- Email: [email protected]
Metabolic and behavioral responses of predators to prey nutrient content
Journal of Insect Physiology 116 (2019) 25–31
Contents lists available at ScienceDirect
Journal of Insect Physiology journal homepage: www.elsevier.com/locate/jinsphys
Metabolic and behavioral responses of predators to prey nutrient content Nicholas A. Koemel, Cody L. Barnes, Shawn M. Wilder
⁎
T
Department of Integrative Biology, Oklahoma State University, 501 Life Science West, Stillwater, OK 74075, USA
ARTICLE INFO
ABSTRACT
Keywords: Nutrition Wolf spider Spider Predator Metabolism Behavior
Predators feed on a diversity of prey that can vary widely in nutrient content. While prey nutrient content is known to have important consequences for life history traits, less is known about how it affects physiology and behavior. The purpose of this study was to test how diet affected the physiology and behavior of the wolf spider Hogna carolinensis. We hypothesized that higher protein intake would result in a lower metabolic rate due to less energy intake. Further, we also expected the high protein group to exhibit increased activity levels and aggression in an attempt to increase energy intake. Spiders were maintained on three different treatment diets in order to simulate prey with differing macronutrient composition: high protein, intermediate, and high lipid. Spider respiration was measured to quantify the baseline metabolic rate (SMR), digestive metabolic rate (SDA), and active metabolic rate (AMR). We found no significant effect of diet on metabolic rates. However, the SDA coefficient (i.e. digestive cost relative to prey content) was higher in the high protein group, meaning that this group metabolized a greater portion of their prey during digestion and had a lower net energy intake from prey. In our behavioral assays, spiders in the high protein group were significantly more active and attacked prey more quickly in their first trial. Our results demonstrate that diet had relatively little effect on predator metabolism but more of an effect on behavior. These findings suggest that diet regulation should be analyzed by studying multiple responses together, including metabolism and behavior, to gain a more comprehensive understanding of the effects of diet on organism performance and fitness.
1. Introduction Predators often consume a diverse selection of prey and prey can vary widely in a number of traits including: size, defenses, activity level, and nutritional content (Mooney et al., 2010; Wilder et al., 2013). Prey traits can influence predator performance by affecting the amounts and types of nutrients gained from prey. Often, the total amount or balance of nutrients in prey may not match the optimal balance of nutrients required by predators, which can lead to reductions in performance or changes in behavior (Raubenheimer et al., 2007; Wilder and Rypstra, 2010; Jensen et al., 2011a, 2012; Wiggins et al., 2018). While the effects of dietary imbalance on life histories, physiology, and behavior have been studied in a number of herbivores and omnivores, much less is known about how dietary imbalances affect predators (Simpson and Raubenheimer, 2012; Simpson et al., 2015). The main macronutrients ingested by predators are lipid and protein (Wilder et al., 2013). Lipids provide a source of energy to fuel behavior or physiological processes (Jensen et al., 2011a). While protein can also be catabolized for energy, protein is primarily important for providing material to build enzymes and tissues (Walter et al., 2017). Individual components of fitness may require particular amounts or balances of ⁎
nutrients, which may not always be the same (Raubenheimer et al., 2007; Jensen et al., 2010, 2011a, 2012; Simpson and Raubenheimer, 2012). Hence, understanding how imbalanced diets affect multiple components of fitness simultaneously is important for gaining a more comprehensive understanding of diet consequences for animals. Body composition, physiology, and behavior may all be affected by diet. In terms of body composition, diet can affect lipid reserves, which may be a critical backup energy source to survive periods without prey. This may be especially important for predators such as spiders since evidence suggests that many species may be food limited in nature (Wise, 1975; Bilde and Toft, 1998; Wilder and Rypstra, 2008). As such, spiders and other predators may need to be efficient in their use of energy during metabolism including baseline metabolism (i.e., standard metabolic rate, SMR), active metabolic rate (AMR), and metabolism during digestion (i.e., specific dynamic action, SDA) (Tanaka and Itô, 1982; O’Connor et al., 2000; McCue, 2006; Secor, 2009; Van Leeuwen et al., 2012). Prey nutrient content could affect predator metabolism either directly (e.g., SDA during digestion) or through its influence on other aspects of physiology (e.g., SMR) behavior (e.g., AMR) (Nespolo et al., 2011). For example, low lipid or energy content of prey could result in increased foraging behavior as predators search for additional
Corresponding author. E-mail address: [email protected] (S.M. Wilder).
https://doi.org/10.1016/j.jinsphys.2019.04.006 Received 28 August 2018; Received in revised form 12 April 2019; Accepted 18 April 2019 Available online 19 April 2019 0022-1910/ © 2019 Elsevier Ltd. All rights reserved.
Journal of Insect Physiology 116 (2019) 25–31
N.A. Koemel, et al.
prey (Hazlett et al., 1975). Spiders are a particularly abundant and diverse group of predators. They are also voracious predators, with an estimated annual consumption of 400–800 metric tons of insects per year worldwide (Nyffeler and Birkhofer, 2017). Spiders have served as models for the study of predator diet because large numbers can be easily maintained in the laboratory, physiology and behavior can be easily measured, treatments have been developed that allow investigators to manipulate the nutrient content of prey, and experiments can be done with less ethical concerns than larger predators (Wilder, 2011). As such, there is a foundation of research on spider nutrition on which studies can build, to provide a more extensive understanding of the nutritional ecology of this common predatory group (Salomon et al., 2008; Wilder and Rypstra, 2010; Jensen et al., 2010, 2011b; Hawley et al., 2014). The overall goal of this study was to test how the macronutrient content of prey of the wolf spider, Hogna carolinensis, affected body composition, metabolism (i.e., SDA and AMR), and behavior. We predicted that spiders fed high lipid diets would incur lower energetic costs associated with digestive metabolic rate when compared to that of spiders on a high protein diet because catabolism of protein has been demonstrated to produce greater digestive costs than lipids and carbohydrates in a range of other taxa (Secor, 2009). As for the locomotor performance, we predicted that the spiders in the high lipid groups would have higher performance in endurance and sprint speed compared to the high protein treatments because of greater available energy in their diet. Lastly, we predicted that spiders consuming the high protein diet would have the lowest lipid reserves and, as a consequence, would be more active and more aggressive response towards prey to increase energy intake.
Following 10 days of starvation, wolf spiders were arranged by increasing mass, then sequentially assigned to one of the three cricket diet treatments. This assortment scheme, rather than one assigned at random, ensured that there were not initial differences between treatments in spider mass. After the wolf spiders were assigned to a treatment group, they were fed 2 treatment crickets twice a week for a total of 6 weeks. The spiders consumed 1 of 3 different cricket diet treatments for the 6-week duration (i.e., high protein, intermediate, or high lipid). All physiological and behavioral experimentation occurred during the week following the 6-week diet acclimation period. Wolf spiders are polyphagous predators that consume a wide variety of prey. While their diet is variable, they may experience periods of time where they consume similar prey due to a seasonal outbreak of a prey species or high density of particular prey in a habitat. Hence, the 6-week trial period may represent a somewhat long, although still realistic, time period for the spiders to feed on a particular diet. Crickets were purchased from a commercial supplier (“¼ inch” in length, © Fluker's Cricket Farm, 2016). Crickets were reared on diet treatments for 7–10 days to allow for body compositional changes prior to being fed to the wolf spiders (Barnes et al., 2019). A prior study (Barnes et al., 2019) developed these diets and showed that feeding them to crickets results in significant differences in the nutrient content of the cricket bodies. Cricket diets treatments consisted of high protein, intermediate, and high lipid groups (Table 1). Diets were constructed by incorporating liquid lipids with the dry powder components and then mixing thoroughly after allowing to dry. Water was provided ad libitum, which was replaced twice weekly. Representative crickets from each diet group were used to analyze prey nutrient content. These crickets were euthanized by freezing and stored at 0 OC, then dried at 60 OC for 24 h, weighed, and analyzed for nutrients.
2. Methods 2.1. Study animals
2.2. Respiration
Hogna carolinensis typically construct burrows used as daytime retreats but will forage away from the burrow at night in search of prey. Mature female wolf spiders, H. carolinensis, were collected from fields in Stillwater, Oklahoma during September 2016 (n = 46; Lipid n = 15, Intermediate n = 15, Protein n = 16) for experiments that tested how diet affected body composition and respiration of spiders. A second subset of spiders was collected (n = 48; Lipid n = 16, Intermediate n = 16, Protein n = 16) in June 2017 to conduct additional experiments to test how diet affected activity and aggression. The timing of collection and experiment duration coincided with the spider reproductive season, so most spiders produced egg sacs in the laboratory. The spiders did not differ in mass between the two collection periods (t 91 = 1.7, p = 0.10; 1.59 ± 0.05 mg versus 1.71 ± 0.05 mg). All egg sacs were removed shortly after their production. Wolf spiders were acclimated to laboratory conditions in individual, plastic deli containers (1420 mL). We cut circular pieces of paper towels to act as substrates for the floor of each container. Then, we cut a 946 mL opaque deli container in half to create a shelter for each spider. Additionally, a moist cotton ball was placed in each spider container and exchanged weekly. Paper substrates were replaced and the spider housing was cleaned each week. Spiders were housed at 25 ± 1 OC and 14L:10D light regime. Housing location within shelving units was rotated weekly. Prior to experiments, the spiders were fed two crickets, Acheta domesticus that were reared on dog-food (Rachel Ray Nutrish) twice during the first week, then fasted for 10 days to clear spider gut contents and standardize the spider hunger level (Barnes et al., 2019). Starvation periods of other spiders in the field have been estimated to be between 4 and 8 days on average (Bilde and Toft, 1998). Although, one study of wolf spiders found that the body condition of field-collected spiders was similar to spiders that had been fed ad libitum and then deprived of food for 3 months (Wilder and Rypstra, 2008).
The SDA coefficient was calculated as SDA divided by the amount of prey energy ingested (i.e. total prey energy minus the energy of prey remains) (Beaupre, 2005). Activity also has an influence on the metabolic rate; this is demonstrated by the increase in metabolism during or after activity known as the active metabolic rate (AMR). AMR can be energetically expensive for many organisms but has not been exhaustively studied for wolf spiders (Ford, 1977). We used closed system respirometry to determine the SMR, AMR, and SDA rates of spiders. The spiders were starved for 10 days after the 6-week treatment diets in order to standardize feeding level. Spiders were placed into individual 946 mL deli containers. Sealable ports were installed on the top of the container to allow CO2 readings to be Table 1 The ingredients incorporated into the three experimental diets fed to the crickets (in grams). The percentage of macronutrients for each diet are designated at the top (Protein:Lipid:Carbohydrate).
