Accessibility / Report Error

Flower transcriptome dynamics during nectary development in pepper (Capsicum annuum L.)

Abstract

The measurement of gene expression can provide important information about gene function and the molecular basis for developmental processes. We analyzed the transcriptomes at three different developmental stages of pepper flower [sporogenous cell division, stage (B1); pollen mother cell meiosis, stage (B2); and open flower (B3)]. In the cDNA libraries for B1, B2, and B3: 82718, 77061, and 91491 unigenes were assembled, respectively. A total of 34,445 unigene sequences and 128 pathways were annotated by KEGG pathway analysis. Several genes associated with nectar biosynthesis and nectary development were identified, and 8,955, 12,182, and 23,667 DEGs were identified in the B2 vs B1, B3 vs B1, and B3 vs. B2 comparisons. DEGs were involved in various metabolic processes, including flower development, nectar biosynthesis, and nectary development. According to the RNA-seq data, all 13 selected DEGs showed similar expression patterns after q-PCR analysis. Sucrose-phosphatase, galactinol-sucrose galactosyltransferase, and sucrose synthase played very important roles in nectar biosynthesis, and CRABS CLAW could potentially be involved in mediating nectary development. A significant number of simple sequence repeat and single nucleotide polymorphism markers were predicted in the Capsicum annuum sequences. The new results provide valuable genetic information about flower development in pepper.

Keywords
Capsicum annuum L.; flower transcriptome; nectar biosynthesis; nectary development; gene

Introduction

Most flowering plants attract pollinators by offering a reward of floral nectar in many plant-pollinator systems. A floral organ called the nectary is responsible for nectar biosynthesis. However, the molecular events associated with the biosynthesis of nectar and nectary development are not clearly understood. To date, only a few individual genes, including BLADE-ON-PETIOLE (BOP) 1, BOP2, and CRABS CLAW (CRC) have been isolated and confirmed to be associated with the development of the nectary (Lee et al., 2005aLee JY, Baum SF, Alvarez J, Patel A, Chitwood DH and Bowman JL (2005a) Activation of CRABS CLAW in the nectaries and carpels of Arabidopsis. Plant Cell 17:25-36.; McKim et al., 2008McKim SM, Stenvik GE, Butenko MA, Kristiansen W, Cho SK, Hepworth SR, Aalen RB and Haughn GW (2008) The BLADE-ONPETIOLE genes are essential for abscission zone formation in Arabidopsis. Development 135:1537-1546.). Previous research has shown that crc knockout mutant lines failed to develop a nectary, whereas bop1/bop2 double mutant plants have significantly smaller nectaries along with aberrant morphologies (Bowman and Smyth, 1999Bowman JL and Smyth DR (1999) CRABS CLAW, a gene that regulates carpel and nectary development in Arabidopsis, encodes a novel protein with zinc finger and helix-loop-helix domains. Development 126:2387-2396.; McKim et al., 2008McKim SM, Stenvik GE, Butenko MA, Kristiansen W, Cho SK, Hepworth SR, Aalen RB and Haughn GW (2008) The BLADE-ONPETIOLE genes are essential for abscission zone formation in Arabidopsis. Development 135:1537-1546.). Furthermore, although CRC expression is essential, it cannot promote ectopic nectary development and previous results have indicated that some additional genetic elements might exist that restrict nectary development (Baum et al., 2001Baum SF, Eshed Y and Bowman JL (2001) The Arabidopsis nectary is an ABC-independent floral structure. Development 128:4657-4667.). Some other organ identity genes, including LEAFY, AGAMOUS (AG), SHATTERPROOF1/2, APETALA2/3 (AP2/3), PISTILLATA (PI), and SEPALLATA1/2/3 (SEP1/2/3), have demonstrated roles in regulating CRC expression (Baum et al., 2001Baum SF, Eshed Y and Bowman JL (2001) The Arabidopsis nectary is an ABC-independent floral structure. Development 128:4657-4667.; Lee et al., 2005bLee JY, Baum SF, Oh SH, Jiang CZ, Chen JC and Bowman JL (2005b) Recruitment of CRABS CLAW to promote nectary development within the eudicot clade. Development 132:5021-5032.). However, most of the genes that participate in de novo nectar production and development of the nectary have not been identified, which limits our understanding of the pathways and cellular processes critical for nectary development and function.

Pepper (Capsicum annuum L.) originated in Central and South America. It is one of the most important economic vegetables in the world and is consumed as a fruit or is processed into other products (Zou, 2002Zou XX (2002) China Capsicum. China Agricultural Press, Beijing.). Nectar is the primary reward offered by plants to attract pollinators. Pepper is a typical allogamy plant and has predominant heterosis. The use of nectaries to attract insect pollination has important potential when attempting to breed peppers using the cytoplasmic male sterile line as the female parent. However, nectar biosynthesis and nectary development in pepper flowers is not clearly understood. Xin et al. (2008)Xin H, Liu HZ, Wang HX and Zhao YJ (2008) Studies on developmental anatomy of floral nectaries of male-sterile-homomaintainer line in "hot pepper 95-1 . J Wuhan Bot Res 26:119-123. reported that the pepper floral nectary is at the base of the ovary and belongs to the ovarial nectary. The nectary is composed of nectariferous tissue and a secretory epidermis, which is covered by cuticle tissue.At the sporogenous cell division stage, the nectary has not yet developed and at the pollen mother cell stage, the nectary is forming, but nectar is not present. Colorless and transparent nectar is produced at the open flower stage.

