nature NEUROSCIENCE  VOLUME 17 | NUMBER 1 | JANUARY 2014	 89
a r t ic l e s
Responding to environmental stimuli is crucial to the survival of organisms
and often manifests as alterations in the structure and function of
the nervous system. When and how information from the environment
results in experience-dependent alteration of nervous system structure
and function are fundamental questions in behavioral neuroscience.
An important, but often ignored, factor that influences adult nervous
systems is exposure of parents to salient environmental stimuli
before the conception of their offspring. Such information transfer
would be an efficient way for parents to ‘inform’ their offspring about
the importance of specific environmental features that they are likely
to encounter in their future environments. However, this would necessitate
the transgenerational inheritance of environmental information
via the germ line by offspring not even conceived at the time. Although
our understanding of such non-Mendelian modes of inheritance is
continually being revised in terms of the epigenetic inheritance of
traits1, empirical data to support transgenerational epigenetic inheritance
of behavioral traits in mammals are beginning to accumulate at
the level of morphological, behavioral and metabolic traits2–15.
We used olfactory fear conditioning to address when and how
the olfactory experience of a parent might influence their offspring.
Specifically, we focused on the olfactory system, given its wellunderstood
molecular biology and neuroanatomy16–18, the ability
to differentially target odorant-receptor pairs in the same modality
for differential and well-controlled behavioral studies, and previous
findings that experience-dependent alterations occur in olfactory
neuroanatomy and behavior following olfactory conditioning19.
We examined how specific features of the parental sensory environment
before conception can influence sensory nervous system
structure and function in a cue-specific manner in subsequently
conceived F1 and F2 generations. Bisulfite sequencing of olfactory
receptor genes in the sperm of the F0 and F1 generations revealed
differences in methylation that may mark the specific olfactory receptor
gene for enhanced transcription in the subsequent generation.
Finally, using in vitro fertilization (IVF), F2 and cross-fostering studies,
we found that the behavior and structural alterations were inherited
and were not socially transmitted from the F0 generation.
RESULTS
Olfactory fear conditioning to study descendant generations
We examined whether olfactory fear conditioning of the F0 generation
leads subsequently conceived adult F1, F2 and IVF-derived
generations to exhibit F0-like behavioral sensitivity toward the F0
conditioned odor, and whether there were neuroanatomical changes
at the level of the main olfactory epithelium (MOE) and olfactory
bulb in these generations (Supplementary Fig. 1). The odors that
we used were chosen on the basis of prior work demonstrating that
the M71 odorant receptors (encoded by the Olfr151 gene) expressed
by olfactory sensory neurons (OSNs) in the MOE are activated by
acetophenone20. The use of a chemical mixture that contained compounds
very similar to propanol did not elicit any responses from
M71 cells, suggesting that propanol does not activate M71 receptors.
In addition, glomerular activity patterns elicited by acetophenone or
propanol (http://gara.bio.uci.edu) are different and non-overlapping,
suggesting that a different population of OSNs primarily responds
to each odor.
In this procedure, 2-month-old sexually inexperienced and odor
naive C57Bl/6J male mice or homozygous M71-LacZ transgenic male
mice were left in their home cage (F0-Home) or conditioned with
acetophenone (F0-Ace) or propanol (F0-Prop). Subsequently conceived
adult male offspring (F1) belonged to three groups: F1-Home,
F1-Ace and F1-Prop (Online Methods). It is important to note that
no F0 males were excluded from the study after training and that
1Department of Psychiatry and Behavioral Sciences, Emory University School of Medicine, Atlanta, Georgia, USA. 2Yerkes National Primate Research Center,
Atlanta, Georgia, USA. 3Howard Hughes Medical Institute, Chevy Chase, Maryland, USA. Correspondence should be addressed to B.G.D. (bdias@emory.edu) or
K.J.R. (kressle@emory.edu).
Received 21 September; accepted 1 November; published online 1 December 2013; corrected online 9 December 2013; doi:10.1038/nn.3594
Parental olfactory experience influences behavior and
neural structure in subsequent generations
Brian G Dias1,2 & Kerry J Ressler1–3
Using olfactory molecular specificity, we examined the inheritance of parental traumatic exposure, a phenomenon that has been
frequently observed, but not understood. We subjected F0 mice to odor fear conditioning before conception and found that
subsequently conceived F1 and F2 generations had an increased behavioral sensitivity to the F0-conditioned odor, but not to
other odors. When an odor (acetophenone) that activates a known odorant receptor (Olfr151) was used to condition F0 mice,
the behavioral sensitivity of the F1 and F2 generations to acetophenone was complemented by an enhanced neuroanatomical
representation of the Olfr151 pathway. Bisulfite sequencing of sperm DNA from conditioned F0 males and F1 naive offspring
revealed CpG hypomethylation in the Olfr151 gene. In addition, in vitro fertilization, F2 inheritance and cross-fostering revealed
that these transgenerational effects are inherited via parental gametes. Our findings provide a framework for addressing how
environmental information may be inherited transgenerationally at behavioral, neuroanatomical and epigenetic levels.
npg©2014NatureAmerica,Inc.Allrightsreserved.
90	 VOLUME 17 | NUMBER 1 | JANUARY 2014  nature NEUROSCIENCE
a r t ic l e s
all of them were mated with naive females. Thus, any findings that
we obtained were not the results of extreme phenotype biasing or a
previously existing genetic sensitivity. Both C57Bl/6J and M71-LacZ
mice possess the M71 odorant receptor in their olfactory epithelium21
and both can consequently detect acetophenone. The main difference
between the strains is that the OSNs of the M71-LacZ mice produce
β-galactosidase in M71-expressing neurons and can therefore be
visualized22. This procedure allowed us to examine a seldom studied
factor that might markedly influence the nervous systems of adults;
namely, the experience of the F0 generation before conception.
Transgenerational olfactory sensitivity after F0 conditioning
Fear-potentiated startle (FPS) is a behavioral test to assay for fear
learning23. FPS manifests as an augmented startle response in the
presence of the aversive conditioned cue. In our case, to assay for
behavioral sensitivity to an odor, we used a modified FPS protocol
that consists of odor presentation before the startle stimuli. An odorpotentiated
startle (OPS) score is computed, in which an enhanced
OPS reflects a greater startle to the odor relative to control, when the
odor is paired with the startle stimulus. Traditionally, FPS tests have
been used to query the emotional state of the animal and the valence
of the stimulus paired with the startle. It is important to note that we
did not use this test as a measure of valence of the odor, but rather
as a readout of the sensitivity toward that odor, similar to FPS tests
that have been used to test the sensitivity of mice to natural odors
such as fox urine24. Enhanced OPS to acetophenone in our experiment
would be interpreted as an enhanced behavioral sensitivity to
acetophenone, not necessarily an increase in fear to acetophenone.
Making any statements about valence specificity and the emotional
value of the odor would necessitate subjecting the F0 generation to
an appetitive odor-conditioning task.
In the F0 generation, we previously reported that olfactory fear
conditioning adult males to acetophenone increases FPS when the
startle stimuli are paired with acetophenone presentation19. In the
F1 generation, we found that C57Bl/6J F1-Ace mice (F1-Ace-C57)
showed enhanced OPS (unconditioned) to acetophenone compared
with C57Bl/6J F1-Home mice (F1-Home-C57) (Fig. 1a). No differences
between groups were found when propanol was paired with
the startle, indicating that the response was specific to acetophenone
(Fig. 1b). Similarly, F1-Ace-M71 showed enhanced OPS to acetophenone,
but not to propanol, compared with F1-Home-M71 and
F1-Prop-M71 (Fig. 1c,d). In contrast, F1-Prop-M71 showed enhanced
OPS to propanol, but not to acetophenone (Fig. 1c,d). These data
suggest a double dissociation and specificity of the odor association,
along with the inheritance of a behavioral sensitivity that is specific
to the F0-conditioned odor.
To further corroborate the enhanced behavioral sensitivity to the
F0-conditioned odor, we conducted an independent behavioral assay
that directly probes behavioral sensitivity using an odor concentration
curve and the association time of the mice with these concentrations.
We found that F1-Ace males were able to detect acetophenone at lower
concentrations than F1-Prop males, whereas F1-Prop males detected
propanol at lower concentrations than F1-Ace males (Fig. 2a,b), further
suggesting an enhanced detection sensitivity that is specific to
the F0-conditioned odor. Although we make a case for both the OPS
and association time assays testing for behavioral sensitivity, we used
OPS in our subsequent experiments because of our ability to carefully
calibrate odor presentation and removal, parameters that might influence
habituation to odors and skew experimental results.
