Synopsis
Transitions between life-history stages are often accompanied by dramatic behavioral switches that result from a shift in motivation to pursue one resource over another. While the neuroendocrine mechanisms that regulate such behavioral transitions are poorly understood, arginine vasotocin (AVT) and neuropeptide Y (NPY) are excellent candidates because they modulate reproductive and feeding behavior, respectively. We asked if seasonal changes in AVT and NPY are concomitant with the seasonal migration to and from the feeding grounds in red-sided garter snakes (Thamnophis sirtalis parietalis). Male and female snakes were collected in different migratory states during both the spring and fall. The total number of AVT- and NPY-immunoreactive (ir) cells was then quantified in each brain region of interest. To correct for potential variation in region volume related to sexually dimorphic body size in this species, we first determined that snout–vent length is an accurate proxy for regional brain volume. We then corrected each individual’s ir cell number by its SVL to directly compare seasonal changes in AVT and NPY between males and females. Within the supraoptic nucleus, both males and females had more AVT-ir cells during the fall compared with spring. As predicted, males had significantly more AVT-ir cells during the spring mating season in the hypothalamus (HYP) and bed nucleus of the stria terminalis, brain regions important in regulating reproductive behavior. Females also had significantly more AVT-ir cells in the HYP during the spring, as well as a significantly higher number of hypothalamic AVT cells than males. During the fall, males had significantly more NPY-ir cells in the cortex and posterior HYP compared with spring, possibly reflecting increased feeding behavior during summer foraging. Females did not exhibit significant main effects of season on NPY-ir cell number in any region. Neither AVT- nor NPY-ir cell number varied significantly with migratory status, but we did observe significant changes related to seasonal transitions in reproductive state. Our results indicate that changes in brain AVT and NPY are associated with seasonal transitions in reproductive and foraging behaviors, and may be involved in mediating sex differences in the timing of life-history events.
Introduction
Many vertebrates exhibit distinct life-history stages that are associated with specific physiological and behavioral functions. Examples of such life-history stages include reproduction, migration, and foraging (Wada 2008). Critically, many animals transition from one life-history stage to another as resource availability changes. For example, to maximize reproductive success, organisms should exhibit reproductive physiology and behavior when conspecifics are in breeding condition. Likewise, the probability of individual survival is maximized if animals exhibit foraging and feeding behavior when food is abundant. For many seasonal organisms, resource availability relegates life-history stages to a specific time of year, and therefore seasonal transitions between life-history stages are often accompanied by dramatic changes in both physiology and appetitive and consummatory behavior. The neuroendocrine mechanisms that regulate these seasonal transitions, however, are poorly understood, particularly with regard to ecologically relevant contexts (Wingfield 2008).
Neuropeptides within specific brain regions likely play an important role in mediating seasonal life-history transitions, because they modulate physiology and behavior as well as vary seasonally. Two neuropeptides that are likely candidates for regulating seasonal transitions between reproduction and foraging are arginine vasotocin (AVT) and neuropeptide Y (NPY). AVT, and its mammalian homolog, arginine vasopressin (AVP), play a critical role in mediating reproductive behavior and are known to have context-dependent effects (Goodson and Bass 2001; Wilczynski et al. 2005). For example, AVT/AVP cell number changes both seasonally and varies with sex in response to reproductive and social behaviors (e.g., Boyd 1991, 1994; Tito et al. 1999; Toyoda et al. 2003; Goodson 2005; Madison et al. 2008). Indeed, treatment of male rough-skinned newts (Taricha granulosa) with exogenous AVT increased courtship behavior (Moore and Miller 1983). Locations of neurons that produce AVT/AVP are also conserved across vertebrates (De Vries and Boyle 1998; Foran and Bass 1999; Goodson and Bass 2001; Lim et al. 2004; Goodson 2005), making it a target peptide for understanding the general mechanisms of reproductive life-history transitions.
By contrast, NPY is a neuromodulatory peptide that regulates appetite and feeding behavior in all vertebrates studied (Maniam and Morris 2012). As one of the most abundant neuropeptides in the central and peripheral nervous systems (Malva et al. 2012), NPY increases both appetitive (e.g., food-seeking) and consummatory (e.g., ingestive) feeding behavior (Morris and Crews 1990; Mercer et al. 2011; Schneider et al. 2013). Sahu et al. (1988) investigated the effects of 2–4 days of food deprivation and 4 days of food deprivation followed by a 1-day ad libitum re-feeding period on NPY-like peptide concentrations in six hypothalamic nuclei of adult male rats (Rattus norvegicus). NPY was extracted from each microdissected nucleus and its NPY-like peptide concentration determined via radioimmunoassay. Within the paraventricular nucleus (PVN), NPY concentrations increased significantly in response to 3 and 4 days of food deprivation and decreased following re-feeding. NPY-like concentrations in the medial preoptic area (POA), however, increased in response to re-feeding. Such results support the hypothesis that the function of NPY is region-specific. For example, NPY not only affects food intake, but it also has inhibitory effects on sexual behavior (e.g., Ammar et al. 2000; Schneider et al. 2013). Morris and Crews (1990) observed increased feeding behavior and decreased courtship behavior in male red-sided garter snakes (Thamnophis sirtalis parietalis) receiving intracerebroventricular treatment with exogenous NPY. Thus, changes in NPY within specific brain regions may play an important role in mediating trade-offs between reproduction and feeding.
Independently, there is an abundance of research examining how both AVT and NPY regulate reproductive and feeding behaviors, respectively (e.g., Goodson and Bass 2001; Maniam and Morris 2012). However, few studies have examined if and how these neuropeptides interact to regulate transitions between reproduction and feeding. Because these behavioral transitions typically occur on a short time scale (i.e., from minutes to hours) in many vertebrates, it is often difficult to disentangle the neuroendocrine mechanisms regulating these distinct behaviors. In this context, organisms that exhibit seasonal transitions between temporally distinct life-history stages can be particularly informative for understanding the factors that regulate the underlying physiology and behavior in all vertebrates.
