J Neurophysiol 114: 2023–2032, 2015.
First published August 12, 2015; doi:10.1152/jn.00076.2015.
Compensatory plasticity in the olfactory epithelium: age, timing, and
reversibility
Casey N. Barber and David M. Coppola
Department of Biology, Randolph-Macon College, Ashland, Virginia
Submitted 23 January 2015; accepted in final form 2 August 2015
naris-occlusion; deprivation; olfactory-receptors; transduction
homeostatically to environmental
perturbations is a defining characteristic of living systems:
Shivering when body temperature declines, pancreatic insulin secretion following a meal, and the emergence of hyperexcitability after muscle fiber denervation are all classic
negative-feedback-assisted mechanisms of maintaining the
“milieu intérieur,” as Claude Bernard termed it (Cannon
1929). Homeostatic processes also exist within our senses,
although this fact may be less widely appreciated because
sensory systems are, foremost, reporters of environmental
change, not promoters of stasis. However, here too, countervailing forces are set in motion when change persists in
the statistical distribution of sensory stimuli. The most
extensively studied example of this phenomenon is the
nearly universal process of sensory adaptation whereby
sensory responses rapidly decline following prolonged exposure to intense stimuli (Fain et al. 2001; Wark et al.
2007). While these examples of compensatory responses to
environmental perturbations certainly point to the “robustness” of life processes, including those in the nervous
THE ABILITY TO RESPOND
Address for reprint requests and other correspondence: D. Coppola, Dept. of
Biology, Randolph-Macon College, 304 Caroline St., Ashland, VA 23005
(e-mail: [email protected]).
www.jn.org
system, which instances meet the strict definition of homeostasis—possessing negative-feedback, set points, error signals, and precision— has been fully determined for very few
(Davis 2006).
This study concerns compensatory plasticity, another homeostatic-like process in sensory systems, which occurs over
more extended periods of change in average stimulus intensity
than is the case of sensory adaptation. The phrase “compensatory plasticity,” originally used to refer to the appropriation of
cortical territory by remaining sensory systems after one system was ablated, has also been extended to describe the gain
changes within a given sensory channel following long-term
(days or weeks) stimulus deprivation or enrichment (cf. Rauschecker 1995; Waggener and Coppola 2007). Examples of
such compensatory gain adjustments abound in the sensory
literature: In the visual system dark-rearing increases the lightevoked synaptic currents of retinal ganglion cells and causes
other morphological and physiological changes in the retina
that appear to be compensatory (He et al. 2011; Tian 2008). In
the auditory system, peripheral deafferentation or sound-induced trauma causes adaptive changes centrally, including
long-term increases in spontaneous activity and decreases in
inhibition (Wang et al. 2011). In the olfactory system, compensatory plasticity appears to be implemented at multiple loci,
although it has been most thoroughly studied in the OE and
bulb (reviewed by Coppola 2012). Centrally, 2 wk of unilateral
naris occlusion (UNO), the primary method for inducing odor
deprivation, leads to increases in the probability of glutamate
release and the amplitude of quantal synaptic currents at the
first- and second-order synapses of the ipsilateral olfactory
bulb of young rats (Tyler et al. 2007). Neonatal UNO causes a
loss (or failure to differentiate) of inhibitory short axon cells in
the external plexiform layer of young adult mice (Hamilton and
Coppola 2003). Both these forms of plasticity in the bulb
would lead to an increase in system gain consistent with a
compensatory process.
Peripherally, neonatal UNO in mice causes an ipsilateral
increase in olfactory-transduction-cascade transcripts (Coppola
and Waggener 2012; Zhao et al. 2013) and proteins (Coppola
et al. 2006; He et al. 2012; Waguespack et al. 2005) within a
few weeks. Presumably these biochemical changes underlie the
observed increase in sensitivity of ipsilateral olfactory sensory
neurons (He et al. 2012; Waggener and Coppola 2007) and
could partly explain—along with central compensatory changes—the above-normal olfactory behavioral capabilities of mice
reared under the UNO regimen (Angely and Coppola 2010).
Although compensatory plasticity in olfactory sensory neurons is by now well-established, fundamental questions concerning this phenomenon remain. All studies of the phenome-
0022-3077/15 Copyright © 2015 the American Physiological Society
2023
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 16, 2017
Barber CN, Coppola DM. Compensatory plasticity in the olfactory
epithelium: age, timing, and reversibility. J Neurophysiol 114: 2023–2032,
2015. First published August 12, 2015; doi:10.1152/jn.00076.2015.—
Like other biological systems, olfaction responds “homeostatically” to
enduring change in the stimulus environment. This adaptive mechanism,
referred to as compensatory plasticity, has been studied almost exclusively in developing animals. Thus it is unknown if this phenomenon is
limited to ontogenesis and irreversible, characteristics common to some
other forms of plasticity. Here we explore the effects of odor deprivation
on the adult mouse olfactory epithelium (OE) using nasal plugs to
eliminate nasal airflow unilaterally. Plugs were in place for 2– 6 wk after
which electroolfactograms (EOGs) were recorded from the occluded and
open sides of the nasal cavity. Mean EOG amplitudes were significantly
greater on the occluded than on the open side. The duration of plugging
did not affect the results, suggesting that maximal compensation occurs
within 2 wk or less. The magnitude of the EOG difference between the
open and occluded side in plugged mice was comparable to adults that
had undergone surgical naris occlusion as neonates. When plugs were
removed after 4 wk followed by 2 wk of recovery, mean EOG amplitudes
were not significantly different between the always-open and previously
plugged sides of the nasal cavity suggesting that this form of plasticity is
reversible. Taken together, these results suggest that compensatory plasticity is a constitutive mechanism of olfactory receptor neurons that
allows these cells to recalibrate their stimulus-response relationship to fit
the statistics of their current odor environment.
2024
COMPENSATORY PLASTICITY IN THE OLFACTORY EPITHELIUM
non, to date, have involved subjects deprived during development leaving unknown whether and to what extent
compensation occurs in adults. Only scant evidence is available on how long it takes for compensation to become manifest, and it is unknown whether the process is reversible or has
a sensitive period. Finally, previous studies of olfactory receptor transcript upregulation following UNO rearing suggest that
compensation may be spatially limited within the olfactory
epithelium (Zhao et al. 2013). Here we use removable nasal
plugs and EOG recordings in adult mice to address these
questions.
