Component A2 is a redox-sensitive archaeal ATPase activated by methyl-coenzyme M reductase

Summary

Methyl-coenzyme M reductase (MCR) is the primary source of biogenic methane on Earth. In the active site of MCR, a nickel (Ni)-containing porphyrin (F430) must be in the Ni¹⁺ oxidation state to initiate catalysis. The reductive activation of MCR, i.e., reduction of F430 to its Ni¹⁺ state, is an ATP-dependent process, but the underlying ATPase and its precise role remain unknown. Component A2 is an ATP-binding protein that associates with MCR but was reported to lack ATPase activity. Hence, it was proposed to function solely as an ATP-carrier protein. However, recent structural insights into the MCR activation complex suggest that component A2 may hydrolyze ATP to drive conformational changes required for enzyme activation. Here, we provide direct biochemical evidence that component A2 is a bona fide ATPase that hydrolyzes ATP under strictly anaerobic conditions and only upon interaction with MCR. Mutational analyses reveal that component A2 must be bound to ATP prior to association with MCR and that residues involved in ATP hydrolysis do not impact protein-protein interaction. The two nucleotide-binding domains of A2 act cooperatively but display asymmetric contributions to ATP hydrolysis and MCR engagement. In addition, a distinctive N-terminal zinc-binding motif (ZBM) is required for maximal ATPase activity but is dispensable for MCR binding. Phylogenetic analyses reveal that this ZBM distinguishes component A2 from related ABC-type ATPases. Together, these findings identify component A2 as a distinct class of remodeling ATPases that powers conformational changes underlying the reductive activation of MCR.

Graphical Abstract

eTOC blurb

Methyl-coenzyme M reductase (MCR) is involved in biological methane production. The maturation of MCR is a multi-step ATP-dependent process. Adler and Nayak have shown that component A2 is redox-sensitive remodeling ATPase dedicated to the biogenesis of MCR and other members of the alkyl-coenzyme M reductase family in archaea.

Introduction

Methyl-coenzyme M reductase (MCR)-encoding archaea plays a prominent role in the global methane cycle by catalyzing a reversible reaction shown below¹:

$${CH}_3-S-CoM+CoB-SH\leftrightarrow {CH}_4+CoM-S-S-CoB$$

In methanogenic archaea, MCR mediates the final step of methanogenesis to generate methane. In anaerobic methanotrophic archaea (ANME), MCR activates methane to methyl-coenzyme M, which is the first step in its eventual oxidation to carbon dioxide^2,3. Despite substantial sequence-level divergence, the overall structure and active site architecture of MCR is extremely well-conserved across archaea⁴. The ~300 kDa MCR heterohexamer is comprised of three subunits arranged in an α₂β₂γ₂ conformation and contains two active sites that are 50 Å apart and proposed to be functionally coupled (Figure 1)^4–6. Each active site has a non-covalently bound prosthetic group, factor 430 (F430), that is only found in MCR and other members of the alkyl-coenzyme M reductase (ACR) enzyme family^2,4,7,8. F430 is a Ni-containing tetrapyrrole that must be reduced to its Ni¹⁺ oxidation state, through a process known as reductive activation, for MCR to be catalytically active^9–11. As the mid-point potential of the Ni²⁺/Ni¹⁺ couple is extremely low (between −600 mV and −700 mV relative to a standard hydrogen electrode)¹², the reductive activation of MCR requires many accessory proteins (Figure 1)^13,14. In the absence of these proteins, the Ni atom in F430 gets rapidly oxidized to Ni²⁺ and MCR becomes catalytically inactive, even when purified under anaerobic conditions¹⁵.

Figure 1. Cartoon depicting the reductive activation and catalytic activity of MCR.

(Top) Reductive activation of MCR is an ATP-dependent process. Component A2 (light blue) and other methanogenesis marker proteins or MMPs (gray) are required to generate the Ni¹⁺ form of F430 (green) for activation of MCR from its Ni²⁺ oxidation state (yellow) to its Ni⁺¹ oxidation state (green). (Bottom) The reaction catalyzed by MCR requires an external input of electrons for the reduction of the methyl-group in methyl-coenzyme M to methane, which is provided by coenzyme B. The Ni¹⁺ form of F430 is regenerated at the end of each reaction cycle so ATP-dependent activation is not required for each molecule of methane generated. See also Data S1.

For nearly fifty years it has been known that the reductive activation of MCR is an ATP-dependent process, but the underlying reason is not clearly understood. In 1979, Gunsalus and colleagues showed that Methanothermobacter thermoautotrophicus cell extracts required trace amounts of ATP for methane production from methyl-coenzyme M and estimated that ~1 mol of ATP was required for ~15 mols of methane produced¹⁶. Subsequently, Rouvier et al. isolated an ATP-binding protein, designated as component A2 or AtwA, that was required for optimal methane production by cell extracts of M. theroautotrophicus¹⁷. Initial reports indicated that component A2 is a soluble, colorless, air-tolerant protein of the ABC-type ATPase family that lacks any detectable ATP hydrolysis activity^17,18. Component A2 could not perform electron transfer either to or from common electron carriers like FAD, NAD, NADP, or F₄₂₀¹⁷. Based on these analyses, component A2 was designated as an ATP-carrier protein^16–18 that delivers ATP to another enzyme, often speculated to be a homolog of the dinitrogenase reductase (NifH)¹⁹, which can hydrolyze ATP and transfer electrons to F430 for the reductive activation of MCR. However, NifH homologs have not, to our knowledge, been shown to copurify with MCR even though they are universally conserved in MCR-encoding archaea because one of the F430 biosynthesis steps requires CfbCD, where CfbC is a homolog of NifH^20,21.

Recently, a cryo-EM structure of the MCR activation complex shed new light on the putative function of component A2¹³. The MCR activation complex contains component A2 bound asymmetrically to one γ subunit (McrG). In addition, the MCR activation complex contains McrC and other methanogenesis marker proteins (MMPs) – Mmp3, Mmp7, and Mmp17 – that coordinate three complex Fe-S clusters, which closely resemble the [8Fe-9S-C] L clusters involved in the maturation of nitrogenases (Figure 1). Even though component A2 does not interact directly with McrC and the MMPs in the activation complex, it is postulated that the structural rearrangements required for the reduction of F430 to its Ni¹⁺ state occur when component A2 hydrolyzes ATP¹³. This hypothesis directly contradicts previous biochemical evidence that component A2 is merely an ATP-carrier protein. Furthermore, since component A2 is bound to ATP —not ADP — in the activation complex, it is still plausible that it delivers ATP to another ATPase domain protein, like MMP15, that could transiently associate with the activation complex, making it difficult to capture structurally²².

Here, we resolve a longstanding conundrum regarding the role of component A2 in the reductive activation of MCR by complementary in vivo and in vitro analysis of the protein from Methanosarcina acetivorans. Our results show that component A2 is a bona fide ATPase that hydrolyzes ATP only upon interaction with MCR under anaerobic conditions, and that interaction with MCR is predicated on A2 first binding ATP. In addition, our evolutionary analyses indicate that component A2 belongs to a unique class of redox-sensitive remodeling ATPases that cluster with members of the ACR family within archaea.

Results

Component A2 is essential and constitutively expressed in Methanosarcina acetivorans

In M. acetivorans, component A2 is encoded by MA_3998 (MA_RS20860) and is found in the vicinity of six genes encoding methanogenesis marker proteins (MMPs) (Figure 2A), some of which have also been implicated in the activation of MCR (Figure 2B) and all of which are highly conserved across genomes containing MCR/ACR^13,23. Component A2 and these six others neighboring MMPs are often referred to as the MCR activation operon, but it is unclear if these genes are co-transcribed and the degree to which they are expressed. First, to test if component A2 is in an operon with the other six MMPs (shown in Figure 2A), we extracted RNA from M. acetivorans and performed RT-PCR with primers that span intergenic regions across the putative operon as shown in Figure S1A. We were able to observe PCR products for all the intergenic regions tested, which experimentally validates that these seven genes are, indeed, organized as an MCR activation operon (Figure S1A). Next, we mapped RNA-sequencing reads from previous transcriptomics experiments with M. acetivorans^24–26 to the MCR activation operon. The average expression of the MCR activation operon does not vary significantly across growth substrates and remains unchanged in response to MCR-limitation (Figures S1B–C). The read depth and coverage of the MCR activation operon is ~40-fold lower than the MCR operon (mcrBDCGA) (Figure 2C and Figure S1D). Taken together, these data suggest that component A2 is part of the MCR activation operon, which is constitutively expressed at much lower levels than MCR operon in M. acetivorans.

