Global climate change is expected to have broad effects on the biodiversity of Earth, and our knowledge about how organisms will respond to ongoing environmental disturbance still needs considerable improvement (IPCC, 2014). Many studies have measured the thermal performance and lethal heat thresholds of species to examine the ecological and evolutionary impact of rising global temperature, with the ultimate goal of making inferences about how population processes will be affected (summarized in Seebacher et al., 2014; Gunderson and Stillman, 2015; Sinclair et al., 2016; Pinsky et al., 2019; Buckley and Kingsolver, 2021). However, these studies tend to focus on diploid adult or sub-adult individuals. Far fewer studies have focused on how sublethal temperatures impact reproductive cells, even though they have direct influence on fertility and fitness. For example, a recent study on 43 Drosophila species showed that male sterility temperatures better predict species global distributions than the upper lethal temperatures of adults (Parratt et al., 2021). This result emphasizes the significant role male gametes can play in shaping ecological pattern and process, and their likely importance in dictating how biodiversity will react to global warming.

Sperm are one of the only cell types that are released into a foreign environment after maturing, traveling through and interacting with the external physical environment and/or the female reproductive tract (Pitnick et al., 2009, 2020). During the life cycle of sperm, from development in the testis to egg fertilization, sperm may experience a wide range of environmental conditions including variation in temperature, pH, ionic state, and viscosity (Reinhardt et al., 2015). Here we focus on temperature, one of the most significant environmental factors relevant to environmental adaptation and climate change that could impact sperm traits and their ability to fertilize. In general, studies of temperature-dependent sperm performance are concentrated on model organisms used for medical or domestic breeding purposes, and therefore lack ecological and evolutionary context (Reinhardt et al., 2015; Lüpold and Pitnick, 2018; Walsh et al., 2019). From these studies, we know the impact of thermal stress on sperm traits can include morphological and behavioral changes, decreased sperm count and longevity, and increased DNA damage (Cameron and Blackshaw, 1980; Foldesy and Bedford, 1982; Flowers, 1997, 2015; Blanckenhorn and Hellriegel, 2002; Reinhardt et al., 2015; Parrish et al., 2017; Peña et al., 2021). Therefore, as thermal niches of populations diverge, we expect the temperature-dependence of sperm performance to undergo adaptive evolutionary change in concert with other aspects of organismal performance. However, considerably less is known about ecologically relevant sperm thermal biology in wild populations. This is slowly changing, however, and exciting results are beginning to emerge.

Here, we review empirical evidence for how temperature, and especially high temperature, affect sperm traits. We focus on data from wild animal populations whenever possible, but include data from plants and domesticated organisms when appropriate. We also attempt to place data within the “thermal performance curve” framework (Huey and Kingsolver, 1989). Thermal performance curves describe how rates of biological processes change with temperature, and are applicable to everything from enzymatic catalysis to population growth rates (Angilletta, 2009; Somero et al., 2017). The crux of thermal performance curves is that they are unimodal and asymmetrical: performance peaks at a temperature closer to the heat tolerance limit than to the cold tolerance limit (Figure 1A). This means that organisms tend to live at body temperatures close to their heat thresholds, making them vulnerable to warming (Deutsch et al., 2008; Huey et al., 2009). Data on sperm thermal traits are rarely discussed within the performance curve context (Figure 1B), but we believe doing so is beneficial for understanding the evolutionary ecology of sperm traits. Throughout, we attempt to synthesize emerging patterns from the available data, and also highlight gaps in our understanding that require greater attention.

Our review includes morphological and performance aspects of sperm phenotypes. We apply the term “performance” broadly, including any aspect of sperm-related phenotypes (number, percent motility, velocity, and fertilization success) that is not explicitly morphological.

Temperature can affect sperm traits in many ways, both pre- and post-ejaculation (Figure 2). Below, we step through examples of how temperature can affect sperm traits at different stages of the sperm life cycle.

