Homeostatic plasticity mechanisms are required for
juvenile, but not adult, ocular dominance plasticity
Adam Ranson, Claire E. J. Cheetham, Kevin Fox1, and Frank Sengpiel1
School of Biosciences and the Neuroscience and Mental Health Research Institute, Cardiff University, Cardiff CF10 3AX, United Kingdom
Edited* by Michael P. Stryker, University of California, San Francisco, CA, and approved December 5, 2011 (received for review July 27, 2011)
Ocular dominance (OD) plasticity in the visual cortex is a classic
model system for understanding developmental plasticity, but the
visual cortex also shows plasticity in adulthood. Whether the
plasticity mechanisms are similar or different at the two ages is
not clear. Several plasticity mechanisms operate during develop-
ment, including homeostatic plasticity, which acts to maintain the
total excitatory drive to a neuron. In agreement with this idea, we
found that an often-studied substrain of C57BL/6 mice, C57BL/
6JOlaHsd (6JOla), lacks both the homeostatic component of OD
plasticity as assessed by intrinsic signal imaging and synaptic
scaling of mEPSC amplitudes after a short period of dark exposure
during the critical period, whereas another substrain, C57BL/6J
(6J), exhibits both plasticity processes. However, in adult mice, OD
plasticity was identical in the 6JOla and 6J substrains, suggesting
that adult plasticity occurs by a different mechanism. Consistent
with this interpretation, adult OD plasticity was normal in TNFα
knockout mice, which are known to lack juvenile synaptic scaling
and the homeostatic component of OD plasticity, but was absent
in adult α-calcium/calmodulin-dependent protein kinase II;T286A
(αCaMKIIT286A) mice, which have a point mutation that prevents
autophosphorylation of αCaMKII. We conclude that increased re-
sponsiveness to open-eye stimulation after monocular deprivation
during the critical period is a homeostatic process that depends
mechanistically on synaptic scaling during the critical period,
whereas in adult mice it is mediated by a different mechanism that
requires αCaMKII autophosphorylation. Thus, our study reveals
a transition between homeostatic and long-term potentiation–like
plasticity mechanisms with increasing age.
spans approximately postnatal days (P) 19–32 (3). During the
critical period, monocular deprivation (MD) causes neurons in
the binocular zone of the primary visual cortex (V1) to undergo
an initial long term depression (LTD)-like decrease in re-
sponsiveness to the closed eye (4), followed by a delayed increase
in responsiveness to both the open and closed eye (5). Many
previous studies have searched for factors that might limit the
critical period to this early developmental time window (for
reviews see refs. 6 and 7). This search is complicated by recent
studies showing that in mice, OD plasticity also occurs in the adult
visual cortex (8, 9) although it declines beyond 110 d of age (10).
There are notable phenomenologic differences between plasticity
in the critical period and adulthood; in adult mice, the response to
MD is slower, is restricted to an increase in open-eye re-
sponsiveness, and is more pronounced in the hemisphere contra-
lateral to the deprived eye (11). These findings raise the question
of whether adult plasticity mechanisms are the same as or differ-
ent than those in operation during the critical period. If the
mechanisms are the same, then the critical period is simply a stage
of development during which the gain or sensitivity of the plas-
ticity mechanism is greatest; if different, then either more plas-
ticity mechanisms are present during the critical period than in
adulthood or new plasticity mechanisms replace critical period
plasticity mechanisms in adulthood.
Homeostatic plasticity has been shown to play a role in syn-
aptic development (12) and in OD plasticity during the critical
he visual cortex demonstrates a clear critical period for oc-
ular dominance (OD) plasticity (1, 2), which in the mouse
period (13, 14). TNFα is an important factor for synaptic scaling
(15) and is required for the increased open-eye responsiveness
observed during juvenile OD plasticity (13). OD plasticity after
MD in the adult visual cortex lacks the depression of the closed-
eye responses observed during the critical period and is instead
characterized by increased cortical responsiveness to the open
eye (9). Whether this potentiation depends on the same ho-
meostatic plasticity mechanism(s) that operate during the critical
period is not known, however. One study has suggested that
a form of synaptic scaling persists into adulthood in layer 2/3
(L2/3) of the visual cortex (16). On the other hand, studies in
the hippocampus have shown that homeostatic plasticity, and
synaptic scaling in particular, are down-regulated during devel-
opment (17, 18).
We report a pronounced homeostatic plasticity deficit in the
commonly used 6JOla substrain of C57BL/6 mice. During the
critical period, these mice exhibit impairments in both the MD-
induced increase in open-eye responsiveness and synaptic scaling
induced by dark exposure (DE). In contrast in adult 6JOla ani-
mals, the MD-induced increase in open-eye responsiveness was
normal, as was OD plasticity in adult TNFα KO (TNFα−/−) mice,
which are also known to lack synaptic scaling and the juvenile
homeostatic component of OD plasticity. Furthermore, adult
OD plasticity was entirely absent in T286A mice. Taken to-
gether, our data suggest that with aging, visual cortex plasticity
becomes increasingly dependent on a mechanism that requires
α-calcium/calmodulin-dependent protein kinase II (αCaMKII)
autophosphorylation and less dependent on the homeostatic
mechanisms that are critical to juvenile plasticity.
The Homeostatic Component of OD Plasticity Is Absent in 6JOla Mice.
A 3-d period of MD is known to result in an LTD-like loss of
cortical responsiveness to the closed eye with no change in open-
eye responses (13). If MD continues for another 2 or 3 d, then
the cortical response to the closed eye increases slightly from its
low point but is still depressed overall, and the response to the
open eye increases substantially (13). There is evidence in-
dicating that both these increases depend on a homeostatic
mechanism (13). We used intrinsic signal imaging (Fig. 1A) to
compare OD plasticity in vivo in the primary visual cortex of
6J and 6JOla mice after 3 d or 5–6 d of MD (Fig. 1B) and an-
alyzed the open-eye and closed-eye responses separately (3). A
two-way ANOVA of closed- and open-eye response magnitudes
found a significant interaction between substrain and depriva-
tion, indicating that MD had different effects on the two sub-
strains of C57BL/6 mice [closed eye: P < 0.01, F2,60 = 6.33
Author contributions: A.R., C.E.J.C., K.F., and F.S. designed research; A.R. and C.E.J.C.
performed research; A.R. and C.E.J.C. analyzed data; and A.R., C.E.J.C., K.F., and F.S. wrote
The authors declare no conflict of interest.