In the present study, we asked if AVT and NPY correlate with the seasonal transition from reproduction to migration and foraging behavior in a well-studied population of red-sided garter snakes (T. sirtalis parietalis) in Manitoba, Canada. Red-sided garter snakes have a life-history cycle that includes eight months of winter dormancy, after which snakes emerge from underground dens in late April to mid-May. Upon emergence, red-sided garter snakes engage in an intense, 4-week mating season at the den, where individual females can be courted by up to 100 males. The mating system of red-sided garter snakes is defined by scramble competition, as garter snakes do not exhibit territorial aggression at any time of year nor do they engage in agonistic encounters to gain access to mates. Males and females do not eat during winter dormancy or the mating season, but rather migrate up to 17 km to summer feeding grounds at the end of the breeding season (Gregory and Stewart 1975). During the spring, female snakes typically migrate to summer feeding grounds soon after emerging from the den, whereas males remain near the den for several weeks to court newly-emerging females (Shine et al. 2001). At the summer feeding grounds, females ovulate and, if they become gravid, give birth to live young; male snakes produce sperm to be used the following spring. Toward the end of the summer activity period, appetitive feeding behavior presumably wanes and snakes migrate back to the den site in preparation for winter dormancy.
The factors that induce spring migration from the den site are not yet known, but our previous studies suggest that it is not simply due to an increase in feeding behavior (D. I. Lutterschmidt, personal observation). Red-sided garter snakes are an ideal system for this study because they exhibit distinct seasonal transitions between the life-history stages of hibernation, reproduction, migration, and foraging. Furthermore, we can exploit naturally-occurring sex differences in the timing of these life-history transitions to better understand if and how AVT and NPY mediate seasonal changes in behavior. We specifically asked if AVT and NPY immunoreactivity in the brain varies with season, sex, or life-history stage (i.e., with migratory and/or reproductive status). We also investigated if changes in AVT and NPY are associated with differences in the timing of dispersal between males and females.
Methods
These experiments were performed in the field with free-ranging red-sided garter snakes in the Interlake region of Manitoba, Canada. Experimental protocols were approved by the Portland State University Animal Care and Use Committee and were performed under the authority of Wildlife Scientific Permit WB12691 issued by the Manitoba Department of Conservation.
Experimental design
Experiment 1. Season and sex variation in AVT and NPY: relation to migratory status
We first asked if there are seasonal and/or sex differences in AVT and NPY. To determine whether these differences are related to changes in migratory behavior, we compared non-migratory snakes collected from the den site to migratory snakes in each season. Similar to Cease et al. (2007), we used a road located approximately 1 km from the den along the migration route to aid in intercepting migrating snakes. All spring animals were collected from May 14–18, 2012; fall animals were collected on September 14, 2012.
During the spring, female snakes collected from the road typically have visible copulatory plug residue, indicating that they mated prior to migrating from the breeding grounds. We therefore compared migrating female snakes (n = 10) to mated females collected from the den (n = 11). Females were collected from actively mating pairs and held in outdoor arenas until copulation was completed; successful mating was confirmed by the presence of a mating plug in the cloaca.
Similarly, we compared courting males collected from the den (n = 10) to migrating males that were still exhibiting courtship behavior (n = 10). During this experiment, we noted that migratory males collected at the road exhibit two different behavioral phenotypes: some males continued to actively court females while others did not. Differences in courtship behavior among migratory males have been described anecdotally (see Lutterschmidt and Maine 2014), but have not yet been investigated experimentally. We used a well-established ethogram of male courtship behavior [Lutterschmidt et al. 2004; modified from Crews (1984) and Moore et al. (2000)] to categorize the reproductive status of each male as courting or non-courting. Of the 22 migrating males collected from the road during the spring, 10 male snakes exhibited courtship scores ≥2, behaviors that are only expressed in the context of reproduction (Crews 1984). These males were classified as “courting” and included in Experiment 1, while the remaining 12 snakes were classified as “non-courting” and reserved for Experiment 2. These behavioral phenotypes allowed us to examine changes in brain neuropeptides related to migratory status without introducing the confounding variable of differences in female reproductive status (mated vs. unmated) or male reproductive behavior (courting vs. non-courting) during the spring.
During the fall, we collected 10 females and 10 males that had migrated to the same road from the summer feeding grounds. We collected an additional 10 females and 10 males from the den site during the fall pre-hibernation period, after snakes returned to the den in preparation for winter dormancy. None of the fall-collected animals exhibited courtship behavior or signs of recent mating activity (i.e., copulatory plug residue).
Experiment 2. Changes in AVT and NPY during the spring: relation to reproductive status
In this study, we asked if variation in AVT and NPY is associated with the seasonal life-history transition from reproductive to non-reproductive status. To address this question, we needed to distinguish the changes in neuropeptides related to migration from those related to changes in reproductive behavior. We therefore focused on the differences between reproductive and post-reproductive snakes while keeping migratory status constant. We compared AVT and NPY between the 10 courting males and 12 non-courting males collected from the road during the spring. To determine changes related to reproductive status in females, we collected an additional 10 females from the den immediately upon spring emergence and prior to mating. We then compared AVT and NPY between these unmated females and the 11 mated females collected from the den during Experiment 1. We confirmed unmated status by verifying the absence of a mating plug in the cloaca.
Tissue collection and animal housing
Immediately upon capture, blood samples (200 µL) were collected within 3 min using tuberculin syringes and heparinized needles. Body mass and snout–vent length (SVL) were recorded for all snakes before they were scale clipped on the ventrum with a unique number. Snakes then received two pulse injections of 100 mg kg−1 body mass 5-bromo-2′-deoxyuridine (BrdU) as in Almli and Wilczynski (2007) and Maine et al. (2014); injections were administered sequentially into two different regions of the peritoneal cavity. BrdU is a thymidine analog that is incorporated into the DNA of mitotic cells. We treated snakes with BrdU to examine whether seasonal changes in migratory and/or reproductive behavior are associated with variation in cell proliferation and cell migration within the brain. These data are the subject of additional analyses not presented here. Our previous studies indicate that injection with BrdU does not alter reproductive behavior or brain neuropeptides in male red-sided garter snakes (D. I. Lutterschmidt, unpublished data; Maine et al. 2014). Thus, it is unlikely that treatment of snakes with BrdU affected the results presented here.
