Effects of Repeated Exposure to Fox Odor on Locomotor Activity ...

Journal of Chemical Ecology, Vol. 25, No. 7, 1999

EFFECTS OF REPEATED EXPOSURE TO FOX ODOR ON LOCOMOTOR ACTIVITY LEVELS AND SPATIAL MOVEMENT PATTERNS IN BREEDING MALE AND FEMALE MEADOW VOLES (Microtus pennsylvanicus) T. S. PERROT-SINAL,1 K.-P. OSSENKOPP, and M. KAVALIERS* Neuroscience Program University of Western Ontario London, Ontario, Canada N6A 5C2 (Received August 12, 1998; accepted March 6, 1999) Abstract—Following five days of baseline activity recording, voles were exposed to fox odor for 3 min each day for five days. Immediately following each daily exposure, locomotor activity levels and spatial movement patterns were assessed using an automated activity monitoring system (Digiscan system). Males displayed a significant reduction in levels of various measures of locomotor activity following exposure to fox odor on each exposure day relative to baseline levels. Males preferred the corner of the testing box significantly more on the second day of fox odor exposure relative to baseline. Although females showed only a brief reduction in the number of movements made on the first day of odor exposure, this response lasted significantly longer on each of the subsequent odor exposure days. The reliability of the reductions in activity levels displayed across days by breeding male voles supports the hypothesis that this response is adaptive. Furthermore, the results suggest that, although female voles do not generally display this behavioral response, it can be elicited in females when the predation threat is repeated in consistent context. Key Words—Locomotor activity, meadow vole, sex difference, predator odor.

INTRODUCTION

Predation represents the leading cause of death for many small rodents and as such has led to a variety of predator detection mechanisms and antipredatory *To whom correspondence should be addressed. 1 Present address: Department of Physiology, School of Medicine, Bressler Building, University of Maryland, 655 W. Baltimore St., Baltimore, Maryland 21201.

1567 0098-0331/99/0700-1567$16.00/0 © 1999 Plenum Publishing Corporation

1568

PERROT-SINAL, OSSENKOPP, AND KAVALIERS

responses (Lima, 1998). Several species have evolved the ability to detect a predator by its odor, thus reducing the chances of actual encounters with the predator. Past work has demonstrated, both in the laboratory and in the field, that rodents display characteristic behavioral responses following exposure to a predator or its odor. In particular, predators are able to influence the levels and patterns of prey motor activity (Lima, 1998). Results of field studies have shown that small mammals avoid vegetation and traps tainted with the odors of predators, including the red fox (Vulpes vulpes) and weasel (Mustela spp.) (Calder and Gorman, 1991; Dickman and Doncaster, 1984; Sullivan et al., 1988). Similarly, increases in defensive behaviors and decreases in nondefensive behaviors have been noted in laboratory rats following exposure to predatory cat odor (Blanchard et al., 1990, 1995a,b). The results of these laboratory and field investigations have further suggested that these predator odor responses are sexually dimorphic, with males apparently displaying enhanced behavioral responses relative to females (Jedrzejewski and Jedrzejewski, 1990). Our recent work has focused on behavioral responses of meadow voles (Microtus pennsylvanicus) to predator odor in the laboratory. We have demonstrated that male meadow voles display a significant reduction in locomotor activity levels following exposure to fox odor (trimethyl thiazoline; main constituent of fox anal gland secretion) but not following exposure to other aversive, pungent control odors (Perrot-Sinal et al., 1996). We have established further that this response is selective, as it is observed only in males that are reproductively active and not in either breeding female or in nonbreeding male or female meadow voles (Perrot-Sinal, 1998; Perrot-Sinal et al., 1996). It was hypothesized that this particular response is adaptive for breeding males since: (1) foxes hunt using sound cues primarily (Jedrzejewski et al., 1993) and reductions in locomotor activity levels would serve to reduce sound, and (2) breeding males may have a greater chance of encountering a predator in the wild because they have increased home-range size, higher activity levels, and likely spend more time in open areas such as runways (Madison, 1985; Madison and McShea, 1987). It may be hypothesized further that this particular behavioral response is not an essential part of the repertoire of defensive behaviors of breeding females. This may be because females either have lower activity levels and smaller home ranges, do not enter open areas as readily as males (where this defense would be most effective), and/or have evolved other behavioral responses that were not measured. It is quite likely that in the wild rodents are repeatedly exposed to predator odor, especially when population levels of predators are high. Most past work has investigated behavioral responses following acute exposure to a predator odor. However, a few studies have examined behavioral responses in laboratory rats to more chronic predation threats. For example, male rats continued to avoid (File et al., 1993) and display a reduced amount of contact time with (Zan-

