Light controls de novo adventitious root regeneration by modulating jasmonate and cytokinin homeostasis in Norway spruce hypocotyls

Vegetative propagation relies on the capacity of plants to regenerate adventitious roots (ARs) de novo. Light plays a central role in this process; however the physiological and molecular components mediating light cues remain largely elusive, especially in conifers. Here, we explore the effect of light spectral quality on AR initiation (ARI) in de-rooted Norway spruce seedlings. We used light emitting diodes (LEDs) to study the effects of different light spectra on ARI, hormone metabolism and gene expression. We used sensitive mass spectrometry-based methods, and, since the Norway spruce genome sequence recently became available, we coupled this to gene expression analysis. We performed extensive anatomical characterization to investigate ARI at the cellular level. We showed that in contrast to constant white light (WL) and constant blue light (cBL), constant red light (cRL) promoted ARI by reducing jasmonate (JA) and JA-isoleucine contents and repressing the accumulation of isopentyl-adenine-type cytokinins and abscisic acid, which are known to inhibit AR development. We found that JA and CKs possibly act in the same pathway to repress ARI. Anatomical analysis showed that JA represses the early events of ARI. In conclusion, cRL promotes ARI by repressing the accumulation of the wound-induced phytohormones JA and CK.


Summary 30
• Vegetative propagation relies on the capacity of plants to regenerate adventitious 31 roots (ARs) de novo. Light plays a central role in this process; however the 32 physiological and molecular components mediating light cues remain largely 33 elusive, especially in conifers. Here, we explore the effect of light spectral quality 34 on AR initiation (ARI) in de-rooted Norway spruce seedlings. 35 • We used light emitting diodes (LEDs) to study the effects of different light spectra 36 on ARI, hormone metabolism and gene expression. We used sensitive mass 37 spectrometry-based methods, and, since the Norway spruce genome sequence 38 recently became available, we coupled this to gene expression analysis. We 39 performed extensive anatomical characterization to investigate ARI at the cellular 40 level. 41 • We showed that in contrast to constant white light (WL) and constant blue light 42 (cBL), constant red light (cRL) promoted ARI by reducing jasmonate (JA) and JA-43 isoleucine contents and repressing the accumulation of isopentyl-adenine-type 44 cytokinins and abscisic acid, which are known to inhibit AR development. We found 45 that JA and CKs possibly act in the same pathway to repress ARI. Anatomical 46 analysis showed that JA represses the early events of ARI. 47 temperature of 4 °C prior to use. The seeds were soaked in tap water for 24 h, in the 122 dark at 4 °C. They were then sown in fine wet vermiculite as described in Ricci et al. 123 (2008) and allowed to germinate and grow in a growth chamber. The seedlings were 124 watered twice a week with tap water and left to grow for three weeks (Fig. S1a) Twenty-one days after sowing, the seedling hypocotyls were severed 2 cm below the 133 apex ( Fig. S1b) and placed in 24 ml vials filled with distilled water (Fig. S1c). Three 134 vials containing 5 seedlings each, constituting technical replicates, were prepared. 135 In order to test different white light conditions, vials containing de-rooted seedlings 136 were kept in either long-day conditions in the growth chamber as described above ( In order to remove the blue region of the spectrum from the white LEDs we added a 145 yellow filter, which almost totally abolished the blue peak at 460 nm (Fig. S2d). 146 The cBL (460 nm) intensity was set at 9 µmol/m 2 /s (Fig. S2e), and cRL (660 nm) was 147 set at an intensity of 9 µmol /m 2 /s (Fig. S2f). 148 The vials were refilled daily with distilled water to replace the water lost by 149 evapotranspiration. The number of roots per cutting and the number of rooted cuttings 150 were monitored for 30 days after cutting (DAC) Fig. S4. To find the putative Arabidopsis orthologs 199 in Picea abies, a BLAST alignment of the coding sequences of selected genes was 200 generated using "Genome tools" from the Congenie database with default settings. 201 Subsequently, the sequences were used to construct phylogenetic trees with the tools 202 available at "http://www.phylogeny.fr". The phylogenetic analysis was based on the 203 neighbor-joining method (with 1000 bootstrap replicates). The putative orthologs were 204 named according to the Congenie database. Candidate genes were identified by BLASTN searches against the spruce database 208 (http://congenie.org) using Arabidopsis protein coding nucleotide sequences. The 209 Norway spruce genes that showed the highest sequence similarity with Arabidopsis 210 genes were further considered for primer design and gene expression analysis. Primers 211 for the selected candidate genes and reference genes to be used for qRT-PCR were 212 designed using Primer3web version 4.1.0 (http://primer3.ut.ee); the primer sequences 213 are listed in Table S1. Specificity of the primers was analyzed by Sanger sequencing of 214 amplified PCR products and cDNA was used as a template for PCR amplification from 215 each primer pair. For each primer pair, PCR reaction efficiency estimates were 216 calculated from a standard curve generated from a serial dilution of cDNA. Based on 217 the average cycle threshold (Ct) values for each five-fold dilution, a standard curve was 218 generated using linear regression. PCR efficiency (E) was derived from the equation: 219 Efficiency % = (10 (-1/slope) -1) ×100%. PCR conditions were as follows: 95°C for 1 220 min, 40 cycles of 95°C for 15 s, 60°C for 30 s, and 72°C for 30 s. Finally, dissociation 221 curves were generated by increasing the temperature from 65 to 95°C, stepwise by 222 0.3°C every 10 s. For gene expression analysis by quantitative PCR (qPCR), total RNA 223 was isolated from the base of spruce hypocotyls (5 mm long from 36 to 38 hypocotyl 224 cuttings) using a Spectrum™ Plant Total RNA Kit (Sigma-Aldrich; 225 https://www.sigmaaldrich.com/). cDNA was synthesized using an iScript™ cDNA 226 Synthesis Kit (Bio-Rad; https://www.bio-rad.com/en-se/) following DNase treatment. To investigate the role of light in adventitious root initiation (ARI) in Norway spruce, 288 we used de-rooted Norway spruce seedlings as a model system. Three-week-old 289 seedlings grown under white light condition were harvested and de-rooted as described 290 in Materials and methods (Fig. S1). They were first kept in hormone free (HF) distilled 291 water, under three different white light (WL) regimes as described in Materials and 292 methods (irradiance 69 to 75 µM/m 2 /s -1 ; Fig. S2a,b,c). In these conditions no AR 293 developed at the base of the hypocotyls regardless of the photoperiod or the source of 294 white light. Since we obtained the same results in all WL conditions, we continued to 295 use constant WL (cWL) provided by LEDs ( Fig. S1d and Fig. S2c) as control 296 conditions for the rest of the study. 297 It is well known that auxin can stimulate adventitious rooting in several plant species 298 (Lakehal & Bellini, 2019), therefore we tested whether different types of auxin could 299 stimulate ARI under cWL. Three-week-old de-rooted seedlings were treated with 1 or 300 5 µM of Indole 3-Acetic Acid (IAA), 1-Naphthalene 3-Acetic Acid (1-NAA) or Indole 301 3-Butyric Acid (IBA), but none of the auxins could induce ARI under cWL conditions 302 (Fig. 1a). These results indicate that auxin is not sufficient to induce ARI in de-rooted 303 Norway spruce seedlings kept under cWL. 304 305

