Next Article in Journal
Evaluation of Cobalt, Nickel, and Palladium Complexes as Catalysts for the Hydrogenation and Improvement of Oxidative Stability of Biodiesel
Next Article in Special Issue
Laccase-Catalyzed Polymerized Natural Bioactives for Enhanced Mushroom Tyrosinase Inhibition
Previous Article in Journal
A Spy Chemistry-Based Method for Purification of Proteins with Authentic N-Termini
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Catalysts
Volume 14
Issue 9
10.3390/catal14090652
catalysts-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Diverse YqeK Diadenosine Tetraphosphate Hydrolases Control Biofilm Formation in an Iron-Dependent Manner

by
Chie Ueda
1,†,
Natalie Chin
1,†,
Qianyi Yang
1,
Luying Pan
1,
Rheann Ponniah
2 and
Maria-Eirini Pandelia
1,*
1
Department of Biochemistry, Brandeis University, Waltham, MA 02453, USA
2
Acton-Boxborough High School, Acton, MA 01720, USA
*
Author to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Catalysts 2024, 14(9), 652; https://doi.org/10.3390/catal14090652
Submission received: 24 July 2024 / Revised: 12 September 2024 / Accepted: 18 September 2024 / Published: 23 September 2024
(This article belongs to the Special Issue The Design of Protein-Based Catalysts)
Browse Figures
Versions Notes

Abstract

:
YqeK is a bacterial HD-domain metalloprotein that hydrolyzes the putative second messenger diadenosine tetraphosphate (Ap4A). Elevated Ap4A levels are primarily observed upon exposure of bacteria to factors such as heat or oxidative stress and cause pleiotropic effects, including antibiotic sensitivity and disrupted biofilm formation. Ap4A thus plays a central role in bacterial physiology and metabolism, and its hydrolysis by YqeK is intimately linked to the ability of these microbes to cope with stress. Although YqeK is reported to hydrolyze Ap4A under aerobic conditions, all four existing crystal structures reveal an active site that consists of a diiron center, portraying a cryptic chemical nature for the active metallocofactor. This study examines two YqeK proteins from two ecologically diverse parent organisms: the obligate anaerobe Clostridium acetobutylicum and the facultative aerobe Bacillus halodurans. Both enzymes utilize Fe-based cofactors for catalysis, while under ambient or oxidative conditions, Bh YqeK hydrolyzes Ap4A more efficiently compared to Ca YqeK. This redox-dependent activity difference stems from the following two molecular mechanisms: the incorporation of mixed-metal, Fe-based bimetallic cofactors, in which the second metal is redox inert (i.e., Fe–Zn) and the upshift of the Fe–Fe cofactor reduction potentials. In addition, three strictly conserved, positively charged residues vicinal to the active site are critical for tuning Ap4A hydrolysis. In conclusion, YqeK is an Fe-dependent phosphohydrolase that appears to have evolved to permit Ap4A hydrolysis under different environmental niches (aerobic vs. anaerobic) by expanding its cofactor configuration and O2 tolerance.

Graphical Abstract

1. Introduction

YqeK belongs to the HD-domain superfamily, a large and catalytically diverse group of proteins that contain a conserved histidine-aspartate (HD) motif to coordinate transition metal ions [1,2]. YqeK symmetrically cleaves diadenosine tetraphosphate (Ap4A), a putative alarmone ubiquitously found in prokaryotes and eukaryotes (Scheme 1) [3]. Elevated Ap4A levels are observed as a stress response to heat, antibiotics, hypoxia, or oxidative damage [4,5,6,7] and are further linked to pleiotropic effects, including antibiotic sensitivity and disrupted biofilm formation [4,8]. High Ap4A levels have adverse effects on Salmonella enterica pathogenesis and Myxococcus xanthus sporulation [9,10], while they influence RNA degradation and transcription initiation via the integration of Ap4A into the 5′ cap of mRNA [11,12]. Although Ap4A exhibits characteristics reminiscent of an alarmone, to date, no signaling cascade in response to changes in Ap4A levels has been recognized [13]. Moreover, its spurious binding to ATP-binding proteins, ability to disrupt Zn homeostasis, and formation through the non-canonical activities of aminoacyl tRNA synthetases have invoked the role of Ap4A as a damage metabolite [4,13].
Under non-stress conditions, low Ap4A levels are maintained by balancing its synthesis through adenylate-forming enzymes and its degradation through hydrolases [14,15].
Hydrolases are thus key enzymes for the ability of microbes to form protective biofilms and cope with stress [5,6,16,17,18]. Bacteria use two types of Ap4A hydrolases that differ in the cleavage position and final products. Enzymes from the nucleoside diphosphates linked to x (NUDIX) or the histidine triad (HIT) families cleave Ap4A asymmetrically to yield ATP and AMP [19,20,21], while metal-dependent hydrolases belonging to the ApaH and YqeK families cleave Ap4A symmetrically to yield 2 ADP molecules [4,13]. ApaH is predominantly found in Gram-negative bacteria, and its activity is dependent on divalent cations like Mn2+ or Co2+ [22]. YqeK is prevalent in Gram-positive bacteria, and while its structure depicts a diiron active site, its role in enzymatic activity has not yet been delineated [3,23].
Staphylococcus aureus YqeK hydrolyzes Ap4A with a catalytic efficiency of 34 × 106 M−1s−1 and Streptococcus pneumoniae SP1746 with 3.2 × 106 M−1s−1, which are both on par with that of Salmonella enterica ApaH (1 × 106 M−1s−1) [3,9,23]. YqeK also symmetrically cleaves other dinucleotides, such as Gp4G, Ap4G, and Ap4U, with similar catalytic efficiencies. Compared to ApaH, YqeK is more selective toward symmetrical hydrolysis, with Ap5A catalysis being one order of magnitude lower than that of Ap4A [3,9,24]. YqeK activity is metal-dependent, and all four available crystal structures (Bacillus halodurans (pdb: 2O08), Clostridium acetobutylicum (pdb: 3CCG), Streptococcus agalactiae (pdb: 2OGI), and Streptococcus pneumoniae (pdb: 8JJK)) harbor a diiron metallocofactor [23]. Considering that Fe introduces an unwanted redox sensitivity, the employment of a diiron site is in apparent contrast with the reported ability of YqeK to hydrolyze Ap4A under ambient conditions, portraying a cryptic chemical nature for the active cofactor form. While diiron cofactors are not uncommon in HD-domain phosphodiesterases, HD-domain diphosphatases usually employ MnII centers, making the Fe-based cofactor of YqeK even more unusual [2,23,25,26,27,28,29,30].
In the present study, we selected the following two YqeK proteins from organisms from different environmental niches: the facultative aerobe Bh and the obligate anaerobe Ca. In both cases, we demonstrate that Fe is preferentially incorporated and a key element for catalysis. Both YqeKs assemble active bimetallic Fe-based cofactors, which can be of different chemical nature, endowing the respective proteins with different catalytic fidelities. Here, we show that YqeK hydrolases utilize the tunable redox potential of Fe, as well as its ability to form homo- and mixed-metal bimetallic cofactors to regulate activity. Collectively, our findings show that Ap4A hydrolysis by Bh YqeK exhibits higher O2 tolerance than that of Ca YqeK and exemplifies the manners by which environmental and evolutionary pressure can tune cofactor selectivity and activity.