1569

FOX ODOR EFFECTS ON VOLES

grossi and File, 1992) a cloth that had been tainted with cat odor, even after five consecutive exposures. Similarly, a detailed study examining more ethologically relevant behaviors demonstrated that increases in the use of tunnels and burrows by male laboratory rats following exposure to predation threat does not diminish after five consecutive exposures (Blanchard et al., 1990) and behavioral habituation is minimal after 20 days of 60-min exposure to a cat (Blanchard et al., 1998). The majority of field investigations of repeated/continuous exposure to predator odors have been conducted with microtine rodents such as the meadow vole. In contrast, laboratory investigations of predator responses have been conducted primarily with laboratory rats and mice (Blanchard et al., 1990; 1995b; File et al., 1993). Here we propose to add to our understanding of defensive behaviors following predation threat by examining repeated exposure in the laboratory setting in a species other than the laboratory rat and by providing evidence for the adaptiveness of these behaviors. In the present study, we examined the effects of repeated daily exposure to fox odor on locomotor activity levels and patterns of space use in male and female meadow voles housed in mixed-sex pairs under a reproductively stimulatory photoperiod (simulating the breeding season). A detailed multivariate analysis of locomotor activity levels and patterns of space use was obtained using an automated activity monitoring system (Ossenkopp and Kavaliers, 1996). It was hypothesized that, if reductions in activity levels are adaptive for breeding males, this behavioral response would not habituate with repeated exposure. Furthermore, although breeding females do not generally display reductions in activity levels following exposure to fox odor, it was hypothesized that this behavioral response may be elicited under repeated predation threat in a consistent environment.

METHODS AND MATERIALS

Subjects Eight male and eight female, adult, laboratory bred (fifth-sixth generation) meadow voles (40-80 g; approximately 12-15 months of age) were housed in mixed-sex pairs throughout the experiment. The animals were kept in polyethylene cages with hardwood (Beta chip) bedding in a colony room maintained at 21 ± 1°C under a long day (i.e., reproductively stimulatory) photoperiod (16L: 8D dark cycle with lights on at 06:00 hr). Since the housing conditions used were meant to simulate the breeding season (Adams et al., 1980; Lee et al., 1970; Seabloom, 1985), voles will be labeled as breeding. Information about reproductive state was gained through examining testes position in males and sex steroid levels in both sexes. Food (Agway Rat Chow and alfalfa pellets)

1570


and water were available ad libitum. Lab chow was supplemented with carrot once a week.

Experimental Apparatus The automated Digiscan activity monitoring system consisting of six Digiscan Animal Activity Monitors (Omnitech model RXY2CM-16, Omnitech Electronics Inc., Columbus, Ohio) was used. Activity test chambers consisted of clear Plexiglas boxes (40 x 40 x 30 cm) that had a grid of infrared beams mounted horizontally every 2.54 cm along the perimeter (16 infrared beams along each side) and 4.5 cm above the floor. The six monitors were connected to a Digiscan Analyzer (Omnitech model DCM-8) that transmitted the activity data to a microcomputer. A number of activity variables, either calculated directly by the Digiscan Analyzer or derived from the measured variables, were examined in order to provide a complete assessment across days of testing (Table 1). The validity and reliability of these measures has been previously verified (Ossenkopp et al., 1987; Sanberg et al., 1985).

TABLE 1. DESCRIPTIONS OF ACTIVITY VARIABLES MEASURED DURING BASELINE ACTIVITY MONITORING AND FOLLOWING Fox ODOR EXPOSURE IN MALE AND FEMALE MEADOW VOLES Activity variable Total distance (cm) Movement time (sec) Number of movements Center time (sec)

Margin time (sec) Center distance (cm) Margin distance (cm) Time in corners (sec)

Brief description The horizontal distance traveled by an animal during a given time sample (dependent on animal's path) The amount of time spent engaged in movement during a given time sample The number of separate movements made by an animal during a given time sample The amount of time spent in the center of the testing box (animal's center of gravity more than 6.35 cm away from any wall) during a given time sample The amount of time spent within 6.35 cm of a wall during a given time sample Distance traveled in the area defined as the center of the testing box (see center time) Distance traveled outside of the area defined as the center of the testing box (see margin time) The amount of time spent in close proximity to 2 adjoining walls of the testing box


