Bacteria-specific modified nucleoside is released and elevated in urine of patients with bacterial infections

Abstract

Over 170 types of chemical modifications have been identified in cellular RNAs across the three domains of life. Modified RNA is eventually degraded to constituent nucleosides, and in mammals, modified nucleosides are released into the extracellular space. By contrast, the fate of modified nucleosides in bacteria remains unknown. In this study, we performed liquid chromatography-mass spectroscopy (LC-MS) analysis of modified nucleosides from the RNA of 23 pathogenic bacteria, revealing 2-methyladenosine (m²A) as a common bacteria-specific modified nucleoside detected in all bacterial RNAs. Under normal culture conditions, bacteria did not actively release most modified nucleoside species, but robustly released nucleosides, including m²A, following addition of antibiotics or immune cells. These results indicate that m²A is released following bacterial lysis. Intraperitoneal injection of mice with m²A increased detectable levels of m²A in the urine, indicating that mammals can effectively excrete m²A. Additionally, mice infected with wild-type E. coli showed higher levels of m²A in their urine than mice infected by m²A-deficient rlmN KO E. coli. This suggests that m²A from the infected bacteria is excreted in the urine. Lastly, clinical studies using urine samples from febrile patients revealed significantly elevated levels of m²A during bacterial infections, and these values did not correlate with inflammation severity markers, such as white blood count (WBC) and C-reactive protein (CRP). This study reports the mammalian metabolism of modified nucleosides derived from bacterial RNA, and the elevation of urinary m²A in patients with bacterial infections.

Importance: This study reveals the differences in the fate and release of modified nucleosides in bacteria and mammals. Additionally, our study highlights that external bacteria-damaging factors, such as antibiotics and phagocytosis by host immune cells, promote the release of bacteria-specific modified nucleosides. Furthermore, we found that m²A was elevated in the urine from animal models of bacterial infection and the urine of patients with bacterial infections. Collectively, this work spans basic biology and clinical science, offering valuable insights into the fate of modified nucleosides in bacterial systems and their relevance to infectious diseases.

Keywords: LC-MS; RNA modification; bacterial infection; biomarker; modified nucleoside.

Fig 1

Profiling RNA modifications in 23 bacterial species and three mammalian cell lines. (A) Venn diagram classified by use of each modified nucleoside by eubacteria and mammals. mcmo⁵U: 5-methoxycarbonylmethoxyuridine, ms²i⁶A: 2-methylthio-N⁶-isopentenyladenosine, s⁴U: 4-thiouridine, s²C: 2-thiocytidine, m³U: 3-methyluridine, m⁴Cm: N⁴,2′-O-dimethylcytidine, m⁴₂cm: N⁴,N⁴,2′-O-trimethylcytidine, m²A: 2-methyladenosine, m⁶t⁶A: N⁶-methyl-N⁶-threonylcarbamoyladenosine, m⁴₂C: N⁴,N⁴-dimethylcytidine, Ψ: pseudouridine, Cm: 2′-O-methylcytidine, m⁵U: 5-methyluridine, m⁵C: 5-methylcytidine, t⁶A: N⁶-threonylcarbamoyladenosine, Am: 2′-O-methyladenosine, Gm: 2′-O-methylguanosine, mnm⁵s²U: 5-methylaminomethyl-2-thiouridine, m¹A: N¹-methyladenosine, m⁶A: N⁶-methyladenosine, D: dihydrouridine, m¹G: N¹-methylguanosine, m²G: N²-methylguanosine, m⁷G: 7-methylguanosine, m²₂G: N²,N²-dimethylguanosine, m⁶₂A: N⁶,N⁶-dimethyladenosine, I: inosine, Um: 2′-O-methyluridine, i⁶A: N⁶-isopentenyladenosine, ac⁴C: N⁴-acetylcytidine, s²U: 2-thiouridine, m³C: 3-methylcytidine, cm⁵U: 5-carboxymethyluridine, m¹I: N¹-methylinosine, mcm⁵U: 5-methoxycarbonylmethyluridine, f⁵C: 5-formylcytidine, hm⁵C: 5-hydroxymethylcytidine, mcm⁵s²U: 5-methoxycarbonylmethyl-2-thiouridine, ncm⁵s²U: 5-carbamoylmethyl-2-thiouridine, and ncm⁵U: 5-carbamoylmethyluridine. (B) Heatmap of RNA modification levels from 23 bacterial species and three mammalian cell lines. LC-MS peak areas of modified nucleosides divided by unmodified nucleosides are shown. The color scale indicates the auto-scaled relative mean of three biological replicates. A. baumannii: Acinetobacter baumannii, A. hydrophila: Aeromonas hydrophila, A. xylosoxidans: Achromobacter xylosoxidans, C. freundii: Citrobacter freundii, C. koseri: Citrobacter koseri, C. striatum: Corynebacterium striatum, E. cloacae: Enterobacter cloacae, E. coli: Escherichia coli, K. aerogenes: Klebsiella aerogenes, K. oxitoca: Klebsiella oxytoca, K. pneumoniae: Klebsiella pneumoniae, M. morganii: Morganella morganii, P. aeruginosa: Pseudomonas aeruginosa, P. vulgaris: Proteus vulgaris, R. ornithinolytica: Raoultella ornithinolytica, S. aureus: Staphylococcus aureus, S. caprae: Staphylococcus caprae, S. dysgalactiae: Streptococcus dysgalactiae, S. epidermidis: Staphylococcus epidermidis, S. haemolyticus: Staphylococcus haemolyticus, S. hominis: Staphylococcus hominis, S. lugdunensis: Staphylococcus lugdunensis, S. maltophila: Stenotrophomonas maltophilia.