2. Results

2.1. Occurrence and Evolution of YqeK Hydrolases

The SSN contains 8028 YqeK sequences that cluster on the basis of their taxonomic order, according to which the nodes have been colored (Figure 1A). YqeK occurs only in prokaryotic organisms, with the most prevalent phylum being the Firmicutes (Figure S1) [3]. The three major clusters are Bacillales (blue), Lactobacillales (red), and Clostridiales (green). These three orders represent 73% of the total sequences, while 83% of the sequences belong to Gram-positive prokaryotes. However, there are notable Gram-negative classes, including Cyanobacteria, Spirochetes, Selenomonadales, Leptospirales, Veillonellales, Tenericutes, and Fusobacteriales. The biased representation of YqeKs in Gram-positive bacteria aligns with the predominant occurrence of the functionally homologous ApaH in Gram-negative prokaryotes [3].
A total of 749 unique sequences from diverse organisms were selected from the major clusters of the SSN to generate the maximum likelihood phylogenetic tree (Figure 1B), in which the HD-domain protein (pdb: 2PQ7) was selected as the outgroup on the basis of its sequence similarity to YqeK [31]. The tree consists of three major events and first branches off to the Deinococcales, aerobic bacteria that are highly resistant to environmental hazards and exhibit attributes of both Gram-negative and Gram-positive bacteria [32]. The second event consists of a split to YqeKs from Cyanobacteria (aerobic photosynthetic bacteria) [33] and a branch collecting the rest of the organisms. The latter segregates into two large clades, one (purple node) that solely contains YqeKs from anaerobic bacteria and a second one (orange node) that collects YqeKs from both aerobic and anaerobic bacteria. The anaerobic clade first splits (blue node) into Eubacteriales of the genus Clostridium and Acetobacterium, mostly found in anaerobic sediments, as well as the pathogenic Fusobacterium responsible for human gut infections. The second split (pink node) of the anaerobic clade collects Selenomonadales (Selenomonas, Sporomusa) that colonize the gastrointestinal tract, the gut microbiota Collinsella (Coriobacteriales), the lactate-fermenting Veillonellales colonizing mammalian intestines and oral mucosa, and the Thermoanaerobacteriales (genus Moorella, Thermoanaerobacter, and Thermovenabulum) that are acetogenic bacteria isolated from terrestrial hydrothermal sources. The clade that contains YqeKs from both aerobic and anaerobic bacteria is depicted with an orange node. The first of its two major branches (green node) contains the anaerobic Oscillospira and Faecalibacterium that belong to the Clostridia class and are important for human gut health, the often-pathogenic Leptospirales aerobes, the Paenibacillus (facultative anaerobes or strictly aerobes) that support agroecology and plant health, and Enterococcus (Lactobacillales) that are facultative anaerobes associated with various human infections. The second major branch (red node) collects YqeKs that segregate into a clade of obligate or facultative anaerobes that colonize the gut, such as Eubacteriales (Lachnobacterium, Clostridium, Ruminococcus, Butyrivibrio, etc.), Acholeplasmatales, and Candidatus Gastranaerophilales, and a second one consisting of the aerobic Bacillales and Lactobacillales (such as Staphylococcus, Streptococcus, and Bacillus) that are opportunistic pathogens. Overall, YqeKs occur in diverse microbes that are central to human health and the environment. While our analyses posit that YqeK likely originated from aerobic Gram-negative-like prokaryotes, the tree suggests that YqeKs emerged in species that are facultative and strict anaerobes or aerobes predominantly Gram-positive in nature. In our studies, we selected Ca from the strictly anaerobic clade and Bh from the strictly aerobic branch to address any common or divergent traits in the chemical makeup of the active site metallocofactors between these phylogenetically diverse enzymes.

