An R2R3-MYB Transcription Factor PgFLP Directly Activates PgPIN10 to Regulate the GSA of Adventitious Root in Pomegranate

Background: The self-rooted seedling is widely used in pomegranate planting industry currently; However, the root system of self-rooted seedling is shallow and poor cold resistance. Therefore, the study of the molecular mechanisms of pomegranate adventitious root gravitropism is very important for developing deep-rooted pomegranate cultivars. Results: We report the pomegranate FOUR LIPS (PgFLP) that play an key role in regulating the gravitropic set-point angle of pomegranate adventitious root in response to gravity signal. In our study, PgFLP directly regulates the transcriptional expression of PgPIN10 by binding to its promoter, thus regulating the GSA of adventitious root in pomegranate. Additionally, the 35S::PgFLP show stronger gravitational response than wild-type, leading to a smaller GSA in Arabidopsis lateral roots, indicating that PgFLP participates in regulating the GSA of adventitious root via PgPIN10 in pomegranate. Conclusion: Our results conrm that the transcriptional regulation of PgPIN10 by R2R3-MYB transcription factor PgFLP in setting the gravitropic set-point angle of pomegranate adventitious root in response to gravity signal.


Background
Pomegranate (Punica granatum L.) is believed to originated in Afghanistan, present-day Iran, India, and Turkmenistan and other central Asia countries [36,23]. In China, many recognized pomegranate varieties with high genetic diversity have been cultivated [33]. The self-rooted seedling is widely used in pomegranate planting industry currently. However, the development of gravitropic set-point angle (GSA) between self-rooted seedling and pomegranate seedling is a great difference, leading to the adventitious root of pomegranate was poor in shallowness and solidity, resulted in poor cold resistance [28].
The plant hormone auxin is explicit for a common integrator to many environmental and endogenous signals regulating lateral root formation [12]. In the lateral root cap, the auxin stimulating prebranch site formation [32]. Subsequently, the dynamic auxin ows orchestrated lateral root patterning and morphogenesis [3]. In addition, it interacts with the surrounding tissue in a complex manner [10,16,19,26]. The gravitropic set-point angle can be de ned by orientation of plant growth with respect to the gravity vector [5]. In addition, auxin controls gravitropic set-point angle of lateral branches through auxin signaling pathway dependent on TIR1/AFB-Aux/IAA-ARF in the higher plant [22].
MYB TFs have been found to regulate plant development and metabolism [7]. In Arabidopsis, the two R2R3-MYB proteins, MYB88 and FOUR LIPS (FLP), can bind directly to the promoters of PINs genes that harbour an [A/T/G][A/T/G]C[C/G][C/G] motif [31]. FLP and MYB88 have been found to play redundantly role in limiting terminal divisions in Arabidopsis. Compared with p single mutants, the stomatal defects of p-1myb88 double mutants are more severe [11]. In addition, FLP and MYB88 have been found to can regulate root gravitropism [27], the late stages of stomatal development [11], cold hardiness [30], female reproductive development [17], and guard mother cell proliferation [11,31]. The PINs have been found to conducive to nearly every step of auxin uxes [3,18,35,20]. Following the primary roots are stimulated by gravity, the PINs have been repolarized [21,29], the auxin ux redirection to the lower side of roots, leading to root tip bending [8,9,22,25]. In addition, the polar auxin transport is bene cial to the root bending under obstacle avoidance [13]. The auxin distribution patterns have been predicted and explained through auxin transport [2,6,1]. The AtPIN3 shows remarkably dynamic expression during the root gravitropism and patterning of stomatal complexes in Arabidopsis [14,27]. However, how regulators regulate PIN expression combining with auxin signalling pathway during root development is nothing known in pomegranate. We investigated the mechanisms regulating PgPIN10 expression during pomegranate adventitious root gravitropism.
The regulation of root gravitropism in pomegranate is very little known currently. In our study, PgFLP participates in regulating the gravitropic set-point angle of pomegranate adventitious root. PgFLP directly regulates the transcriptional expression of PgPIN10 by binding to its promoter, thus regulating the GSA of adventitious root in pomegranate. Our results con rm that the transcriptional regulation of PgPIN10 by PgFLP in setting the gravitropic set-point angle of pomegranate adventitious root in response to gravity signal.

