Huanglongbing (HLB) is a phloem-limited bacterial disease that infects citrus trees of all kinds, caused by Candidatus liberibacter (Jagoueix et al., 1996; Hansen et al., 2008; Duan et al., 2009). It is delivered to the phloem tissue by the Asian citrus psyllid vector and slowly kills trees over a period of several years (Bové, 2006). HLB has devastated the citrus industry in the United States, resulting in severe economic deterioration in affected regions (Dala-Paula et al., 2019; Singerman and Rogers, 2020). Efforts to control and manage HLB have been unsuccessful leading to yearly declines in citrus production. In areas where the disease is not broadly prevalent, management strategies include scouting and destroying infected trees. Aggressive management strategies are recommended for highly infected areas including the use of insecticides, antimicrobials, thermotherapy, antibiotics, and novel therapeutic agents (Davis et al., 2000; Bové, 2006; Das et al., 2014; Gardner et al., 2020). However, effective treatment and a commercially viable cure for the disease have not been found. Defense elicitors in crops have been under investigation as possible solutions to alleviate symptoms of various agronomic diseases. Elicitors upregulate the expression of defense proteins that in turn activate both systemic and local defense responses in treated plants (Montesano et al., 2003). In situations where this activity is observed, the reduction of diseases results in healthier plants and higher yielding crops. In multiple studies, HLB has been observed to interrupt the normal biochemical pathways in citrus. In Citrus limon, HLB infection facilitated the downregulation of catalases, chitinases, lectin-related proteins, and miraculin-like proteins (Nwugo et al., 2013). These metabolic changes ultimately lead to increased bacterial spread and susceptibility to other problematic pathogens. Among its repertoire of virulence tools, HLB encodes a salicylic acid hydroxylase, which degrades the plant hormone salicylic acid (Li et al., 2017). HLB also contains peroxiredoxins that increase oxidative stress tolerance, leading to increased HLB fitness in planta (Jain et al., 2015, 2018). The routine activation of a plants defense in response to an elicitor could perhaps reduce the success of HLB in citrus.

Microbial fermentation application (MFA), a previously described elicitor program, is formulated with a bacterial fermentation media, yeast cell wall extract, and a Cu component. It has been shown to reduce powdery mildew infection in wheat (Twamley et al., 2019). The authors also concluded that MFA upregulated pathogenesis-related protein 1 (PR 1), pathogenesis-related protein 4 (PR 4), pathogenesis-related protein 5 (PR 5), and pathogenesis-related protein 9 (PR 9) gene expression (Twamley et al., 2019). Upregulation of PR1 is an indicator of increased systemic acquired resistance (SAR) in plants (Ryals et al., 1996) and influences salicylic acid accumulation at the site of infection by various pathogens. The expression of PR1 increases plant tolerance to fungal, oomycete, and bacterial pathogenic challenges (Alexander et al., 1993; Jirage et al., 1999; Zhang et al., 1999). PR5 are a group of thaumatin-like proteins that are also associated with SAR and general defense responses. These are induced in response to various fungal and bacterial challenges (Li et al., 2015b; Chen et al., 2017). Peroxidases, such as PR9, are also widely involved in plant defense (Kawano, 2003) and have been observed in response to bacterial and fungal pathogen infections or exposure.

In this study, MFA was investigated for its potential role in activating citrus defenses against HLB. The previous research in wheat and soybean demonstrates that microbial fermentation products increase the expression of defense networks and aid in the improvement of crop quality parameters or can reduce the burden of disease (Twamley et al., 2019; Schulman et al., 2021; Twamley et al., 2021). If the upregulation of a systemic defense network can be achieved in a similar way in citrus, HLB might be suppressed using the MFA program. Additionally, MFA may affect the plant tissue Cu concentration, as a further mechanism for the restriction of HLB population. To this end, citrus gene expression, nutrient composition, and assessment of routine use of MFA on HLB infection levels quantified by PCR in a field study over the course of 20months were examined. This experiment monitored four treatment groups consisting of an uninfected control, infected trees, uninfected trees with MFA treatment, and infected trees with MFA treatment. Citrus petiole samples were observed routinely throughout the trial to determine the level of HLB infection and how it progressed in response to each treatment. Leaf tissue nutrient analysis was conducted to determine whether nutrient concentrations were influenced in each treatment group. Gene expression of PR2 and phenylalanine ammonia lyase (PAL) was assessed to determine the impact of MFA on defense gene expression between 0 and 8h post-MFA treatment. Samples from the hour 6 time point were used for a transcriptomic assessment using microarray.

