Physiological and Comparative Transcriptome Analyses of the High-Tillering Mutant mtn1 Reveal Regulatory Mechanisms in the Tillering of Centipedegrass (Eremochloa ophiuroides (Munro) Hack.)

Tillering is a key factor that determines the reproductive yields of centipedegrass, which is an important perennial warm-season turfgrass. However, the regulatory mechanism of tillering in perennial plants is poorly understood, especially in perennial turfgrasses. In this study, we created and characterised a cold plasma-mutagenised centipedegrass mutant, mtn1 (more tillering number 1). Phenotypic analysis showed that the mtn1 mutant exhibited high tillering, short internodes, long seeds and a heavy 1000-seed weight. Then, a comparative transcriptomic analysis of the mtn1 mutant and wild-type was performed to explore the molecular mechanisms of centipedegrass tillering. The results revealed that plant hormone signalling pathways, as well as starch and sucrose metabolism, might play important roles in centipedegrass tillering. Hormone and soluble sugar content measurements and exogenous treatment results validated that plant hormones and sugars play important roles in centipedegrass tiller development. In particular, the overexpression of the auxin transporter ATP-binding cassette B 11 (EoABCB11) in Arabidopsis resulted in more branches. Single nucleotide polymorphisms (SNPs) were also identified, which will provide a useful resource for molecular marker-assisted breeding in centipedegrass. According to the physiological characteristics and transcriptional expression levels of the related genes, the regulatory mechanism of centipedegrass tillering was systematically revealed. This research provides a new breeding resource for further studies into the molecular mechanism that regulates tillering in perennial plants and for breeding high-tillering centipedegrass varieties.


Introduction
Centipedegrass (Eremochloa ophiuroides (Munro) Hack.) is a perennial warm-season C 4 grass originating from China with wide adaptability, high barrenness and acid tolerance [1,2]. It has great potential use for soil conservation and environmental remediation and as a forage grass [3]. As a perennial turfgrass, centipedegrass is propagated by stolons and seeds, which directly depend on the continuous tillering of stolon nodes [4,5]. However, the regulatory mechanism in the development of tillering in centipedegrass has not been reported.
Tillering is an important breeding trait that determines reproduction efficiency and yields in perennial plants. It appears from the node and grows independently of the mother stem from its own roots [6,7]. Previous studies have indicated that auxin (IAA), gibberellic acid (GA) and cytokinin (CTK), are the most important factors affecting tillering [7][8][9][10]. Auxin has been reported to inhibit tillering. In rice, the overexpression of PIN-FORMED1 (OsPIN1), OsPIN2 and OsPIN9 increases the amount of tillering [8,11]. In Arabidopsis, the atiaa17 mutant shows a high branching phenotype [12]. GA is involved in the regulation

RNA-Seq Data Analysis
A total of 279.00 million raw reads were obtained from the six libraries, and 96.08% of the reads were considered clean reads (Table S1). The GC contents ranged from 53.93% to 54.95%, Q20 (proportion of nucleotides with a quality value greater than 20 in clean reads) ranged from 98.43% to 98.60% and Q30 ranged from 95.34% to 95.69% (Table S1). As a result, 93.70% to 94.90% of the total clean reads and 87.42% to 89.30% of the unique reads were mapped to the centipedegrass genome (Table S2). Among them, 81.52% to 83.46% of clean reads were mapped to exonic regions, 4.34% to 5.04% were mapped to intronic regions and 12.10% to 13.48% were mapped to intergenic regions (Table S3).

Identification of DEGs between the mtn1 Mutant and WT
To explore DEGs involved in high tillering, a comparison was performed between the mtn1 mutant and WT. A total of 3025 DEGs were detected, including 1589 upregulated genes and 1436 downregulated genes (Figure 2A,B). The expression patterns of the DEGs were hierarchically clustered ( Figure 2C).

Identification of DEGs between the mtn1 Mutant and WT
To explore DEGs involved in high tillering, a comparison was performed between the mtn1 mutant and WT. A total of 3025 DEGs were detected, including 1589 upregulated genes and 1436 downregulated genes (Figure 2A,B). The expression patterns of the DEGs were hierarchically clustered ( Figure 2C).

