Citrus plants infected with CLas pathogen recovered through Bacillus subtilis L1-21
In 2020-2021, we tested the efficacy of B. subtilis L1-21 against 120 citrus plants in a citrus field (100% infected) trial in Genma county of Yunnan Province, China. Symptomatology inspections showed that most of the citrus plants were diseased with different severity levels as shown in Fig. 1A. Disease severity were visually rated as healthy, symptomatic, and asymptomatic followed by quantification of CLas pathogen copies through quantitative real-time PCR by targeting the ribosomal protein L12 [rplL] as performed in our previous study [31]. Citrus plants with visible symptoms indicated higher pathogen load suggested CLas was the main source of infection leading to diverse symptoms in diseased citrus fields. We evaluated that the introduction of B. subtilis L1-21 once every month for a period of one year could provide significant exclusion of CLas pathogen from diseased plants (Fig. 1B). Within one month, endophyte-inoculated infected plant contributes to higher pathogen reduction. Initial pathogen load before the start of field experiment was higher with 105 pathogen copies. All the diseased citrus trees before the experiment were treated as control. Results revealed significant decrease of pathogen load to 10 copies per gram with diseased trees displayed more robust growth and plants survived for long time after introduction of endophyte L1-21 (Fig. 1C).
Manipulation of citrus bacterial community composition towards pathogen resistance in the presence of potential Bacillus subtilis L1-21
Further, we explored the effect of introduced B. subtilis L1-21 on the citrus diseased plants for possible enrichment of potential groups of microbial communities that could influence the pathogen resistance. A total of 1,541,254 raw reads with an average of 128,438 reads per sample were obtained from 12 samples through sequencing of V5-V7 variable region of 16S rRNA on an Illumina Hiseq2500 platform. After quality control and chimeras removal a total of 1,198,019, high quality effective reads were obtained with an average of 99,835 reads per sample. The effective reads were then clustered into a total of 2,264 OTUs with an average of 189 OTU per sample (Table S2). First, we assessed the alpha diversity of bacterial communities of citrus bacterial communities before (0 d; CK) and after treatment (3 d, 30 d, and 1 year) of endophyte L1-21 introduction and no significant difference was observed among the treatments (Tuckey-HSD; p > 0.05, Fig. 2A-B). A venn diagram further showed the variation between common and unique OTUs of native bacterial communities (Fig. 2C). Further, we assessed the changes in the structure of bacterial communities through PCoA based on Bray-Curtis dissimilarity matrix under different treatment (0 d, 3 d, 30 d, and 1 year). Results of PCoA based on Bray-Curtis dissimilarity matrix explained a total of the 76.37% and 11.70% variation among bacterial communities (Fig. 2D), indicating that bacterial community composition significantly changed after introduction of B. subtilis L1-21 (PERMANOVA; p = 0.001 and R2 = 0.3804).