Most noteworthy for these data is the fact that the naive F1 mice
had never been exposed to any of the odors with which they were
tested. Taken together, these data indicate that the behavioral sensitivity
to an odor in adult offspring is specific to the odor that the
F0 male was conditioned to, as shown across two different odorants
and two different strains of mice. Furthermore, given the fact that
300a *
200
100
PercentageOPSto
acetophenone
0
–100
F1-Home-C57 F1-Ace-C57
*
300b
200
100
PercentageOPSto
propanol
0
–100
F1-Home-C57 F1-Ace-C57
600d
400
200
PercentageOPSto
propanol
0
–200
F1-H
om
e-M
71
F1-Ace-M
71
F1-Prop-M
71
***
600c
400
200
PercentageOPSto
acetophenone
0
–200
F1-H
om
e-M
71
F1-Ace-M
71
F1-Prop-M
71
Figure 1  Behavioral sensitivity to odor is specific to the paternally
conditioned odor. (a,b) Responses of individual C57Bl/6J F1 male
offspring conceived after the F0 male was fear conditioned with
acetophenone. F1-Ace-C57 mice had an enhanced sensitivity to
acetophenone (a), but not to propanol (control odor, b) compared with
F1-Home-C57 mice (F1-Ace-C57, n = 16; F1-Home-C57, n = 13; t test,
P = 0.043, t27 = 2.123). (c,d) Responses of M71-LacZ F1 male offspring
conceived after the F0 male was fear conditioned with acetophenone or
propanol. F1-Ace-M71 mice had an enhanced sensitivity to acetophenone (c),
but not to propanol (d), compared with F1-Home-M71, and F1-Prop-M71
mice. In contrast, F1-Prop-M71 mice had an enhanced sensitivity to
propanol (d), but not acetophenone (c) (F1-Home-M71, n = 11;
F1-Ace-M71, n = 13; F1-Prop-M71, n = 9; OPS to acetophenone:
ANOVA, P = 0.003, F2,30 = 6.874; F1-Home-M71 versus F1-Ace-M71,
P < 0.05; F1-Ace-M71 versus F1-Prop-M71, P < 0.01; OPS to propanol:
ANOVA, P = 0.020, F2,26 = 4.541; F1-Ace-M71 versus F1-Prop-M71,
P < 0.05). Data are presented as mean ± s.e.m. *P < 0.05, **P < 0.01.
5
a
0
–5
Aversionindex
–10
–15
Acetophenone concentration
0.01%
0.03%
0.06%
**
10
b
0
5
–5
**
Aversionindex
–10
–20
–15
Propanol concentration
0.001%
0.003%
0.006%
F1-Ace-C57 F1-Prop-C57 F1-Ace-C57 F1-Prop-C57
Figure 2  Sensitivity of F1 males toward F0-conditioned odor. Association
time with either the concentration of odor on the x axis or an empty
chamber was recorded. An aversion index was computed by subtracting
the amount of time spent in the open chamber from the time spent in the
odor chamber. (a) When tested with acetophenone, F1-Ace mice detected
acetophenone at a lower concentration (0.03%) than F1-Prop mice, with
both groups eventually showing equal aversion at the 0.06% concentration
(P = 0.005 with Bonferroni correction for multiple comparisons).
(b) When tested with propanol, F1-Prop mice detected propanol at a lower
concentration (0.003%) than F1-Ace mice, with both groups eventually
showing equal aversion at the 0.006% concentration (P = 0.0005 with
Bonferroni correction for multiple comparisons) (F1-Ace-C57, n = 16;
F1-Prop-C57, n = 16). Data are presented as mean ± s.e.m. (**P < 0.01).
npg©2014NatureAmerica,Inc.Allrightsreserved.
nature NEUROSCIENCE  VOLUME 17 | NUMBER 1 | JANUARY 2014	 91
a r t ic l e s
both F0-Ace and F0-Prop mice received shocks during conditioning,
these data suggest that these training-specific effects do not occur
simply as a result of paternal history of the stress of shock exposure
or conditioning to odors in general.
Studies that have examined the effect of parental stress after conception,
either in utero or postnatally, have often found an anxiogenic
phenotype in the offspring25,26. Using an elevated plus maze to assay
for anxiety-like behavior, we found that prior, rather limited foot
shock conditioning of the F0 generation, did not extend to generalized
anxiety-like behavior in the F1 generation (Supplementary Fig. 2a,b).
To test the idea that olfactory fear conditioning of the F0 generation
results in offspring that might be generally deficient in processing
sensory cues and in learning and memory processes, we sought
to examine whether auditory fear conditioning was affected in our
experimental groups. Across all experimental groups, adult male F1
offspring subjected to auditory fear conditioning acquired, consolidated
and extinguished fear similarly (Supplementary Fig. 3a–c).
F0 olfactory experience affects F1 neuroanatomy
Previously19, we reported that the behavioral response (increased
FPS to acetophenone) of F0-Ace conditioned males is complemented
by an increase in the number of acetophenone–responsive
M71-expressing OSNs in the MOE and M71 glomerular area in
the olfactory bulbs. To examine whether alterations in the neuroanatomical
representation of the conditioned odor accompanied
the behavioral sensitivity reported above, we used standard
β-galactosidase staining in naive M71-LacZ F1 males that had neither
been behaviorally tested with, nor exposed to, any of the conditioned
odors. We found that the dorsal and medial M71-specific
glomeruli in the olfactory bulb of F1 offspring of acetophenonetrained
F0 males (F1-Ace-M71) were significantly increased in size
compared with those of the offspring of home cage or propanoltrained
F0 males (F1-Home-M71 and F1-Prop-M71, respectively)
(ANOVA, P < 0.0001 for dorsal and medial glomeruli; Fig. 3a–h).
This increase in M71 glomerular area was accompanied by a significant
increase in the numbers of M71 OSNs in the MOE (ANOVA,
P < 0.0001; Fig. 3i).
These data suggest that the effect of paternal olfactory fear conditioning
on neuroanatomy is associated with increased numbers
of OSNs and increased glomerular area, both specific for the F0­conditioned
odor. We posit that this increased structural representation
in the main olfactory epithelium and olfactory bulb may underlie
the specific enhanced olfactory sensitivity that we observed in the
behavioral experiments (Figs. 1 and 2). We were concerned that performing
behavior would make the offspring no longer odor naive,
and thereby potentially confound the interpretation of the neuroanatomical
results. Thus, all of the neuroanatomy data were generated
using animal cohorts independent of any behavior data. Correlations
between behavior and neuroanatomy within and between generations
present an interesting and important future direction for research.
Inheritance of effects in the F2 and IVF-derived generations
Two mechanisms could explain how information about the F0conditioned
odor could be transferred to the subsequently conceived
male offspring: inheritance via the gametes or transmission via a
social route that is reminiscent of the transmission of maternal care in
rodents27. To begin to dissociate these two possibilities, we conducted
experiments with the F2 generation and with IVF-derived mice. Naive
F1 males (F1-Ace, F1-Prop) were mated with naive females to generate
F2 adults (F2-Ace, F2-Prop) whose F0 ancestors had been conditioned
with either acetophenone or propanol. For the IVF experiment,
sperm from F0 males was collected 10 d after the last conditioning day,
and IVF was performed by the Transgenic Mouse Facility at Emory
University in a location independent of our laboratory at Yerkes where
we conducted all of the other studies reported. Subsequently conceived
IVF offspring (F1-Ace-IVF and F1-Prop-IVF) were raised to
adulthood and tissue was collected in this facility.