Functional Analysis of DEGs
According to gene ontology (GO) subcategories, the DEGs were divided into three main processes: biological processes (BP), cellular components (CC) and molecular functions (MF) ( Figure S1). Some DEGs were annotated with more than one GO term. The small molecule metabolic process, ion transport and response to stress were highly enriched among the biological processes. The top three highly enriched processes in cellular components were nonmembrane-bound organelles, organelle parts and cell peripheries. The top three highly enriched processes in molecular functions were transmembrane transporter activity, transferase activity and hydrolase activity. According to a Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, the plant hormone signal transduction pathways were the most significantly enriched ( Figure 3). In addition, plant-pathogen interactions, phenylpropanoid biosynthesis and the mitogen-activated protein kinase (MAPK) signalling pathway were also significantly enriched. These results indicate that plant hormones might play important roles in the high tillering of the mtn1 mutant.

Functional Analysis of DEGs
According to gene ontology (GO) subcategories, the DEGs were divided into three main processes: biological processes (BP), cellular components (CC) and molecular functions (MF) ( Figure S1). Some DEGs were annotated with more than one GO term. The small molecule metabolic process, ion transport and response to stress were highly enriched among the biological processes. The top three highly enriched processes in cellular components were nonmembrane-bound organelles, organelle parts and cell peripheries. The top three highly enriched processes in molecular functions were transmembrane transporter activity, transferase activity and hydrolase activity. According to a Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, the plant hormone signal transduction pathways were the most significantly enriched ( Figure 3). In addition, plant-pathogen interactions, phenylpropanoid biosynthesis and the mitogen-activated protein kinase (MAPK) signalling pathway were also significantly enriched. These results indicate that plant hormones might play important roles in the high tillering of the mtn1 mutant.

Regulatory Pathways of DEGs Related to Plant Hormones
To identify the contributions of hormone-mediated regulation to high tilling in the mtn1 mutant, we compared the expression of genes in hormone-related pathways. There were 44 DEGs related to auxin (33 DEGs), GA (6 DEGs) and CTK (5 DEGs) ( Figure 4A and Table S4). A total of six auxin transport-related DEGs were found. Among them, ABCB11 and PIN9 were upregulated, and the Auxin1/like aux 4 (AUX1/LAX4) and ABCB19 genes were downregulated in the mtn1 mutant (Table S4). A total of 16 DEGs were annotated as small auxin upregulated RNA (SAUR) genes, 10 of which were upregulated in the mtn1 mutant (Table S4). A total of two Gretchen hagen 3 genes (GH3.8 and GH3.12) were upregulated in the mtn1 mutant (Table S4). These results suggest that auxin is one of the main hormones regulating high tillering in the mtn1 mutant, and the polar transport of this hormone was highly active, while the signal transduction pathway was reactive.

Regulatory Pathways of DEGs Related to Plant Hormones
To identify the contributions of hormone-mediated regulation to high tilling in the mtn1 mutant, we compared the expression of genes in hormone-related pathways. There were 44 DEGs related to auxin (33 DEGs), GA (6 DEGs) and CTK (5 DEGs) ( Figure 4A and Table S4). A total of six auxin transport-related DEGs were found. Among them, ABCB11 and PIN9 were upregulated, and the Auxin1/like aux 4 (AUX1/LAX4) and ABCB19 genes were downregulated in the mtn1 mutant (Table S4). A total of 16 DEGs were annotated as small auxin upregulated RNA (SAUR) genes, 10 of which were upregulated in the mtn1 mutant (Table S4). A total of two Gretchen hagen 3 genes (GH3.8 and GH3.12) were upregulated in the mtn1 mutant (Table S4). These results suggest that auxin is one of the main hormones regulating high tillering in the mtn1 mutant, and the polar transport of this hormone was highly active, while the signal transduction pathway was reactive.
In this study, a total of three DEGs related to CTK metabolism were significantly expressed ( Figure 4C and Table S4). The expression levels of LOG6 and CKX5 were upregulated, and the ZOG3 gene was downregulated in the mtn1 mutant. One DEG related to CK transport (Purine nucleobase transmembrane transport 3, PUP3) was downregulated, and one DEG related to CK signal transduction (Ethylene-responsive transcription factor 4, CRF4) was upregulated in the mtn1 mutant compared with the WT (Table S4).
In this study, a total of three DEGs related to CTK metabolism were significantly expressed ( Figure 4C and Table S4). The expression levels of LOG6 and CKX5 were upregulated, and the ZOG3 gene was downregulated in the mtn1 mutant. One DEG related to CK transport (Purine nucleobase transmembrane transport 3, PUP3) was downregulated, and one DEG related to CK signal transduction (Ethylene-responsive transcription factor 4, CRF4) was upregulated in the mtn1 mutant compared with the WT (Table S4).

