Elucidation and functional characterization of CsPSY and CsUGT promoters in Crocus sativus L.

The dried stigmas of Crocus sativus constitute the saffron, which is considered to be the costliest spice of the world. Saffron is valuable for its constituents, which are mainly apocarotenoids. In order to enhance the production of apocarotenoids, it is imperative to understand the regulation of apocarotenoid biosynthetic pathway. In C. sativus, although the pathway has been elucidated, the information regarding the regulation of the pathwaygenes is scanty. During the present investigation, the characterization of promoters regulating the expression of two important genes i.e. CsPSY and CsUGT was performed. We successfully cloned the promoters of both the genes, which were functionally characterized in Crocus sativus and Nicotiana tabaccum. In silico analysis of the promoters demonstrated the presence of several important cis regulatory elements responding tolight, hormonesand interaction with transcription factors (TFs). Further analysis suggested the regulation of CsPSY promoter by Abscisic acid (ABA) and that of CsUGT by Gibberellic acid (GA). In addition, we also observed ABA and GA mediated modulation in the expression of significant TFs and CsPSY and CsUGT transcripts. Overall, the study addresses issues related to regulation of key genes of apocarotenoid pathway in C.sativus.


Introduction
Crocus sativus, commonly called saffron, belongs to the family Iridaceae. Saffron is one of the most fascinating cash crops due to its dried red stigma, which constitutes saffron of commerce [1]. Iran, Spain, India, Greece, Azerbaijan, Morocco and Italy dominate the world saffron production, with Iran and Spain producing 80% of the world crop. In India, saffron is mainly grown in the state of Jammu and Kashmir [2]. Saffron is a rich source of several therapeutically important carotenoids and apocarotenoids. Apocarotenoids play variety of roles in many biological processes, like flowering and fruit colour development [3], light harvesting and protection of photosynthetic machinery [4], act against biotic and abiotic stresses [5] and in biosynthesis of ABA and Strigolactone (SL) [6].
Apocarotenoids are non-polar isoprenoids that have alternate double and single bond pattern in their structure. Saffron contains mainly three types of chemical components, which a1111111111 a1111111111 a1111111111 a1111111111 a1111111111

Isolation of CsPSY and CsUGT promoters
Genome walking PCR was performed using gene specific primers designed from 5' end of CsPSY and CsUGT genes and adaptor primers AP1 and AP2, specific to ligated adaptor sequences (provided in GenomeWalker kit). All four genomic libraries were used as template for Genome Walking PCR. The primary PCR was set up using the GSP1 and AP1 and four different libraries as template. The primary PCR product was further diluted and used as template for secondary PCR, where AP2 and GSP2 were used as nested primer pair. The secondary PCR products were resolved on 1% agarose gel. All PCR programs were performed following manufacturers' protocols. The primer sequences are listed in S1 Table. Cloning and analysis of promoter sequence All amplified products obtained from secondary PCR were purified from agarose gel using a QIAquick Spin kit (Qiagen) and further cloned into the pTZ57R/T cloning vector system using TA cloning kit (Fermentas). Plasmids were isolated from different clones and were then processed for sequencing after purification. For removal of vector contamination, sequences were trimmed using VecScreen software (https://www.ncbi.nlm.nih.gov/tools/ vecscreen/) and were aligned by BLASTn and ClustalW program. The presence of putative cis-regulatory elements were detected by PLACE (http://www.dna.affrc.go.jp/PLACE/ signalscan.html) and Plant CARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/ html) scan tools [13]).