ARI in de-rooted Norway spruce hypocotyls 307
Next, we wondered whether a particular region of the light spectrum had a negative 308 effect on ARI. To test this hypothesis, three-week old de-rooted seedlings were kept in 309 HF distilled water under either cWL ( Fig. S1d and Fig. S2c), constant blue light (cBL) 310 ( Fig. S1f and Fig. S2d) or constant red light (cRL) (Fig. S1e and Fig. S2e). 311 Interestingly, AR developed only when seedlings were kept under cRL conditions ( Fig.  312   1 b, c, d). In these conditions 100% of the cuttings developed at least 1 AR by 30 days 313 after cutting (DAC) (Fig. 1b). These results raised the possibility that in cWL, the blue 314 light part of the spectrum has a negative effect on ARI. To test this hypothesis, 315 hypocotyl cuttings were kept under cWL but a yellow filter was added to remove the 316 blue wavelength region (Fig. S2f). In these conditions 83% of the cuttings developed 317 at least 1 AR (Fig. 1b). 318 These results indicate that cRL promotes ARI, whereas cBL inhibits this process, and 319 the absence of AR under cWL could be due to the presence of blue light. Under cWL conditions, we observed a 40% increase in the free IAA level 6 h after 337 cutting compared to the level at T0 (Supplemental data set 1), but thereafter the IAA 338 content remained mostly constant over time, with the exception of a slight increase 24 339 h after cutting ( Fig. 2a; Supplemental data set 1). There was no effect of wounding on 340 the endogenous content of SA, as shown in Supplemental data set1, and the SA content 341 remained constant up to 48 h after cutting (Fig. 2b). A slight increase could be observed 342 72 h after cutting (Fig. 2b, Supplemental data set 1). Based on these results, we 343 concluded that the inability to initiate ARs under cWL could not be explained by a 344 reduction in auxin content, and this is in agreement with the fact that exogenous 345 application of IAA or other auxins could not stimulate ARI under these conditions (Fig.  346

