Plant Physiology and Biochemistry 88 (2015) 70e81
Contents lists available at ScienceDirect
Plant Physiology and Biochemistry journal homepage: www.elsevier.com/locate/plaphy
Research article
Water availability influences morphology, mycorrhizal associations, PSII efficiency and polyamine metabolism at early growth phase of Scots pine seedlings €kela € a, b, *, Jaana Vuosku b, Esa La €a €ra € c, Markku Saarinen a, Riina Muilu-Ma d b a €ggman , Tytti Sarjala Juha Heiskanen , Hely Ha a
Natural Resources Institute Finland (Luke), Parkano Research Unit, FI-39700 Parkano, Finland Department of Biology, University of Oulu, P.O. Box 3000, FI-90014 Oulu, Finland Department of Mathematical Sciences, University of Oulu, P.O. Box 3000, FI-90014 Oulu, Finland d Natural Resources Institute Finland (Luke), Suonenjoki Research Unit, FI-77600 Suonenjoki, Finland b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 26 November 2014 Accepted 27 January 2015 Available online 28 January 2015
Scots pine (Pinus sylvestris L.) is adapted to various soil types with diverse water availabilities. However, Scots pine seedlings are vulnerable to abiotic stress during the early growth, when they may be exposed to both dry and wet conditions. Here, we focused on the above and below ground coping strategies of Scots pine seedlings under controlled wet, optimal and dry soil conditions by investigating morphological traits including seedling biomass, number of root tips, proportion of mycorrhizal root tips and brown needles. In addition, we studied metabolic and physiological responses including gene expression involved in biosynthesis and catabolism of polyamines (PA), PSII efficiency and the expression of the catalase (CAT) late-embryogenesis abundant protein (LEA), pyruvate decarboxylase (PDC), glutamatecysteine ligase (GCL) and glutathione synthetase (GS) genes. We found that seedlings invested in shoots by maintaining stable shoot water content and high PSII efficiency under drought stress. Free and soluble conjugated putrescine (Put) accumulated in needles under drought stress, suggesting the role of Put in protection of photosynthesizing tissues. However, the expression of the PA biosynthesis genes, arginine decarboxylase (ADC), spermidine synthase (SPDS) and thermospermine synthase (ACL5) was not affected under drought stress whereas catabolizing genes diamino oxidase (DAO) and polyamine oxidase (PAO) were down-regulated in shoots. The morphology of the roots was affected by peat water content. Furthermore, both drought stress and water excess restricted the seedling ability to sustain a symbiotic relationship. The consistent pattern of endogenous PAs seems to be advantageous to the Scots pine seedlings also under stress conditions. © 2015 Elsevier Masson SAS. All rights reserved.
Keywords: Water stress Mycorrhiza Pinus Polyamine PSII efficiency Scots pine
Abbreviations: ADC, arginine decarboxylase; CAT, catalase; CI, confidence interval; Cp, crossing point; DAO, diamine oxidase; DW, dry weight; FW, fresh weight; F0, minimum fluorescence; Fv, variable fluorescence; Fm, maximum fluorescence; GAPDH, glyceraldehyde-tri-phosphate dehydrogenase; GS, glutathione synthetase; LEA, late embryogenesis abundant protein; MD, mean difference; ODC, ornithine decarboxylase; PA, polyamine; PAO, polyamine oxidase; PDC, pyruvate decarboxylase; PSII, photosystem II; Put, putrescine; SAMDC, S-adenosyl methionine decarboxylase; Spd, spermidine; SPDS, spermidine synthase; Spm, spermine; SPMS, spermine synthase; Tspm, thermospermine; TSPMS, thermospermine synthase; TUBA, a-tubulin; UBQ, ubiquitin. * Corresponding author. Natural Resources Institute Finland (Luke), Kaironiementie 15, FI-39700 Parkano, Finland. €), jaana.vuosku@ E-mail addresses: riina.muilu-makela@luke.fi (R. Muilu-M€ akela €a €ra €), markku.saarinen@luke.fi luke.fi (J. Vuosku), esa.laara@oulu.fi (E. La (M. Saarinen), juha.heiskanen@luke.fi (J. Heiskanen), hely.haggman@oulu.fi €ggman), tytti.sarjala@luke.fi (T. Sarjala). (H. Ha http://dx.doi.org/10.1016/j.plaphy.2015.01.009 0981-9428/© 2015 Elsevier Masson SAS. All rights reserved.
1. Introduction Scots pine (Pinus sylvestris L.) is the most widely distributed Eurasian conifer and an economically important source of timber. Scots pine inhabits large variety of soil types, coping with diverse water availabilities. However, pine seedlings are most vulnerable to abiotic stress factors during their early growth, when they may be exposed to both dry and wet conditions. Especially on organic soils water level and peat water retention capacity has unique impact on pine (Pinus sp.) forest regeneration success (Saarinen, 2013). Imbalance in root water uptake and leaf transpiration leads to dehydration of plant tissues (Aroca et al., 2012). Pines are considered as water-savers. They prevent transpiration and oxidative
€kela € et al. / Plant Physiology and Biochemistry 88 (2015) 70e81 R. Muilu-Ma
damage in tissues by strong stomatal control (Poyatos et al., 2008). Although, Scots pine is a xerophyte, the limited plasticity of xylem and needle properties has been suggested to explain its vulnerability to severe decline of water (Martinez-Vilalta et al., 2009). Moreover, some forest trees increase water uptake by extending their root capacity under arid conditions (Hamanishi and Campbell, 2011). Under water excess, instead, transport of oxygen and carbon dioxide can be hindered, which e.g. reduces photosynthesis (Aroca et al., 2012). The lack of oxygen (O2) in the rhizosphere in flooding conditions switches energy production from mitochondrial respiration to fermentation in plant cells. Up-regulation of pyruvate decarboxylase (PDC) gene expression was used to indicate alcohol fermentation in flooding tolerant grey poplar (Populus canescence) tissues (Kreuzwieser et al., 2009) under hypoxia. In pines, dehydration induces many genes such as lateembryogenesis abundant proteins (LEA), dehydrins (DHN), catalase (CAT), water-stress-inducible proteins (Ip2-3) and early response to drought protein 3 (ERD3) (Lorenz et al., 2005, 2011). LEA proteins, the well-known osmoprotectors (Battaglia et al., 2008), are reported to accumulate under extreme desiccation conditions in higher plants (Yang et al., 2012) including pines (Lorenz et al., 2011; Velasco-Conde et al., 2012). LEA proteins have been grouped to various families on the basis of sequence similarity (Battaglia et al., 2008). Dehydrins belong to group II of LEA proteins and are important during acclimation of pines (Velasco-Conde et al., 2012). Abiotic stresses cause oxidative stress in plant cells, where one of the first responses is accumulation of reactive oxygen species (ROS), like hydrogen peroxide (H2O2). ROS are important signalling molecules, but in high concentrations they are toxic. CAT is an antioxidant enzyme, which dissociates H2O2 to water and O2 (Mhamdi et al., 2012). In plant cells the CAT is mainly localized in peroxisomes and has an important role in ROS homoeostasis regulation (Mhamdi et al., 2012). Increased activity of CAT has been detected in drought-stressed pines (Lorenz et al., 2005). Furthermore, the oxidative stress induces plants to produce glutathione. The balance between reduced and oxidized forms of glutathione is a central component in maintaining redox state in plant cells (Galant et al., 2011). Glutathione is produced by two step reaction where first GCL catalyses the formation of g-glutamylcysteine from cysteine and glutamate and then GS catalyses the addition of glycine to g-glutamylcysteine to yield glutathione. PA metabolic route is well known in plants (see Supplementary Fig. S1). PAs, Put, Spd, spermine (Spm) and thermospermine (Tspm), are small, positively charged nitrogenous compounds having an important but intricate role in stress and developmental pathways in plants, recently reviewed by Tiburcio et al. (2014). PA metabolism related genes of Arabidopsis have been divided to different categories with stress or developmentally related genes (Tiburcio et al., 2014). For instance, expression of ADC2, SPMS and most PAOs correlate positively with stress related genes in Arabidopsis, whereas expression of ADC1, SPDSs and ACL5 co-express with other genes induced during development (Tiburcio et al., 2014). Thus, both PA biosynthesis and catabolism seem to play a role in stress and development of plants (Tiburcio et al., 2014). PA levels are strictly regulated by biosynthesis, degradation, conjugation, back-conversion and transport and by interactions with other pathways in response to stress (Tiburcio et al., 2014). The PA metabolic route intermediates nitrogen metabolism, and has interactions with other metabolites, including stress protective compounds, hormones and signalling molecules (Moschou et al., 2012). For instance, accumulation of PAs by abscisic acid (ABA) treatment enhanced PA oxidation, which in turn launched protective effects such as stomatal closure in grapevine (Vitis vinifera) (Toumi et al., 2010). Furthermore, ACL5 and Tspm have a specific role in vascular development and stem elongation in Arabidopsis
