Leaf-level metabolic changes in response to drought affect daytime CO2 emission and isoprenoid synthesis pathways

Abstract

In the near future, climate change will cause enhanced frequency and/or severity of droughts in terrestrial ecosystems, including tropical forests. Drought responses by tropical trees may affect their carbon use, including production of volatile organic compounds (VOCs), with implications for carbon cycling and atmospheric chemistry that are challenging to predict. It remains unclear how metabolic adjustments by mature tropical trees in response to drought will affect their carbon fluxes associated with daytime CO2 production and VOC emission. To address this gap, we used position-specific 13C-pyruvate labeling to investigate leaf CO2 and VOC fluxes from four tropical species before and during a controlled drought in the enclosed rainforest of Biosphere 2 (B2). Overall, plants that were more drought-sensitive had greater reductions in daytime CO2 production. Although daytime CO2 production was always dominated by non-mitochondrial processes, the relative contribution of CO2 from the tricarboxylic acid cycle tended to increase under drought. A notable exception was the legume tree Clitoria fairchildiana R.A. Howard, which had less anabolic CO2 production than the other species even under pre-drought conditions, perhaps due to more efficient refixation of CO2 and anaplerotic use for amino acid synthesis. The C. fairchildiana was also the only species to allocate detectable amounts of 13C label to VOCs and was a major source of VOCs in B2. In C. fairchildiana leaves, our data indicate that intermediates from the mevalonic acid (MVA) pathway are used to produce the volatile monoterpene trans-β-ocimene, but not isoprene. This apparent crosstalk between the MVA and methylerythritol phosphate pathways for monoterpene synthesis declined with drought. Finally, although trans-β-ocimene emissions increased under drought, it was increasingly sourced from stored intermediates and not de novo synthesis. Unique metabolic responses of legumes may play a disproportionate role in the overall changes in daytime CO2 and VOC fluxes in tropical forests experiencing drought.

Keywords: GC–IRMS; PTR–TOF–MS; daytime respiration; legumes; position-specific isotope labeling; tropical plants; volatile organic compounds.

Figure 1

Schematic representation of drought responses by four plant functional groups identified in the B2WALD drought, modified from Werner et al. 2021. Drought-sensitive canopy trees were responsible for the majority of ecosystem fluxes before the drought, but they rapidly dropped leaves and reduced transpiration demand as upper soil layers dried. Their drought-tolerant counterparts never moved water through their stems as quickly but had much smaller reductions in the activity and leaf water potential in response to drought. Meanwhile, in the understory, drought tolerance was largely driven by microclimate variability, with shaded plants having low fluxes of carbon and water throughout the study period that did not change with drought and were most likely determined by limited light availability. Understory plants in locations with more light availability, either from canopy gaps or windows, were much more sensitive to drought and were the most stressed plants by the end of the experimental drought.

Figure 2

Schematic representation of important processes for the synthesis of isoprenoids and production of CO₂ within plant cells, with several intermediate steps removed for clarity. Green arrows represent possible movement of carbon from the C1 position of pyruvate, and purple arrows represent movement of carbon from the C2 position of pyruvate. Numbers represent the relevant steps discussed. 1: Transport of pyruvate into mitochondria, followed by initial decarboxylation of the C1-position and entrance into the TCA–cycle. 2: Conversion of pyruvate to GA-3-P and transport into the chloroplast. Pyruvate might also be transported directly to the chloroplast. All carbon positions of pyruvate remain in both cases. 3: Entrance of CO₂ into the Calvin cycle and addition of GA-3-P and other assimilation products into the metabolic pool of the chloroplast. 4: Initial reaction of pyruvate and GA-3-P to form DMAPP and IPP via MEP pathway, followed by reaction of two DMAPP/IPP molecules to produce GPP. As GA-3-P is not decarboxylated in this reaction, all carbon positions of the converted GA-3-P remain in the produced isoprene and monoterpenes. Pyruvate, on the other hand, is decarboxylated, resulting in only the C2-position remaining in these products. 5: Entrance of pyruvate into MVA pathway and initial decarboxylation of the C1-position of pyruvate to form DMAPP, IPP and subsequently GPP in the cytosol. After transport of these metabolites incorporating the ¹³C2-label into the chloroplast, DMAPP and IPP can either enter isoprene or monoterpene synthesis, while GPP is used to synthesize monoterpenes. 6: Anaplerotic replenishment of OAA to the TCA cycle from PEP and CO₂, which can support the production of carbon backbones for N-containing compounds.