Figure 2. Component A2 is an oxygen-sensitive ATPase that hydrolyzes ATP upon interaction with MCR.

(A) Genomic arrangement of the MCR activation operon in M. acetivorans with scale bar depicting 1 kilobase pair. (B) Structure of the putative MCR activation complex (PDB: 9H1L) with MCR in orange, component A2 (with relevant domains highlighted) in blue, and the other methanogenesis marker proteins in gray. NBD refers to Nucleotide Binding Domain and ZBM refers to Zinc Binding Motif. (C) Log₂ transformed FPKM (Fragments Per Kilobase of transcript per Million mapped reads) of the MCR activation operon and the MCR operon in M. acetivorans grown in high-salt (HS) minimal medium supplemented with trimethylamine (TMA) at 37 °C. (D) Genotype of an M. acetivorans strain expressing a second copy of component A2 in trans under the control of a tetracycline inducible promotor. (E) Anti-FLAG immunoblot showing the inducible production of component A2 upon the addition of 100 μg/mL tetracycline (tet) to the growth medium in crude and soluble cell lysates with 13.9 μg total protein loaded into each lane (F) Gel showing anaerobic purification of full-length TAP-tagged component A2 (64 kDa with tag). (G) Anti-FLAG immunoblot of aerobic affinity-purification of A2 with a Streptactin resin. The lanes represent the following: (1) ladder, (2) cell lysate, (3) flow-through, (4) first wash, (5) second wash, (6) first elution, (7) second elution, and (8) third elution. (H) Anaerobic ATPase assay of component A2 alone (blue), MCR alone (orange), and component A2 combined with MCR (purple). Each reaction contained 500 μg/mL of each protein indicated with 200 μM ATP, 10 mM MgCl₂, 20 mM HEPES, 300 mM NaCl, and 1% glycerol v/v at pH = 8.0. Reactions were incubated at 37 °C. Inorganic phosphate production was measured at 0, 15, 30, and 60 minutes using the malachite green reagent. (I) Aerobic ATPase assay of component A2 alone (blue), MCR alone (purple), and component A2 combined with MCR (orange). Proteins were purified anaerobically then removed from the anaerobic chamber and reactions were set up on the bench top. The same assay conditions as (C) were used, but time points were taken at 0, 30, and 60 minutes. Error bars represent the standard deviation of three technical replicates for each reaction. See also Figures S1 and S2, Table S1 and Data S1.

Based on their proposed function, component A2 and other genes in the MCR activation operon are likely to be essential in methanogens, but this hypothesis, to the best of our knowledge, has not been explicitly tested. To test the essentiality of the MCR activation operon, we designed CRISPR-editing plasmids to generate in-frame chromosomal deletions of each gene individually²⁷. For component A2, transformation of the CRISPR-editing plasmids did not produce any viable colonies. For the rest of the genes-encoding Mmp5, Mmp6, Mmp17, Mmp7, Mmp3, and Mmp15-in the MCR activation operon, a few antibiotic-resistant transformants were detected (Figures S1E–H). However, PCR with primers flanking the gene of interest revealed either a mixed population of the wildtype (WT) and the mutant allele or only the WT allele, suggesting that genome editing was either incomplete or unsuccessful, respectively (Figures S1E–H). A similar observation was also made in recent attempts to delete nifB²⁸, another essential gene in M. acetivorans. Altogether, we were unable to successfully delete component A2 or any of the neighboring MMPs in M. acetivorans, which corroborates their proposed role in the activation of MCR, an enzyme essential for growth and viability.

Component A2 hydrolyzes ATP after interacting with MCR under anaerobic conditions

Since the native component A2 locus is essential and constitutively expressed at low levels in M. acetivorans (Figure 2C), we generated a strain to overexpress it in trans for facile purification and biochemical characterization of its proposed function (Figure 2D). Expressing component A2 in trans also allowed us to study catalytically inactive mutants as the chromosomal locus retains its original function, which, as noted above, appears to be essential for cell viability. To this end, we fused a tandem-affinity purification (TAP) tag, comprised of a Twin-Strep-1X-FLAG sequence, at the N-terminus of component A2 and expressed it under the control of the PmcrB(tetO1) tetracycline inducible promoter in the pJK027A vector backbone²⁹. We chose to add a TAP tag the N-terminus of component A2 as it is far from the interaction interface with MCR and is also known to not disrupt ATP-binding^13,19. The resulting plasmid (pSAA004) was integrated in the chromosome of M. acetivorans (WWM73) at the φC31 attachment site as described previously²⁹. Whole genome sequencing was performed to verify the genotype of the strain used for protein purification (Table S1) and the production of TAP-tagged component A2 was detected by immunoblotting against the FLAG-tag when expression was induced with 100 μg/mL tetracycline (Figure 2E). Overexpression of component A2 (by ~14-fold²⁶) did not lead to an observable growth defect, which indicated that elevated levels of this protein are not toxic or lethal to M. acetivorans (Figures S2A–B). Anaerobic affinity purification of TAP-tagged component A2 yielded colorless, soluble, full-length protein of ~ 64 kDa (Figures 2F, 2G and Figures S2C–D and S2F). In addition to component A2, we observed nine additional bands that either correspond to McrB, McrG or degradation products of component A2 itself (Figure 2F and Data S1A–I). A band corresponding to Glutamyl-tRNA(Gln) amidotransferase subunit E (Data S1A) was observed in all protein preparations of component A2 (even the mutant forms; see below) hence was not considered to impact any of the data presented here. Since McrA is the same size as component A2, we used an anti-McrA antibody to confirm that it also copurifies with this protein (Figure S2G). A reverse pulldown with TAP-tagged MCR³⁰ (Figure S2E) confirmed the interaction between component A2 and MCR (Data S1I).

While it is well established that component A2 binds ATP^13,17,31, it is unclear if its role in the context of MCR activation is to serve as an ATP carrier for an alternate ATPase or to perform ATP hydrolysis by itself¹⁹. To disentangle these two hypotheses, we explored if component A2 can hydrolyze ATP in isolation or in conjunction with MCR using the Malachite green assay to quantify free inorganic phosphate (P_i) produced during the process (see more details in the Materials and Methods section). When component A2 was purified under anaerobic conditions, we detected reproducible ATPase activity, ~ 38 nmol P_i released/mg protein after 60 minutes of incubation for three independent preparations (Figures 2H, 3J [blue bars]; and Figure S2I). In contrast, when we purified component A2 under aerobic conditions, no free P_i was detected under the same conditions (Figure S2H). To rule out the possibility that the ATP hydrolysis in anaerobic preparations of component A2 was due to co-purifying contaminant(s), we assayed the ATPase activity of anaerobically prepared A2 after air exposure. Treatment with air and oxic buffers abolished ATPase activity (Figure 2I), which suggests that component A2 is an oxygen-sensitive protein and can only hydrolyze ATP under reducing conditions.

Figure 3. Nucleotide binding domains (NBDs) and zinc binding motif (ZBM) of component A2 are highly conserved and disrupt ATPase activity.

(A-C) Sequence alignment of 888 component A2 sequences assigned to HMM (Hidden Markov Model) TIGR03269. The x-axis indicates position number in M. acetivorans. Arrows indicate the (A) mutated residues K43 and K329, (B) D200 and D459, and (C) C70, C73, C86, and C89. (D) Change in melting temperature (ΔT_m) of component A2 variants in buffer with 2 mM ATP versus 125 μM ATP. Error bars represent standard deviation of the mean ΔT_m of three independent preparations of the protein (except for E200A and E459A, which represent the data for two and four protein preparations, respectively) and Welch’s unpaired t-test was performed to assess statistical significance; * indicates P-value < 0.05. (E-I) Anaerobic ATPase assay of each indicated component A2 variant (blue), MCR alone (orange), and the component A2 variant combined with MCR (purple). Each reaction contained 500 μg/mL of each protein with 200 μM ATP, 10 mM MgCl₂, 20 mM HEPES, 300 mM NaCl, and 1% glycerol v/v at pH = 8.0. Reactions were incubated at 37 °C. Inorganic phosphate production was measured at 0, 15, 30, and 60 minutes using malachite green reagent. Error bars represent the standard deviation of three technical replicates. (J) ATPase activity for each component A2 variant alone and upon addition of MCR. Error bars represent the standard deviation of three independent preparations of each variant of component A2, except C70A/C73A/C86A/C89A for which only two independent preparations were performed. See also Data S2 and Figures S3 and S4.