Egg white Micellar Casein Sugar Flour Cellulose Nipagin True Balance™ Vitamin (NOW Foods; Capsule) Cholesterol Fish Oil Lard Olive Oil
26
Lipid (10:45:45)
Intermediate (70:15:15)
Protein (100:00:00)
11 11 55 85 94 1 1
105 105 14 22 43 1 1
137 137 0 0 26 1 1
0.5 3 22 22
0.5 3 5.5 5.5
0.5 3 0 0
Journal of Insect Physiology 116 (2019) 25–31
N.A. Koemel, et al.
collected without opening the container. The amount of CO2 respired into the container was then measured via a Li-840A CO2/H2O respirometer (LiCor; Lincoln, NE, USA; Barnes et al., in press). We measured SMR twice daily (09:00 and 21:00) in the final two days of the spider starvation period. For SDA, the wolf spiders were fed two crickets (a total of 10–15% of spider body mass) from their assigned treatment group. Additionally, prey were size matched across all treatments. To do this, we pre-weighed all crickets and ensured that the range of cricket weights for all treatments was broadly overlapping. The respiration chambers were then re-sealed following the feeding. Respiration was measured at 3-hour intervals for the first 9 h following the introduction of prey, then respiration measures were resumed at 12hour intervals for the next 3 days to monitor the attenuation of the SDA response. We found that the metabolic rate of the spiders returned to baseline within 2–3 days after feeding (see results). Accumulated CO2 was converted to percent by dividing by 10,000. This value was converted to the total volume of CO2 (VCO2) by dividing by 100, and then multiplied by the volume of the container (946 mL). Metabolism was calculated as the final minus initial respiration reading, divided by the sampling duration. Units of respiration were converted to energy using the conversion factor 24.65 kJ L CO2−1(Chown et al., 2007). Control respiration chambers without animals, but otherwise similar handling, were run alongside treatment chambers to test if handling conditions affected CO2 readings, which they did not. Spiders generally completed the extra-oral digestion process and deposited boluses of prey remains within nine hours of introduction of prey and time to first excretion was between 24 and 48 h. From observations of spiders, it appeared that feeding was finished within 24 h and most excreta production occurred within 48 h. Uneaten prey remains were collected within 24 h following each feeding. We were also interested in how diets affected active metabolic rate (AMR). However, rather than measure metabolic rate during activity, we measured metabolic rate during recovery after physical activity. Spiders’ muscles contain few mitochondria (Foelix, 2011). This limits the endurance of spiders, which typically fatigue relatively quickly, and suggests that the metabolic costs of physical activity may be incurred during recovery more so than during the physical activity itself. Hence, we used closed respirometry to measure the metabolic rate of a spider during post-locomotor, recovery (hereafter referred to as AMR). When the spiders had fully completed the locomotor performance trials (described below) they were immediately placed into individual respiration chambers to measure gas exchange. The starting CO2 of the container was measured and then a CO2 measurement was taken in intervals of 12 h. The measurements in parts-per-million (ppm) were converted to volume (mL) using the same protocol from the SDA and SMR trials. AMR trials were repeated twice over two days. At the end of the second trial, the spiders were sacrificed and their lipid and protein composition was measured using the same procedure used for the prey (described below).
however, those who stopped mid-track were encouraged to continue with a light stroke of a paintbrush (Prenter et al., 2010). We filmed the first and last 25 cm of sprints using a Canon Vixia HF R52 camera stationed on a tripod above the track. Once the experiment was completed, the videos were processed using ImageJ 1.48v (National Institutes of Health, USA) to measure the time to complete a 150 cm sprint. At the end of the 75 cm linear track was an opening to a circular arena. The spiders were introduced to the container after completing the 150 cm sprint. Once inside the container, spider endurance was measured by documenting the time elapsed until the spider reached full fatigue. The spiders were periodically tapped with the paintbrush to maintain movement around the perimeter of the container. Spiders were considered to be fatigued upon loss of righting response when spiders were flipped over. The endurance and sprint trials were immediately followed by the assessment of AMR. All of the trials were repeated twice approximately 2 days apart. Spider activity and aggression were measured by conducting a behavioral assay in a circular arena with a filter paper-substrate (25 cm diameter plastic deli containers). The circular arena was separated into two equal semi-circle portions of filter paper. One half had a blank filter paper, and the other half contained filter paper with the prey chemical cues. Heterogeneity of chemical cues in the arena could stimulate predators to continue searching the area. Prey chemical cues were collected by placing a live cricket on the filter paper for 24 h prior to the trial and removing the cricket before the trial began. Each spider was released in the center of the arena prior to the recording. Individual spiders were observed for 35 min after release into the arena. The first 5 min of recording was omitted from data analysis in order to allow for arena acclimation. For the movement assay, the cumulative movement (i.e. total time spent moving), velocity, and distance moved were extracted from video recordings using Ethovision XT13 (Noldus Information Technologies, Wageningen, The Netherlands). After the activity trials, prey capture was observed by introducing a single cricket into the arena and measuring the time lapse before the prey item was attacked and killed (Walker et al., 1999). Prior to measuring the attack latency time, the spiders and prey were acclimated in isolated chambers to the arena for 5 min. The acclimation chambers for prey and predator were placed on opposite sides of the circular arena in each trial. The trials for both attack latency and locomotor behavior were repeated for a total of two trials each. 2.4. Nutrient analyses We measured the lipid content of the whole crickets, remains of cricket bodies after the spider finished feeding, and spiders using a gravimetric assay with chloroform (Wilder et al., 2013). Whole crickets maintained on the experimental diets were previously analyzed in another study (Barnes et al., 2019) but we analyzed samples from the current study to confirm that the treatments were still effective in manipulating cricket nutrient content. Briefly, the items were first dried at 60 OC for 24 h and weighed. Once dried, they were individually placed into 15 mL vials and washed with 10 mL of pure chloroform for 24 h. After the 24-hour period of chloroform washing, the chloroform was drained and replaced with fresh chloroform and repeated for a total of three washes to facilitate full lipid extraction. After extraction, the contents were again dried at 60 OC for 24 h and reweighed. Lipid content was quantified by subtracting the mass of dried lipid-free components from the initial dry mass of the item. Protein was determined in triplicates using the Bradford Assay of lean, ground samples (i.e., after lipid extraction) (Wilder et al., 2013). Carbohydrates were not estimated, as they are typically present in arthropods at low concentrations (Raubenheimer and Rothman, 2013). Protein and lipid dry masses were converted to energy using standard conversion factors (protein = 17 kJ/g and lipid = 37 kJ/g; Raubenheimer and Rothman, 2013).
2.3. Behavior We also examined the effects of diet on behavior. After maintaining the spiders on the experimental diets for the 6-week period, the spiders’ locomotor performance was first individually examined as sprint speed of the spiders using a linear track. The behavior trials were repeated 3 days apart for a total of two trials. We conducted two separate trials for behavior to test if the behavioral responses were consistent and robust. The spiders were placed in a small black box at the start point and allowed 2 min to acclimate before beginning the trial. After the acclimation period, the front box lid was removed. Each trial began with the spiders being released from the acclimation chamber. The spiders were prompted to travel a full 75 cm distance of the track and back to the starting line for a total distance of 150 cm, which was timed using a stopwatch. The majority of the spiders ran the complete track; 27
Journal of Insect Physiology 116 (2019) 25–31
N.A. Koemel, et al.
Nutrient intake of spiders was estimated by subtracting the amount of nutrients present in prey remains from the amounts of nutrients estimated to be in prey before they were fed on by spiders. The initial nutrient content of prey fed to spiders was estimated using linear regression relationships between prey wet mass and nutrient content that were developed using representative crickets from each treatment (as in Wilder and Rypstra, 2010). 2.5. Data analysis Differences in animal composition were tested using ANOVA and Tukey HSD post hoc analysis, while spider performance was evaluated using ANCOVA with body mass as a covariate. Statistical analyses were conducted in JMP 13 software package (SAS Institute, Cary, NC, USA). Respiration was compared using repeated measures analysis of covariance in SAS 9.4 software package (SAS Institute, Cary, NC, USA), with repeated measures over time and spider body mass as a covariate. Behavioral measures were first analyzed for each trial date using MANOVA to test for significant effects of prey nutrient treatments and then we ran a repeated measures analysis of variance. The attack latency response variable was not normally distributed, so we applied a log-transformation. Statistical significance was evaluated at the 0.05 level.
Fig. 2. The lipid composition of the different H. carolinensis wolf spider treatments at the end of the six weeks of feeding (mg 100 mg−1 dry mass). Bars with different letters are significantly different from each other in post hoc analyses.
p = 0.07, Start = 1.5891 ± 0.0610 g, F2, 44 = 2.5, End = 1.47204 ± 0.0658 g). Spider body composition was measured in experiment 1. At the end of the six weeks of diet treatments, there was no significant difference in protein content of the wolf spiders fed different diets (percent protein, mean ± 1 SE: Lipid = 59.6 ± 1.1, Intermediate = 60.8 ± 1.8, Protein = 62.4 ± 1.5; F2, 41 = 0.9; p = 0.42). However, there was a significant difference in the lipid content by treatment. There was a higher lipid content in spiders fed high lipid diets compared to the protein and intermediate treatment groups (F2, 41 = 8.8; p < 0.01 Fig. 2).
