons. Panels A and B represent typical pre-CS measurements. The response of both layer IV C and V P M D neurons to 30 stimulations (10 ms deflections) of the C R F whisker in the presence of 40 nA BIC in the thalamus dramatically increased the n u m b e r of spikes generated (onset of the stimulus occurred at 100 ms). The responses in V P M 1 h after the CS presentation (F) were identical to preCS PSTHs, but the responses in cortex were significantly enhanced 9 For this cortical neuron, the n u m b e r of spikes between 2-15 ms after the stimulus increased by 5l % following the conditioning procedure

259 disinhibition in the thalamus and stimulation of the CRF whisker at 7 Hz elicited a 47 % increase in the number of spikes generated in cortex, yet the VPM response quickly returned to pre-CS measurements (compare Fig. 7A and 7E vs. 7B and 7F). Discussion

The results from this study demonstrate that high frequency (50-200 Hz) activity generated in thalamocortical relay neurons can induce long-term modification in the responsiveness of layer IV neurons to cutaneous stimuli even in adult neocortex. These changes in the cortical responses to whisker deflections were detectable using single-unit recording techniques and documented in adult rats maintained under light urethane anesthesia (Stage III-3 as defined in a previous study, Friedberg et al. 1991). The long-term increase in cortical response magnitude was seen without a concurrent long-term increase in the thalamic responses or a change in the anesthetic state of the animal. Several methodological issues are highly relevant to these conclusions. First, it was impossible to determine conclusively whether a thalamic relay neuron was monosynaptically linked to the cortical unit being studied using extracellular recording techniques. However, for each pair of cortical and VPM units isolated, the degree of cross-correlation was measured before the conditioning procedures. This cross-correlation procedure has been demonstrated previously to be an effective means of estimating how closely two units are anatomically connected (Shin and Chapin 1990; West et al. 1990). All layer IV cortical units which were included in the analysis were strongly correlated in their activity to the spontaneous firing of VPM neurons. However, the recording techniques used here cannot rule out the possibility that the increase in the cortical cell responsiveness may reflect the enhancement of VPM neurons in the same thalamic "barreloid" other than the cell recorded, which may converge on the layer IV neuron tested. Likewise, the conditioning procedure could induce changes in VPM responsiveness which are difficult to detect beyond a given level of significance using extracellular single-unit recording techniques. The sum of these sub-detectable changes converging on layer IV neuron could profoundly affect the excitability of cortical neurons to sensory stimulation. Thus, the possibility exists that the robust changes seen in the responsiveness of layer IV neurons to sensory stimulation may reflect the sum of subthreshold events in VPM only measurable using intracellular recording techniques. The high frequency activity needed to induce potentiation in cortex in this study was generated by two different methods of thalamic activation. First, the thalamocortical relay neurons were directly activated using bursts (100 Hz) of electrical stimuli at a physiologically relevant frequency of 7 Hz. The other method was to stimulate the CRF whisker at the 7 Hz whisking rate while the VPM thalamic neurons were in a state of disinhibition. The strategy behind both methods was to combine a naturally occurring stimulus frequency with

