The authors have declared that no competing interest exists.
Conceived and designed the experiments: NRB AB LW AR JSMS. Performed the experiments: SYA NOG SE MAA HSA SI AMS FME. Analyzed the data: SI AB NRB LW AR JSMS. Wrote the paper: AR JSMS AB NRB LW SI.
Water availability is a major limitation for agricultural productivity. Plants growing in severe arid climates such as deserts provide tools for studying plant growth and performance under extreme drought conditions. The perennial species
We describe the results of studying the metabolomic response of wild
Taken together these data suggest that the ability of
Drought is one of the most serious limitations for agriculture limiting plant growth, photosynthesis and, thus, productivity in many areas of this planet. Given the anticipated climate change, it is expected to worsen in the future thus becoming an even more important threat for global food supply
Essentially, a variety of different approaches have been followed in different plant systems. Starting from the observation that response to drought/drought tolerance is a multigenic trait, quantitative genetic studies taking advantage of natural diversity with respect to drought tolerance have been performed. Many of these studies have been performed in crop species such as corn or rice
As to the molecular responses, numerous studies have been performed in different plant systems such as
Membranes are very sensitive to dehydration and in consequence lipid composition changes to cope with this stress
In this study we describe the metabolomic and lipidomic response of a wild plant species,
In addition
We here describe the metabolomic and lipidomic response of
The experiment was conducted using plants grown wild in the desert near to Jedda. No specific permission was required to perform these experiments as this area is used routinewise by the Colleagues from the King Abdullaziz University. The plant used in these experiments Calotropis procera is not an endangered or protected species.
This experiment was conducted during September 2012 in the desert about 30 km away from Jeddah (latitude 21°26′6.00, longitude 39°28′3.00). The average temperature during the time the experiments were conducted varied from 28–37°C, humidity was 70–75%. In this area,
(a) Representative photo of the plants chosen for this experiment growing in its natural habitat in Saudi Arabia near to Jeddah. For this study representative species of similar size and performance were chosen. (b) Experimental set-up. At day 1 (control) leaves of three independent plants were harvested 1 h post-dawn, at midday and 1 h pre-dusk. One day later (Day 2), plants were watered at dawn and leaves were harvested 1 h post dawn, at midday and 1 h pre-dusk. Harvested leaves were frozen immediately in liquid –N and processes as described in Experimental procedures.
Leaves of the three
RWC (%) equals (FW – DW) divided by ( TW – DW)×100.
Multiple comparisons were performed following the procedure outlined by Duncan's New Multiple Range test.
Samples for metabolomic studies were taken one hour (at dawn), six (at midday) and 12 hours (one hour pre-dusk) after water treatment. In order to be able to identify possible changes in metabolism due to diurnal fluctuations, samples were in addition taken one day before watering at the same three time points. Samples taken were frozen in liquid nitrogen and kept at −80 C until extraction. Three independent but comparable plants were used for this experiment, thus representing three biological replicates.
Leaf samples were extracted and processed for metabolomics analysis as detailed below
Approximately 100 mg of the frozen plant tissue was homogenized in 2-ml Eppendorf tubes twice for 1 min at maximum speed within a Retschmill. The metabolites were extracted from each aliquot in 1 ml of a homogenous mixture of −20°C methanol: methyl-tert-butyl-ether: water (1∶3∶1), with shaking for 30 min at 4°C, followed by another 10 min of incubation in an ice cooled ultrasonication bath. After adding 650 µl of UPLC-grade methanol: water 1∶3, the homogenate was vortexed and spun for 5 min at 4°C in a table-top centrifuge. The addition of methanol: water leads to a phase separation, providing the upper organic phase, containing the lipids, a lower aqueous phase, containing the polar and semipolar metabolites, and a pellet of starch and proteins at the bottom of Eppendorf tube. The separate phases are isolated and dried down in a speed vac and stored at −80°C until use in the different metabolomic or lipidomic analyses.
