Letters pubs.acs.org/acschemicalbiology
Thiamine Pyrophosphate Stimulates Acetone Activation by Desulfococcus biacutus As Monitored by a Fluorogenic ATP Analogue Olga B. Gutiérrez Acosta,†,§ Norman Hardt,‡,§ Stephan M. Hacker,‡ Tobias Strittmatter,‡ Bernhard Schink,*,† and Andreas Marx*,‡ †
Department of Biology and ‡Department of Chemistry, Konstanz Research School Chemical Biology, University of Konstanz, Universitätsstr. 10, 78457 Konstanz, Germany S Supporting Information *
ABSTRACT: Acetone can be degraded by aerobic and anaerobic microorganisms. Studies with the strictly anaerobic sulfate-reducing bacterium Desulfococcus biacutus indicate that acetone degradation by these bacteria starts with an ATPdependent carbonylation reaction leading to acetoacetaldehyde as the first reaction product. The reaction represents the second example of a carbonylation reaction in the biochemistry of strictly anaerobic bacteria, but the exact mechanism and dependence on cofactors are still unclear. Here, we use a novel fluorogenic ATP analogue to investigate its mechanism. We find that thiamine pyrophosphate is a cofactor of this ATP-dependent reaction. The products of ATP cleavage are AMP and pyrophosphate, providing first insights into the reaction mechanism by indicating that the reaction proceeds without intermediate formation of acetone enol phosphate. bond between the β/γ-phosphate, or AMP and pyrophosphate by cleavage of the α/β-phosphate bond. Furthermore, whether cofactors contribute to the process could not be assayed up to now. In this study, using novel fluorogenic ATP analogues, we found that thiamine pyrophosphate (TPP) plays a crucial role in the transformation of acetone by D. biacutus. Furthermore, we found that the products of ATP cleavage are AMP and pyrophosphate, indicating that acetone enol phosphate, a conceivable reaction intermediate, is not formed. For the investigation of acetone transformation in cell-free extracts (CE) we envisioned to utilize an ATP-based Försterresonance energy transfer (FRET)-cassette that should monitor ATP consumption in real time as recently reported in studies on isolated enzymes.8,9 In this concept, an ATP analogue is intramolecularly equipped with a fluorescent donor at the γphosphate and an acceptor attached to the nucleoside part of ATP (Figure 1B). In the noncleaved state, the excited donor fluorophore can transfer its energy to the acceptor via FRET. If a nonfluorescent acceptor dye (dark quencher) is used, this results in a nonfluorescent ATP analogue. Upon cleavage of the probe by an enzyme, the acceptor dye is spatially separated from the donor dye, and the specific fluorescence emission of the donor is restored. In order to employ these analogues as activity probes toward ATP-dependent processes in this specific study on CE of D.
A
cetone is an important solvent used in chemical industry.1 Its microbial degradation has been studied with aerobic and anaerobic bacteria. Nitrate-reducing and aerobic bacteria activate acetone in an ATP-dependent carboxylation reaction that consumes at least two ATP equivalents per molecule acetone and forms acetoacetate as a first reaction product.2−5 Desulfococcus biacutus is a strictly anaerobic bacterium that can grow with acetone as carbon and electron source and sulfate as electron acceptor.6 Unlike aerobic and nitrate-reducing bacteria, D. biacutus activates acetone not by carboxylation but uses carbon monoxide (CO) as a co-substrate in the activation reaction (Figure 1A). The product of this carbonylation reaction is acetoacetaldehyde, which is further transformed to acetoacetyl-CoA and subsequently to two molecules of acetylCoA.7 The mechanism of the novel carbonylation reaction is still under investigation. The product of the carbonylation reaction, acetoacetaldehyde, is highly unstable, and therefore could be identified only as the dinitrophenylhydrazone derivative or after derivatization with guanidine leading to 2-amino-4-methylpyrimidine.7 Different from nitrate-reducing and aerobic bacteria2−5 it is unlikely that D. biacutus invests two ATP equivalents into this activation reaction because the overall energy budget of sulfate-reducing bacteria is very tight. Nevertheless, acetone activation by sulfate reducers does require some energy input by ATP hydrolysis in the initial step, as shown recently.7 Due to the instability of the product acetoacetaldehyde, the activating reaction is difficult to characterize. Therefore, it is not known whether ATP is hydrolyzed during this process to result in ADP and phosphate by cleavage of the phoshoranhydride © 2014 American Chemical Society
Received: February 27, 2014 Accepted: April 29, 2014 Published: April 29, 2014 1263
dx.doi.org/10.1021/cb500152y | ACS Chem. Biol. 2014, 9, 1263−1266
ACS Chemical Biology
Letters
Figure 1. Investigation of the novel carbonylation reaction of acetone in D. biacutus. (A) Pathway proposed by Gutierrez Acosta et al. involving ATPand CO-dependent acetone activation by Desulfococcus biacutus.7 (B) Modified ATP probe, optimized for experiments with cell-free extracts, bearing a fluorescent donor D attached to the γ-phosphate and a quencher Q attached to the C2-position of the nucleobase.
