Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Aug 28;57(34):5076-5087.
doi: 10.1021/acs.biochem.8b00715. Epub 2018 Aug 16.

A New Class of EutT ATP:Co(I)rrinoid Adenosyltransferases Found in Listeria Monocytogenes and Other Firmicutes Does Not Require a Metal Ion for Activity

Free PMC article

A New Class of EutT ATP:Co(I)rrinoid Adenosyltransferases Found in Listeria Monocytogenes and Other Firmicutes Does Not Require a Metal Ion for Activity

Flavia G Costa et al. Biochemistry. .
Free PMC article


ATP:Co(I)rrinoid adenosyltransferases (ACATs) are involved in de novo adenosylcobamide (AdoCba) biosynthesis and in salvaging complete and incomplete corrinoids from the environment. The ACAT enzyme family is comprised of three classes of structurally and evolutionarily distinct proteins (i.e., CobA, PduO, and EutT). The structure of EutT is unknown, and an understanding of its mechanism is incomplete. The Salmonella enterica EutT ( SeEutT) enzyme is the best-characterized member of its class and is known to be a ferroprotein. Here, we report the identification and initial biochemical characterization of an enzyme representative of a new class of EutTs that does not require a metal ion for activity. In vivo and in vitro evidence shows that the metal-free EutT homologue from Listeria monocytogenes ( LmEutT) has ACAT activity and that, unlike other ACATs, the biologically active form of LmEutT is a tetramer. In vitro studies revealed that LmEutT was more efficient than SeEutT and displayed positive cooperativity. LmEutT adenosylated cobalamin, but not cobinamide, showed specificity for ATP and 2'-deoxyATP and released a triphosphate byproduct. Bioinformatics analyses suggest that metal-free EutT ACATs are also present in other Firmicutes.

Conflict of interest statement

CONFLICT OF INTEREST. The authors have no conflict of interest with the content of this article.


