This is the accepted version of the journal article: Huang, Jianbei; Hartmann, Henrik; Ogaya Inurrigarro, Romà; [et al.]. «Hormone and carbohydrate regulation of defense secondary metabolites in a Mediterranean forest during drought». Environmental and Experimental Botany, Vol. 209 (May 2023), art. 105298. DOI 10.1016/j.envexpbot.2023.105298 This version is available at https://ddd.uab.cat/record/284320 under the terms of the license 1 1 Hormone and carbohydrate regulation of defense secondary metabolites in a 2 Mediterranean forest during drought 3 4 Authors: 5 Jianbei Huang 1, Henrik Hartmann 1, Romà Ogaya 2,3 , Ingo Schöning 1, Michael Reichelt 4, 6 Jonathan Gershenzon 4, and Josep Peñuelas 2,3 7 Affiliations: 1 8 Max Planck Institute for Biogeochemistry, Hans-Knöll-Str. 10, 07745 Jena, Germany 2 9 CSIC, Global Ecology Unit CREAF-CSICUAB, Bellaterra, 08193 Catalonia, Spain 3 10 CREAF, Cerdanyola del Vallès, 08193 Catalonia, Spain 4 11 Max Planck Institute for Chemical Ecology, Hans-Knöll-Str. 8, 07745 Jena, Germany 12 13 *Correspondence: Dr. Jianbei Huang 14 Email: hjianbei@bgc-jena.mpg.de 15 Tel: +49 3641 576368 16 Postal address: Hans-Knöll-Str. 10, 077 45, Jena, Germany 17 18 Tables and figures in main body: seven figures, all should be published in colour 19 Supplemental file: four figures and four tables 20 21 Declarations of interest: none 22 2 23 Abstract 24 Plants have evolved complex physiological mechanisms to regulate production of defense 25 secondary metabolites (SMs) to cope with environmental stress such as drought. Yet, these 26 mechanisms remain under-studied in mature trees, limiting our ability to understand effects of 27 climate change on tree productivity and survival. We investigated defense SMs and their 28 relations to growth, nonstructural carbohydrates and phytohormones across seasons, in a 29 Mediterranean forest dominated by two tree species with different resistance to drought. Our 30 results showed seasonality of leaf SMs, with a strong accumulation of SMs during cold and dry 31 conditions in winter. The drought-resistant species Phillyrea latifolia had high levels of SMs in 32 both leaves (ca. 10-20%) and stem phloem (4-6%), but experimental drought decreased SMs and 33 soluble sugars in stem phloem in spring and summer. By contrast, the drought-susceptible 34 species Quercus ilex had relatively low and constant levels of SMs, irrespective of drought. We 35 further showed that P. latifolia leaves had generally higher levels of jasmonic acid and lower 36 salicylic acid than Q. ilex leaves, potentially leading to differences in SMs and growth between 37 the two co-occucrring species. Our study suggests that species- and organ-specific dynamics of 38 defense SMs are driven by phytohormones and carbohydrates in mature trees, and that P. 39 latifolia may become less resistant to abiotic and biotic stress due to reduced phloem defense 40 SMs and carbohydrates in a drying climate. 41 Keywords: carbon allocation; drought; growth-defense trade-offs; jasmonic acid; nonstructural 42 carbohydrates; phytohormones; salicylic acid; secondary metabolites; tree mortality and 43 community shift 44 3 45 Introduction 46 Climate change is expected to increase the frequency and intensity of drought, heat waves, and 47 fires that are likely to threaten forest ecosystem services, including disruption of biogeochemical 48 cycles and loss of biodiversity (Anderegg et al., 2020; McDowell et al., 2020). Of particular 49 concern are drought-induced tree mortality and associated decline in forest productivity, 50 particularly in dry regions of the US (Trugman et al., 2020), Asia (Liu et al., 2013), and Europe 51 (Peñuelas et al., 2018; Senf et al., 2020). Drought may reduce the ability of trees to defend 52 against abiotic and biotic stresses (McDowell et al., 2011). In this context, considerable research 53 has focused on drought impacts on carbohydrate metabolism and in particular nonstructural 54 carbohydrates (NSCs) (Hesse et al., 2021; Jin et al., 2018; Pivovaroff et al., 2021; Rowland et al., 55 2015; Schönbeck et al., 2018; Zhang et al., 2020). In contrast, very few field studies have 56 investigated impacts of drought on production of secondary metabolites (SMs; Trowbridge et al., 57 2021), which are a crucial component of plant antioxidant (Agati et al., 2020) and antiherbivore 58 defense (Huang et al., 2020a). 59 Production of defense SMs under drought is thought to be regulated by carbohydrate availability. 60 Under moderate drought conditions, carbon supply to trees exceeds sink demand, because 61 drought leads to more rapid and sharper decreases in growth than in leaf photosynthesis (sink 62 limitation; Muller et al., 2011; Palacio et al., 2014). This sink limitation may lead to an increase 63 in carbohydrate availability for production of SMs. However, as drought progresses, 64 photosynthesis continues to decrease and cannot meet carbohydrate demands for maintenance 65 metabolism, leading to source limitation for production of defense SMs (Herms and Mattson, 66 1992; McDowell et al., 2011). This may be particularly important in the Mediterranean where 67 stomata frequently close to reduce transpiration but also limit photosynthetic activity. The extent 68 to which drought affects NSCs and their utilization for defense SMs may vary depending on the 69 season and organ types. Drought-induced sink limitation to growth and increases in NSCs and 70 SMs may occur mainly in the growing season. However, severe drought in summer may lead to 71 source limitation and high carbohydrate demands for maintenance (e.g., osmoregulation; 72 Blumstein et al., 2023; Hajek et al., 2022), thereby reducing the availability of carbohydrates for 73 production of SMs. Drought impacts on NSCs and SMs are further complicated by organ- 74 specific responses to drought. Drought may reduce leaf export and phloem transport of 4 75 carbohydrates (Hesse et al., 2018; Joseph et al., 2020; Ruehr et al., 2009), leading to lower 76 carbohydrate availability for production of SMs in stem phloem and roots than in leaves 77 (Mundim and Pringle, 2018). 78 In addition to carbohydrates, phytohormones have been identified as the central regulators of 79 growth and defense, as well as their trade-offs (Huot et al., 2014; Panda et al., 2021; Züst and 80 Agrawal, 2017). Modulation of phytohormones has also been associated with growth and 81 defense regulation under drought stress, such as accumulation of abscisic acid (ABA) and 82 jasmonic acid (JA) as a signaling mechanism to close stomata, inhibit growth, and induce 83 antioxidant and antiherbivore defense (De Ollas and Dodd, 2016; Nguyen et al., 2016; Riemann 84 et al., 2015). Furthremore, the antagonistic interactions between JA and salicylic acid (SA) are 85 known to mediate plant defense (Pieterse et al., 2012; Thaler et al., 2012). However, studies on 86 phytohormonal regulation of tree defense are mostly based on short-lived model species, such as 87 Arabidopsis and crop plants under short-term stress. There is a substantial knowledge gap for 88 long-lived organisms, including trees, that must acclimate and adapt to periodic and episodic 89 environmental stress conditions. 90 To begin filling this gap, we conducted a quantitative analysis of diverse polyphenolics 91 (flavonoids and secoiridoids), NSCs (soluble sugars and starch), and phytohormones (ABA, JA 92 and SA) in leaves and stem phloem of two co-occurring tree species (Quercus ilex and Phillyrea 93 latifolia) in a rainfall exclusion experiment. Polyphenolics have been shown to be prominent 94 antioxidants that defend against UV, heat, fire, and drought stresses in Mediterranean tree 95 species (Laoué et al., 2022; Santacruz-García et al., 2021), including Q. ilex (Tienda-Parrilla et 96 al., 2022) and P. latifolia (Gori et al., 2021; Gori et al., 2019) studied here. It should be noted 97 that P. latifolia can synthesize and store SMs not only in parenchyma and epidermal cells but 98 also in glandular trichomes, and that accumulation of SMs in glandular trichomes can enhance 99 resistance of P. latifolia to high solar radiation (Tattini et al., 2000; Tattini et al., 2005). 100 Polyphenolics were used as an indicator of defense SMs because our previous work showed that 101 both Q. ilex and P. latifolia do not store terpenes (Llusia and Penuelas, 1998). The two tree 102 species, Q. ilex and P. latifolia, are the dominant tree species in this region and known to differ 103 in their resistance and resilience to drought. Both species experienced crown defoliation, growth 104 reduction and/or stem mortality following extreme natural drought (Barbeta et al., 2015) and 5 105 during experimental sustained drought (Barbeta et al., 2013; Liu et al., 2015; Ogaya et al., 2020), 106 but drought impacts tend to be more pronounced in Q. ilex than in P. latifolia. Thus, Q. ilex 107 appears to be more vulnerable to climate stress than P. latifolia. 