Figure 2. Synthesis and fluorescence characteristics of probe 1. (A) Compound 2 was synthesized as described.9 Reagents and conditions: a) Eclipsequencher-NHS, 0.1 M NaHCO3 (pH 8.7), DMF, rt, overnight, 3 in 26%. b) TCEP, H2O/MeOH/NEt3, rt, 4 h, 4 in 57%. c) SulfoCy3-NHS, 0.1 M NaHCO3 (pH 8.7), DMF, rt, overnight, 1 in 80%. (B) Reaction of the fluorogenic ATP activity probe 1 with phosphodiesterase I (SVPD) of C. adamanteus. (C) Fluorescence emission spectra of the ATP probe 1 without SVPD (black) and after digestion with SVPD (gray); excitation wavelength was 532 nm. (D) RP-HPLC analysis of the negative control without SVPD (left) and the SVPD-digested ATP probe 1 (right) were performed according to the general procedures. After 30 min of incubation quantitative conversion is detected. Fractions of negative and positive control were identified by high resolution mass spectrometry (HRMS).
biacutus, it is required to ensure its stability in CE. It was shown recently10 that phosphoester modifications are stable in a wide pH range and are therefore suited to investigate enzyme reactions. However, all recently reported ATP-based FRETcassettes8,9 were not stable in CE of D. biacutus. We found that the fluorogenic ATP analogue 1 was suited for the analysis of enzymatic reactions performed with CE of D. biacutus. The novel probe carries a Sulfo-Cy3 dye at the γ-moiety of the
triphosphate and an eclipse-quencher connected to the C2position of the nucleobase. The synthesis of probe 1 starts with the modified adenosine triphosphate 2 that was reported before9,11 (Figure 2A). The eclipse-quencher was introduced by treatment of the free amine with the corresponding N-hydroxysuccinimide (NHS) ester of the dye to form 3 with 26% yield. The azide in compound 3 was reduced using tris(2-carboxyethyl)phosphine (TCEP), 1264
dx.doi.org/10.1021/cb500152y | ACS Chem. Biol. 2014, 9, 1263−1266
ACS Chemical Biology
Letters
cofactor in acetone activation by D. biacutus. The hydrolysis rate of probe 1 was two times higher when the protein concentration was doubled (data not shown). The hydrolyzing activity of 1 depended furthermore on acetone and CO, indicating the specificity of the reaction. The ATP analogue 1 was cleaved to a minor extent also by other ATP-cleaving enzyme(s) present in the CE, as indicated by the background activity measured in the absence of acetone (Figure 3A). We next investigated the effect of the subsequent oxidation of acetoacetaldehyde to acetoacetyl-CoA on the rate of hydrolysis of probe 1 (Figure 3B). Addition of CoA and NAD+ to the reaction mix containing TPP caused a further increase of the activity (28.7 ± 3.11 nmol min−1 mg−1 total protein), but no significant increase was observed after addition of only CoA or only NAD+. This rate is comparable with the rate of the acetone activation reaction in the presence of nonmodified ATP; a specific activity of 40.1 nmol min−1 mg−1 total protein was measured for acetoacetaldehyde formation with subsequent reaction with guanidine.7 Furthermore, it is even higher than the specific substrate turnover rate of D. biacutus cells growing with acetone and sulfate (19 nmol min−1 mg−1 total protein as calculated from the doubling time and the substrate-specific growth yield).12 Compared to a TPP-free control with CoA plus NAD+, TPP addition caused even a 7fold stimulation of the overall reaction. The increase in the hydrolysis rate of probe 1 after addition of CoA and NAD+ is consistent with the novel pathway for acetone degradation that we proposed before,7 in which the formed acetoacetaldehyde is oxidized further to acetoacetylCoA with CoA and NAD+ as cofactors. Oxidation of acetoacetaldehyde after acetone activation was also shown before by mass spectrometry.7 In the absence of these cofactors, the activity is impaired, perhaps due to accumulation of the highly reactive acetoacetaldehyde. Reactive aldehydes are highly toxic because of their ability to form adducts with functional groups of proteins.13,14 To further characterize the fate of probe 1 during the enzymatic reaction, RP-HPLC analysis of the reaction mixtures was performed. In the absence of CO only probe 1 could be detected (Figure 3C). This shows that probe 1 is stable to the applied conditions and can be reliably