Phenotypic plasticity refers to the ability of a genotype to express different phenotypes based on environmental conditions (Pigliucci, 2001). If plasticity occurs in a manner that improves performance, it is considered adaptive (Huey et al., 1999; Wilson and Franklin, 2002). Adaptive phenotypic plasticity is thought to be an important mechanism that can help organisms cope with temperature fluctuations (Somero, 2010; Seebacher et al., 2014; Gunderson and Stillman, 2015).

Explicit tests of adaptive phenotypic plasticity in sperm traits are rare, but there is compelling evidence that it can occur. Perhaps the best evidence comes from a study of red flour beetles (T. castaneum). Male beetles were acclimated to either standard or warm conditions, and were then mated to standard-reared females under the male acclimation temperature. After mating, females were move into either standard or warm conditions to reproduce. The results showed that females were able to produce significantly more offspring when their post-mating temperature matched male acclimation temperature. In other words, sperm performance was greatest under temperatures to which males were acclimated (Vasudeva et al., 2019). However, adaptive sperm thermal plasticity has not been supported in some other studies. In brown trout (S. trutta) and European whitefish (C. lavaretus), adult males acclimated to warm temperatures did not produce sperm with improved thermal tolerance or relative motility under warm temperatures (Fenkes et al., 2017; Kekäläinen et al., 2018). Similarly, no adaptive plasticity has been found in response to either rearing or adult acclimation temperature in mosquitofish (G. holbrooki; Adriaenssens et al., 2012; Iglesias-Carrasco et al., 2020).

Another approach to inferring adaptive plasticity is to compare the direction of plastic phenotypic change to the direction of evolutionary phenotypic change under the same conditions (Ghalambor et al., 2015). Interestingly, two studies that have compared evolutionary and plastic responses of sperm traits to temperature within the same taxa found that they occur in opposite directions: decreased sperm length was the plastic response to heat, while increased sperm length was the evolutionary response to heat (Breckels and Neff, 2013; Vasudeva et al., 2019). This suggests that some plasticity in temperature-dependent sperm traits could be maladaptive, and evolution acts to overcome that maladaptive response (Ghalambor et al., 2015).

Phenotypic plasticity is typically underlain by changes in gene regulation, and temperature-dependent transcript plasticity has been shown when individuals or their ejaculated sperm experience heat stress. For example, male domesticated bulls (B. taurus) produce sperm with an increased abundance of transcripts for heat shock protein genes hsp60 and hsp70 after exposure to heat (Cheng et al., 2016). In contrast, lower levels of hsp90 but not hsp70 mRNA were found in ejaculated sperm of mussels (M. galloprovincialis) after ejaculated sperm were exposed heat, though no phenotypic changes in sperm motility were observed (Lymbery et al., 2020). Whether differences in transcript abundance were due to gene regulation, degradation of specific RNAs, and/or changes in translation remain unclear (Lymbery et al., 2020). Heat-induced changes in transcript profiles, particularly for hsp genes, could imply an adaptive response since these genes code for molecular chaperones that maintain cellular function under thermal stress and are important in sperm development and function (Huang et al., 2000; Dun et al., 2012; Meccariello et al., 2014). However, further studies will need to be done to establish any adaptive significance.

From an evolutionary perspective, sperm traits are expected to be selected to maximize fertilization success under prevailing environmental conditions (Pitnick et al., 2009; Reinhardt et al., 2015). This view has been supported by studies on sperm morphology. For example, in many species, sperm length co-evolves with different aspects of female reproductive tract (Miller and Pitnick, 2002; Pitnick et al., 2009; Breckels and Neff, 2013). However, we know relatively little about adaptive divergence in sperm physiological traits with respect to abiotic environmental factors such as temperature (Reinhardt et al., 2015). From our review of the literature, there seems to be a correlation between adult thermal tolerance and the heat tolerance of sperm production and/or post-ejaculate performance. For example, several studies have compared sperm thermal performance between two or three species with divergent thermal niches and have found that performance at high temperature is greater in the warmer niche species (Rahman et al., 2009; Tourmente et al., 2011; Lahnsteiner and Mansour, 2012; Poullet et al., 2015; Revathy and Benno Pereira, 2016).

