Genome-wide identification, characterization, interaction network and expression profile of GAPDH gene family in sweet orange (Citrus sinensis)

Citrus GAPDH gene identification and chromosome mapping in sweet orange

The method to identify all the putative GAPDH genes in sweet orange is derived from Zhang et al. (2019). Annotated proteins sequence were downloaded from the C. sinensis annotation project database of Huazhong Agricultural University (http://citrus.hzau.edu.cn/orange/) (Xu et al., 2013), phytozome Cirus sinensis v1.1 database (http://phytozome.jgi.doe.gov/pz/portal.html) (Wu et al., 2014), and the Citrus Genome Database (https://www.citrusgenomedb.org/). Then potential candidates were screened based upon homologous alignment among rice, Arabidopsis and sweet orange. The integrity of the Gp_dh_N and Gp_dh_C domain candidate genes were identified using SMART software with e-value < 0.1 (http://smart.embl-heidelberg.de/). Finally, the correctly predicted genes of sweet orange were named from CsGAPDH1 to CsGAPDH6 according to their phylogenetic relationships with Arabidopsis. Five CsGAPDH genes were mapped onto the chromosomes, and the schematic diagram distribution on the chromosomes was drawn using Photoshop software.

Sequence alignment, phylogenetic analysis, and classification of GAPDH genes in C. sinensis

GAPDH protein sequences of Arabidopsis and Citrus species (https://www.arabidopsis.org/) were aligned using ClustalW program of MEGA 5.1 software. After alignment, GeneDoc was used to display the differential amino acids in the alignment results with default parameters. To classify and analyze the evolutionary relationship of GAPDHs identified from C. sinensis, related genus, and Arabidopsis, the phylogenetic analysis based on sequence alignments was conducted with MEGA 5.1 software using the Neighbor-joining statistical method and the bootstrap test carried out with 1,000 replications.

Putative cis-acting regulatory elements and protein–protein interaction network of CsGAPDH proteins

The up-stream 1,500 bp sequence from the DNA transcription start site was used to analyze the cis-elements. Putative stress or hormone-responsive cis-elements located in the promoter region of CsGAPDH genes were obtained using PlantCARE online (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/). The interacting proteins of CsGAPDHs were searched, employing ortholog-based and domain-based methods from the Orange Genome Annotation Project (http://citrus.hzau.edu.cn/orange/ppi/index.php) (Ding et al., 2014; Xu et al., 2013).

GO analysis

Homologous genes of CsGAPDHs interacting genes in Arabidopsis were taken from the sweet orange annotation project database of Huazhong Agricultural University (http://citrus.hzau.edu.cn/orange/), and then GO analysis for the homologous genes were performed using the agriGO online database (http://bioinfo.cau.edu.cn/agriGO/analysis.php) (Tian et al., 2017).

Seeds germination

Mature seeds, with their seed coat stripped off, were spread evenly in a petri dish containing two layers of filter paper and cultured in darkness at 28 °C. Replenish water in time in the process of culture to ensure the normal germination of seedlings. Three seeds were each taken at 0, 3, 6 and 9 days following germination and were immediately added to liquid nitrogen for RNA extraction.

Plant growth condition and treatments with darkness in C. sinensis and phosphorus deficiency in Poncirus trifoliata

All the citrus seedlings studied were rooted from seeds. The seeds were spread in a petri dish containing moist filter paper and cultured at 28 °C in the dark. When the root length reached 2 cm, the seedlings were transferred into soil to grow and were cultivated in a greenhouse at 28 °C under 14 h-light/10 h-dark growth conditions. Consistent and robust plants were selected after a month of growth to be used in dark treatments to determine CsGAPDH gene expression. Fresh leaves were harvested after dark treatment of 12 h, 24 h, and 48 h respectively and stored in liquid nitrogen immediately. Plants grown under normal photoperiods were used as controls. For phosphorus deficiency treatment, 2-month-old Poncirus trifoliata (Pt) seedlings were irrigated with Hoagland solutions containing 1 μM P, while the controls were irrigated with 1 mM P (Zhang et al., 2019). Detection of CsGAPDH transcriptional levels in roots were measured after 1 week and 4 weeks of starvation treatment.