1a) 347
We have previously shown that JA and its bioactive form Jasmonoyol-Isoleucine (JA- decreased over time (Fig. 2c, d; Supplemental data set 1). Interestingly, although we 354 observed a significant decrease (72%) in total CK content 6 h after cutting compared to 355 the content at T0 (Supplemental data set 2), the total CK content then increased again 356 and CKs accumulated over time at the base of the hypocotyls (Fig. 2). This increase in 357 CKs was mostly due to the accumulation of isopentyl-adenine-type (iP-type) cytokinins 358 including the precursors iP riboside 5′-monophosphate (iPRMP) and iP ribosides (iPR) 359 (Fig. 2e,f,g; Supplemental data set 2). Similarly, we observed a 78 % decrease in the 360 endogenous content of ABA 6 h after cutting compared to the content at T0 361 (Supplemental data set 2), but it increased again over time (Fig. 2h). 362 Interestingly, in cRL conditions, although the free IAA content was significantly 363 reduced compared to that in cWL 24 h after cutting and continued to decrease over time 364 (Fig. 2a), cutting developed ARs. This could be explained by a significant reduction in 365 the amount of JA, Ja-Ile, CKs and ABA compared to cWL at all time points (Fig. 2c-366 h). In contrast to what was observed in cWL, the amounts of CKs and ABA remained 367 constant over time, except for iP which continued to decrease over time (Fig. 2g). 368 Although the interaction between light spectral quality and hormone homeostasis is 369 complex, these results suggest that the positive effect of cRL on ARI cannot be 370 explained by the modification of IAA homeostasis; rather it is a consequence of a 371 decrease in negative regulators such as JA, JA-Ile, and iP. 372 Surprisingly, although cBL had the same effect as cRL on the endogenous levels of 373 hormones ( Fig. 2a-h), ARs could not develop. These results suggest that cBL inhibits 374 adventitious rooting through another pathway yet to be identified. In the remainder of 375 this study we therefore investigated only the role of cRL on ARI. 376 377

Exogenously applied JA and CK inhibit ARI under cRL 378
To get physiological insights into the possible crosstalk between IAA, JA and CKs 379 during cRL-induced AR development, we treated the de-rooted seedlings exogenously 380 with either auxins, JA, or CK alone or in combination. 381 First, we showed that the three types of auxin (IAA, NAA and IBA) enhanced AR 382 formation under cRL conditions in a dose-dependent manner (Fig. 3a-c). IBA, which 383 is often used to induce rooting in recalcitrant species (reviewed in Geiss et al., 2018; 384 Stevens et al., 2018), appeared to be the most efficient auxin (Fig. 3a). When applied 385 exogenously, both JA and 6-Benzylaminopurine (6-BAP) inhibited AR development in 386 a dose-dependent manner (Fig. 3d,e) and they both repressed the positive effect of 387 exogenous IBA (Fig. 3f,g). These results suggest that JA and CK act downstream of 388 auxin signaling to repress ARI, as has been described for intact Arabidopsis hypocotyls 389 (Gutierrez et al., 2012; Lakehal et al., 2020a). When JA and CK were combined, no 390 additive or synergistic effect was observed (Fig. 3h), suggesting that they act in the 391 same pathway. These results are in agreement with our previous data that showed that 392 CK signaling is induced by JA to repress AR formation in intact Arabidopsis 393 hypocotyls (Lakehal et al., 2020a). . We found that the Norway spruce 408 genome contains eleven putative PaCOI1-like genes (Fig. S3a), five putative PaMYC2-409 like genes (Fig. S3b), one and three putative orthologs of PaJAZ3-like and PaJAZ10-410 like respectively (Fig. S3c) and two putative PaAOC-like genes (Fig. S3d). The full-411 length coding sequences are reported in Fig. S4. When more than one putative ortholog 412 was identified we chose that most closely related to the Arabidopsis one for further 413
The 419 relative amounts of PaMYC2-like, PaJAZ3-like, and PaAOC-like transcripts were 420 slightly reduced in cRL compared to those observed in cWL (Fig. 4b). These results 421 are in agreement with the reduced content of JA-Ile, the active form of JA, in cRL 422 compared to cWL (23% less in cRL compared to cWL) ( Fig. 2f and Supplemental data 423 set 2). Twenty-four hours after cutting, PaMYC2-like, PaJAZ10-like, and PaCOI1-like 424 were slightly upregulated in cRL compared to cWL (Fig. 4b), results which also 425 coincided with a slightly higher concentration of JA-Ile (20% more in cRL compare to 426 cWL) (Fig. 2f). Later on, by 48h and 72h, the JA responsive genes were downregulated 427 (Fig. 4), and this coincided with a greater decrease in JA and JA-Ile contents in cRL 428 compared to cWL ( Fig. 2f; Supplemental data set 2). In conclusion, when de-rooted

