Exposure to Chemical Cues from Predator-Exposed Conspecifics Increases Reproduction in a Wild Rodent

Abstract

Abstract Predation involves more than just predators consuming prey. Indirect effects, such as fear responses caused by predator presence, can have consequences for prey life history. Laboratory experiments have shown that some rodents can recognize fear in conspecifics via alarm pheromones. Individuals exposed to alarm pheromones can exhibit behavioural alterations that are similar to those displayed by predator-exposed individuals. Yet the ecological and evolutionary significance of alarm pheromones in wild mammals remains unclear. We investigated how alarm pheromones affect the behaviour and fitness of wild bank voles (Myodes glareolus) in outdoor enclosures. Specifically, we compared the effects of exposure of voles living in a natural environment to a second-hand fear cue, bedding material used by predator-exposed voles. Control animals were exposed to bedding used by voles with no predator experience. We found a ca. 50% increase in litter size in the group exposed to the predator cue. Furthermore, female voles were attracted to and males were repelled by trap-associated bedding that had been used by predator-exposed voles. Movement and foraging were not significantly affected by the treatment. Our results suggest that predation risk can exert population-level effects through alarm pheromones on prey individuals that did not encounter a direct predator cue.

Full text

1
Scientific REPORtS | (2018) 8:17214 | DOI:10.1038/s41598-018-35568-0
www.nature.com/scientificreports
Exposure to Chemical Cues from
Predator-Exposed Conspecics
Increases Reproduction in a Wild
Rodent
M. Haapakoski
1, A. A. Hardenbol2,3 & Kevin D. Matson2
Predation involves more than just predators consuming prey. Indirect eects, such as fear responses
caused by predator presence, can have consequences for prey life history. Laboratory experiments
have shown that some rodents can recognize fear in conspecics via alarm pheromones. Individuals
exposed to alarm pheromones can exhibit behavioural alterations that are similar to those displayed
by predator-exposed individuals. Yet the ecological and evolutionary signicance of alarm pheromones
in wild mammals remains unclear. We investigated how alarm pheromones aect the behaviour and
tness of wild bank voles (Myodes glareolus) in outdoor enclosures. Specically, we compared the
eects of exposure of voles living in a natural environment to a second-hand fear cue, bedding material
used by predator-exposed voles. Control animals were exposed to bedding used by voles with no
predator experience. We found a ca. 50% increase in litter size in the group exposed to the predator cue.
Furthermore, female voles were attracted to and males were repelled by trap-associated bedding that
had been used by predator-exposed voles. Movement and foraging were not signicantly aected by
the treatment. Our results suggest that predation risk can exert population-level eects through alarm
pheromones on prey individuals that did not encounter a direct predator cue.
In 1961, Robert Ardrey wrote, “sex is a sideshow in the world of the animal, for the dominant color of that world
is fear”1. Of course, the purveyors of fear in the world of the animal are predators, and their ultimate threat comes
in the form of killing and eating prey. However, predation involves more than just predators consuming prey2,3.
In fact, when facing only the risk of predation, prey can exhibit an array of anti-predatory responses, which have
arisen through co-evolution with predators4. Such awareness or perception of predators by prey can aect the
physiology, body condition, and reproduction in many prey species (reviewed by Lima 19982). At the population
level, the eects of fear of predation may be of the same magnitude or greater than the actual act of predation5.
While predation fear responses (sometimes placed under the umbrella of “stress”) are oen seen as negative or
costly for the prey, these responses can be adaptive6. For example, fear of predation can result in physiological
changes in pregnant females that prepare ospring for life under high predation pressure7.
Early predator detection is necessary for prey to maximize their own tness8. In many vertebrate predator-prey
systems, the key cue of predator presence is scent9. Prey can assess predation risk by eavesdropping on predator
scent marks le for intraspecic communication. An adaptive prey response requires recognition and correct
interpretation and response to the cue. While fresh cues may suggest predation is imminent and a change in
behaviour is needed, old cues should be ignored10. Behavioural responses by prey to avoid predation can include
reduced activity or escape to another, oen lower-quality, habitat11,12. ese responses can translate to changes in
foraging behaviours, decreased body condition13, and ultimately reduced tness. In females, poor body condition
can diminish ospring quantity and quality14; in males, predator avoidance during the mating season may lead
to missed mating opportunities15 or breeding delays. Hence, if one looks back at the world described by Andrey,
indirect interactions between the sideshow of sex and the world of fear are evident.
1Konnevesi Research Station, Department of Biological and Environmental Science, P.O. Box 35, FI-40014, University
of Jyväskylä, Jyväskylä, Finland. 2Resource Ecology Group, Environmental Sciences Department, Wageningen
University, Droevendaalsesteeg 3a, 6708 PB, Wageningen, The Netherlands. 3Present address: School of Forest
Sciences, University of Eastern Finland, P.O. Box 111, FI-80101, Joensuu, Finland. Correspondence and requests for
materials should be addressed to M.H. (email: marko.j.haapakoski@jyu.)