71
(Vera-Sirera et al., 2010) i.e. in PA mediated developmental processes in plants. Here, we propose that water stress induced responses may be linked to PA metabolism especially during the vulnerable early growth phases of Scots pine. Therefore; we investigated young 6 to 17-week-old pine seedlings in controlled wet, optimal and dry conditions and monitored the changes in biomass, phenotype, and chlorophyll fluorescence (Fv/Fm). We measured the expression of LEA and PDC genes to describe the severity of the drought stress and water excess, respectively. The expression of CAT, GCL and GS genes was measured to indicate oxidative stress reactions under different treatments. The stress induced changes in the PA gene expressions and seedling morphology were compared to the optimal condition to elucidate the role of PA metabolism under different water levels (wet, optimal and dry) in needles, stems and roots. We found that the unfavourable water conditions changed the seedling morphology, disturbed symbiotic relationships and induced stress related gene expression. However, PA contents were mostly maintained at constant levels which underlines the importance of the strict regulation of PA metabolism for the early development of Scots pine seedlings. 2. Materials and methods 2.1. Plant material and growth conditions Scots pine (P. sylvestris L.) seed orchard (Hiirola, Finland 61490 N, 27150 E) seeds were sterilized with 30% H2O2. Seeds, 80 seeds per pot, were sown in twelve 3 dm3, 16.5 cm tall plastic pots (i.e. altogether 960 seeds per 12 pots) (VG-Products, VG-potter, Billund, Denmark) filled with 690 g of 1:9 combination of fertilized (Kekkil€ a Oy, Vantaa, Finland) and non-fertilized (Biolan Oy, Kauttua, Finland) horticultural peat. Before sowing, the peat in the pots was watered properly with 1.5 dm3 of H2O and left on a grate until water no longer drained away. The pots were irrigated two to three times a week during the first six weeks. Nutrients [KH2PO4 25 mg/l, (NH4)2HPO4 12.5 mg/l, CaCl2 2.5 mg/l, NaCl 1.25 mg/g, MgSO4 7H2O 7.5 mg/l, FeCl3 6H2O 0.6 mg/l] were added to irrigation water. Within two weeks the seeds started to germinate and before the first sampling, the average length of the seedlings was 10 ± 4 cm. The photoperiod in the growth chamber was 6 h darkness and 18 h daylight with light intensity of ca. 200 mmol m2 s1 and temperature 25 C during the experiment. Water content of the peat in all pots was ca. 75%. 2.2. Watering treatments and harvesting The watering treatments were based on the water retention characteristics of the peat (Heiskanen, 1993) determined with a pressure plate apparatus (soil moisture Equipment Corp., USA). The water retention curve indicates the level of suction required by the plant to get water from the soil as it dries (see Supplementary Fig. S2). The following watering treatments were planned to describe water availability to the pine seedlings: wet, optimal and dry. In the wet treatment, the volumetric water content of the peat was kept as high as possible without reaching the minimum airfilled porosity level (20 vol.%) (Wall and Heiskanen, 2003). In the dry treatment, the water content was kept above the wilting point (Hillel, 1971), which here was 9.7 vol.%. The optimal water availability was targeted to be around field capacity (10 kPa matric potential) (Hillel, 1971). It is noteworthy that both wet and optimal water treatments are within the limits where growth is unrestricted (see Supplementary Fig. S2). Thus, after first harvesting (week one), four pots were kept at a high water content (75% ¼ wet
72
€kela € et al. / Plant Physiology and Biochemistry 88 (2015) 70e81 R. Muilu-Ma
treatment ¼ water excess) and irrigated two to three times a week. The water content in four pots was mimicking optimal water content (40% ¼ optimal treatment) in natural conditions in northern pine forests. The optimal water content was reached in five weeks. The four remaining pots were allowed to dry one more week until a low water content (15% ¼ dry treatment ¼ drought stress) was reached (Fig. 1). The water content of the peat in the pots was monitored with ThetaProbe (HH2 Moisture meter, Delta T Devices Ltd, Cambridge, UK) and by weighing the peat pots before and after watering to ensure the targeted water contents. The ThetaProbe meter was calibrated gravimetrically at desorption with the same peat and pots as used in the experiment. Samples for free and soluble conjugated PA concentration measurements as well as for gene expression analyses were collected three times during the experiment. The first sampling during the first week of the treatments represents an initial phase, where the water content of the peat was above 75% in all pots. At the initial stage, twelve needles per pot were collected for PA and RNA analyses. Samples of the roots and stems were combined from four seedlings and four to five samples per pot were used to describe the initial phase. On the second (5th week) and third sampling dates (11th week), PA and RNA samples from needles, stems and roots were collected from all twelve pots. The PA and RNA samples were frozen immediately in liquid nitrogen and stored at 80 C until extracted and analysed. 2.3. Morphological parameters After the last harvesting, the fresh weight (FW) and dry weight (DW) (samples dried overnight at þ105 C) of the roots and shoots (¼stem with needles) of four seedlings from each pot (n ¼ 12) were measured. The total number of the root tips was calculated under stereomicroscope also from four seedlings per pot. Because nonsterilized peat was used as growth substrate, mycorrhizal infection could not be excluded; therefore, the number of the mycorrhizal root tips was also counted. The needles of four seedlings per pot were categorized according to their colour: green, greenish brown, brown. 2.4. Chlorophyll fluorescence Chlorophyll fluorescence has been regarded as a tool for interpreting stress tolerance of plants (Peeva and Cornic, 2009). As a non-destructive method, chlorophyll fluorescence, especially the