This study examined the impact of a microbial preparation, MFA, against HLB in young citrus trees. The study assessed four experimental groups: control, infected, MFA treated, and MFA treated + infected. The results presented here indicate that MFA increased the transcriptional activity of citrus defense mechanisms, elevated Cu concentrations in citrus leaf tissue, and stabilized HLB infection in treated plants (Figure 1, Table 1).

Following monthly treatments of MFA over the experimental period (Figure 1), there was a consistent increase in disease severity over time in infected trees which is consistent with previous observations (Li et al., 2018). A DI reduction of 13.3% in MFA-treated trees was observed at 20months, but this infection rate did not significantly deviate from the untreated infected group (Figure 1B). Monitoring of the distribution of change in DI revealed that MFA, when all trees and infected trees were considered, resulted in significant stabilization of disease progression in the MFA-treated groups (Table 1). PR2 was significantly increased 2h after MFA treatment (Figure 2A), and PAL was also significantly increased 6h after treatment in response to MFA (Figure 2B). In addition to this, a transcriptomic assessment revealed an array of defense mechanisms that were upregulated in response to MFA. The previous work examined the impact of plant defense elicitors on HLB in a field evaluation. Li et al. (2015a) and Li et al. (2018) found a consistent reduction in disease severity when β-aminobutyric acid (BABA), benzothiadiazole (BTH), ascorbic acid (AA), and salicylic acid (SA) were applied to citrus. Induction of PR2 was observed after BABA and BTH applications, suggesting its implication in the defense against HLB.

The significant increases in Cu concentrations of citrus leaves following treatment with MFA (Table 2) observed in this study can be attributed to the Cu in the application which significantly increased the mineral in healthy and infected trees to 32.4 and 26.8ppm, respectively. Copper nutrition is critical in citrus development. Its deficiency is commonly associated with rapid growth in non-bearing trees following periods of high nitrogen fertilizer usage, and this leads to distorted leaf and twig growth and visual leaf chlorosis (Yruela, 2009; Hippler et al., 2017). Cu is poorly mobile in the phloem. Its application is recommended during periods of new vegetative growth (Marschner, 2012; Hippler et al., 2017) and should be regularly applied as foliar and ground fertilizer (Hippler et al., 2018). Cu is an important cofactor for many enzymes such as superoxide dismutase, amino oxidase, laccase, blue copper protein, and plastocyanin, but it also plays an important role for the transcription of protein trafficking machinery and oxidative phosphorylation (Yruela, 2005, 2009). Applications that increase Cu uptake in citrus could therefore be deemed valuable for agronomic purposes. Cu as a phytosanitary application has been used in the control of pathogens for many years (Russell, 2005; Hippler et al., 2017) and is considered a potent antimicrobial element (Vincent et al., 2018). The Cu portion of MFA may have contributed to the stabilization of HLB DI. In an investigation into the impact of micronutrients on HLB in C. sinensis, Da Silva et al. (2020) examined the use of copper hydroxide fertilizer on HLB in citrus. These authors reported that although leaf Cu concentrations reached 20ppm, there was no significant impact on HLB concentrations. The authors also reported that individual Cu fertilizer treatments helped mitigate the impact of HLB on starch metabolism a fundamental issue in trees infected with the disease. For this reason, MFA could potentially play a role in alleviating HLB symptoms by delivering higher Cu concentrations in infected citrus trees. However, the added elicitation of defense responses by MFA may have contributed to stabilization of disease progression. A previous investigation which compared CuSO₄, MFA without CuSO₄ and MFA reported that MFA had a significant impact on the induction of defense-related genes and on the control of powdery mildew in wheat (Twamley et al., 2019). The evidence presented suggests that MFA as a combined formulation (fermentation media and CuSO₄) gave the greatest control of disease and was the most effective at priming defense-related responses. This observation was further supported when MFA demonstrated greater antifungal activity against Zymoseptoria tritici compared to individual treatments of CuSO₄ and MFA without CuSO₄ at reduced concentrations in vitro (Twamley et al., 2021). Twamley et al. (2021) also indicated that MFA-treated plants resulted in grain yield and quality improvements in healthy MFA-treated plants. The authors suggest that this could be a function of peptide or amino acid complexes that might have greater bioactive properties and that could help limit the oxidative damage caused by Cu in plants. This suggests that the microbial preparation with CuSO₄ (MFA) possibly responds better than CuSO₄ or the fermentation media in isolation. The formulation may have a unique synergistic mechanism as a combined product. This may suggest that in the current study, the total preparation is conferring the benefits observed in the data for citrus and not just the CuSO₄ or fermentation media in isolation. However, a more comprehensive study should be conducted to confirm this. Excessive Cu use will continue to be regulated against due to its negative environmental impact. Issues such as accumulation and the selection of Cu resistant pathogens in the natural environment are of primary concern (Lamichhane et al., 2018). Environmentally safe microbial preparations that can prime plant natural defenses may play a role in reducing the use of Cu as a phytosanitary application.

