- Email: [email protected]
Do gill parasites influence the foraging and antipredator behaviour of rainbow darters, Etheostoma caeruleum?
Animal Behaviour 82 (2011) 817e823
Contents lists available at ScienceDirect
Animal Behaviour journal homepage: www.elsevier.com/locate/anbehav
Do gill parasites influence the foraging and antipredator behaviour of rainbow darters, Etheostoma caeruleum? Adam L. Crane*, Andrea K. Fritts, Alicia Mathis, John C. Lisek, M. Chris Barnhart Department of Biology, Missouri State University
a r t i c l e i n f o Article history: Received 24 February 2011 Initial acceptance 9 May 2011 Final acceptance 27 June 2011 Available online 11 August 2011 MS. number: A11-00164R Keywords: alarm cue Etheostoma caeruleum glochidia parasite rainbow darter Unionidae
Parasites are known to affect an array of characteristics of their hosts, including morphology, physiology and behaviour. We examined the foraging and antipredator behaviour of rainbow darters, Etheostoma caeruleum, that were parasitized by glochidia larvae of freshwater mussels (Ptychobranchus occidentalis and Venustaconcha pleasii: Unionidae). Glochidia attach to the gills of the host and become encapsulated in host tissue. Over a period of days or weeks the larvae develop into free-living juveniles, which then leave the host. Parasitized darters increased ventilation rates (either early in the infestation or at the height of the infestation), were less active during foraging trials, lost more body size than nonparasitized darters and showed significantly weaker responses to predation risk (signalled by the presence of a chemical alarm cue). Therefore, even for a relatively short-term infection, parasitized darters may pay a cost in terms of decreased growth and decreased probability of survival. Ó 2011 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.
Parasites are defined partly by their production of negative impacts on their hosts (Begon et al. 1990), although the magnitude of these impacts can be highly variable. Increased predation of intermediate hosts due to manipulation by parasites can increase parasite fitness by facilitating their transfer to the next host (e.g. Lafferty & Morris 1996; Bakker et al. 1997; Loot et al. 2002). In contrast, single-host parasites are not expected to manipulate their hosts to increase predation risk, because death of the host would also result in death of the parasite (Smith Trail 1980). Nevertheless, single-host parasites can harm the host by co-opting host resources (overviews: Barber et al. 2000; Bush et al. 2001), and can lead to death of the host when infestations are particularly heavy (e.g. Brown et al. 1995; Northcott et al. 2003). Negative effects of single-host parasites can include anatomical and physiological effects (Bush et al. 2001) and/or changes in behaviour that reduce success of activities such as foraging (Maksimowich & Mathis 2000; Finley & Forrester 2003), aggressive competition (Fox & Hudson 2001), courtship (Pélabon et al. 2005) and parental care (Sasal 2006). Of the studies that have examined the influence of single-host parasites on antipredator behaviour, the results have not been consistent, ranging from no effects on * Correspondence: A. L. Crane, Department of Biology, Missouri State University, 901 S. National, Springfield, MO 65897, U.S.A. E-mail address: [email protected] (A. L. Crane).
predation avoidance (Vaughan & Coble 1975) to increased levels of antipredator responses (Milinski 1985; Parris et al. 2006) to decreased levels of antipredator responses (LaMunyon & Eisner 1990; Krkosek et al. 2011). The life cycle of freshwater mussels (Unionidae) includes a parasitic life stage where large numbers of larvae (i.e. glochidia) attach to fish hosts, usually on the gills (Barnhart et al. 2008). These single-host parasites have evolved specificity to their fish hosts, which are used for dispersal and development (Waller & Mitchell 1989). During the infestation process, glochidia become encapsulated in the hosts’ tissues and then metamorphose over a period of days into free-living juveniles (Rogers-Lowery & Dimock 2006; Barnhart et al. 2008). Little is known of the physiological effects of glochidial infestation on host fish. However, Kaiser (2005) found that respiratory gas exchange of largemouth bass, Micropterus salmoides, was impaired by an infestation of glochidia from the broken ray mussels, Lampsilis reeveiana, resulting in elevated resting ventilation rates, reduced oxygen consumption and lower tolerance of low oxygen conditions. These effects were dependent on intensity of infestation (number of attached glochidia). In extreme cases glochidial infestations can cause mortality (Howerth & Keller 2006); however, the more typical effects that have been observed include the energetic costs of immune responses (Dodd et al. 2005) and/or increased ventilation rates, which appear to be due to reduced surface area for gas exchange over the gills (Kaiser 2005).
0003-3472/$38.00 Ó 2011 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.anbehav.2011.07.015
818
A. L. Crane et al. / Animal Behaviour 82 (2011) 817e823
The objective in this study was to examine the effects of glochidia on the behaviour and growth of a host fish, the rainbow darter, Etheostoma caeruleum. Our study involved two experiments that occurred over the course of approximately 5 months. We used two species of mussels in two experiments (experiment 1: kidneyshell mussel, Ptychobranchus occidentalis; experiment 2: Pleas’ mussel, Venustaconcha pleasii) based on species availability. Both mussels are common at sites in the smaller tributaries of the White River system in the Missouri Ozarks (Oesch 1984) and both use darters and sculpin as the primary hosts for their glochidia (Barnhart & Roberts 1997; Riusech & Barnhart 1998). For both species of mussels, glochidia are produced in the autumn and released in the spring, and females are barren of glochidia during JuneeSeptember (M. C. Barnhart, personal observations). At sites where mussels are abundant, glochidial infestations of darters are common. For example, over 60% of rainbow darters examined from a site in the James River in May were parasitized with Pleas’ mussel glochidia (infestation intensity: mean SD ¼ 16 16.7, range 1e73; Riusech & Barnhart 1998). In our first experiment (kidneyshell mussels), we made four predictions. (1) Because of damage to the host’s gills or reduction in gill surface area available for respiration (Kaiser 2005), ventilation rates should be higher for parasitized darters. (2) Because of respiratory stress, activity levels of parasitized individuals should decrease. (3) Reduced activity of darters should result in lower foraging success in behavioural trials. (4) Reduced foraging success should lead to decreased growth. In the second experiment (Pleas’ mussels), we made two predictions. (1) Increased ventilation rates should also increase in response to infestation by this species. (2) If parasitism has energetic costs, then parasitized darters should be less sensitive to predation risk while foraging (Lima 1998) than nonparasitized darters, which have been observed to decrease foraging activity under conditions of high predation risk (Woods 2008); thus, we predicted that parasitized darters would not decrease foraging behaviour under conditions of high predation risk.
METHODS Experiment 1: Parasitism by Kidneyshell Mussels Effects on activity, ventilation, foraging and body size of rainbow darters In March 2009, we collected 40 rainbow darters and one adult kidneyshell mussel from the James River in Greene County, Missouri, U.S.A. We did not check whether fish were previously parasitized with glochidia; however, no exogenous glochidia were recovered from the darters during monitoring (details below), indicating that they did not carry glochidia when they were collected. To produce infestation levels realistic for wild fishes, we used standard methods (Riusech & Barnhart 1998). This mussel species releases its glochidia in membranous packets (i.e. conglutinates). We separated glochidia from the conglutinates under a dissecting microscope. We counted the viable glochidia in 10 200 ml subsamples of this volume and extrapolated the numbers in these samples to determine the total number of glochidia. Viable glochidia were identified by their closing response to salt (Lefever & Curtis 1912). We then parasitized 20 darters by placing them in a 4-litre container with approximately 4000 viable glochidia per liter for 15 min. During this time we used a rubber-bulb 25 ml syringe (turkey baster) to gently agitate the water and keep the glochidia uniformly suspended. An additional 20 darters served as a control group; they experienced the stress of a sham infestation that followed the same methods, but no glochidia were added.
