Promotion of growth by elevated carbon dioxide is coordinated through a flexible transcriptional network in Arabidopsis

Abstract

Although gibberellins (GAs) promote many developmental responses in plants, little is known about how the hormone interacts with environmental signals at the molecular level for regulating plant growth. Recently, we have demonstrated that inhibition of growth by the GA biosynthesis inhibitor paclobutrazol (PAC) at ambient [CO₂] (350 µmol CO₂ mol^-1) is reverted by elevated [CO₂] (750 μmol CO₂ mol^-1). Our finding points to an important role of elevated [CO₂] as a signal allowing higher growth rates of low-GA plants. GA promotes plant growth via a complex transcriptional network that integrates multiple signaling pathways. Herein, we discuss how elevated [CO₂] stimulates biomass accumulation in a GA-independent manner by regulating the expression of growth-related genes.

Keywords: :

• elevated carbon dioxide concentration
• transcription factor
• gibberellin
• growth
• paclobutrazol

This article refers to:

Extensive molecular genetic and biochemical studies in Arabidopsis thaliana have illustrated a GA signal transduction pathway from the GA INSENSITIVE DWARF 1 (GID1) receptor to DELLA proteins, which integrates brassinosteroid (BR) and light signals.^1-3 DELLAs repress transcriptional activity of the phytochrome-interacting factors (PIFs) and brassinazole-resistant (BZR) transcription factors (TFs) to regulate a large number of genes. Among these are genes encoding members of the paclobutrazol resistance (PRE) family of helix-loop-helix (HLH) TFs, which promote cell elongation by antagonizing several inhibitory HLH factors.⁴ Hence, the DELLA-BZR-PIF framework can explain how plants integrate GA, BR and light signals to optimize growth and development in response to changing environments. The increase of atmospheric [CO₂] during the last decades has promoted a growing interest in the function of this environmental factor on plant growth. Elevated [CO₂] often leads to a stimulation of leaf growth by triggering both, cell expansion and cell division.^5,6 As such, (elongation) growth is also regulated by multiple hormones including GA and BR.^4,7 However, regulatory pathways that firmly connect hormonal control of growth with altered carbon assimilation have yet to be established.

In our original paper⁸ we showed that growth retardation induced by PAC treatment is overcome by elevated [CO₂] through different metabolic and expressional pathways. Elevated [CO₂] increased enzyme activities of the Benson-Calvin cycle and nitrate reductase to the greatest extent in GA-limited plants which also showed the greatest growth stimulation.⁸ In addition, at ambient [CO₂] GA deprivation caused a decreased expression of genes associated with cell wall and lipid metabolism while these genes showed an upregulation after GA treatment.^8,9 However, under elevated [CO₂] the expression of these genes was induced during PAC treatment. Moreover, both under ambient and elevated [CO₂] PAC treatment affected the transcription of GA signaling and biosynthesis genes in a similar manner indicating that restoration of growth of PAC-treated plants by elevated [CO₂] might be driven by a direct effect on the expression of cell elongation genes independent of GA. Of note, enhancement of photosynthetic capacity by BR is associated with increases in gene transcription and the activity of enzymes of the Benson-Calvin cycle.^10,11 In this context, BR could also trigger a transient generation of reactive oxygen species (ROS) which may increase CO₂ assimilation by modifying the redox signal system.¹² Together, these observations indicate the presence of a sophisticated regulation network that coordinates growth depending on carbon supply and plant hormone levels. This regulation might have a major impact for discussion of how, and to what extent, plant growth can be optimized in a fluctuating environment and in particular under conditions of elevated atmospheric [CO₂].

Coordination of growth by GA, BR and light signals involves a common transcriptional network.^1-3 To understand the significance of changes of individual transcripts in response to elevated [CO₂], the expression of growth-related genes was analyzed in plants treated with PAC and/or GA grown under ambient or elevated [CO₂]. The PRE family of small HLH proteins has been characterized as positive regulators of cell elongation acting immediately downstream of BZRs and PIFs.^7,13 Moreover, overexpression of PRE1 suppresses the gibberellin insensitive-1 (gai-1) and ga2-201 phenotypes while all members render PAC resistance when ectopically expressed.¹³ Elevation of [CO₂] itself has only a minor effect on the expression of PRE genes (Fig. 1). Treatment with PAC (leading to GA deprivation) decreased the expression of PRE1 and PRE5 in plants grown at both, ambient and elevated [CO₂]. In elevated [CO₂] the expression of both genes was induced by GA treatment, while under ambient [CO₂] only PRE5 was upregulated. In elevated [CO₂], PRE2 and PRE4 showed a significant upregulation upon GA deprivation, whereas PAC treatment did not significantly affect the expression level of these genes at ambient [CO₂] (Fig. 1). Together, these data indicate that different members of the PRE family can mediate regulation of cell elongation at different [CO₂]. The expression of PRE1, PRE2, PRE5 and PRE6 is induced by short-term GA treatment in wild-type Arabidopsis plants, but their induction is decreased in the BR-signaling mutant bri1-119.² PAC treatment has opposing effects on the expression of PRE2 and PRE4 under ambient and elevated [CO₂]. Therefore, it appears that PRE2 and PRE4 are common nodes for the regulation of cell elongation by multiple growth-promoting hormones under elevated [CO₂]. High [CO₂] does not overcome the repression of PRE1 and PRE5 by PAC treatment, but might compensate for this affect through the activation of PRE2 and PRE4 to induce plant growth. Furthermore, the induction of PRE1 and PRE5 by GA under high [CO₂] might explain the significantly increased rosette area of these plants under these conditions as compared with wild type.⁸

