Range Size and Growth Temperature Influence Eucalyptus Species Responses to an Experimental Heatwave

Macquarie University PURE Research Management System

This is the peer reviewed version of the following article:

Aspinwall, M.J., Pfautsch, S., Tjoelker, M.G., et al. (2019), Range size and growth temperature influence Eucalyptus species responses to an experimental heatwave. Global Change Biology, vol. 25, no. 5, pp. 1665– 1684.

which has been published in final form at: https://doi.org/10.1111/gcb.14590

This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions.

DR. MICHAEL J ASPINWALL (Orcid ID : 0000-0003-0199-2972) DR. JOHN E DRAKE (Orcid ID : 0000-0003-1758-2169) DR. OWEN K ATKIN (Orcid ID : 0000-0003-1041-5202)

Article type : Primary Research Articles

Range size and growth temperature influence Eucalyptus species responses to an experimental heatwave Running title: mechanisms of tree heatwave tolerance

Michael J. Aspinwall1,2*, Sebastian Pfautsch1, Mark G. Tjoelker1, Angelica Vårhammar1, Malcolm Possell3, John E. Drake1,4, Peter B. Reich1,5, David T. Tissue1, Owen K. Atkin6, Paul D. Rymer1, Siobhan Dennison7, Steven C. Van Sluyter7

1Hawkesbury Institute for the Environment, Western Sydney University, Locked Bag 1797, Penrith NSW 2751, Australia 2Department of Biology, University of North Florida, 1 UNF Drive, Jacksonville FL 32224 USA 3School of Life and Environmental Sciences, University of Sydney, Sydney, NSW 2006, Australia 4Forest and Natural Resources Management, SUNY-ESF, 1 Forestry Drive, Syracuse, NY, 13210 USA. 5Department of Forest Resources, University of Minnesota, 1530 Cleveland Ave N., St Paul, MN 55108, USA 6Division of Plant Sciences, Research School of Biology, The Australian National University, Canberra, ACT 2601, Australia 7Department of Biological Science, Macquarie University, North Ryde, NSW 2109 Australia

*Corresponding author:Author Manuscript email: [email protected], phone: +1-904-620-5626

This is the author manuscript accepted for publication and has undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/GCB.14590

Abstract Understanding forest tree responses to climate warming and heatwaves is important for predicting changes in tree species diversity, forest C uptake, and vegetation – climate interactions. Yet, tree species differences in heatwave tolerance and their plasticity to growth temperature remain poorly understood. In this study, populations of four Eucalyptus species, two with large range sizes and two with comparatively small range sizes, were grown under two temperature treatments (cool, warm) before being exposed to an equivalent experimental heatwave. We tested whether the species with large and small range sizes differed in heatwave tolerance, and whether trees grown under warmer temperatures were more tolerant of heatwave conditions than trees grown under cooler temperatures. Visible heatwave damage was more common and severe in the species with small rather than large range sizes. In general, species that showed less tissue damage maintained higher stomatal conductance, lower leaf temperatures, larger increases in isoprene emissions and less photosynthetic inhibition than species that showed more damage. Species exhibiting more severe visible damage had larger

increases in heat shock proteins (HSPs) and respiratory thermotolerance (Tmax). Thus, across

species, increases in HSPs and Tmax were positively correlated, but inversely related to increases in isoprene emissions. Integration of leaf-gas exchange, isoprene emissions, proteomics, and respiratory thermotolerance measurements provided new insight into mechanisms underlying variability in tree species heatwave tolerance. Importantly, warm-grown seedlings were, surprisingly, more susceptible to heatwave damage than cool-grown seedlings, which could be associated with reduced enzyme concentrations in leaves. We conclude that species with restricted range sizes, along with trees growing under climate warming, may be more vulnerable to heatwaves of the future. Keywords: forests, heat shock proteins (HSPs), heat stress, isoprene, photosynthesis, thermal acclimation

Introduction Extreme climatic events (i.e., drought, heatwaves, floods) are the hallmark of contemporary climate change (Collins et al., 2013; Meehl & Tebaldi, 2004). These extreme events are expected to exert stronger selective pressures on organisms than gradual changes in mean climatic conditions (Gutschick & BassiriRad, 2003; Parmesan, 2006; Reyer et al., 2013). Heatwaves,

This article is protected by copyright. All rights reserved 4 loosely defined as a period of consecutive (>3) excessively hot days (Perkins & Alexander, 2013), are extreme climatic events that could negatively impact the function and survival of organisms; in some regions, the frequency, intensity, and duration of heatwaves are increasing (Coumou & Robinson, 2013; Seneviratne, Donat, Mueller & Alexander, 2014). The potential for heatwaves to negatively impact the persistence of species, and the diversity and function of ecosystems, underscores the need for determining the patterns and mechanisms underlying variation in organismal responses to heatwaves (Buckley & Huey, 2016). Plant species with large geographic range sizes may cope better with climate change and extreme events than species with smaller range sizes (Aitken, Yeaman, Holliday, Wang & Curtis-McLane, 2008; González-Orozco et al., 2016; Pacifi et al., 2015; Thuiller, Lavorel & Araújo, 2005). This prediction is based on the expectation that species with broad climatic distributions possess wider environmental tolerances, thereby allowing them to inhabit a broader range of habitats (Hamrick, Godt & Sherman-Broyles,1992; Leimu, Mutikainen, Koricheva & Fischer, 2006; Morin & Thuiller, 2009; Slayter, Hirst & Sexton, 2013). Yet, predicting species responses to climate change based on range size alone has its limitations (Pacifi et al., 2015; Pearson & Dawson, 2003), and empirical tests of the relationship between species range size and vulnerability to climate change are rare (Fensham, Fraser, Macdermott & Firn, 2015; Lacher & Schwartz, 2016). Heatwaves are also occurring under background temperatures that are, on average, warmer (Collins et al., 2013). It is not clear how plants growing and existing in warmer climates will respond to heatwaves. Plants often acclimate photosynthetic and respiratory rates in response to long-term warming (Aspinwall et al., 2016; Atkin & Tjoelker, 2003; Gunderson, Norby & Wullschleger, 2000; Slot & Kitajima, 2015), yet the role of thermal acclimation in conferring high-temperature tolerance is not known. A few studies have shown that trees exposed to short-term warming or high temperature events (minutes, hours) exhibit less tissue damage and physiological stress when exposed to extended heatwaves, compared to trees with no such history (Colombo & Timmer, 1992; Daas, Montpied, Hanchi & Dreyer, 2008; Ghouil et al., 2003; Niinemets, 2010). Yet, Drake et al. (2018) found that Eucalyptus parramattensis trees grown at ambient and warmed (+3 °C) temperatures showed equivalent physiological responses and coped equally well with an experimental heatwave of four consecutive days with

This article is protected by copyright. All rights reserved 5 temperatures > 43 °C. Additional studies are needed to assess how trees living in a warmed climate will respond to heatwaves of the future. Prolonged exposure to excessively high temperatures can impact plant performance at multiple scales (shown conceptually in Figure 1). High air temperatures directly affect leaf temperature (Tleaf), and along with associated increases in vapor pressure deficit of the atmosphere (VPD) influence stomatal conductance (gs), which in union with VPD modifies Tleaf via the extent of transpirational cooling. As Tleaf increases beyond the optimum temperature of photosynthesis, net photosynthesis (A) is reduced by several related processes (Lin, Medlyn &

Ellsworth, 2012) including reduced gs, increased photo- and mitochondrial (i.e. dark) respiration (R, von Caemmerer & Quick, 2000; Peñuelas & Llusià, 2002), inhibition of Rubisco activase and deactivation of Rubisco (Hozain, Salvucci, Fokar & Holaday, 2010; Law & Crafts-Brandner, 1999; Salvucci & Crafts-Brandner, 2004), as well as damage to photosystem II (Allakhverdiev et

al., 2008; Wise, Olson, Schrader & Sharkey, 2004). Excessively high Tleaf also (i) stimulates the

synthesis of isoprene (Is) (among isoprene emitting species) which may help stabilize thylakoid membranes (Monson et al., 1992; Pollastri, Tsonev & Loreto, 2014; Singaas, Lerdau, Winter & Sharkey, 1997; Velikova et al., 2011; Vickers, Gershenzon, Lerdau & Loreto, 2009), and (ii) can increase production of heat shock proteins (HSPs) that act as chaperonins for other proteins, protect against oxidative damage, and stabilize cellular membranes (Figure 1, Heckathorn, Downs, Sharkey & Coleman, 1998; Mittler, 2002; Vierling, 1991). A few studies have demonstrated that tree species may vary in their physiological and biochemical responses to heatwaves (Ameye et al., 2012; Guha, Han, Cummings, McLennan & Warren, 2018; Wujeska-Klause, Bossinger & Tausz, 2015). Even when high temperatures reduce

A, some tree species maintain high gs (and transpiration) which helps cool leaves (Drake et al., 2018; Urban, Ingwers, McGuire, & Teskey, 2017). There is also evidence that trees can tolerate temperatures much higher than they typically experience (O’Sullivan et al., 2017) and may acclimate to high temperatures by increasing the thermal limits of leaf function (e.g. photosynthesis, respiration; Drake et al., 2018; Sastry & Barua, 2017; Zhu et al., 2018). Isoprene

and HSPs may facilitate acclimation to high temperatures, yet associations between Is, HSP production, and changes in the thermal limits of leaf function in response to heatwave conditions have not been tested. In general, species differences in physiological and biochemical responses to heatwaves are not well-understood, especially in relation to species performance during

This article is protected by copyright. All rights reserved 6 heatwaves (Fauset et al., 2018; Teskey et al., 2015). Evidence from multiple studies suggest that species that maintain higher gs, lower Tleaf, and exhibit larger increases in Is, HSPs and leaf

thermotolerance (e.g., leaf respiratory thermotolerance, Tmax) are better in avoiding photosynthetic inhibition and tissue damage during heatwaves (Figure 1). Further experiments are required to test this expectation. We conducted a heatwave experiment using four co-occurring Eucalyptus species; two with large range sizes and two with comparatively smaller range sizes. A single population of each species was obtained from locations in southeast Australia with relatively similar climates. Trees of each species were grown from seed for more than 80 days under two temperature treatments (mean daily air temperature, cool = 18.0 °C, warm = 21.5 °C), before being exposed to an equivalent experimental heatwave. We assessed heatwave tissue damage severity, and measured leaf gas exchange, chlorophyll fluorescence (quantum yield of PSII, quantum efficiency of PSII), isoprene emissions, and concentrations of stress, photosynthetic, and metabolic proteins, before, during and after (i.e., recovery) the heatwave. We also measured the

high temperature threshold of leaf dark respiration (Tmax), a measure of leaf thermotolerance, before and during the heatwave to test whether species differ in physiological acclimation to heatwave conditions. We tested the hypotheses that: 1) species with large range sizes would show less heatwave damage and would maintain higher leaf gas-exchange and chlorophyll fluorescence during the heatwave compared to species with smaller range sizes, 2) trees grown under warmer mean temperatures would exhibit increased tolerance to high temperatures during heatwaves and would maintain higher leaf gas-exchange and chlorophyll fluorescence during the

heatwave compared to trees grown under cool temperatures, and 3) species that maintain high gs and low Tleaf, and show larger increases in Is, HSPs, and Tmax in response to heatwave conditions will show less heatwave damage in terms of impaired physiological function and visible injury.

