Genetic Dissection of Light-Regulated Adventitious Root Induction in Arabidopsis thaliana Hypocotyls

Photomorphogenic responses of etiolated seedlings include the inhibition of hypocotyl elongation and opening of the apical hook. In addition, dark-grown seedlings respond to light by the formation of adventitious roots (AR) on the hypocotyl. How light signaling controls adventitious rooting is less well understood. Hereto, we analyzed adventitious rooting under different light conditions in wild type and photomorphogenesis mutants in Arabidopsis thaliana. Etiolation was not essential for AR formation but raised the competence to form AR under white and blue light. The blue light receptors CRY1 and PHOT1/PHOT2 are key elements contributing to the induction of AR formation in response to light. Furthermore, etiolation-controlled competence for AR formation depended on the COP9 signalosome, E3 ubiquitin ligase CONSTITUTIVELY PHOTOMORPHOGENIC (COP1), the COP1 interacting SUPPRESSOR OF PHYA-105 (SPA) kinase family members (SPA1,2 and 3) and Phytochrome-Interacting Factors (PIF). In contrast, ELONGATED HYPOCOTYL5 (HY5), suppressed AR formation. These findings provide a genetic framework that explains the high and low AR competence of Arabidopsis thaliana hypocotyls that were treated with dark, and light, respectively. We propose that light-induced auxin signal dissipation generates a transient auxin maximum that explains AR induction by a dark to light switch.


Introduction
Light, as with many other environmental factors, is of fundamental importance for the autotrophic growth of plants [1]. Therefore, it is not surprising that plants tune their body plan through the correct positioning of aerial organs or to accelerate elongation to outgrow competing plants to optimally capture and use the available light [2,3]. In fact, light triggers a major developmental transition just after germination [4]. Seedlings that germinated in the dark, grow almost exclusively by elongation in their search for light, which in turn signals that they have reached the soil surface. This is referred to as skotomorphogenesis. Once seedlings emerge into the light, plants develop a de-etiolated morphology (photomorphogenesis), including a short hypocotyl and open, expanded and green cotyledons [5], and the development of adventitious roots (AR) [6] The latter is, however, a poorly understood effect that is associated with the induction of photomorphogenesis. In fact, light has been reported to both stimulate and inhibit rooting [7]. While the dark to light switch is their Aux/IAA counterparts IAA6, IAA9 and IAA17 [61], and downstream transcription factors ARF6, ARF8 [62] and the ARF7/ARF19 pair [63], have been implicated in AR formation. Light is an important regulator of AR formation. However, its effects on AR formation are somewhat ambiguous. It has been hypothesized that light stimulates AR formation via activating ARF6/8 [64], which was corroborated by the direct interaction between CRY1 and PHYB with ARF6 and ARF8 [65]. However, this was contradicted by  who showed that light, through PHYB, suppresses AR formation by stabilizing Aux/IAA proteins and by suppressing ARF7/19 activity [7]. These seemingly conflicting data illustrate the complexity by which light modulates auxin activities to control AR development.
A dark period prior to light treatment has been shown to enhance the rooting capacity of many plant species [66,67]. Insight into the genetic factors involved in light control of rooting is of primordial importance to crop propagation. Moreover, there are seemingly contradicting reports on how light and darkness affect adventitious root formation. The purpose of this study was to establish a mechanistic framework of light-regulated AR formation. Therefore, we used etiolated Arabidopsis thaliana seedlings as a well-established genetic model to explore how different light signaling and photomorphogenesis pathways control AR formation. In addition to the regulation of hypocotyl adventitious rooting capacity by the COP9 signalosome (CSN), SPA1, 2 and 3, and PIFs, we identified the transcription factor HY5 as a strong inhibitor of rooting competence. Together, these findings clearly position key regulators of light signaling and photomorphogenesis in the context of light-controlled AR induction. Based on our observations and known target processes of the respective components, we propose that the transition from dark to light triggers auxin redistribution and auxin sensitivity changes that result in AR inductive auxin activity maxima.