108 The objective of this study is to investigate how drought affects defense SMs and their 109 relationships to carbohydrates and phytohormones in co-cocurring mature trees. Specifically, we 110 addressed the following hypotheses: 1) experimental drought causes an increase in SMs and 111 NSCs in the leaves, especially in the growing season (i.e. spring); 2) experimental drought 112 decreases levels of SMs and NSCs in the stem phloem, especially in the driest season (i.e. 113 summer). We further hypothesized that 3) phytohormones are potentially associated with SMs 114 and their responses to experimental drought. 115 Materials and methods 116 Study site and experimental design 117 The experiment was conducted in a holm oak (Q. ilex) forest, located in the Prades Mountains in 118 southern Catalonia, Spain (41°21'N, 1°2'E) at an elevation of 930 m. The region is characterized 119 by a typical Mediterranean climate, with the majority of vegetation growth occurring during the 120 wet spring and autumn months. Across the 20-year experiment, mean annual air temperature was 121 12.4°C and mean annual precipitation was 610 mm. The tree community of the forest is 122 dominated by Q. ilex, P. latifolia and Arbutus unedo, with understory species, such as Erica 123 arborea and Juniperus oxycedrus. The structure consists of 6–8 m tall, dense multi-stemmed 124 stools, as a result of coppicing activity prior to 1950. Soils comprise Dystric Cambisol over 125 Paleozoic schist, ranging in depth from 35 to 100 cm. Eight 15 × 10 m plots (four plots received 126 ambient rainfall and four received reduced rainfall) were established in March 1999; PVC strips 127 were placed 0.5–0.8 m above the soil surface, to reduce rainfall by c. 30%. Rainfall was reduced 128 by ca. 30%, within anticipated precipitation reductions of ~20% to ~40% in Mediterranean 129 regions under the RCP8.5 warming scenario (Brogli et al., 2019; Tuel and Eltahir, 2020). 130 Upslope runoff was intercepted by c. 0.8 m deep trenches along the upper edges of the plots. In 131 each plot, soil moisture was measured continuously at four randomly selected positions (at soil 132 depth of 25 cm) using automatic sensors connected to a data logger (ECH2O, METER, Pullman, 133 WA, USA). Overall, the rainfall reduction treatment resulted in c. 15% lower soil moisture 6 134 throughout the experimental period (Bogdziewicz et al., 2020). Leaf water potential was 135 measured at around midday in each of the seasons on twigs of two trees per species in each plot, 136 using a Scholander pressure chamber (Fig. S1; PMS Instrument Co., Corvallis, OR, USA). We 137 identified and labeled all living stems with a diameter larger than 2 cm at breast height (c. 530 138 and 880 stems for Q. ilex and P. latifolia, respectively), and measured changes in stem diameter 139 every winter using a metric tape, as described in Liu et al. (2015) and Ogaya et al. (2020). 140 Sampling and processing 141 For each treatment, plant samples were collected from four plots. In each plot, we selected four 142 trees for each of the two dominant tree species (Q. ilex and P. latifolia), and collected current- 143 year leaf from the upper canopy and stem phloem at around noon on 30-31 May 2018 (spring), 144 30-31 August 2019 (summer), 12-13 November 2019 (autumn), and 21-22 February 2019 145 (winter), respectively. In total, this yielded 512 samples (2 treatments × 4 plots × 2 species × 4 146 trees × 2 organ types × 4 seasons). Samples were flash-frozen in liquid N2, transported on dry ice 147 to the Max Planck Institute for Biogeochemistry (Jena, Germany), and stored at -80 °C prior to 148 processing. Samples were freeze-dried and weighed for measurement of biomass. Leaf samples 149 of P. latifolia from each plot were pooled to make a composite sample due to low leaf biomass. 150 We define stem phloem as the inner bark separated from the thin dead bark and xylem. Samples 151 were ground to fine powder using a ball mill (Retsch MM400, Haan, Germany) and stored at - 152 20 °C prior to extraction and analysis of soluble sugars, starch, secondary metabolites, and 153 phytohormones. 154 Soluble sugars and starch 155 Soluble sugars and starch were extracted following Landhäusser et al. (2018). Approximately 10 156 mg of freeze-dried fine powdered samples were extracted using 85% ethanol, and then vortexed 157 and incubated at 90 °C for 10 min. After centrifugation at 13000 g for 1 min, the supernatant was 158 collected, diluted and stored at -20 °C prior to analysis. Glucose, sucrose, and fructose of the 159 supernatant were analyzed using an High Performance Liquid Chromatography coupled to a 160 Pulsed Amperometric Detection (HPLC-PAD). Concentrations of glucose, sucrose, and fructose 161 were quantified based on linear calibration curves from their respective authentic standards. 162 Soluble sugars were calculated as their sum. For starch analysis, the remaining pellets of the 7 163 sugar extraction were oven-dried at 40 °C, digested with α-amylase (Sigma-Aldrich), vortexed, 164 and incubated at 85 °C for 30 min. After centrifugation at 13000 g for 1 min, an aliquot of the 165 supernatants was transferred to a new vial and digested using amyloglucosidase (Sigma-Aldrich) 166 at 55 °C for 30 min. The supernatant aliquot was then washed with chloroform to remove 167 interfering enzymes, diluted, and analyzed again using HPLC-PAD to determine concentrations 168 of hydrolyzed glucose. Starch content was calculated as glucose equivalents, by multiplying by a 169 factor of 0.9 (Landhäusser et al., 2018). Oak standards were measured as a quality control; 170 measurement error for individual soluble sugars and starch was < 0.6 mg/g. 171 Polyphenolics 172 We extracted polyphenolics as described by Huang et al. (2017). Freeze-dried fine powdered 173 samples (c. 10 mg and 30 mg for leaf and stem phloem, respectively) were extracted using 174 methanol, bead-beaten for 40 s at 6.0 ms, vortexed, and centrifuged at 11 000 g for 10 min. The 175 supernatant was collected and the remaining pellet was re-extracted using the same procedure 176 (methanol) to improve extraction efficiency. The supernatants were combined and then stored at 177 -20°C prior to analysis. 178 Interspecific comparisons of SMs require accurate identification and quantification of 179 metabolites in different tree species and organ types (Tables S1, S2). To do so, we identified and 180 quantified the major defense compounds in leaves and stem phloem of the two tree species using 181 a combination of HPLC-UV-ESI-Ion-Trap-MS and UPLC-Q-TOF-MS. An aliquot of methanol 182 extracts (40 µl) was injected into HPLC-UV-ESI-MS (Esquire 6000, Bruker Daltronics, Bremen, 183 Germany). Compounds were separated using a Nucleodur Sphinx RP column (250 × 4.6 mm, 5 184 μm particle size; Macherey Nagel, Dueren, Germany), with mobile phases of 0.2% (v/v) formic 185 acid (A) and acetonitrile (B), with A–B gradient elution profiles at a flow rate of 1 ml/min over c. 186 25 min. The elution profile was: 0 to 5min, 0% B; 5 to 20 min, 0% to 45% B; 20 to 20.1 min, 45% 187 to 100% B; 20.1 to 23 min, 100% B; 23 to 28 min, 0% B. Major phenolic compounds were 188 selected for subsequent identification and quantification using UV wavelengths at 280 and 330 189 nm. ESI-MS was performed in either positive or negative mode at 60 V skimmer voltage, 4200 190 V capillary voltage, 35 psi nebulizer pressure, 11 l/min drying gas at 330 °C, and -121V capillary 191 exit potential. Compounds were identified by comparing the fragmentation patterns with 192 previous studies (Guimarães et al., 2013; Quirantes-Piné et al., 2013; Song et al., 2021) and/or 8 193 five commercial standards (catechin, quercetin 3-glucoside, oleuropein, procyanidin B1, 194 verbascoside; Tables S1, S2). 195 Compounds that displayed a large UV peak, but could not be clearly identified using ESI-Ion- 196 Trap-MS were further investigated using UPLC-Q-TOF-MS. An aliquot of methanol extracts (2 197 µl) was injected into UPLC-Q-TOF-MS, and compounds were separated using a Dionex 198 Ultimate 3000 RS Pump system (Thermo Fisher Scientific) equipped with a Zorbax Eclipse 199 XDB-C18 column (100x2.1 mm,1.8mm; Agilent Technologies). The mobile phases consisted of 200 0.1% (v/v) formic acid in water (A) and acetonitrile (B), with a flow rate of 0.3 mL/min. The 201 column temperature was maintained at 25°C. The elution profile was: 0 to 0.5min, 5% B; 0.5 to 202 11 min, 5% to 60% B; 11.1 to 12 min, 100% B; 12.1 to 15 min, 5%B. The LC system was 203 coupled to a timsTOF mass spectrometer (Bruker Daltonics) equipped with a turbospray ion 204 source (capillary voltage, 3,500 V). Nitrogen was used as drying gas (8 L/min, 230°C) and 205 nebulizer gas (2.8 bar). Analysis was carried out in negative ionization mode, scanning a mass 206 range from m/z 50 to 1,500. Sodium formate adducts were used as internal calibrators. The 207 derived molecular formula was used to identify SMs, based on previous studies included in the 208 SciFinder database. The major SMs that may induce trade-offs with growth were quantified 209 based on linear calibration curves from seven commercial standards, including catechin, 210 coumaric acid, apigenin 7-O-glucoside, quercetin 3-glucoside, oleuropein, procyanidin B1 and 211 verbascoside (Tables S1, S2). 