Materials and methods Study species and experimental design Four co-occurring Eucalyptus species were included in this study: Eucalyptus camaldulensis subsp. camaldulensis (EC, River Red Gum), Eucalyptus tereticornis subsp. tereticornis (ET, Forest Red Gum), Eucalyptus botryoides (EB, Bangalay/Southern mahogany), and Eucalyptus smithii (ES, Blackbutt peppermint). Species were chosen based on differences in their range size

This article is protected by copyright. All rights reserved 7 and climatic distribution (Table 1, Figure S1). EC and ET both have large range sizes. EC is the most broadly distributed eucalypt in Australia, occurring across a wide range of climatic conditions. The distribution of ET spans temperate to tropical climates throughout eastern Australia. EB and ES have smaller range sizes, and both species occupy temperate forests of southeastern Australia (Figure S1). The distribution of EC and ET extends into warmer regions than EB and ES, so they have an overall warmer climatic range. Seed was selected for one provenance of each species from locations of comparable summer climates (Table 1). Thus, we attempted to separate the role of climatic adaptation and species range size in determining species heatwave tolerance. Seed material was sourced from the Australian Tree Seed Centre (CSIRO, Canberra Australia), using temperature, precipitation, latitude, and longitude data gathered for each provenance. The mean daily summertime (November to February) temperature for the selected provenances ranged from 18.4 to 21.2 °C. Seeds were germinated in potting mix in a shade house on the Western Sydney University Hawkesbury campus (Richmond, NSW Australia) during September and October 2014. Eight weeks after sowing, trees were transplanted individually into 6.9 L pots (15 cm diameter, 40 cm in height) filled with 9 kg of alluvial topsoil, sourced from a local quarry in Menangle, NSW Australia. A complete description of soil micro- and macro-nutrients is provided in Drake et al., (2015). Twenty trees of similar basal stem diameter and stem height were selected from each provenance and randomly divided into two groups. Each group was placed into one of two naturally-lit adjacent climate-controlled glasshouse rooms, each set at a different growth temperature. The first growth temperature simulated a mean daily (24-hr) summertime air temperature of 18 °C, the lower value among the four provenances (Table 1). The second growth temperature simulated the mean daily summertime air temperature + 3.5 °C warming (i.e. 21.5 °C); hereafter, we refer to the two temperature treatments as ‘cool’ and ‘warm’, respectively. The warming treatment mimicked a realistic future climate scenario for the year 2100 (Collins et al., 2013) and represents a temperature that shifts the provenances near or beyond their average summertime temperatures (Table 1). The diurnal variation in air temperature within each treatment was controlled by five temperature set points, resulting in a diurnal temperature range of roughly 9 °C. The mean maximum midday temperatures (12:00 – 16:00) of the cool and warm treatments were 23.5 ± 1.9 (standard error throughout) °C and 27.1 ± 1.9 °C, respectively, and

This article is protected by copyright. All rights reserved 8 the mean minimum night-time temperatures (00:00 – 06:00) were 15.6 ± 0.6 °C and 19.1 ± 0.6 °C, respectively. Humidifiers (Carel Humidisk 65, Carel S.p.a, Padova, Italy) were used to maintain relative humidity (RH) in each treatment at or above 50%. Plants were rotated within each treatment room on a fortnightly-basis and confounding of temperature treatment and glasshouse rooms was limited by rotating plants and temperature treatments between rooms every 3 weeks. Trees were well-watered throughout the experiment and fertilized fortnightly with liquid fertilizer (Aquasol, Yates Australia, Padstow NSW, Australia) to prevent water and nutrient limitations.

Heatwave conditions After 85 days of growth under the temperature treatments, six randomly selected trees of each species grown in each temperature treatment were exposed to a controlled heatwave (2 growth temperature treatments × 4 species × 6 replicates = 48 trees). Stem length, measured just before the heatwave, was significantly higher in warm-grown trees (P<0.0001; cool = 122 ± 3.3 cm, warm = 148 ± 3.5 cm). Stem length was highest in EC (172 ± 5.0 cm), lowest in EB (102 ± 4.7 cm), and similar between ET (132 ± 4.9 cm) and ES (134 ± 4.5 cm). The day prior to the start of heatwave (Day 1 of the experiment; 28 January 2015, Austral summer) at 08:00 h, trees were moved into one glasshouse room set to a mean daily Tair of 19 °C and a diurnal range of 9 °C.

Natural heatwaves often begin with a 1-2 day increase in mean daily Tair prior to a series of high temperature days (Furrer, Katz, Walter & Furrer, 2010). We simulated this temperature ‘ramp’ by increasing mean daily Tair to 24.5 °C (diurnal range, 19 – 30 °C) on Day 2, to 30 °C (25 – 38 °C) on Day 3, and to 36 °C (30 – 45 °C) on Day 4, which was then maintained through Day 8 of the experiment. Thus, the ‘heatwave’ portion of the experiment lasted for five consecutive days

(Days 4-8 of the experiment) with mean maximum midday Tair of 45 °C. This heatwave is extreme but not unrealistic. Heatwaves in this region can last 4 days with temperatures 15 °C above average maximum daily summertime temperatures (>40 °C, Perkins & Alexander, 2013). By 2100, heatwaves in this region are expected to be longer and hotter (roughly 3-5 °C above current average maximum heatwave temperatures, Cowan, Purich, Perkins, Pezza, Boschat, &

Sadler, 2014). At the end of Day 8, Tair was decreased over a 12 h period (18:00 – 06:00) and the

pre-heatwave Tair conditions resumed on Day 9 (15 – 23 °C). We mitigated the potentially confounding and compounding effects of water stress and high VPD on heat stress by

This article is protected by copyright. All rights reserved 9 humidifying the room, and by filling pots with water 1-2 times daily. Thus, this experiment focused on the effects of extreme heat alone and without co-occurring effects of soil and atmospheric drought. Mean and maximum (15-minute interval) Tair and VPD during the heatwave are shown in Figure S2.

Tree aboveground injury A subjective visual index was used to evaluate the extent of tissue necrosis through progression of the heatwave. The index ranged from 0 to 2; 0 indicating little or no visible browning or mortality of leaves, branch tips, and stem tips; 1 indicating clearly visible but minor to moderate browning or mortality of leaves, branch tips, and stem tips (see Figure S3a); and 2 indicating heavy to severe browning and mortality of leaves, branch tips and stem tips (see Figure S3b). For simplicity, we report observations of heatwave damage on Day 9 only (1 day post-heatwave).

Leaf gas-exchange, chlorophyll fluorescence, and water potential Leaf gas-exchange and chlorophyll fluorescence were measured on all trees on Days 1 – 8 of the experiment, and Days 9 and 14 (1 and 6 days post-heatwave). Measurements were taken each -2 -1 day between 10:00 and 12:00 h. Light-saturated leaf net photosynthesis (Asat, μmol m s ), -2 -1 stomatal conductance to water vapour (gs, mol m s ), quantum yield of PSII (ΦPSII), and

quantum efficiency of PSII (Fv′/Fm′) were measured on one fully-expanded, mature upper canopy leaf per tree using four portable photosynthesis systems (LI-6400XT, LI-COR, Inc., Nebraska, USA) equipped with modulated chlorophyll fluorometers (6400-40). The measured leaf was tagged on the first measurement day and repeatedly measured throughout the

experiment. ΦPSII quantifies the proportion of absorbed energy being used in photochemistry and is associated with the quantum yield of photosynthesis (Genty, Briantais & Baker, 1989;

Edwards & Baker, 1993). Fv′/Fm′ provides a measure of the efficiency of PSII (Ameye et al., 2012, see Maxwell & Johnson, 2000 for calculation) and quantifies limitations to PSII by

thermal decay processes (i.e. decreasing Fv′/Fm′ indicates greater non-photochemical quenching, Oxborough & Baker, 1997). For all measurements, actinic photosynthetic photon flux density -2 -1 (PPFD) within the cuvette was maintained at 1800 μmol m s , and chamber reference [CO2] -1 was controlled at 415 μmol mol . Cuvette block temperature was set at the target midday Tair for

each day (i.e., 23, 30, 38, or 45 °C). Leaf temperature (Tleaf) was measured using the LI-6400XT

This article is protected by copyright. All rights reserved 10 leaf temperature thermocouple and was recorded concurrently with leaf gas-exchange and chlorophyll fluorescence data. We also used a handheld thermal camera (FLIR Systems, Oregon,

USA) to measure Tleaf outside of the LI-6400XT cuvette on a subset of leaves per species. Water vapour content of incoming air was not scrubbed so that reference RH conditions inside the cuvette were similar to RH conditions within the glasshouse bay. Gas-exchange and chlorophyll fluorescence measurements were recorded after Asat and gs reached steady state, typically within 5 min of closing the leaf within the cuvette. Fluorescence parameters were calculated following the built-in functions of the LI-6400XT system. Midday (12:00 – 14:00 h) leaf water potentials

(Ψmd, MPa) were measured each day on leaves similar to those used for gas-exchange and fluorescence. One leaf per plant was excised and immediately placed in a moist plastic zip-lock bag. Leaf water potentials were measured within 30 minutes after sample collection using a pressure chamber (PMS Instrument Company, Oregon, USA).

Respiratory thermotolerance

The leaf temperature at which leaf dark respiration (R) is maximal (Tmax) was determined by measuring the short-term temperature response of R across a temperature range of 15 to 70 °C.

Tmax is an important measure of leaf thermotolerance. At temperatures above Tmax, the decline in R is irreversible reflecting loss of respiratory function and onset of tissue death (Heskel et al.,

2014; O’Sullivan et al., 2017). Pre-heatwave measures of Tmax occurred across three days, roughly 1-week prior to the heatwave. Measurements of Tmax during the heatwave occurred across the third and fourth day of the ‘extreme’ portion of the heatwave (i.e. day 6 and 7). In total, Tmax was measured on 5 or 6 replicate trees (1 leaf per tree) from each species × growth temperature combination both before and during the heatwave (n=89). Leaves were detached pre-dawn, and leaf area (LA, m2) was quickly measured with a leaf area meter (LI-3100C, Li- Cor Inc., Lincoln, NE, USA). Leaves were immediately placed in sealed plastic bags with moist paper to prevent desiccation and placed in a dark box to ensure they remained dark-adapted. Previous studies using the same technique for measuring the temperature response of leaf R found no difference between attached and detached leaves of related species (Gauthier et al., 2014; O’Sullivan et al., 2013). The short-term temperature response of R was measured using an established heating protocol (O’Sullivan et al., 2013 and 2017), which utilizes a large gas-

This article is protected by copyright. All rights reserved 11 exchange chamber (3010-GWK1, Heinz Walz GmbH, Effeltrich, Germany) connected to an infrared gas analyser (LI-6400XT). We used four 3010-GWK1 chambers to expedite the measurements. From each short-term temperature response of R, the leaf temperature at which R is maximal was identified as Tmax. Respiratory acclimation to the heatwave was also determined by comparing rates of R measured at 20 °C before and during the heatwave. Upon completing each curve, leaves were removed and dried at 70 °C for 72 hrs. Leaf dry mass (g) was determined and leaf dry mass per unit leaf area (LMA, g m-2) for each leaf was calculated by dividing leaf dry mass by LA.

Isoprene measurements

Leaf isoprene emission rates (Is) were measured three days before the heatwave, at Day 6 (peak heatwave), and Day 13 (five days post-heatwave) using a LI-6400XT portable photosynthesis

system with a 2×3 cm leaf chamber. Leaf temperature was set at the ambient Tair, PPFD was set 2 -1 at 1500 µmol m s using the LED red/blue light source, reference CO2 concentration was set at 415 µmol mol-1, and flow rate set at 500 µmol s-1. Isoprene concentration in the air exiting the leaf chamber, for both cuvette blanks and samples, were determined by sampling the air at 200 mL min–1 for 10 minutes into sampling tubes containing Tenax (60–80 mesh; Sigma–Aldrich, Sydney) using an AirChek 2000 sampling pump (SKC Inc., Eighty Four, PA, USA). Isoprene concentrations and calculated emissions rates (nmol m-2 leaf area s-1) were determined by gas chromatography following the protocol described in Methods S1.