*This Direct Submission article had a prearranged editor.
1To whom correspondence may be addressed. E-mail: email@example.com or sengpielf@
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
| January 24, 2012
| vol. 109
| no. 4
(Fig. 1D); open eye: P < 0.01, F2,60= 5.04 (Fig. 1E)] prompting
In both the 6J and 6JOla substrains, 3 d of MD resulted in
a robust OD shift toward the open eye, mediated principally by
a loss of responsiveness to the closed eye (6J: control, 1.16 ±
0.13; 3 d MD, 0.58 ± 0.08; P < 0.01; t = 3.55; 6JOla: control, 1.40 ±
0.07; 3 d MD, 0.75 ± 0.07; P < 0.01; t = 4.83) (Fig. 1 D and G).
However, the cortical response to MD diverged between the two
substrains after 5–6 d. In the 6J substrain, 5–6 d of MD resulted
in a homeostatic increase in the cortical response to stimulation
of the open eye, as has been reported previously (5, 13) (6J:
control, 0.60 ± 0.06; 5–6 d MD, 0.90 ± 0.08; P < 0.05; t = 3.15)
(Fig. 1E). In contrast, the homeostatic increase in open-eye re-
sponsiveness was absent in the 6JOla substrain after 5–6 d of MD
(6JOla: control, 0.96 ± 0.07; 5–6 d MD, 0.77 ± 0.08; P = 0.09; t =
1.79) (Fig. 1E).
An indication that a homeostatic process is induced during OD
plasticity is the otherwise paradoxical “rebound” in cortical re-
sponsiveness to the deprived eye input between 3 d and 5–6 d of
MD (5, 13). This has been observed previously in mice as a slight
increase in cortical responsiveness to the closed eye in both the
binocular cortex and the monocular cortex, which receives input
only from the deprived (contralateral) eye (13). In 6J mice, ho-
meostatic recovery of deprived eye responses occurred in both
the binocular cortex and the monocular cortex after 5–6 d of MD
[binocular: 3 d MD, 0.58 ± 0.08; 5–6 d MD, 0.82 ± 0.08; P < 0.05;
t = 2.16 (Fig. 1D); monocular: 3 d MD, 0.83 ± 0.08; 5–6 d MD,
1.12 ± 0.09; P < 0.05; t = 2.39 (Fig. 1F)]. Strikingly, the opposite
trend was observed in 6JOla mice, whose deprived-eye responses
continued to decrease significantly between 3 d and 5–6 d MD
(binocular: 3 d MD, 0.75 ± 0.07; 5–6 d MD, 0.50 ± 0.07; P < 0.05;
t = 3.30; monocular: 3 d MD, 1.07 ± 0.13; 5–6 d MD, 0.70 ± 0.07;
P < 0.05; t = 2.65) (Fig. 1F). Despite the apparent absence of this
homeostatic plasticity mechanism in 6JOla mice, intrinsic signal
response magnitude was comparable in the two substrains
(control 6J contralateral: 2.43 ± 0.45; 6JOla contralateral: 2.92 ±
0.27; P = 0.37; t = 0.94) (Fig. S1A), as was retinotopic map or-
ganization (6J scatter: 3.02 ± 0.43; 6JOla scatter: 2.90 ± 0.36; P =
0.84; t = 0.21) (Fig. 1C and Fig. S1B).
One known genetic difference between the two substrains is
that the gene encoding α-synuclein, a neuronally expressed
protein associated with neurodegeneration but of uncertain
function, is absent in the 6JOla strain (19). Therefore, we
assessed the homeostatic component of OD plasticity (i.e., the
up-regulation of open-eye and deprived-eye responses between
3 d and 5–6 d of MD) in α-synuclein KO mice and compared
them with WT littermates of the C57BL/6JCrl substrain (which
exhibit juvenile OD plasticity identical to that of the 6J substrain,
including the homeostatic component). We found that the ho-
meostatic component of OD plasticity was normal in the α-syn-
uclein KO mice (Fig. S2), indicating that some other genetic
difference is likely responsible for the absence of homeostatic
plasticity in the 6JOla substrain.
These results indicate that 6JOla mice have a robust OD
plasticity deficit that is specific to the homeostatic component
that manifests itself at the 5–6 d MD time point. Accordingly, the
increase in cortical responsiveness to the nondeprived eye is
completely absent in 6JOla mice during the critical period.
Homeostatic Synaptic Scaling Is Absent in 6JOla Mice. Synaptic
scaling, a key mechanism underlying response homeostasis in the
brain, is typically assayed by measuring the amplitude of minia-
ture excitatory postsynaptic currents (mEPSCs), which result
from the spontaneous release of single vesicles of a neurotrans-
mitter. There is some evidence indicating that synaptic scaling
might underlie the homeostatic component of OD plasticity
that occurs between brief (3 d) and a longer period (5–6 d) of
MD in the mouse visual cortex (13). However, this had not been
previously tested directly by measuring the effect of visual dep-
rivation on mEPSC amplitudes in the visual cortex of mice
lacking homeostatic plasticity. Therefore, because the homeo-
static component of OD plasticity was absent in 6JOla mice (Fig.
