We tested whether cortisol concentrations were associated with CWD status by quantifying cortisol in hair from hunter-harvested deer. We supplemented this analysis with a small sample size (n=9) of serum from captured deer where CWD status was known within 30 days of capture due to mortality. Cortisol in hair is representative of long-term cortisol concentrations. In contrast, serum cortisol represents stress during capture and is highly variable by individual. In hair from hunter-harvested deer, cortisol concentrations were significantly higher in CWD+ animals (P = 0.033). Cortisol concentrations in the hair of CWD+ deer were 1.3 pg/mg hair, (SE = 0.24) and for CWD non-detect (ND) were 0.79 pg/mg, (SE = 0.09). Despite small sample sizes, serum cortisol was still on average higher in CWD+ (mean = 5.9 mg/dl, SE = 1.6) compared to CWD ND deer (mean = 2.9 mg/dl, SE = 1.3), though this relationship was not significant (P = 0.183). Combining these independent tests, there was significant evidence that cortisol concentrations differed between CWD+ and CWD non-detect animals (Fisher’s exact method: P = 0.031).

Hair sample collection

We collected hair from 31 hunter-harvested deer at Ames REC during the regular hunting season. We collected additional samples from 1 collared deer mortality and 1 deer capture mortality (see below for capture description). We checked harvested deer at the Ames REC hunting club lodge where we collected samples and biometrics. We used Oster cordless hair clippers with #10 shears (Oster Professional Products, Sunbeam Products, Miami, FL, USA) to shave an approximate 10 x 10 cm section from the left rear hindquarter. Hair was collected on a paper plate and transferred to a manilla envelope that was stored at room temperature until analysis.

Serum sample collection from hunter-harvested and captured deer

We used cardiac puncture for hunter-harvested deer to collect blood when available. We also collected blood from deer captured via helicopter with net guns (Helicopter Wildlife Services, Austin, TX, USA) and on the ground using baited Clover traps (Industrial Mechanical Inc., Watkinsville, GA, USA), baited Wildlife CaptureÒ magnetic drop nets (Wildlife Capture Services LLC, Flagstaff, AZ, USA), and by free-range darting from a baited ground blind. A Pneu-Dart G2 X-Caliber pneumatic air rifle and barbed 2ml Pneu-Dart transmitter darts were used for darting (Pneu-Dart Inc., Williamsport, PA, USA). Helicopter captured deer were hobbled, blindfolded, and placed in a sling for transport to a central processing station. Upon arrival at the processing station, deer were chemically immobilized. Deer captured with ground methods were chemically immobilized immediately. We used BAM (butorphanol-azaperone-medetomidine 17.3 mg/ ml butorphanol tartrate, 9.1 mg/ml azaperone, 10.9 mg/ml medetomidine; Wedgewood, Swedesboro, NJ, USA), or NalMed-A (40 mg/ml nalbuphine, 10 mg/ml azaperone, 10 mg/ml medetomidine, Wedgewood), or MKT (medetomidine (0.07 mg/kg, 20 mg/ml, Wedgewood), ketamine (1.5 mg/kg, 100 mg/ml, Covetrus, Ocala, FL, USA) and telazol (1.1 mg/kg, 100 mg/ml, Covetrus; Muller et al. 2012) before collecting a blood sample. For BAM or NalMed-A, we gave adult females 2 mls and males 3 mls. Once immobilized, we collected biological samples including blood from the jugular vein. We fitted LiteTrack Iridium Series GPS collars (Lotek Wireless, Newmarket, ON, Canada) on deer at this time. Animal temperature, pulse, and respiration were monitored during workup and supplemental oxygen was provided to anesthetized deer. Upon workup completion, deer were given appropriate antagonists and released (For BAM or NalMed-A, atipamezole (2 ml for every ml of BAM or NalMed-A, 25 mg/ml, Wedgewood) and 0.5 ml naltrexone (50 mg/ml, Wedgewood). For MKT, we used 0.35 mg/kg atipamezole (5 mg/ml, Covetrus)). We centrifuged (1775 G-force) all blood samples for 5 minutes to collect serum. We stored sera at -20 °C until analysis. We monitored collared deer for mortality signals using Lotek Web Service. Mortality alert signals were sent once collars had not moved for 6 hours. We collected mortalities as soon as possible once an alert was received. A licensed veterinarian (D. Grove) performed necropsies on deer that died when possible, including RPLN collection for CWD testing and evaluation of fat stores in body cavity and bone marrow.

Only 9 live captured deer which died within 30 days after capture were used in the serum analysis because we required CWD status for the analysis close to the time of capture. These deer were tested for CWD at necropsy. We assumed within 30 days, the CWD status would be the same as the status during the time of capture. 

Cortisol assay

We measured cortisol in hair and serum using cortisol ELISA kits (DetectX Cortisol ELISA Kits, Arbor Assays, Ann Arbor, MI, USA) at a temporary lab set up in West Tennessee to prevent potential contamination of CWD prions to our home lab in East Tennessee. For hair samples, we followed Arbor Assays hair protocol where 250 mg shaved hair was placed into 50 ml conical polypropylene centrifuge tubes with screw caps. We added 5 ml of isopropanol (Optima for HPLC, Fisher Chemical, Fisher Scientific, www.fishersci.com), mixed with the hair, and incubated for 5 minutes. The liquid was decanted and hair was washed 2 more times. The hair was dried on paper plates at room temperature. We ground the dried hair samples to a fine powder using a Retsch MM400 grinder mill with 50 ml steel screw-top canister and 25 mm grinding ball (Verder Scientific, Inc., Newtown, PA, USA). We weighed 50 mg of powdered hair into a 15 ml polypropylene centrifuge tube and added 1 ml methanol (Methanol (HPLC), Fisher Chemical). We placed the tubes with methanol on a rocker (Corning LSE Platform Rocker, Fisher Scientific) set to rotate 12-16 hours at 100 rpm. We centrifuged the tubes for 5 minutes (1775 G) and pipetted 0.5 ml of the methanol extract into a 1.5 ml centrifuge tube. The methanol was evaported using a Speedvac SPD1030 (Fisher Scientific) for 90 min at vacuum pressure 1 (heat time 30 min, temperature 40^^ °C). We reconstituted the dried extracts using 120 ul Assay Buffer in the cortisol kit. We measured parallelism using a higher concentration hair sample with dilutions from 120, 140, 150, 170, 190 ul buffer). All samples were measured between 20-89 % binding in the linear part of the curve. We reran any samples with CV > 20 % if possible. We measured extraction efficiency by comparing 2 samples with and without an added 5 ml of cortisol standard (32,000 pg/ml) to the spiked sample. We ran a pooled serum sample with every plate to evaluate interassay variability.

We followed Arbor Assays protocol and tested parallelism in deer serum according to Grunwald et al. (2025). All samples were measured between 20-83 % binding in the linear part of the curve. Coefficient of variation was determined by (standard deviation/mean) *100 of replicates measured for each sample. We reran any samples with coefficient of variation (CV) > 20 %.