Heat shock protein gene expression varies among populations but does not strongly track recent environmental conditions: implications for biomarker development

Mary J. Woodruff^1,2,3*Sarah E. Wolf^2,3,4Ethan D. Clotfelter⁵Elizabeth P. Derryberry⁶Mark T. Stanback⁷Kimberly A. Rosvall^2,3*

• ¹Center for Epidemiology and Animal Health USDA-APHIS, Oak Ridge Institute for Science and Education (ORISE), Fort Collins, CO, United States
• ²Department of Biology, Indiana University, Bloomington, IN, United States
• ³Center for the Integrative Study of Animal Behavior, Indiana University, Bloomington, IN, United States
• ⁴Institute of Ecology and Evolution, School of Biological Sciences, University of Edinburgh, Edinburgh, UK, Edinburgh, Scotland, United Kingdom
• ⁵Department of Biology, Amherst College, Amherst, Massachusetts, United States
• ⁶Department of Ecology and Evolutionary Biology, University of Tennessee, Knoxville, Knoxville, Tennessee, United States
• ⁷Department of Biology, Davidson College, Davidson, NC, United States

Global temperatures are rising, and scientists are mobilizing to uncover which birds are most affected by the problem of heat. Heat shock proteins (HSPs), for example, can shed light on this issue because they prevent damage and promote recovery from heat. However, few studies have investigated the relationship between HSPs and heat outside of experimental contexts. Here, we ask whether natural variation in HSP gene expression can serve as a biomarker of recent ambient conditions in wild nestling tree swallows (Tachycineta bicolor). We focused on HSP90AA1 because this HSP increases mRNA abundance in avian blood, after acute heat. Using blood samples collected across ten degrees of latitude, we tested for population differences in constitutive HSP90AA1 gene expression in 12-day-old nestlings. To quantify the specific time period over which ambient conditions best predicted variation in HSP gene expression, we used a climate window analysis, evaluating the predictive value of maximum temperatures and maximum heat index in the hours and days from hatching until sampling. We observed a significant difference in constitutive HSP gene expression between populations, with South Carolina nestlings showing nearly double the HSP90AA1 mRNA abundance compared to those in Massachusetts. There was no relationship between HSP90AA1 and heat index at any time (hours or days), meaning that baseline HSP gene expression is not a reliable biomarker for the combined effects of heat and humidity, at least not when applying existing metrics that were developed for poultry. We found some evidence linking HSP90AA1 gene expression with maximum temperatures three to four days before sampling; however, a permutation test could not rule out the possibility of a false positive. Thus, HSP90AA1 mRNA abundance is not necessarily an effective biomarker of recent heat, and it may instead reflect other inherent population differences. As heat waves intensify, this conclusion could change, and other species could be more reactive to heat. We urge the avian biology community to continue biomarker testing for estimating heat impacts on wild birds, as we seek to better understand and predict avian resilience to environmental challenges.