Snakes were housed in semi-natural outdoor arenas (1 × 1 × 1 m) containing a hide box and water bowl. During these experiments, a large, seasonal pool of water located near the den site provided the population with ad libitum access to water. To mimic these conditions, we provided snakes held in outdoor arenas with regular access to water. Snakes were not offered food because they do not eat during either the spring mating season or fall pre-hibernation period. These housing conditions do not induce stress responses in red-sided garter snakes (Moore and Mason 2001; Cease et al. 2007; Lutterschmidt and Maine 2014).
Four days after their initial capture, a second blood sample was collected before snakes were euthanized with a lethal overdose of 1% sodium Brevital. Male courtship behavior was assessed prior to final tissue collection. We chose this sampling regime because it allowed us to more accurately assess the behavioral phenotypes of migrating males without the influence of capture and handling immediately preceding courtship trials (Cease et al. 2007; Lutterschmidt and Maine 2014), it maximized our chances of observing changes in AVT and NPY related to the post-mating estradiol surge in females (Whittier et al. 1987), and it optimized the labeling of newly proliferated cells by BrdU treatment (Maine et al. 2014). Thus, this experimental design allowed us to minimize animal numbers while simultaneously assessing changes in AVT, NPY, and neurogenesis associated with seasonal changes in migratory and reproductive behaviors.
Brains were immersion-fixed in 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.2) for 16–18 h at 4°C. Tissues were then transferred to 0.1 M phosphate buffer and stored at 4°C until sectioning. Brains were cyroprotected in 30% sucrose in 0.1 M phosphate buffer and cut on a cryostat (Leica 3050 S) into four alternate series of 25-µm coronal sections. Tissues were thaw-mounted onto subbed slides (Fisherbrand Superfrost Plus) and stored at −20°C until immunohistochemical processing.
Radioimmunoassay
Blood samples were stored on ice until centrifuged and the plasma separated and frozen at −10°C. Plasma was transported to Portland State University on dry ice and stored at −80°C until analysis. Plasma samples were assayed for corticosterone, androgens (testosterone and dihydrotestosterone), and estradiol concentrations via direct radioimmunoassay following procedures described and validated for red-sided garter snakes by Lutterschmidt and Mason (2009, 2005). Because our testosterone antiserum cross-reacts significantly with 5-α-dihydrotestosterone (63% cross-reactivity; Fitzgerald Industries International, Acton, MA, USA), our direct assay measures both plasma testosterone and 5-α-dihydrotestosterone. Thus, we report here data for total androgen concentrations.
All hormone concentrations were corrected for individual recovery variation; mean extraction efficiency was 90.8%. Mean intra- and inter-assay coefficients of variation, respectively, were as follows: 11.4% and 17.9% for corticosterone, 12.7% and 14.8% for androgens, and 15.0% and 16.7% for estradiol. Average limits of detectability were 32.8 pg/mL for corticosterone, 19.2 pg/mL for androgens, and 8.4 pg/mL for estradiol. All plasma corticosterone concentrations were above the limits of detectability. When androgen and estradiol concentrations were below the limits of detection for an assay, we assigned those samples the limit of detectability to retain these individuals in our statistical analyses. This occurred in 1 of 98 androgen samples and 34 of 102 estradiol samples.
Immunohistochemistry and cell counting
We examined potential differences in AVT and NPY among snakes using immunohistochemistry methods identical to those described and validated by Lutterschmidt and Maine (2014). AVT immunoreactivity was examined using a rabbit anti-AVP antiserum (item 20069; ImmunoStar, Hudson, WI, USA) at a dilution of 1:5000. NPY immunoreactivity was examined using a rabbit anti-NPY antiserum (item 22940; ImmunoStar) at a dilution of 1:3000. Primary antibody signal was amplified by incubation with biotinylated horse (for AVT) or biotinylated goat (for NPY) anti-rabbit IgG secondary antibody diluted 1:500 (products BA-1100 and BA-1000, respectively, Vector Labs Inc.) followed by avidin conjugated to horseradish peroxidase (Elite ABC peroxidase kit, Vector Labs, Inc.). The NPY primary antibody signal was further amplified with biotin-labeled tyramide (item SAT700; Perkin Elmer, Piscataway, NJ, USA) diluted 1:50 according to manufacturer instructions followed by a second incubation with avidin conjugated to horseradish peroxidase. Primary antibody binding was visualized using 0.25 mg/mL diaminobenzidine in 0.2% hydrogen peroxide in 0.05 M Tris buffer (pH 7.2).
Locations of AVT- and NPY-immunoreactive (ir) cells were mapped onto standard anatomical brain sections adapted from Krohmer et al. (2010) and Martinez-Marcos et al. (2005, 2001). As in Lutterschmidt and Maine (2014), AVT-ir cells were quantified in the supraoptic nucleus (SON), bed nucleus of the stria terminalis (BNST), and posterior hypothalamus (PH; including the ventromedial hypothalamic nucleus, the lateral posterior hypothalamic nucleus, and the periventricular hypothalamic nucleus). Within the anterior hypothalamus (AH), we also observed AVT-ir cell bodies in the anterior POA and PVN. However, it was difficult to establish consistent boundaries among the cell populations located in the POA, PVN, and PH, and therefore AVT-ir cells in these regions were grouped together into one population for analysis. This AVT cell population is hereafter collectively referred to as the hypothalamus (HYP).
NPY-ir cell populations were quantified in the cortex (lateral, medial, and dorsal regions), nucleus sphericus (NS), AH (including cells within the BNST), and PH. Cell counts within the NS include cells located within the anterior and posterior dorsal ventricular ridge and the lateral and medial ventral amygdaloid nucleus. Lastly, the PH population includes NPY-ir cells located in the dorsomedial hypothalamic nucleus, lateral posterior hypothalamic nucleus, periventricular hypothalamic nucleus, and ventromedial hypothalamic nucleus. Example photomicrographs of brain sections showing AVT- and NPY-ir staining are shown in Fig. 1. Detailed atlases identifying the locations of AVT- and NPY-ir cell bodies and fibers throughout the brain of red-sided garter snakes can be found in Lutterschmidt and Maine (2014).