Agrobacterium mediated transient assay in heterologous and homologous system
Agrobacterium tumefaciens strain GV3101 transformed with pCAMBIA1391Z, pCAM-BIA1391Z-CsPSYp::GUS and pPZP200N-35S::GUS were inoculated in 5 ml of LB medium containing Kanamycin (50 μg/ml), Rifampicin (50 μg/ml) and Gentamycin (40 μg/ml). The cultures were incubated at 28˚C for two days at 120 rpm. Similarly, pMDC164, pMDC164-CsUGTp::GUS and pPZP200N-35S:: GUS transformed A. tumefaciens strain GV3101 were inoculated in the above-mentioned medium. 10 μl of primary culture was inoculated in 50ml of LB medium containing 10 mM MES, 30 μM acetosyringone and above-mentioned antibiotics. The secondary culture (OD 0.6) was syringe-infiltrated on abaxial side of leaves of Nicotiana tabaccum. The infiltrated leaves were incubated for two days and further processed for estimation of GUS activity. Similarly, Crocus calli were co-cultivated with A. tumefaciens GV3101 harbouring the above-mentioned constructs and were kept on MS media. After two days of incubation, calli were transferred to selective MS media containing Cefotaxime (250 μg/ml), Kanamycin (100 μg/ml), TDZ (1mg/l) and IBA (0.5 mg/l). Plates were incubated at 15˚C for one monthand further processed for GUS activity.

Histochemical GUS staining
Post-infiltrated tobacco leaves and co-cultivated Crocus calli were processed for GUS activity, following manufacturer's protocol (GUSS-1KT, 1002229482-Sigma). Briefly, detached leaves were kept in fixative buffer solution and incubated for 45 min at 25˚C followed by washing. Samples were vacuum infiltrated for 2-5 minutes in staining solution and incubated overnight at 37˚C. Subsequently, samples were washed with 75% ethanol. Similarly Crocus calli were processed for GUS activity detection.

Abscisic acid and Gibberellic acid treatment
Two months old Crocus plants and calli were treated with ABA (100 μM) and GA (100 μM). For ABA treatment 100 μM aqueous solution was prepared and poured in the pots in which saffron corms were growing. For GA treatment 100 μM solution was prepared in water to which 10% Tween 20 was added. GA solution was sprayed over the plants. These plants were covered with perforated autoclave bags. For mock treatments, plants were treated with autoclaved water (in case of ABA) and water containing 10% Tween 20 (in case of GA). Flowers of treated plants were harvested after 12 and 24 hours along with mock treated plants. The seven days post-treated calli were also harvested along with mock and processed further.

RNA isolation and real time PCR
Total RNA was extracted from flower and calli of Crocus using Trizol reagent (Invitrogen). The cDNA was synthesized from each RNA sample following manufacturer's protocol (GoScript Reverse Transcription System-A5000). The estimation of qRT-PCR based quantitative fold induction was performed using the SYBR Green (Power Syber Green 4367659) and ABI StepOne Real time (Applied Biosystems). Experiments were conducted in three biological and three technical replicates. The reaction was carried out in a total volume of 10 μl that comprised of 5 μl of 2X SYBR Green Master Mix, gene specific primer pairs (0.2 μM each) and 20 ng of cDNA template. The cycling parameters were 95˚C for 20 s, followed by 40 cycles of 95˚C for 15 s and 60˚C for 1 min. The ΔΔ -CT method was used for relative quantification and ACTIN2 (ACT2) was used as an internal control. Primers for RT-PCR were designed from unique region of the selected transcripts as given in S1 Table.

Isolation of CsPSY and CsUGT promoters
PCR based Genome walking was performed, using all the four genomic libraries which resulted in products of different lengths for CsPSYp and CsUGTp. The largest amplicons of CsPSYp (900bp) and CsUGTp (850 bp) were obtained from EcoRV and PvuII genomic library, respectively (S1 and S2 Figs). The isolated sequences showed alignment at 5' region of CsPSY (AY262037.1) and CsUGT (AJ888514.1) sequences, respectively.