212 Phytohormones 213 Phytohormones were extracted, identified, and quantified as described by Huang et al. (2017), 214 with slight modifications. The extraction procedures used for the phenolic compounds were also 215 used for phytohormones, except that freeze-dried ground powder (c. 20 mg and 30 mg for leaf 216 and stem phloem, respectively) was extracted using methanol containing 40 ng/ml of stable 217 isotope-labeled internal standards, including D5-indole-3-acetic-acid (D5-IAA, Olchemin, 218 Olomouc, Czech Republic), D6-abscisic acid (D6-ABA, Toronto Research Chemicals, Toronto, 219 CA), D6-jasmonic acid (D6-JA, HPC Standards GmbH, Cunnersdorf, Germany), and D4-salicylic 220 acid (D4-SA, Santa Cruz Biotechnology, Santa Cruz, CA, USA). The remaining pellet was re- 221 extracted using methanol without internal phytohormone standards, and the supernatants were 222 combined and then stored at -20°C prior to analysis. 9 223 We identified and quantified phytohormones using a UPLC coupled to a triple quadrupole MS 224 (QTRAP 6500+, Sciex, Foster City, CA, USA), where phytohormones were separated using a 225 Zorbax Eclipse XDB-C18 column (50 x 4.6 mm, 1.8 µm, Agilent Technologies), at a flow rate of 226 1.1 ml/min, with mobile phases of 0.05% (v/v) formic acid (A) and acetonitrile (B) comprising 227 linear gradient elution profiles of 0 to 0.5 min, 10% B; 0.5 to 4.0 min, 10–90% B; 4.0 to 4.02 228 min, 90–100% B; 4.02 to 4.50 min, 100% B, 4.50 to 4.51 min, 100–10% B; and, 4.51 to 7.00 229 min, 10% B. ABA, JA, and SA were ionized in negative mode; IAA was ionized in positive 230 mode in separate chromatographic analysis, with collision energy set at 19V and declustering 231 potential at 31V. The parent ion → product ion (m/z) of each hormone was: 263.0 → 153.2 for 232 ABA; 269.0 → 159.2 for D6-ABA; 209.1 → 59.0 for JA; 215.0 → 59.0 and 214.0 → 59.0 for 233 D6-JA; 136.9 → 93.0 for SA; 140.9 → 97.0 for D4-SA; 176 → 130 for IAA; and, 181 → 135, 234 181 → 134, and 181 → 133 for D5-IAA. Concentrations of ABA, JA, SA, and IAA were 235 calculated as the ratio of the peak area of the analyte to that of the corresponding stable isotope- 236 labeled internal standard multiplied by the concentration of the standard. Linearity was 237 confirmed from analysis of serial dilutions of internal standards. IAA concentrations are below 238 the limits of detection in both tree species and thus not reported here. Data were processed using 239 Analyst 3.0.3 software (Sciex). 240 Data analysis 241 Prior to statistical analyses, leaf biomass, stem diameter and metabolite concentrations of the 242 four individual trees were averaged for each plot. Leaf mass and stem diameter were used as a 243 proxy of leaf and stem size, respectively. We calculated changes in leaf biomass over a 30-day 244 period in spring as an indicator of springtime leaf growth, and the 2018-2019 annual increment 245 in stem diameter as an indicator of stem growth. Data normality and variance equality were 246 checked using Shapiro–Wilk and Levene’s tests, respectively. Differences in metabolites (SMs, 247 soluble sugars, starch, NSCs, ABA, JA and SA) among species, treatment and seasons were 248 tested by three-way ANOVA (Table S3, S4). Drought effects on leaf and stem size/growth, and 249 metabolites were tested using Student’s t-test. Data were log-transformed to meet normality, 250 homoscedasticity and/or sphericity when needed. 251 To determine the relationships of SMs to soluble sugars, starch, and different phytohormones, we 252 used Pearson’s correlation and calculated coefficients for each species in each season. We pooled 10 253 treatment data (control + drought plots) for each species and preformed principal component 254 analysis (PCA) to summarize and visualize relationships among biomass/stem diameter and all 255 metabolic data across species and treatments. All data were standardized prior to PCA. All 256 statistical analysis was conducted in R (version 3.6.3, R Development Core Team, 2020). 257 Results 258 Environmental conditions 259 Our field site in Catalonia is characterized by strong seasonality of solar radiation, air 260 temperature and precipitation (Fig. 1). Daily solar radiation and mean air temperature had similar 261 seasonal fluctuations: they increased from early spring (March) to mid summer (July) and then 262 declined until winter (Fig. 1a, b). Precipitation also showed great seasonal variation (Fig. 1c), 263 with 90 % of precipitation (c. 890 mm) in spring (March to May) and autumn (September to 264 November), and almost no precipitation in summer (June to August) and winter (December to 265 February). 266 Leaf biomass and stem diameter 267 P. latifolia had significantly lower leaf biomass and springtime leaf growth than Q. ilex, 268 regardless of drought treatment (Fig. 2a, b). P. latifolia also had smaller stem diameter sizes than 269 Q. ilex, while the two tree species showed similar annual stem diameter growth (Fig. 2c, d). For 270 both species, experimental drought had no significant effects on leaf biomass and stem diameter. 271 Seasonal dynamics of secondary metabolites in leaves and stem phloem of both species 272 We identified and quantified a diverse array of SMs in leaves and stem phloem of both species 273 (Tables S1, S2). The major SMs in P. latifolia leaves were hydroxycoumarin glucoside, flavones 274 (e.g., apigenin and luteolin derivatives), and in particular secoiridoids (e.g., oleuropein). The 275 stem phloem of P. latifolia contained secoiridoids (e.g., oleoside dimethyl ester) and 276 phenylpropanoids (e.g., verbascoside and osmanthuside J). The major SMs in Q. ilex leaves were 277 flavan-3-ols (e.g., catechin and procyanidins) and flavonols (e.g., quercetin and kaempferol 278 derivatives); Q. ilex stem phloem contained flavan-3-ols (e.g., catechin and procyanidin) and 279 flavonols (e.g., taxifolin-hexoside). 11 280 We found much larger differences in concentrations of leaf and phloem SMs between species 281 than within species (Fig. 3; Table S3, S4). Concentrations of leaf and phloem SMs were 282 significantly higher in P. latifolia (up to 18% and 5%, respectively) than in Q. ilex (7% and 0.6%, 283 respectively) (Fig. 3a). In both species, concentrations of leaf SMs varied with seasons, and 284 increased from autumn to winter (Fig. 3a). Experimental drought had no significant effects on 285 concentrations of total SMs in leaves (Fig. 3a; Table S3), although there was a modest decrease 286 in specific groups of SMs, like flavan-3-ols in Q. ilex (Fig. S2) and flavones in P. latifolia (Fig. 287 S3). In stem phloem, concentrations of SMs remained relatively constant throughout the year 288 (Fig. 3b). Experimental drought significantly decreased concentrations of SMs (secoiridoids; Fig. 289 S3) in spring and summer for P. latifolia (P<0.05), but not for Q. ilex (Fig. 3b). 290 Seasonal dynamics of soluble sugars and starch in leaves and stem phloem of both species 291 The two co-occurring tree species showed significant species-specific seasonal varations in 292 concentrations of soluble sugars and starch (Fig. 4a, d; Tables S3, S4). In spring, leaves and stem 293 phloem had significantly higher concentrations of soluble sugars in P. latifolia than in Q. ilex 294 (P<0.05; Fig. 4a, d). For P. latifolia, soluble sugars strongly declined from spring to summer in 295 leaves, and from autumn to winter in both leaves and stem phloem. By contrast, for Q. ilex 296 soluble sugars remained relatively constant in leaves and strongly increased from spring to 297 summer in the stem phloem (Fig. 4a, d). As a result, the interspecific differences in soluble 298 sugars were diminished in summer and autumn, and were reversed in winter (Fig. 4a, d). More 299 specifically, P. latifolia leaves had generally higher concentrations of glucose and lower 300 concentrations of sucrose compared to Q. ilex leaves. In the stem phloem, however, P. latifolia 301 had lower concentrations of glucose and sucrose than Q. ilex across seasons (Fig. S4). 