MATERIALS AND METHODS
OB
*
*
Np
Nt
Mt
2mm
*
ne
Zo
*
1
ne
2
Zo
3
ne
Zo
4
ne
Zo
Fig. 1. Drawing of midsagittal view of adult mouse head illustrating the medial
aspect of endoturbinates [Reproduced from Coppola et al. 2014 with permission. Copyright Company of Biologists Ltd]. Colored stripes illustrate approximate boundaries of olfactory receptor zones (Ressler et al. 1993; Vassar et al.
1993; see Coppola et al. 2013 for discussion). Electroolfactogram (EOG)
recordings were made at the position of the white asterisks on turbinate
IId. Recording position in Zone 1 (yellow) is referred to in the text as
caudal. Recording position in Zone 4 (green) is referred to in the text as rostral.
Color stripes are coextensive with olfactory epithelium (OE) in this region of
nasal cavity. Note that recordings were not made in Zone 2 or Zone 3 in this
study. d, dorsal; Mt, maxilloturbinate; Np, nasopharynx; Nt, nasoturbinate;
OB, olfactory bulb; v, ventral.
J Neurophysiol • doi:10.1152/jn.00076.2015 • www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 16, 2017
All animal care and experimental procedures followed the Guide to
Care and Use of Laboratory Animals (National Institutes of Health)
and were approved by the Randolph-Macon College Institutional
Animal Care and Use Committee.
Adult female (mean age 90 days) and neonatal male and female
CD-1 strain (Charles River Laboratories, Wilmington, MA) mice
were used in three separate experiments. Adult and weanlings were
housed four per cage (191 ⫻ 292 ⫻ 127 mm) with ad libitum access
to food and water in an approved animal care facility.
Naris plugs. Plugs were constructed following the procedures of
Cummings and colleagues (1997). Briefly, plugs were constructed of
4 mm lengths of polyethylene tubing (OD ⫽ 0.965 mm) occluded
with a knotted piece of 3.0 silk suture. A strand of human hair was tied
into the knotted suture to serve as an aid to plug localization and
extraction. Plugs, lubricated with ophthalmic ointment, were completely inserted into either the right or left external naris under brief
isoflurane anesthesia. The hair was then trimmed to protrude 2 mm
from the external naris. Left or right side plug insertion was counterbalanced as much as possible. Plugs were checked daily to confirm
they had not become dislodged. No perforations were detected during
periodic application of a soap solution to the plugged naris.
Surgical naris occlusion. Naris occlusion followed the procedures
of Coppola and colleagues (2006). Briefly, on the day after birth (PND 1)
male and female pups were anesthetized with isoflurane inhalation
followed by hypothermia. Either the right or left naris was occluded
using a surgical cautery to cut a slot across the external naris, the
margins of which were subsequently joined using cyanoacrylate
cement. Mice were checked daily to ensure the cauterized naris
remained occluded. Animals were excluded from the study if their
occluded naris became patent prior to use in the study.
EOG recordings. The procedures followed those of Waggener and
Coppola (2007). Briefly, mice were killed by pentobarbital overdose,
decapitated, and the skulls skinned. A disposable mictrotome blade
was used to cut the skull in half along the midsagittal plane. The nasal
septum and overlying tissues were resected to reveal the medial aspect
of the endoturbinates. Hemisections, thus prepared, remained in a
humidified chamber until recording commenced.
Recordings were conducted inside a Faraday cage covered with
plastic sheeting in a room kept below 20°C to preserve the viability of
the animal preparation. The output of an ultrasonic humidifier and a
forced air humidifier maintained humidity near 100% and maintained
positive pressure in the recording chamber.
Glass recording electrodes were pulled to a diameter of 20 – 40 ␮m
and filled with 0.5% agar dissolved in phosphate-buffered saline. The
indifferent electrode consisted of a plastic 500-␮l pipette tip filled
with 1% agar dissolved in phosphate-buffered saline. Wires made of
Ag/AgCl connected the electrodes to the leads of a dc amplifier (WPI
Iso-Dam 8A, Sarasota, FL). The amplifier’s low-pass filter was set at
10 Hz and its output was sent to a PowerLab/8SP (ADInstruments,
Colorado Springs, CO) for A/D conversion (20 kHz) and recording.
The maximum amplitude of traces was measured semi-manually with
the use of the PowerLab’s LabChart software.
The recording electrode and the output of the odor port were
positioned using three-axis micromanipulators and a stereomicroscope to aid visual guidance. The recording electrode was placed
sequentially in presumptive Zone 1 (caudal position) and presumptive
Zone 4 (rostral position) of dorsal turbinate II (IId; Fig. 1; Liebich
1975). The indifferent electrode was manually placed on the frontal
bone near the cribriform plate and held in place with a magnetic
clamp. The output of the odor port was positioned ⬃10 mm above the
placement of the recording electrode at a 35° angle to the epithelial
surface.
The order of recording, occluded or open nasal cavity, was counterbalanced for all experiments as was location of recording at either
Zone 1 or Zone 4. Once the electrode was positioned at a recording
site, responses to each stimulus were recorded twice with a 50- to 60-s
intertrial interval. Recording sessions were limited to less than 80 min,
consistent with the duration of viability of the preparation (Coppola et
al. 2013).
Stimulus delivery. The stimulus delivery system and justification
for the use of a single concentration were discussed previously
(Waggener and Coppola 2007). Briefly, isoamyl acetate (IAA), carvone (CAR), and eucalyptol (EUC), of the highest purity available
(Sigma-Aldrich, St. Louis, MO), were mixed (vol/vol) with mineral
oil to a concentration of 0.1%. These odorants were chosen because 1)
they have been used extensively in previous olfactory experiments
including our previous EOG studies (Coppola et al. 2013; Waggener
and Coppola 2007), 2) they are known to cause strong responses in the
area of the mucosa where we chose to record, and 3) they are
somewhat dissimilar chemically and thus may involve a nonoverlapping set of olfactory receptors. The order of odor stimulation was
counterbalanced across subjects to the extent possible.
COMPENSATORY PLASTICITY IN THE OLFACTORY EPITHELIUM
RESULTS
The EOG is known to be a highly variable measure of
olfactory receptor activation prompting various methods of
normalization to a standard odor (e.g., Scott et al. 1997). Since
in the mouse all of the odors that we have tested, to date, cause
heterogeneous responses across different locations, we have
chosen to use randomization and ample replication to uncover
treatment effects instead of standardization (Coppola et al.