RNA isolation and qRT-PCR

Total RNA was isolated from each sample using TransZol reagent (TransGen Biotech) according to the manufacturer’s instructions. RNA integrity and concentration were measured using 1% agarose gels and the NanoDrop 2000 (Thermo Scientific, Waltham, MA, USA) respectively. Four micrograms of total RNA was reverse-transcribed in a reaction of 20 μl, using TransCript One-Step gDNA Removal and cDNA Synthesis Super Mix Kit (TransGen Biotech, Beijing, China) to obtain first-strand cDNA. Melting curve analysis was utilized for detecting primer specificity before quantification with the following program: 95 °C for 15 s, 60 °C for 1 min, 95 °C for 15 s, 60 °C for 15 s. The quantitative PCR reaction procedure was performed on an Applied Biosystems^® QuantStudioTM 7 Flex Real-Time PCR System (Life Technologies, Carlsbad, CA, USA) at 95 °C for 10 min, 35 cycles of 95 °C for 15 s and 60 °C for 1 min. The relative expression levels were analyzed using the 2^−∆∆CT method. The C. sinensis Actin gene was used as the internal reference for normalization. All the primers for quantitative real-time PCR are listed in Table S1 and were designed using primer-blast (https://www.ncbi.nlm.nih.gov/tools/primer-blast/).

Subcellular localization analysis

Constructed expression vector pFGC5941-CsGAPDHs-GFP was introduced into Agrobacteriumtumefaciens EHA105, and then the suspensions of Agrobacteria were infiltrated into tobacco leaves (Nicotiana benthamiana). Next, the plants were incubated in a culture room for 2 days. CsGAPDHs-GFP was observed by a Confocal Laser Scanning Microscopy. GFP was excited at 488 nm with an argon laser and fluorescence was detected by 505–550 nm band-pass filter. Band-pass filter 575–615 IR was used to detect chlorophyl auto-fluorescence.

Transcriptional analysis of CsGAPDHs in protoplasts

In order to examine the transcriptional activity of CsGAPDH1/CsGAPDH2, full-length CsGAPDH1/CsGAPDH2 was fuzed to the GAL4 DNA binding domain to generate a GAL4DBD-CsGAPDH1/CsGAPDH2 fusion construct driven by the CaMV 35S promoter as an effector. 4×GAL4 was fuzed to luciferase driven by the CaMV 35S promoter as the reporter. A construct containing the Renilla luciferase gene driven by the Arabidopsis UBIQUITIN3 promoter was used as the internal control. The effector and reporter, and internal control, were co-transfected into Arabidopsis protoplasts in a ratio of 6:6:1 (effector:reporter:reference) respectively, and was then cultured for 12 h at 25 °C in darkness. The luciferase activities were measured with the Dual Luciferase Reporter Assay System (Promega, Madison, WI, USA).