Results
The expression pattern of PgPINs in different rootedness pomegranate self-rooted seedling adventitious root Two different rootedness pomegranate self-rooted seedling are screened in this study, compared with each other, the two self-rooted seedling show different rootedness adventitious roots branch angle ( Table  1). The two self-rooted seedling provide appropriate material development for the study of GSA. We had screened the shallow-rooted pomegranate breed 'Taishanhong'(TSH) and the deep-rooted pomegranate breed: 'Lanbaoshi'(LBS) ( Fig. 1a). The PINs have been found to conducive to nearly every step of auxin uxes [3,18,35,20] and the auxin distribution patterns have been predicted and explained through auxin transport [2,6,1]. In this study, the expression of PgPIN1 and PgPIN10 showed higher in LBS than TSH (Fig. 1b). We also investigated the expression of other PgPINs, PgPIN3 and PgPIN7 showed lower in LBS than TSH (Additional le 1: Figure S1). These results indicated that PgPINs may participate in the regulation of GSA in pomegranate adventitious root.

Molecular cloning of PgFLP
To further characterize the function of PgPINs in pomegranate, we performed the Y1H screening via pomegranate cDNA library. Eventually, we used the promoter sequences of PgPIN10 as baits to obtained a cDNA fragment of PgFLP. To identify the pomegranate PgFLP that is involved in adventitious root gravitropism, a genome-wide analysis was performed. Comparison of amino acid sequences of PgFLP with their close homologs, the PgFLP protein is highly conserved identity to R2R3-MYB protein (Fig. 2a).
To understand the evolutionary relationship of PgFLP, using the neighbor-joining (NJ) method [24], we constructed a phylogenetic tree. As a result, We found that PgFLP had been clustered within the R2R3-MYB clade including AtFLP, MdFLP and others for which root gravitropism have been identi ed. Within the R2R3-MYB clade, PgFLP is closely associated with AtFLP, which regulate lateral root gravitropism, indicating that PgFLP may participate in the regulation of GSA in pomegranate adventitious root (Fig.   2b).

Subcellular localization analysis of PgFLP
To further characterize the function of PgFLP in pomegranate, we used the coding sequence of PgFLP and GFP reporter to constructed a 35S::PgFLP-GFP translational fusion. We found that in contrast to 35S::GFP where a uorescent signal was observed throughout the whole epidermal cell, the 35S::PgFLP-GFP showed a signal in nucleus ( Fig. 3a-b), suggesting that as a transcription factor, PgFLP may be involved in pomegranate adventitious root gravitropism.

PgFLP expression pattern in pomegranate
To further determine the function of PgFLP, we performed the qRT-PCR to determine the expression patterns of PgFLP in pomegranate. We found that the expressing PgFLP were detected in various organs of pomegranate, especially higher expressing in the leaves and the fruit peels (Fig. 4a). The auxin participate in triggering lateral root development [4], we investigated whether PgFLP expression is induced by auxin. The expression of PgFLP was rapidly up-regulated in auxin-treated roots (Fig. 4b).
We also investigated the expression of PgMYB88 was not signi cantly induced by auxin (Additional le 1: Figure S2b). Subsequently, we found that the transcripts of PgFLP in LBS was increased (~2-fold) upon the TSH (Fig. 4c). We also investigated the IAA content in different rootedness pomegranate self-rooted seedling and found that the level of IAA was slightly increased in LBS compared to that in TSH on shoot cutting at 30 d and 40 d (Fig. 4d). These results indicated that PgFLP plays an key role in pomegranate adventitious root gravitropism.

PgFLP directly regulate PgPIN10 transcription
To further characterize the mechanism of PgFLP to regulate GSA of adventitious root in pomegranate, we performed the Y1H assays. Y1H assays showed that PgFLP could interact with the promoter of PgPIN10 to activate its expression (Fig. 5a). To further investigate whether PgFLP could activate the transcriptional expression of PgPIN10 in planta, we conducted a transient expression assay. As a result, the proPgPIN10::GUS plus PgFLP showed signi cantly higher GUS activity in tobacco leaves than the control (Fig. 5b-d). These results showed that PgFLP could activate the expression of PgPIN10 in pomegranate. [31]. We found that there was two potential binding sites including AGCGG and TACCC in the promoter region of PgPIN10 ( Fig. 5e and Additional le 1: Figure S3). To further determine if PgFLP could directly bind to these binding sites in plants, we performed Y1H assays. We found that PgFLP could interact with FBS (AGCGG). On the other hand, PgFLP could not interact with the mutated FBS motif (mFBS) (Fig. 5f). These data demonstrate that PgFLP directly regulate PgPIN10 transcription by binding to the AGCGG motif.