applied in experiments with CE. When the reactions were performed in the presence of CO a single peak eluted at the retention time and with the spectral properties of compound 5. The corresponding cleavage product 6 could not be detected, possibly due to its low absorbance. The identity of cleavage product 5 was further proven by HRMS. Production of 5 bearing the diphosphate moiety shows that ATP is cleaved to AMP and pyrophosphate in the process of acetone carbonylation and reveals the mode of ATP cleavage during this process. Cleavage of ATP to AMP and pyrophosphate indicates that acetone is not phosphorylated to phosphoenol-acetone in this reaction. However, formation of acetonyl-AMP as an intermediate cannot be ruled out at the moment. Our findings strongly indicate that TPP assists the acetone activating enzyme(s) in D. biacutus, by acting as a cofactor in the initial reaction. Furthermore, we also show that acetone is not phosphorylated by ATP to give phosphoenol-acetone as ATP is cleaved to yield AMP and pyrophosphate. Although the mechanism of TPP involvement in acetone activation is still unclear, one may speculate that TPP forms an intermediate with acetone or an acetone analogue that has been previously activated by ATP. Alternatively, one may speculate a reaction
generating the amino group to form 4 in 57% yield. Finally, Sulfo-Cy3 was introduced as described above using NHS ester chemistry, resulting in ATP analogue 1 with 80% yield. To evaluate the optical properties of 1, fluorescence spectra of the noncleaved and cleaved state were measured. The cleavage of probe 1 between the α- and the β-phosphate was promoted using snake venom phosphodiesterase (SVPD) of Crotalus adamanteus (C. adamanteus) that is known to efficiently cleave this type of molecule and liberate both dyes of the FRET cassette8 (Figure 2B−D). The experiment shows that upon cleavage of the donor dye from the acceptor the fluorescence intensity changes dramatically, rendering the probe suitable as a nucleotide-based activity probe toward ATP-consuming enzymes. We next set out to use 1 to study acetone carbonylation in extracts of D. biacutus (Figure 3A and B). We detected a linear
Figure 3. Reaction of ATP analogue 1 in cell-free extract (CE) of D. biacutus. All experiments were performed under anoxic conditions at 30 °C. Concentrations of the substrates were 1 mM acetone, 0.5 mM ATP probe 1, 2 mM TPP, 2 mM CoA, 4 mM NAD+, and 10% CO in the headspace. (A) Time course experiments of the hydrolysis of ATP probe 1 in the acetone activation reaction. (B) Activity of hydrolysis of ATP probe 1 with addition of cofactors after 60 min of enzymatic reaction. (C) RP-HPLC analysis after 60 min of the control under nitrogen (top) gave intact ATP probe with a retention time of 21 min; the positive control (bottom) showed cleavage product 5 with a retention time of 16 min and typical spectral properties. Fractions of negative and positive control were further identified by high resolution mass spectrometry.
increase of Sulfo-Cy3 fluorescence over time in the presence of CE, acetone, and CO. The rate of 1 consumption was determined to be 2.1 ± 0.47 nmol min−1 mg−1 total protein. In order to gain further insight into the reaction mechanism, we checked for a possible requirement of cofactors (Figure 3A and B). Initial analysis in our laboratory indicated that TPP might be involved in the transformation of acetone. Thus, we investigated a possible involvement of TPP as cofactor in the metabolism of acetone. TPP addition caused a nearly 4-fold increase of the rate of hydrolysis of probe 1 (7.77 ± 0.27 nmol min−1 mg−1 total protein), which indicates that TPP acts as a 1265
dx.doi.org/10.1021/cb500152y | ACS Chem. Biol. 2014, 9, 1263−1266
ACS Chemical Biology
Letters
