Notice: Wiley Online Library will be unavailable on Saturday 30th July 2016 from 08:00-11:00 BST / 03:00-06:00 EST / 15:00-18:00 SGT for essential maintenance. Apologies for the inconvenience.
Author for correspondence: Carl K-Y. Ng Tel: +353 1 716 2250 Email: firstname.lastname@example.org
•Physcomitrella patens is a bryophyte belonging to early diverging lineages of land plants following colonization of land in the Ordovician period. Mosses are typically found in refugial habitats and can experience rapidly fluctuating environmental conditions. The acquisition of dehydration tolerance by bryophytes is of fundamental importance as they lack water-conducting tissues and are generally one cell layer thick.
•Here, we show that dehydration induced oscillations in the steady-state transcript abundances of two group 3 late embryogenesis abundant (LEA) protein genes in P. patens protonemata, and that the amplitudes of these oscillations are reflective of the severity of dehydration stress.
•Dehydration stress also induced elevations in the concentrations of abscisic acid (ABA), and ABA alone can also induce dosage-dependent oscillatory increases in the steady-state abundance of LEA protein transcripts. Additionally, removal of ABA resulted in rapid attenuation of these oscillatory increases.
•Our data demonstrate that dehydration stress-regulated expression of LEA protein genes is temporally dynamic and highlight the importance of oscillations as a robust mechanism for optimal responses. Our results suggest that dehydration stress-induced oscillations in the steady-state abundance of LEA protein transcripts may constitute an important cellular strategy for adaptation to life in a constantly changing environment.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
The transition from an aqueous to a terrestrial environment during the colonization of land by plants requires the evolution of strategies for life under constantly fluctuating environmental conditions. The ability to withstand periods of dehydration is of particular importance for survival in the terrestrial environment (Rensing et al., 2008; Mishler & Oliver, 2009). Dehydration tolerance is not limited to plants, as anhydrobiosis has also been reported in bacteria, fungi and animals (Stacy & Aalen, 1998; Browne et al., 2002; Wise & Tunnacliffe, 2004; Goyal et al., 2005). Little is known about how anhydrobiotes survive periods of dehydration, although it is becoming increasingly evident that late embryogenesis abundant (LEA) proteins may play an important role in conferring dehydration tolerance. LEA proteins are intrinsically disordered proteins and are characterized by their ability to function as molecular shields (Chakrabortee et al., 2012). It has been proposed that the ability of LEA proteins to function as molecular shields may result from their ability to occupy space in solution to reduce the collision rate of proteins that would otherwise aggregate following denaturation as a result of dehydration. Alternatively, LEA proteins can form a loose association with other polypeptides, resulting in a dynamic, three-dimensional protective barrier (shield) to reduce aggregation (Chakrabortee et al., 2012). Such a loose interaction may also facilitate refolding of the protein to its native conformation (Tompa & Csermely, 2004).
The moss, Physcomitrella patens, belongs to early diverging lineages of land plants and there is evidence to suggest that P. patens is highly tolerant of abiotic stress (Frank et al., 2005; Cho et al., 2009; Koster et al., 2010). For example, P. patens can survive exposure to 350 mM NaCl and 500 mM sorbitol, indicating that P. patens is highly tolerant of salt and osmotic stress (Frank et al., 2005). P. patens can survive moderate dehydration but it is not desiccation-tolerant (Koster et al., 2010; Pressel & Duckett, 2010). However, pretreatment of P. patens with ABA conferred desiccation tolerance (Koster et al., 2010; Pressel & Duckett, 2010). Moss protonemata pretreated with ABA showed chloronemal cells packed with small vacuoles containing electron-rich deposits and increased cell wall thickness (Pressel & Duckett, 2010). In P. patens, ABA pretreatment decreased electrolyte leakage from dehydrated tissues following rehydration, suggesting the activation of membrane protective mechanisms by ABA, possibly involving accumulation of sucrose and nonreducing sugars (Koster et al., 2010). Additionally, ABA may also induce the accumulation of dehydrin-like proteins (group 2 LEA proteins) which may confer osmotic and salinity stress tolerance in P. patens as a result of their ability to act as radical scavengers and/or membrane stabilizers (Saavedra et al., 2006).