302 Concentrations of leaf starch were generally higher in P. latifolia than in Q. ilex, with the latter 303 having extreme low concentrations in summer (< 5 mg/g; Fig. 4b). In contrast to leaves, stem 304 phloem had lower concentrations of starch in P. latifolia than in Q. ilex (Fig. 4e). In both species, 305 concentrations of phloem starch were lowest in summer and autumn, and strongly increased from 306 autumn to winter (Fig. 4b, e). Concentrations of leaf NSCs were significantly higher in P. 307 latifolia than in Q. ilex during the growing season (spring and early summer), but the differences 308 diminished in autumn and winter. By contrast, concentrations of phloem NSCs were similar 12 309 between the two species during the growing season, but were significantly lower in P. latifolia 310 than in Q. ilex in winter. 311 Experimental drought had no significant effects on concentrations of soluble sugars in leaves of 312 both species (Fig. 4a). In the stem phloem, however, concentrations of soluble sugars were 313 significantly reduced in spring and summer for P. latifolia, but not for Q. ilex (Fig. 4d). There 314 were no significant effects of drought on the concentrations of starch in leaves and stem phloem 315 of both species across seasons (Fig. 4b, e). 316 Seasonal dynamics of phytohormones in leaves and stem phloem of both species 317 Concentrations of phytohormones showed seasonal variations and species- and tissue-specific 318 differences (Fig. 5; Tables S3, S4). In spring, leaf ABA levels were higher in P. latifolia than in 319 Q. ilex, whereas the opposite was found for phloem ABA levels (Fig. 5a, d). Leaves of P. 320 latifolia had generally higher concentrations of JA and lower concentrations of SA compared to 321 leaves of Q. ilex across seasons (Fig. 5b, c; Table S3). Leaf hormone levels showed strong 322 seasonal varations for both species, and were highest in spring and lowest in autumn or winter 323 (Fig. 5a-c). Under experimental drought, Q. ilex showed a slight but nonsignificant increase in 324 leaf SA and ABA as well as in phloem ABA, whereas P. latifolia showed no changes in leaf 325 phytohormones (Fig. 5a-c). Note that both JA and SA were present at very low concentrations in 326 stem phloem (Fig. 5e, f) and did not differ between the two species (Table S4). 327 Correlations of SMs to soluble sugars, starch, and phytohormones within and between species 328 SMs and soluble sugars were not correlated in leaves of both species (Fig. 6a, b), but positively 329 correlated in stem phloem (Fig. 6c, d), especially for P.latifolia. Relationships between SMs and 330 phytohormones differed between species. In Q. ilex, SMs were generally negatively correlated 331 with phytohormones in leaves (Fig. 6a), particularly with ABA and SA in spring. In P. latifolia, 332 however, there were no consistent relationships between SMs and phytohormones in leaves (Fig. 333 6b) and stem phloem (Fig. 6d). 334 Principal component analysis showed that the two tree species were separated on the first axis 335 (PC1) reflecting trade-offs between SMs and leaf biomass/stem diameter across seasons (Fig. 7). 336 Across species and treatments, there were positive relationships between SMs and soluble sugars 13 337 in spring (Fig. 7a, e); however, these relationships were absent or even reversed in the other 338 seasons. The relationships between SMs and starch were relatively consistent but differed 339 between organs, being generally positive in leaves (Fig. 7; upper panels) and negative in stem 340 phloem (Fig. 7; lower panels). Strikingly, leaf SMs were found to be consistently and positively 341 correlated with leaf JA and negatively correlated with leaf SA in all seasons (Fig. 7; upper 342 panels). In stem phloem, however, SMs were negatively correlated with ABA, but not with JA 343 and SA (Fig. 7; lower panels). 344 Discussion 345 Species-specific SMs and their seasonal dynamics 346 Understanding the role of SMs in tree responses to environmental changes requires accurate 347 identification of SMs. By using UPLC-Q-TOF-MS, we found that the two tree species have 348 different types of SMs: P. latifolia has flavones (e.g. luteolin-hexoside) and secoiridoids (e.g. 349 oleuropein and its precursor oleoside), whereas Q. ilex has flavan-3-ols (e.g. catechin), quercetin 350 and kaempferol deriviatives. These compounds are known to have multiple functions, such as 351 anti-herbivore defense by decreasing the suitability of tissues for insects (Hammerbacher et al., 352 2020; Konno et al., 1999), antioxidant defense by scavenging reactive oxygen species (ROS) and 353 photoprotection by screening UV-B radiation (Agati et al., 2020; Agati and Tattini, 2010; 354 Brunetti et al., 2015). 355 Quantification of SMs based on multiple standards showed that concentations of SMs (flavones 356 and secoiridoids) in the drought-resistant species P. latifolia (up to 20% of biomass) were much 357 higher than those of SMs (flavan-3-ols and flavonols) in the drought-susceptible species Q. ilex. 358 High levels of SMs may be attributed to the presence of glandular trichomes for P. latifolia, but 359 this is not the case for Q. ilex. Our result may help explain why P. latifolia experienced less 360 defoliation and mortality following extreme nature drought (Barbeta et al., 2015) and herbivory 361 and diseases than Q. ilex (Paine and Lieutier, 2016). However, high production of SMs may be 362 associated with the low growth capacity of P. latifolia, leading to a trade-off between resistance 363 and growth. This may reflect an evolutionary adaptation of the drought-resistant P. latifolia to 364 harsh environmental conditions and high herbivory/oxidative stress that prevail in Mediterranean 365 semi-arid climates (Di Ferdinando et al., 2014). 14 366 The seasonal dynamics of SMs may in part be driven by their demands for screening UV-B 367 radiation and alleviating oxidative damage caused by environmental stress like excess light, high 368 or low temperatures, and drought (Fig. 1; Agati et al., 2020; Agati and Tattini, 2010; Brunetti et 369 al., 2015). In P. latifolia, concentrations of leaf SMs increased during hot and dry summer (Fig. 370 3), consistent with previous studies on P. latifolia (Gori et al., 2019) where phenylpropanoids 371 accumulated in response to high temperatures. However, this is not case for Q. ilex in summer, 372 where production of SMs could potentially be limited by carbohydrate supply (Huang et al., 373 2020b), as indicated by a severe depletion of leaf starch (Fig. 4). In both species, accumulation of 374 SMs was observed in leaves during dry and cold winter (Fig. 3). We propose that accumulation 375 of SMs may be an important physiological mechanism to enhance leaf tolerance to 376 photoinhibition in winter (Chang et al., 2021), as reflected by relatively low photochemical 377 efficiency (Ogaya and Peñuelas, 2003; Ogaya et al., 2011). 378 Homeostatic levels of leaf defense SMs and NSCs under experimental drought 379 We hypothesized that experimental drought would lead to an increase in leaf SMs, especially 380 during the growing season. However, contrary to our hypothesis, concentrations of total leaf SMs 381 showed a slight and nonsignificant decrease in Q. ilex and remained relatively constant in P. 382 latifolia (Fig. 3), even when leaf water potential showed a great decline in summer (Fig. S1). 383 These results differ from a few glasshouse studies on tree seedlings, in which drought led to 384 significant changes in leaf content of phenolics (Aranda et al., 2018; Mundim and Pringle, 2018). 385 In our study, these mature trees have acclimated to the long-term drought treatment and that 386 droughted trees may have reduced canopy area (Barbeta et al., 2013; Liu et al., 2015) and 387 adjusted root distribution (Barbeta et al., 2015), thereby leading to no significant changes in leaf- 388 level chemical characteristics. This is also supported by non-significant effects of experimental 389 drought on leaf growth (Fig. 2) and NSCs (Fig. 4), as has been observed in other tree species 390 after long-term acclimation to rainfall manipulations (Pivovaroff et al., 2021; Rowland et al., 391 2015; Schönbeck et al., 2018). 392 Within-species variations in SMs were not associated with those in soluble sugars in leaves, most 393 likely due to the multifunctional nature of soluble sugars being not only substrates but also 394 osmolytes (Blumstein et al., 2023; Hajek et al., 2022) and reserves for survival (Herrera-Ramírez 395 et al., 2021; Huang et al., 2021). Another explanation is that turnover of SMs and soluble sugars 15 396 may have very different temporal dynamics, and that SMs had more constant seasonal dynamics 397 than soluble sugars. Substantial amounts of SMs like flavan-3-ols in Q. ilex and secoiridoids in P. 398 latifolia, were produced in spring to protect developing and vulnerable organs (sensu optimal 399 defense theory, Barton et al., 2019; McKey, 1974). These compounds may be stored for a long 400 time until being mobilized or metabolized during stress (Panda et al., 2021), whereas soluble 401 sugar levels are known to be determined by the source-sink balance that can vary from hours and 402 days to seasons (Furze et al., 2018; Martínez-Vilalta et al., 2016; Tixier et al., 2018). Thus, our 403 results highlight that the temporal dynamics of SMs and NSCs should be accounted for when 404 interpreting the physiological linkages between SMs and NSCs, especially for trees operating at 405 seasonal and annual timescales. 