Fig 2

Modified nucleosides in the culture medium decrease during bacterial growth. (A, C) Heatmap and chart analysis of modified nucleoside levels in the culture medium from E. coli (ATCC25922) over time. The color scale indicates the auto-scaled relative mean of four biological replicates. Abbreviations refer to Fig. 1A. (B, D) Heatmap and chart analysis of modified nucleoside levels in E. coli (ATCC25922) lysates at each time point. The color scale shows the auto-scaled relative mean of four biological replicates. Abbreviations refer to Fig. 1A.

Fig 3

Modified nucleosides are secreted following treatment with bactericidal antibiotics. (A) Heatmap analysis of modified nucleoside levels in E. coli (ATCC25922) culture medium following treatment with antibiotics. CLA: clarithromycin, KAN: kanamycin, CAM: chloramphenicol. The color scale indicates the auto-scaled relative mean of four biological replicates. Abbreviations in the heatmap refer to Fig. 1A. (B) Measurement of bacteria-specific modified nucleosides in the culture medium. **: P < 0.01, ****: P < 0.0001, ns: not significant. Data were analyzed using a one-way ANOVA and Dunnett’s multiple comparison test.

Fig 4

Modified nucleosides are released following phagocytosis by differentiated THP-1 cells. (A) Heatmap analysis of modified nucleoside levels in the culture medium of THP-1 cells co-cultured with E. coli (ATCC25922). The color scale indicates the auto-scaled relative mean of six biological replicates. (B) Measurements of bacteria-specific modified nucleosides in culture medium. *: P < 0.05. Data were analyzed using a Mann–Whitney test.

Fig 5

Measurement of m²A levels in the urine of mice after intraperitoneal injection of synthetized m²A or infection with bacteria. (A) Measurement of m²A levels in the urine of mice after intraperitoneal injection of m²A. (B) Measurement of m²A levels in the urine of mice after intraperitoneal injection of E. coli (ATCC25922) or rlmN KO E. coli. n = 4; CPZ, mice orally administered with cefoperazone. (C) Measurement of m²A levels in the urine of mice 1 week after administration of CPZ. n = 4, **: P < 0.01. Data were analyzed using Mann–Whitney test.

Fig 6

m²A level in the urine as a potential biomarker of bacterial infection. (A) Measurements of bacteria-specific modified nucleosides in the urine of patients with bacterial infection at normal phase and febrile phase. LC-MS peak areas of modified nucleosides divided by urine creatinine are shown. *: P < 0.05. Data were analyzed using the Wilcoxon signed-rank test. (B) Measurements of bacteria-specific modified nucleosides in the urine of patients with bacterial infection or healthy volunteers. LC-MS peak areas of modified nucleosides divided by urine creatinine are shown. *: P < 0.05. Data were analyzed using the Wilcoxon signed-rank test. (C) ROC analysis of m²A levels in urine normalized by urine creatinine was performed for calculation of sensitivity and specificity. (D) Correlation of m²A levels in the urine normalized by urine creatinine and WBC or CRP values.