Introduction

Cultivation of citrus plants is widespread due to their delicious taste and high nutritional content (Talon et al. 2020). Citrus sinensis is especially noteworthy for its high commercial value in subfamily Aurantioideae (Biswas et al. 2020; Liu et al. 2022). However, they are often susceptible to diseases, including Huanglongbing (HLB), which was first identified in Asia and has since spread to South Africa and South America (Ferrarezi et al. 2020). The impact of Huanglongbing (HLB) on the citrus industry is significant, with an estimated 90% of affected areas experiencing damage (Shahzad et al. 2020). The disease is caused by the bacterium Candidatus Liberibacter asiaticus (CLas), which is transmitted by the Diaphorina citri (Hu et al. 2021; Alquézar et al. 2022). At the present time, there is unfortunately no cure available for HLB. Possible treatments include antibiotics, thermotherapy, gibberellin, and the use of citrus cultivars that are resistant to HLB (Munir et al. 2018; Ma et al. 2022). The interactions between citrus and CLas have been continuously explored, as several important secreted proteins in CLas have been functionally analyzed, including SDE1, SDE3, and SDE15 (Clark et al. 2018; Pang et al. 2020; Shi et al. 2023). Despite extensive research and findings, the pathogenesis of HLB remains unclear.

Plants are faced with environmental stress from abiotic and biotic factors, and the response to this is greatly influenced by the presence of phytohormones, which are responsible for the regulation of multiple functions and intricate biological processes (Ku et al. 2018). Research has shown that phytohormones are necessary for plants to effectively guard themselves against pathogens (Collum and Culver 2016). The major hormones that are functionally active are abscisic acid (ABA), auxin, brassinolide (BR), cytokinin (CK), ethylene (Eth), gibberellin (GA), jasmonic acid (JA), and salicylic acid (SA) (Checker et al. 2018). Utilizing GA can help to improve the consequences of Huanglongbing (HLB) infection, including reducing blotchy mottle and boosting plant growth (Ma et al. 2022). C. sinensis displayed a higher tolerance to HLB, likely due to the augmented levels of auxins, CKs, and JA, found in citrus affected by HLB (Peng et al. 2021; Suh et al. 2021). However, the levels of SA varied between different citrus varieties during CLas infection (Suh et al. 2021; Zou et al. 2021).

Photosynthesis in plants takes place in chloroplasts, which are capable of converting the sun’s light energy into chemical energy effectively (Kruse et al. 2005). Furthermore, they are instrumental in defending against pathogen infection (Lu and Yao 2018). Consequently, chloroplasts are indispensable for both photosynthesis and independent immune responses when the plant is exposed to pathogen (Jensen and Leister 2014). Research has revealed that CLas infection has an impact on carbon metabolism, resulting in an increase of starch in leaves and the destruction of chloroplasts (Etxeberria et al. 2009). Transcriptome analysis of Citrus spp. showed that the photosynthesis processes were impeded when citrus plants were infected with HLB (Zhong et al. 2016; Hu et al. 2017; Tang et al. 2018; Liu et al. 2019; Zhao et al. 2019).

RNA editing events in plant organellar were documented and studied in the early 2000s (Shikanai 2006). In recent years, the importance of RNA editing in plant development and defense has been increasingly recognized (Hao et al. 2021). It has been suggested that RNA editing is a key mechanism for the regulation of gene expression in chloroplasts (Wang et al. 2022). RNA editing has been shown to be involved in the regulation of chloroplast-localized defense responses, including the expression of defense-related genes, the accumulation of defense-related metabolites, and the modulation of JA and ABA signaling pathways (Jiang and Dehesh 2021; Zhang et al. 2022). Moreover, it has been suggested that RNA editing is involved in the abiotic stress response in C. reticulata (Pan et al. 2021). Nevertheless, a few investigations have indicated correlations between RNA editing events and the emergence of HLB.

