Genome-Wide Identification and Expression Analysis of the CLAVATA3/ESR-Related Gene Family in Tiger Nut

Abstract

CLAVATA3 (CLV3)/EMBRYO SURROUNDING REGION (ESR)-related (CLE) genes encode a group of peptide hormones, which coordinate cell proliferation and differentiation in plants. Tiger nut (Cyperus esculentus L.) is a perennial monocot plant that produces oil-rich tubers. However, the mechanisms regulating tuber development in tiger nut are poorly understood, and nothing is known about CLE genes in tiger nut. In this study, we identified 34 CLE genes in the genomes, proteomes, and transcriptomes of C. esculentus (CeCLE). We analyzed their gene structures and expression profiles in different parts of the plant, at three stages of tuber development and in roots in response to dehydration stress. We found a relatively high expression level of CeCLE13 in growing tuber and suggested that the corresponding CLE peptide could be involved in the regulation of tuberization. We also analyzed CeCLE gene sequences in the genome of the most productive K-17 variety in the N. I. Vavilov All-Russian Institute of Plant Genetic Resources collection and found many single nucleotide polymorphisms, insertions, and deletions. Our data provides fundamental information for future research on tiger nut growth and tuberization.

1. Introduction

Cyperus esculentus L., commonly known as tiger nut, yellow nutsedge, or chufa, is an herbaceous plant belonging to the Cyperaceae family. It is native to Mediterranean countries and Africa and is grown in various European and African countries. In Spain, it is cultivated on a large scale. The globular underground tubers of this plant are edible and serve as a valuable source of oil, starch, fiber, protein, vitamins, and minerals.

CLAVATA3/ESR-related (CLE) genes encode peptide hormones that control the division of stem cells in different types of plant meristems and play a role in symbiosis, parasitism, and responses to abiotic stress (reviewed by [1,2]). They act as mobile molecules that transmit information between neighboring cells or from the roots to the leaves. The CLE peptides are derived from prepropeptides and are 12–13 amino acids long. They undergo post-translational modifications and are targeted by a signal peptide to the apoplast, where they bind to and signal through their receptors. CLE genes are expressed in different tissues and are present in all land plants [3,4], as well as in plant-parasitic nematodes [5]. However, these peptides have not been previously studied in tiger nuts and have not been annotated, possibly due to the fact that many standard gene prediction pipelines miss short genes [3]. At the same time, they could participate in the regulation of tiger nut tuber formation via regulation of stem cell activity, as well as in root xylem and phloem development, and stomatal closure during water deficit.

Here, we have identified CLE genes in the genomes, proteomes, and transcriptomes of C. esculentus (CeCLE). We identified 34 CeCLE genes and analyzed their expression patterns based on available transcriptomic data. We found a relatively high expression level of the CeCLE13 gene in tubers and suggested that the corresponding CLE peptide could be involved in the regulation of tuberization due to the high similarity of its CLE domain to Arabidopsis CLEs, which are involved in the regulation of vascular meristem cell proliferation. We also sequenced the genome of the most productive K-17 variety in the VIR collection and found many single nucleotide polymorphisms, as well as insertions and deletions, which also affected the CeCLE genes.

2. Materials and Methods

2.1. Identification and Phylogenetic Analysis of CeCLE Genes

Transcriptome data of nine tiger nut tissues, including young leaf, mature leaf, leaf sheath, root, rhizome, stem apex, and tubers, were downloaded from the National Center for Biotechnology Information (NCBI) project PRJNA703731 [6]. Two cultivars of tiger nut from Shihezi University in Xinjiang (XJ) and Heshun County in Yunnan (YN) [6] were used. Reads quality control was performed with FastQC v. 0.11.9 [7]. Reads were filtered and aligned on genome of C. esculentus JiYouSha NO. 2 (China National GeneBank DataBase accession number CNA0051961) [8] using HISAT2 v. 2.0.5 [9] and counted with StringTie [10]. Transcripts were additionally assembled with StringTie, and a new merged annotation was produced. Candidate coding regions within transcript sequences from new annotation were identified by Transdecoder v. 5.5.0 (https://github.com/TransDecoder/TransDecoder; accessed on 28 June 2022) and borf (https://github.com/signalbash/borf; accessed on 1 May 2024). To identify CLE genes, BLASTP and TBLASTN analyses with the top 10 most frequently used CLE motifs in monocots [11] as queries were used against predicted proteins in transcriptomes of XJ and YN [6], proteomes [12], JiYouSha NO. 2 [8], and K-17 genomes. Signal peptides were predicted by SignalP 6.0 server [13]. The protein sequences were analyzed with the MEME program v. 5.5.7 (https://meme-suite.org/meme/tools/meme; accessed on 10 June 2024), and TBtools software v. 2.121 [14] was used to visualize CLE domains distribution (https://github.com/CJ-Chen/TBtools-II; accessed on 20 September 2024). Alignment of CLE domain amino acid sequences was performed by MEGA software (v. 11) using the muscle algorithm [15]. A phylogenetic tree with 1000 bootstrap replicates was constructed using the MEGA11 neighbor-joining method with default parameters [15].

