The climate is warming, and this is predicted to result in an increase in extremes of temperature. Understanding how this will affect the survival and distribution of organisms is vital if we are to prevent extinctions, and invasions by harmful pests. The impacts of climate change are often estimated by examining the temperatures that kill animals. However, we may be underestimating the impact that non-lethal high temperatures have on organisms. In most animals, from beetles to birds to badgers, males typically lose their fertility at a far lower temperature than that required to kill them. This also affects livestock, so that breeds which do well in temperate countries can be infertile in the tropics. Even humans, despite being able to control temperatures using clothes, houses, and air conditioning, show evidence of reductions in fertility during heatwaves. If increasing temperatures cause all the males in an animal population to become sterile, then that population will not survive, even if the temperatures are nowhere near high enough to actually kill any animals.

Recently we systematically tested the impact of temperature on the fertility of 43 species of fruit fly. In more than half, males lost fertility at a much cooler temperatures than their lethal limits. We then looked at where the flies were found worldwide. We found that across species and habitats, the highest temperature a species experiences in nature was far better predicted by sterilising temperature limits than by lethal temperature limits. This strongly suggests that temperature-induced sterility really does matter in nature, and that distributions of many species are limited by their ability to remain fertile at high temperatures. If so, then we are likely underestimating the impact of climate change on species, because we are only now realising that losses of fertility may be a major problem.

However, very little is known about the physiological and genetic processes which mean species are differently sensitive to temperature-based infertility. For example, whilst it has long been known that Heat Shock Proteins are critical to protecting cells from high temperatures in terms of survival, we don't yet know if the same proteins protect sperm. Our previous work shows that heat-shock might not affect existing mature sperm, but when these are used, species become infertile and don't seem to be able to recover. This suggests that it is the process of making sperm (spermatogenesis) that is damaged by heat.

We will study 7 species of fruit fly, which cover the range both in absolute differences in temperature tolerance, and the difference between the limits of fertility and survival. We will assess when spermatogenesis breaks down, using microscope images to describe if heat damages the sperm morphology or the DNA packaged inside. To understand which genes are important, we will measure gene expression in testes both before and after heat shock, to assess if some species are primed to respond more quickly. Crucially, we will not just correlate these findings, but use genetic and chemical manipulations to directly test the function of specific genes. Once we know what causes variation between these fly species, this can then be tested across other animals. Ultimately, this information could be used to quickly assess which species will be vulnerable to high temperature and identify particular genes to target in livestock breeding, to generate more thermally robust breeds.

Technical Abstract:
We developed a high-throughput heat shock assay to determine Thermal Fertility Limits (TFL) across Drosophila species. Here, to investigate the range of variation both in absolute TFL and TFL-CTL gap, we will use 7 target species. We will assess spermatogenesis and gene expression, before, during and after heat-shock at 1oC below that when 80% are infertile (TFL80).

We will measure gross testes morphology, and parameters of spermatogenesis (1) number of spermatocytes per cyst, (2) length of the cyst and (3) localisation of spermatocyte nuclei along the cyst, as an indicator of elongation (4) chromatin condensation within sperm nuclei; and (5) number of abnormal individualized complexes. We will perform TUNEL assays to assess DNA damage, and correlate TFL sensitivity with tolerance of an oxidative stress agent.

We will use RNAseq to examine differential expression (DE) in response to heat-shock. We predict genes conferring thermal tolerance are enriched in testes in robust species before stress, and/ or will rapidly change expression during heat stress, and will show less DE compared to controls after heat-stress. We will use a combination of computational approaches including weighted gene co-expression networks, gene ontology and pathway analysis approaches, to identify candidate genes. For 5 of the most promising candidates and test these using RTq-PCR in another 3 robust and 3 sensitive species.

We will attempt to functionally interfere with TFL profiles by using transgenics or chemical inhibitors to interfere with genes or the epigenetic landscape. We expect that if a gene/ epigenetic modifier is critical then we will be able to either generate a more sensitive TFL in a robust species or improve robustness in a sensitive species. In D. melanogaster we will use testes-specific Gal4 drivers to knock down genes of interest. For all target species, we will feed them inhibitors such as sodium butyrate, SAH, BIX, curcumin to modify their epigenome.