406 Species-specific reduction in phloem defense SMs and soluble sugars under experimental 407 drought 408 Consistent with our hypothesis, experimental drought reduced levels of total SMs and NSCs in 409 the stem phloem of P. latifolia during the growing season (spring and early summer; Fig. 3). 410 Given that we measured concentrations of SMs and NSCs including those that are being 411 transported in the sieve tubes and also those stored in parenchyma cells, we propose two possible 412 mechanisms: 1) reduced soluble sugars may indicate locally reduced availability of substrates for 413 production of SMs in the stem phloem (Erbilgin et al., 2021; Huang et al., 2020b; Wiley et al., 414 2016). This is supported by significant relationships between SMs and soluble sugars in P. 415 latifolia stem phloem from spring to autumn; 2) reduced levels of phloem SMs and soluble 416 sugars in P. latifolia may indicate reduced leaf export and phloem transport of soluble 417 compounds (e.g., oleoside dimethyl ester; Table S2) under experimental drought (Savage et al., 418 2016). This explanation is consistent with results of isotopic labeling studies showing that 419 drought can limit transport of recent photosynthates belowground (Dannoura et al., 2018; 420 Hagedorn et al., 2016; Hesse et al., 2018; Joseph et al., 2020; Ruehr et al., 2009). Our study thus 421 provides experimental evidence that drought may reduce defense and carbohydrate metabolisms 422 in the stem phloem of mature P. latifolia trees, implying that with more frequent and severe 423 future spring/summer drought this tree species may be less resistant to oxidative stress (Tattini et 424 al., 2015) and biotic attacks by phloem-feeding insects (Huang et al., 2020a). 16 425 In contrast to P. latifolia, Q. ilex did not reduce levels of phoem SMs and NSCs under 426 experimental drought. The exact mechanisms leading to the differential effects of drought on 427 phloem SMs in P. latifolia and Q. ilex remain uncertain, but are probably related to the 428 concentration of solutes in the stem phloem. Compared to P. latifolia, Q. ilex trees transport 429 and/or store significantly smaller amounts of SMs in stem phloem. The relatively high 430 concentrations of phloem SMs (c. 5 % on a dry weight basis) in P. latifolia trees may 431 theoretically lead to high phloem sap viscosity, making phloem transport of soluble SMs and 432 carbohydrates susceptible to increased drought stress. This may explain why experimental 433 drought decreased levels of phloem SMs in P. latifolia in spring and summer, but not in winter 434 when phloem soluble sugars decreased to very low levels (c. 0.05 %). These results demonstrate 435 that drought impacts on phloem defense and carbohydrate metabolism are seasonal and species- 436 specific, and suggest phloem transport as an important mechanism that needs to be examined by 437 isotopic labeling and tracing (Huang et al., 2019). 438 The potential role of phytohormones in regulating growth and production of SMs within and 439 between species 440 While studies have investigated the hormonal regulation of defense SMs in crop and model plant 441 species (Huot et al., 2014; Panda et al., 2021; Züst and Agrawal, 2017), little information is 442 available for mature trees. We found that stem phloem had extremely low levels of JA and SA 443 across seasons, while leaf hormone levels were highest in spring and declined over time (Fig. 5). 444 Similarly, a recent study found in Scots pine and Norway spruce that leaf hormones declined 445 over time during summer drought (Pashkovskiy et al., 2022). These results suggest that 446 phytohormones may play an important role in regulating leaf metabolism during the early stage 447 of development. Interestingly, experimental drought induced a slight but nonsignificant increase 448 in ABA and SA in Q. ilex leaves (Fig. 5), which was significantly correlated with SMs, 449 especially in spring (Fig. 6). Future studies should investigate whether increased ABA in Q. ilex 450 leaves may induce stomatal closure and reduce photosynthesis (De Ollas and Dodd, 2016; 451 Nguyen et al., 2016), and whether SA-induced abiotic stress responses (i.e. production of 452 antioxidant enzymes; Khan et al., 2015) may induced trade-offs with production of defense SMs 453 (Thaler et al., 2012). 17 454 The divergence in leaf growth and defense SMs between the co-occurring species appears to be 455 associated with phytohormones (Fig. 7). Compared to Q. ilex leaves, P. latifolia leaves had 456 significantly higher levels of ABA in spring, potentially leading to lower growth capacity (Fig. 2; 457 Nguyen et al., 2016). It should be noted that the two species have similar timing of leaf flushing 458 (late April) and at the time that we collected leaf samples (mid-May), flower senescence 459 occurred in P. latifolia but not in Q. ilex. Thus, the high levels of leaf ABA may be associated 460 with flower senescence in P. latifolia (Arrom and Munné-Bosch, 2012; Tripathi and Tuteja, 461 2007). Interestingly, P. latifolia leaves also had significantly higher levels of JA than Q. ilex 462 leaves, which may potentially induce greater production of SMs (Wasternack and Strnad, 2019). 463 In addition, JA appears to antagonistically interact with SA between species (Pieterse et al., 2012; 464 Thaler et al., 2012). Thus, we propose that differentiation in phytohormone production is a 465 potential mechanism driving the divergence in leaf growth, phenology and defense SMs in this 466 forest community. If this holds true for other tree species, investigations of phytohormones 467 would have strong implications for assessing and predicting species-specific defense capacity. 468 Conclusion and outlook 469 Our field drought experiment combined with quantitative metabolomics provide evidence about 470 how drought affects the seasonal dynamics of SMs and their potential interactions with 471 carbohydrates and phytohormones in co-occurring mature trees. We show that the two dominant 472 tree species have different types of SMs, and that the drought-resistant species P. latifolia 473 contained greater levels of SMs compared to the drought-susceptible species Q. ilex. The 474 observed seasonality of SMs may be explained by the seasonal variations in climate. 475 Experimental drought had no significant effects on levels of SMs, NSCs and phytohormones in 476 the leaves of both tree species, even when leaf water potential was strongly reduced in dry 477 summer. The homeostatic leaf-level chemical characteristics corroborate the acclimation of 478 mature trees to long-term drought. By contrast, experimental drought reduced levels of SMs and 479 soluble sugars in the stem phloem of P. latifolia but not of Q. ilex, especially in spring and 480 summer. This suggests that increasing drought stress may reduce the resistance of P. latifolia 481 trees to abiotic and biotic stresses. Our results provide mechanistic insights into organ- and 482 species-specific dynamics of SMs, which may represent an important adaptation component in 18 483 this forest community and may further shape its structure and function under ongoing climate 484 change and biotic attacks (Peñuelas et al., 2018). 485 We acknowledge that our drought treatment (i.e. 30% rainfall exclusion) was relatively moderate, 486 and suggest that future studies should impose more severe drought treatments to examine the 487 relationships of SMs to NSCs and phytohormones. Future studies should investigate the linkage 488 of production of SMs to carbon economy, hydraulic strategies (Limousin et al., 2022) and 489 phenology across tree species under experimental drought. Phytohormoal regulation of defense 490 SMs should be examined for a greater range of tree species and organ types, to unravel the key 491 regulatory mechanisms that determine the defense capacity of different tree species under 492 climate change. 493 Supplementary Information 494 Figure S1 Leaf water potential in Quercus ilex and Phillyrea latifolia. 495 Figure S2 Concentrations of secondary metabolites in leaves and stem phloem of Quercus ilex. 496 Figure S3 Concentrations of secondary metabolites in leaves and stem phloem of Phillyrea 497 latifolia. 