AmericanChemicalSociety.com

The levels of organic sulfur in sediments, which are controlled by the availability of reduced sulfur compounds, have a major impact on the kinetics of petroleum formation and the quality of the harvested product. Like many geochemical processes, the sulfur cycle is controlled by enzymes.  The focus of our work under this grant is to determine the detailed mechanism of sulfur/polysulfide reduction at the molecular level.  Our efforts are divided into 2 areas of emphasis: 1) studies of the polysulfide reductase (PsrABC) complex of Shewanella, and 2) detailed characterization of a FAD and coenzyme A-dependent cytoplasmic or periplasmic enzyme recently discovered in this lab (Npsr) that catalyzes NADH-dependent sulfur reduction; this enzyme is present in two species of Shewanella and homologues are present in other microbes isolated from petroleum reservoirs (while the details of this work was not contained in the grant proposal, because the enzyme catalyzes sulfur reduction via an uncharacterized mechanism we have chosen to pursue this avenue under this funding).

In previously reported work, we have overexpressed and purified the membrane anchor subunit (PsrC) and the PsrB subunit of the PsrABC complex. We determined that the PsrC subunit contained the expected bound quinone and the psrB subunit showed a UV/Vis spectra consistent with the presence of the Fe-S centers observed previously.  Exposure of PsrB to air, however, was shown to result in loss of iron, so we are currently purifying and working with this protein anaerobically.  In order to make characterization of the system easier an additional goal was to reconstitute the complex in a reusable film via electrostatic self-assembly, and we have been able to create a robust layer of PsrB and C in a polyanionic (polyethylimine) film, as shown via quartz crystal microbalance.

We are currently optimizing overexpression of the molybdopterin-containing PsrA subunit.  PsrA is overexpressed in copious quantities; however, the protein precipitates during overexpression.  Attempts to reconstitute this protein with a range of denaturants and detergents and detergent/denaturant concentrations have proven futile.  We are currently developing a homologous overexpression system, a technique which has worked in the past for molybdopterin-dependent enzymes.  Because PsrA is responsible for the S-S bond cleavage whose mechanism we are most interested in, the delay in obtaining usable protein has been vexing.

Within the past year another group published the 3-dimensional structure of this protein (from very different source organism), and it is obvious from their results, which were obtained with protein purified from large quantities of the source organism, overexpression of the PsrABC complex had met with difficulty for this group too.  Because of these difficulties, much of our current effort has been focused on the characterization of S-S bond cleavage using a flavin, NADH and coenzyme A-dependent mechanism via the enzyme Npsr, isolated in our laboratory after the submission of this proposal.  We have overexpressed and purified the enzyme (recently, using a new procedure, yielding >100 mg/l of culture) and created mutants of the enzyme which eliminate the cysteine residues which we believe are essential to the mechanism (Cys43Ser, Cys531Ser, and a Cys43Ser/Cys531Ser double mutant).  We have also crystallized the protein, obtaining crystals that diffract to 2.5 Å, and are in the process of solving the structure.  We fully expect to publish this structure within the next year.

In probably the most significant result of the past year we discovered what we believe to be the actual substrate for this sulfur reducing enzyme.  The sulfur-reducing activity of a homologue of this enzyme had been discovered by the Adams group (University of Georgia), and it was this group that first suggested the role for these enzymes in sulfur reduction.  They were not able to determine, however, the actual substrate for the enzyme (it is hard to imagine that the substrate would be the completely insoluble S8).  We have been able to obtain high velocities of NADH-dependent sulfide production using Coenzyme A persulfide (Co-A-S-S-H), which is freely soluble and binds tightly to the enzyme (based on its low Km).  When combined with the mechanistic details below, we believe that these results are significant enough for publication, and we intend to submit a second manuscript within the next year discussing the substrate specificity and mechanism of Npsr.

During the past year we have characterized the reductive half reaction of this enzyme with static titrations and stopped-flow UV/Vis spectroscopy.  The enzyme, which has a dimeric quaternary structure, shows subunit asymmetry. Each of the subunits accepts a first hydride equivalent from NADH, dithionite or titanium (III).  A second hydride equivalent reduces the FAD of only one of the subunits when NADH or dithionite are used as reductants, indicating that upon reduction of one of the FADs in the dimer that the potential of the other FAD is shifted well below -320 mV.  Titration with titanium (III), however, results in full reduction of both FADs of the dimer.

In order to determine the identity of the non-flavin redox center reduced by the first equivalent, enzyme was reduced with TCEP ((tris-(2-carboxyethyl)phosphine), a compound that reacts only with S-S bonds. When Npsr was titrated with one equivalent of TCEP, a spectral change in the FAD absorbance was observed that is identical to that seen on reduction with NADH (with a blue shift of the 460 nm FAD peak), indicating that a S-S bond, present as a Cys-Cys disulfide, Cys-CoA heterodisulfide, or a Cys-persulfide is being broken, eliminating the possibility that a cysteine sulfenic acid (Cys-SOH, a common intermediate in this class of enzyme) is the oxidized form of the enzyme. 

Reductive and TCEP titrations and steady-state kinetic analysis of the cysteine mutants of Npsr have shown the essential role for both Cys43 and Cys531 in the mechanism.  As mentioned above, with the mechanistic insights we have gained into the mechanism of S-S bond cleavage via Npsr and the structure of Npsr we plan to submit two manuscripts on this enzyme within the next year.