Proteomics We repeatedly sampled three of six trees of each species × growth temperature combination on four days for proteomics analysis. Leaves were similar to those measured for leaf gas-exchange and chlorophyll fluorescence and were collected on Day 1 (pre-heatwave), Day 7 (heatwave), Day 9 (one day post-heatwave), and Day 14 (six days post-heatwave). Leaves, one from each plant, were collected at midday (~12:00 hr), flash frozen in liquid N, and stored at -80 °C. In total, 96 leaves were sampled and analysed. Leaf pieces ranging from 50 to 100 mg fresh weight (2.3 to 5.1 cm2) were taken from each leaf sample and ground to a fine powder. The powder was cold solvent extracted and 2.5 µg ovalbumin per cm-2 leaf area (5.64 × 10-11 mol cm-2) was added as an internal standard. The

This article is protected by copyright. All rights reserved 12 powder was extracted with phenol and hot borate-citrate buffer. Protein was precipitated from the phenol with ethanol and diethyl ether and the resulting pellets dissolved in SDS-urea buffer. Protein concentration was assayed by FluoroProfile Kit (Sigma-Aldrich, St. Louis, MO, USA). Peptides were analyzed on a TripleTOF 6600 (Sciex, Framingham, MA, USA) using a 60 min acetonitrile gradient on a 75 μm × 10 cm C18AQ column at 300 nL/min. A SWATH MS (Gillet et al., 2012) ion library was created from data dependent acquisition (DDA) runs using two sets of 16 randomly pooled samples, four samples from each species per pool, four DDA runs per pool. Results from the DDA runs were matched using ProteinPilot (Sciex, Framingham, MA, USA) to protein sequences from the following sources: predicted proteins from the Eucalyptus grandis genome project (Phytozome; Bartholomé et al., 2015), myrtales and Arabidopsis thaliana mitochondrion encoded proteins (UniProt), Eucalyptus globulus chloroplast encoded proteins (Bayly et al., 2013), and common contaminants (http://www.thegpm.org/crap/). Protein areas were calculated from exported SWATH results by summing the top two ion areas of the top three peptides per protein per sample (Ludwig et al., 2012). Protein concentrations were estimated by dividing each protein area by the ovalbumin protein area for that sample, then multiplying by 5.64 × 10-11 mol cm-2 and the protein molecular weight calculated from the amino acid sequence. Proteins were functionally annotated using the MapMan scheme and the Mercator 3 online tool (Lohse et al., 2014). In total, we identified >2000 proteins. However, we largely focused on examining how broad categories of photosynthetic (e.g., Calvin cycle, light reactions), metabolic (e.g., TCA cycle, glycolysis, and lipid metabolism), and abiotic stress proteins (e.g., heat shock proteins) varied between treatments, among species, and in response to the heatwave conditions. We also examined sub-categories of proteins and specific proteins with well-known functions or sensitivity to heat stress (e.g., Rubisco large sub-unit, HSP70, PSII light harvesting complex). Our analysis was carried out on summed protein concentrations for each category, sub-category, or specific protein. A more detailed step-by-step description of the proteomics methods is provided in Methods S2. A complete list of the proteins (with identifiers) considered in this analysis is provided in Table S1.

Statistical analysis

Statistical analyses were conducted in SAS v.9.3 (SAS Institute, 2010). Growth temperature, species, and growth temperature × species effects on heatwave damage were tested using logistic regression (PROC LOGISTIC), which used maximum likelihood estimation to model the probability of damage based on a χ2 distribution. Tree physiological and proteomic responses over time were examined using a repeated measures mixed model (PROC MIXED) in the form:

Yijk = μ + Di + Tj + Sk + DiTj + DiSk + TjSk + DiTjSk + Rijk + Ɛijk

where Yijk represents the response variable (e.g. Asat, Fv′/Fm′), Di represents the ith day of the

experiment, Tj represents the jth growth temperature treatment (cool, warm), and Sk represents

the kth species. All other terms represent the respective interactions and Rijk and Ɛijk represent the repeated measures term with an autoregressive covariance structure and the residual error term,

respectively. Stomatal conductance (gs) and ΦPSII were log-transformed to fulfil assumptions of normality. All trees experienced the heatwave, so pre-planned contrasts of pre-heatwave (day 1), heatwave (days 4-8), and post-heatwave measurements (days 9 or 14) were used to infer species and temperature treatment effects on tree heatwave performance.

Results Heatwave-induced tissue damage Heatwave damage differed among species (χ2 = 9.0, P=0.03) and between growth temperature treatments (χ2 = 6.6, P=0.01), with no species × temperature interaction (χ2 = 0.84, P=0.84). The species with larger range sizes (EC and ET) showed less heatwave damage (7% damage score ≥ 1) than the species with smaller range sizes (EB and ES, 35% damage score ≥ 1) (Figure 2a). Moreover, neither large range size species exhibited severe damage (damage scores of 2) whereas both small range size species did (Figure 2a). In contrast to our expectations, heatwave damage was more common and severe in warm- grown trees compared to cool-grown trees. Thirty-two percent of warm-grown trees showed some damage (damage score ≥ 1), with 9% showing severe damage (Figure 2b). Only 13% of cool-grown trees showed some damage, with 4% showing severe damage (Figure 2b).

Leaf gas-exchange and chlorophyll fluorescence

Species differed in Asat, gs, and Fv’/Fm’ before the heatwave (Day 1 of experiment, Figure 3). -2 -1 EC showed higher Asat (21.2 ± 0.9 μmol m s ) than the remaining species (P<0.001, ET = 15.1 -2 -1 -2 -1 ± 1.1, EB = 12.8 ± 0.8, ES = 14.4 ± 1.1 μmol m s ) and higher gs (0.76 ± 0.05 mol m s ) than -2 -1 -2 -1 EB (P=0.01, 0.37 ± 0.04 mol m s ), but similar gs to ET (P=0.13, 0.53 ± 0.05 mol m s ) and -2 -1 ES (P=0.49, 0.66 ± 0.08 mol m s ). The large range size species showed higher Fv’/Fm’ (EC = 0.48 ± 0.01, ET = 0.50 ± 0.02) than the small range size species (P<0.05, EB =0.41 ± 0.01, ES = 0.43 ± 0.01).

All species reduced in-situ midday Asat, Fv’/Fm’, and gs during the heatwave (days 4-8, Figure 3), but the extent and pattern of the reduction differed among species (day × species, Table 2, Figure 3). Averaged across heatwave days, EC and EB (a large and small range species, respectively) showed the smallest and largest reduction in Asat (-45% and -99%, respectively),

Fv’/Fm’ (-21% and -40%, respectively) and gs (-15% and -40% respectively) relative to pre- heatwave values (Figure 3). ET and ES showed intermediate reductions in Asat, Fv’/Fm’, and gs (ET = -77%, -36%, and -36%, ES = -89%, -33%, and -29%).

Overall, species with large range sizes sustained higher Asat (P<0.0001, EC = 11.6 ± 0.8, -2 -1 ET = 3.9 ± 0.5, both in μmol m s ) and Fv’/Fm’ (P<0.0001, EC = 0.38 ± 0.01, ET = 0.31 ±

0.01) during the heatwave (days 4-8) than species with smaller range sizes (Asat = 1.7 ± 0.5 (ES) -2 -1 and 0.33 ± 0.3 (EB) μmol m s ; Fv’/Fm’= 0.29 ± 0.01 (ES), 0.25 ± 0.01 (EB), Figure 3). Across -2 heatwave days, EC showed higher gs than the remaining species (P<0.0001, 0.62 ± 0.05 mol m -1 -2 -1 s ), and ES and ET showed higher gs (0.41 ± 0.04 and 0.30 ± 0.02 mol m s , respectively) than -2 -1 EB (0.22 ± 0.02 mol m s , Figure 3). Ψmd showed a significant day × species interaction (Table

2). Ψmd did not differ among species before the heatwave (P=0.34, average Ψmd = -0.72 ± 0.02

MPa), yet species that maintained higher gs showed significantly lower Ψmd (EC = -1.33 ± 0.04

MPa; ES = -1.39 ± 0.04 MPa) during the heatwave than species that maintained lower gs (P=0.007, EB = -0.98 ± 0.04 MPa; ET = -1.25 ± 0.03 MPa, Figure 3).

Notably, all species gradually increased Asat, Fv’/Fm’, and gs between days 4 and 8 of the heatwave (Figure 3), yet the increase in Asat, Fv’/Fm’, and gs per day was higher in large range -2 -1 -1 size species (Asat = 1.4 (ET) to 2.7 (EC) μmol m s day ; Fv’/Fm’= 0.015 (ET) to 0.027 (EC) -1 -2 -1 -1 units day ; gs = 0.02 (ET) to 0.06 (EC) mol m s day ) than in small range size species (Asat = -2 -1 -1 -1 0.76 (ES) to 0.79 (EB) μmol m s day ; Fv’/Fm’ = 0.007 (ES) to 0.012 (EB) units day ; gs = 0.01 (EB) to 0.01 (ES) mol m-2 s-1 day-1; Figure 3, Figure S4). Overall, we observed a tendency

This article is protected by copyright. All rights reserved 15 for large range size species to maintain higher gas-exchange and chlorophyll fluorescence (and smaller reductions in these processes) during the heatwave than species with small range sizes.

Species differences in gs during the heatwave (days 4-8) generated species differences in the Tleaf - Tair differential (P<0.0001, both measured using the LI-6400XT). Higher gs during the heatwave was associated with Tleaf below Tair, while lower gs was associated with Tleaf above Tair

(Figure 4a). As a result of high gs, Tleaf during the heatwave was on average 1.08 (±0.2) °C below Tair in EC (Figure 4b). In contrast, Tleaf during the heatwave was on average 1.3 (±0.2) °C above Tair in EB, which had low gs (Figure 4b). Tleaf during the heatwave was on average 0.02

(±0.1) °C below and 0.3 (±0.1) °C above Tair, respectively, in ES and ET (Figure 4b). Thermal

images taken during the heatwave supported the observed differences in Tleaf among species

(Figure S5). In one image, Tleaf measured on one individual of EC ranged from 41 to 42 °C

(when Tair was ~45 °C). Tleaf measured on an adjacent individual of EB was nearly 50 °C (Figure

S5a). Individuals of ET and ES showed Tleaf values ranging from 44 to 48 °C (Figure S5d). The

absolute values of Tleaf measured with the leaf chamber and thermal images differ, but the data are congruent. Averaged across species and all measurement days, warm-grown trees showed 26% -2 -1 -2 -1 lower Asat (7.2 ± 0.4 μmol m s ) than cool-grown trees (9.7 ± 0.4 μmol m s ) when measured

at a common temperature (no day × treatment interaction; Table 2). Fv’/Fm’ and gs showed a significant growth temperature treatment × species interaction (Table 2). Warm-grown trees of

ET showed reduced Fv’/Fm’ and gs yet warming did not alter Fv’/Fm’ or gs in the remaining species (Table 2; Figure S6).

ΦPSII varied by day, temperature treatment, and species (day × treatment × species, Table

2). Cool- and warm-grown trees of EC reduced ΦPSII during the first 2 or 3 days of the heatwave,

but gradually increased ΦPSII thereafter (day 6 through 8) (Figure 3). Warm-grown trees of the

remaining species showed reduced ΦPSII throughout the heatwave, while cool-grown trees

showed an initial reduction then gradual recovery of ΦPSII as the heatwave progressed. Overall, warm-grown trees tended to maintain lower photosynthetic function, on average, compared to cool-grown trees.

Isoprene emissions

All species increased Is during the heatwave, but species with larger range sizes showed greater increases in Is (ET = average 18-fold increase, EC = average 20-fold increase) than species with restricted ranges (EB = average 10-fold increase, ES = average 14-fold increase; day × species interaction, Table 2), relative to pre-heatwave rates (Figure 5). On average, Is remained 2.7 to 3.8-fold higher in EC, ET, and ES during recovery compared to pre-heatwave. EB showed only

1.1-fold higher Is during recovery compared to pre-heatwave. These results support the hypothesis that species that show larger increases in Is show less photosynthetic inhibition and visible heatwave damage. There was no significant difference in Is between growth temperature treatments, and there were no interactions between treatments, measurement days, or species (Table 2).

