Thermal extremes alter population processes, which can result in part from temperature-induced movement at different spatial and temporal scales. Thermal thresholds for animal movement likely change based on underlying thermal physiology and life-history stage, a topic that requires greater study. The intertidal porcelain crab Petrolisthes cinctipes currently experiences temperatures that can reach near-lethal levels in the high-intertidal zone at low tide. However, the thermal thresholds that trigger migration to cooler microhabitats, and the extent to which crabs move in response to temperature, remain unknown. Moreover, the influence of reproductive status on these thresholds is rarely investigated. We integrated demographic, molecular, behavioral, and physiological measurements to determine if behavioral thermal limits varied due to reproductive state. Demographic data showed a trend for gravid, egg bearing, crabs to appear more often under rocks in the cooler intertidal zone where crab density is highest. In situ expression of 31 genes related to stress, metabolism, and growth in the field differed significantly based on intertidal elevation, with mid-intertidal crabs expressing the gene for the reproductive yolk protein vitellogenin (vg) earlier in the season. Furthermore, we experimentally demonstrate that VG protein levels in female hemolymph increases with density. We tested for temperatures that elicit movement and found that gravid females engage in heat avoidance behavior at lower temperatures (i.e., have a lower voluntary thermal maximum, VT_max) than non-gravid females. VT_max was positively correlated with the temperature of peak firing rate for distal afferent nerve fibers in the walking leg, a physiological relationship that could correspond to the mechanistic underpinning for temperature dependent movement. The vulnerability of marine organisms to global change is predicated by their ability to utilize and integrate physiological and behavioral strategies in response to temperature to maximize survival and reproduction. Interactions between fine-scale temperature variation and reproductive biology can have important consequences for the ecology of species, and is likely to influence how populations respond to ongoing climate change.

Total RNA was extracted from the bodies of the crabs. The walking legs and claws of the frozen crabs were removed, and the remaining bodies were ground to a powder in liquid nitrogen with a mortar and pestle. RNA was extracted from the powder using guanidine isothiocyanate extraction with Tri Reagent (Molecular Research Center, USA) according to the manufacturer's protocol. Tri Reagent was added to each 50 mg powder sample along with nuclease-free stainless-steel beads and shaken in a TissueLyser (Qiagen) at 30 Hz for 10 minutes. Phase separation was performed with BCP (Molecular Research Center, USA), and isopropanol and a high salt buffer were used to precipitate the RNA. The RNA was washed twice with 75% ethanol and dissolved in 50 μL of RNase free water. RNA quality was assessed with A260/280 ratios and 1% agarose gel electrophoresis with ethidium bromide staining. RNA was stored at -80°C.

The measurement of target and housekeeping gene expression was conducted by the University of California, San Francisco Center for Advanced Technology using the Nanostring platform. Nanostring generates read counts for each gene from each sample without cDNA synthesis or PCR amplification. 25 μL of RNA per sample at a concentration of 100 ng/μL was provided to generate expression data of 31 target genes and eight housekeeping genes. Target genes were chosen based on their association with important biological processes including stress physiology (e.g., heat-shock proteins, V-type proton ATPases, genes associated with ubiquitination), metabolism (e.g., cytochromes, acyl-coa synthetase), growth (e.g., cuticular proteins), and reproduction (e.g., vitellogenin, vitelline egg coat protein). Gene expression was standardized relative to the expression of internal positive controls (for target and housekeeping genes) and housekeeping genes (for target genes only) for each individual. For the positive controls, we generated a positive control correction factor for each individual by dividing the geometric mean of positive controls for the individual by the geometric mean of positive controls across all individuals. The counts of all genes for that individual were then multiplied by this factor. Target genes for an individual were standardized to housekeeping genes in a similar manner, whereby target gene expression was multiplied by a correction factor taken as the geometric mean of housekeeping gene expression for that individual divided by the geometric mean of housekeeping gene expression across all individuals.