When tested in our behavioral assay, F2-Ace-C57 mice exposed to
odors for the first time showed increased OPS to acetophenone compared
with F2-Prop-C57 mice, whereas F2-Prop-C57 mice showed
an increased OPS to propanol (Fig. 4a,b). This persistent behavioral
sensitivity to the F0-conditioned odor was accompanied by corresponding
increases in glomerular size in an independent set of F2
M71-LacZ mice that had no previous exposure to the odors used.
a b c
d e f
Dorsal bulb
Medial bulb
F1-Home F1-Prop F1-Ace
g 6,000
4,000
2,000
0
Dorsalglomerulusarea
(pixels)
F1-H
om
e-M
71
F1-Ace-M
71
F1-Prop-M
71
*** *
h
6,000
8,000
4,000
2,000
0
Medialglomerulusarea
(pixels)
F1-H
om
e-M
71
F1-Ace-M
71
F1-Prop-M
71
**** ****
i
F1-H
om
e-M
71
F1-Ace-M
71
F1-Prop-M
71
*** **350
300
250
200
100
NumberofM71OSNs
0
Figure 3  Neuroanatomical characteristics of the olfactory system in
F1 males after paternal F0 olfactory fear conditioning. (a–f) β-galactosidase
staining revealed that offspring of F0 males trained to acetophenone (F1-Ace-M71) had larger dorsal
and medial acetophenone-responding glomeruli (M71 glomeruli) in the olfactory bulb compared with
F1-Prop-M71 and F1-Home-M71 mice. Scale bar represents 1 mm. (g) Dorsal M71 glomerular area in F1
generation (M71-LacZ: F1-Home, n = 38; F1-Ace, n = 38; F1-Prop, n = 18; ANOVA, P < 0.0001, F2,91 =
15.53; F1-Home-M71 versus F1-Ace-M71, P < 0.0001; F1-Ace-M71 versus F1-Prop-M71, P < 0.05).
(h) Medial M71 glomerular area in F1 generation (M71-LacZ: F1-Home, n = 31; F1-Ace, n = 40; F1-Prop,
n = 16; ANOVA, P < 0.0001, F2,84 = 31.68; F1-Home-M71 versus F1-Ace-M71, P < 0.0001; F1-Ace-M71
versus F1-Prop-M71, P < 0.0001). (i) F1-Ace-M71 mice had a larger number of M71 OSNs in the MOE than
F1-Prop-M71 and F1-Home-M71 mice (M71-LacZ: F1-Home, n = 6; F1-Ace, n = 6; F1-Prop, n = 4; ANOVA,
P = 0.0001, F2,13 = 18.80; F1-Home-M71 versus F1-Ace-M71, P < 0.001; F1-Ace-M71 versus F1-Prop-M71,
P < 0.01). Data are presented as mean ± s.e.m. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.
npg©2014NatureAmerica,Inc.Allrightsreserved.
92	 VOLUME 17 | NUMBER 1 | JANUARY 2014  nature NEUROSCIENCE
a r t ic l e s
The dorsal and medial M71-specific glomeruli in the olfactory bulbs
of F2-Ace-M71 mice were significantly increased in size compared
with those of F2-Prop-M71 mice (Fig. 4c–h). Similar results were
obtained in our IVF study, using sperm from F0-Ace and F0-Prop
males to generate offspring. We found that F1 offspring generated
with sperm from acetophenone-trained F0 males (F1-Ace-IVF) had
significantly larger dorsal and medial M71-specific glomeruli in the
olfactory bulb, as compared with offspring generated with sperm from
propanol-trained F0 males (F1-Prop-IVF) (t test, P < 0.001 for dorsal
and medial glomeruli; Fig. 4i,j). We could not perform behavioral
analyses on IVF-generated offspring because of animal quarantine
issues. These data indicate that behavioral sensitivity and neuroanatomical
alterations in the nervous system are specific to the F0conditioned
odor and persist until at least the F2 generation, as well
as in the IVF-derived F1 generation, thereby pointing to an inheritance
of these effects.
Cross-fostering supports inheritance of information
Our observations of the behavioral and structural changes specific
to the F0-conditioned odor being retained in the F2 generation, and
the persistence of the structural effects after IVF, argue against social
transmission and make a strong case for transgenerational inheritance.
Notably, our results are highly specific in the olfactory sensory
modality toward the F0-conditioned odor, and both F0-Ace and F0Prop
males were subjected to the same shock training conditions that
might be deemed stressful and potentially conveyed to the mother.
This argues against the idea that our results might merely be the transmission
of a stressful paternal experience to the mother during the
time of co-habitation. To ensure that our experimental groups were
balanced for both odor and shock exposure, many of our experiments
utilized F0-Prop as the control group rather than F0-Home.
To further address this issue, and to address potential maternal
transmission, we conducted a cross-fostering study. Sexually naive
female mice were conditioned with acetophenone or left in their home
cage. They were then mated with odor-naive males for 10 d, after
which the male was removed. Subsequent offspring were then divided
into the following groups: offspring of home cage mothers (F1-Home),
offspring of acetophenone-conditioned mothers (F1-Ace), offspring
of home cage mothers cross-fostered starting at postnatal day 1 by
mothers conditioned to acetophenone (F1-Home(fostered)), and offspring
of acetophenone conditioned mothers cross-fostered by home
cage mothers (F1-Ace(fostered)) (Supplementary Fig. 4). Notably,
the females were only exposed to the conditioning odor before mating,
and never while pregnant, precluding the possibility that offspring
were directly exposed to any odor-related fear and in utero
learning. We conducted this cross-fostering study in females for two
main purposes. First, we sought to examine whether these effects were
specific to paternal conditioning or could also be inherited via the
female germ line. Second, given the possibility that mating with the
F0 conditioned male in some way altered maternal behavior toward
subsequently born offspring, we wanted to account for any differences
in maternal investment or information transfer about the conditioned
odor that might result from our conditioning protocol.
We found that, similar to the situation in which the F0 male (father)
was conditioned to acetophenone, F1-Ace mice in this maternally
trained experiment had an enhanced OPS to acetophenone compared
with F1-Home controls (Fig. 5a). If our behavioral findings were a
result of a ‘social transmission’ mode of information transfer, we
would have predicted a reversal of the above result. Instead, we found
that the F1-Ace-C57(fostered) male offspring still had a higher OPS
to acetophenone than F1-Home-C57(fostered) offspring (Fig. 5b),
suggesting a biological, rather than social, mode of inheritance.
For the equivalent experiment to visualize neuroanatomy, we performed
a similar cross-fostering experiment using M71-LacZ females,
and used female mice conditioned to propanol as our control group
(offspring labeled as F1-Prop). We found that the increased dorsal
and medial glomerular area persisted in F1-Ace mice even after they
were cross-fostered by mothers conditioned to propanol (F1-AceM71(fostered)).
In contrast, F1-Prop mice cross-fostered by mothers
conditionedtoacetophenone(F1-Prop-M71(fostered))didnotshowany
increases in M71 glomerular area (Fig. 5c–h). In summary, these crossfostering
results, taken together with our IVF and F2 ­studies, strongly
c
e f
dMedial
Dorsal
Propanol
trained F0
Acetophenone
trained F0
300
a
200
100
PercentageOPSto
acetophenone
–100
F2-Prop-C57
*
F2-Ace-C57
0
b
150
100
50
–50
PercentageOPSto
propanol
–100
F2-Prop-C57
*
F2-Ace-C57
0
4,000
3,000
Dorsalglomerulus
area
1,000
F2-Prop-M71
****
F2-Ace-M71
2,000
h 5,000
4,000
3,000
Medialglomerulus
area
1,000
g
F2-Prop-M71
***
F2-Ace-M71
2,000
j
4,000
3,000
Medialglomerulus
area
1,000
F1-Prop-IVF
****
F1-Ace-IVF
2,000
i
3,000
2,500
Dorsalglomerulusarea
1,000
F1-Prop-IVF
***
F1-Ace-IVF
2,000
1,500
Figure 4  Behavioral sensitivity and neuroanatomical changes are inherited
in F2 and IVF-derived generations. (a,b) Responses of F2-C57Bl/6J
males revealed that F2-Ace-C57 mice had an enhanced sensitivity to
acetophenone compared with F2-Prop-C57 mice (a). In contrast,
F2-Prop-C57 mice had an enhanced sensitivity to propanol compared
with F2-Ace-C57 mice (b; F2-Prop-C57, n = 8; F2-Ace-C57, n = 12; OPS
to acetophenone: t test, P = 0.0158, t18 = 2.664; OPS to propanol: t test,
P = 0.0343, t17 = 2.302). (c–f). F2-Ace-M71 mice whose F0 generation
male had been conditioned to acetophenone had larger dorsal and medial
M71 glomeruli in the olfactory bulb than F2-Prop-M71 mice whose
F0 generation had been conditioned to propanol. Scale bar represents
200 µm. (g) Dorsal M71 glomerular area in F2 generation (M71-LacZ:
F2-Prop, n = 7; F2-Ace, n = 8; t test, P < 0.0001, t13 = 5.926).