RNA-seq offers a swift approach to distinguishing differentially expressed genes (DEGs) and discovering gene clusters that are co-expressed (Han et al. 2015; Conesa et al. 2016). Moreover, RNA-seq data can be utilized to identify single nucleotide polymorphisms and can be used to detect RNA-editing events (Pinto and Levanon 2019). To comprehend the events that may be elicited by CLas and uncover the molecular foundation during interaction between CLas and C. sinensis, we conducted a study to investigate the transcriptomic alterations, including the identification of DEGs, the enrichment of gene functions, and the detection of RNA-editing events in mitochondria and chloroplasts during CLas infection in C. sinensis using RNA-seq, which provides valuable insight for further research.

Materials and methods

Measurement the activities of antioxidant enzymes

Leaves from CLas free and -infected two years old C. sinensis cv. “Newhall” plants, grown in the citrus germplasm of Gannan Normal University, were collected for this study. To determine whether the citrus plants were CLas free or -infected, qPCR was employed as previously described (Li et al. 2023). The leaves were immediately ground in phosphate buffer saline solutions and centrifuged to obtain the supernatant for determining the activities of antioxidant enzymes. The activities of ascorbate peroxidase (APX), catalase (CAT), glutathione reductase (GR), peroxidase (POD), and superoxide dismutase (SOD) were measured using commercially available APX activity detection assay kits (BC0225, Solarbio), CAT activity detection assay kits (BC4785, Solarbio), GR activity detection assay kits (BC1165, Solarbio), POD activity detection assay kits (BC0095, Solarbio), and SOD activity detection assay kits (BC0175, Solarbio), respectively.

RNA extraction and sequencing

Total RNA was extracted from citrus leaves using the TaKaRa MiniBEST Plant RNA Extraction Kit (9769, TaKaRa) according to the manufacturer’s instructions. The concentration, purity, and integrity of the extracted total RNA were assessed using NanoDrop one Spectrophotometer (Thermo Fisher Scientific) and Agilent Bioanalyzer 2100 system (Agilent Technologies), respectively. For library construction, the TruSeq RNA Library Prep Kit v2 was used, and the resulting libraries were subsequently sequenced on an Illumina HiSeq 4000 platform.

Transcriptome analysis

Raw sequencing reads were subjected to quality control using trimmomatic v0.36 to remove adapter sequences and low-quality sequences. Subsequently, the clean reads were aligned to the reference genome of C. sinensis (version 3.0) obtained from the CPBD database using Hisat2 with default parameters (Kim et al. 2019; Liu et al. 2022). To estimate the Fragments Per Kilobase of exon model per million mapped fragments (FPKM) value for each gene, a count matrix was generated using StringTie (Pertea et al. 2015). For the analysis of gene expression patterns in the chloroplast and mitochondrion genomes, the reference genomes of chloroplast (NC_008334.1) and mitochondrion (NC_037463.1) were downloaded from the NCBI genome database since this information was not available in the CPBD database (Bausher et al. 2006; Yu et al. 2018). Differential expressed genes (DEGs) were identified using DESeq2 with a significance threshold of |log2(fold change, FC)|≥ 1 and false discovery rate (FDR) < 0.05 (Love et al. 2014). The gene trend cluster was performed using R package “Mfuzz” (Kumar and Futschik 2007).

Functional and pathway analysis of DEGs

The Gene Ontology (GO) enrichment analysis of DEGs was performed using the “GO enrichment” modules in TBtools, with a significance threshold of a p value < 0.05 (Chen et al. 2020). Similarly, the KEGG enrichment analysis of DEGs was conducted using the “KEGG enrichment analysis” modules in TBtools, also with a significance threshold of a p value < 0.05 (Chen et al. 2020). Additionally, MapMan was utilized to display the DEGs on diagrams representing specific metabolic pathways (Thimm et al. 2004).

RNA editing event identification

REDO was used to detect the RNA editing events in chloroplast and mitochondrion (Wu et al. 2018). BWA was employed for RNA-seq mapping (Li and Durbin 2009), while GATK and SAMtools were employed for genome variant calling (Li et al. 2009; McKenna et al. 2010). The output variant call format files and the sequence and gene annotation file of the reference genome were then used as input files for REDO. The output files generated by REDO were manually settled.