Hypocotyl Adventitious Root Initiation Is Triggered by De-Etiolation
Light sensing triggers photomorphogenesis in dark-grown seedlings [68]. At the level of the hypocotyl, this includes the arrest of elongation [69], apical hook opening [70] and induction of adventitious roots (AR) [36]. The role of light and the process of photomorphogenesis on AR formation is, however, the least well understood. Therefore, we chose to study this in Arabidopsis thaliana, in which many mutants in light signaling and photomorphogenesis are readily available. The exposure of etiolated seedlings to light induced AR formation in Arabidopsis thaliana hypocotyls [71,72], while no AR appeared in the hypocotyls of seedlings that were germinated in the light (Figure 1a,b). Light-grown seedlings did not show indications of cell divisions or arrested primordia, which was visualized upon tissue clearing in the etiolated seedlings exposed to light (Figures 1c-e and S1). An average of 1.487 AR primordia (ARP) were observed in hypocotyls of the etiolated seedlings transferred to light conditions (Figure 1f). Etiolation is therefore a critical precondition for light-dependent AR induction in Arabidopsis thaliana hypocotyls.

AR Formation in Etiolated Seedlings Is Stimulated by Blue, but Not by Red Light
Next, we aimed to characterize the effects of different light conditions on AR formation. We found that etiolated seedlings transferred into the light produced more AR than those kept in the dark ( Figure 2). Continuous darkness, however, did not prevent AR induction, and AR primordia were observed in cleared hypocotyls" indicating that light is not essential, yet enhances the capacity to form AR (Figure 2a,c). To determine the sensitivity to different light wavelengths, etiolated seedlings were exposed to white, blue, red and far-red LED light. The transfer of three days etiolated seedlings into the light strongly suppressed hypocotyl elongation with blue light showing the strongest inhibition (Figure 2a,b). Blue and white light stimulated AR formation, whereas red and far-red did not show a significant increase in the number of AR compared to seedlings kept in darkness (Figure 2c). The

AR Formation in Etiolated Seedlings Is Stimulated by Blue, but Not by Red Light
Next, we aimed to characterize the effects of different light conditions formation. We found that etiolated seedlings transferred into the light produced m than those kept in the dark (Figure 2). Continuous darkness, however, did not pre induction, and AR primordia were observed in cleared hypocotyls,, indicating t is not essential, yet enhances the capacity to form AR (Figure 2a,c). To determ

Blue Light Signaling Contributes to AR Formation
Since blue light is perceived by cryptochromes (cry1 [73], cry1cry2 [74]) an phototropins (phot1phot2 [75]), we analyzed the contributions of these receptors t hypocotyl elongation and AR formation. Hypocotyls elongated in the dark (Figures 2 and 3e), and this was suppressed by white ( Figure 2b) and blue (Figures 2b and 3e) ligh as expected. This suppression was significantly less in cry1, cry1cry2 and phot1pho mutants (Figure 3b). In the dark, AR formation in the photoreceptor mutants did not diffe from the wild type controls (Figure 3d,f). In white light, ARP formation was strong reduced in cry1 and cry1cry2 (Figure 3a,c), suggesting that blue light receptors are mor important for AR formation than red and far-red signaling. Therefore, we explored i more detail the blue light signaling components ( Figure 3). In white light, th cryptochrome mutants cry1 and cry1cry2 had longer hypocotyls than WT (Ler) an phot1phot2 (Figure 3a,b), while AR formation in all tested blue light receptors wa significantly reduced (Figure 3a,c). In blue light, a similar response was observed, excep for the phot1phot2 mutant that was responsive to blue light by inhibiting hypocoty elongation (Figure 3d,e) and stimulating AR formation (Figure 3d,f). These data sugge a prominent role for CRY in hypocotyl elongation and AR formation, whereas PHOTs ar not required for elongation and play only a minor role in AR formation. n-far-red = 51). Different letters indicate a significant difference at p ≤ 0.05 (ANOVA and LSD post hoc analysis). Scale bar: 5 mm. All images were taken at the same magnification.

