Available online at www.sciencedirect.com

www.elsevier.com/locate/brainres

b r a i n r e s e a r c h 1 4 8 8 ( 2 0 1 2 ) 5 1 – 5 9

0006-8993/$ - see frohttp://dx.doi.org/10

Abbreviations: A

i.p., intraperitonea

cortex; m-V1, med

of the mean.nCorresponding aut

E-mail address:

Research Report

Comparison of long-term potentiation (LTP) in the medial(monocular) and lateral (binocular) rat primaryvisual cortex

Min-Ching Kuo, Hans C. Dringenbergn

Department of Psychology and The Center for Neuroscience Studies, Queen’s University, Kingston, Ont., Canada K7L 3N6

a r t i c l e i n f o

Article history:

Accepted 2 October 2012

Recent evidence suggests that the primary visual cortex (V1) of rodents expresses

surprisingly high levels of plasticity into adulthood. For example, long-term potentiation

Available online 9 October 2012

Keywords:

Visual cortex

Lateral geniculate nucleus

Long-term potentiation

Long-term depression

NMDA receptor

Rat

nt matter & 2012 Elsevie.1016/j.brainres.2012.10.0

NOVA, analysis of va

l; LTD, long-term depre

ial primary visual cortex

hor. Fax: þ1 613 533 2499.dringenb@queensu.ca (H

a b s t r a c t

(LTP) is readily induced in the mature V1 of adult rodents in vivo. Here, adult, urethane-

anesthetized rats were used for a detailed characterization of LTP in the pathway between

the dorsal lateral geniculate nucleus (dLGN) and ipsilateral V1. Strong theta-burst

stimulation (TBS) of the LGN resulted in LTP of field postsynaptic potentials (fPSPs)

recorded in the lateral (binocular) aspects of V1 (l-V1), but failed to do so in the medial

(monocular) V1 (m-V1). Administration of MK 801 (1 mg/kg, i.p.) blocked LTP in l-V1,

indicative of a critical role of NMDA receptors in this effect. Interestingly, weaker TBS

induction protocols resulted in synaptic depression in both l-V1 and m-V1, an effect that

was blocked by MK 801 only in m-V1. Finally, dLGN stimulation also elicited long-latency

fPSPs in the V1 contralateral to the stimulation site, likely reflecting polysynaptic activity

crossing the midline via callosal fibers. Relative to ipsilateral recordings, contralateral

fPSPs showed greater LTP in both V1 segments, which could not be blocked by MK 801

administration. Together, these data reveal clear differences in the expression of LTP at

synapses in l-V1 and m-V1, with greater plasticity in the lateral V1 segment. Further, we

confirm that NMDA receptors mediate some, but not all forms of synaptic plasticity in the

V1 of adult rodents.

& 2012 Elsevier B.V. All rights reserved.

r B.V. All rights reserved.06

riance; dLGN, dorsal lateral geniculate nucleus; fPSP, field postsynaptic potential;

ssion; LTP, long-term potentiation; NMDA, N-methyl-D-aspartate; V1, primary visual

; l-V1, lateral primary visual cortex; TBS, theta burst stimulation; S.E.M., standard error

.C. Dringenberg).

1. Introduction

A significant amount of evidence now demonstrates that

sensory cortices, including the primary visual cortex (V1),

maintain a significant capacity for synaptic plasticity well

beyond early development. For example, N-methyl-D-

aspartate (NMDA) receptor-dependent long-term potentiation

(LTP) is readily induced in the fully matured V1 of adult

b r a i n r e s e a r c h 1 4 8 8 ( 2 0 1 2 ) 5 1 – 5 952

rodents (Heynen and Bear, 2001; Dringenberg et al., 2007; Kuo

and Dringenberg, 2008; Mainardi et al., 2010). Similarly, ocular

dominance plasticity, initially thought to be temporally

restricted to an early ‘‘critical/sensitive period’’, has now been

shown to occur in adult rodents (Sawtell et al., 2003; Hofer

et al., 2006), even though it appears to involve mechanisms

partially distinct from those mediating ocular dominance

plasticity in young, immature animals (Sato and Stryker, 2008).

Specific forms of visual stimulation are also able to induce

modifications of synaptic connectivity in the mature V1.

Clapp et al. (2006) demonstrated an LTP-like, NMDA-

receptor dependent enhancement of visual evoked potentials

in V1 following rapid delivery of visual (checkerboard) stimuli

to the retina of adult rats, and similar results have been

obtained in humans (Teyler et al., 2005). Also, repeated

presentation of visual (grating) patterns to mice can lead to

a gradual, orientation-specific facilitation of visual responses

over several days, a phenomenon that shares many molecu-

lar mechanisms with electrically-induced LTP and could

provide the basis for certain forms of perceptual learning in

sensory networks (Frenkel et al., 2006; Cooke and Bear, 2010).