2013). However, in an effort to better understand the variability
observed in the EOG and to validate our experimental design,
we recorded responses to repeated stimulations of the epithelial
preparation from normal mice with IAA over ⬃80 min with the
recording electrode placed in Zone 4 (Fig. 1). The results
establish that after an initial period of stability, during which
sequentially recorded traces are virtually identical in amplitude
and waveform, the amplitude of the EOG starts to decline
monotonically with repeated stimulations (Fig. 2). This suggests that the decline is more an effect of the previous stimulation history than the passing of time, a conclusion borne out
by the results of switching to the opposite nasal cavity after 60
min. In this case, one observes only a small depression in
response a full hour after the death of the subject. However,
after a brief period of stability, this side too begins to show
reductions in EOG amplitude with continued stimulation
(Fig. 2). Based on these results, the number of stimulations
at a given location in the OE was kept fewer than 10 for all
subsequent experiments.
Adult compensation. Response magnitudes varied consistently between the Zone 1 and Zone 4 recording locations and
among odor types (ANOVA main effect for location, P ⬍
0.0001; main effect for odor, P ⬍ 0.015; Fig. 3) consistent with
a recent response-mapping study of the mouse OE (Coppola et
al. 2013). However, after plugs were in place for 2 wk or
longer EOG magnitudes on the ipsilateral side were greater
than on the contralateral side for both locations and for all three
odors (ANOVA main effect for side, P ⬍ 0.0001; see Fig. 3 for
individual group comparisons). Comparison between the open
side and untreated controls confirmed that the difference between the open and plugged side of the OE was due to an
increase in EOG magnitude on the plugged side instead of a
decrease in magnitude on the open side (ANOVA open side vs.
untreated controls: main effect for side P ⫽ 0.24; Fig. 3). The
magnitude of the EOG enhancement, which averaged 35.2%
overall, was greater in the Zone 4 than Zone 1 (ANOVA
location by side interaction: P ⬍ 0.02) for all three odors
(Caudal: IAA ⫽ 33%, CAR ⫽ 32%, EUC 35%; Rostral: IAA
37%, CAR ⫽ 35%, EUC ⫽ 39%; Fig. 3).
Compensation timing. The results shown in Fig. 3 are pooled
from mice that had plugs in place for between 14 and 42 days.
Next we asked whether the magnitude of the compensation
observed on the plugged side changed with duration of occlusion (Fig. 4). The ratios of occluded side and open side EOG
magnitudes plotted across duration of plugging failed to show
a significant linear relationship for either the caudal recording
locations (IAA: R2 ⫽ 0.005, P ⫽ 0.67; CAR: R2 ⫽ 0.005, P ⫽
0.70; EUC: R2 ⫽ 0.006, P ⫽ 0.66) or the rostral recording
location (IAA: R2 ⫽ 0.04, P ⫽ 0.27; CAR: R2 ⫽ 0.004, P ⫽
0.72; EUC: R2 ⫽ 0.02, P ⫽ 0.39). To determine if there was
some nonlinear relationship with plug duration and the compensation effect, we then pooled the durations into weekly bins
(Fig. 4, inset) and analyzed bin differences using one-way
ANOVA for each odor and location separately. None of the six
ANOVAs (three odors by two locations) were significant with
alpha levels ranging from P ⬍ 0.58 to P ⬍ 0.96 (Fig. 4, inset).
Thus compensation was evident in most mice irrespective of
plug dwell time.
Adult vs. neonate compensation. Next we asked how the
compensation observed after occlusion in adulthood compared with the compensation in adults unilaterally naris
occluded as neonates. Mice that underwent surgical UNO on
the day after birth showed enhanced ipsilateral EOG magnitudes as adults, consistent with our previous study
(Waggener and Coppola 2007) (ANOVA main effect for
side, P ⬍ 0.0001; Fig. 5). Comparisons of the mean compensatory responses (plugged side/open side) between adult-
J Neurophysiol • doi:10.1152/jn.00076.2015 • www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 16, 2017
Ten milliliters of the odor liquid was placed in 25 ml glass reservoir
vials and a portion of the headspace of the vials was conveyed through
an odor port to the epithelial surface in a 0.5-s pulse of charcoal
filtered and humidified room air (700 ml/min). A custom-made olfactometer system controlled stimulus timing and duration (Knosys,
Lutz, FL). The odor delivery port was a 3-cm long ⫻ 3.5-mm
diameter glass tube connected by a 3-cm long Teflon tube to the odor
reservoir vial.
Stimulation of the mouse OE with the selected odor concentration
results in an EOG between the 2nd and 3rd quartile of the maximum
response magnitude for the three stimuli used in this study. The actual
concentration of gas-phase odor reaching the OE was unknown and
immaterial to the goals of the study.
Experimental design. In a preliminary experiment we investigated
potential sources of variability in the EOG to validate the experimental plan. The OE of two normal mice was repeatedly stimulated with
isoamyl acetate so that EOGs could be recorded from the Zone 4
position at intervals of ⬃50 s for 80 min. For one mouse, responses
were recorded from the right half of the nasal cavity for approximately
1 h followed by recordings from the left half. This experiment was
repeated in a second mouse with the order of recordings reversed.
For the remaining experiments, all three odorants were used at both
locations in each mouse. EOGs were recorded from both hemisections
of plugged and cauterized mice after a predetermined duration of naris
occlusion. Normal mice of similar ages to the plugged or cauterized
mice served as controls. For the plugged adult mice, the duration of
occlusion lasted between 14 and 42 days. In addition, recordings were
made from neonatally cauterized mice between 34 and 46 days of age
as in our previous study (Waggener and Coppola 2007). In a third
experiment, nasal plugs were removed from adult mice after 4 wk of
occlusion by gently tugging on the extruding hair to release the plug
from the naris. This group was then allowed to recover without plugs
for 2 wk after which EOGs were recorded.