UPLC-FT-MS measurement of lipids and semipolar metabolites and GC-TOF analysis of primary metabolites
UPLC separation of the semipolar fraction of the fractionated metabolite extract is performed using a Waters Acquity UPLC system, using an HSS T3 C18 reversed-phase column (100 mm ×2.1 mm ×1.8 µm particles; Waters). The mobile phases are 0.1% formic acid in H2O (Buffer A, ULC MS grade; BioSolve,
The lipid fraction of the fractionated metabolite extract is performed on the same UPLC system using a C8 reversed-phase column (100 mm ×2.1 mm ×1.7 µm particles; Waters). The mobile phases are water (UPLC MS grade; BioSolve) with 1% 1 M NH4Ac, 0.1% acetic acid (Buffer A,) and acetonitrile: isopropanol (7∶3, UPLC grade; BioSolve) containing 1% 1 M NH4Ac, 0.1% acetic acid (Buffer B). A 2-µl sample (the dried-down organic fraction was re-suspended in 500 µl of UPLC-grade acetonitrile: isopropanol 7∶3) is loaded per injection, and the gradient, which was taken out with a flow rate of 400 µl min−1, is: 1 min 45% A, 3-min linear gradient from 45% A to 35% A, 8-min linear gradient from 25% A to 11% A, 3-min linear gradient from 11% A to 1% A. After washing the column for 3 min with 1% A, the buffer is set back to 45% A and the column is re-equilibrated for 4 min (22-min total run time).
The mass spectra are acquired using an Exactive mass spectrometer. The spectra are recorded alternating between full-scan and all-ion fragmentation-scan modes, covering a mass range from 100 to 1500 m/z. The resolution is set to 10 000, with 10 scans per second, restricting the loading time to 100 ms. The capillary voltage is set to 3 kV with a sheath gas flow value of 60 and an auxiliary gas flow of 35 (values are in arbitrary units). The capillary temperature is set to 150°C, whereas the drying gas in the heated electrospray source is set to 350°C. The skimmer voltage is set to 25 V, whereas the tube lens is set to a value of 130 V. The spectra are recorded from 1 min to 17 min of the UPLC gradients.
The polar phase is analyzed for primary metabolites using an established GC-TOF ms protocol
In a preliminary experiment leaf samples were taken from three independent plants of similar stature and developmental stage (cf.
The absence of a significant effect of watering on the metabolism of the treated plants could have two explanations: either metabolism of
In order to see whether changes in metabolism are detectable at the earlier time-points, metabolite data from samples harvested within 12 hours after watering and obtained at the corresponding time points before watering were subjected to a principal component analysis.
The results are shown in
(a) PCA (upper part) and ANOVA (lower part) for primary metabolites. (b) PCA (upper part) and ANOVA (lower part) for complex lipids. (c) PCA (upper part) and ANOVA (lower part) for secondary metabolites. Shown are always three independent samples per time point (dawn (1 hour post dawn/after watering), midday (6 hours after dawn/after watering) and pre-dusk (12 hours after dawn/after watering). Watered samples are shown in blue, non-watered in red. The lower part shows the results of a Bonferroni corrected ANOVA displaying the influence of treatment (watering) for all samples and of harvesting time for primary and secondary metabolites.
There is a very clear and significant effect of watering on metabolism detectable on the level of primary, secondary and lipid metabolite content.
This effect is of a highly transitory nature. Thus, samples taken one and six hours after watering clearly differ from the non-treated samples, on one hand, and from each other, on the other hand. Besides, the effect of watering on metabolism vanishes after 12 hours (sample: at pre-dusk), thus confirming the results from the pilot experiment.
Some metabolites measured by the GC-MS platform (primary metabolism) vary with sampling time with the samples taken at dawn being different from those taken at the pre-dusk and midday. This was also evidenced by ANOVA, whereas no such influence is seen for the lipids (
In total, 357 primary metabolites could be detected via GC-TofMS analysis of which 118 could be annotated (cf.
(
Shown is the average abundance of several complex lipids visualized in a false-color heatmap at the three time points before and after watering ordered according to their presence in photosynthetic membranes, in cellular membranes or representing storage lipids.
