An in vitro protocol for rapidly assessing the effects of antimicrobial compounds on the unculturable bacterial plant pathogen, Candidatus Liberibacter asiaticus

Abstract

Background: Most bacteria are not culturable, but can be identified through molecular methods such as metagenomics studies. Due to specific metabolic requirements and symbiotic relationships, these bacteria cannot survive on typical laboratory media. Many economically and medically important bacteria are unculturable; including phloem-limited plant pathogens like Candidatus Liberibacter asiaticus (CLas). CLas is the most impactful pathogen on citrus production, is vectored by the Asian citrus psyllid (ACP, Diaphorina citri), and lacks an effective treatment or resistant cultivars. Research into CLas pathogenicity and therapy has been hindered by the lack of persistent pure cultures. Work to date has been mostly limited to in planta studies that are time and resource intensive.

Results: We developed and optimized an in vitro protocol to quickly test the effectiveness of potential therapeutic agents against CLas. The assay uses intact bacterial cells contained in homogenized tissue from CLas-infected ACP and a propidium monoazide (PMA) assay to measure antimicrobial activity. The applicability of PMA was evaluated; with the ability to differentiate between intact and disrupted CLas cells confirmed using multiple bactericidal treatments. We identified light activation conditions to prevent PCR interference and identified a suitable positive control for nearly complete CLas disruption (0.1% Triton-X 100). Isolation buffer components were optimized with 72 mM salt mixture, 1 mM phosphate buffer and 1% glycerol serving to minimize unwanted interactions with treatment and PMA chemistries and to maximize recovery of intact CLas cells. The mature protocol was used to compare a panel of peptides already under study for potential CLas targeting bactericidal activity and identify which were most effective.

Conclusion: This psyllid homogenate assay allows for a quick assessment of potential CLas-disrupting peptides. Comparison within a uniform isolate largely eliminates experimental error arising from variation in CLas titer between and within individual host organisms. Use of an intact vs. disrupted assay permits direct assessment of potential therapeutic compounds without generating pure cultures or conducting extensive in planta or field studies.

Keywords: Anti-microbial peptide; Asian citrus psyllid; Bioassay; Candidatus Liberibacter asiaticus; Citrus; Huanglongbing; Unculturable bacteria.

Fig. 1

Procedure for antimicrobial peptide assay on Candidatus Liberibacter asiaticus with insect homogenate

Fig. 2

Time trial for full activation of PMAxx with LED illuminator and PMAxx concentration test. a CLas containing ACP homogenate was aliquoted and treated either with 25 μM PMAxx or water and illuminated for 0, 2.5, 5, 7.5, 10, 12.5, 15, 17.5 or 20 min. Samples were immediately taken for DNA extraction after treatment and taken as template for qPCR analysis of CLas 16 s rDNA. Three replicates for each treatment at each time point were run with the values averaged and shown with standard error of the means. The difference between PMAxx treated and untreated samples levels off after 12.5 min. b A group of homogenate aliquots was subjected to the final protocol PMAxx treatment conditions (25 μM PMAxx or water and 15 min illumination). All samples were supplemented with 20 pg of plasmid DNA after PMAxx treatment and qPCR amplified with plasmid specific primers. Shown are the average C_t values from 7 replicates with standard error of the means. Plasmid amplification as measured by C_t was not statistically different in the presence of PMAxx treated samples compared to water, indicating no effect of residual PMAxx

Fig. 3

Dose response curve to Triton-X 100 positive control and comparison to other cell lysis methods. a, b CLas containing ACP homogenate was treated with a serial dilution of the detergent Triton-X 100 (1–0.001%) or a control of sterile water (0%) for 4 h. C_t values for v-qPCR from each combination of treatments shown in a as the average of six replicates. Relative quantification of cells under each treatment compared to the untreated control and ΔC_t in PMAxx averages is shown in b. CLas cells in homogenate were also heat stressed for 15 min @ 95 °C or subjected to three rounds of freezing on liquid nitrogen followed by thawing in a 37 °C water bath. c is a comparison between the v-qPCR C_t values for each treatment compared to 4 h of 0.1% Triton-X 100. Each treatment is paired with a non-treatment control; Triton-X 100 with water and each temperature stress to an identical time period @ 25 °C. All three cell lysis treatments (marked with “*”) were statistically different than the paired control (P < 0.05)

Fig. 4

Effects of glycerol buffer concentration on CLas viability. a, b Homogenate was prepared with saline buffer or saline buffer + 12.5% glycerol. Relative quantification of intact cells by v-qPCR from two trials shown in a with the corresponding average qPCR values of raw homogenate, light without PMAxx and light with PMAxx in b. The average viable cell signal from saline + glycerol was statistically higher than saline only in non-parametric analysis when comparing six biological replicates. c Viability of CLas cells before and after Triton-X 100 treatment was tested in homogenates with 1% and 12.5% glycerol. Values and statistical groupings (P = 0.05 in non-parametric analysis). There was no statistical change on total cell isolation (No PMAxx control), viable cell (Water/PMAxx) or lysis through detergent (Triton-X/PMAxx) between glycerol concentrations. Values shown are the average of five biological replicates

Fig. 5

Testing of synthetic antimicrobial peptides on CLas. Summary data from assessment of three putative AMPs, 1 mM streptomycin and 0.1% Triton-X 100. All samples underwent a 4 h incubation and received PMAxx @ 25 μM with light @15 min. The results from three separate experiments are summarized here. Each experiment contained twelve replicate samples per treatment and control. Values (ΔΔC_t) are shown as the average difference between the intact/disrupted cell ratio (ΔC_t of CLas 16 s rDNA from PMAxx compared to no PMAxx) for the experimental treatment compared to the intact/disrupted cell ratio of the no peptide (water) control. * Results with statistically significant increase in ΔΔC_t compared to no-peptide control (P < 0.05) by non-parametric comparison