First historical genome of a crop bacterial pathogen from herbarium specimen: Insights into citrus canker emergence

Abstract

Over the past decade, ancient genomics has been used in the study of various pathogens. In this context, herbarium specimens provide a precious source of dated and preserved DNA material, enabling a better understanding of plant disease emergences and pathogen evolutionary history. We report here the first historical genome of a crop bacterial pathogen, Xanthomonas citri pv. citri (Xci), obtained from an infected herbarium specimen dating back to 1937. Comparing the 1937 genome within a large set of modern genomes, we reconstructed their phylogenetic relationships and estimated evolutionary parameters using Bayesian tip-calibration inferences. The arrival of Xci in the South West Indian Ocean islands was dated to the 19th century, probably linked to human migrations following slavery abolishment. We also assessed the metagenomic community of the herbarium specimen, showed its authenticity using DNA damage patterns, and investigated its genomic features including functional SNPs and gene content, with a focus on virulence factors.

Fig 1. Citrus sp. specimen MAU 0015151 (HERB_1937), Mauritius Herbarium.

MAU 0015151 Citrus sp. specimen (HERB_1937) was collected from Mauritius in 1937 and deposited in the Mauritius Herbarium. Leaf areas displaying typical symptoms of Asiatic citrus canker are highlighted with blue dotted frames.

Fig 2. Major steps performed for characterization and integration of our herbarium sample into genomic analyzes.

See Material & Methods for more details on the workflow processed for HERB_1937 in this study. Abbreviations: CTAB, cetyl trimethylammonium bromide; DNA, deoxyribonucleic acid; BAM file, binary alignment map file; BLAST, basic local alignment search tool; NCBI nt database: national center for biotechnology information nucleotide database; SNP, single-nucleotide polymorphism.

Fig 3. Metagenomic composition of HERB_1937 historical specimen.

Proportions of reads assigned to Homo sapiens (5.4%), Citrus sp. (21.0%), Xci (1.2%). Others (12.4%): reads unassigned at the species level; unassigned reads (60.1%). Table: reads unassigned at the species level were assigned to the family (for Beijerinckiaceae) or genus level (for all others) and belong, for 0.18% to 1.57% of the aligned reads to the domain bacteria; “Others (<0.25% all reads)” include reads assigned to different plant, fungi, vertebrate, bacteria and phage genera (each identified genus totalizing less than 0.25% of all reads).

Fig 4. Coverage plots for the reconstructed HERB_1937_Xci chromosome and plasmids (pXAC33, pXAC64) sequences.

From inside to outside, a light to dark blue scale (delimited by a white line) represents 1, 1–5, 5–15, 15-35-fold coverage (Xci chromosome) and 1, 1–5, 5–35, 35-90-fold coverage (plasmids). Red rings indicate no identified coverage (depth = 0). SNP positions between the respective reconstructed and reference sequences are indicated (orange line). Accession numbers for Xci reference strain IAPAR 306: NC_003919.1 (chromosome), NC_003921.3 (plasmid pXAC33) and NC_003922.1 (plasmid pXAC64).

Fig 5. HERB_1937_Xci post-mortem DNA damage patterns.

Post-mortem DNA damage patterns were measured on historical HERB_1937_Xci (full, dotted or dashed blue lines for chromosome, pXAC33 and pXAC64 respectively) and compared with three modern Xci strains isolated from SWIO in 2012, 2013 and 2015 respectively (red lines, see results and S1 Table for full description). (A) Fragment length distribution (nucleotides; relative frequency in arbitrary units). (B) Deamination percentages of the first 25 nucleotides from the 5’ (C to T substitutions) and 3’ (G to A substitutions) ends, respectively. Dots: five most extreme nucleotides of the reads, showing a significant increase (towards the extremity) between each nucleotide along the five first or last positions of all HERB_1937_Xci reads. Along the five extreme nucleotides, reads matching to HERB_1937_Xci harboured significantly higher values than modern controls, and reads matching to the HERB_1937_Xci chromosome harboured significantly lower deamination rates than sequences matching to either plasmid (see results for statistics).

Fig 6. Tip-dating Bayesian inferences on historical and modern Xci genomes from the SWIO islands.

(A) Dated BEAST tree of 116 Xci modern strains sampled from the SWIO islands between 1978 and 2015 with historical HERB_1937_Xci (highlighted in green) built from 2,632 non recombining SNPs. Node support values are displayed by diamonds, in white for Posterior Probabilities below 0.9, in black for values above 0.9; node bars cover 95% Highest Probability Density of node height. The tree is structured in three lineages (A, B & C). Branches are collapsed and colored, according to the sample’s geographic origin, except lineage A, which is cartooned to help visualization. Tip labels include the geographic origin and, in cases of collapsed or cartooned branches, the number of samples. Map layer is from Natural Earth, available from https://www.naturalearthdata.com. (B) Linear regression of root-to-tip distance on year of sampling (tip date) test for temporal signal. Regression lines are plotted in blue when integrating historical HERB_1937_Xci genome and in red (dotted lines) when not. Grey areas correspond to their confidence interval. Associated values are the regression equation, adjusted R² (Adj R²) and p-value. (C) Boxplot distribution of root age, with (left) and without (right) integrating historical HERB_1937_Xci in the dating inference, and associated statistical comparisons. Boxes represent 25^th to 75^th percentiles, Minimum-Maximum intervals are displayed by a vertical bar and outliers as circles.