In silico analysis of CsPSY and CsUGT promoters
In silico analysis showed presence of various important cis-regulatory elements involved in different signaling cascades. The analysis of the promoter sequences demonstrated presence of light responsive, TF binding and hormone signaling elements. Besides these, other elements involved in signaling network for phosphate starvation, carbohydrate metabolism and defense response were also revealed by this analysis (Fig 1a and 1b). In case of CsPSYp, prediction analysis exhibited the presence of a potential TATA box 213 bp upstream to ATG and 27 bp upstream to TSS (+1). In CsUGTp, a potential TATA box, observed at 241 bp upstream to ATG and 27 bp upstream to TSS (+1) was predicted. A detailed description about cis regulatory elements present in CsPSYp and CsUGTp are given in Tables 1 and 2. CsPSY and CsUGT promoter activity assay GUS activity was detected in pCAMBIA1391Z-CsPSYp::GUS and pMDC164-CsUGT::GUS (Promoter-Reporter construct) infiltrated leaves while in mock treated plants no GUS activity was observed. The transformed Crocus calli harbouring pCAMBIA1391Z-CsPSYp::GUS and

Effect of ABA and GA on CsPSY and CsUGT promoter activity
In silico analysis as reported above demonstrated the presence of ABA and GA responsive elements (ABRE and P-Box) in both CsPSYp and CsUGTp. Several reports suggest that GA and ABA treatment modulates the expression of a number of transcripts either involved in biosynthesis or other signaling cascades. Recently, Mandal et al. [20] have shown positive regulation of ARM protein by GA treatment. In the present case, the ABA and GA treatment enhanced CsPSYp and CsUGTp driven GUS activity as compared to mock treated calli (Fig 3). However, GUS activity was comparatively high in case of CsPSYp transformed calli compared to CsUGTp transformed calli after ABA and GA treatment. It was also observed that in CsUGTp transformed calli GUS activity was higher in case of ABA as compared to GA treatment.   and CsUGT transcripts was observed after ABA and GA treatment as compared to mock. Interestingly, the peak expression was observed in flowers 12 hrs post-GA treatment. However, the CsUGT was highly up-regulated in flowers 12 hrs post-ABA treatment.

Effect of ABA and GA on expression of transcription factors
Six transcription factors i.e. CsbHLH, CsARF, CsMYB, CsWRKY, CsSNF and CsAUX were selected for expression analysis after 12 and 24 hrs in ABA and GA treated flower samples. Significant modulation in expression of these TFs was observed. The CsbHLH exhibited

Discussion
The understanding of regulation of various genes and enzymes involved in carotenoid/apocarotenoid pathway may help in its engineering to enhance the production of various bioactive compounds. One of the possible means in C.sativus involves the overexpression and functional knock-out of apocarotenoid pathway genes [21]. Other interesting way is to explore the regulatory regions of native promoters and utilize the generated information to attain desired metabolite production. In C. sativus, characterization of promoters of CsCCD2 and CsLycB2a has been reported earlier [10]. The PSY and UGT are the important enzymes responsible for apocarotenoid production. In some plants such as tomato, citrus and sweet pepper, light and development dependent expression of PSY and UGT genes and its correlation with apocarotenoid content has been demonstrated [12,1]. In citrus fruit, PSY transcript level increases with rise in fruit carotenoid content [22]. Two UGT isoforms (UGTCs2 and UGTCs3) in C. sativus have been identified [9]. Among them, higher expression of UGTCs2 in stigma was observed, which was further correlated to crocetin production [9]. Therefore, for the present investigation we explored the promoter regions of PSY and UGT genes genes so as to identify characteristic elements. Detailed information about various important genes of carotenoid and apocarotenoid biosynthetic pathway and related transcription factors in C. sativus was extensively reviewed recently [1].

