While significant efforts have been dedicated to study the bases of tolerance/resistance to the devastating HLB disease, we are still a long way from a comprehensive understanding of this process. To gain further insights into the underlying mechanisms of the disease, omics technologies constitute an indispensable research tool to understand plant tolerance to this pathogen. In this study, we use metabolomics to try to gain insight into the mechanism underpinning the resistance, by analysing the composition of phloem sap in citrus varieties with different level of susceptibility to HLB (Alves et al., 2021; 2022).
Regarding organic acids, we observed a global upward trend in both the resistant and partially resistant cultivars. This aligns with the finding of Killiny et al. (2017), who conducted a metabolite profile analysis of the “Sugar Belle” mandarin hybrid to investigate its relative tolerance to HLB in comparison to some of its ancestors. Their research revealed that “Sugar Belle” exhibited elevated levels of phosphoric and some organic acids, including malic and threonic acid. Furthermore, previous studies have also reported an increase in the presence of various organic acids in resistant varieties (Killiny, 2017), and CLas-infected leaves from susceptible cultivars (Albrecht et al., 2016). Therefore, our data, in line with these previous studies, suggest that organic acids may have a role in nutrient uptake from the soil and serve as a priming strategy for those cultivars to be more tolerant.
In our analysis, we found that malic and quinic acid were the most abundant organic acids, consistent with prior research (Jones et al., 2012). Notably, only quinic acid showed a trend of higher content in PR and R samples compared to S samples. This metabolite has previously been link to defence responses in citrus leaves and has been detected in fruits of semi-tolerant varieties to CLas (Jones et al., 2012; Killiny, 2017), although its specific role in countering this pathogen remains unknown. Additionally, the average percentage of each organic acid within the three categories revealed trends that may be associated with the level of tolerance. Among them, citric acid and anthranilic acid, found at higher accumulation levels in resistant cultivars in our analysis, have been reported as metabolites involved in stress tolerance and plant protection (Zahan et al., 2021; Köllner et al., 2010).
Higher levels of sugars have been previously associated with HLB susceptibility (Albrecht et al., 2016; Killiny, 2016). Although there was not a consistent pattern in sugar composition among the different categories in our analysis, we did observe an upward trend in total sugar levels in the susceptible genotypes. However, these metabolites are unlikely to be the limiting factor for the HLB vector, since no clear correlation between glucose, fructose, sucrose, and citrus susceptibility to CLas was found (Killiny, 2016, 2017). It´s worth mentioning that effects on carbohydrate metabolism have been previously described in different phloem-related plant-pathosystems, including HLB (Albrecht et al., 2016).
Irrespective of their role in protein biosynthesis, numerous amino acids have been reported to participate in plant response to different stresses (Trovato et al., 2021). In our study, we have observed distinct trends in the levels of certain amino acids that could be coupled to the degree of HLB resistance, with a majority of them belonging to the glutamate and aspartate families. Higher constitutive concentrations of amino acids from the glutamate family have previously been associated with tolerance and resistance to CLas. In our analysis, glutamate and 4-hydroxyproline showed an upward tendency in resistant cultivars. Both of these amino acids have been formerly correlated with defence responses and resistance to plant pathogens (Albrecht et al., 2020; Qiu et al., 2020; Kim et al., 2021; Deepak et al. 2007, 2010). In contrast, two other members of this family, ornithine and proline, displayed a downward trend in the resistant samples. Supporting this, ornithine has been associated with susceptibility in various pathosystems (Dhodary et al. 2022; Jiménez-Bremont et al., 2014). Proline upregulation, on the other hand, has been connected to CLas tolerance, as well as other biotic and abiotic stresses, and has been observed in infected plants compared to controls (Albrecht et al., 2020; Chin et al., 2020). However, in a metabolic profiling of the phloem sap from fourteen varieties with different levels of tolerance to CLas, only one tolerant variety exhibited higher proline levels (Killiny, 2017). The variability in the metabolic composition of the different varieties and the conditions used in the different studies make it challenging to draw consistent conclusions. Nonetheless, it is worth noting that CLas lacks the ability to synthesize proline, phenylalanine, tryptophan, cysteine, tyrosine, and histidine and other essential translation components. These components are crucial for the bacterium's replication and metabolic activity, and it must acquire them from the host (Mendonça et al., 2017; Zuñiga et al., 2020).
The aspartate family pathway encompasses the synthesis of amino acids such as lysine, asparagine, and threonine (Yang & Ludewig, 2014). Our metabolomics analysis revealed higher levels of threonine in the resistant cultivars as well as increased levels of asparagine in both the resistant and partially resistant cultivars. The essential role of threonine as a sensor of different metabolic and environmental signals and translator of these signals into specific functional outputs, including stress tolerance, has been previously reported (Muthuramalingam et al., 2018; Killiny, 2017; Killiny et al., 2017). However, it's worth noting that lower amounts of threonine in resistant Rutaceae genotypes in comparison to susceptible cultivars, have also been described (Killiny, 2016). Pepper asparagine synthase 1, the enzyme required for the production of asparagine from aspartate, was reported as essential for stress response (Hwang et al., 2011).