2.2. LB Bh and Ca YqeK Harbor Redox and Hydrolytically Active Homo- and Mixed-Metal Fe-Based Cofactors

Bh and Ca YqeK share 42% sequence identity and 58% sequence similarity (Figure S2) and are homodimers of approximately 45 kDa (Figure 2A,B). Irrespective of the expression media, both proteins copurify with a significant amount of Fe (Table 1). When isolated from cells grown in non-metal-specific LB media, Bh and Ca YqeK contain an appreciable amount of Zn, which for Bh YqeK, is stoichiometric and for Ca YqeK, is sub-stoichiometric with respect to Fe (Table 1). Bh and Ca YqeK isolated from cells grown in minimal media with Fe supplementation (M9-Fe) are enriched in iron (~1.3–1.5 molar eq of Fe per monomer), with the rest of the transition metals present in trace levels (Table 1). These results agree with the available crystal structures of both proteins that depict a diiron active site (Figure 2C). However, the relevance and active cofactor redox states are enigmatic when considering that under aerobic conditions, YqeK exhibits high catalytic rates of ~100 s−1 that are on a par with those of the functionally homologous ApaH [3,9,23].
The EPR spectra of the as-isolated LB Ca and Bh YqeK exhibit high-spin signals suggestive of the formation of mixed-metal bimetallic FeIII-MII complexes (Figure 2D,E) [34,35,36,37,38]. The spectrum of Bh YqeK exhibits a resonance at g ~ 9.33 and anisotropy at the g = 4.23 signal (often associated with adventitiously bound ferric species). The spectrum of Ca YqeK exhibits similar features, but with slightly different g-values (g ~ 7.97) and anisotropy. These signals arise from the ±3/2 and ±1/2 doublets of an S = 5/2 high-spin FeIII in a rhombic environment and are reminiscent of the signals detected in kidney bean PAPs that have been attributed to FeIII–ZnII heteronuclear cofactors (Figure S3) [34,35,36,38]. The assignment of these high-spin signals to FeIII–ZnII sites is consistent with the elemental analysis that shows both proteins to copurify with Fe and Zn (Table 1). The low-field region of the Bh YqeK EPR spectrum contains a signal with a gav < 2 that is reminiscent of antiferromagnetically (AF)-coupled FeIII–FeII (S = 1/2) centers (Figure 2F) that amount to ~7% of the total Fe in the sample [39]. The presence of mixed-valent (MV) FeIII–FeII centers demonstrates the assembly of diiron sites, which agrees with the published crystallographic data (Figure 2C). In contrast, the low-spin region of the Ca YqeK spectrum does not exhibit detectable FeIII–FeII signals, which may be skewed by the presence of a mononuclear MnII contaminating signal evidenced by the characteristic 6-line splitting at g ~ 2 (Figure 2G). Collectively, the EPR experiments show that YqeK isolated from cells grown in LB contains a mixture of Fe–Zn and Fe–Fe centers. Next, we examined any redox dependence of Ap4A hydrolysis by carrying out end-point assays under three different conditions, as follows: as-isolated, H2O2-oxidized, and dithionite-reduced. The YqeK from the aerobe Bh retains full activity under ambient conditions, which is in contrast to the YqeK from the anaerobic Ca that only exhibits ~33% of substrate conversion (Figure 2H,I). Oxidation with H2O2 adversely impacts hydrolysis resulting in a six-fold activity reduction in Bh YqeK, when compared to ambient conditions, and almost complete inactivation in the case of Ca YqeK. Dithionite reduction enhances Ap4A hydrolysis by Ca YqeK (70% conversion) but not by Bh YqeK (Figure 2H,I). Bh YqeK thus maintains higher activity than Ca YqeK under ambient and oxidative conditions, but not reducing conditions. The observed redox dependence of the YqeK activity underscores that iron and its redox state are essential catalytic elements, while it highlights that enzymes from aerobic organisms exhibit a more O2-tolerant activity compared to those expressed in anaerobic organisms.

2.3. The Diiron Cofactors of Bh and Ca YqeK

Bh and Ca YqeK were enriched in Fe by expressing them in minimal media with Fe supplementation to specifically study their diiron forms. The EPR spectra of the as-isolated M9-Fe Bh and Ca YqeK are drastically different in the high-spin region when compared to those of the LB proteins, as they only contain an isotropic signal at g = 4.23 corresponding to adventitiously bound mononuclear FeIII (S = 5/2) (Figure 3A). The lack of additional resonances and anisotropy in the FeIII signal is consistent with directed enrichment with Fe and elemental analyses (Table 1) that support (i) the assignment of the high-spin resonances in the LB samples to FeIII–ZnII centers and (ii) the absence of detectable Fe–Zn sites in the M9-Fe samples (Figure 2D,E). Upon reduction with sodium ascorbate, the FeIII–FeII signal (gav < 2) gains in intensity and corresponds to ~40–60% of the diiron sites (Figure 3B,C). The EPR spectra of the mixed-valent forms of M9-Fe Ca and Bh YqeK exhibit more than three components (i.e., the upper limit for rhombic symmetry), suggestive of the presence of multiple FeIII–FeII species with different electronic parameters, the relative intensity of which varies among different preparations (Figure S4). Because Bh and Ca YqeK coordinate small molecules to the diiron site (e.g., dGDP or Pi, Figure 2C), we speculated that this species heterogeneity may be reflective of populations with or without a small-molecule ligand. To examine this scenario, both YqeKs were chelated to remove the metallocofactor (and any bound small molecule) and then reconstituted with Fe. The EPR spectra of the reconstituted Bh and Ca YqeK consist of a single component with principal g-values, (1.96, 1.81, and 1.74) and (1.96, 1.81, and 1.61), respectively (Figure 3B,C). In both cases, reconstitution recovered the signal with the smaller g-anisotropy, termed here Component 1. Subtracting the contribution of Component 1 in the reconstituted enzyme from the spectrum of the as-purified enzyme yields Component 2, which is characterized by a higher g-anisotropy and downshifted principal g-values (1.96, 1.64, and 1.52). The more rhombic spectrum of Component 2 reflects a more asymmetric electronic distribution that could stem from the binding of a small molecule [30,39,40]. This hypothesis is supported by experiments in which a series of (d)/NTPs were added to the as-purified Bh YqeK. (d)/NTPs were chosen because they are poor substrates and would allow the accumulation of any small molecule-bound FeIII–FeII states [3,23]. In all cases, an observable decrease in the intensity of Component 1 (assigned to a ligand-free FeIII–FeII cofactor form) was observed, followed by a concomitant increase in the intensity of Component 2 (assigned to a small-molecule bound form of the FeIII–FeII cofactor) (Figure 3D). Therefore, Component 2 must represent a form of the cofactor that copurifies with a bound nucleotide.
The chemical nature and redox behavior of the diiron cofactor were probed using Mössbauer spectroscopy on the 57Fe enriched Bh YqeK (Figure 4). The 4.2-K/53-mT Mössbauer spectrum of the as-isolated Bh YqeK is dominated by three quadrupole doublets that have parameters characteristic of AF-coupled FeIII–FeIII centers, i.e., δ1 = 0.56 mm/s, ΔEQ1 = 1.99 mm/s (dotted line, 32% of the total Fe), δ2 = 0.49 mm/s, ΔEQ3 = 1.02 mm/s (black solid line, 36% of the total Fe), and δ3 = 0.48 mm/s and ΔEQ3 = 0.51 mm/s (gray solid line, 32% of the total Fe) [39,41] (Figure 4, top). These quadrupole doublets suggest the presence of two or three different FeIII–FeIII species that may differ in the protonation state of a bridging water ligand or coordination of the Fe sites by protein or inorganic ligands [28]. Such heterogeneity has been previously observed for other diiron HD-domain hydrolases and has been generally considered unrelated to enzymatic activity [2,28,29,30].
The ascorbate-reduced Bh YqeK exhibits broad and complex signals that originate from the one-electron reduced paramagnetic FeIII–FeII state (Figure 4, middle) [41]. The spectral components for the high-spin FeIII site could be fitted with parameters δ = 0.58 mm/s, ΔEQ = 1.03 mm/s, η = 1, and a/gnβn = (−37.1, −53.4, −60) T, and for the high-spin FeII site, δ = 1.23 mm/s, ΔEQ = 2.71 mm/s, η = −0.18, and a/gnβn = (17.2, 22.5, −6.8) T [41]. The amount of the FeIII–FeII state corresponds to ~60% of the total Fe, while the rest of the Fe is in the FeII–FeII state (vide infra). Treatment with sodium dithionite affords homogeneous enrichment of the FeII–FeII form (Figure 4, bottom). The spectrum is best fitted considering two quadrupole doublets with parameters (δ1 = 1.27 mm/s, ΔEQ1 = 2.67 mm/s, and δ2 = 1.08 mm/s, ΔEQ2 = 2.75 mm/s) that are characteristic of high-spin FeII sites with O/N coordination and demonstrate complete reduction to the FeII–FeII form [39,41,42].
Following the paradigm of diiron HD-domain proteins [2,28,39,40,43], the diiron cofactor in YqeK attains three different redox states (i.e., FeIII–FeIII, FeIII–FeII, FeII–FeII). Considering that YqeK is a hydrolase, the FeIII–FeIII state must be hydrolytically inactive due to the usual requirement of at least one divalent metal ion to activate the water for nucleophilic attack and subsequent bond cleavage [44]. Therefore, the semi-reduced FeIII–FeII and/or fully reduced FeII–FeII states are likely the catalytically active forms.