Cis-regulatory elements in CsPSY and CsUGT promoters
In silico analysis of CsPSY and CsUGT promoters showed the presence of diverse cis-regulatory elements involved in different signaling cascades. The active form of promoter was validated by the presence of TATA box and transcription initiation site. W-Box element which is a potential binding site of WRKY group of TFs, was present in promoters of CsPSY and CsUGT genes. According to Mandal et al. [23] WRKY group of TFs regulates the genes involved in plant development, by binding to W-Box elements present in their promotersunder stress condition. The light responsive elements present in CsPSY and CsUGT promoters have been predicted to be involved in different light regulated signaling response in many plant species [15].
In CsPSY promoter presence of light responsive elements viz. GATA box, CAAT box, GT1 and Box 4 was observed. GATA box is a GATA-binding site and is involved in light-regulated developmental cues [24]. GT1 is important cis-regulatory element involved in control of light regulated transcription [12]. In CsUGT promoter, light responsive cis-elements like SORLI-P2AT (Phy A response element), CATT motif, G-boxand GAG motif were present. SORLI-P2AT element has been reported to be involved in phytochromeA regulated gene expression [25]. The G-box element is responsible for PIF (bHLH) binding, a central mediator in a variety of light-mediated responses and also in repression of photomorphogenic development in dark [15]. Saffron is a geophyte with a very short aboveground growth period. It breaks its dormancy in winter resulting in shoot and flower bud formation. Stigma development starts in unopened flower bud. Initially the stigma is yellowish, followed by orange and acquires scarlet colour during the last stage i.e. fully bloomed flower. At scarlet stage the amount of apocarotenoids is the highest. Presence of light responsive cis-elements in the CsPSYp and CsUGTp provide a clue about the regulation of the genes involved in carotenoid/apocarotenoid pathway. Thus, it can be predicted that for the formation of scarlet stigma, besides, TFs and phytohormones, light is a necessary factor, which triggers the expression of pathwaygenes, resulting in the maximum concentration of apocarotenoids (Fig 6). Besides, several elements required for phytohormone mediated signaling were also present in CsPSYp (ABRE, ERE and GARE) and CsUGTp (ARR1, ABRE and P-Box). The ABRE element has earlier been identified in the PSY promoter of Oryza sativa and Zea mays [26,27,28]. Besides, ABRE has been reported to play a crucial role in the regulation of NAC TFs under stressed condition [29]. The phytohormones play significant role in development and defense and hence presence of these elements in CsPSYp and CsUGTp point towards the possibility of their phytohormone mediated regulation. In C. sativus, flower development and maturation is correlated with its metabolite production, therefore it is likely that ABA, GA, ethylene and auxin treatment may also modulate metabolite levels by influencing the expression of CsPSY and CsUGT transcripts. In current study, the induced expression of CsPSY and CsUGT was also observed after ABA and GA treatment in C. sativus flowers. The probable reason for this induced expression is the presence of ABA and GA phytohormone binding factors in the promoter sequence of CsPSY and CsUGT genes. Presence of these binding factors may have resulted in increased transcription of these genes.
Transcription factors are other imperative candidates involved in biosynthetic pathway regulation, which affect the expression of pathway genes under different conditions. CsPSYp and CsUGTp sequences showed binding sites for MYB, MYC, MADS, CBF, WRKY, BELL homeodomain, ARF, RAP2.2, transcription factors. Besides, CsPSYp and CsUGTp contain sites involved in carbohydrate metabolic signaling (Pyrimidine box, CMSRE1IBSPOA), fungal elicitor responsive element (Box-W1) andphosphate starvation response etc. As CsUGT is involved in glucosylation, the presence of carbohydrate metabolic signaling element in CsUGTp seems reasonable. Earlier workers have suggested that sugar accumulation in Crocus corms is positively correlated with flower development [30]. Hence, it can be argued that presence of sugar responsive element might be a part of carbohydrate metabolic signaling in relation to flower development. In cotton, promoter of GhGlcAT1 gene was responsive to glucose and sucrose as their treatment increased the expression of GUS fusion protein as compared to other sugars [25]. Therefore, by analyzing these elements one can identify the upstream candidates involved in regulation of gene expression like hormones, TFs, etc. under such conditions.