3. Results 3.1. Body composition Crickets from the high lipid treatment had significantly higher lipid content compared to crickets from the intermediate and high protein treatments (F2, 48 = 30.3, p < 0.01; Fig. 1). There were no differences in cricket protein content between treatment groups (F2, 54 = 2.6; p = 0.11; Fig. 1). The energetic content of crickets from the high lipid treatment (1.74 ± 0.01 kJ) was greater than the intermediate (1.42 ± 0.02) and protein (1.18 ± 0.01) diet treatments (F2, 38 = 248.6, p < 0.01). There was no significant difference between diet treatments in the mass of spiders at the start of the experiment in either the first (F2, 45 = 0.0, p = 0.97) or second sets of experiments (F2, 47 = 0.0, p = 0.99). The increase in mass of the spiders over the course of the 6week feeding period was not significantly different between treatments in the first or second experiment (Experiment 1: F2, 44 = 6.7, p = 0.97, Start = 1.7083 ± 0.0522 g, End = 1.9660 ± 0.0552 g; Experiment 2:
3.2. Respiration Following prey capture and consumption, there was a significant increase in metabolic rate of spiders above baseline levels that peaked after about 4 h and lasted for a duration of 48 h (Fig. 3). Larger spiders respired more (F1, 30 = 22.2, p < 0.01), but we did not find significant differences in SDA between treatments (F2, 30 = 2.5, p = 0.10; Fig. 3). The SDA coefficient, which is the proportion of prey energy used during the SDA response, was significantly higher for spiders in the protein prey treatment (F2, 38 = 3.8, p = 0.03; Fig. 4). The post-activity respiration trials taken for AMR were significantly higher than the SMR (F4, 87 = 4.8, p < 0.01). We found that AMR increased with spider mass (F4, 87 = 12.2, p < 0.01). However, we did not find an effect of diet treatment on AMR (mL CO2/hr, mean ± 1 SE: Lipid = 0.074 ± 0.011, Intermediate = 0.083 ± 0.011, Protein = 0.056 ± 0.011; F4, 87 = 1.5, p = 0.23) 3.3. Behavior When analyzing the sprint times in a repeated measures ANOVA analysis, we did not find significant differences in endurance (i.e., time to fatigue) between treatments (F5, 38 = 0.4, p = 0.82) nor did sprint time differ between treatments (F5, 37 = 0.8, p = 0.59). There was also no effect of time in these trials (Endurance; F2, 38 = 0.0, p = 0.93, Sprint; F1, 32 = 1.1, p = 0.30). When we examined these performance trials in a separate ANOVA for each trial individually, there were no significant differences by treatment (Endurance; F2, 38 = 0.6, p = 0.55, Fig. 5a; Sprint; F2, 32 = 0.6, p = 0.53, Fig. 5b). When we used MANOVA to test if behavioral responses (i.e. total distance, velocity, time spent moving, and attack latency) were dependent on prey nutrient treatments, we found an effect of nutrition on day 2 (F8,66 = 2.3, p = 0.03) but not day 1 (F8,82 = 1.9, p = 0.07). A repeated measures ANOVA was conducted to analyze velocity data
Fig. 1. The lipid and protein composition of the cricket treatments (mg 100 mg−1 dry mass). There was no significant effect of diet on protein content. For lipid, bars with different letters are significantly different from each other in post hoc analyses. 28
Journal of Insect Physiology 116 (2019) 25–31
N.A. Koemel, et al.
Fig. 3. Change in metabolic rate (i.e., elevation of metabolic rate above the baseline, which is shown in the dashed line) of the wolf spider Hogna carolinensis following prey consumption of a specific treatment cricket (SDA; mL CO2 hr−1). Diet treatment macronutrient ratios (P:L:C), protein diet (100:00:00), lipid (10:45:45), and intermediate (70:15:15).
Fig. 4. The proportion of the SDA to the amount of prey energy ingested by the wolf spiders (SDA coefficient) between the different treatment groups.
from both trial days together. We found an overall significant effect of treatment on spider velocity (F1, 46 = 8.8, p < 0.01). There was also an overall time effect between the two trials, with higher average velocity on day 2 (F1, 46 = 5.9, p = 0.02). In a separate analysis of data from each day, there was not a statistically significant difference in velocity by treatment on day 1 (F2, 47 = 2.7, p = 0.08; Fig. 6a). On day 2, the difference in velocity was statistically significant between treatments, where the protein treatment had a significantly higher velocity compared to the lipid and intermediate treatment groups (F2, 47 = 6.2, p < 0.01; Fig. 6a). For the total distance traveled by spiders, a repeated measures ANOVA showed that there was an overall significant treatment effect (F1, 46 = 8.2, p < 0.01) and no effect of time (i.e., between the days) (F1, 46 = 1.9, p = 0.18. Spiders in the high protein diet treatment traveled a significantly greater total distance than spiders in the intermediate and high lipid treatments (Fig. 5b). Separate ANOVA analysis for each day of trials confirmed that distance traveled was higher in the high protein treatment on both day 1 (F2, 46 = 4.2, p = 0.02; Fig. 6b) and day 2 (F2, 46 = 5.5, p = 0.01; Fig. 6b). For cumulative time spent moving, there was an overall significant treatment effect when looking at both trials together in a repeated measures ANOVA (F1, 46 = 3.6, p = 0.01). There was also an overall time effect with higher average time spent moving on the second day of trials (F1, 46 = 5.9, p = 0.02). When analyzing each trial separately, spiders in the high protein treatment spent more time moving compared to intermediate and lipid treatment for both day 1 (F2, 47 = 3.8, p = 0.03; Table 2) and day 2 (F2, 47 = 5.7, p = 0.01; Table 2).
Fig. 5. a) The endurance performance in seconds (i.e., time until full exhaustion) of the spiders in their respective treatment groups. b) The sprint performance (i.e., time to travel 150 cm) of the spiders on the diet treatments.
For attack latency, there was a significant difference between treatments. In trial one, the protein group caught prey faster than the lipid and intermediate group (F1, 46 = 5.9, p = 0.02; Table 2), but no differences between treatments in trial two (F1, 38 = 0.4, p = 0.50; Table 2). There was no overall significant treatment effect when combining the two days in a repeated measures ANOVA analysis (F1, 44 = 0.0, p = 0.96). However, there was an overall time effect between the two trials with longer latencies on the second day (F1, 44 = 6.9, p = 0.01).
29
Journal of Insect Physiology 116 (2019) 25–31
N.A. Koemel, et al.
metabolism had we fed the spiders a more limited amount of food. Overall, these results suggest that diet can have multiple simultaneous impacts on animal composition, physiology, and behavior and that a more complete understanding of the effects of diet on animals may require simultaneously considering these multiple measures. While prey varied in lipid and energy content, these treatments had no effect on the standard or digestive metabolic rates of spiders. This result was similar to that of a previous study that found no effect of prey lipid and protein content on the SDA of a wolf spider (Jensen et al., 2010). Previous studies have also found that SDA is relatively high in proportion to prey energy (i.e. SDA coefficient; Zaidan and Beaupre (2003), Bessler et al. (2010)). Overall, the amount of digestive energy expended by spiders in our study was approximately 5–7% of the total meal energy (Fig. 4). More specifically, although SDA did not differ between lipid-based and protein-based diets, the SDA coefficient was greater in spiders fed protein-biased diets. We found that spiders used a greater amount of protein-biased prey energy during digestion compared to lipid-based prey-fed spiders. The differences in the SDA coefficient appear to be driven by differences in prey energy content (i.e., spiders ingested more energy from high lipid prey) since overall metabolic costs of digesting prey did not differ between treatments. There were no significant effects of diet on AMR, sprint speed, or endurance. These three measures likely comprise a small fraction of the energy and time budget of this study species. Spiders are extremely efficient predators with a very low amount of metabolically active physiological components compared to other arthropods (Anderson, 1970). The wolf spider Hogna carolinensis is a sit-and-wait predator that appears to spend much of its time waiting to encounter prey that pass near their burrows (Suter et al., 2011). When we collected spiders in the field, they were often within 25 cm of their burrows, which means they would only have a short distance to run from predators. In addition, they typically waited for prey to be within 5–10 cm before attacking in prey capture trials, which would also represent a short time spent sprinting. Hence, in addition to not being significantly different between treatments, these three measures are likely relatively less important components of the overall fitness of these spiders in nature. Conversely, in other species of spiders, such as male web-building spiders that wander in search of females, AMR can be a more important component of fitness (Kasumovic and Seebacher, 2013). Two opposite predictions could be made as to the effects of diet on spider movement. First, high lipid spiders could be predicted to move more since they have higher stored energy and, hence, the surplus energy that they are able to expend on movement, including exploration of new habitats. Alternatively, spiders on the high protein diet could be predicted to move more because they have less energy, thus greater movement may increase the chances of encountering prey (Pekár et al., 2010). In our study, spiders on the high protein diet were significantly more active in terms of total distance moved, cumulative movement, and velocity. The high protein fed spiders were also captured prey the quickest in the first foraging trial. Given these two results and that the arenas for the behavioral trials contained prey cues, it seems likely that the higher activity of protein fed spiders was prey searching behavior. The differences that we observed were likely a result of differences among treatments in the motivation of spiders to move and not the availability of energy for movement. All of the treatments were provided with moderate amounts of food providing adequate energy for spiders to maintain or gain mass in all treatments. Higher activity levels due to prey searching behavior could increase predation risk in nature, as wolf spiders are prey to a wide range of visually hunting predators (Cloudsley-Thompson, 1995). This study was conducted on adult female wolf spiders during the reproductive season. Nearly all spiders produced an egg sac, which was immediately removed, during the period on which they were maintained on their treatment diets. It is unclear if the results of this study apply to only reproductive females or might also apply to juveniles and adult males. We would predict that high protein diets might contribute
Fig. 6. a) Mean velocity of the spider (cm/s). b) Total distance traveled (cm) of the spiders. Comparisons were made between diet treatments. The changes from trial one and trial two are also shown. Values with unlike letters represent significant treatment differences in the given trial (P < 0.05). Table 2 The mean behavior of spiders fed different diets in the experimental trials. Comparisons were made between diet treatments. The changes from trial one and trial two are also shown. Values with unlike letters represent significant differences between treatments (P < 0.05). Diet Treatment
Protein
Intermediate
Lipid
Cumulative Movement (s) Trial 1 59.11a ± 18.12 Trial 2 95.61a ± 26.19
1.17b ± 0.48 9.70b ± 6.79
14.79b ± 8.54 5.42b ± 2.36
Attack Latency (s) Trial 1 Trial 2
21.0 ± 4.59 70.8 ± 23.21
39.0 ± 11.34 40.7 ± 14.69
17.0 ± 3.17 30.7 ± 14.90
4. Discussion Overall, our results show that diets biased in lipid versus protein affect some measures of body composition (i.e., lipid reserves), physiology (i.e., respiration), and behavior (i.e., movement). For body composition, high lipid diets increased lipid reserves. In terms of metabolism, diets had no effect on SDA or AMR; although spiders on the high lipid treatment had lower SDA coefficients. For behavior, while there were no effects of diet on sprint speed or endurance, spiders in the high protein treatment were more active, on average, and captured prey more quickly in their first trial than spiders in the other diet treatments. One explanation for the significant effects of diet on behavior but not metabolism is that behavior is more sensitive to diet than metabolism. In the current study, spiders had a moderate quantity of food that varied in nutrient content. Hence, the spiders may have shifted their behavior in an attempt to change their diet but may not have been stressed by the diet to the point that it affected their metabolism. It is possible that we would have seen effects on both behavior and 30
Journal of Insect Physiology 116 (2019) 25–31
N.A. Koemel, et al.
to higher growth of juveniles, whereas high lipid diets might be especially important to adult males since they spend much of their time actively searching for females (Wilder, 2011). Relatively little is known about the comparative dietary requirements of male and female spiders. Although, studies of crickets support the idea that female fitness is enhanced on more protein biased diets while males benefit more from more energy (i.e., carbohydrate) rich diets (Maklakov et al., 2008). Furthermore, relatively little is known about the comparative nutritional ecology of different species (Wilder and Eubanks, 2010). Spiders differ widely in foraging strategies from actively pursuing prey to spending most of their time in sit-and-wait foraging on a web. Evidence suggests that these foraging strategies could be associated with overall nutrient requirements as a more active jumping spider grew larger on higher lipid diets (Wiggins et al., 2018) while a more sedentary wolf spider grew larger on more protein biased diets (Jensen et al., 2011b). Further research on diverse species and life stages are needed to gain a more general understanding of predator diet requirements, how they vary among predators, and how they compare to animals at other trophic levels.