Immediately following the exposure period, darters from both treatment groups (parasitized and nonparasitized) were moved to individual 1.5-litre monitoring containers (Aquatic Habitats, Inc., Apopka, FL, U.S.A.) that received continuously flowing water (ca. 26 C). Water from each container flowed into a common holding chamber, and a sump pump was used to pump the water through two filters (UV sterilizer and a 5 mm particulate filter) before it was recirculated. Although some chemical stimuli from individual darters may have remained in the recirculating water, these chemicals should have been distributed equally to darters in each container. To capture adult glochidia that did not remain attached to gills and to capture juveniles that detached after transformation, the outflow from each container entered a corresponding filter cup (125 mm mesh); our methods were similar to those of Dodd et al. (2005). The minimum size of glochidia for the species that we used in this study was 200 mm (Barnhart et al. 2008), so it is unlikely that any glochidia would pass through the filter. However, any glochidia or metamorphosed juveniles that escaped would be removed by the filters attached to the common holding chamber before the water was recirculated. At 8 days postinoculation (DPI) we began examining the filter cups of the parasitized darters by temporarily stopping the water flow, removing each filter cup, and rinsing its contents into a Bogorov tray placed under a microscope. We then counted the unmetamorphosed glochidia and metamorphosed juveniles. After 8 days, we checked the filters and counted the recovered mussels at 2-day intervals until no more were recovered (ca. 3 weeks). Because behavioural and physiological effects can be dependent on infestation intensity (Kaiser 2005), we calculated the intensity of the infestation that each fish experienced (i.e. total number of attached glochidia) by the total numbers of glochidia and metamorphosed juveniles that we recovered. Immediately following exposure to glochidia, and also at the end of the experiment, we measured the volume of darters as the amount of water that each darter displaced in a 5 ml graduated cylinder (0.1 ml precision). We used volumetric displacement in water as a measure of body size because it is likely to be less stressful to fish than other measures that require a longer removal from the water. Beginning on 2 DPI, we fed darters with brine shrimp, Artemia salina (1 ml of brine shrimp solution, ca. 10 total brine shrimp), every other day between 1500 and 1700 hours. We conducted behavioural trials on 5 different days that included both during (DPI: 2, 8, 14, 20) and after (DPI: 28) the parasitic period. Before the trials began, we lined the outer front of the tanks with dark window tinting and placed lights behind the tanks to allow us to see the darters easily while minimizing our visibility to the darters. We also placed cardboard barriers between the tanks to eliminate visual cues between adjacent fish. Each behavioural trial first consisted of recording the ventilation rate for each darter prior to feeding. At the beginning of the observation period, we stood quietly approximately 0.25 m in front of tanks. Most individuals were relatively stationary in the tanks. Once we were confident that we could clearly see at least one operculum (within 0e2 min), we counted ventilation rates for 60 s. Then, we immediately fed each darter and observed it for 120 s while recording the number of moves and the total number of brine shrimp consumed. These feeding trials were done in accordance with the darters’ routine feeding schedule. After the trials on 28 DPI, we measured the volumetric displacement for each darter. Experiment 2: Parasitism by Pleas’ Mussels Effects on activity, ventilation and antipredator behaviour of rainbow darters During December of 2008, we collected 103 rainbow darters from Bull Creek in Taney County, Missouri, U.S.A. After 2 days in
A. L. Crane et al. / Animal Behaviour 82 (2011) 817e823
captivity, we parasitized darters with glochidia from a single gravid Pleas’ mussel collected from the James River in Greene County, Missouri, U.S.A. This species of mussel holds its glochidia inside the marsupial gills. To extract glochidia, we used a hooked probe to cut the marsupial gills of the mussel and flushed the glochidia into a measured volume of water in a collection beaker. All other procedures for infecting darters with glochidia were identical to experiment 1. We parasitized 44 darters and gave a sham infestation to a group of 44 control darters (see experiment 1 methods for details). Again, no exogenous glochidia were recovered from the fish during monitoring. The set-up of tanks was as in experiment 1 except that the water temperature was 24 C. We fed darters with blackworms (Lumbriculus variegatus) or brine shrimp (A. salina) in a water solution at 2-day intervals between 0900 and 1700 hours throughout the time they were housed in the monitoring containers. The amount of prey varied depending on the amount that individuals could consume in approximately 5 min. Two days following the infestation, we began monitoring ventilation rates (i.e. opercula beats) on a subset of randomly selected darters (10 parasitized and 10 nonparasitized). We repeated measurements on the same darters between 1500 and 1700 hours at 2-day intervals for 22 days. For tests of antipredator responses, we used a chemical alarm cue as a stimulus indicating predation risk. Chemical alarm cues are contained in the skin of many fishes and can be released when a fish’s skin is damaged during a predator attack; nearby conspecifics respond with antipredator behaviour (reviewed in: Smith 1992; Mathis 2009). Rainbow darters have been shown to respond to conspecific alarm cues with reduced activity in both laboratory (Commens & Mathis 1999) and field (Crane et al. 2009) studies; these studies also showed that the antipredator responses are specific to conspecific extract and are not responses to general disturbance or to extracts from other species. We randomly assigned parasitized and nonparasitized darters to exposure to one of two stimuli: (1) conspecific alarm cue, or (2) control water. The alarm cues used in this experiment were collected using standard procedures for darters (e.g. Smith 1979; Haney et al. 2001). Donor fish (N ¼ 15) were killed by a blow to the head, and shallow cuts (approximately 5 mm long) were made in the skin of the donor fish (25 cuts per fish) using a razor blade. The carcass of the fish was placed in a glass dish containing 60 ml of dechlorinated tap water and swirled with the razor blade for approximately 60 s. Because we used 20 ml of stimulus per trial, the stimulus collected from one donor fish was used for three trials, and skin extracts were less than 30 min old at the time of use. All behavioural trials were performed at 14 DPI between 1000 and 1600 hours, near the expected completion of glochidial metamorphosis (indicated by drop-off from unpublished data on darters parasitized with Pleas’ mussel glochidia). Immediately before the first trial, we stopped the water circulation in the entire aquarium system so that no stimulus solutions would circulate among tanks; we did not restart the water circulation until all trials were completed (within 6 h). At the beginning of each trial, a stimulus aliquot (20 ml of either the alarm cue or control) was added into the centre of the tank by injecting it through a hole in the top of the tank lid from a syringe attached to a 10 cm length of airline tubing. Immediately following the injection, we used a pipette to add exactly five brine shrimp (in a water solution) to the tank. We then recorded two measures of activity: (1) general activity: the number of moves (each time the centre of the body changed locations; darter movements are typically discrete and easily quantified as hops or ‘darts’ on the substrate) and (2) foraging activity: the number of brine shrimp consumed
819
(maximum of five) for 5 min; decreased foraging is often considered as an antipredator response for fishes (e.g. Morgan 1987; Brown & Smith 1997). Ethical Note For our experiments, we collected adult and juvenile rainbow darters (143 total; unsexed) from the wild using a seine (Missouri Department of Conservation, permits 13966 and 13946). Rainbow darters are widely distributed across the southeastern U. S. and are abundant at our study site. We transported the darters to our laboratory in buckets with aerated stream water. In the laboratory, we exposed 64 darters to a short-term parasitic infestation (approximately 3 weeks long) from mussel glochidia, using mussel species that commonly use rainbow darters as hosts in natural habitats. During our experiments there was some mortality of darters (N ¼ 5e7) due to escape of small individuals from their tanks through gates covering the water out-flow. We manipulated predation risk by exposing darters to chemical alarm cues rather than staging encounters with predators. Damage to the skin is necessary to release the alarm cue, so we euthanized 15 individuals to obtain this cue following the American Veterinary Medical Association (AVMA 2007) guidelines. We minimized the number of darters that were euthanized by using stimuli from each sacrificed darter in multiple trials. After the end of the parasitic period and within 2 months of collection, we returned darters to their home sites; both mussel species are present in these streams. Enrichment of aquaria was not provided because cover (e.g. plants or other shelters) would have interfered with visibility of the focal individuals’ behaviour during testing; holding tanks received gravel substrates, which are present in natural stream habitats. This research was approved by Missouri State University’s Institutional Animal Care and Use Committee (protocol number 2009X). Statistical Analyses Ventilation and foraging data from experiment 1 did not meet the normality assumption of parametric testing, so we used twotailed ManneWhitney U tests to evaluate differences between treatments (parasitized versus nonparasitized) at each time interval. To reduce the probability of a type II error, we minimized the number of statistical tests using the following approach modelled after Zar (1984) for performing nonparametric multiple comparisons. We tabulated the differences between the ranked sums of each treatment at each time period. Then, beginning with the largest difference, we conducted comparison tests in the order of decreasing magnitude. Finally, when a test revealed no statistical difference, we concluded that no differences existed on the remaining days (i.e. those having smaller differences between the ranked sums of each treatment). To determine whether there was a relationship between the intensity of infestation (total number of attached glochidia) and the change in size (body volume) of the fish, we used a nonparametric Spearman rank correlation test. To analyse ventilation rate data in experiment 2 we performed two-tailed ManneWhitney U tests using the procedure described above for data from experiment 1. For foraging data, we used two-way ANOVAs with treatment (parasitized or nonparasitized) and stimulus (alarm or control) as factors; because data were not normally distributed, we first transformed the data with the aligned-rank transformation (ART; Higgins & Tashtoush 1994). We used Minitab 16 to analyse data from both experiments, and we used alpha ¼ 0.05 for statistical significance.
820
A. L. Crane et al. / Animal Behaviour 82 (2011) 817e823
RESULTS
Effects on activity, ventilation, foraging and body size of rainbow darters An average SD of 31.5 23.7 kidneyshell mussel glochidia initially attached per fish. Of these, 13.2 16.7 per fish developed into live juveniles. Mussel drop-off peaked around 17 DPI (Fig. 1). Parasitized darters had significantly higher ventilation rates than nonparasitized darters on 14 DPI (ManneWhitney U test: U ¼ 90, N1 ¼ N2 ¼ 19, P ¼ 0.009; Fig. 2a). Ventilation rates for darters in both treatments tended to decline over the course of the experiment (Fig. 2a), and we attribute this decline to acclimation to the testing conditions. Parasitized darters also differed from nonparasitized darters in feeding trials. Parasitized darters made fewer moves than nonparasitized darters on 3 days: 8 (U ¼ 119.5, N1 ¼ N2 ¼ 20, P ¼ 0.030), 20 (U ¼ 86.5, N1 ¼ N2 ¼ 18, P ¼ 0.017) and 28 (U ¼ 76.5, N1 ¼ 16, N2 ¼ 17, P ¼ 0.033; Fig. 2b). However, the reduction in moves made by parasitized darters did not result in fewer prey consumed (P ¼ 0.631 on the day with the largest rank-sum difference; Fig. 2c). At the conclusion of the experiment, infestation intensity (total number of attached glochidia) was significantly correlated with loss of host body volume (Spearman rank correlation: rS ¼ 0.37, N ¼ 33, P ¼ 0.032; Fig. 3). Mean SD volume lost was 0.138 0.135 ml for parasitized darters and 0.035 0.162 ml for nonparasitized darters.