Next, we tested if the addition of MCR impacts ATP hydrolysis by component A2 in vitro. Anaerobic preparations of MCR from stationary-phase cultures of M. acetivorans had no detectable ATPase activity whereas ATP hydrolysis was observed in MCR preparations from exponential-phase cultures, likely because component A2 copurifies with it (Figure S2J). Hence, we only used MCR derived from stationary-phase cultures to test its impact on ATPase activity of component A2 (Figure 2H). The addition of an equal amount of MCR by weight to component A2, which represents a ca. 5:1 molar ratio of component A2: MCR, increased its ATPase activity under anaerobic conditions by 2.2-fold to ~ 84 nmol P_i released/mg component A2 protein after 60 minutes (Figures 2H and 3J [purple bars]; and Figure S2I). In contrast, no detectable ATPase activity was observed when MCR was added to component A2 under aerobic conditions (Figure 2I). These data are consistent with the hypothesis that the ATPase activity of component A2 alone is due to the small amount of MCR that copurifies with it, which is further enhanced by the addition of MCR. Taken together, our in vitro analyses suggest that component A2 is a redox-sensitive ATPase that hydrolyzes ATP only after it interacts with MCR.

At least one nucleotide-binding domain and the zinc-binding motif are important for ATP hydrolysis by component A2

Component A2 from M. acetivorans contains two nucleotide binding domains (NBDs) as well as a CXXCX₁₂CXXC Zn²⁺-binding motif (ZBM) (Figures 3A–3C). The two NBDs are highly conserved in component A2 sequences across archaea and each of them contains an A-loop (either Y or F), a Walker A motif (GxxGxGK[T/S]) and Walker B motif (hhhhDE) (Figures 3A and 3B). The lysine residue (K) of the Walker A motif is known to be critical for ATP-binding³², and the glutamate residue (E) of the Walker B motif deprotonates water in the first step of ATP hydrolysis and is essential for catalysis³³. To eliminate ATP-binding, we mutated the lysine residues in the two Walker A motifs to alanine by generating a K43A/K329A double point mutant of component A2. To specifically abolish ATPase activity, without impacting ATP-binding, we generated a E200A/E459A double point mutant of component A2. Additionally, we generated single point mutants, K43A, K329A, E200A, and E459A, to determine if catalytic activity is dependent on both NBDs and if the two NBDs act asymmetrically. All point mutants were expressed and purified from M. acetivorans using the strategy described for the WT allele in Figure 2 (Figure S3).

To validate the proposed role of each of these NBD residues, we performed differential scanning fluorimetry (DSF) (or thermal shift assays) in the presence of either low (0.125 mM) or high (2 mM) ATP in the buffer³⁴. We chose these ATP concentrations based on prior work that had been conducted with component A2^13,17. All DSF assays were performed aerobically with anaerobically purified enzyme, and the first differential of the melt curve had a single, major peak, which we infer corresponds to the melting temperature (‘T_m’) of component A2 (Figure 3D and Figures S4A–H). Switching from 0.125 mM to 2 mM ATP concentrations in the buffer increased the T_m (referred to as the ΔT_m) of all mutants of components A2 by > 5.0 °C, apart from the K43A/K329A double mutant (Figure 3D and Figures S4A–H). These data suggest that ATP-binding stabilizes component A2 and that the K43A/K329A double mutant, as expected, does not experience this shift in T_m because it cannot bind ATP. Like the NBDs, the CXXCX₁₂CXXC ZBM is highly conserved across all component A2 sequences too (Figure 3C), which suggests that it is functionally relevant. To test the role of the ZBM, we generated a C70A/C73A/C86A/C89A mutant. The loss of the ZBM seems to have no impact on the structural integrity of component A2 and the C70A/C73A/C86A/C89A mutant was stabilized by the presence of ATP in the buffer (Figure 3D), which indicates that it can still bind ATP.

Next, we measured the anaerobic ATPase activity of all the point mutants either by themselves or in the presence of an equal amount of MCR by weight. The K43A/K329A double mutant produced <6 nmol P_i/mg protein after 60 minutes of incubation, which did not increase upon the addition of MCR (Figures 3E, 3J and Figure S4I). ATP hydrolysis was similarly attenuated in the E200A/E459A mutant (Figures 3F, 3J and Figure S4J). The C70A/C73A/C86A/C89A mutant had ~50% activity of the WT by itself, which did not increase upon the addition of MCR (Figures 3G, 3J and Figure S4K). Additionally, we used Ellman’s assay³⁵ to assess the oxidation state of these Cys residues in component A2. We were able to detect a significant decrease in free thiols, likely due to the formation of disulfide bridges, after exposing the protein to oxygen for 10 minutes, indicating that the ZBM is oxygen sensitive (Figure S4L). These data suggest that the NBDs are essential whereas the ZBM is critical for optimal ATPase activity.

To test if the two NBDs work synergistically or independently, we assayed the ATPase activity of the single point mutants. While neither single mutant behaved like WT, the K329A mutant had ~ 2.9 times higher activity than the K43A mutant in the presence of MCR (Figures 3H, 3J and Figures S4M–N). Similarly, the E200A mutant had ~ 3 times higher activity than the E459A mutant (Figures 3I, 3J and Figures S4O–P). These data suggest that the NBDs have an asymmetric but synergistic contribution to ATP hydrolysis.

Only ATP-bound component A2 can interact with MCR

There are two models for ATP hydrolysis by the component A2-MCR complex that are equally consistent with the activity data noted above. One is that component A2 binds ATP first, ATP-bound component A2 interacts with MCR, and then ATP hydrolysis occurs (Figure 4A). Alternately, component A2 binds MCR first, ATP binds to the component A2-MCR complex, which is followed by hydrolysis (Figure 4A). To distinguish between these two possibilities, we leveraged DSF to observe complex formation between component A2 and MCR either in the presence or absence of different adenine nucleotides. All DSF assays were performed aerobically with anaerobically purified enzyme, so that the interaction between component A2 and MCR could be decoupled from ATP hydrolysis. In the presence of ATP and MgCl₂, the first differential of the melt curve for component A2 and MCR individually showed a single peak at 45.5 ± 1.3 °C and 67.2 ± 0.8 °C respectively, which corresponds to the T_m of these two proteins (Figures 4B, 4C, and Figures S5A–B). When component A2 and MCR are combined in the presence of ATP and MgCl₂, a third peak at 60.8 ± 0.6 °C appears in the melt curve (Figure 4D and Figure S5C). We interpret that this third peak corresponds to the T_m of the complex of component A2 and MCR. This third peak is absent when MCR and component A2 are combined in the absence of ATP or in the presence of ADP (Figures 4D, 4E, and Figures S5D–F). All tested concentrations of ATP showed some presence of this component A2-MCR peak (Figure 4E and Figure S5G). In contrast, a peak corresponding to the component A2-MCR complex was not detected regardless of the [ADP] (Figure 4E and Figure S5H).

Figure 4. Only ATP-bound component A2 can interact with MCR.

(A) Cartoons showing two models of ATP-based interaction between component A2 and MCR. In Model 1, component A2 (blue) binds MCR (in orange) and the component A2-MCR complex interact with ATP. In Model 2, component A2 is bound to ATP (in blue) prior to engagement with MCR (in range). (B-D) Differential scanning fluorimetry (DSF) of (B) component A2, (C) MCR, and (D) component A2 combined with MCR. Samples contained 2 mM ATP (left panel) or ADP (right panel), 10 mM MgCl₂, 20 mM HEPES, 300 mM NaCl, and 1% glycerol v/v at pH = 8.0. (e) DSF of component A2 combined with MCR with increasing concentrations as indicated of ATP (left) or ADP (right) and the same assay conditions as (B-D). The peak corresponding to the component A2-MCR complex (purple arrow) only appears in the presence of ATP. All samples in panels b-e contain 0.5 mg/mL of each protein indicated. See also Figure S5.

