In plants, triacylglycerols (TAGs), which provide energy for seed germination and seedling growth, consist of three fatty acids bound to a glycerol backbone (Lu et al. 2018). These TAGs are stored within specialized organelles known as oil bodies (OBs). OBs typically range from 0.2 to 2.0 µm in diameter and are found in the seeds, pollen, and tapetum of higher plants (Lu et al. 2018; Yuan et al. 2021). Enclosed within a single layer of phospholipids, these OBs are stabilized by a group of embedded proteins, including oleosins (OLEs), caleosins, and steroleosins (Hyun et al. 2013). OLEs, accounting for 80%-90% of the structural proteins in OBs, are a group of hydrophobic proteins predominantly located on the surfaces of OBs in seeds and pollen (Cao et al. 2014; Siloto et al. 2006). Loss-of-function mutations in Arabidopsis OLE1 affected OB size and reduced seed oil contents (Siloto et al. 2006), whereas the overexpression of Carthamus tinctorius OLEs (Lu et al. 2018) and Glycine max OLE1 (Zhang et al. 2019) increased seed oil contents. These findings indicate that OLEs play a crucial role in regulating the size and number of OBs and in regulating lipid accumulation (Shimada and Hara-Nishimura 2010). In addition, Arachis hypogaea OLE3 (AhOLE3) was identified as a bifunctional enzyme that could exhibit both monoacylglycerol acyltransferase and phospholipase activities and was regulated by serine/threonine/tyrosine protein kinases (Parthibane et al. 2012a; 2012b). During seed germination, the phosphorylation level of AhOLE3 was found to increase, possibly as a result of elevated PLA2 activity. This activity induced the hydrolysis of phosphatidylcholine to lysophosphatidylcholine, resulting in the disturbance of the membrane structure and facilitating the release of TAGs from OBs. This finding suggests that OLEs are crucial for the biosynthesis and degradation of TAGs.

Given the essential role of OLEs in the biosynthesis and degradation of TAGs, many OLE genes have been identified and characterized in various plant species, including safflower (Carthamus tinctorius), rice (Oryza sativa), flax (Linum usitatissimum), castor bean (Ricinus communis), Arabidopsis

(Arabidopsis thaliana), tiger nut (Cyperus esculentus), and rapeseed (Brassica napus) (Chen et al. 2019; Hyun et al. 2013; Kim et al. 2002; Liu et al. 2012; Lu et al. 2018; Zhu et al. 2023). This indicates that the OLE family is highly abundant and diverse in land plants. Nevertheless, to gain a more comprehensive understanding of OLEs across different plant species, OLEs within different plant genomes must be systematically identified and analyzed.

In this study, we used publicly available databases and bioinformatic tools and conducted a comprehensive genome-wide analysis of balloon flower (Platycodon grandiflorus), a highly significant medicinal crop. Phylogenetic classification and domain analyses to predict the specific functions of OLEs from P. grandiflorus (PlgOLEs). In addition, the expression profiles of PlgOLEs in response to abiotic stresses suggested that they are stress-inducible genes. Our systematic analysis provides a foundation for further functional dissection of OLE genes in balloon flowers and could help to elucidate the function of these genes in higher plants.

Identification and characterization of the OLE family in P. grandiflorus

To identify members of the PlgOLE family, the genome sequence of P. grandiflorus (Kim et al. 2020) was searched using OLE protein sequences from Arabidopsis and rice through Basic Local Alignment Search Tool (BLAST) algorithms. Putative PlgOLEs underwent analyses including conserved domain identification, determination of molecular weight, phylogenetic analysis, determination of isoelectric point (pI), and subcellular localization (WoLF PSORT, https://wolfpsort.hgc.jp/). A Kyte-Doolittle hydropathic plot (Kyte and Doolittle 1982) was generated using ProtScale (http://web.expasy.org/protscale/) with the Kyte-Doolittle option. For cis-element analysis, all 1.5-kb upstream of PlgOLEs were compared with known cis-regulatory elements using the PlantCARE database (https://bioinformatics.psb.ugent.be/webtools/plantcare/html/).

Analysis of tissue-specific expression profiles

Tissue-specific expression of PlgOLEs was analyzed using RNA-seq data of eight different tissues downloaded from NCBI GenBank (SRR8712510-SRR8712517). Gene expression levels were estimated as fragments per kilobase of transcript per million mapped reads (FPKM) using methods similar to those described by Kim et al. (2020).

Plant growth and treatment

Balloon flower seeds were germinated and cultivated in a growth chamber under controlled conditions (24°C temperature and 50% relative humidity). To analyze the expression pattern of PlgOLEs in response to abiotic stresses, 2-month-old plants were treated with 250 mM NaCl or effectively wounded using a sterile syringe needle. For heat stress treatment, the plants were treated at 45°C.

