Abstract
A combination of NRVS and FT-IR spectroscopies and DFT calculations was used to observe and characterize Fe-H/D bending modes in CrHydA1 [FeFe]-hydrogenase Cys-to-Ser variant C169S. Mutagenesis of cysteine to serine at position 169 changes the functional group adjacent to the H-cluster from a –SH to –OH, thus altering the proton transfer pathway. C169S has a significant reduction in catalytic activity compared to the native CrHydA1, presumably due to less efficient transfer of protons to the H-cluster. This mutation allowed effective capture of a hydride/deuteride intermediate and facilitated direct detection of the Fe-H/D normal modes. We find a significant shift to higher frequency in a Fe-H bending mode of the C169S variant, as compared to previous findings with reconstituted native and oxadithiolate (ODT) substituted CrHydA1. Rationalized by DFT calculations, we propose that this shift is caused by a stronger interaction between the –OH of C169S with the bridgehead –NH– of the active site, as compared to the –SH of C169 in the native enzyme.
Keywords: hydrogenase, NRVS, NIS, DFT, FT-IR, catalysis, hydride
Graphical Abstract
Combination of NRVS measurements, DFT calculations, and mutagenesis identified Fed-H bond vibrational modes and observed effects of amino acid environment that surrounds the [FeFe]-hydrogenase H-cluster.
Hydrogenases are enzymes that catalyze the reversible interconversion of molecular hydrogen with protons and electrons: H2 ⇄ 2H+ + 2e−.[1] Under optimal electron delivery hydrogenases require little or no driving force while utilizing base transition metals as catalytic cofactors. The [FeFe]-hydrogenases ([FeFe]-H2ases) are known to possess exceptionally high turnover rates of >104 s−1.[1a, 2] Resolving the mechanistic properties of [FeFe]-H2ase requires identification of the catalytically relevant intermediates,[3] and is of interest to advancing catalyst design for developing H2 fuel-cells and renewable energy production technologies.[4]
The active site for all known [FeFe]-H2ases consists of an ‘H-cluster’ that is composed of [4Fe-4S]H and [2Fe]H subclusters, linked by a cysteine thiolate (Figure 1).[6] The main site of catalytic H2 bond activation is the Fed iron site of [2Fe]H distal to [4Fe-4S]H, where the azadithiolate (ADT) group has a fundamental role in proton transfer. The nitrogen NADT of the ADT bridge is involved in transporting solvent protons to and from Fed through a relay of conserved amino acids that lead to the surface of the protein.[7] In one of the simplest structural forms of the enzyme, Chlamydomonas reinhardtii [FeFe]-H2ase HydA1 (CrHydA1), the conserved relay is composed of residues C169, E141, S189, and E144 (Figure 1). The C169 side chain plays a crucial role in catalysis due to the –SH group positioned within H-bonding distance (≈2 Å) to the bridgehead amine of ADT.[7b]
Alteration of the ADT bridge or critical amino acids along the proton transfer pathway can abolish or severely reduce the H2 reduction activity. Examples include [FeFe]-H2ase variants C169A and C169S of CrHydA1,[3b, 3e] C299S of Clostridium pasteurianum (CpI), C298S of Clostridium acetobutylicum (CaHydA), and C178A of Desulfovibrio desulfuricans (DdHydAB).[7b, 8] Furthermore, using these modified proteins, one can accumulate significant amounts of an intermediate known as Hhyd, which contains a trapped terminal hydride Fed-H (Figure 1), and features a [4Fe-4S]H+-Fep(II)Fed(II) oxidation state of the H-cluster.[3b, 3e] [3c] [9]
Nuclear Resonance Vibrational Spectroscopy (NRVS), also known as Nuclear Inelastic Scattering (NIS), is a synchrotron-based X-ray technique that probes vibrational modes using nuclear excitation in Mössbauer-active isotopes.[10] When applied to 57Fe-labeled proteins, it selectively probes 57Fe nuclei vibrational motion. In previous works on CrHydA1 and DdHydAB [FeFe]-H2ases,[3c] [9] we have shown that 57Fed-H bending modes can be observed by NRVS. The NRVS signature of the terminal Fed-H hydride in Hhyd contains two bands >670 cm−1, rationalized by density functional theory (DFT) as H– motions perpendicular to and within the approximate mirror symmetry plane of [2Fe]H.