Lysine is a limiting essential amino acid in plants, and its biosynthesis has been a matter of study aimed at enhancing the nutritional value of crops. As mentioned above, lysine catabolism research has recently focused on tolerance to biotic and abiotic stresses. Lysine can be catabolized through several metabolic pathways (Fig. 6). In some plants, lysine is converted into the alkaloid cadaverine (Hartmann & Zeier, 2018; Arruda & Barreto, 2020). A second ubiquitous catabolic pathway in plants is the saccharopine pathway rendering acetyl-CoA and glutamate. Furthermore, the third pathway, with a central role in plant immunity, leads to the generation of N-hydroxypipecolic acid (NHP), a metabolite recently involved in plant SAR (Hartmann et al., 2018). Contradictory results have been reported describing the lysine accumulation levels in response to pathogens (e.g. Návarová et al., 2013; Albrecht et al., 2016, 2020; Killiny & Hijaz, 2016; Killiny, 2016). According to our results, among the essential amino acids, lysine represented on average 19.26%, 12.66%, and 11.40% in susceptible, partially resistant, and resistant samples, respectively. The significant difference in lysine levels between the resistant and susceptible cultivars, suggests a potential key role in plant defense.
Regarding the synthesis of NHP (N-hydroxypipecolic acid), several enzymes have been associated with its production. The aminotransferase ALD1 (AGD2-like defense response protein 1), (Hartmann & Zeier, 2018; Holmes et al., 2019), a reductase, SARD4 (Ding et al., 2016); and FMO1, an NHP-synthesizing pipecolate N-hydroxylase (Hartmann & Zeier, 2018; Holmes et al., 2019). Additionally, the saccharopine pathway, which is associated with both abiotic and biotic stress responses, can produce pipecolate through two enzymatic reactions catalyzed by lysine-ketoglutarate reductase/saccharopine dehydrogenase (LKR/SDH) and pyrroline-5-carboxylate reductase (P5CR) (Arruda & Barreto, 2020). These authors explored whether both pathways contribute to SAR activation by comparing the transcriptional response of key genes, including LKR/SDH, ALD1, SARD4, and FMO1 in Arabidopsis thaliana under various biotic and abiotic conditions. Under most biotic stresses, all the analyzed genes were upregulated. Conversely, only the saccharopine pathway was upregulated under abiotic conditions.
The observed differences in lysine levels between susceptible and resistant categories led us to study the potential involvement of both the saccharopine and NHP pathways in HLB tolerance. We monitored transcriptional response of key genes, including LKR/SDH, ALDH7B4, P5CR, SARD 4, and FMO1, in the leaves of citrus plants. Our results revealed significant upregulation of FMO1 in two of the resistant samples compared to the susceptible ones. As previously mentioned, FMO1 is the final enzyme of a pathogen-inducible L-lysine catabolic pathway in Arabidopsis thaliana. Koch et al. (2006) reported that the overexpression of FMO1 under the control of the 35S promoter increases the basal resistance to Pseudomonas syringae. Additionally, Mishina & Zeier (2006) demonstrated the crucial role of this enzyme in the establishing of SAR. Afterward, supporting this, it was reported that FMO1 catalyzes the conversion of pipecolic acid to NHP, which is a critical amino acid with a central role in the establishment of SAR (Hartmann et al., 2018). Interestingly, these authors did not detect this metabolite in unstressed Arabidopsis thaliana; however, NHP strongly accumulated in the infected leaves. This metabolite was also not detected in any healthy phloem sap samples we analysed. Concerning the pipecolic acid, we did not obtain any conclusive result indicating significant differences among categories.
Another lysine-catabolite is the α-amino adipic acid (Fig. 6). Its biosynthesis is dependent on the LKR/SDH and ALDH7B4 saccharopine enzymes. In our analysis, ALDH7B4 exhibited a decreased transcriptional response in the resistant genotypes. It's interesting to note that Návarová et al. (2013) previously suggested that the saccharopine pathway leading to α-amino adipic acid has no critical role in the resistance and SAR establishment after Pseudomonas syringae infection. Upstream of ALDH7B4, LKR/SDH converts lysine to α-aminoadipate semialdehyde and glutamate, which is a precursor for several metabolites related to stress including pipecolic acid (Arruda & Barreto, 2020).
Although this is a small picture especially due to the number of samples analysed, it allows us to speculate about the potential roles of the upregulation of FMO1 and the downregulation of ALDH7B4 transcriptional responses. This differential regulation could create a more favourable scenario to combat the bacteria, potentially leading to increase resistance in certain cultivars. Somehow, it could be a plant strategy to anticipate defence mechanisms and gain an advantage over the pathogen. By identifying specific metabolic steps and defining the pathways involved in tolerance and resistance, we can develop powerful tools not only for discovering potential antimicrobial compounds against CLas but also for expediting the development of new citrus cultivars with enhanced resistance to this devastating disease.