It is well established that FLP can bind directly to the promoters of PINs genes that harbour an [A/T/G] [A/T/G]C[C/G][C/G] motif
Ectopic overexpression of PgFLP in Arabidopsis seedlings results in a smaller GSA of lateral root As it is di cult to obtain transgenic pomegranate plants overexpressing PgFLP, PgFLP was ectopically expressed in Arabidopsis to further characterize if PgFLP functions in the regulation GSA of pomegranate self-rooted seedling adventitious root in planta. Three independent transgenic lines, OE3, OE4 and OE7, were chosen for further analysis (Fig. 6a). To investigate phenotypes of the WT and PgFLP overexpression lines grown for 21 days on MS media, we found that PgFLP (OX-3, OX-4, OX-7) (Fig. 6b-e), exhibited a smaller GSA and grew downwards faster than WT. For example, over 35% of PgFLP-OX lateral root increased within a 30-50° range, meanwhile, within a 50-70° range, over 30% of lateral root was decreased (Fig. 6f). These results indicated that PgFLP functions in the regulation GSA of pomegranate self-rooted seedling adventitious root in planta.

Discussion
PgFLP is involved in the pomegranate self-rooted seedling adventitious root gravitropism The self-rooted seedling is widely used in pomegranate planting industry currently. However, the development of gravitropic set-point angle (GSA) between self-rooted seedling and pomegranate seedling is a great difference, leading to the adventitious root of pomegranate was poor in shallowness and solidity, resulted in poor cold resistance [28]. FLP is angle-inducible gene in response to gravity in Arabidopsis and apple [27,28]. We found that PgFLP had been clustered within the R2R3-MYB clade including AtFLP, MdFLP and others for which root gravitropism have been identi ed. Within the R2R3-MYB clade, PgFLP is closely associated with AtFLP, which regulate lateral root gravitropism, indicating that PgFLP may participate in the regulation of GSA in pomegranate adventitious root (Fig. 2b).
The PINs have been found to conducive to nearly every step of auxin uxes [3,18,35,20] and the auxin distribution patterns have been predicted and explained through auxin transport [2, 6, 1]. FLP and MYB88 have been found to can regulate root gravitropism [27], the late stages of stomatal development [11], cold hardiness [30], female reproductive development [17], and guard mother cell proliferation [11,31]. To further understand the function of PgFLP, the expression of PgFLP was rapidly up-regulated in auxintreated roots (Fig. 4b) and the transcripts of PgFLP in LBS was increased (~2-fold) upon the TSH (Fig.  4c). We also investigated the IAA content in different rootedness pomegranate self-rooted seedling and found that the level of IAA was slightly increased in LBS compared to that in TSH on shoot cutting at 30 d and 40 d (Fig. 4d). PgFLP (OX-3, OX-4, OX-7) (Fig. 6b-e), exhibited a smaller GSA and grew downwards faster than WT. In addition, PgMYB88 transcripts were detected in various pomegranate organs, but the expression of PgMYB88 was not signi cantly induced by auxin (Additional le 1: Figure S2a-b). We also found that the transcripts of PgMYB88 in LBS was increased (~3-fold) upon the TSH (Additional le 1: Figure S2c). Whether PgMYB88 is also involved in transcriptional regulation of adventitious root gravitropism in pomegranate requires further investigation. These results suggested that PgFLP participates in regulating the GSA of adventitious root via the basal expression of PgPIN10 in pomegranate.
The PgFLP is participated in the regulatory network during the pomegranate adventitious root gravitropism

It is well established that FLP can bind directly to the [A/T/G][A/T/G]C[C/G][C/G] motif [31]. Because
PgFLP is closely associated with AtFLP, which regulate lateral root gravitropism, and PgFLP functions in the regulation GSA of pomegranate self-rooted seedling adventitious root in planta (Fig. 6). Y1H experiments showed that PgFLP can interact with FBS (AGCGG). We also found the expression of PgFLP was rapidly up-regulated in auxin-treated roots (Fig. 4b), indicating that the function of PgFLP may be more diverse compared with FLP in Arabidopsis.
We found that FLP can activate the expression of PIN3 and PIN7 in Arabidopsis [27]. In our study, we found that PgFLP could activate the expression of PgPIN10 in pomegranate (Fig. 5) and PgFLP could interact with FBS (AGCGG) (Fig. 5f). These data demonstrate that PgFLP directly regulate PgPIN10 transcription by binding to the AGCGG motif. How PgFLP is participated in the feed-forward transcriptional regulation of PgPIN10 during the pomegranate adventitious root gravitropism needs further investigation.
In our study, a model of PgFLP in response to gravity signal during pomegranate self-rooted seedling adventitious root gravitropism was proposed (Fig. 7). Under gravity, PgFLP directly regulates the expression of PgPIN10 by binding to the AGCGG motif to regulate GSA. On the other hand, other PgPINs may also in response to IAA affecting auxin re ux to regulate GSA. Overall, our results con rm that the transcriptional regulation of PgPIN10 by R2R3-MYB transcription factor PgFLP in setting the gravitropic set-point angle of pomegranate adventitious root in response to gravity signal.