Pepper nectar contains fructose, glucose, and sucrose. Therefore, the sugar metabolism pathway plays an important role in the biosynthesis of pepper nectar (Rabinowitch et al., 1993Rabinowitch HD, Fahn A, Meir T and Lensky Y (1993) Flower and nectar attributes of pepper (Capsicum annuum L.) plants in relation to their attractiveness to honeybees (Apis mellifera L.). Ann Appl Biol 123:221-232.; Roldán-Serrano and Guerra-Sanz, 2004Roldán-Serrano S and Guerra-Sanz JM (2004) Dynamics and sugar composition of sweet pepper (Capsicum annuum, L.) nectar. J Hortic Sci Biotech 79:717-722.; Greco et al., 2011Greco MK, Spooner-Hart RN, Beattie AGAC, Barchia I and Holford P (2011) Australian stingless bees improve greenhouse capsicum production. J Apic Res 50:102-115.; Pereira et al., 2015Pereira ALC, Taques TC, Valim JOS, Madureira AP and Campos WG (2015) The management of bee communities by intercropping with flowering basil (Ocimum basilicum) enhances pollination and yield of bell pepper (Capsicum annuum). JInsect Conserv 19:479-486.).

Next-generation sequencing technologies (e.g., RNA-seq) permit that the whole transcriptome can be sequenced, and they are convenient and rapid methods that can be used to investigate gene expression at the whole-genome level and define putative gene function (Ozsolak and Milos, 2011Ozsolak F and Milos PM (2011) RNA sequencing: advances, challenges and opportunities. Nat Rev Genet 12:87-98.; Jain, 2012Jain M (2012) Next-generation sequencing technologies for gene expression profiling in plants. Brief Funct Genomics 11:63-70.). Over the past few years, many studies have confirmed the efficiency and sensibility of RNA-seq in various biological contexts (González-Ballester et al., 2010González-Ballester D, Casero D, Cokus S, Pellegrini M, Merchant SS and Grossman AR (2010) RNA-seq analysis of sulfurdeprived Chlamydomonas cells reveals aspects of acclimation critical for cell survival. Plant Cell 2:2058-2084.; Li et al., 2010Li P, Ponnala L, Gandotra N, Wang L, Si Y, Tausta SL, Kebrom TH, Provart N, Patel R, Myers CR et al. (2010) The developmental dynamics of the maize leaf transcriptome. Nat Genet 42:1060-1067.; Zenoni et al., 2010Zenoni S, Ferrarini A, Giacomelli E, Xumerle L, Fasoli M, Malerba G, Bellin D, Pezzotti M and Delledonne M (2010) Characterization of transcriptional complexity during berry development in Vitis vinifera using RNA-Seq. Plant Physiol 152:1787-1795.; Yang et al., 2011Yang H, Lu P, Wang Y and Ma H (2011) The transcriptome landscape of Arabidopsis male meiocytes from high-throughput sequencing: the complexity and evolution of the meiotic process. Plant J 65:503-516.). Rapid progress has been made towards understanding the transcriptional programs associated with the specific development processes of many plant species, but only a few studies have investigated pepper (Ashrafi et al., 2012Ashrafi H, Hill T, Stoffel K, Kozik A, Yao J, Chin-Wo SR and Van Deynze A (2012) De novo assembly of the pepper transcriptome (Capsicum annuum): A benchmark for in silico discovery of SNPs, SSRs and candidate genes. BMC Genomics 30:571.; Liu et al., 2013Liu S, Li W, Wu Y, Chen C and Lei J (2013) De novo transcriptome assembly in chili pepper (Capsicum frutescens) to identify genes involved in the biosynthesis of capsaicinoids. PLoS One 8:e48156.; Martínez-López et al., 2014Martínez-López LA, Ochoa-Alejo N and Martínez O (2014) Dynamics of the chili pepper transcriptome during fruit development. BMC Genomics 21:143.).

To better understand the molecular mechanisms of nectar biosynthesis and nectary development in pepper, the sequencing data from the young pepper flowers obtained from RNA-seq was used to investigate differential gene expressions at three pepper flower development stages. We obtained high-quality reads and assembled unigenes that allowed us to identify the genes involved in pepper flower development, nectar biosynthesis, and nectary development. By comparing the expression patterns of genes at different stages of pepper flower development, we were able to identify several pathways and a large number of differentially expressed genes (DEGs). Some DEGs involved in nectar biosynthesis and nectary development were selected and analyzed by quantitative real-time PCR (qPCR). Our study provides valuable genetic information for the elucidation of pepper flower nectar biosynthesis and nectary development.

Materials and Methods

Plant material collection

The pepper cultivar 9704B was grown in experimental fields at the College of Horticulture and Landscape, Yunnan Agricultural University, China. Nectary development was divided into three stages (B1: sporogenous cell division, where the nectary has not yet developed; B2: pollen mother cell, where the nectary is forming, but no nectar is produced; and B3: open flower with nectar production). These stages were based on the cytology results reported by Xin et al. (2008)Xin H, Liu HZ, Wang HX and Zhao YJ (2008) Studies on developmental anatomy of floral nectaries of male-sterile-homomaintainer line in "hot pepper 95-1 . J Wuhan Bot Res 26:119-123.. The materials were frozen in liquid nitrogen and then stored at -80 °C until needed.

RNA extraction and library preparation for transcriptome analysis

Total RNA was isolated using the CTAB reagent method (Invitrogen). RNA from the B1, B2, and B3 stages of flower development was used to construct the sequence libraries. A NanoDrop 1000 spectrophotometer and an Agilent 2100 Bioanalyzer were used to verify RNA quality and quantity prior to further processing. The total RNA was treated with DNase I prior to library construction. Magnetic Oligo(dT) beads were used to purify the poly-(A) mRNA. Adaptor-ligated fragments were generated according to manufacturer's instructions for T4 DNA polymerase, T4 polynucleotide kinase, Klenow 3' to 5' exo-polymerase, and T4 DNA ligase. When the adaptor-ligated fragments were separated on a 1.0% agarose gel, the desired range of cDNA fragments (200 ± 25 bp) were excised from the gel and then the cDNA fragments were enriched and amplified using PCR. The cDNA library was subjected to Solexa sequencing using an Illumina HiSeq2000 sequencing platform (BGI-Shenzhen, Shenzhen, China).

