Mycobacterial phosphodiesterase Rv0805 is a virulence determinant and its cyclic nucleotide hydrolytic activity is required for propionate detoxification

Abstract

Cyclic AMP (cAMP) signaling is essential to Mycobacterium tuberculosis (Mtb) pathogenesis. However, the roles of phosphodiesterases (PDEs) Rv0805, and the recently identified Rv1339, in cAMP homeostasis and Mtb biology are unclear. We found that Rv0805 modulates Mtb growth within mice, macrophages and on host-associated carbon sources. Mycobacterium bovis BCG grown on a combination of propionate and glycerol as carbon sources showed high levels of cAMP and had a strict requirement for Rv0805 cNMP hydrolytic activity. Supplementation with vitamin B12 or spontaneous genetic mutations in the pta-ackA operon restored the growth of BCGΔRv0805 and eliminated propionate-associated cAMP increases. Surprisingly, reduction of total cAMP levels by ectopic expression of Rv1339 restored only 20% of growth, while Rv0805 complementation fully restored growth despite a smaller effect on total cAMP levels. Deletion of an Rv0805 localization domain also reduced BCG growth in the presence of propionate and glycerol. We propose that localized Rv0805 cAMP hydrolysis modulates activity of a specialized pathway associated with propionate metabolism, while Rv1339 has a broader role in cAMP homeostasis. Future studies will address the biological roles of Rv0805 and Rv1339, including their impacts on metabolism, cAMP signaling and Mtb pathogenesis.

Keywords: TB complex bacteria; cAMP signaling; carbon metabolism; cyclic nucleotide phosphodiesterase.

Figure 1:. Biochemical activities of Rv0805 and CpdA_Ec.

(A) Purification of recombinant N-His-Rv0805 and N-His-CpdA_Ec. Lane M, broad range protein MW marker (Promega). (B) Activities of Rv0805 and CpdA_Ec using BnPP as a substrate at indicated pH. (C) Activities of Rv0805 and CpdA_Ec using BnPP as a substrate in the presence of 0.1 mM of indicated metal ions. (D) Cleavage of 2′, 3′-cAMP by Rv0805 and CpdA_Ec. 10 mM 2′, 3′-cAMP, 2′AMP, and 3′-AMP were served as standards. (E) Cleavage of 2′, 3′-cGMP by Rv0805 and CpdA_Ec. 10 mM 2′, 3′-cGMP and 2′GMP/ 3′-GMP mixtures were served as standards. (F) Cleavage of 3′, 5′-cAMP by Rv0805 and CpdA_Ec. 10 mM 3′, 5′-cAMP, 3′AMP, and 5′-AMP were served as standards. (G) Cleavage of 3′, 5′-cGMP by Rv0805 and CpdA_Ec. 10 mM 3′, 5′-cGMP and 5′-GMP were served as standards. (H) Reactions of 3′, 5′-cAMP with purified Rv0805 or CpdA_Ec were separated using HPLC and monitored at 254 nm. (I) Detection of choline cleaved from GPC incubated with various concentrations of Rv0805, CpdA_Ec, or BSA. (J) Comparison of effect of divalent cations and Rv0805 point mutations on GDPD activity. Statistical analysis of Panel C using two-way ANOVA with the Bonferroni correction. Compared each column to No metal control. ***, P <0.001.

Figure 2:. Mtb Rv0805 is required for optimal survival in an infected host

(A) Growth of WT BCG, BCGΔRv0805, and BCGΔRv0805::Rv0805 in Mycomedia adjusted pH at 5.5. (B) Infection of J774.16 macrophages with MtbH37Rv WT, H37RvΔRv0805, and H37RvΔRv0805::Rv0805. (C) Eight to ten-week-old C3HeB/FeJ female mice were aerogenically infected with 50–100 CFU of Mtb H37Rv WT, H37RvΔRv0805, and H37RvΔRv0805::Rv0805. At the indicated time points after infection, lung bacterial loads were assessed. A group of 8 to10-week-old C57BL/6 mice were used for enumerating the bacterial loads in the lungs at 16–24 hours post-infection to determine the inoculum size. Four mice per group per time point were studied. The bacterial burden was significantly lower in the lungs of mice infected with the ΔRv0805 deletion mutant compared with WT H37Rv strain at 1, 3 and 6 months post infection. (D and E) BCG (D) or H37Rv (E) strains were grown in Mycomedia or 7H12 media containing 10 mM glycerol, 100 μM cholesterol, or 10 mM propionate in St/Amb conditions. Data shown are means of at least 2 independent experiments, except for d10 datapoints in (D) Glycerol, Cholesterol, or Propionate conditions, which are from one experiment, and Mycomedia condition in (E), which is representative of two similar experiments. Error bars denote SD. *, P <0.05; **, P <0.01; ***, P <0.001; ****, P <0.0001. In panels A, D, and E, p values displayed are between BCG and BCGΔRv0805, unless otherwise shown. p values between KO and complement strains are all <0.05, except for d15 of BCG in media with cholesterol, d5 and 14 of H37Rv in media with propionate, and d21 of H37Rv in media with glycerol where there is no significant difference.

