Skip to main content
The Journal of Biological Chemistry logoLink to The Journal of Biological Chemistry
. 2025 Jun 19;301(8):110394. doi: 10.1016/j.jbc.2025.110394

CryoEM structure of Rv2531c reveals cofactor-induced tetramer–dimer transition in a tuberculin amino acid decarboxylase

Jyoti Gupta 1, Tina Izard 1,2,
PMCID: PMC12329521  PMID: 40543586

Abstract

The survival of Mycobacteriumtuberculosis relies on its ability to adapt to dynamic and hostile host environments. Amino acid decarboxylases play a crucial role in these adaptations, but their structural and mechanistic properties are not fully understood. Bioinformatic analyses revealed that these enzymes exist in three distinct forms based on their domain organization. We used cryoEM at 2.76 Å resolution to show that Rv2531c exhibits unexpected oligomeric and conformational flexibility. The enzyme forms a tetramer with distinct open and closed conformations in its apo state, suggesting dynamic intersubunit interactions. Upon binding pyridoxal 5′-phosphate, the enzyme undergoes a dramatic structural rearrangement, transitioning into a dimer. These findings reveal a novel mechanism of oligomeric plasticity. We also uncover an amino-terminal domain that might play a role in this process. Our results provide critical insights into the structural adaptations that support bacterial persistence under intracellular stress. By elucidating the apo and pyridoxal 5′-phosphate–bound states of Rv2531c, we contribute to a deeper understanding of how M. tuberculosis navigates its challenging intracellular environment. These insights into the unique structural features of Rv2531c offer a foundation for targeting metabolic resilience in tuberculosis and open avenues for future studies on the role of this domain in pathogenesis.

Keywords: cryogenic electron microscopy, glutamate decarboxylase, Mycobacterium tuberculosis, pyridoxal 5'-phosphate, γ-aminobutyric acid


Tuberculosis continues to be a major global zoonotic disease that kills more than one million people every year and affects millions. This is due to the emergence of multi- and extensive drug-resistant tuberculosis strains, which make treatment challenging, and thus, there is a need to develop new treatment strategies. Although much has been done to understand the biology of tuberculosis, a significant fraction of the Mycobacterium tuberculosis genome is not well defined in terms of function, which means that more work has to be done to identify new possible drug targets.

Recent developments in M. tuberculosis research have been based on enzymatic pathways. For instance, the total ergothioneine biosynthetic pathway was recently elucidated, which revealed the importance of this essential thiol compound (1). Other developments, such as a glycerol-phosphate phosphatase (2), a terpene-containing nucleoside, and enzymes responsible for branched-chain amino acid catabolism and itaconate dissimilation (3, 4, 5), have also helped to understand the metabolism of the bacterium. Identifying a bifunctional C–C bond lyase involved in the catabolism of l-leucine and the detoxification of the host-derived antimicrobial itaconate shows the flexible defense strategy of M. tuberculosis (5, 6).

Pyridoxal 5′-phosphate (PLP)–dependent enzymes are particularly noteworthy for their remarkable catalytic versatility. These enzymes catalyze many reactions, including decarboxylation, transamination, Claisen condensation, and others, where PLP coordinates carbanions to enable the reactions (7, 8). This catalytic flexibility makes it difficult to comprehend substrate specificity and reaction outcomes because slight variations in the active site can significantly affect the enzyme's activity.

PLP-dependent enzymes also have entirely different three-dimensional structures according to their functions (9, 10). Some are common to other species, but others are new and associated with specific metabolic functions (11, 12, 13). These enzymes in M. tuberculosis could be essential for the pathogen to grow and propagate in the limited and hostile conditions of the host.

Specifically, glutamate decarboxylases (GADs) exhibit substantial structural and functional diversity across species, reflecting their organism-specific physiological roles. Bacterial GADs, including those from Escherichia coli (Protein Data Bank [PDB] entries 1pmm (14), 1pmo (14), and 1xey (15)), typically form pH-responsive hexamers optimized for function under acidic conditions. In contrast, eukaryotic GADs (PDB entries 3fz7, 2okk (16), 2okj (16), 3hbx (17), and 5gp4 (18)) are predominantly dimeric and often contain regulatory elements enabling calcium/calmodulin binding or membrane tethering. Despite these architectural differences, all GADs retain a conserved PLP binding site, where a lysine residue (e.g., K276 in E. coli GAD, K396/405 in human isoforms) forms a Schiff base with the cofactor. Structural variation in loop conformations and ligand occupancy further distinguishes these enzymes, with forms ranging from apo (e.g., PDB entry 3hbx (17)) to PLP- or substrate-bound states (e.g., PDB entry 2okk (16)). These adaptations serve context-specific functions: human GAD65 localizes to synaptic vesicles to regulate neurotransmitter pools, whereas GAD67 is constitutively active and cytosolic. GAD integrates γ-aminobutyric acid (GABA) synthesis with plant calcium-mediated stress signaling (e.g., PDB entry 5gp4 (18)). These examples underscore the functional specialization of GAD enzymes and support our rationale for examining M. tuberculosis Rv2531c under neutral cytosolic pH conditions, which are physiologically relevant for the bacterium during infection.

The structural features we identified in Rv2531c, an extended polypeptide chain and an unusual quaternary structure, indicate that it is a distinct member of the PLP-dependent decarboxylase family. Using cryoEM, we determined that Rv2531c forms a tetramer in two different conformations, open and closed, and undergoes a PLP-induced transition to a dimeric state, a process not previously observed in related decarboxylases. These findings reveal unique oligomeric plasticity and domain architecture, suggesting adaptation to M. tuberculosis' metabolic and regulatory environment.

To our knowledge, Rv2531c has not been biochemically characterized or structurally studied to date. Proteomic studies have identified Rv2531c in a triton-insoluble fraction (19), suggesting potential membrane association. Our study provides the first structural insight into this uncharacterized enzyme and lays the groundwork for future investigation into its function in M. tuberculosis biology and pathogenesis.

Results

Rv2531c is a novel extended form of decarboxylase

Rv2531c shares ∼30% sequence identity with other bacterial decarboxylases (Fig. 1A, Fig. S1) and represents a structurally distinct member of the decarboxylase family, characterized by its extended polypeptide sequence and unique domain architecture. In contrast to canonical bacterial decarboxylases, Rv2531c contains an additional amino-terminal domain that adopts a four-helix bundle fold and a central response regulator–like domain, neither of which are present in classical glutamate, lysine, or arginine decarboxylases. These features distinguish Rv2531c from known decarboxylases and suggest specialized structural or regulatory functions that may be adapted to M. tuberculosis' metabolic or environmental demands.