The phyla Firmicutes, Proteobacteria, and Actinobacteria dominated the citrus bacterial communities and accounted for more than 95% of the citrus microbiome (Fig. 2E). The relative abundance of Proteobacteria and Actinobacteria was significantly higher at 0 d (before endophyte L1-21 treatment; CK) than in after 3 d, 30 d, and 1 year of endophyte L1-21 treatment (Tuckey-HSD, p < 0.05; Fig. 2E). The relative abundance of Firmicutes was significantly increased with introduction of endophytic treatment (3 d, 30 d, and 1 year) compared with 0 d and relative abundance of Proteobacteria and Actinobacteria significantly decreased after L1-21 treatment (3 d, 30 d, and 1 year) compared with 0 d (Tuckey-HSD, p < 0.05). At the genera level the patterns of relative abundance differences and taxonomic distribution for citrus bacterial communities became more obvious under different treatments (Fig. 2F). There were several genera such as Candidatus, Pseudomonas, Tsukamurella, and Shewanella were present in significantly high relative abundance and unique to CK compared to 3 d, 30 d, and 1 year (Tuckey-HSD, p < 0.05). In contrast, the relative abundance of Bacillus was significantly increased and relative abundance of Candidatus Liberibacter, Pseudomonas, Tsukamurella, and Shewanella was significantly decreased after application of endophyte L1-21 at 3 d, 30 d, and 1 year compared with CK (0 d) (Tuckey-HSD, p < 0.05). This suggested that bacterial community composition and structure of citrus plant significantly changed after application of B. subtilis L1-21. The application of endophyte to diseased host resulted in complex network after one year compared to untreated diseased plants. The co-occurrence network involving healthier microbe interaction with diverse Bacillus might have directly or indirectly reduced the CLas pathogen as reported in field experiments (Fig. 3A, B). Therefore, we reasoned that significant changes in microbial communities towards positive trends are due to some unknown positive interactions that are involved inside the diseased citrus plants in presence of B. subtilis L1-21. These interactions significantly increased the abundance of other beneficial Bacillus or unknown potential strains which reduced the disease incidence. Overall research results show that potential endophyte leads to restructuring of bacterial endophytic microbiome in diseased citrus groves and its application reduces the pathogen at genera level which are similar with results of qPCR (data not shown) that application of endophyte L1-21 significantly reduced the population of CLas in the citrus endosphere.
Citrus transcriptomic response to Bacillus subtilis L1-21 inoculant introduction
RNA-seq results provided here provided an appropriate data set for further exploration of citrus transcriptome. A total of 38.1 million to 57.4 million reads generated from the all the samples were directly mapped to the C. sinensis genome (http://citrus.hzau.edu.cn/orange/) with 82.08 % to 83.43 % were unique mapped reads matched annotated citrus genes, and the mapping ratio of these reads is 83.04-84.23% (Table S3). Notably, samples treated with endophytic bacteria L1-21 suggested obvious molecular changes. Importantly, the correlation coefficient of all the samples were high indicating endophytes treatment showed marked molecular changes in each of the healthy and diseased trees, the molecular difference between health and diseased plants (Fig. S1). Hence, the genes in response to citrus defense response need to be characterized. Differences in Gene ontology gene enrichment were studied for three main categories, namely, the biological process, cellular component, and molecular function for H (healthy) versus D (Disease) (Fig. S2). The endophyte B. subtilis L1-21 was applied to both healthy and diseased citrus plants (Fig. S2A), and similar genes among both plants were sorted out (Fig. S2B). We found that the most significant biological process were the metabolic process and single organism process in which six and three genes were upregulated respectively. We choose the most significantly enriched 17 biological processes after treatment with endophyte L1-21 (padj<0.05). Significantly enriched differentially expressed genes were found in the diseased (D vs LD) and healthy citrus trees (H vs LH) treated with endophytes (Fig. S2C, S2D).
Bacillus subtilis L1-21 manipulated disease responsive genes in infected citrus