high frequency bursts demonstrated previously to be effective in inducing LTP in neocortical slices in this system (Lee et al. 1991). Both methods were effective in inducing LTP in barrel field layer IV neurons to stimulations of the CRF whisker (73.3 % vs. 88.9 %, respectively). It should be noted that although the precise frequency for obtaining maximal potentiation was not systematically measured in this study, bursts of electrical stimuli delivered at somewhat higher frequencies, e.g. 30 to 40 Hz, tended to diminish the response level of cortical neurons to further whisker stimulation. Results from studies of guinea pig slice preparations have shown that thalamic neurons possess intrinsic membrane properties which facilitate the generation of high frequency activity (Jahnsen and Llinfis 1984a, b). The high frequency activity is reported to occur by one of two mechanisms: 1) opening of a low-threshold Ca 2+ channel and 2) activation of a Na+-dependent plateau potential. The low-threshold Ca 2+ channel becomes deinactivated at hyperpolarized membrane potentials and under these conditions it can generate a low-threshold spike (LTS) which is accompanied by a "burst" of action potentials at its crest. The Na +-dependent plateau potential, which has been shown to occur at depolarized potentials (Jahnsen and Llinfis 1984a, b), causes repetitive firing of action potentials for the duration of a stimulus, e.g. direct depolarizing current injection, at 40 to 300 Hz interspike frequencies. In previous studies from this laboratory (Lee et al. 1992a, b), we have reported that the loss or blockade of inhibition causes the majority (85-90%) of VPM neurons tested to respond repetitively in a sustained discharge for the duration of the stimulus. We predict that the control of inhibition through the thalamic reticular nucleus, which in rats appears to be the sole source of GABAergic inhibition in VPM (Barbaresi et al. 1986; Harris and Hendrickson 1987), regulates the types of thalamic responses that can lead to long-term changes in the responsiveness of cortical neurons to cutaneous stimulation. Several recent studies have indicated that the simultaneous activation of cholinergic neurons in the basal forebrain with cutaneous stimulation leads to longterm enhancement in the responsiveness of neurons in SI cortex (Metherate et al. 1987; Rasmusson and Dykes 1988; Tremblay et al. 1990). It is interesting to note that a number of anatomical studies have provided strong evidence that the cholinergic neurons in the basal forebrain also project to the thalamic reticular nucleus (Hallanger et al. 1987; Levey et al. 1987). Further, brainstem cholinergic nuclei, e.g. laterodorsal tegmental nucleus (LDT) and midbrain reticular formation (MRF), send projections to both the basal forebrain and the thalamic reticular nucleus (Scheibel and Scheibel 1958; Nauta and Kuyers 1958; Edwards and de Olmos 1976; Sofroniew et al. 1985; Satoh 1986) and may participate in the functions of the ascending reticular activating system (Moruzzi and Magoun 1949; Jasper 1949; Steriade and Llimis 1988; Buzs~tki et al. 1988). The functional role of cholinergic projections to the thalamic reticular nucleus is complex, but appears to be predominantly inhibitory (Ben-Ari et al. 1976; Dingledine and Kelly 1977). Thus, the activation of brainstem

260 cholinergic neurons m a y decrease the inhibitory influence of the thalamic reticular neurons on the thalamic relay neurons by decreasing the level of GABAergic inhibition. This disinhibition o f sensory transmission through the thalamus has been postulated to be responsible for the facilitated transmission of cutaneous information to cortex at different states of vigilance (Steriade 1970; Singer 1973; Singer et al. 1976; Steriade and Morin 1981). In view o f these findings, one influence of M R F activation on sensory transmission would be to alter the firing pattern of thalamic relay neurons. Together, the increased cholinergic activity f r o m the basal forebrain plus high frequency responses induced by disinhibition would provide the conditions necessary for long-term modification o f thalamic influence on cortical layer IV neurons. The incorporation of L T P - m e d i a t e d changes in synaptic strength into the overall function of sensory cortex remains to be fully investigated, but recent theories in this area propose two i m p o r t a n t consequences: i) mechanisms that support L T P could drastically alter the top o g r a p h y of m a p representations in sensory neocortex or ii) potentiating mechanisms m a y result in an enhanced response to selective sensory experiences. A recent report f r o m our l a b o r a t o r y using an in vitro slice preparation has demonstrated that the horizontal spread o f activity in sensory cortex can be markedly enhanced by the induction of L T P (Lee et al. 1991). This facilitation of the horizontal spread o f activity effectively increases the area of neocortex in which cells can be influenced by sensory activity. The extent o f this enlargement can be manipulated by altering the pattern o f the conditioning stimulus used to elicit LTP. These results support the intriguing speculation that, under the proper conditions, high frequency activity in the thalamocortical p a t h w a y can increase dramatically the n u m b e r of cortical neurons which responds to a sensory stimulus. The second hypothesis for the role of L T P in sensory processing is the possibility that L T P can lead to an enhanced responsiveness of cortical neurons to specific types of sensory stimuli. An example of this enhanced processing in the rat trigeminal system could be manifest as cortical neurons poorly sensitive to the m o v e m e n t of whiskers in a particular direction becoming highly selective. Such a shift in sensitivity m a y increase the effectiveness o f certain types of sensory input or may, in effect, decrease the n u m b e r of cortical neurons required to process sensory information at a given level of precision. An analysis o f the changes in receptive field properties before and after L T P m a y clarify its role in sensory processing.

Acknowledgements. We would like to thank Drs. Mark Bear, James Simmons, Barry Connors, Robert Dykes and Marc Friedberg for helpful discussion and comments. This study was supported by grants NS-13031 and NS-25907 from the NIH.