Figure 1.

Figure 1

Domain organization and oligomerization of Rv2531c.A, amino-terminal domain (NTD; residues 1–131), response regulator domain (RRD; 134–288) aka receiver domain, catalytic domain (291–741), carboxy-terminal domain (CTD; 748–947). Mycobacterium abscessus has no equivalent NTD, and Escherichia coli only has the catalytic and carboxy-terminal domains. B, size-exclusion chromatogram (SEC; pH 7.5) of the apo tetramer (red), which elutes as a tetramer and the PLP-bound Rv2531c (blue), which elutes as a dimer. C, apo Rv2531c elutes as a tetramer at pH 8.5 (top) or 7.5 (bottom). D, SDS-PAGE of SEC fractions from the chromatogram at the bottom of C (at pH 7.5). Lane 1, molecular weight marker; 2, nickel–nitrilotriacetic acid fraction prior to SEC; 3 to 5, void peak fractions; and 6 to 13, peak fractions. PLP, pyridoxal 5′-phosphate.

Rv2531c undergoes a PLP-dependent tetramer–dimer transition in solution

Decarboxylases are important PLP-dependent enzymes. Nearly all published studies have focused on the medium-length E. coli form. The longer form, represented by M. tuberculosis, has not been studied in detail. PLP-containing proteins have been classified into seven clusters or folds (10, 20), with family members forming oligomeric structures, such as dimers, tetramers, or pentamers.

The E. coli GAD GadB forms a pH-dependent hexamer, with structural states characterized by PDB entries 1pmm (14) (acidic, active form) and 1pmo (neutral, inactive form) (14). GadB adopts an open hexameric conformation at low pH with ordered amino-terminal helices (residues 3–15), forming intersubunit bundles that stabilize the active site. In contrast, the amino terminus becomes disordered at neutral pH, and the structured carboxy-terminus folds over the active site, effectively blocking it. This reversible structural rearrangement, involving the amino and carboxy termini, functions as a molecular switch that regulates enzymatic activity in response to environmental pH. GadB is cytosolic at neutral pH but relocalizes to the membrane under acidic conditions, a behavior driven by this amino-terminal conformational change (14). We determined the oligomeric state of Rv2531c in solution (Fig. 1, BD) by size-exclusion chromatography. We found that Rv2531c is a tetramer and that PLP-bound Rv2531c is a dimer in solution (Fig. 1B). Thus, PLP binding induces a transition from a tetramer to a dimer.

The PLP-bound dimer cryoEM structure of Rv2531c

Some GADs function as dimers, similar to Rv2531c. In mammals, both isoforms of GADs (GAD65 and GAD67; PDB entries 2okk (16) and 2okj, respectively) (16) are homodimers. Dimerization is essential for their function. The active site is located at the interface between the two subunits, as we observed for Rv2531c (Fig. 2, AC). Some bacterial GADs, like those from E. coli (GAD-A, PDB entry 1xey) (15), form hexameric structures. The hexameric assembly in E. coli GAD is pH dependent (14), with a triple helical bundle formed by the amino termini contributing to its stability under acidic conditions. Our Rv2531c GAD is a tetramer at pH 7.5 and 8.5 (Fig. 1C). While many GADs, including mammalian GAD65 and GAD67, function as dimers, others, such as the E. coli GAD, assemble into hexamers (15, 16). This structural diversity reflects adaptations to different physiological roles and environmental conditions.

Figure 2.

Figure 2

CryoEM structure of the PLP-bound Rv2531c dimer.A, cartoon drawing of the PLP-bound dimer. PLP is shown as spheres. Each Rv2531c protomer is colored spectrally (NTD, residues 1–131, green; RRD, 134–288, orange; catalytic domain [CD], 291–741, magenta; and CTD, 748–947, blue). B, cryoEM map of the Rv2531c dimer with C2 symmetry, front and back view as indicated. Each domain is colored differently. C, interface residues within the PLP-linked Rv2531c dimer. The catalytic domain of one protomer is colored purple. The other protomer is colored gray. CTD, carboxy-terminal domain; NTD, amino-terminal domain; PLP, pyridoxal 5′-phosphate; RRD, response regulator domain.

The unique amino-terminal domain of Rv2531c

To gain molecular insights into Rv2531c, we determined its apo and PLP-bound structures by cryoEM (Figs. S2, S3; Table 1; Movie S1). Initial attempts were hindered by particle denaturation at the air–water interface, leading to only a partial structure. To overcome this challenge, we employed three key strategies: (1) collecting micrographs near the edge of the hole, where the ice layer is thicker and beam-induced motion is minimized, (2) performing multiple rounds of heterogeneous classification to exclude denatured particles, and (3) applying C2 symmetry during data processing. These approaches allowed us to resolve the structure to 2.76 Å resolution to gain insights into the architecture and dynamics of this important decarboxylase.

Table 1.

Data collection and model refinement statistics for our Rv2531c structures

Parameter Open tetramer Closed tetramer PLP-bound dimer
Magnification 60,000 60,000 60,000
Total dose (e-per Å2) 60 60 60
Defocus range (μm) −0.8 to −2.4 −0.8 to −2.4 −0.8 to −2.4
Pixel size (Å per pixel) 0.36 0.36 0.36
No. of frames per movie 50 50 50
No. of curated micrographs 17,749 17,749 16,144
No. of particles in the final map 113,770 108,734 379,820
PDB entry 9N0P 9N0N 9N0O
EMDB entry EMD-48789 EMD-48787 EMD-48788
Map resolution (Fourier shell correlation of 0.143) 3.4 Å 3.3 Å 2.76 Å
Refinement statistics
 Polypeptide chains 4 4 2
 No. of non-H atoms 41,599 38,340 26,110
 Protein residues 2622 2417 1642
 Ligands 0 0 2 PLP
Map correlation coefficient (CC)
 CC (mask) 0.76 0.80 0.85
 CC (box) 0.64 0.64 0.68
 CC (peaks) 0.53 0.56 0.66
 CC (volume) 0.74 0.77 0.79
 Mean CC for ligands 0.87
Validation
 MolProbity score 1.80 1.65 1.52
 Clash score 8.71 5.28 5.71
 Rotamer outliers 0.0023 0 0
 Cβ outliers 0 0 0
 CαBLAM outliers 0.0235 0.0213 0.0075
Ramachandran plot
 Allowed 0.0475 0.0508 0.0332
 Favored 0.9525 0.9471 0.9668
 Outliers 0 0.0021 0
RMSD
 Bond length (>4σ) 0.003 (0) Å 0.004 (0) Å 0.003 (0) Å
 Bond angle (>4σ) 0.583° (4) 0.652° (3) 0.585° (0)

All data were collected on a JEOL cryoARM300 at 300 kV with a GATAN K3 detector using Au Flat R 1.2/1.3 grids.