Of course, evolutionary adaptation cannot be inferred from two-species studies (Garland and Adolph, 1994). More compelling evidence comes from studies of divergence in male thermal sterility thresholds. For example, male whole-organism heat tolerance, a proxy for thermal niche, highly corelates with male sterility temperature across many Drosophilinae fly species (Figure 3; David et al., 2005; Parratt et al., 2021; van Heerwaarden and Sgrò, 2021). In addition, male fertility limits were found to correlate strongly with geographic distribution (and thus thermal environment) across many Drosophila species, providing further evidence for adaptive divergence in temperature-dependent male reproductive traits (Parratt et al., 2021). Interestingly, in almost every species, male fertility thermal (heat) limits (FTL) are lower than the adult heat limits (Figure 4), indicating that male fertility is more sensitive to heat than the whole organism (David et al., 2005; Jørgensen et al., 2006; Parratt et al., 2021; van Heerwaarden and Sgrò, 2021).

At the population level, divergence in sperm morphological traits is commonly reported (Kuramoto, 1996; Pitnick et al., 2003; Hettyey and Roberts, 2006; Minoretti and Baur, 2006; Kustra et al., 2019). Data on population-level divergence in environment-dependent performance is much more limited. For example, Green et al. (2019) found evidence of adaptive sperm evolution in response to local salinity conditions in two invasive populations of round goby (Neogobius melanostomus). Sperm from the high salinity population had greater sperm motility under high salinity conditions than those from the low salinity population (Green et al., 2019). Few studies have tested for intraspecies divergence in sperm thermal performance, but examples do exist. Sperm from a low latitude (i.e., warmer climate) population of Drosophila subobscura were less affected by heat than those from a high latitude population (Porcelli et al., 2017). Similarly, intraspecific variation in the heat sensitivity of fertility is largely consistent with habitat temperature in D. melanogaster (Rohmer et al., 2004) and the nematodes C. briggsae and C. tropicalis (Prasad et al., 2011; Poullet et al., 2015).

For sperm traits to evolve under different thermal regimes, there must be genetic variation in the traits among individuals. Individual-level variation in sperm thermal performance has been reported in Atlantic cod (Gadus morhua), where a genotype by temperature interaction was shown for sperm swimming speed (Purchase et al., 2010). Similarly, individual variation in sperm trait thermal plasticity has been reported in domesticated animals that experience scrotal insulation (Vogler et al., 1991; Barth and Bowman, 1994). Individual variation in sperm performance opens the door for temperature-dependent sperm competition. Furthermore, if adult thermal traits are genetically linked to sperm thermal traits, selection via sperm competition could increase rates of thermal adaptation when habitat temperatures change. Although rarely tested in animals, temperature selection on the gamete level has been applied in plants. For example, pollen selection has been shown to increase chilling tolerance in chickpea (Cicer arietinum; Clarke et al., 2004) and heat tolerance in maize (Zea mays L.; Mohapatra et al., 2020).

Laboratory experimental evolution experiments have shown that sperm traits can rapidly evolve under divergent thermal regimes, though these experiments are rare. In the two studies we are aware of, evolution under warm temperature led to a decrease in sperm length over time in guppies (8 generations, Breckels and Neff, 2013) and in red flour beetles (over 50 generations, Vasudeva et al., 2019). Whether the evolutionary response of sperm length is adaptive, or is simply linked with other adaptive traits, remains unclear. In contrast, laboratory selection experiments in Drosophila found no evidence for evolution in male fertility thermal limits, and heat-induced infertility often leads to population extinction (van Heerwaarden and Sgrò, 2021).