(h) Medial M71 glomerular area in F2 generation (M71-LacZ: F2-Prop,
n = 6; F2-Ace, n = 10; t test, P = 0.0006, t14 = 4.44). (i) Dorsal M71
glomerular area in IVF offspring (F1-Prop-IVF, n = 23; F1-Ace-IVF, n = 16;
t test, P < 0.001, t37 = 4.083). (j) Medial M71 glomerular area in
IVF offspring (F1-Prop-IVF, n = 16; F1-Ace-IVF, n = 19; t test, P < 0.001,
t33 = 5.880). Data are presented as mean ± s.e.m.*P < 0.05,
**P < 0.01, ***P < 0.001, ****P < 0.0001.
npg©2014NatureAmerica,Inc.Allrightsreserved.
nature NEUROSCIENCE  VOLUME 17 | NUMBER 1 | JANUARY 2014	 93
a r t ic l e s
suggest that our behavioral and structural data
are a consequence of biological inheritance.
Altered epigenetic signature at Olfr151 (M71) locus in sperm
Given that our data suggests a biological inheritance of our behavioral
and structural effects via parental gametes, we sought to examine
sperm of the F0 generation males for epigenetic clues that might
explain an enhanced representation for the M71 receptor (Fig. 6).
CpG methylation is one mechanism by which a particular genetic
locus can be marked for altered transcription, with less CpG
di-nucleotide methylation typically being associated with more
transcription. Bisulfite sequencing around the Olfr151 (M71) locus
and the non-acetophenone–responsive Olfr6 locus (Supplementary
Fig. 5) was conducted by Active Motif on DNA extracted from
sperm of F0-Prop and F0-Ace mice. Olfr6 converges at a glomerular
space that is distinct from glomerular activity patterns elicited by
acetophenone or propanol (http://gara.bio.uci.edu) and we therefore
used it as a control odorant receptor for bisulfite sequencing studies.
We found that the Olfr151 (P = 0.0323; Fig. 6a), but not Olfr6
(P = 0.54; Fig. 6c), locus was significantly less methylated in sperm
from F0-Ace males compared with F0-Prop males. In addition, after
correcting for multiple comparisons, one particular CpG di-nucleotide
at the 3′ end of Olfr151 was significantly less methylated in F0-Ace
males than in F0-Prop males (P = 0.003; Fig. 6b,d).
These findings led us to hypothesize that relative hypomethylation
of Olfr151 in F0 sperm may lead to inheritance of the hypomethylated
Olfr151 in F1 MOE and F1 sperm, creating an inheritance cascade.
A related idea would be that, during the stochastic odorant receptor
choice process in the MOE18,28, Olfr151 (M71) would be more likely to
be expressed in the next generation as a consequence of the epigenetic
signature around that locus in the sperm. When bisulfite-converted
DNA from sperm of the F1 generation was sequenced, we found that,
similar to the F0 scenario, the Olfr151 locus was hypomethylated in
F1-Ace sperm compared with F1-Prop controls (Fig. 6e). In addition,
after correcting for multiple comparisons, two particular CpG
di-nucleotides in Olfr151 were significantly less methylated in F1-Ace
sperm compared with F1-Prop sperm (P = 0.002; Fig. 6f). These data
suggest that inheritance of an epigenetic signature around a salient
genetic locus accompanies our transgenerational effects. At the level
of the MOE, we did not find any differences in the methylation at
the Olfr151 locus of either the F1 or F2 generations (Supplementary
Fig. 6). This is perhaps unsurprising given that other modes of epigenetic
modifications have been implicated in the marking of olfactory
receptor loci in the MOE29, and mandates future investigation. For
example, DNA methylation and histone modifications are known to
be dependent on each other30, and changes in the methylation pattern
in Olfr151 in sperm DNA that we observe may potentially result in
histone modifications around Olfr151 in MOE DNA.
Published data also support the idea of epigenetic marking in sperm
by indicating that sperm-associated histones are retained with chromatin
of the paternal genome at the one-cell embryo stage31,32. To
investigate the possibility that histone modifications mark the Olfr151
(M71) locus, we collected sperm from F0-Ace and F0-Prop males
10 d after the last day of conditioning and performed native-chromatin
immunoprecipitation (N-ChIP) on the sperm chromatin. Briefly,
chromatin was extracted from sperm and immunoprecipitated
with antibodies that recognize histone modifications, after which
a b200
* *
*
DorsalglomerulusareaMedialglomerulusarea
*
**
**
**
***
***
F1-Home-C57
F1-H
om
e-C
57
F1-H
om
e-C
57
(fostered)
F1-Ace-C57
Dorsal
Medial
Propanol
cross-fostered
Acetophenone
cross-fostered
F1-Ace-C
57
F1-Ace-C
57
(fostered)
F1-Prop-M
71
F1-Prop-M
71
(fostered)
F1-Ace-M
71
F1-Ace-M
71
(fostered)
F1-Prop-M
71
F1-Prop-M
71
(fostered)
F1-Ace-M
71
F1-Ace-M
71
(fostered)
PercentageOPS
toacetophenone
PercentageOPS
toacetophenone
100
c d
e f
g
4,000
h
100
0
–100
50
0
–50
–100
3,000
2,000
1,000
0
4,000
3,000
2,000
1,000
0
Figure 5  Behavioral sensitivity and
neuroanatomical changes persist after crossfostering.
(a) F1 offspring of mothers that
had been fear conditioned with acetophenone
(F1-Ace-C57) showed enhanced sensitivity to
acetophenone compared with F1-Home-C57
controls (F1-Home-C57, n = 13; F1-Ace-C57,
n = 16; t test, P = 0.0256, t27 = 2.362).
(b) Cross-fostering behavior. F1-Ace-C57
males had higher OPS to acetophenone
than F1-Home-C57 males (P < 0.01). F1-AceC57(fostered)
males still had higher OPS to
acetophenone than F1-Home-C57(fostered)
males (P < 0.05) (ANOVA, P = 0.0011, F3,18 =
6.874, planned post hoc comparisons).
(c–f) Cross-fostering neuroanatomy. F1-AceM71
males cross-fostered by mothers conditioned
to propanol (F1-Ace-M71(fostered)) continued
to have larger M71 glomeruli than F1-Prop-M71
males cross-fostered by mothers conditioned
to acetophenone (F1-Prop-M71(fostered)).
Scale bar represents 100 µm. (g) Dorsal M71
glomerular area in F1 cross-fostered generation
(M71-LacZ: F1-Prop, n = 6; F1-Ace, n = 4; F1Prop(fostered),
n = 5; F1-Ace(fostered), n = 3;
ANOVA, P < 0.0001, F3,14 = 17.52; F1-Prop
versus F1-Ace, P < 0.001; F1-Prop(fostered)
versus F1-Ace(fostered), P < 0.01). (h) Medial
M71 glomerular area in F1 cross-fostered
generation (M71-LacZ: F1-Prop, n = 4; F1-Ace, n
= 3; F1-Prop(fostered), n = 8; F1-Ace(fostered),
n = 4; ANOVA, P < 0.01, F3,15 = 5.933; F1-Prop
(fostered) versus F1-Ace(fostered), P < 0.01).
Data are presented as mean ± s.e.m. *P < 0.05,
**P < 0.01, ***P < 0.001.
npg©2014NatureAmerica,Inc.Allrightsreserved.
94	 VOLUME 17 | NUMBER 1 | JANUARY 2014  nature NEUROSCIENCE
a r t ic l e s
­quantitative PCR was performed for the Olfr151 gene. We did not
observe any differences in histone-mediated epigenetic signatures
around the M71 locus when chromatin was immunoprecipitated
with antibodies that recognize histone modifications that either permit
(acetylated H3) or repress (H3trimethyl K27) to transcription
(Supplementary Fig. 7). The fact that the M71 locus was not epigenetically
marked via histones in the F0 sperm could indicate that we
did not immunoprecipitate with the relevant histone-modification
antibody or that the epigenetic basis of this inheritance might not be
histone based, instead relying on other mechanisms, such as DNA
methylation (as reported above) or non-coding RNA, as has been
demonstrated for the Kit locus33.
