The dark-eyed junco subspecies live in sympatry and flock together at Mountain Lake Biological Station (MLBS), Virginia, from early fall to spring and start to diverge as soon as the photoperiod starts increasing (Figures 1a,b). In the wintering ground, the resident and migrant juncos are identified by their beak color. The migrants are pink-billed, and residents have blue-billed resident birds that breed at MLBS (Greives et al., 2016; Fudickar et al., 2016). The migrant juncos start to prepare for departure to the North as day length increases. Whereas resident juncos start showing gonadal recrudescence (Figure 1).

The distal most secondary feather from each individual was collected and stored in the coin envelopes and transported to the Indiana University, Bloomington. All feathers were cleaned using a chloroform: methanol in a 2:1 ratio to remove any external contaminants. A small fraction from the tip of the feather was cut, weighed to approximately 0.5 mg, placed in a 3 × 5 mm silver capsule, and shipped to a laboratory located in the US Geological Survey in Denver, CO, to analyze the hydrogen isotope (δ²H) values. Hydrogen isotope ratios were measured using established methods of mass spectrometry (Wunder et al., 2012; Fudickar et al., 2016; Singh et al., 2021; Wanamaker et al., 2020). The non-exchangeable δ²H values were reported in parts per mil notation (‰) with respect to VSMOW (Vienna Standard Mean Oceanic Water) using Caribou (−157‰) and Kudu (−35.3‰) standards. We used δ²H values to estimate the latitude and photoperiod where each bird spent the breeding, molting season and incorporated local δ²H signatures in the growing feathers (Singh et al., 2021). To estimate the latitude and photoperiod for each corresponding δ²H value, we obtained North America hydrogen isotope ratios for August month precipitation (δ²H_p) from the OPIC 3.2 database (Bowen, 2020). For each bird, we found all feather δ²H values in the North America compatible to δ²H precipitation values (for detailed methods see Singh et al., 2021). The δ²H values were used as a continuous variable against all the physiological, hormone measures, and expressed transcripts.

We measured circulating baseline testosterone levels in blood plasma by puncturing the wing vein and collecting blood into heparinized capillary tubes. The blood samples were stored at 4 ^°C and plasma was extracted by centrifuging blood samples at 3000 rpm for 15 min at room temperature. The plasma was separated using a Hamilton syringe and stored at −20 °C until assayed to measure testosterone. The plasma samples were thawed on ice, and 20 μL of plasma from each individual was aliquoted in glass tubes. Hormone was extracted using diethyl-ether mixed by agitating and incubation at room temperature for 20 min. After incubation, all tubes were snap frozen and the supernatant was immediately transferred to a fresh tube. This procedure was repeated three times to extract the hormone, and finally the tube was dried using high pressure liquid nitrogen. Assay buffer (250 μL) was added to each tube, and 100 μL of extracted hormone was used in duplicate wells to run the testosterone assay. Plasma testosterone was measured using high sensitivity testosterone ELISA kit (Enzo ADI-900-176) as per manufacturer’s protocol. All samples were placed on one plate, testosterone ELISA with intra-assay coefficient variability of 0.6% ± 0.06 (mean ± SE) and sensitivity of 2.6 pg/mL.

Every year from fall to early spring, dark-eyed junco subspecies are sympatric and flock together at Mountain Lake Biological Station (MLBS), Virginia (Figures 1a,b). The pink-billed migrants arrive in the beginning of fall and depart to the north as spring approaches. Whereas blue-billed resident birds live year-round and breed at MLBS. In early spring, the migrants start preparing for migration. By the time resident juncos start showing gonadal recrudescence, the migrants are departing for their breeding ground. (Figure 1). The residents breeding and growing feathers at lower latitude had heavier δ²H than migrants (Mann-Whitney U test, p = 0.0012, Figure 2a). Also, there were significantly higher blood testosterone levels in resident as compared to migratory birds (Mann-Whitney U test, p = 0.014, Figure 2b). We did not find any significant difference in the body weight or subcutaneous fat scores of migrant and resident birds.

A total of 11,454 expressed sequenced tags (ESTs) were found after mapping to the dark-eyed junco draft genome assembly homologous to protein sequences from zebra finch (Taeniopygia guttata) as summarized in Supplementary Table S1. Further, we found three DEGs (FDR <0.05; Figure 3; Supplementary Table S1). Among the three DEGs, only one, gene macrophage mannose receptor 1-like (MMR1), was downregulated in resident birds (FDR = 0.0135; Figure 3a). Unlike MMR1, uncharacterized protein C21orf58 homolog (C7H21orf58; FDR = 0.0135; Figure 3b), gonadotropin releasing hormone 1 (GnRH1; FDR = 0.0216; Figure 3c) were upregulated in resident birds.