AlphaFold³⁶ predicted structures of apo-component A2 compared to the ATP- and ADP-bound versions reveal significant conformational changes (Figures S5I–M). Additionally, the AlphaFold predicted structure of ATP-bound component A2 is in high agreement with the MCR-bound component A2 from the activation complex¹³ (Figure S5M). These structural predictions further support model 1 in Figure 4A that only ATP-bound component A2 can interact with MCR.

A catalytically dead component A2 can still interact with MCR

Based on our model so far, mutations in the Walker A motif of the NBDs that disrupt ATP-binding in component A2 would abolish interaction with MCR. However, whether the Walker B motif of the NBD or the ZBM contributes to the interaction between component A2 and MCR is not as intuitive. To test the role of each of these motifs in MCR interaction, we assessed the amount of MCR that copurifies with each component A2 mutant by immunoblotting against a previously described anti-McrA antibody²⁶. As expected, the K43A/K329A double mutant, which cannot bind ATP, copurifies with practically no MCR compared to WT (Figure 5A and Figures S6A–C). However, each of the single mutants (K43A and K329A) could still interact with MCR but to slightly lesser degree than WT. The E459A, E200A/E459A and C70A/C73A/C86A/C89A mutants co-purified with similar amounts of MCR as WT (Figure 5A and Figures S6A–C). However, the E200A mutant had diminished interaction with MCR, like the K43A and K329A mutants. To further corroborate these findings, we performed DSF assays of each of these mutants alone and with MCR in the presence of ATP. No evidence of the component A2-MCR complex could be detected when the K43A/K329A mutant was combined with MCR in the presence of ATP (Figure 5C and Figure S6D). In contrast, a peak corresponding to the T_m of the component A2-MCR complex was observed for all the other mutants (Figures 5D–I and Figures S6E–J). These data indicate that the interaction between MCR and component A2 is dependent on ATP-binding but not linked to ATP hydrolysis.

Figure 5. Interaction of MCR and component A2 is not coupled to ATP hydrolysis.

(A) Immunoblots with anti-McrA specific antibodies to evaluate the amount of MCR that copurifies with variants of component A2: K43A/K329A, K43A, and K329A (top), E200A/E459A, E200A, and E459A (middle) and C70A/C73A/C86A/C89A (bottom). Each blot shows a 2X dilution series for each indicated protein starting at 2.5 μg. (B-I) Differential scanning fluorimetry (DSF) of component A2 variants alone (left) or combined with MCR (right). Peaks with a blue background correspond to component A2, with an orange background correspond to MCR, and with a purple background correspond to component A2-MCR complex. Samples contained 2 mM ATP, 10 mM MgCl₂, 20 mM HEPES, 300 mM NaCl, 0.5 mg/mL of each indicated protein, and 1% glycerol v/v at pH = 8.0. Panel B shows data from Figure 4B and 4D and is repeated here for ease of comparison to the point mutants of component A2. See also Figure S6.

Component A2 clusters with members of the alkyl-coenzyme M reductase (ACR) superfamily

To study the evolutionary history of component A2, we built a phylogenetic tree using 80 sequences that capture the sequence diversity of the protein as well as the taxonomic breadth of archaea that encode them (Figure 6 and Data S2). Many archaea, including M. acetivorans, also encode a component A2-like protein that contains both NBDs but lacks the ZBM, and we designated these proteins as the outgroup to root the component A2 tree. Despite the sequence similarity with component A2, these component A2-like proteins have never been shown to co-purify with MCR,^13,19,22 nor are they designated as methanogenesis marker proteins²³. Hence, we hypothesize that component A2-like proteins are functionally unrelated to component A2 but are a valid outgroup for phylogenetic analyses. The phylogeny of component A2 is not consistent with vertical inheritance and indicates that it has undergone rampant HGT within archaea. For example, component A2 sequences from archaea within the Halobacteriota can be found in at least 8 distinct clades that have a bootstrap support of 100 (Figure 6). Methanoliparia encode both MCR and ACR³⁷ and two distantly related component A2 sequences. Each component A2 sequence from Methanoliparia either clusters with sequences derived from archaea that encode MCR or ACR. Methanogens that encode multiple MCR isozymes³⁸ (members of the Methanobacteriota) only encode one component A2 (Data S2) and there are no obvious distinctions between the component A2 sequences derived from ANME and methanogens either. Altogether, these observations suggest that the MCR-specific component A2 is likely to be compatible with all MCRs but not with ACRs and vice versa. Finally, even though component A2 sequences from ACR-encoding archaea are more closely related to each other than to their counterparts from MCR-encoding archaea (Figure 6), they do not form a highly divergent monophyletic clade like ACRs do relative to MCR⁷. This is likely because all component A2 sequences perform the same core function of ATP hydrolysis regardless of which ACR they associate with, in contrast with the significant sequence divergence ACR has undergone to accommodate divergent substrates.

Figure 6. Phylogenetic analysis of component A2 across alkyl-coenzyme M reductase (ACR)-encoding archaea.

Phylogenetic tree of component A2 amino acid sequences derived from genomes encoding alkyl coenzyme-M reductase (ACR) in green, methyl-coenzyme M reductase (MCR) in white, and both in blue. All 888 protein sequences with the TIGRFAM Hidden Markov Model (HMM) TIGR0326 assigned to component A2 were obtained from the Genome Taxonomy Database (GTDB) R214. These sequences were clustered at 70% sequence similarity and any truncated sequences or sequences with large insertions (as described in the Methods) were removed, which resulted in a final set of 80 component A2 sequences. The outgroup comprises of 20 sequences of a component A2-like protein lacking a zinc-binding motif (ZBM) that was identified using the sequence from M. acetivorans (MA_3967 or MA_RS20695) as a search query. The clade containing the component A2 sequence from M. acetivorans is indicated by a star. See also Data S2.

Discussion

In this study, we established an over-expression system in M. acetivorans to experimentally evaluate the role of component A2 in the reductive activation of MCR. Many hypotheses regarding the role of component A2 have been made with protein purified from E. coli or from indirect measurements using methane production rates as a proxy for ATPase activity. We suspect that the use of heterologously produced protein and an insensitive assay technique have led to confounding results for decades. By directly measuring P_i production by component A2 derived from M. acetivorans, we show that this enzyme is a bona fide ATPase, whose activity is dependent on redox conditions (Figures 2H and 2I) and is modulated by its association with MCR (Figure 2H). In our ATPase assays, ~8 nmol of component A2 produces ~ 80 nmol of P_i in the presence of MCR (Figure 3J), which implies that there are at least 10 catalytic turnover events. These data reinforce our central hypothesis that Component A2 is acting as a true catalyst rather than a stoichiometric reactant for the ATPase reaction. Even though both NBDs of component A2 are required for interaction with MCR and for optimal ATP hydrolysis they are not functionally equivalent (Figures 3E–J). Mutations in the Walker B motif of NBD2 impede ATP hydrolysis far more than mutations in the Walker B motif of NBD1 (Figure 3I). In contrast, mutations in the Walker B motif of NBD1 substantially lower component A2-MCR interactions whereas mutations in the Walker B motif of NBD2 do not affect protein-protein interaction at all (Figure 5A). We also hypothesize that the universally conserved ZBM renders component A2 redox-sensitive (Figure 3C). It is quite likely that the ZBM acts as a “redox-switch” where reduced cysteine residues bind Zn²⁺ and oxidation of these residues reduces rates of ATP hydrolysis (Figure 3G)³⁹. This redox switch in component A2 might ensure that the reductive activation of MCR only occurs under environmental conditions that are conducive for catalysis, i.e., in the absence of oxygen or other oxidants.