Received: 22 June 2018
Accepted: 1 November 2018
Published: xx xx xxxx
OPEN
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
2
Scientific REPORtS | (2018) 8:17214 | DOI:10.1038/s41598-018-35568-0
Many organisms, including vertebrates, invertebrates, and plants, respond to predation threats by producing
alarm signals. (reviewed by Verheggen et al.)16. For example, some rodents are known to produce alarm phero-
mones (AP), chemical cues that warn conspecics about possible dangers17,18. Neurobiological and psycholog-
ical experiments show that individuals can recognise stressed conspecics from the AP le behind following
a stressful situation19. e receivers of this indirect signal increase their vigilance and risk assessment behav-
iour, i.e., change their behaviour in ways that are similar to individuals that encountered a direct predator cue18.
Furthermore, murine AP is biochemically related to predator-produced scent cues common to most carnivores20;
both contain similar sulfur-containing volatiles21.
In general, a role for fear in ecology of animals is recognized22; however, the ecological and evolutionary sig-
nicance of AP as a fear cue in wild mammals remains unclear. To the best of our knowledge, only two studies
have investigated the role of AP in wild mammals. In an experimental indoor arena, Cabrera voles (Microtus
cabrerae) avoided areas with the scent marks of conspecics that experienced various predation risks23. ese
predation risks included handling (simulation of capture), audio playbacks of an avian predator, and visual con-
tact with a mammalian predator. In the same study, but under eld conditions, Cabrera voles tended to decrease
their activity near sites treated with AP. In another study, black-tailed deer (Odocoileus hemionus columbianus)
produced AP when disturbed or alarmed24. In the presence of AP, female black-tailed deer became more alert
and le the site more oen than in the presence of control odours, odourless air, or deer urine24. Most other AP
studies have used lab-strain rodents and unnatural stimuli, such as electric foot-shocks17,18.
Wild rodents are ideally suited for studying the ecology of predator cues. ese animals have a highly devel-
oped olfactory system, and odours play a key role in their behavioural decision-making25. We tested the hypoth-
esis that wild bank voles (Myodes glareolus), a model prey species in many studies of predator-prey interactions,
would show antipredator-like responses when exposed to AP11,26, since the murine AP shares structural simi-
larity with predator scent21. More specically, we made the following predictions. First, bank voles are expected
to decrease movement in response to AP11. Second, voles are expected to be more fearful and thus forage less
eciently in response to AP15,27. ird, these behavioural and ecological responses should lead to poorer body
condition, which in turn should lead to a smaller proportion of breeding females and a smaller litter sizes on
average26,28.
Results
Litter size was signicantly greater by 1.9 ± 0.7 pups in AP enclosures compared to control enclosures (p = 0.013;
Table1, Fig.1). Previous pregnancy and the interaction between treatment and previous pregnancy were not
signicant. None of these three terms, which also comprised the pregnancy model, accounted signicantly for
variation in the proportion of pregnant females (Table1). Treatment did not have a signicant eect on the female
survival (control = 55 ± 12%, treatment = 50 ± 10%; Table1).
e interaction between treatment and sex signicantly aected bait-less trapping (p = 0.019; Table1, Fig.2a).
Females were trapped more and males less in the presence of AP. Neither treatment, sex, nor their interaction
signicantly aected trappability (Table1). Only sex and the number of trapping instances signicantly aected
movement: females moved less than males (p = 0.012, females 70 ± 13 m2, males 177 ± 45 m2; Table1, Fig.2b)
and more trapping instances correlated with greater movement areas (p < 0.001; 63.48 ± 11.33 m2 per trapping
instance, Table1).
Litter size Pregnant
females Female survival
d.f. χ2Pχ2Pχ2P
Treatment*Pregnancy 1 2.31 0.129 1.36 0.243 — —
Treatment 1 6.23 0.013 0.55 0.456 0.11 0.741
Pregnancy 1<0.01 0.980 2.90 0.089 — —
Movement area Trappability Bait-less trapping
d.f.χ2Pχ2Pχ2P
Treatment*Sex 1 0.24 0.627 1.57 0.210 5.46 0.019
Treatment 1 0.11 0.740 0.87 0.350 — —
Sex 1 6.34 0.012 0.42 0.516 — —
Trappings 1 36.93 <0.001 x x x x
Giving Up Density
d.f.χ2P
Treatment*Time 1 2.16 0.142
Treatment 1 0.10 0.749
Time 1 0.34 0.561
Vegetation 1 67.51 <0.001
No. of Voles 1 0.41 0.520
Table 1. A detailed summary of statistical analysis with the χ2 and P values for all xed factors.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
3
Scientific REPORtS | (2018) 8:17214 | DOI:10.1038/s41598-018-35568-0
GUD was signicantly higher in areas where the vegetation was open compared to grass-covered areas (by
on average 38.8 ± 3.8 seeds, p < 0.001; Table1, Fig.3). None of the other model terms accounted signicantly for
variation in GUD.