Blue Light Signaling Contributes to AR Formation
Since blue light is perceived by cryptochromes (cry1 [73], cry1cry2 [74]) and phototropins (phot1phot2 [75]), we analyzed the contributions of these receptors to hypocotyl elongation and AR formation. Hypocotyls elongated in the dark (Figures 2b and 3e), and this was suppressed by white ( Figure 2b) and blue (Figures 2b and 3e) light as expected. This suppression was significantly less in cry1, cry1cry2 and phot1phot2 mutants (Figure 3b). In the dark, AR formation in the photoreceptor mutants did not differ from the wild type controls (Figure 3d,f). In white light, ARP formation was strongly reduced in cry1 and cry1cry2 (Figure 3a,c), suggesting that blue light receptors are more important for AR formation than red and far-red signaling. Therefore, we explored in more detail the blue light signaling components ( Figure 3). In white light, the cryptochrome mutants cry1 and cry1cry2 had longer hypocotyls than WT (Ler) and phot1phot2 (Figure 3a,b), while AR formation in all tested blue light receptors was significantly reduced (Figure 3a,c). In blue light, a similar response was observed, except for the phot1phot2 mutant that was responsive to blue light by inhibiting hypocotyl elongation (Figure 3d,e) and stimulating AR formation (Figure 3d,f). These data suggest a prominent role for CRY in hypocotyl elongation and AR formation, whereas PHOTs are not required for elongation and play only a minor role in AR formation.

CSN Subunits Play Differential Roles in AR Initiation
Since de-etiolation by white and blue light increased the ARP number, we asked whether activation of photomorphogenesis was sufficient for AR induction. To test this, constitutively photomorphogenic mutants of the COP9 signalosome were analyzed. Constitutive photomorphogenesis mutants produce short hypocotyls in the dark [76,77] and since hypomorphic alleles have been identified, namely, csn2-5 [78], csn3-3 [79], csn5a-1 [80], csn5a-2 [80] and csn5b-1 [80], these were analyzed. These weak alleles did not show the short hypocotyl phenotype and instead, with the exception of csn3-3, hypocotyls were longer than the corresponding WTs (Figure 4a,b).
Previously, a weak allele of CSN4 (csn4-2035) was identified as a suppressor for excessive AR production in the auxin-overproducing superroot2-1 (sur2-1) mutant [36]. The csn2-5 and csn3-3 mutants formed fewer ARP than WT. The full and partial CSN5A knock-out alleles, csn5a-1 and csn5a-2, respectively [80], hardly formed any ARPs. In contrast, the full knock-out in CSN5B (csn5b-1) [80] formed significantly more ARPs than WT (Figure 4a,c). This antagonistic function of CSN was also observed previously and reflects the differential contributions of the CSN5 subunits in AR initiation [36]. Taken together, these results suggest that activation of photomorphogenesis is not sufficient for AR stimulation.  Previously, a weak allele of CSN4 (csn4-2035) was identified as a suppressor for excessive AR production in the auxin-overproducing superroot2-1 (sur2-1) mutant [36]. The csn2-5 and csn3-3 mutants formed fewer ARP than WT. The full and partial CSN5A knock-out alleles, csn5a-1 and csn5a-2, respectively [80], hardly formed any ARPs. In contrast, the full knock-out in CSN5B (csn5b-1) [80] formed significantly more ARPs than WT (Figure 4a,c). This antagonistic function of CSN was also observed previously and reflects the differential contributions of the CSN5 subunits in AR initiation [36]. Taken together, these results suggest that activation of photomorphogenesis is not sufficient for AR stimulation.

COP1/SPA Complex Plays a Role in Dark-Light-Induced AR Initiation
The suppression of photomorphogenesis in dark-grown Arabidopsis thaliana seedlings also requires, next to COP1, the activity of members of the SUPPRESSOR OF PHYA-105 (SPA) kinase family [81]. SPA kinases interact with and activate COP1 [82,83], forming a complex that promotes ubiquitination and degradation of ELONGATED HYPOCOTYL5 (HY5) to repress photomorphogenesis in the dark. Knocking out this COP1/SPA pathway is lethal or leads to developmental defects including dwarfism and early flowering [84], precluding the assessment of post-embryonic processes such as AR development. Therefore, we focused on mutants of the SPA family [22] that can complete the life cycle [82].

