Temperature is a key environmental factor influencing reproduction and geographic distribution of insects and other ectotherms (e.g., Karl et al., 2009; Kellermann et al., 2012; Colinet et al., 2015, Ørsted et al., 2018, MacLean et al., 2019, Walsh et al., 2019, Terblanche and Hoffmann, 2020, van Heerwaarden and Sgrò, 2021). For instance, reproductive success in populations decreases strongly under high-temperature stress (Miwa et al., 2018, Parratt et al., 2021). Mating success is one of the most inclusive components of reproductive fitness (Brooks and Endler, 2001), and the ability to mate at elevated temperatures is likely to be one of the direct selection targets for thermal adaptation in ectotherms (e.g., Dolgin et al., 2006).

A wide range of genetic variability in courtship behaviors and mating success has been documented in Drosophila (e.g., Greenspan and Ferveur, 2000, Mackay et al., 2005, Gaertner et al., 2015). Selection for mating success under heat stress can also affect heat knockdown resistance as a correlated selection response in flies (Stazione et al., 2020). Under heat stress, heat-sensitive flies show lower performance than heat-resistant flies in many physiological functions including cardiac performance (e.g., Rodriguez et al., 2021), which could affect the ability to mate at elevated temperature. Previous work revealed a positive genetic correlation between heat-stress resistance and mating success at high temperature in Drosophila (Dolgin et al., 2006, Sambucetti and Norry, 2015, Stazione et al., 2020). In addition, male Drosophila melanogaster have higher mating success when adapted to the thermal environment (Dolgin et al., 2006).

The evolution of complex traits like courtship behaviors and mating success is thought to reflect evolution at many loci (e.g., Gaertner et al., 2015). Quantitative trait locus (QTL) analysis is a rather conservative approach to estimate the number and locations (usually as marker intervals) of genome regions (QTLs) that influence phenotypic variation of a trait (Falconer and Mackay, 1996, Lynch and Walsh, 1998). In D. melanogaster, copulatory success at elevated temperature was found to be associated to a thermotolerance QTL in the middle of chromosome 2 in a field-caged experiment (Loeschcke et al., 2011). The hypothesis that multiple rather than only one large-effect QTL are significant for mating success at elevated temperature remains to be tested in D. meanogaster. This is an interesting hypothesis because the number of QTLs is an important feature of the genetic architecture of quantitative traits (Falconer and Mackay, 1996, Lynch and Walsh, 1998, Mackay et al., 2005). Another hypothesis to be tested in thermal-adaptation studies is whether or not the QTLs for mating success at elevated temperature mostly differ from QTLs for mating success at benign temperature (Stazione et al., 2020), as QTLs influencing the reproduction at elevated temperature only, with no effects on mating success at benign temperature, should be relevant for thermal adaptation itself.

Here we tested three mating traits in D. melanogaster at elevated temperature with and without a heat-hardening pre-treatment, as well as at a benign temperature, by using a set of recombinant inbred lines (RIL) that were constructed from populations artificially selected for high and low knockdown resistance to high temperature (KRHT). The traits tested were copulation latency, copulation duration, and mating frequency as an index of mating success. Several aims were addressed. First, correlations between the traits were tested across RIL as an index of genetic correlations between the traits in each RIL set. Across-RIL genetic correlations between traits are expected to be significant if the traits share one or more QTLs. Second, QTLs were mapped on the three major chromosomes for each trait at both elevated and benign (control) temperatures to conservatively estimate the number and genomic regions of QTLs. Mating behavior was expected to be a polygenic rather than a monogenic trait in this study, as mating behavior is a complex trait influenced by many genes (Mackay et al., 2005). Third, QTLs for mating success at high temperature were compared to QTLs for mating success at benign temperature, in addition to comparisons with previously identified QTLs for thermotolerance in adult flies. In these wide-genome comparisons, for instance, a lack of overlapping between QTL regions across temperatures will indicate a different genetic basis for mating success at elevated versus benign temperature.