Example photomicrographs of brain sections showing (A) AVT-ir cells in the preoptic area (POA) and (B) NPY-ir cells in the cortex of red-sided garter snakes (Thamnophis sirtalis parietalis). Scale bars = 50 µm. See Lutterschmidt and Maine (2014) for a detailed atlas of AVT- and NPY-ir cell bodies and fibers throughout the brain of red-sided garter snakes.
Within each region of interest, ir cells were counted manually following the methods of Lutterschmidt and Maine (2014); animals were coded so that the observer was blind to the identity of individuals. The number of AVT- and NPY-ir cells in each region was totaled for each individual. We followed the calculations and criteria described by Laberge et al. (2008) and used in Maine et al. (2014) to account for missing and/or damaged sections. Within each region of interest, the number of AVT- and NPY-ir cells for an individual was calculated as the total number of labeled cells divided by the total number of usable brain sections multiplied by the average number of brain sections for a given region across all animals. We calculated the total number of labeled cells in males and females separately, because body size (and therefore brain size) is sexually dimorphic. These counting methods allowed us to retain more animals in the statistical analyses [in contrast to eliminating an animal if two or more consecutive sections were missing or damaged (e.g., Lutterschmidt and Wilczynski 2012)]. The mean number of missing or damaged sections across all animals and regions was 0.65 (±0.06 SE) and 1.86 (±0.09 SE) for AVT and NPY, respectively. Labeled cells were distributed relatively evenly across tissue sections within each region of interest. For example, the mean number of AVT-ir cells per section within the SON and across all animals was 6.61; the mean SE across all individuals was 1.23 cells per section. Similarly, the mean number of NPY-ir cells per section within the NS and across all animals was 4.26; the mean SE across all individuals was 0.71 cells per section. Thus, a small number of missing sections within a region of interest for any one individual is unlikely to affect the calculations of ir-cell number or the comparisons presented in this study. Any animal missing greater than 30% of its sections within a region of interest was excluded from statistical analyses. Final sample sizes for each region of interest are listed in the figures.
Finally, two NPY-ir cell populations are localized within large, bilateral brain regions: the cortex and NS. To determine if cell counts from one hemisphere could be used for analysis, we compared the total number of NPY-ir cells between the left and right hemispheres using a subset of males and females in this study (n = 9 per sex). As reported in the results, there was no difference in NPY-ir cell number between the left and right hemispheres within either the cortex or NS. We therefore used NPY-ir cell counts from one hemisphere for all subsequent analyses within these regions.
Normalization of sexual size dimorphism for direct sex comparisons
Body size in red-sided garter snakes is sexually dimorphic, with females being larger (mean female SVL ± SE in this study: 56.5 ± 0.87 cm) than males (mean SVL in this study: 47.2 ± 0.68 cm). This produces concomitant sex dimorphism in overall brain size and likely regional brain volumes. Thus, we first needed to correct the number of ir cells for differences in region volume before directly comparing males and females. For these comparisons, we asked if SVL could be used to accurately represent differences in regional brain volume. To address this question, we counterstained a series of tissue from this experiment using 0.25% toluidine blue in 70% ethanol. We then quickly rinsed the tissues in nanopure H2O (two dips) followed by 95% ethanol (four dips) and then covered the tissues with Permount and coverslips. We used a subset of 18 females and 20 males for this analysis, with differences in season and migratory status equally represented within each sex.
We chose two brain regions for this analysis that are both seasonally variable and sexually dimorphic: the medial cortex and NS (Crews et al. 1993; Kabelik et al. 2006; Krohmer et al. 2011; Holding et al. 2012). We distinguished the boundaries of these regions using the same brain atlas we used for counting NPY cells (Martínez-Marcos et al. 2001; Krohmer et al. 2010; Lutterschmidt and Maine 2014). We then outlined the region of interest and measured its area (µm2) using the calibrated measure feature of ImageJ software (NIH); we repeated this procedure for every tissue section containing the region of interest. We calculated the total region volume (µm3) for each animal by first summing the areas measured on each tissue section, multiplying by 4 to account for all tissue series, and then multiplying by 25 µm to account for section thickness.
Statistical analyses
Data were transformed where necessary, data were transformed using log, square root, or natural log to meet the assumptions of parametric analysis. When data transformation could not correct for non-normality and/or unequal variance, we used a nonparametric Mann–Whitney U-test or a Scheirer–Ray–Hare extension of the Kruskal–Wallis ANOVA. We used SigmaPlot 12.0 (Systat Software) for all statistical analyses.
For the bilateral NPY-ir cell populations in the cortex and NS, we verified that we could use cell counts from one brain hemisphere for analysis by comparing the number of NPY-ir cells between the left and right hemispheres using paired t-tests. We determined if SVL could be used as a proxy for regional brain volume by examining the relationship between the volume of the medial cortex or NS and SVL using linear regressions. We also compared results obtained from a two-way ANOVA (with season and migratory status as factors) on NPY-ir cell number corrected by SVL versus brain region volume.
In Experiment 1, we used a two-way ANOVA within each sex to confirm that body size did not vary with season or migratory status. We corrected our cell counts for sex differences in brain volume by dividing the total number of ir cells for each individual by its SVL. We then compared these SVL-corrected data using a general linear model three-way ANOVA with season, sex, and migratory status as between-subjects factors for each neuropeptide and brain region of interest. These initial analyses allowed us to directly examine possible sex differences in ir cell number within each region. Because the main effects of sex were statistically non-significant in all but one region, we separated the ir cell counts by sex and reanalyzed the data using a general linear model two-way ANOVA with season and migratory status as factors; data were not SVL-corrected in these analyses. Rather than collapsing sex as a factor and combining males and females for analysis, we opted to analyze potential differences in neuropeptides within each sex separately due to a priori knowledge that males and females exhibit differences in the onset and timing of seasonal migration (Shine et al. 2001).
Lastly, we used linear regression to determine if variation in ir cell number is related to seasonal changes in plasma steroid concentrations. We limited our comparisons to those brain regions with significant seasonal variation in ir cell number: AVT within the SON, HYP, and BNST of males; AVT within the HYP of females; NPY within the cortex and PH of males. We used hormone concentrations measured immediately before brain collection for these tests.