DISCUSSION
Focusing on classical conditioning in an F0 generation before conception
and using specific odors as the conditioned stimuli allowed us to
tag a specific olfactory experience and follow the salience of that experience
at the level of behavior and neuroanatomy through subsequent
generations. We found that the F1 and F2 generations were extremely
sensitive to the specific odors used to condition F0 mice. Using a transgenic
mouse in which OSNs expressing a specific odorant receptor can
be visualized, we found that the behavioral sensitivity was accompanied
by an altered olfactory neuroanatomy for the conditioned odor.
The fact that these changes persisted after IVF, cross-fostering and
across two generations is indicative of biological inheritance. Finally,
we observed that the sperm of the F0 and F1 generation males bear
epigenetic marks that could be the basis for such inheritance.
There have been other studies that examined the transmission of
stimulus-specific behavioral and structural adaptations in the nervous
system from parents to their offspring, albeit with substantial
differences from our experimental design. For example, in utero taste
aversion learning affects the offspring’s preference and avoidance of
flavors and odors in the mother’s diet during gestation34. In addition,
quality of maternal care is transmitted across generations in
rodents27. Furthermore, fetal origins of diseases have been proposed
for a multitude of disorders as having their roots in the experience
of the fetus to the parental environment while in utero35. From a
chemosensory perspective, anti-predatory behavior is transmitted
from gravid female crickets to their offspring when the females are
exposed to a high density of a predator36. Finally, indirectly related to
our study is a report that supplementing the mouse maternal diet with
acetophenone at various stages of gestation increases M71 glomerular
area and preference for acetophenone in adolescent offspring37. This
last study exemplifies how the olfactory sensitivity and neuroanatomy
of offspring can bear imprints of parental experience.
However, it is imperative to realize that all of the aforementioned
manipulations of the parental condition have occurred when the pups
or embryos are in utero, thereby assaying behavior and neuroanatomy
in animals that are extremely different from those conceived after
perturbation to the parent. In other words, the fetuses in the cited
studies were directly exposed to the environmental perturbation. This
important point about perturbation of the parental (F0) environment
affecting the F1 embryo directly, as well as the F2 germ line, has been
used to argue that true transgenerational inheritance should manifest
itself in the F3 generation38. It is important to note that the F2 mice
that we tested are a full and complete generation removed from the
environmental perturbation of their parent; as such, our observations
suggest a transgenerational phenomenon. Our IVF data complement
this point further.
Most recently, several studies have factored paternal effects and
transgenerational inheritance of behavior and metabolic states into
their experimental design. First, paternal diet has been shown to have
marked effects on the metabolic physiology of offspring conceived
after the father’s diet had been manipulated7. Second, exposure to the
anti-androgenic endocrine disruptor vinclozolin during embryonic
gonadal sex determination affects fertility and behavior in at least four
subsequent generations, and it is associated with epigenetic changes
in the sperm of descendant male offspring2,9,39. A recent study used
a social defeat procedure in mice and found paternal transmission of
depressive-like behavior in subsequently conceived adult offspring.
100
a b
80
60
F0-Prop-
Sperm
F0-Ace-
Sperm
Olfr151 (over all CpG sites) Olfr151 (individual CpG sites)
*
*
Percentagemethylation
100
80
60
C
pG
1C
pG
2C
pG
3C
pG
4C
pG
5C
pG
6C
pG
7C
pG
8C
pG
9
Percentagemethylation
F0-Prop-Sperm
F0-Ace-Sperm
Olfr6 (over all CpG sites) Olfr6 (individual CpG sites)
F0-Ace-
Sperm
dc
100
80
60
C
pG
1C
pG
2C
pG
3C
pG
4C
pG
5C
pG
6C
pG
7C
pG
8C
pG
9
Percentagemethylation
100
80
60
Percentagemethylation
F0-Prop-Sperm
F0-Ace-Sperm
F0-Prop-
Sperm
F1-Prop-
Sperm
F1-Ace-
Sperm
fe Olfr151 (individual CpG sites)
*
*
100
80
60
C
pG
1C
pG
2C
pG
3C
pG
4C
pG
5C
pG
6C
pG
7C
pG
8
Percentagemethylation
100
80
60
Olfr151 (over all CpG sites)
*Percentagemethylation
F1-Prop-Sperm
F1-Ace-Sperm
Figure 6  Methylation of odorant receptor genes in sperm DNA from
conditioned F0 and odor naive F1 males. (a) Bisulfite sequencing of
CpG di-nucleotides in the Olfr151 (M71) gene in F0 sperm revealed that
F0-Ace mouse DNA (n = 12) was hypomethylated compared with
that of F0-Prop mice (n = 10) (t test, P = 0.0323, t16 = 2.344).
(b) A particular CpG di-nucleotide in the Olfr151 (M71) gene in
F0 sperm was hypomethylated in F0-Ace mice (n = 12) compared with
F0-Prop mice (n = 10) (P = 0.003, Bonferroni corrected). (c) We found
no differences in methylation between F0-Ace (n = 12) and F0-Prop
(n = 10) mice across all of the CpG di-nucleotides queried in the Olfr6
gene in F0 sperm (P > 0.05). (d) Across specific CpG di-nucleotides in
the Olfr6 gene, we found no differences in methylation between F0-Ace
(n = 12) and F0-Prop (n = 10) mice (Bonferroni corrected). (e) Bisulfite
sequencing of the Olfr151 (M71) gene in F1 sperm revealed that F1-Ace
mouse DNA (n = 4) was hypomethylated compared with that of F1-Prop
mice (n = 4) (t test, P = 0.0153, t14 = 2.763). (f) Bisulfite sequencing
of CpG di-nucleotides in the Olfr151 (M71) gene in F1 sperm revealed
that two particular CpG di-nucleotides in the Olfr151 (M71) gene were
hypomethylated in F1-Ace mice (n = 4) compared with F1-Prop mice
(n = 4) (P = 0.002, Bonferroni corrected). Data are presented as
mean ± s.e.m. *P < 0.05 after correction.
npg©2014NatureAmerica,Inc.Allrightsreserved.
nature NEUROSCIENCE  VOLUME 17 | NUMBER 1 | JANUARY 2014	 95
a r t ic l e s
These authors found an (epi)genetic inheritance of depression-like
behavior in the forced swim test using IVF with sperm from socially
defeated fathers, indicating that behavior in offspring can be affected
by paternal experience even if the offspring have not been conceived
at the time of paternal trauma5. There was also a recent report of
epigenetic inheritance of a cocaine-resistance phenotype in rats sired
by F0 males that self-administered cocaine40. Finally, behavioral and
epigenetic changes have been shown in generations of normally raised
offspring whose parents had been subjected to maternal separation
procedures4,41. These data, including ours, emphasize that transgenerational
epigenetic inheritance does occur in mammals, supporting
findings of such inheritance in organisms ranging from flies
to worms42–44.
How olfactory stimulation in the F0 generation comes to be linked
to sperm is an intriguing question for which we can only offer speculation.
Evidence exists for blood-borne odorants activating odorant
receptors in the nose45. Thus, it is also conceivable that the odorants
used in the F0 fear conditioning protocol enter the circulatory stream
and activate odorant receptors that are expressed on sperm46. With
mouse spermatogenesis occurring over, on average, 26 d (ref. 47),
we reasoned that the interval of 13 d between the first conditioning
day and breeding would be enough time for any mature sperm
to be cleared from the system and for any information to be inherited
by sperm precursors that were 13 d into the maturation process.
However, at this point, we cannot and do not claim to know what
fraction of sperm precursors, and consequently mature sperm, transferred
to the female carry the pertinent information. Future studies
would be well served by examining such information storage across
spermatogenesis using irradiation-based approaches.
Behavioral sensitivity to odors might be linked to the olfactory
topography in the MOE and bulb. For example, animals that have 95%
of their OSN population dominated by the M71 receptor show deficits
in odor detection48 as well as increased anxiety49. We hypothesize that
a substantial increase in the number of M71 neurons in the MOE (but
not nearly to the extent of the ‘monoclonal nose’ animals referred to
above), and the subsequent enlarged M71-specific glomeruli, are a
direct structural mechanism for the enhanced olfactory sensitivity phenotype.
We chose to query the overall unconditioned sensitivity of the
F1 and F2 generations to the F0 conditioned odor. It would be equally
interesting to examine how the F1 and F2 generations respond to direct
conditioning with the F0 conditioned odor. However, we felt that this
nuance would be better addressed after the parameters and mechanisms
underlying any unconditioned responses were appreciated.
In summary, we have begun to explore an under-appreciated influence
on adult behavior—ancestral experience before conception.