COP1/SPA Complex Plays a Role in Dark-Light-Induced AR Initiation
The suppression of photomorphogenesis in dark-grown Arabidopsis thaliana seedlings also requires, next to COP1, the activity of members of the SUPPRESSOR OF PHYA-105 (SPA) kinase family [81]. SPA kinases interact with and activate COP1 [82,83], forming a complex that promotes ubiquitination and degradation of ELONGATED HYPOCOTYL5 (HY5) to repress photomorphogenesis in the dark. Knocking out this COP1/SPA pathway is lethal or leads to developmental defects including dwarfism and early flowering [84], precluding the assessment of post-embryonic processes such as AR development. Therefore, we focused on mutants of the SPA family [22] that can complete the life cycle [82].
proteolysis of bZIP transcription factors LONG HYPOCOTYL5 (HY5) HOMOLOGUE (HYH), and in the light they are stabilized, activating lightgene expression and photomorphogenesis [89]. Consistent with this dark-light system, hy5 and hyh hypocotyls were longer than WT (Figure 5a,b) and formed (Figure 5a,c). These results affirm that photomorphogenesis primarily supp formation.

Skotomorphogenesis PIF Factors Are Required for AR Formation
Phytochrome-Interacting Factors (PIFs) are basic helix-loop-helix transcription factors (TFs) that accumulate in the dark to promote skotomorp [90,91] and act antagonistically to HY5 in light responses [41,92,93]. To determin the etiolation factors PIF control AR formation, phenotypic responses in da growth conditions were analyzed for single mutants pif1-1, pif3-7, pif4-2 and pif multiplex mutant pifQ [42]. The single mutants developed slightly longer hypo the wild type (Figure 6a,b). In contrast, pifQ had shorter hypocotyls (Fig  consistent with PIF

functional redundancy in suppressing photomorphogenes
Similarly, the single knock-out mutants showed moderate to no chan formation, while pifQ formed very few or no ARP four days after transfer into (Figure 6a,b). These results indicate that etiolation-induced AR competence d PIFs.  Similarly, the single knock-out mutants showed moderate to no changes in AR formation, while pifQ formed very few or no ARP four days after transfer into white light (Figure 6a,b). These results indicate that etiolation-induced AR competence depends on PIFs.

Light Has Contrasting Effects on AR Formation
Light plays a pivotal role in plant growth and development [96], with other environmental factors [97]. Darkness and low light conditions induce etiolation and light triggers photomorphogenic growth. The transition from dark to light causes the arrest of hypocotyl elongation, opening of the apical hook, activation of the shoot meristem, greening and the induction of adventitious roots [8,[98][99][100]. While many of these developmental processes have been studied intensively, AR induction is less well understood, and moreover, light has been reported to both stimulate and inhibit rooting [7].
The notion that photomorphogenesis and light signaling inhibit AR formation comes from the inhibition of AR formation in light-grown seedlings (Figure 1b). Mutants that express (partial) photomorphogenesis characteristics in the dark, e.g., cns, pifQ and spa1/2/3 triple ( Figure 5), and mutants expressing a constitutive active PHYB [7], show a severely reduced capacity of AR formation. Consistently, the incapacity to execute photomorphogenesis in the hy5 mutant [101] resulted in increased AR formation.
This complementarity leads to the interpretation that skotomorphogenesis is associated with increased AR competence. In Arabidopsis thaliana hypocotyls, there seems to be an optimal etiolation time for AR induction, after which the light-induced AR production becomes less efficient [7]. This is reflected in the lack of AR formation in pifQ mutants that are defective in the major skotomorphogenesis transcription factors ( Figure 6). Consistent with enhanced AR competence in etiolated seedlings, a dark treatment is often used to improve AR formation of cuttings as shown for Petunia hybrida and Prunus avium [66,67], and in micropropagation of various horticulture and tree crops as in, e.g., Acacia mangium and Malus domestica (Borkh.) Likhonos [102,103]. Therefore, this improved AR formation might be explained by a partial reversal of the photomorphogenetic state during the dark treatment, which alleviates the suppression of AR formation.
The rooting responses of skotomorphogenesis and photomorphogenesis Arabidopsis thaliana mutants illustrate how darkness installs a state of AR competence and light reduces AR formation. Light, however, stimulates AR formation when applied to etiolated seedlings within a limited time period [7]. Blue light and red light, but not far-red light, are AR inductive signals in etiolated hypocotyls [9]. In our experiments, stronger stimulation of AR formation by blue light than with red light was observed, presumably because we used higher intensities than in previous studies. Light intensity may therefore also be a factor determining the capacity to induce AR formation in etiolated seedlings.
The AR inducing effect by light has been proposed to be related to the activation of photosynthesis [104]. However, in our experiments, the blue-light effect was entirely dependent on the blue light receptors CRY1 and CRY2, and partially on PHOT1/PHOT2, indicating that photosynthesis played only a minor role. Photosynthesis may however contribute to AR formation during further development of the AR primordia by activating the small GTPase ROP2 and TOR kinase [105], leading to AR formation in potato [106].
In conclusion, our data show that light acts as an inhibitor of AR competence when applied to light-grown seedlings, while it functions as an AR stimulus in dark-grown seedlings.