The evolutionary ties between the maturation machinery for nitrogenases and MCR have been discussed previously¹³ but are worth revisiting considering our findings. Nitrogenases requires ATP hydrolysis for enzyme maturation as well as electron transfer during catalytic turnover⁴⁰. In contrast, MCR requires ATP input only for cofactor reduction during enzyme maturation. Thus, the net demand for ATP hydrolysis is much lower for MCR than nitrogenases. Furthermore, in nitrogenases, ATP hydrolysis is directly coupled to a redox gating mechanism that triggers electron flow from a [4Fe-4S] cluster in NifH to the P-cluster in NifDK, which is the terminal electron donor for N₂ reduction⁴¹. Our data suggest that component A2, by itself, is unlikely to be involved in a redox gating mechanism for MCR activation but might facilitate this process through a series of cascading events that are somehow linked to ATP hydrolysis. This observation is also consistent with the cryo-EM structure of the MCR activation complex where component A2 seems to primarily interact with MCR rather than the rest of the activation complex¹³. Our DSF assays reveal a distinct peak for the component A2-MCR complex, which indicates that the two proteins interact even in the absence of McrC and the other MMPs that are putatively involved in transferring electrons to F430 in the active site of MCR (Figure 4D, 4E). That said, even when an equal amount of MCR is added to component A2, the rate of ATP hydrolysis only increases by 2.2-fold, and the reaction does not proceed to completion (Figure 2H, 3J). These results suggest that ATP hydrolysis is only triggered when component A2 interacts with a certain sub-population of MCR, likely MCR bound to the electron transfer components of the activation complex. By decoupling MCR interaction from ATP hydrolysis, component A2 might be able to rapidly associate with MCR but only hydrolyze ATP when MCR is primed for reductive activation. Our proposal is also consistent with the mode of action for type I/II ABC transporters where conformational changes in the membrane-bound transporter that are triggered by the interaction of the substrate with the substrate-binding protein ultimately led to hydrolysis by the ATPase component⁴². The conformational cues in MCR induced by other components of the activation complex that ultimately trigger ATP hydrolysis by component A2 are currently unknown but an exciting area of investigation for future studies.

In summary, this work provides clear experimental evidence that component A2 is an unusual yet bona fide ATPase involved in the activation of MCR. We also establish a pipeline for holistic and mechanistic analyses of essential components involved in the biogenesis of MCR in vivo. A detailed understanding of how each component of the activation complex interacts with MCR will be necessary for a complete understanding of the intricate pathway involved in the maturation of MCR within the cell.

Resource availability

Lead contact

Further information and request for resources and reagents should be directed to and will be fulfilled by the lead contact, Dipti D. Nayak (dnayak@berkeley.edu)

Material availability

Newly generated plasmids and strains from this manuscript are available from Dipti D. Nayak (dnayak@berkeley.edu)

Data and code availability

• Sequencing data have been deposited in the Sequencing Reads Archive and the Bioproject number is listed in the key resources table. All other data generated in this study are provided in the manuscript.

• This paper does not report any original code.

• Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Anti-FLAG M2 HRP-conjugated antibody	Millipore Sigma	Cat#A8592
Anti-McrA polyclonal rabbit antibodies	Chadwick et al.26
Anti-Rabbit HRP conjugate antibody	Promega	Cat#W401B
Bacterial and virus strains
Escherichia coli WM4489	Nayak et al27	Laboratory of William W. Metcalf
Biological samples
Chemicals, peptides, and recombinant proteins
Trimethylamine hydrochloride, 98%	Millipore Sigma	Cat#T72761
Puromycin hydrochloride	Cayman Chemicals	Cat#13884
Tetracycline hydrochloride	Millipore Sigma	Cat#T7660
Lysogeny broth (LB), Miller	Fisher Scientific	Cat#BP1426
Chloramphenicol, >98%	Millipore Sigma	Cat#C0378
L-Rhamnose monohydrate	Millipore Sigma	Cat#R3875
DOTAP Liposomal transfection reagent	Roche	Cat#11202375001
TRIzol	Invitrogen	Cat#15-596-026
Bovine Serum Albumin (BSA), lyophilized powder	Millipore Sigma	Cat#A9418
D-Desthiobiotin	Millipore Sigma	Cat#D1411
β-mercaptoethanol or 2-mercaptoethanol	VWR Life Science	Cat#97064-878
Ellman’s reagent or DTNB, 5,5’-dithio-bis-(2-nitrobenzoic acid)	Millipore Sigma	Cat#D8130
L-Cysteine hydrochloride monohydrate, >98%	Millipore Sigma	Cat#C7880
Deoxyribonuclease I from bovine pancreas	Millipore Sigma	Cat#D-4263
cOmplete EDTA-free Protease Inhibitor Cocktail	Roche	Cat#11836170001
Strep-Tactin Superflow Plus resin	QIAGEN	Cat#30004
SYPRO Orange Protein Gel Stain	Invitrogen	Cat#S6650
Precision Plus Prestained Protein Standards	Bio-Rad	Cat#1610373
Coomassie GelCode Blue Stain Reagent	Thermo Scientific	Cat#24592
Immobilon Western Chemiluminescent HRP Substrate	Millipore Sigma	Cat#WBKLS0050
Agar	Millipore Sigma	Cat#A1296
Agarose LE	Goldbio	Cat#A-201-100
Ethidium bromide	Millipore Sigma	Cat#E8751
Sodium dodecyl sulfate (SDS)	Bio-Rad	Cat#1610301
Laemmli Sample Buffer	Bio-Rad	Cat#1610737
Glycine	Fisher bioreagents	Cat#BP381
Sodium citrate monobasic	Millipore Sigma	Cat#71497
Boric Acid	Millipore Sigma	Cat#B6768
Ethylenediaminetetraacetic acid (EDTA) disodium salt dihydrate	Millipore Sigma	Cat#E4884
Adenosine 5′-triphosphate disodium salt hydrate (ATP)	TCI Chemicals	Cat#34369-07-08
Adenosine 5′-diphosphate disodium salt (ADP)	Millipore Sigma	Cat#20398-34-9
Nonfat dry milk, blotting grade	Apex bioresearch products	Cat#20-241
Bromophenol blue	Millipore Sigma	Cat#114391
Glycerol	Millipore Sigma	Cat#G7893
Trizma hydrochloride	Millipore Sigma	Cat#RDD009
Trizma base	Millipore Sigma	Cat#T6066
Sodium phosphate dibasic heptahydrate (Disodium phosphate)	Millipore Sigma	Cat#S9390
Sodium phosphate monobasic monohydrate (Monosodium phosphate)	Millipore Sigma	Cat#S9638
Potassium phosphate monobasic	Millipore Sigma	Cat#P0662
Sodium chloride	Fisher bioreagents	Cat#012314A3
Potassium chloride	Millipore Sigma	Cat#P9541
Sodium bicarbonate	Millipore Sigma	Cat#S6014
Magnesium chloride hexahydrate	Fisher bioreagents	Cat#BP214
Calcium chloride dihydrate	Millipore Sigma	Cat#C3306
Resazurin sodium salt	Millipore Sigma	Cat#199303
Ammonium chloride	Millipore Sigma	Cat#A4514
Sodium sulfide nonahydrate	Millipore Sigma	Cat#431648
Zinc sulfate heptahydrate	Millipore Sigma	Cat#221376
Nitrilotriacetic acid (N,N-bis[carboxymethyl]glycine, NTA) trisodium salt	Millipore Sigma	Cat#N0253
Ammonium iron(II) sulfate hexahydrate	Millipore Sigma	Cat#F3754
Sodium selenite	Millipore Sigma	Cat#214485
Cobalt(II) chloride hexahydrate	Millipore Sigma	Cat#202185
Manganese (II) sulfate monohydrate	Millipore Sigma	Cat#M7634
Sodium molybdate dihydrate	Millipore Sigma	Cat#331058
Sodium tungstate dihydrate	Millipore Sigma	Cat#223336
Nickel(II) chloride hexahydrate	Millipore Sigma	Cat#223387
Copper(II) sulfate pentahydrate	Millipore Sigma	Cat#C7631
p-Aminobenzoic acid (PABA)	Millipore Sigma	Cat#A9878
Nicotinic acid (Vitamin B3)	Millipore Sigma	Cat#N4126
D-Pantothenic acid hemicalcium salt (Vitamin B5)	Millipore Sigma	Cat#21210
Pyridoxine hydrochloride (Vitamin B6)	Millipore Sigma	Cat#P9755
Riboflavin (Vitamin B2)	Millipore Sigma	Cat#47861
Thiamine hydrochloride (Vitamin B1)	Millipore Sigma	Cat#T4625
Biotin (Vitamin B7)	Millipore Sigma	Cat#B4639
Folic acid (Vitamin M)	Millipore Sigma	Cat#F8758
(±)-α-Lipoic acid	Millipore Sigma	Cat#T1395
Hydroxocobalamin hydrochloride (Vitamin B12a)	Millipore Sigma	Cat#H7126
GoTaq Green Master Mix	Promega	Cat#M7122
Phusion High-Fidelity DNA Polymerase	New England Biolabs	Cat#M0530S
Malachite Green	Echelon Biosciences	Cat#K1501
HEPES	Fisher Bioreagents	Cat#BP3101
Tween-20	Millipore Sigma	Cat#9005645
Ammonium bicarbonate	Millipore Sigma	Cat#1066337
Dithiothreitol (DTT)	Millipore Sigma	Cat# 3483123
Acetonitrile	Millipore Sigma	Cat#75058
Trypsin
Formic acid	Millipore Sigma	Cat#141537
Critical commercial assays
DNeasy Blood & Tissue kit	QIAGEN	Cat#69504
Zyppy plasmid miniprep kit	Zymo Research	Cat#D4020
Zymo DNA Clean & Concentrator kit	Zymo Research	Cat#D4004
Zymoclean Gel DNA Recovery kit	Zymo Research	Cat#D4002
Pierce Bradford Plus Protein Assay Reagent	Thermo Scientific	Cat#A55866
10 kDa MWCO Amicon centrifugal filter	Millipore Sigma	Cat#UFC9010
RNeasy Mini Kit	QIAGEN	Cat#74104
Midi PVDF Transfer Pack	Bio-Rad	Cat#1704157
Random Hexamers	Thermo Scientific	Cat#N8080127
SuperScript III Reverse Transcriptase	Thermo Scientific	Cat#18080044
TURBO DNA-free Kit	Thermo Scientific	Cat#AM1907
Deposited data
Genome resequencing data	This paper	NCBI (Bioproject ID: PRJNA1460427)
Transcriptomic data	Chadwick et al.26 Shalvarjian et al.24 Downing et al.25
Experimental models: Cell lines
Experimental models: Organisms/strains
WM60	Guss et al.29	Laboratory of William W. Metcalf
WM1086	Nayak et al.30	Laboratory of William W. Metcalf
WWM73	Guss et al.29	Laboratory of William W. Metcalf
Oligonucleotides
All oligonucleotides used in this study are listed in Table S4
Recombinant DNA
All plasmids and recombinant DNA used in this study are listed in Table S5
Software and algorithms
Prism v10.3.1	GraphPad	https://www.graphpad.com/
WebLogo v3	Crooks et al53	https://weblogo.threeplusone.com/
Geneious v2024.0.2	Geneious	https://www.geneious.com/
AlphaFold 3	Abramson et al.36	https://alphafold.ebi.ac.uk/
Breseq v0.35.5	Deatherage and Barrick46	https://github.com/barricklab/breseq
BLASTp v2.15.0	Camacho et al.55	https://blast.ncbi.nlm.nih.gov/Blast.cgi?PROGRAM=blastp&PAGE_TYPE=BlastSearch&LINK_LOC=blasthome
Genome Taxonomy Database (GTDB) R214	Parks et al.51	https://gtdb.ecogenomic.org/
ChimeraX v1.7.1	Meng et al.49	https://www.rbvi.ucsf.edu/chimerax/
InterPro v108.0	Blum et al.48	https://www.ebi.ac.uk/interpro/
iTOL v7	Letunic et al.57	https://itol.embl.de/help.cgi
Randomized Axelerated Maximum Likelihood (RAxML) algorithm v8.2.11	Stamatakis et al.56	https://github.com/stamatak/standard-RaxML
Cluster Database at High Identity with Tolerance (CDHIT)	Li et al.54	https://www.bioinformatics.org/cd-hit/
MUSCLE v5.1	Edgar et al.52	https://github.com/rcedgar/muscle
Other
ChemiDoc™ MP Imaging System	Bio-Rad	Cat#170-8280
Heratherm General Protocol Microbiological Incubator	Thermo Fisher Scientific	Cat#840-298000
UV-Vis spectrophotometer	Thermo Fisher Scientific	Cat#29187193
Epoch2 Microplate reader	BioTek	Cat#BTEPPCH2NS