Despite the progress of colonization of citrus endophyte and CLas pathogen in the phloem as same niche [14], citrus defense mechanism in the presence of indigenous citrus endophyte to mitigate this pathogen is still unknown. Further, molecular mechanisms in citrus plants after introduction of endophytes to diseased trees has not been studied so far. Whole transcriptomic profile analysis showed that diseased (Endo+HLB affected) and healthy (Endo+HLB free) citrus plants subjected to endophyte L1-21. More expression matrix were revealed after 6 h compared to control group as depicted through principal component analysis (PCA) (Fig. 4A). Consistently, approx. 3000 DEGs were regulated in Endo+HLB free plants (1965 upregulated; 979 downregulated) while more than 2000 DEGs were found in Endo+ HLB affected plants (1442 up regulated; 715 downregulated) (Fig. 4B). Our untargeted analysis points out that endophyte L1-21 application in plant-pathogen interaction pathway in both diseased and healthy citrus plants induces significant genes with more pronounced defense response compared to control group (Fig. 4C), including important genes (FLS2, WRKY33, PR1, PR4, RPS5, and RBOHD) with significant expression statistics (Fig. 4D). Further, we selected the top 10 abundant KEGG pathways that were triggered following application of the endophyte (Fig. 5) on diseased trees. The major upregulated KEGG pathways included biosynthesis of secondary metabolites; plant-pathogen interaction; and phenylpropanoid biosynthesis (padj<0.05) (Fig. 5A). In treatment of healthy citrus trees, major KEGG enrichment pathways comprised biosynthesis of secondary metabolites; metabolic pathways; biosynthesis of amino acids; and, phenylpropanoid biosynthesis (padj<0.05) (Fig. 5B). Upregulated pathogen resistance genes (in response to CLas) following endophyte application are pathogenesis-related 4 (PR4: Ciclev10029328m.g, Ciclev10029327m.g, Ciclev10029528m.g, Ciclev10029536m.g), disease resistance protein (CC-NBS-LRR class) family (Ciclev10030667m.g, Ciclev10024849m.g, Ciclev10024854m.g), chitin elicitor receptor kinase 1 (LYSM RLK1) (Ciclev10017678m.g), and respiratory burst oxidase homologue D (RBOHD, Ciclev10027774m.g) (Fig. 6A: additional file Table S4) and downregulated genes involved in pathogen resistance were heat shock protein 70 (Ciclev10027981m.g) and 17.6 kDa class II heat shock protein (Ciclev10009756m.g), which are responsible for protein folding. The control citrus plants in the absence of endophytes indicating clear differences were observed compared to diseased and healthy plants treated with B. subtilis L1-21.
Secondary metabolites in citrus defense to CLas pathogen
Secondary metabolites play an important role in defense mechanism of citrus trees and other plants. Significant regulation was noted in the diseased citrus leaves after application with endophyte indicating the positive effect of these agents against CLas pathogen. Significant regulation of secondary metabolites genes such as terpenoids and polyketides were noted in the diseased citrus leaves after treatment with endophytes indicating the positive affect of these agents against CLas pathogen. Geranylgeranyl pyrophosphate synthase 1 (Ciclev10012067m.g), isopentenyl diphosphate isomerase 1 (Ciclev10012312m.g), hydroxymethylglutaryl-CoA synthase/HMG-CoA synthase/3-hydroxy-3-methylglutaryl coenzyme A synthase (Ciclev10020042m.g), and squalene synthase 1 (Ciclev10028537m.g) are the important genes upregulated in the diseased citrus trees after treatment with endophytes (padj<0.05). Lipoxygenase 2 (Ciclev10014199m.g) involved in jasmonic acid mediated response in leaves, aldehyde dehydrogenase 3F1 (Ciclev10025492m.g), and GroES-like zinc-binding dehydrogenase family protein (Ciclev10020620m.g) are responsible for metabolism of lipids (Fig. 6B: additional file Table S5). The oxidative stress created by CLas pathogen inside citrus trees are detoxified through genes such as glutathione S-transferases (Ciclev10008944m.g, Ciclev10032702m.g, Ciclev10033001m.g, Ciclev10005808m.g, Ciclev10005812m.g), thus helping the citrus trees to show tolerance to CLas pathogen.
CLas pathogen negatively affected the protein folding, chaperones, and heat shock proteins
The most important mechanism involved in the disease symptoms of HLB are the down regulation of heat shock proteins, and product of these genes protect the protein folding and function during pathogen attack. The correct function is maintained in the phloem and leaves in the presence of these genes. The most important mechanism involved in the disease symptoms of HLB are the down regulation of heat shock proteins, and product of these genes protect the protein folding and function during pathogen attack. Up regulation in the citrus trees were observed treated with endophyte such as Chaperone DnaJ-domain superfamily protein (Ciclev10002683m.g, Ciclev10016883m.g, Ciclev10009810m.g), and Chaperone protein htpG family protein (Ciclev10030743m.g). The fold change for all these genes were more than 1 and the FDR ratio was (padj<0.05). Ubiquitin mediated protein degradation plays an important role in plant-pathogen interactions. There are 10 ubiquitin related genes regulated after endophyte treatment inside the citrus trees, 9 of them were up-regulated and 1 was down-regulated (Fig. 6C: additional file Table S6).