498 Figure S4 Concentrations of glucose, sucrose, and fructose in leaves and stem phloem of 499 Quercus ilex and Phillyrea latifolia. 500 Table S1 Method of identification and quantification of the major secondary metabolites in 501 leaves of Quercus ilex and Phillyrea latifolia. 502 Table S2 Method of identification and quantification of the major secondary metabolites in stem 503 phloem of Quercus ilex and Phillyrea latifolia. 504 Table S3 Summary of the three-way ANOVA testing effects of species, treatment, seasons, and 505 their interactions on different metabolic pools in leaves. 506 Table S4 Summary of the three-way ANOVA testing effects of species, treatment, seasons, and 507 their interactions on different metabolic pools in the stem phloem. 19 508 Acknowledgements 509 We thank Dr. Henriette Uthe for her assistance with analysis of the phytohormones, and 510 Savoyane Lambert, Anett Enke and Iris Kuhlmann for their help with sample processing and 511 analysis of NSCs. J.H. acknowledge support from the Max Planck Society, and Deutsche 512 Forschungsgemeinschaft (437772218). 513 Author contributions 514 J.H., H.H., and J.P. contributed to the plan and design of the work. R.O. collected samples in the 515 field. J.H., M.R., H.H., and J.G. proformed compound analysis. J.H. and I.S analysed the data. 516 J.H. wrote the manuscript and all authors contributed to revisions. 517 518 References 519 Agati, G., Brunetti, C., Fini, A., Gori, A., Guidi, L., Landi, M., Sebastiani, F., Tattini, M., 2020. Are 520 flavonoids effective antioxidants in plants? Twenty years of our investigation. Antioxidants 9, 1098. 521 Agati, G., Tattini, M., 2010. Multiple functional roles of flavonoids in photoprotection. New Phytologist 522 186, 786-793. 523 Anderegg, W.R.L., Trugman, A.T., Badgley, G., Anderson, C.M., Bartuska, A., Ciais, P., Cullenward, D., 524 Field, C.B., Freeman, J., Goetz, S.J., Hicke, J.A., Huntzinger, D., Jackson, R.B., Nickerson, J., Pacala, S., 525 Randerson, J.T., 2020. Climate-driven risks to the climate mitigation potential of forests. Science 368, 526 eaaz7005. 527 Aranda, I., Sánchez-Gómez, D., Cadahía, E., de Simón, B.F., 2018. Ecophysiological and metabolic 528 response patterns to drought under controlled condition in open-pollinated maternal families from a 529 Fagus sylvatica L. population. Environ. Exp. Bot. 150, 209-221. 530 Arrom, L., Munné-Bosch, S., 2012. Hormonal changes during flower development in floral tissues of 531 Lilium. Planta 236, 343-354. 532 Barbeta, A., Mejía-Chang, M., Ogaya, R., Voltas, J., Dawson, T.E., Peñuelas, J., 2015. The combined 533 effects of a long-term experimental drought and an extreme drought on the use of plant-water sources 534 in a Mediterranean forest. Global Change Biology 21, 1213-1225. 535 Barbeta, A., Ogaya, R., Peñuelas, J., 2013. Dampening effects of long-term experimental drought on 536 growth and mortality rates of a Holm oak forest. Global Change Biology 19, 3133-3144. 537 Barton, K.E., Edwards, K.F., Koricheva, J., 2019. Shifts in woody plant defence syndromes during leaf 538 development. Functional Ecology 33, 2095-2104. 539 Blumstein, M., Gersony, J., Martinez-Vilalta, J., Sala, A., 2023. Global variation in nonstructural 540 carbohydrate stores in response to climate. Glob Chang Biol 29, 1854-1869. 541 Bogdziewicz, M., Fernández-Martínez, M., Espelta, J.M., Ogaya, R., Penuelas, J., 2020. Is forest fecundity 542 resistant to drought? Results from an 18-yr rainfall-reduction experiment. New Phytologist 227, 1073- 543 1080. 544 Brogli, R., Sørland, S.L., Kröner, N., Schär, C., 2019. Causes of future Mediterranean precipitation decline 545 depend on the season. Environmental Research Letters 14, 114017. 20 546 Brunetti, C., Guidi, L., Sebastiani, F., Tattini, M., 2015. Isoprenoids and phenylpropanoids are key 547 components of the antioxidant defense system of plants facing severe excess light stress. Environ. Exp. 548 Bot. 119, 54-62. 549 Chang, C.Y.-Y., Bräutigam, K., Hüner, N.P.A., Ensminger, I., 2021. Champions of winter survival: cold 550 acclimation and molecular regulation of cold hardiness in evergreen conifers. New Phytologist 229, 675- 551 691. 552 Dannoura, M., Epron, D., Desalme, D., Massonnet, C., Tsuji, S., Plain, C., Priault, P., Gérant, D., 2018. The 553 impact of prolonged drought on phloem anatomy and phloem transport in young beech trees. Tree 554 Physiol. 39, 201-210. 555 De Ollas, C., Dodd, I.C., 2016. Physiological impacts of ABA–JA interactions under water-limitation. Plant 556 Mol. Biol. 91, 641-650. 557 Di Ferdinando, M., Brunetti, C., Agati, G., Tattini, M., 2014. Multiple functions of polyphenols in plants 558 inhabiting unfavorable Mediterranean areas. Environ. Exp. Bot. 103, 107-116. 559 Erbilgin, N., Zanganeh, L., Klutsch, J.G., Chen, S.H., Zhao, S., Ishangulyyeva, G., Burr, S.J., Gaylord, M., 560 Hofstetter, R., Keefover-Ring, K., Raffa, K.F., Kolb, T., 2021. Combined drought and bark beetle attacks 561 deplete non-structural carbohydrates and promote death of mature pine trees. Plant, Cell & 562 Environment 44, 3636-3651. 563 Furze, M.E., Huggett, B.A., Aubrecht, D.M., Stolz, C.D., Carbone, M.S., Richardson, A.D., 2018. Whole- 564 tree nonstructural carbohydrate storage and seasonal dynamics in five temperate species. New 565 Phytologist 221, 1466-1477. 566 Gori, A., Brunetti, C., dos Santos Nascimento, L.B., Marino, G., Guidi, L., Ferrini, F., Centritto, M., Fini, A., 567 Tattini, M., 2021. Photoprotective role of photosynthetic and non-photosynthetic pigments in Phillyrea 568 latifolia: is their “antioxidant” function prominent in leaves exposed to severe summer drought? 569 International Journal of Molecular Sciences 22, 8303. 570 Gori, A., Tattini, M., Centritto, M., Ferrini, F., Marino, G., Mori, J., Guidi, L., Brunetti, C., 2019. Seasonal 571 and daily variations in primary and secondary metabolism of three maquis shrubs unveil different 572 adaptive responses to Mediterranean climate. Conservation Physiology 7. 573 Guimarães, R., Barros, L., Dueñas, M., Carvalho, A.M., Queiroz, M.J.R.P., Santos-Buelga, C., Ferreira, 574 I.C.F.R., 2013. Characterisation of phenolic compounds in wild fruits from Northeastern Portugal. Food 575 Chem. 141, 3721-3730. 576 Hagedorn, F., Joseph, J., Peter, M., Luster, J., Pritsch, K., Geppert, U., Kerner, R., Molinier, V., Egli, S., 577 Schaub, M., Liu, J.-F., Li, M., Sever, K., Weiler, M., Siegwolf, R.T.W., Gessler, A., Arend, M., 2016. 578 Recovery of trees from drought depends on belowground sink control. Nature plants 2, 16111. 579 Hajek, P., Link, R.M., Nock, C.A., Bauhus, J., Gebauer, T., Gessler, A., Kovach, K., Messier, C., Paquette, A., 580 Saurer, M., Scherer-Lorenzen, M., Rose, L., Schuldt, B., 2022. Mutually inclusive mechanisms of drought- 581 induced tree mortality. Glob Chang Biol 28, 3365-3378. 582 Hammerbacher, A., Wright, L.P., Gershenzon, J., 2020. Spruce phenolics: biosynthesis and ecological 583 functions, in: Porth, I.M., De la Torre, A.R. (Eds.), The Spruce Genome. Springer International Publishing, 584 Cham, pp 193-214. 585 Herms, D.A., Mattson, W.J., 1992. The dilemma of plants: to grow or defend. The Quarterly Review of 586 Biology 67, 283-335. 587 Herrera-Ramírez, D., Sierra, C.A., Römermann, C., Muhr, J., Trumbore, S., Silvério, D., Brando, P.M., 588 Hartmann, H., 2021. Starch and lipid storage strategies in tropical trees relate to growth and mortality. 589 New Phytologist 230, 139-154. 590 Hesse, B.D., Goisser, M., Hartmann, H., Grams, T.E.E., 2018. Repeated summer drought delays sugar 591 export from the leaf and impairs phloem transport in mature beech. Tree Physiol. 39, 192-200. 21 592 Hesse, B.D., Hartmann, H., Rötzer, T., Landhäusser, S.M., Goisser, M., Weikl, F., Pritsch, K., Grams, T.E.E., 593 2021. Mature beech and spruce trees under drought – Higher C investment in reproduction at the 594 expense of whole-tree NSC stores. Environ. Exp. Bot. 191, 104615. 595 Huang, J., Hammerbacher, A., Gershenzon, J., van Dam, N.M., Sala, A., McDowell, N.G., Chowdhury, S., 596 Gleixner, G., Trumbore, S., Hartmann, H., 2021. Storage of carbon reserves in spruce trees is prioritized 597 over growth in the face of carbon limitation. Proceedings of the National Academy of Sciences 118, 598 e2023297118. 599 Huang, J., Hammerbacher, A., Weinhold, A., Reichelt, M., Gleixner, G., Behrendt, T., van Dam, N.M., Sala, 600 A., Gershenzon, J., Trumbore, S., Hartmann, H., 2019. Eyes on the future – evidence for trade-offs 601 between growth, storage and defense in Norway spruce. New Phytologist 222, 144-158. 602 Huang, J., Kautz, M., Trowbridge, A.M., Hammerbacher, A., Raffa, K.F., Adams, H.D., Goodsman, D.W., 603 Xu, C., Meddens, A.J.H., Kandasamy, D., Gershenzon, J., Seidl, R., Hartmann, H., 2020a. Tree defence and 604 bark beetles in a drying world: carbon partitioning, functioning and modelling. New Phytologist 225, 26- 605 36. 