* 200 175 150 125 100 75 2 30
14 12
28
* *
20 15 * 10 5
2
8 14 20 Days postinoculation
2
8 14 20 Days postinoculation
28
8
No. of prey consumed
(c) 6
4
2
0
28
Figure 2. Experiment 1. Median (a) ventilation rates (opercula beats/60 s), (b) number of moves made and (c) number of prey (brine shrimp) consumed in 2 min for rainbow darters (N ¼ 16e20 per group) parasitized (closed circles) and not parasitized (open circles) with kidneyshell mussel glochidia. Upper quartiles are given for upper points and lower quartiles are given for lower points. Other quartiles were omitted for clarity, but note that upper and lower quartiles around each median are asymmetrical. *P < 0.05.
10 No. of mussels
14 20 Days postinoculation
(b)
0
Effects on activity, ventilation and antipredator behaviour of rainbow darters An average SD of 73.5 35.4 Pleas’ mussel glochidia initially attached per fish. Of these, 13.2 11.8 developed into live juveniles. Mussel drop-off peaked around 16 DPI (Fig. 4). Parasitized darters had higher ventilation rates than nonparasitized darters throughout most of the experiment, but this difference was significant only at 2 DPI (U ¼ 12.5, N1 ¼ N2 ¼ 10, P ¼ 0.005; Fig. 5). For the number of prey consumed, a significant interaction term indicated that the parasitized darters did not respond to the alarm cue as strongly as nonparasitized darters (ANOVA: F1,43 ¼ 9.35, P ¼ 0.003). Whereas nonparasitized darters greatly reduced foraging
8
25
Experiment 2: Parasitism by Pleas’ Mussels
8 6 4 2 0 0
(a)
225
No. of moves
Experiment 1: Parasitism by Kidneyshell Mussels
Ventilation rate (per min)
250
2
4
6
8 10 12 14 16 Days postinoculation
18
20
22
Figure 1. Experiment 1. Mean SE numbers of kidneyshell mussel glochidia (closed circles) and live juveniles (open circles) recovered from host fish (N ¼ 44) at 2-day intervals.
in the alarm cue treatment, parasitized darters foraged at nearly the same levels in the alarm treatment and control treatments (Fig. 6a). There was a significant main effect of the alarm cue (ANOVA: F1,43 ¼ 6.09, P ¼ 0.016) and no main effect of the infestation (ANOVA: F1,43 ¼ 1.64, P ¼ 0.204). For the number of moves, only the alarm cue effect was statistically significant (alarm cue treatment: ANOVA: F1,43 ¼ 4.52, P ¼ 0.036; infestation treatment: ANOVA: F1,43 ¼ 2.79, P ¼ 0.098; interaction term: ANOVA: F1,43 ¼ 0.77, P ¼ 0.383; Fig. 6b).
A. L. Crane et al. / Animal Behaviour 82 (2011) 817e823
200 rS = −0.37, P = 0.032
0.4 0.2 0 −0.2 −0.4 −0.6
0
20
40
60
80
100
Ventilation rate (per min)
Changing in volume (ml)
0.6
180
*
160 140 120 100 80 60 40
2
4
6
8
10 12 14 16 18 20 22
Days postinoculation
Infestation intensity
DISCUSSION Ventilation rates of darters in our experiments were increased by glochidial infestations, either at the height of the infestation or early in the infestation. In experiment 1 (kidneyshell mussels), ventilation rates of parasitized darters were significantly higher than nonparasitized darters just prior to glochidial drop-off (Fig. 2a), whereas in experiment 2 (Pleas’ mussels), differences between the parasitized and nonparasitized fish were evident at the start of the infestation, and parasitized darters tended to have somewhat higher ventilation rates until after glochidial drop-off (Fig. 5). This difference may have been due to the glochidal species used or to the infestation intensity, which was about two times higher in experiment 2. Compared to our study, Kaiser (2005) found that increased ventilation rates of glochidia-parasitized fish persisted longer (both throughout the infestation period and even 3 months postinfestation), perhaps because infestation intensities (per gram of fish) were, on average, 53.5% higher than in our experiments, or because different host and glochidial species were used in that study. Increased ventilation rates of parasitized fishes appear to be an effort to compensate for decreased oxygen conductance due to gill damage (Kaiser 2005; Howerth & Keller 2006). In Kaiser (2005), increased ventilation did not completely compensate for the damage because oxygen consumption decreased despite an increase in ventilation rate.
Figure 5. Experiment 2. Median ventilation rates (opercula beats/60 s) for rainbow darters (N ¼ 10 per group) parasitized (closed circles) and not parasitized (open circles) with Pleas’ mussel glochidia. Upper quartiles are given for upper points and lower quartiles are given for lower points. Other quartiles were omitted for clarity, but note that upper and lower quartiles around each median are asymmetrical. *P < 0.05.
Our experimental inoculations for Pleas’ mussel glochidia produced infestation intensities that were at the high end of the range of natural infestation levels reported by Riusech & Barnhart (1998). For kidneyshell glochidia, no data have been reported on natural infestation rates, but our experimental rates were lower for kidneyshell mussel glochidia than for Pleas’ mussels. Therefore, it is likely that our infestation rates have relevance for natural populations.
70
No. of prey consumed (ART transformed)
Figure 3. Experiment 1. Effect of infestation intensity (total number of attached glochidia) on change in body size (displacement volume) of 33 rainbow darters.
No. of moves (ART transformed)
8 6 4
(a) Two-way ANOVA Parasite treatment: P = 0.204 Chemical stimuli: P = 0.016* 60 Interaction: P = 0.003* 50
40
30
20 80
10
No. of mussels
821
(b) Two-way ANOVA Parasite treatment: P = 0.098 70 Chemical stimuli: P = 0.036* Interaction: P = 0.383 60 50 40
2 30 0
32 10
12
14 16 18 Days postinoculation
20
Figure 4. Experiment 2. Mean SE numbers of Pleas’ mussel glochidia (closed circles) and live juveniles (open circles) recovered from host fish (N ¼ 44) at 2-day intervals.
Parasitized
Nonparasitized
Figure 6. Experiment 2. Mean SE transformed (a) number of prey (brine shrimp) consumed and (b) number of moves made in 5 min by parasitized and nonparasitized rainbow darters (N ¼ 22 per group) exposed to alarm cues (solid bar) or control water (open bar) on postinoculation day 14. An asterisk denotes statistical significance.
822
A. L. Crane et al. / Animal Behaviour 82 (2011) 817e823
In both of our experiments, parasitized and nonparasitized darters consumed similar amounts of prey under conditions of low predation risk (Figs 2c, 6a). However, parasitized darters in experiment 1 reduced activity (number of moves; Fig. 2b), presumably to conserve energy to fight the negative effects of parasitism (Hart 1988). Qualitatively, parasitized darters appeared to adopt a sit-and-wait strategy whereas nonparasitized darters were more active. Reduced activity may indicate energy conservation by parasitized individuals. Why would nonparasitized darters fail to adopt a similar energy conservation strategy? We hypothesize that prey are likely to be less readily available in natural stream habitats than in our laboratory trials, resulting in lower overall foraging success for less active foragers under natural conditions. It is puzzling that parasitism did not also affect activity in experiment 2 (Fig. 6b), but feeding regimes (different prey types and more variable delivery and amounts) and infestation intensities differed between the two experiments and these differences may have influenced the results. Although parasitized darters presumably used less energy to capture prey, they still showed substantial loss of body size (volume) over the course of the study in comparison to nonparasitized darters (Fig. 6b). Losses in body size may have resulted from the energetic cost of increased ventilation (Kaiser 2005; this study) and/or increased immune function (Dodd et al. 2005). Fry of the coho salmon, Oncorhynchus kisutch, parasitized with glochidia of the freshwater mussel Anodonta oregonensis also showed substantial weight loss, with heavily parasitized individuals failing to survive (Moles 1983). Low survival following experimental infestation of glochidia has also been reported for other salmonid fishes (Myers & Millemann 1977). Other single-host parasites also have been shown to cause reduced growth (Finley & Forrester 2003), loss of body fat (Tocque 1993) and decreased long-term survival of hosts (Brown et al. 1995). Parasitized darters, unlike controls, did not decrease their rate of foraging when threatened with predation (i.e. exposed to a chemical alarm cue). In natural habitats, this strategy could increase the risk of predation via decreased vigilance and increased visibility to predators. Results of the relatively few studies on the effects of single-host parasites on foraging behaviour have been mixed. Both threespine sticklebacks, Gasterosteus aculeatus (Milinski 1985) and pink salmon, Oncorhynchus gorbuscha (Krkosek et al. 2011) increased risk taking to obtain foraging opportunities, similar to the darters in our study. However, studies of other species have reported decreases in overall foraging success by parasitized individuals (Maksimowich & Mathis 2000; Finley & Forrester 2003). Our results suggest that fishes parasitized with glochidia may experience decreased fitness through weight loss and decreased predatory vigilance resulting from an increased focus on foraging to compensate for increased energetic demands. Mortality of glochidia-parasitized darters under laboratory conditions is generally negligible (M. C. Barnhart, unpublished data); however, the effects of energetic demands, weight loss and vulnerability to predation could have more serious fitness consequences under natural conditions. Acknowledgments We thank Missouri State University for financial support, and Mike Pillow, Jingjing Miao and Zac Beussink for help with experimental observations and collecting. References AVMA (American Veterinary Medical Association) 2007. Guidelines on Euthanasia. http://www.avma.org. Bakker, T. C. M., Mazzi, D. & Zala, S. 1997. Parasite-induced changes in behaviour and color make Gammarus pulex more prone to fish predation. Ecology, 78, 1098e1104.