The unique amino-terminal domain of Rv2531c of unknown function encompasses the first 133 amino acids of Rv2531c. It adopts a characteristic four-helix bundle, whereby four α-helices are arranged in an antiparallel up–down–up–down orientation (Fig. 3A). This four-helix bundle domain sits against the following response regulator domain (RRD; residues 134–288), aka receiver domain. The bundle is otherwise mainly solvent exposed to varying degrees in our three structures.

Figure 3.

Figure 3

CryoEM Rv2531c structure.A, cartoon drawing of the PLP-bound Rv2531c protomer. The four α-helices of the amino-terminal domain (NTD, colored in green) are labeled. The following response regulator domain (RRD) is shown in orange. The catalytic domain is colored and labeled in magenta. The carboxy-terminal domain (CTD) is shown in blue. B, Coulomb potential map of the PLP binding site. C, predicted binding mode of glutamate to the Rv253c1 dimer, obtained through molecular docking using AutoDock Vina within ChimeraX (36, 42). Glutamate and PLP are shown in gray; the catalytic domain of Rv253c1 is depicted in magenta, and the carboxy-terminal domain in blue. The most favorable binding conformation, exhibiting a docking score of −4.5, is shown. PLP, pyridoxal 5′-phosphate.

To assess structural similarity, we submitted the coordinates of the amino-terminal domain of Rv2531c (residues 1–133) to the DALI server (21). The top matches include four-helix bundle-containing proteins involved in varied functions, such as respiratory complexes (e.g., cytochrome bc1 subunit; PDB entry 6hu9 (22); Z = 9.2), flagellar assembly (e.g., FlgK; PDB entry 3kli (23); Z = 9.1), and de novo designed scaffolds (e.g., PDB entry 6w6x (24); Z = 8.7). Root mean square deviation values ranged from 2.1 to 4.1 Å across 80 to 100 aligned residues despite low sequence identities (2–10%), confirming the presence of a canonical antiparallel four-helix bundle. Given the wide range of functions associated with structurally similar domains, the specific role of the Rv2531c amino-terminal domain remains to be determined.

The response regulator/receiver domain structure of Rv2531c

The RRD, also known as the receiver domain, of Rv2531c (residues 134–288) typically adopts the Rossmann fold, which is characteristic of two-component signal transduction systems (Fig. 3A). This fold plays an important role in recognizing and binding cofactors in many enzymes, including PLP-dependent decarboxylases (25). The domain comprises five parallel β-strands forming a central β-sheet, flanked by α-helices on both sides. The β-strands are arranged in a β1–β5 core topology, where the β-sheet is sandwiched between α-helices in a βαβ motif. The RRD in Rv2531c adopts the canonical structure of many PLP-dependent decarboxylases.

The catalytic domain structure of Rv2531c

The catalytic domain of Rv2531c (residues 291–741) adopts a PLP-dependent decarboxylase fold characterized by a combination of α-helices and β-sheets forming a core structure that accommodates the PLP cofactor. This fold is a typical structural arrangement for enzymes catalyzing decarboxylation reactions and consists of a large β-sheet flanked by α-helices. The β-sheet acts as a scaffold to stabilize PLP, whereas the α-helices properly position active site residues for efficient catalysis. The PLP cofactor is embedded within the fold, interacting primarily with the lysine residue (K570) through a Schiff base linkage (Fig. 3B). This arrangement is characteristic of decarboxylases, facilitating the transfer of the amino group from the substrate and the decarboxylation process. The active site contains critical residues (such as lysine 570 and histidine 596), which help stabilize the reaction intermediate and assist in the proton transfer steps of the decarboxylation reaction (Fig. 3B). The main interacting protomer has the backbone nitrogen of T380 interacting with the O2P atom of PLP (3.1 Å). S381 forms two hydrogen bonds whereby its backbone amide interacts with O3P (3.3 Å), and its hydroxyl group forms a bond with O3P (2.9 Å).

In addition, S567 contributes two hydrogen bonding interactions to the O2P (3.2 Å) and O4P (3.2 Å) atoms. T606′ from the two-fold related subunit of Rv2531c (indicated by the prime) forms a hydrogen bond with the O1P group of PLP (2.9 Å). In comparison, S607′ contributes hydrogen bonds from O1P (2.5 Å) and O2P (3.2 Å) (Fig. 2C). The pyridine ring of PLP also engages in a combination of hydrophobic and electrostatic interactions. The imidazole side chain of H405 engages in π-stacking interactions with the imidazole ring and hydrophobic interactions with the methyl group of A495. The indole side chain of W496 is positioned near the pyridine ring, whereas the carboxylate group of D493 establishes an electrostatic interaction with the nitrogen atom of the ring. These interactions orient the pyridine ring for catalysis (Movie S2). The residues surrounding PLP create a precisely shaped binding cleft that accommodates the phosphate and pyridine groups of PLP, orienting it for decarboxylation of the external aldimine.

In our PLP-bound Rv2531c structure, the phosphate group of PLP participates in several hydrogen bonds, with residue alignments referenced to the E. coli GAD α (GAD-A, PDB entry 1xey) (15) (Fig. S4). The OP1 atom forms hydrogen bonds with T380 (aligned with S126) and H596 NE2 (aligned with H275) (Fig. S1). The OP2 atom interacts via hydrogen bonds with the N and OG atoms of S607 (aligned with S318) from the adjacent dimer subunit. The OP3 atom establishes hydrogen bonds with S567 N (aligned with S126) and S381 N (aligned with S127). D493 OD2 (aligned with D243) forms a hydrogen bond with the N1 atom of PLP. A hydrophobic interaction is observed between the side chain of H405 (aligned with Q163) and the pyridine ring of PLP.

Notably, the catalytic domain of Rv2531c shares structural features with GADs, especially in how PLP is coordinated at the active site (26). GADs also adopt a similar fold with an α/β barrel structure to support cofactor binding and the decarboxylation of glutamate to GABA. In these enzymes, lysine is critical in forming a Schiff base with PLP, as seen in Rv2531c. The active site residues are highly conserved compared with the GAD (Fig. S1).