1571

Voles were exposed individually to the fox odor in a varnished wooden box that was divided into six equal compartments (32 x 15 x 15 cm each). The box had a Plexiglas front and lid, which was closed during exposure. Procedure Baseline Activity Monitoring. At mid-photoperiod, each vole was placed individually in an activity monitor and the activity measures were accumulated in 5-min time "bins" over a 30-min session (a total of six time samples). Baseline activity was monitored in this manner on five consecutive days prior to the beginning of the repeated daily odor exposures in order to habituate the voles to the apparatus. Activity Monitoring Following Odor Exposure. On the day immediately following the end of baseline recording, voles were exposed to fox odor (2,5dihydro-2,4,5-trimethyl thiazoline; Phero-Tech Inc.) for five consecutive days. Since previous experiments have demonstrated that activity levels following a variety of control odors, including butyric acid (Perrot-Sinal et al., 1996) and extract of almond (Perrot-Sinal, 1998), are not different from activity levels monitored during the second day of baseline activity recording, no control odor condition was used in the present experiment. Additionally, levels of activity during baseline did not change significantly across the five days of testing for males or females. Thus, activity levels following exposure to fox odor were compared to baseline levels of activity averaged across the five baseline days. This average was not different from levels of activity measured previously during baseline (Perrot-Sinal et al., 1996) and therefore was assumed not to be different from levels of activity previously measured following exposure to control odors. Cotton swabs saturated with the odor solution were attached with masking tape inside the Plexiglas front of each compartment of the exposure box. Once all cotton swabs were in place, animals were removed from their home cages, individually placed into the compartments, and the lid was closed. Odor exposure lasted 3 min, after which the animals were removed from the compartments, placed in their home cages, and transported to an adjacent room where the activity monitoring apparatus was located. This process required approximately 1-2 min. Activity was monitored for 30 min in the same manner as during the baseline recording (see above) and took place at mid-photoperiod. Following each odor exposure trial, the wooden box was thoroughly washed using an odorless Alconox (VWR Canlab) solution and then rinsed with a baking soda solution. Following activity monitoring, voles were returned to their home cages. The Plexiglas activity monitors were thoroughly cleaned with an Alconox solution and rinsed with a baking soda solution between each activity recording session. Radioimmunoassay of Sex Steroids. The reproductive state of male and female voles was assessed by assaying plasma levels of sex steroid hormones.

1572


Following the completion of the fifth day of odor-exposure testing, voles were given sodium pentobarbital (65 mg/kg, intraperitoneally) and once animals were completely anesthetized, blood was taken by transcardial puncture. Blood samples were centrifuged at 14,000 rpm (Eppendorf, model 5402) for 10 min and the resulting plasma was stored at -50°C until the time of assay. For male and female voles, plasma samples (100 ul) were assayed in duplicate for testosterone or estradiol, respectively, using commercially available 125I RIA kits (Coat-aCount, Diagnostic Products, Los Angeles, California). For the testosterone assay, the antiserum had a cross-reactivity with 5a-dihydrotestosterone of 5%, and the sensitivity of the assay was 0.025 ng/ml as calculated from the standard curve. The intraassay coefficient of variation was measured in triplicate from low, medium, and high pools and ranged from 7% to 15%. For the estradiol assay, the sensitivity of the assay was calculated to be 10.5 pg/ml and the intraassay coefficient of variation, as measured in triplicate from low, medium, and high pools, ranged from 3% to 18%. Data and Statistical Analyses To assess differences in activity levels across the five days of fox odor exposure, repeated-measures analysis of variance (ANOVA) was performed for each activity variable separately with condition [baseline (average across five days of baseline activity recording) and five exposure days] and time (six time samples) as within-subject factors. These were performed separately for male and female voles. Post-hoc tests consisted of Tukey's HSD test for all tests. A significance level of a = 0.05 was used for all tests. RESULTS

Reproductive Condition and Plasma Levels of Sex Steroids. All males had scrotal testes, which is a good indicator of heightened reproductive activity in small mammals (McCravy and Rose, 1992). One male and one female were excluded from the testosterone and estradiol analyses, respectively, because the plasma samples obtained were not large enough to perform the assays. The mean level of testosterone in plasma of males was 1.37 ± 0.6 ng/ml (N = 7), while the average level of estradiol was 15.5 + 1.6 pg/ml in plasma of females (N= 6). One female was removed from all analyses because she gave birth during the course of the experiment. Her partner was retained in the experiment but was housed alone for the odor-exposure portion of the study. The plasma testosterone level of this male did not appear to be different from the other males. Levels of estradiol in females measured here were indicative of reproductively active females and higher than levels reported for reproductively quiescent females (Perrot-Sinal,


