https://doi.org/10.

1093/braincomms/fcae390 BRAIN COMMUNICATIONS 2024: fcae390 | 1

BRAIN COMMUNICATIONS

Olfactory stimulation with multiple odorants

prevents stress-induced cognitive
and psychological alterations

Downloaded from https://academic.oup.com/braincomms/article/6/6/fcae390/7876424 by guest on 16 July 2025

Bruno Bandiera,1,* Francesca Natale,1,2,* Marco Rinaudo,1,2 Raimondo Sollazzo,1
Matteo Spinelli,1,3 Salvatore Fusco1,2,† and Claudio Grassi1,2,†

* These authors contributed equally to this work.

† These authors contributed equally to this work.

Acute and chronic stress markedly affects behavior by triggering sympathetic nervous system activation and several hypothalamus-
pituitary-adrenal-dependent responses. Brain regions of the limbic system are responsible for the regulation of stress response, and
different reports have demonstrated that their activity can be influenced by olfactory stimuli. Here we report that, in mice exposed
to acute restraint stress, olfactory stimulation employing a combination of three odorants, i.e. vanillin, limonene and green odor
(trans-2-hexenal and cis-3-hexenol) decreased anxiety behavior, assessed in the elevated plus maze, and halted recognition and spatial
memory deficits, as appraised in two different object recognition tasks. Of note, when applied singularly, the same odorants were un­
able to block the detrimental effects of stress. We also found that the multiple odorants stimulation prevented the development of
depressive symptoms assessed by the sucrose splash test and forced swim test in an experimental model of depression, i.e. mice exposed
to a chronic unpredictable stress paradigm, and reduced interleukin 1β levels in the prefrontal cortex of depressed mice. Collectively,
our data indicate that olfactory stimulation counteracts the detrimental effects of acute and chronic stress on mood regulation and
cognitive functions, thus representing a potential tool for the treatment of stress-induced disorders.

1 Department of Neuroscience, Università Cattolica del Sacro Cuore, Rome 00168, Italy
2 Fondazione Policlinico Universitario ‘A. Gemelli’ IRCCS, Rome 00168, Italy
3 Department of Biomedical Sciences, Università degli studi di Sassari, Sassari 07100, Italy
Correspondence to: Marco Rinaudo, Ph.D
Department of Neuroscience, Università Cattolica del Sacro Cuore
00168 Rome, Italy; Fondazione Policlinico Universitario ‘A. Gemelli’ IRCCS
00168 Rome, Italy
E-mail: marco.rinaudo@unicatt.it
Keywords: stress; odorants; anxiety; memory; depression

Received June 11, 2024. Revised October 02, 2024. Accepted November 06, 2024. Advance access publication November 5, 2024
© The Author(s) 2024. Published by Oxford University Press on behalf of the Guarantors of Brain.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse,
distribution, and reproduction in any medium, provided the original work is properly cited.
2 | BRAIN COMMUNICATIONS 2024, fcae390 B. Bandiera et al.

Graphical Abstract

Downloaded from https://academic.oup.com/braincomms/article/6/6/fcae390/7876424 by guest on 16 July 2025

hypothalamus also reduce the activity of corticotropin-
Introduction releasing hormone neurons, which are activated in the pres­
Environmental threats to physical and/or psychological in­ ence of a stressor.10 Different reports have shown that olfac­
tegrity induce the activation of specific stress-related path­ tory stimuli can influence limbic activity, in terms of both
ways allowing the individual to better respond or adapt to stress response/emotional regulation and learning and mem­
the source of stress.1 Stress-inducing stimuli activate the sym­ ory processes in animal models as well as in humans.11
pathetic nervous system and the hypothalamus-pituitary- Aromatherapy has been employed by humans for centur­
adrenal axis, which in turn induce the production of stress ies to improve different psychological functions.12 Indeed,
hormones. The net effects of this complex nervous and hor­ some aromatic components derived from plants can modu­
monal response include increased heart rate and arterial late different receptors in the brain or they can, as well,
blood pressure, glucose mobilization and increased respira­ stimulate olfactory perception. Different reports have shown
tory rate, preparing the body for the fight or flight response.2 that olfactory stimulation using particular odorants is able to
However, duration and intensity of the stressor can lead to impact on body physiology.13 In humans, it has been demon­
different behavioral outcomes, rapidly precipitating their strated that exposure to a pleasant odorant, such as black tea
short-term beneficial effects to more pronounced long-term aroma, can reduce the levels of salivary chromogranin-A,
aberrations causing anxiety, cognitive alterations and in­ which is used as a marker of stress levels,14 or lavender
creasing the risk for the development of mood disorders.3 and rosemary smell induce an increase in free radical scaven­
The limbic system, an ensemble of cortical and subcortical ging activity and reduce cortisol levels in saliva.15 Similar re­
brain regions including the medial prefrontal cortex, hippo­ sults have been obtained in animal models, where different
campus, amygdala, and hypothalamus, is the main regulator odorants have been reported to modulate stress responses.
of the stress response and it is highly influenced by sensory In mice, exposure to roman chamomile essential oil com­
inputs, and especially olfactory stimuli.4 Olfactory informa­ bined with clomipramine improves behavioral symptoms
tion is first encoded within the olfactory bulb, which subse­ in an experimental model of depression.16 Limonene inhal­
quently projects to the piriform cortex.5 From the piriform ation reduces anxiety through the modulation of serotonin
cortex, odorant-induced activity is then conveyed to amyg­ and dopaminergic receptors, whereas α-pinene can enhance
dalar nuclei, the entorhinal cortex that is connected to hippo­ the mRNA expression of neurotrophins.17,18 Other works
campus, to the prefrontal cortex and hypothalamus.6,7 A have shown that green odor, a mixture of 3-trans-hexenal
direct pathway connecting the olfactory bulb to the cortical and 2-cis-hexenol, can improve anxiety symptoms in a
amygdala has been reported to mediate innate odor-driven mouse model of post-traumatic stress disorder while vanillin
behavioral responses in mice.8 Direct projections from has been shown to possess antidepressant activity.19,20
hippocampus to the anterior olfactory nucleus have been de­ However, most works have focused on stimulation with a
monstrated to be necessary for the formation of odor-place single odorant molecule or essential oil, and a direct com­
associations.9 Olfactory inputs to the ventromedial parison between the effects of different odorants when
Odorants modulate inflammatory signaling BRAIN COMMUNICATIONS 2024, fcae390 | 3

inhaled singularly or presented in sequence has never been (CUS) was employed.22 Specifically, animals were subjected
performed, nor an evaluation of their efficacy in halting daily, for 6 weeks, to different stressors in an unpredictable
stress-induced alterations on memory. fashion. Stressors employed were: 1) 3-h restraint stress;
In this work, we sought to determine whether olfactory 2) 6 h food or water deprivation; 3) tilted cage at 45° for
stimulation with a combination of different odorants could 6 h; 4) soiled bedding for 6 h; 5) tail pinch or tail suspension
better impact and halt the detrimental effects of stress on for 3 min; 6) cold water swim for 5 min; 7) cage with no bed­
mood and memory compared to the stimulation with a single dings for 6 h. A couple of stressors were administered
odorant. We found that olfactory stimulation with a com­ each day. Concerning the olfactory stimulation, the
bination of vanillin, limonene and green odor prevented following odorants were employed: 1) vanillin, 1.8 mg/mL;
the development of anxiety, recognition and spatial memory 2) R-limonene, 10−3% v/v; 3) trans 2-hexen-1-al and cis
deficits in a mouse model of acute restraint stress that we 3 hexen-1-ol (green odor) at 10−3% v/v. All odorants were
characterized at biochemical, molecular, and behavioral le­ dissolved in 5% tween-aqueous solution and sprayed on