In Experiment 2, we used t-tests to confirm that body size did not differ between reproductive states. We then used a t-test within each region of interest to compare ir cell number between snakes in different reproductive conditions. For each neuropeptide, we compared mated and unmated females collected from the den and courting and non-courting males collected from the road. Each of these comparisons allowed us to assess potential changes in AVT and/or NPY in relation to changes in reproductive behavior while holding migratory status constant.
Results
There were no significant differences in the total number of NPY-labeled cells between the left and right hemispheres within the cortex (t = 0.61, P = 0.55) or NS (t = 0.99, P = 0.34). We therefore used cell counts from one hemisphere for all subsequent analyses of NPY-ir cell number within these regions. There was a significant positive relationship between both medial cortex volume and SVL (r2 = 0.30, P < 0.001) and NS volume and SVL (r2 = 0.42, P < 0.001). Further, results obtained from a two-way ANOVA using NPY-ir cell number divided by region volume did not differ from those obtained using SVL-corrected cell number (data not shown). Thus, we used SVL to correct for sex differences in region volume for all analyses directly comparing ir cell number in males and females.
Experiment 1. Season and sex variation in AVT and NPY: relation to migratory status
Body size did not vary significantly with season or migratory status in either sex; interaction terms were non-significant. For all regions except AVT in the HYP, we did not observe a significant sex difference in ir cell number nor any significant interactions between sex and the other factors (all P-values > 0.10; from general linear model three-way ANOVAs on SVL-corrected data). Thus, we report sex differences only for the HYP. Unless otherwise noted, all interaction terms were statistically non-significant.
AVT immunoreactivity
Supraoptic nucleus. Male snakes had significantly more AVT-ir cells in the SON during fall compared with spring (F1,34 = 5.98, P = 0.02); migratory status did not influence cell number (Fig. 2A). Female snakes did not show a significant main effect of season or migratory status on AVT-ir cell number (Fig. 2B), but a statistically significant interaction between these factors was observed in the SON (F1,31 = 5.33, P = 0.03).
Effects of season and migratory status on the total number of AVT-ir cells in the (A, B) supraoptic nucleus (SON), (C, D) hypothalamus (HYP), and (E, F) bed nucleus of the stria terminalis (BNST) of male and female red-sided garter snakes (Thamnophis sirtalis parietalis), respectively. The x-axis is arranged to reflect the migration of snakes over time, from the den to the feeding grounds during the spring and from the feeding grounds back to the den during the fall. Similar to Cease et al. (2007), we intercepted migratory snakes on a road located approximately 1 km from the den site. Each bar is the mean number of AVT-labeled cells ± 1 SEM. Capital letters indicate significant differences between seasons in male and female snakes (statistics from two-way ANOVAs). There was a significant interaction between season and migratory status in the number of AVT-ir cells in the SON of females (P = 0.03). Final sample sizes within each region of interest are shown above the x-axes.
Hypothalamus. A three-way ANOVA indicated that females had more AVT-ir cells in the HYP compared with males (F1,69 = 13.12, P < 0.001; data not shown), even after accounting for differences in body size and variation due to season and migratory status. Females had a mean ± SE of 6.9 ± 0.34 AVT-ir cells per cm SVL in the HYP; males, 5.5 ± 0.28 AVT-ir cells per cm SVL. There were no significant interactions in any combination of sex, season, and migratory status on SVL-corrected AVT-ir cell number (all P-values ≥ 0.10).
After separating these data by sex, two-way ANOVAs on the uncorrected data indicated that both male (Fig. 2C;F1,34 = 8.76, P < 0.01) and female (Fig. 2D;F1,34 = 5.87, P = 0.02) snakes had significantly more AVT-ir cells in the HYP during spring compared with fall. Migratory status did not significantly influence the total number of AVT-ir cells in either sex.
Bed nucleus of the stria terminalis. Male snakes had significantly more AVT-ir cells in the BNST during spring compared with fall (F1,32 = 5.87, P = 0.02); migratory status did not influence cell number (Fig. 2E). Female snakes did not show a statistically significant effect of season (F1,28 = 3.28, P = 0.08) or migratory status on AVT-ir cell number (Fig. 2F).
NPY immunoreactivity
Cortex. Male snakes had significantly more NPY-ir cells in the cortex during fall compared with spring (Fig. 3A;F1,35 = 12.74, P = 0.001), whereas female snakes did not show a significant seasonal difference (Fig. 3B). Migratory status did not significantly influence the total number of NPY-ir cells in the cortex of either sex.
Effects of season and migratory status on the total number of NPY-ir cells in the (A, B) cortex, (C, D) nucleus sphericus (NS), (E, F) anterior hypothalamus (AH), and (G, H) posterior hypothalamus (PH) of male and female red-sided garter snakes (Thamnophis sirtalis parietalis), respectively. The x-axis is arranged to reflect the migration of snakes over time, from the den to the feeding grounds during the spring and from the feeding grounds back to the den during the fall. Similar to Cease et al. (2007), we intercepted migratory snakes on a road located approximately 1 km from the den site. Each bar is the mean number of NPY-labeled cells ± 1 SEM. Capital letters indicate significant differences between seasons in male snakes (statistics from two-way ANOVAs). There was a significant interaction between season and migratory status in the number of NPY-ir cells in the PH of females (P = 0.03). Final sample sizes within each region of interest are shown above the x-axes.
Nucleus sphericus and anterior hypothalamus. Neither season nor migratory status significantly influenced NPY-ir cell number in the NS or AH of male or female snakes (Fig. 3C–F; results in Fig. 3D from a nonparametric two-way ANOVA).
Posterior hypothalamus. Male snakes had significantly more NPY-ir cells in the PH during fall compared with spring (H1,31 = 6.78, P < 0.01; from a nonparametric two-way ANOVA); migratory status did not influence cell number (Fig. 3G). Female snakes did not exhibit a significant main effect of season or migratory status on NPY-ir cell number (Fig. 3H), but there was a significant interaction between factors (F1,29 = 5.17, P = 0.03).