Transcriptomic analysis gives an insight into what cellular processes are dormant or active in response to either a disease, challenge, or a nutritional treatment. In the control vs. infected trees, there were several differentially expressed genes associated with HLB infection. In previous studies conducted in greenhouses, HLB infection significantly impacted between 604and 633 genes (245–589 upregulated and 22–350 downregulated; Kim et al., 2009; Mafra et al., 2013; Fu et al., 2016; Hu et al., 2017). In the present study, HLB infection impacted a total of 171 genes (150 upregulated and 21 downregulated), in a field infection model. The most common issues associated with HLB are alterations in transcriptional processes related to defense, oxidative stress, carbohydrate metabolism, cell structure and organization, transcription factors, hormone signaling, and phloem blockage (Kim et al., 2009; Albrecht and Bowman, 2012; Mafra et al., 2013; Martinelli et al., 2013; Fu et al., 2016; Hu et al., 2017). Each of these processes are believed to contribute to the rapid development of disease symptoms, ultimately leading to declining tree health and death. Similar patterns for many of the previously reported correlations (Table 3 and Supplementary File 1) were observed in this study. Peroxidase, thaumatin-like protein, beta-galactosidase 1-like, hypersensitive-induced response protein 2-like, and elicitor-induced gene product homologue were upregulated in response to HLB infection. These are indicators of defense and oxidative stress-related responses in citrus and have previously been associated with HLB infection (Kim et al., 2009; Fu et al., 2016). Pathways associated with secondary metabolite biosynthesis were also upregulated, including multiple PAL genes, flavanone synthase genes, isoflavone 4′-O-methyltransferase-like gene, (E)-beta-farnesene synthase, and a trans-resveratrol di-O-methyltransferase-like gene. Phenylpropanoids are both associated with defense and the syntheses of flavonoids and terpenes which both play a role in the attraction of pollinating insects, defense against herbivories, and in the establishment of disease resistance. Previous studies have indicated that HLB impacts phenylpropanoid and terpene synthesis (Albrecht and Bowman, 2012; Fu et al., 2016; Hu et al., 2017).

Genes involved with carbohydrate metabolism and cell wall modification were also significantly impacted. Early nodulin, extensins, probable polygalacturonase, sucrose synthase, xyloglucan endotransglucosylase/hydrolase, cellulose synthase, and pectinesterase inhibitor were included in the group of upregulated genes responding to HLB infection in this study. The previous work has identified transcriptomic signatures that demonstrate the impact of HLB on carbohydrate metabolism (Fu et al., 2016, Hu et al., 2017) where expansins, xyloglucan endotransglucosylase/hydrolase proteins, cellulose synthesis, and pectinase-related genes were previously reported as differentially impacted in response to HLB. Genes involved in hormone signaling were also significantly impacted between control and HLB-infected trees; specifically, three gibberellin-regulated genes and a HVA22-like gene were upregulated in the HLB-infected tissue. Auxin-responsive protein, brassinosteroid-regulated gene, ethylene-induced esterase homologue, and an ethylene-responsive transcription factor were all downregulated. Phytohormones are believed to be impacted due to the cellular processes that are distorted as a result of changes in cell growth and defense responses. Zheng and Zhao (2013) saw similar changes in phytohormone production including auxin-, ethylene-, and gibberellin-related genes which may play a role in the mediation of citrus responses to HLB.