Downloaded from https://academic.oup.com/braincomms/article/6/6/fcae390/7876424 by guest on 16 July 2025

vels. The above-mentioned combinations of odorants also the bedding of mice home cages. Mice subjected to only
prevented the development of a depressive phenotype during the ARS protocol received a total of 12 sprays (80–100 µL
exposure to chronic stress, by improving behavioral and mo­ for each squirt) with vehicle solution, delivered at 20 min in­
lecular hallmarks of depression. tervals during the 2 h of ARS procedure, for a total of
72 sprays. Animals exposed to olfactory stimulation received
either 12 sprays of a single odorant (single-odorant condi­
Material and methods tion) or 4 sprays of each odorant (Multiple odorants condi­
tion, M.O., for a total of 12 sprays administered in a random
Animals order) every 20 min during the 2 h ARS procedure. During
the 2 h of ARS protocol, both groups received a total of 72
Wild-type C57BL/6 male mice (3–5 months of age), derived sprays in their home cage. Mice subjected to the CUS proto­
from the Animal Facility of Catholic University, were em­ col were daily exposed to olfactory stimulation consisting of
ployed for this study. Mice were housed in groups of three 4 sprays of each odorant (for a total of 12 sprays/day pre­
to five animals per cage. The animals were kept at a con­ sented at 5 min intervals in a random order) directly on the
trolled temperature of 24°C under a 12 h light/dark cycle bedding of their home cage for the whole duration of the
with unrestricted access to food (Mucedola 4RF21, Milan, CUS protocol. Nesting material was changed only during
Italy) and water. Animals within the same litter were allo­ the weekly cage change. All the odorants were sprayed
cated to different groups. in the cage in the morning, between 09:00 and 11:00 a.m.
for the CUS procedure, while they were sprayed for the
Ethics whole duration of ARS procedure. Animals were housed in
ventilated cage racks (IVS system, Tecniplast company)
All animal procedures were approved by the Ethics where the air enters and exits from each singular cage with
Committee of Università Cattolica and the Italian Ministry its own unique set of pipes, so that each cage has its own ven­
of Health (experimental protocol number 847/2021-PR). tilation system. In this way, the possibility of cross-exposure
They were fully compliant with Italian (Legislative Decree to other odorants beyond those employed was excluded. All
No. 26/2014) and European Union (Directive No. 2010/ stress procedures were performed under veterinary staff
63/UE) legislation on animal research. All efforts were supervision to control for animal health.
made to limit the number of animals used and to minimize
their suffering.
Behavioral paradigms
Stress procedures and olfactory Behavioral tests were performed as previously reported,22,23
with slight modifications. All tests were performed by experi­
stimulation menters blind to treatment, from 9:00 a.m to 14:00 p.m and
Stress procedures were performed as previously reported.21 using the ANY-maze tracking system (StoeltingTM). Animals
Briefly, for the acute restraint stress (ARS) procedure, ani­ from different experimental conditions were tested sequen­
mals were placed inside a 50 mL conic Falcon tubes for tially. Briefly, for anxiety analysis, animals were placed on
two hours and placed in their home cage. For the elevated the elevated plus maze (EPM) and allowed to explore the ap­
plus maze, animals were allowed to recover for 30 min prior paratus for 10 min. Time spent in the open and closed arms
to the beginning of the test. For the novel object recognition was recorded, as well as open arms entries. For the novel ob­
and object place recognition, ARS was administered after the ject recognition (NOR) and object place recognition (OPR)
training phase. For experiments involving blood sampling paradigms, animals were allowed to explore two identical
and brain tissue harvesting, blood was collected immediately objects placed within a square arena (33 × 33 cm) for
after the ARS procedure. At the end of blood collection ani­ 10 min. Twenty-four hours later, in the case of the NOR
mals were sacrificed to isolate the ventral hippocampus. For test, one of the objects was substituted with a novel one
the chronic stress paradigm, the chronic unpredictable stress and animals were allowed to explore the arena and the
4 | BRAIN COMMUNICATIONS 2024, fcae390 B. Bandiera et al.

objects for 5 min. For the OPR, instead, one of the objects Louis, MO, USA), 1 mM sodium orthovanadate (S6508,
was moved to a different position in the arena. Time spent Sigma-Aldrich, Saint Louis, MO, USA), and 1 mM sodium
exploring the novel/displaced object to total exploration fluoride (201154, Sigma-Aldrich, Saint Louis, MO, USA).
time was then reported as preference index (PI) and as dis­ After lysis, tissues were spun down at 22,000× g and 4°C,
crimination index (D.I., new/displaced minus old/stationary and the supernatants were quantified for protein content
divided by total exploration time). For the EPM, NOR and (500006, DC protein assay; Bio-Rad, Hercules, CA, USA).
OPR procedures, different cohorts of mice were employed. Equal amounts of protein were diluted in Laemmli buffer,
Between each animal, all the apparatuses and objects were boiled, and resolved by SDS-PAGE. The primary antibodies
cleaned with 70% ethanol. For the chronic stress procedure, were incubated overnight at 4°C and revealed with
the same cohort of mice was evaluated using the forced swim HRP-conjugated secondary antibodies (#7074 and #7076,
test (FST) and the sucrose splash test (SST), to measure Cell Signaling Technology Inc., Danvers, MA, USA).
stress-coping response and apathic/anhedonic behavior. Primary antibodies for phospho-ERK1/2 and total ERK1/2

Downloaded from https://academic.oup.com/braincomms/article/6/6/fcae390/7876424 by guest on 16 July 2025

For the FST, animals were placed in a baker filled with water (catalogue references, respectively: #9101 and #9102 from
at a temperature of 25° ± 1°C for six minutes, and total Cell Signaling Technology Inc.), pNF-kB (AB11226,
immobility time was recorded. Baker size was of 16 cm Immunological sciences), total NF-kB (ab16502, Abcam),
height × 12 cm diameter. Water depth was of 10 cm, leaving PSD-95 (#3450, Cell Signaling Technology Inc.), pGSK3β
6 cm from water surface to baker borders. All animals were (#9336, Cell Signaling Technology Inc.) and total GSK3β
tested using the same apparatus. The water was changed (#12456, Cell Signaling Technology Inc.) were diluted
every 10 animals to avoid temperature shifts. Concerning 1:1000. Primary antibodies for actin (ab8227, Abcam) and
the SST, animals were sprayed on their coat with 7 squirts Hsp90 (#4877, Cell Signaling) were diluted at 1:10,000.
of a 10% sucrose solution and grooming time was recorded Changes in protein phosphorylation were evaluated and
for the following 5 min. For the open field test assessing loco­ documented using UVItec Cambridge Alliance (Uvitec,
motor activity in the chronic stress model, animals were Cambridge, UK). Images shown were cropped for presenta­
placed inside a 45 × 45 cm square arena and allowed to ex­ tion without any other manipulation (see Supplementary
plore for 10 min. materials for uncropped images). A total of 22 mice (n = 9
control mice, n = 8 CUS mice, n = 5 ARS mice) were used
for western blot analyses.
ELISA
For corticosterone measurements the ELISA kit ADI-900-
097 from Enzo life sciences was employed. Briefly, blood
Immunofluorescence experiments
was collected, in the presence of EDTA, from the subman­ For immunofluorescent labeling, sections were processed as
dibular vein at the end of the ARS procedure. Samples previously described.27 Animals were intraperitoneally in­
were then centrifuged at room temperature at 600 g for jected with bromodeoxyuridine (BrdU; Sigma, St. Louis,
10 min, and plasma was subsequently isolated and stored MO, USA; 100 mg/kg dissolved in 0.9% NaCl solution)
at −80°C for the analysis. ELISA assay was then performed for the last 5 days of CUS exposure. Animals were deeply an­
according to the manufacturer instructions. For the esthetized and were transcardially perfused with PBS (0.1 M,
BDNF assay (kit IK-10146, Immunological Sciences) pH 7.4) followed by 4% PFA. Brains were collected, post-
and IL 1β ELISA (kit number IK-4205, Immunological fixed overnight at 4°C in PFA, and then transferred to a so­
Sciences) the measurements were performed following lution of 30% sucrose in 0.1 M PBS. Coronal brain sections
manufacturer instructions and as previously reported.24,25 (30-μm-thick) were cut with a vibratome (VT1000S, Leica
A total of 15 mice (n = 6 control mice, n = 6 CUS mice, Microsystems, GmbH, Wetzlar, Germany). Hippocampal
n = 3 CUS + M.O.) were used for both BDNF and IL 1β dos­ slices were incubated sequentially with 2N HCl for DNA hy­
age by ELISA. drolysis and epitope retrieval, 1×PBS with 0.3% Triton
X-100 (Sigma, St. Louis, MO, USA) and 5% NGS at RT
for permeabilization and blocking and then with BrdU anti­
Western blot body (1:500; Abcam, ab6362), overnight at 4°C. The next
Western blot analyses were performed as previously de­ day, tissues were incubated for 90 min at RT with the sec­
scribed.23 After ARS exposure, animals were sacrificed ondary antibody: Alexa Fluor-488 anti-rat (1:600;
through cervical dislocation for brain tissues collection. Invitrogen, a11006). Finally, nuclei were counterstained
Briefly, the brain was placed inside a brain matrix and sliced with DAPI (0.5 μg/mL for 10 min; Invitrogen), and slices
coronally at 4 mm from the olfactory bulbs, to collect the were coverslipped with ProLong Gold anti-fade reagent.
prefrontal cortex from anterior slices. Then, the ventral Images (1024 × 1024 pixels) were acquired at 20× magnifi­
hippocampus was isolated as reported in,26 with slight mod­ cation with a Nikon A1 MP confocal system (Tokyo,
ifications. The tissues were lysed in ice-cold lysis buffer Japan). For analyses, DAPI+/BrdU+, cells were counted.
(150 mM NaCl, 50 mM pH 8 Tris-HCl, and 2 mM EDTA) During brain sectioning, hippocampal slices were placed in
containing 1% Triton X-100, 0.1% sodium dodecyl sulfate, a 6-wells plate, by sequentially adding 1 slice per well from
1 × protease inhibitor cocktail (P8340, Sigma-Aldrich, Saint the anterior part of the hippocampus (−1.70 mm from
Odorants modulate inflammatory signaling BRAIN COMMUNICATIONS 2024, fcae390 | 5