STAR Methods

Experimental model and study participant details

All M. acetivorans strains were grown at 37 °C without shaking in hermetically sealed Balch tubes or 1-liter anaerobic bottles in bicarbonate-buffered high-salt (HS) medium containing either 50 mM (for standard passaging) or 100 mM (for protein purification) Trimethylamine.HCl (TMA) as the growth substrate. All E. coli strains were grown shaking at 37 °C in lysogeny broth (LB) with 20 μg/mL chloramphenicol.

Method Details

Plasmid construction

The vector to overexpress component A2 (MA_3998 or MA_RS20860) in M. acetivorans was constructed using the pJK027A vector backbone. Briefly, MA_3998 was amplified from the M. acetivorans genome and the 2X Strep-1X FLAG TAP tag was added to the N-terminus of the protein via primer overhangs. The amplified PCR product was assembled into pJK027A linearized with NdeI and HindIII via Gibson Assembly as previously described⁴³. Site directed mutagenesis of component A2 to generate the K43A, K329A, K43A/K329A, E200A, E459A, and E200A/E459A variants was conducted with primers containing the desired point mutations. The C70A/C73A/C86A/C89A allele of A2 was generated as a synthetic construct (Twist Biosciences, South San Francisco, CA, USA). All plasmids were transformed into E. coli strain WWM4489⁴⁴, a derivative of DH10B, by electroporation (MicroPulser Electroporator, Bio-Rad, Hercules, CA, USA) and the resulting transformants were grown in LB supplemented with 20 μg/mL chloramphenicol and 10 mM rhamnose to induce high copy number of the plasmid⁴⁴. All plasmids were extracted using the Zyppy Miniprep Kit (Zymo Research, Tustin, CA) and verified by Sanger sequencing at the UC Berkeley DNA Sequencing Facility. All primers used were obtained from Integrated DNA Technologies, Coralville, IA, USA. All primers and plasmids used in this study can be found in Table S2 and Table S3.

Transformation of Methanosarcina acetivorans

M. acetivorans strain WWM73²⁹ was transformed with the component A2 overexpression plasmids by liposome-mediated transformation as previously described⁴⁵. Each transformation reaction was performed with 2 μg plasmid DNA and 20 mL of M. acetivorans cells grown to late-exponential phase high-salt (HS) minimal medium with 50 mM trimethylamine.hydrochloride (TMA) as the sole carbon and energy source containing HS-media. Transformants were then plated on agar-solidified HS-medium with 50 mM TMA and 2 μg/mL puromycin. Plates were incubated at 37 °C for 2–3 weeks in an intrachamber anaerobic incubator with an H₂S/CO₂/N₂ (1,000 ppm/20%/balance) headspace. Colonies were picked into 10 mL of HS-medium with 50 mM TMA and 2 μg/mL puromycin and incubated at 37 °C without shaking. Once grown, integration of the plasmid onto the chromosome was confirmed by PCR using diagnostic primers listed in Table S2. PCR product sequences were verified by Sanger sequencing at the UC Berkeley DNA Sequencing Facility. Strains generated for this study are listed in Table S3.

Genomic DNA extraction and whole-genome sequencing

Genomic DNA was extracted from 2 mL of culture grown to saturation using the QIAGEN DNeasy Blood & Tissue kit (QIAGEN, Hilden, Germany) according to manufacturer’s protocol. Library preparation and Illumina sequencing was conducted by SeqCenter (Pittsburgh, PA, USA). The resulting sequencing reads were analyzed breseq v0.35.5⁴⁶ using default settings (see Table S1).

Growth curves of Methanosarcina acetivorans

Growth curves were performed by preculturing strains in HS-TMA media without shaking at 37 °C (HeraTherm General Protocol Microbiological Incubator, Thermo Fisher Scientific, Waltham, MA) in sealed Balch tubes. From early stationary phase cells, 0.5 mL culture was transferred into three replicate tubes with 100 μg/mL tetracycline and three replicate tubes without tetracycline. All tubes contained 2 μg/mL puromycin to maintain the integrated plasmid. Growth was measured by monitoring the optical density at 600 nm (Genesys 50, Thermo Fisher Scientific, Waltham, MA). All tubes containing tetracycline were wrapped in aluminum foil to prevent light-based degradation of the chemical. The growth rate was calculated from the slope of the linear fit of the log₁₀-transformed optical density versus time with a minimum five points in the exponential phase with the highest R² value (minimum cut-off ≥ 0.99).

