4. Discussion

Significant effects of [CO2] and N application rate and their interaction on tillering were founded in this study (Fig. 1). Gradual N enrichment promoted tillering under both ambient CO2 and eCO2 confirmed that N availability was a constraint to the rice tillering process. However, N promotion effect on tillering diminished at the N10 under the ambient CO2, but still showed an enhancement effect at the N15 under the eCO2. These differential responses suggested that N limitation was aggravated under the eCO2. This was further corroborated by the declining N content in leaf and sheath and the increment of carbohydrates under the eCO2. It implies that more N input is needed to take advantage of the CO2 fertilization effect in future eCO2condition in order to achieve more tillers. The new-born tillers actually had comparable or higher N contents under the eCO2 condition than under the ambient CO2 (Fig. 2), whereas leaf and sheath showed a significant decline of N content. The more dramatic decline of N content in the sheath than in the leaf was consistent with previous findings (Jilta et al, 1997; Huang et al, 2004). Therefore, there could be a redistribution or reallocation of N favoring towards the SAM (new-born tillers include novel SAM) under the eCO2. As the N constraint is persistent in an unmanaged niche (Reich et al, 2006; Mueller et al, 2013), the plants likely adapt to distribute more of their acquired N to cell proliferating active zones rather than maintaining the original distribution pattern. This implies that future production requires more N input to take advantage of eCO2, detail may vary in different species and varieties.
To reveal the molecular mechanism underlying this change, we conducted RNA-sequencing for a complete profile of transcriptomic alteration between the SAM and leaf tissues under four combinations of [CO2] and N application (Fig. 4). Though tissue differences predominantly defined the transcription profile globally, N application generally showed a more influential effect than the [CO2]. This is reasonable as N has been proved to be a major limiting factor in plant growth regulation (Xuan et al, 2017; Zhang et al, 2019). For a specific group of the N metabolism related genes, they displayed a consistency with the general transcriptomic differences, i.e., N application overrode the [CO2] influences as a more significant effector. However, when tiller-related genes were specifically picked out, they were affected more by the [CO2] than by the N application. We speculate that the redistribution of N favoring the SAM under the eCO2strengthened the impact on tiller-related gene regulations.
Further check on the putative functions of those N metabolism related genes altered by eCO2 revealed that a major portion of N receptors and transporters was inversely regulated in leaf and SAM. Among them, the LOC_Os10g40600 (Fig. 5F) encoded OsNRT1.1a/b, which have been proved to function in nitrate sensing, transport and remobilization (Hu et al, 2015; Fan et al, 2016; Wang et al, 2018b). In addition, there might be other genes that regulating the N signaling and transportation (Feng et al, 2011; Wang et al, 2019c) in response to eCO2. As evidenced here, gene OsNRT2.3a/b(LOC_Os01g50820, Fig. 5A) had a similar expression level (FPKM) withOsNRT1.1a/b , and two glutamate receptors (LOC_Os06g08910 and LOC_Os09g26160, Fig. 5G and I, respectively) also showed decent expression levels in both leaf and SAM, which were differently regulated in the tissues by [CO2] (Fig. 5). This suggests that they were coordinately regulated in the procedure of eCO2 adaptation. There is an integrative co-expression network that determines the N effect and NUE (Zhang et al, 2019.). Engineering one target gene could bring in certain effect on crop yield improvement or NUE (Zeng et al, 2017; Xu et al, 2019). However, stacking (pyramiding) more targets together, especially targets in certain important network pathways may eventually create substantial yield breakthrough. Our results here provided a list of potential targets for such an approach.
Knowledge of the underlying mechanism of eCO2 adaptation in crops can provide us useful information for breeders in genetic engineering (Schimel, 2006; Ainsworth and Ort, 2010; Fletcher, 2018). Our preliminary results showed that the N sensor and transporter genes play important roles in the response to eCO2. We believe that our results can be applied to certain other plants, as under the eCO2 conditions, the element stoichiometry reveals a generally consistent reduction of N content in most terrestrial plants, but their growth and biomass accumulation are accelerated to a variating degree (Reich et al, 2006; Norby et al, 2010; Deng et al, 2015). The plausible reason probably lies in that plants redistribute N to the division active zones such as SAM and new-branches in a different way. This might be a universal adaptation strategy in most terrestrial plants. Consequently, genetic engineering on these targets that facilitate the N redistribution may improve the CO2harness and yield in those plants.
In summary, eCO2 increased carbohydrates in all organs of rice plant and lowered N content in leaves and sheath, as expected. However, the new-born tillers maintained N content comparable to or higher than those under control CO2 condition. The redistribution among the organs was likely accomplished by coordinated regulations of N metabolism receptor and transporter expression in leaf and SAM. The preliminary understanding of the mechanism provides a group of putative genes for further functional investigation and breeding targets.