References Armstrong-James M, Millar J (1979) Carbon fibre microelectrodes. J Neurosci Meth 1 : 279-287

Artola A, Br6cher S, Singer W (1990) Different voltage-dependent thresholds for inducing long-term depression and long-term potentiation in slices of rat visual cortex. Nature (Lond) 347: 69-72 Artola A, Singer W (1987) Long-term potentiation and NMDA receptors in rat visual cortex. Nature (Lond) 330:649-652 Barbaresi P, Spreafico R, Frassoni C, Rustioni A (1986) GABAergic neurons are present in the dorsal column nuclei but not in the ventroposterior complex of rats. Brain Res 382:305-326 Ben-Ari Y, Dingledine R, Kanazawa I, Kelly JS (1976) Inhibitory effects of acetylcholine on neurones in the feline nucleus reticularis thalami. J Physiol (Lond) 261 : 647-671 Berry RL, Teyler TJ, Taizhen H (1989) Induction of LTP in rat primary visual cortex: tetanus parameters. Brain Res 481:221-227 Bindman LJ, Murphy KPSJ, Pockett SJ (1988) Postsynaptic control of the induction of long-term changes in efficacy of transmission at neocortical synapses in slices of rat brain. J Neurophysiol 60:1053-1065 Bland BH, Andersen P, Ganes T, Sveen O (1980) Automated analysis of rhythmicity of physiologically identified hippocampal formation neurons. Exp Brain Res 38:205-219 Bliss TVP, Gardner-Medwin AR (1973) Long-lasting potentiation of synaptic transmission in the dentate area of the unanaesthetized rabbit following stimulation of the perforant pathway. J Physiol (Lond) 232:357-374 Bliss TVP, Lomo T (1973) Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant pathway. J Physiol (Lond) 232:331-356 Buzsfiki G, Bickford RG, Ponomareff G, Thal LJ, Mandel R, Gage F (1988) Nucleus basalis and thalamic control of neocortical activity in the freely moving rat. J Neurosci 8:4007-4026 Dingledine R, Kelly JS (1977) Brain stem stimulation and the acetylcholine-evoked inhibition of neurones in the feline nucleus reticularis thalami. J Physiol (Lond) 271 : 135-154 Edwards SB, de Olmos JS (1976) Autoradiography study of the projections of the midbrain reticular formation: ascending projections of nucleus cuneiformis. J Comp Neurol 165: 417-432 Friedberg MH, Lee SM, Ebner FF (1991) Anesthetic stage as a determinant of VPM receptive field properties. Soc Neurosci Abst 17:839 Guedel AE (1920) Signs of inhalational anesthesia. A fundamental guide. In: Guedel AE (ed) Inhalational anesthesia. Macmillan, New York, pp 10-52 Hallanger AE, Levey AI, Lee HJ, Rye DB, Wainer BH (1987) The origins of cholinergic and other subcortical afferents to the thalamus in the rat. J Comp Neurol 262:105-124 Harris RM, Hendrickson AE (1987) Local circuit neurons in the rat ventrobasal thalamus - a GABA immunocytochemical study. Neuroscience 21 : 22%36 Jahnsen H, Llin~is R (1984a) Electrophysiological properties of guinea-pig thalamic neurones: an in vitro study. J Physiol Lond 349 : 205-226 Jahnsen H, Llin/ts R (1984b) Ionic basis for the electroresponsivehess and oscillatory properties of guinea-pig thalamic neurones in vitro. J Physiol Lond 349:227-247 Jasper H (1949) Diffuse projection systems: the integrative action of the thalamic reticular system. Electroencephagr. Clin Neurophysiol 1: 405-420 Jensen KF, Killackey HP (1987) Terminal arbors of axons projecting to the somatosensory cortex of the adult rat. I. The normal morphology of specific thalamocortical afferents. J Neurosci 7: 352%3543 Killackey HP, Leshin S (1975) The organization of specific thalamocortical projections to the posteromedial barrel subfield of the rat somatic sensory cortex. Brain Res 86:469-472 Land PW, and Simons DJ (1985) Cytochrome oxidase staining in the rat SmI barrel cortex. J Comp Neurol 238:225-235