The V1 of rodents is composed of two functional segments,

which can be distinguished by the types of retinal inputs

relayed to the cortex. The medial aspect of V1 (m-V1) receives

inputs (by means of the dorsal lateral geniculate nucleus,

dLGN) that carry information almost exclusively from the

contralateral retina, while the lateral V1 (l-V1) responds to

signals from both the ipsi- and contralateral eye (Zilles et al.,

1984; Reid and Jaruska, 1991; Sefton et al., 2004). The question

Fig. 1 – Experimental set-up and examples of field postsynaptic

ipsilateral and contralateral to the stimulation site in the dorsal

in the medial (monocular) V1 (V1M) and lateral (binocular) V1 (V

burst stimulation (TBS; 4 bursts). Note the greater amplitude of

V1B. Scale bars: 12.5 ms and 0.5 mV. Numbers indicate distan

Watson, 1998). (For interpretation of the references to color in th

this article.)

of whether properties and mechanisms of synaptic plasticity

differ between m-V1 and l-V1 has received relatively little

attention in previous work. It is of interest to note, however,

that LTP of visual evoked potentials following rapid retinal

light stimulation appears to be anatomically restricted, with

LTP expressed in l-V1, while no significant potentiation was

induced in m-V1 (Clapp et al., 2006). Also, McCoy et al. (2008)

recently demonstrated that activation of muscarinic acetyl-

choline receptors in V1 slice preparations produced a form of

persistent LTP of layer 2/3 synapses in the binocular (l-V1)

segment, while long-term depression (LTD) was induced in

the monocular component of V1 (m-V1). These studies

suggest that thresholds for plasticity induction, as well as

the direction of plasticity (i.e., LTP vs. LTD) initiated by the

same receptor population might be different between the

lateral and medial V1 segments. Nevertheless, in many of the

previous in vivo studies of LTP or LTD in V1, a clear distinction

or comparison of l-V1 and m-V1 was not undertaken (e.g.,

Heynen and Bear, 2001; Dringenberg et al., 2007; Kuo and

Dringenberg, 2008).

With the present experiments, we provide a more detailed

comparison of LTP in the l-V1 and m-V1 of adult rats for a

number of induction protocols of different strengths. Further,

in addition to examining plasticity in the direct (ipsilateral)

geniculo-cortical pathway, LTP of long-latency responses in the

V1 contralateral to the dLGN stimulation site were also exam-

ined. Recent work suggests that these long-range responses

might reflect activity that crosses the midline by means of

callosal projections and exhibits plasticity properties distinct

potentials (fPSPs) recorded in the primary visual cortex (V1)

lateral geniculate nucleus (dLGN). Recordings were obtained

1B) before (black) and after (blue and red) application of theta

baseline fPSPs and the greater potentiation following TBS in

ce (in mm) posterior from bregma (based on Paxinos and

is figure legend, the reader is referred to the web version of

b r a i n r e s e a r c h 1 4 8 8 ( 2 0 1 2 ) 5 1 – 5 9 53

from those observed in the uncrossed geniculo-cortical fiber

system (Gagolewicz and Dringenberg, 2009; Kuo et al., 2009).

Fig. 2 – Amplitude of field postsynaptic potentials (fPSP) in

the medial (A) and lateral (B) segment of the primary visual

cortex (V1) ipsilateral to the dorsal lateral geniculate

nucleus (dLGN) stimulation site. (A) Delivery (at dashed line)

of weak theta burst stimulation (3 TBS) to the dLGN resulted

in depression, while strong 10 TBS failed to induce LTP in

the medial V1. n indicate significant differences (po0.05,

t-test) between 0 vs. 3 TBS. (B) In the lateral V1, weak TBS

(3 TBS) induced depression, while strong 10 TBS resulted in

LTP. ‘�’ indicates significant differences (po0.05, t-test)

between 0 vs. 3 TBS and 0 vs. 10 TBS. Insets in A and B

show amplitude means during the 4th (last) hour of the

experiment for all TBS conditions.

2. Results

2.1. Characterization of fPSPs

Extracellular recordings in the superficial layers (I–II) of V1

revealed that single pulse stimulation of the dLGN reliably

elicited field postsynaptic potentials (fPSPs) in both the ipsi-

and contralateral V1 of urethane-anesthetized rats (Fig. 1). All

fPSPs consisted of a primarily negative-going deflection,

which appears to reflect excitatory current sinks originating

in cortical layers II/III (Heynen and Bear, 2001). In the V1

ipsilateral to the dLGN stimulation site, latencies to the

maximal, negative fPSP peak were equivalent for recordings

in the medial (monocular, n¼37) and lateral (binocular, n¼32)

segment (�13–15 ms). However, fPSPs in the l-V1 consistently

showed larger amplitude relative to the m-V1 (0.79 and

0.60 mV, respectively; Fig. 1). Applying two successive stimu-

lation pulses (100 ms interval) resulted in an augmentation of

the second fPSP in a pair, with similar levels of augmentation

observed in m-V1 and l-V1 (ratio of the second fPSP ampli-

tude over amplitude of the first fPSP: 1.90 and 1.86 for l-V1

and m-V1, respectively).

Similar to previous work (Gagolewicz and Dringenberg,

2009; Kuo et al., 2009), stimulation of the dLGN also elicited

fPSPs in the contralateral V1 (Fig. 1). These potentials exhib-

ited considerably longer latencies to peak (�23–25 ms) and

smaller amplitudes (�50%) relative to ipsilateral fPSPs. Given

the apparent absence of direct projections from dLGN to the

contralateral V1, it is likely that these long-latency potentials

reflect polysynaptic activity that crosses the midline by

means of callosal projections between the V1 segments of

both cortical hemispheres (see Sefton et al., 2004; Gagolewicz

and Dringenberg, 2009). Interestingly, paired stimulation

(100 ms interval) resulted in much greater augmentation of

contralateral fPSPs (ratio of the second over the first fPSP

approximately 2.5), suggesting differences in the capacity for

short-tern synaptic enhancements in the fiber systems

between dLGN and the ipsi- and contralateral V1.