Hawley, J., Simpson, S.J., Wilder, S.M., 2014. Effects of prey macronutrient content on body composition and nutrient intake in a web-building spider. PLoS One 9 (6), E99165. Kasumovic, M., Seebacher, F., 2013. The active metabolic rate predicts a male spider's proximity to females and expected fitness. Biol. Lett. 9 (2), 20121164. Maklakov, A.A., Simpson, S.J., Zajitschek, F., Hall, M.D., Dessmann, J., Clissold, F., Brooks, R.C., 2008. Sex-specific fitness effects of nutrient intake on reproduction and lifespan. Curr. Biol. 18 (14), 1062–1066. McCue, M.D., 2006. Specific dynamic action: a century of investigation. Comp. Biochem. Physiol. Part A: Mol. Integr. Physiol. 144 (4) 381394. Mooney, K.A., Gruner, D.S., Barber, N.A., Van Bael, S.A., Philpott, S.M., Greenberg, R., 2010. Interactions among predators and the cascading effects of vertebrate insectivores on arthropod communities and plants. Proc. Nat. Acad. Sci. U.S.A 107, 7335–7340. Nespolo, Correa, Pérez-Apablaza, Cortés, Bartheld, 2011. Energy metabolism and the postprandial response of the Chilean tarantulas, Euathlus truculentus (Araneae: Theraphosidae). Comp. Biochem. Physiol. Part A 159 (4), 379–382. Nyffeler, M., Birkhofer, K., 2017. An estimated 400-800 million tons of prey are annually killed by the global spider community. Sci. Nat. https://doi.org/10.1007/s00114017-1440-1. O'Connor, K.I., Taylor, A.C., Metcalfe, N.B., 2000. The stability of standard metabolic rate during a period of food deprivation in juvenile Atlantic salmon. J. Fish Biol. 57 (1), 41–51. Pekár, Mayntz, Ribeiro, Herberstein, 2010. Specialist ant-eating spiders selectively feed on different body parts to balance nutrient intake. Anim. Behav. 79 (6), 1301–1306. Prenter, J., Pérez-Staples, D., Taylor, P.W., 2010. The effects of morphology and substrate diameter on climbing and locomotor performance in male spiders. Funct. Ecol. 24 (2), 400–408. Raubenheimer, D., Rothman, J.M., 2013. Nutritional ecology of entomophagy in humans and other primates. Annu. Rev. Entomol. 58, 141–160. Raubenheimer, D., Mayntz, D., Simpson, S.J., Tøft, S., 2007. Nutrient specific compensation following dipause in a predator: implications for Intraguild predation. Ecology 88 (10), 2598–2608. Salomon, M., Mayntz, D., Lubin, Y., 2008. Colony nutrition skews reproduction in a social spider. Behav. Ecol. 19 (3), 605–611. Secor, S.M., 2009. Specific dynamic action: a review of the postprandial metabolic response. J. Comp. Physiol. B 179 (1), 1–56. Simpson, S., Raubenheimer, D., 2012. The Nature of Nutrition: A Unifying Framework from Animal Adaptation to Human Obesity. Princeton University Press, Princeton. Simpson, S.J., Le Couteur, D.G., Raubenheimer, D., 2015. Putting the balance back in diet. Cell 161 (1), 18–23. Suter, R.B., Stratton, G.E., Miller, P.R., 2011. Mechanics and energetics of excavation by burrowing wolf spiders, Geolycosa spp. J. Insect Sci. 11, 22. https://doi.org/10. 1673/031.011.0122. Tanaka, K., Itô, Y., 1982. Decrease in respiratory rate in a wolf spider, Pardosa astrigera (L. Koch), under starvation. Res. Popul. Ecol. 24 (2), 360–374. Walker, S., Marshall, S., Rypstra, A., Taylor, D., 1999. The effects of hunger on locomotory behaviour in two species of wolf spider (Araneae, Lycosidae). Anim. Behav. 58, 515–520. Walter, A., Bechsgaard, J., Scavenius, C., Dyrlund, T., Sanggaard, K., Enghild, J., Bilde, T., 2017. Characterisation of protein families in spider digestive fluids and their role in extra-oral digestion. BMC Genomics 18. Wiggins, W., Bounds, D., Wilder, S., 2018. Laboratory-reared and field-collected predators respond differently to same experimental treatments. Behav. Ecol. Sociobiol. 72 (2), 1–8. Wilder, S.M., 2011. Spider nutrition: an integrative perspective. Adv. Insect Physiol. 40, 87–136. Wilder, S.M., Eubanks, M.D., 2010. Might nitrogen limitation promote omnivory among carnivorous arthropods? Comment. Ecol. 91 (10), 3114–3117. Wilder, S.M., Rypstra, A., 2008. Diet quality affects mating behaviour and egg production in a wolf spider. Anim. Behav. 76 (2), 439. Wilder, S.M., Rypstra, A.L., 2010. Males make poor meals: a comparison of nutrient extraction during sexual cannibalism and predation. Oecologia 162 (3), 617–625. Wilder, S.M., Norris, M., Lee, R.W., Raubenheimer, D., Simpson, S.J., 2013. Arthropod food webs become increasingly lipid-limited at higher trophic levels. Ecol. Lett. 16, 895–902. Wise, D., 1975. Food limitation of the spider Linyphia marginata: experimental Field Studies. Ecology 56 (3), 637–646. Van Leeuwen, T.E., Rosenfeld, J.S., Richards, J.G., 2012. Effects of food ration on SMR: influence of food consumption on individual variation in metabolic rate in juvenile coho salmon (Onchorhynchus kisutch). J. Anim. Ecol. 81 (2), 395–402. Zaidan, F., Beaupre, S., 2003. Effects of body mass, meal size, fast length, and temperature on specific dynamic action in the timber rattlesnake (Crotalus horridus). Ecol. Evol. Physiol. Biochem. Zool.
Acknowledgements We would like to thank the Wilder Lab group for assistance in the lab and constructive comments on the paper. This research was supported by the Department of Integrative Biology at Oklahoma State University and the Wentz Research Grant from the Lew Wentz Foundation awarded to NAK. References Anderson, J.F., 1970. Metabolic rates of spiders. Comp. Biochem. Physiol. 33 (1), 51–72. Barnes, C.L., Hawlena, D., Wilder, S.M., 2019. Predators buffer the effects of variation in prey nutrient content for nutrient deposition. Oikos 128 (3), 360–367. Barnes, C.L., Hawlena, D., McCue, M.D., Wilder, S.M., in press. Consequences of prey exoskeleton content for predator feeding and digestion: black widow predation on larval versus adult mealworm beetles. Oecologia. Beaupre, S., 2005. Ratio representations of specific dynamic action (MassSpecific SDA and SDA Coefficient) do not standardize for body mass and meal size. Physiol. Biochem. Zool. 78 (1), 126–131. Bessler, S., Stubblefield, M., Ultsch, G., Secor, S., 2010. Determinants and modeling of specific dynamic action for the Common Garter Snake (Thamnophis sirtalis). Can. J. Zool. Revue Canadienne De Zoologie 88 (8), 808–820. Bilde, T., Toft, S., 1998. Quantifying food limitation of arthropod predators in the field. Oecologia 115 (1), 54–58. Chown, S.L., Marais, E., Terblanche, J.S., Klok, C.J., Lighton, J.R.B., Blackburn, T.M., 2007. Scaling of insect metabolic rate is inconsistent with the nutrient supply network model. Funct. Ecol. 21, 282–290. Cloudsley-Thompson, J.L., 1995. A review of the anti-predator devices of spiders. Bull. Br. Arachnol. Soc. 10 (3), 81–96. Foelix, R., 2011. Biology of Spiders, third ed. Oxford University Press, Oxford; New York. Ford, M., 1977. Metabolic costs of the predation strategy of the spider Pardosa amentata (Clerck) (Lycosidae). Oecologia 28 (4), 333–340. Hazlett, B., Rubenstein, D., Rittschof, D., 1975. Starvation, energy reserves, and aggression in the crayfish Orconectes virilis (Hagen, 1870) (Decapoda, Cambaridae). Crustaceana 11–16. Jensen, Mayntz, Wang, Simpson, Overgaard, 2010. Metabolic consequences of feeding and fasting on nutritionally different diets in the wolf spider Pardosa prativaga. J. Insect Physiol. 56 (9), 1095–1100. Jensen, K., Mayntz, D., Toft, S., Raubenheimer, S., Simpson, J., 2011a. Prey nutrient composition has different effects on Pardosa wolf spiders with dissimilar life histories. Oecologia 165 (3), 577–583. Jensen, K., Mayntz, D., Toft, S., Raubenheimer, S., Simpson, J., 2011b. Nutrient regulation in a predator, the wolf spider Pardosa prativaga. Anim. Behav. 81 (5), 993–999. Jensen, K., Mayntz, D., Toft, S., Clissold, F.J., Hunt, J., Raubenheimer, D., Simpson, S.J., 2012. Optimal foraging for specific nutrients in predatory beetles. Proc. R. Soc. London B Biol. Sci rspb20112410.