1), we tested whether homeostatic synaptic scaling of mEPSC
amplitudes was present in the 6JOla substrain after visual dep-
rivation during the critical period. A previous study showed that
brief periods (1–3 d) of DE result in scaling-up of mEPSC am-
plitude in V1 of juvenile rats and 6J mice, which can be detected
ex vivo in acute brain slices (16, 20). We chose DE over MD
because the latter would be expected to have different effects on
different synapses depending on whether they represent the
open eye or the closed eye, whereas the former would be
expected to affect the synapses of both eyes on a given neuron
similarly. Therefore, we recorded mEPSCs from L2/3 pyramidal
neurons in V1 of acute brain slices from 6J and 6JOla mice that
had been either dark-exposed for 3 d (P25–P28) or reared on
a normal 12-h light/12-h dark cycle (Fig. 2A).
Two-way ANOVA indicated significant effects of DE (P =
0.008; F1,20= 8.91) and substrain (P = 0.002; F1,20= 13.66), as
well as an interaction between DE and substrain (P = 0.049;
F1,20= 4.39). Using post hoc pairwise comparisons, we found
that, as expected, mEPSC amplitude was significantly larger in
the dark-exposed 6J mice compared with age-matched 6J
Intrinsic signal imaging
Control 3d MD 5-6d MD
Control 3d MD5-6d MD
Control3d MD5-6d MD
Control3d MD5-6d MD
Closed eye (monocular zone)
Closed eye (binocular zone)
Open eye (binocular zone)
ODI (binocular zone)
static component of OD plasticity during the critical period in the 6JOla
substrain. (A) Schematic of the intrinsic signal imaging setup. (B) Deprivation
and imaging timeline. (C) Representative retinotopic maps from the two
C57BL/6 substrains investigated overlaid over the cortical vasculature. (Scale
bar: 1 mm.) (D) Binocular zone responsiveness to closed-eye stimulation
during MD. (E) Binocular zone responsiveness to open-eye stimulation dur-
ing MD. (F) Monocular zone responsiveness to closed-eye stimulation during
MD. (G) ODI in the binocular zone during MD. For 6J: control, n = 10; 3 d MD,
n = 8; 5–6 d MD, n = 9. For 6JOla: control, n = 22; 3 d MD, n = 6; 5–6 d MD,
n = 11. **P < 0.01; *P < 0.05 for comparisons between time points. In D–F,
response values are mean ΔR/R normalized to control values ± SEM.
In vivo intrinsic signal imaging reveals the absence of the homeo-
| www.pnas.org/cgi/doi/10.1073/pnas.1112204109Ranson et al.
controls, indicating the presence of synaptic scaling (6J control:
10.4 ± 0.4 pA; 6J dark-exposed: 12.7 ± 0.2 pA; 22% increase;
P = 0.010; t = 3.68) (Fig. 2 A and B). In contrast, there was no
difference in mEPSC amplitude between dark-exposed and
control 6JOla mice (6JOla control: 9.7 ± 0.2 pA; 6JOla dark-
exposed: 10.1 ± 0.7 pA; 4% increase; P = 0.55, t = 0.62), in-
dicating a lack of synaptic scaling in this substrain (Fig. 2 A
The difference in DE-induced mEPSC scaling between the
substrains could not be explained by differences in baseline
mEPSC amplitudes, which were similar in the 6J and 6JOla mice
(P = 0.30; t = 1.11) (Fig. 2B). Neither substrain nor DE had any
effect on mEPSC rise time (effect of substrain: P = 0.87, F1,20=
0.02; effect of DE: P = 0.79, F1,20= 0.07; interaction: P = 0.46;
F1,20= 0.57; two-way ANOVA) (Fig. 2C) or mEPSC frequency
(effect of substrain: P = 0.51 F1,20= 0.44; effect of DE: P = 0.97,
F1,20 = 0.00; interaction, P = 0.71, F1,20 = 0.14; two-way
ANOVA) (Fig. 2D). Resting membrane potential and input
resistance were also similar across all groups (Table S1).
The conclusions derived from the statistical comparisons be-
tween individual animals described above were corroborated by
further analysis of the distributions of mEPSC amplitudes be-
tween the 6J and 6JOla substrains. The distribution of mEPSC
amplitudes for individual neurons was shifted toward larger
values after DE in 6J mice [D = 0.49; P = 0.002, Kolmogorov–
Smirnov (KS) test] (Fig. 2E), but not in 6JOla mice (D = 0.23;
P = 0.60, KS test). We also found that the distribution of raw
mEPSC amplitudes was shifted toward larger values in dark-
exposed 6J mice than in control 6J mice (D = 0.24; P < 0.001, KS
test) (Fig. 2F). In contrast, the distribution of mEPSC ampli-
tudes was similar in dark-exposed and control 6JOla mice (D =
0.03; P = 0.32, KS test) (Fig. 2G). In summary, DE resulted in
robust homeostatic synaptic scaling in the 6J mice, but not in the
Homeostatic synaptic scaling is thought to involve a pro-
portional change in the strength of all inputs to a neuron, re-
ferred to as multiplicative scaling (12). Multiplicative scaling has
been reported in cultured neurons after pharmacologic activity
blockade (12) and in L2/3 pyramidal neurons in the visual cortex
after DE during the critical period, but not in adulthood (16).
Therefore, we investigated whether the increase in mEPSC
amplitude in dark-exposed 6J mice that we report here resulted
from multiplicative scaling. Mean mEPSC amplitude was 22%
larger in the 6J dark-exposed mice compared with the 6J control
mice; thus, we multiplied 6J control mEPSC amplitudes by 1.22
to generate a scaled distribution (6J CTRLscaled), and compared
these values with the amplitudes of mEPSCs recorded in brain
slices from dark-exposed 6J mice (Fig. S3). There was no signifi-
cant difference between the 6J dark-exposed and 6J CTRLscaled
distributions (D = 0.03; P = 0.22, KS test) (Fig. S3), consistent
with multiplicative scaling-up of inputs to L2/3 pyramidal neurons
in the 6J mice after DE during the critical period.