Molecular docking of glutamate to the PLP-linked Rv2531c dimer revealed binding of glutamate's N1 atom via a hydrogen bond with PLP's O3 atom (Fig. 3C). Notably, PLP's phosphate oxygens are involved in a network of hydrogen bonds intermolecularly with S607 and T606 of one protomer and intramolecularly with S381 and S567 of the other protomer within the dimer.

Collectively, the catalytic domain of Rv2531c adopts a fold typical of PLP-dependent decarboxylases, resembling the structure of known decarboxylases (16, 27, 28). The fold and its cofactor-binding strategy are conserved across different decarboxylases, with the lysine residue playing a pivotal role in the enzyme's catalytic mechanism.

The carboxy-terminal domain structure of Rv2531c

The carboxy-terminal domain (residues 748–947) adopts an α/β fold that consists of β-strands forming an antiparallel sheet, with surrounding α-helices (Fig. 3A). This fold is less common among decarboxylases. Still, similar motifs are seen in other enzymes' regulatory and interaction domains (29). While the catalytic domain of decarboxylases, including GADs, is often highly conserved in structure and function, the carboxy-terminal domain in Rv2531c may represent a unique adaptation in M. tuberculosis.

Apo Rv2531c is a dimer of dimers in two states

We obtained two distinct tetramer Rv2531c structures with C2 symmetry (Figs. S2, S5; Fig. 4). The center in one of them has a solvent channel of over 26 Å (Fig. 4A). We call this structure the “open” tetramer, which is ∼16 Å taller than the “closed” tetramer (Fig. 4B). In our closed tetramer structure, the solvent channel and the respective catalytic domains of two protomers are in contact to fill that solvent channel. Two of the amino-terminal domains within either tetramer are disordered. This alternating flexibility of the amino-terminal domains suggests a functional role contributing to the tetramer's asymmetric structural arrangement as a dimer of dimers. Specifically, residues R56, R59, and S106 in the closed tetramer interact intermolecularly with D548′, E454′, and D666′. E545 engages in electrostatic intermolecular interactions with R59’’ of the other dimer within the tetramer (Fig. S5). Comparison of the two tetramers highlights a relative shift of about 9 Å of residues 603 to 635 for residues Q622, E626, and T568 interacting intermolecularly with R621′, T568′, and E626′ in the closed tetramer.

Figure 4.

Figure 4

Rv2531c has two stable tetrameric configurations. View of Rv2531c as an open (A) or a closed (B) tetramer. Each domain is colored differently (NTD, residues 1-131, green; RRD, 134-288, orange; catalytic domain, 291-741, magenta; and CTD, 748-947, blue). Top row, Coulomb potential maps. Bottom row, two opposite residing protomers showing all bonds. A, the open tetramer measures about 145 Å by 113 Å by 77 Å. B, the closed tetramer measures about 129 Å by 120 Å by 77 Å. CTD, carboxy-terminal domain; NTD, amino-terminal domain; RRD, response regulator domain.

The intramolecular interactions between the catalytic domains are similar in the PLP-bound dimer and the respective dimers of the tetrameric structures. In contrast, the other domains engage in unique intermolecular and intramolecular interactions in our three structures. Specifically, the catalytic and carboxy-terminal domains (residues 291–947) of one protomer of the dimer with the polypeptide chain superimpose well with the open (RMSD of 1.322 Å for 6,511 atoms) or closed (RMSD of 1.556 Å for 6,558 atoms) tetramers, respectively (Fig. 5A). Such superposition results in similar positions for the two carboxy-terminal domains. However, the amino-terminal domains in our three structures are distinct. They are pivoted toward one side by about 100° in the open structure, whereas they are rotated in the other direction by about −80° (Fig. 5, BD).

Figure 5.

Figure 5

Distinct protomer arrangements determine the oligomeric Rv2531c architecture. Residues 291 to 947 are shown in the same orientation in all panels. The catalytic and carboxy (CTD) terminal domains of the protomer of our dimeric (A and B) and tetrameric open (C) and closed (D) structures are colored spectrally from green to red as indicated. Residues 291 to 947 are almost identical in all three structures, with RMSD of 1.322 Å for 6511 atoms for the protomer of the open or closed (RMSD of 1.556 Å for 6558 atoms) tetramer relative to the protomer in the PLP-bound dimer shown in B. However, the amino-terminal domain (NTD, residues 1–131) and the response regulator domain (RRD; 134–288; blue), aka receiver domain, are pivoted toward one side by about 100° in (C) the open tetramer structure, whereas they are rotated in (D) the other direction by about −80° in the closed tetramer structure. PLP, pyridoxal 5′-phosphate.

Preservation of the active site in the dimer and its disruption in the tetramers

Superimposition of the respective dimers reveals a relative shift of H405 by ∼3.5 Å in the open tetramer (Fig. S6A) and ∼2 Å in the closed tetramer (Fig. S6B) compared with its position in the PLP-bound dimer (Fig. S6C). Similarly, W496 is displaced by ∼6 Å in the open tetramer and ∼4 Å in the closed tetramer (Fig. S6D). H569 undergoes substantial displacements of ∼15 Å in the open tetramer and ∼9 Å in the closed tetramer (Fig. S6E). Notably, active site residues T606 and S607 show significant relative movements, with T606 shifting by ∼10.5 Å in the open tetramer (Fig. S6F) and ∼6 Å in the closed tetramer (Fig. S6G) and S607 moving by ∼16 Å in the open tetramer and ∼6 Å in the closed tetramer (Movie S3). Thus, tetramerization alters the relative orientation of active site residues, effectively collapsing the active site geometry.

Discussion

The discovery of Rv2531c as a novel, extended form of decarboxylases in M. tuberculosis is important to understanding the diversity and specificity of PLP-dependent enzymes in this pathogen. Thus, our study describes the unique structural characteristics of Rv2531c that differentiate it from canonical decarboxylases, such as GADs, which may be associated with its function in M. tuberculosis.

Short and intermediate decarboxylases do not have an additional amino-terminal domain with a novel four-helix bundle fold, which is not seen in typical GADs. This domain may be regulatory or structural and may be an evolution adaptation to the pathogen's environment. We further explored this domain by performing a DALI search using its atomic coordinates. Despite low sequence identity, the top structural matches include proteins with diverse biological functions, including signaling, protein transport, and structural scaffolding. This confirms that the domain adopts a canonical four-helix bundle architecture, a structurally versatile motif in many cellular contexts. Although the structural homologs do not reveal a straightforward, functional assignment for the Rv2531c amino-terminal domain, the variety of similar folds suggests that it could contribute to the enzyme's oligomeric plasticity or be a regulatory interface. The RRD also highlights the functional complexity of Rv2531c, which has a Rossmann fold, characteristic of two-component signal transduction systems that may couple decarboxylation activity to environmental signals.