Barber, I., Hoare, D. & Krause, J. 2000. Effects of parasites on fish behaviour: a review and evolutionary perspective. Reviews in Fish Biology and Fisheries, 10, 131e165. Barnhart, M. C. & Roberts, A. 1997. Reproduction and fish hosts of unionids from the Ozark uplift. In: Conservation and Management of Freshwater Mussels II: Initiatives for the Future (Ed. by K. S. Cummings, A. C. Buchanan, C. A. Mayer & T. J. Naimo), pp. 16e20. Rock Island, Illinois: Upper Mississippi River Conservation Committee. Barnhart, M. C., Haag, W. R. & Roston, W. N. 2008. Adaptations to host infestation and larval parasitism in Unionoida. Journal of the North American Benthological Society, 27, 370e394. Begon, M., Harper, J. L. & Townsend, C. R. 1990. Ecology: Individuals, Populations and Communities. 2nd edn. Boston, Massachusetts: Blackwell Scientific. Brown, C. E., Brown, M. B. & Rannala, B. 1995. Ectoparasites reduce long-term survival of their avian host. Proceedings of the Royal Society B, 262, 313e319. Brown, G. E. & Smith, R. J. F. 1997. Conspecific skin extract elicits antipredator behaviour in juvenile rainbow trout (Oncorhynchus mykiss). Canadian Journal of Zoology, 75, 1916e1922. Bush, A. O., Fernandez, J. C., Esch, G. W. & Seed, J. R. 2001. Parasitism: the Diversity and Ecology of Animal Parasites. New York: Cambridge University Press. Commens, A. M. & Mathis, A. 1999. Alarm pheromones of rainbow darters: responses to skin extracts of conspecifics and congeners. Journal of Fish Biology, 55, 1359e1362. Crane, A. L., Woods, D. & Mathis, A. 2009. Behavioural responses to alarm cues by free-ranging rainbow darters (Etheostoma caeruleum). Behaviour,146,1565e1572. Dodd, B. J., Barnhart, M. C., Rogers-Lowery, C. L., Fobian, T. B. & Dimock, R. V., Jr. 2005. Cross-resistance of largemouth bass to glochidia of unionid mussels. Journal of Parasitology, 95, 1064e1072. Finley, R. J. & Forrester, G. E. 2003. Impact of ectoparasites on the demography of a small reef fish. Marine Ecology Progress Series, 248, 305e309. Fox, A. & Hudson, P. J. 2001. Parasites reduce territorial behaviour in red grouse (Lagopus lagopus scoticus). Ecology Letters, 4, 139e143. Haney, D. C., Vokoun, J. C. & Noltie, D. B. 2001. Alarm pheromone recognition in a Missouri darter assemblage. Journal of Fish Biology, 59, 810e817. Hart, B. L. 1988. Biological basis of the behavior of sick animals. Neuroscience and Biobehavioral Reviews, 12, 123e137. Higgins, J. J. & Tashtoush, S. 1994. An aligned rank transform test for interaction. Nonlinear World, 1, 201e211. Howerth, E. W. & Keller, A. E. 2006. Experimentally induced glochidiosis in smallmouth bass (Micropterus dolomieu). Veterinary Pathology, 43, 1004e1007. Kaiser, B. E. 2005. The effects of glochidiosis on fish respiration. M.S. thesis, Missouri State University. Krkosek, M., Connors, B. M., Ford, H., Peacock, S., Mages, P., Ford, J. S., Morton, A., Volpe, J. P., Hilborn, R. & Dill, L. M., et al. 2011. Fish farms, parasites, and predators: implications for salmon population dynamics. Ecological Applications, 21, 897e914. Lafferty, K. D. & Morris, A. K. 1996. Altered behaviour of parasitized killifish increases susceptibility to predation by bird final hosts. Ecology, 77, 1390e1397. LaMunyon, C. & Eisner, T. 1990. Effect of mite infestation on the anti-predator defenses of an insect. Psyche, 97, 31e41. Lefever, G. & Curtis, W. C. 1912. Studies on the reproduction of artificial propagation of freshwater mussels. U. S. Bureau of Fisheries Bulletin, 30, 105e209. Lima, S. L. 1998. Stress and decision making under the risk of predation: recent developments from behavioral, reproductive, and ecological processes. Advances in the Study of Behavior, 27, 215e290. Loot, G., Aulagnier, S., Lek, S., Thomas, F. & Guégan, J. 2002. Experimental demonstration of a behavioural modification in a cyprinid fish, Rutilus rutilus (L.), induced by a parasite, Ligula intestinalis (L.). Canadian Journal of Zoology, 80, 738e744. Maksimowich, D. S. & Mathis, A. 2000. Parasitized salamanders are inferior competitors for territories and food resources. Ethology, 106, 319e329. Mathis, A. 2009. Alarm responses as a defense: chemical alarm cues in nonostariophysan fishes. In: Fish Defenses. Vol. 2: Pathogens, Parasites and Predators (Ed. by G. Zaccone, C. Perrière & A. Mathis), pp. 323e386. Enfield, New Hampshire: Science. Milinski, M. 1985. Risk of predation of parasitized sticklebacks (Gasterosteus aculeatus L.) under competition for food. Behaviour, 93, 203e216. Moles, A. 1983. Effect of parasitism by mussel glochidia on growth of coho salmon. Transactions of the American Fisheries Society, 112, 201e204. Morgan, M. J. 1987. The influence of hunger, shoal size and predator presence on foraging in bluntnose minnows. Animal Behaviour, 36, 1317e1322. Myers, T. R. & Millemann, R. E. 1977. Glochidiosis of salmonid fishes. I. Comparative susceptibility to experimental infestation with Margaritifera margaritifera (L.) (Pelecypoda: Margaritanidae). Journal of Parasitology, 63, 728e733. Northcott, S. J., Lyndon, A. R. & Campbell, A. D. 2003. An outbreak of freshwater fish lice, Argulus foliaceus L., seriously affecting a Scottish stillwater fishery. Fisheries Management and Ecology, 4, 73e75. Oesch, R. D. 1984. Missouri Naiads: a Guide to the Mussels from Missouri. Jefferson City: Missouri Department of Conservation. Parris, M. J., Reese, E. & Storfer, A. 2006. Antipredator behaviour of chytridiomycosis-infested northern leopard frog (Rana pipiens) tadpoles. Canadian Journal of Zoology, 84, 58e65. Pélabon, C., Borg, A. A., Bjelvenmark, J., Barber, I., Forsgren, E. & Trond Amundsen, T. 2005. Do microsporidian parasites affect courtship in two-spotted gobies? Marine Biology, 148, 1432e1793.
A. L. Crane et al. / Animal Behaviour 82 (2011) 817e823 Riusech, F. A. & Barnhart, M. C. 1998. Host suitability in Venustaconcha ellipsiformis and Venustaconcha pleasii (Bivalvia: Unionidae) from Ozark plateaus. In: Proceedings of the Conservation, Captive, and Propagation of Freshwater Mussels Symposium (Ed. by T. A. Tankersley, T. Watters, B. Armitage & D. Warmolts), pp. 83e91. Columbus: Ohio University Press. Rogers-Lowery, C. L. & Dimock, R. V., Jr. 2006. Encapsulation of attached ectoparasitic larvae of freshwater mussels by epithelial tissue on fins of native and resistant host fish. Biological Bulletin, 210, 51e63. Sasal, P. 2006. Nest guarding in a damselfish: evidence of a role for parasites. Journal of Fish Biology, 68, 1215e1221. Smith, R. J. F. 1979. Alarm reaction of Iowa and Johnny darters (Etheostoma, Percidae, Pisces) to chemicals from injured conspecifics. Canadian Journal of Zoology, 57, 1278e1282. Smith, R. J. F. 1992. Alarm signals in fishes. Reviews in Fish Biology and Fisheries, 2, 33e63.
823
Smith Trail, D. R. 1980. Behavioural interactions between parasites and hosts: host suicide and the evolution of complex life cycles. American Naturalist, 116, 77e91. Tocque, K. 1993. The relationship between parasite burden and host resources in the desert toad (Scaphiopus couchii), under natural environmental conditions. Journal of Animal Ecology, 62, 683e693. Vaughan, G. E. & Coble, D. W. 1975. Sublethal effects of three ectoparasites on fish. Journal of Fish Biology, 7, 283e294. Waller, D. L. & Mitchell, L. G.1989. Gill tissue reactions in walleye Stizostedion vitreum vitreum and common carp Cyprinus carpio to glochidia of the freshwater mussel Lampsilis radiata siliquoidea. Diseases of Aquatic Organisms, 6, 81e87. Woods, D. A. 2008. Sub-lethal effects of copper on foraging and alarm behaviour of rainbow darters (Etheostoma caeruleum). M.S. thesis, Missouri State University. Zar, J. H. 1984. Biostatistical Analysis. 4th edn. Upper Saddle River, New Jersey: Prentice Hall.