RT-qPCR analysis

Total RNA of the C. sinensis leaves was extracted using M5 HiPer Plant RNeasy Complex Mini Kit (Mei5bio, China). Then, first strand cDNA was reverse translated by EasyScript® First-Strand cDNA Synthesis SuperMix (Transgen, China). The qPCR experiments were conducted on ABI StepOne PLUS Real-Time PCR System (ABI, USA) with 2X M5 HiPer UltraSYBR Mixture (Low ROX) (Mei5bio, China). All primers were listed in Table S1. The relative expression profiles of candidate genes were calculated using 2−ΔΔCT method with CsGAPDH as internal gene (Livak and Schmittgen 2001).

Statistical analysis

The statistical significance of experimental data was import into SPSS 25.0 and analyzed by one-way ANOVA using Tukey’s (HSD) test or by Student’s unpaired two-sided t-test.

Results

CLas infection caused alterations in physiological indices of C. sinensis

To assess the changes in physiological indices of C. sinensis during CLas infection, the levels of chlorophyll and the activities of APX, CAT, GR, POD, and SOD were evaluated (Fig. 1). Results showed that the chlorophyll content of leaves of citrus plants affected by HLB was significantly reduced (Fig. 1A). Upon CLas infection, the activities of APX and GR were diminished (Fig. 1A, D), whereas CAT, POD, and SOD activities were enhanced (Fig. 1C, E, F).

Fig. 1
figure 1

Determination analysis of chlorophyll content and oxidation–reduction enzyme activity of CLas free and -infected citrus plants. A Content of chlorophyll, B activity of ascorbate peroxidase (APX), C activity of catalase (CAT), D activity of glutathione reductase (GR), E activity of peroxidase (POD), E activity of superoxide dismutase (SOD). Statistical analysis revealed significant differences (p value < 0.05) between the groups, as indicated by different letters

Full size image

Analysis of transcriptome data obtained from RNA-seq

In this study, the amount of filtered clean reads of sequenced samples for transcriptome analysis varied from 35,920,788 to 51,884,560. The correlation and PCA analysis revealed that biological repeatability is of a good quality (Fig. 2A, B). The DEGs were identified based on a significance threshold of |log2(fold change, FC)|≥ 1 and a false discovery rate (FDR) of < 0.05 (Fig. 2C). A total of 3604 DEGs were identified between CLas free and -infected citrus plants, with 1371 downregulated and 2233 upregulated (Fig. 2D and Table S2).

Fig. 2
figure 2

Comparison transcriptome analysis of CLas free and -infected citrus plants. A Cluster analysis, B principal component analysis, C volcano map analysis, D number of differentially expressed genes. “MOCK” referred to CLas free Citrus sinensis samples and “HLB” referred to CLas infected samples. the number after MOCK or HLB indicated the number of biological replicates

Full size image

Clustering and GO enrichment analysis of DEGs

The R package “Mfuzz” was used to analyze the expression tendency of DEGs, and the tendency was discernible. Analysis revealed that DEGs were mainly divided into two distinct categories, with either down- or upregulation, although there were genes with sample specificity (Fig. 3A). The GO enrichment analysis revealed that the majority of the top 20 enriched GO terms were associated with either “chloroplast” or “photosynthesis,” suggesting a significant effect on photosynthesis (Fig. 3B and Table S3). The enriched GO terms included “GO:0009535, chloroplast thylakoid membrane,” “GO:0055035, plastid thylakoid membrane,” “GO:0019684, photosynthesis, light reaction,” and “GO:0015979, photosynthesis,” (Fig. 3B).

Fig. 3
figure 3

Gene expression tendency and GO enrichment analysis of differentially expressed genes between CLas free and -infected citrus plants. A Gene expression tendency analysis and B GO enrichment analysis. The size of each circle was proportional to the gene numbers of the related GO term

Full size image

KEGG enrichment analysis of DEGs

To investigate the particular pathway of the downregulated and upregulated DEGs, KEGG enrichment analysis was conducted (Fig. 4). The top 25 KEGG enrichment pathways of downregulated DEGs, including “photosynthesis,” “Fructose and mannose metabolism,” “Carotenoid biosynthesis,” and “Nitrogen metabolism,” suggest that these pathways were inhibited during CLas infection (Fig. 4A and Table S4). Conversely, the top 25 KEGG enrichment pathways of upregulated DEGs encompass “Amino sugar and nucleotide sugar metabolism,” “Phenylpropanoid biosynthesis,” “Oxidative phosphorylation,” and “Starch and sucrose metabolism,” signifying that these pathways were augmented during CLas infection (Fig. 4B and Table S5).

