Cytokinin perception in potato: New features of canonic players

Potato is the most economically important non-cereal food crop. Tuber formation in potato is regulated by phytohormones, cytokinins (CKs) in particular. The present work was aimed to study CK signal perception in potato. The sequenced potato genome of doubled monoploid Phureja was used for bioinformatic analysis and as a tool for identification of putative CK receptors from autotetraploid potato cv. Désirée. All basic elements of multistep phosphorelay (MSP) required for CK signal transduction were identified in Phureja genome, including three genes orthologous to three CK receptor genes (AHK 2-4) of Arabidopsis. As distinct from Phureja, autotetraploid potato contains at least two allelic isoforms of each receptor type. Putative receptor genes from Désirée plants were cloned, sequenced and expressed, and main characteristics of encoded proteins, firstly their consensus motifs, structure models, ligand-binding properties, and the ability to transmit CK signal, were determined. In all studied aspects the predicted sensor histidine kinases met the requirements for genuine CK receptors. Expression of potato CK receptors was found to be organ-specific and sensitive to growth conditions, particularly to sucrose content. Our results provide a solid basis for further in-depth study of CK signaling system and biotechnological improvement of potato.


Introduction
Fidelity DNA polymerase (Thermo Scientific). The primer design was performed to amplify the full-  Supplementary Table S2. PCR products were gel 145 purified and cloned using the PCR Cloning Kit (Thermo Scientific) into the plasmid pJET1.2/blunt 146 according to the manufacturer's instructions followed by transformation of E. coli strain DH10B 147 (Invitrogen). StHK4 was amplified using StHK4_truncated primers. The product was inserted into the 148 construction of pB7FWG2-AHK3 instead of AHK3. The latter was removed at the BcuI and EcoRI 149 restriction sites (Lomin et al., 2015). The nucleotide sequences of the cloned genes were confirmed by 150 DNA sequencing.

151
StHK2 and StHK3 sequences were subcloned into the plasmid pDONR TM 221 in BP reaction with 152 Gateway® BP Clonase® II Enzyme mix (Thermo Scientific). Then, using the LR reaction with the LR 153 Clonase® II Plus enzyme (Thermo Scientific), the cloned sequence was transferred into the expression 154 vector pB7FWG2 (Karimi et al., 2007) where it was fused at the 3'-terminus to the eGFP gene. For  Table S2). The product was then inserted into the plasmid 157 pCOLD IV (Takara BioInc.) at the XhoI and XbaI restriction sites for StHK2 and SacI and EcoRI 158 restriction sites for StHK4, followed by transformation of the E. coli DH10B strain. 160 The transient transformation of tobacco (Nicotiana benthamiana Domin) leaves was accomplished 161 according to Sparkles et al. (2006). Eight week-old tobacco plants were infiltrated with a mixture of 162 Agrobacterium tumefaciens carrying CK receptor genes fused to GFP and the A. tumefaciens helper 163 strain p19 (Voinnet et al., 2003), and the expression of receptor genes was checked after 5-6 days on 164 fluorescence microscope Axio Imager Z2 (Carl Zeiss Microscopy GmbH) before leaves were proceeded 165 further for microsome isolation. 167 All manipulations were done at 4 °C. Tobacco leaves 6 days after infiltration were homogenized in 168 buffer (3 ml per 1 g of fresh weight) containing 100 mM Tris-HCl (pH 8.0), 2 mM Na 2 - EDTA, 50 mM al., 2015). Studies of pH influence on hormone binding were performed in 50 mM MES-KOH (pH 5-7) 176 or Tris-HCl (pH 7-9) buffers with 50 mM KCl. K d for [ 3 H]tZ binding to different receptors were 177 determined in saturation assays followed by data analysis in Scatchard plots. 179 Plasmids pCOLD IV with StHK coding sequences were transferred for the expression into E. coli strain 180 KMI001 (Suzuki et al., 2001). In this strain, HK receptor→YojN→RcsB→cps::lacZ pathway can be 181 activated by external CKs (Takeda et al., 2001). The activation of the signaling pathway was monitored 182 by measuring β-galactosidase activity of E. coli cells. Cultivation of clones on Petri dishes containing 40 183 mM glucose, 40 μg ml -1 X-gal, 100 μM IPTG, 50 μg ml -1 ampicillin at 15 °C was performed for 4 days. 184 The individual clones were then streaked onto new Petri dishes containing 40 mM glucose, 40 μg ml -1 185 X-gal, 100 μM IPTG, 50 μg ml -1 ampicillin ± trans-zeatin at a concentration of 0.5 μM. cDNA and genomic DNA. The band derived from genomic DNA was absent in the separating gel.