Statistical analyses. The analysis strategy followed that of Coppola
and colleagues (2013) utilizing the Fit Model routines within JMP
statistical software (SAS Institute, Cary, NC). Briefly, mean EOG
magnitude was the dependent variable for all statistical analyses using
the average response to two applications of each stimulus. Withinsubject comparisons (open vs. plugged) were evaluated using a
three-way factorial ANOVA with location (Zone 1 or Zone 4), odor
(IAA, CUR, EUC) and side (open or plugged) as main effects and
subject as a random effect (Sall et al. 2007). A between-subject
ANOVA with the same factors in the model, as above, with the
random effect for subject omitted, was used to compare EOG responses from open and normal mice. Post hoc comparisons were made
using Bonferroni-corrected paired or unpaired Student t-tests or in the
case of ratios the Mann-Whitney test. The relationship between the
measured compensatory response and duration of plugging was evaluated with linear regression and one-way ANOVA for plug duration
binned by week.
2025
2026
COMPENSATORY PLASTICITY IN THE OLFACTORY EPITHELIUM
A
1 mV
1 min
1s
1st
Trace
10th
Trace
Right side
Left side
Fig. 2. A: raw EOG traces from one mouse stimulated repeatedly at approximately 1-min intervals with the headspace above 0.1% isoamyl acetate. Note that
through 8 traces the responses show very little variability in amplitude or waveform. Horizontal bars above the traces show timing and duration of stimulus. B:
plot of EOG amplitude vs. time-after-death data from one mouse showing the effect of repeated stimulations (0.1% isoamyl acetate). Note that while
time-after-death has some effect on amplitude (compare the EOG amplitudes from the left and right nasal vault) most of the variance in EOG amplitude is caused
by number of stimulations at a particular location. Dashed circles show relatively stable condition of early stimulations.
and neonatally-occluded mice failed to reveal significant
differences for the caudal recording location (IAA: U ⫽
123.5, P ⫽ 0.1; CAR: U ⫽ 175, P ⫽ 0.76; EUC: U ⫽ 172.5,
P ⫽ 0.70) or the rostral recording location (IAA: U ⫽ 185.5,
P ⫽ 0.65; CAR: U ⫽ 188, P ⫽ 0.70; EUC: U ⫽ 186, P ⫽
0.88). Likewise, comparisons of the distributions of compensation ratios by Kolmogorov-Smirnov test between
adult- and neonatally-occluded mice showed no difference
for the caudal recording location (IAA: D ⫽ 0.37, P ⫽ 0.16;
CAR: D ⫽ 0.24, P ⫽ 0.69; EUC: D ⫽ 0.24, P ⫽ 0.62) or
the rostral recording location (IAA: D ⫽ 0.19, P ⫽ 0.87;
CAR: D ⫽ 0.21, P ⫽ 0.78; EUC: D ⫽ 0.09, P ⫽ 0.99)
Compensation reversibility. In this experiment, adult mice
had plugs placed in one naris for 4 wk followed by plug
removal and recovery for 2 wk before electrophysiological
recordings were performed. EOG magnitudes on the reopened
side of the nasal cavity were not statistically different from
those on the always-open side (ANOVA main effect for side,
P ⫽ 0.68; see Fig. 6 for post hoc comparisons) confirming that
a reversal must have occurred in the compensatory response to
the previous plugging.
Unexpectedly, the unplugged control mice had statistically smaller average amplitudes than the always-opened
mice (ANOVA comparing open side plugged mice with
normal mice, P ⬍ 0.006). However, post hoc analysis
revealed that this was predominantly due to the IAA and
CAR responses in the caudal location (see Fig. 6 for post
hoc results).
J Neurophysiol • doi:10.1152/jn.00076.2015 • www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 16, 2017
B
COMPENSATORY PLASTICITY IN THE OLFACTORY EPITHELIUM
2027
DISCUSSION
A
Isoamyl Acetate
ctrl
open
plug
ctrl
open
plug
p < 0.006
p = 0.45
p < 0.003
p = 0.34
Carvone
ctrl
open
plug
open
ctrl
plug
p < 0.006
p = 0.34
p < 0.003
p = 0.09
C
Eucalyptol
ctrl
open
plug
ctrl
open
plug
p < 0.006
p = 0.25
p < 0.003
p = 0.09
Fig. 3. EOG responses are shown for isoamyl acetate (A), carvone (B), and
eucalyptol (C). Representative EOG traces are shown from adult mice that had
nasal plugs inserted in 1 naris for 2– 6 wk or were normal controls. The open
and plugged traces are from the same mouse for a given recording location and
odor. Column graphs show the mean ⫾ SE of EOG response magnitudes for
the different odors at the caudal location or the rostral location of turbinate IId.
The omnibus test of the open side vs. the plugged side was a within-subjects,
factorial ANOVA with main effect for location, F(1,363) ⫽ 173.2, P ⬍
0.0001; main effect for odor, F(2,363) ⫽ 4.22, P ⬍ 0.015; main effect for side,
F(1,363) ⫽ 59.9, P ⬍ 0.0001; only the location ⫻ side interaction was
significant, F(1,363) ⫽ 5.1, P ⬍ 0.02. The omnibus test of open side vs. the
control (ctrl) side was a between-subjects factorial ANOVA with main effect
for location, F(1,293) ⫽ 45.8, P ⬍ 0.0001; main effect for odor, F(2,293) ⫽
1.10, P ⫽ 0.012; main effect for side, F(1,293) ⫽ 1.41, P ⫽ 0.24; none of the
interactions were significant. For the post hoc test probabilities shown,
open vs. control means were compared by two-tailed unpaired t-test with
comparison-wise error rates. Open vs. plugged means were compared by
one-tailed paired t-test with Bonferroni correction (see MATERIALS AND
METHODS).
J Neurophysiol • doi:10.1152/jn.00076.2015 • www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 16, 2017
B
1 mV
1s
The EOG was used in this study to compare olfactory
receptor response magnitudes between the two sides of the
nasal cavity of adult mice in which nasal plugs had been
inserted unilaterally. A group of age-matched mice, that never
received nasal plugs, served as negative controls and another
group that was neonatally naris occluded by cautery served as
positive controls. While the EOG, an ensemble recording of the
summed generator potential of olfactory receptor neurons, is a
labor intensive and variable method of mapping responses over
large regions of the OE, there is currently no practical alternative (Scott and Scott-Johnson 2002). Our preliminary studies
demonstrated that most of the variability in the EOG arises
from rundown in the tissue after repeated stimulation rather
than from the passage of time, per se, in the ex vivo preparation
used in this and previous work (e.g., Scott et al. 1997). Our
experimental design exploited this situation by keeping the
number of stimulations of each area of the OE to a minimum.
