Phenotypic plasticity is a major mechanism whereby organisms adjust their traits within generations to complement environmental conditions.  In the context of range expansions, plasticity is thought to be especially important, as plastic changes in traits can lead to rapid adaptation.  For these reasons, there has been growing interest in the roles of molecular epigenetic processes in the context of range expansions.  One epigenetic process in particular, DNA methylation, enables organisms to adjust gene expression contingent on the environment, which probably explains why it has played a role in some invasions.  Nevertheless, we still know very little about how methylation is regulated in wildlife, especially the expression of the enzymes responsible for altering methyl marks on the genome.  The most important forms of these enzymes in vertebrates are DNA methyltransferase 1, which largely maintains existing methyl marks, DNA methyltransferase 3, which creates most de novo methyl marks, and TET2, which is a major demethylator of CpG motifs, genomic regions where most methyl marks occur.  In this study, we compared the expression of these genes in three different tissues (i.e., gut, liver, and spleen) of house sparrows (Passer domesticus) from 9 locations.  Some populations derived from the native range of the species (i.e., Israel, the Netherlands, Norway, Spain, and Vietnam), whereas others came from areas where birds were introduced <150 years ago (i.e., Australia, Canada, New Zealand, and Senegal).  We hypothesized that non-native birds and/or birds from sites with comparatively unpredictable climates would express the highest levels of all three genes.  We found, however, that DNMT expression differences, while extensive, were reversed from predictions: Sparrows from the native range and from areas with more stable temperatures expressed more of all three genes.  Surprisingly, too, expression of all three genes was strongly correlated among countries and within individuals.  Our results reveal no simple role for these enzymes in range expansions, but they do indicate appreciable among and within-population variation, which we hope motivates more detailed investigations of these enzymes in other wildlife.

Bird capture, husbandry, and tissue collection:

We captured adult house sparrows using mist nets from sunrise until ~1100 at each location (Table 1). Upon capture, we measured wing chord (to the nearest 1 mm), tarsus length (to the nearest 1 mm), and body mass (to the nearest 0.1 g). We also collected approximately 50 µl of blood from the brachial vein of each bird, which was stored in 300 µl of DNA/RNA Shield (Zymo R1100-50). Immediately after, each bird was injected subcutaneously with 100 µl of 1 mg ml⁻¹ LPS (from E. coli 055; Fisher L4005) in sterile saline over the breast muscle. Birds were housed individually in wire songbird cages (approx. 35.6 x 40.6 x 44.5 cm) with food and water provided ad libitum, while maintaining visual and vocal contact. Forty-eight hours post-injection, between 0700 and 1000, birds were euthanized via isoflurane overdose followed by rapid decapitation. We collected liver, spleen, and gut samples in 500 µl of DNA/RNA Shield, and all samples were stored at -80°C until further analysis. We chose these tissues in particular because the goal of the larger project for which we collected sparrows involves epigenetic regulation of immune gene expression; these tissues are among the most active lymphoid tissues in the body.  All animal procedures complied with local ethical guidelines, approved by the USF IACUC (IS00011653) and relevant authorities in the countries of capture. Export and import of animal tissues followed all relevant U.S. regulations, including USDA-APHIS permits.

RNA extraction and gene expression analyses:

We extracted RNA from liver, gut, and spleen tissue samples of each sparrow using a standard phenol: chloroform protocol (Sambrook and Russell, 2012). Reverse transcription was carried out using the iScript cDNA Synthesis kit (Bio-Rad 1708891) according to the manufacturer’s instructions. We then quantified the absolute copy numbers of DNMT1, DNMT3, and  TET2 using droplet digital PCR (ddPCR). Each ddPCR reaction contained 5 µl ddPCR Multiplex Supermix (12005909, Bio-Rad), 2.25 µl of forward and reverse primers (10 µM), 0.63 µl of probe FAM, 0.63 µl of probe HEX, and 0.63 µl of FAM + HEX probe mixture (for 50% FAM + HEX, 0.31 µl of each), and 1 µl of cDNA sample (3500 ng/µl). The reactions were run on a C1000 Touch™ Thermal Cycler with a 96–Deep Well Reaction Module (1851197, Bio-Rad). After amplification, droplets were separated and analyzed as positive (containing the target sequence) or negative (without the target sequence) using the QXDx Droplet Reader (12008020, Bio-Rad). Expression data were analyzed using QuantaSoft™ Analysis Pro software (version 1.05).