ABA and GA regulates CsPSY and CsUGT promoter activity
The presence of ABA and GA responsive elements in CsPSY and CsUGT suggested the ABA and GA mediated regulation of gene expression. In accordance, the enhanced GUS expression in Crocus calli and tobacco leaves was observed after hormone treatments. These results thus justify the functional relevance of ABA and GA responsive elements in CsPSYp and CsUGTp. Again, the enhanced expression of CsPSY and CsUGT transcripts in ABA and GA treated flowers and leaves further validated the impact of these hormones on the activity of CsPSYp and CsUGTp. Expression of PSY is highly regulated by stress, development and light signaling [31,32]. Apart from this, as PSY catalyzes the rate, limiting step of this pathway so its modulation can affect carotenoid metabolic flux. Dong et al. [33] reported increase in the transcript level of two closely related homologs of UGT (UGT71B7 and UGT71B8) in ABA treated Arabidopsis. Thus, understanding of the regulatory aspects of PSY promoter may help to manipulate the carotenoid/apocarotenoid biosynthesis. In present study we reported that initially at 12 hours of post GA and ABA treatments CsPSY expression was induced. Further CsPSY transcript level decreased after 24 hrs of post GA and ABA treatment. Probably, by regulatory feedback inhibition is responsible for down regulation of CsPSY transcript as earlier mentioned by Zeng et al. [12].

ABA and GA modulates the expression of TFs
The possible binding of various transcription factors to the promoter of the pathway genes has been reported to modulate their expression. In this study, the analysis of CsPSYp and CsUGTp revealed presence of binding elements for many regulatory agents including transcription factors. Moreover, there are other reports demonstrating that TFs may itself be regulated by phytohormones. To work on this possibility, six potential transcription factors preferentially expressed in stigma [34] and having potential binding sites in CsPSY and CsUGT promoters were selected for expression analysis. The significant modulation in expression of these TFs in C. sativus suggests co-regulation of transcription factors and apocarotenoid biosynthestic pathway genes. As shown in Fig 5 the relative expression pattern of CsARF, CsbHLH, CsMYB, CsSNF and CsWRKY transcription factors was almost similar as in case of CsPSY and CsUGT transcripts after GA and ABA treatment. ARF and Aux TFs are the components of Auxin signaling pathway where Aux TF represses activity of ARF TF. It has been reported in literature that Auxin treatment decreases the level of Aux TF. According to Bjorklund S and Wang T, GA and ABA treatment increases Auxin biosynthesis and polar auxin transport [35,36]. So this justifies that GA and ABA treatment decreases the level of Aux TF by enhancing the Auxin biosynthesis. Again a correlation between expression of CsbHLH and CsPSY transcript can be observed as the expression of CsPSY decreased at 24 hours post GA and ABA treatments (Fig 4) while CsbHLH TF showed increased expression in GA and ABA treated flowers (Fig 4) at similar time point (24 hours). Toledo-Ortiz et al. [14] reported suppression of PSY transcript by PIF a CsbHLH transcription factor in Arabidopsis thaliana which justifies above mentioned correlation between CsbHLH and CsPSY transcripts. Earlier, Gómez-Gómez et al. [37] identified a MYB TF in C. sativus whichhas been demonstrated to play role in the regulation of stigma morphology. The wound inducible PsWRKY of Papaver somniferum has been reported to enhance benzylisoquinoline alkaloid biosynthesis metabolism, which also exhibited modulation after MJ and ABA treatment [38]. This TF is known to play significant role in hormone signaling pathways and metabolism [39]. Further, most of the TFs utilized in this study are known to regulate secondary metabolism [40] and are specific to stigma [33]. Hence, possibly these TFs may be regulating apocarotenoid biosynthesis in Crocus stigma.

Conclusion
The present study elucidates the functional validation of CsPSY and CsUGT promoter sequences in heterologous as well as homologous systems, thereby enhancing our understanding of the regulation of apocarotenoid biosynthetic pathway. Based on the results of various experiments, it appears that ABA and GA may be regulating the expression of these promoters. Besides, increased expression of stigma specific TFs was also observed in response to ABA and GA. Co-regulation of TFs, CsPSY and CsUGT promoters by ABA and GA is a pointer towards their important role in regulation of apocarotenoid pathway genes in C.sativus. So, for further studies to understand regulatory signaling mechanism of Apocarotenoid biosynthetic pathway bHLH, WRKY and MYB TFs can be potential candidates. This study can prove to be highly useful to identify upstream trans-acting elements of these promoters as well as for genetic engineering to enhance metabolite production in saffron.