Contents lists available at ScienceDirect
Animal Behaviour journal homepage: www.elsevier.com/locate/anbehav
Do gill parasites influence the foraging and antipredator behaviour of rainbow darters, Etheostoma caeruleum? Adam L. Crane*, Andrea K. Fritts, Alicia Mathis, John C. Lisek, M. Chris Barnhart Department of Biology, Missouri State University
a r t i c l e i n f o Article history: Received 24 February 2011 Initial acceptance 9 May 2011 Final acceptance 27 June 2011 Available online 11 August 2011 MS. number: A11-00164R Keywords: alarm cue Etheostoma caeruleum glochidia parasite rainbow darter Unionidae
Parasites are known to affect an array of characteristics of their hosts, including morphology, physiology and behaviour. We examined the foraging and antipredator behaviour of rainbow darters, Etheostoma caeruleum, that were parasitized by glochidia larvae of freshwater mussels (Ptychobranchus occidentalis and Venustaconcha pleasii: Unionidae). Glochidia attach to the gills of the host and become encapsulated in host tissue. Over a period of days or weeks the larvae develop into free-living juveniles, which then leave the host. Parasitized darters increased ventilation rates (either early in the infestation or at the height of the infestation), were less active during foraging trials, lost more body size than nonparasitized darters and showed significantly weaker responses to predation risk (signalled by the presence of a chemical alarm cue). Therefore, even for a relatively short-term infection, parasitized darters may pay a cost in terms of decreased growth and decreased probability of survival. Ó 2011 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.
Parasites are defined partly by their production of negative impacts on their hosts (Begon et al. 1990), although the magnitude of these impacts can be highly variable. Increased predation of intermediate hosts due to manipulation by parasites can increase parasite fitness by facilitating their transfer to the next host (e.g. Lafferty & Morris 1996; Bakker et al. 1997; Loot et al. 2002). In contrast, single-host parasites are not expected to manipulate their hosts to increase predation risk, because death of the host would also result in death of the parasite (Smith Trail 1980). Nevertheless, single-host parasites can harm the host by co-opting host resources (overviews: Barber et al. 2000; Bush et al. 2001), and can lead to death of the host when infestations are particularly heavy (e.g. Brown et al. 1995; Northcott et al. 2003). Negative effects of single-host parasites can include anatomical and physiological effects (Bush et al. 2001) and/or changes in behaviour that reduce success of activities such as foraging (Maksimowich & Mathis 2000; Finley & Forrester 2003), aggressive competition (Fox & Hudson 2001), courtship (Pélabon et al. 2005) and parental care (Sasal 2006). Of the studies that have examined the influence of single-host parasites on antipredator behaviour, the results have not been consistent, ranging from no effects on * Correspondence: A. L. Crane, Department of Biology, Missouri State University, 901 S. National, Springfield, MO 65897, U.S.A. E-mail address: [email protected] (A. L. Crane).
predation avoidance (Vaughan & Coble 1975) to increased levels of antipredator responses (Milinski 1985; Parris et al. 2006) to decreased levels of antipredator responses (LaMunyon & Eisner 1990; Krkosek et al. 2011). The life cycle of freshwater mussels (Unionidae) includes a parasitic life stage where large numbers of larvae (i.e. glochidia) attach to fish hosts, usually on the gills (Barnhart et al. 2008). These single-host parasites have evolved specificity to their fish hosts, which are used for dispersal and development (Waller & Mitchell 1989). During the infestation process, glochidia become encapsulated in the hosts’ tissues and then metamorphose over a period of days into free-living juveniles (Rogers-Lowery & Dimock 2006; Barnhart et al. 2008). Little is known of the physiological effects of glochidial infestation on host fish. However, Kaiser (2005) found that respiratory gas exchange of largemouth bass, Micropterus salmoides, was impaired by an infestation of glochidia from the broken ray mussels, Lampsilis reeveiana, resulting in elevated resting ventilation rates, reduced oxygen consumption and lower tolerance of low oxygen conditions. These effects were dependent on intensity of infestation (number of attached glochidia). In extreme cases glochidial infestations can cause mortality (Howerth & Keller 2006); however, the more typical effects that have been observed include the energetic costs of immune responses (Dodd et al. 2005) and/or increased ventilation rates, which appear to be due to reduced surface area for gas exchange over the gills (Kaiser 2005).
0003-3472/$38.00 Ó 2011 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.anbehav.2011.07.015
818
A. L. Crane et al. / Animal Behaviour 82 (2011) 817e823
The objective in this study was to examine the effects of glochidia on the behaviour and growth of a host fish, the rainbow darter, Etheostoma caeruleum. Our study involved two experiments that occurred over the course of approximately 5 months. We used two species of mussels in two experiments (experiment 1: kidneyshell mussel, Ptychobranchus occidentalis; experiment 2: Pleas’ mussel, Venustaconcha pleasii) based on species availability. Both mussels are common at sites in the smaller tributaries of the White River system in the Missouri Ozarks (Oesch 1984) and both use darters and sculpin as the primary hosts for their glochidia (Barnhart & Roberts 1997; Riusech & Barnhart 1998). For both species of mussels, glochidia are produced in the autumn and released in the spring, and females are barren of glochidia during JuneeSeptember (M. C. Barnhart, personal observations). At sites where mussels are abundant, glochidial infestations of darters are common. For example, over 60% of rainbow darters examined from a site in the James River in May were parasitized with Pleas’ mussel glochidia (infestation intensity: mean SD ¼ 16 16.7, range 1e73; Riusech & Barnhart 1998). In our first experiment (kidneyshell mussels), we made four predictions. (1) Because of damage to the host’s gills or reduction in gill surface area available for respiration (Kaiser 2005), ventilation rates should be higher for parasitized darters. (2) Because of respiratory stress, activity levels of parasitized individuals should decrease. (3) Reduced activity of darters should result in lower foraging success in behavioural trials. (4) Reduced foraging success should lead to decreased growth. In the second experiment (Pleas’ mussels), we made two predictions. (1) Increased ventilation rates should also increase in response to infestation by this species. (2) If parasitism has energetic costs, then parasitized darters should be less sensitive to predation risk while foraging (Lima 1998) than nonparasitized darters, which have been observed to decrease foraging activity under conditions of high predation risk (Woods 2008); thus, we predicted that parasitized darters would not decrease foraging behaviour under conditions of high predation risk.
METHODS Experiment 1: Parasitism by Kidneyshell Mussels Effects on activity, ventilation, foraging and body size of rainbow darters In March 2009, we collected 40 rainbow darters and one adult kidneyshell mussel from the James River in Greene County, Missouri, U.S.A. We did not check whether fish were previously parasitized with glochidia; however, no exogenous glochidia were recovered from the darters during monitoring (details below), indicating that they did not carry glochidia when they were collected. To produce infestation levels realistic for wild fishes, we used standard methods (Riusech & Barnhart 1998). This mussel species releases its glochidia in membranous packets (i.e. conglutinates). We separated glochidia from the conglutinates under a dissecting microscope. We counted the viable glochidia in 10 200 ml subsamples of this volume and extrapolated the numbers in these samples to determine the total number of glochidia. Viable glochidia were identified by their closing response to salt (Lefever & Curtis 1912). We then parasitized 20 darters by placing them in a 4-litre container with approximately 4000 viable glochidia per liter for 15 min. During this time we used a rubber-bulb 25 ml syringe (turkey baster) to gently agitate the water and keep the glochidia uniformly suspended. An additional 20 darters served as a control group; they experienced the stress of a sham infestation that followed the same methods, but no glochidia were added.