Many of the genes impacted following MFA treatment are associated with a strong localized and systemic defense responses. These included direct defense genes and numerous genes involved in the regulation of biotic defense responses. The defense stimulatory response may have supported the 13.3% reduction in DI (Figure 1). Disease progression was also significantly stabilized (Table 1) in MFA-treated trees. The transcriptome analysis revealed an increase in many direct defense-related enzymes but also a series of pathways involved in secondary metabolite biosynthesis and the production of antimicrobial compounds (Tables 4 and 5, and Supplementary File 1). Microbial elicitors have been successful in the past in the induction of resistance to bacterial, fungal, and oomycete pathogens (Thakur and Sohal, 2013). MFA contains both bacterial and yeast fermentation media, and both components may play a role in initiating gene transcription responses and have benefited plants in the past (Wiesel et al., 2014). The application of a yeast suspension to Arabidopsis saw the development of resistance to both Pseudomonas syringae and Botrytis cinerea infections (Raacke et al., 2006). The study also uncovered the ability of the yeast application to stimulate the expression of several plant defense systems including SAR, detoxification, and the jasmonate/ethylene pathways (Raacke et al., 2006). Yeast cell wall extracts are also well-documented inducers of plant defense, observed to regulate plant stomatal closure and mediate ROS responses following their use in various plant models (Khokon et al., 2010; Wiesel et al., 2014). A study that investigated ROS generation in rice in response to a N-acetylchitooligosaccharide demonstrated its role in plant defense elicitation (Kuchitsu et al., 1995). In Arabidopsis, chitin elicitors also increased ROS generation while contributing to both fungal and bacterial suppression (Egusa et al., 2015). The benefits of bacterial derived elicitors for plants have also been reported. They trigger induced systemic resistance (ISR) in plants and have other secondary benefits to plant health (Choi and Klessig, 2016; Ek-Ramos et al., 2019). ISR is understood to be important in cell wall thickening or destruction of infected cells, which helps cut off nutrients and access to invading pathogens. This has been effective against a range of plant pathogens in greenhouse and field settings (Lugtenberg et al., 2002; Aloo et al., 2019).

Phenylpropanoids produce an array of secondary metabolites derived from intermediates of the shikimate pathway (Fraser and Chapple, 2011). They contribute to a plant’s response to biotic and abiotic stimuli. PAL catalyzes the non-oxidative deamination of phenylalanine to trans-cinnamate and directs carbon flow from the shikimate pathway to general phenylpropanoid biosynthesis (Vogt, 2010). Phenylpropanoids are also understood to be involved in a plants overall defense strategy. They form preformed defenses, inducible defense responses, physical barriers and act as signaling molecules (Dixon et al., 2002). They are also well documented in plants in response to pathogens and in the development of resistance to disease. This data provides evidence that MFA may result in the beneficial upregulation of the phenylpropanoid pathways, and several phenylalanine ammonia-lyase-like genes were upregulated and complimented with shikimate O-hydroxycinnamoyltransferase-like, 4-coumarate--CoA ligase-like, and feruloyl CoA ortho-hydroxylase. There were significant increases in the lignin synthetic genes including caffeoyl-CoA O-methyltransferase-like, cinnamoyl-CoA reductase 1-like, and caffeic acid 3-O-methyltransferase-like which could be indicative of enhanced lignin formation. Laccase-7-like and blue copper protein was among the top 10 highest upregulated genes in healthy and infected trees where MFA was applied (Tables 4 and 5). Laccases are multi-copper enzymes that catalyze substrate oxidation and that reduce molecular oxygen to water. Laccases are also involved in Cu-facilitated lignification of cell walls contributing to defense, structure, and rigidity (Janusz et al., 2020). Xu et al. (2019) indicated that citrus laccases respond to environmental stress and are involved in lignin synthesis, and Cu ions are bound in several sites in laccases by Type 1 blue copper proteins (Printz et al., 2016), making blue copper proteins essential in the formation and function of laccases. Based on the transcriptomic evidence provided in this study, MFA potentially may play a crucial role in the activation of the phenylpropanoid pathway and facilitates lignin formation in citrus cell walls.

