ARTICLE No apparent trade-offs associated with heat tolerance in a reef-building coral Liam Lachs 1 ✉, Adriana Humanes 1, Daniel R. Pygas 2,3, John C. Bythell 1, Peter J. Mumby 4,5, Renata Ferrari 2, Will F. Figueira 3, Elizabeth Beauchamp 1, Holly K. East 6, Alasdair J. Edwards 1, Yimnang Golbuu 5, Helios M. Martinez 1, Brigitte Sommer 3,7, Eveline van der Steeg 1 & James R. Guest 1 As marine species adapt to climate change, their heat tolerance will likely be under strong selection. Yet trade-offs between heat tolerance and other life history traits could compro- mise natural adaptation or assisted evolution. This is particularly important for ecosystem engineers, such as reef-building corals, which support biodiversity yet are vulnerable to heatwave-induced mass bleaching and mortality. Here, we exposed 70 colonies of the reef- building coral Acropora digitifera to a long-term marine heatwave emulation experiment. We tested for trade-offs between heat tolerance and three traits measured from the colonies in situ – colony growth, fecundity, and symbiont community composition. Despite observing remarkable within-population variability in heat tolerance, all colonies were dominated by Cladocopium C40 symbionts. We found no evidence for trade-offs between heat tolerance and fecundity or growth. Contrary to expectations, positive associations emerged with growth, such that faster-growing colonies tended to bleach and die at higher levels of heat stress. Collectively, our results suggest that these corals exist on an energetic continuum where some high-performing individuals excel across multiple traits. Within populations, trade-offs between heat tolerance and growth or fecundity may not be major barriers to natural adaptation or the success of assisted evolution interventions. https://doi.org/10.1038/s42003-023-04758-6 OPEN 1 School of Natural and Environmental Sciences, Newcastle University, Newcastle upon Tyne NE1 7RU, UK. 2 Australian Institute of Marine Sciences, Townsville, QLD 4810, Australia. 3 School of Life and Environmental Sciences, University of Sydney, Sydney, NSW 2006, Australia. 4 Marine Spatial Ecology Lab, School of Biological Sciences, University of Queensland, St. Lucia, QLD 4072, Australia. 5 Palau International Coral Reef Center, Koror 96940, Palau. 6 Department of Geography and Environmental Sciences, Northumbria University, Newcastle upon Tyne, UK. 7 School of Life Sciences, University of Technology Sydney, Sydney, NSW 2007, Australia. ✉email: [email protected] COMMUNICATIONS BIOLOGY | (2023) 6:400 | https://doi.org/10.1038/s42003-023-04758-6 | www.nature.com/commsbio 1 1234567890():,; O cean warming is causing profound changes in marine ecosystems, and to keep pace and avoid extirpation, species must either migrate or adapt. The adaptive capacity of ecosystem engineers, such as reef-building corals, will play a disproportionately large role in the future biodiversity and function of marine ecosystems. Coral reefs continue to face unprecedented declines due to mass coral bleaching and mortality events caused by marine heatwaves 1–3. These extended periods of anomalously high ocean temperatures are increasing in frequency and intensity under climate change 4,5. The ability of individual corals to survive levels of heat stress suf ficient to induce mass bleaching and mortality, hereafter ‘heat tolerance ’, will likely emerge as an important trait under natural selection during the coming decades. Survivorship rather than just bleaching is a core component of heat tolerance as bleached corals can still recover and persist 6,7. Considerable variability in coral heat tolerance exists between individuals, even within a single coral population on a single reef8. Currently, there are growing efforts to test novel restorative interventions, such as assisted evolution, which aims to enhance the heat tolerance of coral populations by seeding reefs with more tolerant coral colonies 9,10. Understanding variation in heat tol- erance is crucial to estimating the capacity for natural adaptation to climate change and the ef ficacy of assisted evolution inter- ventions. Under both natural selection and assisted evolution, the relationships between coral heat tolerance and fitness traits (i.e., reproduction and survival — even in non-heat stress years) or other ecological traits (e.g., growth and fecundity which may or may not affect fitness) are of fundamental importance to future population persistence. Organisms are limited by resource availability, forcing them to balance resource allocation between different physiological