2.2. fPSPs between dLGN and the ipsilateral V1

In control animals that did not receive theta burst stimula-

tion (0 TBS), fPSPs recorded in m-V1 showed a slight upward

drift over the course of the experiment (Fig. 2A), with

amplitude during the last hour of the experiment at

123716.3% of baseline (n¼5), a trend that reached statistical

significance (po0.001). In the l-V1 (Fig. 2B), fPSPs remained

stable, with amplitude during the last hour of the experiment

at 10577% of baseline (n¼5).

For recordings in m-V1 (Fig. 2A), weak TBS protocols (3 TBS

and 4 TBS; n¼6/group) resulted in a significant depression of

fPSPs relative to animals not receiving TBS (0 TBS; analysis of

variance (ANOVA) comparing 3 TBS vs. 0 TBS, Fgroup� time

26,260¼5.3, po0.001; 4th hour mean of 8479% for 3 TBS; 4 TBS

vs. 0 TBS, Fgroup� time 26,260¼6.2, po0.001, 4th hour mean of

8379% for 4 TBS). Stronger induction protocols (5 and 10 TBS;

n¼5/group) did not result in any significant changes in fPSP

amplitude relative to control (0 TBS) rats (Fig. 2A; 5 TBS,

p¼0.057; 4th hour mean, 104717%; 10 TBS, p¼0.64; 4th hour

mean, 12173%).

For fPSPs in l-V1 (Fig. 2B), the weak 3 TBS protocol (n¼5)

also elicited significant depression relative to 0 TBS animals

(Fgroup� time 26,234¼4.5, po0.001; 4th hour mean, 8174%),

while 4 TBS (n¼5) did not result in any significant changes

in fPSP amplitude (p¼0.12; 4th hour mean, 11278%).

However, the strong 10 TBS protocol (Fig. 2; n¼6) elicited

clear potentiation of fPSP amplitude relative to control

animals (Fig. 2B; Fgroup 1,10¼5.13, po0.05; 4th hour mean,

118712%).

Fig. 3 – Effect of MK 801 treatment (1 mg/kg, i.p., injection at

arrow) on changes in field postsynaptic potential (fPSP)

amplitude elicited by weak and strong theta burst stimula-

tion (TBS, at dashed line) of the ipsilateral dorsal lateral

geniculate nucleus in the medial (A) and lateral (B and C)

segments of the primary visual cortex (V1). A) MK 801

reversed the effect of 3 TBS to depress fPSP amplitude in

the medial V1. ‘�’ indicates significant differences (po0.05, t-

test) between 3 TBS vs. 3 TBSþMK. (B) MK 801 did not block

the depression of fPSPs induced by 3 TBS in the lateral V1. ‘�’indicates significant differences (po0.05, t-test) between 0

TBS vs. 3 TBSþMK. (C) MK 801 blocked LTP in the lateral V1

induced by strong 10 TBS. ‘�’ indicates significant differences

(po0.05, t-test) between 10 TBS vs. 10 TBSþMK (Note that the

0 TBS, 3 TBS, and 10 TBS groups are the same as in Fig. 2).

b r a i n r e s e a r c h 1 4 8 8 ( 2 0 1 2 ) 5 1 – 5 954

2.3. Effect of MK 801 treatment on ipsilateral LTP andLTD

To assess the role of NMDA receptors in the bidirectional

synaptic changes induced by different TBS protocols applied

to the dLGN, additional groups of animals were administered

the non-competitive NMDA receptor antagonist MK 801

(1.0 mg/kg, i.p.). After baseline recording, these groups of

animals received MK 801 and fPSP recordings continued for

another 30 min (see Fig. 3), after which independent groups

received either the 3 TBS or 10 TBS induction protocols.

As mentioned above, the 3 TBS protocol induced significant

depression of fPSPs in both m-V1 and l-V1. MK 801 treatment

was effective in blocking the depressant effect in m-V1

(Fig. 3A; n¼5; ANOVA comparing 3 TBS groups with and

without MK 801, Fgroup� time 26,234¼2.98, po0.001; 4th hour

mean of 114712% for MK 801 animals), but not in l-V1

(Fig. 3B; n¼6; group by time effect, p¼0.1; 4th hour mean of

85713% for MK 801 group).

As mentioned, application of the strong induction protocol

(10 TBS) resulted in LTP in the l-V1. However, in the presence

of MK 801 (n¼5), 10 TBS failed to induce potentiation and

fPSP amplitude in this group was not different from that in

control rats not given TBS (Fig. 3C; group by time effect,

p¼0.995; 4th hour mean, 104713%). Further, in m-V1, where

10 TBS did not results in significant changes in fPSP ampli-

tude in untreated rats, this protocol produced significant

depression when applied in the presence of MK 801 (data

not shown; n¼5; ANOVA comparing drug-free and MK 801

rats receiving 10 TBS, Fgroup� time 26,208¼1.759, po0.05; 4th

hour mean, 95711%). Thus, NMDA receptors not only play a

critical role in LTP in the l-V1, but also influence the direction

of plasticity induction in m-V1 under the present, experi-

mental conditions.