ratio of variable (Fv) to maximum fluorescence (Fm), has been widely used for assessing plant physiological status and the state of Photosystem II (PSII) (Krause and Weis, 1991). Chlorophyll fluorescence emission from green needles was measured by means of a Plant Efficiency Analyzer (Hansatech Instruments Ltd., Norfolk, England). The ratio of variable (Fv) to maximum fluorescence (Fm) was measured from three seedlings per pot. Needles were dark adapted at RT in leaf clips supplied with the analyser for at least 30 min before measurement. The minimum fluorescence (F0), the maximum fluorescence (Fm) and the variable fluorescence (Fv ¼ FmeF0) as well as the ratio Fv/Fm were recorded with equipment settings of 15 s at 100% intensity level of photon flux density (4000 mmol m2 s1). 2.5. Polyamine analysis The PA samples, consisting of about 100 mg of plant material per sample, were extracted in 5% (w/v) perchloric acid. Crude extracts for free PAs and hydrolysed supernatant for perchloric acid-soluble conjugated PAs were dansylated and separated by HPLC according et al. (1999). The PA to Sarjala and Kaunisto (1993) and Fornale concentrations were expressed as nmol g1 FW of plant tissue. The present study does not distinguish between Spm and possible thermospermine (Tspm) (see Supplementary Fig. S1), rather they are both included in the same fraction and referred to as Spm. At the end of the experiment the PA concentrations were estimated also in dry weight basis by using the average dry weight percentage values of the roots and shoots. Thus, the PA concentrations in dry weight basis are not as accurate estimations as the values in FW basis. 2.6. RNA isolation and reverse transcription The total RNA of Scots pine needle, root, and stem tissues was extracted from 100 mg of plant material using total RNA purification PureLink™ Plant RNA Reagent (Invitrogen Corporation, California, USA) according to the manufacturer's instructions. The RNA samples were treated with rDNase set (MachereyeNagel, Duren, Germany) at 37 C for 10 min in order to eliminate contaminating genomic DNA. The amount of DNase used to produce DNA-free RNA samples was three times higher than recommended by the manufacturer. The RNA samples were purified with the NucleoSpin® RNA Clean-Up kit (MachereyeNagel, Duren, Germany). The RNA yields were measured three times with OD260 analysis using spectrophotometer. cDNA was prepared from 1 mg of total RNA, which was reverse transcribed by SuperScript VILO™ cDNA synthesis kit (Invitrogen Corporation, California, USA). 2.7. Quantitative real-time PCR analysis
Fig. 1. Water content (%) of the peat during the experiment indicating wet, optimal and dry growing conditions. Initial sampling in the 1st week, second sampling in the 5th week, third sampling in the 11th week.
The real-time PCR primers for the gene expression studies (see Supplementary Table S1) on PA biosynthesis genes ADC, ODC, SPDS, ACL5, DAO and PAO were designed against Scots pine PA gene sequences ADC [GenBank: HM236823.1], ornithine decarboxylase (ODC) [GenBank: HM236831], SPDS [GenBank: HM236827], ACL5 [GenBank: HM236828], DAO [GenBank: HM236829] and PAO [GenBank: HM236830]. The primers of the housekeeping genes ubiquitin (UBQ), glyseraldehyde-3-phosphate dehydrogenase (GAPDH) and alfa-tubulin (TUBA) were based on the maritime pine (Pinus pinaster Ait.) putative UBQ sequence [GenBank: AF461687], Scots pine GAPDH sequence [GenBank: L07501] and Scots pine TUBA sequence [GenBank: FN546172], respectively. Primers for the LEA gene were based on Scots pine sequence [GenBank: FJ201571], which has been used as a candidate gene to cold acclimation due to its function in dehydrative stress (Wachowiak et al., 2009).
€kela € et al. / Plant Physiology and Biochemistry 88 (2015) 70e81 R. Muilu-Ma
73
Fig. 2. Relative gene expressions of LEA and CAT related to reference gene UBQ in needles, stems and roots of the Scots pine seedlings under three watering treatments (solid circle, wet; asterisk, optimal; open circle, dry). The geometric mean values of the four replicates are connected with solid, dashed, and dotted lines for the wet, optimal, and dry treatments, respectively. There are four replicates per treatment and sampling date except at the first sampling, where growth conditions in all pots were still similar. The replicates represent the seedlings grown in different pots but under the same treatment. The arrows indicate significant differences compared to optimal treatment (see Supplementary Table S2).
According to its translated protein sequence the LEA gene belongs to the same LEA II superfamily with loblolly pine (Pinus taeda) LEA protein [AAX68990.1] expressed under drought stress (Lorenz et al., 2011). The primers for the CAT gene were based on the Scots pine sequence [GenBank: EU513163]. The primers for the PDC gene were based on putative PDC mRNA sequence of root flooded loblolly pine [GenBank: CO161777.1], which was similar with sequence of black cottonwood (Populus trichocarpa) PDC family protein [GenBank: XM_002308194.2]. The primers for the GCL were based on Norway spurce (Picea abies) sequence [GenBank: AJ132540.1] and for GS on the putative GS mRNA sequence of well-watered loblolly pine roots [GenBank: CF478539.1], which was similar with the grey poplar GS sequence [Potri.013G026800.2]. The expression of the PA genes and the stress related genes were normalized by the gene expression derived from the three different housekeeping genes UBQ, GAPDH and TUBA, which showed very similar expression trends. The UBQ had the most stable Cp values during the experiment and therefore the UBQ normalized and statistically analysed gene expression results are presented. The quantification of mRNA with real-time PCR was performed in 20 mL of reaction mixture composed of 2 mL of cDNA, 10 mL Light Cycler® 480 SYBR Green I Master Mix, and 100 nM gene-specific primers. PCR amplification was initiated by incubation at 95 C for 10 min followed by 40 cycles: 30 s at 95 C, 1 min at 58 C, and 1 min at 72 C (LightCycler® 480/96, Roche Diagnostics Ltd.). The PCR conditions were optimized for high amplification efficiency 90% for all primer pairs used. Every PCR reaction was done in triplicate and the results were produced with LightCycler® 480 version 1.5.0.39 software.
2.8. Statistical methods The number of root tips, the proportion of mycorrhizal root tips and PA contents in DW basis were compared by computing the pairwise mean differences between the treatment groups and their 95% confidence intervals based on t-distribution. The amount of green, brown and greenbrown needles under different treatments were statistically analysed by Pearson's Chi-squared test, where the needle colour composition under optimal treatment was compared with wet and dry treatments at the week 11. The effects of treatment and time under treatment, and their interactions on PA concentrations and gene expression were analysed by linear mixed modelling (Fitzmaurice et al., 2004). Individual models were fitted for free and soluble conjugated fractions of Put, Spd and Spm, separately in needles, stems and roots. Similar models were fitted for relative expression of ADC, SPDS, ACL5, DAO, PAO, LEA, PDC, GCL, GS and CAT, respectively, each again in needles, stems and roots. Before model fitting, the original response variables (PA level or relative expression) were transformed to the natural logarithm scale. Both treatment and time under treatment were included as categorical factors with fixed effects, such that their reference levels were optimal treatment and week 5, respectively. Accordingly, the two main effect parameters for watering treatment refer to the mean differences in the log-response: wet vs. optimal, and dry vs. optimal, respectively, both at week 5. Similarly, the two main effects of time under treatment describe analogous contrasts: week 1 vs. week 5, and week 11 vs. week 5, respectively, both under optimal treatment. Treatment time interactions were specified by four product terms of the pertinent
€kela € et al. / Plant Physiology and Biochemistry 88 (2015) 70e81 R. Muilu-Ma
74
Table 1 Water content and dry weight (mg) of the seedlings at the end of the experiment. Treatment
Wet Optimal Dry
Water content (%)
Dry weight (mg)
Shoot
Shoot
Root
3. Results 3.1. Expression of LEA gene indicated drought stress
Root
Mean
Range
Mean
Range
Mean
Range
Mean
Range
62 61 46
58e65 57e64 21e63
82 69 55
75e87 49e78 41e64
35 32 32
26e46 22e49 16e58
12 12 12
7e17 5e19 6e22
main effects. Thus, for example, an interaction term “dry vs. opt: wk 11 vs. 5” represents the difference between dry and optimal treatments in the mean change of the log-response from week 5 to week 11. To allow the likely correlation across the repeated measurements within each pot, random intercept terms for the 12 pots were included in the model upon the fixed effects above. The models were fitted using function lme in package nlme (Pinheiro et al., 2012) of the R environment, version 2.15.1 (R Development Core Team, 2012). The estimated regression coefficients and their 95% confidence intervals (CI) were back-transformed to the original scale of PA concentration or gene expression, such that apart from the overall intercept (“grand mean”), each estimate describes the relative change associated with the pertinent main effect or interaction effect with respect to the relevant reference level.