And, as in previous EOG studies from our laboratory, replication and randomization were used to overcome the vagaries of
the method rather than standardizing to a particular odor, a
common practice in other laboratories (cf. Coppola et al. 2013;
Scott et al. 1997).
Our results confirm a marked increase in the odor response
of the OE on the ipsilateral side after nasal plugs were in place
for 2– 6 wk, compared with either the contralateral side or
aged-matched controls. The proportional increase in EOG
magnitudes were similar for the three odor stimuli, IAA, CAR,
and EUC and for the caudal (Zone 1) and rostral (Zone 4)
recording sites, ranging from 32–39%, although the Zone 4 had
a statistically larger proportional response. These results are
consistent with our previous EOG study of mice that had been
neonatally naris occluded by the permanent cautery method
and subsequently tested as adults (Waggener and Coppola
2007). In this previous study, using the same concentration and
battery of odors as in the current experiments, we found a 27%
to 43% increase, depending on the stimulus, in EOG magnitudes on the occluded side compared with the open side. In the
current study, since there were few significant differences in
average magnitude of the EOG responses between the open
side OE and those from control mice, the difference between
the plugged and open sides is attributed to an increase in the
former’s response rather than a decrease in the latter’s re-
2028
COMPENSATORY PLASTICITY IN THE OLFACTORY EPITHELIUM
A
B
10
IAA
CAR
EUC
3.0
2.5
2.0
IAA
1.5
CAR
1.0
EUC
2
3
4
5
Weeks
15
20
25
30
35
3.0
2.5
2.0
1.5
1.0
2
40
45
3
4
5
Weeks
0.1
0.1
10
1
Occluded EOG Mag / Open EOG Mag
Occluded EOG Mag / Open EOG Mag (Log10)
Occluded EOG Mag / Open EOG Mag
1
10
15
Days Plugged
20
25
30
35
40
45
Days Plugged
Fig. 4. Ratios of plugged-side to open-side EOG amplitude vs. the duration of plugging are plotted. Note that the greater the value the more the plugged side
response exceeded the open side response. Different odors are represented by symbol shape. A and B show responses at the caudal and rostral recording sites,
respectively. Trend lines are excluded for simplicity but there were no statistically significant linear trends with duration for any odor or either location (see text
for regression statistics). Inset graphs show same plug duration data binned by weeks. None of the one-way ANOVA results were significant. Caudal location:
isoamyl acetate (IAA), F(3,30) ⫽ 0.41, P ⫽ 0.74; carvone (CAR), F(3,30) ⫽ 0.17, P ⫽ 0.92; eucalyptol (EUC), F(3,30) ⫽ 0.10, P ⫽ 0.96. Rostral location:
IAA, F(3,29) ⫽ 0.67, P ⫽ 0.58; CAR, F(3,29) ⫽ 0.35, P ⫽ 0.79; EUC, F(3,29) ⫽ 0.42, P ⫽ 0.74 (Kruskal-Wallis tests produced same nonsignificant results
for all odor-location combinations).
sponse. However, to confirm that the compensatory process on
the occluded side of the nasal cavity observed in the adult
plugged-animal was similar to what we had previously reported in adults that had been neonatally naris-occluded, we
replicated our earlier result in a small group of mice. The
results confirmed that there was no difference in the means or
distribution of compensatory responses comparing plugged
with neonatally cauterized adults.
We next asked if the magnitude of the compensatory response changed with the duration of plugging. Over the 2- to
6-wk period of plugging used in this study we did not find a
consistent change in EOG responses with time for any of the
stimuli or either recording location. Thus it appears that the
compensatory response is fully manifest throughout the OE (or
at least across a large region of it) within 2 wk, although even
longer durations of plugging than attempted here would be
needed to establish this definitively.
We also showed that in addition to being inducible within 2
wk, the compensatory response is reversible. Mice that wore
nasal plugs for 4 wk and then had them removed for 2 wk had
no difference in EOG magnitudes between the always-open
and formerly-plugged OEs. This reversibility was manifest for
all three stimuli tested and for both the caudal and rostral
recording locations. However, there was one unexpected result
from the reversibility study: the control mice had statistically
smaller mean EOGs than those from the open nasal cavity of
plugged mice. Indeed, while not statistically significant, the
same pattern appeared in the first experiment. Post hoc analyses of the reversibility data revealed that this difference was
only pronounced at the caudal recording location for IAA and
CAR. And this difference was not manifest in our previous
study of neonatally naris occluded mice—indeed controls
tended to have larger EOGs than those from the open nasal
cavity of occluded mice (Waggener and Coppola 2007). If the
J Neurophysiol • doi:10.1152/jn.00076.2015 • www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 16, 2017
Occluded EOG Mag / Open EOG Mag (Log10)
10
COMPENSATORY PLASTICITY IN THE OLFACTORY EPITHELIUM
A
Isoamyl Acetate
Rostral
Caudal
Neonate
Adult
Neonate
Occlusion Age
B
Carvone
Caudal
Adult
Neonate
Rostral
Adult
Neonate
Occlusion Age
C
Eucalyptol
Rostral
Caudal
Adult
Neonate
Adult
Occlusion Age
Neonate
current result is valid, it might suggest that the open side OE of
unilaterally occluded mice may not be “normal,” a point we
have argued previously (Coppola 2012).
The first evidence, albeit indirect, of compensatory plasticity
in the OE emerged from the observation that olfactory marker
protein (OMP) was upregulated on the occluded side of UNO
mice and downregulated on the open-side (Tian and Ma 2008;
Waguespack et al. 2005). Since the occluded side is deprived
of sensory stimuli and the open-side is hyperstimulated—
having to carry twice its normal volume of air—the OMP
response suggested a compensatory process. However, OMP’s
upregulation with deprivation may not be related to a compensation process given what has subsequently been learned about
this protein’s function in olfactory receptors neurons (Kass et
al. 2013; Lee et al. 2011; Reisert and Margolis 2008). More
direct proof of compensation emerged from observations that
key elements of the olfactory transductory cascade in olfactory
receptor neurons, including adenylyl cyclase and the cyclicnucleotide-gated cation channel, show the same pattern of upand downregulation as OMP following UNO (Coppola et al.