199
Expression of genes encoding predicted proteins of CK signaling system was determined by qRT PCR.

228
CK receptors of flowering plants can be grouped into three main clades, corresponding to the 229 Arabidopsis AHK2, AHK3, and CRE1/AHK4 receptors (Pils and Heyl, 2009;Lomin et al., 2012;230 Steklov et al., 2013). Predicted potato and tomato receptors are unequivocally distributed among these 231 three clades. Evolutionally, they are closer to Arabidopsis than to rice receptors, what was expected the cognate genes as well as occurrence and position of functional domains in the receptor proteins were 236 analyzed. Known CK receptors share a common organization, including (from N to C termini) sensory 237 module with CHASE domain, catalytic module with HisKA and ATPase domains, and receiver module 238 with pseudoreceiver and receiver domains (Kakimoto, 2003;Steklov et al., 2013). The sensory module 239 is flanked by predicted transmembrane (TM) α-helices. There is always a single TM-helix C-terminal 240 (downstream) of module while the number of TM-helices N-terminal (upstream) of module is variable.  receptor. This insert is located, however, on the opposite side from the ATP binding site. This structural 282 feature distinguishes not only potato receptors but also CK receptors of other species.

283
The CKI1 histidine kinase receiver domain (RD), used as the template for CK receptor RD, adopts a 284 fold typical for the REC (or CheY-like) superfamily proteins. It is formed by five α-helices and β-sheet 285 composed of five parallel β-strands. Two α-helices are located on one side of the β-sheet, and remaining 286 three on the other side. The same fold is characteristic for the model of the Arabidopsis CRE1/AHK4 287 receptor RD. As distinct from this, an additional small helix is present in the region between α3 helix 288 and β4 strand in the models of potato and other Arabidopsis receptors AHK2 and AHK3 RDs.

289
Conserved aspartate residue, serving as a phosphate acceptor in RD, is located at the N terminus of the 290 β3 sheet (Fig. 4).

291
Deviations from canonic CHASE motifs in sensory modules of putative potato CK receptors do not  Sequencing Consortium, 2011). Such differences in phenotype are underlain by considerable sequence 303 and structural genome variations between potato haplotypes. Therefore, the results of genome study of 304 monoploid Phureja do not mirror exactly more complex genomes of common potato cultivars.

305
Our experimental study of CK receptors was performed using the autotetraploid potato cv. Désirée, 306 widely used for commercial and scientific purposes (Aksenova et al., 2000;Kolachevskaya et al., 2015). 307 We cloned the putative receptor genes using primers designed according to Phureja gene sequence data.

308
Distinct from Phureja genome, at least six genes of putative CK receptors were cloned from cDNA of StHK2b have three and two aa substitutions, respectively, relative to Phureja receptor (Table 2).

317
Of two cloned genes of HK3-clade, StHK3a is identical to its counterpart of Phureja, whereas 318 StHK3b differs by 20 SNPs together with a 3-nucleotide deletion. These differences result in the 319 absence of one aa and nine aa substitutions in StHK3b compared to its Phureja ortholog. Similar data 320 were obtained for HK4-clade: StHK4a was fully identical to that of Phureja whereas StHK4b differs by 321 28 SNPs and a 3-nucleotide deletion. Correspondingly, StHK4b differs from its Phureja ortholog, as 322 well as from StHK4a, by deletion of one aa and substitution of 13 ones ( Table 2)  First, we determined the pH-dependence of hormone binding to these receptors within the pH range 339 of 5-9 (Fig. 5). All StHKs exhibited maximal trans-zeatin binding at the neutral-mildly basic pH: 340 StHK2a at pH 7.5, StHK3a SM at pH 7, StHK4a at pH 7.5-8, and StHK4b at pH 8-9. All StHKs showed 341 a decrease in ligand binding at acid pH: StHK2a and StHK3a SM reduced their binding at pH 5 compared 342 to pH 7 by a factor of 2 and 5, respectively. Ligand binding by StHK4a and StHK4b decreased at pH 5 343 about three times compared to maximal values. Although the StHK3a was represented in this study only 344 by its sensory module, a control experiment with the full-length StHK2a and its sensory module showed 345 a similar pH-dependence of hormone binding (data not shown). This means that an isolated sensory 346 module is sufficient to determine the pH-dependence of hormone binding by the receptor.

347
The interaction of a hormone with a receptor is characterized by the equilibrium dissociation constant   reporter gene driven by cps promoter (Suzuki et al., 2001;Takeda et al., 2001). This design allows 381 assessment of the ability of hybrid histidine kinases to initiate signaling over the MSP pathway.

382
Activation of MSP signaling in the bacteria leads to the expression of the reporter galactosidase (LacZ), 383 whose activity is manifested by blueing of clones growing on X-Gal-supplemented medium. We 384 expressed the cloned genes of the putative potato CK receptors in E. coli ΔrcsC. In the clones 385 expressing the StHKs but not in the control clone, blue staining was observed (Fig. 6). The degree of 386 blueing was greatly increased in the presence of CK. It confirms the ability of the cloned potato proteins 387 to transmit the CK signal to the primary response genes via the canonic MSP pathway. 389 To assess the functionality of a gene in vivo, it is important to know the level and pattern of its  Table S1). These primers were 395 complementary to both alleles of the same clade owing to a great similarity of these gene sequences.

396
The relative amounts of putative receptors of distinct clades in potato organs were judged by comparing 397 the levels of transcripts of the cognate genes.