bregma) to the posterior (−3.16 mm from bregma). Every 6 (22.9 ± 5.28 versus 163.4 ± 13.4 ng/mL, respectively;
slices, this procedure was repeated starting from the first Student’s t-test, t = 10.880, P < 0.001; Fig. 1D). After blood
well. Then, a single well, containing 8 slices, was employed collection, animals were sacrificed, and the ventral hippo­
for the analysis of the number of proliferating cells. All la­ campus and prefrontal cortex (PFC) were isolated for west­
beled cells within the subgranular zone (SGZ) of the dentate ern blot analyses. A significant increase of the extracellular
gyrus of each slice were counted separately. The sum of pro­ signal-regulated kinases (ERK) activation was observed in
liferating cells counted in 8 slices was then multiplied by 6 to the ventral hippocampus of ARS-exposed animals (43 ±
estimate the total number of labeled cells in the whole hippo­ 0.1% increase in ARS-exposed animals compared to control
campus. Image acquisition and analysis were carried out animals; Student’s t-test, t = 5.345, P < 0.001; Fig. 1E
using the software NIS Elements AR 5.30.01. A z-stack and F). In the PFC no significant difference in nuclear factor
analysis, which allows to evaluate the fluorescence kappa B (NF-kB) activatory phosphorylation was observed
intensity of BrdU signal cell-by-cell along the 3 cross-sections between the two groups (5% decrease in ARS-exposed

Downloaded from https://academic.oup.com/braincomms/article/6/6/fcae390/7876424 by guest on 16 July 2025

XY, XZ, and YZ, was then performed. A total of 7 mice animals compared to control animals; Student’s t test,
(n = 3 control mice, n = 4 CUS mice) were used for histo­ t = 0.416, P = 0.691; Fig. 1G and H), whereas a significant
logical analyses. decrease in ERK phosphorylation was found in
ARS-exposed mice (13% decrease in ARS-exposed animals
compared to control animals; Student’s t test, t = 2.457,
P = 0.049; Fig. 1G and I).
Statistical analyses
Sample sizes were calculated with adequate power (0.8)
based on pilot studies and literature data. All statistical ana­
lyses were performed by using SigmaPlot 14 software (Systat Olfactory stimulation with multiple
Software, Palo Alto, CA, USA). Data distribution was first odorants halts the effects of ARS on
evaluated for equal variance and normality (Shapiro–Wilk
test). All statistical tests used (Student’s t test, one-way
anxious behavior
ANOVA, one-way ANOVA on Ranks, two-way ANOVA Once tested the efficacy of our stress paradigm, we investi­
and post-hoc tests) are reported in the main text. The sample gated the effects of olfactory stimulation on ARS-induced be­
sizes (n) are reported in the figure legends. Significance was havioral alterations. Specifically, animals were exposed to
set at 0.05, and all tests were two-tailed. The results are re­ three different odorants, i.e. vanillin, limonene and green
ported as means ± sem. No animals were excluded from odor to compare the effects of a single versus multiple odor­
the study. ant stimulation (Fig. 2A). Exposure to a single odorant did
not halt the behavioral alterations induced by ARS proced­
ure in terms of time spent in the closed and open arms
(closed arms time: CTRL 452.6 ± 10.1 s; ARS 532.7 ±
Results 7.8 s; ARS-Vanillin 501.6 ± 11.7 s; ARS-Limonene 526.7
± 18.4 s; ARS-Green Odor 511.3 ± 7.6 s; One-way
ARS induces a stress phenotype ANOVA, F(5,64) = 7.948, P < 0.001; post-hoc Holm-Sidak
To characterize the stress response in our experimental mod­ method: CTRL versus ARS: P < 0.001; CTRL versus ARS-
el of acute stress, we performed biochemical, molecular and Limonene: P < 0.001; CTRL versus ARS-Green Odor: P =
behavioral analyses in mice subjected to ARS for 2 h. Thirty 0.010; CTRL versus ARS-Vanillin: P = 0.019; open arms
minutes after the end of the ARS procedure, animals were time: CTRL 57.8 ± 6.6 s; ARS 21.7 ± 4 s; ARS-Vanillin
tested in the Elevated Plus Maze (EPM). ARS-exposed ani­ 39.3 ± 9.6 s; ARS-Limonene 23.4 ± 10.5 s; ARS-Green
mals showed a significant increase in time spent in closed Odor 30.6 ± 6.9 s; ANOVA on Ranks, P < 0.001; post-hoc
arms compared to control mice (525.0 ± 11.8 s versus Dunn’s method: CTRL versus ARS: P = 0.010; CTRL versus
464.0 ± 8.2 s, respectively; Student’s t-test, t = 4.502 P < ARS-Limonene P = 0.011), as well as the number of open
0.001, Fig. 1A) and a significant decrease in both time spent arms entries (CTRL 21.0 ± 3.1; ARS 8.8 ± 1.3; ARS-
in open arms (23.0 ± 6.3 s versus 44.0 ± 3.4 s; Student’s Vanillin 15.2 ± 2.4; ARS-Limonene 8.3 ± 2.4; ARS-Green
t-test, t = 3.132, P = 0.016; Fig. 1B) and number of open Odor 11.5 ± 2.2; One-way ANOVA, F(5,64) = 5.197, P <
arms entries (6.8 ± 1.2 versus 10.7 ± 0.9; Student’s t-test, 0.001; post-hoc Holm-Sidak method: CTRL versus ARS:
t = 2.711, P = 0.007, Fig. 1C). To avoid any possible P = 0.005; CTRL versus ARS-Limonene: P = 0.011;
influence of either EPM testing on the studied molecular Fig. 2C and D). However, stressed mice exposed to all odor­
parameters or of the blood collection on behavioral assess­ ants showed a significant decrease in time spent in closed
ment, a different cohort of mice exposed to the same ARS arms (466.6 ± 13.4 s, P = 0.003 versus ARS), a significant in­
protocol was employed for biochemical experiments. crease in time spent in the open arms (55 ± 7.1 s, P = 0.038
Blood was collected soon after the end of the 2-hour ARS versus ARS) as well as increased number of entries in the
procedure. Corticosterone levels in ARS-exposed animals open arms (19.7 ± 3.5, P = 0.031 versus ARS) compared to
were significantly higher compared to control animals ARS animals (Fig. 2B and C and D).
6 | BRAIN COMMUNICATIONS 2024, fcae390 B. Bandiera et al.