The increase in concentration for most amino acids seen here in case of
We do not know the reason underlying these differences however except the use of a different plant system the experiments described here were performed under natural conditions in the field and furthermore the time intervals at which samples were taken after drought respectively rehydration treatment were as a rule longer in these studies as compared to our study. As described above all changes are of a highly transient nature.
Another aspect of our study worth mentioning is that the pattern of the amino acid concentration with respect to the three time points is very similar in the control and the watered conditions, i.e., the highest level is reached 1 hour after dawn in both control and watered samples with the concentrations subsequently decreasing towards midday and pre-dusk. Specifically the fact that the amino acid abundance in the samples taken at dawn before watering is also higher as compared to the midday and dusk samples is an independent confirmation of the response of the amino acids to watering.
Concerning TCA cycle intermediates the situation is similar though less pronounced. Whereas pyruvate and fumarate and, to a lesser extent, succinate display a significant increase as a result of watering, oxo-glutaric and malic acids essentially remain unchanged, whereas citric acid actually shows a decrease (cf.
With respect to sugars and sugar alcohols, a more complex picture emerges. Glucose and fructose largely remain unchanged with only the one hour value being lower in the watered as compared to the non-watered control. Maltose is initially reduced in the water control which could be due to either an increase in maltose consumption or a decreased starch degradation. Sucrose, raffinose and maltitol are believed to serve as osmoprotectant. All three compounds display a significant reduction in watered as compared to non-watered control at the first two time points. This observation could be taken as indication that
With respect to osmoprotecant sugars and sugar alcohols most studies analyzed their behavior in response to drought stress and not surprisingly an increase has been described in most studies
The significant transitory decrease in malonic acid described in our study is interesting when connecting this observation with the lipidomics data where we observed a transitory decrease in storage lipids (triacylglycerides). Taken together, this might suggest a reduced flux into storage lipids as an early response to watering.
Finally, it is noteworthy to comment on the behavior of glycolate and glycine (
The non-polar phase of the extracts was subjected to UPLC-MS measurements and we identified 133 lipids belonging eight different classes: Diacylglycerol (DAG), Mono-galactosyl-diacylglycerol (MGDG), Di-galactosyl-diacylglycerol (DGDG), Sulfoquinovosyl-diacylglycerol (SQDG), Phosphatidylcholine (PC), Phosphatidylserine (PS), Phosphatidylinositol (PI), Phsophatidylethanolamine (PE), Triacylglycerols TAG). Lipid species within each class are characterized by the number of C-atoms and by the number of double bonds in the acyl-chains.
As visible from
Boxplot-visualizations for a subset of complex lipids as determined for the three independent samples for the different time points and treatments as indicated on the x-axis of three replicates that were harvested at 1 h post-dawn, midday and 1 h pre-dusk for control and rehydrated plants.
A contrasting picture emerges for the storage lipids, namely TAG's. Here for all classes, we observe a fast and significant decrease for the first two time points after watering and a reversion to the non-watered condition at the third time point.
These results largely agree with data described for other plant systems. Thus galactolipids have been described consistently to be reduced as a result of dehydration (thus mirroring the decline after the first transient increase after watering) although a change in the ratio between MGDG's and DGDG's is not obvious in our case. Also the transient decrease observed for TAG's as a result of watering is in agreement with the described data (an increase in TAG's as a result of drought stress;
Extracts of
As shown in
As evident from
A time-resolved metabolomics and lipidomics response of
Key observations are the transitory decrease in maltitol and raffinose (indicating a reduced drought stress), an increase in essentially all amino acids which might suggest that increased protein biosynthesis takes place, an increase in all structural lipids of the photosynthetic membranes (DGDG's, MGDG's, SQDG's) which may suggest that the plant prepares itself for increasing its photosynthetic capacity as well as an increase in most other membrane lipids.
Thus most changes observed and specifically their kinetics suggest that water availability in the natural habitat of
GC measured metabolite list and ANOVA values for individual metabolites.
(XLSX)
Lipid list and ANOVA values for individual lipids.
(XLSX)