Leaf respiration 20 Leaf respiration per unit area (at 20 °C, Rarea ) measured during the heatwave was significantly 20 lower (-28%) than Rarea measured before the heatwave (Table 2; Figure 6a-d). Species differed 20 20 in Rarea (Table 2; Figure 6a-d). ET showed significantly lower Rarea than EC (P=0.003) and 20 ES (P=0.002). Rarea did not differ between temperature treatments, and there were no interactions between days, treatments, and species. LMA averaged 62.6 ±1.4 g m-2 and was similar between treatments (P=0.44) and species (P=0.19).

On average, Tmax measured during the heatwave (55.0 ± 0.4 °C) was nearly 3 °C higher than Tmax measured before the heatwave (52.2 ± 0.4 °C, P<0.0001) (Table 2, Figure 6e-h, Figure

S7). Tmax did not differ between treatments or species (Table 2). Although a day × species interaction was not significant for Tmax (Table 2, P=0.14), species with smaller range sizes showed larger increases in Tmax (4 – 4.5 °C) measured during the peak heatwave than species with larger range sizes (1 – 2 °C, Figure 6e-h). This finding contrasts with our hypothesis that species that show larger increases in Tmax will show the least visible damage and physiological inhibition.

Proteomics Across all samples, total leaf protein concentrations averaged 7059 ± 215 mg m-2 (Figure S8). Proteins directly involved in photosynthesis made up 61% of total leaf protein. Proteins involved

This article is protected by copyright. All rights reserved 17 in metabolic processes and abiotic stress responses made up a smaller percentage (14%) of total leaf protein. Proteins not considered in this analysis (e.g. protein folding) made up 25% of total leaf protein (Figure S8). Concentrations of individual proteins within each of the twelve broad protein classes considered in this analysis are shown in Figure S9. On average, the following groups of proteins showed small but significant declines during the heatwave but recovered after the heatwave: photorespiration (mostly peroxisomal and glycine cleavage proteins), TCA cycle (e.g., ATP citrate lyase), carbohydrate metabolism, mitochondrial ATP synthesis, lipid metabolism, and hormone metabolism (Table 3, Figure 7, Figure S10). In contrast, heat shock, secondary metabolism (especially terpene synthesis), and glycolysis proteins increased during or after the heatwave (Table 3, Figure 7, Figure S10). The average 2.5x increase in total HSP concentrations during the heatwave and decline during recovery was particularly notable (Figure 7). Light reaction proteins were slightly lower during and after the heatwave than before the heatwave. Averaged across species and treatments, Calvin Cycle and redox proteins (except superoxide dismutase increased) showed no significant change over time (Table 3, Figure 7, Figure S10). Importantly, HSP and secondary metabolism protein concentrations both showed a significant day × species interaction (Table 3). Species with small range sizes showed a 3x increase in HSP concentrations during the heatwave whereas species with large range sizes showed a large but comparatively smaller (~2x) increase in HSP concentrations (Figure 8a). This finding contrasts with our hypothesis that species that show larger increases in HSPs will show the least damage and physiological inhibition. The day × species interaction for secondary metabolism proteins was driven mostly by the degree to which species increased terpene synthesis proteins during the heatwave (day × species, P<0.01). We found that EC, ET, and ES increased terpene synthesis proteins by 2.3x, 2.8x, and 3.5x, respectively, in response to the heatwave, while EB increased terpene proteins by only 1.3x (Figure 8b). Concentrations of photorespiration and glycolysis proteins differed among species, with EC showing significantly higher concentrations than the remaining species (Table 3, 4). All classes of leaf proteins tended to be lower in warm-grown trees than cool-grown trees (Table 3,4), although the effect of warming often depended upon day and species. Nonetheless, warming consistently reduced the concentration of proteins involved in photorespiration,

This article is protected by copyright. All rights reserved 18 secondary metabolism, and glycolysis (Table 3, 4). Light reaction proteins showed a species × treatment interaction (Table 3). Warming reduced light reaction proteins for ET and EB (-41% and -29%, respectively), but did not change light reaction protein concentrations in EC or ES. Calvin cycle, TCA cycle, redox, carbohydrate metabolism, mitochondrial ATP synthesis, and lipid and hormone metabolism proteins showed a day × treatment × species interaction (Table 3, Figure S11). For these protein classes, cool- and warm-grown trees of EC and ES showed relatively similar concentrations of most proteins through time, but warm-grown trees of ET and EB often showed significantly lower protein concentrations than cool-grown trees during and immediately following the heatwave (Figure S11). Species physiological and biochemical responses to the heatwave are summarized in Figure 9.

Discussion We grew four Eucalyptus species from similar climate zones under two temperature treatments and exposed them to a common experimental heatwave. We found that the two species with large range sizes showed less frequent and severe tissue damage than the two species with comparatively smaller range sizes. These results support the expectation that species with smaller range sizes may be more vulnerable to extreme climate events than species with large range sizes. Species differences in heatwave damage were associated with contrasting physiological and biochemical responses to high temperatures. The species’ that showed less visible damage tended to show smaller reductions in leaf gas exchange and chlorophyll fluorescence, larger increases in isoprene, and smaller increases in HSP and respiratory thermotolerance. These findings shed new light into factors influencing variation in tree species responses to heatwaves. Contrary to our hypothesis that warming would prime trees for heatwaves, we found that warmed trees had lower photosynthetic rates and lower protein concentrations than cool-grown seedlings and were more susceptible to heatwave damage than cool-grown seedlings independent of species range size. If this phenomenon is common, trees growing under warmer climates may be more vulnerable to heatwaves of the future.

Species range size and heatwave vulnerability Species are hypothesized to differ in their range sizes partly because they differ in their environmental tolerances (Brown, 1984; Slatyer et al., 2013); that is, the idea is that species with

This article is protected by copyright. All rights reserved 19 inherently broader environmental tolerances have larger range sizes than species with narrower environmental tolerances. Thus, shifting climates and more extreme climate events are expected to more negatively impact the performance and habitat suitability of species with small rather than large range sizes (González-Orozco et al., 2016; Pacifi et al., 2015; Thuiller et al. 2005). Two complementary (and not mutually exclusive) hypotheses are that (i) climate of seed origin and (ii) average climate of range (independent of climate of seed origin), also determine population responses to climate change (Oleksyn et al., 1998, Reich & Oleksyn, 2008, Reich et al., 2015). Thus, populations from similar climates and species with similar climate distributions should exhibit similar responses to climate warming and heatwaves. Using climatically similar provenances of four Eucalyptus species, we found that species with large range sizes showed less visible heatwave damage and greater physiological tolerance of heatwave conditions than species with smaller range sizes. Other studies have observed inconsistent associations between plant species range size and environmental tolerance (Hirst et al., 2017; Lacher & Schwartz, 2016), although none have examined heatwave tolerance. Separate from species range size, other aspects of species adaptation could also influence species tolerance of heatwaves. For example, the distribution of E. camaldulensis (EC) extends into drier and more open woodland habitats than the remaining species. Under these conditions, leaf cooling and heat stress avoidance are likely adaptive and may explain the exceptional heat tolerance of EC. Nonetheless, species range size may partly explain species differences in heatwave tolerance. Yet, accurately predicting individual tree species vulnerability to heatwaves requires understanding differences in traits governing physiological tolerance of high temperatures.

Physiological and biochemical factors associated with species differences in heatwave damage In general, we found that species differences in heatwave tissue damage paralleled their physiological and biochemical responses to the heatwave (Figure 9). In particular, E. camaldulensis (EC) showed no heatwave damage, had the highest average gs, showed the smallest reduction in gs during the heatwave, and Tleaf averaged >1 °C below Tair during the hottest part of the heatwave. In comparison, more than half of E. botryoides (EB) seedlings showed signs of heatwave damage, and the species showed the lowest average gs, the largest reduction in gs during the heatwave, and Tleaf was often more than 1 °C higher than Tair under peak heatwave conditions. This species also exhibited the least negative Ψmd, underlining its

This article is protected by copyright. All rights reserved 20 conservative use of water and associated lack of evaporative cooling. The two remaining species showed intermediate levels of damage, and similar average gs and Tleaf during the heatwave. Placing leaves inside chambers alters the light environment and boundary layer conductance, and thus Tleaf. Yet, thermal images taken during the heatwave supported the observed differences in Tleaf among species (Figure S5). Therefore, despite growing side-by-side with adequate water supply, species differed in stomatal regulation of Tleaf during the heatwave, causing species to experience ‘different’ heatwaves. Our results add to the growing number of studies showing that species may vary in their stomatal responses to high temperatures (Ameye et al., 2012; Drake et al., 2018; Fauset et al., 2018; Hammerlynck & Knapp 1996; Urban et al., 2017; Wujeska-Klause et al., 2015). Our results suggest that further work is required to identify patterns in stomatal responses to heatwave conditions across species, and associations with species performance during heatwaves. Species stomatal responses to heatwave conditions influenced the degree to which

species reduced Asat during the heatwave, but other mechanisms are also likely given that declines in gs were smaller than declines in Asat. Averaged across treatment groups and heatwave days, large-ranged EC showed little change in ΦPSII (relative to pre-heatwave) while small-

ranged EB showed the largest reduction in ΦPSII, indicating that electron transport to photochemistry was not substantially altered in EC, but was reduced in EB. Species-specific

changes in ΦPSII during the heatwave were strongly influenced by the leaf temperatures experienced by each species, which is mediated by species variation in gs and transpiration at

high air temperatures. For instance, during the peak heatwave, EC showed high gs and Tleaf

values below 45 °C, whereas EB showed much lower gs and Tleaf >45 °C (Figure 4). Importantly, temperatures >45 °C can significantly inhibit PSII capacity by increasing thylakoid membrane permeability resulting in proton leakage and reduced electron transport (Allakhverdiev et al., 2008; Duarte et al., 2016; Sharkey, 2005). These results suggest impairment of energy transfer and production at high temperatures, particularly in EB since that species was most negatively affected by the heatwave. Guha et al., (2018) identified similar relationships between heatwave

tissue damage, reductions in Asat, and inhibition of PSII among four northern temperate tree

species. We also found that the decline in PSII efficiency (Fv’/Fm’) was smallest in EC and largest in EB, indicating that EB was dissipating more excited energy to non-photochemical quenching or possibly photorespiration than EC. Indeed, it is well-known that high temperatures

This article is protected by copyright. All rights reserved 21 increase rates of photorespiration and non-photochemical quenching via the xanthophyll cycle (Demmig-Adams & Adams 2006; Gamon & Pearcy 1989), both of which reduce net C fixation. Calvin Cycle proteins concentrations showed a complex response to the heatwave that depended upon species and treatment. EC and ES trees (with large and small range sizes, respectively) of both treatments generally showed little change in Calvin Cycle proteins over time, while warmed trees of EB and ET showed reduced concentrations during and immediately after the heatwave. These responses could partly explain why reductions in Asat differed among species. However, short-term heat exposure is more likely to inhibit Rubisco activase and cause temporary deactivation of Rubisco than substantially reduce Calvin Cycle enzyme concentrations (Haldimann & Feller 2004; Hozain et al., 2010; Salvucci & Crafts-Brandner; 2004). Indeed, the temperatures reached in this experiment were likely high enough to cause deactivation of