2.4. fPSPs between dLGN and the contralateral V1

In control rats that did not receive TBS (0 TBS), long-latency

fPSPs elicited in m-V1 contralateral to the dLGN stimulation

site showed a slight, but significant enhancement over time

in the 0 TBS group (Fig. 4A; n¼5; mean amplitude of 11876%

of baseline during the last hour, po0.001). A small, but non-

significant increase in fPSP amplitude was also apparent in

the l-V1 in the 0 TBS condition (Fig. 4B; n¼5; amplitude at

11077% of baseline, p¼0.25).

The contralateral m-V1 segment exhibited minor, transient

potentiation following delivery of the 3 TBS (n¼6; 4th hour

mean, 11876%), 4 TBS (n¼6; 4th hour mean, 99716%), and 5

TBS (n¼5; 4th hour mean, 130712%) protocols (Fig. 4A;

ANOVAs comparing TBS animals against controls (0 TBS): 3

TBS, Fgroup� time 26,234¼3.1, po0.001; 4 TBS, Fgroup� time

26,234¼4.1, po0.001; 5 TBS, Fgroup� time 26,208¼2.0, po0.01). In

all cases, fPSP amplitude returned to levels similar to those in

0 TBS rats by the end of the experiment. The strongest

protocol (10 TBS, n¼5) also appeared to elicited synaptic

potentiation, but this effect was quite variable and did not

reach statistical significance (Fig. 4A; Fgroup� time 26,208¼1.3,

p¼0.14; 4th hour mean, 140710%).

In contrast, recordings of fPSPs in the contralateral l-V1

(Fig. 4B) showed significant potentiation in response to all

tested TBS protocols (3 TBS, n¼5, Fgroup� time 26,234¼4.4,

po0.001; 4 TBS, n¼5, Fgroup� time 26,234¼3.2, po0.001; 10 TBS,

n¼6, Fgroup� time 26,260¼3.7, po0.001). The potentiation

Fig. 4 – Amplitude of field postsynaptic potentials (fPSP) in the medial (A) and lateral (B) segment of the primary visual cortex

(V1) contralateral to the dorsal lateral geniculate nucleus (dLGN) stimulation site. (A) Delivery (at dashed line) of weak theta

burst stimulation (4 TBS) and strong 10 TBS to the dLGN resulted in minor, non-significant enhancements of fPSP amplitude

in the medial V1. (B) In the lateral V1, 4 TBS and 10 TBS induced transient and stable LTP, respectively. ‘�’ indicates

significant differences (po0.05, t-test) between 0 TBS vs. 10 TBS. Inserts in A and B show amplitude means during the 4th

(last) hour of the experiment for all TBS conditions.

b r a i n r e s e a r c h 1 4 8 8 ( 2 0 1 2 ) 5 1 – 5 9 55

elicited by 3 TBS (4th hour mean, 12376%) and 4 TBS (4th

hour mean, 121710%) was transient, while 10 TBS (4th hour

mean 161719%) induced stable enhancement (Fig. 4B). Over-

all, it appears that fPSPs elicited in the V1 contralateral to the

dLGN stimulation site are more prone to exhibit synaptic

potentiation relative to synaptic responses in the ipsilateral

V1, and fPSPs in l-V1 show more enhancement than those in

m-V1.

2.5. Effect of MK 801 treatment on contralateral LTP

In untreated rats, the weak (3 TBS) and strong (10 TBS)

induction protocols resulted in minor, transient and strong,

persistent potentiation in m-V1, respectively (see above).

Both protocols continued to elicit potentiation in the pre-

sence of MK 801 (n¼5 for 3 TBSþMK 801; data not shown),

with statistical analyses confirming that potentiation follow-

ing 10 TBS was not different for untreated and MK 801-

treated rats (Fig. 5A; n¼5; 4th hour mean, 13076%; p¼0.14).

Similar observations were made for potentiation in l-V1

(Fig. 5B). Again, MK 801 failed to block LTP elicited by either

the weak 3 TBS (data not shown; n¼6) or the strong 10 TBS

protocol (Fig. 5B; n¼5; 4th hour mean, 150720%; p¼0.86),

even though there appeared to be a minor reduction in

synaptic enhancement for the 10 TBS group given MK 801.

Together, these data suggest that synaptic enhancement in

the long-range connection system between dLGN and the

contralateral V1 is, for the most part, an NMDA receptor-

independent process.

3. Discussion

The present study confirms that TBS of the dLGN effectively

elicits LTP in the mature V1 of adult, urethane anesthetized

rats. However, clear differences in LTP were seen between the

m-V1 and l-V1 (corresponding to the monocular and binocu-

lar segments of V1, respectively; see Paxinos and Watson,

Fig. 5 – Effect of MK 801 treatment (1 mg/kg, i.p., injection at arrow) on changes in field postsynaptic potential (fPSP) amplitude

in the medial (A) and lateral (B) segments of the primary visual cortex (V1) elicited by theta burst stimulation (TBS, at dashed

line) of the contralateral dorsal lateral geniculate nucleus. (A) MK 801 did not reversed the effect of strong 10 TBS to potentiate

fPSP amplitude in the medial V1. ‘�’ indicates significant differences (po0.05, t-test) between 0 TBS vs. 10 TBSþMK. (B) MK 801

appeared to reduced, but did not block LTP induced by 10 TBS in the lateral V1. ‘�’ indicates significant differences (po0.05,

t-test) between 0 TBS vs. 10 TBSþMK (note that the 0 TBS and 10 TBS control groups are the same as in Fig. 4).

b r a i n r e s e a r c h 1 4 8 8 ( 2 0 1 2 ) 5 1 – 5 956

1998; Sefton et al., 2004), with LTP restricted to the lateral

segment. Interestingly, weak TBS protocols elicited LTD in

both segments, a surprising finding, given that TBS is not

typically used for LTD induction. Together, these results

demonstrate clear differences in the susceptibility to express

synaptic modifications between different anatomical and

functional subdivisions of V1, with potentiation occurring

much more readily in the lateral V1 segment.

