The<i>Escherichia coli</i>metal-binding chaperone SlyD interacts with the large subunit of [NiFe]-hydrogenase 3

HycE physically interacts with SlyD by cross linking study (view interaction 1, 2) Hydrogenase enzymes, found in many bacteria and archaea, catalyze the heterolytic cleavage of hydrogen gas into protons and electrons [1]. Escherichia coli express at least three [NiFe]-hydrogenase isoforms that all require auxiliary proteins for the assembly and insertion of metal cofactors into the active sites of the large subunits [2, 3]. The biosynthesis of the Ni–Fe(CO)(CN)2 metallocenter is believed to occur in two stages with iron insertion preceding nickel delivery [4]. The hydrogenase pleiotropic DEF (HypDEF) proteins, responsible for preparing and delivering the iron center, and the nickel-binding protein HypB, are common to the maturation of all hydrogenases 1–3 [3]. Recently, the E. coli sensitive to lysis D (SlyD) protein was also implicated in nickel insertion because deletion of the slyD gene resulted in decreased hydrogenase activity that was rescued by supplementing the growth media with nickel [5]. SlyD interacts with HypB [5, 6] and the two proteins, along with HypA/HybF, participate in nickel delivery [2, 3]. E. coli SlyD is a member of the FK506-binding protein family of peptidyl-prolyl isomerases (PPIases) and is composed of two well-defined domains: the PPIase domain catalyzes proline isomerization and the molecular chaperone domain named IF (insert in the flap) prevents protein aggregation [7, 8]. In addition, E. coli SlyD contains a non-conserved C-terminal metal-binding domain that can bind up to 7 nickel ions, contributes to nickel uptake, and is required for optimal hydrogenase production [5, 6, 9]. Furthermore, SlyD can stimulate nickel release from HypB [6], suggesting a role during hydrogenase nickel insertion beyond that of nickel source. Deletion of a loop (residues 107–111) in the IF domain abrogated stimulation of metal release from HypB in vitro and abated hydrogenase activity in vivo, presumably due to elimination of the interaction with HypB [6]. Defining the protein–protein interactions that occur during hydrogenase biosynthesis would facilitate a deeper understanding of this process, but such a goal is challenging to meet because of the labile and transient nature of these complexes. In an effort to identify proteins associated with the large subunit of hydrogenase 3, HycE, a tagged version was created and cell-permeable crosslinkers were applied to capture in vivo protein complexes. This strategy enabled identification of HycE complexes that include the nickel insertion proteins HypB, HypA, and SlyD. The SlyD–HycE interaction was analyzed, and a model for SlyD involvement in hydrogenase maturation in the context of both its HypB-mediated role and its interaction with HycE is discussed. The E. coli strains used are listed in Table S1. All strains were transformed with pBAD24-Strep-tag II HycE. The strains ΔhyaBΔhybCΔhycE and DHP-B (ΔhypB) were also separately transformed with pBAD24-Strep-tag II HypB. Cells were grown anaerobically in sealed 1 l bottles of buffered TGYEP medium [5] supplemented with 1 μM sodium molybdate, 1 μM sodium selenite, 30 mM sodium formate, 0.8% glycerol, 100 μM arabinose, and 100 μg/ml ampicillin at 37 °C for 16 h after inoculation with 1% (v/v) overnight culture. The cells were harvested by centrifugation, washed with 50 mM potassium phosphate, pH 7.6 (buffer A), and resuspended in buffer A containing 0.2 mM phenylmethylsulfonyl fluoride and trace amounts of DNase. Crosslinker, 5 mM 1,5-difluoro-2,4-dinitrobenzene (DFDNB) (Pierce), was added and the cells were incubated at room temperature for 30 min, followed by quenching with 100 mM Tris–HCl, pH 7.5. The cells were sonicated on ice or immediately stored at −80 °C for later use. Following sonication and centrifugation for 30 min at 14 000 rpm at 4 °C, the supernatant was applied to a Strep-Tactin Superflow column (1 ml bed volume, IBA). The column was washed with 25 ml of buffer A, followed by 1 ml each of 100 mM, 500 mM, and 1 M NaCl in buffer A, and 2 ml buffer A. The proteins were eluted with 3 ml of buffer A containing 2.5 mM desthiobiotin. Proteins were resolved on 10% or 12.5% SDS–polyacrylamide gels and transferred to polyvinylidene difluoride membranes (Millipore). The blots were probed with the appropriate polyclonal antibody at a 1:1000 dilution. The 2° goat anti-rabbit antibody (BioRad) was used at a dilution of 1:30 000. Enhanced chemiluminescence (Pierce) was used for detection. Additional materials and methods including plasmid construction, antibody sources, hydrogenase activity assays, and nano-LC–MS/MS are described in the Supplementary data. In order to isolate the [NiFe]-hydrogenase 3 large subunit, HycE, an N-terminal Strep-tag II variant called hycEStr was cloned into the arabinose-inducible