2.2. Sample Collection, DNA Extraction, and Biochemical Analysis

Tubers for C. esculentus accession K-17 were obtained from the VIR (Saint-Petersburg, Russia) collection. Tubers were germinated in vermiculite under a 16 h photoperiod with day/night temperatures of 22 °C/22 °C. DNA was extracted from the leaves [16]. The biochemical analysis was performed using methods edited by Ermakov [17]. Starch content was measured using the Ewers polarimetric method. Two grams of the material were hydrolyzed in 25 mL of 1% hydrochloric acid solution for 15 min in boiling water and then cooled to room temperature. An amount of 2.5 mL of phosphotungstic acid was added to precipitate polysaccharides, proteins, etc. Distilled water was added up to 250 mL. After filtering, the extract was placed in a polarizing cuvette, and the rotation angle was measured on a SAC-I automatic polarimeter/saccharimeter (Saitama, Japan). The conversion coefficient was 203. The pigment absorption isolated with 100% acetone was measured at various wavelengths on an Ultrospec II spectrophotometer: 662 nm and 645 nm for chlorophylls a and b, respectively; 440 nm for carotenoids (followed by the calculation of pigment concentrations according to Wetstein and Holm levels); and 454 nm for carotenes and β-carotene.

2.3. Genome Sequencing and Variant Detection

The library was prepared with a NEB NEBNext^® Ultra™ II DNA Library Prep Kit for Illumina according to the manufacturer’s instructions. Sequencing was performed using an Illumina HiSeq2500 System in Rapid Run mode. The raw reads were checked by FastQC v. 0.11.9 [7]. Mitochondrion and chloroplast reads were removed by mapping the Illumina reads to the C. esculentus mitochondrion and chloroplast genomes (NC_058697.1 and NC_058698.1, respectively) using bbduk program from bbtools suite v. 37.23 (https://sourceforge.net/projects/bbmap/; accessed on 1 February 2017). Reads were mapped against the tiger nut JiYouSha NO. 2 reference genome [8] using BWA v. 0.7.17 [18]. The mapped bam file was used for variant calling using Freebayes v. 1.3.6 [19]. PCR duplicates were marked with samtools markdup [20]. The VCF file was filtered by vcffilter from vcflib v. 1.0.3. FEATnotator [21] was used for variant annotation and functional effect prediction.

2.4. RNA-Seq Data Analysis

Transcriptome data were downloaded from the NCBI projects PRJNA703731 and PRJNA821655. Two cultivars of tiger nut were used: XJ [4] and Jisha 1 [22]. Reads quality control, alignment, and counting were performed as described above. Differential gene expression analysis was performed using DeSeq2 R package (v. 1.40.2) [23]. For heatmap construction, the heatmap.2 function of the gplots R package (v. 3.1.3.1) [24] was used.

3. Results

3.1. CeCLE Genes Identification

For the identification of previously unknown CLE genes in C. esculentus, sequences of the top 10 most frequently used CLE motifs in monocots [11] were used as queries to perform BLASTP and TBLASTN searches against the predicted proteins in the transcriptomes [6], proteomes [12], and genomes of C. esculentus. Transcriptomes of two tiger nut cultivars (XJ and YN), proteomes of four stages of tuber development (freshly harvested, dried, rehydrated for 48 h, and sprouted), and two genomes (JiYouSha NO. 2, sequenced by [8], and K-17, sequenced in this study) were used. K-17 (VIR catalogue No. 17) tiger nut sample originated from Ivory Coast and is the most productive variety among the varieties in the VIR collection [25] (Figure 1).

In total, 34 CeCLE genes were identified (Figure 2; Document S1). All detected CeCLE genes lacked introns. Signal peptides were predicted for all CeCLE, except CeCLE4, CeCLE13, CeCLE20–22, and CeCLE29 (Figure 3). Several CeCLE proteins (CeCLE1, CeCLE2, CeCLE32, and CeCLE33) contained more than one CLE domain (Figure 3) that is quite common in plants [3]. The presence of multiple CLE domains within a single gene may enhance the efficiency of CLE peptide production.

In the K-17 genome, 7,465,786 single nucleotide polymorphisms, as well as 1,290,427 insertions and deletions, were found, which also affected the CeCLE genes. Nonsense mutations affected the CeCLE26, CeCLE32, and CeCLE33 genes and resulted in the loss of the CLE domain in the CeCLE26 gene or the loss of part of the CLE domains in multidomain CeCLE32 and CeCLE33 genes. Frameshifts associated with deletions or insertions in the CeCLE2, CeCLE5, CeCLE13, and CeCLE32 genes were also detected (Table S1).