LEA gene was up-regulated in all parts of the Scots pine seedlings which were suffering from drought stress (Fig. 2 and Supplementary Table S2). The estimated relative changes on expression of LEA and their 95% confidence intervals associated with the effects of watering treatment, time, and their interactions are reported in Supplementary Table S2. Compared to the optimal treatment, the relative expression of LEA under the drought stress was 3.8-, 2.4-, and 3.8-fold higher in needles, stems and roots, respectively. The expression of CAT remained stable in needles but decreased in stems and roots towards the end of the experiment (Fig. 2 and Supplementary Table S2). Generally, CAT expression was not induced by drought stress or water excess. Only in roots, CAT expression was slightly induced by dry treatment being 1.8-fold higher than in optimal treatment. The expression of GCL was too low to be investigated (See low Cp values in Supplementary Table S3) and the GS expression levels remained stable in all plant organs throughout the experiment (Supplementary Table S2). This reveals that glutathione synthesis pathway was not transcriptionally up-regulated in drought-stressed seedlings. The PDC gene, which indicates alcohol fermentation under hypoxia, was not up-regulated in the roots under wet treatment. In the needles the
Fig. 3. (a) The number of the root tips in Scots pine seedlings per mg of dry weight and (b) the proportion of mycorrhizal root tips relative to the total number of root tips (%) at the end of the experiment.
Fig. 4. The effects of the dry (open circle), optimal (asterisk) and wet (solid circle) treatments (a) on needle colour of the Scots pine seedlings and (b) on the ratio of variable to maximum fluorescence (Fv/Fm) in the needles as a function of time. The number of replicates is four and each symbol represents the mean value of one to three measurements from seedlings grown in the same pot. The values are connected with solid, dashed, and dotted lines for the wet, optimal and dry treatments, respectively.
€kela € et al. / Plant Physiology and Biochemistry 88 (2015) 70e81 R. Muilu-Ma
75
Fig. 5. Free (a) and soluble conjugated PA (b) concentrations (nmol/gFW) in needles, stems and roots of the Scots pine seedlings under wet (solid circle), optimal (asterisk), and dry (open circle) treatments. The geometric mean values of the four replicates are connected with solid, dashed, and dotted lines for the wet, optimal, and dry treatments, respectively. There are four replicates per treatment and sampling date except at the first sampling, where growth conditions in all pots were still similar. The replicates represent the seedlings grown in different pots but under the same treatment. The arrows indicate significant differences compared to optimal treatment (see Supplementary Table S4).
expression of PDC increased towards the end of the experiment in all treatments (Supplementary Table S2). 3.2. Water levels affected root morphology and mycorrhizal association Physiological and morphological parameters indicated a perturbation of the early development of Scots pine seedlings under drought stress. The water content % (w/w) of the shoots and roots of the seedlings decreased under dry treatment, but the dry mass of the seedlings was rather equal (similar) in all treatments (Table 1). The water content in pine shoots remained relatively stable during the experiment, whereas the roots showed a linear
decrease in the water content along the decreasing water gradient in peat. Extreme drought in peat at the end of the experiment (week eleven) finally led to a collapse of water content (21e63%) also in the shoots. The morphology of the root system was affected by the watering treatments. The total number of root tips was lower under the drought stress than in the optimal (mean difference MD 3.3, 95% confidence interval, CI 5.4 to 1.2) or in the wet treatment (MD 4.0, 95% CI 6.6 to 1.4; Fig. 3a). In dry treatment root branches were also withered compared to the optimal and wet treatments. The highest proportion of mycorrhizal root tips was observed in the optimal treatment (compared to wet treatment: MD 13.2%, 95% CI 8.6%e17.8%; compared to dry treatment: MD
76
€kela € et al. / Plant Physiology and Biochemistry 88 (2015) 70e81 R. Muilu-Ma
Fig. 5. (continued).
10.1%, 95% CI 6.3%e13.9%; Fig. 3b) suggesting that both drought stress and water excess restricted the ability to sustain a symbiotic relationship. The proportion of green needles was over 65% in all treatments just after the target water content was reached in peat (week five) (Fig. 4a). Six weeks later (week eleven), the proportion of green needles was 63%, 42% and 32% under optimal, wet and dry treatments, respectively. When compared to optimal treatment, the composition of green, brown and greenbrown needles was statistically different under wet [X-squared value ¼ 11.5982, df ¼ 2 and p-value ¼ 0.00303] and dry treatment [X-squared ¼ 20.8404, df ¼ 2, p-value ¼ 0.00] at the week 11. However, relatively high chlorophyll fluorescence (Fv/Fm) values were detected in the green needles under dry and wet treatments (Fig. 4b). In addition to the decreased number of green needles, increased variation in Fv/Fm
between the seedlings in the wet treatment was apparent. Nevertheless, the average decrease in Fv/Fm was not as great as in the dry treatment (Fig. 4b). Under the drought stress, Fv/Fm remained quite stable until crashed when the peat water potential had decreased (9.7 vol.%) below the wilting point. 3.3. Changes in PA content The results on PA levels in the wet, optimal and dry treatments in FW basis are displayed in Fig. 5a and b. The estimated relative changes with 95% confidence intervals associated with the effects of watering treatment and time and their interactions are reported in Supplementary Table S4. In general, a decreasing trend was apparent in the free PA levels of needles and roots over time under all treatments, but in stems the downward trend appeared only
€kela € et al. / Plant Physiology and Biochemistry 88 (2015) 70e81 R. Muilu-Ma
after five weeks (Fig. 5a and Supplementary Table S4). Put was the most abundant PA in the seedlings (100e500 nmol/gFW). Free Put levels increased in needles under drought stress from week five onwards, being 1.6-fold concentration in dry than optimal treatment. In the roots free Put levels were slightly higher under the dry than optimal treatment throughout the experiment. The concentration of free Spd was higher in needles and roots (10e100 nmol/ gFW) than in stems (5e20 nmol/gFW) (Fig. 5a). In needles, free Spd had a decreasing trend under all treatments. In stems, the amount of free Spd decreased from week five onwards under the optimal and dry treatments, whereas it remained stable under the water excess (see Supplementary Table S4). In roots, the concentration of free Spd decreased more sharply under the optimal than other treatments from week five onwards. Overall free Spm concentrations were low in the seedlings (
Contents lists available at ScienceDirect
Plant Physiology and Biochemistry journal homepage: www.elsevier.com/locate/plaphy
Research article
Water availability influences morphology, mycorrhizal associations, PSII efficiency and polyamine metabolism at early growth phase of Scots pine seedlings €kela € a, b, *, Jaana Vuosku b, Esa La €a €ra € c, Markku Saarinen a, Riina Muilu-Ma d b a €ggman , Tytti Sarjala Juha Heiskanen , Hely Ha a
Natural Resources Institute Finland (Luke), Parkano Research Unit, FI-39700 Parkano, Finland Department of Biology, University of Oulu, P.O. Box 3000, FI-90014 Oulu, Finland Department of Mathematical Sciences, University of Oulu, P.O. Box 3000, FI-90014 Oulu, Finland d Natural Resources Institute Finland (Luke), Suonenjoki Research Unit, FI-77600 Suonenjoki, Finland b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 26 November 2014 Accepted 27 January 2015 Available online 28 January 2015