The catalytic domain of Rv2531c has all the features of PLP-dependent decarboxylases, such as an α/β barrel fold and conserved active site residues. A detailed comparison with GADs from E. coli and mammalian enzyme isoforms shows that PLP-binding interactions are well conserved, particularly with lysine residues that form a Schiff base. However, the extended architecture of Rv2531c is altogether different and is likely to influence the function and oligomerization of the enzyme differently. Surprisingly, our cryoEM structures revealed that Rv2531c forms a stable tetramer in solution, unlike the dimeric forms of mammalian GAD65 and GAD67 and the hexameric forms of bacterial GAD. To our knowledge, no apo structures of mammalian GAD65 or GAD67 have been reported, as all published structures include PLP (16). Thus, our comparison refers to the PLP-bound forms, where mammalian GADs form dimers stabilized by interprotomer interactions around the PLP-binding site. In contrast, PLP-bound Rv2531c forms a stable tetramer and exhibits a cofactor-induced tetramer-to-dimer transition. These observations suggest that Rv2531c possesses unique oligomeric plasticity not observed in other structurally characterized GADs.

We also found two distinct tetrameric states of Rv2531c, which we termed “open” and “closed,” with different interdomain interactions and solvent channel accessibility. This structural plasticity may allow Rv2531c to change its activity in response to environmental or metabolic signals in the host, which may be an advantage to the pathogen.

To our knowledge, this is the first structural characterization of a GAD from M. tuberculosis, and the observed oligomeric transitions may reflect species-specific adaptations. M. tuberculosis is known to reprogram central metabolism under host-imposed stress and nutrient limitation (30, 31), and this unique conformational flexibility could contribute to the dynamic regulation of decarboxylase activity in such conditions. While it is tempting to speculate that this structural plasticity may contribute to environmental responsiveness in vivo, further biochemical or genetic experiments will be needed to test whether such transitions directly affect M. tuberculosis pathogenesis.

Surprisingly, our cryoEM structures revealed that Rv2531c undergoes a drastic tetramer to dimer transition upon binding of PLP. Unlike conventional GADs with a stable oligomeric structure, Rv2531c has a cofactor-induced structural change from a tetramer to a dimer. Although the functional consequences of this transition remain unknown, the data suggest that PLP binding stabilizes the active site and promotes a catalytically competent state. This hypothesis awaits validation by enzymatic and kinetic assays.

Docking studies of glutamate with the Rv2531c dimer revealed that the PLP's phosphate group interacts with serine and threonine residues. In contrast, the glutamate's amino group forms a hydrogen bond with PLP. This interaction positions glutamate for transamination with PLP, the first step in decarboxylation. The observed binding mode implies a dynamic, active site where PLP may reorient upon substrate binding to achieve catalytic competence, a phenomenon common in PLP-dependent enzymes (32, 33). These findings suggest an active form of Rv2531c.

Rv2531c contributes to the knowledge of the structural and functional diversity of PLP-dependent enzymes. Though GAD has been shown to decarboxylate l-glutamate to form GABA, Rv2531c has an extended sequence and oligomeric variability, suggesting it may have a more diverse set of functions. Whether this structural variability correlates with functional specialization in M. tuberculosis remains to be tested. Interestingly, the structural dynamics between its open and closed tetrameric states suggest potential regulation mechanisms. Furthermore, the substrate specificity of Rv2531c and catalytic efficiency compared with canonical decarboxylases, such as GAD, may elucidate its potential as a therapeutic target.

Collectively, Rv2531c is a novel class of PLP-dependent decarboxylases in M. tuberculosis with an extended sequence, a novel domain architecture, and a tetrameric structure that dissociates into a dimer upon binding to PLP. Our findings provide a structural framework for future biochemical and functional studies. However, conclusions regarding regulatory function, oligomeric transitions, and the role of the amino-terminal domain remain speculative at this stage. Further experimental validation through enzymatic assays, biophysical oligomerization studies, and in vivo functional analyses will be essential to establish the physiological role of the structural features described here.

Experimental procedures

Protein purification

Full-length Rv2531c (accession I6X4K0 in pET28 with six histidine residues at the amino terminus) was expressed in E. coli strain BL21(DE3) Rosetta2 (Novagen) at 18 °C for 20 h. Six one-liter cultures were grown and induced with 1 mM isopropyl β-d-1-thiogalactopyranoside at 18 °C for 20 h. Cells were harvested by centrifugation (6000 rpm, 25 min) and lysed by sonication in 180 ml lysis buffer (20 mM triethanolamine, pH 8, 500 mM NaCl, and EDTA-free protease inhibitor cocktail). Sonication was performed on ice at 75% amplitude for 3 min (5 s on, 10 s off cycles).

The initial lysate (180 ml) was clarified by centrifugation at 35,000 rpm for 30 min. The resulting pellet was resuspended and subjected to a second lysis in a 180 ml fresh lysis buffer. SDS-PAGE analysis indicated a significant amount of Rv2531c remained in the pellet after the first lysis.

The combined lysate (360 ml) was loaded onto a nickel–nitrilotriacetic acid column pre-equilibrated with lysis buffer using an ÄKTA chromatography system. The column was washed with buffer A (20 mM triethanolamine, pH 8, 300 mM NaCl) until the absorbance returned to baseline. Bound His-6-Rv2531c protein was then eluted with a linear imidazole gradient (20 mM triethanolamine, pH 8, 300 mM NaCl, 0.5 M imidazole, pH 8). Fractions of 1.5 ml were collected during the elution and analyzed by SDS-PAGE.

After the SDS-PAGE analysis, Rv2531c fractions were pooled and concentrated with 5% added glycerol. Aliquots were frozen in liquid nitrogen and stored at −80  °C. Before cryoEM grid preparation, Rv2531c was subjected to a Superose 6 Increase 10/300 column equilibrated with 20 mM phosphate buffer, 150 mM NaCl, 1 mM Tris(2-carboxyethyl)phosphine, pH 7.5. To generate the PLP-linked dimer, a 20-fold molar excess of PLP was added to the peak fraction in phosphate buffer (pH 7.5) and incubated for 2 h. The sample was then subjected to a Superose 6 Increase 10/300 column equilibrated with phosphate buffer (pH 7.5). The peak fraction obtained was used for grid preparation.