Quantitative real-time PCR (qRT-PCR) analysis

The expression of PlgOLEs was analyzed using qRT-PCR. Total RNA was extracted using RNA extraction kit following the manufacturer’s instructions (Favorgen, Ping Tung, Taiwan) and reverse transcribed into cDNA. qRT-PCR was performed using Toyobo SYBR Green Master Mix (Toyobo, Co., Ltd., Osaka, Japan), with the balloon flower actin gene serving as an internal reference. Expression levels were normalized to the constitutive expression level of the actin and calculated relative to values at mock treatment. Expression levels are represented as log2 ratios. Primer sequences are listed in Table 1.

Table 1 . Primer sequences used for qRT-PCR analysis.

Primer name	Sequence (5’-3’)
PlgOLE-1	F-CTTGATTGGGCTCGAGAGAG
	R-TTAAGAACCCGCCTGTTGAC
PlgOLE-2	F-CTGAGCACCATCCTCTACAC
	R-AGGCCAGAGAGTACCAGTAG
PlgOLE-3	F-GGACCGATTTGACTATGCGA
	R-GCAGCGTCCTTAACCTTACT
PlgOLE-4	F-GTCCTTACTCCCACAGCAAG
	R-AGATGATGAAGAGCGGTGTG
PlgOLE-5	F-CCTCCCACAGAGTCAGAGAG
	R-GAGTGATGCCAGCAAGTAGG
PlgActin	F-CCATACAGTCCCCATTTATGAAG
	R-GCTAACTTCTCCTTCATGTCTCTCA

Identification of OLEs in balloon flowers

The availability of the draft balloon flower genome sequence has enabled the identification of putative OLEs in this plant species. For the identification of putative OLEs, BLAST was used to search the genomic database using OLEs previously identified in Arabidopsis and rice as queries. Redundant sequences were eliminated through self-BLAST analysis, followed by manual curation, resulting in the identification of five putative OLE genes (Table 2). These genes were designated as PlgOLE-1 to PlgOLE-5. OLEs are characterized as small alkaline proteins, typically ranging from 13 to 23 kDa (Hyun et al. 2013). The putative PlgOLE proteins exhibited molecular masses ranging from 14.89 to 18.69 kDa and theoretical pIs ranging from 9.70 to 10.11 (Table 2). In A. hypogaea, AhOLE3 has been found to be associated with chloroplast OBs, which are structurally analogous to seed OBs (Parthibane et al. 2012a). In addition, OLEs are located in the cell membrane before OB formation in Brassica napus (Chen et al. 2019). Similar to these findings, subcellular localizations predicted using WoLF PSORT indicated that PlgOLEs were potentially localized in the chloroplast and plasma membrane (Table 2).

Table 2 . Oleosin family in P. grandiflorus.

Name	Locus name	CDS (bp)	AA	pI	kDa
PlgOLE-1	PGJG135130	444	147	9.70	15.50
PlgOLE-2	PGJG228120	429	142	9.86	14.89
PlgOLE-3	PGJG249670	507	168	9.83	17.96
PlgOLE-4	PGJG068290	543	180	9.99	18.69
PlgOLE-5	PGJG335340	450	149	10.11	15.84

The central hydrophobic OB-anchoring domains of OLEs are characterized by a proline knot motif (PX₅SPX₃P), which typically includes three proline residues and one serine residue, forming a hairpin-like structure within the OB-TGA matrix (Hyun et al. 2013). Analysis of putative PlgOLEs revealed the presence of the proline knot motif (Fig. 1A). OLEs are categorized into H-, U-, and L-isoforms, with the distinction being an additional 18-residue segment in the C-terminal region of H-OLE (Zhi et al. 2017). To classify PlgOLEs, we aligned the C-terminal regions around the insertion site, categorizing them into three classes, with PlgOLE-4 and PlgOLE-5 being identified as H-OLEs (Fig. 1A). This classification was corroborated by the phylogenetic analysis of PlgOLEs (Fig. 1A). Gymnosperms have been found to exclusively possess L-OLEs (Wu et al. 1999), suggesting that L-OLE is a primitive isoform (Tai et al. 2002). In vitro analysis of OB assembly revealed that OBs reconstituted with L-OLE are more stable than those reconstituted with H-OLE (Tai et al. 2002), implying that PlgOLE-1 and PlgOLE-2 are promising candidates for enhancing OB stability.