Relationship between neuropeptides and steroid hormones
There were no significant relationships between androgen concentrations and AVT-ir cell number within the SON, HYP, or BNST of male snakes. There were also no significant relationships between corticosterone and AVT-ir cell number within these regions. In female snakes, AVT-ir cell number in the HYP was positively related to estradiol concentrations (Fig. 4A;r2 = 0.11, P = 0.05) and negatively related to corticosterone concentrations (Fig. 4B;r2 = 0.14, P = 0.03). However, when all estradiol samples assigned the assay limit of detection are removed from the analysis, the relationship between AVT-ir cell number in the HYP and estradiol is no longer significant (r2 = 0.10, P = 0.17). Finally, neither androgen nor corticosterone concentrations were significantly related to NPY-ir cell number in the cortex or PH of males.
Relationship between the number of AVT-ir cells in the HYP and (A) plasma estradiol and (B) plasma corticosterone concentrations in female red-sided garter snakes (Thamnophis sirtalis parietalis). Undetectable estradiol samples that were assigned the assay sensitivity are shown by solid gray circles. Note that when these individuals are removed from the analysis, the relationship between AVT-ir cell number and estradiol is no longer statistically significant (P = 0.17).
Experiment 2. Changes in AVT and NPY during the spring: relation to reproductive status
We asked if changes in AVT- or NPY-ir cell number in the brain are related to the transition from courting to non-courting in males or from unmated to mated status in females. Within each sex, body size (SVL) did not differ significantly between groups. Non-courting males had significantly more AVT-ir cells compared with courting males in both the SON (Fig. 5A;t = −2.64, P = 0.02) and HYP (Fig. 5C;t = −2.22, P = 0.04); AVT-ir cell number did not differ between groups within the BNST (Fig. 5E). Non-courting males also had significantly more NPY-ir cells in the cortex compared with courting males (Fig. 6A;t = −2.07, P = 0.05); NPY-ir cell number did not differ between groups in the NS, AH, or PH. There were no significant differences in AVT- or NPY-ir cell number between unmated and mated females in any region (Figs. 5 and 6).
Effects of reproductive status on the total number of AVT-ir cells in the (A, B) SON, (C, D) HYP, and (E, F) BNST of male and female red-sided garter snakes (Thamnophis sirtalis parietalis), respectively. All males were collected during spring migration from the breeding grounds; behavioral trials were used to assess courtship behavior and categorize snakes as courting or non-courting. Unmated and mated female snakes were collected from the den. Within each sex, we therefore examined differences related specifically to reproductive status while controlling for differences in migratory behavior. Each bar is the mean number of AVT-labeled cells ± 1 SEM. Asterisks indicate significant differences in AVT-ir cell number between courting and non-courting males in the SON and the POA. Final sample sizes within each region of interest are shown above the x-axes.
Effects of reproductive status on the total number of NPY-ir cells in the (A, B) cortex, (C, D) NS, (E, F) AH, and (G, H) PH of male and female red-sided garter snakes (Thamnophis sirtalis parietalis), respectively. All males were collected during spring migration from the breeding grounds; behavioral trials were used to assess courtship behavior and categorize snakes as courting or non-courting. Unmated and mated female snakes were collected from the den. Thus, within each sex, we examined differences related specifically to reproductive status while controlling for differences in migratory behavior. Each bar is the mean number of NPY-labeled cells ± 1 SEM. The asterisk indicates a significant difference in NPY-ir cell number between courting and non-courting males in the cortex. Final sample sizes within each region of interest are shown above the x-axes.
Discussion
The current study demonstrates that the number of AVT- and NPY-ir cells within brain regions important for regulating reproduction, migration, and feeding varies with season, sex, and life-history stage. During the spring mating season, males had more AVT-ir cells in two brain regions important for modulating reproductive behavior: the HYP and BNST. In contrast, male snakes had more AVT-ir cells during the fall in the SON, the brain region that synthesizes AVT/P for systemic release from the posterior pituitary in mammals and other vertebrates. Males also had more NPY-ir cells during the fall in the cortex and PH, regions thought to be important in regulating spatial memory and feeding behavior, respectively. Similar to males, females had significantly more AVT-ir cells in the HYP during the spring. Further, we observed a higher number of AVT-ir cells in the HYP of females compared with males, even after correcting for potential differences in brain volume. In male snakes, the number of ir cells varied with reproductive status in the SON, HYP, and cortex, but not directly with migratory status in any region. Lastly, ir cell number did not vary significantly with either migratory or reproductive status in females. These data suggest that AVT and NPY are involved in the seasonal life-history transition from reproduction to foraging behavior in males, but probably not in changes in migratory behavior per se.
Seasonal variation
From late April to late May, northern populations of red-sided garter snakes emerge from underground hibernacula and engage in an intense mating period at the den, after which they migrate to summer feeding grounds. We compared snakes collected during the spring mating season and the fall to examine if variation in ir cell number is related to seasonal changes in behavior. We hypothesized that there would be more AVT-ir cells during the spring mating season in brain regions that are important for reproductive behavior. Indeed, male and female snakes had more AVT-ir cells during the spring in the HYP, and males had more AVT-ir cells within the BNST during the spring versus the fall. Both of these regions are known to play important roles in reproductive behavior in many vertebrates (Krohmer and Crews 1987a, 1987b; Lanuza and Halpern 1997; Kabelik et al. 2008; Xie et al. 2010; Ball and Balthazart 2011). For example, Toyoda et al. (2003) observed increased courtship behavior and release of female-attracting pheromone following treatment with exogenous AVT in male red-bellied newts (Cynops pyrrhogaster). Additionally, Kabelik et al. (2013) used cFos labeling to observe activation of AVT neurons within the POA and BNST in adult male brown anoles (Anolis sagrei) engaging in courtship behavior. The seasonal differences we observed in AVT-ir cell number in the HYP and BNST of males suggest that AVT is associated with seasonal changes in courtship behavior in red-sided garter snakes.