Recent genome-wide analysis identified several group 3 LEA protein genes whose expression levels are highly up-regulated during dehydration stress and ABA treatments in P. patens (Cuming et al., 2007). In this study, we showed that dehydration can induce oscillatory increases in group 3 LEA protein transcripts in P. patens protonemata that have lost c. 90% of total water content. We also demonstrate that dehydration induced elevations in the concentrations of ABA, and that ABA alone can also induce dosage-dependent oscillatory increases in the steady state abundances of LEA protein transcripts. Interestingly, we show that removal of ABA resulted in rapid attenuation of these oscillatory increases. In the context of the moss, temporal dynamics in steady-state transcript abundances in the form of oscillations that is dosage responsive, or responsive to the severity of the dehydration stress or ABA concentrations are likely to have fundamental implications for surviving periodic fluctuations in water availability and confer evolutionary advantages during the colonization of land by plants.
Materials and Methods
Plants and growth conditions
Physcomitrella patens (Hedw.) B.S.G. ecotype ‘Gransden 2004’ was propagated on cellophane overlay plates containing BCDAT media (1 mM MgSO4, 1.84 mM KH2PO4 (pH 6.5 adjusted with KOH), 10 mM KNO3, 45 μM FeSO4, 5 mM ammonium tartrate, 1 mM CaCl2, and supplemented with a trace element solution (alternative TES- 10 μM H3BO3, 2 μM MnCl2, 0.22 μM CuSO4, 0.23 μM CoCl2, 0.19 μM ZnSO4, 0.1 μM Na2MoO4, and 0.17 μM KI) and 0.8% w/v agar as previously described (Nishiyama et al., 2000) under controlled conditions: light intensity, 50 μmol s−1 m−2; 16 h light : 8 h darkness; 75% relative humidity (RH); and temperature, 23°C in a growth chamber (Sanyo MLR-351 H Versatile Environmental Test Chamber, Tokyo, Japan).
Dehydration stress and abscisic acid treatments
Ten- to 12-d-old P. patens protonemata were used for dehydration (acute and medium) stress treatments over a duration of 24 h. Acute dehydration stress treatments were carried out by transferring thin lawns of moss protonemata to plastic Petri dishes (9 cm diameter). The dishes (uncovered) were then placed in a controlled chamber. For medium dehydration treatments, lawns of cellophane-grown moss protonemata were transferred directly to plastic Petri dishes (9 cm diameter) and the dishes (uncovered) were placed in transparent plastic containers containing saturated sodium chloride (NaCl) solutions in order to maintain the RH of the atmosphere within the container at 75% as previously described (Cuming et al., 2007). The plastic containers were then kept in a growth chamber under controlled conditions. The extent of FW loss by protonemata was monitored by weighing the samples at regular 2 h time intervals (0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, and 24 h) following the onset of dehydration treatments: 0 h in the time series represent the start of the 16 h light regime. Tissue samples were also harvested at regular 2 h time intervals (0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, and 24 h) and snap-frozen using liquid N2 before RNA isolation. ABA ((±)-cis,trans-ABA) pretreatment was carried out by transferring lawns of 10- to 12-d-old protonemata grown on cellophane overlay BCDAT media on to fresh agar plates of BCDAT medium containing 10 or 100 μM ABA, followed by incubation at 23°C for 24 h under controlled growth conditions as described earlier. Control and ABA pretreated protonemata were then subjected to a 24 h dehydration stress treatment before being transferred to fresh BCDAT media and cultured for 6 d to determine desiccation tolerance. To determine the effects of ABA alone on gene expression, protonemata were transferred to BCDAT agar plates containing 10 and 100 μM ABA. Samples were harvested at regular 2 h time intervals (0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, and 24 h) and snap-frozen using liquid N2 before RNA isolation. ABA rescue experiments were performed by transferring protonemata pretreated for 24 h with 10 and 100 μM ABA to fresh BCDAT media, and samples were harvested at regular 2 h time intervals (0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, and 24 h) and snap-frozen using liquid N2 before RNA isolation.