Recordings obtained from l-V1 consistently yielded fPSPs of

larger amplitude than those obtained in m-V1. Some anato-

mical differences between the monocular and binocular

regions of the rat V1 have been observed, including a more

striated appearance, greater acetylcholinesterase staining,

and higher granule cell density in layers II, IV, and V of the

monocular V1, with the binocular V1 exhibiting a greater

number of myelinated fibers (Zilles et al., 1984; Sefton et al.,

2004). While clear, cytoarchitectonic boundaries between the

medial and lateral regions are lacking (Reid and Jaruska, 1991;

Zilles et al., 1984), functional mapping studies have demon-

strated that stimulation of the ipsilateral eye results in an

evoked response that is confined to the lateral portion of V1,

thereby confirming its binocular response property (both l-V1

and m-V1 respond to the contralateral eye; Sakai et al., 1983;

Sawtell et al., 2003; Hofer et al., 2006; Kuo and Dringenberg,

2009).

The differences in LTP induction noted here are consistent

with other work that has employed alternative approaches to

modify synaptic strength in the rodent V1. Clapp et al. (2006),

delivering photic stimulation to rats using checkerboard

visual stimuli, observed that LTP of visual evoked potentials

could be induced in the binocular, but not the monocular V1

segment. More recently, McCoy and colleagues (2008) exam-

ined slices of V1 of tree shrews, which has a clear cytoarch-

itectonic boundary separating the monocular and binocular

regions. Using this in vitro approach, McCoy et al. found that

the activation of muscarinic acetylcholine receptors (using

carbachol) resulted in the induction of LTP and LTD in the

binocular and monocular V1 regions, respectively, a pattern

of results that is similar to our observation of a greater

likelihood of inducing LTP in the l-V1. It is of importance to

note, however, that the difference in LTP between V1 seg-

ments appears to be one of induction threshold, rather than a

b r a i n r e s e a r c h 1 4 8 8 ( 2 0 1 2 ) 5 1 – 5 9 57

complete inability of m-V1 to express LTP. Previous work has

shown that acute (2–5 h) sensory deprivation by means of

dark exposure can lower this threshold, allowing successful

LTP induction in m-V1 by thalamic TBS or retinal light

stimulation (Kuo and Dringenberg, 2009). Thus, the recent

history of sensory stimulation and related levels of synaptic

activity appear to play a particularly important role in gating

plasticity in the medial/monocular segment of the rodent V1

(see Bear, 2003).

We were surprised to see LTD-like synaptic changes follow-

ing weak (3–4 bursts) TBS in both V1 segments. The LTD

developed gradually and often required about 60 min to show

significant fPSP depression, which lasted until the end of

experiment (4 h). Several authors have commented on the

fact that the intact V1 displays a strong resistance against the

induction of LTD for protocols that readily elicit LTD in

various in vitro preparations (Jiang et al., 2003; Hager and

Dringenberg, 2010). Hence, our results demonstrate a possi-

ble, novel stimulation protocol to elicit LTD in V1 under

in vivo conditions. Pre-treatment with MK 801 blocked LTD

in m-V1, but not in l-V1, suggesting differences in NMDA-

dependency between these cortical regions. Previous work

has shown that some forms of LTD in the binocular V1

require activation of endocannabinoid receptors (Liu et al.,

2008; Smith et al., 2009) and future work should assess if this

mechanism plays a role in the LTD observed here.

Electrical stimulation of the dLGN also elicited evoked

responses in the hemisphere contralateral to the stimulation

site. These long-range, polysynaptic responses have been

observed in previous work (Gagolewicz and Dringenberg,

2009; Kuo et al., 2009) and likely reflect activity that crosses

the midline by means of callosal fibers, given the absence of

direct projections between dLGN and the contralateral V1

(Silveira et al., 1989). Callosal fibers are known to play impor-

tant roles in binocular responses of V1 neurons in rodents and

other species (Diao et al., 1983; Silveira et al., 1989; Sun et al.,

1994) and preliminary experiments in our laboratory have

shown that contralateral fPSPs can be reduced or abolished

by callosal transections (P. Gagolewicz and H. Dringenberg,

unpublished observations). Thus, callosal connections are a

likely candidate mediating the long-range, contralateral

responses observed in the present experiments, even though

contributions of other, cortical or subcortical pathways (see

Sefton et al., 2004) cannot be ruled out.