De novo assembly, assessment, and annotation

The raw data processing and functional annotations were performed according to Zhang et al. (2015)Zhang GH, Ma CH, Zhang JJ, Chen JW, Tang QY and He MH (2015) Transcriptome analysis of Panax vietnamensis var. fuscidicus discovers putative ocotillol-type ginsenosides biosynthesis genes and genetic markers. BMC Genomics 16:159-172.. In brief, the raw sequencing data were transformed by base calling into sequence data, which were stored in fastq format. Adaptor fragments were removed from the raw reads to obtain the clean reads. De novo transcriptome assembly of these short reads was performed using the SOAP de novo assembling program, which first combined reads with a certain length of overlap to form longer fragments without Ncalled contigs. The reads were mapped back to the contigs. Paired-end reads were used to check the contigs that had come from the same transcript. Secondly, using "N" to represent unknown sequences, SOAP de novo connected the contigs to produce the scaffolds. Paired-end reads were then used again to fill in the gaps between the scaffolds. These were designated as unigenes.

For further analysis, we used BLASTX (E-value < 10-5) to search the unigene sequences against many protein databases, including the non-redundant database (Nr), SwissProt, the Kyoto Encyclopedia of Genes and Genomes (KEGG), and the Orthologoue Group of proteins (COG) (Natale et al., 2000Natale DA, Shankavaram UT, Galperin MY, Wolf YI, Aravind L and Koonin EV (2000) Towards understanding the first genome sequence of a crenarchaeon by genome annotation using clusters of orthologous groups of proteins (COGs). Genome Biol 1:RESEARCH0009.). Unigene sequences were not used to search the subsequent database(s) if they have hits in one of the named databases. BLAST results then used to extract the coding sequences (CDSs) from the unigene sequences and translate them into peptide sequences. The BLAST results were also used to train ESTScan. If the unigene CDSs had no BLAST hits, then they were predicted by ESTScan, and translated into peptide sequences. Unigene annotation provided information on the expression patterns and functional annotations of the identified genes. For Nr annotation, the BLAST2GO program was used to obtain the GO annotations for the unigenes (Conesa et al., 2005Conesa A, Gz S, García-Gez JM, Terol J, Tal M and Robles M (2005) Blast2GO: A universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics 21:3674-3676.), and WEGO software was used to perform the GO functional classifications for all the unigenes (Ye et al., 2006Ye J, Fang L, Zheng H, Zhang Y, Chen J, Zhang Z, Wang J, Li S, Li R, Bolund L et al. (2006) WEGO: a web tool for plotting GO annotations. Nucleic Acids Res 34:W293-297.). The WEGO software was also used to explore the macro-distribution of gene functions for this species.

Analysis of metabolic pathway genes identified from pepper flowers

The KEGG database and related software applications (http://www.genome.jp/kegg/kegg4.html) were used to analyze the metabolic pathways (Nakao et al., 1999Nakao M, Bono H, Kawashima S, Kamiya T, Sato K, Goto S and Kanehisa M (1999) Genome-scale gene expression analysis and pathway reconstruction in KEGG. Genome Inform Ser Workshop Genome Inform 10:94-103.; Kanehisa et al., 2008Kanehisa M, Araki M, Goto S, Hattori M, Hirakawa M, Itoh M, Katayama T, Kawashima S, Okuda S, Tokimatsu T et al. (2008) KEGG for linking genomes to life and the environment. Nucleic Acids Res 36:480-484.). The variations that were specific to particular organisms and information about the networks of molecular interactions within cells were obtained from the PATHWAY database. The genes involved in flower development, nectar biosynthesis, and nectary development were selected and analyzed using BLAST annotation of KEGG and the other databases mentioned above.

Differential gene expression analysis

The KEGG database and related software applications (http://www.genome.jp/kegg/kegg4.html) were used to analyze the metabolic pathways (Nakao et al., 1999Nakao M, Bono H, Kawashima S, Kamiya T, Sato K, Goto S and Kanehisa M (1999) Genome-scale gene expression analysis and pathway reconstruction in KEGG. Genome Inform Ser Workshop Genome Inform 10:94-103.; Kanehisa et al., 2008Kanehisa M, Araki M, Goto S, Hattori M, Hirakawa M, Itoh M, Katayama T, Kawashima S, Okuda S, Tokimatsu T et al. (2008) KEGG for linking genomes to life and the environment. Nucleic Acids Res 36:480-484.). Transcript expression was calculated by the RPKM method using the following formula: RPKM(A) = 106C / (N × L / 103), where RPKM(A) is the expression of unigene A, C is the number of fragments uniquely aligned to unigene A, N is the total number of fragments uniquely aligned to all the unigenes, and L is the base number in the coding region (CDS) of unigene A. The p-value corresponding to the differential transcript expression in two samples was determined according to Audic and Claverie (1997)Audic S and Claverie JM (1997) The significance of digital gene expression profiles. Genome Res 7:986-995. and the FDR method was used to determine the threshold p-values in a number of tests. We used FDR ≤ 0.001 and the absolute value of log2 ratio ≥ 1 as thresholds to consider whether the gene expression difference was significant.

Gene validation and expression analysis

For the qPCR analysis we selected 13 unigenes that had potential roles in nectar biosynthesis and nectary development for validation. Specific primers for qPCR were designed using Primer Premier 5.0 (Table 1). Total RNA was extracted individually from the B1, B2, and B3 developmental stages. Total RNA and first-strand cDNA synthesis of the samples were carried out according to Lv et al. (2016)Lv JH, Zhao K, Huo JL and Deng MH (2016) Molecular cloning sequence characterization of a novel pepper gene CaNDPK and its effect on cytoplasmic male sterility. Res J Biotech 11:36-41..

Table 1
Primer pairs for qRT-PCR.

Phylogenetic analysis

Alignment of the nucleotide sequences was computed using ClusterX, and the phylogenetic trees were created using the ClustalX and Mega 4.0 software packages with standard parameters.

Putative molecular markers

Potential simple sequence repeat (SSR) markers were detected using the MISA Perl script (http://pgrc.ipkgatersleben.de/misa/). Mono-, di-, tri-, tetra-, penta- and hexa-nucleotide sequences with minimum repeat numbers of 10, 6, 5, 5, 5, and 5, respectively, were used as the search criteria. SOAPaligner software (Release 2.21.08-13-2009) was used to mine for the single nucleotide polymorphism (SNP) markers. The thresholds for SNP identification were carried out according to Liu et al. (2013)Liu S, Li W, Wu Y, Chen C and Lei J (2013) De novo transcriptome assembly in chili pepper (Capsicum frutescens) to identify genes involved in the biosynthesis of capsaicinoids. PLoS One 8:e48156..