EM sample preparation and data collection

Au-Flat 1.2/1.3 300 mesh grids (Protochips) were glow discharged for 120 s at 15 mA using a PELCO easiglow (Ted Pella, Inc) and mounted on Leica GP2 cryogenic plunger (95% humidity, 8 °C). Rv2531c (3.5 μl at 0.7 mg/ml for the tetramer of 0.4 mg/ml for the PLP-bound dimer) was dispensed unto the carbon film side of the grid, incubated for 60 s, followed by blotting for 7 s and plunge freezing in liquid ethane, which was maintained at −183 K by liquid nitrogen. The grids were then clipped into cartridges. JEOL cryoARM300 was used to collect the movies. The condenser aperture was set to 20 μm with an Omega in-column energy filter, a slit width of 20 eV, and a zero-loss peak that aligned every 6 h. A total of 17,000 movies were collected for each sample. The movies were recorded using SerialEM (34) at −0.8 to −2.4 μm defocus using a K3 direct electron detector (Gatan) with super-resolution mode at a physical pixel size of 0.72 Å per pixel (super-resolution 0.36 Å per pixel) and magnification of 60,000. Fifty frames were captured per specimen area with 2 to 3 s exposure with 7 to 9 e/s/pixel and total specimen dose, 60 e/A2.

Data processing

The data were processed using cryoSPARC, v 4.3.1. (35). Super-resolution image stacks were aligned and binned by a factor of two and motion corrected using cryoSPARC to give a final pixel size of 0.72 Å per pixel. The contrast transfer function (CTF) was estimated using patchCTF in cryoSPARC. After manually curating images based on the ice thickness and CTF fit, the blob picker picked particles in cryoSPARC from the curated subsets of micrographs. Multiple rounds of 2D classification classified these particles. Featureless classes were removed at each step, resulting in cleaned particle stacks for further processing. An ab initio reconstruction allowed us to build five (for the tetramer) or three (for the PLP-bound dimer) initial 3D maps (C1 symmetry) from the complete dataset. The particles from classes with similar volumes were pooled and further subjected to separate heterogenous refinements. By visualizing in ChimeraX (36), one good volume was obtained after three heterogeneous refinements. Each volume was then further subjected to homogeneous refinement with applied C2 symmetry. These particles were subjected to symmetry expansion with applied C2 symmetry and a local refinement using C1 symmetry. C4 and D2 symmetry were also evaluated and triaged.

Model building and refinement

The AlphaFold (37) model of Rv2531c was used as the starting model. For both states of the Rv2531c tetramer, the AlphaFold model for residues 1 to 355 and 356 to 947 were separately fitted as amino-terminal and carboxy-terminal domains by rigid body docking into the cryoEM map using ChimeraX (36) followed by molecular dynamic flexible fitting using ISOLDE (38). The map for the open Rv2531c tetramer was flipped in the Z-axis to fit the AlphaFold model. For the Rv2531c dimer, the AlphaFold model for residues 1 to 947 was supplied by rigid body docking into the cryoEM map using ChimeraX (36). The subsequent model-building processes were performed manually using Coot (39). Further iterative structure refinements were done using the phenix.real_space_refine subroutine (40) from the PHENIX suite (41).

Molecular docking analysis

Molecular docking was performed to investigate the interaction between glutamate and the PLP-bound active site of Rv2531c. The cryoEM–derived structure of Rv2531c was prepared using Chimera (36, 42, 43) by removing nonstandard residues and water molecules, adding hydrogens, and optimizing ionization states. The structure of glutamate was retrieved from PubChem (CID entry 33032) and prepared using Chimera's Dock Prep tool. Docking was carried out with AutoDock Vina integrated with Chimera. The search grid encompassed the PLP-binding region with center coordinates (200 Å, 185 Å, and 195 Å) and dimensions of 40 Å × 40 Å × 30 Å. Default parameters were used. The best-ranked pose based on binding energy was selected for analysis. To contextualize the binding orientation, we aligned the catalytic core of Rv2531c with the glutarate-bound E. coli GAD structure (PDB entry 1xey (15)). Despite the flipped orientation of the docked glutamate relative to glutarate, both ligands occupy similar positions near the PLP cofactor. This alignment supports the plausibility of the docked conformation and suggests functional conservation in substrate binding.

Data availability

The final coordinates and the Coulomb potential maps for the PLP-bound Rv2513c dimer (PDB entry 9n0o and EMD-48788), open Rv2531c tetramer (PDB entry 9n0p, EMD-48789), and closed Rv2531c tetramer (PDB entry 9n0n, EMD-48787) structures have been deposited with the PDB and EMDB, respectively.

Supporting information

This article contains supporting information.

Conflict of interest

The authors declare that they have no conflicts of interest with the contents of this article.

Acknowledgments

T. I. is grateful for the support from grants from the National Institutes of Health (grant no.: R35 GM139604), the National Science Foundation (grant no.: MCB 2232523), and start-up funds provided to The Scripps Research Institute by the State of Florida. We are deeply indebted to Dr. Luiz de Carvalho (UF Scripps) for gifting the Rv2531c construct and for countless insightful discussions.

Author contributions

J. G. and T. I. conceptualization; J. G. methodology; J. G. validation; J. G. formal analysis; J. G. and T. I. investigation; T. I. resources; J. G. data curation; J. G. and T. I. writing–original draft; J. G. and T. I. writing–review & editing; J. G. and T. I. visualization; T. I. supervision; T. I. funding acquisition.