Figure 1. Putative oleosin proteins in P. grandiflorus. (A) A phylogenetic tree of the P. grandiflorus oleosins (PlgOLEs) was constructed using the neighbor-joining method in MEGA7. In the proline knot motif, the four invariable residues are highlighted within a red box, while the H-form insertion in C-terminal regions is indicated within a yellow box. (B) PlgOLEs were evaluated using the Kyte-Doolittle hydrophobicity plot, employing a moving window of nine residues

An OLE molecule comprises three segments: an amphipathic N-terminal portion, a central hydrophobic portion typically around 70 residues in length, and an amphipathic C-terminal portion of variable length (Hyun et al. 2013; Zhi et al. 2017). To provide further evidence, the hydrophobicity of putative PlgOLEs was analyzed using the Kyte-Doolittle method (Kyte and Doolittle 1982). As shown in Fig. 1B, an approximately 76-residue-long conserved central hydrophobic region was observed across all putative PlgOLEs. For example, PlgOLE-1 exhibited a 72-residue-long central hydrophobic region, flanked by N and C termini having variable lengths and differing amphipathic properties. The similarity of these characteristics to those of other plant OLEs suggests that all putative PlgOLEs belong to the OLE family.

Expression profiles of PlgOLEs in different balloon flower tissues

Analysis of tissue-specific expression patterns is crucial for understanding the potential functions of a gene within distinct tissues. In this study, we assessed the expression profiles of PlgOLEs across various tissues, including leaf, root, stem, seed, petal, pistil, sepal, and stamen. As shown in Fig. 2, most PlgOLEs were primarily expressed in seeds. Similarly, OLEs have been reported to be exclusively expressed in Arabidopsis, castor bean, and flax seeds (Hyun et al. 2013; Kim et al. 2002). The predominant expression of PlgOLEs suggests that they play a role in regulating seed OB stability. Extensive research on cytosolic OBs in seeds has revealed their presence in pollen and various vegetative organs (Bouchnak et al. 2023). In balloon flowers, the transcription levels of PlgOLE-1, -2, and -5 were also detected in other tissues but with low expression level compared to than those in seeds (Fig. 2). Taken together, these findings indicate that PlgOLEs may have broader functions beyond their primary role in regulating seed OB stability and size.

Figure 2. Tissue-specific expression patterns of PlgOLEs. The results represent the FPKM values of RNA-seq data generated from eight different tissues of P. grandiflorus

Expression of PlgOLEs in response to abiotic stresses

To characterize the general features of the promoter regions of PlgOLEs, 1.5-kb sequences upstream of the ATG start codon were searched against known cis-regulatory elements in the PlantCARE database. In total, 18 potential cis-acting elements associated with hormonal response, including abscisic acid (ABA) response; abiotic stress response; light response; and developmental regulation were identified in the promoters of PlgOLEs (Fig. 3A). Several light-responsive elements, including the GT1-motif, ACE motif, TCCC-motif, G-box, TCT-motif, Sp1-binding site, and AE-box, were also identified in the promoters of PlgOLEs. Among these, the G-box was the most abundant light-responsive element. The ABRE motif, which is an ABA-responsive element (Kim et al. 2011), was identified in all PlgOLEs, with one to five copies being detected in each gene. In addition, ARE, which is a cis-acting regulatory element essential for anaerobic induction (Yin et al. 2017), was identified in PlgOLE-1, -2, -4, and -5 promoters. These findings suggest that PlgOLEs respond to environmental conditions through a complex mechanism mediated by various phytohormones, including ABA.

Figure 3. Stress-responsive transcription of PlgOLEs. (A) Expression patterns of PlgOLEs in response to abiotic stresses. The heatmap illustrates the gene expression levels determined using qRT-PCR. (B) Cis-element analysis of PlgOLEs. The various colored markers indicate the predicted number of cis-elements

While TAGs typically do not accumulate to significant levels in vegetative tissues under optimal growth conditions, various stressors, such as drought, extreme temperatures, and nutrient deprivation, can induce their production, particularly in leaves (Lu et al. 2020). In addition, OLE was down-regulated by heat stress in embryogenic carrot cell (Milioni et al. 2001), whereas sorghum OLE was induced by ABA, NaCl and PEG treatment (Buchanan et al. 2005). This indicates that OB-associated proteins, including OLEs, are stress-responsive proteins. To determine the involvement of PlgOLEs in balloon flowers in response to abiotic stresses, we analyzed the expression patterns of PlgOLEs in response to external stimuli, such as salt, heat, or wounding. In balloon flower plants, the expression level of PlgOLE-1 increased in response to salt stress but decreased in response to heat stress (Fig. 3B). In addition, PlgOLE-2, -4, and -5 transcripts were strongly induced by wounding, heat, and salt stresses. However, the expression level of PlgOLE-3 was reduced by heat and salt stresses (Fig. 3B). Taken together, these expression patterns indicate a divergence in the function of PlgOLEs in response to different stimuli.

In this study, we conducted a comprehensive analysis of the OLE family in balloon flowers; this included genome-wide, phylogenetic tree, protein motif, and hydrophobicity analyses. The expression profiles suggested that members of the PlgOLE family have evolved diverse physiological functions in response to stresses, laying the foundation for future research on the physiological roles of PlgOLEs. An additional challenge would be to analyze the protein activities and functions using genetic modification approaches.

This work was supported by a funding for the academic research program of Chungbuk National University in 2023.