Discussion
Conspecic alarm cues impact tness. In the current study, exposure to conspecic AP increased litter
size by about 50% compared to control treatment. is result supports our hypothesis that voles can receive dif-
ferent information from predator-exposed and unexposed male conspecics, and it suggests that female and male
voles can respond dierently to this information. Chemical cues can serve as an important means of communica-
tion among animals, both within and between species. Some rodents AP to warn conspecics about possible dan-
gers17,18,23. AP produced by mice are known to be molecularly similar to chemical cues produced by predators21.
e observed eects of AP in voles may also be rooted in biochemical similarity, for example, to other compounds
that signal danger natural settings, but this possibility requires further investigation.
Figure 1. Litter size ± SE in the control treatment, which received bedding materials of non-predator-exposed
male bank voles, and in the AP treatment, which received beddings materials of predator- exposed male voles.
Sample size inside the bars. Letters denote statistically signicant dierence between bars.
Figure 2. Bait-less trapping proportion ± SE (a) and movement area of voles ± SE (b) in AP and control
enclosures. White bars represent females; grey bars represent males. Sample size inside the bars. Letters denote
statistically signicant dierence between bars.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
4
Scientific REPORtS | (2018) 8:17214 | DOI:10.1038/s41598-018-35568-0
Increased investment in reproduction by female bank voles faced with a predator cue is in line with several
classic hypotheses concerning residual reproductive value29 and terminal investment30. We conducted the exper-
iment late in the breeding season with the prediction that females should heavily invest resources into current
reproduction. Such investment is expected because future reproduction opportunities are improbable for several
interacting reasons. e most important are likely the approach of winter and poor overwinter survival of females
that have already reproduced31, combined with high overall mortality risks from predation32. In comparison to
AP females, females in the control treatment did not increase their litter size and were still prepared to produce
one or two litters during ongoing breeding season. Reproduction should continue as long as there are sucient
resources for doing so; bank vole breeding season in our latitude usually lasts until September33. e litter size
eect of AP could be mediated by males (e.g., via sperm quality), but examples of such mechanisms are not cur-
rently described in the literature. Simple physiological mechanisms could allow bank voles females to increase
litter size: bank voles are induced ovulators34, meaning that females can produce more pups by mating more with
one or several dierent males35. Many animals, including sticklebacks (Gasterosteus aculeatus)36 and some song-
birds, have been found to increase investment in reproduction when predation risk is experimentally increased37.
However, such increases are not universal. In some cases, cessation or delay of reproduction may result from high
levels of predation. is phenomenon has been previously described in bank voles in laboratory studies38,39 and
in studies with challenging ecological conditions (e.g., variable food availability and predation risk) during the
rst reproductive window aer winter26. Reduced reproduction, however, is expected to be an unlikely alternative
late in a breeding season. Quantifying fear eect on prey tness has been criticized because it is dicult to assess
the magnitude of experimenter-induced predator cues40. While ecological experiments can deviate from natural
conditions in some way, we factored in vole biology as much as possible (e.g., by using naturally occurring sex
ratios and population densities and by adding scent cues produced by a single unique male per enclosure41.
Males and females respond dierently to scent cues. Chemical cues produced by prey faced with a
predator might communicate more than just alarm. For example, male lab mice exposed to either a predator (cat
urine) or a predator/competitor (rat urine) produce cues that make themselves more attractive to female mice42,43.
e element of our study involving trapping without seeds as bait oered support for this idea: female voles were
more attracted to traps in the AP enclosures, where those females were presented with bedding used by the male
voles that had encountered a predator. In contrast, male voles avoided traps in the AP enclosures and were more
frequently found in traps in the control enclosures. e fact that only male voles had been used to produce both
types of bedding materials in our study may be useful in linking our results to possible mechanisms. One possible
candidate attracting females to the bedding materials from predator-exposed AP males could be the major uri-
nary protein (MUP). Mice are able to regulate the total amount of MUP in their urine during social contacts, and
female mice advertise their reproductive state by varying the concentration of MUP during the oestrous cycle44.
MUP can function as a pheromone and stimulate sexual attraction45 and oestrus in female mice46. e chemical
can also promote aggressive behaviour in male mice47 and extend the longevity of olfactory cues that they leave
behind48. One or more urine components (e.g., MUP), and not urine itself, most likely drive our observed dier-
ences. Control and AP bedding was collected aer being used by a vole for about 48 hours, and since bank voles
are assumed to urinate roughly constant rate, both control and AP bedding should have been thoroughly contam-
inated with urine and feces. Furthermore, prairie voles (M. ochrogaster) and woodland voles (M. pinetorum) urine
mark at the same rate with and a without predator cue49.
If the bedding material used by predator-exposed male voles contained AP that resembled direct odour
cues from predators themselves, then we predicted trappability would be higher under control conditions (i.e.,
unstressed male used bedding), since the bedding materials were distributed in the chimneys near the traps.
However, neither the two groups nor the two sexes diered signicantly in trappability when baited traps were in
use. e absence of an experimental eect might be due to the highly attractive food rewards (i.e., energetically
rich sunower seeds) that were used as a bait in the traps. During the breeding season, voles, especially female
ones, have high energy demands, which may be met (at least partially) by consuming the sunower seed bait.