Funding and additional information

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Reviewed by members of the JBC Editorial Board. Edited by Wolfgang Peti

Supporting information

Supplement
mmc1.pdf (15.7MB, pdf)
Movie 1
Download video file (1.8MB, mp4)
Movie 2
Download video file (1.7MB, mp4)
Movie 3
Download video file (1.4MB, mp4)

References

  • 1.Seebeck F.P. In vitro reconstitution of Mycobacterial ergothioneine biosynthesis. J. Am. Chem. Soc. 2010;132:6632–6633. doi: 10.1021/ja101721e. [DOI] [PubMed] [Google Scholar]
  • 2.Larrouy-Maumus G., Biswas T., Hunt D.M., Kelly G., Tsodikov O.V., de Carvalho L.P. Discovery of a glycerol 3-phosphate phosphatase reveals glycerophospholipid polar head recycling in Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. U. S. A. 2013;110:11320–11325. doi: 10.1073/pnas.1221597110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Young D.C., Layre E., Pan S.J., Tapley A., Adamson J., Seshadri C., et al. In vivo biosynthesis of terpene nucleosides provides unique chemical markers of Mycobacterium tuberculosis infection. Chem. Biol. 2015;22:516–526. doi: 10.1016/j.chembiol.2015.03.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Buter J., Cheng T.Y., Ghanem M., Grootemaat A.E., Raman S., Feng X., et al. Mycobacterium tuberculosis releases an antacid that remodels phagosomes. Nat. Chem. Biol. 2019;15:889–899. doi: 10.1038/s41589-019-0336-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Wang H., Fedorov A.A., Fedorov E.V., Hunt D.M., Rodgers A., Douglas H.L., et al. An essential bifunctional enzyme in Mycobacterium tuberculosis for itaconate dissimilation and leucine catabolism. Proc. Natl. Acad. Sci. U. S. A. 2019;116:15907–15913. doi: 10.1073/pnas.1906606116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Kreutzfeldt K.M., Jansen R.S., Hartman T.E., Gouzy A., Wang R., Krieger I.V., et al. CinA mediates multidrug tolerance in Mycobacterium tuberculosis. Nat. Commun. 2022;13:2203. doi: 10.1038/s41467-022-29832-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Eliot A.C., Kirsch J.F. Pyridoxal phosphate enzymes: mechanistic, structural, and evolutionary considerations. Annu. Rev. Biochem. 2004;73:383–415. doi: 10.1146/annurev.biochem.73.011303.074021. [DOI] [PubMed] [Google Scholar]
  • 8.Du Y.L., Ryan K.S. Pyridoxal phosphate-dependent reactions in the biosynthesis of natural products. Nat. Prod. Rep. 2019;36:430–457. doi: 10.1039/c8np00049b. [DOI] [PubMed] [Google Scholar]
  • 9.Izard T., Fol B., Pauptit R.A., Jansonius J.N. Trigonal crystals of porcine mitochondrial aspartate aminotransferase. J. Mol. Biol. 1990;215:341–344. doi: 10.1016/s0022-2836(05)80355-9. [DOI] [PubMed] [Google Scholar]
  • 10.Liang J., Han Q., Tan Y., Ding H., Li J. Current advances on structure-function relationships of pyridoxal 5′-phosphate-dependent enzymes. Front Mol. Biosci. 2019;6:4. doi: 10.3389/fmolb.2019.00004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Morris V.K., Izard T. Substrate-induced asymmetry and channel closure revealed by the apoenzyme structure of Mycobacterium tuberculosis phosphopantetheine adenylyltransferase. Protein Sci. 2004;13:2547–2552. doi: 10.1110/ps.04816904. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Brown K.L., Morris V.K., Izard T. Rhombohedral crystals of Mycobacterium tuberculosis phosphopantetheine adenylyltransferase. Acta Crystallogr. D Biol. Crystallogr. 2004;60:195–196. doi: 10.1107/s0907444903025988. [DOI] [PubMed] [Google Scholar]
  • 13.Primi M.C., Tavares M.T., Klein L.L., Izard T., Sant'Anna C.M.R., Franzblau S.G., et al. Design of novel phosphopantetheine adenylyltransferase inhibitors: a potential new approach to tackle Mycobacterium tuberculosis. Curr. Top Med. Chem. 2021;21:1186–1197. doi: 10.2174/1568026621666210728094804. [DOI] [PubMed] [Google Scholar]
  • 14.Capitani G., De Biase D., Aurizi C., Gut H., Bossa F., Grutter M.G. Crystal structure and functional analysis of Escherichia coli glutamate decarboxylase. EMBO J. 2003;22:4027–4037. doi: 10.1093/emboj/cdg403. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Dutyshev D.I., Darii E.L., Fomenkova N.P., Pechik I.V., Polyakov K.M., Nikonov S.V., et al. Structure of Escherichia coli glutamate decarboxylase (GADalpha) in complex with glutarate at 2.05 angstroms resolution. Acta Crystallogr. D Biol. Crystallogr. 2005;61:230–235. doi: 10.1107/S0907444904032147. [DOI] [PubMed] [Google Scholar]
  • 16.Fenalti G., Law R.H., Buckle A.M., Langendorf C., Tuck K., Rosado C.J., et al. GABA production by glutamic acid decarboxylase is regulated by a dynamic catalytic loop. Nat. Struct. Mol. Biol. 2007;14:280–286. doi: 10.1038/nsmb1228. [DOI] [PubMed] [Google Scholar]
  • 17.Gut H., Dominici P., Pilati S., Astegno A., Petoukhov M.V., Svergun D.I., et al. A common structural basis for pH- and calmodulin-mediated regulation in plant glutamate decarboxylase. J. Mol. Biol. 2009;392:334–351. doi: 10.1016/j.jmb.2009.06.080. [DOI] [PubMed] [Google Scholar]
  • 18.Huang J., Fang H., Gai Z.C., Mei J.Q., Li J.N., Hu S., et al. Lactobacillus brevis CGMCC 1306 glutamate decarboxylase: crystal structure and functional analysis. Biochem. Biophys. Res. Commun. 2018;503:1703–1709. doi: 10.1016/j.bbrc.2018.07.102. [DOI] [PubMed] [Google Scholar]
  • 19.Mawuenyega K.G., Forst C.V., Dobos K.M., Belisle J.T., Chen J., Bradbury E.M., et al. Mycobacterium tuberculosis functional network analysis by global subcellular protein profiling. Mol. Biol. Cell. 2005;16:396–404. doi: 10.1091/mbc.E04-04-0329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Percudani R., Peracchi A. The B6 database: a tool for the description and classification of vitamin B6-dependent enzymatic activities and of the corresponding protein families. BMC Bioinformatics. 2009;10:273. doi: 10.1186/1471-2105-10-273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Holm L., Rosenstrom P. Dali server: conservation mapping in 3D. Nucleic Acids Res. 2010;38:W545–W549. doi: 10.1093/nar/gkq366. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Marek M., Shaik T.B., Heimburg T., Chakrabarti A., Lancelot J., Ramos-Morales E., et al. Characterization of histone deacetylase 8 (HDAC8) selective inhibition reveals specific active site structural and functional determinants. J. Med. Chem. 2018;61:10000–10016. doi: 10.1021/acs.jmedchem.8b01087. [DOI] [PubMed] [Google Scholar]