Figure 1. Expression pattern of PREs, PIFs and BR-related genes in shoots of Arabidopsis plants treated with PAC and/or GA₄₊₇ grown at 350 (left) or 750 (right) µmol CO₂ mol^-1 as determined by quantitative RT-PCR. The color scale represents the Log2 fold change values of expression changes compared with control (blue, reduced; yellow, increased). For plants treated at ambient [CO₂], non-treated plants were used as control. For untreated plants at high [CO₂], the relative expression is shown as compared with ambient [CO₂] while expression data for plants treated at high [CO₂] were normalized to non-treated plants at elevated [CO₂]. Data represent means of three independent biological replicates. Heatmap was generated with the MultiExperiment Viewer (MeV) software (www.tm4.org/mev). AGI codes: PRE1 (AT5G39860); PRE2 (AT5G15160); PRE4 (AT3G47710); PRE5 (AT3G28857); PRE6 (AT1G26945); PIF1 (AT2G20180); PIF3 (AT1G09530); PIF4 (AT2G43010); PIF5 (AT3G59060); PIF6 (AT3G62090); PIF7 (AT5G39860); BZR1 (AT1G75080); BZR2 (AT1G19350); IBH1 (AT2G43060); ARL (AT2G44080); DWF4 (AT3G50660); BLH1 (AT2G35940).

PIFs have been shown to function in GA signaling and they are required for growth in response to time of day, high temperature and other stimuli.^3,14,15 To evaluate to what extent elevated [CO₂] affects PIF genes in response to GA, we quantified the transcript levels of PIF family members. Elevated [CO₂] alone did not significantly affect the expression level of PIFs as compared with ambient [CO₂] (Fig. 1). Treatment with PAC led to an increase in the expression of PIF6 while GA treatment decreased the expression of this gene in plants grown at both, ambient or elevated [CO₂]. Thus, the GA sensing pathway might regulate the plant’s response via selective control of PIF6 abundance independent of the [CO₂]. At ambient [CO₂], expression of PIF1, -3, -4 and -5 was only slightly affected by GA deprivation, whereas PAC treatment increased the expression of these genes at elevated [CO₂] (Fig. 1). Overall, expression of PIF1, -3, -4 and -5 was lower in plants treated with GA, under ambient [CO₂] than in plants grown at elevated [CO₂]. PIF4 and PIF5 seem to be particularly important for growth as the pif4pif5 double mutant has a short-hypocotyl phenotype.¹⁶ Since in Arabidopsis growth inhibition induced by PAC is overcome by elevated [CO₂],⁸ factors in addition to PIF4 and PIF5 may play a significant role in promoting growth under elevated [CO₂]. Taken together, our results indicate that elevated [CO₂] functions as a positive modulator of members of the PIF family in low-GA plants. Such a dynamic change of four members of the PIF family would trigger a robust downstream gene expression response, which could contribute to growth of low-GA plants under elevated [CO₂].

Many developmental responses induced by GAs are similar to those induced by BR.¹⁷ Furthermore, direct crosstalk between GA and BR signaling pathways has been demonstrated in Arabidopsis.² In order to evaluate the effect of elevated [CO₂] on the relationship between BR and GA, the expression of BR-related genes was analyzed in plants treated with PAC or GA grown under ambient or elevated [CO₂]. An increase of [CO₂] itself has only a minor effect on the expression of BR-related genes (Fig. 1). At ambient [CO₂], treatment with PAC and/or GA resulted in a slight decrease in the expression of genes associated with BR. Interestingly, almost all BR-associated genes were upregulated in low- and high-GA plants grown under elevated [CO₂]. This highlights the exceptional flexibility in these hormone pathways in response to elevated [CO₂]. In other words, hormone sensing and signaling mechanisms permit a flexible and appropriate modulation of plant growth to changes in carbohydrate availability. At elevated [CO₂], there was a substantial stimulation of sugar and starch synthesis in plants treated with PAC and/or GA, similar to control.⁸ Furthermore, a restoration of growth of plants affected by PAC treatment under ambient [CO₂] was observed upon elevated [CO₂]. Thus, growth was faster at elevated [CO₂] in GA-limited plants because the storage and allocation of assimilates was altered.⁸

Together with the fact that elevated [CO₂] affected the BR-related genes in low- and high-GA plants, our findings imply that plants adjust their growth patterns to a change in carbon supply by involving multiple hormonal signaling pathways. Moreover, the availability of carbon seems to override hormone-regulated growth by immediately acting on the expression of effector genes. It appears that under carbon limiting conditions the action of plant hormones and other regulators to steer plant development is more strict and conservative. In some way the restoration of growth of PAC-treated plants by elevated [CO₂] is similar to observations made for photorespiratory mutants. These mutants show a strong inhibition of plant growth under ambient [CO₂], but are indistinguishable from wild-type plants in elevated [CO₂] as e.g., shown recently for mutants lacking functional mitochondrial transporter A BOUT DE SOUFFLE (BOU).¹⁸ Potentially, the reduced size of several hormone mutants might be rescued by elevated CO₂, however, to our knowledge this has not been reported yet.

In summary, biomass accumulation of plants with restriction of GA biosynthesis under elevated [CO₂] appears to be regulated by a flexible transcription network that integrates multiples signaling pathways. Thus, strategies to identify genes which orchestrate these signaling pathways will probably represent promising ways by which to identify new markers for growth potential, especially for future climate conditions.

Abbreviations:
TF	=	transcription factor
GA	=	gibberellin
PAC	=	paclobutrazol
[CO2]	=	carbon dioxide concentration

Acknowledgments

D.M.R. was supported by the Alexander-von-Humboldt Foundation.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.