Confirmation of RNA-Seq Data
Eight of the DEGs (PIN9, ABCB11, SAUR36, SAUR50, IAA14, GH3.12, GA2ox1, CKX5) were randomly selected to confirm the reliability of the RNA-seq data ( Figure S2). The results indicated that the expression levels of the genes evaluated by qRT-PCR were consistent with those evaluated by RNA-seq. Therefore, the results of the RNA-seq were reliable.

IAA, GA, CTK and Soluble Sugar Contents and Their Influence on the Tillering Number
The results showed that the IAA and GA contents in the mtn1 mutant were obviously lower than those in the WT ( Figure S3A,B), and isopentenyl adenosine (IPA) and soluble sugar contents in the mtn1 mutant were higher than those in the WT ( Figure S3C,D). The application of IAA and GA 3 decreased the tillering number ( Figure S3E), while 6-BA and starch increased the tillering number of the mtn1 mutant and WT ( Figure S3E).

Overexpression of EoABCB11 in Arabidopsis Affected Branch Number
To further explore gene function, the EoABCB11 gene was transferred into wild-type Arabidopsis. A total of eight transgenic lines were acquired, and six lines showed similar phenotypes. Three of the six transgenic lines were selected to measure the branch number. Compared with the wild-type (WT) plants, the overexpressed transgenic lines showed higher branch numbers ( Figure 6).

Overexpression of EoABCB11 in Arabidopsis Affected Branch Number
To further explore gene function, the EoABCB11 gene was transferred into wi Arabidopsis. A total of eight transgenic lines were acquired, and six lines showed phenotypes. Three of the six transgenic lines were selected to measure the branch n Compared with the wild-type (WT) plants, the overexpressed transgenic lines s higher branch numbers ( Figure 6).

Identification of SNPs
A total of 80,384 and 8,2247 SNPs in mtn1 and WT were identified, mainly inc transitions and transversions ( Table 2). Transitions accounted for 64.07% in mt 64.24% in WT. Transversions accounted for 35.93% in mtn1 and 35.76% in WT, which, the most common base substitution was C/T, followed by A/G, while the was A/T ( Table 2).

Identification of SNPs
A total of 80,384 and 8,2247 SNPs in mtn1 and WT were identified, mainly including transitions and transversions ( Table 2). Transitions accounted for 64.07% in mtn1 and 64.24% in WT. Transversions accounted for 35.93% in mtn1 and 35.76% in WT, among which, the most common base substitution was C/T, followed by A/G, while the rarest was A/T ( Table 2).

Discussion
The molecular mechanism underlying tillering in centipedegrass is unknown. In this work, we studied a high-tillering mutant of centipedegrass that exhibited high tillering, fast bud outgrowth, short internodes and large seeds. The high-tillering phenotype of the mtn1 mutant was the result of fast bud outgrowth. We used RNA-seq to explore the internal reason for high tillering in the mtn1 mutant. The results showed that the DEGs involved in plant hormone signal transduction pathways and sugar metabolism were highly enriched, indicating that plant hormones and sugars likely play important roles in regulating tillering in centipedegrass.