3.2. CeCLE Genes Expression Analysis

To provide some clues on the roles of the CeCLE genes in tiger nut growth, expression profiles of these genes in different parts of the plant (young leaves, mature leaves, leaf sheaths, roots, rhizomes, stem apexes), at three stages of tuber development (40, 80, and 120 days after sowing), and in roots in response to dehydration stress were analyzed. The transcriptomic data was obtained from the NCBI database for XJ and Jisha 1 varieties of tiger nuts (Figure 4; Table S2) [6,22]. Two genes (CeCLE32 and CeCLE33) were not included in the heatmap as they had zero TPM values in all analyzed tissues. Six other CeCLEs (CeCLE11, CeCLE4, CeCLE14, CeCLE22, CeCLE23, and CeCLE15) showed high TPM levels in almost all analyzed tissues. Several CeCLE gene expressions were specific to certain tissues; for example, the CeCLE10 gene was expressed highly in roots, suggesting that its corresponding CLE peptide may be involved in regulating root growth. CeCLE13 is expressed in all tissues and has a CLE domain similar to that of the TDIF peptide (Figure 2), which plays a crucial role in controlling vascular meristem cell proliferation and differentiation [27]. CeCLE9 and CeCLE12 are expressed in young tubers and shoot apexes and have domains identical to those of AtCLE45 and AtCLE25, respectively (Figure 2), which regulate phloem differentiation [28,29]. We did not find differentially expressed CeCLE genes in the roots of plants grown under drought stress.

4. Discussion

Tubers are thickened parts of plants that serve as storage organs for nutrients. These organs help plants survive unfavorable conditions, such as cold weather and drought [30,31]. In agriculture, tuber crops like potato and tiger nuts are grown for their nutritional value and serve as an important source of food for both humans and animals. Regulation of tuber development in potato has been studied extensively, while there have been few studies on the development of tiger nut tubers. At the same time, tiger nut tubers contain large amounts of starch, oil, carbohydrates, vitamins C and E, flavonoids, and minerals and could be an important source of these nutrients in the human diet.

Interestingly, the anatomy of potato tubers and tiger nut tubers differs. In a potato tuber, the perimedullary zone occupies most of the tuber’s volume, while in tiger nut, it is the cortex and pith that occupy the most space [32]. It has also been found that auxins, jasmonic acids, brassinosteroids, G-proteins, MAPK signaling pathways, and ubiquitin protein pathways regulate cell division and expansion in tiger nut tubers [32]. Recent studies have emphasized the role of peptide hormones in coordinating plant growth, with CLE peptides being the most studied and characterized regulators of growth and development. Previously, we identified 41 CLE genes in potato Solanum tuberosum (StCLE) and analyzed their structure and expression patterns, finding a relatively high level of expression for StCLE8 and StCLE12 in stolons and tubers [33]. We found that StCLE8 and StCLE12 have an identical CLE domain to the Arabidopsis TDIF peptides, which are known to play a crucial role in vascular meristem maintenance [27]. Overexpression of StCLE8 promotes vascular cell proliferation and reduces tuber weight in potato [34]. In this study, we identified 34 CLE genes in tiger nut and analyzed their expression profiles in different organs. Some CeCLE genes (CeCLE4, CeCLE7, CeCLE9, CeCLE11, CeCLE12, CeCLE13, CeCLE14, CeCLE15, CeCLE20, CeCLE22, CeCLE23, CeCLE24, and CeCLE25) are expressed at relatively high levels in the stolon and in the growing tuber. Phylogenetic analysis revealed that the CeCLE13 peptide is more closely related to Arabidopsis TDIF peptides, suggesting that this CLE peptide could be involved in tiger nut tuber growth regulation. We also found that several CeCLE proteins (CeCLE1, CeCLE2, CeCLE32, and CeCLE33) contain more than one CLE domain, which could potentially increase the efficiency of CLE peptide production. However, the expression of genes encoding these peptides was not detected or was detected at very low levels in all tissues analyzed. Further experiments are needed to test whether such proteins are involved in the regulation of any processes in the plant and whether functional CLE peptides are formed from such multidomain sequences.

We also performed biochemical analysis and sequencing of the genome of the most productive tiger nut K-17 variety in the VIR collection. Analyzing the genome sequences of the K-17 and JiYouSha NO. 2, we identified a large number of single nucleotide polymorphisms, insertions, and deletions. Some of these changes affected the CeCLE genes and may be responsible for the high productivity of the K-17 tiger nut. The results obtained provide a basis for future genetic and genomic research on tiger nut growth and tuberization.