261 Larson J, Wong D, Lynch G (1986) Patterned stimulation at the theta frequency is optimal for the induction of hippocampal long-term potentiation. Brain Res 368: 347-350 Lee KS (1982) Sustained enhancement of evoked potentials following brief, highfrequency stimulation of the cerebral cortex in vitro. Brain Res 239:617-623 Lee SM, Friedberg MH, Ebner FF (1992a) The role of GABAmediated inhibition in the rat ventroposterior medial (VPM) thalamus I: Quantitative assessment of receptive field changes following excitotoxic lesion of the thalamic reticular nucleus. J Neurophysiol Lee SM, Friedberg MH, Ebner FF (1992b) The role of GABAmediated inhibition in the rat ventroposterior medial (VPM) thalamus. II. Differential effects of GABAA and GABAB receptor antagonists to somatic sensory stimuli. J Neurophysiol Lee SM, Weisskopf MG, Ebner FF (1991) Horizontal long-term potentiation of responses in rat somatosensory cortex. Brain Res 544:303-310 Levey AI, Hallanger AE, Wainer BH (1987) Cholinergic nucleus basalis neurons may influence the cortex via the thalamus. Neurosci Lett 74: 7-13 Metherate R, Tremblay N, Dykes RW (1987) Acetylcholine permits long-term enhancement of neuronal responsiveness in cat primary somatosensory cortex. Neuroscience 22:75-81 Moruzzi G, Magoun HW (1949) Brain stem reticular formation and activation of the EEG. Electroencephalogr Clin Neurophysiol 1 : 455473 Nauta WJH, Kuypers HCJM (1958) Some ascending pathways in the brain stem reticular formation. In: Jasper HH (ed) Reticular formation of the brain. Little, Brown, Boston, pp 3-30 Perkins AT, Teyler TJ (1988) A critical period for long-term potentiation in the developing rat visual cortex. Brain Res 439: 222-229 Rasmusson DD, Dykes RW (1988) Long-term enhancement of evoked potentials in cat somatosensory cortex produced by co-activation of the basal forebrain and cutaneous receptors. Exp Brain Res 70:276-86 Satoh K, Fibiger HC (1986) Cholinergic neurons of the laterodorsal tegmental nucleus: efferent and afferent connections. J Comp Neurol 153 : 277-302

Scheibel ME, Scheibel AB (1958) Structural substrates for integrative patterns in brain stem reticular core. In: Jasper HH (ed) Reticular formation of the brain. Little, Brown, Boston, pp 31-55 Shin HC, Chapin JK (1990) Movement induced modulation of afferent transmission to single neurons in the ventroposterior thalamus and somatosensory cortex in rat. Exp Brain Res 81 : 515-522 Singer W (1973) The effect of mesencephalic reticular stimulation on intracellular potentials of cat lateral geniculate neurons. Brain Res 61 : 35-54 Singer W, Tretter F, Cynader M (1976) The effect of reticular stimulation on spontaneous and evoked activity in the cat visual cortex. Brain Res 102:71 90 Sofroniew MV, Priestly JV, Consolazione A, Eckenstein E, Cuello AC (1985) Cholinergic projections from the midbrain and pons to the thalamus in the rat, identified by combined retrograde tracing and choline acetyltransferase immunohistochemistry. Brain Res 329 : 213-223 Steriade M, Morin D (1981) Reticular influences on primary and augmenting responses in the somatosensory cortex. Brain Res 205: 6280 Steriade M, Llinfis RR (1988) The functional states of the thalamus and the associated neuronal interplay. Physiol Rev 68 : 649-742 Steriade M (1970) Ascending control of thalamic and cortical responsiveness. Int J Neurobiol 12:87-144 Tremblay N, Warren RA, Dykes RW (1990) Electrophysiological studies of acetylcholine and the role of the basal forebrain in the somatosensory cortex of the cat. II. Cortical neurons excited by somatic stimuli. J Neurophysiol 64:1212-1222 Welker WI (1964) Analysis of sniffing of the albino rat. Behaviour 22: 223-244 West MO, Carelli RM, Pomerantz M, Cohen SM, Gardner JP, Chapin JK, Woodward DJ (1990) A region in the dorsolateral striatum of the rat exhibiting single-unitcorrelations with specific locomotor limb movements. J Neurophysiol 64:1233-1246 Wilson DA (1984) A comparison of the postnatal development of post-activation potentials in the neocortex and dentate gyrus of the rat. Dev Brain Res 16:61-68 Wong-Riley MTT (1979) Changes in the visual system of monocularly sutured or enucleated cats demonstrable with cytochrome oxidase histochemistry. Brain Res 171:11-28