Regulatory Role of Plant Hormones in Centipedegrass Tillering
Auxin plays an important role in plant growth and development [8]. The balance of auxin content controlled by auxin transporters is essential for tillering [8,29,30]. In this work, the expression levels of auxin efflux carriers related genes ABCB11 and PIN9 were upregulated, and the expression levels of the auxin influx carriers related genes AUX1/LAX4 and ABCB19 were downregulated in the mtn1 mutant. Those results were similar to those of Hou et al. [8], who reported that the overexpression of OsPIN9 increases tillering in rice. The same result was also found in Arabidopsis; the atabcb19 mutant showed a low level of auxin and more branches [31]. In this study, we also found that the GH3.8 gene, which promotes the degradation of auxin, was highly expressed in the mtn1 mutant. This result was in agreement with a previous study on rice, which indicated that the overexpression of the OsGH3.8 gene reduces auxin content and increases tillering numbers [32]. IAA14 is a transcriptional repressor that can interact with ARF5 to negatively regulate auxin response gene expression [33]. In this study, the expression level of IAA14 was upregulated, which was consistent with Liu et al. [34], who found that the overexpression of the PtrIAA14 gene in Arabidopsis results in more branches. This result suggested that IAA14 might have a positive effect on centipedegrass tillering. We speculated that these genes maintain the low level of auxin that promotes tillering in centipedegrass.
Gibberellin 2-oxidases (GA2oxs) are important GA metabolism enzymes that reduce the GA content [35][36][37]. In this work, the expression level of GA2ox1 was upregulated in the mtn1 mutant. This result was in agreement with two previous studies on bahiagrass and hybrid aspen, which indicated that the overexpression of PnGA2ox1 and AtGA2ox1 leads to more branches [38,39]. We inferred that the mtn1 mutant may accumulate less GA relative to GA2ox1, which might underlie the differences in tillering ability between the mtn1 mutant and wild-type. Therefore, the overexpression of GA2oxs is a clear way to reduce GA levels and improve tillering in centipedegrass.
CTKs exist widely in plant tissues, directly promoting tillering by activating axillary bud development [40][41][42]. In this study, we found that the LOG6 gene, which promotes the accumulation of CTK, was highly expressed. This result was consistent with Kurakawa et al. [43], who reported that the OsLOG1 gene is required to maintain meristem activity and that its loss of function causes low tillering in rice. Otherwise, we found that the expression level of the ZOG1 gene, which regulates cytokinin glucosylation levels, was downregulated in the mtn1 mutant. The result was consistent with Shang et al. [18], who found that the knockdown of OsZOG1 expression improves the tillering of rice. Taken together, these data suggested that CTK may act as a positive regulator of high tillering in centipedegrass, and the mtn1 mutant accumulated more CTK, mainly as a consequence of the high expression level of LOG6 and low expression level of ZOG1.

Regulatory Role of Sugars in Centipedegrass Tillering
In addition to plant hormones, sugars are also necessary for tillering, as they can act as an early signal and provide energy during the tillering development process [44][45][46]. In this work, the soluble sugar content in the mtn1 mutant was higher than that of the wild-type. This result was in agreement with the previous opinion that plants with more tillering showed more sugar accumulation [46]. During the starch and sucrose metabolism process, TPPs catalyse the transformation of trehalose-6P into trehalose [23]. In this work, the expression level of TPP7 was upregulated in the mtn1 mutant. This result was in agreement with previous studies on Arabidopsis [24], maize [25] and upland cotton [45], which indicated that a high expression level of TPPs could increase the tillering/branch number. The high expression level of the TPP7 gene was consistent with the high soluble sugar concentration and suggested that sugars are an energy source or nutrient aiding in bud growth in high-tillering centipedegrass.
Based on expression analysis of hormone and sugar-related genes, hormone and sugar contents and the results of previous studies, we propose a regulatory network model to explain the role of plant hormones and sugars in the control of high tillering in centipedegrass (Figure 7). We propose that plant hormones and sugars play primary regulatory roles in centipedegrass tillering. IAA and GA are negative regulators, and CTK is the positive regulator. The biosynthesis, transport and signal transduction of IAA, GA and CTK are required for fast cell expansion, division and tissue differentiation in high tillering. The increase in soluble sugar content provides a greater energy source, which improves protein and cell wall synthesis in the tillering of the mtn1 mutant. The interaction between the hormones and sugars results in the tillering of the mtn1 mutant. centipedegrass (Figure 7). We propose that plant hormones and sugars play primary regulatory roles in centipedegrass tillering. IAA and GA are negative regulators, and CTK is the positive regulator. The biosynthesis, transport and signal transduction of IAA, GA and CTK are required for fast cell expansion, division and tissue differentiation in high tillering. The increase in soluble sugar content provides a greater energy source, which improves protein and cell wall synthesis in the tillering of the mtn1 mutant. The interaction between the hormones and sugars results in the tillering of the mtn1 mutant.