Downloaded from https://academic.oup.com/braincomms/article/6/6/fcae390/7876424 by guest on 16 July 2025

Figure 1 Acute restraint stress exposed animals show behavioral, molecular, and biochemical hallmarks of acute stress
response. Animals were subjected to 2 h acute restraint stress (ARS). A, B and C) Thirty minutes after the end of the ARS protocol animals
underwent the EPM paradigm. ARS-exposed animals showed a significant increase in time spent in closed arms (A) and a significant decrease in
time and entries into the open arms (B and C, n = 8 for both groups). D) Corticosterone levels after ARS exposure. ARS-exposed animals show a
significant increase in blood corticosterone levels compared to control animals (n = 5 for both groups). E and F) ERK phosphorylation at residues
Thr202/Tyr204 was significantly higher in ARS-exposed animals (n = 5 for both groups). G, H and I) NF-kB activatory phosphorylation at Ser311
and ERK phosphorylation at residues Thr202/Tyr204 in the PFC of CTRL and ARS-exposed mice. No differences were observed in NF-kB
phosphorylation while a statistically significant decrease in ERK phosphorylation was observed (n = 4 for both groups). All data are reported as
mean ± s.e.m. Dots represent the number of samples (studied animals). Student’s t test for all comparisons; *P < 0.05; **P < 0.01; ***P < 0.001. See
supplementary materials for uncropped blots.

Olfactory stimulation with multiple observed in all ARS mice exposed to a single odorant (D.I.:
CTRL 0.27 ± 0.04; ARS −0.11 ± 0.1; ARS-Vanillin 0.02 ±
odorants halts the effects of ARS on 0.09; ARS-Limonene 0.04 ± 0.05; ARS-Green Odor 0.01 ±
recognition memory 0.06; ANOVA on Ranks, P < 0.001; post-hoc Dunn’s
It has been reported that stressful stimuli also affect recognition Method: CTRL versus ARS: P = 0.009, Fig. 3B). Moreover,
memory.21 Thus, we investigated memory performance after as shown in Fig. 3C, the P.I. for the novel and old object did
exposure to ARS protocol associated with olfactory stimula­ not differ in ARS-exposed animals among vehicle- and each
tion using either a single or multiple odorants. Animals were single odorant-treated mice (P.I. New versus P.I. Old, respect­
exposed to the ARS procedure during the consolidation phase ively: CTRL 63.7 ± 1.9% versus 36.3 ± 1.9%; ARS 44.5 ±
of the NOR training, right after the training phase, and they 5.2% versus 55.5 ± 5.2%; ARS-Vanillin 51.1 ± 4.7% versus
were tested 24 h later (Fig. 3A). ARS-exposed mice showed 48.8 ± 4.7%; ARS-Limonene 51.7 ± 2.4% versus 48.2 ±
memory impairment with a significant reduction of the D.I. 2.4%; ARS-Green Odor 50.6 ± 2.9% versus 50.6 ± 2.9%;
compared to control animals, and the same pattern was Student’s t-test, P.I. novel versus P.I. old: CTRL, t = 10.785,
Odorants modulate inflammatory signaling BRAIN COMMUNICATIONS 2024, fcae390 | 7

Downloaded from https://academic.oup.com/braincomms/article/6/6/fcae390/7876424 by guest on 16 July 2025

Figure 2 Olfactory stimulation using multiple odorants prevents stress induced anxiety. (A) Schematic representations of the
experimental protocol. (B, C and D) Evaluation of the behavioral profile of animals exposed to the acute restraint stress (ARS) protocol in the
presence of either a single or multiple odorants (ARS-M.O.). ARS-M.O. exposed animals were the only group showing a significant reduction of
time spent into closed arms (B) and a significant increase in time spent and entries into open arms C and D; CTRL n = 13; ARS n = 11; ARS + M.O.
n = 10; ARS + Vanillin n = 13; ARS + Limonene n = 9; ARS + Green Odor n = 9). All data are reported as mean ± s.e.m. Dots represent the
number of samples (studied animals). Statistics by One-Way ANOVA followed by post-hoc Holm-Sidak (B, D), One-Way ANOVA on Ranks,
post-hoc Dunn’s methods (C); *P < 0.05; **P < 0.01; ***P < 0.001.

P < 0.001; ARS, t = 1.59, P = 0.133; Vanillin, t = 0.368, Olfactory stimulation with multiple
P = 0.718; Limonene, t = 1.112, P = 0.284; Green Odor,
t = 0.323, P = 0.750, Fig. 3C). Of note, animals exposed to
odorants halts the effects of ARS on
ARS along with multiple odorants showed a full recovery in spatial memory consolidation
both the preference and discrimination indexes (D.I. Next, we employed the same paradigm to evaluate the im­
ARS-M.O. 0.31 ± 0.05 post-hoc Dunn’s Method: ARS versus pact of olfactory stimulation on spatial memory consolida­
ARS-M.O.: P = 0.003; P.I. New versus P.I. Old: ARS-M.O. tion under stressful conditions. Specifically, animals were
65.3 ± 2.3% versus 34.7 ± 2.3%; Student’s t-test, subjected to ARS and olfactory stimulation since the end of
ARS-M.O., t = 9.673, P < 0.001, Fig. 3B and C). Finally, total OPR training phase for 2 h, which is considered the time
exploration was not altered by either olfactory stimulation or for the consolidation process,28 and they were tested 24 h la­
stress exposure, with the only significant difference observed ter (Fig. 4A). Again, ARS-exposed animals, as well as
between Limonene- and Green Odor ARS-exposed animals ARS-Limonene- and ARS-Green Odor exposed animals,
(Table 1, One-way ANOVA, F(5,45) = 2.741, P = 0.030; showed spatial memory impairment, as revealed by a signifi­
post-hoc Holm-Sidak method: Limonene versus Green Odor: cant reduction of the discrimination index (DI) compared to
P = 0.034). CTRL animals (D.I.: CTRL 0.38 ± 0.05; ARS 0.02 ± 0.5;
8 | BRAIN COMMUNICATIONS 2024, fcae390 B. Bandiera et al.

Downloaded from https://academic.oup.com/braincomms/article/6/6/fcae390/7876424 by guest on 16 July 2025

Figure 3 Stress induced recognition memory deficits are prevented by exposure to multiple odorants. (A) Schematic
representations of the experimental protocol. (B) Analysis of the discrimination index of the test phase reveals a significant difference between
CTRL and acute restraint stress (ARS) exposed animals and between ARS and ARS-M.O. exposed animals, suggesting functional recovery. (C)
Preference index for the novel (P.I. New, green) and old (P.I. Old, orange) objects. Only CTRL and ARS-M.O. animals spent a significantly higher
proportion of time exploring the novel compared to the already known object (CTRL n = 9; ARS n = 8; ARS + M.O. n = 10; ARS + Vanillin n = 8;
ARS + Limonene n = 8; ARS + Green Odor n = 8). All data are reported as mean ± s.e.m. Dots represent the number of samples (studied animals).
One-Way ANOVA on Ranks followed by post-hoc Dunn’s method (B), Student’s t test (C); **P < 0.01; ***P < 0.001.