Normal Adult OD Plasticity in 6JOla and TNFα−/−Mice. Induction of
adult OD plasticity requires a longer period of altered visual
experience and is mediated primarily by an increased cortical
response to the open eye (9, 11, 21). Having found that 6JOla
mice lack a (homeostatic) increase in open-eye responsiveness
during the critical period, we tested whether the same deficit
persisted into adulthood (Fig. 3A). Interestingly, the two sub-
strains showed nearly identical plasticity profiles in adulthood
after 7 d of MD (substrain-deprivation interaction: P = 0.69;
F1,20= 0.16) (Fig. 3B), with a clear shift in the OD index (ODI)
in both substrains (deprivation effect: P < 0.001, F1,20= 28.18;
substrain effect: P = 0.24, F1,20= 1.46; two-way ANOVA). The
shift in ODI was mediated purely by an increase in the open-eye
response (6J: control, 1.29 ± 0.17; 7 d MD, 2.05 ± 0.28; P < 0.05;
t = 2.29; 6JOla: control, 1.86 ± 0.26; 7 d MD, 3.02 ± 0.43; P <
0.05; t = 2.31) (Fig. 3C), whereas the deprived eye responses
were unchanged in both strains (6J: control, 2.43 ± 0.45; 7 d MD,
2.13 ± 0.33; P = 0.60; t = 0.54; 6JOla: control, 2.92 ± 0.66; 7 d
MD, 2.65 ± 0.33; P = 0.54; t = 0.64). No change in deprived
monocular area responses was observed in either the 6J mice
(control, 2.41 ± 0.32; 7 d MD, 2.40 ± 0.17) or the 6JOla mice
(control, 3.12 ± 0.24; 7 d MD, 2.81 ± 0.44).
These findings suggest that the increase in the open-eye re-
sponse in adults might operate by a mechanism distinct from the
homeostatic process that appears to underlie increased open-eye
responsiveness during the critical period. To further explore this
possibility, we measured adult OD plasticity in mice lacking
TNFα (bred on a 6J background), which has been shown to be
required for the homeostatic component of OD plasticity in ju-
venile mice and homeostatic synaptic scaling in organotypic slice
cultures (13, 15). Strikingly, OD plasticity appeared completely
Example traces showing mEPSCs recorded in brain slices from 6J and 6JOla
control (CTRL) and dark-exposed (DE) mice. (Scale bars: 10 pA; 200 ms.) (B)
Effect of DE on mEPSC amplitude. Data are shown as mean for each mouse
(small symbols) and grand mean ± SEM (large symbols); n = 6 mice/group.
**P < 0.01 for 6J DE vs. all other groups. (C) Effect of DE on mEPSC rise time.
Symbols are as in B. (D) Effect of DE on mEPSC frequency. Symbols are as in B.
(E) Cumulative distribution plot of mean mEPSC amplitude for individual
neurons in the four groups. 6J control, n = 27; 6J DE, n = 25; 6JOla control,
n = 20; 6JOla DE, n = 21. (F) Cumulative distribution plot showing the effect
of DE on raw mEPSC amplitude for 6J mice. n = 50 mEPSCs/neuron. mEPSCs
>30 pA are not shown. (G) As in F for 6JOla mice.
Ex vivo homeostatic synaptic scaling is absent in 6JOla mice. (A)
Ranson et al.PNAS
| January 24, 2012
| vol. 109
| no. 4
normal in adult TNFα−/−mice, further indicating that adult OD
plasticity is not mediated by the same homeostatic mechanisms
that operate during juvenile plasticity (ODI: control, 0.25 ± 0.05;
7 d MD, -0.04 ± 0.05; P < 0.01; t = 4.04; deprived eye: control,
3.61 ± 0.32; 7 d MD, 2.93 ± 0.12; P = 0.11; t = 1.95; open eye:
control, 2.20 ± 0.28; 7 d MD, 3.23 ± 0.27; P < 0.05; t = 2.60)
(Fig. 3 B and C).
Previous studies have shown that αCaMKII autophosphor-
ylation is critical for experience-dependent potentiation in adult
somatosensory cortex (22), and that the dependence of cortical
long-term potentiation (LTP) on αCaMKII increases with age in
the visual cortex (23). This raises the possibility that the expe-
rience-dependent cortical plasticity seen in adult visual cortex
depends on αCaMKII. Thus, we examined adult plasticity in the
visual cortex of αCaMKII mutants that lack the ability to auto-
phosphorylate at T286. We found a complete absence of adult
OD plasticity in T286A mice (ODI: control, 0.29 ± 0.03; 7 d MD,
0.31 ± 0.05; P = 0.76; t = 0.32; deprived eye: control, 2.34 ± 0.30;
7 d MD, 2.32 ± 0.21; P = 0.96; t = 0.05; open eye: control, 1.33 ±
0.22; 7 d MD, 1.28 ± 0.21; P = 0.88; t = 0.15) (Fig. 3 B and C),
supporting the concept that αCaMKII autophosphorylation is
vital for the increase in responsiveness to the open eye in mon-
ocularly deprived adults. The fact that our T286A mice had been
bred on a mixed 6J and 6JOla background is unlikely of conse-
quence here, given that the WT in both the 6J and 6JOla sub-
strains exhibited normal adult OD plasticity (in contrast to the
In this study, we have shown that a previously unknown ho-
meostatic plasticity deficit exists in a commonly studied substrain
of C57BL/6 mouse, the 6JOla substrain. The deficit in this sub-
strain was observed both at the systems level using intrinsic signal
imaging and at the cellular level using acute brain slices ex vivo,
and could be revealed by either DE or monocular eyelid closure.
