Environmental enrichment inadulthood promotes amblyopiarecovery through a reductionof intracortical inhibitionAlessandro Sale1,3, Jose Fernando Maya Vetencourt1,3,Paolo Medini2, Maria Cristina Cenni2, Laura Baroncelli1,Roberto De Pasquale1 & Lamberto Maffei1,2

Loss of visual acuity caused by abnormal visual experience

during development (amblyopia) is an untreatable pathology

in adults. We report that environmental enrichment in adult

amblyopic rats restored normal visual acuity and ocular

dominance. These effects were due to reduced GABAergic

inhibition in the visual cortex, accompanied by increased

expression of BDNF and reduced density of extra-

cellular-matrix perineuronal nets, and were prevented

by enhancement of inhibition through benzodiazepine

cortical infusion.

An abnormal visual experience during binocular mammals’ develop-ment, resulting from a functional imbalance between the two eyes,induces an ocular dominance shift of visual cortical neurons in favor ofthe normal eye and a loss of visual acuity in the deprived eye(amblyopia). Recovery of visual acuity and ocular dominance ispossible only if normal visual experience is re-established at early

stages of brain development1, even though some improvements invisual functions have been reported in amblyopic human beings foramblyopia caused by anisometropia or strabismus2,3. The lack ofrecovery from amblyopia in the adult is due to a decline of neuralplasticity during late postnatal development that is attributed, forinstance, to a shift in the composition of NMDA receptor subunits4

and the activity of the CRE-CREB system5. Recently, attention hasbeen focused on the maturation of intracortical inhibitory circuits,which sets the threshold for both the start and the end of the visualcortical plasticity window6–8, and on the condensation of extra-cellular matrix molecules in perineuronal nets (PNNs) mainly aroundinhibitory interneurons9.

We explored environmental enrichment10, a condition characterizedby an increased exploratory behavior and sensory-motor stimulation,as a strategy for enhancing plasticity in the adult. All of the proceduresused were approved by the Italian Ministry of Public Health. We firstassessed the effects of environmental enrichment on the recovery ofvisual acuity in adult rats that were rendered amblyopic by long-termmonocular deprivation and then reverse-sutured (see SupplementaryMethods online). We measured visual acuity using electrophysiologicalrecordings of visual evoked potentials (VEPs) from the primary visualcortex. In reverse-sutured rats housed in standard conditions (RS-SC),visual acuity of the deprived eye remained significantly lower (0.62 ±0.01 cycles per degree, c deg–1) with respect to the other eye (1.06 ± 0.06c deg–1) (Fig. 1a). In contrast, we found a full visual acuity recovery inreverse sutured rats housed under environmental enrichment condi-tions (RS-EE rats) (1.01 ± 0.07 c deg–1, see Fig. 1a). A behavioralmeasure (visual water-box task) of visual acuity confirmed the electro-physiological data: a full recovery was evident in RS-EE animals

RS-SC

Vis

ual a

cuity

(c

deg

–1)

*

RS-SC RS-EE

*

RS-SC0

*

Ocular dominance

Fellow eye Deprived eye

*

0

Electrophysiologicalvisual acuity

RS-EE

1.2

1.0

0.8

0.6

0.4

0.2

Behavioral visual acuity

Normaladult values

Brain microdialysis

RS-EE

RS-SC

[GA

BA

] (µm

ol l–1

)

2.5

1.5

0.5

Per

cent

age

of b

asel

ine

140

120

100

80

WM-LTP

SC EE

EESC TBS

RS-EE

Vis

ual a

cuity

(c

deg

–1)

C/I

VE

P r

atio

1.2

1.0

0.8

0.6

0.4

0.2

3.0

2.0

1.0

–20 –10 10 20 30

Time (min)

a b c d e

Figure 1 Environmental enrichment promotes amblyopia recovery and reduces intracortical inhibition in adult rats. (a–c) Amblyopia recovery in RS-EE rats.Electrophysiological (a) and behavioral (b) assessment of visual acuity, which was lower in the formerly deprived eye than in the other eye in RS-SC rats