Exogenously applied JA inhibits ARI in Norway spruce hypocotyl in cRL 436
Our data indicated that JA signaling was rapidly downregulated (within 6 hours) in the 437 hypocotyls of cuttings kept under cRL compared to cWL, suggesting that JA signaling 438 repressed an early event in AR formation. To assess the repressive effect of JA at the 439 cellular level, we analyzed and compared the anatomy of hypocotyls kept under cRL, 440 in hormone free distilled water or in the presence of 1 µM IBA or 20 µM JA, at different 441 time points after cutting ( Fig. 5 and 6). On day 0 (i.e. immediately after cutting the 442 hypocotyl of three-week-old Norway spruce seedlings), the hypocotyl had a primary 443 structure consisting of a single layer of epidermis, and six to seven cortex cell layers 444 composed of isodiametric parenchyma cells (Fig. 5a). We could guess at the presence 445 of the endodermis layer although it was not always morphologically distinctive (Fig.  446   5a). The central cylinder consisted of seven to eight layers of parenchyma cells which 447 were smaller compared to the cortex parenchyma cells. The vascular system was 448 composed of a continuous procambium and differentiated elements of the protophloem 449 and protoxylem. The pith region showed small isodiametric parenchyma cells (Fig. 5a). 450 Three days after cutting, in HF distilled water and in the presence of 1 µM IBA, a few 451 cells in the cambial zone and in the adjacent phloem cells developed dense cytoplasm 452 and large nuclei indicating the initiation of the first cell divisions (Fig. 6 a,b). In 453 contrast, no obvious histological modifications were observed in the presence of 20 μM 454 JA, and the histological organization remained unchanged compared to day 0 (Fig. 6c). 455 By five days after cutting, periclinal and anticlinal divisions were clearly observed in 456 the cambial zone and in the outermost layers of the phloem region (Fig. 6d). In 457 longitudinal section these cells were seen to be organized into vertical files external, 458 but adjacent, to the vascular cylinder (Fig. 5b). In the presence of 1 μM IBA it was 459 clearly apparent that the number of cell divisions was higher than under HF conditions 460 (Fig. 6e), whereas no divisions were observed in the presence of 20 μM JA (Fig. 6f). 461 Ten days after cutting, in HF conditions, cambial derivatives divided and formed radial 462 rows of tracheal elements around the xylem, but no AR meristemoids were observed 463 ( Fig. 6g), while in the presence of 1 μM IBA several AR root meristemoids were visible 464 ( Fig. 5c and Fig. 6h); no cell divisions were observed in the presence of 20 μM JA (Fig.  465

6i) 466
It was only by thirteen days after cutting that AR meristemoids began to form at the 467 periphery of the trachea elements in HF conditions (Fig. 6j), while in the presence of 468 IBA, AR had already emerged (Fig. 5d and Fig. 6k); still no cell division activity was 469 observed in the presence of JA (Fig. 6l). In HF, the first emergence of ARs was 470 observed only 15 days after cutting (Fig. 5e). These data clearly indicate that 471 exogenously applied JA repressed the very early stage of ARI, whereas auxin not only 472 promoted but also speeded up the process. poorly understood. In this study, we showed that under cWL conditions de-rooted 485 Norway spruce hypocotyls were unable to produce AR and that this was unlikely to be 486 due solely to an increase in endogenous CK levels, as suggested by (Strömquist & 487 Eliasson, 1979;Bollmark & Eliasson, 1990) since exogenously applied auxin was 488 unable to stimulate AR development. This result led us to hypothesize that specific 489 wavelengths within the white spectrum could be responsible for this strong inhibition, 490 and we therefore grew seedlings under white or monochromatic LEDs. In these 491 conditions, we showed that AR developed under cRL even in the absence of exogenous 492 auxin application, while they did not develop under cBL. 493 The cRL-mediated ARI is possibly due to a faster reduction in JA and JA-Ile contents 494 compared to cWL. cRL also repressed the accumulation of CKs and ABA which was Our data also showed that total CKs accumulated at the base of cuttings when the de-540 rooted seedlings were exposed to cWL, whereas their amount remained constant over 541 time under cRL conditions. Under cWL, the increasing amounts of the iP-type CKs 542 could be due to de novo biosynthesis since we also observed an increased level of the 543 precursors iPR and iPRMP over time. (Bollmark & Eliasson, 1990) showed that de-544