2006; He et al. 2012). These observations were confirmed and
extended by studies of gene transcripts. Coppola and
Waggener (2012), using whole-genome microarrays, confirmed that RNA transcripts coding for olfactory transduction
proteins tended to be upregulated in the OE from the occluded
side of the nasal cavity and downregulated on the open side.
This study was the first to show that most olfactory receptor
RNAs are upregulated on the occluded side of the OE consistent with a compensatory response, a result recently confirmed
and extended with in situ hybridization (Zhao et al. 2013).
However, in this latter study all of the sampled OR genes from
the lateral-ventral part of the nasal cavity, corresponding to our
Zone 4 recording site, showed upregulation on the occluded
side while OR genes from other zones showed inconsistent
patterns of differential regulation. These results suggested the
possibility that olfactory receptor neurons from different parts
of the nasal cavity vary in their susceptibility to modulation by
sensory experience. To address this possibility we recorded
from two distant locations on the OE, one in a caudal, presumed Zone 1 location and one in a rostral, presumed Zone 4
location (Ressler et al. 1993; Vassar et al. 1993). While there
were differences in the average response magnitude at these
locations to each odor, in agreement with our recent mapping
study of odor responses in the mouse OE (Coppola et al. 2013),
the proportional increases in EOG magnitudes from the
plugged side compared with the open side were similar beFig. 5. EOG responses are shown for isoamyl acetate (A), carvone (B),
eucalyptol (C). Solid lines are representative traces from the occluded-side;
dashed lines are the open-side traces from the same subject. The column graphs
compare ratios of occluded-side to open-side EOG amplitude for plugged adult
mice (circles) and neonatally naris occluded mice examined as adults
(squares). Individual values and the means (⫾ SEs) are plotted on a log10 scale
for clearer visualization of the individual values. The data from plugged adults
are the same as in Fig. 3. For the neonatally naris occluded mice, the omnibus
test of the open side vs. the plugged side was a within-subjects, factorial
ANOVA with main effect for location, F(1,118) ⫽ 55.5, P ⬍ 0.0001; main
effect for odor, F(2,118) ⫽ 18.1, P ⬍ 0.003; main effect for side, F(1,118) ⫽
35.8, P ⬍ 0.0001. None of the interaction effects were significant. The ratios
of occluded-side to open-side EOG amplitude for plugged adult mice and
neonatally naris occluded mice were compared for differences in median and
distribution by the Mann-Whitney and Kruskal-Wallis tests, respectively, but
none were found to be significant (see text).
J Neurophysiol • doi:10.1152/jn.00076.2015 • www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 16, 2017
Adult
1 mV
1s
2029
2030
COMPENSATORY PLASTICITY IN THE OLFACTORY EPITHELIUM
A
Isoamyl Acetate
ctrl
open
plug
ctrl
open
plug
1 mV
1s
p = 0.12 p = 0.39
p = 0.36
p < 0.002
Carvone
ctrl
open
plug
open
ctrl
plug
p = 0.43 p = 0.25
p = 0.39
p < 0.006
C
Eucalyptol
ctrl
open
plug
ctrl
open
plug
p = 0.73 p = 0.29
p = 0.08 p = 0.36
Fig. 6. Results are shown for isoamyl acetate (A), carvone (B), and eucalyptol
(C). Representative EOG traces are from adult mice that had plugs inserted in
one naris for 4 wk followed by 2 wk of recovery after plug removal or were
normal controls. The open and plugged traces are from the same mouse for a
given recording location and odor. Column graphs show the means ⫾ SE of
EOG response magnitudes for the different odors at the caudal recording
location or the rostral recording location on turbinate IId. The omnibus test of
the open side vs. the plugged side comparison was a within-subjects, factorial
ANOVA with main effect for location, F(1,132) ⫽ 45.1, P ⬍ 0.0001; main
effect for odor, F(2,132) ⫽ 20.1, P ⬍ 0.0001; main effect for side was not
significant, F(1,132) ⫽ 0.16, P ⫽ 0.69; none of the interaction effects were
significant. The omnibus test of open side vs. the control (ctrl) side was a
between-subjects factorial ANOVA with main effect for location, F(1,168) ⫽
44.1, P ⬍ 0.0001; main effect for odor, F(2,168) ⫽ 8.7, P ⫽ 0.0003; main
effect for side, F(1,168) ⫽ 12.1, P ⬍ 0.0006; none of the interactions were
significant. The post hoc test probabilities shown above are for comparisons of
the open vs. control means by two-tailed unpaired t-test with comparison-wise
error rates. The open vs. plugged means were compared by one-tailed paired
t-test with comparison-wise error rates (see MATERIALS AND METHODS).
J Neurophysiol • doi:10.1152/jn.00076.2015 • www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 16, 2017
B
tween zones, although statistically greater rostrally. Thus we
conclude that the compensatory response is ubiquitous in the
OE or, at least, not specific to one zone.
Studies of critical period plasticity in the visual (e.g., Hubel
and Wiesel 1963), auditory (e.g., Zhang et al. 2002) and
somatosensory (e.g., Woolsey and Wann 1976) cortices form a
central pillar in the canon of developmental neurobiology
(Hensch 2005). Likewise in the olfactory system, there is
evidence of a critical period in early postnatal life for synaptic
plasticity at sensory synapses of olfactory cortex (Poo and
Isaacson 2007). Analogously, many of the effects of UNO on
the olfactory bulb and OE were originally thought to be
restricted to a sensitive period in the weeks after birth (reviewed by Coppola 2012). However, subsequent work has
established, at least for many experimental endpoints, that
similar effects of UNO occur in adults (Coppola 2012). Consistent with these findings, the current study is the first to
demonstrate that compensatory plasticity at the receptor level
is not limited to a critical period. The increase in EOG
magnitude observed in nasal plugged adults was similar to that
seen in adult mice that had undergone neonatal UNO, an effect
that was completely reversible, in the former group, within 2
wk. Thus, the yet unknown mechanism underlying this phenomenon seems to be a constitutive homeostatic-like pathway
of the normal olfactory receptor neuron that appears to be
“adaptive” in the evolutionary sense.