pro- cesses leading to trade-offs between resource-intensive traits. For instance, successful strategies to deal with drought stress are well known in long-lived birds, where less energy is allocated to reproduction in drought years to preserve cell maintenance and growth11. Such trade-offs will always occur between resource- limited processes. However, sometimes apparent positive asso- ciations can be found between resource-intensive traits, even when a trade-off might be expected. The variability of total resource budgets among individuals can explain this phenomenon12, and can be associated with positive correlations across multiple traits and co-tolerance to the impacts of multiple stressors among individuals 13, despite the presence of more nuanced resource trade-offs within individuals. For instance, in oysters there is genetic evidence for trade-offs (negative correla- tions) between reproductive effort and both survival and growth14, likely due to resource allocation. However, these negative correlations turn positive when these traits are measured across numerous oysters under feeding treatments 14. This can occur due to variability in resource acquisition among indivi- duals, which in turn can lead to differences in their total resource budgets. Subtle trait trade-offs are then easily masked by the broader population-scale energetic continuum (i.e., gradient of total resource budgets among individuals). Such positive pheno- typic correlations can also be associated to genetic correlations among traits which manifest as co-tolerance of individual organisms to multiple biotic and abiotic stressors 15. Heat tolerance in reef-building corals has been shown to have negative associations with growth 16–18, suggesting a resource trade-off. This can lead to considerable negative impacts on coral reefs at the ecosystem level 19. For instance, Acropora spp. and Pocillopora damicornis corals dominated by thermally tolerant Durusdinium spp. symbiotic microalgae show considerable reductions in vital cell processes, such as carbon storage 20, photosynthetic ef ficiency, and energetics 20,21. Ultimately this results in reduced coral growth in terms of calci fication rates16,22. Notably, this growth disadvantage can be eliminated under warming of 1.5 –3 °C, as growth rates decline disproportionally with increasing temperature for corals hosting Cladocopium symbionts compared to those hosting Durusdinium symbionts23. The presence of mixed symbiont communities and symbiont shuffling post-bleaching can lead to flexibility in the magnitude of heat tolerance-growth trade-offs 22,24. However, for other coral genera (e.g., Montipora spp.), there is mixed evidence on whether (see ref. 25) or not (see ref. 26) Durusdinium spp. symbionts (rather than Cladocopium spp.) in fluence coral host physiology and metabolism. Technological advances in photogrammetry now allow completely non-invasive determination of colony growth. This has some speci fic logistical advantages for repeated monitoring of corals in the field compared to other techniques which require removing corals from the substrate (e.g., buoyant weight) or causing potential harm (e.g., linear extension using staining)27. Many coral populations are dominated by a single symbiont taxon or a single symbiont community type 28,29. In these cases, do trade-offs between heat tolerance and other traits persist? Recent genomic evidence based on corals from contrasting thermal environments suggests that the shift in allele fre- quencies associated with coral host-derived heat tolerance are often associated with a fitness cost 28.H o w e v e r ,i ti sy e tt ob e tested whether trade-offs between heat tolerance and other ecological traits exist for corals that share the same Symbiodi- niaceae community. Considering their prevalence in numerous other taxa, including Crust acea, Insecta, and Chordata 30–32,i ti s likely that heat tolerance-related trait trade-offs also affect coral hosts. As coral adaptation occurs locally, not globally, and since endosymbiont communities are re latively uniform across local scales, it is important to resolve the extent of host-derived heat tolerance trade-offs to better predict coral adaptation to climate change. Egg and sperm development are resource intensive processes and have been suggested as potential costs to growth and heat tolerance33. It is reasonable to expect trade-offs between heat tolerance and fecundity given the evidence for heat tolerance- growth trade-offs in corals, and the fact that growth and gamete production are both resource-intensive processes. Evidence