RNA isolation, real-time, quantitative PCR (qPCR)
Total RNA was isolated from moss tissue using the RNeasy® Plant Mini Kit (Qiagen) according to the manufacturer’s recommendations. Total RNA was treated with Ambion® TURBO DNAse I (Invitrogen) following the manufacturers’ protocol. First-strand cDNA synthesis was carried out using random primers (Promega) and M-MLV reverse transcriptase enzyme (Invitrogen). For real-time PCR (qPCR) amplification reactions were run in triplicate for each experimental time interval. Reactions were carried in 96-well plates (Agilent Technologies, Cork, Ireland) using gene-specific primers (Supporting Information, Table S1) and Kapa SYBR® fast qPCR Universal Kit (Anachem, Luton, UK) on a Mx3000P QPCR System (Agilent Technologies) according to the manufacturers’ instructions with the cycling programme consisting of an initial step at 95°C for 3 min followed by 40 cycles of 95°C for 10 s followed by 60°C for 30 s and one cycle of 95°C for 60 s. For amplification product specificity, a melt curve was routinely generated at the end of each run. The generated data were analysed with Mx3000P qPCR software (Agilent Technologies) to determine Ct values. To determine amplification efficiency of reference and target gene primers, qPCR reactions were carried out using serially diluted cDNA templates. The average Ct and standard deviation (SD) determined from the triplicate Ct values, the natural log of the RNA (cDNA) dilutions was then plotted against the Ct values at each dilution for every primer pair separately, and the slope values used to determine amplification efficiencies using the formula, E = 10−1/slope and expressed as percentage efficiency by using formula %E = (E − 1) × 100% as previously described (Pfaffl, 2001) (see Table S1 for PCR amplification efficiencies). Relative expression was determined by normalizing the transcript abundances of Pp_166566, Pp_211998 and Pp_184622 genes with 18s RNA transcripts. Relative expression or fold change was determined using (efficiency corrected) the −ΔΔCt method (Pfaffl, 2001). Three technical replicates were carried out for each data point in the time series and variations ranged from 0.03 to 2.77% for all genes tested.
Extraction of ABA from P. patens protonemata
Abscisic acid was extracted from P. patens protonemata as previously described with modifications (Ross et al., 2004). Briefly, control and dehydration stress-treated moss samples were harvested at 2 h intervals over a 24 h period, flash-frozen immediately in liquid N2 before freeze-drying. Fifty milligrams of freeze-dried samples were ground into a fine powder with 2 mm ball bearings using a mixer mill (Mixer Mill MM 400, Retsch, Haan, Germany). The samples were spiked with 10 μl of d6-ABA (internal standard, 0.1 μg ml−1) before extraction with 1 ml of 80% acetone containing 1% glacial acetic acid by vortexing for 1 min followed by sonication for 15 min. The supernatants were collected after centrifugation. The extraction procedure was repeated once with the same solvent and the supernatants from both extraction steps were pooled. The supernatant was then evaporated to dryness using a vacuum evaporator, resuspended in 1.2 ml of 10% methanol containing 1% glacial acetic acid, and purified by solid phase extraction (SPE) (Chromabond C18ec, 3 ml per 200 mg, Machery-Nagel, Düren, Germany) using a GX-271 Aspec SPE system controlled by the Trilution LH software (Gilson Scientific Ltd, Luton, UK). SPE cartridges were conditioned and equilibrated with 1 ml of 80% methanol containing 1% glacial acetic acid followed by 1 ml of 10% methanol containing 1% glacial acetic acid. Following cartridge equilibration, 1 ml of sample supernatant was applied and washed with 2 ml of 10% methanol containing 1% glacial acetic acid. ABA was then eluted from the SPE cartridge with 1 ml of 80% methanol containing 1% glacial acetic acid. The purified extract was dried using a vacuum evaporator and reconstituted with 100 μl of mobile phase (solvent A: solvent B, 1:1 v/v), and 5 μl of sample was subjected to LC-MS/MS analysis.
Quantitative analysis of ABA by LC-MS
The high-pervormance liquid chromatography system consists of a binary pump, a temperature-controlled autosampler, an online degasser unit, and a column oven (1200 RRLC, Agilent Technologies, Santa Clara, CA, USA). The HPLC was coupled to an Agilent 6460 triple quadrupole mass spectrometer (Agilent Technologies) equipped with a Jet Stream ion source. Electrospray ionization (ESI) was performed in negative ionization mode with N2 as the nebulizing agent. The gas temperature and flow rate were 350°C and 8 l min−1, respectively, and the sheath gas temperature and flow rate were 350°C and 10 l min−1, respectively. The ESI needle voltage was adjusted to 3.5 KV and the optimum fragmentor voltage (100 volts) and collision energy (12 volts) for ABA and d6-ABA were determined by analysis of reference compounds in selected and product ion scanning mode. Multiple-reaction monitoring (MRM) detection was applied using N2 as the collision gas, with a dwell time of 200 ms for each transition of the deprotonated molecules at m/z 263 and m/z 269 to product ions at m/z 153 and m/z 159 for ABA and d6-ABA, respectively. Data acquisition and analysis were controlled by MassHunter software (Agilent Technologies). The chromatographic separation of compounds was achieved using a narrow-bore analytical column GeminiNX (2.0 mm ID × 100 mm, particle size 3 μm; Phenomenex, Macclesfield, UK) in binary gradient mode at a flow rate of 0.3 ml min−1, and column oven and autosampler temperatures were maintained at 40 and 4°C, respectively. The mobile phase consisted of Solvent A (0.1% formic acid) and Solvent B (acetonitrile supplemented with 0.1% formic acid). The initial elution condition was A-B (85 : 15, v/v), linearly changed to A-B (50 : 50, v/v) at 12 min, and the gradient was changed to A-B (20 : 80, v/v) at 14 min and maintained in this condition up to 21 min before returning to the initial condition at 22 min, followed by 8 min of re-equilibration. Eluent from the column was diverted to waste for the initial 1 min.