Table 1 Exploration times in the test phase of the NOR and OPR tests

CTRL ARS ARS + M.O. ARS + Vanillin ARS + Limonene ARS + Green Odor
NOR 19.3 ± 1.7 s 16.1 ± 1.4 s 15.0 ± 2 s 15.2 ± 2.9 s 12.3 ± 1.3 s 22.3 ± 3.1 s
OPR 15.1 ± 1.7 s 10.6 ± 1.1 s 14.4 ± 2 s 11.6 ± 1.3 s 14.3 ± 2.4 s 14.9 ± 1 s

ARS-Limonene −0.01 ± 0.06; ARS-Green Odor −0.02 ± versus 48.7 ± 2.4%; ARS-Limonene 49.5 ± 3.1% versus
0.05; One-way ANOVA, F(5,55) = 7.189, P < 0.001; post- 50.4 ± 3.1%; ARS-Green Odor 49.0 ± 2.5% versus 51.0 ±
hoc Holm-Sidak method: CTRL versus ARS P = 0.007; 2.5%; Student’s t-test, P.I. Displaced versus P.I. Stationary:
CTRL versus ARS-Limonene P = 0.003; CTRL versus CTRL, t = 10.786, P < 0.001; ARS, t = 0.762, P = 0.456;
ARS-Green Odor P = 0.004, Fig. 4B), further confirmed by ARS-Limonene, t = 0.224, P = 0.825; ARS-Green Odor,
the analysis of the P.I. for the displaced and stationary ob­ t = 0.57, P = 0.577, Fig. 4C). Of note, when analyzing the
jects (P.I. Displaced versus P.I. Stationary, respectively: P.I. for the displaced and stationary objects, but not the
CTRL 68.9 ± 2.6% versus 31.1 ± 2.6%; ARS 51.2 ± 2.4% D.I., ARS-Vanillin exposed animals spent significantly
Odorants modulate inflammatory signaling BRAIN COMMUNICATIONS 2024, fcae390 | 9

Downloaded from https://academic.oup.com/braincomms/article/6/6/fcae390/7876424 by guest on 16 July 2025

Figure 4 Stress induced recognition memory deficits are prevented by exposure to multiple odorants. (A) Schematic
representations of the experimental protocol. (B) CTRL and acute restraint stress (ARS) M.O. exposed animals had a significantly higher
discrimination index compared to ARS-exposed exposed mice. (C) Analysis of the P.I.s for the displaced (P.I. Displaced, green) and stationary (P.I.
Stationary, orange) objects revealed that CTRL, ARS-M.O. and ARS-Vanillin mice spent a significantly higher proportion of time exploring the
displaced object (CTRL n = 9; ARS n = 9; ARS + M.O. n = 11; ARS + Vanillin n = 12; ARS + Limonene n = 8; ARS + Green Odor n = 8). All data are
reported as mean ± s.e.m. Dots represent the number of samples (studied animals). One-Way ANOVA followed by post-hoc Holm-Sidak’s
method (B), Student’s t test (C); **P < 0.01; ***P < 0.001.

higher time exploring the displaced object compared to the Olfactory stimulation counteracts the
stationary object, suggesting memory recovery (D.I.
ARS-Vanillin 0.16 ± 0.08; P.I. Displaced versus Stationary:
unpredictable chronic stress-related
ARS-Vanillin 57.7 ± 4% versus 42.2 ± 4%; t = 2.851, P = behavioral alterations
0.009). In agreement with the results of the NOR test, Next, we tested whether the combination of the studied odor­
ARS-M.O. animals showed an increased preference for the ants prevented mood alterations induced by a chronic unpre­
displaced object compared to the stationary one, which dictable stress paradigm to evaluate a possible translational
was not observed in ARS animals, as well as a significant in­ efficacy of said mixture against the long-term consequences
crease in the D.I. compared to ARS-exposed animals (D.I. of stress. First, we characterized our CUS model at behavior­
ARS-M.O. 0.29 ± 0.05, ARS versus ARS-M.O. P = 0.027; al, molecular and cellular levels. Six-weeks of CUS protocol
P.I. Displaced versus Stationary: ARS-M.O. 64.4 ± 2.5% induced a significant increase in the immobility time of
versus 35.5 ± 2.5%; t = 8.413, P < 0.001). Finally, total ex­ CUS-exposed mice compared to controls in the forced swim­
ploration was not altered by olfactory stimulation nor by ming test (192 ± 11 s versus 134 ± 16 s respectively,
stress exposure (Table 1, One-way ANOVA, F(5,55) = Student’s t test, t = 3.158, P = 0.006, Fig. 5A), as well as a sig­
1.480, P = 0.214). nificant reduction of time spent performing self-grooming in
10 | BRAIN COMMUNICATIONS 2024, fcae390 B. Bandiera et al.

Downloaded from https://academic.oup.com/braincomms/article/6/6/fcae390/7876424 by guest on 16 July 2025

Figure 5 CUS-exposed animals show behavioral, molecular, and biochemical hallmarks of chronic stress response. Animals were
subjected to 6 weeks of chronic unpredictable stress (CUS). (A) and (B) At the end of the 6 weeks period, CUS-exposed animals showed a
significant decrease in the time spent performing self-grooming in the SST (A) and a significant increase in time of immobility in the FST (B, n = 8 for
both groups). (C, D and E) Representative blots showing alterations in expression and phosphorylation of different molecular substrates.
CUS-exposed animals show reduced PSD-95 levels (n = 4 control mice, n = 8 CUS mice), as well as decreased GSK3β inhibitory phosphorylation
on Ser9 (n = 3 control mice, n = 6 CUS mice). (G) IL 1β and BDNF levels after CUS-exposure. Depressed mice show an increase in IL 1β in the PFC
as well as a decrease in BDNF levels (n = 3 control mice, n = 3 CUS mice). (H and I) Representative images and quantification of the total number
of proliferating neural stem cells in the hippocampus of both CTRL and CUS-exposed animals (see ‘Materials and Methods’ section for the
quantification method; n = 3 control mice, n = 4 CUS mice; Scale bar: 100 µm); sgz: subgranular zone. Dots represent the number of samples
(studied animals). Student’s t test was performed for all comparisons. All data are reported as mean ± s.e.m. *P < 0.05; **P < 0.01, ***P < 0.001.
See supplementary materials for uncropped blots.
Odorants modulate inflammatory signaling BRAIN COMMUNICATIONS 2024, fcae390 | 11

Downloaded from https://academic.oup.com/braincomms/article/6/6/fcae390/7876424 by guest on 16 July 2025

Figure 6 Olfactory stimulation prevents the development of a depressive phenotype in CUS-exposed mice. (A) Schematic
representation of the experimental protocol. Animals were divided into four groups: 1) control unstressed mice exposed to vehicle stimulation; 2)
control unstressed mice exposed to multiple odorants stimulation; 3) CUS-exposed mice treated with vehicle stimulation; 4) CUS-exposed mice
treated with multiple odorants stimulation. (B) Time spent in the grooming during the sucrose splash test for CTRL + vehicle, CTRL + M.O.,
CUS + vehicle and CUS + M.O. groups (CTRL + vehicle n = 13; CTRL + M.O. n = 10; CUS + vehicle n = 17; CUS + M.O. n = 19). (C) Time of
immobility during the forced swim test for CTRL, CUS + Vehicle and CUS + M.O. groups (CTRL + vehicle n = 12; CTRL + M.O. n = 10; CUS +
vehicle n = 16; CUS + M.O. n = 17). (D) IL 1β levels in the PFC of CTRL + vehicle, CUS + vehicle and CUS + M.O. stimulation (n = 3 each group).
Dots represent the number of samples (studied animals). Two-way ANOVA followed by post-hoc Holm-Sidak method (B and (C); One-way
ANOVA followed by post-hoc Holm-Sidak method (D). All data are reported as mean ± s.e.m. *P < 0.05; **P < 0.01.