The evidence reviewed above has obvious implications for the ability of species to tolerate anthropogenic global change. High temperatures can have strong negative effects on sperm performance. Importantly, these effects can develop at temperatures not far above normal body temperatures, and far below the lethal body temperatures of mature adults that are often used to model susceptibility to warming (Figure 4). The ability of sperm traits to rapidly evolve under high temperatures does provide some hope for adaptation to rapid climate change. So too does the potential for adaptive thermal plasticity in sperm traits. However, the relative dearth of studies that have investigated questions of thermal evolvability and adaptive plasticity in sperm traits leaves us with much to learn before we can draw any solid conclusions.

One approach that could greatly increase our understanding of how climate change and sperm thermal performance interact is greater integration of ecologically meaningful temperature data into experimental design. The conditions that organisms and sperm are experimentally exposed to should reflect what they experience in the wild, as well as what they may experience in the future (Helmuth et al., 2010; Sunday et al., 2014; Buckley et al., 2018). In many of the papers described above, habitat-specific environmental or, more importantly, body temperature data were not presented nor discussed with respect to experimental treatments. As a result, it is difficult to interpret the results in an ecological context. Studies that use unrealistically high treatment temperatures could be biased toward finding a negative effect of warming, pushing organisms above the thermal optimum of their thermal performance curve. Similarly, studies that use unrealistically cool temperatures could incorrectly predict that warming will increase sperm performance if they span temperatures below the thermal optimum of the thermal performance curve. Ideally, experiments will occur across a wide range of temperature such that a thermal performance curves can be estimated and ecologically relevant temperatures can be mapped onto that curve. Then, one can predict the effects of warming within populations and estimate relative vulnerability to warming among taxa (Huey et al., 2009).

Furthermore, treatments should subject experimental units (individuals or sperm) to ecologically relevant thermal fluctuations, not just constant temperatures. Constant-temperature treatments are common in physiological experiments (Gunderson et al., 2016), but exposure to more ecologically relevant fluctuating temperatures often elicit different physiological responses (Colinet et al., 2015; Marshall and Sinclair, 2015; Marshall et al., 2021). Only when ecologically relevant thermal data are more consistently incorporated into studies of sperm thermal physiology will we be able to determine how close wild organisms are to experiencing sperm performance detriments, how great performance detriments may be when temperatures change, and how the effects of warming may differ geographically and taxonomically.

The weight of evidence makes it clear that temperature has a profound effect on sperm traits. These effects can manifest at many points in the life cycle of sperm and the tissues that produce and support them. Given the thermal sensitivity of sperm combined with the fact that they are so directly tied to reproductive fitness, we expect sperm traits to be under intense selection as species adapt to new thermal environments. Studies on the evolution of temperature-dependent sperm traits are relatively rare, but the data available suggest that divergence is common and may often be adaptive. Furthermore, experimental evolution experiments demonstrate that sperm traits can evolve rapidly when thermal conditions change.

Most studies of temperature-dependent sperm performance measured traits at only two or three temperatures, making inferences about thermal performance curve difficult or impossible. However, studies that applied a wide range of temperature treatments often found the classic thermal performance curve response for traits related to both morphology and performance (Figure 1B). Moving forward, we recommend that the thermal performance curve framework be used as a guide to design studies of thermal effects on sperm traits. Doing so will allow us to better understand the full spectrum of thermal effects on sperm and elucidate the ecological and evolutionary consequences of sperm phenotypes, as is the case for whole-organism thermal performance (Sinclair et al., 2016). In addition, application of the thermal performance curve perspective may help resolve inconsistencies observed among studies. For example, whether a study finds that warming increases, decreases, or has no effect on a trait of interest will depend on whether the treatment temperatures chosen fall on the rising (left), falling (right), or plateau section of the traits thermal performance curve. Ultimately, greater integration of sperm thermal responses to a wide range of ecologically relevant temperatures will be crucial to predict the effects of global warming on sperm and the subsequent consequences for reproduction and population processes.

WW and AG conceived the project and wrote the manuscript. Both authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.