Therefore this study demonstrates directly a mechanistic asso-
ciation between visual deprivation-induced homeostasis at the
systems level and visual deprivation-induced synaptic scaling at
the cellular level. We exploited this finding to ask whether adult
visual cortex plasticity, although phenomenologically similar to
juvenile plasticity, is mechanistically distinct. Although the in-
creased cortical responsiveness to open-eye stimulation is en-
tirely absent in juvenile 6JOla mice, it appears to be normal in
adult 6JOla mice, indicating that the two processes occur by
different mechanisms. Consistent with this interpretation, we
found normal adult OD plasticity in TNFα−/−mice, which were
previously shown to completely lack the increase in both open-
eye and closed-eye responsiveness after 5–6 d of MD during the
critical period, and also to lack synaptic scaling (13). Conversely,
visual cortex plasticity as assessed by 4 d of MD during the
critical period is markedly impaired but not abolished in T286A
mice, which lack autophosphorylatable αCaMKII (24, 25),
whereas we show here that OD plasticity in response to MD is
entirely absent in adult T286A mice, indicating a difference in
the degree of involvement of αCaMKII autophosphorylation in
juvenile and adult plasticity. Thus, our data are consistent with
a model in which both TNFα and αCaMKII autophosphorylation
are involved in juvenile plasticity, but αCaMKII autophosphor-
ylation is dominant in adult plasticity.
Strain Differences in Cortical Plasticity. C57BL/6 substrains are
often considered interchangeable, and a lack of reference to
which specific substrain was used is common in experimental
methodologies. However, there is growing awareness of the
significant genetic and phenotypic differences between even
closely related mouse strains and substrains (19, 26–28), and the
present study demonstrates the complete absence in a C57BL/6
substrain of what is generally considered a ubiquitous plasticity
mechanism in both mice and rats (14, 16, 20, 29). This finding
underscores the importance of reporting mouse background
substrains and of using appropriate littermate controls. This
finding also prompts the question of the critical genetic differ-
ence in the 6JOla substrain. The most prominent genetic dif-
ference between the two substrains is the absence of α-synuclein
in 6JOla mice; however, we have shown that this protein is not
required for the homeostatic component of OD plasticity. The
numerous other genetic differences between the two substrains
include other absent or mutated genes and a multitude of copy
number variations (26, 27, 30). The analysis of these differences
may allow identification of a unique molecular mechanism of
homeostatic plasticity as more is discovered about the signaling
Homeostasis in Critical Period Cortical Plasticity. Several previous
studies have provided evidence for the existence of homeostatic
plasticity in response to reduced neuronal drive in a number of
central brain regions, including the hippocampus and neocortex
(12–14, 20, 31). Theoretically, a homeostatic response to visual
deprivation could occur through a change in intrinsic membrane
properties (32), a change in the excitatory/inhibitory balance at
the circuit level (33), or scaling of the efficacy of individual
synapses at presynaptic (34, 35) and/or postsynaptic loci (15, 36–
41). Although there is evidence indicating that each of these
mechanisms exists, our data (showing a change in mEPSC am-
plitude but not in frequency) imply that postsynaptic scaling at
excitatory to excitatory cell synapses is an essential mechanism
* ** *
Closed eye Open eye
Intrinsic signal imaging
in T286A mice. (A) Adult deprivation and imaging timeline. (B) ODI shift
after 7 d of MD. (C) Responsiveness to closed-eye (Left) and open-eye (Right)
stimulation before and after 7 d of MD. For 6J and 6JOla control and 7 d
MD, n = 6; T286A control, n = 5; 7 d MD, n = 6; TNFα−/−control, n = 5; 7 d
control, n = 5; 7 d MD, n = 6. **P < 0.01; *P < 0.05. Response values are
mean ΔR/R normalized to control values ± SEM. Gray box indicates control 6J
mean ± SEM.
Adult OD plasticity is normal in 6JOla and TNFα−/−mice but absent
| www.pnas.org/cgi/doi/10.1073/pnas.1112204109 Ranson et al.
for homeostatic plasticity in the visual cortex. This conclusion is
consistent with studies showing that TNFα−/−mice lack ho-
meostatic plasticity in the visual cortex during the critical period,
and that organotypic slice cultures derived from the same mutant
lack synaptic scaling (13, 15). Similarly, all aspects of OD plas-
ticity, including homeostatic plasticity, are absent in Arc KO
mice during the critical period, and Arc also has been strongly
implicated in synaptic scaling (42, 43).
The apparent normality of visual response magnitudes and
visual cortex map formation, as well as baseline mEPSC ampli-
tude, in the 6JOla mice is surprising given the apparent severity
of the homeostatic plasticity deficit that we observed. Curiously,
this is a recurrent theme in studies of homeostatic plasticity
in vivo, because TNFα and Arc knockouts are also reported to
develop normally (13, 42), which might reflect the fact that
multiple homeostatic mechanisms operate in parallel (33). In
this regard, it may be noteworthy that although we have found
evidence indicating that the 6JOla mouse lacks a homeostatic
mechanism that up-regulates synaptic gain, this mouse may have
an intact mechanism for down-regulation (12, 44). Homeostatic
down-regulation may mitigate the more significant risk of exci-
totoxicity during the intense synaptogenesis that occurs during
early postnatal development.
Distinct Mechanisms of Adult Visual Cortical Plasticity. The consen-
sus opinion is that induction of adult mouse OD plasticity
requires ∼7 d and is mediated by an increased cortical response
to the open eye (9, 11). We have shown that 6JOla mice lack
both the MD-induced increase in open eye responsiveness and
synaptic scaling during the critical period for OD plasticity, but
have normal open-eye potentiation in adults. We interpret these
findings to indicate that juvenile OD plasticity operates through
a different mechanism than adult OD plasticity. An alternative
explanation could be that homeostatic plasticity matures later in
6JOla mice than in 6J mice, and that adult open-eye potentiation
in both strains occurs via a homeostatic mechanism. However,
we believe that this is unlikely for at least two reasons. First, we
observed no homeostatic increase in the response of deprived
adult monocular cortex in either substrain, which would be
predicted if adult plasticity were due to homeostatic plasticity
(13, 20). Second, delayed development seems unlikely, given that
previous studies in the mouse visual cortex have found that
synaptic scaling is typically present in mice at least 2 wk younger
than those in the present study. Visual deprivation-induced
synaptic scaling in the cortex has been observed as early as P14 in
rats and P21 in mice, with earlier observations complicated by
the fact that the eyes do not open until P14 (16, 20).