Rubisco (Sharkey, 2005). We posit that species-specific reductions in Asat during the heatwave were likely explained by reduced PSII capacity and inhibition of Rubisco activase, as well as increased transfer of electrons to alternative sinks. Yet, species stomatal regulation of Tleaf via transpirational cooling was also a key determinant of species photosynthetic responses to high temperature. Stomatal responses to high temperatures affected photosynthesis in two ways: by regulating the physical diffusion of CO2 through the stomata, and by regulating the temperature at which photosynthetic processes were operating. Many eucalypt species emit isoprene (He, Murray & Lyons, 2000; Winters et al., 2009), and the observed differences in photosynthetic responses to heat stress in the four species may also be associated with the degree to which species increased isoprene emissions. Photosynthetic processes and isoprene production are biochemically linked (Logan, Monson & Potosnak, 2000; Loreto, Pinelli, Brancaleoni & Ciccioli, 2004; Sharkey, Wiberley & Donohue, 2008). There is evidence that isoprene emissions depend principally upon the balance between electron supply and demand for C fixation (Morfopoulos et al., 2013; Niinements, Tenhunen, Harley & Steinbrecher, 1999). Concurrently, isoprene is thought to stabilize thylakoid membranes during heat stress, thereby protecting against oxidative stress (Peñuelas, Llusià, Asensio & Munné- Bosch, 2005; Singaas et al., 1997; Siwko et al., 2007). In our study, the species that showed the largest increase in isoprene emissions (large-ranged EC) also showed the least damage and smallest reductions in Asat and ΦPSII, perhaps supporting the expected protective function of isoprene. We also observed that the species which showed the smallest increase in isoprene

(small-ranged EB) also showed the smallest increase in terpene synthesis proteins during the heatwave, representing a possible link between isoprene synthesis and emissions (Figure 8b). Previous studies have found that trees can acclimate to warmer temperatures and heatwaves by increasing the high temperature threshold for leaf photosynthetic and respiratory function (Drake et al., 2018; Knight & Ackerly, 2002; Zhu et al., 2018). We also tested whether species acclimate to heatwave conditions by increasing Tmax and reducing basal respiration rates. Assuming high temperature acclimation is adaptive, we expected stronger acclimation responses in species that experienced less damage. All species acclimated to heatwave conditions by 20 reducing base rates of leaf dark respiration (at 20 °C) and increasing Tmax. Lower rates of Rarea during the heatwave could be the result of substrate limitation caused by reduced in situ photosynthesis and increased respiration at warmer heatwave temperatures. Yet, in contrast to our expectations, species that experienced more heatwave damage and stress showed stronger acclimation responses (namely, increases in Tmax) than species that showed less heatwave damage and stress (Figure 9). We interpret our findings to mean that physiological acclimation of Tmax to heatwave conditions partly depends upon Tleaf, and the degree of stress. Thus, accurately predicting species acclimation responses to heatwaves may require measurements of

Tleaf during heatwaves. Measurements of leaf HSP concentrations provided additional insight into relationships between species heatwave tolerance, isoprene emissions, and physiological acclimation. Large and small molecular weight HSPs have been shown to facilitate acclimation to stressful conditions by protecting against oxidative damage and stabilizing cellular membranes (Heckathorn et al., 1998; Vierling, 1991). Therefore, we expected that species that were more tolerant of heatwave conditions would show larger increases in HSP concentrations than less tolerant species. However, we observed the opposite; HSP concentrations increased to a greater extent in species that experienced greater damage and physiological stress (e.g., EB), than the species that experienced the least damage and stress (e.g., EC). Moreover, the degree to which species increased HSPs was positively correlated with the degree to which species increased Tmax during the heatwave (r=0.99, P=0.01, Figure 9), supporting the idea that acquired thermotolerance involves the production of HSPs (Larkindale & Vierling, 2008). Across species, we also observed that HSP production and isoprene emissions during the heatwave were inversely related (Figure 9). Our interpretation is that maintenance of lower Tleaf and higher

This article is protected by copyright. All rights reserved 23 isoprene emissions during heatwave conditions results in less damage and stress, a reduced requirement for protection from HSPs, and weaker acclimation of Tmax (Fini et al., 2017; Siwko et al., 2007). Together, these data provide a integrative framework linking biochemical, physiological, and whole-tree responses to species capacity to tolerate heatwave conditions.

Trees growing in warmer climate are more vulnerable to heatwave damage Previous studies have indicated that acclimation to warmer temperatures or exposure to short- term heat events may improve tree performance during extended heat events (e.g., Colombo & Timmer 1992). We observed the opposite. When exposed to a common heatwave treatment, trees acclimated to warmer temperatures were more vulnerable to extreme heatwave conditions than trees acclimated to cooler temperatures. What might explain this result? One of the largest and clearest effects of warming was on leaf protein concentrations. In general, all classes of proteins were lower in warm-grown trees than in cool-grown trees (Table 4). Across days and species, warming consistently reduced photorespiration, secondary metabolism, and glycolysis protein concentrations. Lower Calvin Cycle and light reaction protein concentrations in warmed trees may partly reflect a reduced requirement for enzymes, as enzyme activity is greater at warmer temperatures. Lower protein concentrations may also explain why warm-grown trees showed reduced photosynthetic rates and lower chlorophyll fluorescence (Fv’/Fm’, ΦPSII) at a common measurement temperature, compared to cool-grown trees. Previous studies have found evidence of positive, neutral, and negative responses of photosynthetic capacity with increasing temperatures (Aspinwall et al., 2016; Kurepin et al., 2018; Way & Yamori, 2014), which may be associated with changes in leaf enzyme concentrations. Most importantly, lower concentrations of metabolic (i.e. respiratory) and stress avoidance (i.e. redox, superoxide dismutase) proteins may have hindered warm-grown trees ability to cope with heatwave conditions (Alscher, Erturk & Heath, 2002; Wang, Heckathorn, Mainali & Hamilton, 2008; Wang, Zhang, Goatley & Ervin, 2014). Thus, warmer growth temperatures might reduce leaf enzyme requirements, but may result in greater vulnerability to tissue damage during heatwaves. There are other possible explanations for greater heatwave damage in warm-grown trees, but we find little support for these explanations. For example, warmed trees generally had slightly lower gs than cool-grown trees (cool gs = 0.49 ± 0.03, warm gs = 0.43 ± 0.03), yet Tleaf

This article is protected by copyright. All rights reserved 24 measured during the heatwave (with the LI-6400XT) was comparable between cool- and warm- grown trees (cool Tleaf = 45.5 ± 0.1, warm Tleaf = 45.7 ± 0.1). Thus, growth at warmer temperatures did not substantially influence leaf cooling during the heatwave. In comparison, prior warming treatments of a larger magnitude or prior heat events have been shown to reduce gs during extreme heatwaves (Bauweraerts et al. 2015; Duarte et al., 2016). Prior growth temperature had no significant effect on isoprene emissions or physiological acclimation to the heatwave, indicating that these factors cannot explain greater heatwave damage in warm-grown trees. On average, warm-grown trees were larger (i.e. taller) than cool-grown trees, which could possibly make them more susceptible to hydraulic dysfunction during heatwaves, as has been found in some drought studies (e.g., Nepstad, Tohver, Ray, Moutinho & Cardinot, 2007). Yet, the well-watered cool- and warm-grown trees in our study showed similar Ψmd throughout the experiment, and Ψmd returned to pre-heatwave values during recovery. Moreover, if larger trees are typically more vulnerable to heatwave damage, then EC seedlings which were significantly larger than the other species should have experienced the most damage, but that did not occur. It is possible that tree size influenced heatwave vulnerability, but the mechanisms are unclear. Finally, growth temperature can influence leaf thickness, and thinner leaves are often more susceptible to high temperature damage, especially when light is high and wind speed is low (Curtis, Leigh & Rayburg, 2012; Leigh et al., 2012; Sastry & Barua 2017). Yet, leaves collected for our respiration measurements provided no evidence that warm- and cool-grown leaves differed in LMA. There may be other explanations for greater heatwave tissue damage in warm-grown trees, but our data indicate that reduced enzyme concentrations was likely a contributor to increased vulnerability to heatwave conditions in warm-grown trees. Our results provide new insight into mechanisms underlying variability in heatwave tolerance among tree species. They demonstrate that related co-occurring tree species may differ in their tolerance of heatwaves, which may be partly associated with species range size. We find evidence that species regulation of leaf temperature and isoprene emissions are both important determinants of species vulnerability to heatwave tissue damage. Broader studies, including more species, are required to test these linkages and identify common factors that are predictive of tree species vulnerability to heatwaves. This work is imperative for predicting climate change impacts on forest diversity and function, as well as land surface – atmosphere interactions.

Acknowledgements The authors have no conflicts of interest to declare. This research was supported by the Australian Research Council (Discovery, DP140103415), the Science Industry Endowment Fund (SIEF project code RP04–122), and Hawkesbury Institute for the Environment, Western Sydney University. We thank Danielle Creek for assistance with leaf water potential measurements.

Author contributions MJA co-led the experimental design, collected leaf gas-exchange data, coordinated the experimental implementation, and led the data analysis and writing. SP co-led the experimental design, collected leaf gas-exchange data, coordinated the experimental implementation, took thermal images of leaves during the heatwave to assess variability in leaf temperature, and contributed to analysis and writing. MGT was the senior scientific lead; he co-led the experimental design and made large contributions to analysis, interpretation, and writing. AV contributed to experimental design, assisted with experimental implementation, collected leaf gas-exchange data, and contributed to writing. MP led the collection and analysis of isoprene emission data and assisted with writing. JED, PBR, and DTT assisted with the conceptual framework of the study and assisted with writing. OKA provided additional gas-exchange chambers for respiratory thermotolerance measurements and assisted with writing. PDR collected samples for proteomics analysis and assisted with writing. SCVS and SD developed the proteomics platform and methods, assisted with analysis of leaf protein concentrations, and assisted with writing.

References Aitken S.N., Yeaman S., Holliday J.A., Wang T., Curtis-McLane S. (2008). Adaptation, migration or extirpation: climate change outcomes for tree populations. Evolutionary Applications, 1, 95-111.

Allakhverdiev S., Kreslavski V.D., Klimov V.V., Los D.A., Carpentier R., Mohanty P. (2008). Heat stress: an overview of molecular responses in photosynthesis. Photosynthesis Research, 98, 541-550.

Alscher R.G., Erturk N., Heath L.S. (2002). Role of superoxide dismutases (SODs) in controlling oxidative stress in plants. Journal of Experimental Botany, 53, 1331-1341.

Ameye M., Wertin T.M., Bauweraerts I., McGuire M.A., Teskey R.O., Steppe K. (2012). The effect of induced heat waves on Pinus taeda and Quercus rubra seedlings in ambient and elevated CO2 atmospheres. New Phytologist, 196, 448-461.

Aspinwall M.J., Drake J.E., Campany C., Vårhammar A., Ghannoum O., Tissue D.T., Reich P.B., Tjoelker M.G. (2016). Convergent acclimation of leaf photosynthesis and respiration to prevailing ambient temperatures under current and warmer climates in Eucalyptus tereticornis. New Phytologist, 212, 354-367.

Atkin O.K., Tjoelker M.G. (2003). Thermal acclimation and the dynamic response of plant respiration to temperature. Trends in Plant Science, 8, 343-351.

Backhaus S., Kreyling J., Grant K., Beierkuhnlein C., Walter J., Jentsch A. (2014). Recurrent mild drought events increase resistance toward extreme drought stress. Ecosystems, 17, 1068- 1081.

Bartholomé J., Mandrou E., Mabiala A., Jenkins J., Nabihoudine I., Klopp C., Schmutz J., Plomion C., Gion J-M. (2015). High-resolution genetic maps of Eucalyptus improve Eucalyptus grandis genome assembly. New Phytologist, 206, 1283–96. doi:10.1111/nph.13150.

Bauweraerts I., Ameye M., Wertin T.M., McGuire M.A., Teskey R.O., Steppe K. (2015).

Acclimation effects of heat waves and elevated [CO2] on gas exchange and chlorophyll fluorescence of northern red oak (Quercus rubra L.) seedlings. Plant Ecology, 215, 733-746.

Bayly M.J., Philippe R., Antanas S., Ladiges P.Y., Ades P.K., Anderson C., Bossinger G., Merchant A., Udovicic F., Woodrow I.E., Tibbits J. (2013). Chloroplast genome analysis of

Australian eucalypts--Eucalyptus, Corymbia, Angophora, Allosyncarpia and Stockwellia (Myrtaceae). Molecular Phylogenetics and Evolution 69, 704-716.

