Oxidative stress in response to natural and experimentally elevated reproductive effort is tissue dependent

Summary Oxidative stress is a potential proximal physiological cost of reproduction. Detecting this cost may be performed in several different ways – manipulating reproductive status, correlating natural variations in effort to oxidative stress or manipulating reproductive effort. Here, we manipulated reproductive status and studied oxidative stress due to natural and experimental variation in reproductive effort in Brandt's voles (Lasiopodomys brandtii), using a variety of markers and tissues. We measured markers of oxidative stress in lactating (raising 6 to 8 offspring) and non-reproductive voles (Experiment I) and found that a marker of oxidative protection [serum total-superoxide dismutase (SOD) activity] was reduced, and a marker of oxidative damage (protein carbonyls) was increased, in the serum, in lactating compared with non-reproductive voles. However, protein carbonyls in the liver were lower in lactating compared with non-reproductive voles, consistent with increased liver SOD activity. Lipid damage [malonaldehyde (MDA)] in both serum and liver was unrelated to reproductive status. We compared these markers of oxidative stress between natural large (n ≥ 9) and small litter sizes (n ≤ 5; Experiment II), and between manipulated large (11–13) and small litter sizes (2–3; Experiment III) and found that liver MDA and SOD activity was higher in voles with natural large compared with natural small litter sizes, but there were no differences in other markers. There was no effect of litter size on all measures when it was experimentally manipulated. The effects of reproductive status on oxidative stress were critically dependent on the exact markers and tissues used. The effects of natural variation in reproductive effort suggested that there might be an oxidative stress cost associated with large litter sizes, but this effect was not replicated in the experimentally manipulated litters.