Statistical analysis

Three biological replicates were generated for each developmental stage (each sample containing flowers from 10 individual plants), with technical triplicates for each sample. An actin gene was chosen as internal control for normalization, and relative gene expression was calculated using the 2-ΔΔCt method.

Results

Flower transcriptome sequencing output and de novo assembly

By Illumina sequencing 57,826,112, 59,057,626, and 59,459,964 raw reads were generated for the B1, B2, and B3 development stages, respectively. After filtering of low-complexity reads, low-quality reads, and repetitive reads, more than 53 million 49nt clean reads were obtained for the B1, B2, and B3 cDNA libraries, respectively (Table 2).

Table 2
Output statistics for pepper flower cDNA libraries.

The Trinity program was used for de novo assembly of the high-quality reads. The number of contigs and information about the pepper flower contigs is shown in Table 3. A total of 82,718, 77,061, and 91,491 unigenes, respectively, were assembled, with an average unigene length of 704, 657, and 716 nt. Details about the pepper unigenes are shown in Table 3.

Table 3
Statistics of assembly quality for pepper flower.

The BLASTX program (E-value < 10-5) was used to obtain 59,285 significant BLAST hits. The size distribution for the CDSs is shown in Figure S1 Figure S1 - Size distribution of CDSs produced by searching unigene sequences against various protein databases (Nr, SwissProt, KEGG, and COG, in order) using BLASTX (E-value 10-5). . When the CDSs of the unigenes did not result in any BLAST hits, they were predicted using ESTScan (Iseli et al., 1999Iseli C, Jongeneel CV and Bucher P (1999) ESTScan: A program for detecting, evaluating, and reconstructing potential coding regions in EST sequences. Proc Int Conf Intell Syst Mol Biol 1999:138-148.). A total of 2303 unigenes were analyzed using this method (the size distribution of the ESTs is shown in Figure S2 Figure S2 - Size distribution of ESTs obtained from the ESTScan results. ).

Functional annotation

BLASTX was used to search for distinct gene sequences against the Nr database. the hit rate of 61,365 unigenes exceeded the E-value cutoff. Similarly, 37,039 unigenes were identified in the SwissProt database. A total of 76,833 unigenes were annotated using one or more of the databases, suggesting they had relatively well conserved functions.

The unigenes were searched against the COG database in order to predict and classify possible functions. Out of 61,365 Nr hits, 41,683 sequences had COG classifications that were distributed across 25 COG categories (Figure S3 Figure S3 - COG function classification of the unigenes. ). Among the 25 COG categories, "General function prediction only" represented the largest group (7196, 17.3%), followed by "Replication, recombination and repair" (4008, 9.6%), "Transcription" (3806, 9.1%), and "Signal transduction mechanisms" (3132, 7.5%). The smallest groups were "Nuclear structure" (4 unigenes) and "Extracellular structures" (15 unigenes).

BLAST2GO was used to assign 33,8092 unigenes and summarized the terms into three main GO categories and 55 sub-categories (functional groups) (Figure S4 Figure S4 - GO function classification of the unigenes. ). In each of the three main categories of the GO classification system (Biological process, Cellular component, and Molecular function), the dominant terms were "Cellular process", "Metabolic process", "Cell", "Cell part", "Organelle", "Binding", and "Catalytic activity". About half of the genes were in the Biological process category.

The annotated sequences were mapped to the reference canonical pathways contained in the KEGG database. In total, 34,445 unigenes were assigned to 128 KEGG pathways. The most highly represented category was "Metabolic pathways" with 7658 (22.23%) members. The "Biosynthesis of secondary metabolites" and "Plant-pathogen interaction" pathways were also well represented, with 4110 (11.93%) members and 1969 (5.72%) members, respectively.

Statistical analysis of the differential expression genes (DEGs) during flower development

RNA samples from the B1, B2, and B3 stages were used as the libraries for nectar production and nectary development. Pairs of the three libraries (B2 vs. B1, B3 vs. B1, and B3 vs. B2) were compared. For each different stage, we identified thousands of DEGs, indicating that there had been substantial changes. The B2 vs. B1 comparison showed 8,955 DEGs, with 2666 up-regulated and 6289 down-regulated (Figure 1). In the B3 vs. B1 comparison, 12,182 DEGs were found, of which 7454 were up-regulated and 4728 down-regulated (Figure 1). A total of 17,246 DEGs were up-regulated and 6421 were down-regulated in the B3 vs. B2 comparison (Figure 1). Hence, the number of DEGs in the B2 vs. B3 comparison was the largest, and the number of B2 vs. B1 differences were the lowest in the three pair comparisons.

Next, we mapped the DEGs to the KEGG databases and compared them to our transcriptome data. A number of genes were significantly enriched at the B3 stage. The B2 vs. B1 comparison showed that the most highly represented category was "Metabolic pathways" with 795 (21.75%) members. The "Biosynthesis of secondary metabolites", "RNA transport", and the "Plant-pathogen interaction" pathways were also well represented, with 495 (12.54%) members, 255 (6.97%) members, and 239 (6.53%) members, respectively (Figure S5 Figure S5 - Histogram of the GO classifications for genes that were differentially expressed between B2 and B1. ). In the B3 vs. B1 comparison, the most highly represented category was "Metabolic pathways", with 1180 (25.11%) members. The "Biosynthesis of secondary metabolites" and "Plant-pathogen interaction" pathways were also well represented, with 693 (14.75%) members and 370 (7.87%) members, respectively (Figure S6 Figure S6 - Histogram of the GO classifications for genes that were differentially expressed between B3 and B1. ). In the B3 vs. B2 comparison, the most highly represented category was "Metabolic pathways", with 1872 (23.45%) members. The "Biosynthesis of secondary metabolites" and "Plant-pathogen interaction" pathways were also well represented, with 1069 (13.39%) members and 583 (7.3%) members, respectively (Figure S7 Figure S7 - Histogram of the G O classifications for genes that were differentially expressed between B3 and B2 ).