Herein, we extend our methodology to identify the Fed-H/D modes in the C169S CrHydA1 variant. Specifically, we report the Hhyd 57Fe-NRVS spectra of the biosynthesized [457Fe-4S]H-[257Fe]H C169S CrHydA1 (†C169S) species. We compare this new data to previously reported Hhyd spectra for reconstituted [257Fe]H native[9] (†RN) and [257Fe-ODT]H bridgehead-altered[3c] (†ODT) enzymes. The NRVS data reveal surprising spectral shifts produced by the –SH (†RN) to –OH (†C169S) substitution in the side chain functional group. The results are interpreted using DFT modeling of the H-cluster and its environment.
NRVS spectra for the †C169S samples prepared under reducing conditions for hydride enrichment (documented by FT-IR in Figure S1) in either H2O (†C169SH) or D2O (†C169SD) buffer are illustrated in Figures 2 and S2. Features in the 200–400 cm−1 range are primarily due to Fe-S motion, and hence they show a little change between either H or D isotopologues (Figure 2, top). A benefit from having an 57Fe-enriched [457Fe-4S]H subcluster is a presence of redox-sensitive [4Fe-4S]-specific bands in the NRVS data.[12] As illustrated in Figure 2, the positions of the Fe-S stretch bands at 268 and 362 cm−1 align nicely with a previously recorded spectrum of D14C variant of Pyrococcus furiosus (Pf) ferredoxin (Fd) in its reduced [4Fe-4S]+ form.[11] This is consistent with the assignment of Hhyd to the [4Fe-4S]H+-Fep(II)Fed(II) state having a reduced, paramagnetic Fe-S cubane.[3b, 3e] The NRVS features in the 400–600 cm−1 range, having weaker intensities and higher vibrational energies than the Fe-S bands, are mostly from modes with Fe-CN/CO character, as reported previously.[13]
At still higher energies are bands at 673 (A’) and 772 (A”) cm−1, which have been assigned to the Fed-H bending modes in the Hhyd state.[3c, 9] DFT modeling reveals that these modes contain only ≈8% 57Fe and as much as ≈60–80% H– contributions to their vibrational energies; see animations for the key normal modes in Supporting Information. The A’ and A” features red-shift under D2O conditions, and an Fed-D band (B) appears at 630 cm−1 (Figure 2, bottom) with its intensity enhanced due to coupling to the Fe-CO/CN motions. Band B is thus the most direct evidence for identifying an isotope effect on the Fed-H/D modes. The position of band B of the †C169SD sample changes insignificantly compared with previously published NRVS data on bands B from samples †RND and †ODTD. The relative positions of bands A’ and A” of †ODTH, †RNH, and †C169SH show very different behavior (see additionally Table S1). Band A’ of †C169SH, compared with †ODTH and †RNH, exhibits only small shifts within 5 cm−1 (Figure 2, bottom), less than the 8 cm−1 instrumental resolution of our spectra. However, band A” of †C169SH shows a dramatic 28 cm−1 upshift relative to †RNH, and an even larger 45 cm−1 upshift relative to †ODTH.
Since the NRVS spectra for the Hhyd state accumulated in the three CrHydA1 samples exhibit prominent shifts in their Fed-H bands, a computational model explaining these differences has to include the sample-specific structural variations. Recently, we proposed three levels of DFT modeling for the H-cluster in its Hhyd state: S, L, and L’.[9] S is a small model that contains the [2Fe]H subcluster only, sufficient to rationalize the variation between the †ODTH and †RNH spectra. L and L’ are larger models, both incorporating as well the [2Fe]H-surrounding amino acids including the one at position 169; L’ additionally includes the [4Fe-4S]H cluster (see Figures S4–S7). Model L is superior in reproducing the NRVS spectra >400 cm−1 containing the Fed-H/D bands, and model L’ is optimal for the spectra simulation <400 cm−1 (see Figure S2). In our previous work on the †RN system, the DFT modeling predicted that the Fed-H band structure is indicatively sensitive to the redox level of the [2Fe]H subcluster; Fe-H vibrations in both [FeFe]-H2ase and its model system[14] were found to be sensitive to the details of the environment. A model of Hhyd having the –NH– amino form with its HADT in an ‘Axial’ position pointing towards the Hh hydride (Hhyd-A) as illustrated in Figure 3, top was found to be the best fit to the observed NRVS spectra.[9]
The DFT modeling further indicates that inclusion of the NADT⋯HC/HS hydrogen bond between the ADT bridgehead and C/S169 side chain effectively strengthens the Hh⋯HADT dihydrogen interaction (Table 1, S vs L values) in both the native and C169S variant. The Hh⋯HADT interaction, absent in the bridgehead-altered ODT (–O–) species, is a crucial step for H2 formation from the Hhyd state in the proposed [FeFe]-H2ase mechanism. Interestingly, the C169S variant is predicted to have a ≈0.1 Å shorter Hh⋯HADT distance than the native enzyme (Table 1, L values). Moreover, the optimized C169S L-model has noticeably shorter NADT⋯HS and NADT⋯OS distances, which characterize the hydrogen bonding between ADT and S169. Overall the delocalized interactions involving Fed–Hh⋯HADT–NADT⋯HC/HS–SC/OS have shorter non-covalent bonding (⋯) distances in C169S versus the native enzyme and a ≈0.8 Å contraction of Fed⋯OS to 5.64 Å versus 6.24 Å for Fed⋯Sc.