Figure 1
Figure 1. Chemical structures of vitamin B12, coenzyme B12, and the pathway that converts the vitamin to the coenzyme
A. Structures of cyanocobalamin (a.k.a. CNCbl, vitamin B12) and adenosylcobalamin a.k.a. AdoCbl, coenzyme B12). Vitamin B12 is converted to coenzyme B12 in two one-electron reduction steps. The first step is thought to be enzyme independent, and can be driven by dihydroflavins; the second electron can also be donated by dihydroflavins once ATP and the corrinoid substrate are bound to the active site of the ATP:Co(I)rrinoid adenosyltranferase (ACAT). The ACAT catalyzes the nucleophilic attack of Co(I) to the 5′ C of ATP. B. General corrinoid adenosylation pathway catalyzed by ACATs. ATP binds before the 5-coordinate (5c) Co(II)balamin substrate. The latter is converted to four-coordinate (4c) Co(II)balamin upon binding by displacement of the lower ligand base. The 4c Co(II)balamin species is reduced by dihydroflavins to generate the Co(I)balamin supernucleophile that attacks ATP, releasing a phosphate byproduct.
Figure 2
Figure 2. A EutT sequence alignment reveals two classes of EutT proteins
A portion of interest from the multiple sequence alignment was selected for this figure, the complete sequence alignment is presented in figure S1. Homologues from strains in bold were tested for preliminary in vitro activity. Amino acids that participate in metal-binding are highlighted in yellow. The two classes of EutTs are indicated by the labels ‘Class I’ and ‘Class II’. The numbers above the sequence alignment correspond to the EutT amino acid sequence of S. enterica.
Figure 3
Figure 3. LmEutT is active in vivo
S. enterica strains were grown in minimal medium with ethanolamine as carbon and energy source (panel A) or minimal medium with glycerol as carbon and energy source and ethanolamine as nitrogen source (panel B). Strains were grown in 200 μL of medium in microtiter dishes and growth was monitored at 630 nm with a plate reader. The experiment was repeated three times, and the error bars represent the standard error between three biological replicates of one experiment.
Figure 4
Figure 4. LmEutT synthesized sufficient AdoCbl to activate the transcription of the eut operon
All strains were grown in NCE plus lactose minimal medium. Strains were grown in 200 μL of medium as described in Figure 3. The experiment was repeated three times, and the error bars represent the standard error between three biological replicates. The ΔACAT eutE::MudJ (lacZ+) strain did not grow with lactose as the carbon and energy source (closed circles, open circles respectively). Strains carrying a chromosomal eutE::MudJ (lacZ+) fusion and a plasmid encoding an active EutT grew with lactose as carbon and energy source.
Figure 5
Figure 5. LmEutT is active in vitro
A. The product of reaction mixtures containing SeEutT and LmEutT proteins incubated with the indicated substrates was fed to a strain that required AdoCbl for growth on ethanolamine (ΔcobA Δpdu Δeut). Authentic AdoCbl and HOCbl standards were added as controls; + hν indicates that the reaction was exposed to light. Growth was monitored in a 96-well plate reader. The experiment was repeated twice, and error bars represent the standard error of a biological triplicate. B. UV-vis spectra of the AdoCbl product of the LmEutT-catalyzed reaction, before (dark gray line) and after (light gray line) irradiation. Authentic AdoCbl (Sigma) was used for comparison before (dark gray dotted line) and after (light gray dotted line) irradiation (inset graph).
Figure 6
Figure 6. LmEutT is specific for cobalamin
LmEutT activity with cobalamin or cobinamide were tested using the Co1+ assay [Ti(III)citrate)], the Fre-dependent Co(II) assay, or the FldA-dependent Co(II) assay. SeEutT and SeCobA were also assayed as controls. LmEutT adenosylated HOCbl under all assay conditions, but only adenosylated Co(I)Cbi.
Figure 7
Figure 7. LmEutT releases triphosphate as the by-product
All samples were in Tris•HCl buffer (100 mM Tris, pH 7.5 at 37 °C) containing KCl (500 mM), MgCl2 (2 mM) and Ti(III)citrate (1 mM). A. In the presence of LmEutT (20 μg mL−1), HOCbl (100 μM) and ATP (3 mM) the formation of a triphosphate product was evident, (−6.03, −6.15; −20.57, −20.70, and −20.82 ppm). Integration of these peaks gave values of 1.05 (doublet), 1.08 (triplet), and 1.00 (doublet) for the α, β, and γ phosphate of ATP; and 0.35 (doublet) and 0.14 (triplet) for the α and β phosphate of the PPPi byproduct, respectively. B. This triphosphate product was not present when HOCbl was omitted from the reaction. Regions of interest are magnified in the NMR spectra inset. Integration of these peaks gave values of 1.01 (doublet), 0.97 (triplet), and 1.00 (doublet) for the α, β, and β phosphate of ATP, respectively. C. ATP and HOCbl concentrations were as in the experiment shown in panel A, but phosphate product standards were each added to a final concentration of 1 mM. Shifts for the phosphate standards were the following: Pi (−2.46 ppm), PPi (−6.87 ppm), PPPi (−6.07, −6.19, −20.62, −20.75, −20.87 ppm). Integration of these peaks provided the following: 1.01 (doublet), 0.99 (triplet), and 1.00 (doublet) for the α, β, and γ phosphate of ATP, respectively; 0.21 (doublet) and 0.11 (triplet) for the PPPi standard; 0.12 (singlet) for the PPi standard; 0.27 (singlet) for the Pi standard.
Figure 8
Figure 8. LmEutT kinetics
The kinetics of AdoCbl formation was analyzed with A. Varying concentrations of Cbl (x-axis) and saturating ATP (1 mM); B. Varying concentrations of ATP (x-axis) and saturating Cbl (50 μM); C. Varying concentrations of dATP (x-axis) and saturating Cbl (50 μM).
Figure 9
Figure 9. LmEutT forms a tetramer
Purified LmEutT was applied onto a size exclusion chromatography column (Superose 12 10/300 GL, GE) and the elution time was compared to markers of known molecular masses (γ-globulin, 158 kDa; ovalbumin, 44 kDa; myoglobin, 17 kDa, vitamin B12, 1.35 kDa). A. Data were plotted as the log10 of the molecular mass as a function of elution time. The white squares denote the standards, and the gray circle represents the mean elution time of LmEutT in three different experiments. B. A sample trace of LmEutT eluting from the size exclusion chromatography column is shown. The large peak corresponds to an LmEutT tetramer. The small peak appears to be a protein aggregate >1000 kDa. The identity of the aggregated protein was not established.

Similar articles

See all similar articles

Cited by 2 articles

Publication types