Auxin Plays a Central Role in the Dual Effect of Light on AR Formation
Auxin is the central signaling hormone for AR induction. Its accumulation in the pericycle activates asymmetric cell division and subsequent AR organogenesis [61]. Light is therefore expected to control AR formation via auxin. One of the major molecular differences between light and darkness lies in the antagonism between PIF and HY5 transcription factors that are stable in darkness and light, respectively. A direct target of PIFs are auxin biosynthesis genes of the YUCCA family [107,108]. In the dark, PIFs stimulate auxin biosynthesis and act cooperatively with the auxin ARF6 and brassinosteroid BZR1 transcription factors [109]. ARF6 mediates auxin regulation of AR induction [34,62] and brassinosteroids stimulate AR formation in the auxin-overproducing sur2-7 mutant [110]. A similar cooperation between PIFs, ARF6 and BZR1 may control AR competence in the dark.
In light, PIFs are degraded and HY5 is stabilized, activating photomorphogenesisrelated genes [111]. We found that dark-grown pifQ and hy5 mutants form, respectively, less and more ARs upon transfer to light. One possible explanation for this phenomenon can be found in their respective known targets. While PIFs enhance auxin levels in the dark [112], HY5 suppresses auxin signaling in the light by inducing Aux/IAA signaling repressors SLR/IAA14 and AXR2/IAA7 [113] and suppressing YUCCA9 [114]. Consistently, it was recently found that extended darkness leads to the formation of ARs independent of a light stimulus [7]. This underlines the central role of auxin and darkness in determining the competence to form AR.
Next to auxin biosynthesis, auxin receptivity also determines AR formation. This follows from the observed reduction in AR in csn mutants (Figure 4) [36], which display auxin resistance [31] due to a direct impact on auxin co-receptor SCF TIR1/AFB activity [115] via its role in deconjugation of NEDD8/RUB1 of the CUL1 subunit in SCF complexes. The nuclear localization and HY5-degrading activity of COP1 in the dark depends on CSN [25], causing the constitutive photomorphogenic phenotype of csn mutants, revealing that CSN contributes to AR competence in etiolated seedlings possibly via two separate signaling routes.
In blue light, auxin signaling is mediated via cryptochromes and phototropins. CRY1 interacts with the transcription factors PIF4 and PIF5 [43,[116][117][118] and with SPA1 to suppress COP1-dependent degradation of the transcription factor HY5 [119,120]. In addition, CRY1 counteracts the association of TIR1 and AUX/IAAs [121] and represses DNA binding of ARF6 and ARF8 [65]. Cryptochrome-mediated blue light signaling thereby suppresses auxin signaling, and thus AR formation, in the hypocotyl. However, we found that CRY1 is required for blue light-induced AR formation ( Figure 3). CRYs, therefore, also positively influence auxin signaling in the context of AR formation, a process that awaits elucidation.
A hint for the mechanism by which blue light might regulate AR induction comes from phot mutant analysis (Figure 3). PHOT1-mediated blue light perception in AR formation was proposed to result in the modulation of auxin transport via PIN3 activity [9]. PHOT1 also inhibits the auxin transporter ABCB19, changing auxin flux from the shoot to the hypocotyl [122]. This is consistent with a model in which blue light affects auxin flux from the shoot to the hypocotyl and modulates PIN3-mediated auxin transport to the pericycle to induce AR formation.