Evoked responses in the contralateral V1 showed more

potentiation than ipsilateral fPSPs, regardless of V1 segment,

even though the contralateral l-V1 showed the greatest level

of synaptic enhancement. Further, contralateral response

enhancements were not blocked by systemic MK 801 treat-

ment. Several forms of NMDA receptor-independent LTP in V1

have been described (Ohmura et al., 2003). For example,

Aroniadou and Teyler (1991), applying white matter (100 Hz)

stimulation to V1 slices, observed LTP of layer III responses

that was not blocked (and, in fact, further enhanced) in the

presence of an NMDA receptor antagonist. Alternatively, it is

also possible that the contralateral enhancement noted here,

as well as some forms of potentiation seen in previous work,

constitute phenomena different from classic, input-specific,

and NMDA receptor-dependent LTP, such as general increases

in excitability or disinhibition, or cellular modification at the

dLGN stimulation site following various forms of high-

frequency stimulation or TBS (see Hirata and Castro-

Alamancos, 2006). Future work will be required to assess

these different hypotheses.

All fPSPs recorded in the present study were analyzed by

means of amplitude measures, consistent with the large

majority of previous work assessing LTP or LTD in the rodent

V1 (e.g., Heynen and Bear, 2001; Clapp et al., 2006; Dringenberg

et al., 2007; McCoy et al., 2008; Kuo et al., 2009; Cooke and Bear,

2010; Mainardi et al., 2010). Alternative measures such as fPSP

slopes appear to be rarely used for field recordings in V1, even

by research groups that use such slope measures to analyze

hippocampal field potentials (e.g., Whitlock et al., 2006;

Dringenberg et al., 2008). Current-source density analyses have

shown that fPSPs elicited by LGN stimulation and recorded at

or close to the surface of V1 (as was done in the present study)

largely reflect excitatory current sinks in cortical layers II/III,

which are relatively uncontaminated by discharge-related

events, such as populations spikes (Heynen and Bear, 2001;

M.-C. Kuo, unpublished observations). However, to the best of

our knowledge, a detailed comparison of current-source den-

sity profiles for fPSPs in the medial and lateral aspects of V1

has never been carried out. Such work is currently ongoing in

our laboratory and will be instrumental in characterizing the

precise synaptic (and possibly other) mechanisms that con-

tribute to the plasticity phenomena observed in the present set

of experiments.

In conclusion, the experiments described here show that, in

adult rats, the lateral/binocular segment of V1 readily

expresses both LTD and LTP, depending in the strength of the

TBS protocol used. In contrast, the medial/monocular V1

expresses LTD, but fails to exhibit LTP under the present,

experimental conditions. Further, some, but not all forms of

bidirectional, synaptic plasticity in V1 require NMDA receptor-

dependent processes. LTP and LTD in V1 have been proposed as

mechanisms mediating ocular dominance plasticity and

aspects of perceptual learning in rodents (Frenkel et al., 2006;

Hofer et al., 2006; Smith et al., 2009; Cooke and Bear, 2010). The

results of the present experiments lead to several, intriguing

hypotheses regarding the effectiveness of perceptual learning

in different parts of the visual field of rodents, represented in

the distinct anatomical V1 segments (i.e., central visual field

with binocular overlap vs. peripheral visual field without

binocular overlap; e.g., Sefton et al., 2004); it will be of interest

to develop and test these hypotheses with future work.

4. Experimental procedure

4.1. Animals and surgical preparation

Experiments were conducted on male, adult Long-Evans rats

(300–500 g, Charles River Laboratories Inc., St. Constant,

Quebec) housed in groups of 4 or 5 (reversed 12 h light/dark

cycle, 7 am light off-7 pm light on) with food and water

available ad libitum. All experiments were performed in

accordance with published guidelines of the Canadian Coun-

cil on Animal Care and approved by the Queen’s University

Animal Care Committee.

b r a i n r e s e a r c h 1 4 8 8 ( 2 0 1 2 ) 5 1 – 5 958

Animals were under deep urethane anesthesia (1.5 g/kg,

given as three intraperitoneal (i.p.) injections of 0.5 g/kg each,

every 20 min; supplements as needed) before being placed in

a stereotaxic apparatus. Body temperature was maintained

between 35 and 37 1C with an electrical heating blanket

throughout the experiment. Coordinates for electrode place-

ments were based on the anatomical work of Paxinos and

Watson (1998). The skull was exposed and small skull holes

were drilled overlying the dLGN (AP �3.5, Lþ3.7, V �4.5) and

the ipsilateral and contralateral V1 (Fig. 1). For different

experiments, both V1 electrodes were aimed at either the

medial (m-V1, monocular; AP�7.5, L72.8, V�1.0) or lateral

(l-V1, binocular; AP�7.5, L74.0, V�1.0) segment of V1 (Fig. 1).

The final ventral placements of the dLGN stimulation elec-

trode (concentric bipolar electrode, Rhodes Medical Instru-

ments, Series 100, David Kopf, Tujunga, CA) and both V1

recording electrodes (125 mm diameter Teflon-insulated stain-

less steel wire) were adjusted to yield the maximum fPSP

amplitude and augmenting responses with paired-pulse sti-

mulation (100 ms inter-pulse intervals) of the dLGN. Holes

over the cerebellum and olfactory bulb were used to secure

ground and reference (jewelry screws) electrodes.

4.2. Electrophysiology recordings

Single pulse (0.2 ms duration) stimulation of the dLGN was

used to elicit fPSPs in the ipsi- and contralateral V1. All fPSPs

were recorded differentially against the cerebellar reference

connection. The signal was amplified, filtered (0.3 Hz–1 kHz),

digitized (10 kHz), and stored for offline analyses (PowerLab/

4s system, ADInstruments, Toronto, Canada). Stimulation

pulses were provided by a stimulus isolation unit that

provided a constant current output (PowerLab system with

ML 180 Stimulus Isolator, ADInstruments).