  • 23.Lam W.W., Woo E.J., Kotaka M., Tam W.K., Leung Y.C., Ling T.K., et al. Molecular interaction of flagellar export chaperone FliS and cochaperone HP1076 in Helicobacter pylori. FASEB J. 2010;24:4020–4032. doi: 10.1096/fj.10-155242. [DOI] [PubMed] [Google Scholar]
  • 24.Polizzi N.F., DeGrado W.F. A defined structural unit enables de novo design of small-molecule-binding proteins. Science. 2020;369:1227–1233. doi: 10.1126/science.abb8330. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Medvedev K.E., Kinch L.N., Dustin Schaeffer R., Pei J., Grishin N.V. A fifth of the protein world: rossmann-like proteins as an evolutionarily successful structural unit. J. Mol. Biol. 2021;433 doi: 10.1016/j.jmb.2020.166788. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Ueno H. Enzymatic and structural aspects on glutamate decarboxylase. J. Mol. Catal. B: Enzymatic. 2000;10:67–79. [Google Scholar]
  • 27.Burkhard P., Dominici P., Borri-Voltattorni C., Jansonius J.N., Malashkevich V.N. Structural insight into Parkinson's disease treatment from drug-inhibited DOPA decarboxylase. Nat. Struct. Biol. 2001;8:963–967. doi: 10.1038/nsb1101-963. [DOI] [PubMed] [Google Scholar]
  • 28.Chen H.J., Ko T.P., Lee C.Y., Wang N.C., Wang A.H. Structure, assembly, and mechanism of a PLP-dependent dodecameric L-aspartate beta-decarboxylase. Structure. 2009;17:517–529. doi: 10.1016/j.str.2009.02.013. [DOI] [PubMed] [Google Scholar]
  • 29.Kamtekar S., Hecht M.H. Protein Motifs. 7. The four-helix bundle: what determines a fold? FASEB J. 1995;9:1013–1022. doi: 10.1096/fasebj.9.11.7649401. [DOI] [PubMed] [Google Scholar]
  • 30.Chang D.P.S., Guan X.L. Metabolic versatility of Mycobacterium tuberculosis during infection and dormancy. Metabolites. 2021;11:88. doi: 10.3390/metabo11020088. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Warner D.F. Mycobacterium tuberculosis metabolism. Cold Spring Harb Perspect. Med. 2014;5 doi: 10.1101/cshperspect.a021121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Ngo H.P., Nguyen D.Q., Park H., Park Y.S., Kwak K., Kim T., et al. Conformational change of organic cofactor PLP is essential for catalysis in PLP-dependent enzymes. BMB Rep. 2022;55:439–446. doi: 10.5483/BMBRep.2022.55.9.090. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Ngo H.P., Cerqueira N.M., Kim J.K., Hong M.K., Fernandes P.A., Ramos M.J., et al. PLP undergoes conformational changes during the course of an enzymatic reaction. Acta Crystallogr. D Biol. Crystallogr. 2014;70:596–606. doi: 10.1107/S1399004713031283. [DOI] [PubMed] [Google Scholar]
  • 34.Mastronarde D.N. Automated electron microscope tomography using robust prediction of specimen movements. J. Struct. Biol. 2005;152:36–51. doi: 10.1016/j.jsb.2005.07.007. [DOI] [PubMed] [Google Scholar]
  • 35.Punjani A., Rubinstein J.L., Fleet D.J., Brubaker M.A. cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination. Nat. Methods. 2017;14:290–296. doi: 10.1038/nmeth.4169. [DOI] [PubMed] [Google Scholar]
  • 36.Meng E.C., Goddard T.D., Pettersen E.F., Couch G.S., Pearson Z.J., Morris J.H., et al. UCSF ChimeraX: tools for structure building and analysis. Protein Sci. 2023;32 doi: 10.1002/pro.4792. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Jumper J., Evans R., Pritzel A., Green T., Figurnov M., Ronneberger O., et al. Highly accurate protein structure prediction with AlphaFold. Nature. 2021;596:583–589. doi: 10.1038/s41586-021-03819-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Croll T.I. ISOLDE: a physically realistic environment for model building into low-resolution electron-density maps. Acta Crystallogr. D Struct. Biol. 2018;74:519–530. doi: 10.1107/S2059798318002425. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Emsley P., Cowtan K. Coot: model-building tools for molecular graphics. Acta Crystallogr. D Biol. Crystallogr. 2004;60:2126–2132. doi: 10.1107/S0907444904019158. [DOI] [PubMed] [Google Scholar]
  • 40.Afonine P.V., Poon B.K., Read R.J., Sobolev O.V., Terwilliger T.C., Urzhumtsev A., et al. Real-space refinement in PHENIX for cryo-EM and crystallography. Acta Crystallogr. D Struct. Biol. 2018;74:531–544. doi: 10.1107/S2059798318006551. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Adams P.D., Grosse-Kunstleve R.W., Hung L.W., Ioerger T.R., McCoy A.J., Moriarty N.W., et al. PHENIX: building new software for automated crystallographic structure determination. Acta Crystallogr. D Biol. Crystallogr. 2002;58:1948–1954. doi: 10.1107/s0907444902016657. [DOI] [PubMed] [Google Scholar]
  • 42.Goddard T.D., Huang C.C., Meng E.C., Pettersen E.F., Couch G.S., Morris J.H., et al. UCSF ChimeraX: meeting modern challenges in visualization and analysis. Protein Sci. 2018;27:14–25. doi: 10.1002/pro.3235. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Pettersen E.F., Goddard T.D., Huang C.C., Couch G.S., Greenblatt D.M., Meng E.C., et al. UCSF Chimera--a visualization system for exploratory research and analysis. J. Comput. Chem. 2004;25:1605–1612. doi: 10.1002/jcc.20084. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement
mmc1.pdf (15.7MB, pdf)
Movie 1
Download video file (1.8MB, mp4)
Movie 2
Download video file (1.7MB, mp4)
Movie 3
Download video file (1.4MB, mp4)

Data Availability Statement

The final coordinates and the Coulomb potential maps for the PLP-bound Rv2513c dimer (PDB entry 9n0o and EMD-48788), open Rv2531c tetramer (PDB entry 9n0p, EMD-48789), and closed Rv2531c tetramer (PDB entry 9n0n, EMD-48787) structures have been deposited with the PDB and EMDB, respectively.


Articles from The Journal of Biological Chemistry are provided here courtesy of American Society for Biochemistry and Molecular Biology

RESOURCES