Brown J.H. (1984). On the relationship between abundance and distribution of species. American Naturalist, 124, 255-279.

Buckley L.B., Huey R.B. (2016). Temperature extremes: geographic patterns, recent changes, and implications for organismal vulnerabilities. Global Change Biology, 22, 3829-3842. von Caemmerer S., Quick W.P. (2000). Rubisco: Physiology in Vivo. In: Leegood R.C., Sharkey T.D., & von Caemmerer S. (eds) Photosynthesis. Advances in Photosynthesis and Respiration, vol 9. Springer, Dordrecht.

Colombo S.J., Timmer V.R. (1992). Limits of tolerance to high temperatures causing direct and indirect damage to black spruce. Tree Physiology, 11, 95-104.

Collins, M., Knutti R., Arblaster J., Dufresne J.-L., Fichefet T., Friedlingstein P., Gao X., Gutowski W.J., Johns T., Krinner G., Shongwe M., Tebaldi C., Weaver A.J., Wehner M. (2013). Long-term Climate Change: Projections, Commitments and Irreversibility. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker T.F., Qin D., Plattner G.-K., Tignor M., Allen S.K., Boschung J., Nauels A., Xia Y., Bex V., Midgley P.M. (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.

Coumou D., Robinson A. (2013). Historic and future increase in the global land area affected by monthly heat extremes. Environmental Research Letters, 034018, doi:10.1088/1748- 9326/8/3/034018.

Cowan T., Purich A, Perkins S., Pezza A., Boschat G., Sadler K. (2014) More frequent, longer, and hotter heat waves for Australia in the twenty-first century. Journal of Climate, 27, 5851– 5871.

Curtis E.M., Leigh A., Rayburg S. (2012). Relationships among leaf traits of Australian arid zone plants: alternative modes of thermal protection. Functional Plant Biology, 60, 471-483.

Daas C., Montpied P., Hanchi B., Dreyer E. (2008). Responses of photosynthesis to high temperatures in oak saplings assessed by chlorophyll-a fluorescence: inter-specific diversity and temperature-induced plasticity. Annals of Forest Science, 65, 305 doi: 10.1051/forest:2008002.

Demmig-Adams B., Adams W.W. (2006). Photoprotection in an ecological context: the remarkable complexity of thermal energy dissipation. New Phytologist, 172, 11-21.

Drake J.E., Tjoelker M.G., Vårhammar A., Medlyn B.E., Reich P.B., Leigh A., Pfautsch S., Blackman C.J., Lópex R., Aspinwall M.J., Crous K.Y., Duursma R.A., Kumarathunge D., De Kauwe M.G., Jiang M., Nicotra A.B., Tissue D.T., Choat B., Atkin O.K., Barton C.V.W. (2018). Trees tolerate an extreme heatwave via sustained transpirational cooling and increased leaf thermal tolerance. Global Change Biology, 24, 2390-2402.

Duarte A.G, Katata G., Hoshika Y., Hossain M., Kreuzwieser J., Arneth A., Ruehr N.K. (2016). Immediate and potential long-term effects of consecutive heat waves on the photosynthetic performance and water balance in Douglas-fir. Journal of Plant Physiology, 205, 57-66.

Edwards G.E., Baker N.R. (1993). Can CO2 assimilation in maize leaves be predicted accurately from chlorophyll fluorescence analysis? Photosynthesis Research, 37, 89-102.

Fauset S., Freitas H.C., Galbraith D.R., Sullivan M.J.P., Aidar M.P.M., Joly C.A., Phillips O.L., Vieira S.A., Gloor M.U. (2018). Differences in leaf thermoregulation and water use strategies between three co‐occurrin g Atlantic forest tree species. Plant, Cell and Environment doi: 10.1111/pce.13208.

Fensham R.J., Fraser J., MacDermott H.J., Firn J. (2015). Dominant tree species are at risk from exaggerated drought under climate change. Global Change Biology, 21, 3777-3785.

Fini A., Brunetti C., Loreto F., Centritto M., Ferrini F., Tattini M. (2017). Isoprene responses and functions in plants challenged by environmental pressures associated to climate change. Frontiers in Plant Science, 8, 1281, doi: 10.3389/fpls.2017.01281.

Furrer E.M., Katz R.W., Walter M.D., Furrer R. (2010). Statistical modelling of hot spells and heat waves. Climate Research, 43, 191-205.

Gamon J.A., Pearcy R.W. (1989). Leaf movement, stress avoidance and photosynthesis in Vitis californica. Oecologia, 79, 475-481.

Gauthier P.P.G., Crous K.Y., Ayub G., Duan H., Weerasinghe L.K., Ellsworth D.S., Tjoelker M.G., Evans J.R., Tissue D.T., Atkin O.K. (2014). Drought increases heat tolerance of leaf respiration in Eucalyptus globulus saplings grown under both ambient and elevated atmospheric

[CO2] and temperature. Journal of Experimental Botany, 65, 6471-6485.

Genty B., Briantais J-M., Baker N.R. (1989). The relationship between the quantum yield of photosynthetic electron transport and quenching of chlorophyll fluorescence. Biochimica et Biophysica Acta, 990, 87-92.

Ghouil H., Montpied P., Epron D., Ksontini M., Hanchi B., Dreyer E. (2003). Thermal optima of photosynthetic functions and thermostability of photochemistry in cork oak seedlings. Tree Physiology, 23, 1031-1039.

Gillet L.C., Navarro P., Tate S., Röst H., Selevsek N., Reiter L., Bonner R., Aebersold R. (2012). Targeted data extraction of the MS/MS spectra generated by data-independent acquisition: a new Concept for consistent and accurate proteome analysis. Molecular & Cellular Proteomics 11, O111.016717, doi:10.1074/mcp.O111.016717.

González-Orozco C.E., Pollock L.J., Thornhill A.H., Mishler B.D., Knerr N., Laffan S.W., Miller J.T., Rosauer D.F., Faith D.P., Nipperess D.A., Kujala H., Linke S., Butt N., Külheim C.,

Crisp M.D., Gruber B. (2016). Phylogenetic approaches reveal biodiversity threats under climate change. Nature Climate Change, 6, 1110-1114.

Guha A., Han J., Cummings C., McLennan D.A., Warren J.M. (2018). Differential ecophysiological responses and resilience to heat wave events in four co-occurring temperate tree species. Environmental Research Letters, 13, 065008.

Gunderson C.A., Norby R.J., Wullschleger S.D. (2000). Acclimation of photosynthesis and respiration to simulated climatic warming in northern and southern populations of Acer saccharum: laboratory and field evidence. Tree Physiology, 20, 87-96.

Gutschick V.P., BassiriRad H. (2003). Extreme events as shaping physiology, ecology, and evolution of plants: toward a unified definition and evaluation of their consequences. New Phytologist, 160, 21-42.

Haldimann P., Feller U. (2004). Inhibition of photosynthesis by high temperature in oak (Quercus pubescens L.) leaves grown under natural conditions closely correlates with a reversible heat dependent reduction of the activation state of ribulose-1,5-bisphosphate carboxylase/oxygenase. Plant, Cell and Environment, 27, 1169-1183.

Hammerlynck E., Knapp A.K. (1996). Photosynthetic and stomatal responses to high temperature and light in two oaks at the western limit of their range. Tree Physiology, 16, 557- 565.

Hamrick J.L., Godt M.J.W., Sherman-Broyles SL. (1992). Factors influencing levels of genetic diversity in woody plant species. New Forests, 6, 95-124.

He C., Murray F., Lyons T. (2000). Monoterpene and isoprene emissions from 15 Eucalyptus species in Australia. Atmospheric Environment, 34, 645-655.

Heckathorn S.A., Downs C.A., Sharkey T.D., Coleman J.S. (1998). The small, methionine-rich chloroplast heat-shock protein protects photosystem II electron transport during heat stress. Plant Physiology, 116, 439-444.

Hijmans R.J., Cameron S.E., Parra J.L., Jones P.G., Jarvis A. (2005). Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology, 25, 1965-1978.

Hirst M.J., Griffin P.C., Sexton J.P., Hoffmann A.A. (2017). Testing the niche-breadth-range- size hypothesis: habitat specialization vs. performance in Australian alpine daisies. Ecology, 98, 2708-2724.

Hozain M.I., Salvucci M.E., Fokar M., Holaday A.S. (2010). The differential response of photosynthesis to high temperature for a boreal and temperate Populus species relates to differences in Rubisco activation and Rubisco activase properties. Tree Physiology, 30, 32-44.

Knight C.A., Ackerly D.D. (2003). Evolution and plasticity of photosynthetic thermal tolerance, specific leaf area and leaf size: congeneric species from desert and coastal environments. New Phytologist, 160, 337-347.

Kurepin L.V., Stangl Z.R., Ivanov A.G., Bui V., Mema M., Hüner N.P.A., Öquist G., Way D., Hurry V. (2018). Contrasting acclimation abilities of two dominant boreal conifers to elevated

CO2 and temperature. Plant, Cell and Environment, doi: 10.1111/pce.13158.

Lacher I., Schwartz M.W. (2016). Empirical test on the relative climatic sensitivity between individuals of narrowly and broadly distributed species. Ecosphere, 7, e01227, doi: 10.1002/ecs2.1227.

Larkindale J., Vierling E. (2008). Core genome responses involved in acclimation to high temperature. Plant Physiology, 146, 748-761.

Law R.D., Crafts-Brandner S.J. (1999). Inhibition and acclimation of photosynthesis to heat stress is closely correlated with activation of ribulose-1,5-bisphosphate carboxylase/oxygenase. Plant Physiology, 120, 173-181.

Leigh A., Sevanto S., Ball M.C., Close J.D., Ellsworth D.S., Knight C.A., Nicotra A.B., Vogel S. (2012). Do thick leaves avoid thermal damage in critically low wind speeds? New Phytologist, 194, 477-487.

Leimu R., Mutikainen P., Koricheva J., Fischer M. (2006). How general are positive relationships between plant population size, fitness and genetic variation? Journal of Ecology, 94, 942-952.

Lin Y-S., Medlyn B.E., Ellsworth D.S. (2012). Temperature responses of leaf net photosynthesis: the role of component processes. Tree Physiology, 32, 219-231.

Logan B., Monson R.K., Potosnak M.J. (2000). Biochemistry and physiology of foliar isoprene production. Trends in Plant Science 5, 477-481.

Lohse M., Nagel A., Herter T., May P., Schroda M., Zrenner R., Tohge T., Fernie A.R., Stitt M., Usadel B. (2014). Mercator: a fast and simple web server for genome scale functional annotation of plant sequence data. Plant, Cell & Environment 37, 1250–58. doi:10.1111/pce.12231.

Loreto F., Pinelli P., Brancaleoni E., Ciccioli P. (2004). 13C labelling reveals chloroplastic and extrachloroplastic pools of dimethylallyl pryrophosphate and their contribution to isoprene formation. Plant Physiology, 135, 1903-1907.

Maxwell K., & Johnson G.N. (2000). Chlorophyll fluorescence – a practical guide. Journal of Experimental Botany, 51, 659-668.

Meehl G.A., Tebaldi C. (2004). More intense, more frequent, and longer lasting heat waves in the 21st century. Science, 305, 994-997.

Mittler R. (2002). Oxidative stress, antioxidants and stress tolerance. Trends in Plant Science, 7, 405-410.

Monson R.K., Jaeger C.H., Adams W.W., Driggers E.M., Silver G.M., Fall R. (1992). Relationships among isoprene emission rate, photosynthesis, and isoprene synthase activity as influenced by temperature. Plant Physiology, 98, 1175-1180.

Morfopoulos C., Prentice I.C., Keenan T.F., Friedlingstein P., Medlyn B.E., Peñuelas J., Possell M. (2013). A unifying conceptual model for the environmental responses of isoprene emissions from plants. Annals of Botany, 112, 1223-1238.

Morin X., Thuiller W. (2009). Comparing niche- and process-based models to reduce prediction uncertainty in species range shifts under climate change. Ecology, 90, 1301-1313.