Immediately following the exposure period, darters from both treatment groups (parasitized and nonparasitized) were moved to individual 1.5-litre monitoring containers (Aquatic Habitats, Inc., Apopka, FL, U.S.A.) that received continuously flowing water (ca. 26 C). Water from each container flowed into a common holding chamber, and a sump pump was used to pump the water through two filters (UV sterilizer and a 5 mm particulate filter) before it was recirculated. Although some chemical stimuli from individual darters may have remained in the recirculating water, these chemicals should have been distributed equally to darters in each container. To capture adult glochidia that did not remain attached to gills and to capture juveniles that detached after transformation, the outflow from each container entered a corresponding filter cup (125 mm mesh); our methods were similar to those of Dodd et al. (2005). The minimum size of glochidia for the species that we used in this study was 200 mm (Barnhart et al. 2008), so it is unlikely that any glochidia would pass through the filter. However, any glochidia or metamorphosed juveniles that escaped would be removed by the filters attached to the common holding chamber before the water was recirculated. At 8 days postinoculation (DPI) we began examining the filter cups of the parasitized darters by temporarily stopping the water flow, removing each filter cup, and rinsing its contents into a Bogorov tray placed under a microscope. We then counted the unmetamorphosed glochidia and metamorphosed juveniles. After 8 days, we checked the filters and counted the recovered mussels at 2-day intervals until no more were recovered (ca. 3 weeks). Because behavioural and physiological effects can be dependent on infestation intensity (Kaiser 2005), we calculated the intensity of the infestation that each fish experienced (i.e. total number of attached glochidia) by the total numbers of glochidia and metamorphosed juveniles that we recovered. Immediately following exposure to glochidia, and also at the end of the experiment, we measured the volume of darters as the amount of water that each darter displaced in a 5 ml graduated cylinder (0.1 ml precision). We used volumetric displacement in water as a measure of body size because it is likely to be less stressful to fish than other measures that require a longer removal from the water. Beginning on 2 DPI, we fed darters with brine shrimp, Artemia salina (1 ml of brine shrimp solution, ca. 10 total brine shrimp), every other day between 1500 and 1700 hours. We conducted behavioural trials on 5 different days that included both during (DPI: 2, 8, 14, 20) and after (DPI: 28) the parasitic period. Before the trials began, we lined the outer front of the tanks with dark window tinting and placed lights behind the tanks to allow us to see the darters easily while minimizing our visibility to the darters. We also placed cardboard barriers between the tanks to eliminate visual cues between adjacent fish. Each behavioural trial first consisted of recording the ventilation rate for each darter prior to feeding. At the beginning of the observation period, we stood quietly approximately 0.25 m in front of tanks. Most individuals were relatively stationary in the tanks. Once we were confident that we could clearly see at least one operculum (within 0e2 min), we counted ventilation rates for 60 s. Then, we immediately fed each darter and observed it for 120 s while recording the number of moves and the total number of brine shrimp consumed. These feeding trials were done in accordance with the darters’ routine feeding schedule. After the trials on 28 DPI, we measured the volumetric displacement for each darter. Experiment 2: Parasitism by Pleas’ Mussels Effects on activity, ventilation and antipredator behaviour of rainbow darters During December of 2008, we collected 103 rainbow darters from Bull Creek in Taney County, Missouri, U.S.A. After 2 days in
A. L. Crane et al. / Animal Behaviour 82 (2011) 817e823
captivity, we parasitized darters with glochidia from a single gravid Pleas’ mussel collected from the James River in Greene County, Missouri, U.S.A. This species of mussel holds its glochidia inside the marsupial gills. To extract glochidia, we used a hooked probe to cut the marsupial gills of the mussel and flushed the glochidia into a measured volume of water in a collection beaker. All other procedures for infecting darters with glochidia were identical to experiment 1. We parasitized 44 darters and gave a sham infestation to a group of 44 control darters (see experiment 1 methods for details). Again, no exogenous glochidia were recovered from the fish during monitoring. The set-up of tanks was as in experiment 1 except that the water temperature was 24 C. We fed darters with blackworms (Lumbriculus variegatus) or brine shrimp (A. salina) in a water solution at 2-day intervals between 0900 and 1700 hours throughout the time they were housed in the monitoring containers. The amount of prey varied depending on the amount that individuals could consume in approximately 5 min. Two days following the infestation, we began monitoring ventilation rates (i.e. opercula beats) on a subset of randomly selected darters (10 parasitized and 10 nonparasitized). We repeated measurements on the same darters between 1500 and 1700 hours at 2-day intervals for 22 days. For tests of antipredator responses, we used a chemical alarm cue as a stimulus indicating predation risk. Chemical alarm cues are contained in the skin of many fishes and can be released when a fish’s skin is damaged during a predator attack; nearby conspecifics respond with antipredator behaviour (reviewed in: Smith 1992; Mathis 2009). Rainbow darters have been shown to respond to conspecific alarm cues with reduced activity in both laboratory (Commens & Mathis 1999) and field (Crane et al. 2009) studies; these studies also showed that the antipredator responses are specific to conspecific extract and are not responses to general disturbance or to extracts from other species. We randomly assigned parasitized and nonparasitized darters to exposure to one of two stimuli: (1) conspecific alarm cue, or (2) control water. The alarm cues used in this experiment were collected using standard procedures for darters (e.g. Smith 1979; Haney et al. 2001). Donor fish (N ¼ 15) were killed by a blow to the head, and shallow cuts (approximately 5 mm long) were made in the skin of the donor fish (25 cuts per fish) using a razor blade. The carcass of the fish was placed in a glass dish containing 60 ml of dechlorinated tap water and swirled with the razor blade for approximately 60 s. Because we used 20 ml of stimulus per trial, the stimulus collected from one donor fish was used for three trials, and skin extracts were less than 30 min old at the time of use. All behavioural trials were performed at 14 DPI between 1000 and 1600 hours, near the expected completion of glochidial metamorphosis (indicated by drop-off from unpublished data on darters parasitized with Pleas’ mussel glochidia). Immediately before the first trial, we stopped the water circulation in the entire aquarium system so that no stimulus solutions would circulate among tanks; we did not restart the water circulation until all trials were completed (within 6 h). At the beginning of each trial, a stimulus aliquot (20 ml of either the alarm cue or control) was added into the centre of the tank by injecting it through a hole in the top of the tank lid from a syringe attached to a 10 cm length of airline tubing. Immediately following the injection, we used a pipette to add exactly five brine shrimp (in a water solution) to the tank. We then recorded two measures of activity: (1) general activity: the number of moves (each time the centre of the body changed locations; darter movements are typically discrete and easily quantified as hops or ‘darts’ on the substrate) and (2) foraging activity: the number of brine shrimp consumed
819
(maximum of five) for 5 min; decreased foraging is often considered as an antipredator response for fishes (e.g. Morgan 1987; Brown & Smith 1997). Ethical Note For our experiments, we collected adult and juvenile rainbow darters (143 total; unsexed) from the wild using a seine (Missouri Department of Conservation, permits 13966 and 13946). Rainbow darters are widely distributed across the southeastern U. S. and are abundant at our study site. We transported the darters to our laboratory in buckets with aerated stream water. In the laboratory, we exposed 64 darters to a short-term parasitic infestation (approximately 3 weeks long) from mussel glochidia, using mussel species that commonly use rainbow darters as hosts in natural habitats. During our experiments there was some mortality of darters (N ¼ 5e7) due to escape of small individuals from their tanks through gates covering the water out-flow. We manipulated predation risk by exposing darters to chemical alarm cues rather than staging encounters with predators. Damage to the skin is necessary to release the alarm cue, so we euthanized 15 individuals to obtain this cue following the American Veterinary Medical Association (AVMA 2007) guidelines. We minimized the number of darters that were euthanized by using stimuli from each sacrificed darter in multiple trials. After the end of the parasitic period and within 2 months of collection, we returned darters to their home sites; both mussel species are present in these streams. Enrichment of aquaria was not provided because cover (e.g. plants or other shelters) would have interfered with visibility of the focal individuals’ behaviour during testing; holding tanks received gravel substrates, which are present in natural stream habitats. This research was approved by Missouri State University’s Institutional Animal Care and Use Committee (protocol number 2009X). Statistical Analyses Ventilation and foraging data from experiment 1 did not meet the normality assumption of parametric testing, so we used twotailed ManneWhitney U tests to evaluate differences between treatments (parasitized versus nonparasitized) at each time interval. To reduce the probability of a type II error, we minimized the number of statistical tests using the following approach modelled after Zar (1984) for performing nonparametric multiple comparisons. We tabulated the differences between the ranked sums of each treatment at each time period. Then, beginning with the largest difference, we conducted comparison tests in the order of decreasing magnitude. Finally, when a test revealed no statistical difference, we concluded that no differences existed on the remaining days (i.e. those having smaller differences between the ranked sums of each treatment). To determine whether there was a relationship between the intensity of infestation (total number of attached glochidia) and the change in size (body volume) of the fish, we used a nonparametric Spearman rank correlation test. To analyse ventilation rate data in experiment 2 we performed two-tailed ManneWhitney U tests using the procedure described above for data from experiment 1. For foraging data, we used two-way ANOVAs with treatment (parasitized or nonparasitized) and stimulus (alarm or control) as factors; because data were not normally distributed, we first transformed the data with the aligned-rank transformation (ART; Higgins & Tashtoush 1994). We used Minitab 16 to analyse data from both experiments, and we used alpha ¼ 0.05 for statistical significance.
820
A. L. Crane et al. / Animal Behaviour 82 (2011) 817e823
RESULTS
Effects on activity, ventilation, foraging and body size of rainbow darters An average SD of 31.5 23.7 kidneyshell mussel glochidia initially attached per fish. Of these, 13.2 16.7 per fish developed into live juveniles. Mussel drop-off peaked around 17 DPI (Fig. 1). Parasitized darters had significantly higher ventilation rates than nonparasitized darters on 14 DPI (ManneWhitney U test: U ¼ 90, N1 ¼ N2 ¼ 19, P ¼ 0.009; Fig. 2a). Ventilation rates for darters in both treatments tended to decline over the course of the experiment (Fig. 2a), and we attribute this decline to acclimation to the testing conditions. Parasitized darters also differed from nonparasitized darters in feeding trials. Parasitized darters made fewer moves than nonparasitized darters on 3 days: 8 (U ¼ 119.5, N1 ¼ N2 ¼ 20, P ¼ 0.030), 20 (U ¼ 86.5, N1 ¼ N2 ¼ 18, P ¼ 0.017) and 28 (U ¼ 76.5, N1 ¼ 16, N2 ¼ 17, P ¼ 0.033; Fig. 2b). However, the reduction in moves made by parasitized darters did not result in fewer prey consumed (P ¼ 0.631 on the day with the largest rank-sum difference; Fig. 2c). At the conclusion of the experiment, infestation intensity (total number of attached glochidia) was significantly correlated with loss of host body volume (Spearman rank correlation: rS ¼ 0.37, N ¼ 33, P ¼ 0.032; Fig. 3). Mean SD volume lost was 0.138 0.135 ml for parasitized darters and 0.035 0.162 ml for nonparasitized darters.