31
Contents lists available at ScienceDirect
Journal of Insect Physiology journal homepage: www.elsevier.com/locate/jinsphys
Metabolic and behavioral responses of predators to prey nutrient content Nicholas A. Koemel, Cody L. Barnes, Shawn M. Wilder
⁎
T
Department of Integrative Biology, Oklahoma State University, 501 Life Science West, Stillwater, OK 74075, USA
ARTICLE INFO
ABSTRACT
Keywords: Nutrition Wolf spider Spider Predator Metabolism Behavior
Predators feed on a diversity of prey that can vary widely in nutrient content. While prey nutrient content is known to have important consequences for life history traits, less is known about how it affects physiology and behavior. The purpose of this study was to test how diet affected the physiology and behavior of the wolf spider Hogna carolinensis. We hypothesized that higher protein intake would result in a lower metabolic rate due to less energy intake. Further, we also expected the high protein group to exhibit increased activity levels and aggression in an attempt to increase energy intake. Spiders were maintained on three different treatment diets in order to simulate prey with differing macronutrient composition: high protein, intermediate, and high lipid. Spider respiration was measured to quantify the baseline metabolic rate (SMR), digestive metabolic rate (SDA), and active metabolic rate (AMR). We found no significant effect of diet on metabolic rates. However, the SDA coefficient (i.e. digestive cost relative to prey content) was higher in the high protein group, meaning that this group metabolized a greater portion of their prey during digestion and had a lower net energy intake from prey. In our behavioral assays, spiders in the high protein group were significantly more active and attacked prey more quickly in their first trial. Our results demonstrate that diet had relatively little effect on predator metabolism but more of an effect on behavior. These findings suggest that diet regulation should be analyzed by studying multiple responses together, including metabolism and behavior, to gain a more comprehensive understanding of the effects of diet on organism performance and fitness.
1. Introduction Predators often consume a diverse selection of prey and prey can vary widely in a number of traits including: size, defenses, activity level, and nutritional content (Mooney et al., 2010; Wilder et al., 2013). Prey traits can influence predator performance by affecting the amounts and types of nutrients gained from prey. Often, the total amount or balance of nutrients in prey may not match the optimal balance of nutrients required by predators, which can lead to reductions in performance or changes in behavior (Raubenheimer et al., 2007; Wilder and Rypstra, 2010; Jensen et al., 2011a, 2012; Wiggins et al., 2018). While the effects of dietary imbalance on life histories, physiology, and behavior have been studied in a number of herbivores and omnivores, much less is known about how dietary imbalances affect predators (Simpson and Raubenheimer, 2012; Simpson et al., 2015). The main macronutrients ingested by predators are lipid and protein (Wilder et al., 2013). Lipids provide a source of energy to fuel behavior or physiological processes (Jensen et al., 2011a). While protein can also be catabolized for energy, protein is primarily important for providing material to build enzymes and tissues (Walter et al., 2017). Individual components of fitness may require particular amounts or balances of ⁎
nutrients, which may not always be the same (Raubenheimer et al., 2007; Jensen et al., 2010, 2011a, 2012; Simpson and Raubenheimer, 2012). Hence, understanding how imbalanced diets affect multiple components of fitness simultaneously is important for gaining a more comprehensive understanding of diet consequences for animals. Body composition, physiology, and behavior may all be affected by diet. In terms of body composition, diet can affect lipid reserves, which may be a critical backup energy source to survive periods without prey. This may be especially important for predators such as spiders since evidence suggests that many species may be food limited in nature (Wise, 1975; Bilde and Toft, 1998; Wilder and Rypstra, 2008). As such, spiders and other predators may need to be efficient in their use of energy during metabolism including baseline metabolism (i.e., standard metabolic rate, SMR), active metabolic rate (AMR), and metabolism during digestion (i.e., specific dynamic action, SDA) (Tanaka and Itô, 1982; O’Connor et al., 2000; McCue, 2006; Secor, 2009; Van Leeuwen et al., 2012). Prey nutrient content could affect predator metabolism either directly (e.g., SDA during digestion) or through its influence on other aspects of physiology (e.g., SMR) behavior (e.g., AMR) (Nespolo et al., 2011). For example, low lipid or energy content of prey could result in increased foraging behavior as predators search for additional
Corresponding author. E-mail address: [email protected] (S.M. Wilder).
https://doi.org/10.1016/j.jinsphys.2019.04.006 Received 28 August 2018; Received in revised form 12 April 2019; Accepted 18 April 2019 Available online 19 April 2019 0022-1910/ © 2019 Elsevier Ltd. All rights reserved.
Journal of Insect Physiology 116 (2019) 25–31
N.A. Koemel, et al.
prey (Hazlett et al., 1975). Spiders are a particularly abundant and diverse group of predators. They are also voracious predators, with an estimated annual consumption of 400–800 metric tons of insects per year worldwide (Nyffeler and Birkhofer, 2017). Spiders have served as models for the study of predator diet because large numbers can be easily maintained in the laboratory, physiology and behavior can be easily measured, treatments have been developed that allow investigators to manipulate the nutrient content of prey, and experiments can be done with less ethical concerns than larger predators (Wilder, 2011). As such, there is a foundation of research on spider nutrition on which studies can build, to provide a more extensive understanding of the nutritional ecology of this common predatory group (Salomon et al., 2008; Wilder and Rypstra, 2010; Jensen et al., 2010, 2011b; Hawley et al., 2014). The overall goal of this study was to test how the macronutrient content of prey of the wolf spider, Hogna carolinensis, affected body composition, metabolism (i.e., SDA and AMR), and behavior. We predicted that spiders fed high lipid diets would incur lower energetic costs associated with digestive metabolic rate when compared to that of spiders on a high protein diet because catabolism of protein has been demonstrated to produce greater digestive costs than lipids and carbohydrates in a range of other taxa (Secor, 2009). As for the locomotor performance, we predicted that the spiders in the high lipid groups would have higher performance in endurance and sprint speed compared to the high protein treatments because of greater available energy in their diet. Lastly, we predicted that spiders consuming the high protein diet would have the lowest lipid reserves and, as a consequence, would be more active and more aggressive response towards prey to increase energy intake.
Following 10 days of starvation, wolf spiders were arranged by increasing mass, then sequentially assigned to one of the three cricket diet treatments. This assortment scheme, rather than one assigned at random, ensured that there were not initial differences between treatments in spider mass. After the wolf spiders were assigned to a treatment group, they were fed 2 treatment crickets twice a week for a total of 6 weeks. The spiders consumed 1 of 3 different cricket diet treatments for the 6-week duration (i.e., high protein, intermediate, or high lipid). All physiological and behavioral experimentation occurred during the week following the 6-week diet acclimation period. Wolf spiders are polyphagous predators that consume a wide variety of prey. While their diet is variable, they may experience periods of time where they consume similar prey due to a seasonal outbreak of a prey species or high density of particular prey in a habitat. Hence, the 6-week trial period may represent a somewhat long, although still realistic, time period for the spiders to feed on a particular diet. Crickets were purchased from a commercial supplier (“¼ inch” in length, © Fluker's Cricket Farm, 2016). Crickets were reared on diet treatments for 7–10 days to allow for body compositional changes prior to being fed to the wolf spiders (Barnes et al., 2019). A prior study (Barnes et al., 2019) developed these diets and showed that feeding them to crickets results in significant differences in the nutrient content of the cricket bodies. Cricket diets treatments consisted of high protein, intermediate, and high lipid groups (Table 1). Diets were constructed by incorporating liquid lipids with the dry powder components and then mixing thoroughly after allowing to dry. Water was provided ad libitum, which was replaced twice weekly. Representative crickets from each diet group were used to analyze prey nutrient content. These crickets were euthanized by freezing and stored at 0 OC, then dried at 60 OC for 24 h, weighed, and analyzed for nutrients.
2. Methods 2.1. Study animals
2.2. Respiration
Hogna carolinensis typically construct burrows used as daytime retreats but will forage away from the burrow at night in search of prey. Mature female wolf spiders, H. carolinensis, were collected from fields in Stillwater, Oklahoma during September 2016 (n = 46; Lipid n = 15, Intermediate n = 15, Protein n = 16) for experiments that tested how diet affected body composition and respiration of spiders. A second subset of spiders was collected (n = 48; Lipid n = 16, Intermediate n = 16, Protein n = 16) in June 2017 to conduct additional experiments to test how diet affected activity and aggression. The timing of collection and experiment duration coincided with the spider reproductive season, so most spiders produced egg sacs in the laboratory. The spiders did not differ in mass between the two collection periods (t 91 = 1.7, p = 0.10; 1.59 ± 0.05 mg versus 1.71 ± 0.05 mg). All egg sacs were removed shortly after their production. Wolf spiders were acclimated to laboratory conditions in individual, plastic deli containers (1420 mL). We cut circular pieces of paper towels to act as substrates for the floor of each container. Then, we cut a 946 mL opaque deli container in half to create a shelter for each spider. Additionally, a moist cotton ball was placed in each spider container and exchanged weekly. Paper substrates were replaced and the spider housing was cleaned each week. Spiders were housed at 25 ± 1 OC and 14L:10D light regime. Housing location within shelving units was rotated weekly. Prior to experiments, the spiders were fed two crickets, Acheta domesticus that were reared on dog-food (Rachel Ray Nutrish) twice during the first week, then fasted for 10 days to clear spider gut contents and standardize the spider hunger level (Barnes et al., 2019). Starvation periods of other spiders in the field have been estimated to be between 4 and 8 days on average (Bilde and Toft, 1998). Although, one study of wolf spiders found that the body condition of field-collected spiders was similar to spiders that had been fed ad libitum and then deprived of food for 3 months (Wilder and Rypstra, 2008).
The SDA coefficient was calculated as SDA divided by the amount of prey energy ingested (i.e. total prey energy minus the energy of prey remains) (Beaupre, 2005). Activity also has an influence on the metabolic rate; this is demonstrated by the increase in metabolism during or after activity known as the active metabolic rate (AMR). AMR can be energetically expensive for many organisms but has not been exhaustively studied for wolf spiders (Ford, 1977). We used closed system respirometry to determine the SMR, AMR, and SDA rates of spiders. The spiders were starved for 10 days after the 6-week treatment diets in order to standardize feeding level. Spiders were placed into individual 946 mL deli containers. Sealable ports were installed on the top of the container to allow CO2 readings to be Table 1 The ingredients incorporated into the three experimental diets fed to the crickets (in grams). The percentage of macronutrients for each diet are designated at the top (Protein:Lipid:Carbohydrate).
