Coexpression Analysis Identied PcMYB25 as a Patchoulol Synthase Gene Activator to Enhance Patchouli Alcohol Biosynthesis in Pogostemon Cablin

Patchouli alcohol is an effective component of the medicinal plant patchouli. Similar to other 32 secondary metabolites, its synthesis is likely also regulated by transcription factors. Although 33 the biosynthetic pathway of patchouli alcohol has been characterized, the regulatory mechanism 34 of patchouli alcohol biosynthesis has not been fully revealed. This study combined the transcriptome data of patchouli leaves treated with different 37 hormones and WGCNA to establish a coexpression network. The modules correlated to 38 patchouli alcohol content were identified, and PcMYB25 played a crucial role in regulating 39 patchouli alcohol biosynthesis. The overexpression of PcMYB25 can promote the expression 40 of patchouli alcohol synthase (PTS) , thereby increasing the content of patchouli alcohol. 41 Conclusions This is the first report that MYB25 regulates the secondary metabolism of patchouli. These 43 experimental results lay the foundation for further analysis of the regulatory mechanism of 44 patchouli alcohol synthesis. that the quality of is high be used for further analysis. To access the

Pogostemon cablin (Blanco) Benth. is a kind of herbaceous plant that belongs to Labiatae 52 and is distributed in Southern and Southeast Asia (Swamy and Sinniah 2016). The oil extracted 53 from patchouli has been widely used for medical treatment, fragrances, and cosmetics. 54 Patchouli oil is a special volatile oil, which contains more than 20 kinds of sesquiterpenes 55 (Deguerry et al. 2006). Patchouli alcohol is the main component in the volatile oil of patchouli 56 transcription factor gene, MYB25, was identified from these modules. Transient 118 overexpression analysis confirmed that MYB25 is a positive regulator in the accumulation of 119 patchouli alcohol. This study was the first that identified a transcription factor that positively 120 regulates patchouli alcohol synthesis, laying the foundation for further research on the patchouli 121 alcohol synthesis regulatory network in the future. hormones (dissolved in 0.5% ethanol) were applied on patchouli leaves by spraying. After 135 spraying, the different treatment groups were enclosed by see-through plastic containers to 136 prevent hormones from dissipating. Eight hours after spraying, leaves collected from each 137 treatment group were frozen immediately in liquid nitrogen, and store at -80°C, used for RNA-138   177 The method of PA extraction and quantification used has been described previously (Wang 178 et al. 2019b). Patchouli leaves were ground in liquid nitrogen and 200 mg of powder, after 179 adding 1.5 mL hexane, was extracted by ultrasonic at 60 Hz for 30 min. Then, the mixture was 180 placed in a 56 °C water bath for 1 h. After a short centrifugation, the supernatant was filtered 181 with a 0.22 μm organic microporous membrane, then the filtrate was taken as the test solution 182

PA extraction and quantification
for GC-MS detection. 183 Agilent 7890B/5977A GC-MS and HP5-ms capillary column (30 m × 250 μm × 0.25 μm) 184 was used for the separation and detection of PA. the procedure was set with an initial 185 temperature of 50℃ and kept for 2 min, then the temperature is increased to 130℃ at a rate of 186 20℃/min, and the temperature is increased to 150℃ at 2℃/min, and maintained for 5 min at 187 150 ℃. After that, the temperature increased to 230℃ at rate of 20℃/min. In addition, the 188 patchouli alcohol purchased from Feiyu (Nantong, China) was used as a standard.

Identification of differentially expressed genes (DEGs) 227
To find the potential transcription factors involved in regulation of the PA biosynthetic 228 pathway, we identified DEGs among the five hormone treatment groups by comparing MandE,229 MeJA, SA, ABA, and ETH with control; and by comparing MandE with ETH or MeJA. Here, 230 DESeq2 was used for differential expression analysis, and genes with p<0.05 and FC≥2 or 231 participated in sesquiterpene and triterpenoid biosynthesis (Fig. 1B). 250

Construction of the WGCNA and the identification of hub genes 251
The DEGs were evaluated using WGCNA. Here, genes whose expression level is less than 252 one and whose coefficient of variation is less than 0.1 are filtered. Then, the soft power is 253 adjusted to 12, and the module are identified. Here, the processed genes are divided into 10 254 modules ( Fig. 2A). Association of the PA content (trait) with 10 modules was analyzed to 255 identified key modules and potentially involved transcription factors (Fig. 2B). Among these 256 10 modules, four modules, MEmagenta (r 2 =-0.643, p-value=0.004), MEgreen (r 2 =-0.591, p-257 value=0.0098), MEpink (r 2 =0.544, p-value=0.0196), and MEblack (r 2 =-0.525, p-value=0.0253), 258 were regarded as the key modules associated with the patchouli alcohol content (P<0.05) (Fig.  259   2B). Notably, MEmagenta (r 2 =-0.643, p-value=0.004) has the strongest correction to patchouli 260 alcohol among these 10 modules, but there was not any transcription factor being found in this 261 module. Therefore, we focus on analysis of MEgreen, which is significantly related to the To reveal the function of PcMYB25 in patchouli, the pJLTRBO vector was used to 277 overexpress PcMYB25 in the leaves of patchouli, which is mediated by agrobacteria (Fig. 3A). 278