4.3. Experimental procedures

For each experiment, a 30-min stabilization period was given

after final electrode positioning. Subsequently, an input-output

curve (0.1–1.0 mA with 0.1 mA increments) was established and

a stimulation intensity that yielded about 50–60% of the

maximal fPSP amplitude was chosen for the experiment.

Initially, augmenting responses of fPSPs in the ipsi- and

contralateral V1 were characterized by applying two succes-

sive stimulation pulses to the dLGN (100 ms interval, repeated

10 times every 5 s). For the following LTP experiment, 60

baseline fPSPs were recorded (every 30 s) and recordings

continued until a stable baseline was established (fPSP

amplitude between 95–105% of the average baseline ampli-

tude). Following baseline recordings, LTP was induced by

applying different TBS induction protocols to independent

groups of animals. The TBS consisted of brief stimulation

bursts (5 pulses at 100 Hz/burst), which were repeated at a

theta-frequency of 5 Hz. To vary the strength of the induction

protocol, independent groups of animals received either 1, 3,

4, or 10 TBS bursts (a further 5 TBS burst group was added for

the m-V1 group). Control animals did not receive any TBS

(0 TBS). Following TBS delivery, recordings of fPSPs (every 30 s)

continued for 4 h.

In order to assess the role of NMDA receptors in changes of

synaptic strength induced by TBS, independent groups of rats

received the non-competitive NMDA receptor antagonist MK-

801 (1 mg/kg, i.p.; Sigma Chemicals, Oakville, Ont., Canada).

After 30 min of stable baseline recording, the drug was

administered and fPSPs were recorded for another 30 min

before delivery of either 3 or 10 TBS for both m-V1 and l-V1

experiments. Again, recordings of fPSPs (every 30 s) contin-

ued for 4 h after TBS.

4.4. Histology

At the end of data collection, animals were perfused through

the heart with 0.9% saline, followed by 10% formalin. Brains

were extracted and immersed in formalin before sectioning

(40 mm) using a cryostat. Brain sections were then mounted

onto microscope slides to verify all electrode placements

using standard histological techniques. Data obtained with

inaccurate placements were excluded from the statistical

analysis.

4.5. Data analysis

All data are expressed as mean7standard error of the mean

(S.E.M.). The amplitude of fPSPs was computed offline by the

Scope software (v.3.6.5, ADInstruments). Augmenting responses

elicited by paired pulse stimulation were quantified by dividing

the amplitude of the second fPSP in a pair by that of the first

fPSP. For LTP experiments, fPSP amplitude values were averaged

over 10 min intervals and these averages were normalized by

dividing them by the averaged baseline amplitude of individual

animals.

All data were statistically evaluated using repeated mea-

sures analysis of variance (ANOVA) and, if statistically appro-

priate, by simple effects tests using the CLR ANOVA software

package (v.1.1, Clear Lake Research Inc., Houston, Texas). The

level of significance for all statistical analyses was set at

po0.05.

Acknowledgments

This work was supported by the Natural Sciences and

Engineering Research Council (NSERC) of Canada.

r e f e r e n c e s

Aroniadou, V.A., Teyler, T.J., 1991. The role of NMDA receptors inlong-tem potentiation (LTP) and depression (LTD) in rat visualcortex. Brain Res. 562, 136–143.

Bear, M.F., 2003. Bidirectional synaptic plasticity: from theory toreality. Phil. Trans. R. Soc. Lond. B 358, 649–655.

Clapp, W.C., Eckert, M.J., Teyler, T.J., Abraham, W.C., 2006. Rapidvisual stimulation induces N-methyl-D-aspartate receptor-dependent sensory long-term potentiation in the rat cortex.Neuroreport 17, 511–515.

Cooke, S.F., Bear, M.F., 2010. Visual experience induced long-termpotentiation in the primary visual cortex. J. Neurosci. 30,16304–16313.

b r a i n r e s e a r c h 1 4 8 8 ( 2 0 1 2 ) 5 1 – 5 9 59

Diao, Y.C., Wang, Y.K., Pu, M.L., 1983. Binocular responses ofcortical cells and the callosal projection in the albino rat. Exp.Brain Res. 49, 27–36.

Dringenberg, H.C., Hamze, B., Wilson, A., Speechley, W., Kuo,M.C., 2007. Heterosynaptic facilitation of in vivo thalamocor-tical long-term potentiation in the adult rat visual cortex byacetylcholine. Cereb. Cortex 17, 839–848.

Dringenberg, H.C., Oliveira, D., Habib, D., 2008. Predatory (cathait)-induced enhancement of hippocampal long-term poten-tiation in rats: involvement of acetylcholine. Learn. Mem. 15,112–116.

Frenkel, M.Y., Sawtell, N.B., Diogo, A.C., Yoon, B., Neve, R.L., Bear,M.F., 2006. Instructive effect of visual experience in mousevisual cortex. Neuron 51, 339–349.

Gagolewicz, P.J., Dringenberg, H.C., 2009. Selective potentiation ofcrossed vs. uncrossed inputs from the lateral geniculatenucleus to visual cortex by the basal forebrain: potentialfacilitation of rodent binocularity. Neurosci. Lett. 463, 130–134.