Table 1.
DFT Modeling Level | Internuclear Distance (Å), C169 / C169S Variant[a] | ||||||
---|---|---|---|---|---|---|---|
Fed–Hh | Hh⋯HADT | NADT–HADT | NADT⋯HC / HS | SC / OS–HC / HS | NADT⋯SC / OS | Fed⋯SC / OS | |
S[b] | 1.51 | 2.14 | 1.03 | NA | NA | NA | NA |
L | 1.52 / 1.52 | 2.01 / 1.92 | 1.03 / 1.03 | 2.26 / 2.15 | 1.37 / 0.98 | 3.59 / 3.04 | 6.24 / 5.64 |
Atomic labels follow those used in Figures 3 top and S4–S7. Bonding and non-bonding interactions are marked in column headers as ‘-’ and ‘⋯’, correspondingly.
At the modeling level S, the C/S169 residue is not included. The internuclear distances within the [2Fe]H subcluster therefore take their single values, and ‘NA’ implies distances not applicable.
Even though the Fed⋯SC/OS distance is >5 Å, DFT was able to predict the effects of the change in functional groups on the Fed-Hh vibrational modes. Specifically, there is a ≈20 cm−1 calculated upshift of the 57Fe-PVDOS band A”, produced by the delocalized ‘in-plane’ normal mode (involving hydrogen nuclei Hh of hydride, HADT of ADT, and HC/S of C/S169) from 749 (C169) to 768 cm−1 (C169S), see Figures 3 bottom and S3. In contrast, the lower energy band A’, produced by the localized ‘out-of-plane’ normal mode (which is an essentially a pure Fed-Hh motion decoupled from other nuclei) is predicted to be two-fold less affected by ≈10 cm−1, from 665 (C169) to 675 (C169S) cm−1. The calculations therefore match sufficiently well with shifts observed by NRVS. Notably, the optimized Fed–Hh–HADT angle is ≈110° (Figure 3, top); the Hh displacement vector is therefore approximately inline to the Hh⋯HADT bonding for vibrational mode A”, and perpendicular to it for mode A’. Notably, the band B position is essentially invariant to the †ODTD/†RND/†C169SD system selection, from both the experiment and theory; the Fed-Dh normal mode responsible for B is decoupled from the cofactor bridgehead and C/S169.
Implicated in the catalysis, the Hh⋯HADT dihydrogen interaction of the Hhyd state is therefore stronger in the C169S variant than in the native enzyme. This can be rationalized due to the secondary coordination sphere substitution of the more polar –SH thiol group in cysteine (pKa ≈ 8) for the less polar –OH hydroxyl group in serine (pKa ≈ 13), which sets off a cascade of H-bonding effects linked to the active site. Due to the large difference in pKa between the two groups, this structural difference has also been proposed to alter essential proton-transfer steps required for the catalytic mechanism.[8c]
The above results indicate that the architecture of the amino acid environment around the H-cluster is critical for attaining an optimal form of the Fed-Hh hydride that enables the fast kinetics of reversible H2 activation observed in [FeFe]-H2ases. Analogous model compound studies indeed show that the geometry and precise orientation of the Fe-H Lewis acid-base pair is required to achieve heterolytic H2 bond cleavage.[15]
Through the combined application of protein mutagenesis, NRVS, FT-IR, and DFT calculations we have elucidated the effects of the modification of the proton transport chain of [FeFe]-H2ase on hydride formation. The results reveal that the Cys-to-Ser change modifies structural interactions near the H2 binding site and H-cluster ADT ligand. This suggests a broader role for the C169 residue through vibrational coupling in fine-tuning the hydrogen bonding interactions around the H-cluster required for rapid binding and turnover of H2. The observation of these effects highlights the potential of NRVS in providing detailed structural information on the [FeFe]-H2ase active site environment and, potentially, on other iron-hydride systems. In combination with DFT calculations and site-directed mutagenesis, we have found a synergistic approach to probing the Fe-H vibrational modes and uncovering how they are coupled to the secondary coordination sphere around the active site. We have shown that even an amino acid >5 Å away, such as C/S169, modulates the Fed-H/D bond in CrHydA1.