the sucrose splash test (114 ± 8 s versus 162 ± 7 s, Student’s t test, two-way ANOVA revealed no significant effect of
test, t = 4.439, P < 0.001, Fig. 5B). Both behavioral changes CUS- or M.O.-exposure (Fig. 6B). Instead, a significant inter­
did not depend on altered locomotor activity, as no difference action CUS × M.O. was observed (Two-way ANOVA, F =
between total distance traveled was recorded in the open field 5.552, P = 0.022). Post-hoc tests revealed that mice exposed
test (Supplementary Fig. 1). Animals were then sacrificed for to both CUS and vehicle sprays had a significant reduction
molecular and immunofluorescence analyses. CUS-treated of their self-grooming activity compared to control animals
animals showed in the PFC several characteristic molecular as well as compared to animals exposed to both CUS and ol­
alterations of chronic stress exposure22,29 including a signifi­ factory stimulation, which recovered to the level of control
cant reduction of PSD-95 levels (38% reduction in animals (CTRL + vehicle: 162.1 ± 8.6 s, CTRL + M.O.:
CUS-exposed animals compared to control mice, Student’s 151.5 ± 15.6 s, CUS + vehicle: 121.7 ± 7.7 s, CUS + M.O.:
t test, t = 3.240, P = 0.010, Fig. 5C and D) and a decrease 159 ± 10.1 s; post-hoc Holm-Sidak method: CTRL versus
of GSK3β serine 9 phosphorylation (81% reduction in CUS-vehicle: P = 0.005; CUS-vehicle versus CUS-M.O.:
CUS-exposed mice compared to control animals, Student’s P = 0.005). Similar results were obtained in the forced swim
t test, t = 3.831, P = 0.006, Fig. 5C and E). A significant re­ test, where an effect of the interaction between stress
duction in BDNF levels as well as a significant increase in IL and M.O. was observed (Two-way ANOVA, F = 4.635,
1β concentration were also observed in the PFC of P = 0.036), as immobility time significantly increased in ve­
CUS-exposed mice (BDNF: 50.7 ± 7.1 pg/mg versus 150.7 hicle stimulated-CUS animals compared to both control
± 13.2 pg/mg, Student’s t test, t = 5.771, P = 0.004, Fig. 5F; and olfactory stimulated-CUS animals (CTRL + vehicle:
IL 1β: 36 ± 5.7 pg/mg versus 10 ± 1.8 pg/mg, Student’s 141.1 ± 17.2 s, CTRL + M.O.: 143.2 ± 15.9 s, CUS + ve­
t test, t = 3.782, P = 0.019, Fig. 5G). Furthermore, analysis hicle: 188.7 ± 6.2 s, CUS + M.O.: 135.8 ± 13.2 s, post-hoc
of CUS hippocampi also revealed a reduction of neural stem Dunn’s Method: CTRL versus CUS-vehicle: P = 0.010;
cells proliferation compared to control mice (3601 ± 441 ver­ CUS-vehicle versus CUS-M.O.: P = 0.002, Fig. 5C). No dif­
sus 4579 ± 225 BrdU positive cells, Student’s t test, t = 2.711, ferences were observed between control-vehicle and
P = 0.042, Fig. 5H and I). Subsequently, we evaluated control-M.O. mice in both tests. Finally, we investigated if
whether the multiple odorants stimulation rescued behavior­ the M.O. stimulation also reverted one of the molecular
al alterations in CUS-exposed animals. In the sucrose splash changes potentially involved in the development of
12 | BRAIN COMMUNICATIONS 2024, fcae390 B. Bandiera et al.

CUS-induced depressed phenotype. Indeed, IL 1β levels in the associated with a depressive phenotype. When animals
PFC were significantly reduced by M.O. exposure compared were exposed to all the odorants, they showed no depressive
to CUS-vehicle animals, although they still were significantly behaviors in both the sucrose splash test and in the forced
higher compared to control animals (One-way ANOVA, swim test, thus confirming that the selected mixture also
F(2,8) = 21.189; P = 0.002; CTRL + vehicle: 10 ± 1.7 pg/mg; holds a prophylactic efficacy not only against the short-term
CUS + vehicle: 30.6 ± 2.9 pg/mg; CUS + M.O.: 18.4 ± 0.8 consequences of acute stress, but also toward long-term
pg/mg; post-hoc Holm-Sidak method: CTRL + vehicle versus stress-induced alterations. Furthermore, a representative
CUS + vehicle: P = 0.002; CUS-vehicle versus CUS-M.O.: P stress-induced molecular change such as IL 1β levels was re­
= 0.018; CTRL + vehicle versus CUS + M.O.: P = 0.038, verted in the PFC of CUS-M.O. exposed animals compared
Fig. 6D). to the CUS-vehicle group, therefore pointing toward a poten­
tial role of odorants stimulation in counteracting stress-
dependent alterations including inflammatory signaling.

Downloaded from https://academic.oup.com/braincomms/article/6/6/fcae390/7876424 by guest on 16 July 2025

Our data add a novel layer of knowledge about the effects
Discussion of olfactory stimulation on limbic system activity and stress-
Several reports have shown that vanillin, limonene, or green dependent behavioral alterations. Compared to previous
odor molecules, when either inhaled or orally/intraperitone­ works, some discrepancies may arise and three important
ally administered, induce beneficial effects on brain function. distinctions must be defined in the context of literature re­
It has been reported that green odor exposure can increase ports: i) whether the stress is present or not; ii) stress duration
the time spent in open arms of the EPM, and both dopamine and iii) odorant concentration. Some works have evaluated
and serotonin levels in the brain in non-stressed animals.30-32 the effect of olfactory stimulation under basal condition
It can also reduce hypothalamic parvalbumin positive neu­ (i.e. no stress), reporting a positive effect for single odorants
rons activity and adrenocorticotropic hormone release33,34 on anxious behavior which doesn’t automatically
as well as inducing a different activity pattern in limbic sys­ translates into a single odorant being beneficial in halting
tem regions.35 Limonene inhalation, both as single molecule stress-induced anxiety. Stress duration may also represent
and in essential oils obtained from bergamot or orange, has an important variable in understanding how olfactory re­
been reported as well to reduce anxious behaviors under no sponses shape stress effects. For example, Lee et al. has de­
stress conditions36-38 or to reduce depressive behaviors in monstrated that exposure to a single odorant, such as
chronic stress mouse model.17 Finally, vanillin has been 2-phenylethanol or hinokitiol, can reduce behavioral altera­
shown to impact mood alterations in a model of depression20 tions induced by different types of stress, through the regula­
but it failed to show an anxiolytic effect under no stress con­ tion of corticotropin releasing hormone neurons in the
dition in different tests, such as the EPM, the open field test hypothalamus.10 However, the restraint stress protocol
and the FST.39 In this work, we report that olfactory stimu­ they used lasted 10 min, which may induce a milder response
lation with a combination of multiple odorants applied dur­ compared to a 2 h stress paradigm which could not allow for
ing stressful conditions counteracts the detrimental outcome a recovery through single odorant exposure. Finally, odorant
of stress exposure on behavior, whereas single odorant concentration also plays a role. For example, it has been re­
stimulation is either ineffective or scarcely effective in an ex­ ported that the effect of bergamot essential oil on anxious be­
perimental model of acute stress. Furthermore, we show that havior can be modulated by the odorant concentration, with
the selected mix of odorants prevents the development of de­ more behavioral parameters being modified by higher con­
pressive symptoms in a mouse model of chronic stress. centrations of the volatile compounds.37 Again, other re­
In the ARS model, we observed biochemical, molecular ports have shown a dose-dependent efficacy of limonene
and behavioral alterations that are typical of the acute stress containing mixtures on behavioral parameters.36 In this
response. Therefore, we sought to determine which odorant work, we employed low concentrations, which were calcu­
and/or whether a combination of the three could better halt lated based on our previous studies40 to evaluate if the com­
stress-induced alterations. In the EPM, the stimulation with bination of different odorants per se could still produce an
multiple odorants was able to induce a functional recovery effect on stress response.
on the different parameters recorded, which was not ob­ We do not claim that the specific mixture of odorants we
tained when using any of the single odorants. In the NOR employed is the only one effective in counteracting the effects
and OPR tests, stimulation with multiple odorants induced of acute and chronic stress. It is possible that other combina­
a full recovery of recognition and spatial memory. Of note, tions of multiple odorants may exert similar additive or syn­
vanillin was the only odorant that was able to induce a sig­ ergistic actions, resulting in even a greater effect. For
nificant recovery of function in the spatial memory test. example, measuring the effect size of our olfactory stimula­
Overall, exploratory activity of the objects was reduced by tion between CUS-vehicle and CUS-M.O. mice in the FST re­
the stress paradigm and, in some cases, such as green odor veals a Cohen’s d value of 1.08. A recent meta-analysis on the
and limonene stimulation in the NOR test, the single odor­ effect size of classical antidepressant drugs evaluated in
ants were able to induce an increase in the exploratory func­ chronic stress models in the FST revealed an average effect
tion. In the chronic stress paradigm, we observed behavioral, size of 2.44.41 While we did not add a positive control group
cellular and molecular alterations previously shown to be (i.e. a group treated with antidepressants) for a more
Odorants modulate inflammatory signaling BRAIN COMMUNICATIONS 2024, fcae390 | 13