From a translational perspective, our results allow us to appreciate
how the experiences of a parent, before even conceiving offspring,
markedly influence both structure and function in the nervous system
of subsequent generations. Such a phenomenon may contribute
to the etiology and potential intergenerational transmission of risk
for neuropsychiatric disorders, such as phobias, anxiety and posttraumatic
stress disorder50. To conclude, we interpret these results
as highlighting how generations can inherit information about the
salience of specific stimuli in ancestral environments so that their
behavior and neuroanatomy are altered to allow for appropriate
stimulus-specific responses.
Methods
Methods and any associated references are available in the online
version of the paper.
Note: Any Supplementary Information and Source Data files are available in the
online version of the paper.
Acknowledgments
We would like to thank the animal care staff in the Yerkes Neuroscience Vivarium
for assistance with animal husbandry. A. Magklara, S. Lomvardas, B. Carone,
O. Rando and A.F.H.M. Peters provided invaluable input on the ChIP experiments.
We would like to thank H. Zhang and the staff of the Emory Transgenic Mouse/
Gene Targeting Core Facility for assistance with IVF studies. Bisulfite conversion of
sperm DNA and sequencing was carried out by Active Motif and we especially thank
P. Labhart for addressing our data interpretation queries. Finally, we are grateful to
S. Banerjee, R. Andero-Gali, D. Choi, J. Goodman and F. Morrison for help with
ensuring double-blindness of data acquisition and analysis, and members of the
Ressler laboratory, S. Gourley and M. Davis for helpful feedback on the manuscript.
Funding for this study was provided by the Howard Hughes Medical Institute and
the Burroughs Wellcome Fund to K.J.R., and a US National Institutes of Health
NCRR base grant (P51RR00-0165) to Yerkes National Primate Research Center.
AUTHOR CONTRIBUTIONS
B.G.D. conceived of the project, designed and performed experiments, analyzed
the data, and wrote the paper. K.J.R. obtained funds, designed experiments,
analyzed the data, wrote the paper and supervised the project.
COMPETING FINANCIAL INTERESTS
The authors declare no competing financial interests.
Reprints and permissions information is available online at http://www.nature.com/
reprints/index.html.
1.	 Jirtle, R.L. & Skinner, M.K. Environmental epigenomics and disease susceptibility.
Nat. Rev. Genet. 8, 253–262 (2007).
2.	 Anway, M.D. Epigenetic transgenerational actions of endocrine disruptors and male
fertility. Science 308, 1466–1469 (2005).
3.	 Weaver, I.C.G. Epigenetic programming by maternal behavior and pharmacological
intervention. Nature versus nurture: let’s call the whole thing off. Epigenetics 2,
22–28 (2007).
4.	 Franklin, T.B. et al. Epigenetic transmission of the impact of early stress across
generations. Biol. Psychiatry 68, 408–415 (2010).
5.	 Dietz, D.M. et al. Paternal transmission of stress-induced pathologies. Biol.
Psychiatry 70, 408–414 (2011).
6.	 Morgan, H.D., Sutherland, H.G., Martin, D.I. & Whitelaw, E. Epigenetic inheritance
at the agouti locus in the mouse. Nat. Genet. 23, 314–318 (1999).
7.	 Carone, B.R. et al. Paternally induced transgenerational environmental
reprogramming of metabolic gene expression in mammals. Cell 143, 1084–1096
(2010).
8.	 Ng, S.-F. et al. Chronic high-fat diet in fathers programs β-cell dysfunction in female
rat offspring. Nature 467, 963–966 (2010).
9.	 Crews, D. et al. Transgenerational epigenetic imprints on mate preference.
Proc. Natl. Acad. Sci. USA 104, 5942–5946 (2007).
10.	Kaati, G., Bygren, L.O. & Edvinsson, S. Cardiovascular and diabetes mortality
determined by nutrition during parents’ and grandparents’ slow growth period.
Eur. J. Hum. Genet. 10, 682–688 (2002).
11.	Rakyan, V.K. et al. Transgenerational inheritance of epigenetic states at the murine
Axin(Fu) allele occurs after maternal and paternal transmission. Proc. Natl. Acad.
Sci. USA 100, 2538–2543 (2003).
12.	Morgan, C.P. & Bale, T.L. Early prenatal stress epigenetically programs
dysmasculinization in second-generation offspring via the paternal lineage.
J. Neurosci. 31, 11748–11755 (2011).
13.	Harris, A. & Seckl, J. Glucocorticoids, prenatal stress and the programming of
disease. Horm. Behav. 59, 279–289 (2011).
14.	Meaney, M.J., Szyf, M. & Seckl, J.R. Epigenetic mechanisms of perinatal
programming of hypothalamic-pituitary-adrenal function and health. Trends Mol.
Med. 13, 269–277 (2007).
15.	Rodgers, A.B., Morgan, C.P., Bronson, S.L., Revello, S. & Bale, T.L. Paternal stress
exposure alters sperm microRNA content and reprograms offspring HPA stress axis
regulation. J. Neurosci. 33, 9003–9012 (2013).
16.	Buck, L. & Axel, R. A novel multigene family may encode odorant receptors: a
molecular basis for odor recognition. Cell 65, 175–187 (1991).
17.	Ressler, K.J., Sullivan, S.L. & Buck, L.B. A zonal organization of odorant receptor
gene expression in the olfactory epithelium. Cell 73, 597–609 (1993).
18.	Mombaerts, P. Molecular biology of odorant receptors in vertebrates. Annu. Rev.
Neurosci. 22, 487–509 (1999).
19.	Jones, S.V., Choi, D.C., Davis, M. & Ressler, K.J. Learning-dependent structural
plasticity in the adult olfactory pathway. J. Neurosci. 28, 13106–13111 (2008).
20.	Bozza, T., Feinstein, P., Zheng, C. & Mombaerts, P. Odorant receptor expression
defines functional units in the mouse olfactory system. J. Neurosci. 22, 3033–3043
(2002).
npg©2014NatureAmerica,Inc.Allrightsreserved.
96	 VOLUME 17 | NUMBER 1 | JANUARY 2014  nature NEUROSCIENCE
a r t ic l e s
21.	Ressler, K.J., Sullivan, S.L. & Buck, L.B. Information coding in the olfactory system:
evidence for a stereotyped and highly organized epitope map in the olfactory bulb.
Cell 79, 1245–1255 (1994).
22.	Vassalli, A., Rothman, A., Feinstein, P., Zapotocky, M. & Mombaerts, P. Minigenes
impart odorant receptor-specific axon guidance in the olfactory bulb. Neuron 35,
681–696 (2002).
23.	Davis, M., Falls, W.A., Campeau, S. & Kim, M. Fear-potentiated startle: a neural
and pharmacological analysis. Behav. Brain Res. 58, 175–198 (1993).
24.	Hebb, A.L.O., Zacharko, R.M., Gauthier, M. & Drolet, G. Exposure of mice to a
predator odor increases acoustic startle but does not disrupt the rewarding properties
of VTA intracranial self-stimulation. Brain Res. 982, 195–210 (2003).
25.	Weaver, I.C.G., Meaney, M.J. & Szyf, M. Maternal care effects on the hippocampal
transcriptome and anxiety-mediated behaviors in the offspring that are reversible
in adulthood. Proc. Natl. Acad. Sci. USA 103, 3480–3485 (2006).
26.	Weinstock, M. Intrauterine factors as determinants of depressive disorder. Isr. J.
Psychiatry Relat. Sci. 47, 36–45 (2010).
27.	Champagne, F.A. & Meaney, M.J. Stress during gestation alters postpartum maternal
care and the development of the offspring in a rodent model. Biol. Psychiatry 59,
1227–1235 (2006).
28.	Lomvardas, S. et al. Interchromosomal interactions and olfactory receptor choice.
Cell 126, 403–413 (2006).
29.	Magklara, A. et al. An epigenetic signature for monoallelic olfactory receptor
expression. Cell 145, 555–570 (2011).
30.	Cedar, H. & Bergman, Y. Linking DNA methylation and histone modification: patterns
and paradigms. Nat. Rev. Genet. 10, 295–304 (2009).
31.	van der Heijden, G.W. et al. Sperm-derived histones contribute to zygotic chromatin
in humans. BMC Dev. Biol. 8, 34 (2008).
32.	Brykczynska, U. et al. Repressive and active histone methylation mark distinct
promoters in human and mouse spermatozoa. Nat. Struct. Mol. Biol. 17, 679–687
(2010).