606 Huang, J., Reichelt, M., Chowdhury, S., Hammerbacher, A., Hartmann, H., 2017. Increasing carbon 607 availability stimulates growth and secondary metabolites via modulation of phytohormones in winter 608 wheat. Journal of Experimental Botany 68, 1251-1263. 609 Huang, J., Rucker, A., Schmidt, A., Gleixner, G., Gershenzon, J., Trumbore, S., Hartmann, H., 2020b. 610 Production of constitutive and induced secondary metabolites is coordinated with growth and storage in 611 Norway spruce saplings. Tree Physiol. 40, 928-942. 612 Huot, B., Yao, J., Montgomery, B.L., He, S.Y., 2014. Growth–defense tradeoffs in plants: A balancing act 613 to optimize fitness. Mol. Plant. 7, 1267-1287. 614 Jin, Y., Li, J., Liu, C., Liu, Y., Zhang, Y., Sha, L., Wang, Z., Song, Q., Lin, Y., Zhou, R., Chen, A., Li, P., Fei, X., 615 Grace, J., 2018. Carbohydrate dynamics of three dominant species in a Chinese savanna under 616 precipitation exclusion. Tree Physiol. 38, 1371-1383. 617 Joseph, J., Gao, D., Backes, B., Bloch, C., Brunner, I., Gleixner, G., Haeni, M., Hartmann, H., Hoch, G., Hug, 618 C., Kahmen, A., Lehmann, M.M., Li, M.-H., Luster, J., Peter, M., Poll, C., Rigling, A., Rissanen, K.A., Ruehr, 619 N.K., Saurer, M., Schaub, M., Schönbeck, L., Stern, B., Thomas, F.M., Werner, R.A., Werner, W., 620 Wohlgemuth, T., Hagedorn, F., Gessler, A., 2020. Rhizosphere activity in an old-growth forest reacts 621 rapidly to changes in soil moisture and shapes whole-tree carbon allocation. Proceedings of the National 622 Academy of Sciences 117, 24885-24892. 623 Khan, M.I.R., Fatma, M., Per, T.S., Anjum, N.A., Khan, N.A., 2015. Salicylic acid-induced abiotic stress 624 tolerance and underlying mechanisms in plants. Frontiers in Plant Science 6, 17. 625 Konno, K., Hirayama, C., Yasui, H., Nakamura, M., 1999. Enzymatic activation of oleuropein: A protein 626 crosslinker used as a chemical defense in the privet tree. Proceedings of the National Academy of 627 Sciences 96, 9159-9164. 628 Landhäusser, S.M., Chow, P.S., Dickman, L.T., Furze, M.E., Kuhlman, I., Schmid, S., Wiesenbauer, J., Wild, 629 B., Gleixner, G., Hartmann, H., Hoch, G., McDowell, N.G., Richardson, A.D., Richter, A., Adams, H.D., 2018. 630 Standardized protocols and procedures can precisely and accurately quantify non-structural 631 carbohydrates. Tree Physiol. 38, 1764-1778. 632 Laoué, J., Fernandez, C., Ormeño, E., 2022. Plant flavonoids in Mediterranean species: A focus on 633 flavonols as protective metabolites under climate stress. Plants 11, 172. 634 Limousin, J.M., Roussel, A., Rodriguez-Calcerrada, J., Torres-Ruiz, J.M., Moreno, M., Garcia de Jalon, L., 635 Ourcival, J.M., Simioni, G., Cochard, H., Martin-StPaul, N., 2022. Drought acclimation of Quercus ilex 636 leaves improves tolerance to moderate drought but not resistance to severe water stress. Plant, Cell & 637 Environment 45, 1967-1984. 22 638 Liu, D., Ogaya, R., Barbeta, A., Yang, X., Peñuelas, J., 2015. Contrasting impacts of continuous moderate 639 drought and episodic severe droughts on the aboveground‐biomass increment and litterfall of three 640 coexisting Mediterranean woody species. Global Change Biology 21, 4196-4209. 641 Liu, H., Park Williams, A., Allen, C.D., Guo, D., Wu, X., Anenkhonov, O.A., Liang, E., Sandanov, D.V., Yin, Y., 642 Qi, Z., Badmaeva, N.K., 2013. Rapid warming accelerates tree growth decline in semi-arid forests of 643 Inner Asia. Global Change Biology 19, 2500-2510. 644 Llusia, J., Penuelas, J., 1998. Changes in terpene content and emission in potted Mediterranean woody 645 plants under severe drought. Canadian Journal of Botany 76, 1366-1373. 646 Martínez-Vilalta, J., Sala, A., Asensio, D., Galiano, L., Hoch, G., Palacio, S., Piper, F.I., Lloret, F., 2016. 647 Dynamics of non-structural carbohydrates in terrestrial plants: a global synthesis. Ecol. Monogr. 86, 495- 648 516. 649 McDowell, N.G., Allen, C.D., Anderson-Teixeira, K., Aukema, B.H., Bond-Lamberty, B., Chini, L., Clark, J.S., 650 Dietze, M., Grossiord, C., Hanbury-Brown, A., Hurtt, G.C., Jackson, R.B., Johnson, D.J., Kueppers, L., 651 Lichstein, J.W., Ogle, K., Poulter, B., Pugh, T.A.M., Seidl, R., Turner, M.G., Uriarte, M., Walker, A.P., Xu, C., 652 2020. Pervasive shifts in forest dynamics in a changing world. Science 368, eaaz9463. 653 McDowell, N.G., Beerling, D.J., Breshears, D.D., Fisher, R.A., Raffa, K.F., Stitt, M., 2011. The 654 interdependence of mechanisms underlying climate-driven vegetation mortality. Trends Ecol. Evol. 26, 655 523-532. 656 McKey, D., 1974. Adaptive patterns in alkaloid physiology. The American Naturalist 108, 305-320. 657 Muller, B., Pantin, F., Genard, M., Turc, O., Freixes, S., Piques, M., Gibon, Y., 2011. Water deficits 658 uncouple growth from photosynthesis, increase C content, and modify the relationships between C and 659 growth in sink organs. Journal of Experimental Botany 62, 1715-1729. 660 Mundim, F.M., Pringle, E.G., 2018. Whole-plant metabolic allocation under water stress. Frontiers in 661 Plant Science 9, 852. 662 Nguyen, D., Rieu, I., Mariani, C., van Dam, N.M., 2016. How plants handle multiple stresses: hormonal 663 interactions underlying responses to abiotic stress and insect herbivory. Plant Mol. Biol. 91, 727-740. 664 Ogaya, R., Liu, D., Barbeta, A., Peñuelas, J., 2020. Stem mortality and forest dieback in a 20-years 665 experimental drought in a Mediterranean holm oak forest. Frontiers in Forests and Global Change 2. 666 Ogaya, R., Peñuelas, J., 2003. Comparative field study of Quercus ilex and Phillyrea latifolia: 667 photosynthetic response to experimental drought conditions. Environ. Exp. Bot. 50, 137-148. 668 Ogaya, R., Peñuelas, J., Asensio, D., Llusià, J., 2011. Chlorophyll fluorescence responses to temperature 669 and water availability in two co-dominant Mediterranean shrub and tree species in a long-term field 670 experiment simulating climate change. Environ. Exp. Bot. 71, 123-127. 671 Paine, T.D., Lieutier, F., 2016. Insects and Diseases of Mediterranean Forest Systems. Springer 672 International Publishing, Cham. 673 Palacio, S., Hoch, G., Sala, A., Körner, C., Millard, P., 2014. Does carbon storage limit tree growth? New 674 Phytologist 201, 1096-1100. 675 Panda, S., Kazachkova, Y., Aharoni, A., 2021. Catch-22 in specialized metabolism: balancing defense and 676 growth. Journal of Experimental Botany 72, 6027-6041. 677 Pashkovskiy, P.P., Vankova, R., Zlobin, I.E., Dobrev, P., Kartashov, A.V., Ivanova, A.I., Ivanov, V.P., 678 Marchenko, S.I., Nartov, D.I., Ivanov, Y.V., Kuznetsov, V.V., 2022. Hormonal responses to short-term and 679 long-term water deficit in native Scots pine and Norway spruce trees. Environ. Exp. Bot. 195, 104789. 680 Peñuelas, J., Sardans, J., Filella, I., Estiarte, M., Llusià, J., Ogaya, R., Carnicer, J., Bartrons, M., Rivas-Ubach, 681 A., Grau, O., Peguero, G., Margalef, O., Pla-Rabés, S., Stefanescu, C., Asensio, D., Preece, C., Liu, L., 682 Verger, A., Rico, L., Barbeta, A., Achotegui-Castells, A., Gargallo-Garriga, A., Sperlich, D., Farré-Armengol, 683 G., Fernández-Martínez, M., Liu, D., Zhang, C., Urbina, I., Camino, M., Vives, M., Nadal-Sala, D., Sabaté, S., 684 Gracia, C., Terradas, J., 2018. Assessment of the impacts of climate change on Mediterranean terrestrial 23 685 ecosystems based on data from field experiments and long-term monitored field gradients in Catalonia. 686 Environ. Exp. Bot. 152, 49-59. 687 Pieterse, C.M., Van der Does, D., Zamioudis, C., Leon-Reyes, A., Van Wees, S.C., 2012. Hormonal 688 modulation of plant immunity. Annu. Rev. Cell. Dev. Biol. 28, 489-521. 689 Pivovaroff, A.L., McDowell, N.G., Rodrigues, T.B., Brodribb, T., Cernusak, L.A., Choat, B., Grossiord, C., 690 Ishida, Y., Jardine, K.J., Laurance, S., Leff, R., Li, W., Liddell, M., Scott Mackay, D., Pacheco, H., Peters, J., 691 de J. Sampaio Filho, I., Souza, D.C., Wang, W., Zhang, P., Chambers, J., 2021. Stability of tropical forest 692 tree carbon-water relations in a rainfall exclusion treatment through shifts in effective water uptake 693 depth. Global Change Biology 27, 6454–6466. 694 Quirantes-Piné, R., Lozano-Sánchez, J., Herrero, M., Ibáñez, E., Segura-Carretero, A., Fernández- 695 Gutiérrez, A., 2013. HPLC–ESI–QTOF–MS as a powerful analytical tool for characterising phenolic 696 compounds in olive-leaf extracts. Phytochem. Anal. 24, 213-223. 697 Riemann, M., Dhakarey, R., Hazman, M., Miro, B., Kohli, A., Nick, P., 2015. Exploring jasmonates in the 698 hormonal network of drought and salinity responses. Frontiers in Plant Science 6, 1077. 699 Rowland, L., da Costa, A.C.L., Galbraith, D.R., Oliveira, R.S., Binks, O.J., Oliveira, A.A.R., Pullen, A.M., 700 Doughty, C.E., Metcalfe, D.B., Vasconcelos, S.S., Ferreira, L.V., Malhi, Y., Grace, J., Mencuccini, M., Meir, 701 P., 2015. Death from drought in tropical forests is triggered by hydraulics not carbon starvation. Nature 702 528, 119-122. 