* 200 175 150 125 100 75 2 30
14 12
28
* *
20 15 * 10 5
2
8 14 20 Days postinoculation
2
8 14 20 Days postinoculation
28
8
No. of prey consumed
(c) 6
4
2
0
28
Figure 2. Experiment 1. Median (a) ventilation rates (opercula beats/60 s), (b) number of moves made and (c) number of prey (brine shrimp) consumed in 2 min for rainbow darters (N ¼ 16e20 per group) parasitized (closed circles) and not parasitized (open circles) with kidneyshell mussel glochidia. Upper quartiles are given for upper points and lower quartiles are given for lower points. Other quartiles were omitted for clarity, but note that upper and lower quartiles around each median are asymmetrical. *P < 0.05.
10 No. of mussels
14 20 Days postinoculation
(b)
0
Effects on activity, ventilation and antipredator behaviour of rainbow darters An average SD of 73.5 35.4 Pleas’ mussel glochidia initially attached per fish. Of these, 13.2 11.8 developed into live juveniles. Mussel drop-off peaked around 16 DPI (Fig. 4). Parasitized darters had higher ventilation rates than nonparasitized darters throughout most of the experiment, but this difference was significant only at 2 DPI (U ¼ 12.5, N1 ¼ N2 ¼ 10, P ¼ 0.005; Fig. 5). For the number of prey consumed, a significant interaction term indicated that the parasitized darters did not respond to the alarm cue as strongly as nonparasitized darters (ANOVA: F1,43 ¼ 9.35, P ¼ 0.003). Whereas nonparasitized darters greatly reduced foraging
8
25
Experiment 2: Parasitism by Pleas’ Mussels
8 6 4 2 0 0
(a)
225
No. of moves
Experiment 1: Parasitism by Kidneyshell Mussels
Ventilation rate (per min)
250
2
4
6
8 10 12 14 16 Days postinoculation
18
20
22
Figure 1. Experiment 1. Mean SE numbers of kidneyshell mussel glochidia (closed circles) and live juveniles (open circles) recovered from host fish (N ¼ 44) at 2-day intervals.
in the alarm cue treatment, parasitized darters foraged at nearly the same levels in the alarm treatment and control treatments (Fig. 6a). There was a significant main effect of the alarm cue (ANOVA: F1,43 ¼ 6.09, P ¼ 0.016) and no main effect of the infestation (ANOVA: F1,43 ¼ 1.64, P ¼ 0.204). For the number of moves, only the alarm cue effect was statistically significant (alarm cue treatment: ANOVA: F1,43 ¼ 4.52, P ¼ 0.036; infestation treatment: ANOVA: F1,43 ¼ 2.79, P ¼ 0.098; interaction term: ANOVA: F1,43 ¼ 0.77, P ¼ 0.383; Fig. 6b).
A. L. Crane et al. / Animal Behaviour 82 (2011) 817e823
200 rS = −0.37, P = 0.032
0.4 0.2 0 −0.2 −0.4 −0.6
0
20
40
60
80
100
Ventilation rate (per min)
Changing in volume (ml)
0.6
180
*
160 140 120 100 80 60 40
2
4
6
8
10 12 14 16 18 20 22
Days postinoculation
Infestation intensity
DISCUSSION Ventilation rates of darters in our experiments were increased by glochidial infestations, either at the height of the infestation or early in the infestation. In experiment 1 (kidneyshell mussels), ventilation rates of parasitized darters were significantly higher than nonparasitized darters just prior to glochidial drop-off (Fig. 2a), whereas in experiment 2 (Pleas’ mussels), differences between the parasitized and nonparasitized fish were evident at the start of the infestation, and parasitized darters tended to have somewhat higher ventilation rates until after glochidial drop-off (Fig. 5). This difference may have been due to the glochidal species used or to the infestation intensity, which was about two times higher in experiment 2. Compared to our study, Kaiser (2005) found that increased ventilation rates of glochidia-parasitized fish persisted longer (both throughout the infestation period and even 3 months postinfestation), perhaps because infestation intensities (per gram of fish) were, on average, 53.5% higher than in our experiments, or because different host and glochidial species were used in that study. Increased ventilation rates of parasitized fishes appear to be an effort to compensate for decreased oxygen conductance due to gill damage (Kaiser 2005; Howerth & Keller 2006). In Kaiser (2005), increased ventilation did not completely compensate for the damage because oxygen consumption decreased despite an increase in ventilation rate.
Figure 5. Experiment 2. Median ventilation rates (opercula beats/60 s) for rainbow darters (N ¼ 10 per group) parasitized (closed circles) and not parasitized (open circles) with Pleas’ mussel glochidia. Upper quartiles are given for upper points and lower quartiles are given for lower points. Other quartiles were omitted for clarity, but note that upper and lower quartiles around each median are asymmetrical. *P < 0.05.
Our experimental inoculations for Pleas’ mussel glochidia produced infestation intensities that were at the high end of the range of natural infestation levels reported by Riusech & Barnhart (1998). For kidneyshell glochidia, no data have been reported on natural infestation rates, but our experimental rates were lower for kidneyshell mussel glochidia than for Pleas’ mussels. Therefore, it is likely that our infestation rates have relevance for natural populations.
70
No. of prey consumed (ART transformed)
Figure 3. Experiment 1. Effect of infestation intensity (total number of attached glochidia) on change in body size (displacement volume) of 33 rainbow darters.
No. of moves (ART transformed)
8 6 4
(a) Two-way ANOVA Parasite treatment: P = 0.204 Chemical stimuli: P = 0.016* 60 Interaction: P = 0.003* 50
40
30
20 80
10
No. of mussels
821
(b) Two-way ANOVA Parasite treatment: P = 0.098 70 Chemical stimuli: P = 0.036* Interaction: P = 0.383 60 50 40
2 30 0
32 10
12
14 16 18 Days postinoculation
20
Figure 4. Experiment 2. Mean SE numbers of Pleas’ mussel glochidia (closed circles) and live juveniles (open circles) recovered from host fish (N ¼ 44) at 2-day intervals.
Parasitized
Nonparasitized
Figure 6. Experiment 2. Mean SE transformed (a) number of prey (brine shrimp) consumed and (b) number of moves made in 5 min by parasitized and nonparasitized rainbow darters (N ¼ 22 per group) exposed to alarm cues (solid bar) or control water (open bar) on postinoculation day 14. An asterisk denotes statistical significance.
822
A. L. Crane et al. / Animal Behaviour 82 (2011) 817e823
In both of our experiments, parasitized and nonparasitized darters consumed similar amounts of prey under conditions of low predation risk (Figs 2c, 6a). However, parasitized darters in experiment 1 reduced activity (number of moves; Fig. 2b), presumably to conserve energy to fight the negative effects of parasitism (Hart 1988). Qualitatively, parasitized darters appeared to adopt a sit-and-wait strategy whereas nonparasitized darters were more active. Reduced activity may indicate energy conservation by parasitized individuals. Why would nonparasitized darters fail to adopt a similar energy conservation strategy? We hypothesize that prey are likely to be less readily available in natural stream habitats than in our laboratory trials, resulting in lower overall foraging success for less active foragers under natural conditions. It is puzzling that parasitism did not also affect activity in experiment 2 (Fig. 6b), but feeding regimes (different prey types and more variable delivery and amounts) and infestation intensities differed between the two experiments and these differences may have influenced the results. Although parasitized darters presumably used less energy to capture prey, they still showed substantial loss of body size (volume) over the course of the study in comparison to nonparasitized darters (Fig. 6b). Losses in body size may have resulted from the energetic cost of increased ventilation (Kaiser 2005; this study) and/or increased immune function (Dodd et al. 2005). Fry of the coho salmon, Oncorhynchus kisutch, parasitized with glochidia of the freshwater mussel Anodonta oregonensis also showed substantial weight loss, with heavily parasitized individuals failing to survive (Moles 1983). Low survival following experimental infestation of glochidia has also been reported for other salmonid fishes (Myers & Millemann 1977). Other single-host parasites also have been shown to cause reduced growth (Finley & Forrester 2003), loss of body fat (Tocque 1993) and decreased long-term survival of hosts (Brown et al. 1995). Parasitized darters, unlike controls, did not decrease their rate of foraging when threatened with predation (i.e. exposed to a chemical alarm cue). In natural habitats, this strategy could increase the risk of predation via decreased vigilance and increased visibility to predators. Results of the relatively few studies on the effects of single-host parasites on foraging behaviour have been mixed. Both threespine sticklebacks, Gasterosteus aculeatus (Milinski 1985) and pink salmon, Oncorhynchus gorbuscha (Krkosek et al. 2011) increased risk taking to obtain foraging opportunities, similar to the darters in our study. However, studies of other species have reported decreases in overall foraging success by parasitized individuals (Maksimowich & Mathis 2000; Finley & Forrester 2003). Our results suggest that fishes parasitized with glochidia may experience decreased fitness through weight loss and decreased predatory vigilance resulting from an increased focus on foraging to compensate for increased energetic demands. Mortality of glochidia-parasitized darters under laboratory conditions is generally negligible (M. C. Barnhart, unpublished data); however, the effects of energetic demands, weight loss and vulnerability to predation could have more serious fitness consequences under natural conditions. Acknowledgments We thank Missouri State University for financial support, and Mike Pillow, Jingjing Miao and Zac Beussink for help with experimental observations and collecting. References AVMA (American Veterinary Medical Association) 2007. Guidelines on Euthanasia. http://www.avma.org. Bakker, T. C. M., Mazzi, D. & Zala, S. 1997. Parasite-induced changes in behaviour and color make Gammarus pulex more prone to fish predation. Ecology, 78, 1098e1104.
