Surviving heatwaves: thermal experience predicts life and death in a Southern Ocean diatom

Extreme environmental fluctuations such as marine heatwaves (MHWs) can have devastating effects on ecosystem health and functioning through rapid population declines and destabilisation of trophic interactions. However, recent studies have highlighted that population tolerance to MHWs is variable, with some populations even benefitting from MHWs. A number of factors can explain variation in responses between populations including their genetic variation, previous thermal experience and the intensity and duration of the heatwave itself. We disentangle the contributions of these factors on population survival and post-heatwave growth rates by experimentally simulating heatwaves (7.5 or 9.2 °C, for up to nine days) for three genotypes of the Southern Ocean diatom Actinocyclus actinochilus. The effects of simulated heatwaves on mortality and population growth varied with both genotype and thermal experience. Firstly, hotter and longer heatwaves increased mortality and decreased post-heatwave growth rates relative to milder, shorter heatwaves. Secondly, growth above the thermal optimum before heatwaves exacerbated heatwave-associated negative effects, leading to higher mortality during heatwaves and slower growth after heatwaves. Thirdly, hotter and longer heatwaves resulted in more pronounced changes to thermal optima (Topt) immediately following heatwaves. Finally, there is substantial intraspecific variation in mortality during heatwaves and in post-heatwave growth. Our findings shed light on the potential of Southern Ocean diatoms to tolerate MHWs, which will increase both in frequency and in intensity under future climate change.


31
Extreme temperature fluctuations in terrestrial and marine systems are occurring 32 with higher frequency and duration than previously, and will increase further with   From a physiological perspective, two key mechanisms affect the responses of 69 organisms to thermal extremes, a) the cellular stress response (Schroda et al., 2015) 70 and b) acclimation, which can result in "heat hardening" (Bowler, 2005). The cellular 71 stress response, defined as the upregulation of stress response genes including 72 those that express heat shock proteins, enhances tolerance to stressful conditions 73 (Schroda et al., 2015). Acclimation, or gradual phenotypic plasticity (Kremer et al.,74 2018), describes the effect of altered gene expression and epigenetic modifications 75 to adjust a phenotype (e.g. growth rate) in response to an environmental change 76 (Angilletta, 2009, Kronholm & Ketola, 2018. In phytoplankton, acclimation can alter  increases temperature below thermal optima, the temperature at which growth rate is 89 maximised, can be beneficial by enhancing metabolic activity (Angilletta, 2009). 90 However, environmental warming that raises temperatures near or above thermal the thermal extreme, ranging from acute (hours to days) to chronic (days to weeks) 94 (Huey & Bennett, 1990). Energy and resource investment into the expression of 95 acclimation and stress response genes, such as those that produce heat shock 96 proteins, incur fitness costs (Geider et al., 2009, Kingsolver & Woods, 2016 & Feder, 1997, Viant et al., 2003) and if these are high they can limit responses to 98 future environmental change (Sokolova et al., 2012). In the context of marine 99 heatwaves, the impact of elevated temperatures will be dependent upon the thermal 100 niche of the organisms present, which is subject to both interspecific and 101 intraspecific variation (Boyd et al., 2013). Furthermore, the state of cellular condition 102 before heatwaves has the potential to affect population resilience to heatwaves when  To understand how intraspecific variation and thermal experience interact in 121 determining survival and subsequent population growth, we investigated the growth 122 response of the Antarctic diatom Actinocyclus actinochilus to heatwaves. We 123 examined the impact of heatwaves on growth rates using three thermal variables: 1) 124 normal temperature, 2) heatwave temperature and 3) heatwave duration. We used 125 three genotypes of A. actinochilus, all isolated from the Ross Sea, to assess the 126 potential for intraspecific variation in resilience to heatwaves. Experimental 127 populations were grown either below or above their thermal optima for three weeks 128 until they reached stationary phase, whereupon they were subjected to heatwaves of 129 7.5 or 9.2 o C for 0, 1, 3, 6, or 9 days, during which we measured mortality. Directly 130 after heatwave exposure, we produced acute thermal performance curves for each 131 genotype.

133
We examined whether previous acclimation at high temperatures dampens the 134 negative consequences (increased mortality and/or decreased post-heatwave 135 growth rates) of heatwave exposure by "heat hardening", or if previous growth at a 136 higher temperature exacerbates these negative effects which is consistent with 137 deteriorating cellular condition rather than heat hardening. In addition to looking at          We have accounted for this in two ways. Firstly, for each heatwave duration (0-9 231 days), each experimental population with differing thermal experiences was exposed 232 to nutrient limitation for the same length of time. As such, differences in mortality and 233 post-heatwave growth rates between treatment groups after a specific heatwave 234 duration are directly comparable. Secondly, to quantify the effect of nutrient limitation 235 on mortality in the experiment, we performed control treatments that maintained 236 constant temperatures between the first and second phase of our experimental 237 design, so that they were exposed to nutrient limitation only, without heatwaves.

267
Simulated heatwave temperatures for the control treatment were 2.73 and 6.19 o C.

268
Lighting was provided from above using cool white aquarium LED lights, and light 269 intensity was maintained at 45-55 μmol m -2 s -1 , measured using a 2-pi sensor.   The maximum slope gradient was estimated from the growth curves using a sliding 309 window approach across the two-week growth period, with the window providing the

373
Population survival during heatwaves 374 The interaction of acclimation temperature with heatwave temperature and duration 375 on the survival of experimental populations varied between genotypes (A4, B7 and 376 D8), but general trends could be identified (Figure 2). Acclimation at either 2.5 or 5.8   Since not all temperature response curves had the same shape, we described the before (day 0) and after heatwaves of 3 or 9 days in duration, for each genotype.

491
Thermal experience combines acclimation temperature (2.5 or 5.8 o C) and heatwave 492 temperature (7.5 or 9.2 o C; no temperature noted for cases with no heatwave  Analysis of a key thermal trait, Topt, revealed that A.actinochilus can alter its plastic 720 responses to enhance growth in a warming environment, but that not all genotypes 721 do so ( Figure 5). Upward shifts in Topt of up to +1.5 o C occurred in experimental 722 populations of genotypes A4 and B7 grown at warmer acclimation temperatures and 723 exposed to hotter heatwaves. However, we also found evidence of an upper limit to  We measured the influence of thermal experience on the survival and post-heatwave 783 growth of a Southern Ocean diatom exposed to simulated heatwaves. We found that 784 acclimation temperatures below Topt enabled higher survival in experimental 785 populations than did acclimation temperatures above Topt, which supports the 786 hypothesis that previous thermal stress exacerbates heatwave stress. However, 787 determination of changes in thermal response curves and Topt after heatwaves that 788 varied in both duration and temperature exposed a complex relationship between 789 thermal experience prior to heatwaves, heatwave duration and temperature