Additionally, the voles in this study may have adjusted to the relatively constant presence of AP, which was present
Figure 3. Giving up density (GUD) ± SE measured as number of seed le in the seed-trays in AP and control
enclosures. All seed trays contained 80 sunower seeds mixed with two liters of sand. White bars represent
GUD boxes in dense vegetation; grey bars represent GUD boxes in open areas. Sample size inside the bars.
Letters denote statistically signicant dierence between bars.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
5
Scientific REPORtS | (2018) 8:17214 | DOI:10.1038/s41598-018-35568-0
when the voles were added to the enclosure and regularly refreshed until the pregnant females were removed.
Ultimately, a behavioural shi favouring the gain of food rewards over the risks of predation may have occurred;
such an adjustment matches well with the risk allocation hypothesis50.
We also hypothesised that the addition of bedding materials from predator-exposed males would restrict the
movement area of voles in the experimental enclosures51. Decreasing movement area should eectively decrease
the risk of encountering and being killed by a predator4. However, we were unable to document an eect of our
experimental treatment on movement. Our results contrast with those of an earlier experiment on Cabrera voles,
which tended to decrease activity near AP patches23. In our enclosures, movement may have been suppressed
by the avian predators to such an extent that voles were unable to further reduce their movement in response
to our AP treatment52. Overall low trappability (around 50%) compared to previous summer experiments (see
for example Haapakoski et al., 201353,54) supports this nding. Not surprisingly, males were moving more than
females regardless of predation risk. Males need to move over several females territory in search of potential
mates to increase their tness55.
Perceived predation risk is expected to increase at the expense of foraging56. us, we expected a higher GUD in
enclosures with bedding materials from the predator-exposed voles; however, we were unable to document dier-
ences in GUD between treatment and control enclosures. GUD was higher in open areas compared to more densely
vegetated areas in our experimental enclosures, suggesting that the GUD method worked. With voles seeming to
avoid the avian predation risk associated with foraging in open areas, this result resembles predator facilitation57. In
fact, previous research suggests that eld voles (M. agrestis) prefer open habitats when facing a mammalian predator
(least weasel) and densely vegetated areas when facing an avian predator (kestrel falcon, Falco tinnunculus)52.
e pregnancy rate was similar in the experimental and control enclosures. Under experimentally increased
predation risk in laboratory studies, bank voles have been observed to suppress breeding with females actively
avoiding copulation38. Yet breeding suppression has never been reported during the peak of the breeding season
in the eld58, and the breeding suppression in the lab might be a laboratory artefact. Breeding suppression or
avoidance might only occur when resources are limited26,59.
e survival of females did not dier between treatment and control groups. Previous studies of bank voles
have not found an eect of predation risk on survival during summer58,60, but an inverse relationship between
predation risk and survival in winter has been reported26. In that case, the assumption was that reduced survival
was a consequence of reduced foraging. In our current study, foraging and movement were not signicantly
aected by the experimental treatment, which might explain the uniformity in survival.
Conclusions
e eects of predation are multifaceted and extend far beyond the death and consumption of prey. For example,
we found that bedding materials used by predator-exposed male voles contained cues AP that could exert strong
life-history eects in eld settings. Specically, indirect eects of predation risk, which here was communicated
by the presence of the scent cues, translated to 1) increased litter size and 2) an interaction with sex that aected
the attractiveness of traps. us, AP can impact the tness and behaviour of animals. Continued study of these
cues and others in both eld and lab settings will contribute to a more nuanced understanding of interactions
between predators and prey and between the sideshow of sex and the world of fear.
Material and Methods
Experimental setup. We conducted our experiment in twelve uncovered outdoor enclosures (0.25 ha per
enclosure, see Haapakoski et al.26 for complete details) and in the laboratory of Konnevesi Research Station in cen-
tral Finland (62°37′N, 26°20′E). Experimental status (AP vs. control, six enclosures each) was randomly assigned.
In each enclosure, 25 multiple capture traps (Ugglan Special Nr. 2, Grahnab, Hillerstorp, Sweden) were arranged
in a 5 × 5 grid with 10 m intervals between traps (Fig.4). e traps were covered with bottomless metal chimneys
(40 × 40 × 50 cm3) on which a metal lid was placed to protect voles in the traps from direct sunlight and rain.
Figure 4. Enclosures used in our experiment were 2500 m2, surrounded by a steel wall, and contained 25 traps
that had a distance of 10 m between them. At four trapping locations (circles) in each enclosure, we placed GUD
boxes, two of which had a non-altered dense vegetation surrounding the boxes and two had an mown open
areas surrounding the boxes.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
6
Scientific REPORtS | (2018) 8:17214 | DOI:10.1038/s41598-018-35568-0
Study animals. e bank vole (Myodes glareolus) is a small common boreal microtine rodent. Vole popula-
tions cycle in Scandinavia, and specialist predators play a large role in driving these cycles61. One such predator
is the least weasel (Mustela nivalis nivalis), which specializes in hunting small mammals. Due to its small size,
the weasel is able to hunt in the burrows of voles, leaving few safe places for the voles to escape predation51. As
a consequence of high predation pressure in the wild, antipredator behaviours and physiology are important in
bank voles.
