Hybrids can experience genetic incompatibilities that lead to cellular dysfunction and infertility. Previous analyses of mitochondrial function in the lizards Urosaurus graciosus and Urosaurus ornatus found that hybrids, which had introgressed mtDNA, experienced higher ATP production which increases reactive oxygen species (ROS) production, one of the main causes of damage to DNA. Hybrid populations of these lizards have persisted so we predicted they may have mechanisms for protecting their DNA from damage. To test this, we exposed liver cells of these lizards to hydrogen peroxide, a source of ROS, and compared DNA damage using single-cell gel electrophoresis. Hybrid cells showed significantly less DNA damage. We then examined potential mechanisms to explain this result by comparing melanin content of livers and expression levels of antioxidant genes. If functioning to deal with ROS, these should be expressed higher in hybrids. Melanin content of livers was highly variable among U. graciosus individuals and therefore not highest in hybrids as expected. Eighteen antioxidant genes showed significant differential expression but were not expressed consistently higher in hybrids. The overall gene expression profiles of hybrids and U. graciosus were very similar while that of the mtDNA donor species, U. ornatus, was highly divergent from both. Three antioxidant genes with different expression levels in hybrids and that warrant further study were SOD2, SOD3, and PRDX2. These results show how species can compensate for mtDNA introgression and highlight the complexity of genetic interactions encountered when closely related species exchange genetic material in zones of secondary contact.

Urosaurus graciosus and U. ornatus lizards were sampled from 2 different populations of each type of lizard at the following locations in Arizona during June 2020 (AZ permit #SP403988, IACUC Protocol #2019-0001): U. ornatus morphology and mtDNA n = 4: Burro Creek (n = 2, -113^o 27.00W, 34^o32.55N), Wickenburg Wash (n = 2, -112° 42' 39.13W, 33° 57' 2.12N); U. graciosus morphology with U. ornatus mtDNA “Hybrid ” n = 5: Detritial Wash (n = 3, -114^o30.84W, 35^o49.06N), Date Creek (n = 2, -113^o9.966W, 34^o9.160N), U. graciosus morphology and mtDNA n = 5: Walnut Creek (n = 3, - 114^o9.619W, 35^o2.17N), Hassayampa Wash (n = 2, -112° 40' 54.35"W, 33° 44.35"N).

MtDNA type was verified using a three primer species specific PCR protocol that amplifed either a 81 bp fragment from U. ornatus/hybrid type mtDNA from the ND1 gene or a 141 bp fragment from U. graciosus type mtDNA (see Haenel and Del Gaizo Moore, 2018 for details). The conserved reverse primer was 5’-GGTCGTACTGGTTCTTTTGTGAATA-3’, the U. ornatus and hybrid specific primer was 5’-CCCAAATGTAGTTGGCCCT-3’, and the U. graciosus specific primer was 5’-CGTCGCCTTCCTTACCCTAC-3’.

Liver tissue was dissected out immediately following euthanasia by rapid freezing in liquid nitrogen, separated into approximately 0.05 gram aliquots and stored in a -80ºC freezer. The liver was examined here because it is very metabolically active and previous work on mitochondrial function found evidence of dysfunction for mitochondria taken from liver cells in these lizards (Haenel and Del Gaizo Moore, 2018). Comet assays followed the alkaline protocol described in detail in Collins (Collins, 2014) with minor modifications. Solutions can be found in Collins (Collins, 2014) except where noted below. Each sample was coded by a 4 digit number such that the experimenters did not know which species the sample belonged to until it was categorized for statistical analyses. Liver cells were isolated by gently passing approximately 0.05 g liver tissue through a 100 µm cell strainer (CELLTREAT^® Scientific Products) in 3 ml Merchant’s Medium (0.14M NaCl, 1.47mM KH₂PO₄, 2.7mM KCl, 8.1mM Na₂HPO₄, 10mM Na₂EDTA, pH 7.4) on ice. Preliminary tests found this amount of tissue gave an appropriate number of cells per image. Isolated cells (15 µL) were added to 90 µL Merchants Medium then incubated at 37^oC for 20 minutes. Hydrogen peroxide (0.0878M H₂O₂) was added to treatments as a source of ROS while controls did not have any H₂O₂ added. Following incubation, 15 µL of the cells were added to 90 µL 1% LMP agarose (R&D Systems) and 50 µL of this agarose mixture was pipetted onto microscope slides (CometSlide™ R&D Systems). Slides were incubated at 4^oC for 10 minutes then transferred to lysis buffer (CometAssey^® Lysis Solution, R&D Systems) and incubated for 1 hour to overnight at 4^oC. Slides were then incubated in electrophoresis buffer (0.3M NaOH, 1mM EDTA) at 4^oC for 40 minutes.

Single cell electrophoresis was carried out for 1 hour at 500 mA per cm² in the same electrophoresis buffer at 4^oC. Following electrophoresis, slides were rinsed twice in distilled water for 5 minutes, placed in 70% ETOH for 5 minutes then dried in a 37^oC incubator for 15 minutes. Slides were then stained with Ethidium Bromide for 30 minutes and rinsed twice with distilled water. Comets were viewed and photographed with an EVOS_fi (ThermoFisher Scientific) inverted digital microscope using the 10x objective and RTF filter. Photographs were taken of at least 12 randomly selected fields of each treatment and control. Photos of comets were scored using OpenComet (Gyori et al., 2014), a plugin for ImageJ (Schneider et al., 2012). The default settings in OpenComet were used and all scorable nuclei were included from each field. Following the initial run with OpenComet, each image was viewed to find additional outliers that needed to be removed before statistical analyses. These were mainly where OpenComet scored two comets as one due to an overlap on the image.

Absorption spectroscopy is a commonly used method to quantify melanin in cells and tissue (Chung et al., 2019) and was used here. Samples for melanin quantification were obtained from liver tissue homogenized for the comet assay. Samples ranged in color from light tan to dark brown. Melanin standards ranging from 2 to 75 mg/mL were made from melanin (Sigma-Aldrich) dissolved in mitochondrial isolation buffer (Moore and Haenel, 2022). Absorbance at 475 nm was measured for each standard using an Omega BMG Labteck spectrophotometer, and a standard curve was plotted by subtracting out the background of the buffer alone. Then, 100 µL of each sample was dissolved in a 1 M NaOH solution, and absorbance was measured at 475 nm.