pBAD24 vector. This system was adopted to allow efficient one-step purification [10]. Additionally, expression and activity of HycEStr was verified (see Fig. S1) as well as expression of HypA, HypB, and SlyD (data not shown). The construct was incorporated into mutant E. coli strains (Table S1) with the intention of isolating HycE and any interacting proteins with particular attention directed at the nickel insertion step. To facilitate identification of weakly-interacting proteins the membrane-permeable homobifunctional crosslinker DFDNB [11] was added prior to cell lysis, followed by pull-down assays of HycEStr and analysis by Western blotting. Although partner proteins could be pulled down with HycEStr in the absence of crosslinker, the amounts observed were enhanced by crosslinking and multiple higher molecular weight bands were only detected in the blots upon use of DFDNB (Fig. S2). Therefore subsequent experiments included application of the crosslinker. Isolation of HycEStr from HD705 (ΔhycE) revealed the presence of HypA, HypB, and SlyD (Fig. 1 A and Table 1 ) indicating either that each protein interacts with HycEStr or the formation of one or more multi-protein complexes. The higher molecular weight bands in the HycE Western blots do not all correspond to those observed in the HypA, HypB, and SlyD blots, possibly due to incomplete crosslinking such that only some components of each complex are trapped prior to electrophoresis, or to sensitivity differences between the polyclonal antibodies. In addition, the multi-step process of hydrogenase large subunit maturation involves transient interactions with a variety of proteins and, as a result, it is likely that the observed HycE bands represent a mixture of states of the enzyme with various protein complexes. The focus of this study was therefore to identify the presence or absence of these nickel insertion proteins in a complex with HycE, confirmed by nano-LC–MS/MS, rather than the specific components of each band. Control experiments with deletion strains for each of the nickel insertion proteins resulted in no detectable signals with the corresponding antibody (data not shown). In addition, crosslinking and pull-down assays from wild-type extracts without expression of HycEStr did not produce detectable Western blot signals with any of the antibodies (data not shown), eliminating the possibility of non-specific protein interactions with the Strep-tactin column and supporting the formation of multi-protein complexes with HycE. Nano-LC–MS/MS was used to positively identify HycEStr in the pull-down fractions, as well as SlyD, HypB, and HypA (Table S2). The complex with HypA is currently under investigation and will not be discussed further. Finally, although the focus of these experiments was the nickel insertion event, the presence of accessory proteins from an upstream step cannot be ruled out, and other Hyp proteins, such as HypC, were occasionally detected by nano-LC–MS/MS. SlyD was first identified as a maturation factor in hydrogenase biosynthesis when it was pulled out of cell lysates in a complex with HypB [5]. It was unknown whether this HypB–SlyD interaction occurs in the cytoplasm prior to nickel insertion or, alternatively, with the enzyme precursor acting as a scaffold for the accessory proteins. To investigate the timing of this interaction with respect to hydrogenase maturation, pull-down assays were performed with HypBStr in the absence of the hydrogenase large subunits (ΔhyaBΔhybCΔhycE). The HypB–SlyD interaction was maintained (Fig. S3), consistent with complex formation prior to association with hydrogenase. Nano-LC–MS/MS confirmed the presence of SlyD in HypBStr pull-down experiments (Table S2). It is possible that SlyD forms a complex with HycEStr through interaction with HypB and/or HypA. To test this requirement, pull-down assays were performed in DHP-B (ΔhypB) and DPABF (hypA[ATG→TAA]ΔhybF) strains. SlyD was detected in both cases, and a similar pattern of bands was observed in the Westerns (Fig. 1B, E and Table 1), indicating that neither HypB nor HypA is necessary. Furthermore, pull-down experiments in the HYD723 (ΔnikA) strain revealed an interaction between SlyD and HycEStr in the absence of cytoplasmic nickel (Fig. 1C and Table 1). Contact between SlyD and HycE during events upstream of nickel insertion was also demonstrated by performing experiments in DHP-C (ΔhypC) and ΔhypD strains (Fig. 1D and Table 1) suggesting that SlyD participates in hydrogenase maturation early on in the pathway. To roughly map out the region of SlyD that interacts with HycE, a variety of SlyD mutants were used. Reconstitution of the HycE–SlyD interaction was possible by adding purified SlyD to ΔslyD crude cell lysates containing HycEStr (Fig. 2 and Table 1) or to affinity-purified HycEStr from ΔslyD cells (in the presence or absence of nickel), followed by