Hager, A.M., Dringenberg, H.C., 2010. Assessment of differentinduction protocols to elicit long-term depression (LTD) in therat visual cortex in vivo. Brain Res. 1318, 33–41.

Heynen, A.J., Bear, M.F., 2001. Long-term potentiation of thala-mocortical transmission in the adult visual cortex in vivo.J. Neurosci. 21, 9801–9813.

Hirata, A., Castro-Alamancos, M., 2006. Relief of synaptic depres-sion produces long-term enhancement in thalamocorticalnetworks. J. Neurophyiol. 95, 2479–2491.

Hofer, S.B., Mrsic-Flogel, T.D., Bonhoeffer, T., Hubener, M., 2006.Lifelong learning: ocular dominance plasticity in mouse visualcortex. Curr. Opin. Neurobiol. 16, 451–459.

Jiang, B., Akaneya, Y., Hata, Y., Tsumoto, T., 2003. Long-termdepression is not induced by low-frequency stimulation in ratvisual cortex in vivo: a possible preventing role of endogenousbrain-derived neurotropic factor. J. Neurosci. 23, 3761–3770.

Kuo, M.-C., Dringenberg, H.C., 2008. Histamine facilitates in vivothalamocortical long-term potentiation in the mature visualcortex of anesthetized rats. Eur. J. Neurosci. 27, 1731–1738.

Kuo, M.-C., Dringenberg, H.C., 2009. Short-term (2 to 5 h) darkexposure lowers long-term potentiation (LTP) inductionthreshold in rat primary visual cortex. Brain Res. 1276, 58–66.

Kuo, M.-C., Rasmusson, D.D., Dringenberg, H.C., 2009. Input-selective potentiation and rebalancing of primary sensorycortex afferents by endogenous acetylcholine. Neuroscience163, 430–441.

Liu, C.-H., Heynen, A.J., Shuler, M.G.H., Bear, M.F., 2008. Cannabi-noid receptor blockade reveals parallel plasticity mechanismsin different layers of mouse visual cortex. Neuron 58, 340–345.

Mainardi, M., Landi, S., Gianfranceschi, L., Baldini, S., De Pasquale,R., Berardi, N., Maffei, L., Caleo, M., 2010. Environmental enrich-ment potentiates thalamocortical transmission and plasticity inthe adult rat visual cortex. J. Neurosci. Res. 88, 3048–3059.

McCoy, P., Norton, T.T., McMahon, L.L., 2008. Layer 2/3 synapses inmonocular and binocular regions of tree shrew visual cortexexpress mAChR-dependent long-term depression and long-term potentiation. J. Neurophysiol. 100, 336–345.

Ohmura, T., Ming, R., Yoshimura, Y., Komatsu, Y., 2003. Age andexperience dependence of N-methyl-D-aspartate receptor-independent long-term potentiation in rat visual cortex.Neurosci. Lett. 341, 95–98.

Paxinos, G., Watson, C., 1998. The Rat Brain in StereotaxicCoordinates, fourth ed. Academic Press, San Diego, USA.

Reid, S.N.M., Jaruska, J.M., 1991. The cytoarchitectonic boundariesof the monocular and binocular areas of the rat primary visualcortex. Brain Res. 563, 293–296.

Sakai, M., Yagi, F., Ikeda, Y., 1983. Evoked potential in the visualcortex as modified by enucleation of one eye in the albino rat.Physiol. Psych. 11, 141–146.

Sato, M., Stryker, M.P., 2008. Distinctive features of adult oculardominance plasticity. J. Neurosci. 28, 10278–10286.

Sawtell, N.B., Frenkel, M.Y., Philpot, B.D., Nakazawa, K., Tonegawa, S.,Bear, M.F., 2003. NMDA receptor-dependent ocular dominanceplasticity in adult visual cortex. Neuron 38, 977–985.

Sefton, A.J., Dreher, B., Harvey, A., 2004. Visual system. In:System, G., Paxinos (Eds.), The Rat Nervous. Elsevier AcademicPress, San Diego, CA, pp. 1083–1165.

Silveira, L.C., Heywood, C.A., Cowey, A., 1989. Direct and trans-callosal contributions of the cortical evoked response in rats.Behav. Brain Res. 31, 291–294.

Smith, G.B., Heynen, A.J., Bear, M.F., 2009. Bidirectional synapticmechanisms of ocular dominance plasticity in visual cortex.Phil. Trans. R. Soc. B 364, 357–367.

Sun, J.S., Li, B., Ma, M.H., Diao, Y.C., 1994. Transcallosal circuitryrevealed by blocking and disinhibiting callosal input in thecat. Vis. Neurosci. 11, 189–197.

Teyler, T.J., Hamm, J.P., Clapp, W.C., Johnson, B.W., Corballis, M.C.,Kirk, I.J., 2005. Long-term potentiation of human visual evokedresponses. Eur. J. Neurosci. 21, 2045–2050.

Whitlock, J.R., Heynen, A.J., Shuler, M.G., Bear, M.F., 2006. Learninginduces long-term potentiation in the hippocampus. Science313, 1093–1097.

Zilles, K., Wree, A., Schleicher, A., Divac, I., 1984. The monocularand binocular subfields of the rat’s primary visual cortex: aquantitative morphological approach. J. Comp. Neurol. 226,391–402.