33.	Rassoulzadegan, M. et al. RNA-mediated non-mendelian inheritance of an epigenetic
change in the mouse. Nature 441, 469–474 (2006).
34.	Stickrod, G., Kimble, D.P. & Smotherman, W.P. In utero taste/odor aversion
conditioning in the rat. Physiol. Behav. 28, 5–7 (1982).
35.	Hales, C.N. & Barker, D.J. The thrifty phenotype hypothesis. Br. Med. Bull. 60,
5–20 (2001).
36.	Storm, J.J. & Lima, S.L. Mothers forewarn offspring about predators:
a transgenerational maternal effect on behavior. Am. Nat. 175, 382–390
(2010).
37.	Todrank, J., Heth, G. & Restrepo, D. Effects of in utero odorant exposure on
neuroanatomical development of the olfactory bulb and odour preferences.
Proc. Biol. Sci. 278, 1949–1955 (2011).
38.	Skinner, M.K. What is an epigenetic transgenerational phenotype? F3 or F2. Reprod.
Toxicol. 25, 2–6 (2008).
39.	Guerrero-Bosagna, C., Settles, M., Lucker, B. & Skinner, M.K. Epigenetic
transgenerational actions of vinclozolin on promoter regions of the sperm epigenome.
PLoS ONE 5, e13100 (2010).
40.	Vassoler, F.M., White, S.L., Schmidt, H.D., Sadri-Vakili, G. & Pierce, R.C. Epigenetic
inheritance of a cocaine-resistance phenotype. Nat. Neurosci. 16, 42–47
(2012).
41.	Roth, T.L., Lubin, F.D., Funk, A.J. & Sweatt, J.D. Lasting epigenetic influence of
early-life adversity on the BDNF Gene. Biol. Psychiatry 65, 760–769 (2009).
42.	Greer, E.L. et al. Transgenerational epigenetic inheritance of longevity in
Caenorhabditis elegans. Nature 479, 365–371 (2011).
43.	Rechavi, O., Minevich, G. & Hobert, O. Transgenerational inheritance of an acquired
small RNA-based antiviral response in C. elegans. Cell 147, 1248–1256
(2011).
44.	Cavalli, G. & Paro, R. Epigenetic inheritance of active chromatin after removal of
the main transactivator. Science 286, 955–958 (1999).
45.	Maruniak, J.A., Silver, W.L. & Moulton, D.G. Olfactory receptors respond to bloodborne
odorants. Brain Res. 265, 312–316 (1983).
46.	Vanderhaeghen, P., Schurmans, S., Vassart, G. & Parmentier, M. Specific repertoire
of olfactory receptor genes in the male germ cells of several mammalian species.
Genomics 39, 239–246 (1997).
47.	Oakberg, E.F. Duration of spermatogenesis in the mouse and timing of stages of
the cycle of the seminiferous epithelium. Am. J. Anat. 99, 507–516 (1956).
48.	Fleischmann, A. et al. Mice with a “monoclonal nose”: perturbations in an olfactory
map impair odor discrimination. Neuron 60, 1068–1081 (2008).
49.	Glinka, M.E. et al. Olfactory deficits cause anxiety-like behaviors in mice.
J. Neurosci. 32, 6718–6725 (2012).
50.	Jovanovic, T. et al. Physiological markers of anxiety are increased in children of
abused mothers. J. Child Psychol. Psychiatry 52, 844–852 (2011).
npg©2014NatureAmerica,Inc.Allrightsreserved.
nature NEUROSCIENCEdoi:10.1038/nn.3594
ONLINE METHODS
Mice. All experiments on adult offspring were conducted with 2-month-old male
mice. When F0 males and F0 females were fear conditioned with odor, 2-monthold
sexually inexperienced and odor-inexperienced mice were used. C57Bl/6J
mice (parents) were procured from Jackson Laboratory. M71-IRES-tauLacZ mice
(parents) maintained in mixed 129/Sv × C57Bl/6J background were bred in the
Yerkes Neuroscience animal facility. Mice were housed on a 12-h light/dark cycle
in standard groups cages (≤5/cage) with ad libitum access to food and water,
with all experiments conducted during the light half of the cycle. All procedures
were approved by the Institutional Animal Care and Use Committee of Emory
University, and followed guidelines set by the US National Institutes of Health.
Behavior. All behavior was performed in a double-blind manner and data
acquired using automated computer software programs. We are grateful to
S. Banerjee, R. Andero-Gali, D. Choi, J. Goodman and F. Morrison for help with
ensuring double-blindness of data acquisition and analysis.
Elevated plus maze. The elevated-plus maze consists of an elevated platform
with two walled, closed arms and two non-walled, open arms connected by an
open center. The mice were placed onto the center between the plus maze arms
and were recorded exploring the plus maze for 5 min. The amount of time spent
in the closed and open arms is viewed as a measure of anxiety.
Olfactory fear conditioning of parents. Mice were trained to associate acetophenone
or propanol presentation with mild foot shocks. For this purpose,
the Startle-Response system (SR-LAB, San Diego Instruments) was modified to
deliver discrete odor stimuli as previously described19. The mice were trained on
3 consecutive days, with each training day consisting of 5 trials of odor presentation
for 10-s co-terminating with a 0.25-s 0.4-mA foot shock with an average
inter-trial interval of 120 s. Both acetophenone and propanol (from Sigma) were
used at a 10% concentration diluted with propylene glycol.
OPSofadultoffspring.Micewerehabituatedtothestartlechambersfor5–10min
on three separate days. On the day of testing, mice were first exposed to 15 startlealone
(105-dB noise burst) trials (leaders), before being presented with ten odor +
startle trials randomly intermingled with ten startle-alone trials. The odor +
startle trials consisted of a 10-s odor presentation co-terminating with a 50-ms,
105-dB noise burst. For each mouse, an OPS score was computed by subtracting
the startle response in the first odor + startle trial from the startle response in the
last startle-alone leader. This OPS score was then divided by the last startle-alone
leader and multiplied by 100 to yield the percent OPS score (% OPS) reported in
the results. Mice were exposed to the acetophenone-potentiated startle (acetophenone
+ startle) and propanol-potentiated startle (propanol + startle) procedures
on independent days in a counter-balanced fashion.
Auditory fear conditioning. Mice were pre-exposed to sound attenuated
conditioning chambers (San Diego Instruments) for three consecutive days
before training. On the day of auditory fear conditioning, mice received five
conditioned-unconditioned stimulus pairings (conditioned stimulus: 30-s, 6kHz,
75-dB tone; unconditioned stimulus: 500-ms, 0.6-mA foot shock) with a
5-min inter-trial interval. The percentage of time spent freezing to the tones was
measured by SR-LAB software (San Diego Instruments). The consolidation of fear
memory was tested 24 h after fear conditioning in a novel context (modular test
chambers, Med Associates) when mice were exposed to 15 conditioned stimulus
tones with a 1.5-min inter-trial interval. Freezing during the tone presentations
was measured with FreezeView software (Coulbourn Instruments). The extinction
retention test occurred 24 h after extinction training and consisted of 30
conditioned stimulus tone presentations to the mice.
Odor sensitivity. Mice were placed in a three-chambered box and allowed to
explore between all three chambers for 10 min. A particular concentration of
odor (Fig. 2, either acetophenone or propanol) contained in a 1.5-ml Eppendorf
tube was placed in one of the chambers with the middle chamber empty and an
empty Eppendorf tube in the farthest chamber. Association time with either the
odor or the empty chamber was recorded. An aversion index was computed by
subtracting the amount of time spent in the open chamber from the time spent in
the odor chamber. Pilot experiments on independent mice revealed an increasing
aversion for either acetophenone or propanol as the concentration increased.
Independent mice were used in the acetophenone and propanol experiments
(F1-Ace-C57, n = 16; F1-Prop-C57, n = 16).
IVF. IVF was carried out by the Emory Transgenic Mouse Facility (TMF) located
in a different building from our colony (http://med.emory.edu/research/core_
labs/transgenic_mouse/) across the Emory Campus. Briefly, F0-M71 males were
fear conditioned either to acetophenone or propanol as outlined above. 10 d
after fear conditioning, sperm was collected from the caudal epididymis and vas
deferens of these males in our facility, and then transported to the TMF wherein
IVF was conducted by TMF personnel blinded to the experimental conditions
of the sperm samples according to protocols followed by Jackson Laboratory51.