What could be the biochemical pathway of this compensatory process? The elements and interactions that constitute the
olfactory transduction pathway are reasonably well understood. Olfactory receptor proteins are linked to ion channels by
a classical G protein-coupled process involving the enzyme
adenylyl cyclase and its product cAMP (for reviews see, e.g.,
Firestein 2001; Kleene 2008). Amplification occurs both at the
step between binding of odor to receptor and cyclic nucleotide
gated (CNG) channel opening and between this event and
the opening of Ca2⫹-activated Cl⫺ channels (Firestein 2001).
The previously reported upregulation of several components of
this positive-feedback cascade in ipsilateral olfactory receptor
neurons of UNO mice could explain our earlier and current
EOG results (assuming that these transductory cascade modifications also occur in adults). However, there is relatively little
previous work on which to base speculations about the pathway that might link reduced activity in the olfactory receptor
COMPENSATORY PLASTICITY IN THE OLFACTORY EPITHELIUM
ACKNOWLEDGMENTS
We thank D. Cummings for graciously providing step-by-step instructions
and actual examples of nasal plugs at different stages of construction.
GRANTS
This work was supported by grants from the National Science Foundation
(IOS-0641433), Chenery Endowment, and Rashkind Endowment to D. M.
Coppola.
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the author(s).
AUTHOR CONTRIBUTIONS
Author contributions: C.N.B. performed experiments; C.N.B. and D.M.C.
interpreted results of experiments; C.N.B. and D.M.C. drafted manuscript;
C.N.B. and D.M.C. edited and revised manuscript; C.N.B. and D.M.C. approved final version of manuscript; D.M.C. conception and design of research;
D.M.C. analyzed data; D.M.C. prepared figures.
REFERENCES
Angely C, Coppola DM. How does long-term odor deprivation affect the
olfactory capacity of adult mice? Behav Brain Func 6: i–ij, 2010.
Cannon WB. Organization for physiological homeostasis. Physiol Rev 9:
399 – 431, 1929.
Coppola DM. Studies of olfactory system neural plasticity: the contribution of
the unilateral naris occlusion technique. Neural Plast 2012: 351752, 2012.
Coppola DM, Craven BA, Seeger J, Weiler E. The effects of naris occlusion
on mouse nasal turbinate development. J Exp Biol 217: 2044 –2052, 2014.
Coppola DM, Waggener CT. The effects of unilateral naris occlusion on gene
expression profiles in mouse olfactory mucosa. J Mol Neurosci 47: 604 –
618, 2012.
Coppola DM, Waggener CT, Radwani SM, Brooks DA. An electroolfactogram study of odor response patterns from the mouse olfactory epithelium
with reference to receptor zones and odor sorptiveness. J Neurophysiol 109:
2179 –2191, 2013.
Coppola DM, Waguespack A, Reems M, Butman ML, Cherry J. Naris
occlusion alters transductory protein immunoreactivity in olfactory epithelium. Histol Histopath 21: 487–501, 2006.
Cummings DM, Henning HE, Brunjes PC. Olfactory bulb recovery after
early sensory deprivation. J Neurosci 17: 7433–7440, 1997.
Cygnar KD, Collins SE, Ferguson CH, Bodkin-Clarke C, Zhao H. Phosphorylation of adenylyl cyclase III at serine1076 does not attenuate olfactory
response in mice. J Neurosci 32: 14557–14562, 2012.
Davis G. Homeostatic control of neural activity: from phenomenology to
molecular design. Annu Rev Neurosci 29: 307–323, 2006.
Fain GL, Mathews HR, Cornwall MM, Koutalos Y. Adaptation in vertebrate photoreceptors. Physiol Rev 81: 117–151, 2001.
Firestein S. How the olfactory system makes sense of scents. Nature 413:
211–218, 2001.
Gold JR, Banjo VM. Insult-induced adaptive plasticity of the auditory system.
Front Neurosci 8: 1–31, 2014.
Hamilton KA, Coppola DM. Distribution of GluR1 is altered in the olfactory
bulb following neonatal naris occlusion. J Neurobiol 54: 326 –336, 2003.
He J, Tian H, Lee AC, Ma M. Postnatal experience modulates functional
properties of mouse olfactory sensory neurons. Eur J Neurosci 36: 2452–
2460, 2012.
He Q, Wang P, Tian N. Light-evoked synaptic activity of retinal ganglion and
amacrine cells is regulated in developing mouse retina. Eur J Neurosci 33:
36 – 48, 2011.
Hensch TK. Critical period plasticity in local cortical circuits. Nat Rev
Neurosci 6: 877– 888, 2005.
Hubel DH, Wiesel TN. Receptive fields of cells in striate cortex of very
young, visually inexperienced kittens. J Neurophysiol 26: 994 –1002, 1963.
Kass MD, Moberly AH, Rosenthal MC, Guang SA, McGann JP. Odorspecific, olfactory marker protein-mediated sparsening of primary olfactory
input to the brain after odor exposure. J Neurosci 33: 6594 – 6602, 2013.
Kleene SJ. The electrochemical basis of olfactory transduction in vertebrate
olfactory cilia. Chem Senses 33: 839 – 859, 2008.
Lee AC, He J, Ma M. Olfactory marker protein is critical for functional
maturation of olfactory sensory neurons and development of mother preferences. J Neurosci 31: 2974 –2982, 2011.
Liebich HG. Zum Bau der oberen Luftwege der weissen Ratte (Mus rattus
norvegicus, var. albinos) [Structure of the upper airways of the white rat
(Mus rattus norvegicus, var. albinos)]. Anat Anz 138: 170 –179, 1975.
J Neurophysiol • doi:10.1152/jn.00076.2015 • www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 16, 2017
neuron to subsequent upregulation of the transductory cascade.
While a thorough treatment of olfactory sensory neuron signal
adaptation and termination—a short-term form of homeostasis—is beyond the scope of the current discussion, research on
this process suggests a candidate “integrator” of neuronal
activity that might be a component of the olfactory compensatory plasticity mechanism.
