Partially replacing sole blue light by green light reduced elongation independent of cry1a
Tomato stem length was significantly reduced by partially (20%) replacing sole blue light by green light, whereas partially replacing sole red or red/blue mixture with G did not affect length (Fig. 1). These effects were due to elongation of internodes as leaf number was not affected (Fig. 7). CRYs were reported to mediate hypocotyl elongation inhibition driven by sole blue light or sole green light compared to darkness in Arabidopsis, and G acts additively with B to drive cryptochrome-mediated inhibition of elongation (Wang et al., 2013). However, in our study partially replacing sole B by green light, indicated that G reversed the B-induced stem elongation in tomato (Fig. 1). CRY1a-deficient and CRY2 overexpressing lines (CRY2-OX3 and CRY2-OX8) showed similar responses of stem length to partially replacing sole B by G as the wild-type MM (Fig. 1). This might be attributed to the overlapping role of CRY1 and CRY2 for the inhibition of elongation (Azari et al., 2010; Giliberto et al., 2005), suggesting that the role of CRY2 is partially redundant with that of CRY1 in the control of stem elongation (Giliberto et al., 2005). This indicated that this green light response was independent from cry1a, probably independent from cry2 as well.
The stem length of the cry1a mutant was remarkably higher than for the other genotypes under the same light treatment (Fig. 1), confirming the involvement of CRY1a in the inhibition of internode elongation (Ninu et al., 1999). The overexpression of CRY2 in CRY2-OX3 and CRY2-OX8 induced shorter stems (Fig. 1), confirming also the involvement of cry2 under all light treatments (Yang et al., 2017).
Through blue light the neutral FAD chromophore in crys is converted into a photoexcited state (FADH·), absorbing green light which converts the cryptochromes into a fully reduced and inactive state (Lin, 2003; Banerjee et al., 2007; Bouly et al., 2007). Green light partially inhibits cry2 oxidation by blue light (Banerjee et al., 2007; Zeugner et al., 2005; Bouly et al., 2007; Frechilla et al., 2000), contributing to reduced levels of FADH·. However, this photocycle model could not explain all interactions between blue and green light on stem length, like the finding that G acts additively not reversely with B to drive cryptochrome-mediated stem growth inhibition in Arabidopsis (Wang et al., 2013). Green light did not inhibit blue light-mediated cry2 degradation and the expression of the FLOWERING LOCUS T gene (Li et al., 2011), and it induced similar response on stem elongation in CRY2 overexpressing lines to wild-type (Fig. 1). These results suggest that this photocycle between two FAD protein forms cannot explain the photoperception of crys in plant cells, and the specific mechanism underlying crys photo-excitation has not been identified (for review, see Yang et al., 2017).
The involvement of CRY2 in regulating plant photomorphogenesis
In contrast with MM and cry1a mutant, in CRY2-OX3 and CRY2-OX8 stem length was reduced when partly replacing sole R by G (Fig. 1), while partly replacing B by G induced a lower shoot: root ratio and smaller leaf area (not significant in CRY2-OX3) (Figs. 2 and 3). These results indicate the involvement of CRY2 in green light effects on stem length, shoot: root ratio and leaf area.
Comparing tomato transgenic CRY2 overexpressing lines with wild-type plants, CRY2 may control vegetative development and photosynthesis as suggested by high throughput transciptomic and proteomic analyses by Lopez et al. (2012), and by the overproduction of chlorophylls in CRY2 overexpressors (Giliberto et al. 2005). However, we did not observe significant differences between CRY2-OX3/OX8 and MM on SLA, photosynthesis, and chlorophyll content (Fig. 9, 5, and 10). We conclude that effects of CRY2 on phenotype are limited, which might result from its redundant role with CRY1a.