(paired t-test, P o 0.05) but not in RS-EE rats (paired t-test, P ¼ 0.864 for a and P ¼ 0.100 for b). C/I VEP ratio was statistically lower in RS-SC than in

RS-EE rats (t-test, P o 0.05), but did not differ between RS-EE and normal (non-deprived) adult rats (P ¼ 0.907) (c). (d,e) Reduced levels of inhibition

in RS-EE adult rats. Extracellular GABA levels in the visual cortex contralateral to the deprived eye were lower in RS-EE than in RS-SC adult rats (t-test,

P o 0.05, d). WM-LTP was restored in environmental enrichment adult rats (e). WM-LTP (measured 20–30 min after theta-burst stimulation, TBS) was

significantly greater in adult rats given environmental enrichment than in those maintained in standard conditions (two-way repeated-measures ANOVA,

P ¼ 0.002). Averages of ten traces before (thin line) and 25–30 min after (thick line) TBS are also shown. Scale bars are 25% of baseline amplitude and

2.5 ms. *, statistical significance. Error bars, s.e.m.

Received 6 March; accepted 26 March; published online 29 April 2007; doi:10.1038/nn1899

1Scuola Normale Superiore, Piazza dei Cavalieri, I-56100 Pisa, Italy. 2Institute of Neuroscience, CNR, Via Moruzzi 1, I-56100 Pisa, Italy. 3These authors contributed equallyto this work. Correspondence should be addressed to A.S. (a.sale@sns.it).

NATURE NEUROSCIENCE VOLUME 10 [ NUMBER 6 [ JUNE 2007 679

BR I E F COMMUNICAT IONS©

2007

Nat

ure

Pub

lishi

ng G

roup

ht

tp://

ww

w.n

atur

e.co

m/n

atur

eneu

rosc

ienc

e

(Fig. 1b, see also Supplementary Fig. 1 online). Visual acuity recoveryin the formerly deprived eye measured 2 weeks after the end of theenvironmental enrichment period persisted unaltered.

To control whether environmental enrichment per se modifies visualfunctions in adult animals, we assessed behavioral visual acuity in adifferent group of normal (not deprived) adult rats under standardconditions and with environmental enrichment. Visual acuity valuesremained completely unaltered after 2 weeks of environmentalenrichment in adulthood (before environmental enrichment, 0.92 ±0.03 c deg–1; after environmental enrichment, 0.93 ± 0.04 c deg–1;paired t-test, P ¼ 0.867; controls, 0.89 ± 0.01; t-test environmentalenrichment versus controls, P ¼ 0.406).

To determine ocular dominance (binocularity), we measured thecontralateral to ipsilateral (C/I) VEP ratio (see Supplementary Meth-ods online) in the same animals in which we assessed visual acuityrecovery. The C/I ratio is in the 2.5–3.0 range for adult normal rats,reflecting the predominance of crossed fibers in rat retinal projections.In RS-SC rats, there was no recovery of binocularity in the visual cortexcontralateral to the formerly deprived eye (C/I VEP ratio ¼ 1.11 ± 0.20,Fig. 1c). In contrast, RS-EE rats showed full recovery of binocularity,with a C/I VEP ratio of 2.47 ± 0.50.

Because there is evidence that the maturation of cortical inhibitorycircuits ends plasticity in the visual system11, we investigated whetherthe restored plasticity under environmental enrichment conditions wasaccompanied by a reduction of visual cortex inhibition. In vivo brainmicrodialysis showed that basal levels of GABA were diminished by afactor of 3 in the binocular visual cortex (Oc1B) contralateral to theformerly deprived eye of RS-EE rats (0.64 ± 0.26 mM) compared withRS-SC rats (1.83 ± 0.18 mM) (Fig. 1d). We did not detect a significantdifference (t-test, P = 0.855) in basal levels of glutamate between RS-EE(2.56 ± 1.08 mM) and RS-SC (2.87 ± 0.64 mM) animals (Supplemen-tary Fig. 2 online). The decrease of intracortical inhibition in the visualcortex can also be evaluated electrophysiologically, by assessing long-term potentiation of layer II-III field potentials induced by theta-burststimulation from the white matter (WM-LTP). The WM-LTP is notpresent in the adult as a result of the maturation of inhibitorycircuits7,12, but it can be restored if GABA-mediated inhibition isreduced12,13. Notably, WM-LTP was fully restored in the visual cortexof environmental enrichment adult rats (Fig. 1e). No WM-LTP waspresent in animals kept under standard conditions.