Model for Dual Role of Light in AR Formation
Light and darkness have contrasting effects on auxin homeostasis and signaling. Here, we propose that the switch from darkness to light results in a transient and local accumulation of auxin in the pericycle that reconciles both effects on the induction of AR.
During skotomorphogenesis, auxin biosynthesis rates are high to stimulate rapid growth and to maintain the apical hook [123,124] (Figure 7a). The transfer to light and activation of photomorphogenesis causes an important relocalization of auxin and suppression of auxin signaling in the entire hypocotyl, resulting in an arrest of hypocotyl growth and the induction of apical hook opening (Figure 7b). We anticipate that the redistribution of auxin, further facilitated via blue light effects on auxin transport, results in local auxin maxima in the pericycle that are sufficiently strong to trigger asymmetric division and AR formation. Over time, light signaling and photomorphogenesis install auxin resistance at the level of the signaling machinery, precluding further AR induction (Figure 7c). Future studies to test this hypothesis require the analysis of temporal changes of auxin levels and transport in the Arabidopsis thaliana hypocotyl.

Light Sources
For light-response experiments, seedlings were grown in darkness for 3 days at 22 • C with 16 h/8 h light/dark cycles after stratified 4 days at 4 • C in dark and followed by stimulation under blue light (peak: 470 nm, half band width: 30 nm, 59.75 µmol/m 2 s), red light (peak: 660 nm, half band width: 20 nm, 41.62 µmol/m 2 s), far-red light (peak: 740 nm, half band width: 25 nm, 1.32 µmol/m 2 s) or darkness for 4 days for further analysis. The light condition was applied by Philips Greenpower light-emitting diode (LED) chambers, which are spectrally controllable by Philips GrowWise control system. Total incident light intensity and the spectral distributions of the different light sources were measured using a spectroradiometer (SS-110, Apogee Instruments, Logan, UT, USA).

Hypocotyl Phenotypic Analysis
Images of seedlings on vertical plates were taken with D7000 Nikon camera, AF-S VR Micro-Nikkor 105 mm f/2.8 G IF-ED lens, followed by manual analysis using the ImageJ software plugin (http://www.imagescience.org/meijering/software/neuronj/, accessed on 26 March 2021) [126] for hypocotyl length measurement.

Preparation and Observation of Cleared Seedings
For light microscopy, plant seedlings were cleared with methanol and NaOH and mounted as described in [127]. Seedlings were harvested and fixed in acetone (90%) overnight at 4 • C. After fixation, seedlings were transferred to 0.5 M phosphate buffer for 30 min at 37 • C, followed by 45 min in clearing solution I (0.24 N HCl in 20% methanol) at 60 • C and 15 min in Clearing Sol II (7% NaOH in 60% EtOH) at room temperature. Seedings were then rehydrated sequentially with ethanol series (40,20 and 10%) at room temperature for 5 min each step and infiltrated for at least 1 h with 5% ethanol in 25% glycerol. Subsequently, the entire seedlings were mounted in 50% glycerol (our trick to keep the seedlings straight: putting the cleared seedlings on the coverslip rather than on the microscopy slide) and adventitious root primordia were inspected by BX51 microscope (Olympus, Tokyo, Japan) using differential interference contrast (DIC) optics.

Statistical Analyses
Statistical analysis was performed using GraphPad Prism 8. One way ANOVA and post-hoc tests were used to assess differences between the mutant lines (p < 0.05). Nonpaired Student t-test was used for two group comparison (Figure 1f). All mutants in Figures 5-7, were analyzed in the same runs, resulting in duplication of the WT data, upon splitting the data up over different figures. For all comparisons, at least three independent experiments were performed, each with more than 20 seedlings for mutant lines.

Conclusions
Despite AR formation being intimately connected to light signaling and photomorphogenesis, two heavily studied physiological processes, the mechanism by which these processes converge on AR formation has remained elusive. Here, we analyzed the AR competence of different mutants in key positions in these pathways, providing a first outline of the genetic framework that controls light-controlled AR formation. Integrating these findings with known targets of these components led us to propose the above model based on the contrasting effects of light and darkness on auxin signaling and homeostasis. Validation of this model will require mapping and modeling of the spatio-temporal characteristics of auxin signaling and homeostasis in tissues relevant for dark to light-induced AR formation, and how this is affected in the used light signaling and photomorphogenesis mutants. A deeper understanding of the molecular mechanism of light-regulated AR formation may inspire novel strategies for improving AR formation during clonal propagation of crop and ornamental species.

Informed Consent Statement: Not applicable.
Data Availability Statement: Raw data supporting reported results are stored on a server in accordance to rules outlined by the data management plan of Ghent University (www.ugent.be/en/ research/datamanagement/before-research/datamanagementplan.htm, accessed on 6 April 2022).