Figure 1
Differentially expressed genes (DGEs) between different developmental stages of flowers.

Analysis of nectar biosynthesis associated genes and qPCR validation

Sugars are the principal solutes in most nectars. Genes involved in sugar metabolism were expected to be well represented within the flower transcriptome, which was the case in this study. A total of 18 sugar metabolizing and modifying genes were identified from the flower transcriptome (Table S1 Table S1 - Sugar metabolism unigenes derived from the flower transcriptomes ), and 17 of those unigenes belonged to four genes families, which were sucrose-phosphatase 2 (CaSPP2) (four unigenes), alkaline alpha-galactosidase (CaAGA) (four unigenes), sucrose-phosphatase 1 (CaSPP1) (three unigenes), and sucrose synthase 7-like (CaSUS7) (six unigenes). The last unigene was galactinol-sucrose galactosyltransferase 5-like (CaGSG5). The B2 vs. B1 libraries comparison showed that seven unigenes were up-regulated and eight were down-regulated (Table S2 Table S2 - Sugar metabolism unigenes expression comparison between stages B2 and B1. ). In the B3 vs. B1 comparison, 15 unigenes were up-regulated and one was down-regulated (Table S3 Table S3 - Sugar metabolism unigenes expression comparison between stages B3 and B1. ), whereas 13 unigenes were up-regulated and no unigenes were down-regulated in the B3 vs B2 comparison (Table S4 Table S4 - Sugar metabolism unigenes expression comparison between stages B3 and B2. ).

In the sugar metabolism category, five unigenes were investigated further. In general, the identification of the genes with different expressions that are putatively associated with sucrose biosynthesis are summarized in Figure 2. For all the selected unigenes, most of their expressions continuously increased and peaked at the nectar production stage (B3).

Figure 2
Real time PCR analysis of selected genes involved in sucrose biosysnthesis. 1: Unigene14855_All; 2: CL2479.Contig12_All; 3: CL444.Contig2_All; 4: CL2479.Contig7_All; 5: CL1440.Contig1_All.

Analysis of nectary development association genes and qPCR validation

Forty seven known Arabidopsis nectary-enriched unigenes were identified from the pepper flower transcriptomes (Table S5 Table S5 - Nectary enriched unigenes derived from the flower transcriptomes. ). Two belonged to the transcription factor CRC, two were cupin family proteins (CUPIN), eight unigenes were part of putative beta-fructosidase (βFRA), one unigene belonged to agamous-like MADS box protein AGL5, and the other unigenes belonged to the L-ascorbate oxidase (AO). The B2 vs. B1 library comparison showed that 20 unigenes were up-regulated and eight were down-regulated (Table S6 Table S6 - Nectary-enriched unigenes expression comparison between stages B2 and B1. ). In the B3 vs. B1 comparison, 26 unigenes were up-regulated and four were down-regulated (Table S7 Table S7 - Nectary-enriched unigenes expression comparison between stages B3 and B1. ); and in the B3 vs. B2 comparison, 35 unigenes were up-regulated and three were down-regulated (Table S8 Table S8 - Nectary-enriched unigenes expression comparison between stages B3 and B2. ).

Some known Arabidopsis nectary-enriched genes were used to select eight unigenes for qPCR analysis. Unigene16506 belongs to the cupin family protein group, and was named CaCUPIN. It had a lower expression in B1, was up-regulated in B2, and then down-regulated in B3 samples. CL4573.Contig1 (beta-fructosidase, named CaβFRA), CL7013.Contig2 (transcription factor CRC, named CaCRC), Unigene21647, and CL3588.Contig1 and CL2403.Contig1 (L-ascorbate oxidase, named CaAO1, CaAO2, and CaAO3) were up-regulated in their expression at the flower development stage (B3). The CL2191.Contig2 (beta-fructofuranosidase, named CaβFFA) and CL4219. Contig2 (agamous-like MADS box protein AGL5, named CaAGL5) displayed similar expression patterns, and were up-regulated at the nectary forming stage (B2) (Figure 3).

Figure 3
Real time PCR analysis of selected genes involved in nectary development. I: Unigene16506_All; II: CL4573.Contig1_All; III: CL2191. Contig2_All; IV: CL7013.Contig2_All; V: CL4219.Contig2_All; VI: Unigene21647_All; VII: CL3588.Contig1_All; VIII: CL4586.Contig1_All.

Putative molecular markers

A total of 11,654 SSRs were identified in 97,475 unigenes. Mono-nucleotide SSRs represented the largest fraction (35.9%; 4180), followed by tri-nucleotides (34.7%; 4044), di-nucleotides (24.0%; 2802), hexa-nucleotide tandem repeats (2.3%; 271), penta-nucleotides (0.18%; 207), and tetra-nucleotides (0.13%; 150) (Table 4). Additionally, a total of 17,068, 14,407, and 20,350 putative SNPs were detected in the B1, B2, and B3 libraries, respectively (Table 5).

Table 4
Distribution of SSRs identified using MISA software.
Table 5
Distribution of SNPs identified using SOAPaligner software.