Scots pine (Pinus sylvestris L.) is adapted to various soil types with diverse water availabilities. However, Scots pine seedlings are vulnerable to abiotic stress during the early growth, when they may be exposed to both dry and wet conditions. Here, we focused on the above and below ground coping strategies of Scots pine seedlings under controlled wet, optimal and dry soil conditions by investigating morphological traits including seedling biomass, number of root tips, proportion of mycorrhizal root tips and brown needles. In addition, we studied metabolic and physiological responses including gene expression involved in biosynthesis and catabolism of polyamines (PA), PSII efficiency and the expression of the catalase (CAT) late-embryogenesis abundant protein (LEA), pyruvate decarboxylase (PDC), glutamatecysteine ligase (GCL) and glutathione synthetase (GS) genes. We found that seedlings invested in shoots by maintaining stable shoot water content and high PSII efficiency under drought stress. Free and soluble conjugated putrescine (Put) accumulated in needles under drought stress, suggesting the role of Put in protection of photosynthesizing tissues. However, the expression of the PA biosynthesis genes, arginine decarboxylase (ADC), spermidine synthase (SPDS) and thermospermine synthase (ACL5) was not affected under drought stress whereas catabolizing genes diamino oxidase (DAO) and polyamine oxidase (PAO) were down-regulated in shoots. The morphology of the roots was affected by peat water content. Furthermore, both drought stress and water excess restricted the seedling ability to sustain a symbiotic relationship. The consistent pattern of endogenous PAs seems to be advantageous to the Scots pine seedlings also under stress conditions. © 2015 Elsevier Masson SAS. All rights reserved.
Keywords: Water stress Mycorrhiza Pinus Polyamine PSII efficiency Scots pine
Abbreviations: ADC, arginine decarboxylase; CAT, catalase; CI, confidence interval; Cp, crossing point; DAO, diamine oxidase; DW, dry weight; FW, fresh weight; F0, minimum fluorescence; Fv, variable fluorescence; Fm, maximum fluorescence; GAPDH, glyceraldehyde-tri-phosphate dehydrogenase; GS, glutathione synthetase; LEA, late embryogenesis abundant protein; MD, mean difference; ODC, ornithine decarboxylase; PA, polyamine; PAO, polyamine oxidase; PDC, pyruvate decarboxylase; PSII, photosystem II; Put, putrescine; SAMDC, S-adenosyl methionine decarboxylase; Spd, spermidine; SPDS, spermidine synthase; Spm, spermine; SPMS, spermine synthase; Tspm, thermospermine; TSPMS, thermospermine synthase; TUBA, a-tubulin; UBQ, ubiquitin. * Corresponding author. Natural Resources Institute Finland (Luke), Kaironiementie 15, FI-39700 Parkano, Finland. €), jaana.vuosku@ E-mail addresses: riina.muilu-makela@luke.fi (R. Muilu-M€ akela €a €ra €), markku.saarinen@luke.fi luke.fi (J. Vuosku), esa.laara@oulu.fi (E. La (M. Saarinen), juha.heiskanen@luke.fi (J. Heiskanen), hely.haggman@oulu.fi €ggman), tytti.sarjala@luke.fi (T. Sarjala). (H. Ha http://dx.doi.org/10.1016/j.plaphy.2015.01.009 0981-9428/© 2015 Elsevier Masson SAS. All rights reserved.
1. Introduction Scots pine (Pinus sylvestris L.) is the most widely distributed Eurasian conifer and an economically important source of timber. Scots pine inhabits large variety of soil types, coping with diverse water availabilities. However, pine seedlings are most vulnerable to abiotic stress factors during their early growth, when they may be exposed to both dry and wet conditions. Especially on organic soils water level and peat water retention capacity has unique impact on pine (Pinus sp.) forest regeneration success (Saarinen, 2013). Imbalance in root water uptake and leaf transpiration leads to dehydration of plant tissues (Aroca et al., 2012). Pines are considered as water-savers. They prevent transpiration and oxidative
€kela € et al. / Plant Physiology and Biochemistry 88 (2015) 70e81 R. Muilu-Ma
damage in tissues by strong stomatal control (Poyatos et al., 2008). Although, Scots pine is a xerophyte, the limited plasticity of xylem and needle properties has been suggested to explain its vulnerability to severe decline of water (Martinez-Vilalta et al., 2009). Moreover, some forest trees increase water uptake by extending their root capacity under arid conditions (Hamanishi and Campbell, 2011). Under water excess, instead, transport of oxygen and carbon dioxide can be hindered, which e.g. reduces photosynthesis (Aroca et al., 2012). The lack of oxygen (O2) in the rhizosphere in flooding conditions switches energy production from mitochondrial respiration to fermentation in plant cells. Up-regulation of pyruvate decarboxylase (PDC) gene expression was used to indicate alcohol fermentation in flooding tolerant grey poplar (Populus canescence) tissues (Kreuzwieser et al., 2009) under hypoxia. In pines, dehydration induces many genes such as lateembryogenesis abundant proteins (LEA), dehydrins (DHN), catalase (CAT), water-stress-inducible proteins (Ip2-3) and early response to drought protein 3 (ERD3) (Lorenz et al., 2005, 2011). LEA proteins, the well-known osmoprotectors (Battaglia et al., 2008), are reported to accumulate under extreme desiccation conditions in higher plants (Yang et al., 2012) including pines (Lorenz et al., 2011; Velasco-Conde et al., 2012). LEA proteins have been grouped to various families on the basis of sequence similarity (Battaglia et al., 2008). Dehydrins belong to group II of LEA proteins and are important during acclimation of pines (Velasco-Conde et al., 2012). Abiotic stresses cause oxidative stress in plant cells, where one of the first responses is accumulation of reactive oxygen species (ROS), like hydrogen peroxide (H2O2). ROS are important signalling molecules, but in high concentrations they are toxic. CAT is an antioxidant enzyme, which dissociates H2O2 to water and O2 (Mhamdi et al., 2012). In plant cells the CAT is mainly localized in peroxisomes and has an important role in ROS homoeostasis regulation (Mhamdi et al., 2012). Increased activity of CAT has been detected in drought-stressed pines (Lorenz et al., 2005). Furthermore, the oxidative stress induces plants to produce glutathione. The balance between reduced and oxidized forms of glutathione is a central component in maintaining redox state in plant cells (Galant et al., 2011). Glutathione is produced by two step reaction where first GCL catalyses the formation of g-glutamylcysteine from cysteine and glutamate and then GS catalyses the addition of glycine to g-glutamylcysteine to yield glutathione. PA metabolic route is well known in plants (see Supplementary Fig. S1). PAs, Put, Spd, spermine (Spm) and thermospermine (Tspm), are small, positively charged nitrogenous compounds having an important but intricate role in stress and developmental pathways in plants, recently reviewed by Tiburcio et al. (2014). PA metabolism related genes of Arabidopsis have been divided to different categories with stress or developmentally related genes (Tiburcio et al., 2014). For instance, expression of ADC2, SPMS and most PAOs correlate positively with stress related genes in Arabidopsis, whereas expression of ADC1, SPDSs and ACL5 co-express with other genes induced during development (Tiburcio et al., 2014). Thus, both PA biosynthesis and catabolism seem to play a role in stress and development of plants (Tiburcio et al., 2014). PA levels are strictly regulated by biosynthesis, degradation, conjugation, back-conversion and transport and by interactions with other pathways in response to stress (Tiburcio et al., 2014). The PA metabolic route intermediates nitrogen metabolism, and has interactions with other metabolites, including stress protective compounds, hormones and signalling molecules (Moschou et al., 2012). For instance, accumulation of PAs by abscisic acid (ABA) treatment enhanced PA oxidation, which in turn launched protective effects such as stomatal closure in grapevine (Vitis vinifera) (Toumi et al., 2010). Furthermore, ACL5 and Tspm have a specific role in vascular development and stem elongation in Arabidopsis