Identification and phylogenetic analysis of CsGAPDH proteins

In order to run a complete search identifying GAPDH genes in the genome of sweet orange, all annotated proteins of the genome from the sweet orange annotation project database of Huazhong Agricultural University (http://citrus.hzau.edu.cn/orange/) (Xu et al., 2013), phytozome C. sinensis v1.1 datebase (https://phytozome.jgi.doe.gov/pz/portal.html) (Wu et al., 2014) and the Citrus Genome Database (https://www.citrusgenomedb.org/) were considered for analysis. After determining the integrity of the GAPDH protein domains using the online program SMART (http://smart.embl-heidelberg.de/) alongside sequence alignments with known Arabidopsis GAPDH protein sequences, 46 GAPDH proteins were identified in six species (clementine mandarin, pummelo, citron, papeda, sweet orange, and Chinese box orange; Data S1). Considering C. sinensis is the most cultivated variety of citrus with the most extensive ongoing research, the 12 CsGAPDH proteins identified from this species were the focus of this study. Protein sequences identified from two different genomic reference databases were largely duplicated (Fig. S1), resulting in six GAPDHs that were finally identified in C. sinensis. The detailed information of the GAPDH genes are listed in Table S2. In order to study the evolutionary relationships of the GAPDH gene family, a phylogenetic tree was built from the alignment of all six full-length GAPDH protein sequences in sweet orange and seven GAPDH protein sequences in Arabidopsis (Fig. 1A). CsGAPDH genes were designated CsGAPDH1, CsGAPDH2, CsGAPDH3, CsGAPDH4, CsGAPDH5 and CsGAPDH6 according to their phylogenetic relationship with Arabidopsis, which could further be clustered into three subgroups, subgroup I, II, and III (Fig. 1A). Proteins in the same subgroup may perform similar functions. Subgroup I corresponds with Arabidopsis GAPC proteins (CsGAPDH1, CsGAPDH2 and CsGAPDH3), subgroup II with GAPCp proteins (CsGAPDH4), and subgroup III with GAPA/B proteins (CsGAPDH5 and CsGAPDH6).

To examine the conserved amino acids of Gp_dh_N and Gp_dh_C subunits, multiple sequence alignments of both subunits in Arabidopsis and sweet orange were performed. All of the CsGAPDH proteins, excluding CsGAPDH3, showed highly conserved Gp_dh_N and Gp_dh_C subunits (Figs. 1A–1C). Moreover, two Gp_dh_C subunits were detected in CsGAPDH3 protein, both of which were much shorter than Gp_dh_C subunits found in other CsGAPDH proteins, indicating that CsGAPDH3 may have a distinct origin and function from the other CsGAPDH proteins.

Chromosomal distribution of CsGAPDH genes

The physical positions of the GAPDH genes were obtained from Huazhong Agricultural University’s sweet orange annotation database (http://citrus.hzau.edu.cn/orange/). The position and transcriptional direction of each gene are shown in Figs. 2A–2D, and the exact positions on C. sinensis chromosome pseudomolecules are given in Table S2. The data shows that five CsGAPDH genes were found to be dispersed across four chromosomes. CsGAPDH1, CsGAPDH2, and CsGAPDH5 were localized on chromosome 7, chromosome 5, and chromosome 3, respectively, while both CsGAPDH3 and CsGAPDH6 were located on chromosome 2. There was no information on the position of CsGAPDH4 within the sweet orange annotation database.

Expression profiling of CsGAPDH genes

To reveal the transcriptional accumulation of CsGAPDH genes, the expression of six CsGAPDH genes were examined by real-time PCR in leaf, flower, fruit and callus tissues of the sweet orange (Figs. 2E–2J). Notably, transcriptional accumulation of CsGAPDH genes was found to demonstrate tissue-specific expression. CsGAPDH3 showed extremely low expression levels in the tissues analyzed compared with the other five CsGAPDH genes, except for in calluses where it demonstrated a significantly higher expression level (Fig. 2G), indicating that it might be involved in the development of undifferentiated tissues/cells. In contrast, CsGAPDH genes in subgroup III (CsGAPDH5 and CsGAPDH6) showed extremely high transcript accumulations in leaf tissue, but were much lower in the other three tissues (Figs. 2I and 2J). Therefore, suggesting that these genes might play roles in differentiated tissues/cells, especially in green tissues/cells. The other CsGAPDH genes were expressed constitutively and were found to be highly enriched in most of the tissues. CsGAPDH1 showed a higher transcription level in callus tissues than in other three, while CsGAPDH2 illustrated the opposite pattern (Figs. 2E and 2F). CsGAPDH4 had high transcriptional levels in both callus and flower (Fig. 2H). These results suggest that CsGAPDH genes might play various roles at different stages throughout sweet orange development.