Calibration curve for ABA quantitation
Standard stock solutions of (±)-cis,trans-ABA (AG Scientific, Inc., San Diego, CA, USA) and d6-ABA (Plant Biotechnology Institute, National Research Council, Saskatchewan, Canada) were prepared at a concentration of 1 mg ml−1 in 70% methanol before serial dilution with 70% methanol to obtain working solutions of concentrations ranging from 0.02 to 2 μg ml−1. Ten microlitres of each working solution and internal standard (d6-ABA, 0.1 μg ml−1) were added to 50 mg of freeze-dried control moss (nonstress) samples and extracted with 990 μl of 80% acetone containing 1% glacial acetic as described earlier in this section. The calibration curve was prepared using seven calibration standards, giving final ABA concentrations of 0.2, 0.5, 1, 2, 5, 10 and 20 ng ml−1. Control samples spiked with 10 μl of internal standard (d6-ABA, 0.1 μg ml−1) were extracted in parallel to account for the contribution of endogenous ABA. The calibration curve was constructed by plotting the peak area ratio of analyte (ABA) to internal standard (d6-ABA) by subtracting the peak area ratio of endogenous ABA to internal standard vs concentration in the standard-spiked control sample by least-squares linear regression (r2 = 0.999). The concentrations of ABA in control and dehydrated moss protonemata were determined from the calibration curve.
Results and Discussion
As bryophytes are limited to refugial habitats and can experience rapidly fluctuating environmental conditions, we first determined the effects of two different dehydration regimes (acute and medium) on the rate of water loss from P. patens protonemata. We observed a rapid loss of water of c. 94% within 4 h following onset of acute dehydration, whereas the rate of water loss was slower (c. 94% after 10 h) under medium dehydration (Fig. 1a). Next, we examined the steady-state transcript abundances of the group 3 LEA protein gene (Phypa_166566) under these two dehydration regimes over a 24 h period. Microarray analysis by Cuming et al. (2007) showed that Phypa_166566 transcripts were up-regulated by 16- and 7.7-fold following dehydration (c. 84% water loss 24 h after onset of dehydration) and 10 μM ABA treatment, respectively. We observed that transfer of protomenata to acute and medium dehydration conditions induced increases in steady-state abundances of Phypa_166566 transcripts and that these dehydration-induced increases took the form of oscillations (Fig. 1b,c). We did not observe increases in Phypa_166566 transcripts when protonemata were transferred to nonstress (control) conditions (Fig. 1b,c), suggesting that the physical transfer of protonemata was not sufficient to induce changes in gene expression.
We also determined the expression of another group 3 LEA protein gene (Phypa_211998). Similar to the expression profile of Phypa_166566, oscillatory increases in the steady-state abundances of Phypa_211998 transcripts were observed following acute and medium dehydration treatments (Supporting Information Fig. S1a,b). We did not observe increases in Phypa_211998 transcripts when protonemata were transferred to nonstress (control) conditions (Fig. S1a,b). Additionally, we observed that dehydration-induced increases in steady-state transcript abundances are reflective of the severity of dehydration stress experienced by the moss protonemata, with acute dehydration stress inducing a greater increase in steady-state transcript abundances than that induced by medium dehydration stress for both group 3 LEA protein genes (Figs 1b,c, S1a,b). We did not observe any induction or temporal dynamics in the steady-state transcript abundance of the glutathione S-transferase gene, Phypa_184622, following dehydration stress (data not shown), consistent with the observation that Phypa_184622 is nondehydration-responsive (Cuming et al., 2007). Taken together, these data highlight the temporally dynamic nature of dehydration-induced increases in group 3 LEA gene transcripts.