71
(Vera-Sirera et al., 2010) i.e. in PA mediated developmental processes in plants. Here, we propose that water stress induced responses may be linked to PA metabolism especially during the vulnerable early growth phases of Scots pine. Therefore; we investigated young 6 to 17-week-old pine seedlings in controlled wet, optimal and dry conditions and monitored the changes in biomass, phenotype, and chlorophyll fluorescence (Fv/Fm). We measured the expression of LEA and PDC genes to describe the severity of the drought stress and water excess, respectively. The expression of CAT, GCL and GS genes was measured to indicate oxidative stress reactions under different treatments. The stress induced changes in the PA gene expressions and seedling morphology were compared to the optimal condition to elucidate the role of PA metabolism under different water levels (wet, optimal and dry) in needles, stems and roots. We found that the unfavourable water conditions changed the seedling morphology, disturbed symbiotic relationships and induced stress related gene expression. However, PA contents were mostly maintained at constant levels which underlines the importance of the strict regulation of PA metabolism for the early development of Scots pine seedlings. 2. Materials and methods 2.1. Plant material and growth conditions Scots pine (P. sylvestris L.) seed orchard (Hiirola, Finland 61490 N, 27150 E) seeds were sterilized with 30% H2O2. Seeds, 80 seeds per pot, were sown in twelve 3 dm3, 16.5 cm tall plastic pots (i.e. altogether 960 seeds per 12 pots) (VG-Products, VG-potter, Billund, Denmark) filled with 690 g of 1:9 combination of fertilized (Kekkil€ a Oy, Vantaa, Finland) and non-fertilized (Biolan Oy, Kauttua, Finland) horticultural peat. Before sowing, the peat in the pots was watered properly with 1.5 dm3 of H2O and left on a grate until water no longer drained away. The pots were irrigated two to three times a week during the first six weeks. Nutrients [KH2PO4 25 mg/l, (NH4)2HPO4 12.5 mg/l, CaCl2 2.5 mg/l, NaCl 1.25 mg/g, MgSO4 7H2O 7.5 mg/l, FeCl3 6H2O 0.6 mg/l] were added to irrigation water. Within two weeks the seeds started to germinate and before the first sampling, the average length of the seedlings was 10 ± 4 cm. The photoperiod in the growth chamber was 6 h darkness and 18 h daylight with light intensity of ca. 200 mmol m2 s1 and temperature 25 C during the experiment. Water content of the peat in all pots was ca. 75%. 2.2. Watering treatments and harvesting The watering treatments were based on the water retention characteristics of the peat (Heiskanen, 1993) determined with a pressure plate apparatus (soil moisture Equipment Corp., USA). The water retention curve indicates the level of suction required by the plant to get water from the soil as it dries (see Supplementary Fig. S2). The following watering treatments were planned to describe water availability to the pine seedlings: wet, optimal and dry. In the wet treatment, the volumetric water content of the peat was kept as high as possible without reaching the minimum airfilled porosity level (20 vol.%) (Wall and Heiskanen, 2003). In the dry treatment, the water content was kept above the wilting point (Hillel, 1971), which here was 9.7 vol.%. The optimal water availability was targeted to be around field capacity (10 kPa matric potential) (Hillel, 1971). It is noteworthy that both wet and optimal water treatments are within the limits where growth is unrestricted (see Supplementary Fig. S2). Thus, after first harvesting (week one), four pots were kept at a high water content (75% ¼ wet
72
€kela € et al. / Plant Physiology and Biochemistry 88 (2015) 70e81 R. Muilu-Ma
treatment ¼ water excess) and irrigated two to three times a week. The water content in four pots was mimicking optimal water content (40% ¼ optimal treatment) in natural conditions in northern pine forests. The optimal water content was reached in five weeks. The four remaining pots were allowed to dry one more week until a low water content (15% ¼ dry treatment ¼ drought stress) was reached (Fig. 1). The water content of the peat in the pots was monitored with ThetaProbe (HH2 Moisture meter, Delta T Devices Ltd, Cambridge, UK) and by weighing the peat pots before and after watering to ensure the targeted water contents. The ThetaProbe meter was calibrated gravimetrically at desorption with the same peat and pots as used in the experiment. Samples for free and soluble conjugated PA concentration measurements as well as for gene expression analyses were collected three times during the experiment. The first sampling during the first week of the treatments represents an initial phase, where the water content of the peat was above 75% in all pots. At the initial stage, twelve needles per pot were collected for PA and RNA analyses. Samples of the roots and stems were combined from four seedlings and four to five samples per pot were used to describe the initial phase. On the second (5th week) and third sampling dates (11th week), PA and RNA samples from needles, stems and roots were collected from all twelve pots. The PA and RNA samples were frozen immediately in liquid nitrogen and stored at 80 C until extracted and analysed. 2.3. Morphological parameters After the last harvesting, the fresh weight (FW) and dry weight (DW) (samples dried overnight at þ105 C) of the roots and shoots (¼stem with needles) of four seedlings from each pot (n ¼ 12) were measured. The total number of the root tips was calculated under stereomicroscope also from four seedlings per pot. Because nonsterilized peat was used as growth substrate, mycorrhizal infection could not be excluded; therefore, the number of the mycorrhizal root tips was also counted. The needles of four seedlings per pot were categorized according to their colour: green, greenish brown, brown. 2.4. Chlorophyll fluorescence Chlorophyll fluorescence has been regarded as a tool for interpreting stress tolerance of plants (Peeva and Cornic, 2009). As a non-destructive method, chlorophyll fluorescence, especially the
ratio of variable (Fv) to maximum fluorescence (Fm), has been widely used for assessing plant physiological status and the state of Photosystem II (PSII) (Krause and Weis, 1991). Chlorophyll fluorescence emission from green needles was measured by means of a Plant Efficiency Analyzer (Hansatech Instruments Ltd., Norfolk, England). The ratio of variable (Fv) to maximum fluorescence (Fm) was measured from three seedlings per pot. Needles were dark adapted at RT in leaf clips supplied with the analyser for at least 30 min before measurement. The minimum fluorescence (F0), the maximum fluorescence (Fm) and the variable fluorescence (Fv ¼ FmeF0) as well as the ratio Fv/Fm were recorded with equipment settings of 15 s at 100% intensity level of photon flux density (4000 mmol m2 s1). 2.5. Polyamine analysis The PA samples, consisting of about 100 mg of plant material per sample, were extracted in 5% (w/v) perchloric acid. Crude extracts for free PAs and hydrolysed supernatant for perchloric acid-soluble conjugated PAs were dansylated and separated by HPLC according et al. (1999). The PA to Sarjala and Kaunisto (1993) and Fornale concentrations were expressed as nmol g1 FW of plant tissue. The present study does not distinguish between Spm and possible thermospermine (Tspm) (see Supplementary Fig. S1), rather they are both included in the same fraction and referred to as Spm. At the end of the experiment the PA concentrations were estimated also in dry weight basis by using the average dry weight percentage values of the roots and shoots. Thus, the PA concentrations in dry weight basis are not as accurate estimations as the values in FW basis. 2.6. RNA isolation and reverse transcription The total RNA of Scots pine needle, root, and stem tissues was extracted from 100 mg of plant material using total RNA purification PureLink™ Plant RNA Reagent (Invitrogen Corporation, California, USA) according to the manufacturer's instructions. The RNA samples were treated with rDNase set (MachereyeNagel, Duren, Germany) at 37 C for 10 min in order to eliminate contaminating genomic DNA. The amount of DNase used to produce DNA-free RNA samples was three times higher than recommended by the manufacturer. The RNA samples were purified with the NucleoSpin® RNA Clean-Up kit (MachereyeNagel, Duren, Germany). The RNA yields were measured three times with OD260 analysis using spectrophotometer. cDNA was prepared from 1 mg of total RNA, which was reverse transcribed by SuperScript VILO™ cDNA synthesis kit (Invitrogen Corporation, California, USA). 2.7. Quantitative real-time PCR analysis