There is evidence suggesting that the capacity for synaptic
scaling might be down-regulated during postnatal development.
For example, in the rat hippocampus, a period of in vivo activity
blockade by TTX application results in scaling-up of pyramidal
neuron mEPSC amplitudes in juvenile animals, but has no effect
on mEPSC amplitude in adults (18), whereas the susceptibility to
TTX incubation-induced scaling-up of mEPSC amplitude
declines significantly between P4 and P8 (17). Interestingly,
a nonmultiplicative form of DE-induced synaptic plasticity,
which comprises a nonuniform increase in synaptic weights, has
been reported to persist into adulthood in L2/3 of the mouse
visual cortex (16). Which subset of synapses are preferentially
strengthened by this mechanism is unknown, although one pos-
sibility is that the subset is more active (in the case of DE, active
because of spontaneous activity). In the case of adult MD, the
more active inputs are those relaying input from the open eye,
which would be scaled preferentially. This interpretation is
reminiscent of the idea of “metaplasticity” and a homeostatically
regulated threshold for synaptic modification (45), and is con-
sistent with previous experimental findings of enhanced LTP and
adult OD plasticity after dark rearing (46, 47). Regardless of the
nonmultiplicative scaling mechanism that supports adult OD
plasticity, our data demonstrate that multiplicative scaling de-
pendent on TNFα is not required. A developmentally regulated
transition between synaptic scaling during the critical period and
homosynaptic LTP in adulthood is also consistent with the in-
creased importance of αCaMKII activity in adult experience-
dependent plasticity, given the known crucial role of αCaMKII in
several forms of cortical LTP (48).
Materials and Methods
Subjects. All animal procedures were performed in accordance with the U.K.
Animals (Scientific Procedures) Act 1986. The C57BL/6J (Jackson) mice were
supplied by Charles River Laboratories. The C57BL/6JOlaHsd (Harlan) mice
were supplied by Harlan. The α-synuclein knockout mice were kindly sup-
plied by Prof. Vladimir Buchman (Cardiff University, Cardiff, United King-
dom.) and were congenic C57BL/6JCrl (backcrossed for 12 generations).
TNFα−/−mice were kindly supplied by Dr. Denise McDonald (Queens Uni-
versity, Belfast, Northern Ireland) and were inbred on a homozygous C57BL/
6J strain originally sourced from Bantin & Kingman and generated by tar-
geting C57BL/6 ES cells. T286A mice were obtained from Prof. Alcino Silva
(University of California, Los Angeles, CA). These mice were originally con-
genic C57BL/6J (backcrossed for five generations) and were then inbred
(cousin matings) over 14 y, during which time they were outbred with
C57BL/6JOlaHsd mice on three separate occasions. For the ex vivo experi-
ments, six mice per group were used (24 mice total). For in vivo experiments,
a total of 92 juvenile mice and 43 adult mice were used.
Visual Deprivation. Mice were normally reared on a 12-h light/12-h dark cycle.
For in vivo experiments, mice were monocularly deprived by eyelid suture
under isoflurane anesthesia (2% in O2, 0.6 L/min). For critical period studies,
deprivation began at P26–P27 and lasted for 3 d or 5–6 d, whereas for adult
studies, deprivations began at P90–P120 and lasted for 7 d. The integrity of
the deprivation was checked daily and immediately before intrinsic signal
imaging. The experiment was discontinued if the deprivation was impaired.
For ex vivo experiments, mice were housed in a dark room for 3 d beginning
at P25–P28. Night vision goggles were used for animal care. On the exper-
imental day, mice were anesthetized with isoflurane in the dark before
brain slice preparation.
Ex Vivo Electrophysiology. Mice were anesthetized with isoflurane and de-
capitated. Brain slices were prepared and mEPSC recordings from L2/3 py-
ramidal neurons in V1 were made as described previously (20, 49). Then ≥100
mEPSCs per cell were detected and measured with the MiniAnalysis program
(Synaptosoft). All experiments and analyses were performed blind to the
sensory experience and background strain of the animal.
In Vivo Intrinsic Signal Imaging. Acute intrinsic signal imaging was performed
in the primary visual cortex contralateral to the deprived eye. The visual
cortex was imaged transcranially using 0.8–1% isoflurane in O2at 0.3 L/min,
supplemented with 25 μg chlorprothixene as described previously to mea-
sure OD (13, 21) and retinotopic map organization (13, 21). For quantifica-
tion of OD, visual responses were elicited using a 0.03-cycles/degree square-
wave grating drifting at 2 cycles per second presented in the binocular or
monocular visual field, with the stimulated eye determined by computer-
controlled eye shutters. The ODI was calculated by the formula (C − I)/(C + I),
where C and I are the contralateral and ipsilateral response magnitudes,
respectively. Response magnitudes are presented as ΔR/R values, where R is
light reflected. For measurement of retinotopic map organization, the stim-
ulus was a white horizontal bar of height 1–2 degrees drifting at a rate of
0.125 Hz. Map scatter was calculated by comparing each pixel in the map with
its immediate neighbors in a 5 × 5 pixel box as described previously (50, 51).
Statistics. For ex vivo mEPSC data, mean values were calculated for each
neuron, and grand mean ± SEM values were then calculated for each mouse.
In vivo data are shown as mean ± SEM. Datasets were compared using two-
way ANOVA, the t test, or the KS test. All tests were two-tailed, and α was
set at 0.05.
ACKNOWLEDGMENTS. We thank Thomas Mrsic-Flogel for critical comments
on the manuscript. This work was supported by funding from European Com-
munity’s Seventh Framework Programme FP2007-2013, under Grant Agree-
ment 223326 (to F.S.), the Medical Research Council (to K.F.), and National
Institute of Mental Health’s Conte Center Grant P50-MH0779720 (to K.F.).