The reduction of cortical inhibition in RS-EE rats was paralleled byvariations of other molecular factors involved in cortical plasticity. The

expression of the neurotrophin BDNF was increased in the visualcortex contralateral to the formerly deprived eye in RS-EE comparedwith RS-SC rats (Fig. 2a). Furthermore, the typical organization ofchondroitin sulfate proteoglycans in PNNs was still present in RS-EErats. PNN density, however, was lower compared with RS-SC rats(Fig. 2b) in the visual cortex contralateral to the long-term deprived eye(see also Supplementary Fig. 3 online).

Finally, to test whether the reduction of intracortical inhibition inenvironmental enrichment was causally linked to the recovery ofnormal visual functions, we chronically infused (via osmotic mini-pumps) a different group of RS-EE animals with the benzodiazepineagonist diazepam (2 mg ml–1) during the environmental enrichmentperiod. Cortical diazepam administration totally prevented the envi-ronmental enrichment–induced recovery effect in both visual acuityand the C/I VEP ratio (Fig. 2c,d).

Our findings show that environmental enrichment promotes acomplete recovery of visual acuity and ocular dominance in adultamblyopic animals, and that decreased intracortical inhibition is acrucial molecular mechanism underlying this effect. Reduced expressionof GABAA receptors has been recently shown in adult rats exposed todarkness, a treatment that reactivates ocular dominance plasticity in thevisual cortex14. Both findings point toward reduced GABAergic neuro-transmission as a key element for restoring visual cortex plasticity inadulthood. An interaction between BDNF and the GABAergic system isthought to regulate visual cortical plasticity during development7,11.Using this logic, our results may be explained by a model in whichenhanced sensory-motor activity under environmental enrichmentconditions decreases cortical basal levels of inhibition and, in parallelor in series, increases BDNF expression, upregulating the genes thatpromote plasticity. This could provide a permissive environment forvisual cortical plasticity, allowing a structural and functional rewiring ofcortical circuits, which underlie the environmental enrichment–inducedamblyopia recovery. Notably, we found that environmental enrichmentled to a reduced density of extracellular matrix PNNs in the visualcortex. As recently shown, a pharmacological removal of crucialcomponents of PNNs from the mature extracellular matrix promotesrecovery from the effects of early visual deprivation in adult rats15.

Our results highlight a therapeutic potential for environmentalenrichment as a physiological non-invasive approach for promotingthe recovery of normal visual functions in adult amblyopic animals.

Note: Supplementary information is available on the Nature Neuroscience website.

0

*

*

*

BDNF

BD

NF

pos

itive

ne

uron

s m

m–2

PN

Ns

posi

tive

neur

ons

mm

–2

350

250

150

50

II-III IV-V-VI

160

120

80

40

II-III IV-V-VI

Fellow eye Deprived eyeRS-SC RS-EE

RS-EE RS-EE-Dz

Visual acuity recovery

* ** *

Vis

ual a

cuity

(c

deg

–1)

C/I

VE

P r

atio

1.2

1.0

0.8

0.6

0.4

0.2RS-SC

a b c dWFA

Ocular dominance recovery

Normal adult values

RS-EE

3.0

2.0

1.0

0RS-EE-Dz RS-SC

Figure 2 Recovery from amblyopia in environmental enrichment is due to a reduced GABAergic inhibition in the visual cortex, accompanied by increased

BDNF expression and reduced density of PNNs. (a,b) Environmental enrichment–induced changes of plasticity factors in the visual cortex. BDNF expression

was higher (a) in layers IV-V-VI and the density of extracellular matrix PNNs stained for WFA was lower (b) in layers II-III and IV-V-VI (t-test, P o 0.05 in both

cases) in the visual cortex of RS-EE compared to RS-SC rats. (c,d) Infusion of diazepam into the visual cortex prevented recovery of vision in RS-EE rats. VEP

visual acuity was statistically different between the formerly deprived eye and the other eye in RS-SC and RS-EE infused with diazepam (RS-EE-Dz, paired

t-test, P o 0.05), but not in RS-EE rats (paired t-test, P ¼ 0.864) (c). The C/I VEP ratio was statistically lower in RS-SC and in RS-EE-Dz compared with

RS-EE rats (one-way ANOVA, Tukey test, P o 0.05), but did not differ between RS-EE-Dz and RS-SC rats (P ¼ 0.971) (d). *, statistical significance.