Macrophage mannose receptor 1 (MMR1) expression level showed a significant negative correlation to both δ²H isotope values (r ² = 0.7724, p < 0.0001) and T levels (r ² = 0.4574, p = 0.011; Figures 4a,b). In contrast, GnRH1 showed a positive correlation to both δ²H (r ² = 0.4692, p = 0.0098) and T levels (r ² = 0.3693, p = 0.0276; Figures 4c,d), indicating latitudinal differences in the HPG axis activation in anticipation to spring.

Populations of dark eyed junco are distributed across different latitudes in North America and vary widely in their phenotypic (body size, feather color, eye color) and migratory behavior as well as their reproductive phenologies (Nolan et al., 2002). Those breeding at higher latitudes are more likely to be migratory, and in spring, these higher latitude birds increase body fattening in preparation for migration. Juncos breeding at lower latitudes are more likely to be resident, and these non -migratory birds start to grow their gonads earlier in the year in response to the increase in early spring daylengths (Rogers et al., 1993; Fudickar et al., 2016). After autumn migration, high and low latitude populations typically overwinter together in sympatry.

Feather stable hydrogen isotope ratios have been used as a proxy for breeding latitude in movement ecology studies (Rubenstein et al., 2002; Hobson, 2003; Maggini et al., 2016). Because juncos molt their feathers late in the breeding season before they migrate, analysis of feather hydrogen isotope (δ²H) ratios in sympatric dark-eyed juncos during the non-breeding season can provide an estimate of their breeding latitudes (Fudickar et al., 2016; Singh et al., 2021; Wanamaker et al., 2020). Birds with lighter hydrogen isotope ratios likely breed at higher latitudes and those with heavier hydrogen isotope ratios likely breed at lower latitudes (Fudickar et al., 2016; Singh et al., 2021).

Several studies on junco populations have used feather hydrogen isotope ratios to show the latitudinal cline in spring preparedness of reproductive processes on the overwintering ground (Fudickar et al., 2016; Kimmitt et al., 2019; Wanamaker et al., 2020). These studies comparing overwintering resident and migrant junco populations from two separate locations, Virginia and Colorado, indicate that lower latitude juncos begin to grow their gonads earlier, thus on shorter days, than high latitude juncos (Fudickar et al., 2016; Wanamaker et al., 2020). This difference is likely related to the difference in the threshold photoperiod required to stimulate the neuroendocrine system regulating reproduction (Singh et al., 2021).

Temperate birds preparing to reproduce sense the change in day length and begin to rise in circulating testosterone, but activation of the hypothalamic-pituitary-gonad axis precedes that. Populations breeding at higher latitudes need relatively longer critical photoperiod than lower latitude birds to activate their HPG axis, grow gonads, and increase the circulating testosterone levels (Singh et al., 2021). Several field and captive studies in junco populations show co-variation between hydrogen isotope ratios and elevation of circulating testosterone levels in response to external GnRH 1 challenge, suggesting a latitudinal cline in reproductive timing (Fudickar et al., 2016; Greives et al., 2016; Wanamaker et al., 2020; Singh et al., 2021). In our study, we found higher variability in circulating baseline testosterone and GnRH-1 transcript in resident juncos, suggesting a plausible effect of non-photic cues, such as food availability and exposure to predators, regulating the pace of reproductive maturity among resident birds (Jacobs and Wingfield, 2000; Chmura et al., 2019).

In spring temperate-zone birds induce their gonads to grow in response to increasing day length. As noted, birds breeding at higher latitudes typically delay gonadal recrudescence until the days are longer because they migrate prior to reproducing (Dawson and Goldsmith, 1983; Silverin et al., 1993; Dawson, 2013; Greives et al., 2016). This additional spring migratory phenology in migratory populations breeding at higher latitudes delays the activation of the HPG axis and uses all energy in preparing for the migratory process.