Abscisic acid has been shown to be an important component of the dehydration signalling system in plants (Lee & Luan, 2011). Additionally, there is increasing evidence that ABA may also be an important stress hormone in animal cells (Scarfi et al., 2008; Sturla et al., 2009; Bruzzone et al., 2012). We therefore wondered if ABA is involved in regulating the temporally dynamic oscillatory response to dehydration stress in P. patens protonemata. We show that acute dehydration induced a rapid increase in ABA concentrations in moss protonemata, with ABA concentrations reaching as high as 76.6 ± 17.0 ng g−1 DW within 6 h and remaining elevated for up to 24 h (Fig. 2). This increase in ABA concentrations in P. patens protonemata is in agreement with previous observations of dehydration-induced increases in ABA concentrations in Funaria hygrometrica (Werner et al., 1991).
Importantly, we show that ABA alone (in the absence of dehydration stress) can induce dosage-dependent oscillatory increases in steady-state abundance of both group 3 LEA protein transcripts, Phypa_166566 (Fig. 3a) and Phypa_211998 (Fig. 3b), with 100 μM ABA inducing greater abundance of steady-state transcripts than 10 μM ABA. We did not observed oscillatory fluctuations in the concentrations of ABA (Fig. 2), implying that ABA-induced oscillatory increases in the group 3 LEA protein transcripts (Fig. 3) are not associated with oscillatory fluctuations in cellular ABA concentrations. These oscillatory increases in steady-state transcript abundance are unlikely to be regulated by circadian rhythms as P. patens protonemata grown under continuous light also showed ABA-induced oscillatory increases in Phypa_166566 and Phypa_211998 transcripts (Fig. S2). However, we cannot completely rule out some contributions by circadian rhythms to the oscillations for protonemata grown under a 16 h : 8 h light : dark regime. Additionally, these oscillatory increases are unlikely to be because of the physical transfer of protonemata (Figs 1b,c, S1a,b). P. patens can survive water loss > 90%, although it cannot recover from desiccation (Koster et al., 2010; Pressel & Duckett, 2010; Khandelwal et al., 2011). However, pretreatment of F. hygrometrica (Werner et al., 1991) and P. patens protonemata (Koster et al., 2010; Pressel & Duckett, 2010; Khandelwal et al., 2011) with ABA can confer desiccation tolerance. We demonstrate that pretreatment of P. patens protonemata with 10 or 100 μM ABA for 24 h conferred tolerance to acute dehydration, with 100 μM ABA conferring a greater degree of desiccation tolerance, as evidenced by the robust recovery of growth following transfer to nonstressed conditions (Fig. 3e).
A measure of a highly robust signalling system is its ability to respond rapidly to changing conditions (Nurse, 2008; Brent, 2009). Here, we show that transfer of ABA-pretreated protonemata to fresh media resulted in rapid attenuation of the oscillatory increases in steady-state abundance of LEA protein transcripts, indicating that this temporally dynamic system can respond rapidly to the presence and absence of the ABA stimulus (Fig. 4a,b). We also observed a gradual re-establishment of oscillatory elevations in steady-state transcript abundance, albeit to a lesser degree (Fig. 4a,b), which can be attributed to the presence of residual amounts of ABA following transfer to fresh medium. Together, these results lend credence to the importance of ABA-regulated temporal dynamics in steady-state transcript abundance in the form of oscillations in desiccation tolerance.
In plants, noncircadian cyclic (oscillatory) changes in gene expression have been implicated in developmental regulation of periodic root branching (Moreno-Risueno et al., 2010; Moreno-Risueno & Benfey, 2011). In mouse cells and whole animals, ultradian oscillations in gene expression have been linked with pulsatile releases of glucocorticoids with important implications in transcriptional reprogramming (Stavreva et al., 2009). Robust and highly regulated temporal dynamics in gene expression have been suggested to be an important strategy for the maintenance of cellular processes in eukaryotes in response to perturbations in the environment (Murrat et al., 2007; Tiana et al., 2007; Hager et al., 2009; Paszek et al., 2010; Yosef & Regev, 2011). Additionally, temporal dynamics in the form of oscillations may confer responsiveness (in speed and efficiency) to perturbations (Tiana et al., 2007; Yosef & Regev, 2011). In the context of the moss, dehydration- and ABA-induced temporal dynamics in steady-state transcript abundance in the form of oscillations that are dosage-responsive or responsive to the severity of the dehydration stress is likely to have fundamental implications for surviving periodic fluctuations in water availability and to confer evolutionary advantages during the colonization of land by plants.
This study was supported by a Science Foundation Ireland (SFI) Research Frontiers Programme Grant (08/SFI/EOB1087) and a SFI Equipment Grant (06/SFI/GEN034ES) to C.K-Y.N.