EoABCB11 Was Involved in the Tillering of Centipedegrass
In total, four pathways were identified and examined in detail, revealing that centipedegrass tillering could be regulated by complex mechanisms mediated by plant hormones and/or sugar metabolism. Among these factors, we chose the IAA transport protein ABCB11 for further study, as the tillering number can be increased by increases in IAA auxin efflux. In this study, the branch number was increased by the overexpression of the EoABCB11 gene in Arabidopsis. Similar results were found in Arabidopsis, which indicated that atabcb19 and atabcb1abcb19 mutants showed high branches [28,31]. However, the role of the ABCB11 gene in tillering is still unknown. The ABCB protein can interact with PIN proteins to affect the stability of PIN in the plasma membrane region, thus improving the substrate specificity of the transporter. Previous studies found that ABCB1 and ABCB19 can interact with PIN1 and PIN2, and plants expressing PIN1-ABCB1 showed an increase in

EoABCB11 Was Involved in the Tillering of Centipedegrass
In total, four pathways were identified and examined in detail, revealing that centipedegrass tillering could be regulated by complex mechanisms mediated by plant hormones and/or sugar metabolism. Among these factors, we chose the IAA transport protein ABCB11 for further study, as the tillering number can be increased by increases in IAA auxin efflux. In this study, the branch number was increased by the overexpression of the EoABCB11 gene in Arabidopsis. Similar results were found in Arabidopsis, which indicated that atabcb19 and atabcb1abcb19 mutants showed high branches [28,31]. However, the role of the ABCB11 gene in tillering is still unknown. The ABCB protein can interact with PIN proteins to affect the stability of PIN in the plasma membrane region, thus improving the substrate specificity of the transporter. Previous studies found that ABCB1 and ABCB19 can interact with PIN1 and PIN2, and plants expressing PIN1-ABCB1 showed an increase in auxin output [27,47]. In this study, consistent with ABCB11, the expression level of PIN9 was upregulated, indicating that EoABCB11 and PIN9 might have an interactive relationship to regulate the tillering of centipedegrass. The molecular mechanism of EoABCB11 and its governing effect on centipedegrass tillering requires further study.

SNPs Identification from mtn1 Mutant and WT
SNPs developed from RNA-seq have been widely used as powerful molecular markers for breeding and genetic research on plants [48,49]. In this study, SNPs were in agreement with previous studies, which indicated that the high content of A/G and C/T transitions in Chinese cabbage and cucumber [48,49]. The results provide a useful resource for future studies on gene function and breeding in centipedegrass.

Plant Materials
The centipedegrass cultivar 'Yuxi' (wild-type, WT) was got from the Institute of Botany in Jiangsu Province and the Chinese Academy of Sciences, Nanjing, China. A more tillering number (mtn1) mutant was obtained from calli of the WT radiated by cold plasma, a new radiation method.

Experimental Design
Experiments were performed at the Institute of Botany in Jiangsu Province and the Chinese Academy of Sciences, Nanjing, China (118 • 46 E, 32 • 03 N). Uniform stolons of WT and mtn1 with the first three nodes were planted in a polyethylene plastic pot (35 cm in height and 30 cm in diameter) to investigate phenotypic and yield characteristics. Each pot was filled with 5 kg of soil:sand mixture (w/w, 4:1). The physical and chemical properties of the soil were: pH, 6.20; total nitrogen, 0.44 g kg −1 ; available phosphorus, 4.2 mg kg −1 ; available potassium, 137.01 mg kg −1 ; and dissolved organic carbon, 610.1 mg kg −1 . Each pot contained one stolon. The pots were placed in a greenhouse with a rain shelter, and the temperature and humidity varied from 25-35 • C and 65-80%, respectively. Each type of material was planted in 20 pots. Ten pots of plants were used for phenotype and RNA-seq analysis, and another ten pots were used for yield analysis. Tillering phenotype characteristics, including tiller bud number and tillering bud length, were measured after 20 d of planting, and other characteristics were measured after 40 d of planting. Tiller buds of the mtn1 mutant and wild-type were collected after 40 d of planting, frozen in liquid nitrogen and stored at −80 • C for transcriptome and hormone analyses.

Phenotypic Characteristics and Yields
Tillering phenotype: Tiller bud number (a tiller bud length greater than 1 cm was considered a tiller) and tiller bud length were measured after 20 d of planting [15]. The tiller number, length and diameter of the third internode and the dry weights of the shoots and roots were measured after 40 d of planting. Ten individuals from each sample were randomly selected. Additionally, tiller buds (including nodes) with a length of 1-2 cm from the mtn1 mutant and WT were collected separately as different samples with three replicates. The samples were frozen with liquid nitrogen and then stored at −80 • C for transcriptome and hormone analysis.
Yield characteristics: the length and width of seeds and 1000-seed weight were measured at the mature stage. Ten individuals were measured for each material.