2.4. The Active Forms of the Diiron Bh and Ca YqeK

Under ambient conditions, the M9-Fe Ca YqeK exhibits residual activity (10% of the substrate hydrolyzed), but Ap4A hydrolysis is almost completely abolished after treatment with hydrogen peroxide (Figure 5A). Treatment with sodium dithionite leads to the consumption of almost all the substrate. On the other hand, M9-Fe Bh YqeK is more active when compared to Ca YqeK under both ambient and oxidative conditions, suggesting that it retains a higher amount of an active cofactor state. Treatment with dithionite leads to increased product conversion (65%), demonstrating that reduction activates the enzyme in a comparable manner to that of M9-Fe Ca YqeK (Figure 5B). EPR spectroscopy demonstrates that under ambient conditions M9-Fe Bh YqeK accumulates the semi-reduced FeIII–FeII form, but M9-Fe Ca YqeK hardly does so, providing a possible explanation for the higher observed activity of Bh YqeK (Figure S5). To note, no high-spin FeIII–ZnII signals could be detected in either of the M9-Fe YqeK forms, showing that activity in M9-Fe Bh YqeK must originate from reduced diiron forms rather than mixed-metal cofactors like in the case of the LB protein (Figure 2 and Figure 3).
To evaluate the catalytic ability of the semi-reduced (FeIII–FeII) and fully reduced (FeII–FeII) cofactor states, multiple turnover assays were carried out for both M9-Fe Ca and Bh YqeK under three different conditions, as follows: (i) ambient conditions, (ii) reduced with ascorbate that affords circa 40–60% of the FeIII–FeII state, and (iii) reduced with sodium dithionite that affords a homogeneously reduced FeII–FeII cofactor form. Ap4A hydrolysis is enhanced with either ascorbate or dithionite, demonstrating that enrichment of the reduced Fe cofactor states positively correlates with activity (Figure 5C,D). The dithionite-reduced M9-Fe Ca YqeK (FeII–FeII-enriched) hydrolyzes Ap4A with an apparent initial rate of 135 ± 3 s−1, while the M9-Fe Bh YqeK has an apparent initial rate of 41 ± 3 s−1. These rates are on par with those reported for the LB forms of Sa YqeK [3] and Sp SP1746 [23]. Reduction with ascorbate results in apparent rates for Ca YqeK that are fourfold lower (33.4 ± 2 s−1), whereas for Bh YqeK, Ap4A hydrolysis rates are slightly enhanced (59.4 ± 3 s−1). The apparently divergent activity profile of the two YqeKs with respect to the reducing agent employed alludes to differences between the reduction potentials of the two redox couples, rather than one of the two reduced forms being more active in the two different YqeKs.

2.5. Bh YqeK Accesses Its Reduced Cofactor Forms at Higher Reduction Potentials than Ca YqeK

We proceeded to estimate the reduction potentials of the diiron cofactor by monitoring the EPR signal of the MV form as a function of the electrochemical potential (Figure 6A,B). The titrations were carried out with the as-purified Bh and Ca YqeK for procuring the redox behavior of both spectral components (Components 1 and 2) as small-molecule binding may have an impact on redox properties. As a note, Component 1 has been assigned to a ligand-free FeIII–FeII cofactor form, and Component 2 has been assigned to a small molecule-bound form of the FeIII–FeII cofactor. Titration of a reducing or oxidizing agent resulted in Gaussian-like curves representing the accumulation (FeIII–FeIII/FeIII–FeII couple, EoxA) and decay (FeIII–FeII/FeII–FeII couple, EredB) of the MV diiron cofactor and was fitted to the Nernst equation corresponding to two sequential one-electron reduction processes (see Materials and Methods). The midpoint potentials for Eox,Comp1 and Ered,Comp1 in Bh YqeK are upshifted by 65 mV and 91 mV, respectively, when compared to those for Component 1 in Ca YqeK (Figure 6C,D). The midpoint potentials of Component 2 exhibit a smaller difference between the two proteins and are approximately 20 mV more positive in Bh YqeK when compared to Ca YqeK. Overall, the redox titrations demonstrate that the diiron cofactor in Bh YqeK can access the reduced cofactor states (i.e., FeIII–FeII and FeII–FeII) at more positive reduction potentials than Ca YqeK, in line with the (i) detectable accumulation of the MV form only in the aerobically isolated Bh YqeK (Figure 2 and Figure S4), and (ii) ability of the M9-Fe Bh YqeK to retain higher activity under ambient conditions or with the mild reductant sodium ascorbate when compared to Ca YqeK (Figure 3 and Figure 6).