has shown that temperature stress can reduce hard and soft coral fecundity (i.e., egg density and volume) 34,35, and has suggested that corals may reabsorb their oocytes to divert energy away from reproduction and into growth under certain types of stress such as fragmentation 36,37. However, there has not yet been an assessment of the relationship between fecundity and heat toler- ance to test for associations or trade-offs which could be crucial to understand population fitness and performance. Here, we tested for ecologically relevant associations and trade- offs between heat tolerance and three ecological traits in a com- mon species of Indo-Paci fic coral: colony growth, fecundity, and symbiont community composition. To measure these traits we combined: (i) a long-term ( sensu38) 5-week marine heatwave emulation experiment to measure heat tolerance; (ii) interannual comparisons of 3D models of individual coral colonies to mea- sure growth (change in live surface area and colony volume); (iii) polyp counts and dissections to measure fecundity; and (iv) ITS2 sequencing to determine Symbiodiniaceae community composition. We employ Bayesian methods for solving simple trait trade-off linear regressions (in the form: heat tolerance ~ β0 + β1 × trait + error) to allow the quanti fication of uncertainty via inspection of posterior distributions, speci fically testing the odds of no trade-off occurring (i.e., β1 slope value > 0). ARTICLE COMMUNICATIONS BIOLOGY | https://doi.org/10.1038/s42003-023-04758-6 2 COMMUNICATIONS BIOLOGY | (2023) 6:400 | https://doi.org/10.1038/s42003-023-04758-6 | www.nature.com/commsbio Results Heat tolerance variability. Within the studied coral population, we found substantial variability in coral heat tolerance measured as bleaching and mortality responses throughout a marine heat- wave emulation experiment. At the beginning of the experiment all colonies had all replicate fragments healthy, corresponding to BSI (bleaching and survival index) values of 1. The final heat stress exposure reached a DHW (degree heating weeks) of 10.7 °C-weeks (Fig. 1a, b), a level which would likely induce a mass bleaching and mortality event in nature. By this final exposure, 47% of colonies had all replicate fragments dead (BSI = 0), while most remaining colony fragments were bleached (25% of all fragments), translating to BSI values < 0.56 (Fig. 1c). Meanwhile, in the unstressed procedural control tanks, all representative fragments from each colony remained alive, showing that the experimental setup (aquarium lights, flow 0 200 400 600 800 250 500 750 1000 Egg geometric mean diameter (μm) Counts MMMadj + 1°C MMMadj 28 30 32 34 36 0 1 2 3 4 5 Time (weeks) Temperature (°C) 0 4 8 0 1 2 3 4 5 Time (weeks) DHW (°C−weeks) Stress Tanks T2 T3 T5 T6 T1 T4 0 1 0 4 8 DHW throughout exposure ( C−weeks) BSI Symbiont ITS2 type C40.C15h.C3.C115.C40h (5) C40.C3.C115.C40h (34) C40.C3.C40i.C40j (2) C40.C3.C40i.C40j & D1.D4.D4c.D1c.D1h.D2 (1) C40.C3.C40j (5) C40.C3.C40j & C15h.C15hf.C15hg (1) No data (22) 0 1 0 1 2 3 4 5 Time (weeks) BSI 0.5 0.6 0.7 0.8 0.9 Colony heat tolerance (average BSI) 2017 2018 2019 Fast growth Slow growth Partial mortality A12 A27 A7 a b c e f d Fig. 1 Univariate exploration of traits among Acropora digitifera colonies showing marine heatwave emulation, 3D colony growth, and fecundity. a The experimental marine heatwave exposure conducted on 6 fragments per colony lasted 5 weeks reaching approximately +3.5 °C above the local climatological baseline (MMMadj). Colour legend is shared with panel b with heated tanks and procedural control tanks (T1 and T4). b This translated to accumulated degree heating weeks (DHW) of ~10 °C-weeks. c Bleaching and mortality responses (BSI) of each colony (individual lines) are shown throughout the experiment with a horizontal jitter to separate overlapping lines. d The BSI-DHW relationship (where BSIs are corrected for DHW drift among tanks) was unaffected by symbiont ITS2 type, showing number of colonies with each symbiont ITS2 type in brackets. e Interannual comparisons of coral colony structure-from-motion 3D models revealed marked variability in growth rates, with examples from some common growth types shown here, and all other colony models shown in Fig. S4. f Fecundity measurements from 2 fragments per colony match closely to previous estimates of egg diameter for Acropora spp. (red dashed line)39. COMMUNICATIONS BIOLOGY | https://doi.org/10.1038/s42003-023-04758-6 ARTICLE COMMUNICATIONS BIOLOGY | (2023) 6:400 | https://doi.org/10.1038/s42003-023-04758-6 | www.nature.com/commsbio 3