Following the summer activity period, male and female snakes return to the den site in preparation for winter dormancy. During this fall pre-hibernation period, we observed more AVT-ir cells in the SON of both male and female snakes, although there was a significant season-by-migratory status interaction in females. Numerous studies in mammals have demonstrated that neurons in the SON synthesize and transport AVP from the HYP to the posterior pituitary for systemic release (reviewed in Landgraf and Neumann 2004). An evolutionarily conserved function of the SON is supported in another snake species, Jararaca vipers (Bothrops jararaca) (Zambotti-Villela et al. 2007). As a peripheral neurohormone, AVT/P primarily regulates water balance (e.g., Ludwig et al. 1996), but it has also been shown to directly stimulate release of adrenocorticotropic hormone and increase glucocorticoid synthesis (e.g., Madison et al. 2008). For example, Dunham and Wilczynski (2014) reported a significant increase in plasma glucocorticoids in male green anoles (A. carolinensis) treated with exogenous AVT. In addition to projecting to the posterior pituitary, however, both the SON and PVN of mammals contain AVP-positive fibers that project to other hypothalamic and extra-hypothalamic sites (reviewed in Albers 2015). Thus, central release of AVT/P from the SON may additionally function in a neuromodulatory context. The higher AVT-ir cell number we observed in the SON during the fall in red-sided garter snakes could therefore be associated with modulating seasonally-appropriate behaviors, maintaining water balance during extended terrestrial activity, and/or establishing the higher baseline plasma glucocorticoid concentrations observed during the fall in this population (Moore et al. 2001; Lutterschmidt and Mason 2005).
Compared with the spring mating season, we observed significantly more NPY-ir cells within the PH of males and a significant interaction between season and migratory status on NPY-ir cell number in females during the fall (i.e., after summer foraging). The PH contains NPY-sensitive nuclei that are activated by NPY-secreting cells, including the lateral posterior hypothalamic nucleus, ventromedial hypothalamic nucleus, and periventricular hypothalamic nucleus. When these nuclei were treated with exogenous NPY in rodents, food intake increased (reviewed by Chee and Colmers 2008). Indeed, male red-sided garter snakes increase feeding behavior following intracerebroventricular treatment with exogenous NPY (Morris and Crews 1990). Together, these studies support the hypothesis that modulation of NPY-ir cell number in the PH is related to seasonal changes in foraging behavior.
During the fall pre-hibernation period, males also had a higher number of NPY-ir cells in the cortex compared with spring-collected males. The reptilian cortex is composed of lateral, dorsal, and medial regions. The medial and, to a lesser extent, dorsal cortex is considered a structural and functional homolog of the hippocampus of birds and mammals (Butler and Hodos 1996) and is important for spatial memory in snakes (Roth et al. 2006; Holding et al. 2012). Roth et al. (2006) reported larger medial cortex volumes in male cottonmouths (Agkistrodon piscivorus), which maintain and navigate through larger home ranges compared with females. The hippocampus (and its vertebrate homologs) is well known for its seasonal plasticity (e.g., Migaud et al. 2011). For example, Barnea and Nottebohm (1994) observed greater neuronal recruitment in the hippocampus of black-capped chickadees (Poecile atricapillus) during peak spatial use in the fall.
The higher NPY-ir cell number we observed in the cortex of male snakes during the fall could be related to the spatial memory required for migration back to the den site. However, if true, why didn’t females also exhibit a significant increase in NPY-ir cell number during fall migration? We speculate that NPY plays a neuromodulatory role in the cortex, and that this role is more pronounced in males (e.g., Danger et al. 1990; McShane et al. 1992; Wójcik-Gładysz and Polkowska 2006; Schneider et al. 2013). For example, hippocampal NPY influences the sensitivity of the hypothalamic-pituitary-adrenal (HPA) axis in rats (Heilig and Thorsell 2002; Thorsell et al. 2000). In red-sided garter snakes, the sensitivity of the HPA axis is both seasonally variable and sexually dimorphic (Moore et al. 2000, 2001; Lutterschmidt and Mason 2005; Dayger and Lutterschmidt 2016). During the spring mating season, male snakes are less sensitive to capture stress compared with the fall (Moore et al. 2001). Further, courting males are less sensitive to adrenocorticotropic hormone during the spring than are females (Dayger and Lutterschmidt 2016). Thus, it is possible that seasonal changes in NPY in the cortex of male snakes are related to neuromodulation of the HPA axis, with resulting changes in stress sensitivity. Such a hypothesis is supported by studies in laboratory mice and rats (Morales-Medina et al. 2010).
Sex differences
Numerous studies have demonstrated a higher number of AVT-ir cells in males compared with females in multiple brain regions (reviewed in Goodson and Bass 2001). In contrast, our results revealed significantly more AVT-ir cells in the HYP of female red-sided garter snakes compared with males, even after correcting for differences in body size. While generally uncommon, there is precedence for these results. Maruska et al. (2007) reported that female halfspotted gobies (Asterropteryx semipunctata) had larger and more AVT-ir cells in the POA compared with males during both peak-spawning (egg development and ovulation) and non-spawning stages. In red-sided garter snakes, the POA is important for reproduction and contains sex steroid concentrating neurons (Halpern et al. 1982; Krohmer and Crews 1987a, 1987b). Krohmer and Crews (1987b) observed reduced courtship behavior in male red-sided garter snakes with bilateral lesions to the AH-POA. In this study, we observed a positive relationship between AVT-ir cell number in the HYP and estradiol concentrations, although this relationship is strongly influenced by a large number of females with low or undetectable estradiol concentrations (Fig. 4A). In contrast, AVT-ir cell number in the HYP was significantly negatively related to corticosterone concentrations in female snakes. Corticosterone is known to inhibit female mating behavior in red-sided garter snakes (Dayger et al. 2013), and therefore AVT-ir cells in the HYP may play a role in modulating female receptivity. The observed relationship between hypothalamic AVT cell number and both estradiol and corticosterone would be clarified and potentially bolstered by incorporating additional seasonal time points (i.e., animals with different hormone concentrations) into similar future studies. For example, it is unknown if or how female reproductive condition is associated with AVT or NPY immunoreactivity in this species.
While males exhibited seasonal differences in AVT and NPY in almost all regions of interest, seasonal differences in ir cell number were generally lacking in female snakes. However, sample sizes of fall-collected, migrating females are relatively low in some regions of interest, and therefore these negative results must be interpreted with caution. In two regions of interest (AVT in the SON and NPY in the PH), the seasonal changes in females depended upon migratory status, as evidenced by significant interactions between factors. While male snakes remain near the den for several weeks during the spring to court newly emerging females, females typically disperse from the den within 1–2 days to begin migration to summer feeding grounds (Shine et al. 2001). Thus, our results may reflect the sexually dimorphic timing of dispersal from the den site and the earlier activation of foraging behavior in females.