2.6. Positively Charged Residues Tune Ap4A Hydrolysis

The active site of YqeKs contains one lysine and two arginines that are at an appropriate distance to form hydrogen-bonding contacts with the substrate phosphate oxygens (Figure 7). We explored the effect of alanine substitutions of the highly conserved R18, K54, and R133 in Bh YqeK on the (a) assembly of the cofactor, and (b) ability to hydrolyze Ap4A. All three variants assembled a diiron cofactor to similar extents as the WT protein on the basis of quantification of the MV form in EPR spectra and elemental analyses (Figure 7, Table S2). These results confirm that none of these substitutions compromise cofactor assembly, although there are observed differences in the relative ratio of Components 1 and 2 in the variant EPR spectra. These differences may suggest a possible role for these residues in altering active site affinity for nucleotides (substrate or product) and thus tuning chemistry, although the variability noted among different WT preparations precludes the present unambiguous conclusions (Figure S5). All three variants showed no conversion of Ap4A to ADP under standard assay conditions (Figure 7C). Increasing the protein concentration to 1 µM (67-fold increase) led to complete Ap4A hydrolysis by R18A and R133A but not K54A, demonstrating that this substitution has the most critical role in enzymatic activity. These results show that chemistry can still proceed, albeit slowly, presumably due to the loss of important hydrogen-bonding interactions that may alter substrate and/or product binding.

3. Discussion

YqeK is a bacterial HD-domain metalloprotein that symmetrically cleaves the putative alarmone Ap4A. Although YqeK is predominantly found in Gram-positive bacteria, our bioinformatic analyses demonstrate its significant occurrence in Gram-negative bacteria, as well. The YqeK phylogeny narrates a story that YqeK first appeared in Deinococcales, which are extremophilic bacteria that can withstand a variety of environmental stressors. These are obligate aerobes that exhibit both Gram-positive and Gram-negative characteristics. YqeK is then found in Cyanobacteria, which are Gram-negative, oxygenic photosynthetic organisms that can thrive in diverse and extreme habitats. The next event is the segregation of YqeKs into two large branches, one that consists solely of anaerobic organisms and one that consists of a mixture of aerobic and anaerobic prokaryotes, suggesting that these may have evolved a different cofactor chemical nature to allow for Ap4A hydrolysis under different ecological niches. To examine whether different YqeKs have adopted different strategies or conserved a unified mechanism for Ap4A hydrolysis, we focused our studies on the following two YqeKs from the two different branches of the phylogenetic tree: the YqeK from the obligate anaerobe Ca and the facultative aerobe Bh.
The crystal structures of Bh and Ca YqeK show a diiron center at the active site. However, YqeK has been reported to hydrolyze Ap4A under ambient conditions, which is typically unusual within the context of redox-sensitive Fe–Fe cofactors [3,23]. We thus investigated the chemical nature of the active metallocofactor and addressed (a) whether Fe is participating in catalysis and (b) the types of metallocofactors and redox states that are catalytically active.
When expressed in LB, Bh YqeK retains appreciable activity under aerobic conditions, which agrees well with previously published reports on the YqeK from the aerobic pathogens Sa and Sp [3,23]. Under the same conditions, the YqeK from the anaerobe Ca is almost twofold less active than Bh YqeK. The common cellular oxidant H2O2 is a potent inhibitor of both Bh and Ca YqeK and significantly reduces activity, although Ap4A hydrolysis by Bh YqeK is detectably more oxidatively tolerant. The present findings are best explained by considering an active mixed-metal FeIII–ZnII cofactor form in the LB proteins that is preferentially assembled in Bh YqeK. Perhaps the most well-described case of mixed-metal FeIII–MII cofactors are those found in purple acid phosphatases (PAP), in which M is invariably a divalent metal ion (e.g., Fe, Co, Mn, or Cu) [34,35,36,44,45,46]. In the case in which M is Fe, the cofactor is hydrolytically active in the reduced state (i.e., FeIII–FeII) and inactive in the oxidized state FeIII–FeIII). In PAPs, the FeIII is tyrosine-ligated and resistant to reduction, making the FeII–FeII state inaccessible. However, in calcineurin, a serine/threonine phosphatase, the role of the Fe redox state in catalysis is unclear, with dithionite reduction leading to reduced activity [44]. Low or absent activity from a mixed-metal FeII–MII form may explain the reduced Ap4A hydrolysis observed in the dithionite-reduced Bh YqeK that, based on elemental and EPR analyses, has a larger fraction of mixed-metal Fe sites than Ca YqeK. Ca YqeK exhibits a slightly different activity profile that is more consistent with the preferential assembly of diiron sites over mixed-metal Fe-based cofactors. Considering the different ecological niches of Ca and Bh YqeK, the incorporation of an Fe–Zn cofactor in Bh YqeK may be an evolutionary strategy of aerobic organisms to sustain YqeK activity under the more oxidative conditions the parent organism experiences.
When Ca and Bh YqeK are expressed in Fe-specific media, they are enriched in diiron cofactors. Bh YqeK retains activity under both ambient and oxidative conditions, whereas Ca YqeK is hardly active. This result positively correlates with the presence of semi-reduced FeIII–FeII cofactor forms in the as-isolated Bh YqeK. Reduction with dithionite (FeII–FeII form) or ascorbate (mixture of FeIII–FeII and FeII–FeII forms) affords similar apparent rates for Ap4A hydrolysis by Bh YqeK, demonstrating that both FeII–FeII and FeIII–FeII are comparably active. Redox titrations monitored by EPR show an upshift in the reduction potentials of the diiron cofactor of Bh YqeK when compared to that of Ca YqeK. The appreciable activity under ambient conditions of Bh YqeK is therefore a result of the accumulation of the semi-reduced FeIII–FeII state, while the lower activity of Ca YqeK stems from the persistence of inactive FeIII–FeIII forms. The presence of a small-molecule ligand affects the overall redox potential differences, suggesting that substrate binding may influence reduction potentials. The ability of YqeK from the Bh aerobe to maintain higher activity under ambient conditions when compared to the YqeK from the anaerobe Ca is a consequence of the higher O2 tolerance of the diiron cofactor (reflected in the higher reduction potentials), which represents a second possible strategy employed by aerobic organisms to retain Ap4A hydrolysis under more O2-rich environments.
Apart from the chemical makeup of the active site, we established that three conserved basic residues, i.e., R18, K54, and R133, are important for activity. The presence of positively charged residues in the active site is a shared feature among several other HD phosphohydrolases, including SAMHD1, EF1143, and OxsA [47,48,49]. It has been proposed that the overall positive charge may modulate the electrophilicity of the phosphate to favor hydrolysis [49]. In SAMHD1, the substitution of the cognate active-site arginine to R18 (R164) completely abolishes activity [47]. In Bh YqeK, R18 and R133 make hydrogen bonds with the β-phosphate of the dGDP molecule in the crystal structure, and their substitution leads to a loss of activity, which, however, can be rescued by increasing the enzyme concentration (i.e., 67-fold). The K54 residue appears to be more critical for catalysis than the arginines, even at higher protein concentrations, alluding to its ability to either act as a potential base in the reaction or tune substrate or product binding and release.
Adaptation and sensitivity to oxygen are highly important for cells because oxidative stress is linked to other forms of environmental triggers, such as antibiotics or heat. For example, in Sc and Ec, heat shock induces ROS generation [50,51,52], while aminoglycoside antibiotics trigger ROS accumulation [53,54,55]. In such cases, the employment of a redox-active metal to sense incoming oxidative stress is a common strategy that is often observed in transcriptional repressors, such as SoxR and PerR [56,57,58]. The incorporation of a redox-active cofactor in YqeK is apparently paradoxical, considering that the protein under oxidative stress would inactivate and be unable to efficiently relieve the exogenous stress caused by elevated Ap4A levels, arguing against a classical redox-sensing function. The presence of Fe in YqeK may be related to different cellular and molecular pathways in the parent organisms, as several virulence properties of microbes are iron-dependent, and pathogens have evolved sophisticated mechanisms for Fe acquisition [59,60,61]. Regardless of the exact molecular underpinnings, the inclusion of Fe increases the chemical fidelity of the YqeK active site and adds functionalities beyond catalysis, which appears to be highly tunable across different species, with either incorporation of mixed-metal bimetallic Fe-based sites or upshifts in the reduction potentials of diiron sites.