The efficiency of this overexpression was evaluated by RT-qPCR four days after the injection. 279
Compared with that of the control group, the expression level in the experimental groups 280 increased by 2.55-fold (Fig. 3B). The transcript levels of some genes in the upstream patchouli 281 alcohol biosynthetic pathway, such as AACT and PMK, were significantly increased; on the 282 contrary, the HMGR level was considerably reduced. Notably, the expression level of PcPTS 283 was significantly increased, by 309%, compared with that of the control group (Fig. 4). 284 Interestingly, the PA content was increased by 85% in patchouli leaves overexpressing 285 PcMYB25, which is in accordance with the upregulation of PcPTS (Fig. 3C and Fig. 3D). 286

PcMYB25 codes a R2R3 MYB protein 287
We cloned the full-length coding sequence of PcMYB25 from patchouli cDNAs. After 288 sequence analysis, the ORF including the stop codon of PcMYB25 was determined to be 969 289 bp in length and encodes 322 aa. Amino acid sequence alignment confirms that PcMYB25 290 shares high similarities with the PcMYB25-related proteins retrieved from Rehmannia 291 glutinosa, Salvia splendens, and Actinidia rufa. PcMYB25 harbors the conserved R2 and R3 292 domains in N-terminus, typical of a R2R3-MYB transcription factor. To help predicting 293 potential functions for PcMYB25, a phylogenetic tree was constructed by MEGA 7 (Fig. 5B). 294 Results showed that the closest MYB proteins were from Rehmannia glutinosa, Salvia 295 splendens and Salvia miltiorrhiza. PcMYB25 also displays high similarity (95%) to the R2-R3 conserved domains with Arabidopsis AtMYB25 (Fig. 5B). work on PcMYB25, however, is the first study showing involvement of a MYB25 in secondary 346 metabolism. By WGCNA, we identified a MEgreen module, which shows a tight correlation 347 between gene expression patterns and PA content. Numerous genes from this module were 348 related to terpene synthesis. Prior to this study, the co-expression network had not been used to 349 identify genes and transcription factors related to PA biosynthesis. Although, genes involved in 350 the PA biosynthetic pathway have been identified, the regulatory network response has not been 351 fully elucidated. Identifying more critical genes from this module and their interplays will be a 352 topic in the future research to further examine regulatory network for PA biosynthesis. 353 In the combined correlation between the expression level of PcMYB25 and patchouli alcohol 354 content, the Pearson correlation coefficient was 0.926. Similarly, the correlation coefficients of activator to enhance biosynthesis of patchouli alcohol in its leaves. In addition, after the 357 overexpression of PcMYB25, the PA content in plant leaves was increased by 85%. After testing 358 the gene expression in the patchouli alcohol synthetic pathway, it was found that PcMYB25 had 359 a negligible effect on the upstream genes of the pathway, but it specifically increased the 360 expression of PTS, which was 4.09-fold greater than the original expression level. However, 361 module and trait correlation heat map show a negative correction (r2=-0.591, p-value=0.0098) 362 for green module with PA (Fig. 2B) In addition, constructing a co-expression network, the correlation between gene expression 372 level and trait is considered, but the influence of gene interaction on trait is not considered. 373 Modules clustering heat map showed that, genes in the green module highly expressed in 374 MandE treatment group, lowly expressed in ETH treatment group, and expressed irregularly in 375 other treatment groups. However, the PA content was the lowest under the MandE treatment, 376 so the WGCNA analysis showed that the module was negatively correlated with the gene. In 377 addition, the module is negatively correlated with genes, but there may be differences of genes 378 in green module, which could be the reason of the positive correlation between the PcMYB25 379 and PA. 380

381
In this study, through WGCNA, we identified that the MEgreen module was related to 382 patchouli alcohol biosynthesis. This module provides candidate genes related to patchouli 383 alcohol, and the hub transcription factor PcMYB25 was able to upregulate PTS to increase the 384 content of patchouli alcohol. This is the first time that MYB25 has been reported as a 385 transcriptional activator to regulate the biosynthesis of secondary metabolism. This discovery lays the foundation for further research on the transcriptional regulatory network of patchouli