In this study, several PR genes were upregulated by MFA, including chitinase, endochitinase, endo-1,3-beta-glucosidase, osmotin, peroxidases, and thaumatin-like proteins (PR2, PR3, PR5, PR8, and PR9). Plants produce several direct defense mechanisms when initially challenged by pathogens. The importance of PR gene expression in citrus biotic responses has previously been reported (Campos et al., 2007) and highlights their role in pathogen defense responses. PR gene expression was expected to be upregulated as the data generated in the initial RT-qPCR experiment revealed significant upregulation of PR2 prior to microarray transcriptomic analysis (Figure 2).

Plant hormones are known to be important in plant growth, development, and cell signaling. They are also understood to be crucial in plant defense responses (Bari and Jones, 2009). In this study, there were several differentially expressed genes involved with plant hormone synthesis and regulation in response to MFA. This included genes associated with IAA, ethylene, and abscisic acid. There were also a small number of genes associated with brassinosteroids, salicylic acid, and jasmonic acid. Ethylene is produced from methionine by ACC synthase (ACS) and ACC oxidase (ACO) from its precursor 1-aminocyclopropane-1-carboxylic acid (ACC). The results presented herein revealed that S-adenosylmethionine synthase-like genes ACC synthase and a homologue were upregulated in response to MFA. Two ACC oxidases and ACC oxidase homologues were upregulated in response to MFA. Ethylene-responsive transcription factor-like genes 1A (2), 1B (2), 5, and 113 were also upregulated in MFA-treated plants. These ethylene-responsive transcription factors bind to the GCC-box pathogenesis-related promoter responding to pathogens and increasing the expression of PR genes (Zhou et al., 1997; Fujimoto et al., 2000; Sakuma et al., 2002).

This work provides evidence that MFA when applied to infected and healthy citrus trees increases Cu concentration, stabilizes HLB disease progression, and increases the expression of defense-related genes 2h after its application in a field setting. Transcriptomic assessment revealed that MFA increased the expression of 565 genes in healthy citrus trees and 909 genes in HLB-infected trees. This included traits associated with plant growth, development, and defense. There are very few successful applications that can be integrated into citrus production systems to help manage HLB. Provided herein are promising data suggesting a microbial preparation may help slow the progression of HLB in a field scenario. Further work is required to determine the mode of action of the application. Due to the limits in phytosanitary use of Cu, future work will address whether the efficacy of CuSO₄ contained in MFA is enhanced by its microbe-derived fraction, thus allowing for a reduced Cu input in the environment. Work will also aim to determine the impact of MFA on mature HLB-infected citrus trees in multiyear studies. MFA may also play a role in the management of other problematic pathogens; future research will aim to determine whether MFA has a broader application in agronomy.

The original contributions presented in the study are publicly available. This data can be found here: ArrayExpress accession E-MTAB-10919.

RL, KDo, SB, and KDa designed set up and managed the experiment and its logistics. RL, KDo, UC, WS, and ED applied treatments, sampled the trial, and prepared the samples for mineral and qPCR analysis. RL, KDo, and KE prepared and ran the transcriptomic samples. RL, RM, SB, KH, and KDa managed and supported the overall study. RL, KH, and KDa prepared and wrote the manuscript. All authors contributed to the interpretation of the data prior to preparing the manuscript. All authors contributed to the article and approved the submitted version.

This work was funded in full by Alltech and Alltech Crop Science. Alltech provided the research facilities and financial support for this work.

All the authors were paid employees of Alltech during their contributions to this project. Alltech innovate and manufacture crop input products for the agronomic industry.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.