Error bars, s.e.m.

680 VOLUME 10 [ NUMBER 6 [ JUNE 2007 NATURE NEUROSCIENCE

BR I E F COMMUNICAT IONS©

2007

Nat

ure

Pub

lishi

ng G

roup

ht

tp://

ww

w.n

atur

e.co

m/n

atur

eneu

rosc

ienc

e

ACKNOWLEDGMENTSWe thank F. Clementi for kindly providing us the diazepam and A. Viegi for histechnical assistance with high-performance liquid chromatography. This workwas supported by grants from Ministero dell’Universita e della Ricerca (MIUR),Programmi di Ricerca di Rilevante Interesse Nazionale (PRIN) and FondoIntegrativo Speciale Ricerca (FISR).

AUTHOR CONTRIBUTIONSA.S. and J.F.M.V. contributed equally to this work. A.S. carried out the in vivoelectrophysiology, behavioral experiments, immunohistochemistry and assistedin the in vivo brain microdialysis. J.F.M.V. carried out the in vivo brainmicrodialysis, high performance liquid chromatography and assisted in thein vivo electrophysiology. P.M. carried out the in vivo electrophysiology andM.C.C., the immunohistochemistry experiments. L.B. performed the behavioralassessment of visual acuity in normal rats and R.D.P., the in vitroelectrophysiology. A.S., J.F.M.V. and L.M. wrote the manuscript. All authorsdiscussed the results and commented on the manuscript.

COMPETING INTERESTS STATEMENTThe authors declare no competing financial interests.

Published online at http://www.nature.com/natureneuroscience

Reprints and permissions information is available online at http://npg.nature.com/

reprintsandpermissions

1. Mitchell, D.E. & MacKinnon, S. Clin. Exp. Optom. 85, 5–18 (2002).2. Levi, D.M., Polat, U. & Hu, Y.S. Invest. Ophthalmol. Vis. Sci. 38, 1493–1510

(1997).3. Polat, U., Ma-Naim, T., Belkin, M. & Sagi, D. Proc. Natl. Acad. Sci. USA 101, 6692–

6697 (2004).4. Erisir, A. & Harris, J.L. J. Neurosci. 23, 5208–5218 (2003).5. Pham, T.A., Impey, S., Storm, D.R. & Stryker, M.P. Neuron 22, 63–72 (1999).6. Hensch, T.K. et al. Science 282, 1504–1508 (1998).7. Huang, Z.J. et al. Cell 98, 739–755 (1999).8. Fagiolini, M. & Hensch, T.K. Nature 404, 183–186 (2000).9. Pizzorusso, T. et al. Science 298, 1248–1251 (2002).10. van Praag, H., Kempermann, G. & Gage, F.H. Nat. Rev. Neurosci. 1, 191–198 (2000).11. Hensch, T.K. Nat. Rev. Neurosci. 6, 877–888 (2005).12. Kirkwood, A. & Bear, M.F. J. Neurosci. 14, 1634–1645 (1994).13. Artola, A. & Singer, W. Nature 330, 649–652 (1987).14. He, H.Y., Hodos, W. & Quinlan, E.M. J. Neurosci. 26, 2951–2955 (2006).15. Pizzorusso, T. et al. Proc. Natl. Acad. Sci. USA 103, 8517–8522 (2006).

NATURE NEUROSCIENCE VOLUME 10 [ NUMBER 6 [ JUNE 2007 681

BR I E F COMMUNICAT IONS©

2007

Nat

ure

Pub

lishi

ng G

roup

ht

tp://

ww

w.n

atur

e.co

m/n

atur

eneu

rosc

ienc

e