Discussion

Global gene expression patterns in pepper flowers

High-throughput mRNA sequencing technology is highly suitable for gene expression profiling in non-model organisms. Before this study, most pepper sequence studies were based on EST sequencing, very few tags had been reported in public databases, and there was little available genetic or genomic information. This study used RNA-Seq technology on the Illumina HiSeqTM 2000 platform to profile the pepper flower transcriptomes. We obtained 53,911,440, 55,408,880, and 55,241,582 clean reads, and identified 82,718, 77,061 and 91,491 unigenes from de novo assembly in the B1, B2, and B3 libraries, respectively. The gene or protein names and descriptions were assessed, and their putative conserved domains, gene ontology terms, and potential metabolic pathways were annotated. This study will help to improve our understanding of the processes involved in regulating flower development, nectar biosynthesis, and nectary development in pepper flowers. More contigs and unigenes were reported compared to previous transcriptomic studies in other plants, such as tomato (Pandurangaiah et al., 2016Pandurangaiah S, Ravishankar KV, Shivashankar KS, Sadashiva AT, Pillakenchappa K and Narayanan SK (2016) Differential expression of carotenoid biosynthetic pathway genes in two contrasting tomato genotypes for lycopene content. J Biosci 41:257-264.), potato (Cao et al., 2016Cao Q, Gao S, Li A, Chen J, Sun Y, Tang J, Zhang A, Zhou Z, Zhao D, Ma D et al. (2016) Transcriptome sequencing of the sweet potato progenitor (Ipomoea Trifida (H.B.K.) G. Don.) and discovery of drought tolerance genes. Trop Plant Biol 9:63-72.), and eggplant (Ramesh et al., 2016Ramesh KR, Hemalatha R, Vijayendra CA, Arshi UZ, Dushyant SB and Dinesh KB (2016) Transcriptome analysis of Sola-num melongena L. (eggplant) fruit to identify putative allergens and their epitopes. Gene 15:64-71.), which indicated that pepper contains abundant gene resources. We believe that our data will help to provide important new insights and facilitate further studies on pepper genes and their functions.

Nectar biosynthesis association genes

As we analyzed progressive flowering stages, not surprisingly a large number of genes associated with sugar metabolism and processing were found differentially expressed. This is in agreement with the hypothesis that simple sugars are the principal solutes in most nectars (Davis et al., 1998Davis AR, Pylatuik JD, Paradis JC and Low NH (1998) Nectar-carbohydrate production and composition vary in relation to nectary anatomy and location within individual flowers of several species of Brassicaceae. Planta 205:305-318.; Baum et al., 2001Baum SF, Eshed Y and Bowman JL (2001) The Arabidopsis nectary is an ABC-independent floral structure. Development 128:4657-4667.; Pacini et al., 2007Pacini E and Nepi M (2007) Nectar production and presentation. In: Nicolson SW, Nepi M and Pacini E (eds) Nectaries and nectar. Springer Netherlands, Dordrecht, pp 167-214.; Ren et al., 2007Ren G, Healy RA, Klyne AM, Horner HT, James MG and Thornburg RW (2007) Transient starch metabolism in ornamental tobacco floral nectaries regulates nectar composition and release. Plant Sci 173:277-290.). Sugar modifying enzymes and sugar transporters control sugar transport and metabolism in plant cells and tissues. They also play key roles in establishing and maintaining sugar concentrations across membranes (Roitsch, 1999Roitsch T (1999) Source-sink regulation by sugar and stress. Curr Opin Plant Biol 2:198-206.). For example, sucrose synthase (SUS, EC 2.4.1.13) is a known glycosyltransferase in plants. The enzyme catalyzes the reversible transfer of a glucosyl moiety between fructose and a nucleoside diphosphate (NDP) (NDP-glucose + fructose ↔ sucrose + NDP) (Diricks et al., 2015Diricks M, De Bruyn F, Van Daele P, Walmagh M and Desmet T (2015) Identification of sucrose synthase in nonphotosynthetic bacteria and characterization of the recombinant enzymes. Appl Microbiol Biotechnol 99:8465-8474.). In this study, CaSUS7 (sucrose synthase 7-like family proteins: CL1440.Contig1, CL1440.Contig2, CL1440.Contig3, CL1440.Contig7, CL1440.Contig8, and CL1440.Contig18) were highly homologous to Nicotiana tabacum (XM_016585183), Solanum pennellii (XM_015210288), S. tuberosum (XM_006348118), Ipomoea nil (XM_019321835), Beta vulgaris (XM_010676936), Phoenix dactylifera (XM_008806164), Jatropha curcas (XM_012220325), and Sesamum indicum (XM_011088670) (Figure 4) which strongly suggests that they all might encode SUS7 proteins that catalyze the reversible reaction of fructose to a nucleoside diphosphate. SUS has been shown to play a crucial role in nectar production (Kram et al., 2009Kram BW, Xu WW and Carter CJ (2009) Uncovering the Arabidopsis thaliana nectary transcriptome: Investigation of differential gene expression in floral nectariferous tissues. BMC Plant Biol 9:92-107.). All of the genes associated with CaSUS were strongly up-regulated, which indicates that CaSUS may play a crucial role in pepper nectar production.

Figure 4
Phylogenetic tree of sucrose synthase 7. Nicotiana tabacum: XM_016585183; Solanum pennellii: XM_015210288; Solanum tuberosum: XM_006348118; Ipomoea nil: XM_019321835; Beta vulgaris: XM_010676936; Phoenix dactylifera: XM_008806164; Jatropha curcas: XM_012220325; Sesamum indicum: XM_011088670

In addition to CaSUS, we identified a number of other genes that were up-regulated at the flower development and were involved in simple sugar metabolism. These were CaGSG5 (Unigene14855), 4 CaS6PP (CL2479.Contig12, CL2479.Contig11, CL2479.Contig5, and CL2479.Contig2), 3 CaSPP1 (CL2479.Contig1, CL2479.Contig6, and CL2479.Contig7), 4 CaGSG1 (CL444.Contig4, CL444.Contig1, CL444.Contig2, and CL444.Contig5). Significantly, the TAIR AraCyc databaseTAIR AraCyc database, http://www.arabidopsis.org/biocyc/index.jsp.
http://www.arabidopsis.org/biocyc/index....
(http://www.arabidopsis.org/biocyc/index.jsp) suggested that these genes can be tentatively assigned functions in sucrose metabolism (synthesis/degradation). These results showed that the canonical sucrose biosynthesis pathway was represented by genes that were differentially expressed within the flower at the nectar producing stage (Figure S5 Figure S5 - Histogram of the GO classifications for genes that were differentially expressed between B2 and B1. ).