Subcellular localization of CsGAPDH proteins

In order to examine CsGAPDHs localization in plant cells, full-length cDNAs of CsGAPDH1, CsGAPDH2, CsGAPDH5 and CsGAPDH6 were fuzed to Green Fluorescent Protein (GFP) driven by the CaMV 35S promoter, which was then transiently expressed through tobacco leaf injections. CsGAPDH1 and CsGAPDH2 were both identified within the nucleus and cytoplasm while CsGAPDH5 and CsGAPDH6 were observed solely in the chloroplast as shown in Figs. 3A–3P. The singular locations of CsGAPDH5 and CsGAPDH6 were consistent with previous findings of leaf-specific expression and group III-specific classification (Figs. 1A, 2I, and 2J).

Transcriptional activation of CsGAPDH proteins

Recent research has shown that GAPDH proteins within the nucleus have the ability to bind DNA and play a role in transcriptional activation (Testard et al., 2016; Zhang, Zhao & Zhou, 2017). To determine the functions of nuclear localized CsGAPDHs in gene expression regulation, CsGAPDH1 and CsGAPDH2 were selected for dual-luciferase transient assays using Arabdopsis protoplasts (Figs. 3Q and 3R). CsGAPDH proteins were fuzed with 35S-GAL4DB as effectors, and 4×GAL4 was fuzed to luciferase driven by the CaMV 35S promoter as a reporter. Ranilla luciferase driven by the Arabidopsis UBIQUITIN3 promoter was used as an internal control. When CsGAPDH1, CsGAPDH2 and the control vector were transformed into protoplasts, it was discovered that CsGAPDH1 and CsGAPDH2 were capable of enhancing luciferase activity at levels about 2-fold higher than the internal control, indicating that both of these proteins function as transcriptional activators (Figs. 3Q and 3R).

Light-responsive expression dynamics of CsGAPDH genes

Cis-elements are important components for protein binding sites on promoters in order to regulate transcription in response to phytohormones and other stress factors. A total of 1,500 bp upstream flanks of CsGAPDH genes were analyzed using the online software PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/). Nine types of cis-acting elements were detected in the promoter regions of five CsGAPDH genes, including ABA, Me-JA, auxin, and a GA responsive element (Fig. 4A). In addition, stress responsive elements related to low temperature and drought conditions were also been identified. Notably, light-responsive elements were the most abundant elements and were present in multiple locations of each CsGAPDH promoter region, indicating that these CsGAPDH genes are potentially regulated by light.

Next, the expression dynamics of CsGAPDH genes in response to light were examined. Expression of CsGAPDH genes in seedlings undergoing dark treatment for 12 h, 24 h, and 48 h were compared with an untreated control. CsGAPDH1 was down-regulated at 24 h and 48 h during dark treatment (Fig. 4B), whereas CsGAPDH2 presented a slightly lower expression at 24 h (Fig. 4C). CsGAPDH3 showed an increased expression level at 24 h (Fig. 4D), while CsGAPDH4 demonstrated this level at only 12 h (Fig. 4E). CsGAPDH4 was down-regulated in response to darkness treatment at 48 h (Fig. 4E). Subgroup III genes (CsGAPDH5 and CsGAPDH6) showed much lower transcriptional levels than the control at all the time points (Figs. 4F and 4G), which were consistent with their leaf specific expression (Figs. 2I and 2J). These results align with the abundance of light-responsive elements identified, indicating that CsGAPDH proteins may play roles within metabolism pathways related to photosynthesis and other light response pathways.

Interaction analysis of CsGAPDH protein

To discern the functions of CsGAPDHs in sweet orange, protein–protein interaction (PPI) networks of CsGAPDH proteins were built using the C. sinensis annotation project database. CsGAPDH1, CsGAPDH2, and CsGAPDH4 shared more than 90% of their interacting proteins, and in turn, all three proteins can interact with each other (Fig. 5A; Figs. S2 and S3). Meanwhile, it is predicted that CsGAPDH1, CsGAPDH2, and CsGAPDH4 can interact with themselves. CsGAPDH3 and CsGAPDH6 were predicted to interact with ten of the same proteins (Fig. 5B; Fig. S4).