4. Materials and Methods

Materials. All chemicals, unless specified, were obtained from Fisher Scientific (Newington, NH, USA) and were of high-purity grade. Diadenosine tetraphosphate (Ap4A) was purchased from MilliporeSigma (Burlington, MA, USA).
Sequence Similarity Network (SSN). The SSN was generated using the web-based Enzyme Function Initiative-Enzyme Similarity Tool (EFI-EST) [62] using the YqeK sequence from Ca (WP_010964575.1) as an input. The SSN was calculated with an alignment score of 60 and further refined to only contain sequences between 170 and 200 amino acids long. The final SSN contains 3740 nodes and 8028 sequences.
Phylogenetic analyses. We selected 749 unique and nonredundant sequences from the major protein clusters of the YqeK SSN that represent the bulk of the organisms that encode for YqeK. The sequences were aligned with the MAFFT software version 7 (https://mafft.cbrc.jp/alignment/software/, accessed on 1 July 2024) [63,64], and the maximum likelihood phylogenetic rooted tree was subsequently computed with the IQ-tree software (http://www.iqtree.org, accessed on 1 July 2024) employing the LG+F+R4 model [65]. The uncharacterized HD-domain protein (pdb: 2PQ7) was used as an outgroup (8 sequences) [31].
Protein Expression. The genes encoding for the Bh and Ca YqeK were synthesized and inserted in a pET28a(+) vector via NdeI/XhoI (Genscript, Piscataway, NJ, USA) that allows for expression of the YqeK protein with an N-terminal His6-tag. The plasmids were transformed into T7 express (DE3) competent cells (New England Biolabs, Ipswitch, MA, USA). Cells were grown in standard Luria Bertani (LB) Lennox or minimal (M9) media when directing specific metal incorporation in the presence of 0.5% (w/v) glucose, 0.1 mM CaCl2, and 2 mM MgSO4·7H2O. Then, 50 μg/L of kanamycin was added, and the cultures were incubated at 37 °C with shaking (220 rpm) until OD600 reached a value between 0.6 and 0.8. Protein expression was induced by the addition of 0.5 mM of isopropyl β-d-1-thiogalactopyranoside (IPTG) and 250 μM of Fe(NH4)2(SO4) when directing Fe incorporation in M9 media. Cells were incubated at 18 °C for 18–20 h and then harvested using centrifugation at 7000 rpm for 15 min at 4 °C. Cell pellets were frozen in liquid N2 and stored at −80 °C. To obtain the apo-form (devoid of metals) of YqeK, the cells were grown in LB media, but upon induction with IPTG (0.5 mM), the chelator 1,10-Phenanthroline was added to a final concentration of 0.5 mM.
Holo Bh YqeK Purification. Bh YqeK cell pellets were resuspended in lysis buffer (100 mM HEPES, 300 mM NaCl, 10 mM imidazole, pH 7.5). Phenylmethylsulfonyl fluoride (PMSF) was added to a final concentration of 45 μg/mL. The suspension was lysed using a QSonica sonicator (Newton, CT, USA) (~30 min per 100 mL of suspension). The clarified lysate was loaded onto a Ni2+-NTA immobilized affinity chromatography column (~100 mL resin per 500 mL lysate) equilibrated with the lysis buffer. The column was first washed with lysis buffer and then with wash buffer (100 mM HEPES, 300 mM NaCl, 25 mM imidazole, pH 7.5). The bound protein was eluted with elution buffer (100 mM HEPES, 300 mM NaCl, 10% glycerol, 250 mM imidazole, pH 7.5). Fractions containing the protein of interest were pooled and concentrated at 3500× g using a 30 K Amicon Centrifugal Filter Unit (Millipore, St. Louis, MO, USA). The concentrated protein was loaded onto a size exclusion HiLoad 16/600 Superdex column (Cytiva, Melbourne, Australia), which was equilibrated with the storage buffer (100 mM HEPES, 300 mM NaCl, 10% glycerol, pH 7.5). Fractions containing the pure protein of interest were pooled and concentrated. The protein purity was estimated using SDS-PAGE with Coomassie staining, and the protein concentration was determined using the Bradford assay.
Ca YqeK Purification. Ca YqeK was purified following the exact same protocols as those used for Bh YqeK, except for some changes in the lysis buffer (100 mM HEPES, 500 mM NaCl, 10 mM imidazole, pH 7.5), the wash buffer (100 mM HEPES, 500 mM NaCl, 25 mM imidazole, pH 7.5), the elution buffer (100 mM HEPES, 500 mM NaCl, 250 mM imidazole, 15% glycerol, pH 7.5), and the storage buffer (100 mM HEPES, 500 mM NaCl, 15% glycerol, pH 7.5).
Purification and reconstitution of the apo forms of Bh and Ca YqeK. The apo forms of Bh and Ca YqeK (devoid of metallocofactor) were isolated following the exact same procedures as those of the holo proteins, with the exception that 0.5 mM of EDTA was added to the lysis buffers, and 5 mM of EDTA was added to the elution buffers. The apo-proteins were degassed and then reconstituted under O2-free conditions, with the addition of 5 mM of dithiothreitol (DTT) for 2 min and 2 molar equivalents of Fe(NH4)SO4 for 20 min. The reconstituted proteins were then exposed to the air to oxidize the cofactor and passed through the HiLoad 16/600 Superdex column (Cytiva) under ambient conditions.
Generation of YqeK variants. Single-point variants of Bh YqeK were generated using the Q5 Hot Start Site Directed Mutagenesis Kit (NEB, Ipswitch, MA, USA) with the primers listed in Table S1. The variant sequences were confirmed using Sanger sequencing (Azenta Inc., Burlington, MA, USA). The YqeK variants were expressed and isolated following the same procedures as those for the wild-type protein.
Metal Quantification. A metal analysis of the as-purified proteins was carried out using inductively coupled atomic emission spectrometry (ICP-AES) at the Environmental Sustainability Laboratories (EESL) at the Pennsylvania State University (University Park, PA, USA).
Electron Paramagnetic Resonance (EPR) Spectroscopy. All samples were prepared in a storage buffer under O2-free conditions in an anaerobic glovebox (CoyLab, Grass Lake, MI, USA). The samples were reacted with sodium ascorbate for 10 min at 22 °C prior to being frozen in liquid N2. EPR spectra were recorded on a Bruker E500 Elexsys continuous wave (CW) X-band spectrometer (operating at approximately 9.36 GHz, Bruker, Karlsruhe, Germany) equipped with a rectangular resonator (TE102) and a continuous-flow cryostat (Oxford 910) with a temperature controller (Oxford ITC 503). The spectra were recorded at 10 K and at a microwave power of 0.2 mW using a modulation amplitude of 1 mT and a microwave frequency of 9.36 GHz.