Nectary development association genes

Transcription process genes were also highly represented within nectary expressed genes. Forty-seven unigenes showed open flower (nectary-enriched) expression profiles (Table-S5 Table S5 - Nectary enriched unigenes derived from the flower transcriptomes. ). There were six members of the YABBY transcription factor gene families in Arabidopsis (CRC, FILAMENTOUS FLOWER, YABBY3, INNER NO OUTER, YABBY2, and YABBY5) and they are determinants of abaxial cell fate in the lateral floral organs (Siegfried et al., 1999Siegfried KR, Eshed Y, Baum SF, Otsuga D, Drews GN and Bowman JL (1999) Members of the YABBY gene family specify abaxial cell fate in Arabidopsis. Development 126:4117-4128.). CRABS CLAW (At1g69180, CRC) encodes a transcription factor associated with the regulation of carpel and nectary development (Alvarez and Smyth, 1999Alvarez J and Smyth DR (1999) CRABS CLAW and SPATULA, two Arabidopsis genes that control carpel development in parallel with AGAMOUS. Development 126:2377-2386.). CRC is currently the only known gene to be absolutely required for nectary development (Lee et al., 2005bLee JY, Baum SF, Oh SH, Jiang CZ, Chen JC and Bowman JL (2005b) Recruitment of CRABS CLAW to promote nectary development within the eudicot clade. Development 132:5021-5032.; Alvarez and Smyth, 1999Alvarez J and Smyth DR (1999) CRABS CLAW and SPATULA, two Arabidopsis genes that control carpel development in parallel with AGAMOUS. Development 126:2377-2386.). This study identified two CaCRCs that were highly homologous to N. tabacum (AY854799), S. tuberosum (XM_006348604), Malus x domestica (XM_008391173), Prunus mume (XM_008245598), Gossypium raimondii (XM_012625767), Theobroma cacao (XM_018114782), and A. thaliana At1g69180 (BT008618, DQ446412) (Figure 5). Both were differentially expressed at the pepper nectary forming stage, which suggests that they may be involved in nectary development or function.

Figure 5
Phylogenetic tree of CRC. Arabidopsis thaliana(1): BT008618; Arabidopsis thaliana(2): DQ446412; Gossypium raimondii: XM_012625767; Malus x domestica: XM_008391173; Nicotiana tabacum: AY854799; Prunus mume: XM_008245598; Solanum tuberosum: XM_006348604; Theobroma cacao: XM_018114782

Acknowledgments

This study was supported by the Natural Science Foundation of China (Grant No. 31560556, 31760575), the Natural Science Foundation of Yunnan Province (Grant No. 2013FB038), and the New Products Program of Yunnan Province (Grant no. 2018BB020).