Fig. 4
figure 4

KEGG enrichment analysis of differentially expressed genes between CLas free and -infected citrus plants. A Downregulated genes and B upregulated genes

Full size image

DEGs associated with biotic stress in C. sinensis during CLas infection

Analysis of C. sinensis during CLas infection uncovered a range of DEGs associated with biotic stress, including those involved in hormone signaling, cell wall, proteolysis, redox state, transcription factors, and secondary metabolites (Fig. 5A and Table S6). Specifically, analysis of differentially expressed genes (DEGs) involved in gibberellin signaling (Fig. 5A and Table S6) revealed that two ent-kaurene oxidase genes were upregulated, while the third showed downregulation. Additionally, two gibberellin 20-oxidase genes were downregulated, while two gibberellin receptor GID1 genes were upregulated and one gibberellin 2-oxidase gene demonstrated downregulation. Furthermore, the isochorismate synthase gene was downregulated, while two salicylic acid 3-hydroxylase genes were upregulated in relation to salicylic acid metabolism. Lastly, the majority of DEGs connected to cell wall metabolism and redox state, such as expansin, peroxidases, and glutathione S-transferase genes, showed an upregulation (Fig. 5A and Table S6).

Fig. 5
figure 5

MapMan analysis of differentially expressed genes between HLB-uninfected and -infected citrus plants. A Biotic stress-related genes and B photosynthesis-related genes. PEX referred to peroxisome and MTC referred to mitochondria

Full size image

DEGs associated with photosynthesis in C. sinensis during CLas infection

In order to evaluate the influence of photosynthesis on CLas infection, MapMan software was used to analyze the expression levels of DEGs in photosynthesis (Fig. 5B and Table S7). Results indicated a decrease in expression of DEGs associated with Photosystem I and II, such as LHCa1/2/3/4, PsaK, PsaL, PsaN, LHCb1/2/3, LHCb4/5/6, PsbTn, PsbW, and PsbQ (Fig. 5B and Table S7). Most of the DEGs that form part of the Calvin cycle also showed a decrease in expression, apart from a gene encoding the RuBisCo heterodimer small subunit and a gene encoding fructose 1,6-bisphosphate aldolase (Fig. 5B and Table S7). Moreover, four out of five DEGs associated with ATP synthase had decreased expression.

Examination of RNA editing events and their frequency alteration in chloroplast and mitochondrion of C. sinensis during CLas infection

A systematic analysis of DEGs encoded by the genome of chloroplast and mitochondrion were conducted, and 13 and 9 DEGs were identified respectively (Fig. 6 and Figure S1). The DEGs in the chloroplast were associated with photosystem I and II, including psbK, ndhB, petA, and psaJ genes, all of which were downregulated (Fig. 6C). In contrast, the only DEG in the mitochondrion, nad6, which is involved in oxidative phosphorylation, was upregulated (Figure S1C). To generate reliable RNA editing events, the common RNA editing events from MOCK and HLB samples were utilized for further study (Fig. 7A and S2A). Most of the common RNA editing events were cytimidine mutant to uracil, which was marked as thymine (Table S8, S9, and S10). Significantly decreased RNA editing frequencies were observed for some RNA editing events, such as the rps14 gene in chloroplast and the cob gene in mitochondrion, during CLas infection (Fig. 7B and S2B). Bioinformatic analysis generated RNA editing events, which were verified by PCR amplification and Sanger sequencing. The peaks of the sequencing results indicated the presence of the editing events (Fig. 8).