strains and comparison of acetone carboxylase enzymes. Appl. Environ. Microb. 77, 6821−6825. (5) Schühle, K., and Heider, J. (2012) Acetone and butanone metabolism of the denitrifying bacterium ″Aromatoleum aromaticum″ demonstrates novel biochemical properties of an ATP-dependent aliphatic ketone carboxylase. J. Bacteriol. 194, 131−141. (6) Platen, H., Temmes, A., and Schink, B. (1990) Anaerobic degradation of acetone by Desulfococcus biacutus spec. nov. Arch. Microbiol. 154, 355−361. (7) Gutierrez Acosta, O. B., Hardt, N., and Schink, B. (2013) Carbonylation as a key reaction in anaerobic acetone activation by Desulfococcus biacutus. Appl. Environ. Microb. 79, 6228−6235. (8) Hacker, S. M., Pagliarini, D., Tischer, T., Hardt, N., Schneider, D., Mex, M., Mayer, T. U., Scheffner, M., and Marx, A. (2013) Fluorogenic ATP analogues for online monitoring of ATP consumption: observing ubiquitin activation in real time. Angew. Chem., Int. Ed. 52, 11916−11919. (9) Hardt, N., Hacker, S. M., and Marx, A. (2013) Synthesis and fluorescence characteristics of ATP-based FRET probes. Org. Biomol. Chem. 11, 8298−8305. (10) Hacker, S. M., Mex, M., and Marx, A. (2012) Synthesis and stability of phosphate modified ATP analogues. J. Org. Chem. 77, 10450−10454. (11) Hacker, S. M., Hardt, N., Buntru, A., Pagliarini, D., Mockel, M., Mayer, T. U., Scheffner, M., Hauck, C. R., and Marx, A. (2013) Fingerprinting differential active site constraints of ATPases. Chem. Sci. 4, 1588−1596. (12) Janssen, P. H., and Schink, B. (1995) Catabolic and anabolic enzymes activities and energetics of acetone metabolism of the sulfatereducing baterium Desulfococcus biacutus. J. Bacteriol. 177 (2), 277−82. (13) Schauenstein, E. (1967) Autoxidation of polyunsaturated esters in water: chemical structure and biological activity of the products. J. Lipid Res. 8, 417−428. (14) Esterbauer, H., Zollner, H., and Scholz, N. (1975) Reaction of glutathione with conjugated carbonyls. Z. Naturforsch. C 30, 466−473.
between TPP and CO and a subsequent addition to (activated) acetone. In any case, the reaction would result in the consumption of only one ATP equivalent for the synthesis of acetoacetyl-CoA, whereas acetone-carboxylating aerobic and nitrate-reducing bacteria have to invest at least two ATP equivalents to obtain the same product. This would be favorable for a sulfate-reducing bacterium with its very tight energy budget. The reaction mechanism is to be further elucidated in our lab, with special focus on a TPPacetoacetaldehyde intermediate. Conclusions. In the present study, we investigated the acetone carbonylation reaction in D. biacutus using a novel fluorogenic ATP analogue optimized for experiments in cell extracts. We find that ATP is cleaved to AMP and pyrophosphate, and that thiamine pyrophosphate is a cofactor that enhances the enzymatic acetone activation. These findings suggest that phosphorylation of acetone to phosphoenolacetone is not involved in this specific enzymatic reaction, but acetone is probably activated by ATP in a different manner resulting in the formation of AMP and pyrophosphate.
■ ■
METHODS
Full experimental details are given in the Supporting Information.
ASSOCIATED CONTENT
S Supporting Information *
This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Authors
*E-mail:
[email protected]. *E-mail:
[email protected]. Author Contributions §
These authors contributed equally to this work.
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We gratefully acknowledge funding by the Deutsche Forschungsgemeinschaft within the SFB 969 and the SPP 1319 priority program, the Studienstiftung des Deutschen Volkes and the Zukunftskolleg of the University of Konstanz for a stipend to S.M.H. and the Konstanz Research School Chemical Biology for fellowships granted to O.B.G.A., N.H., and S.M.H. We thank A. Wiese for preparation of bacterial growth media.
■
REFERENCES
(1) Sifniades, S., Levy, A. B., and Bahl, H. (2011) Acetone, in Ullmann’s Encyclopedia of Industrial Chemistry; Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim. (2) Sluis, M. K., Small, F. J., Allen, J. R., and Ensign, S. A. (1996) Involvement of an ATP-dependent carboxylase in a CO2-dependent pathway of acetone metabolism by Xanthobacter strain Py2. J. Bacteriol. 178, 4020−4026. (3) Sluis, M. K., Larsen, R. A., Krum, J. G., Anderson, R., Metcalf, W. W., and Ensign, S. A. (2002) Biochemical, molecular, and genetic analyses of the acetone carboxylases from Xanthobacter autotrophicus strain Py2 and Rhodobacter capsulatus strain B10. J. Bacteriol. 184, 2969−2977. (4) Dullius, C. H., Chen, C. Y., and Schink, B. (2011) Nitratedependent degradation of acetone by Alicycliphilus and Paracoccus 1266
dx.doi.org/10.1021/cb500152y | ACS Chem. Biol. 2014, 9, 1263−1266