Nepstad D.C., Tohver I.M., Ray D., Moutinho P., Cardinot G. (2007). Mortality of large trees and lianas following experimental drought in an amazon forest. Ecology, 88, 2259-2269.

Niinemets Ü., Tenhunen J.E., Harley P.C., Steinbrecher R. (1999). A model of isoprene emission based on energetic requirements for isoprene synthesis and leaf photosynthetic properties for Liquidambar and Quercus. Plant, Cell and Environment, 22, 1319-1335.

Niinemets Ü. (2010). Responses of forest trees to single and multiple environmental stresses from seedlings to mature plants: Past stress history, stress interactions, tolerance and acclimation. Forest Ecology and Management, 260, 1623-1639.

O’Sullivan O.S., Weerasinghe L.K., Evans J.R., Egerton J.J.G., Tjoelker M.G., Atkin O.K. (2013). High-resolution temperature responses of leaf respiration in snow gum (Eucalyptus pauciflora) reveal high-temperature limits to respiratory function. Plant, Cell and Environment, 36, 1268-1284.

O’Sullivan O.S., Heskel M.A., Reich P.B., Tjoelker M.G., Weerasinghe L.K., Penillard A., Zhu L., Egerton J.J.G., Bloomfield K.J., Creek D., Bahar N.H.A., Griffin K.L., Hurry V., Meir P., Turnbull M.H., Atkin O.K. (2017). Thermal limits of leaf metabolism across biomes. Global Change Biology, 23, 209-223.

Oxborough K., Baker N.R. (1997). Resolving chlorophyll a fluorescence images of photosynthetic efficiency into photochemical and non-photochemical components – calculation of qP and Fv’/Fm’ without measuring Fo’. Photosynthesis Research, 54, 135-142.

Pacifici M., Foden W.B., Visconti P., Watson J.E.M., Butchart S.H.M., Kovacs K.M., Scheffers B.R., Hole D.G., Martin T.G., Akçakaya A.R., Corlett R.T., Huntley B., Bickford D., Carr J.A., Hoffmann A.A., Midgley G.F., Pearce-Kelly P., Pearson R.G., Williams S.E., Willis S.G., Young B., Rondinini C. (2015). Assessing species vulnerability to climate change. Nature Climate Change, 5, 215-225.

Parmesan C. (2006). Ecological and evolutionary response to recent climate change. Annual Review in Ecology, Evolution and Systematics, 37, 637-69.

Pearson R.G., Dawson T.P. (2013). Predicting the impacts of climate change on the distribution of species: are bioclimate envelope models useful? Global Ecology & Biogeography, 12, 361- 371.

Peñuelas J., Llusià J. (2002). Linking photorespiration, monoterpenes and thermotolerance in Quercus. New Phytologist, 155, 227-237.

Peñuelas J., Llusià J., Asensio D., Munné-Bosch S. (2005). Linking isoprene with plant thermotolerance, antioxidants and monoterpene emissions. Plant, Cell and Environment, 28, 278-286.

Perkins S.E., Alexander L.V. (2013). On the measurements of heat waves. Journal of Climate, 26, 4500-4517.

Pollastri S., Tsonev T., Loreto F. (2014). Isoprene improves photochemical efficiency and enhances heat dissipation in plants at physiological temperatures. Journal of Experimental Botany, 65, 1565-1570.

Reyer C.P.O., Leuzinger S., Rammig A., Wolf A., Bartholomeus R.P., Bonfante A., de Lorenzi F., Dury M., Gloning P., Aboujaodé R., Klein T., Kuster T.M., Martins M., Niedrist G., Riccardi M., Wohlfahrt M., de Angelis P., de Dato G., François L., Menzel A., Pereira M. (2013). A plant’s perspective of extremes: terrestrial plant responses to changing climatic variability. Global Change Biology, 19, 75-89.

Salvucci M.E., Crafts-Brandner S.J. (2004). Inhibition of photosynthesis by heat stress: the activation state of Rubisco as a limiting factor in photosynthesis. Physiologia Plantarum, 120, 179-186.

Sastry A., Barua D. (2017). Leaf thermotolerance in tropical trees from a seasonally dry climate varies along the slow-fast resource acquisition spectrum. Scientific Reports, 7, 11246, doi: 10.1038/s41598-017-11343-5.

Seneviratne S.I., Donat M.G., Mueller B., Alexander L.V. (2014). No pause in the increase of hot temperature extremes. Nature Climate Change, 4, 161-163.

Sharkey T.D. (2005). Effects of moderate heat stress on photosynthesis: importance of thylakoid reactions, rubisco deactivation, reactive oxygen species, and thermotolerance provided by isoprene. Plant, Cell and Environment, 28, 269-277.

Sharkey T.D., Wiberley A.E., Donohue A.R. (2008). Isoprene emissions from plants: why and how. Annals of Botany, 101, 5-18.

Singsaas E.L., Lerdau M., Winter K., Sharkey T.D. (1997). Isoprene increases thermotolerance of isoprene-emitting species. Plant Physiology, 115, 1413-1420.

Siwko M.E., Marrink S.J., de Vries A.H., Kozubek A., Anton J.M., Uiterkamp S., Mark A.E. (2007). Does isoprene protect plant membranes from thermal shock? A molecular dynamics study. Biochimica et Biophysica Acta, 1768, 198-206.

Slayter R.A., Hirst M., Sexton J.P. (2013). Niche breadth predicts geographical range size: a general ecological pattern. Ecology Letters, 16, 1104-1114.

Slot M., Kitajima K. (2015). General patterns of acclimation of leaf respiration to elevated temperatures across biomes and plant types. Oecologia, 177, 885-900.

Teskey R., Wertin T., Bauweraerts I., Ameye M., McGuire M.A., Steppe K. (2015). Responses of tree species to heat waves and extreme heat events. Plant, Cell and Environment, 38, 1699- 1712.

Thuiller W., Lavorel S., Araújo M.B. (2005). Niche properties and geographical extent as predictors of species sensitivity to climate change. Global Ecology and Biogeography, 14, 347- 357.

Urban J., Ingwers M.W., McGuire M.A., Teskey R.O. (2017). Increase in leaf temperature opens stomata and decouples net photosynthesis from stomatal conductance in Pinus taeda and Populus deltoides x nigra. Journal of Experimental Botany, 68, 1757-1767.

Velikova V., Várkonyi Z., Szabó M., Maslenkova L., Nogues I., Kovács L., Peeva V., Busheva M., Garab G., Sharkey T.D., Loreto F. (2011). Increased thermostability of thylakoid membranes in isoprene-emitting leaves probed with three biophysical techniques. Plant Physiology, 157, 905-916.

Vickers C.E., Gershenzon J., Lerdau M.T., Loreto F. (2009). A unified mechanism of action for volatile isoprenoids in plant abiotic stress. Nature Chemical Biology, 5, 283-291.

Vierling E. (1991). The roles of heat shock proteins in plants. Annual Review of Plant Physiology and Plant Molecular Biology, 42, 579-620.

Wang D., Heckathorn S.A., Mainali K., Hamilton E.W. (2008). Effects of N on plant response to heat-wave: a field study with prairie vegetation. Journal of Integrative Plant Biology, 50, 1416- 1425.

Wang X., Cai J., Liu F., Dai T., Cao W., Wollenweber B., Jiang D. (2014). Multiple heat priming enhances thermo-tolerance to a later high temperature stress via improving subcellular antioxidant activities in wheat seedlings. Plant Physiology and Biochemistry, 74, 185-192.

Way D.A., Yamori W. (2014). Thermal acclimation of photosynthesis: on the importance of adjusting our definitions and accounting for thermal acclimation of respiration. Photosynthesis Research, 119, 89-100.

Winters A.J., Adams M.A., Bleby T.M., Rennenberg H., Steigner D., Steinbrecher R., Kreuzwieser J. (2009). Emissions of isoprene, monoterpene and short-chained carbonyl compounds from Eucalyptus spp. in southern Australia. Atmospheric Environment, 43, 3035- 3043.

Wise R.R., Olson A.J., Schrader S.M. Sharkey T.D. (2004). Electron transport is the functional limitation of photosynthesis in field-grown Pima cotton plants at high temperature. Plant, Cell and Environment, 27, 717-724.

Wujeska-Klause A., Bossinger G., Tausz M. (2015). Responses to heatwaves of gas exchange, chlorophyll fluorescence and antioxidants ascorbic acid and glutathione in congeneric pairs of Acacia and Eucalyptus species from relatively cooler and warmer climates. Trees, 29, 1929- 1941.

Zhu L., Bloomfield K.J., Hocart C.H., Egerton J.J.G., O’Sullivan O.S., Penillard A., Weerasinghe L.K., Atkin O.K. (2018). Plasticity of photosynthetic heat tolerance in plants adapted to thermally contrasting biomes. Plant, Cell and Environment, doi: 10.1111/pce.13133.

Figure captions

Figure 1. Conceptual figure showing hypothesized linkages between leaf-scale, biochemical, and whole-tree responses to heatwaves. Rising air temperatures during a heatwave and associated increases in vapor pressure deficit of the atmosphere (VPD) directly affect leaf temperature (Tleaf) as well as stomatal (gs) regulation of Tleaf. Changes (Δ) in Tleaf directly influence photosynthesis and isoprene emissions. Increasing Tleaf can also lead to physiological inhibition (stress), tissue damage and organ failure. However, increased temperatures may trigger increased production of heat shock proteins (HSPs) and increased leaf respiratory thermotolerance (Δ Tmax) which is expected to improve leaf and whole-plant function during heat stress. The timeframe of these processes is represented by colors (green: seconds-minutes; orange: hours to days; red: several days or plant mortality). We expect that species that maintain higher gs, lower Tleaf, and show larger increases in isoprene emissions, HSPs, and leaf respiratory thermotolerance will maintain higher photosynthetic rates and display less tissue damage.

Figure 2. Percentage of trees of each species (EC = Eucalyptus camaldulensis (n=11), ET = E. tereticornis (n=12), EB = E. botryoides (n=11), ES = E. smithii (n=12)) and temperature treatment (Cool (n=24), Warm (n=22)) showing different levels of heatwave-induced damage on a scale of 0 - 2; 0 indicating little or no visible browning or mortality of leaves and branch/stem tips, 1 indicating clearly visible but minor browning or mortality of leaves and branch/stem tips, and 2 indicating heavy to severe browning and mortality of leaves, branch tips and stem tips.

Figure 3. Mean (± standard error, n=6) values of leaf-level light-saturated net photosynthesis

(Asat), stomatal conductance to water vapour (gs), midday leaf water potential (Ψmd), efficiency of PSII in light-adapted leaves (Fv’/Fm’) and the quantum yield of PSII (ΦPSII) of four Eucalyptus species (EC = Eucalyptus camaldulensis, ET = E. tereticornis, EB = E. botryoides, ES = E. smithii) grown under two growth temperatures (cool, warm) and measured at prevailing temperatures before, during and after an experimental heatwave. Lightly shaded areas represent the ‘ramp-up’ phase of the heatwave and darker shaded areas represent the ‘extreme’ temperature phase of the heatwave. See Fig. S2 for prevailing air temperatures each day.

Figure 4. (a) The relationship between stomatal conductance to water vapor (gs) and the leaf temperature (Tleaf) – air temperature (Tair) differential in four Eucalyptus species (EC = Eucalyptus camaldulensis, ET = E. tereticornis, EB = E. botryoides, ES = E. smithii) across five days of an experimental heatwave (days 4-8 of experiment). Higher gs was associated with Tleaf below Tair, while lower gs was associated with Tleaf above Tair during the heatwave. (b) Box-plots of the leaf temperature (Tleaf) – air temperature (Tair) differential in four Eucalyptus species (abbreviated as in (a)) across five days of an experimental heatwave. Species differed strongly in

Tleaf – Tair (P<0.0001). Different lower-case letters under box-plots indicate significant inter- specific differences in Tleaf – Tair. Error bars represent 95% confidence intervals, dotted lines within each box-plot represent the mean response ratio, and solid lines represent the median response ratio. Response ratios falling outside the 95% confidence intervals are shown as filled circles. Tleaf and Tair were both measured with the LI-6400XT system.

