Role of succinyl substituents in the mannose-capping of lipoarabinomannan and control of inflammation in Mycobacterium tuberculosis infection

Abstract

The covalent modification of bacterial (lipo)polysaccharides with discrete substituents may impact their biosynthesis, export and/or biological activity. Whether mycobacteria use a similar strategy to control the biogenesis of its cell envelope polysaccharides and modulate their interaction with the host during infection is unknown despite the report of a number of tailoring substituents modifying the structure of these glycans. Here, we show that discrete succinyl substituents strategically positioned on Mycobacterium tuberculosis (Mtb) lipoarabinomannan govern the mannose-capping of this lipoglycan and, thus, much of the biological activity of the entire molecule. We further show that the absence of succinyl substituents on the two main cell envelope glycans of Mtb, arabinogalactan and lipoarabinomannan, leads to a significant increase of pro-inflammatory cytokines and chemokines in infected murine and human macrophages. Collectively, our results validate polysaccharide succinylation as a critical mechanism by which Mtb controls inflammation.

Fig 1. Detail of the covalent substituents modifying the arabinan domains of AG and LAM in Mtb.

The various chemical modifications found on AG, LM, and LAM are shown in colored boxes: Mannoside caps: yellow boxes; succinyl substituents: green; methylthioxylose (MTX): orange, galactosamine: gray. The succinylation of the mycolylated chains in AG has been reported to be diminished or absent compared to that of non-mycolylated chains [22]. Ara₅ motifs, not shown on this figure, are thought to be extended linear Ara₄ motifs harboring one additional Araf residue [10].

Fig 2. Biochemical analysis of LAM and AG prepared from WT Mtb CDC1551, the sucT mutant and the complemented sucT mutant strains.

(A) Electrophoretic mobility of LM and LAM. Lipoglycans purified from WT Mtb CDC1551, Mtb sucT::Tn, and Mtb sucT::Tn/pMVGH1-Rv1565c (Mtb sucT::Tn comp) were run on a 10–20% Tricine gel followed by periodic acid-silver staining. The results presented are representative of two independent SDS-PAGE runs using different lipoglycan preparations from each strain. (B) Succinate content of AG and LAM. GC/MS-based quantification of succinates and arabinose residues in the same LAM and mAGP samples prepared from the WT, mutant and complemented mutant strains. An assumption is made here based on other structural analyses (alditol acetate and glycosyl linkage analyses presented in S1 through S4 Tables) that the arabinan domains of LAM and AG are not significantly altered in the mutant. Results are expressed as average ± SD succinate/arabinose molar ratios from three technical replicates. Asterisks denote statistical differences (*p < 0.05; **p < 0.005; ***p < 0.0005, Student’s t-test). The two complemented mutant strains shown here are: Mtb sucT::Tn/pMVGH1-Rv1565c (sucT expressed under control of the hsp60 promoter from a replicative plasmid; Mtb sucT::Tn comp) and Mtb sucT::Tn/pNIP40b-Rv1565c (sucT expressed under control of the hsp60 promoter from an integrative plasmid; Mtb sucT::Tn comp-int). (C) NMR analysis of LAM. Shown are the 1D ¹H (i, iii and v) and 2D ¹H-¹³C (ii, iv and vi) HMQC NMR spectra of LAM purified from Mtb CDC1551 WT, Mtb sucT::Tn and Mtb sucT::Tn comp. Arrows point to the signals typifying succinates. The 1D ¹H spectra of the WT and complemented mutant LAM show the characteristic two pseudo-triplets (J = 6.5 Hz) of similar intensity at 2.50 and 2.65 ppm assigned to methylene groups of succinyl units (panels i and v). Their corresponding carbons were characterized at 34.7 and 33.3 ppm, respectively, on the 2D ¹H-¹³C HMQC spectra (panels ii and vi). These signals are absent from the sucT mutant 1D ¹H and 2D ¹H-¹³C HMQC NMR spectra (panels iii and iv). (D) Quantification of mannooligosaccharide caps in the LAM from WT Mtb, the sucT mutant and the complemented mutant strain, Mtb sucT::Tn comp. Left panel: Capillary electrophoresis cap profile of LAM prepared from the three strains. Right panel: Abundance of the different cap motifs per LAM molecule. White bars, WT Mtb CDC1551; black bars, Mtb sucT::Tn; grey bars, Mtb sucT::Tn comp. IS, internal standard, mannoheptose-APTS; AM, Manp-(α1→5)-Ara-APTS (mono-mannoside cap); AMM, Manp-(α1→2)-Manp-(α1→5)-Ara-APTS (di-mannoside cap); AMMM, Manp-(α1→2)-Manp-(α1→2)-Manp-(α1→5)-Ara-APTS (tri-mannoside cap). The results are shown are averages and standard deviations from three technical replicates. Asterisks denote statistical differences pursuant to the Student’s t-test (***p<0.0001; **p<0.005; *p<0.05. ns = not significant).

Fig 3. Evaluation of immune activation and chemokine/cytokine secretion by C3HeB/FeJ BMMΦ infected with Mtb CDC1551 WT, the sucT mutant and the complemented mutant strain.

C3HeB/FeJ BMMΦ were infected with either WT Mtb CDC1551, Mtb sucT::Tn (“SucT”) or Mtb sucT::Tn comp (“SucT comp”) and allowed to adhere for 2 h. Cells were washed to remove extracellular bacteria and subsequently harvested 48 h post-infection for activation marker, chemokine/cytokine and NO release analysis. (A) Levels of expression of activation markers for MHCII, CD80, CD86 and CD40 was determined by flow cytometry. (B) Culture supernatants were analyzed for chemokine/cytokine secretion by multiplex immunoassay using Luminex. (C) NO production was determined using the Griess Reagent test. Immune activation, chemokine/cytokine release and NO data for triplicate samples were analyzed using an ordinary one-way ANOVA with *p≤0.05, **p≤0.01, ***p≤0.005 and ****p≤0.001. Error bars represent the mean spanning the SD of each replicate. Experiment was repeated two times.

Fig 4. Evaluation of immune activation and chemokine/cytokine secretion by M-CSF-differentiated MDM infected with Mtb CDC1551 WT, the sucT mutant and the complemented mutant strain.

Human monocyte derived macrophages (MDM) were infected with either WT Mtb CDC1551, Mtb sucT::Tn (“SucT”) or Mtb sucT::Tn comp (“SucT comp”) for 2 h. Cells were then washed to remove extracellular bacteria and were subsequently harvested 48 h post-infection for activation marker, chemokine/cytokine and NO release analysis. (A) Levels of expression of activation markers for HLA-DR, CD80, CD86 and CD40 was determined by flow cytometry. (B) Cytokines and chemokines in culture supernatants were analyzed by multiplex immunoassay using Luminex. (C) NO secretion was determined by the Griess Reagent kit. Immune activation, chemokine/cytokine release and NO data for triplicate samples were analyzed using an ordinary one-way ANOVA with *p≤0.05, **p≤0.01, ***p≤0.005 and ****p≤0.001. Error bars represent the mean spanning the SD of each replicate. Experiment was repeated two times.