In vitro fertilization culture medium, Mouse Vitro Fert (MVF, Cook Medical),
was used for sperm isolation, IVF and zygote culture. Superovulated C57BL/6
female mice were used as oocyte donors. Sperm were co-incubated with oocytes
in MVF for 4 h in a 5% CO2 incubator, the presumptive zygotes were washed, and
were cultured overnight in a 150-µl MVF drop in the incubator. In the second
morning, two-cell embryos were scored and washed in MVF drop; pseudopregnant
CD-1 female mice of 9–13 weeks of age were used as embryo recipients.
15–20 embryos were transferred into one oviduct of each female. Pups were
born after 19 d and weaned from their foster moms at 3.5 weeks of age, they were
reared to 2 months of age and then tissue was collected in the TMF facility for
β-galactosidase staining on the MOE and olfactory bulb.
b-galactosidase staining, quantitation of MOE OSN number and glomerular
area in bulb. The MOE and olfactory bulbs of 2-month-old M71-LacZ mice
were processed for β-galactosidase staining, and then M71 OSN number and
glomerular area were quantitated using previously published protocols19. Briefly,
lateral whole-mount MOE and brains were rapidly dissected and placed into 4%
paraformaldehyde (wt/vol) for 10 min at ~23 °C, after which they were washed
three times in 1× phosphate-buffered saline (PBS) for 5 min. M71-LacZ was
stained for β-galactosidase using 45 mg of X-gal (1 mg ml−1) dissolved in 600 µl of
DMSO and 45 ml of a solution of 5 mM potassium ferricyanide, 5 mM potassium
ferrocyanide and 2 mM MgCl in 1× PBS, and incubated at 37 °C for 3 h.
Quantitation of MOE M71-positive OSN number. The lateral whole mount
MOE was imaged using a microscope-mounted digital camera. β-galactosidase–
stained blue OSNs were manually counted by an experimenter blinded to the
experimental groups.
Measurementofglomerularareaintheolfactorybulb.A microscope-mounted
digital camera was used to capture high-resolution images of theβ-galactosidase–
stained M71 glomeruli at 40× magnification. Images were converted to grayscale
and equalized for background brightness. The distribution of pixel brightness was
exported in ImageJ as gray levels from 0 = black to 255 = white. X-gal–labeled
glomerular area was quantified as pixels, less than a set threshold gray level of
150 (optimized for axon versus background). Each glomerulus was traced using
the lasso tool in Photoshop and the area was recorded from the histogram tool.
This quantitation was conducted by two experimenters both blinded to the
experimental groups.
N-ChIPonsperm. N-ChIP was conducted on sperm chromatin using previously
described procedures52. Briefly, the cauda epidydymis was dissected into 1 ml of
M2 medium (Sigma), and sperm were allowed to swim into the medium for 1 h
at 37 °C. Five epidydymis were used per sample, and each experimental group
had three samples. At least 3 × 106 sperm were used for each ChIP. Sperm were
then collected by centrifugation at 4 °C for 10 min at 500g, and resuspended in
1× PBS, 1 mM PMSF. Sperm were then lysed on ice for 10 min in 1× PBS, 1 mM
PMSF, 0.5% Triton X-100 (vol/vol). Nuclei were pelleted by centrifugation at
4 °C for 10 min at 371g. The pellet was then suspended in 1× PBS, 1 mM PMSF,
10 mM DTT, and incubated at 37 °C for 30 min, before the addition of 0.6 mM
CaCl2 and MNase (Sigma) to yield mono-, di- and tri-nucleosomal chromatin.
Immunoprecipitation was then carried out as described for the MOE and
followed the previously established protocol52. The antibodies were used at
1:1,000 and were specific to H3 trimethyl lysine-27 (07-449) and acetyl histone H3
(06-599) from Upstate. Immunoprecipitated DNA was isolated by phenolchloroform
extraction and ethanol precipitation and used in quantitative PCR
npg©2014NatureAmerica,Inc.Allrightsreserved.
nature NEUROSCIENCE doi:10.1038/nn.3594
reactions on an ABI 7900 Real-Time PCR machine. 5 mM sodium butyrate was
added to all buffers and wash solutions to inhibit histone deacetylases. Primers
for the control genes were the same as those used in ref. 32. ChIP on sperm was
conducted on two independent sets of samples with similar results.
NGS analysis of bisulfite PCR amplicons from sperm and MOE DNA. F0-C57
males were fear conditioned to either acetophenone (F0-Ace-Sperm, n = 12) or
propanol (F0-Prop-Sperm, n = 10) as outlined above. 10 d after fear conditioning,
sperm was collected from these males in our facility. As outlined above, a
separate group of F0-Ace-C57 and F0-Prop-C57 males sired F1 offspring. At 2
months of age, sperm and MOE were collected from F1-Ace-C57 and F1-PropC57
mice (n = 4 each). In yet another independent experiment, F1-Ace or F1Prop
males sired F2-Ace-C57 and F2-Prop-C57 mice, respectively. MOE from
this F2 generation (F2-Ace-C57, and F2-Prop-C57) were collected at 2 months
of age. All samples were shipped to Active Motif (http://www.activemotif.com)
for genomic DNA extraction, bisulfite conversion, PCR-based library generation
and sequencing. We coded the samples, and personnel at Active Motif were
blinded to this code.
PCR primers to the target regions were designed with the MethPrimer software
(http://www.urogene.org/cgi-bin/methprimer/methprimer.cgi) (Supplementary
Fig. 5). For the F1 sperm, F1 MOE and F2 MOE samples, a shorter amplicon
was generated, and we queried eight CpG sites in this analysis, compared with
nine sites in the F0 generation. The queried CpG sites 1–8 were the same in the
F0 and F1 generations. Genomic DNA was isolated from the sperm samples
using Quick-gDNA miniprep kit (Zymo Research), and bisulfite-converted using
MethylDetector (Active Motif). PCR reactions (40–55 cycles) were performed
using Invitrogen’s Platinum PCR supermix.
DNA samples containing approximately the same amounts of two (or
four) bisulfite PCR products (~300 ng DNA total) were treated with T4 DNA
polymerase, Klenow large fragment and T4 polynucleotide kinase to generate
5′-phophorylated blunt ends. After concatemerization with T4 DNA ligase,
the sample was sonicated to an average fragment length of 150–300 bp using a
Misonix cuphorn sonicator 3000. Libraries were generated from these sonicated
DNA samples using the standard Illumina protocol. The 8 (16) samples were
indexed with 6-bp barcodes (independent Illumina index read). Sequencing on
Hi-Seq generated ~5–10 million reads per sample. Reads were aligned to chr7 and
chr9 reference sequences (mm9 assembly) using the Bismark software (version
0.7.7)53. Alignment and methylation information was captured in BAM files, and
percentage methylation and read coverage at each CpG site was determined by
running the appropriate Bismark scripts. Alignments to the strands and genomic
locations not expected to be present in the PCR products were filtered out using
a combination of SAMtools54 and standard UNIX commands.
51.	Ostermeier, G.C., Wiles, M.V., Farley, J.S. & Taft, R.A. Conserving, distributing and
managing genetically modified mouse lines by sperm cryopreservation. PLoS ONE
3, e2792 (2008).
52.	Umlauf, D., Goto, Y. & Feil, R. Site-specific analysis of histone methylation and
acetylation. Methods Mol. Biol. 287, 99–120 (2004).
53.	Krueger, F. & Andrews, S.R. Bismark: a flexible aligner and methylation caller for
Bisulfite-Seq applications. Bioinformatics 27, 1571–1572 (2011).
54.	Li, H. et al. 1000 Genome Project Data Processing Subgroup. The sequence
alignment/map (SAM) format and SAMtools. Bioinformatics 52, 2078–2079
(2009).
npg©2014NatureAmerica,Inc.Allrightsreserved.
nature neuroscience
CORRI G EN DA
Corrigendum: Parental olfactory experience influences behavior and neural
structure in subsequent generations
Brian G Dias & Kerry J Ressler
Nat. Neurosci.; doi:10.1038/nn.3594; corrected online 9 December 2013
In the version of this article initially published online, the base grant to the Yerkes National Primate Research Center was omitted from the
Acknowledgments. The error has been corrected for the print, PDF and HTML versions of this article.
npg©2014NatureAmerica,Inc.Allrightsreserved.