Anaerobic purification of proteins

500 mL cultures of M. acetivorans were grown in HS-media supplemented with 100 mM TMA in 1-liter anaerobic bottles sealed with butyl rubber stoppers (Chemglass Life Sciences, Vineland, NJ, USA) at 37 °C. Cultures were supplemented with 2 μg/mL puromycin and 100 μg/mL tetracycline to induce protein expression and grown to late-exponential phase (for component A2) or stationary phase (for methyl-coenzyme M reductase; MCR). All the steps of protein purification were conducted under anaerobic conditions in a Coy chamber (Coy Lab Products, Grass Lake, MI, USA) with a N₂/H₂ (96–97%/ 3–4%) headspace. All buffers were made anaerobic by sparging with 100% N₂ for 30 minutes and were filter sterilized following sparging. Cells were harvested by spinning (Sorvall Legend XTR, Thermo Fisher Scientific, Waltham, MA, USA) at 6000 RPM at 4 °C for 10 minutes in a gas-tight sealed polypropylene Nalgene centrifuge bottles. After harvesting, cells were anaerobically lysed via osmotic pressure in 20 mM HEPES, 1% glycerol v/v buffer, pH =8.0 (Buffer A) with the addition of 0.25X Roche cOmplete EDTA-free Protease Inhibitor Cocktail (MilliporeSigma, Burlington, MA, USA) and DNase I (Thermo Fisher Scientific, Waltham, MA, USA). After lysis, 5 M NaCl was added to crude cell lysate for a final concentration of 300 mM NaCl. Crude cell lysate was cleared by spinning at 10,000 RPM at 4 °C for 30 minutes. To purify protein, cleared cell lysate was applied twice to 0.25–0.5 mL Strep-Tactin Superflow plus resin (Qiagen, Hilden, Germany) equilibrated with 10 mL anaerobic Buffer A + 300 mM NaCl. After washing the resin with 12 mL of Buffer A + 300 mM NaCl, the protein of interest was eluted with Buffer A + 300 mM NaCl + 2.5 mM desthiobiotin (Sigma-Aldrich, St. Louis, MO, USA). Purified protein was stored at RT in the anaerobic chamber, unless otherwise stated⁴⁷. Protein concentration was determined by Pierce Bradford Plus Protein Assay Reagent (Thermo Fisher Scientific, Waltham, MA, USA) using Bovine Serum Albumin (BSA) for calibration curve generation.

ATPase assays

All ATPase assays were performed under anaerobic conditions in the Coy anaerobic chamber unless otherwise stated. Assays were performed with component A2 within 48 hours of purification. Component A2 was stored anaerobically at room temperature (RT) prior to use. ATPase reactions were performed in anaerobically in 110 μL reaction volume incubated at 37 °C containing 20 mM HEPES, 1% glycerol v/v, 300 mM NaCl, 10 mM MgCl₂, 200 μM ATP, pH = 8.0, and 500 μg/mL of either component A2, MCR, or both. Reactions were started with addition of protein. Reactions were set up in technical triplicate and 25 μL aliquots were taken at 0, 15, 30, and 60-minute time points. To quench reaction time points, aliquots were immediately added to an equivalent volume of ice cold 20 mM EDTA, pH =8.0, and removed from the anaerobic chamber. Aliquots were stored on ice until the final time point was taken. ATPase activity was evaluated immediately following the final time point by measuring the production of inorganic phosphate using malachite green solution (Echelon Biosciences, Salt Lake City, UT, USA). Each 50 μL aliquot was plated in duplicate in a 96-well plate and 80 μL of malachite green solution was added to each well and mixed by pipetting. After allowing color development for 25 minutes at RT, the absorbance was measured at 620 nm on an Epoch 2 microplate reader (BioTek, Winooski, VT, USA). Absorbance was converted to phosphate concentration using a 0–100 μM phosphate calibration curve and plotted against time to determine ATPase activity over the course of the reaction. Reactions were internally normalized to the 0-minute time point to only measure change in phosphate production over the course of the 60-minute incubation.

Quantification of free thiols

Free thiols were measured using Ellman’s reagent to estimate the number of reduced cysteines when component A2 is maintained anaerobic or exposed to oxygen. Purified component A2 from two individual preparations was combined and dispensed in triplicate in 96-well plates at a concentration of 0.8 mg/mL in 50 μL 20 mM HEPES, 1% glycerol v/v, 300 mM NaCl at pH = 8.0. BSA (Pierce Albumin Standard, Thermo Fisher Scientific, Waltham, MA, USA) was also analyzed at a concentration of 0.8 mg/mL protein in equivalent buffer. Ellman’s reagent was prepared both aerobically and anaerobically. For aerobic preparation, 0.5 mM 5,5’-dithio-bis-(2-nitrobenzoic acid) solution was made in 150 mM Tris pH = 8.0 on the bench top. For anaerobic preparation, 0.5 mM 5,5’-dithio-bis-(2-nitrobenzoic acid) solution was prepared in 150 mM Tris pH = 8.0 sparged with 100% N2 for 30 minutes. For each preparation, 200 μL of fresh Ellman’s reagent was added to each well containing protein and incubated at RT for 10 minutes before absorbance was read at 412 nm in a microplate reader (BioTek Epoch 2, Winooski, VT, USA). For aerobic preparation, the protein was removed from the anaerobic chamber and exposed to air for ten minutes prior to the addition of Ellman’s. The number of free thiols per protein was determined relative to a cysteine (0–300 μM) standard curve. Absorbance readings were normalized to blank wells containing aerobic or anaerobic Ellman’s reagent per relevant condition.

Differential Scanning Fluorimetry

Anaerobically purified component A2 or MCR was removed from the anaerobic chamber to set up reactions for differential scanning fluorimetry at RT on the bench top. Each sample contained 500 μg/mL indicated protein, 20 mM MgCl_2, ATP at concentration indicated in Figure legend (ranging from 0 – 2 mM), 20 mM HEPES, 300 mM NaCl, and 1% glycerol v/v at pH = 8.0. SYPRO Orange Protein Gel Stain (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA) was added to a final concentration of 5X and each sample was gently pipetted to mix and dispensed into 96-well PCR plates (Applied Biosystems, Thermo Fisher Scientific, Waltham, MA, USA) and sealed with adhesive optical film. Using a CFX96 Touch Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA) fluorescence was measured in the FRET channel (450–730 nm range of excitation/emission wavelengths) along a temperature gradient from 25 °C to 95 °C in increments of 0.5 °C every 30 seconds. The first derivative of the fluorescence emission as a function of temperature (dRFU/dT) was plotted and the melting temperature (T_m) was defined as the peak(s) of the first derivative. Melt curves were repeated with biological triplicate (independent protein purifications) as indicated.

SDS-PAGE and Immunoblotting of FLAG-tagged proteins and of MCR

Samples containing protein as indicated in Figure captions were combined with Laemmli sample buffer (Bio-Rad, Hercules, CA, USA) and 2.5% β-mercaptoethanol and then heated at 95 °C for 8 minutes. Samples were then loaded into 12% precast Tris-Glycine denaturing gel (Mini-PROTEAN TGX, Bio-Rad, Hercules, CA, USA). Bio-Rad Precision Plus Prestained Protein Standard was used as a molecular weight standard. Gels were run at 100–150 V until the dye front reached the bottom of the cassette. Gels were stained with Coomassie GelCode Blue (Thermo Fisher Scientific, Waltham, MA, USA). For immunoblotting, protein was transferred from the gel to the PVDF (polyvinylidene fluoride) membrane with the Trans-Blot Turbo Transfer System and Midi-PVDF Transfer Pack (Bio-Rad, Hercules, CA, USA) following manufacturer’s instructions. Following transfer, membranes were rinsed for 5 minutes with water, 5 minutes with PBS (137 mM NaCl, 2.7 mM KCl, 10 mM Na₂HPO₄, 1.8 mM KH₂PO₄), and blocked for 1 hour at RT with 5% nonfat milk dissolved in PBS. For visualization of FLAG-tagged proteins, membranes were then rinsed in PBS 4X for 5 minutes prior to incubation with mouse monoclonal anti-FLAG M2 HRP-conjugated antibody (Sigma-Aldrich, St Louis, MO) diluted 1:66666 in PBS-T (PBS with 0.05% Tween-20) for 1 hour at RT. Following incubation with antibody, membranes were rinsed 3X in PBS-T and then 3X in PBS for 5 minutes each. Alternatively, for visualizing McrA, membranes were rinsed in PBS 4X for 5 minutes following blocking and then incubated overnight at 4 °C with polyclonal rabbit antibodies raised against McrA (1:10000 dilution) (GenScript, Piscataway, NJ, USA) in PBS-T. The following day membranes were rinsed in PBS-T 4X for 5 minutes each and then incubated with anti-rabbit horseradish peroxidase (HRP) conjugate antibodies (1:100000 dilution) (Promega, Madison, WI, USA) for 2 hours at RT and finally rinsed 3X in PBS-T and then 3X in PBS for 5 minutes each. For both FLAG and McrA visualization, the signal was developed by addition of the Immobilon Western Chemiluminescent HRP Substrate (MilliporeSigma, Burlington, MA, USA) in the dark. Both Coomassie gels and immunoblots were visualized on a ChemiDoc MP Imaging System (Bio-Rad, Hercules, CA, USA).