The preoptic area (POA), a key regulatory nucleus located in the hypothalamus, has neuronal populations that synthesize and release a neuropeptide, gonadotropin-releasing hormone (GnRH1). Transition from short to long day lengths is known to induce GnRH1 release from the hypothalamus that in turn stimulates the release of luteinizing hormone (LH) from gonadotroph cells in the anterior pituitary (Dawson et al., 1985; Dawson and Goldsmith, 1997; Millar, 2005). Increased levels of LH coming to the systemic blood circulation further stimulate gonadal recrudescence and synthesis of testosterone, a gonadal steroid (Wingfield and Farner, 1993; Norris, 2007). Based on the studies on seasonal breeding songbirds, GnRH1 is the most critical top hierarchical molecule in stimulation of HPG axis with its elevation eventually leading to gonadal recrudescence.

In many avian species, levels of hypothalamic GnRH1 have been shown to be elevated significantly in the breeding season as compared to the non-breeding season (Dawson, 2013; Bauer et al., 2018). Significant seasonal changes in the abundance of hypothalamic GnRH1 transcript/peptide have been found in several songbird species, such as house sparrow (Passer domesticus; Hahn and Ball, 1995; Stevenson and MacDougall-Shackleton, 2005), American tree sparrow (Spizella arborea; Reinert and Wilson, 1996), house finches (Carpodacus mexicanus; Cho et al., 1998); dark-eyed juncos (Junco hyemalis; Deviche et al., 2006; Greives et al., 2016; Fudickar et al., 2016) and rufous-winged sparrows (Aimophila carpolis; Small et al., 2008).

A prior study comparing captive resident and migrant juncos in a common garden demonstrated differential response of the HPG axis to the same day lengths despite exposure to similar photoperiod, food, and temperature in captive condition. At a given early spring day length, resident juncos showed higher GnRH mRNA expression levels than migrant juncos, providing evidence for difference in the HPG axis activation in individuals originating from different latitudes (Bauer et al., 2018).

In our study, the hypothalamic transcriptome data collected from early spring free-living overwintering junco populations showed higher GnRH mRNA expression in resident compared to migrant juncos. This indicates stimulation of GnRH neurons leading to earlier recrudescence in low-latitude residents as compared to high-latitude migratory juncos. The higher GnRH1 transcript levels in resident junco than migrants could be involved in earlier activation of the hypothalamic gonadal axis in residents than in migrant juncos. Thus, differential GnRH1 synthesis in junco populations experiencing similar environmental conditions suggests difference in the photic sensitivity of the neuroendocrine machinery regulating the timing of reproduction in junco populations breeding at different latitudes.

In a similar study of the neuroendocrine tissue proteome of resident and migrant junco populations collected in early spring showed key proteins involved in GnRH 1 transcription, translation, and post-translational modification, stability, and synthesis (Singh et al., 2020). In addition, a proteome study on the juncos reported on here showed differential expression of Vasoactive Intestinal Peptide (VIP), a key neuropeptide molecule that has been implicated in the photoperiodic reproductive response and as a synchronizer of key circadian pacemaker cells in both birds and mammals (Singh et al., 2020; Rastogi et al., 2014).

Research on female phenology as compared to males is lacking. A study comparing female junco populations in Virginia showed similar trends of higher expression of key reproductive genes in the ovary and ovary mass in resident juncos than high-latitude migrants (Kimmitt et al., 2019). Future work should focus on investigating the role of GnRH1 and other key genes discussed above in the seasonal divergence of reproductive timing, comparing sex and population-level differences in both free-living and captive conditions.

In addition to GnRH 1, we found two more genes differentially expressed in the hypothalamus of resident and migrant juncos. Macrophage mannose receptor 1-like (MMR1), which encodes a glycan-binding protein expressed in macrophages, was downregulated in resident juncos. The MMR1 role has been studied in several biological processes that include the regulation of circulating reproductive hormone levels, homeostasis, innate immune response, and neuroinflammation (Cummings, 2022). The protein encoded from Mannose receptor gene is involved in the clearance of circulating Luteinizing hormone (LH) from the bloodstream which could lead to a decrease in the circulating testosterone levels (Cummings, 2022). Though in our study, we found a negative correlation of MMR1 transcripts level with testosterone levels. But given the low sample size and high variability in resident testosterone levels, we would avoid deriving any mechanistic relationship between MMR1 transcript levels regulating circulating testosterone levels. Another gene, C21orf58 homolog (C7H21orf58), which is an uncharacterized protein, showed higher transcript levels in resident birds. Not much is known about the function of the C7H21orf58 gene and its role in regulating the seasonal phenologies of birds.