Nucleotidyltransferase toxin MenT extends aminoacyl acceptor ends of serine tRNAs to control Mycobacterium tuberculosis growth

Abstract

Toxins of toxin-antitoxin systems use diverse mechanisms to inhibit bacterial growth. In this study, we characterize the translation inhibitor toxin MenT3 of Mycobacterium tuberculosis, the bacterium responsible for tuberculosis in humans. We show that MenT3 is a robust cytidine specific tRNA nucleotidyltransferase in vitro, capable of modifying the aminoacyl acceptor ends of most tRNA but with a marked preference for tRNA^Ser, to which long stretches of cytidines are added. Furthermore, transcriptomic-wide analysis of MenT3 targets in M. tuberculosis identifies tRNA^Ser as the sole target of MenT3 and reveals significant detoxification attempts by the essential CCA-adding enzyme PcnA in response to MenT3. Finally, under physiological conditions, only in the presence the native menAT3 operon, an active pool of endogenous MenT3 targeting tRNA^Ser in M. tuberculosis is detected, likely reflecting the importance of MenT3 during infection.

Fig. 1. MenT3 NTase activity in vitro.

a MenT3 preferentially adds CTP to tRNA. Total tRNA (100 ng) of M. smegmatis (Msm) were incubated at 37 °C for 5 min with MenT3 (0.2 µM) and α-32P labeled nucleoside triphosphates (NTPs). b Promiscuous tRNA alteration by MenT3. About 1 µg of total RNA from M. tuberculosis (Mtb) or human cells, or 100 ng of tRNA extracts from E. coli were incubated with 0.2 µM MenT3 in the presence of α-32P labeled CTPs for 5 min at 37 °C. c Comparison of tRNA NTase activity among MenT toxins. MenT3 (0.2 µM) or MenT1 (5 µM), or MenT4 (5 µM) were incubated with 1 µg of total RNA of Msm for 20 min at 37 °C. Given the robust activity of MenT3, the sample was serial diluted 1/10, 1/100, or 1/1000 to facilitate visualization of MenT1 and MenT4 activity through phosphor exposure. Reactions were conducted in the presence of α-32P labeled CTP for MenT1 and MenT3, and α-32P labeled GTP for MenT4. Representative results of triplicate experiments are shown in panels (a–c). d, e tRNA 3′-end mapping. Five µg total RNA from Msm were incubated with MenT3 (2.5 µM) and 1 mM CTP for 20 min at 37 °C, and the tRNA-seq library was prepared and sequenced. Modified tRNA reads were quantified in samples with or without MenT3 and the modifications are given as a percentage of the total tRNA 3′-ends. Note that tRNA^Sec is not present in M. tuberculosis. The tRNA detected from two independent experiments are shown on the bottom of panel (e) and the number of cytidines added is shown in the inset at the right of the same panel (duplicate is shown in Source Data file). Source data are provided as a Source Data file.

Fig. 2. tRNA structural and sequence determinants for MenT3 activity in vitro.

a MenT3-tRNA preference in vitro. Purified α-32P labeled tRNA^Ser-4 and tRNA^Met-2 were incubated in the presence of 1 mM CTP with MenT3 (0.2 µM) at 37 °C for different time points, separated on a 10% urea gel and revealed by autoradiography. b tRNA-Seq mapping of purified tRNA^Ser-4 and tRNA^Met-2. Purified tRNA^Ser-4 and tRNA^Met-2 (20 ng µl⁻¹) were incubated with MenT3 (1 µM) and 1 mM CTP for 10 min at 37 °C and subjected to tRNA-seq. c Effect of MenT3 on tRNA^Ser-4 3’-end length variants. Purified α-32P labeled 3′∆CCA, ∆CA, and ∆A truncated tRNA^Ser-4 were incubated with MenT3 (0.2 µM) for 5 min at 37 °C, separated as in (a). d tRNA-seq mapping of truncated tRNA^Ser-4 ∆CCA, ∆CA, and ∆A variants from panel (b). e Impact of tRNA^Ser variable loop on MenT3 activity. Purified labeled tRNA^Ser-4 deleted for its long variable loop (ΔVL), and the tRNA^{Met-2 (VLSer-4)} chimera with the long variable loop of tRNA^Ser-4 (+VL^Ser-4) were incubated with MenT3 (0.2 µM) for 5 min at 37 °C, separated as in (a). f tRNA-Seq mapping of purified tRNA^Met-2 and tRNA^{Met-2(VLSer-4)} chimera from (e). Purified tRNA (20 ng µl⁻¹) were incubated with MenT3 (1 µM) and 1 mM CTP for 10 min at 37 °C and subjected to tRNA-seq. Representative results of triplicate experiments are shown in (a, c, e), and sequencing data from two independent experiments are shown in panels in (b, d, f). Source data are provided as a Source Data file.