Conclusions
An R2R3-MYB transcription factor PgFLP directly activates PgPIN10 by binding to the AGCGG motif of its promoter region to regulate the GSA of adventitious root in pomegranate. Overall, our results con rm that the transcriptional regulation of PgPIN10 by PgFLP in setting the gravitropic set-point angle of pomegranate adventitious root in response to gravity signal.

Plant materials and growth conditions
The deep-rooted pomegranate breed: 'Lanbaoshi'(LBS) and shallow-rooted pomegranate breed: 'Taishanhong'(TSH), provided by Professor Chuanduo Shi (Shandong Institute of Pomology), were for the study of GSA. The roots of 'Taishanhong'(TSH) were used for gene cloning. Arabidopsis thaliana (ecotype 'Columbia'), provided by Professor Xiang Shen (Shandong Agricultural University), was for genetic transformation. The Arabidopsis seeds were grown on MS medium at 22°C for 16 hours of light / 8 hours of dark.

RNA extraction and qRT-PCR analysis
We used Total RNA Isolation System to extract total RNA from pomegranate tissues. Then, an appropriate amount of total RNA was taken to synthesize cDNA with the PrimeScript®1st Strand cDNA Synthesis Kit (Takara, Japan). The qRT-PCR was performed as previously described [15,28]. The relative expression level of each target gene was normalized to that of the ACTIN gene. Three biological replicates were performed for each analysis. Details primers used in this study are shown in Additional le 2: Table S1.
Sequence and phylogenetic tree analysis The corresponding amino acid sequences were obtained from the NCBI nucleotide database and aligned using ClustalX  [24].
Yeast one-hybrid assays Y187 (Clontech) was used for Y1H assays. The PgFLP was recombined into the pGADT7 vector to obtain the AD-PgFLP plasmid, and the PgPIN10 promoter was cloned into the pHIS2 vector. Yeast Y187 cells containing different combinations of recombinant vectors were placed on medium SD/-Leu-Trp-His with different concentration of 3-AT for detection.

GUS transient assays
The tobacco leaves were used to conduct transient expression assays. The promoter of PgPIN10 was cloned into pCAMBIA1300-GUS to obtain the proPgPIN10::GUS recombinant vector. The coding sequence of PgFLP were inserted into pRI101 plant transformation vector downstream of 35S promoter. The corresponding combinations were co-injected into tobacco leaves according to previously described [28]. The tobacco of the experimental group was cultured normally for 2 days. Then, we detected GUS activity as previously described [28].

The construction of PgFLP overexpression vector and genetic transformation
The complete PgFLP coding region was integrated into the pRI101 plant transformation vector downstream of 35S promoter. The 35S::PgFLP vector was transformed into Arabidopsis plants [27]. The seeds of the transgenic Arabidopsis plants were individually harvested. Homozygous transgenic lines were used for further investigation.

Quanti cation of plant hormones
The pomegranate self-rooted seedling adventitious roots were immediately frozen with liquid nitrogen. Then, we used ELISA (Enzyme-Linked Immuno Sorbent Assay) to measure hormones IAA according to Zhang et al [34].

Measurement of gravity set-point angle
Transgenic Arabidopsis thaliana roots were used to measure GSA. The GSA angles were divided into 0-30º, 30-50º, 50-70º, 70-90º and 90-110º. The Image J software was used to measure GSA. Experiments were repeated independently three times. Statistical signi cance: **P < 0.01. The funders had no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.

Availability of data and materials
The datasets used and analysed during the current study are available from the corresponding author on reasonable request.
Authors' contributions YLY conceived the project. YLY, JHT and ZHW designed the experiments. ZHW performed most of the experiments, and ZHW wrote the paper. JLL helped with the transient expression assays; JLL, XMY, HXT, LJF and FW helped with the phenotypic observations and the paper revision. All authors read and approved the nal manuscript.
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.