Introduction
Herbivores in mid to high latitudes, are dependent on seasonal food availability and use photoperiod (i.e. day length) to synchronize their reproduction with primary food production. Primary production in the food web is temperature dependent(Malyshev, Henry, & Kreyling, 2014; Peacock, 1976; Robson, 1967). Since temperature is a notoriously noisy environmental signal, herbivores use photoperiod as a proxy to adjust seasonal timing of reproduction(Baker, 1938). The annual relationship between photoperiod and ambient temperature is primarily dependent on latitude(Hut, Paolucci, Dor, Kyriacou, & Daan, 2013), but will also change with altitude and longitude, depending on local climatic conditions. As a consequence, for the Northern hemisphere, primary food production and biological spring will generally start later in the year at higher latitudes, coinciding with longer photoperiods than at low latitudes. Therefore, selection pressure on timing of reproduction has caused adaptive evolution of seasonal timing mechanisms in herbivores, such that reproduction in spring starts at longer photoperiods in more northern populations in the Northern hemisphere.
Latitudinal clines in annual timing have been described in many insect species and in some bird and plant species, but rarely in mammals(Hut et al., 2013). For photoperiodic induction of diapause in insects, the critical photoperiod (CPP) increases with latitude(W. E. Bradshaw, Holzapfel, & Mathias, 2006; William E. Bradshaw & Holzapfel, 2010; William E. Bradshaw & Lounibos, 1977; Hut et al., 2013; S. Paolucci, van de Zande, & Beukeboom, 2013; Saunders, 1973). Furthermore, bird species change their annual breeding frequency patterns at different latitudes, with the peak shifting to a later time point in the year at higher latitudes(Baker, 1938). These findings suggest that there is latitudinal adaptation of photoperiodic timing mechanisms driving reproduction. A genetic basis for variation in photoperiodic responsiveness has been demonstrated in the parasitoid Nasonia vitripennis where it is associated with clinal allelic variation of theperiod gene(Benetta, Beukeboom, & van de Zande, 2019; Silvia Paolucci, Salis, Vermeulen, Beukeboom, & van de Zande, 2016), in pitcher-plant mosquitoes Wyeomyia smithii (William E. Bradshaw & Holzapfel, 2001b; D. Mathias, Jacky, Bradshaw, & Holzapfel, 2005; Derrick Mathias, Jacky, Bradshaw, & Holzapfel, 2007), in the deer micePeromyscus leucopus (Heideman & Bronson, 1991), andPeromyscus maniculatus (Desjardines, Bronson, & Blank, 1986), and in Siberian hamsters Phodopus sungorus (Kliman & Lynch, 1992; Lynch, Lynch, & Kliman, 1989). Although, latitude of origin influences photoperiodic responses in deer mice(Dark, Johnston, Healy, & Zucker, 1983), the underlying genetics for adaptation of such photoperiodic mechanisms in mammals is not clear.
Laboratory experiments have revealed that annual rhythms in physiology and reproduction are driven by the photoperiodic neuroendocrine system(Nakane & Yoshimura, 2019). This mechanism is well conserved among vertebrates, including the common vole, Microtus arvalis (Król et al., 2012; van Rosmalen et al., 2021; van Rosmalen, van Dalum, Hazlerigg, & Hut, 2020), and comprises a seasonal timing mechanism which synchronizes to changes in photoperiod using the central circadian clock (i.e. the suprachiasmatic nucleus, SCN). Photoperiod is inversely related to the duration of the nocturnal melatonin release by the pineal gland(Bittman, Dempsey, & Karsch, 1983; Carter & Goldman, 1983; Hoffman & Reiter, 1965). Under short photoperiod, pineal melatonin is present in the morning hours (12h after lights OFF) and binds to melatonin receptors in the pars tuberalis of the pituitary, causing suppression of thyroid-stimulating hormone beta-subunit (TSHβ)(Dardente et al., 2010; Masumoto et al., 2010). Under long photoperiod, melatonin is absent in the morning (12h after lights OFF) which allows the transcriptional coactivator EYA3 and subsequently TSHβ to increase. TSHβ forms an active heterodimer with glycoprotein hormone alpha-subunit (αGSU)(Magner, 1990), which then binds as TSH to its receptor, TSHR, in the tanycytes around the third ventricle of the brain where it increases the production of iodothyronine deiodinase 2 (DIO2)(Guerra et al., 2010; Hanon et al., 2008; Nakao et al., 2008; Ono et al., 2008; Yoshimura et al., 2003). This leads to increased active thyroid hormone levels (T3)(Lechan & Fekete, 2005) acting on GnRH neurons regulating gonadotropin release to control reproductive behavior(Hanon et al., 2008; Nakao et al., 2008; Yoshimura et al., 2003).