In the study area, the breeding season of bank voles usually extends from the end of April to September.
During the breeding season, female bank voles are strictly territorial and need an exclusive territory to breed.
e territories of breeding males usually overlap with the territories of several females62. Bank voles are easy to
trap in the eld and maintain and breed in laboratory, thereby making the species an ideal subject for the study
of predation risk.
We used sexually mature bank voles that were either enclosure- or laboratory-born. These voles were
rst-generation descendants of wild bank voles that had been originally trapped in the Konnevesi region. Prior
to the experiment, these voles were housed individually under standard conditions60. Voles grouped together in
an outdoor enclosure were unrelated. Enclosure groups were balanced based on vole body size and number of
females who had previously given birth.
Production of alarm pheromone and control bedding materials. To produce AP and control bed-
ding materials, we used six pairs of bank vole brothers from litters produced by six dierent parental pairs. Each
of these twelve male ospring was placed individually into a standard rodent cage containing 3 l of wood shavings.
A randomly selected vole from each parental pair was used for AP production, and these six were housed in the
same room. e six that produced the control bedding were housed in a separate room.
In order to induce AP production, we exposed the six AP voles to a weasel. For the rst two weeks, a caged
weasel was placed daily on top of the cage of each AP vole for 2 minutes. Weasel cage was equipped with solid
bottom in order to prevent weasel urine and feces mixing in to AP bedding. Aer this procedure, we captured
each AP vole with our hands and released it back into its cage to simulate predator capture. e AP production
procedure was changed two weeks aer the start of the experiment in order to minimize stress for the weasel.
With this new method, a bank vole was rst captured by hand from its cage and placed individually inside a small
wire mesh cage, which was then placed inside the weasel’s cage for 2 minutes. Aerwards, each vole was released
back into its original cage. e weasel used in this procedure was housed under standard conditions60 in a cage
that was kept in a dierent building from all voles.
About 48 h old bedding material originating from one specic male vole (either AP or control, see below)
was assigned to a specic outdoor enclosure for the duration of the experiment. In all enclosures, 0.1 l of bedding
material was placed in each trap chimney and around the giving-up density tray (see below). Bedding material
(AP or control) was placed at the start of the experiment and replaced every other day during the experiment.
e experiment started with the spreading AP and control bedding materials into the enclosures at the end of
July 2015. On the same day bedding materials were added to an enclosure, four females were released to allow for
habituation to the environment and its cue of indirect predation risk or not. ree days aer adding the females,
four males were released into each enclosure. In order to determine female survival, the proportion of pregnant
females, parturition date, and litter size, females were trapped and moved to the laboratory around 18 days aer
initiation of the experiment.
Giving-up density. To measure the voles’ perception of predation risk in each enclosure, giving-up density
(GUD56) was measured using four boxes (20 × 20 × 15 cm3 covered with a transparent lids) accessible via two
2 cm diameter entrances. Two GUD boxes were placed in open vegetation, and two were placed under the grass
and were thus more protected from avian predators (see Fig.1). Open vegetation areas were created by mowing
the grass in a 1 m diameter area where the GUD boxes were placed. GUD boxes contained 80 sunower seeds
mixed with 2 L of sand. GUD, quantied here the as the number of seeds remaining, was recorded aer two
consecutive 24 hr periods, which started on the aernoon of day 4 and ending on the aernoon of day 6. If voles
perceive that predation risk outweighs foraging gains, then they will give up searching for seeds, and GUD will be
high. In situations where they feel safer, the voles will spend more time searching for the diminishing seeds, and
GUD will be lower.
Movement and trappability. To quantify movement area, all traps were activated and regularly checked,
and the location of any captured individual was recorded before release at the point of capture. Movement area
trapping was conducted on days 10 (evening only), 11 and 12 (both morning and evening), and 13 (morning
only). e rst movement area trapping (day 10 evening) was conducted without bait (hereaer “bait-less trap-
ping”) to measure the eect of AP treatment, since the food reward might have otherwise lured voles into traps.
In bait-less trapping traps were activated immediately aer distributing AP and control bedding materials in
the chimneys. Bait-less trapping is possible since voles are using trap chimneys and traps as hiding places when
traps are deactivated. Movement areas (100% convex polygons) were calculated from the trapping data with the
program Ranges VI (Anatrack ltd. Wareham, UK). Movement area calculations excluded individuals that were
caught only once. Trappability was calculated per vole as the number of captures divided by six (the number of
baited trap checks). Bait-less trapping, used for the calculation of the movement areas but not for trappability.
At the start of the intensive trapping period, two individuals were found dead in one trap in one enclosure; these
individuals were replaced.