Variation related to migratory and reproductive status
Overall, migratory status did not have a significant main effect on neuropeptide cell number. We did observe a significant interaction between season and migratory status in two regions of interest in females, suggesting further research is needed to understand if and how AVT in the SON and NPY in the PH are related to seasonal changes in migratory behavior. Because of the limitations associated with interpreting differences (or a lack thereof) in ir cell number, future studies using techniques that would allow for quantification of peptide or mRNA levels (e.g., Western blot or qPCR) would be particularly helpful in understanding whether migratory status is associated with seasonal changes in the AVT and NPY systems.
In male snakes, we observed significant variation in immunoreactivity related to changes in reproductive status. The majority of these differences mirrored those related to the observed seasonal variation. For example, both fall-collected and non-courting male snakes had more AVT-ir cells in the SON compared with spring-collected and courting male snakes, respectively. Our results suggest that the seasonal change in AVT within the SON (Fig. 2A) is concomitant with the seasonal transition from reproductive to non-reproductive status in males (Fig. 5A). Similar to our findings, Canada geese (Branta canadensis) have more intense AVT staining in the anterior median eminence and neurohypophysis prior to spring migration compared with post-migration (George and John 1987). Our results in red-sided garter snakes suggest that seasonal changes in AVT in the SON may be more closely related to seasonal transitions in reproductive life-history stages than migration per se. Interestingly, the seasonal patterns of AVT immunoreactivity in the HYP and BNST did not appear to reflect changes in reproductive status. Male snakes had a higher number of AVT-ir cells in the HYP during the spring compared with fall, but courting snakes had fewer AVT-ir cells than non-courting males. Further, AVT-ir cell number in the BNST did not differ significantly between courting and non-courting male snakes, but did vary seasonally. We do not yet know if AVT influences courtship behavior in any snake species, and such studies will be critical for understanding if and how AVT in the HYP and BNST regulate male reproductive behavior. Our results support those of Lutterschmidt and Maine (2014), who suggested that unchanging brain AVT during the seasonal transition to foraging behavior may enable males to maximize reproductive opportunities during the initial stages of dispersal to the feeding grounds.
Finally, we observed more NPY-ir cells in the cortex of non-courting males compared with courting males, a change that mirrored the higher number of NPY-ir cells observed in the cortex during fall. These results are also similar to those of Lutterschmidt and Maine (2014), in which a higher NPY-ir cell number in the cortex was observed in feeding snakes. Together, our results suggest that the seasonal change in NPY in the male cortex is concomitant with the transition from reproductive to feeding behavior. Specifically, as male snakes migrate to summer feeding grounds from the den during the spring, reproductive behavior wanes and snakes transition to foraging behavior (Gregory and Stewart 1975; O’Donnell et al. 2004; Cease et al. 2007).
Conclusions
We examined seasonal and sex differences in the number of AVT and NPY cells in relation to changes in migratory and reproductive status. During the spring mating season, we observed a higher number of AVT-ir cells in the HYP of males and females and the BNST of males, supporting a role for AVT in regulating reproductive behavior in red-sided garter snakes. Future studies are needed to determine if AVT modulates male courtship behavior, which could explain the sexually dimorphic seasonal changes in AVT within the BNST. In contrast, both males and females exhibited overall higher numbers of AVT-ir cells in the SON during the fall. Such results could be related to the demands of maintaining water balance during terrestrial activity, modulation of plasma glucocorticoid concentrations during fall migration and pre-hibernation, and/or central neuromodulation. Following summer foraging, both males and females had an overall higher number of NPY-ir cells in the PH, supporting the hypothesis that NPY regulates seasonal changes in appetitive and consummatory feeding behavior. In the cortex, however, significant seasonal changes in NPY were observed only in males, and therefore NPY’s role within the cortex may be more related to well-documented seasonal modulation of the stress axis.
The majority of the seasonal variation we observed in AVT and NPY immunoreactivity was not related to changes in migratory behavior. Rather, some key differences were more related to the cessation of reproductive behavior in males, suggesting that at least some of the observed seasonal variation in ir cell number is specifically related to the transition from reproduction to foraging during the spring. Collectively, our results suggest that region-specific changes in brain AVT and NPY play a more prominent role in modulating seasonal transitions in reproductive and foraging behaviors than initiating migration per se, although additional studies are needed to understand the observed region-specific interactions between season and migratory status in females. If correct, then the factors that specifically induce migration in this population, and interact with AVT and NPY to alter behavior, are currently unknown. Further studies are needed to examine how changes in brain AVT and NPY (and other neuroendocrine factors) relate to the timing of seasonal changes in reproductive and feeding behavior, especially during the migration to and from the feeding grounds. It will also be critical to incorporate additional methods of assessing changes in the activity of the AVT and NPY systems. For example, ir-cell number is not necessarily indicative of the quantity of peptide produced or released from such cells, and it is likely that variation in receptor availability, AVT and NPY mRNA expression, peptide synthesis, and/or hormone release versus storage all contribute to seasonal changes in behavior. Together with the current results, such a systematic approach would help reveal the basic neuroendocrine mechanisms that mediate changes in an animal’s motivation to pursue one resource over another.
Acknowledgments
The authors thank Jill Schneider and Pierre Deviche for the invitation to present this research and for organizing the “Molecular and Neuroendocrine Approaches to the Study of Evolutionary Tradeoffs” Symposium. They also thank SICB for their support and the Manitoba Department of Conservation and Dave Roberts for logistical support in Manitoba, Canada. Catherine Dayger, Chris Friesen, and Robert Mason provided assistance in the field.
Funding
This work was partially supported by a Forbes-Lea Student Research Grant from the PSU Biology Department [to A.R.L.] and a PSU Faculty Enhancement Grant [to D.I.L.]. The National Science Foundation [grant IOS-1355203 to D.I.L.] and Lehigh University provided partial financial support for participation at the meeting.
References
Author notes
From the symposium “Molecular and Neuroendocrine Approaches to the Study of Evolutionary Tradeoffs: Food, Sex, Stress, and Longevity” presented at the annual meeting of the Society for Integrative and Comparative Biology, January 4–8, 2017 at New Orleans, Louisiana.