Barber, I., Hoare, D. & Krause, J. 2000. Effects of parasites on fish behaviour: a review and evolutionary perspective. Reviews in Fish Biology and Fisheries, 10, 131e165. Barnhart, M. C. & Roberts, A. 1997. Reproduction and fish hosts of unionids from the Ozark uplift. In: Conservation and Management of Freshwater Mussels II: Initiatives for the Future (Ed. by K. S. Cummings, A. C. Buchanan, C. A. Mayer & T. J. Naimo), pp. 16e20. Rock Island, Illinois: Upper Mississippi River Conservation Committee. Barnhart, M. C., Haag, W. R. & Roston, W. N. 2008. Adaptations to host infestation and larval parasitism in Unionoida. Journal of the North American Benthological Society, 27, 370e394. Begon, M., Harper, J. L. & Townsend, C. R. 1990. Ecology: Individuals, Populations and Communities. 2nd edn. Boston, Massachusetts: Blackwell Scientific. Brown, C. E., Brown, M. B. & Rannala, B. 1995. Ectoparasites reduce long-term survival of their avian host. Proceedings of the Royal Society B, 262, 313e319. Brown, G. E. & Smith, R. J. F. 1997. Conspecific skin extract elicits antipredator behaviour in juvenile rainbow trout (Oncorhynchus mykiss). Canadian Journal of Zoology, 75, 1916e1922. Bush, A. O., Fernandez, J. C., Esch, G. W. & Seed, J. R. 2001. Parasitism: the Diversity and Ecology of Animal Parasites. New York: Cambridge University Press. Commens, A. M. & Mathis, A. 1999. Alarm pheromones of rainbow darters: responses to skin extracts of conspecifics and congeners. Journal of Fish Biology, 55, 1359e1362. Crane, A. L., Woods, D. & Mathis, A. 2009. Behavioural responses to alarm cues by free-ranging rainbow darters (Etheostoma caeruleum). Behaviour,146,1565e1572. Dodd, B. J., Barnhart, M. C., Rogers-Lowery, C. L., Fobian, T. B. & Dimock, R. V., Jr. 2005. Cross-resistance of largemouth bass to glochidia of unionid mussels. Journal of Parasitology, 95, 1064e1072. Finley, R. J. & Forrester, G. E. 2003. Impact of ectoparasites on the demography of a small reef fish. Marine Ecology Progress Series, 248, 305e309. Fox, A. & Hudson, P. J. 2001. Parasites reduce territorial behaviour in red grouse (Lagopus lagopus scoticus). Ecology Letters, 4, 139e143. Haney, D. C., Vokoun, J. C. & Noltie, D. B. 2001. Alarm pheromone recognition in a Missouri darter assemblage. Journal of Fish Biology, 59, 810e817. Hart, B. L. 1988. Biological basis of the behavior of sick animals. Neuroscience and Biobehavioral Reviews, 12, 123e137. Higgins, J. J. & Tashtoush, S. 1994. An aligned rank transform test for interaction. Nonlinear World, 1, 201e211. Howerth, E. W. & Keller, A. E. 2006. Experimentally induced glochidiosis in smallmouth bass (Micropterus dolomieu). Veterinary Pathology, 43, 1004e1007. Kaiser, B. E. 2005. The effects of glochidiosis on fish respiration. M.S. thesis, Missouri State University. Krkosek, M., Connors, B. M., Ford, H., Peacock, S., Mages, P., Ford, J. S., Morton, A., Volpe, J. P., Hilborn, R. & Dill, L. M., et al. 2011. Fish farms, parasites, and predators: implications for salmon population dynamics. Ecological Applications, 21, 897e914. Lafferty, K. D. & Morris, A. K. 1996. Altered behaviour of parasitized killifish increases susceptibility to predation by bird final hosts. Ecology, 77, 1390e1397. LaMunyon, C. & Eisner, T. 1990. Effect of mite infestation on the anti-predator defenses of an insect. Psyche, 97, 31e41. Lefever, G. & Curtis, W. C. 1912. Studies on the reproduction of artificial propagation of freshwater mussels. U. S. Bureau of Fisheries Bulletin, 30, 105e209. Lima, S. L. 1998. Stress and decision making under the risk of predation: recent developments from behavioral, reproductive, and ecological processes. Advances in the Study of Behavior, 27, 215e290. Loot, G., Aulagnier, S., Lek, S., Thomas, F. & Guégan, J. 2002. Experimental demonstration of a behavioural modification in a cyprinid fish, Rutilus rutilus (L.), induced by a parasite, Ligula intestinalis (L.). Canadian Journal of Zoology, 80, 738e744. Maksimowich, D. S. & Mathis, A. 2000. Parasitized salamanders are inferior competitors for territories and food resources. Ethology, 106, 319e329. Mathis, A. 2009. Alarm responses as a defense: chemical alarm cues in nonostariophysan fishes. In: Fish Defenses. Vol. 2: Pathogens, Parasites and Predators (Ed. by G. Zaccone, C. Perrière & A. Mathis), pp. 323e386. Enfield, New Hampshire: Science. Milinski, M. 1985. Risk of predation of parasitized sticklebacks (Gasterosteus aculeatus L.) under competition for food. Behaviour, 93, 203e216. Moles, A. 1983. Effect of parasitism by mussel glochidia on growth of coho salmon. Transactions of the American Fisheries Society, 112, 201e204. Morgan, M. J. 1987. The influence of hunger, shoal size and predator presence on foraging in bluntnose minnows. Animal Behaviour, 36, 1317e1322. Myers, T. R. & Millemann, R. E. 1977. Glochidiosis of salmonid fishes. I. Comparative susceptibility to experimental infestation with Margaritifera margaritifera (L.) (Pelecypoda: Margaritanidae). Journal of Parasitology, 63, 728e733. Northcott, S. J., Lyndon, A. R. & Campbell, A. D. 2003. An outbreak of freshwater fish lice, Argulus foliaceus L., seriously affecting a Scottish stillwater fishery. Fisheries Management and Ecology, 4, 73e75. Oesch, R. D. 1984. Missouri Naiads: a Guide to the Mussels from Missouri. Jefferson City: Missouri Department of Conservation. Parris, M. J., Reese, E. & Storfer, A. 2006. Antipredator behaviour of chytridiomycosis-infested northern leopard frog (Rana pipiens) tadpoles. Canadian Journal of Zoology, 84, 58e65. Pélabon, C., Borg, A. A., Bjelvenmark, J., Barber, I., Forsgren, E. & Trond Amundsen, T. 2005. Do microsporidian parasites affect courtship in two-spotted gobies? Marine Biology, 148, 1432e1793.
A. L. Crane et al. / Animal Behaviour 82 (2011) 817e823 Riusech, F. A. & Barnhart, M. C. 1998. Host suitability in Venustaconcha ellipsiformis and Venustaconcha pleasii (Bivalvia: Unionidae) from Ozark plateaus. In: Proceedings of the Conservation, Captive, and Propagation of Freshwater Mussels Symposium (Ed. by T. A. Tankersley, T. Watters, B. Armitage & D. Warmolts), pp. 83e91. Columbus: Ohio University Press. Rogers-Lowery, C. L. & Dimock, R. V., Jr. 2006. Encapsulation of attached ectoparasitic larvae of freshwater mussels by epithelial tissue on fins of native and resistant host fish. Biological Bulletin, 210, 51e63. Sasal, P. 2006. Nest guarding in a damselfish: evidence of a role for parasites. Journal of Fish Biology, 68, 1215e1221. Smith, R. J. F. 1979. Alarm reaction of Iowa and Johnny darters (Etheostoma, Percidae, Pisces) to chemicals from injured conspecifics. Canadian Journal of Zoology, 57, 1278e1282. Smith, R. J. F. 1992. Alarm signals in fishes. Reviews in Fish Biology and Fisheries, 2, 33e63.
823
Smith Trail, D. R. 1980. Behavioural interactions between parasites and hosts: host suicide and the evolution of complex life cycles. American Naturalist, 116, 77e91. Tocque, K. 1993. The relationship between parasite burden and host resources in the desert toad (Scaphiopus couchii), under natural environmental conditions. Journal of Animal Ecology, 62, 683e693. Vaughan, G. E. & Coble, D. W. 1975. Sublethal effects of three ectoparasites on fish. Journal of Fish Biology, 7, 283e294. Waller, D. L. & Mitchell, L. G.1989. Gill tissue reactions in walleye Stizostedion vitreum vitreum and common carp Cyprinus carpio to glochidia of the freshwater mussel Lampsilis radiata siliquoidea. Diseases of Aquatic Organisms, 6, 81e87. Woods, D. A. 2008. Sub-lethal effects of copper on foraging and alarm behaviour of rainbow darters (Etheostoma caeruleum). M.S. thesis, Missouri State University. Zar, J. H. 1984. Biostatistical Analysis. 4th edn. Upper Saddle River, New Jersey: Prentice Hall.