Supplementary Material
Acknowledgements
This work was supported by NIH GM-65440 (S.P.C.) and the Cluster of Excellence “Unifying Concepts in Catalysis” initiative of DFG (V.P.). This work was authored in part by Alliance for Sustainable Energy, LLC, the manager and operator of the National Renewable Energy Laboratory for the U.S. Department of Energy (DOE) under Contract No. DE-AC36–08GO28308. Funding for protein purification and FT-IR provided by U.S. Department of Energy, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences, and Biosciences (D.W.M., M.W.R., and P.W.K.). The views expressed in the article do not necessarily represent the views of the DOE or the U.S. Government. The U.S. Government retains and the publisher, by accepting the article for publication, acknowledges that the U.S. Government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this work, or allow others to do so, for U.S. Government purposes. NRVS experiments were supported by APS (under proposals 44733, 43032, and 49765) and SPring-8 (under JASRI proposals 2015B0103, 2015B1134, 2016A0103, 2016A1154, 2016B1347, 2017A0141, 2017A1115, 2017B0141, 2017B1331, and RIKEN proposals 20150048, 20160063, 20170048.
Footnotes
Supporting information for this article is given via a link at the end of the document.
References
- [1].a) Lubitz W, Ogata H, Rudiger O, Reijerse E, Chem. Rev 2014, 114, 4081–4148; [DOI] [PubMed] [Google Scholar]; b) Peters JW, Schut GJ, Boyd ES, Mulder DW, Shepard EM, Broderick JB, King PW, Adams MWW, Biochim. Biophys. Acta 2015, 1853, 1350–1369. [DOI] [PubMed] [Google Scholar]
- [2].Madden C, Vaughn MD, Díez-Pérez I, Brown KA, King PW, Gust D, Moore AL, Moore TA, J. Am. Chem. Soc 2011, 134, 1577–1582. [DOI] [PubMed] [Google Scholar]
- [3].a) Sommer C, Adamska-Venkatesh A, Pawlak K, Birrell JA, Rudiger O, Reijerse EJ, Lubitz W, J. Am. Chem. Soc 2017, 139, 1440–1443; [DOI] [PubMed] [Google Scholar]; b) Mulder DW, Guo Yisong, Ratzloff MW, King PW, J. Am. Chem. Soc 2017, 139, 83–86; [DOI] [PubMed] [Google Scholar]; c) Reijerse EJ, Pham CC, Pelmenschikov V, Gilbert-Wilson R, Adamska-Venkatesh A, Siebel JF, Gee LB, Yoda Y, Tamasaku K, Lubitz W, Rauchfuss TB, Cramer SP, J. Am. Chem. Soc 2017, 139, 4306–4309; [DOI] [PMC free article] [PubMed] [Google Scholar]; d) Winkler M, Senger M, Duan JF, Esselborn J, Wittkamp F, Hofmann E, Apfel UP, Stripp ST, Happe T, Nat. Commun 2017, 8; [DOI] [PMC free article] [PubMed] [Google Scholar]; e) Mulder DW, Ratzloff MW, Bruschi M, Greco C, Koonce E, Peters JW, King PW, J. Am. Chem. Soc 2014, 136, 15394–15402. [DOI] [PubMed] [Google Scholar]
- [4].De Poulpiquet A, Ciaccafava A, Szot K, Pillain B, Infossi P, Guiral M, Opallo M, Giudici-Orticoni MT, Lojou E, Electroanalysis 2013, 25, 685–695. [Google Scholar]
- [5].Pandey AS, Harris TV, Giles LJ, Peters JW, Szilagyi RK, J. Am. Chem. Soc 2008, 130, 4533–4540. [DOI] [PubMed] [Google Scholar]