Fig. 1. Water content (%) of the peat during the experiment indicating wet, optimal and dry growing conditions. Initial sampling in the 1st week, second sampling in the 5th week, third sampling in the 11th week.
The real-time PCR primers for the gene expression studies (see Supplementary Table S1) on PA biosynthesis genes ADC, ODC, SPDS, ACL5, DAO and PAO were designed against Scots pine PA gene sequences ADC [GenBank: HM236823.1], ornithine decarboxylase (ODC) [GenBank: HM236831], SPDS [GenBank: HM236827], ACL5 [GenBank: HM236828], DAO [GenBank: HM236829] and PAO [GenBank: HM236830]. The primers of the housekeeping genes ubiquitin (UBQ), glyseraldehyde-3-phosphate dehydrogenase (GAPDH) and alfa-tubulin (TUBA) were based on the maritime pine (Pinus pinaster Ait.) putative UBQ sequence [GenBank: AF461687], Scots pine GAPDH sequence [GenBank: L07501] and Scots pine TUBA sequence [GenBank: FN546172], respectively. Primers for the LEA gene were based on Scots pine sequence [GenBank: FJ201571], which has been used as a candidate gene to cold acclimation due to its function in dehydrative stress (Wachowiak et al., 2009).
€kela € et al. / Plant Physiology and Biochemistry 88 (2015) 70e81 R. Muilu-Ma
73
Fig. 2. Relative gene expressions of LEA and CAT related to reference gene UBQ in needles, stems and roots of the Scots pine seedlings under three watering treatments (solid circle, wet; asterisk, optimal; open circle, dry). The geometric mean values of the four replicates are connected with solid, dashed, and dotted lines for the wet, optimal, and dry treatments, respectively. There are four replicates per treatment and sampling date except at the first sampling, where growth conditions in all pots were still similar. The replicates represent the seedlings grown in different pots but under the same treatment. The arrows indicate significant differences compared to optimal treatment (see Supplementary Table S2).
According to its translated protein sequence the LEA gene belongs to the same LEA II superfamily with loblolly pine (Pinus taeda) LEA protein [AAX68990.1] expressed under drought stress (Lorenz et al., 2011). The primers for the CAT gene were based on the Scots pine sequence [GenBank: EU513163]. The primers for the PDC gene were based on putative PDC mRNA sequence of root flooded loblolly pine [GenBank: CO161777.1], which was similar with sequence of black cottonwood (Populus trichocarpa) PDC family protein [GenBank: XM_002308194.2]. The primers for the GCL were based on Norway spurce (Picea abies) sequence [GenBank: AJ132540.1] and for GS on the putative GS mRNA sequence of well-watered loblolly pine roots [GenBank: CF478539.1], which was similar with the grey poplar GS sequence [Potri.013G026800.2]. The expression of the PA genes and the stress related genes were normalized by the gene expression derived from the three different housekeeping genes UBQ, GAPDH and TUBA, which showed very similar expression trends. The UBQ had the most stable Cp values during the experiment and therefore the UBQ normalized and statistically analysed gene expression results are presented. The quantification of mRNA with real-time PCR was performed in 20 mL of reaction mixture composed of 2 mL of cDNA, 10 mL Light Cycler® 480 SYBR Green I Master Mix, and 100 nM gene-specific primers. PCR amplification was initiated by incubation at 95 C for 10 min followed by 40 cycles: 30 s at 95 C, 1 min at 58 C, and 1 min at 72 C (LightCycler® 480/96, Roche Diagnostics Ltd.). The PCR conditions were optimized for high amplification efficiency 90% for all primer pairs used. Every PCR reaction was done in triplicate and the results were produced with LightCycler® 480 version 1.5.0.39 software.
2.8. Statistical methods The number of root tips, the proportion of mycorrhizal root tips and PA contents in DW basis were compared by computing the pairwise mean differences between the treatment groups and their 95% confidence intervals based on t-distribution. The amount of green, brown and greenbrown needles under different treatments were statistically analysed by Pearson's Chi-squared test, where the needle colour composition under optimal treatment was compared with wet and dry treatments at the week 11. The effects of treatment and time under treatment, and their interactions on PA concentrations and gene expression were analysed by linear mixed modelling (Fitzmaurice et al., 2004). Individual models were fitted for free and soluble conjugated fractions of Put, Spd and Spm, separately in needles, stems and roots. Similar models were fitted for relative expression of ADC, SPDS, ACL5, DAO, PAO, LEA, PDC, GCL, GS and CAT, respectively, each again in needles, stems and roots. Before model fitting, the original response variables (PA level or relative expression) were transformed to the natural logarithm scale. Both treatment and time under treatment were included as categorical factors with fixed effects, such that their reference levels were optimal treatment and week 5, respectively. Accordingly, the two main effect parameters for watering treatment refer to the mean differences in the log-response: wet vs. optimal, and dry vs. optimal, respectively, both at week 5. Similarly, the two main effects of time under treatment describe analogous contrasts: week 1 vs. week 5, and week 11 vs. week 5, respectively, both under optimal treatment. Treatment time interactions were specified by four product terms of the pertinent
€kela € et al. / Plant Physiology and Biochemistry 88 (2015) 70e81 R. Muilu-Ma
74
Table 1 Water content and dry weight (mg) of the seedlings at the end of the experiment. Treatment
Wet Optimal Dry
Water content (%)
Dry weight (mg)
Shoot
Shoot
Root
3. Results 3.1. Expression of LEA gene indicated drought stress
Root
Mean
Range
Mean
Range
Mean
Range
Mean
Range
62 61 46
58e65 57e64 21e63
82 69 55
75e87 49e78 41e64
35 32 32
26e46 22e49 16e58
12 12 12
7e17 5e19 6e22
main effects. Thus, for example, an interaction term “dry vs. opt: wk 11 vs. 5” represents the difference between dry and optimal treatments in the mean change of the log-response from week 5 to week 11. To allow the likely correlation across the repeated measurements within each pot, random intercept terms for the 12 pots were included in the model upon the fixed effects above. The models were fitted using function lme in package nlme (Pinheiro et al., 2012) of the R environment, version 2.15.1 (R Development Core Team, 2012). The estimated regression coefficients and their 95% confidence intervals (CI) were back-transformed to the original scale of PA concentration or gene expression, such that apart from the overall intercept (“grand mean”), each estimate describes the relative change associated with the pertinent main effect or interaction effect with respect to the relevant reference level.