Figure 3

Representative leaves (a) and rates of (b) carbon assimilation A, (c) transpiration E and (d) instantaneous WUE_i measured on individual leaves of C. fairchildiana, P. auritum, H. rosa sinensis and P. dioica under pre-drought (green) and drought (brown) conditions. Boxes represent the median and 25–75% range of 7–12 replicate leaves. Whiskers and outliers calculated using Tukey’s method. Significant differences between pre-drought and drought conditions for each species are indicated as ^* when P < 0.05, ^** when P < 0.01 and ^*** when P < 0.001.

Figure 4

Emission rates of (a) isoprene and (b) monoterpenes measured on individual leaves of C. fairchildiana and P. auritum under pre-drought (green) and drought (brown) conditions. Fluxes from C. fairchildiana are plotted on the left y-axis and fluxes from P. auritum on the right y-axis. Boxes represent the median and 25–75% range of 7–12 replicate leaves. Whiskers and outliers are calculated using Tukey’s method. Significant differences between pre-drought and drought conditions for each species are indicated as: ^* when P < 0.05.

Figure 5

The % of ¹³C from ¹³C1-labeled pyruvate (left column) and ¹³C2-labeled pyruvate (right column) that was released as CO₂ by C. fairchildiana, P. auritum, H. rosa sinensis and P. dioica under pre-drought (a) and drought (b) conditions. Boxes represent the median and 25–75% range of three to six replicate leaves. Whiskers are calculated using Tukey’s method. Significant differences among groups as determined by three-way ANOVA with Tukey’s post hoc test are indicated as: ^* when P < 0.05, ^** when P < 0.01, ^*** when P < 0.001 and ^**** when P < 0.0001.

Figure 6

(a) ¹³C enrichment of CO₂ from ¹³C1-labeled pyruvate (y-axis) plotted relative to ¹³C enrichment of CO₂ from ¹³C2-labeled pyruvate (x-axis) during pre-drought (open symbols) and drought (closed symbols). (b) Same as panel a, but with the % of ¹³C from each position of the pyruvate label. Symbols represent mean values, and error bars represent 1 SD of three to six replicate leaves per species.

Figure 7

¹³C from ¹³C-labeled pyruvate incorporation into VOCs from C. fairchildiana. (a) ¹³C excess isoprene emissions relative to natural background ¹³C abundance for isoprene and (b) ¹³C enrichment of the monoterpene trans-β-ocimene relative to unlabeled measurements during pre-drought (green) and drought (brown) (four to six replicates per treatment). Boxes represent median and 25–75% range of 7–12 replicate leaves. Whiskers and outliers are calculated using Tukey’s method.

Figure 8

Schematic representation of pyruvate carbon partitioning among biosynthetic processes involved in the synthesis of isoprenoids and production of CO₂ within C. fairchildiana cells as well as proposed adjustments of these processes during drought. Several intermediate steps and involved enzymes are removed for clarity. Arrow thickness represents the relative amount of ¹³C1-/¹³C2-labeled pyruvate used for the indicated pathway during pre-drought conditions. Gray arrows indicate no changes in carbon utilization during drought, while brown arrows indicate a down-regulation of the pathway. Blue-colored ¹³C1/¹³C2 indicate an increase in excess ¹³C emissions of isoprene or monoterpenes during drought, while red color indicates a decline in excess ¹³C emissions. Numbers represent the relevant pathways discussed, with descriptions of pathways 1–5 described in the caption of Figure 1.