References

  • Alvarez J and Smyth DR (1999) CRABS CLAW and SPATULA, two Arabidopsis genes that control carpel development in parallel with AGAMOUS. Development 126:2377-2386.
  • Ashrafi H, Hill T, Stoffel K, Kozik A, Yao J, Chin-Wo SR and Van Deynze A (2012) De novo assembly of the pepper transcriptome (Capsicum annuum): A benchmark for in silico discovery of SNPs, SSRs and candidate genes. BMC Genomics 30:571.
  • Audic S and Claverie JM (1997) The significance of digital gene expression profiles. Genome Res 7:986-995.
  • Baum SF, Eshed Y and Bowman JL (2001) The Arabidopsis nectary is an ABC-independent floral structure. Development 128:4657-4667.
  • Bowman JL and Smyth DR (1999) CRABS CLAW, a gene that regulates carpel and nectary development in Arabidopsis, encodes a novel protein with zinc finger and helix-loop-helix domains. Development 126:2387-2396.
  • Cao Q, Gao S, Li A, Chen J, Sun Y, Tang J, Zhang A, Zhou Z, Zhao D, Ma D et al. (2016) Transcriptome sequencing of the sweet potato progenitor (Ipomoea Trifida (H.B.K.) G. Don.) and discovery of drought tolerance genes. Trop Plant Biol 9:63-72.
  • Conesa A, Gz S, García-Gez JM, Terol J, Tal M and Robles M (2005) Blast2GO: A universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics 21:3674-3676.
  • Davis AR, Pylatuik JD, Paradis JC and Low NH (1998) Nectar-carbohydrate production and composition vary in relation to nectary anatomy and location within individual flowers of several species of Brassicaceae. Planta 205:305-318.
  • Diricks M, De Bruyn F, Van Daele P, Walmagh M and Desmet T (2015) Identification of sucrose synthase in nonphotosynthetic bacteria and characterization of the recombinant enzymes. Appl Microbiol Biotechnol 99:8465-8474.
  • González-Ballester D, Casero D, Cokus S, Pellegrini M, Merchant SS and Grossman AR (2010) RNA-seq analysis of sulfurdeprived Chlamydomonas cells reveals aspects of acclimation critical for cell survival. Plant Cell 2:2058-2084.
  • Greco MK, Spooner-Hart RN, Beattie AGAC, Barchia I and Holford P (2011) Australian stingless bees improve greenhouse capsicum production. J Apic Res 50:102-115.
  • Iseli C, Jongeneel CV and Bucher P (1999) ESTScan: A program for detecting, evaluating, and reconstructing potential coding regions in EST sequences. Proc Int Conf Intell Syst Mol Biol 1999:138-148.
  • Jain M (2012) Next-generation sequencing technologies for gene expression profiling in plants. Brief Funct Genomics 11:63-70.
  • Kanehisa M, Araki M, Goto S, Hattori M, Hirakawa M, Itoh M, Katayama T, Kawashima S, Okuda S, Tokimatsu T et al. (2008) KEGG for linking genomes to life and the environment. Nucleic Acids Res 36:480-484.
  • Kram BW, Xu WW and Carter CJ (2009) Uncovering the Arabidopsis thaliana nectary transcriptome: Investigation of differential gene expression in floral nectariferous tissues. BMC Plant Biol 9:92-107.
  • Lee JY, Baum SF, Alvarez J, Patel A, Chitwood DH and Bowman JL (2005a) Activation of CRABS CLAW in the nectaries and carpels of Arabidopsis. Plant Cell 17:25-36.
  • Lee JY, Baum SF, Oh SH, Jiang CZ, Chen JC and Bowman JL (2005b) Recruitment of CRABS CLAW to promote nectary development within the eudicot clade. Development 132:5021-5032.
  • Li P, Ponnala L, Gandotra N, Wang L, Si Y, Tausta SL, Kebrom TH, Provart N, Patel R, Myers CR et al. (2010) The developmental dynamics of the maize leaf transcriptome. Nat Genet 42:1060-1067.
  • Liu S, Li W, Wu Y, Chen C and Lei J (2013) De novo transcriptome assembly in chili pepper (Capsicum frutescens) to identify genes involved in the biosynthesis of capsaicinoids. PLoS One 8:e48156.
  • Lv JH, Zhao K, Huo JL and Deng MH (2016) Molecular cloning sequence characterization of a novel pepper gene CaNDPK and its effect on cytoplasmic male sterility. Res J Biotech 11:36-41.
  • Martínez-López LA, Ochoa-Alejo N and Martínez O (2014) Dynamics of the chili pepper transcriptome during fruit development. BMC Genomics 21:143.
  • McKim SM, Stenvik GE, Butenko MA, Kristiansen W, Cho SK, Hepworth SR, Aalen RB and Haughn GW (2008) The BLADE-ONPETIOLE genes are essential for abscission zone formation in Arabidopsis. Development 135:1537-1546.
  • Nakao M, Bono H, Kawashima S, Kamiya T, Sato K, Goto S and Kanehisa M (1999) Genome-scale gene expression analysis and pathway reconstruction in KEGG. Genome Inform Ser Workshop Genome Inform 10:94-103.
  • Natale DA, Shankavaram UT, Galperin MY, Wolf YI, Aravind L and Koonin EV (2000) Towards understanding the first genome sequence of a crenarchaeon by genome annotation using clusters of orthologous groups of proteins (COGs). Genome Biol 1:RESEARCH0009.
  • Ozsolak F and Milos PM (2011) RNA sequencing: advances, challenges and opportunities. Nat Rev Genet 12:87-98.
  • Pacini E and Nepi M (2007) Nectar production and presentation. In: Nicolson SW, Nepi M and Pacini E (eds) Nectaries and nectar. Springer Netherlands, Dordrecht, pp 167-214.
  • Pandurangaiah S, Ravishankar KV, Shivashankar KS, Sadashiva AT, Pillakenchappa K and Narayanan SK (2016) Differential expression of carotenoid biosynthetic pathway genes in two contrasting tomato genotypes for lycopene content. J Biosci 41:257-264.
  • Pereira ALC, Taques TC, Valim JOS, Madureira AP and Campos WG (2015) The management of bee communities by intercropping with flowering basil (Ocimum basilicum) enhances pollination and yield of bell pepper (Capsicum annuum). JInsect Conserv 19:479-486.
  • Rabinowitch HD, Fahn A, Meir T and Lensky Y (1993) Flower and nectar attributes of pepper (Capsicum annuum L.) plants in relation to their attractiveness to honeybees (Apis mellifera L.). Ann Appl Biol 123:221-232.
  • Ramesh KR, Hemalatha R, Vijayendra CA, Arshi UZ, Dushyant SB and Dinesh KB (2016) Transcriptome analysis of Sola-num melongena L. (eggplant) fruit to identify putative allergens and their epitopes. Gene 15:64-71.
  • Ren G, Healy RA, Klyne AM, Horner HT, James MG and Thornburg RW (2007) Transient starch metabolism in ornamental tobacco floral nectaries regulates nectar composition and release. Plant Sci 173:277-290.
  • Roldán-Serrano S and Guerra-Sanz JM (2004) Dynamics and sugar composition of sweet pepper (Capsicum annuum, L.) nectar. J Hortic Sci Biotech 79:717-722.
  • Roitsch T (1999) Source-sink regulation by sugar and stress. Curr Opin Plant Biol 2:198-206.
  • Siegfried KR, Eshed Y, Baum SF, Otsuga D, Drews GN and Bowman JL (1999) Members of the YABBY gene family specify abaxial cell fate in Arabidopsis. Development 126:4117-4128.
  • Xin H, Liu HZ, Wang HX and Zhao YJ (2008) Studies on developmental anatomy of floral nectaries of male-sterile-homomaintainer line in "hot pepper 95-1 . J Wuhan Bot Res 26:119-123.
  • Yang H, Lu P, Wang Y and Ma H (2011) The transcriptome landscape of Arabidopsis male meiocytes from high-throughput sequencing: the complexity and evolution of the meiotic process. Plant J 65:503-516.
  • Ye J, Fang L, Zheng H, Zhang Y, Chen J, Zhang Z, Wang J, Li S, Li R, Bolund L et al. (2006) WEGO: a web tool for plotting GO annotations. Nucleic Acids Res 34:W293-297.
  • Zenoni S, Ferrarini A, Giacomelli E, Xumerle L, Fasoli M, Malerba G, Bellin D, Pezzotti M and Delledonne M (2010) Characterization of transcriptional complexity during berry development in Vitis vinifera using RNA-Seq. Plant Physiol 152:1787-1795.
  • Zhang GH, Ma CH, Zhang JJ, Chen JW, Tang QY and He MH (2015) Transcriptome analysis of Panax vietnamensis var. fuscidicus discovers putative ocotillol-type ginsenosides biosynthesis genes and genetic markers. BMC Genomics 16:159-172.
  • Zou XX (2002) China Capsicum. China Agricultural Press, Beijing.

Internet Resources

Edited by

Associate Editor: Houtan Noushmehr

Publication Dates

  • Publication in this collection
    29 May 2020
  • Date of issue
    2020

History

  • Received
    27 Sept 2018
  • Accepted
    12 June 2019
Sociedade Brasileira de Genética Rua Cap. Adelmio Norberto da Silva, 736, 14025-670 Ribeirão Preto SP Brazil, Tel.: (55 16) 3911-4130 / Fax.: (55 16) 3621-3552 - Ribeirão Preto - SP - Brazil
E-mail: editor@gmb.org.br