Quantitative Real-Time PCR Analysis of Expression of Candidate Genes
Based on our experimental results, we randomly selected 8 DEGs-PIN9, ABCB11, SAUR36, SAUR50, IAA14, GH3.12, GA2ox1, and CKX5-to verify the reliability of the transcriptome data. The total RNA was extracted from tiller buds of the WT and the mtn1 mutant using an RNAprep Pure Plant Kit (Takara, Japan). Primers for the 8 DEGs were designed using Primer 5.0 software (Table S6). Quantitative real-time PCR (qRT-PCR) was carried out on the Bio-Rad CFX Connect PCR Detection System (Bio-Rad, Hercules, CA, USA) [53]. Each sample contained three biological replicates.

Analysis of Plant Hormones and Total Soluble Sugar Contents
IAA, GA, CTK and soluble sugar contents were measured in tiller buds (including nodes) of the WT and the mtn1 mutant. A total of 0.50 g of each sample was used to measure IAA, cytokinin (IPA) and gibberellin acid (GA 1 + GA 3 ) contents, as described previously [54,55]. IAA, IPA, GA 1 and GA 3 were measured with high-performance liquid chromatography-electrospray ionization tandem mass spectrometry (HPLC-ESI-MS/MS; Agilent 1290 and TripleQuad 6500 HPLC/MS system, Agilent Technologies Inc., Santa Clara, CA, USA). Hormones were extracted with acetonitrile solution and purified using the QUECH method. Then, the solvent was concentrated with a nitrogen purge. Samples were subjected to HPLC-ESI-MS/MS analysis. Soluble sugar contents were measured through anthrone colourimetry [56]. Each sample consisted of three biological replicates.

Influence of IAA, GA, CTK and Soluble Sugar on Tillering Number
Uniform stolons of the WT and the mtn1 mutant with the first three nodes were selected and planted in a bucket (20 cm in height and 15 cm in diameter). Each bucket contained 2.5 L of 1/2 Hoagland nutrient solution. The bucket was covered by a polyethylene plate with 14 spaced holes, and one stolon was planted in each hole. Starting one week after planting, the plant leaves were sprayed with 10 µM IAA, 25 µM GA 3 , 10 µM 6-benzylaminopurine (6-BA) and 1 µM sucrose once a week. Control plants were treated with distilled water. Buckets were placed in a greenhouse with a rain shelter. The nutrient solution was changed once a week. After three weeks, the tillering number was measured. Experiments were conducted following a completely randomised design with three replicates.

EoABCB11 Genetic Transformation
Based on the sequencing of the centipedegrass transcriptome, the EoABCB11-F/R primer pair (Table S7) was designed using the Primer 5.0 software, and then the coding sequence (CDS) was amplified. The CDS of EoABCB11 was first cloned into a pBI121 vector, and the vector was transformed into the Agrobacterium tumefaciens strain EHA105. The transgenic plants were produced via the floral dip method [57]. The primers used to detect the overexpression of transgenic plants were Bar-F/R (Table S7). Homozygous T 3 progeny were used for branch analysis. The branch numbers of the transgenic lines and the wild-type (each line contained five plants) were analysed.

Data Analysis
The morphological and hormone data are presented as the means ± standard errors. SPSS statistical software (version 16.0, IBM, Armonk NY, USA) and Origin 8.2 software (OriginLab, Northampton, MA, USA) were used to perform the statistical analyses. The variance (p < 0.05) of the data was analysed by one-way ANOVA (Sage Publications Inc., Newbury Park, CA, USA) (Duncan's test).

Conclusions
Overall, the mtn1 mutant exhibited high tillering, short internodes, long seeds and a heavy 1000-seed weight. The transcriptome results revealed that plant hormone signal transduction and starch and sucrose metabolism have interactive effects that contribute to the difference in tillering between the mtn1 mutant and the WT. ABCB transporters might be involved in auxin transport to regulate tillering development. The transgenic results showed that overexpressing EoABCB11 in Arabidopsis increases the branch number. The results of this study provide essential information for future studies of specific genes regulating tillering in centipedegrass and other perennial plants.