Redox titrations monitored by EPR. For the titration experiments, 4–5 mL of protein was maintained under O2-free conditions in a titration cell, like the one described by Dutton et al. [66]. The protein solution was maintained O2-free during the course of the titration by continuous flushing with hydrated argon gas. The titration was carried out by adjusting the redox potential of the solution with sub-stoichiometric amounts of sodium dithionite (Na2S2O4) and potassium ferricyanide (K3[Fe(CN)6]). The temperature was maintained at 15 °C by cold water passing through the glass body of the cell. The potentials were measured with a pH/redox meter (GPHR 1400, Greisinger, Münzbach, Germany) using a combination Pt/Ag/AgCl micro-electrode (3 M KCl, Mettler Toledo, Port Melbourne, Australia) and quoted relative to the normal standard hydrogen electrode. The electrode was calibrated with a saturated quinhydrone (Qh) solution at a pH of 9 (1 M CHES buffer) based on the following formula: E = EQh − 0.1984 (273.16 + T) pH, where EQh is the measured potential of the saturated quinhydrone solution in mV and T is the temperature in K. For the titration at pH 9, the buffer solution was 25 mM of CHES and 100 mM of KCl, and the following redox mediators were present at a final concentration of 100 µM, with the exception of phenazine methosulfate which was at a final concentration of 50 µM: 2,6-dichlorophenolindophenol (+220 mV), 1,2-naphthoquinone (+145 mV), phenazine methosulfate (+90 mV), 1,4-napthoquinone (+60 mV), methylene blue (+11 mV), 5-hydroxy-1,4-napthoquinone (−93 mV), indigo carmine (−130 mV), 2- hydroxyl-1,4 naphthoquinone (−137 mV), anthraquinone-1,5-disulfonate (−170 mV), phenosafranine (−252 mV), and safranine T (−290 mV). Values are quoted with respect to the standard hydrogen electrode potential (SHE) and the error in the determination of the oxidation-reduction potentials is ±20 mV. At each selected equilibration potential, 300 µL samples were quickly loaded in calibrated EPR X-Band tubes under a stream of argon gas and frozen in a cold liquid ethanol–nitrogen mixture. For the determination of the midpoint reduction potentials of the redox couples FeIII–FeIII/FeIII–FeII and FeIII–FeII/FeII–FeII were obtained by fitting the intensity of the EPR signal of the mixed-valence cofactor at the respective potentials to the Nernst equation, considering two one-electron transfer events using the following equation:
I n t ( M V ) = e x p ⁡ ( F R T · ( E − E r e d ) ) 1 + exp ⁡ F R T · E − E r e d · ( 1 + exp ⁡ F R T · E − E o x
E is the measured potential, F is the Faraday constant (96,486 J∙V−1), R is the universal gas constant (8.314472 J∙K−1∙mol−1), T is the temperature (K), Eox = Em(FeIII–FeIII/FeIII–FeII), and Ered = Em(FeIII–FeII/FeII–FeII).
The plotted intensities correspond either to the peak or peak-to-peak intensities of the EPR signals of the two mixed-valent components, and these were used for plotting the redox dependence of the two different mixed-valent forms
Mössbauer Spectroscopy. Mössbauer spectra were recorded on a WEB Research (Edina, MN, USA) instrument that has been described previously [30]. The spectrometer used to acquire the weak-field spectra is equipped with a Janis SVT-400 variable-temperature cryostat (Lakeshore, Westerville, OH, USA). The external magnetic field was applied parallel to the γ beam. All isomer shifts are quoted relative to the centroid of the spectrum of α-iron metal at room temperature. Mössbauer spectra were fitted using the WMOSS spectral analysis software (www.wmoss.org, version 4F, accessed on 1 June 2023, WEB Research).
Ap4A hydrolysis by high-performance liquid chromatography (HPLC).
YqeK isolated from cells grown in LB or M9-Fe media. All activity assays were carried out in 25 mM of CHES and 100 mM of KCl at a pH of 9.0 unless otherwise stated. The pH for the activity assays was selected on the basis of a pH screen, under which both Ca and Bh YqeK exhibited maximal activity at pH 9 (Figure S1). The reactions contained 15 nM Bh or Ca YqeK and 0.25 mg/mL of bovine serum albumin (BSA, Chatswood, Australia). Hydrolysis of Ap4A by Bh or Ca YqeK was examined under the following three different redox conditions: oxidized (treated with 1 mM H2O2 for 10 min), as-isolated, and reduced (treated with 2 mM sodium dithionite for 10 min). Activity assays of the dithionite-reduced YqeK were carried out under O2-free conditions in an anaerobic glovebox. Reactions were initiated with 300 μM of Ap4A and stopped after 5 min by the addition of 0.6 M of HClO4. The samples were then neutralized with the addition of 0.8 M of K2CO3. The samples were then centrifuged at 17,000× g for 10 min, and the supernatant was filtered with a 0.22 μM nylon Spin-X Centrifuge Filter (Corning Inc, Corning, NY, USA). The filtered samples were analyzed on a 1260 Infinity Liquid Chromatography system (Agilent, Santa Clara, CA, USA) equipped with a reverse-phase C18-A Polaris column (Agilent, particle size of 5 μm, 150 mm × 4.6 mm) by monitoring the absorbance at 254 nm. The reaction products and substrates were eluted with a linear gradient using a water-based mobile phase consisting of 10 mM of KH2PO4 and 10 mM of tetrabutylammonium hydroxide (TBAH) at a pH of 6.0 and an organic-based mobile phase containing 10 mM of TBAH in methanol. The relative integrated peak intensities were used to quantify Ap4A and ADP concentrations, accounting for their different extinction coefficients, 25.4 and 15.4 mM−1 cm−1, respectively, as well as for the stoichiometry of the reaction (i.e., the hydrolysis of one molecule of Ap4A results in two molecules of ADP). The actual concentrations of the two nucleotides were calculated by considering that the sum of the integrated peak intensities (accounting for the different extinction coefficients) should correspond to the initial concentration of Ap4A in the reaction and that the ADP formed or Ap4A consumed is the fractional ratio of its integrated peak intensity to the sum of the intensities, respectively.
Time-dependent Ap4A hydrolysis by the M9-Fe YqeK forms. To estimate the apparent rate of hydrolysis by the M9-Fe enzyme forms under different redox conditions, we carried out time-dependent assays at saturating substrate concentrations (two orders of magnitude higher than the reported KM for Ap4A) [3,23]. For the M9-Fe proteins, all reactions were performed in the anaerobic glovebox with degassed buffers. The M9-Fe Bh YqeK and Ca YqeK (25 nM) were assayed either in the presence of sodium ascorbate (1 μM) or dithionite (2 mM). The samples were quenched, neutralized, filtered, and analyzed as described above.