Ranson et al. PNAS
| January 24, 2012
| vol. 109
| no. 4
1. Hubel DH, Wiesel TN (1970) The period of susceptibility to the physiological effects of
unilateral eye closure in kittens. J Physiol 206:419–436.
2. Olson CR, Freeman RD (1980) Profile of the sensitive period for monocular deprivation
in kittens. Exp Brain Res 39:17–21.
3. Gordon JA, Stryker MP (1996) Experience-dependent plasticity of binocular responses
in the primary visual cortex of the mouse. J Neurosci 16:3274–3286.
4. Heynen AJ, et al. (2003) Molecular mechanism for loss of visual cortical responsiveness
following brief monocular deprivation. Nat Neurosci 6:854–862.
5. Frenkel MY, Bear MF (2004) How monocular deprivation shifts ocular dominance in
visual cortex of young mice. Neuron 44:917–923.
6. Hensch TK (2005) Critical period mechanisms in developing visual cortex. Curr Top Dev
7. Daw NW, Reid SN, Wang XF, Flavin HJ (1995) Factors that are critical for plasticity in
the visual cortex. Ciba Found Symp 193:258–276.
8. Tagawa Y, Kanold PO, Majdan M, Shatz CJ (2005) Multiple periods of functional
ocular dominance plasticity in mouse visual cortex. Nat Neurosci 8:380–388.
9. Sawtell NB, et al. (2003) NMDA receptor-dependent ocular dominance plasticity in
adult visual cortex. Neuron 38:977–985.
10. Lehmann K, Löwel S (2008) Age-dependent ocular dominance plasticity in adult mice.
PLoS ONE 3:e3120.
11. Sato M, Stryker MP (2008) Distinctive features of adult ocular dominance plasticity.
J Neurosci 28:10278–10286.
12. Turrigiano GG, Leslie KR, Desai NS, Rutherford LC, Nelson SB (1998) Activity-de-
pendent scaling of quantal amplitude in neocortical neurons. Nature 391:892–896.
13. Kaneko M, Stellwagen D, Malenka RC, Stryker MP (2008) Tumor necrosis factor-α
mediates one component of competitive, experience-dependent plasticity in de-
veloping visual cortex. Neuron 58:673–680.
14. Mrsic-Flogel TD, et al. (2007) Homeostatic regulation of eye-specific responses in
visual cortex during ocular dominance plasticity. Neuron 54:961–972.
15. Stellwagen D, Malenka RC (2006) Synaptic scaling mediated by glial TNF-α. Nature
16. Goel A, Lee HK (2007) Persistence of experience-induced homeostatic synaptic plas-
ticity through adulthood in superficial layers of mouse visual cortex. J Neurosci 27:
17. Huupponen J, Molchanova SM, Taira T, Lauri SE (2007) Susceptibility for homeostatic
plasticity is down-regulated in parallel with maturation of the rat hippocampal syn-
aptic circuitry. J Physiol 581:505–514.
18. Echegoyen J, Neu A, Graber KD, Soltesz I (2007) Homeostatic plasticity studied using
in vivo hippocampal activity-blockade: Synaptic scaling, intrinsic plasticity and age-
dependence. PLoS ONE 2:e700.
19. Specht CG, Schoepfer R (2001) Deletion of the α-synuclein locus in a subpopulation of
C57BL/6J inbred mice. BMC Neurosci 2:11.
20. Desai NS, Cudmore RH, Nelson SB, Turrigiano GG (2002) Critical periods for experi-
ence-dependent synaptic scaling in visual cortex. Nat Neurosci 5:783–789.
21. Hofer SB, Mrsic-Flogel TD, Bonhoeffer T, Hübener M (2006) Prior experience enhances
plasticity in adult visual cortex. Nat Neurosci 9:127–132.
22. Glazewski S, Giese KP, Silva A, Fox K (2000) The role of α-CaMKII autophosphorylation
in neocortical experience-dependent plasticity. Nat Neurosci 3:911–918.
23. Kirkwood A, Silva A, Bear MF (1997) Age-dependent decrease of synaptic plasticity in
the neocortex of α-CaMKII mutant mice. Proc Natl Acad Sci USA 94:3380–3383.
24. Taha SA, Stryker MP (2005) Ocular dominance plasticity is stably maintained in the
absence of α calcium calmodulin kinase II (αCaMKII) autophosphorylation. Proc Natl
Acad Sci USA 102:16438–16442.
25. Taha S, Hanover JL, Silva AJ, Stryker MP (2002) Autophosphorylation of αCaMKII is
required for ocular dominance plasticity. Neuron 36:483–491.
26. Zurita E, et al. (2011) Genetic polymorphisms among C57BL/6 mouse inbred strains.
Transgenic Res 20:481–489.
27. Egan CM, Sridhar S, Wigler M, Hall IM (2007) Recurrent DNA copy number variation in
the laboratory mouse. Nat Genet 39:1384–1389.
28. Heimel JA, Hermans JM, Sommeijer JP, Levelt CN; Neuro-Bsik Mouse Phenomics
Consortium (2008) Genetic control of experience-dependent plasticity in the visual
cortex. Genes Brain Behav 7:915–923.
29. Turrigiano GG (2008) The self-tuning neuron: Synaptic scaling of excitatory synapses.
30. Specht CG, Schoepfer R (2004) Deletion of multimerin-1 in α-synuclein–deficient mice.
31. Thiagarajan TC, Lindskog M, Tsien RW (2005) Adaptation to synaptic inactivity in
hippocampal neurons. Neuron 47:725–737.
32. Breton JD, Stuart GJ (2009) Loss of sensory input increases the intrinsic excitability of
layer 5 pyramidal neurons in rat barrel cortex. J Physiol 587:5107–5119.
33. Maffei A, Turrigiano GG (2008) Multiple modes of network homeostasis in visual
cortical layer 2/3. J Neurosci 28:4377–4384.