Figure 3:. Rv0805 facilitates bacterial growth in GP.

(A) The methyl citrate cycle (MCC) and methylmalonyl-CoA (MMCoA) pathways are two mechanisms of propionyl-CoA detoxification. Both pathways can generate TCA intermediates, but the TCA anaplerotic generating step of the MMCoA pathway requires B12. Rv0998 can acetylate and inactivate ACS, it is unknown if Rv0998 contributes to the growth defect of BCGΔRv0805. It is unknown if Pta and AckA in mycobacteria can generate propionyl-CoA in an analogous pathway to acetyl-CoA formation. PrpC, methyl citrate synthase; PrpD, methyl citrate dehydratase; Icl, isocitrate lyase; AckA, acetate kinase; Pta, phosphate acetyltranferase; ACS, acetyl-CoA synthetase; DHAP, dihydroxyacetone phosphate; α-KG, alpha-ketoglutarate. (B) Bacterial growth measured by OD₆₂₀ was determined for BCG strains (BCG, ΔRv0805, ΔRv0805::Rv0805) at day 7 in 10 mM propionate media with additives listed. Data shown are means of at least 2 independent experiments. Error bars denote SD. *, P <0.05; **, P <0.01; ***, P <0.001; ****, P <0.0001.

Figure 4:. BCGΔRv0805 growth defect is specific to GP.

BCG strains (BCG, ΔRv0805, and ΔRv0805::Rv0805) were grown under St/Amb condition in 10 mM glycerol alone, or 10 mM glycerol with 10 mM of specified carboxylic acid. 10 mM glycerol with 10 mM pyruvate served as a control for a non-carboxylic acid carbon. Only glycerol and propionate (GP) caused a large growth defect of BCGΔRv0805, indicating Rv0805 is required for a propionate specific pathway. Data shown are means of 2 independent experiments. Error bars denote SD. *, P <0.05. P values displayed are between BCG and BCGΔRv0805, unless otherwise shown. Growth of BCG and BCGΔRv0805 was not statistically significant on d10. P values between KO and complement strains on d7, 10, and 13 are all <0.05.

Figure 5:. Mutations within acetate metabolic genes and Rv0806c are associated with suppression of BCGΔRv0805 phenotype

(A) BCG, BCGΔRv0805, or BCGΔRv0805 suppressor isolates were grown to 7 days in GP. BCGΔRv0805 displayed its characteristic growth defect, while BCGΔRv0805 strains with suppressor mutations (ΔS) clones 1–13 displayed WT like growth. BCG and BCGΔRv0805 are representative of 3 independent experiments. * above isolate indicates those that were sequenced. (B) Schematic showing the fdg1-pta-ackA locus (Rv0407-Rv0409, BCG_0446-BCG_0448) or Rv0805-Rv0806c (BCG_0857-BCG_0858c). Base and amino acid changes associated with each of the four mutations are highlighted beneath schematic for affected locus. Each mutation is boxed. REF refers to BCG reference genome while ALT indicates mutant sequence. BCG, BCGΔRv0805, and ΔS isolates were grown to day 7 in 7H12 supplemented with 10 mM acetate (C) or propionate (D). Notably these isolates displayed less growth on acetate (C) as primary carbon source as they have mutations within acetate metabolic genes but grew better than the KO on propionate. (E and F) Growth of BCGΔRv0805 expressing an empty or ackA containing overexpression vector were grown in glycerol (E) or GP (F). Data shown are from at least 2 independent experiments. Error bars denote SD. *, P <0.05; **, P <0.01; ***, P <0.001.