Statistical analysis. R version 3.2.263 with package lme464 was used to perform linear mixed eects anal-
yses, which accounted for the eight voles per enclosure. We used log-likelihood ratio tests and χ2 statistics to
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
7
Scientific REPORtS | (2018) 8:17214 | DOI:10.1038/s41598-018-35568-0
evaluate statistical signicance (α = 0.05). Litter size was analysed by using a model that included treatment (AP/
control), previous pregnancy (yes/no) and the interaction between the two. Movement area, trappability and
bait-less trapping were analysed by using a model that included treatment (AP/control), sex (male/female), and
the interaction between the two. In addition, the number of captures per individual was used as a covariate in the
movement area analysis to account for dierences in movement area due trappability. Female survival and female
pregnancy were analysed with generalized linear mixed models (GLMM) with binomial error distribution. In the
analysis of survival, treatment was used as a xed factor. e pregnancy model, like the litter size model, included
as xed treatment, previous pregnancy (yes/no) and the interaction between the two. GUD was analysed using
repeated measures analyses because GUD was measured on two days. For GUD, models included treatment, time
(day one/two), their interaction, vegetation (open/covered), and the minimum number of voles known to be
alive in the enclosure. Enclosure identity was included as a random factor in all analyses. Residuals were visually
inspected for deviations from homoscedasticity and normality. In the movement area analysis, data were cube
root transformed to meet the assumptions. For unknown reasons, four voles in one control enclosure died during
the movement area trapping. erefore, we excluded this enclosure from all analyses of data from aer the point
of mortality. (Data from this enclosure were used in analyses of GUD and bait-less trapping, which were collected
before the mortality event.) We also excluded two individuals (IDs 4885 and 4886) in an AP treatment enclosure
from our movement analyses, since their tag numbers were indistinguishable in the eld.
Ethics. All work was conducted according Finnish legislation with the animal experimentation permission
from Jyväskylä University No. ESAVI/6370/04.10.07/2014.
Data Accessibility
Data will be deposited to Dryad Digital Repository aer acceptance.
References
1. Ardrey, . African Genesis: A Personal Investigation Into the Animal Origins and Nature of Man (Book). (Atheneum, 1963).
2. Lima, S. L. Nonlethal Eects in the Ecology of Predator-Prey Interactions. Bioscience 48, 25–34 (1998).
3. MacLeod, . J., rebs, C. J., Boonstra, . & Sheri, M. J. Fear and lethality in snowshoe hares: the deadly eects of non-consumptive
predation ris. Oikos 127, 375–380 (2018).
4. Lima, S. L. & Dill, L. M. Behavioral decisions made under the ris of predation: a review and prospectus. Can. J. Zool. 68, 619–640
(1990).
5. Preisser, E. L., Bolnic, D. I. & Benard, M. F. Scared to death? e eects of intimidation and consumption in predator-prey
interactions. Ecology 86, 501–509 (2005).
6. Bo onstra, . eality as the leading cause of stress: rethining the impact of chronic stress in nature. Funct. Ecol. 27, 11–23 (2013).
7. Sheri, M. J., rebs, C. J. & Boonstra, . e sensitive hare: Sublethal eects of predator stress on reproduction in snowshoe hares.
J. Anim. Ecol. 78, 1249–1258 (2009).
8. Ferrari, M. C. O., Wisenden, B. D. & Chivers, D. P. Chemical ecology of predator–prey interactions in aquatic ecosystems: a review
and prospectuse present review is one in the special series of reviews on animal–plant interactions. Can. J. Zool. 88, 698–724
(2010).
9. Apfelbach, ., Blanchard, C. D., Blanchard, . J., Hayes, . A. & McGregor, I. S. e eects of predator odors in mammalian prey
species: a review of eld and laboratory studies. Neurosci. Biobehav. Rev. 29, 1123–44 (2005).
10. Bytheway, J. P., Carthey, A. J. . & Bans, P. B. is vs. reward: How predators and prey respond to aging olfactor y cues. Behav. Ecol.
Sociobiol. 67, 715–725 (2013).
11. Sundell, J. & Ylönen, H. Behaviour and choice of refuge by voles under predation ris. Behav. Ecol. Sociobiol. 56, 263–269 (2004).
12. Mäeläinen, S., Trebaticá, L., Sundell, J. & Ylönen, H. Dierent escape tactics of two vole species aect the success of the hunting
predator, the least weasel. Behav. Ecol. Sociobiol. 68, 31–40 (2014).
13. Creel, S. & Christianson, D. elationships between direct predation and ris eects. Trends Ecol. Evol. 23, 194–201 (2008).
14. Lindström, J. Early development and tness in birds and mammals. Trends in Ecology and Evolution 14, 343–348 (1999).
15. Brown, J. S. Patch us e as an indicator of habitat preference, predation ris, and competition. Behav. Ecol. Sociobiol. 22, 37–47 (1988).
16. Verheggen, F. J., Haubruge, E. & Mescher, M. C. Alarm pheromones-chemical signaling in response to danger. Vitam. Horm. 83,
215–39 (2010).
17. Zalaquett, C. & iessen, D. e eects of odors from stressed mice on conspecic behavior. Physiol. Behav. 50, 221–227 (1991).
