**1. Introduction**

The red deer (*Cervus elaphus* L., 1758) species, as the majority of the cervid taxa, shows a very deep relationship between its peculiar reproductive physiology and the their living environment [1]. It is widely diffused in almost all of continental Europe; in Italy it can be identified in a large Alpine area that extends from Cuneo to Udine, while in the Apennines the red deer occupies four distinct areas: the first corresponds to most of the mountain territory of the provinces of Pistoia, Prato, Florence, and Bologna, the second to the Tuscan-Romagna Apennines, the third is represented by the Abruzzo National Park together with the neighboring territories and the fourth by the mountain massif of the Maiella. Lately, some specimens have also been found in southern regions, such as Sardinia [2]. The original habitat of the red deer is constituted by wooded areas with clearings or areas of little thick bush, generally in a flat environment or at low altitudes. However, over the years it has successfully adapted to live in di fferent areas, from the heath to the forest of conifers, due to the increasing human competitive pressure [3].

In order to guarantee a good survival rate and growth of the o ffspring, hinds need to give birth at an appropriate time [4]. In the Mediterranean area, calves are usually born in spring and lactation goes on during summer [1], in order to exploit the most favorable time of the year, implying an extremely high synchronization between male and female reproductive activity during the rutting period. Red deer hinds are seasonally polyestrous, with a mean length of the estrous cycle of 18 ± 7 days and a gestation of approximately 231 days [5]. The reproductive cycle begins in October and if pregnancy does not occur, ovulation continues until March showing up to eight cycles [5]. The red deer, as well as fallow (*Dama dama*) and sika (*Cervus nippon*) deer, is a photoperiodic species: starting from the summer solstice, the decrease in daylight hours causes an increase in the duration of the nocturnal peak in melatonin secretion [6]. This event is responsible for the activation of the reproductive axis, leading to an increased secretion of LH releasing hormone, known as GnRH, and subsequent release of LH and FSH by the pituitary gland [7].

The afore-mentioned physiological events makes melatonin a major coordinator of the reproductive system and the annual rhythm [8]. The same endogenous circannual rhythm has been confirmed in red deer males, called stags, that show highly synchronized testicular cycles with a peak that overlaps the females' ovulation [5,9,10]. The almost complete suspension in spermatogenesis, during the non-rutting season (February–April), is further proof of such synchronization process [11]. Progesterone (P4) secreted by corpora lutea is necessary for the maintenance of pregnancy [12,13], while the contribution of placental progesterone is still to be assessed. P4 blood concentrations remain high (-4 ng/mL) for the duration of gestation, and start decreasing one to two days before parturition in order to reach baseline levels after delivery [4,14].

In mammals, many other hormones undergo changes in blood concentration during the period of pregnancy, but how these relate to reproductive processes and their e ffects are poorly understood in wildlife. Cortisol (CORT) is another important hormone, generally used as stress marker also in wild animals [15], but the evaluation of its plasma concentrations is considered as poorly reliable since the hypothalamic–pituitary–adrenal axis is instantaneously activated upon stressful stimuli, such as restraint and blood sampling [16]. This is partially why, in the last years, alternative matrices for hormones' and other analytes' quantification have been proposed, including hair and feces, capable of providing di fferent information regarding a longer timespan [17–19], also in red deer [20,21] and other frequently hunted ungulates [22]. For example, fecal CORT was proven to increase during late pregnancy with 10-year old females showing higher levels in comparison to five-year old hinds. It is interesting to note that fecal cortisol levels in wild animals increased during the calving period between May and June, potentially because of the stress related to delivery [23]. Despite the interest of the scientific community towards the physiology of reproduction of di fferent species of cervids, there is currently only very few information regarding wild red deer hinds and pregnancy. Studies have been performed on farmed hinds [24], but such data could potentially di ffer from wildlife animals in light of the strong influences of the environment on such species. This lack of data can be partially imputed to the di fficult nature of sampling. Since these animals can only be killed during certain periods of the year according to the local hunting regulation and usually undergo biometric examinations immediately after, collaborations with control centers for the collection of biometric records and hunters may help increasing the chances of sampling [19].

In light of the afore-mentioned reasons, the aim of this research was to analyze and describe hair and plasma levels of progesterone and cortisol from samples collected from pregnan<sup>t</sup> wild red deer hinds killed during the hunting season in the south-western Bologna Apennines (Italy) area. Data were analyzed to highlight potential correlations with the age of the animals and the date of death. Hopefully, this will help gaining more knowledge regarding the reproduction of this wild species and its relationship with the environment.