1573

1998). Levels of testosterone in males were lower than previously reported for breeding males from the same colony but were higher than levels reported for reproductively quiescent male voles (Perrot-Sinal, 1998) and comparable to levels for reproductively active wild-caught voles (Perrot-Sinal et al., 1998). The discrepancy in testosterone levels in laboratory-bred breeding males may be due to the length of time that males were housed with the same female (two weeks in present study and 72 hr in past study). Alternatively, levels of testosterone reported here may have been suppressed by the repeated exposure to predator odor stressor. For this reason, no attempt will be made to correlate hormone levels with behavior. Activity Levels and Patterns of Space Use Following Odor Exposure for Males. Significant condition x time interactions were found for total difference traveled [F (25, 175) = 2.33, P < 0.001; Figure 1A], amount of time spent in movement [F (25, 175) = 2.04, P = 0.004] and the number of movements made [F (25, 175) = 2.01, P = 0.005; Figure 1B]. The results of post-hoc analyses are summarized in Table 2. Generally, activity levels were reduced for 15-20 min of activity recording following exposure to fox odor on exposure days 1 and 2 relative to baseline. Activity also was reduced during the initial 5 min following exposure to fox odor on exposure days 3-5 relative to baseline levels. Significant condition x time interactions also were noted for the distance traveled in the center of the test box [F (25, 175) = 1.64, P = 0.04; Table 1 and Figure 2A] and the distance traveled around the perimeter of the test box [F (25, 175) = 2.61, P < 0.001; Figure 3A]. Results of post-hoc analyses are summarized in Table 2. There was a trend for a main effect of condition for both center time (P = 0.07; Figure 2B) and margin time (P = 0.07; Figure 3B). A significant main effect of condition was noted for comer time [F (5, 35) = 3.41, P = 0.01; Figure 4] and post-hoc analyses revealed that more time was spent in the four corners of the test box on day 2 of exposure relative to baseline, day 1 and day 3 of exposure. Figure 5 summarizes the time spent in the different areas of the testing box for baseline and each day of odor exposure for male voles. Activity Levels and Patterns of Space Use Following Odor Exposure for Females. A significant condition x time interaction was found for the number of movements made [F (25, 150) = 1.6, P = 0.04; see Figure 6A). Results of the post-hoc analyses are summarized in Table 3. Generally, reductions in the number of movements made were most pronounced on the fifth day of odor exposure relative to baseline. There were no significant differences in levels of the other activity variables measured following exposure to the fox odor relative to baseline for females (see Figure 6B for an example of total distance traveled). Female voles did not alter their patterns of space use significantly within the testing boxes following exposure to fox odor on any exposure day relative to baseline.

1574


FIG. 1. (A) Total distance traveled and (B) number of movements made over 5-min intervals following exposure to fox odor on five consecutive days relative baseline in male meadow voles (N = 8). For ease of presentation, lines representing standard error of the mean have been omitted. Significant differences are summarized in Table 2.

DISCUSSION The results of this study showed that repeated daily exposure to predator odor significantly affected the locomotor activity levels and patterns of space use by reproductive meadow voles in a sexually dimorphic manner. The major findings can be summarized as follows: (1) Breeding male meadow voles displayed

1575


TABLE 2. SUMMARY OF ALL SIGNIFICANT REDUCTIONS IN LEVELS OF ACTIVITY AT SPECIFIC TIME POINTS FOR EACH DAY OF ODOR EXPOSURE RELATIVE TO BASELINE LEVELS FOR MALE MEADOW VOLES (N = 8)a Exposure day Activity variable

Total distance (cm)

Movement time (sec)

Number of movements

Center distance (cm)

Margin distance (cm)

aAll

One

Two

5 min 10 min 15 min 5 min 10 min 15 minb 5 min 10 min 15 minb 20 min 5 min 10 min

5 min 10 min 15 min 5 min 10 min 15 min 5 min 10 min 15 min

5 min 10 min 15 min

5 min 10 min 15 min 5 min 10 min 15 min

Three

Four

Five

5 min

5 min 10 minb

5 min 10 min

5 minb

5 min

5 min 10 minb

15 min

5 min 10 min 15 minb

10 min

5 minb 10 minb

5 min

5 min 10 min

5 min 10 min

significant differences are relative to baseline for same time point at P < 0.05 significance level.

b P