Studies of the CobA-type ATP:Co(I)rrinoid adenosyltransferase enzyme of Methanosarcina mazei strain Go1

Abstract

Although methanogenic archaea use B(12) extensively as a methyl carrier for methanogenesis, little is known about B(12) metabolism in these prokaryotes or any other archaea. To improve our understanding of how B(12) metabolism differs between bacteria and archaea, the gene encoding the ATP:co(I)rrinoid adenosyltransferase in Methanosarcina mazei strain Gö1 (open reading frame MM3138, referred to as cobA(Mm) here) was cloned and used to restore coenzyme B(12) synthesis in a Salmonella enterica strain lacking the housekeeping CobA enzyme. cobA(Mm) protein was purified and its initial biochemical analysis performed. In vitro, the activity is enhanced 2.5-fold by the addition of Ca(2+) ions, but the activity was not enhanced by Mg(2+) and, unlike the S. enterica CobA enzyme, it was >50% inhibited by Mn(2+). The CobA(Mm) enzyme had a K(m)(ATP) of 3 microM and a K(m)(HOCbl) of 1 microM. Unlike the S. enterica enzyme, CobA(Mm) used cobalamin (Cbl) as a substrate better than cobinamide (Cbi; a Cbl precursor); the beta phosphate of ATP was required for binding to the enzyme. A striking difference between CobA(Se) and CobA(Mm) was the use of ADP as a substrate by CobA(Mm), suggesting an important role for the gamma phosphate of ATP in binding. The results from (31)P-nuclear magnetic resonance spectroscopy experiments showed that triphosphate (PPP(i)) is the reaction by-product; no cleavage of PPP(i) was observed, and the enzyme was only slightly inhibited by pyrophosphate (PP(i)). The data suggested substantial variations in ATP binding and probably corrinoid binding between CobA(Se) and CobA(Mm) enzymes.

FIG. 1.

Alignment of CobA from S. enterica with archaeal homologs. The primary amino acid sequence of CobA_Se is 31% identical and 52% similar to that of CobA_Mm. The dark-shaded box indicates the P-loop ATP-binding motif, while the light-gray box indicates regions of conservation. Asterisks denote absolutely conserved amino acid residues. Sen, Salmonella enterica; Mma, Methanosarcina mazei strain Gö1; Mac, Methanosarcina acetivorans strain C2A; Has, Halobacterium salinarum strain NRC-1; Pae, Pyrobaculum aerophilum; Pab, Pyrococcus abyssi; Pfu, Pyrococcus furiosus; Pho, Pyrococcus horikoshii; Tac, Thermoplasma acidophilum; Tvo, Thermoplasma volcanium. Alignment was generated by using DNA* software package v.1.66 (DNASTAR, Inc.) without adjustments.

FIG. 2.

cobA⁺_Mm compensates for the lack of cobA_Se under conditions that require low levels of AdoCbl for growth. (A) Growth behavior of S. enterica strains grown in minimal medium containing glycerol as sole carbon and energy source. (B) Growth in minimal medium containing ethanolamine as sole carbon and energy source. Derivatives of strain JE7180 [cobA366::Tn10d(cat⁺) eut1141 (ΔeutT)] carrying plasmids pMmaCOBA4 cobA⁺_Mm (strain JE7954) or pT7-7 (VOC, strain JE7955) were used to investigate CobA_Mm function in vivo. Strains TR6583 (cobA⁺ eut⁺) and JE1293 [cobA366::Tn10d(cat⁺)] were used as controls. VOC, vector-only control.

FIG. 3.

CobA_Mm protein purity. CobA_Mm protein was isolated as described in Materials and Methods. After Coomassie blue staining, protein purity was quantified by using a Fotodyne.

FIG. 4.

CobA_Mm enzyme activity is enhanced by the addition of Ca²⁺ ions. Various metal salts were tested for their effect on CobA_Mm activity in vitro. Salts were added to the reaction mixture containing 0.2 M Tris-Cl buffer (pH 8, 37°C), CobA_Mm (2 μm), ATP (100 mM), HOCbl (50 μM), and Ti(III) citrate (2.5 mM) as a reductant.

FIG. 5.

Kinetic analysis of the CobA_Mm-catalyzed reaction. (A) HOCbl was held constant at 0.05 mM, while the concentration of ATP was varied. V_max was determined to be 3 nmol of AdoCbl formed min⁻¹ mg⁻¹; the apparent K_m^ATP was calculated at 3 μM. (B) The ATP concentration was held constant at 0.50 mM, while the HOCbl concentration was varied. V_max was 3 nmol of AdoCbl formed min⁻¹ mg⁻¹, and the apparent K_m^HOCbl was calculated to be 1 μM.

FIG. 6.

CobA_Mm protein interacts with the 2′-OH and the γ phosphate of ATP but is not inhibited by PPP_i. (A) Specificity of the CobA_Mm enzyme for its nucleoside triphosphate substrate. When added, ADP was present in the reaction mixture at 500 μM. The reaction rate for a mixture containing ATP was 2.3 nmol of AdoCbl min⁻¹ ± 0.1. Ado, adenosine. (B) When added to the reaction mixture, PPP_i, PP_i, or P_i was present at 1 mM. The reaction rate for a mixture with no additions was 1.8 nmol of AdoCbl min⁻¹.

FIG. 7.

³¹P-NMR analysis of CobA_Mm reaction by-products. Strong signals for the β phosphate of ATP (centered at −21.51 ppm) and PPP_i (centered at −22.39 ppm) were observed after incubation of CobA_Mm with its substrates. No signal was detected from PP_i at −7.41 ppm. (A) Complete reaction mixture; (B) ATP standard; (C) PPP_i standard; (D) mixture of ATP, PPP_i, PP_i, P_i, and HOCbl.