accurate evaluation of the different effect sizes, a comparison simple mixture of three odorants, can prevent depressive be­
with the above cited results suggests that the proposed mix­ havior induced by chronic stress. Furthermore, our results
ture can still be improved to match standard antidepressant point toward the existence of a tight connection between ol­
therapies, by adding other odorants in the stimulation proto­ factory function and the regulation of limbic inflammatory
col or employing other types of odorants. Indeed, stimula­ signaling. Our data underline the possibility that the use dif­
tion by different odorants may overcome the adaptation of ferent mixtures of odorants may be more effective in treating
olfactory receptors thus triggering a significantly higher sen­ stress-induced conditions than exposure to a single odorant.
sory input through the olfactory nerve. Olfactory habitu­ Further experiments will be necessary to obtain a thorough
ation is a well-known phenomenon in which an odorant is insight on the mechanisms of action of olfactory stimulation,
not perceived after prolonged exposure, due to the desensi­ to better characterize which brain areas are more affected by
tization of its receptors.42 The stimulation with a single such treatments and how they can be exploited in clinical
odorant may therefore induce the deactivation of the specific settings.

Downloaded from https://academic.oup.com/braincomms/article/6/6/fcae390/7876424 by guest on 16 July 2025

receptor, leading to reduced olfactory stimulation. On the
other hand, alternatively using multiple odorants could
maintain a higher stimulation level compared to single Supplementary material
stimulation. Nonetheless, each odorant may exert a specific
action thus inducing a different physiological effect. For in­ Supplementary material is available at Brain Communications
stance, limonene inhalation, but not Melaleuca alternifolia online.
essential oils, have been shown to impact on anxious behav­
ioral profiles.38 Here we report that vanillin prevents the det­
rimental effect of stress on spatial memory consolidation. Acknowledgements
This effect may be explained by vanillin affinity for the tran­
sient receptor potential vanilloid 1,43 which is involved in The authors would like to acknowledge the contribution of
pain perception and in memory storage and is thoroughly ex­ Experimental Models Research Core Facility G-STeP of
pressed in the hippocampus.44,45 Indeed, it is possible that Fondazione Policlinico Universitario ‘A. Gemelli’ IRCCS,
odor stimulation exerts its action either by stimulation of Ministero della Salute - Ricerca Corrente 2024 Fondazione
the olfactory pathway leading to the regulation of limbic re­ Policlinico Universitario ‘A. Gemelli’ IRCCS. The graphical
gions activity or by absorption of volatile molecules in the abstract, as well as Fig. 2A and 3A and 4A and 6A were cre­
nasal cavity through the nasal mucosa, which can then enter ated using BioRender.com. Regione Lazio and Sensosan s.r.l.
the blood flow to reach and bind specific receptors. co-funded Dr. Bandiera‘s PhD fellowship. Ministero
Neuroinflammation is one of the multiple pathways in­ dell‘Università e della Ricerca funded Dr. Rinaudo‘s
volved in the etiopathogenesis of stress-induced mood disor­ Assistant Professor fellowship through the PNRR-PON
ders, as also shown by our characterization of the (Piano Nazionale di Ripresa e Resilienza-Programma
CUS-model. Among the multiple pro-inflammatory cyto­ Operativo Nazionale) programs, with the support of
kines, IL 1β is at the crossroad of different signaling cascades Sensosan s.r.l.
regulating learning and memory, neurogenesis, glial reactiv­
ity, neuronal excitability etc., both in physiological and
pathological conditions. A recent work has demonstrated Funding
that lavender essential oils can reduce the levels of multiple
The authors report no funding for the present research.
pro-inflammatory cytokines, such as IL 1β and tumor necro­
sis factor α, in the central amygdala of a mouse model of vis­
ceral pain.46 These data are in agreement with our findings,
as we also observed a reduction of IL 1β in CUS-M.O. ex­ Competing interest
posed mice compared to CUS-vehicle mice, although in a dif­ The authors declare no competing interests.
ferent brain area. Given its importance, we focused our
attention on IL 1β but we cannot exclude the role of other
pro-inflammatory molecules nor the role of other molecular
pathways mediating the effects of M.O. stimulation. Indeed,
Data availability
M.O. stimulation induces only a partial recovery of Data are available on request to the corresponding author.
PFC IL 1β levels, being significantly lower compared to
CUS-vehicle animals yet higher than control animals, thus
suggesting that M.O. stimulation involves other mechanisms
beyond IL 1β reduction.
References
1. Herman JP. Neural control of chronic stress adaptation. Front
Collectively, our data suggest that, even in acute condi­
Behav Neurosci. 2013;7:61.
tions, olfactory stimulation using a mixture of different 2. Herman JP, McKlveen JM, Ghosal S, et al. Regulation of
odorants can have a prophylactic action on the detrimental the hypothalamic-pituitary-adrenocortical stress response. In:
consequences of an acute stressor and show that, even a Comprehensive physiology. Wiley; 2016:603-621.
14 | BRAIN COMMUNICATIONS 2024, fcae390 B. Bandiera et al.

3. Kendler KS, Kessler RC, Walters EE, et al. Stressful life events, gen­ infected mice by downregulating the expression of synaptic
etic liability, and onset of an episode of major depression in women. plasticity-related genes via the epigenetic MeCP2/HDAC4 complex.
Am J Psychiatry. 1995;152(6):833-842. Cell Mol Life Sci. 2023;80(6):172.
4. Soudry Y, Lemogne C, Malinvaud D, Consoli SM, Bonfils P. 25. Cocco S, Rinaudo M, Fusco S, et al. Plasma BDNF levels following
Olfactory system and emotion: Common substrates. Eur Ann transcranial direct current stimulation allow prediction of synaptic
Otorhinolaryngol Head Neck Dis. 2011;128(1):18-23. plasticity and memory deficits in 3×Tg-AD mice. Front Cell Dev
5. Blazing RM, Franks KM. Odor coding in piriform cortex: Biol. 2020;8:541.
Mechanistic insights into distributed coding. Curr Opin 26. Jaszczyk A, Stankiewicz AM, Juszczak GR. Dissection of mouse
Neurobiol. 2020;64:96-102. hippocampus with its dorsal, intermediate and ventral subdivisions
6. Murata K, Kinoshita T, Fukazawa Y, et al. GABAergic neurons in combined with molecular validation. Brain Sci. 2022;12(6):799.
the olfactory cortex projecting to the lateral hypothalamus in 27. Natale F, Leone L, Rinaudo M, et al. Neural stem cell-derived extra­
mice. Sci Rep. 2019;9(1):7132. cellular vesicles counteract insulin resistance-induced senescence of
7. Strauch C, Manahan-Vaughan D. Orchestration of hippocampal in­ neurogenic niche. Stem Cells. 2022;40(3):318-331.
formation encoding by the piriform Cortex. Cerebral Cortex. 2020; 28. Villar ME, Martinez MC, da Cunha PL, Ballarini F, Viola H.
30(1):135-147. Memory consolidation and expression of object recognition are sus­

Downloaded from https://academic.oup.com/braincomms/article/6/6/fcae390/7876424 by guest on 16 July 2025