Egg white Micellar Casein Sugar Flour Cellulose Nipagin True Balance™ Vitamin (NOW Foods; Capsule) Cholesterol Fish Oil Lard Olive Oil
26
Lipid (10:45:45)
Intermediate (70:15:15)
Protein (100:00:00)
11 11 55 85 94 1 1
105 105 14 22 43 1 1
137 137 0 0 26 1 1
0.5 3 22 22
0.5 3 5.5 5.5
0.5 3 0 0
Journal of Insect Physiology 116 (2019) 25–31
N.A. Koemel, et al.
collected without opening the container. The amount of CO2 respired into the container was then measured via a Li-840A CO2/H2O respirometer (LiCor; Lincoln, NE, USA; Barnes et al., in press). We measured SMR twice daily (09:00 and 21:00) in the final two days of the spider starvation period. For SDA, the wolf spiders were fed two crickets (a total of 10–15% of spider body mass) from their assigned treatment group. Additionally, prey were size matched across all treatments. To do this, we pre-weighed all crickets and ensured that the range of cricket weights for all treatments was broadly overlapping. The respiration chambers were then re-sealed following the feeding. Respiration was measured at 3-hour intervals for the first 9 h following the introduction of prey, then respiration measures were resumed at 12hour intervals for the next 3 days to monitor the attenuation of the SDA response. We found that the metabolic rate of the spiders returned to baseline within 2–3 days after feeding (see results). Accumulated CO2 was converted to percent by dividing by 10,000. This value was converted to the total volume of CO2 (VCO2) by dividing by 100, and then multiplied by the volume of the container (946 mL). Metabolism was calculated as the final minus initial respiration reading, divided by the sampling duration. Units of respiration were converted to energy using the conversion factor 24.65 kJ L CO2−1(Chown et al., 2007). Control respiration chambers without animals, but otherwise similar handling, were run alongside treatment chambers to test if handling conditions affected CO2 readings, which they did not. Spiders generally completed the extra-oral digestion process and deposited boluses of prey remains within nine hours of introduction of prey and time to first excretion was between 24 and 48 h. From observations of spiders, it appeared that feeding was finished within 24 h and most excreta production occurred within 48 h. Uneaten prey remains were collected within 24 h following each feeding. We were also interested in how diets affected active metabolic rate (AMR). However, rather than measure metabolic rate during activity, we measured metabolic rate during recovery after physical activity. Spiders’ muscles contain few mitochondria (Foelix, 2011). This limits the endurance of spiders, which typically fatigue relatively quickly, and suggests that the metabolic costs of physical activity may be incurred during recovery more so than during the physical activity itself. Hence, we used closed respirometry to measure the metabolic rate of a spider during post-locomotor, recovery (hereafter referred to as AMR). When the spiders had fully completed the locomotor performance trials (described below) they were immediately placed into individual respiration chambers to measure gas exchange. The starting CO2 of the container was measured and then a CO2 measurement was taken in intervals of 12 h. The measurements in parts-per-million (ppm) were converted to volume (mL) using the same protocol from the SDA and SMR trials. AMR trials were repeated twice over two days. At the end of the second trial, the spiders were sacrificed and their lipid and protein composition was measured using the same procedure used for the prey (described below).
however, those who stopped mid-track were encouraged to continue with a light stroke of a paintbrush (Prenter et al., 2010). We filmed the first and last 25 cm of sprints using a Canon Vixia HF R52 camera stationed on a tripod above the track. Once the experiment was completed, the videos were processed using ImageJ 1.48v (National Institutes of Health, USA) to measure the time to complete a 150 cm sprint. At the end of the 75 cm linear track was an opening to a circular arena. The spiders were introduced to the container after completing the 150 cm sprint. Once inside the container, spider endurance was measured by documenting the time elapsed until the spider reached full fatigue. The spiders were periodically tapped with the paintbrush to maintain movement around the perimeter of the container. Spiders were considered to be fatigued upon loss of righting response when spiders were flipped over. The endurance and sprint trials were immediately followed by the assessment of AMR. All of the trials were repeated twice approximately 2 days apart. Spider activity and aggression were measured by conducting a behavioral assay in a circular arena with a filter paper-substrate (25 cm diameter plastic deli containers). The circular arena was separated into two equal semi-circle portions of filter paper. One half had a blank filter paper, and the other half contained filter paper with the prey chemical cues. Heterogeneity of chemical cues in the arena could stimulate predators to continue searching the area. Prey chemical cues were collected by placing a live cricket on the filter paper for 24 h prior to the trial and removing the cricket before the trial began. Each spider was released in the center of the arena prior to the recording. Individual spiders were observed for 35 min after release into the arena. The first 5 min of recording was omitted from data analysis in order to allow for arena acclimation. For the movement assay, the cumulative movement (i.e. total time spent moving), velocity, and distance moved were extracted from video recordings using Ethovision XT13 (Noldus Information Technologies, Wageningen, The Netherlands). After the activity trials, prey capture was observed by introducing a single cricket into the arena and measuring the time lapse before the prey item was attacked and killed (Walker et al., 1999). Prior to measuring the attack latency time, the spiders and prey were acclimated in isolated chambers to the arena for 5 min. The acclimation chambers for prey and predator were placed on opposite sides of the circular arena in each trial. The trials for both attack latency and locomotor behavior were repeated for a total of two trials each. 2.4. Nutrient analyses We measured the lipid content of the whole crickets, remains of cricket bodies after the spider finished feeding, and spiders using a gravimetric assay with chloroform (Wilder et al., 2013). Whole crickets maintained on the experimental diets were previously analyzed in another study (Barnes et al., 2019) but we analyzed samples from the current study to confirm that the treatments were still effective in manipulating cricket nutrient content. Briefly, the items were first dried at 60 OC for 24 h and weighed. Once dried, they were individually placed into 15 mL vials and washed with 10 mL of pure chloroform for 24 h. After the 24-hour period of chloroform washing, the chloroform was drained and replaced with fresh chloroform and repeated for a total of three washes to facilitate full lipid extraction. After extraction, the contents were again dried at 60 OC for 24 h and reweighed. Lipid content was quantified by subtracting the mass of dried lipid-free components from the initial dry mass of the item. Protein was determined in triplicates using the Bradford Assay of lean, ground samples (i.e., after lipid extraction) (Wilder et al., 2013). Carbohydrates were not estimated, as they are typically present in arthropods at low concentrations (Raubenheimer and Rothman, 2013). Protein and lipid dry masses were converted to energy using standard conversion factors (protein = 17 kJ/g and lipid = 37 kJ/g; Raubenheimer and Rothman, 2013).
2.3. Behavior We also examined the effects of diet on behavior. After maintaining the spiders on the experimental diets for the 6-week period, the spiders’ locomotor performance was first individually examined as sprint speed of the spiders using a linear track. The behavior trials were repeated 3 days apart for a total of two trials. We conducted two separate trials for behavior to test if the behavioral responses were consistent and robust. The spiders were placed in a small black box at the start point and allowed 2 min to acclimate before beginning the trial. After the acclimation period, the front box lid was removed. Each trial began with the spiders being released from the acclimation chamber. The spiders were prompted to travel a full 75 cm distance of the track and back to the starting line for a total distance of 150 cm, which was timed using a stopwatch. The majority of the spiders ran the complete track; 27
Journal of Insect Physiology 116 (2019) 25–31
N.A. Koemel, et al.
Nutrient intake of spiders was estimated by subtracting the amount of nutrients present in prey remains from the amounts of nutrients estimated to be in prey before they were fed on by spiders. The initial nutrient content of prey fed to spiders was estimated using linear regression relationships between prey wet mass and nutrient content that were developed using representative crickets from each treatment (as in Wilder and Rypstra, 2010). 2.5. Data analysis Differences in animal composition were tested using ANOVA and Tukey HSD post hoc analysis, while spider performance was evaluated using ANCOVA with body mass as a covariate. Statistical analyses were conducted in JMP 13 software package (SAS Institute, Cary, NC, USA). Respiration was compared using repeated measures analysis of covariance in SAS 9.4 software package (SAS Institute, Cary, NC, USA), with repeated measures over time and spider body mass as a covariate. Behavioral measures were first analyzed for each trial date using MANOVA to test for significant effects of prey nutrient treatments and then we ran a repeated measures analysis of variance. The attack latency response variable was not normally distributed, so we applied a log-transformation. Statistical significance was evaluated at the 0.05 level.
Fig. 2. The lipid composition of the different H. carolinensis wolf spider treatments at the end of the six weeks of feeding (mg 100 mg−1 dry mass). Bars with different letters are significantly different from each other in post hoc analyses.
p = 0.07, Start = 1.5891 ± 0.0610 g, F2, 44 = 2.5, End = 1.47204 ± 0.0658 g). Spider body composition was measured in experiment 1. At the end of the six weeks of diet treatments, there was no significant difference in protein content of the wolf spiders fed different diets (percent protein, mean ± 1 SE: Lipid = 59.6 ± 1.1, Intermediate = 60.8 ± 1.8, Protein = 62.4 ± 1.5; F2, 41 = 0.9; p = 0.42). However, there was a significant difference in the lipid content by treatment. There was a higher lipid content in spiders fed high lipid diets compared to the protein and intermediate treatment groups (F2, 41 = 8.8; p < 0.01 Fig. 2).