Fig. 6
figure 6

Analysis of differentially expressed genes in the chloroplast genome between CLas free and -infected citrus plants. A Volcano map analysis, B heatmap analysis, and C MapMan analysis

Full size image
Fig. 7
figure 7

RNA editing events observed in genes coded in the chloroplast genome. A Venn diagram analysis and B example genes with RNA editing events. “MOCK” referred to CLas free Citrus sinensis samples and “HLB” referred to CLas infected samples. The percentage number that followed each bar graph was representative of the frequency of RNA editing events

Full size image
Fig. 8
figure 8

PCR verification of RNA editing events in the chloroplast genome. A ndhD, B rpoA, C atpA, and D ropB. Labels indicating the genome location of RNA editing events were placed beneath the sequencing peak diagrams. “MOCK” referred to PCR products amplified from the cDNA of CLas free Citrus sinensis samples while “HLB” referred to CLas infected samples. “Reference” referred to PCR products amplified from the genomic DNA of CLas free C. sinensis samples

Full size image

RT-qPCR verification of transcriptome data

To validate the accuracy of the transcriptome results, we conducted RT-qPCR analysis and compared the expression tendency of nine candidate genes using their RT-qPCR and RNA-seq data. Results showed that eight of the nine genes were upregulated and one gene, Cs_ont_6g016140, was downregulated (Fig. 9). The expression patterns of these genes were consistent with the FPKM value of RNA-seq analysis. For instance, Cs_ont_5g006680, a WRKY transcription factor, was induced about 3.50-fold during CLas infection as determined by RT-qPCR analysis (Fig. 9). Similarly, the log2FC of Cs_ont_5g006680 was 3.79-fold according to the RNA-seq analysis. This suggests that the RNA-seq data is highly reliable.

Fig. 9
figure 9

RT-qPCR verification analysis of differentially expressed genes between CLas free and -infected citrus plants. A Cs_ont_1g026400, B Cs_ont_1g029550, C Cs_ont_2g012750, D Cs_ont_3g016900, E Cs_ont_3g016840, F Cs_ont_5g006060, G Cs_ont_5g006680, H Cs_ont_6g016140, and I Cs_ont_8g025430. The heatmaps above the histograms displayed the average FPKM value of both “MOCK” and “HLB” samples. “MOCK” referred to PCR products amplified from the cDNA of CLas free Citrus sinensis samples while “HLB” referred to CLas infected samples. Statistical analysis revealed significant differences (p value < 0.05) between the groups, as indicated by asterisks

Full size image

Discussion

The utilization of high-throughput sequencing technology has enabled researchers to analyze the interactions between pathogens and plants, providing an effective means of analyzing HLB (Wang et al. 2016; Naidoo et al. 2018). Studies on this subject have revealed that photosynthesis, phytohormone metabolism, and signaling pathways are significantly altered due to HLB, with marked responses observed in biotic stress related differentially expressed genes (DEGs) (Balan et al. 2018; Arce-Leal et al. 2020; Weber et al. 2022; Liu et al. 2023).

CLas infection caused a significant decrease in photosynthesis process of C. sinensis

The starch accumulation in citrus plants infected with HLB is abnormally high, resulting in the destruction of chloroplasts and degradation of chlorophyll (Etxeberria et al. 2009). Accordingly, CLas infection in citrus plants caused a blockage in photosynthesis processes. The expression profiles of photosynthesis related genes were dramatically influenced. The photosynthesis processes enriched by DEGs in MapMan were downregulated in HLB-diseased fruits (Zhao et al. 2019). Remarkably, the DEGs associated with the light reactions of photosynthesis displayed a decrease in susceptible C. sinensis, while not showing any alteration in tolerant C. hystrix under CLas infection (Hu et al. 2017). Most DEGs related to photosynthesis were downregulated in CLas infected periwinkle (Liu et al. 2019). Through our study, we identified 125 DEGs that are related to photosynthesis (Table S6). Out of these, only nine were upregulated, while the rest associated with photosystem I and II were downregulated, suggesting that photosynthesis in the leaves of C. sinensis “Newhall” has been significantly suppressed.