TSH dependent sensitivity to photoperiod, which is defined as a shift of photoperiodic response curves, can be modulated by TSHR abundance and function. TSHR is therefore an essential protein in the mammalian seasonal reproduction pathway. In addition, Ho and colleagues showed that Tshr mutations can change the signaling efficiency of the receptor(Ho, Sande, Lefort, Vassart, & Costagliola, 2001). Selection on the Tshr gene has been shown in domestic chicken and suggests that the transmembrane domain is especially important in modulating photoperiodic responses(Karlsson et al., 2016; Rubin et al., 2010). Therefore, to assess seasonal adaptation of photoperiodic mechanisms, we focus on the TS­­H receptor (Tshr ).
The protein encoded by the Tshr gene belongs to the glycoprotein hormone receptor family(Smits et al., 2003). TSHR is a transmembrane domain (TMD) G protein-coupled receptor (GPCRs) with a large extracellular N-terminal part, containing leucine-rich repeats (LRRs). LRRs form a baseball glove-like structure, which is responsible for hormone (TSH) recognition and binding(Kleinau, Neumann, Grüters, Krude, & Biebermann, 2013). TSH binding causes a conformational change in the TSHR, which activates G-protein dependent signaling transduction. The extracellular LRR domain is connected to a large transmembrane helix, the hinge region.
The Tshr gene is very large owing to its exceptionally large introns. This offers many possibilities for cis-regulatory elements, that may modulate transcription(Wittkopp & Kalay, 2012). Mutations in intronic regions may disrupt transcription factor binding, which may lead to altered TSHR expression. Human TSHR mRNA splice variants encoding a TSHR without TMD have been reported(Graves, Tomer, & Davies, 1992), which may hint at alternative splice sites within intronic regions prior to the exon encoding for the TMD. Since the hinge region and the first TMD are important for ligand binding and for signaling transduction(Mizutori, Chen, McLachlan, & Rapoport, 2008) and mutations in the hinge region of the TSHR are known to change the signaling efficiency of the receptor(Ho et al., 2001), we consider this region as a potential target for natural selection and functional adaptive variation.
In this study we evaluate genetic adaptation to local climate conditions in the common vole (Microtus arvalis ) Tshr gene by comparing genetic variance over the large European geographical range of this herbivorous rodent. For this purpose, we focus on the end of intron 8 and the beginning of exon 9, encoding for the hinge region and the first part of the transmembrane domain of TSHR. Because the reproductive response at high latitudes in the Northern hemisphere requires longer photoperiods, and therefore higher TSH levels, allelic variation of theTshr gene may be associated with reduced TSHR signaling at high latitudes. Tissue samples were collected from 43 different localities over a large European geographical distribution (Fig. 1A, Table S1). The large variation in latitude (42° 21’ 36” N - 59° 17’ 60” N), longitude (5° 31’ 48” W - 38° 23’ 24” E) and altitude (4 – 2146 m above sea level), allowed us to assess whether location specific annual photoperiod-temperature ellipsoid patterns can explain the distribution of genetic variation in coding and non-coding parts of Tshr .