18. iyoawa, Y., odama, Y., ubota, T., Taeuchi, Y. & Mori, Y. Alarm pheromone is detected by the vomeronasal organ in male rats.
Chem. Senses 38, 661–668 (2013).
19. ottman, S. J. & Snowdown, C. T. Demonstration and analysis of an alarm pheromone in mice. J. Comp. Physiol. Psychol. 81,
483–490 (1972).
20. Ferrero, D. M. et al. Detection and avoidance of a carnivore odor by prey. Proc. Natl. Acad. Sci. USA 108, 11235–40 (2011).
21. Brechbühl, J. & Moine, F. Mouse alarm pheromone shares str uctural similarity with predator scents. Proc. R. Soc. B Biol. Sci. 112, 1–6
(2013).
22. Boissy, A. Fear and Fearfulness in Animals. e Quarterly Review of Biology 70, 165–191 (1995).
23. Gomes, L. A. P., Salgado, P. M. P., Barata, E. N. & Mira, A. P. P. Alarm scent-maring during predatory attempts in the Cabrera vole
(Microtus cabrerae omas, 1906). Ecol. Res. 28, 335–343 (2013).
24. Müller-Schwarze, D., Altieri, . & Porter, N. Alert odor from sin gland in deer. J. Chem. Ecol. 10, 1707–1729 (1984).
25. Carthey, A. J. . & Bans, P. B. Naïveté in novel ecological interactions: Lessons from theory and experimental evidence. Biol. Rev.
89, 932–949 (2014).
26. Haapaosi, M., Sundell, J., Ylönen, H., Ylo, H. & Ylönen, H. Predation ris and food: opposite eects on overwintering survival and
onset of breeding in a boreal rodent. J. Anim. Ecol. 81, 1183–92 (2012).
27. Bedoya-Perez, M. A., Carthey, A. J.  ., Mella, V. S. A., McArthur, C. & Bans, P. B. A practical guide to avoid giving up on giving-up
densities. Behav. Ecol. Sociobiol. 67, 1541–1553 (2013).
28. orpimäi, E., Norrdahl, . & Valama, J. eproductive investment under uctuating predation ris: Microtine rodents and small
mustelids. Evol. Ecol. 8, 357–368 (1994).
29. Williams, G. C. Natural S election, the Costs of eproduction, and a enement of Lac’s Principle. Am. Nat. 100, 687–690 (1966).
30. Clutton-Broc, T. H. eproductive Eort and Terminal Investment in Iteroparous Animals. Am. Nat. 123, 212–229 (1984).
31. Prévot-Julliard, A. C., Henttonen, H., Yoccoz, N. G. & Stenseth, N. C. Delayed maturation in female ban voles: Optimal decision or
social constraint? J. Anim. Ecol. 68, 684–697 (1999).
32. Tadlec, E. & Zejda, J. Small rodent population uctuations: e eects of age structure and seasonality. Evol. Ecol. 12, 191 (1998).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
8
Scientific REPORtS | (2018) 8:17214 | DOI:10.1038/s41598-018-35568-0
33. Eccard, J. A., lemme, I., Horne, T. J. & Ylönen, H. Eects of competition and season on survival and maturation of young ban vole
females. Evol. Ecol. 16, 85–99 (2002).
34. Clare, J. ., Clulow, F. V. & Grieg, F. Ovulation in the ban vole, Clethrionomys glareolus. J. Reprod. Fertil. 23, 531 (1970).
35. Moonen, M., osela, E., Mappes, T. & Mills, S. C. Sexual antagonism for testosterone maintains multiple mating behaviour. J.
Anim. Ecol. 81, 277–283 (2012).
36. Candolin, U. eproduction under predation ris and the trade-o between current and future reproduction in the threespine
sticlebac. Proc. R. Soc. B Biol. Sci. 265, 1171–1175 (1998).
37. Mönönen, M., Forsman, J. T., ananoja, T. & Ylönen, H. Indirect cues of nest predation ris and avian reproductive decisions.
Biol. Lett. 5, 176–8 (2009).
38. Ylönen, H. & onainen, H. Breeding suppression in the ban vole as antipredatory adaptation in a predictable environment. Evol.
Ecol. 8, 658–666 (1994).
39. Ylönen, H. et al. Small mustelids and breeding suppression of cyclic microtines: adaptation or general sensitivity? Ann. Zool. Fennici
32, 171–174 (1995).
40. Peers, M. J. L. et al. Quantifying fear eects on prey demography in nature. Ecology 99, 1716–1723 (2018).
41. Norrdahl, . & orpimäi, E. Changes in individual quality during a 3-year population cycle of voles. 239–249, https://doi.
org/10.1007/s004420100795 (2002).
42. Lixing, J. Z. Æ., evin, S. Æ. & Novotny, M. V. Chronic exposure of cat odor enhances aggression, urinary attractiveness and sex
pheromones of mice. J. Ethol. 279–286, https://doi.org/10.1007/s10164-007-0060-1 (2008).
43. Liu, Y.-J. et al. Chronic Co-species Housing Mice and ats Increased the Competitiveness of Male Mice. Chem. Senses 42, 247–257
(2017).