PHYs play role in blue light effects on elongation
Besides mediation by CRYs, the blue light effects might also be mediated by PHYs. The PSS value which is an indicator of phytochrome status, was lower under sole blue than that under all other light treatments; green light had little effect on the PSS value (Table 1). CRYs and PHYs converge blue and red light signals at different levels to co-regulate physiological responses, such as root greening, de-etiolation, shade avoidance symptoms, photoperiodic flowering, etc (Su et al., 2017). In contrast to the expectation that blue light triggered shorter plants due to involvement of cry, stems under sole B were not shortest (Fig. 1), indicating elongation might be counteracted by phytochrome action. The phytochrome effect may dominate stem elongation of cry, therefore longer plants were observed under sole blue resulting from less reduction in elongation due to cryptochromes.
Strikingly, similar to 100% B, 100% R also induced significantly longer cry1a mutant plants compared to MM (Fig. 1), consistent with the results of Fantini et al. (2019). On the contrary, Ninu et al. (1999) found that 8-days old CRY1a antisense tomato plants did not show an elongated hypocotyl under red light but under blue light (both approximately 8 µmol m− 2 s− 1). These different results might be caused by the fact that CRY1a gene is not knocked out but only downregulated in CRY1a antisense plants, or by differences in development stage or light intensity. Accumulating evidence in the model plant Arabidopsis has revealed that CRYs and PHYs share two mechanistically distinct pathways that coordinately regulate transcriptional changes in response to light. However, the role of photoreceptor interactions and the mechanism responsible for the direct convergence of CRYs and PHYs signals on the COP1/SPA complex or phytochrome-interacting factors (PIFs) remain elusive (Su et al., 2017).
Arabidopsis cryptochrome 1, phytochromes A, B1 and B2 are all capable of mediating responses to B under some circumstances (Weller et al., 2000). CRYs may act in a blue-light independent manner to affect PHY regulation of gene expression and development, resulting in different protein expression between the WT and cry1cry2 mutant Arabidopsis in red light as well as in blue light (Yang et al. 2008; Lopez et al., 2012). Arabidopsis CRY1 interacts directly with PIF4 in a blue light-dependent manner to repress the transcription activity of PIF4 (Ma et al., 2016). This indicates that stem elongation in cry1a mutants under sole R could be mediated by downstream genes shared by CRYs and PHYs (Facella et al., 2012; Su et al., 2017). However, the extent and relative importance of their individual contributions differ depending on irradiance, on which other photoreceptors are present, and on which plant process is examined.
Replacing 20% of red, red/blue or blue light by green had no significant effect on biomass production
McCree (1972) measured the instantaneous response of leaf photosynthesis to different spectra, finding that the quantum yield of photosynthesis of green photons (525 nm) can be about 25–30% less than that of red photons (675 nm), while the quantum yield of green is comparable to that of blue photons (450 nm). However, this may not be representative of whole plants or plant communities grown at high PPFD under mixed colors of light. Green light could drive carbon fixation deep within leaves (Sun et al., 1998), even more efficient than R or B (Nishio, 2000), because it could penetrate deep into the mesophyll layers (Smith et al., 2017). In our study where the light contained 0 or 20% green, the leaf photosynthesis rate, stomatal conductance and transpiration rate were not significantly affected by green light (Figs. 5 and 10). Similarly, the contents of chlorophyll a and b and carotenoids as well as their ratios, were hardly affected by green light (Fig. 9).
Partially replacing sole R or B or R/B mixture by green light did not cause differences in leaf area, SLA, shoot: root ratio and biomass of MM and cry1a mutant. This contradicts previous findings on green light responses, but in those studies PPFD also increased when adding G (Kim, 2005; Samuolienė et al., 2012; Novičkovas at al., 2012). Zhang et al. (2011) reported that 40% green light induced a shade avoidance response in Arabidopsis seedlings, whereas 10% did not. Too much G (51%) or too little (0%) decreased lettuce growth, while about 24% resulted in the highest growth rate (Kim et al., 2004). However, in our study 20% G did not induce much effects, which is comparable to the study of Hernández and Kubota (2015) who analyzed the effect of 28% G in cucumber. Kaiser et al. (2019) found that replacing 32% of a red/blue mixture spectrum by green light significantly increased plant biomass and yield. These different observations among studies suggest that G effects might be genotype-specific and dependent on and/or interact with other environmental conditions.