Mass spectrometry Sample Identification

Mass spectrometry-based identification of proteins extracted from gels were performed by either Applied Biomics, Inc (Hayward, CA, USA) or QB3/Chemistry Mass Spectrometry Facility at UC Berkeley. Applied Biomics performed sample preparation in house. For samples analyzed at the UC Berkeley QB3/Chemistry Mass Spectrometry Facility sample preparation was performed as follows. Bands from Coomassie stained gels were excised with a razor blade and diced, and gel pieces were washed for 20 minutes in 500 μL of 100 mM NH₄HCO₃ and the supernatant was discarded. Gel pieces were then incubated in 150 μL of 100 mM NH₄HCO₃ and 10 μL of 45 mM DTT for 15 minutes at 50 °C. Next 10 μL of 100 mM iodoacetamide was added and the mixture was incubated for 15 minutes in the dark at RT. The supernatant was discarded gel pieces were washed with 500 μL of a 50:50 mix of acetonitrile and 100mM NH₄HCO₃ with shaking for 20 minutes. After the supernatant was discarded, gel pieces were incubated with 50 μL of acetonitrile for 15 minutes. Solvent was removed and gel fragments were dried in a speed vac. Gel pieces were then reswelled with 10 μL of 25 mM NH₄HCO₃ containing 0.1 μg sequencing-grade modified trypsin (Promega, Madison, WI, USA). After 15 minutes, 20 μL of additional NH₄HCO₃ was added and the reaction was incubated overnight at 37 °C. Remaining peptides from the gel pieces were extracted twice with 50 μl of 60% acetonitrile/0.1% formic acid for 20 minutes, then once with 25 μl acetonitrile, and samples were dried in a speed-vac.

RNA extraction and cDNA synthesis

M. acetivorans strain WWM60 was grown to mid-exponential phase and 1 mL of culture was added to an equivalent volume of Trizol (Life Technologies, Carslbad, CA) prewarmed to 37 °C. Following incubation for 5 minutes at RT, 2 mL of 100% cold ethanol was added, and RNA was extracted according to the manufacturer’s instructions using the Qiagen RNeasy Mini Kit (Qiagen, Hilden, Germany). Concentration of the RNA was determined using a Nanodrop One UV Spectrophotometer (Thermo Fisher Scientific, Waltham, MA) and stored at −80 °C. To eliminate any minor amounts of contaminating genomic DNA, RNA was treated with DNase according to the manufacturer’s instructions using TURBO-DNA free kit (Thermo Fisher Scientific, Waltham, MA, USA). Next, to synthesize cDNA, reactions were set up with 2.5 ng/μL random hexamers, 10 ng/μL DNase-treated RNA, 0.5 mM dNTPs, 5 mM DTT, First-strand buffer, and 10 U/μL Superscript III. Prior to the addition of DTT, First-strand buffer, and Superscript III, the reaction was incubated at 65 °C for 5 minutes and then placed on ice for 1 minute. Following the addition of the rest of the reagents, the reaction was incubated at 25 °C for 5 minutes, then 50 °C for 60 minutes, and finally 70 °C for 15 minutes. The resulting cDNA was stored at −20 °C. All reagents used for cDNA synthesis were obtained from Thermo Fisher Scientific, Waltham, MA, USA.

Protein structural prediction and visualization

AlphaFold 3³⁶ was used for structural models of M. acetivorans component A2 (MA_3998) alone and with ATP or ADP. Protein domains were identified using InterPro v108⁴⁸. Protein structures were visualized using ChimeraX v1.7.1⁴⁹.

Protein sequence alignment

Protein sequences with the assigned TIGRFAM Hidden Markov Model (HMM)⁵⁰ TIGR03269 were downloaded from the Genome Taxonomy Database (GTDB) R214⁵¹, which included 888 sequences. A multiple sequence alignment (MSA) MUSCLE v5.1 alignment⁵² was performed using Geneious software v 2024.0.2. The MSA of individual motifs of interest (NBD1 walker-A, NBD2 walker-A, NBD1 walker-B, NBD2 walker-B, and zinc binding motif) were extracted from the full sequence alignment. Each individual motif contained the following number of sequences from the MSA: 867 in NBD1 walker-A, 888 in NBD2 walker-A, 883 in NBD1 walker-B, 880 in NBD2 walker-B, and 870 in the zinc binding motif. Individual sequence motifs were visualized with WebLogo v3⁵³.

Phylogenetic analysis

All 888 component A2 sequences were clustered based on 70% sequence similarity using Cluster Database at High Identity with Tolerance (CD-HIT)⁵⁴ with default parameters. Clustering provided a list of 86 sequences that were further trimmed based on sequence length to omit truncated sequences or sequences with large insertions. Sequences were retained if they fell within ±10% of the mean sequence length, and the component A2 sequence from M. acetivorans was manually added in, due to its relevance to this work. This resulted in a final set of 80 sequences (see Data S2). The tree was built using these 80 sequences along with 20 sequences for the outgroup. The outgroup was determined by using Basic Local Alignment Search Tool (BLAST)⁵⁵ to find 20 sequences that are like the component A2-like protein from M. acetivorans MA_3967 (MA_RS20695). A multiple sequence alignment (MSA) was performed using the MUSCLE v5.1 plug-in⁵² in Geneious software v2024.0.2. A tree was then built using the Randomized Axelerated Maximum Likelihood (RAxML) algorithm v8.2.11⁵⁶ with protein model GAMMABLOSUM62 and 100 bootstrap replicates. The tree was visualized using iTOL v7 (Interactive Tree of Life)⁵⁷.

Quantification and statistical analysis

Statistical details of experiments can be found in the corresponding figure legends. Growth experiments were conducted in triplicate cultures and statistical analyses of growth parameters were performed in GraphPad Prism v10.3.1 using unpaired t-tests with Welch’s correction. Enzyme activity assays were conducted in technical triplicate and biological replicates which represent multiple independent purifications of component A2. All statistical tests were performed in GraphPad Prism v10.3.1. Error bars represent the standard deviation of the mean of three technical replicates (individual enzyme assays) or biological replicates (bulk activities calculations).

Supplementary Material

1

Data S1. Tryptic peptides and identified proteins by mass spectrometry of SDS-PAGE bands from component A2 and MCR purifications. Related to Figure 1 and Figure S2. A) Proteins identified in 70 kDa band of component A2 purification. B) Proteins identified in 60 kDa band of component A2 purification. C) Proteins identified in 50 kDa band of component A2 purification. D) Proteins identified in 45 kDa band of component A2 purification. E) Proteins identified in 30 kDa band of component A2 purification. F) Proteins identified in 28 kDa band of component A2 purification. G) Proteins identified in 27 kDa band of component A2 purification. H) Proteins identified in 27 kDa band of component A2 purification. I) Proteins identified in 50 kDa band of component A2 purification. J) Proteins identified in 60 kDa band of MCR purification.

2

Data S2. Sequences used to build protein sequence alignment and phylogenetic tree. Related to STAR Methods and Figures 3 and 6. A) Sequences used to build protein alignment and input for clustering analysis. B) Sequences used to build phylogenetic tree.

Document S1. Figures S1–S6, Tables S1–S3, and Supplemental references.

Highlights.

• Component A2 is a distinct ABC-type ATPase found in methane-metabolizing archaea

• Component A2 is a redox-sensitive ATPase

• Component A2 hydrolyzes ATP when it binds methyl-coenzyme M reductase (MCR)

• ATP hydrolysis by component A2 is an essential step in the maturation of MCR