44. Stopa, P., Janotova, . & Heyrovsy, D. e advertisement role of major urinary proteins in mice. Physiol. Behav. 91, 667–670
(2007).
45. oberts, S. A. et al. Darcin: a male pheromone that stimulates female memory and sexual attraction to an individual male’s odour.
BMC Biol. 8, 75 (2010).
46. Marchlewsa-oj, A., Cavaggioni, A., Mucignat-Caretta, C. & Olejnicza, P. Stimulation of Estrus in Female Mice by Male Urinary
Proteins. J. Chem. Ecol. 26, 2355–2366 (2000).
47. Chamero, P. et al. Identication of protein pheromones that promote aggressive behaviour. Nature 450, 899–902 (2007).
48. Hurst, J. L., obertson, D. H. L., Tolladay, U. & Beynon, . J. Proteins in urine scent mars of male house mice extend the longevity
of olfactory signals. Anim. Behav. 55, 1289–1297 (1998).
49. Wol, J. O. Scent maring by voles in response to predation ris: a eld-laboratory validation. Behav. Ecol. 15, 286–289 (2004).
50. Lima, S. L. & Bedneoff, P. A. Temporal Variation in Danger Drives Antipredator Behavior: The Predation is Allocation
Hypothesis. Am. Nat. 153, 649–659 (1999).
51. Norrdahl, . & orpimäi, E. Does Mobility or Sex of Voles Aect is of Predation by Mammalian Predators? Ecology 79, 226–232
(1998).
52. orpimäi, E., oivunen, V. & Haarainen, H. Microhabitat use and behavior of voles under weasel and raptor predation ris:
predator facilitation? Behav. Ecol. (1996).
53. Haapaosi, M., Sundell, J. & Ylönen, H. Mammalian predator-prey interaction in a fragmented landscape: Weasels and voles.
Oecologia 173, 1227–1235 (2013).
54. Haapaosi, M., Lensu, A., Sundell, J., Vihervaara, H. & Ylönen, H. Infanticide eects on behavior of the ban vole (Myodes
glareolus) in the fragmented breeding habitat. Behav. Ecol. Sociobiol. 69, 49–59 (2015).
55. osela, E., Mappes, T. & Ylonen, H. Territorial Behaviour and eproductive Success of Ban Vole Clethrionomys glareolus
Females. J. Anim. Ecol. 66, 341–349 (1997).
56. Brown, J. S. Vigilance, patch use and habitat selection: Foraging under predation ris. Evol. Ecol. Res. 1, 49–71 (1999).
57. otler, B., Blaustein, L. & Brown, J. Predator facilitation: the combined eect of snaes and owls on the foraging behavior of gerbils.
Annales Zoologici Fennici 47, 465–468 (1992).
58. Trebaticá, L., Suortti, P., Sundell, J. & Ylönen, H. Predation ris and reproduction in the ban vole. Wildl. Res. (2012).
59. Jochym, M. & Halle, S. To breed, or not to breed? Predation ris induces breeding suppression in common voles. Oecologia 170,
943–953 (2012).
60. Haapaosi, M., Sundell, J. & Ylönen, H. Conservation implications of change in antipredator behavior in fragmented habitat:
Boreal rodent, the ban vole, as an experimental model. Biol. Conserv. 184 (2015).
61. Hansi, I., Henttonen, H., orpimäi, E., Osanen, L. & Turchin, P. Small odent Dynamics and Predation. Ecology 82, 1505–1520
(2001).
62. Bujalsa, G. e role of spacing behavior among females in the regulation of reproduction in the ban vole. J. Reprod. Fertil. Suppl.
19, 465–74 (1973).
63.  Development Core Team, . : A L anguage and Environment for Statistical Computing. R Foundation for Statistical Computing 1,
409 (2011).
64. Bates, D., Mächler, M., Boler, B. & Waler, S. Fitting Linear Mixed-Eects Models using lme4. (2014).
Acknowledgements
We thank the mammalian ecology and life histories research group lead by Prof. Hannu Ylönen for providing
all of the facilities for our study and commenting the manuscript. Technicians at the Konnevesi Research Station
did an excellent job of maintaining the enclosures. Alexandra iel’s help in the lab and in the eld was valuable.
Ines Klemme provided comment to the manuscript. Financial support came from the Jenny and Antti Wihuri
Foundation (to M.H.). A.H. and K.D.M. thank the sta of the Resource Ecology Group for their support.
Author Contributions
M.H. and A.A.H. conceived the study; A.A.H. and M.H. carried out eld experiment; M.H., K.D. M. and A. A. H.
performed statistical analyses; and M.H., K.D.M. and A.A.H. wrote the manuscript.
Additional Information
Competing Interests: e authors declare no competing interests.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
9
Scientific REPORtS | (2018) 8:17214 | DOI:10.1038/s41598-018-35568-0
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-
ative Commons license, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not per-
mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the
copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
© e Author(s) 2018
Content courtesy of Springer Nature, terms of use apply. Rights reserved

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com