#### **2. Materials and Methods**

#### *2.1. Animals and Sampling*

Ten (n = 10) pregnan<sup>t</sup> red deer hinds were killed between 26 January and 1 March 2018 in the south-western Bologna Appennines (Italy) according to the regional hunting calendar (Resolution No. 473 of the Emilia Romagna Regional Executive, 10 April 2017). Upon death, hinds were immediately transferred to the pertinent control centers for the collection of biometric records where the samplings were performed. Ages were estimated from analyses of teeth eruption and wear patterns [25]. Pregnancy was confirmed by direct visualization of the uterus and identification of corpora lutea in the ovaries. Blood was collected from the jugular vein in sterile Lithium-heparin tubes, while hair was clipped from the dorsal-caudal region of the animals. All samples were refrigerated (5 ± 1 ◦C) and transferred in a cooler, within 24 h, to the physiology laboratories (ANFI-ASA) of the Department of Veterinary Medical Sciences of the University of Bologna (Ozzano dell'Emilia, Italy) as previously described [19].

#### *2.2. Samples Preparation and Extraction*

Upon arrival, blood samples (1 mL) were centrifuged (4000× *g*, 20 min) and stored at −20 ◦C until analysis for plasma hormone levels. Progesterone and cortisol were extracted by mixing 0.2 mL of plasma with 5 mL of petroleum ether or diethyl ether, respectively, for 30 min on a rotary mixer as described by Bono et al. [26]. The tubes were centrifuged at 2000× *g* for 15 min and the supernatants evaporated to dryness at 37 ◦C under an air-stream suction hood.

The hair samples were handled and analyzed as previously described [27]. Briefly, hair samples (250 mg) were washed with water and isopropanol in order to remove any organic residue from the surface. Once fully dried, samples were finely pulverized (180 mg) and incubated overnight with 6 mL of methanol for steroid extraction. After centrifugation, methanol was collected and evaporated to dryness under an air-stream suction hood.

#### *2.3. Progesterone and Cortisol Radioimmunoassay*

The dry extracts were stored at −20 ◦C until reconstitution in assay bu ffer for measurement of P4 (20 μL plasma equivalent, 6 mg hair equivalent) or cortisol (40 μL plasma equivalent, 60 mg hair equivalent) by radioimmunoassay; tritiated P4 (30 pg/0.1 mL; 96.6 Ci/mmol; PerkinElmer Inc., Boston, MA, USA) or tritiated cortisol (30 pg/0.1 mL; 94.6 Ci/mmol; PerkinElmer inc. Boston, MA, USA) were added, followed by rabbit anti-progesterone serum (0.1 mL, 1:10,000; antiserum produced in our laboratory) or rabbit anti-cortisol serum (0.1 mL, 1:20,000; produced in our laboratory), respectively. After incubation and separation of antibody-bound and -unbound steroid by charcoal-dextran solution (charcoal 0.25%, dextran 0.02% in phosphate bu ffer), tubes were centrifuged (15 min, 3000× *g*), the supernatant was decanted, and radioactivity immediately measured using a β-scintillation counter (Packard C1600, Perkin Elmer, USA).

The sensitivity of the P4 assay was 3.87 pg/tube, the intra- and inter-assay coe fficients of variation were 5.7% and 9.4%, respectively. The sensitivity of the cortisol assay was 5.5 pg/tube, the intra- and inter-assay coe fficients of variation were 4.9% and 8.7%, respectively.

Cross reactions of other steroids with antiserum raised against P4 were progesterone (100%), <sup>11</sup>α-hydroxyprogesterone (9.7%), 5α-pregnan-3-20-dione (4.4%), <sup>17</sup>α-hydroxyprogesterone (1.5%), 20α-hydroxyprogesterone (0.3%), cortisol (0.05%), testosterone, and 17β-estradiol (<0.001%).

Cross reactions of various steroids with antiserum raised against cortisol were cortisol (100%), cortisone (5.3%), <sup>11</sup>α-deoxycortisol (5.0%), corticosterone (9.5%), 20α-dihydrocortisone (0.4%), prednisolone (4.60%), progesterone, and testosterone (<0.001%).

In order to determine the parallelism between hormone standards and endogenous hormones in red deer hair, a pool sample containing high concentrations of cortisol and progesterone was serially diluted (1:1–1:8) with assay buffer. A regression analysis was used to determine parallelism between the two hormone levels in the same assay. A high degree of parallelism was confirmed by regression test (*R*<sup>2</sup> = 0.98, *p* < 0.01).

The assay results for both hormones were expressed as ng/mL for plasma and as pg/mg for hair.