CLas infection caused a significant dynamic change in hormone signaling of C. sinensis

Plant hormones are essential for plants, as they are involved in the responses to a variety of pathogenic infections, insect infestations, and abiotic stresses (Bari and Jones 2009). Our study revealed a great number of DEGs related to hormone signaling were observed during CLas infection (Table S7). Auxin functions as a balancing agent between JA and SA, connecting the defense response and development of plants when they interact with pathogens (Kazan and Manners 2009). The Aux/IAA genes act as transcriptional repressors, controlling the auxin signaling transduction process (Reed 2001). The PIN proteins facilitate the movement of auxin (Adamowski and Friml 2015). We identified 22 auxin-related DEGs (Table S7), 8 of which were downregulated, including YUCCA, Aux/IAA, MES17, and PIN. The other 14 upregulated DEGs also contained these genes, demonstrating a divergence in gene function during CLas infection in auxin signaling metabolism. Five SA-related DEGs were identified, including one ICS, two NPR3/4, and two DLO, as well as ten JA-related DEGs with seven upregulated, such as LOX, AOC, JAT, and ACS (Table S7). There were 14 DEGs related to GA, 6 of which were downregulated, and 14 DEGs related to ABA, 8 of which were downregulated. Additionally, 15 DEGs related to CK were identified, 10 of which were downregulated, including IPT, LOG, AHP, and IPT. Plant hormones play an important role in balancing the development and defense responses of plants when they encounter pathogens (Kazan and Manners 2009). Generally, SA and JA related DEGs were both affected due to the close relationship between these two plant hormones and plant immunity (Hou and Tsuda 2022); however, in our study, we found that the majority of DEGs were related to auxin, followed by CK, GA, and ABA. It is possible that HLB is caused by an immune response that has been provoked by a pathogen (Ma et al. 2022).

CLas infection caused activation in ROS-related genes of C. sinensis

ROS are highly toxic and play a key role in various processes related to plant growth, development, and defense (Shetty et al. 2008). ROS are also responsible for systemic reactions during plant-pathogen interactions (Bolwell and Daudi 2009). In C. sinensis, ROS were elicited in response to CLas infection (Ma et al. 2022). A previous study documented that the expression of RbohB, RbohC, RbohD, and RbohF was increased, while SOD, APX, CAT, and GR were decreased in both Valencia and SB mandarin (Ribeiro et al. 2023). Results of our study demonstrated a significant increase in the activity of CAT, POD, and SOD and a decrease in the activity of APX and GR in response to CLas infection (Fig. 1). According to a recent study, CLas infection stimulated the activity of CAT, POD, and SOD in C. maxima and C. reticulata (Wu et al. 2023). However, another study revealed that CAT and SOD activities decreased, with no change in POD activity in C. junos (Chen et al. 2022). It appears that various Citrus species reacted differently to CLas infection. Correspondingly, the expression of Rboh, APX, MDAR, Trxh, and multiple GST genes was upregulated in the presence of HLB (Fig. 5A and Table S6).

RNA-editing events were influenced during CLas infection in the plant organellar of C. sinensis

RNA-editing events were ubiquitous in cells and participated in a lot of biological processes. Studies conducted by Yang et al. (2020) found that NbMORF8 reduces plant immunity to Phytophthora via RNA editing, whereas AtSLO2 improves plant resistance to Botrytis cinerea through the same mechanism (Zhu et al. 2014). In our research, we identified a substantial amount of RNA editing events in the chloroplast and mitochondrion and observed that the editing frequencies of some of these events were impacted by CLas infection. Pentatricopeptide repeat (PPR) proteins were responsible for plant organellar RNA editing (Small et al. 2020). However, few PPR proteins were identified and functional analyzed in C. sinensis. We hypothesized that CLas infection could disrupt the functioning of PPR proteins during RNA editing. Nevertheless, the correlation between these RNA-editing events and the pathogenesis of CLas is yet to be determined, and further experiments must be conducted to gain a better understanding of this relationship.

Conclusions

Our research has revealed that citrus plants affected by HLB experience considerable physiological changes, including chlorophyll content and the activity of APX, CAT, GR, POD, and SOD. In addition, HLB affects the expression of genes related with the metabolism of phytohormone such as Auxin, CK, GA, and JA and inhibited most of the genes participated in the photosynthesis pathway. Furthermore, the RNA editing events were disturbed as the editing frequency of specific nucleotide sites was changed by CLas infection. Our discovery provides a new perspective on revealing the pathogenic mechanism of HLB.