LEA gene was up-regulated in all parts of the Scots pine seedlings which were suffering from drought stress (Fig. 2 and Supplementary Table S2). The estimated relative changes on expression of LEA and their 95% confidence intervals associated with the effects of watering treatment, time, and their interactions are reported in Supplementary Table S2. Compared to the optimal treatment, the relative expression of LEA under the drought stress was 3.8-, 2.4-, and 3.8-fold higher in needles, stems and roots, respectively. The expression of CAT remained stable in needles but decreased in stems and roots towards the end of the experiment (Fig. 2 and Supplementary Table S2). Generally, CAT expression was not induced by drought stress or water excess. Only in roots, CAT expression was slightly induced by dry treatment being 1.8-fold higher than in optimal treatment. The expression of GCL was too low to be investigated (See low Cp values in Supplementary Table S3) and the GS expression levels remained stable in all plant organs throughout the experiment (Supplementary Table S2). This reveals that glutathione synthesis pathway was not transcriptionally up-regulated in drought-stressed seedlings. The PDC gene, which indicates alcohol fermentation under hypoxia, was not up-regulated in the roots under wet treatment. In the needles the
Fig. 3. (a) The number of the root tips in Scots pine seedlings per mg of dry weight and (b) the proportion of mycorrhizal root tips relative to the total number of root tips (%) at the end of the experiment.
Fig. 4. The effects of the dry (open circle), optimal (asterisk) and wet (solid circle) treatments (a) on needle colour of the Scots pine seedlings and (b) on the ratio of variable to maximum fluorescence (Fv/Fm) in the needles as a function of time. The number of replicates is four and each symbol represents the mean value of one to three measurements from seedlings grown in the same pot. The values are connected with solid, dashed, and dotted lines for the wet, optimal and dry treatments, respectively.
€kela € et al. / Plant Physiology and Biochemistry 88 (2015) 70e81 R. Muilu-Ma
75
Fig. 5. Free (a) and soluble conjugated PA (b) concentrations (nmol/gFW) in needles, stems and roots of the Scots pine seedlings under wet (solid circle), optimal (asterisk), and dry (open circle) treatments. The geometric mean values of the four replicates are connected with solid, dashed, and dotted lines for the wet, optimal, and dry treatments, respectively. There are four replicates per treatment and sampling date except at the first sampling, where growth conditions in all pots were still similar. The replicates represent the seedlings grown in different pots but under the same treatment. The arrows indicate significant differences compared to optimal treatment (see Supplementary Table S4).
expression of PDC increased towards the end of the experiment in all treatments (Supplementary Table S2). 3.2. Water levels affected root morphology and mycorrhizal association Physiological and morphological parameters indicated a perturbation of the early development of Scots pine seedlings under drought stress. The water content % (w/w) of the shoots and roots of the seedlings decreased under dry treatment, but the dry mass of the seedlings was rather equal (similar) in all treatments (Table 1). The water content in pine shoots remained relatively stable during the experiment, whereas the roots showed a linear
decrease in the water content along the decreasing water gradient in peat. Extreme drought in peat at the end of the experiment (week eleven) finally led to a collapse of water content (21e63%) also in the shoots. The morphology of the root system was affected by the watering treatments. The total number of root tips was lower under the drought stress than in the optimal (mean difference MD 3.3, 95% confidence interval, CI 5.4 to 1.2) or in the wet treatment (MD 4.0, 95% CI 6.6 to 1.4; Fig. 3a). In dry treatment root branches were also withered compared to the optimal and wet treatments. The highest proportion of mycorrhizal root tips was observed in the optimal treatment (compared to wet treatment: MD 13.2%, 95% CI 8.6%e17.8%; compared to dry treatment: MD
76
€kela € et al. / Plant Physiology and Biochemistry 88 (2015) 70e81 R. Muilu-Ma
Fig. 5. (continued).
10.1%, 95% CI 6.3%e13.9%; Fig. 3b) suggesting that both drought stress and water excess restricted the ability to sustain a symbiotic relationship. The proportion of green needles was over 65% in all treatments just after the target water content was reached in peat (week five) (Fig. 4a). Six weeks later (week eleven), the proportion of green needles was 63%, 42% and 32% under optimal, wet and dry treatments, respectively. When compared to optimal treatment, the composition of green, brown and greenbrown needles was statistically different under wet [X-squared value ¼ 11.5982, df ¼ 2 and p-value ¼ 0.00303] and dry treatment [X-squared ¼ 20.8404, df ¼ 2, p-value ¼ 0.00] at the week 11. However, relatively high chlorophyll fluorescence (Fv/Fm) values were detected in the green needles under dry and wet treatments (Fig. 4b). In addition to the decreased number of green needles, increased variation in Fv/Fm
between the seedlings in the wet treatment was apparent. Nevertheless, the average decrease in Fv/Fm was not as great as in the dry treatment (Fig. 4b). Under the drought stress, Fv/Fm remained quite stable until crashed when the peat water potential had decreased (9.7 vol.%) below the wilting point. 3.3. Changes in PA content The results on PA levels in the wet, optimal and dry treatments in FW basis are displayed in Fig. 5a and b. The estimated relative changes with 95% confidence intervals associated with the effects of watering treatment and time and their interactions are reported in Supplementary Table S4. In general, a decreasing trend was apparent in the free PA levels of needles and roots over time under all treatments, but in stems the downward trend appeared only
€kela € et al. / Plant Physiology and Biochemistry 88 (2015) 70e81 R. Muilu-Ma
after five weeks (Fig. 5a and Supplementary Table S4). Put was the most abundant PA in the seedlings (100e500 nmol/gFW). Free Put levels increased in needles under drought stress from week five onwards, being 1.6-fold concentration in dry than optimal treatment. In the roots free Put levels were slightly higher under the dry than optimal treatment throughout the experiment. The concentration of free Spd was higher in needles and roots (10e100 nmol/ gFW) than in stems (5e20 nmol/gFW) (Fig. 5a). In needles, free Spd had a decreasing trend under all treatments. In stems, the amount of free Spd decreased from week five onwards under the optimal and dry treatments, whereas it remained stable under the water excess (see Supplementary Table S4). In roots, the concentration of free Spd decreased more sharply under the optimal than other treatments from week five onwards. Overall free Spm concentrations were low in the seedlings (