- [6].a) Berggren G, Adamska A, Lambertz C, Simmons TR, Esselborn J, Atta M, Gambarelli S, Mouesca JM, Reijerse E, Lubitz W, Happe T, Artero V, Fontecave M, Nature 2013, 499, 66–69; [DOI] [PMC free article] [PubMed] [Google Scholar]; b) Esselborn J, Muraki N, Klein K, Engelbrecht V, Metzler-Nolte N, Apfel U-P, Hofmann E, Kurisu G, Happe T, Chem. Sci 2015, 7, 959–968. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].a) Fontecilla-Camps JC, Volbeda A, Cavazza C, Nicolet Y, Chem. Rev 2007, 107, 5411–5411; [DOI] [PubMed] [Google Scholar]; b) Cornish AJ, Gartner K, Yang H, Peters JW, Hegg EL, J. Biol. Chem 2011, 286, 38341–38347; [DOI] [PMC free article] [PubMed] [Google Scholar]; c) Peters JW, Lanzilotta WN, Lemon BJ, Seefeldt LC, Science 1998, 282, 1853–1858; [DOI] [PubMed] [Google Scholar]; d) Long H, King PW, Chang CH, J. Phys. Chem. B 2014, 118, 890–900; [DOI] [PubMed] [Google Scholar]; e) Ginovska-Pangovska B, Ho MH, Linehan JC, Cheng YH, Dupuis M, Raugei S, Shaw WJ, Biochim. Biophys. Acta, Bioenerg 2014, 1837, 131–138. [DOI] [PubMed] [Google Scholar]
- [8].a) Morra S, Giraudo A, Nardo G. Di, King PW, Gilardi G, Valetti F, PLOS One 2012, 7; [DOI] [PMC free article] [PubMed] [Google Scholar]; b) Lautier T, Ezanno P, Baffert C, Fourmond V, Cournac L, Fontecilla-Camps JC, Soucaille P, Bertrand P, Meynial-Salles I, Leger C, Faraday Disc 2011, 148, 385–407; [DOI] [PubMed] [Google Scholar]; c) Knörzer P, Silakov A, Foster CE, Armstrong FA, Lubitz W, Happe T, J. Biol. Chem 2012, 287, 1489–1499. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Pelmenschikov V, Birrell JA, Pham CC, Mishra N, Wang HX, Sommer C, Reijerse E, Richers CP, Tamasaku K, Yoda Y, Rauchfuss TB, Lubitz W, Cramer SP, J. Am. Chem. Soc 2017, 139, 16894–16902. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].a) Chumakov A, Sturhahn W, Hyperfine Interact 1999, 123, 781–808; [Google Scholar]; b) Sturhahn W, Toellner TS, Alp EE, Zhang X, Ando M, Yoda Y, Kikuta S, Seto M, Kimball CW, Dabrowski B, Phys. Rev. Lett 1995, 74, 3832–3835. [DOI] [PubMed] [Google Scholar]
- [11].Mitra D, Pelmenschikov V, Guo Y, Case DA, Wang H, Dong W, Tan M-L, Ichiye T, Francis J.Jenney E, Adams MWW, Yoda Y, Zhao J, Cramer SP, Biochemistry 2011, 50, 5220–5235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].a) Adamska-Venkatesh A, Simmons TR, Siebel JF, Artero V, Fontecave M, Reijerse E, Lubitz W, Phys. Chem. Chem. Phys 2015, 17, 5421–5430; [DOI] [PubMed] [Google Scholar]; b) Esselborn J, Lambertz C, Adamska-Venkatesh A, Simmons T, Berggren G, Nothl J, Siebel J, Hemschemeier A, Artero V, Reijerse E, Fontecave M, Lubitz W, Happe T, Nature Chem. Biol 2013, 9, 607–609; [DOI] [PMC free article] [PubMed] [Google Scholar]; c) Barton BE, Olsen MT, Rauchfuss TB, Curr. Opin. Biotechnol 2010, 21, 292–297. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].a) Gilbert-Wilson R, Siebel JF, Pham CC, Adamska-Venkatesh A, Reijerse E, Wang H, Cramer SP, Lubitz W, Rauchfuss TB, J. Am. Chem. Soc 2015, 137, 8998–9005; [DOI] [PMC free article] [PubMed] [Google Scholar]; b) Lunsford AM, Beto CC, Ding SD, Erdem OF, Wang N, Bhuvanesh N, Hall MB, Darensbourg MY, Chem. Sci 2016, 7, 3710–3719. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Carlson MR, Gray DL, Richers CP, Wang W, Zhao PH, Rauchfuss TB, Pelmenschikov V, Pham CC, Gee LB, Wang H, Cramer SP, Inorg. Chem 2018, 57, 1988–2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].Liu TB, Wang XP, Hoffmann C, DuBois DL, Bullock RM, Angew. Chem. Int. Ed 2014, 53, 5300–5304. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.