3. Results 3.1. Body composition Crickets from the high lipid treatment had significantly higher lipid content compared to crickets from the intermediate and high protein treatments (F2, 48 = 30.3, p < 0.01; Fig. 1). There were no differences in cricket protein content between treatment groups (F2, 54 = 2.6; p = 0.11; Fig. 1). The energetic content of crickets from the high lipid treatment (1.74 ± 0.01 kJ) was greater than the intermediate (1.42 ± 0.02) and protein (1.18 ± 0.01) diet treatments (F2, 38 = 248.6, p < 0.01). There was no significant difference between diet treatments in the mass of spiders at the start of the experiment in either the first (F2, 45 = 0.0, p = 0.97) or second sets of experiments (F2, 47 = 0.0, p = 0.99). The increase in mass of the spiders over the course of the 6week feeding period was not significantly different between treatments in the first or second experiment (Experiment 1: F2, 44 = 6.7, p = 0.97, Start = 1.7083 ± 0.0522 g, End = 1.9660 ± 0.0552 g; Experiment 2:
3.2. Respiration Following prey capture and consumption, there was a significant increase in metabolic rate of spiders above baseline levels that peaked after about 4 h and lasted for a duration of 48 h (Fig. 3). Larger spiders respired more (F1, 30 = 22.2, p < 0.01), but we did not find significant differences in SDA between treatments (F2, 30 = 2.5, p = 0.10; Fig. 3). The SDA coefficient, which is the proportion of prey energy used during the SDA response, was significantly higher for spiders in the protein prey treatment (F2, 38 = 3.8, p = 0.03; Fig. 4). The post-activity respiration trials taken for AMR were significantly higher than the SMR (F4, 87 = 4.8, p < 0.01). We found that AMR increased with spider mass (F4, 87 = 12.2, p < 0.01). However, we did not find an effect of diet treatment on AMR (mL CO2/hr, mean ± 1 SE: Lipid = 0.074 ± 0.011, Intermediate = 0.083 ± 0.011, Protein = 0.056 ± 0.011; F4, 87 = 1.5, p = 0.23) 3.3. Behavior When analyzing the sprint times in a repeated measures ANOVA analysis, we did not find significant differences in endurance (i.e., time to fatigue) between treatments (F5, 38 = 0.4, p = 0.82) nor did sprint time differ between treatments (F5, 37 = 0.8, p = 0.59). There was also no effect of time in these trials (Endurance; F2, 38 = 0.0, p = 0.93, Sprint; F1, 32 = 1.1, p = 0.30). When we examined these performance trials in a separate ANOVA for each trial individually, there were no significant differences by treatment (Endurance; F2, 38 = 0.6, p = 0.55, Fig. 5a; Sprint; F2, 32 = 0.6, p = 0.53, Fig. 5b). When we used MANOVA to test if behavioral responses (i.e. total distance, velocity, time spent moving, and attack latency) were dependent on prey nutrient treatments, we found an effect of nutrition on day 2 (F8,66 = 2.3, p = 0.03) but not day 1 (F8,82 = 1.9, p = 0.07). A repeated measures ANOVA was conducted to analyze velocity data
Fig. 1. The lipid and protein composition of the cricket treatments (mg 100 mg−1 dry mass). There was no significant effect of diet on protein content. For lipid, bars with different letters are significantly different from each other in post hoc analyses. 28
Journal of Insect Physiology 116 (2019) 25–31
N.A. Koemel, et al.
Fig. 3. Change in metabolic rate (i.e., elevation of metabolic rate above the baseline, which is shown in the dashed line) of the wolf spider Hogna carolinensis following prey consumption of a specific treatment cricket (SDA; mL CO2 hr−1). Diet treatment macronutrient ratios (P:L:C), protein diet (100:00:00), lipid (10:45:45), and intermediate (70:15:15).
Fig. 4. The proportion of the SDA to the amount of prey energy ingested by the wolf spiders (SDA coefficient) between the different treatment groups.
from both trial days together. We found an overall significant effect of treatment on spider velocity (F1, 46 = 8.8, p < 0.01). There was also an overall time effect between the two trials, with higher average velocity on day 2 (F1, 46 = 5.9, p = 0.02). In a separate analysis of data from each day, there was not a statistically significant difference in velocity by treatment on day 1 (F2, 47 = 2.7, p = 0.08; Fig. 6a). On day 2, the difference in velocity was statistically significant between treatments, where the protein treatment had a significantly higher velocity compared to the lipid and intermediate treatment groups (F2, 47 = 6.2, p < 0.01; Fig. 6a). For the total distance traveled by spiders, a repeated measures ANOVA showed that there was an overall significant treatment effect (F1, 46 = 8.2, p < 0.01) and no effect of time (i.e., between the days) (F1, 46 = 1.9, p = 0.18. Spiders in the high protein diet treatment traveled a significantly greater total distance than spiders in the intermediate and high lipid treatments (Fig. 5b). Separate ANOVA analysis for each day of trials confirmed that distance traveled was higher in the high protein treatment on both day 1 (F2, 46 = 4.2, p = 0.02; Fig. 6b) and day 2 (F2, 46 = 5.5, p = 0.01; Fig. 6b). For cumulative time spent moving, there was an overall significant treatment effect when looking at both trials together in a repeated measures ANOVA (F1, 46 = 3.6, p = 0.01). There was also an overall time effect with higher average time spent moving on the second day of trials (F1, 46 = 5.9, p = 0.02). When analyzing each trial separately, spiders in the high protein treatment spent more time moving compared to intermediate and lipid treatment for both day 1 (F2, 47 = 3.8, p = 0.03; Table 2) and day 2 (F2, 47 = 5.7, p = 0.01; Table 2).
Fig. 5. a) The endurance performance in seconds (i.e., time until full exhaustion) of the spiders in their respective treatment groups. b) The sprint performance (i.e., time to travel 150 cm) of the spiders on the diet treatments.
For attack latency, there was a significant difference between treatments. In trial one, the protein group caught prey faster than the lipid and intermediate group (F1, 46 = 5.9, p = 0.02; Table 2), but no differences between treatments in trial two (F1, 38 = 0.4, p = 0.50; Table 2). There was no overall significant treatment effect when combining the two days in a repeated measures ANOVA analysis (F1, 44 = 0.0, p = 0.96). However, there was an overall time effect between the two trials with longer latencies on the second day (F1, 44 = 6.9, p = 0.01).
29
Journal of Insect Physiology 116 (2019) 25–31
N.A. Koemel, et al.
metabolism had we fed the spiders a more limited amount of food. Overall, these results suggest that diet can have multiple simultaneous impacts on animal composition, physiology, and behavior and that a more complete understanding of the effects of diet on animals may require simultaneously considering these multiple measures. While prey varied in lipid and energy content, these treatments had no effect on the standard or digestive metabolic rates of spiders. This result was similar to that of a previous study that found no effect of prey lipid and protein content on the SDA of a wolf spider (Jensen et al., 2010). Previous studies have also found that SDA is relatively high in proportion to prey energy (i.e. SDA coefficient; Zaidan and Beaupre (2003), Bessler et al. (2010)). Overall, the amount of digestive energy expended by spiders in our study was approximately 5–7% of the total meal energy (Fig. 4). More specifically, although SDA did not differ between lipid-based and protein-based diets, the SDA coefficient was greater in spiders fed protein-biased diets. We found that spiders used a greater amount of protein-biased prey energy during digestion compared to lipid-based prey-fed spiders. The differences in the SDA coefficient appear to be driven by differences in prey energy content (i.e., spiders ingested more energy from high lipid prey) since overall metabolic costs of digesting prey did not differ between treatments. There were no significant effects of diet on AMR, sprint speed, or endurance. These three measures likely comprise a small fraction of the energy and time budget of this study species. Spiders are extremely efficient predators with a very low amount of metabolically active physiological components compared to other arthropods (Anderson, 1970). The wolf spider Hogna carolinensis is a sit-and-wait predator that appears to spend much of its time waiting to encounter prey that pass near their burrows (Suter et al., 2011). When we collected spiders in the field, they were often within 25 cm of their burrows, which means they would only have a short distance to run from predators. In addition, they typically waited for prey to be within 5–10 cm before attacking in prey capture trials, which would also represent a short time spent sprinting. Hence, in addition to not being significantly different between treatments, these three measures are likely relatively less important components of the overall fitness of these spiders in nature. Conversely, in other species of spiders, such as male web-building spiders that wander in search of females, AMR can be a more important component of fitness (Kasumovic and Seebacher, 2013). Two opposite predictions could be made as to the effects of diet on spider movement. First, high lipid spiders could be predicted to move more since they have higher stored energy and, hence, the surplus energy that they are able to expend on movement, including exploration of new habitats. Alternatively, spiders on the high protein diet could be predicted to move more because they have less energy, thus greater movement may increase the chances of encountering prey (Pekár et al., 2010). In our study, spiders on the high protein diet were significantly more active in terms of total distance moved, cumulative movement, and velocity. The high protein fed spiders were also captured prey the quickest in the first foraging trial. Given these two results and that the arenas for the behavioral trials contained prey cues, it seems likely that the higher activity of protein fed spiders was prey searching behavior. The differences that we observed were likely a result of differences among treatments in the motivation of spiders to move and not the availability of energy for movement. All of the treatments were provided with moderate amounts of food providing adequate energy for spiders to maintain or gain mass in all treatments. Higher activity levels due to prey searching behavior could increase predation risk in nature, as wolf spiders are prey to a wide range of visually hunting predators (Cloudsley-Thompson, 1995). This study was conducted on adult female wolf spiders during the reproductive season. Nearly all spiders produced an egg sac, which was immediately removed, during the period on which they were maintained on their treatment diets. It is unclear if the results of this study apply to only reproductive females or might also apply to juveniles and adult males. We would predict that high protein diets might contribute
Fig. 6. a) Mean velocity of the spider (cm/s). b) Total distance traveled (cm) of the spiders. Comparisons were made between diet treatments. The changes from trial one and trial two are also shown. Values with unlike letters represent significant treatment differences in the given trial (P < 0.05). Table 2 The mean behavior of spiders fed different diets in the experimental trials. Comparisons were made between diet treatments. The changes from trial one and trial two are also shown. Values with unlike letters represent significant differences between treatments (P < 0.05). Diet Treatment
Protein
Intermediate
Lipid
Cumulative Movement (s) Trial 1 59.11a ± 18.12 Trial 2 95.61a ± 26.19
1.17b ± 0.48 9.70b ± 6.79
14.79b ± 8.54 5.42b ± 2.36
Attack Latency (s) Trial 1 Trial 2
21.0 ± 4.59 70.8 ± 23.21
39.0 ± 11.34 40.7 ± 14.69
17.0 ± 3.17 30.7 ± 14.90
4. Discussion Overall, our results show that diets biased in lipid versus protein affect some measures of body composition (i.e., lipid reserves), physiology (i.e., respiration), and behavior (i.e., movement). For body composition, high lipid diets increased lipid reserves. In terms of metabolism, diets had no effect on SDA or AMR; although spiders on the high lipid treatment had lower SDA coefficients. For behavior, while there were no effects of diet on sprint speed or endurance, spiders in the high protein treatment were more active, on average, and captured prey more quickly in their first trial than spiders in the other diet treatments. One explanation for the significant effects of diet on behavior but not metabolism is that behavior is more sensitive to diet than metabolism. In the current study, spiders had a moderate quantity of food that varied in nutrient content. Hence, the spiders may have shifted their behavior in an attempt to change their diet but may not have been stressed by the diet to the point that it affected their metabolism. It is possible that we would have seen effects on both behavior and 30
Journal of Insect Physiology 116 (2019) 25–31
N.A. Koemel, et al.
to higher growth of juveniles, whereas high lipid diets might be especially important to adult males since they spend much of their time actively searching for females (Wilder, 2011). Relatively little is known about the comparative dietary requirements of male and female spiders. Although, studies of crickets support the idea that female fitness is enhanced on more protein biased diets while males benefit more from more energy (i.e., carbohydrate) rich diets (Maklakov et al., 2008). Furthermore, relatively little is known about the comparative nutritional ecology of different species (Wilder and Eubanks, 2010). Spiders differ widely in foraging strategies from actively pursuing prey to spending most of their time in sit-and-wait foraging on a web. Evidence suggests that these foraging strategies could be associated with overall nutrient requirements as a more active jumping spider grew larger on higher lipid diets (Wiggins et al., 2018) while a more sedentary wolf spider grew larger on more protein biased diets (Jensen et al., 2011b). Further research on diverse species and life stages are needed to gain a more general understanding of predator diet requirements, how they vary among predators, and how they compare to animals at other trophic levels.