crosslinking and Strep-tactin chromatography (Fig. S4 and Table 1). This interaction was maintained with SlyD(1-146), which lacks the metal-binding domain, and with the PPIase-inactive SlyD mutant, I42S and F132Y, which alters the substrate-binding pocket [12, 13]. In contrast, the IF domain of SlyD is required because pure SlyD mutant with a deletion of residues 107–111 (SlyDΔflap) did not form a detectable complex with HycEStr in reconstitution experiments (Fig. 2 and Table 1). The importance of E. coli SlyD in hydrogenase maturation has thus far been attributed to its nickel-binding ability and the activation of nickel release from HypB [5, 6]. In this study, we introduce an additional role for SlyD in the hydrogenase maturation pathway as a possible chaperone for HycE. The combination of Strep-tag II technology with a membrane permeable crosslinker revealed HycE complexes containing SlyD, HypB, and/or HypA, providing the first evidence, to our knowledge, that all of these nickel insertion proteins interact with the hydrogenase precursor protein. Furthermore the interaction between SlyD and HycE in the absence of HypB or HypA, and the observations that this complex is also independent of HypC, HypD, and nickel, indicate that the complex extends to steps in the enzyme maturation pathway prior to both nickel and iron insertion, suggesting that SlyD has an additional role apart from nickel delivery. SlyD is also a chaperone for twin-arginine translocation (Tat) signal sequences that enable translocation of proteins through the bacterial cytoplasmic membrane [14]. Typically, Tat signals can be found on the small subunits of hydrogen-oxidizing hydrogenases [15], and these associate with the corresponding fully-matured large subunits prior to translocation across the cytoplasmic membrane by Tat transport [15]. It has been suggested that SlyD may have a role in bringing the large and small subunits together [16], or in linking the timing of nickel insertion to export of mature hydrogenase via interaction with the Tat signal commonly found on the small subunit [17]. However, a Tat sequence was not found on the small subunit of E. coli hydrogenase 3. Alternatively, it is quite possible that SlyD is responsible for keeping the small subunit from associating with HycE prior to full maturation or acts to chaperone the large subunit to the membrane in some manner independent of the Tat pathway. The IF domain of SlyD, which binds to unfolded hydrophobic protein sequences and is responsible for the chaperone activity [7, 8, 18], is required for reconstituting a complex with HycE. While this result establishes the IF domain as a binding site on SlyD for HycE, a role as a chaperone is supported by the prominent detection of GroEL and DnaK with tagged HycE in ΔslyD cell extracts by nano-LC–MS/MS (Table S2). Participation of a metal-binding chaperonin in hydrogenase maturation was demonstrated for Helicobacter pylori HspA [19], and the presence of chaperones to prevent aggregation during maturation of the enzyme, or to actively keep the precursor enzyme in a partially unfolded state to allow metal insertion, would not be surprising. Previous studies demonstrated that the IF domain of SlyD is important for complex formation with HypB [6]. This raises questions as to how, or if, SlyD is capable of interacting with both HypB and HycE during biosynthesis of the metallocenter. The results presented here suggest that the HypB–SlyD interaction is independent of the hydrogenase large subunits, so it is possible that two or more molecules of SlyD are involved in hydrogenase maturation, one as chaperone for the precursor enzyme and one dedicated to nickel delivery in association with HypB. It is perhaps not a coincidence that E. coli utilizes SlyD, with its metal-storage capabilities, to chaperone hydrogenase. Although the metal-binding domain of SlyD is not required for complex formation with the large subunit of hydrogenase 3 or HypB, it would concurrently provide a local source of nickel for the maturation process before proteolytic cleavage, conformational rearrangement, and final association with the small subunit. The role of SlyD in hydrogenase maturation has proven to be more complex than initially believed. With this newly found interaction between SlyD and HycE, further investigation is required to delineate all of the details of the function of SlyD in this complex pathway. We thank Prof. A. Böck (MC4100 strains, anti-HypB, and anti-HycE antibodies), Prof. A. Emili (ΔslyD and ΔhypD W3110 strains), and Prof. T. K. Wood (ΔhyaBΔhybCΔhycE BW25113 strain), Prof. A. Wheeler and H. Yang for use of the LC–MS/MS, and Zamble lab members for helpful discussions. This work was funded by CIHR and the CRC program. Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.febslet.2010.12.024. Supplementary data. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.