To determine the functions of these interacting proteins, GO enrichment analysis for these proteins was performed with the AgriGO tools (http://bioinfo.cau.edu.cn/agriGO/analysis.php), reflecting the possible biological process that they involved in (Fig. 5C). Arabidopsis is considered one of the most important model organisms for plants and the majority of its genes have been functionally characterized. This makes it a good starting point to determine the functionality of the proteins found to interact with CsGAPDHs in sweet orange. According to their homologous proteins in Arabidopsis, these proteins appear to play roles in multiple biological metabolism processes such as glycolysis, generation of precursor metabolites and energy, carbohydrate catabolic process, glucose catabolic process, and more (Fig. 5C). Moreover, these proteins also are involved in response to inorganic substances such as cadmium ions or metal ions. Through extensive analysis, these findings suggest that CsGAPDHs may interact with each other to form complexes for cooperation or feedback networks.

Expression profiles of CsGAPDH genes in response to phosphorus deficiency in the root of Poncirus trifoliata

Sweet orange is generally used as a scion in agricultural production, and Poncirus trifoliata (Pt) is the main rootstock for sweet orange due to its ideal growth characteristics such as a strong root system, drought and pest resistance. Phosphorus (Pi) is important macronutrient for plant growth and development, and the CsGAPDH genes involved in response to inorganic substance (Fig. 5C). To gain a deeper understanding of CsGAPDH genes, the expression of CsGAPDH genes in roots exposed to phosphorus deficiency for 1 week and 4 weeks, with the seedling in the normal condition as control (Data S2) were investigated. CsGAPDH5 and CsGAPDH6 were predominantly expressed in young leaves, with CsGAPDH3 showing significantly low expression in all the tissues (Figs. 2G, 2I, and 2J). Consistent with this, CsGAPDH3, CsGAPDH5, and CsGAPDH6 showed much lower expression when compared with CsGAPDH1, CsGAPDH2, and CsGAPDH4 at the first and fourth week with or without Pi (Figs. 6A–6F). In addition, CsGAPDH5 and CsGAPDH6 were relatively highly expressed during the Pi starvation treatment by the fourth week (Figs. 6E and 6F), whereas CsGAPDH3 showed opposite pattern (Fig. 6C). However, CsGAPDH1, CsGAPDH2 and CsGAPDH4 showed high transcriptional accumulation with or without Pi starvation treatment (Figs. 6A, 6B, and 6D). In conclusion, the expression of CsGAPDH3, CsGAPDH5, and CsGAPDH6 exhibited a time-dependent response to Pi deficiency in Pt roots.

Expression of CsGAPDH genes during seeds germination

GO analysis revealed that the CsGAPDH genes may be involved in energy metabolic processes, providing energy for plant growth and development. Previous studies showed that glycolysis and TCA cycle provide considerable energy for germination through increase GAPDHs levels (Kim et al., 2009). In addition, plastidial glyceraldehyde-3-phosphate dehydrogenases are involved in ABA signal transduction, affecting seed germination (Muñoz-Bertomeu et al., 2011). Citrus seeds are relatively sensitive to salt stress during germination (Ziogas et al., 2017). The expression levels of CsGAPDHs during germination were monitored to uncover additional functions of the CsGAPDH genes. The expression of CsGAPDH1, CsGAPDH5 and CsGAPDH6 were increased at the third, sixth, and ninth day (Figs. 7A, 7E, and 7F). On the other hand, the expression of CsGAPDH2 and CsGAPDH4 decreased significantly subsequent to 3 days after germination (Figs. 7B and 7D). Therefore, it is speculated that some CsGAPDHs may be involved in energy metabolism during seed germination.