Endophyte induced changes in photosynthesis and carbohydrate metabolism
The HLB affected citrus leaves results in downregulation of important genes involved in photosynthesis processes. In the present study, we also showed that when diseased citrus leaves were treated with indigenous endophyte, only one of the gene ferrodoxin 3 (Ciclev10029499m.g) related to photosynthesis was up-expressed while two genes, APE1 (Ciclev10012434m.g) and PSB28 (Ciclev10022322m.g) related to acclimation of photosynthesis to environment and photosystem II reaction center, respectively were down-expressed indicating the pathogen is present inside the diseased leaves. Significant genes responsible for ethylene (Ciclev10000608m.g, Ciclev10014617m.g, Ciclev10031204m.g) were up-regulated. CLas pathogen causes accumulation of starch inside the phloem and other photosynthetic cells, resulting in blockage of important nutrients inside leaves. We found that B. subtilis L1-21 treatment resulted in downregulation of beta glucosidase 46 (Ciclev10014887m.g) and beta glucosidase 11 (Ciclev10019719m.g), which are responsible for starch accumulation (Fig. 6D: additional file Table S7). It has been reported that CLas pathogen negatively affected the metabolism of carbohydrate inside citrus trees. Significant genes are aldehyde dehydrogenase 3F1 (Ciclev10025492m.g), Galactose mutarotase (Ciclev10012004m.g),GroES-like zinc-binding dehydrogenase (Ciclev10020620m.g), hydroxymethylglutaryl-CoA synthase / HMG-CoA synthase / 3-hydroxy-3-methylglutaryl coenzyme A synthase (Ciclev10020042m.g), Alpha amylase(Ciclev10007401m.g), arginosuccinate synthase (Ciclev10019860m.g), and phosphoglucose isomerase 1 (Ciclev10000603m.g) with log fold change (LFC) of 1.77, 1.69, 1.57, 1.47, 1.45, 1.14, and 1.01, respectively, with FDR (padj<0.05) (Additional file Table S7).
Endophyte modulated the host cell wall genes
The genes involved in cell wall breakdown are mostly expressed in the citrus leaves affected with HLB, indicating the symptoms development are associated with these genes in HLB progression. The cellulose/transferases are all associated with the cell wall breakdown. Our study indicated that cellulose synthase/transferases (Ciclev10007586m.g, Ciclev10023570m.g, Ciclev10014586m.g) genes are down regulated after applying endophytes, which is the main genes in breakdown of cell wall (Additional file Table S8).
Host transcription factors induced during endophyte introduction
A total of 23 important transcription factors were identified when citrus trees were treated with endophytes. Among them 18 TFs are upregulated and 5 are down-expressed. These transcription factors has an important role in plant defense response, biotic and abiotic stress, plant immunity, leaf senescence, stomatal movement, and jasmonate metabolism. Several families of transcription factors, such as WRKY (11, 28, 33, 40, 50, 55), MYB (1, 15, 116), and EIN3 (Ciclev10000608m.g) are associated with the plant immunity and defense up-expressed in the citrus trees after endophyte treatments (Fig. 6E: additional file Table S9). The WRKYs transcription factors are also involved in the tolerance to CLas pathogen in citrus. The other important TFs responsible for plant defense mechanism are respiratory burst oxidase homologue D (Ciclev10027774m.g) and leucine-rich repeat protein kinase family protein (Ciclev10019897m.g), expressed to a log fold change of 4.02 and 2.29, respectively (padj<0.05) (Table S9). The expression profiles of 10 genes were compared with qPCR and RNA-seq data, and the results were consistent with corresponding log2fold values. Most of the genes exhibited the similar expression patterns using both methods. Similar results indicated that data are reproducible and reliable though the samples were collected from different batches (Fig. S3).