34. Han EB, Stevens CF (2009) Development regulates a switch between post- and pre-
synaptic strengthening in response to activity deprivation. Proc Natl Acad Sci USA 106:
35. Zhao C, Dreosti E, Lagnado L (2011) Homeostatic synaptic plasticity through changes
in presynaptic calcium influx. J Neurosci 31:7492–7496.
36. Aoto J, Nam CI, Poon MM, Ting P, Chen L (2008) Synaptic signaling by all-trans retinoic
acid in homeostatic synaptic plasticity. Neuron 60:308–320.
37. Gainey MA, Hurvitz-Wolff JR, Lambo ME, Turrigiano GG (2009) Synaptic scaling re-
quires the GluR2 subunit of the AMPA receptor. J Neurosci 29:6479–6489.
38. Goel A, et al. (2006) Cross-modal regulation of synaptic AMPA receptors in primary
sensory cortices by visual experience. Nat Neurosci 9:1001–1003.
39. Goel A, et al. (2011) Phosphorylation of AMPA receptors is required for sensory
deprivation-induced homeostatic synaptic plasticity. PLoS ONE 6:e18264.
40. O’Brien RJ, et al. (1998) Activity-dependent modulation of synaptic AMPA receptor
accumulation. Neuron 21:1067–1078.
41. Wierenga CJ, Ibata K, Turrigiano GG (2005) Postsynaptic expression of homeostatic
plasticity at neocortical synapses. J Neurosci 25:2895–2905.
42. McCurry CL, et al. (2010) Loss of Arc renders the visual cortex impervious to the effects
of sensory experience or deprivation. Nat Neurosci 13:450–457.
43. Shepherd JD, et al. (2006) Arc/Arg3.1 mediates homeostatic synaptic scaling of AMPA
receptors. Neuron 52:475–484.
44. Goold CP, Nicoll RA (2010) Single-cell optogenetic excitation drives homeostatic
synaptic depression. Neuron 68:512–528.
45. Bear MF (2003) Bidirectional synaptic plasticity: From theory to reality. Philos Trans R
Soc Lond B Biol Sci 358:649–655.
46. Kirkwood A, Rioult MC, Bear MF (1996) Experience-dependent modification of syn-
aptic plasticity in visual cortex. Nature 381:526–528.
47. He HY, Hodos W, Quinlan EM (2006) Visual deprivation reactivates rapid ocular
dominance plasticity in adult visual cortex. J Neurosci 26:2951–2955.
48. Lisman J, Schulman H, Cline H (2002) The molecular basis of CaMKII function in syn-
aptic and behavioural memory. Nat Rev Neurosci 3:175–190.
49. Kaneko M, et al. (2010) Constitutively active H-ras accelerates multiple forms of
plasticity in developing visual cortex. Proc Natl Acad Sci USA 107:19026–19031.
50. Smith SL, Trachtenberg JT (2010) The refinement of ipsilateral eye retinotopic maps is
increased by removing the dominant contralateral eye in adult mice. PLoS ONE 5:
51. Cang J, et al. (2005) Ephrin-As guide the formation of functional maps in the visual
cortex. Neuron 48:577–589.
| www.pnas.org/cgi/doi/10.1073/pnas.1112204109 Ranson et al.
Ranson et al. 10.1073/pnas.1112204109
reflectance/reflectance x 10
Map scatter (deg)
absolute intrinsic signal response magnitude is similar in control animals of the two substrains. (B) Retinotopic map scatter is comparable in the two substrains.
ΔR/R (R, reflectance) and scatter values are shown as mean ± SEM.
In vivo intrinsic signal imaging shows similar V1 intrinsic signal response magnitude and retinotopic map in the 6J and 6JOla substrains. (A) The
Closed eye Open eye
Control 5-6d MD
s i l a
animals with α-synuclein and without α-synuclein. (A) Closed eye. (B) Open eye. 6Crl: control, n = 9; 5–6 d MD, n = 6. α-synuclein: control, n = 5; 5–6 d MD, n = 5.
**P < 0.01; *P < 0.05. Response values are shown as mean response normalized to control value ± SEM.
In vivo intrinsic signal imaging shows a normal late homeostatic component of OD plasticity in α-synuclein KO mice. OD plasticity is comparable in
mEPSC amplitude for 6J mice. n = 50 mEPSCs/neuron. The scaled distribution (6J CONscaled) was generated by multiplying 6J control (CON) values by 1.22, the
difference in mean mEPSC amplitude between 6J control and 6J dark-exposed groups. (B) Histogram showing the distribution of raw mEPSC amplitudes for 6J
dark-exposed and 6J CONscaledgroups. mEPSC amplitudes are shown in 1-ms bins. The blue line indicates the difference between the 6J dark-exposed and 6J
CONscaleddistributions (CONscaledsubtracted from DE).
Dark exposure (DE) results in multiplicative scaling of mEPSC amplitudes in 6J mice. (A) Cumulative distribution plot showing the effect of DE on raw
Ranson et al. www.pnas.org/cgi/content/short/11122041091 of 2
Table S1.Dark exposure (DE) does not affect passive properties of L2/3 pyramidal neurons
Resting membrane potential at break-in, mVInput resistance, MΩ
−77 ± 1
−77 ± 0
−76 ± 1
−77 ± 1
71 ± 9
88 ± 6
79 ± 9
85 ± 11
Data are shown as grand mean ± SEM. There was no difference in resting membrane potential among the
groups (effect of strain: P = 0.20, F1,20= 1.79; effect of DE: P = 0.55, F1,20= 0.38; interaction: P = 1.00, F1,20= 0.00;
two-way ANOVA). There also was no difference in input resistance between groups (effect of strain: P = 0.78,
F1,20= 0.08; effect of DE: P = 0.23, F1,20= 1.51; interaction: P = 0.60, F1,20= 0.29; two-way ANOVA).
Ranson et al. www.pnas.org/cgi/content/short/1112204109 2 of 2