Calcium/calmodulin-dependent protein kinase II (CaMKII)
is a serine/threonine protein kinase found throughout the brain
but most intensively studied in hippocampal synapses where it
plays a critical role inducing long-term potentiation (review by
Lisman et al. 2014). Interestingly, this Ca2⫹ activated molecule
is concentrated in olfactory sensory neurons and has been
implicated in olfactory adaptation (Wei et al. 1998; Zufall and
Leinder-Zufall 2000; cf. Cygnar et al. 2012). Calmodulin,
CaMKII’s Ca2⫹ sensor, activates the holoenzyme in the presence
of Ca2⫹, which subsequently undergoes autophosphorylation to
become Ca2⫹ independent (Lisman et al. 2014). In the hippocampus the amount of Ca2⫹-autonomous CaMKII is governed by the
frequency, duration, and amplitude of Ca2⫹ transients (Lisman et
al. 2014; Robison 2014). This Ca2⫹ -transient-integrating characteristic of CaMKII makes it ideally suited as a central component
in long-term potentiation, thought to be the cellular basis of some
forms of memory. We suggest that this same characteristic would
make it well suited as the integrator of activity in olfactory sensory
neurons such that it could negatively feedback on transcriptional
and/or translational controls governing the synthesis of olfactory
transduction-cascade components. Thus olfactory receptor neurons in “noisy” odor environments would have components of the
olfactory cascade downregulated by CaMKII negative feedback
while receptors in odor-sparse environments (such as under deprivation) would have this negative feedback removed by the
relative scarcity of autonomous CaMKII. Admittedly, this model
is incomplete as stated and untested; however, its falsification
would be straightforward experimentally through the measurement and manipulation of CaMKII in olfactory receptor neurons
under different stimulation regimes.
The sequelae of UNO—a technique for inducing sensory
deprivation used for more than 100 years—are legion including
effects on the OE, olfactory bulb and olfactory cortex (reviewed by Coppola 2012). Indeed, it seems doubtful that all the
effects of this form of deprivation can be subsumed under one
theoretical construct (e.g., homeostatic plasticity, see Coppola
2012). In other sensory systems, stimulus deprivation or superabundance can lead to both adaptive and maladaptive forms
of plasticity (Gold and Banjo 2014). There is no reason to
suspect that olfaction is any different. Nevertheless, compensatory plasticity would seem to be a highly adaptive mechanism affording olfactory receptor neurons the ability to recalibrate their stimulus-response relationship to fit the statistics of
their current odor regime, which is likely to be temporally and
spatially variable in nature. Elucidating the activity-integrating
and negative feedback mechanisms that must underlie this
process will require a substantial amount of additional effort.
2031
2032
COMPENSATORY PLASTICITY IN THE OLFACTORY EPITHELIUM
Vassar R, Ngai J, Axel R. Spatial segregation of odorant receptor expression
in the mammalian olfactory epithelium. Cell 74: 309 –318, 1993.
Waggener C, Coppola DM. Naris occlusion alters the electro-olfactogram:
Evidence for compensatory plasticity in the olfactory system. Neurosci Let
427: 112–116, 2007.
Waguespack A, Reems M, Butman ML, Cherry J, Coppola DM. Olfactory
receptor neurons have enhanced olfactory marker protein immunoreactivity
in the nasal cavity ipsilateral to naris occlusion. Brain Res 1044: 1–7, 2005.
Wang H, Brozoski TJ, Caspary DM. Inhibitory neurotransmission in animal
models of tinnitus: Maladaptive plasticity. Hearing Res 279: 111–117, 2011.
Wark B, Lundstrom BN, Fairhall A. Sensory adaptation. Curr Opin Neurobiol 17: 423– 429, 2007.
Wei J, Zhao AZ, Chan GC, Baker LP, Impey S, Beavo JA, Storm DR.
Phosphorylation and inhibition of olfactory adenylyl cyclase by CaM kinase
II in neurons: a mechanism for attenuation of olfactory signals. Neuron 21:
495–504, 1998.
Woolsey TA, Wann JR. Areal changes in mouse cortical barrels following
vibrissal damage at different postnatal ages. J Comp Neurol 170: 53– 66,
1976.
Zhang LI, Bao S, Merzenich MM. Disruption of primary auditory cortex by
synchronous auditory inputs during a critical period. Proc Natl Acad Sci
USA 99: 2309 –2314, 2002.
Zhao S, Tian H, Ma L, Yuan Y, Yu CR, Ma M. Activity-dependent
modulation of odorant receptor gene expression in the mouse olfactory
epithelium. PLoS One 8: e69862, 2013.
Zufall F, Leinder-Zufall T. The cellular and molecular basis of odor adaptation. Chem Senses 25: 473– 481, 2000.
J Neurophysiol • doi:10.1152/jn.00076.2015 • www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 16, 2017
Lisman J, Yasuda R, Raghavachari S. Mechanisms of CaMKII action in
long-term potentiation. Nat Rev Neurosci 13: 169 –182, 2014.
Poo C, Isaacson JS. An early critical period for long-term plasticity and
structural modification of sensory synapses in olfactory cortex. J Neurosci
27: 7553–7558, 2007.
Rauschecker JP. Compensatory plasticity and sensory substitution in the
cerebral cortex. Trends Neurosci 18: 36 – 43, 1995.
Reisert J, Margolis FL. Olfactory marker protein: a gift to molecular
biologists, an enigma to physiologists. Physiol News 73: 27–29, 2008.
Ressler KJ, Sullivan SL, Buck LB. A zonal organization of odorant receptor
gene expression in the olfactory epithelium. Cell 73: 597– 609, 1993.
Robison AJ. Emerging role of CaMKII in neuropsychiatric disease. Trends
Neurosci 37: 653– 662, 2014.
Sall J, Creighton L, Lehman A. JMP Start Statistics: A Guide to Statistical
Data Analysis Using JMP (4th ed.). Cary, NC: SAS Institute, 2007.
Scott JW, Scott-Johnson PE. The electroolfactogram: a review of its history
and uses. Microsc Res Tech 58: 152–160, 2002.
Scott JW, Shannon DE, Charpentier J, Davis LM, Kaplan C. Spatially
organized response zones in rat olfactory epithelium. J Neurophysiol 77:
1950 –1962, 1997.
Tian H, Ma M. Activity plays a role in eliminating olfactory sensory neurons
expressing multiple odorant receptors in the mouse septal organ. Mol Cell
Neurosci 38: 484 – 488, 2008.
Tian N. Synaptic activity, visual experience and the maturation of retinal
synaptic circuitry. J Physiol 586: 4347– 4355, 2008.
Tyler WJ, Petzold GC, Pal SK, Murthy VN. Experience-dependent modification of primary sensory synapses in the mammalian olfactory bulb. J
Neurosci 27: 9427–9438, 2007.