398
Expression levels differed significantly depending on StHK group, organ and growth conditions (Fig. content. In the case of 1.5% sucrose medium, the highest expression of StHK3 genes was observed in 401 roots, while in the case of 5% sucrose medium, the maximal expression of StHK3 tended to occur in 402 leaves. In the low sucrose grown plants, StHK4 gene was much weaker expressed in leaves than in 403 stems or roots, whereas at the higher sucrose content levels of StHK4 expression in different organs 404 were more equalized. In the StHK2 group, noticeable organ-specific differences were detected when 405 plants were grown on 5%, but not on 1.5% sucrose. The lowest expression level of all StHK groups was 406 usually observed in tubers compared to other organs (Fig. 7A).

407
Within each organ, expression of StHK3 undoubtedly dominated in leaves, regardless of the sucrose 408 content (Fig. 7B). Expression of StHK2 and especially StHK4 genes in leaves was much weaker. In  Although the primers used for qRT PCR did not distinguish closely related isoforms of the CK 421 receptor genes, it is still possible to approximately estimate the relative expression of these alleles. To 422 achieve this goal, data on cDNA clone numbers can be used (Table 2). Within the same clade, relative 423 quantity of cDNA clones harboring a distinct isoform should reflect the relative occurrence of cognate 424 mRNAs. According to the last column of Table 2 corresponding to aerial part of potato seedlings, two 425 mRNA isoforms of the HK2 clade were in the 1:1 ratio; among mRNA isoforms of HK3 clade, StHK3a 426 was approx. two-fold more frequent than StHK3b; in the case of HK4 clade, StHK4b was expressed 427 about one order of magnitude more intensively than StHK4a. 429 Treatment of potato plants with N 6 -benzyladenine had a small effect on the expression of the CK 430 receptor genes, and the hormonal impact, when occurred, was only local and not always reproduced. At 431 1.5% sucrose, the upregulation (on average, 2.5-fold) of StHK4 expression was regularly recorded, but 432 only in leaves (Fig. 8). It can be stated that the CK effect on the expression of potato receptor genes, if any, is mostly limited to StHK4 and depends on both organ/tissue type and conditions of plant 434 cultivation.

435
To validate the results of CK treatment experiments, the effect of CK administration on the transcript 436 level of the genes of type A response regulator (RR-A) genes was analyzed. These genes in other species 437 (Arabidopsis, maize) represent genes of primary response to CK, so it might be expected that in potato 438 too they would be responsive to CK. Indeed, our experiments showed a rapid and reliable increase in the 439 expression of StRR-A genes, in contrast to the receptor genes, after plant treatment with BA (Fig. 8).

440
These results prove the reliability of design and implementation of experiments and, on the other hand, 441 corroborate the common mode of functioning of the CK signaling system in different plant species.

442
Analysis of promoter structures of the studied genes ( Fig. 9) was mostly consistent with the gene 443 expression data. Long CK-sensitive cis-regulatory elements or blocks of 4 or more short elements near 444 the transcription start (~300 bp area) were found in promoters of almost all StRR-A, but not StHK genes.   et al., 2000;Romanov et al., 2000). In non-potato plants, CKs alone were 456 able to induce the emergence of tuber-like structures (Guivarc'h et al., 2002;Eviatar-Ribak et al., 2013;457 Frugier et al., 2008;Miri et al., 2016). Apart the impact on the formation of tubers, CKs are known to 458 regulate overall plant architecture, biomass partitioning as well as resistance to biotic and abiotic stress-459 factors (Aksenova et al., 2000;Abelenda and Prat, 2013;Zwack and Rashotte, 2015;Brütting et al., 460 2017;Thu et al., 2017). All these point to the importance to investigate CK signaling system in plants, 461 in particular in tuber crops like potato.

462
Herein, we present first results of detailed study of CK receptors from potato plants. Two different 463 potato forms were examined: doubled monoploid Phureja and tetraploid potato of Désirée variety.

464
Phureja plants possess, like Arabidopsis, three CK receptor orthologs. By contrast, in Désirée plants two 465 allelic forms of each receptor type (StHK2a/b, StHK3a/b and StHK4a/b) have been found belonging to the three known phylogenetic clades. Our data indicated that this receptor abundance is characteristic of further in-depth study of the role of the CK signaling system in potato ontogenesis and provide new 501 biotechnological tools to optimize hormonal regulation of tuber formation.  Table S1. Sequence identity of modeled receptor domains and corresponding templates. 504 Table S2. Primers used in this work.

506
This work was supported by the Russian Science Foundation, grants no. 14-14-01095 (before 31.12           with 1.5% sucrose for 5-6 weeks under standard LD conditions. L, S, R signify leaves, stems and roots, respectively. More than two-fold prevalence of transcripts in BA-treated over control plants is considered as significant induction, bars corresponding to induced genes are outlined red. Fig. 9. CK-responsive cis-regulatory elements in promoters of CK receptor genes (upper part) and response regulators type A genes (lower part) of potato. Elements are shown on both DNA strands.
Promoter area proximal to transcription start is boxed.    and roots, respectively. More than two-fold prevalence of transcripts in BA-treated over control plants is considered as significant induction, bars corresponding to induced genes are outlined red. Promoter area proximal to transcription start is boxed.