Fig. 3. The length and orientation of the tRNA^Ser variable loop contribute to MenT3 preference.

a Model of the interaction between MenT3 and tRNA^Ser-4, or tRNA^Leu-3 generated using AlphaFold3. The pTM/ipTM of MenT3-tRNA^Ser-4 and MenT3-tRNA^Leu-3 were 0.91/0.85 and 0.89/0.77, respectively. A close-up view of the possible interaction between MenT3 and the variable loop of tRNA^Ser-4 and tRNA^Leu-3 is shown on the right panel. Residues in the arm structure of the tRNA^Ser-4 variable loop are depicted as sticks. Potential hydrogen bonds are indicated with black dashed lines, highlighting key interactions. b The four deletion mutants of tRNA^Ser-4 used to assess the importance of the length of the variable loop for MenT3-mediated alterations are presented. c Effect of MenT3 on tRNA^Ser-4, tRNA^Leu-3 wild-type and mutants. Purified α-32P labeled tRNAs were incubated with MenT3 (0.02 µM) in the presence of 1 mM CTP at 37 °C for 2 min. The modified tRNAs were separated on a 10% urea gel and visualized by autoradiography. A representative gel of three independent experiments is shown. d Insertion or deletion between the T-arm and the VL or tRNA^Ser-4 and tRNA^Leu-3, respectively. The tRNA^Ser-4 + GU has a GU dinucleotide insertion between the VL and the T-arm to mimic tRNA^Leu-3, and the tRNA^Leu-3ΔGU has a deletion of GU between the VL and T-arm to mimic tRNA^Ser-4. The effect of MenT3 on the different mutants is shown (c). Source data are provided as a Source Data file.

Fig. 4. MenT3 specifically targets tRNA^Ser in M. tuberculosis.

a Experimental conditions for tRNA-seq in M. tuberculosis. M. tuberculosis (Mtb) wild-type H37Rv strain and its isogenic mutant ∆menAT3 expressing MenT3 from the integrative pGMC vector were individually grown at 37 °C in 7H9 medium supplemented with 10% albumin-dextrose-catalase (ADC, Difco) and 0.05% Tween 80. When the OD₆₀₀ reached to about 0.5, the anhydrotetracycline inducer (Atc, 200 ng ml⁻¹) was added, and cells were collected after 0, 3, or 24 h incubation at 37 °C. Total RNA was extracted, and tRNA-seq was performed. b Percentage of modified tRNA per tRNA species identified for the mutant overexpressing MenT3 and panel c for the wild type strain. The names of the identified tRNA for both strains are shown on the left of (b). The data were presented as the mean value obtained from three independent experiments. A detailed view of the different modifications obtained for tRNA^Ser is shown in Supplementary Fig. S3. Source data are provided as a Source Data file.

Fig. 5. CCA-adding enzyme PcnA counteracts MenT3.

a Suppression of MenT3 toxicity in M. smegmatis by the antitoxin MenA3 (control) and PcnA. M. smegmatis strain containing the plasmid pGMC-MenT3 was transformed with pLAM12-based plasmids harboring RpH or PcnA of M. smegmatis. Cultures were serially diluted and spotted on LA plates with or without inducers (Atc, 100 ng ml⁻¹; Ace, 0.01%). Plates were incubated for 3 days at 37 °C. Shown are representative results from triplicate experiments. b Northern blot analysis of RNA extracts from M. smegmatis co-transformed with various plasmid combinations: pGMC/pLAM, pGMC-MenT3/pLAM, pGMC-MenT3/pLAM-PcnA, pGMC-MenT3/pLAM-RpH, pGMC/pLAM-RpH, and pGMC/pLAM-PcnA. Cultures of transformants were grown until an OD600 of 0.1 in fresh LB medium and induced with 100 ng ml⁻¹Atc and 0.2% Ace for 3 h at 37 °C. RNA was extracted and analyzed by Northern blot to detect the presence of specific tRNA^Ser species. Shown are representative results from triplicate experiments. c In vitro activity of the M. tuberculosis PcnA enzyme. tRNA^Ser-4∆CCA was incubated with 0.5 µM PcnA in the presence of 1 mM NTP at 37 °C for 10 min to test the CCA-addition nucleotidyltransferase (NTase) activity. To examine whether PcnA can counteract MenT3 by its pyrophosphorolysis activity, tRNA^Ser-4 and MenT3-modified tRNA^Ser-4 were incubated with 0.5 µM PcnA in the presence of 1 mM inorganic pyrophosphate (PPi) at 37 °C for 10 min. Shown are representative results from triplicate experiments. Source data are provided as a Source Data file.

Fig. 6. Steady-state tRNA^ser modification in M. tuberculosis by endogenous MenT3.

M. tuberculosis wild-type H37Rv, the isogenic ∆menAT3 mutant and ∆menAT3-menT3 complementation strain were individually grown at 37 °C in 7H9 medium supplemented with 10% albumin-dextrose-catalase (ADC, Difco) and 0.05% Tween 80. When the OD₆₀₀ reached to 0.5, the cells were collected. Total RNA was extracted to perform tRNA-seq. The percentage of modified tRNA per tRNA species identified is shown. The data were presented as the mean value obtained from three independent experiments. A detailed view of the different species obtained for tRNA^Ser is shown in Supplementary Fig. S6. Source data are provided as a Source Data file.