Additional file 1 of Bacterial genome-wide association study substantiates papGII of Escherichia coli as a major risk factor for urosepsis

Additional file 1: Fig. S1. Distribution of E. coli phylogroups (left) and Sequence Types (ST) (right) in male (n=251) (upper row) and female (n=574) (lower row) patients. Fig. S2. Distribution of E. coli phylogroups (left) and Sequence Types (ST) (right) in in female patients (n=574) younger than 40 years (n=52) (upper row) and older than 40 years (n=522) (lower row) patients. Fig. S3. Distribution of E. coli phylogroups (left) and Sequence Types (ST) (right) in invasive infections (n=261) (upper row) and non-invasive infections (n=574) (lower row) patients. Fig. S4. Core genome phylogeny of 825 E. coli strains. Columns represent (from left to right): the assigned phylogroup, the sequence type, phenotypic resistance against ceftriaxone, meropenem, fosfomycin, nitrofurantoin and ciprofloxacin. Fig. S5. Within host genetic diversity of E. coli strains isolated from the same clinical cases. a: core genome phylogeny of E. coli strains (n=225), isolated from the same clinical case (n=106), coloured by phylogroup. The numbers correspond to the case identifier and strains were only labelled, if they exhibited < 99.9% Average Nucleotide Identity to the strain isolated from the same clinical case. b: papG variant encoded by isolates which exhibited < 99.9% Average Nucleotide Identity to the strain isolated from the same clinical case. c: Average Nucleotide Identity of strains isolated from case 3. d: SNV of 10 picked isolates from three cases, either from urine or blood culture samples. Fig. S6. Average Nucleotide Identity values for unitigs identified in our bGWAS and mapping to papG (X-axis) and the reference sequences for the five papG variants (Y-axis). Fig. S7. a: Significance level and average effect size of genes with mapping unitigs identified as significant in a bGWAS including all clinical cases (n=751 complete observations) (right) and including cases for which the port of entry for bacteraemia could be assigned to the urinary tract (n=612 complete observations) (right). In the right figure only genes with a maximum -log10(p-value) > 11 are labelled. Genes with locus tags 100888-20_01189, 100033-19_04615 and 100033-19_04621 are labelled as papJ_2, papJ_3 and papI_2, respectively, as they were identified as such. The gene with the locus tags 100033-19_03452 are labelled as ‘hp’ (= hypothetical protein); Odds ratio estimates with 95% confidence intervals for b: Typical urinary tract infection symptoms (n = 717 complete observations with 213 events); c: Admission to the intensive care unit (n = 751 complete observations with 172 events); d: 30-day all cause mortality (n = 749 complete observations with 45 events); using the generalised linear model (GLM). e: Performance of GLM classifiers using ‘Invasive disease’ as outcome variable and the same dataset as and variables as in the GLM as input (751 complete observations with 210 events), either including the presence of papGII as a predictor or not. Error bars indicate the standard deviation and the means were compared using paired Wilcoxon tests. OR = odds ratio; CI = confidence interval; CCI = Charlson Comorbidity Index; ‘AUROC’: area under the receiver operating curve; ‘NPV’: negative predictive value; ‘PPV’: positive predictive value; ‘ns’ = not significant; ‘*’ = p-value < 0.05; ‘**’ = p-value < 0.01; ‘***’ = p-value < 0.001. Fig. S8. C-reactive protein concentration (a) and leucocyte count (b) measured in blood samples of cases, for which a papGII positive or a papGII negative E. coli strain was isolated from a urine or a blood culture samples. Leucocyte counts were measured on the day the urine / blood culture samples were taken. Fig. S9. (a) Bacterial cell count, (b), leucocyte count divided by bacterial cell count (c) nitrite status and (d) erythrocyte count divided by bacterial cell count measured in urine samples of cases, for which a papGII positive or a papGII negative E. coli strain was isolated from a urine or a blood culture samples. Fig. S10. Relative occurrence of papGII in isolates from patients younger vs. older than 40 years, in isolates from male vs. female patients and in isolates from patients which were immunosuppressed vs. patients which were not immunosuppressed. Fig. S11. Occurrence of MALDI-TOF mass peaks in spectra acquired from E. coli strains encoding no papG gene, encoding a papG variant other than papGII and encoding papGII. ‘Occurrence’ refers to the percentage of spectra per group in which a peak was detected. Each strain was measured in quadruplicate either on a Microflex Biotyper device, or an Axmina Confidence device. Masses are only depicted if detected in > 30% or < 25% of spectra for one or more of the groups. Fig. S12. Occurrence of MALDI-TOF mass peaks in spectra acquired from E. coli strains of different phylogroups. ‘Occurrence’ refers to the percentage of spectra per group in which a peak was detected. Each strain was measured in quadruplicate either on a Microflex Biotyper device, or an Axmina Confidence device. Phylogroups for which less than five strains were available (E1, E2 and G) were excluded from the plot. Masses are only depicted if detected in > 50% or < 25% of spectra for one or more of the groups. Fig. S13. Core genome phylogeny of the E. coli strains collected for this study (one strain per clinical case, n=825). Phylogroup assignment, Sequence Type (ST) (eight most frequent ones coloured, more rare STs in grey), papG variant, mass of HdeA, predicted from the amino acid sequence. Fig. S14. Results of the endpoint PCR assay (a) to test the functionality of the primers designed at centre 1. This also includes tests for the cross reactivity between papGII and papGIII primers. (b) to test the functionality of the rpoD primers designed at centre 2. Fig. S15. Evaluating the efficiency of primers and probes used in our qPCR assay (a) qPCR standard curves and values for the primer pairs gapC_2, papC_1, uidA and papGII_2 tested at centre 1. Each measurement was performed in triplicate. (b) Amplification plots for the two rpoD probes designed at centre 2. Measurements performed in quadruplicate. Fig. S16. Variants of primer and probe sequences detected in our genome collection (n=1,076). Sequences used in the qPCR assay are indicated in blue and alternative variants detected in the genomes are depicted in black. Variants were called using the variantcaller Freebayes via snippy and using a minimum coverage of 20x. Fig. S17. (a) Efficiency of the primer pairs in the single reaction (blue) and in a triplex reaction (orange) for the primers used at center 1 (gapC, papC and papGII). (b) Amplification curves of primers used at center 2: rpoD and papGII in duplex reactions and of papGII in a triplex reaction with rpoD and papC. Fig. S18. Comparison of the Ct-value yielded when processing urine pellets (n=24) using the QIAamp DNA Mini Kit and after boiling for 10 minutes. Supplementary Methods. Endpoint PCR, Quantitative PCR, Multiplexing the qPCR, Applying qPCR assay directly to urine samples, Screening of patient samples. Supplementary PCR Data: Evaluation of primer functionality.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.