Figure 6:. Rv0805 cNMP catalysis is required for growth and cAMP levels are restored by ΔRv0805 genetic suppression

(A) Intracellular cAMP from WT BCG grown St/Amb in 7H12 with indicated carbon sources [glycerol (10 mM), propionate (10 mM), GP (both 10 mM) or, GP with 10 μg/mL B12]. (B) To determine if a propionate dependent metabolic defect contributed to Rv0805 dependent growth defects, 10 μg/mL B12 was added. Addition of B12 restored growth of BCGΔRv0805 (OD₆₂₀ measured at day 13). All conditions were compared simultaneously in the same experiment at least once. Two additional biological repeats for propionate and GP are replotted from Fig. 2D or Fig. S7A, respectively. (C and D) Intracellular (C) and extracellular (D) fractions of cAMP measured from BCG strains grown to day 7 in GP media. (E and F) Intracellular (E) and extracellular (F) fractions of cAMP measured from BCG, BCGΔRv0805, or BCGΔRv0805 ΔS pta frameshift strains grown to day 7 in GP media were compared. Data shown are from at least 2 independent experiments. Error bars denote SD. *, P <0.05; **, P <0.01; ***, P <0.001. One sample from BCG WT in panel A was reanalyzed and used for comparison against BCGΔRv0805 samples in panel C, they were grown in the same experiment.

Figure 7:. Partial growth increase by depletion of cAMP by Rv1339 overexpression suggests an importance of Rv0805 localization (A and B)

Effect of overexpressing Rv1339 or Rv1339-NTT on intracellular (A) and extracellular (B) cAMP levels in BCGΔRv0805 grown to 10 days in GP. (C) Comparison of growth between BCG and BCGΔRv0805 containing either empty or Rv1339 overexpressing constructs. (D) Western blot to confirm the truncation of Rv0805 CTE (removal of G279-D318). (E) Role of Rv0805 CTE in bacterial growth in GP media. (F) comparison of total (intracellular + extracellular) cAMP levels of WT BCG, BCGΔRv0805, BCGΔRv0805::Rv0805, and BCGΔRv0805::Rv0805ΔCTE. Error bars represent SD. *, P <0.05; **, P <0.01

Figure 8:. Proposed model of Rv0805 function

Growth on GP promotes metabolic stress in TB complex mycobacteria that results in increased levels of cAMP (depicted by star) through activation of undefined ACs (Fig. 6A, Fig. S11B). Up and down arrows above the AC represent the unknown localization (cytoplasmic or membrane associated) of the AC or ACs that are responsible for the cAMP production in GP conditions. B12 and Pta-AckA pathways have opposing effects on this metabolic stress, as B12 limited and the Pta-AckA pathway promoted cAMP levels (Fig. 6A, E, F; Fig. S13). Rv0805 localization is mediated through its CTE (represented by dotted line) [42], which may enable Rv0805 to regulate a local pool of functional cAMP or other unknown cNMP substrate/s that are modulated in response to the GP metabolic stress. The Rv0805 regulated cNMP may influence an unknown effector to respond to the GP induced metabolic stress. This signaling cascade may be overstimulated in the absence of Rv0805, or additional cNMP pathways may be turned on. The addition of B12 or disruption of pta-ackA, which both serve to lower cAMP levels, can compensate for loss of Rv0805 (Fig. 6E, F; Fig. S13). cAMP regulated by Rv0805 is secreted through an unknown mechanism (Fig. 6D). Rv1339 is an additional mycobacterial PDE described herein that can regulate other pools of cAMP (additional cAMP) and may have some overlap with the Rv0805 regulated cAMP, as growth was improved but not fully restored by overexpressing Rv1339 in BCGΔRv0805 (Fig. 7A–C). The presence of an additional PDE that is not functionally interchangeable leads to the possibilities that the cNMP regulated by each PDE serves disparate cellular fates, a phenomenon exemplified by study of eukaryotic PDEs [90], that the cognate substrates of Rv0805 or Rv1339 differ in vivo, or that the two PDEs are present in different regions of the bacteria. Subsequent studies to test these possibilities should focus on comparison between the subcellular distribution of Rv0805 and Rv1339 and substrate preferences of each PDE. In addition, the possibility of Rv0805 interacting with a protein complex, like how PDEs function within eukaryotes, can be probed through co-immunoprecipitation experiments. Image generated using BioRender.