703 Ruehr, N.K., Offermann, C.A., Gessler, A., Winkler, J.B., Ferrio, J.P., Buchmann, N., Barnard, R.L., 2009. 704 Drought effects on allocation of recent carbon: from beech leaves to soil CO2 efflux. New Phytologist 184, 705 950-961. 706 Santacruz-García, A.C., Bravo, S., del Corro, F., García, E.M., Molina-Terrén, D.M., Nazareno, M.A., 2021. 707 How do plants respond biochemically to fire? The role of photosynthetic pigments and secondary 708 metabolites in the post-fire resprouting response. Forests 12, 56. 709 Savage, J.A., Clearwater, M.J., Haines, D.F., Klein, T., Mencuccini, M., Sevanto, S., Turgeon, R., Zhang, C., 710 2016. Allocation, stress tolerance and carbon transport in plants: how does phloem physiology affect 711 plant ecology? Plant, Cell & Environment 39, 709-725. 712 Schönbeck, L., Gessler, A., Hoch, G., McDowell, N.G., Rigling, A., Schaub, M., Li, M.-H., 2018. 713 Homeostatic levels of nonstructural carbohydrates after 13 yr of drought and irrigation in Pinus 714 sylvestris. New Phytologist 219, 1314-1324. 715 Senf, C., Buras, A., Zang, C.S., Rammig, A., Seidl, R., 2020. Excess forest mortality is consistently linked to 716 drought across Europe. Nature Communications 11, 6200. 717 Song, X.C., Canellas, E., Dreolin, N., Nerin, C., Goshawk, J., 2021. Discovery and characterization of 718 phenolic compounds in bearberry (Arctostaphylos uva-ursi) leaves using liquid chromatography-ion 719 mobility-high-resolution mass spectrometry. J. Agric. Food Chem. 720 Tattini, M., Gravano, E., Pinelli, P., Mulinacci, N., Romani, A., 2000. Flavonoids accumulate in leaves and 721 glandular trichomes of Phillyrea latifolia exposed to excess solar radiation. New Phytologist 148, 69-77. 722 Tattini, M., Guidi, L., Morassi-Bonzi, L., Pinelli, P., Remorini, D., Degl'Innocenti, E., Giordano, C., Massai, 723 R., Agati, G., 2005. On the role of flavonoids in the integrated mechanisms of response of Ligustrum 724 vulgare and Phillyrea latifolia to high solar radiation. New Phytologist 167, 457-470. 725 Tattini, M., Loreto, F., Fini, A., Guidi, L., Brunetti, C., Velikova, V., Gori, A., Ferrini, F., 2015. Isoprenoids 726 and phenylpropanoids are part of the antioxidant defense orchestrated daily by drought-stressed 727 Platanus × acerifolia plants during Mediterranean summers. New Phytologist 207, 613-626. 728 Thaler, J.S., Humphrey, P.T., Whiteman, N.K., 2012. Evolution of jasmonate and salicylate signal crosstalk. 729 Trends in Plant Science 17, 260-270. 730 Tienda-Parrilla, M., López-Hidalgo, C., Guerrero-Sanchez, V.M., Infantes-González, Á., Valderrama- 731 Fernández, R., Castillejo, M.-Á., Jorrín-Novo, J.V., Rey, M.-D., 2022. Untargeted MS-based metabolomics 24 732 analysis of the responses to drought stress in Quercus ilex L. leaf seedlings and the identification of 733 putative compounds related to tolerance. Forests 13, 551. 734 Tixier, A., Orozco, J., Roxas, A.A., Earles, J.M., Zwieniecki, M.A., 2018. Diurnal variation in nonstructural 735 carbohydrate storage in trees: remobilization and vertical mixing. Plant Physiology 178, 1602-1613. 736 Tripathi, S.K., Tuteja, N., 2007. Integrated signaling in flower senescence: an overview. Plant signaling & 737 behavior 2, 437-445. 738 Trowbridge, A.M., Adams, H.D., Collins, A., Dickman, L.T., Grossiord, C., Hofland, M., Malone, S., Weaver, 739 D.K., Sevanto, S., Stoy, P.C., McDowell, N.G., 2021. Hotter droughts alter resource allocation to chemical 740 defenses in piñon pine. Oecologia. 741 Trugman, A.T., Anderegg, L.D.L., Shaw, J.D., Anderegg, W.R.L., 2020. Trait velocities reveal that mortality 742 has driven widespread coordinated shifts in forest hydraulic trait composition. Proceedings of the 743 National Academy of Sciences 117, 8532-8538. 744 Tuel, A., Eltahir, E.A.B., 2020. Why is the Mediterranean a climate change hot spot? J. Clim. 33, 5829- 745 5843. 746 Wasternack, C., Strnad, M., 2019. Jasmonates are signals in the biosynthesis of secondary metabolites — 747 Pathways, transcription factors and applied aspects — A brief review. New Biotechnology 48, 1-11. 748 Wiley, E., Rogers, B.J., Hodgkinson, R., Landhäusser, S.M., 2016. Nonstructural carbohydrate dynamics of 749 lodgepole pine dying from mountain pine beetle attack. New Phytologist 209, 550-562. 750 Zhang, P., Zhou, X., Fu, Y., Shao, J., Zhou, L., Li, S., Zhou, G., Hu, Z., Hu, J., Bai, S.H., McDowell, N.G., 2020. 751 Differential effects of drought on nonstructural carbohydrate storage in seedlings and mature trees of 752 four species in a subtropical forest. Forest Ecology and Management 469, 118159. 753 Züst, T., Agrawal, A.A., 2017. Trade-offs between plant growth and defense against insect herbivory: an 754 emerging mechanistic synthesis. Annual Review of Plant Biology 68, 513-534. 755 Figure legends 756 Figure 1 Daily solar radiation (a), mean air temperature (b) and precipitation (c) during the study 757 period (2018–2019). Grey lines indicate the sample dates. Thick lines are smoothing splines with 758 a span value of 0.7. 759 Figure 2 Leaf mass (a) and springtime leaf mass increment (b), stem diameter (c) and annual 760 stem diameter increment (d) in Quercus ilex (green) and Phillyrea latifolia (blue) under control 761 (filled circles and solid lines) and drought (open circles and dashed lines) treatment. Shaded 762 areas represent 95% confidence intervals for each species. Each point represents the mean of 763 four plots, and error bars are SE. 764 Figure 3 Concentrations of total secondary metabolites (SMs; the sum of individual SMs) in 765 leaves (a) and stem phloem (b) in Quercus ilex (green) and Phillyrea latifolia (blue) under 766 control (filled circles and solid lines) and drought (open circles and dashed lines) treatment. 767 Shaded areas represent 95% confidence intervals for each species. Each point represents the 25 768 mean of four plots, and error bars are SE. Significant differences between control and drought 769 plots were indicated by asterisks (Student’s test, P<0.05). Levels of statistical significance (P- 770 values) of species, treatment, seasons, and their interactions on SMs are in shown Table S3 and 771 S4. 772 Figure 4 Concentrations of soluble sugars, starch, and nonstructural carbohydrates (NSCs) in 773 leaves (upper panels) and stem phloem (lower panels) in Quercus ilex (green) and Phillyrea 774 latifolia (blue) under control (filled circles and solid lines) and drought (open circles and dashed 775 lines) treatment. Shaded areas represent 95% confidence intervals for each species. Each point 776 represents the mean of four plots, and error bars are SE. Levels of statistical significance (P- 777 values) of species, treatment, seasons, and their interactions on sugars, starch and NSCs are 778 shown in Table S3 and S4. 779 Figure 5 Concentrations of abscisic acid (ABA), jasmonic acid (JA), and salicylic acid (SA) in 780 leaves (upper panels) and stem phloem (lower panels) in Quercus ilex (green) and Phillyrea 781 latifolia (blue) under control (filled circles and solid lines) and drought (open circles and dashed 782 lines) treatment. Shaded areas represent 95% confidence intervals for each species. Each circle 783 represents the mean of four plots, and error bars are SE. Significant differences between control 784 and drought plots were indicated by asterisks (Student’s test, P<0.05). Levels of statistical 785 significance (P-values) of species, treatment, seasons, and their interactions on ABA, JA and SA 786 are shown in Table S3 and S4. 787 Figure 6 Coefficients for relationships of secondary metabolites (SMs) to soluble sugars, starch, 788 abscisic acid (ABA), jasmonic acid (JA), and salicylic acid (SA) in leaves (upper panels) and 789 stem phloem (lower panels) for Quercus ilex (left panels) and Phillyrea latifolia (right panels). 790 Data are not shown for SA in leaves of Phillyrea latifolia and JA in the stem phloem of both 791 species due to very low concentrations. Significant relationships were indicated by asterisks 792 (P<0.05). 793 Figure 7 Principal component analysis showing the distribution of the variation among secondary 794 metabolites (SMs; blue arrows), leaf biomass or stem diameter (green arrows), soluble sugars 795 and starch (brown arrows), phytohormones (red arrows; abscisic acid, ABA; jasmonic acid, JA; 796 salicylic acid, SA) in leaves (upper panels) and stem phloem (lower panels). Each circle 26 797 represents the mean of four individual trees of each species (Quercus ilex, green; Phillyrea 798 latifolia, blue) for each plot (control, filled circles; drought, open circles). Ellipses represent the 799 95% confidence interval of the bi-variate distribution.