Fig 1.
Mycobacterial YCC protein/protein interactions.
(A) Binary, hetero-dimeric protein/protein interactions are indicated by double circles when found reciprocally, by using reverse bait/prey protein pairs, and by simple circles when identified only in one of the two possible combinations. Homo-oligomeric assemblies were found for all proteins tested (dotted patterns). Mycobacterial YCC CT subunits AccD1 to AccD6 are denoted “D1” to “D6” (grey), BT subunits AccA1 to AccA3 are denoted “A1” to “A3” (no background), and the AccE ε-subunit is labeled “E” (black). (B) Identification of YCC holo complexes. Arrows point to the protein component used as prey; protein/protein assemblies, for which binary interactions were identified by the reverse use of bait/prey are indicated by thick double arrows. For further details, see S2 Table.
Fig 2.
AccD1-AccA1 holo complex formation.
(A) SDS-PAGE showing the Ni-affinity eluate (lane 2), and after gel filtration (panel B) in lane 3; molecular markers are shown in Lane 1. (B) Size-exclusion chromatogram of purified AccD1-AccA1. The elution volume and molecular mass of the molecular weight calibration standards are indicated. The equation describing the linear regression line is f(x) = -0.16x + 4.6 and has a R2 = 0.99.
Fig 3.
Identification and biochemical characterization of the AccD1-AccA1 substrate 3-methylcrotonyl-CoA.
(A) Untargeted CoA-profiles of YCC knockout strains. Mass spectra of Co-A profiles of M. smegmatis wt and knockout strains. Insets show spectra at dashed line. The peak evident in the ΔaccD1-ΔaccA1 spectrum is that of 3-methylcrotonyl-CoA. (B) Steady-state kinetics of AccD1-AccA1 incubated with 3-methylcrotonyl-CoA, propionyl-CoA and acetyl-CoA. (C) Time course of the consumption of 3-methylcrotonyl-CoA by purified M. tuberculosis AccD1-AccA1 and AccD5-AccA3.
Fig 4.
Metabolite analysis upstream and downstream of 3-methylcrotonyl-CoA.
(A) The leucine degradation pathway in eubacteria, modified from pathway ko00280 of the KEGG database (www.kegg.jp/kegg-bin/show_pathway?map00280). (B) Metabolites upstream of the putative MCC activity. The metabolite 3-methylcrotonyl-CoA accumulates during leucine degradation in ΔaccD1-ΔaccA1 M. smegmatis. However, this does not lead to backlog accumulation of isovaleryl-CoA upstream of MCC catalysis during leucine degradation. An increased pool of 3-methylcrotonyl-CoA upon MCC blockage also triggers accumulates of hydroxyisovaleryl-CoA. (C) Metabolites downstream of the AccD1-AccA1 MCC activity: Both methylglutaconyl-CoA and HMG-CoA show generally higher levels during leucine-degrading conditions. The amounts are comparable between wt and ΔaccD1-ΔaccA1 strains, as one would expect for metabolites downstream of a blocked enzymatic function. For compounds that are commercially available, chemical standard metabolite concentrations in the metabolite extracts are given in μM. Otherwise, relative concentrations are indicated and the compounds are annotated based on their mass. Average concentrations and standard deviations were calculated from four independent bacterial cultures of each strain.
Fig 5.
Growth phenotype characterization and complementation of mycobacterial accD1 and accA1 genes.
(A) Growth of M. smegmatis wt and knockout strains in minimal 7H9 media supplemented with isovalerate (IVAL). The wt is in green; ΔaccD1-ΔaccA1 is in red; ΔaccD2-ΔaccA2 (control) is in blue. Error bars are in black. (B) Rescue of the M. smegmatis growth phenotype observed in isovalerate-containing minimal media by complementing ΔaccD1-ΔaccA1 (blue line) with the M. tuberculosis homolog of AccD1-AccA1 (red line). The wt is in green. Average values for the culture density were calculated from four replicates.
Table 1.
Dimensions of AccD1-AccA1 and YCC complexes with known high-resolution structures.
Fig 6.
Architecture of the Mycobacterium tuberculosis AccD1-AccA1 complex.
(A) Electron micrograph of negatively stained AccD1-AccA1 complex. (B) Selected class averages representing the most abundant views of the AccD1-AccA1 complex. (C) Histograms of particle dimensions. The individual dimensions are schematically indicated in relation to the particle appearance in the class averages.
Fig 7.
Structural conservation of the M. tuberculosis AccD1 CT active site.
(A) Overall MCC quaternary arrangement, as determined previously for MCC from P. aeruginosa [17]. The BT subunits of the two distal trimeric tiers are shown in different green shadings. The central hexameric CT subunit assembly is shown in different colors. One dimeric CT subunit assembly in grey/cyan has been depicted in the lower panel and the CT active site formed within the dimeric interface is indicated by sphere presentation of the modeled substrate 3-methylcrotonyl-CoA with atom-specific colors (carbon, green) [17]. (B) Sequence alignment of the predicted M. tuberculosis AccD1 3-methylcrotonyl-CoA substrate-binding segment with the respective sequence segments of the P. aeruginosa MCC and M. tuberculosis AccD2 of unknown function. The residues numbers indicated refer to the AccD1 sequence. Conserved residue positions are indicated with asterisks. Residue positions that are conserved in the two upper confirmed MCC sequences (Phe151, Phe158, Phe161 in M. tuberculosis AccD1) but not in the AccD2 sequence are indicated by arrows. The complete sequence alignment is shown in S9 Fig. (C) Residues that interact with 3-methylcrotonyl-CoA in the P. aeruginosa MCC model [17], as determined with PISA [47], are shown in stick presentation and are labeled. Color codes are as in panel A. (D) Homology model of the M. tuberculosis AccD1 CT active site. Residues that are invariant in an alignment of the CT sequences from P. aeruginosa and M. tuberculosis (S9 Fig.) are in red, demonstrating a very high level of conservation of the AccD1 active site; conserved residues are in orange, non-conserved residues are in grey.