Hawley, J., Simpson, S.J., Wilder, S.M., 2014. Effects of prey macronutrient content on body composition and nutrient intake in a web-building spider. PLoS One 9 (6), E99165. Kasumovic, M., Seebacher, F., 2013. The active metabolic rate predicts a male spider's proximity to females and expected fitness. Biol. Lett. 9 (2), 20121164. Maklakov, A.A., Simpson, S.J., Zajitschek, F., Hall, M.D., Dessmann, J., Clissold, F., Brooks, R.C., 2008. Sex-specific fitness effects of nutrient intake on reproduction and lifespan. Curr. Biol. 18 (14), 1062–1066. McCue, M.D., 2006. Specific dynamic action: a century of investigation. Comp. Biochem. Physiol. Part A: Mol. Integr. Physiol. 144 (4) 381394. Mooney, K.A., Gruner, D.S., Barber, N.A., Van Bael, S.A., Philpott, S.M., Greenberg, R., 2010. Interactions among predators and the cascading effects of vertebrate insectivores on arthropod communities and plants. Proc. Nat. Acad. Sci. U.S.A 107, 7335–7340. Nespolo, Correa, Pérez-Apablaza, Cortés, Bartheld, 2011. Energy metabolism and the postprandial response of the Chilean tarantulas, Euathlus truculentus (Araneae: Theraphosidae). Comp. Biochem. Physiol. Part A 159 (4), 379–382. Nyffeler, M., Birkhofer, K., 2017. An estimated 400-800 million tons of prey are annually killed by the global spider community. Sci. Nat. https://doi.org/10.1007/s00114017-1440-1. O'Connor, K.I., Taylor, A.C., Metcalfe, N.B., 2000. The stability of standard metabolic rate during a period of food deprivation in juvenile Atlantic salmon. J. Fish Biol. 57 (1), 41–51. Pekár, Mayntz, Ribeiro, Herberstein, 2010. Specialist ant-eating spiders selectively feed on different body parts to balance nutrient intake. Anim. Behav. 79 (6), 1301–1306. Prenter, J., Pérez-Staples, D., Taylor, P.W., 2010. The effects of morphology and substrate diameter on climbing and locomotor performance in male spiders. Funct. Ecol. 24 (2), 400–408. Raubenheimer, D., Rothman, J.M., 2013. Nutritional ecology of entomophagy in humans and other primates. Annu. Rev. Entomol. 58, 141–160. Raubenheimer, D., Mayntz, D., Simpson, S.J., Tøft, S., 2007. Nutrient specific compensation following dipause in a predator: implications for Intraguild predation. Ecology 88 (10), 2598–2608. Salomon, M., Mayntz, D., Lubin, Y., 2008. Colony nutrition skews reproduction in a social spider. Behav. Ecol. 19 (3), 605–611. Secor, S.M., 2009. Specific dynamic action: a review of the postprandial metabolic response. J. Comp. Physiol. B 179 (1), 1–56. Simpson, S., Raubenheimer, D., 2012. The Nature of Nutrition: A Unifying Framework from Animal Adaptation to Human Obesity. Princeton University Press, Princeton. Simpson, S.J., Le Couteur, D.G., Raubenheimer, D., 2015. Putting the balance back in diet. Cell 161 (1), 18–23. Suter, R.B., Stratton, G.E., Miller, P.R., 2011. Mechanics and energetics of excavation by burrowing wolf spiders, Geolycosa spp. J. Insect Sci. 11, 22. https://doi.org/10. 1673/031.011.0122. Tanaka, K., Itô, Y., 1982. Decrease in respiratory rate in a wolf spider, Pardosa astrigera (L. Koch), under starvation. Res. Popul. Ecol. 24 (2), 360–374. Walker, S., Marshall, S., Rypstra, A., Taylor, D., 1999. The effects of hunger on locomotory behaviour in two species of wolf spider (Araneae, Lycosidae). Anim. Behav. 58, 515–520. Walter, A., Bechsgaard, J., Scavenius, C., Dyrlund, T., Sanggaard, K., Enghild, J., Bilde, T., 2017. Characterisation of protein families in spider digestive fluids and their role in extra-oral digestion. BMC Genomics 18. Wiggins, W., Bounds, D., Wilder, S., 2018. Laboratory-reared and field-collected predators respond differently to same experimental treatments. Behav. Ecol. Sociobiol. 72 (2), 1–8. Wilder, S.M., 2011. Spider nutrition: an integrative perspective. Adv. Insect Physiol. 40, 87–136. Wilder, S.M., Eubanks, M.D., 2010. Might nitrogen limitation promote omnivory among carnivorous arthropods? Comment. Ecol. 91 (10), 3114–3117. Wilder, S.M., Rypstra, A., 2008. Diet quality affects mating behaviour and egg production in a wolf spider. Anim. Behav. 76 (2), 439. Wilder, S.M., Rypstra, A.L., 2010. Males make poor meals: a comparison of nutrient extraction during sexual cannibalism and predation. Oecologia 162 (3), 617–625. Wilder, S.M., Norris, M., Lee, R.W., Raubenheimer, D., Simpson, S.J., 2013. Arthropod food webs become increasingly lipid-limited at higher trophic levels. Ecol. Lett. 16, 895–902. Wise, D., 1975. Food limitation of the spider Linyphia marginata: experimental Field Studies. Ecology 56 (3), 637–646. Van Leeuwen, T.E., Rosenfeld, J.S., Richards, J.G., 2012. Effects of food ration on SMR: influence of food consumption on individual variation in metabolic rate in juvenile coho salmon (Onchorhynchus kisutch). J. Anim. Ecol. 81 (2), 395–402. Zaidan, F., Beaupre, S., 2003. Effects of body mass, meal size, fast length, and temperature on specific dynamic action in the timber rattlesnake (Crotalus horridus). Ecol. Evol. Physiol. Biochem. Zool.
Acknowledgements We would like to thank the Wilder Lab group for assistance in the lab and constructive comments on the paper. This research was supported by the Department of Integrative Biology at Oklahoma State University and the Wentz Research Grant from the Lew Wentz Foundation awarded to NAK. References Anderson, J.F., 1970. Metabolic rates of spiders. Comp. Biochem. Physiol. 33 (1), 51–72. Barnes, C.L., Hawlena, D., Wilder, S.M., 2019. Predators buffer the effects of variation in prey nutrient content for nutrient deposition. Oikos 128 (3), 360–367. Barnes, C.L., Hawlena, D., McCue, M.D., Wilder, S.M., in press. Consequences of prey exoskeleton content for predator feeding and digestion: black widow predation on larval versus adult mealworm beetles. Oecologia. Beaupre, S., 2005. Ratio representations of specific dynamic action (MassSpecific SDA and SDA Coefficient) do not standardize for body mass and meal size. Physiol. Biochem. Zool. 78 (1), 126–131. Bessler, S., Stubblefield, M., Ultsch, G., Secor, S., 2010. Determinants and modeling of specific dynamic action for the Common Garter Snake (Thamnophis sirtalis). Can. J. Zool. Revue Canadienne De Zoologie 88 (8), 808–820. Bilde, T., Toft, S., 1998. Quantifying food limitation of arthropod predators in the field. Oecologia 115 (1), 54–58. Chown, S.L., Marais, E., Terblanche, J.S., Klok, C.J., Lighton, J.R.B., Blackburn, T.M., 2007. Scaling of insect metabolic rate is inconsistent with the nutrient supply network model. Funct. Ecol. 21, 282–290. Cloudsley-Thompson, J.L., 1995. A review of the anti-predator devices of spiders. Bull. Br. Arachnol. Soc. 10 (3), 81–96. Foelix, R., 2011. Biology of Spiders, third ed. Oxford University Press, Oxford; New York. Ford, M., 1977. Metabolic costs of the predation strategy of the spider Pardosa amentata (Clerck) (Lycosidae). Oecologia 28 (4), 333–340. Hazlett, B., Rubenstein, D., Rittschof, D., 1975. Starvation, energy reserves, and aggression in the crayfish Orconectes virilis (Hagen, 1870) (Decapoda, Cambaridae). Crustaceana 11–16. Jensen, Mayntz, Wang, Simpson, Overgaard, 2010. Metabolic consequences of feeding and fasting on nutritionally different diets in the wolf spider Pardosa prativaga. J. Insect Physiol. 56 (9), 1095–1100. Jensen, K., Mayntz, D., Toft, S., Raubenheimer, S., Simpson, J., 2011a. Prey nutrient composition has different effects on Pardosa wolf spiders with dissimilar life histories. Oecologia 165 (3), 577–583. Jensen, K., Mayntz, D., Toft, S., Raubenheimer, S., Simpson, J., 2011b. Nutrient regulation in a predator, the wolf spider Pardosa prativaga. Anim. Behav. 81 (5), 993–999. Jensen, K., Mayntz, D., Toft, S., Clissold, F.J., Hunt, J., Raubenheimer, D., Simpson, S.J., 2012. Optimal foraging for specific nutrients in predatory beetles. Proc. R. Soc. London B Biol. Sci rspb20112410.
31