Figure 5. Mean (± standard error, n = 3) isoprene emission rates (Is) of four Eucalyptus species (EC = Eucalyptus camaldulensis, ET = E. tereticornis, EB = E. botryoides, ES = E. smithii) grown under two temperature treatments (Cool, Warm) measured at ambient midday temperatures before (3 days pre-heatwave, 20 °C), during (day 6, 45 °C), and 5 days after (i.e., recovery, 20 °C) an experimental heatwave. There was a significant day × species interaction (Table 2). There was no significant effect of growth temperature, and growth temperature did not interact with species or experimental stage.

Figure 6. (a-d) Mean (± standard error) rates of leaf dark respiration per unit area measured at 20 20 °C (Rarea ), and (e-h) mean (± standard error) leaf temperatures at which leaf respiration reaches

its maximum rate (Tmax), i.e. the temperature at which damage to leaf function is irreversible, measured prior to the heatwave (pre-heatwave, blue bars) and during peak heatwave conditions (day 6 and 7, red bars) on four Eucalyptus species (EC = Eucalyptus camaldulensis, ET = E. tereticornis, EB = E. botryoides, ES = E. smithii). There was no significant effect of growth temperature, and growth temperature did not interact with species or experimental stage (Table 2).

Figure 7. Box-plots of response ratios showing how different groups of proteins responded to an experimental heatwave. The statistical effect of ‘day of experiment’ is given in Table 3. Error bars represent 95% confidence intervals and solid lines within each box-plot represent the median response ratio. Response ratios falling outside the 95% confidence intervals are shown as open circles. Response ratios were calculated by summing the protein concentrations for each protein class for each tree on day 1 (pre-heatwave), then dividing the summed concentrations for each tree on each subsequent measurement day by the pre-heatwave concentration (e.g. Heatwave response ratio = day 7 heatwave concentration / pre-heatwave concentration; 1 day post-heatwave response ratio = 1 day post-heatwave concentration / pre-heatwave concentration, etc.). Thus, the individual responses of each tree (n=24), from all species and both treatments, are captured in each box-plot.

Figure 8. Mean (± standard error, n=6) total concentrations of leaf heat shock proteins and terpene synthesis proteins in four Eucalyptus species (EC = Eucalyptus camaldulensis, ET = Eucalyptus tereticornis, EB = Eucalyptus botryoides, ES = Eucalyptus smithii) exposed to an experimental heatwave. Lightly shaded areas represent the ‘ramp-up’ phase of the heatwave and darker shaded areas represent the ‘extreme’ temperature phase of the heatwave (see Fig. S2).

Figure 9. (a) Mean (± standard error) response ratios summarizing the physiological response of four Eucalyptus species (EC = Eucalyptus camaldulensis, ET = Eucalyptus tereticornis, EB = Eucalyptus botryoides, ES = Eucalyptus smithii) to an experimental heatwave. Mean response ratios were calculated by averaging the ratio of values measured on each heatwave day (days 4 –

8) to values on day 1 (pre-heatwave). Note: * the response ratio for isoprene (Is) is natural-log- transformed so that data fit on the same axis as the other response variables. Thus, if for example the response ratio of Is is 3, the actual response is a 20x or 1000%. Panel (b) shows the mean (± standard error) absolute temperature difference/change of two temperature variables; ‘heatwave

Tleaf – Tair’ is the average temperature difference between Tleaf and Tair across all five heatwave days, and Tmax is the change in the high temperature threshold of leaf dark respiration during the heatwave (day 6-7) relative to the pre-heatwave measure of Tmax.

Table 1. Geographic and climatic information for the species and provenances included in this study.

Eucalyptus Eucalyptus Eucalyptus Eucalyptus camaldulnesis tereticornis botryoides smithii Species range size (km2)a 6,040,600 792,575 74,175 95,750 Species latitudinal range (°S)b 12-38 9-38 32-39 34-38 Species mean annual temperature range 11-29 10-26 10,18 10,16 (minimum – maximum) (°C)c Species 2.5% percentile / median / 97.5% 13/17/27 13/17/22 14/16/18 11/14/16 percentile for mean annual temperature (°C)c Species 2.5% percentile / median / 97.5% 233/562/970 662/1010/1695 729/1118/1476 868/949/1565 percentile for mean annual precipitation (mm)c

Geographic and climatic origin of provenances in this study Latitude (°S) 36.36 35.4 35.02 36.1 Longitude (°E) 146.47 150.07 150.27 150.04 MAT (°C) 14.5 13.2 13.3 15.8 MAP (mm) 693 948 1153 1030

Mean daily Tair during summer (°C) 21.2 18.4 18.5 20.1

Mean max daily Tair during summer (°C) 29.6 25.3 25.1 24.7

Mean daily Tair of driest quarter (°C) 21.1 8.4 8.5 11.8 Precipitation of driest month (mm) 38 58 68 50 Precipitation of warmest quarter (mm) 131 254 322 332 aSpecies range size in 2014 (from González-Orozco et al., 2016) bAtlas of Living Australia (ALA.org), excluding records > 50 years old, and coordinate uncertainty > 50 km cMean annual temperature and mean annual precipitation data from WorldClim.org (Hijmans et al., 2005) matched to species occurrence data (latitude and longitude) from ALA.org

Table 2. Analysis of variance of day (i.e., day of experiment, D), growth temperature treatment (T), and species (S) effects on leaf-level physiological traits. Numerator and denominator degrees of freedom (df) are similar for all traits except isoprene which is reported separately. F- values denoted with ‘***’, ‘**’, and ‘*’, are significant at P<0.001, P<0.01, and P≤0.05, respectively.

20 df Asat g s Ψmd Fv'/Fm'' ΦPSII df Is df Rarea Tmax Day (D) 9,336 65.5*** 21.2*** 38.6*** 96.9*** 20.0*** 2,40 322.4*** 1,35 67.3*** 25.1*** Treatment (T) 1,38 21.1*** 5.2* 0.6 11.8** 40.0*** 1,24 4.0 1,38 2.5 0.1 D × T 9,336 0.9 1.6 0.7 1.4 1.4 2,40 1.2 1,35 2.1 0.2 Species (S) 3,38 72.2*** 24.3*** 39.7*** 58.8*** 56.4*** 3,24 3.5* 3,38 9.0** 1.6 D × S 27,336 1.9** 1.8** 2.0** 2.2** 1.2 6,40 3.2* 3,35 2.7 2.0 T × S 3,38 1.7 4.0* 0.7 3.1* 1.3 3,24 0.8 3,38 0.3 2.2 D × T × S 27,336 1 0.7 1 1.4 1.9** 6,40 0.9 3,35 1.6 0.1

Variable descriptions: Asat, light-saturated net photosynthetic rate; gs, stomatal conductance to

water vapour; Ψmd, midday leaf water potential; Fv′/Fm′, quantum efficiency of PSII; ΦPSII, 20 quantum yield of photosystem II; Is, isoprene emission rate; Rarea , rate of leaf dark respiration

per unit area measured at 20 °C; Tmax, high temperature threshold for leaf dark respiration. Note:

gs, isoprene, and ΦAuthor PSII were log-transformed to fulfil assumptions of normality.

Table 3. Analysis of variance results (F-values) indicating the effects of day (D, day of experiment), prior growth temperature treatment (T), and species (S) on broad classes of leaf proteins. F-values denoted with ‘***’, ‘**’, and ‘*’, are significant at P<0.001, P<0.01, and P≤0.05, respectively. Protein class Day (D) Species (S) D × S Treatment (T) D × T S × T D × S × T df 3,48 3,16 9,48 1,16 3,48 3,16 9,48

Calvin Cycle 1.0 25.2*** 2.0 20.8** 0.7 2.6 2.7* Light Reactions 7.8** 61.5*** 2.0 34.7*** 0.2 6.2** 1.7

Photorespiration 3.4* 27.8*** 1.5 22.4** 0.1 2.1 1.6 Heat shock proteins 275.3*** 7.7** 3.8** 0.6 2.6 1.6 0.8 TCA cycle 12.0*** 7.1** 2.5* 21.0** 1.3 0.7 2.5* Redox 1.7 22.9*** 2.3* 19.4** 0.4 1.2 2.4* Carbohydrate metabolism 30.8*** 42.5*** 4.5** 17.3** 1.1 2.5 2.6* Secondary metabolism 15.4*** 13.0** 2.4* 10.7** 2.1 1.2 0.8 Glycolysis 3.9* 6.2** 1.5 13.1** 1.6 0.4 1.4 Mitochondrial ATP synthesis 6.8** 13.5** 2.7* 28.9*** 2.8* 2.3 2.6* Lipid metabolism 6.3** 2.6 5.3*** 15.83** 3.8* 0.8 3.2** Hormone metabolism 51.5*** 6.7** 3.4** 9.4** 2.91* 0.3 3.0**

Table 4. Total mean (± standard error) concentrations of different leaf protein classes averaged over four collection days for trees of four Eucalyptus species (EC = Eucalyptus camaldulensis, ET = E. tereticornis, EB = E. botryoides, ES = E. smithii) grown under ‘cool’ (18 °C) or ‘warm’ (21.5 °C) daily average temperatures. Note: significant day × treatment, day × species, or day × treatment × species interactions exist for many of the protein classes (see Table 3). Differences in protein concentrations between treatments and among species should be interpreted with caution. Treatment Species Protein class Cool Warm EC ET EB ES (units = mg m-2) (n=48) ( n=48) ( n=24) ( n=24) ( n=24) ( n=24) Calvin Cycle 2343.5 ± 110.8 1933.5 ± 103.7 3115.6 ± 125.1 1747.1 ± 83.8 1781.3 ± 126.5 1910.1 ± 82.1 Light Reactions 2111.6 ± 91.7 1738.0 ± 112.1 2939.2 ± 93.3 1518.2 ± 95.3 1501.8 ± 101.0 1740.1 ± 51.8 Photorespiration 283.0 ± 14.2 234.2 ± 12.9 391.3 ± 14.4 219.2 ± 10.5 213.2 ± 15.4 210.8 ± 5.8 Heat shock proteins 248.0 ± 17.5 228.3 ± 14.4 215.8 ± 12.2 199.6 ± 17.2 252.2 ± 29.4 285.1 ± 24.8 TCA cycle 178.2 ± 7.3 150.5 ± 9.8 239.6 ± 10.6 134.8 ± 9.1 132.6 ± 9.4 150.4 ± 4.5 Redox 171.3 ± 5.7 144.4 ± 6.9 217.5 ± 6.2 130.0 ± 6.1 133.6 ± 7.0 150.4 ± 4.3 Carbohydrate metabolism 152.3 ± 5.5 130.0 ± 5.9 190.7 ± 5.2 121.2 ± 5.7 115.5 ± 7.7 137.2 ± 4.0 Secondary metabolism 100.8 ± 4.8 82.4 ± 3.5 99.3 ± 4.4 73.8 ± 4.3 75.0 ± 3.3 118.2 ± 7.0 Glycolysis 75.1 ± 2.3 62.9 ± 1.9 84.4 ± 2.4 60.6 ± 2.3 60.0 ± 2.7 71.0 ± 2.7 Mitochondrial ATP synthesis 64.7 ± 1.5 53.7 ± 1.7 70.2 ± 1.4 53.1 ± 2.2 54.4 ± 3.0 58.9 ± 1.8 Lipid metabolism 53.4 ± 1.9 41.2 ± 1.8 54.1 ± 1.7 47.0 ± 4.0 45.4 ± 3.1 42.8 ± 1.9 Hormone metabolism 25.7 ± 1.1 19.8 ± 0.7 25.2 ± 0.8 17.9± 1.0 25.3 ± 2.0 22.5 ± 1.0