8. Root CM, Denny CA, Hen R, Axel R. The participation of cortical ceptible to retroactive interference. Neurobiol Learn Mem. 2017;
amygdala in innate, odour-driven behaviour. Nature. 2014; 138:198-205.
515(7526):269-273. 29. Cohen JW, Louneva N, Han L, et al. Chronic corticosterone exposure
9. Aqrabawi AJ, Kim JC. Hippocampal projections to the anterior ol­ alters postsynaptic protein levels of PSD-95, NR1, and synaptopodin
factory nucleus differentially convey spatiotemporal information in the mouse brain. Synapse. 2011;65(8):763-770.
during episodic odour memory. Nat Commun. 2018;9(1):2735. 30. Tokumo K, Tamura N, Hirai T, Nishio H. Effects of (Z)-3-hexenol, a
10. Lee EJ, Saraiva LR, Hanchate NK, et al. Odor blocking of stress hor­ major component of green odor, on anxiety-related behavior of the
mone responses. Sci Rep. 2022;12(1):8773. mouse in an elevated plus-maze test and biogenic amines and their
11. Hedigan F, Sheridan H, Sasse A. Benefit of inhalation aromatherapy metabolites in the brain. Behav Brain Res. 2006;166(2):247-252.
as a complementary treatment for stress and anxiety in a clinical set­ 31. Kako H, Kobayashi Y, Yokogoshi H. Effects of n-hexanal on dopa­
ting – A systematic review. Complement Ther Clin Pract. 2023;52: mine release in the striatum of living rats. Eur J Pharmacol. 2011;
101750. 651(1–3):77-82.
12. Sattayakhom A, Wichit S, Koomhin P. The effects of essential oils on 32. Nakatomi Y, Yokoyama C, Kinoshita S, et al. Serotonergic medi­
the nervous system: A scoping review. Molecules. 2023;28(9):3771. ation of the antidepressant-like effect of the green leaves odor in
13. Masuo Y, Satou T, Takemoto H, Koike K. Smell and stress response mice. Neurosci Lett. 2008;436(2):167-170.
in the brain: Review of the connection between chemistry and 33. Kim J. Effects of green odor on expression of fos-immunoreactivity
neuropharmacology. Molecules. 2021;26(9):2571. in the paraventricular nucleus of the thalamus in forced swimming
14. Yoto A, Fukui N, Kaneda C, et al. Black tea aroma inhibited in­ rats. Chem Senses. 2005;30(Supplement 1):i266-i267.
crease of salivary chromogranin-A after arithmetic tasks. J Physiol 34. Nikaido Y, Miyata S, Nakashima T. Mixture of cis-3-hexenol and
Anthropol. 2018;37(1):3. trans-2-hexenal attenuates behavioral and stress responses induced
15. Atsumi T, Tonosaki K. Smelling lavender and rosemary increases by 2,5-dihydro-2,4,5-trimethylthiazoline and electric footshock
free radical scavenging activity and decreases cortisol level in saliva. stress in rats. Physiol Behav. 2011;103(5):547-556.
Psychiatry Res. 2007;150(1):89-96. 35. Nikaido Y, Nakashima T. Different patterns of neuronal activities
16. Hashikawa-Hobara N, Otsuka A, Ishikawa R, Hashikawa N. in the infralimbic and prelimbic cortices and behavioral expression
Roman chamomile inhalation combined with clomipramine treat­ in response to two affective odors, 2,5-dihydro-2,4,5-
ment improves treatment-resistant depression-like behavior in trimethylthiazoline and a mixture of cis-3-hexenol and trans-2-
mice. Biomed Pharmacother. 2019;118:109263. hexenal, in the freely moving rat. Behav Brain Res. 2011;218(1):
17. Zhang LL, Yang ZY, Fan G, Ren JN, Yin KJ, Pan SY. 218-227.
Antidepressant-like effect of Citrus sinensis (L.) osbeck essential 36. Lima NGPB, De Sousa DP, Pimenta FCF, et al. Anxiolytic-like activ­
oil and its main component limonene on mice. J Agric Food ity and GC–MS analysis of (R)-(+)-limonene fragrance, a natural
Chem. 2019;67(50):13817-13828. compound found in foods and plants. Pharmacol Biochem Behav.
18. Kasuya H, Okada N, Kubohara M, Satou T, Masuo Y, Koike K. 2013;103(3):450-454.
Expression of BDNF and TH mRNA in the brain following inhaled 37. Saiyudthong S, Marsden CA. Acute effects of bergamot oil on
administration of α-pinene. Phytother Res. 2015;29(1):43-47. anxiety-related behaviour and corticosterone level in rats.
19. Nikaido Y, Yamada J, Migita K, et al. cis-3-Hexenol and trans-2- Phytother Res. 2011;25(6):858-862.
hexenal mixture prevents development of PTSD-like phenotype in 38. Faturi CB, Leite JR, Alves PB, Canton AC, Teixeira-Silva F.
rats. Behav Brain Res. 2016;297:251-258. Anxiolytic-like effect of sweet orange aroma in Wistar rats. Prog
20. Xu J, Xu H, Liu Y, He H, Li G. Vanillin-induced amelioration of Neuropsychopharmacol Biol Psychiatry. 2010;34(4):605-609.
depression-like behaviors in rats by modulating monoamine neuro­ 39. Ueno H, Shimada A, Suemitsu S, et al. Comprehensive behavioral
transmitters in the brain. Psychiatry Res. 2015;225(3):509-514. study of the effects of vanillin inhalation in mice. Biomed
21. Torrisi SA, Rizzo S, Laudani S, Ieraci A, Drago F, Leggio GM. Acute Pharmacother. 2019;115:108879.
stress alters recognition memory and AMPA/NMDA receptor sub­ 40. Mastrodonato A, Barbati SA, Leone L, et al. Olfactory memory is
units in a sex-dependent manner. Neurobiol Stress. 2023;25: enhanced in mice exposed to extremely low-frequency electromag­
100545. netic fields via Wnt/β-catenin dependent modulation of subventricu­
22. Aceto G, Colussi C, Leone L, et al. Chronic mild stress alters synap­ lar zone neurogenesis. Sci Rep. 2018;8(1):262.
tic plasticity in the nucleus accumbens through GSK3β-dependent 41. Smalheiser NR, Graetz EE, Yu Z, Wang J. Effect size, sample size
modulation of Kv4.2 channels. Proceedings of the National and power of forced swim test assays in mice: Guidelines for inves­
Academy of Sciences. 2020;117(14):8143-8153. tigators to optimize reproducibility. PLoS One. 2021;16(2):
23. Spinelli M, Natale F, Rinaudo M, et al. Neural stem cell-derived e0243668.
exosomes revert HFD-dependent memory impairment via 42. Kim K, Bae J, Jin Y, Moon C. Odor habituation can modulate very early
CREB-BDNF signalling. Int J Mol Sci. 2020;21(23):8994. olfactory event-related potential. Sci Rep. 2020;10(1):18117.
24. Li Puma DD, Colussi C, Bandiera B, et al. Interleukin 1β triggers 43. Jayant S, Sharma BM, Sharma B. Protective effect of transient recep­
synaptic and memory deficits in Herpes simplex virus type-1- tor potential vanilloid subtype 1 (TRPV1) modulator, against
Odorants modulate inflammatory signaling BRAIN COMMUNICATIONS 2024, fcae390 | 15

behavioral, biochemical and structural damage in experimental 45. Iglesias LP, Fernandes HB, de Miranda AS, et al. TRPV1 modula­
models of Alzheimer’s disease. Brain Res. 2016;1642:397-408. tion of contextual fear memory depends on stimulus intensity and
44. Bashiri H, Hosseini-Chegeni H, Sharifi KA, Sahebgharani M, endocannabinoid signalling in the dorsal hippocampus.
SalariAA. Activation of TRPV1 receptors affects memory Neuropharmacology. 2023;224:109314.
function and hippocampal TRPV1 and CREB mRNA expression 46. Liao WC, Yao RA, Chen LY, et al. Olfactory stimulation successful­
in a rat model of biliary cirrhosis. Neurol Res. 2018;40(11): ly modulates the neurochemical, biochemical and behavioral pheno­
938-947. types of the visceral pain. Molecules. 2022;27(21):7659.

Downloaded from https://academic.oup.com/braincomms/article/6/6/fcae390/7876424 by guest on 16 July 2025