5. Conclusions

Ap4A is a stress metabolite of growing interest due to its potential as a target for antibiotic resistance therapies. YqeK is the major Ap4A hydrolase in Gram-positive bacteria and increases the ability of bacteria to cope with stress. We characterized YqeKs from two different environmental niches, the facultative aerobe Bh and the obligate anaerobe Ca, and showed that both utilize Fe-based metal active sites for hydrolysis. While Fe is a common trait of the chemical makeup of their active sites, Bh YqeK has adopted two strategies that allow for sustained activity under more oxidative conditions. The more O2-tolerant hydrolysis is afforded by either the preferential incorporation of Fe–Zn mixed-metal sites (i.e., Fe–Zn) or upshifted reduction potentials of the diiron sites (Fe–Fe). The present results further our understanding of the adaptation of YqeK hydrolases to more oxidative environments in aerobic bacteria and demonstrate an evolutionary tuning of the metal preference and redox potentials of bimetallic Fe-based cofactors to endorse their catalytic capabilities.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/catal14090652/s1: Figure S1: Percentage distribution of the major Phyla that are present in the SSN in the main text that consists of 8028 sequences; Figure S2: Sequence alignment of the Ca and Bh YqeK. Colored boxes represent sites that are 100% conserved among the two proteins, which share 42% sequence identity. Sequence alignment was generated with MAFFT; Figure S3: Representative rhombogram assuming a negative zero-field splitting parameter D for a system with an S = 5/2 similar to that observed in purple acid phosphatases; Figure S4: CW X-Band EPR spectra of different Bh YqeK preparations demonstrates a variability in the two mixed-valent species (i.e., Component 1, Component 2). All samples have been reduced with 2 molar excess of sodium ascorbate. Experimental conditions: temperature = 10 K, microwave frequency = 9.36 GHz, modulation amplitude = 1 mT, microwave power = 0.2 mW; Figure S5: CW X-Band EPR spectra of the ascorbate-reduced M9-Fe Bh and Ca YqeK as isolated or reduced with 2 molar excess of sodium ascorbate. Experimental conditions: temperature = 10 K, microwave frequency = 9.36 GHz, modulation amplitude = 1 mT, microwave power = 0.2 mW; Figure S6: End-point activity assay to examine the pH dependence of the hydrolysis by Bh and Ca YqeK expressed in LB media. Experimental conditions: [YqeK] = 5 nM, [Ap4A] = 300 μM, reaction time 300 s under air; Table S1: Primers utilized for the generation of the Bh YqeK variants; Table S2: Fe content of the variant Bh YqeK as estimated by ferrozine assays. Ref. [67] is cited in the Supplementary Materials.

Author Contributions

Conceptualization, M.-E.P.; Methodology, C.U. and M.-E.P.; Validation, M.-E.P.; Formal analysis, C.U., N.C., L.P., R.P. and M.-E.P.; Investigation, C.U., N.C., Q.Y. and R.P.; Data curation, C.U., Q.Y., L.P., R.P. and M.-E.P.; Writing—original draft, C.U.; Writing—review & editing, C.U. and M.-E.P.; Project administration, M.-E.P.; Funding acquisition, M.-E.P. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the National Institutes of Health (Grant No. GM111978 and Grant No. GM126303 to M.-E.P.).

Data Availability Statement

Data are contained within the article and Supporting Information. The original contributions presented in this study are included in the article; further inquiries can be directed to the corresponding author.

Acknowledgments

The authors thank Paul S. Ralifo for providing us access to the EPR spectrometer at the Chemical Instrumentation Center of Boston University (Boston, MA, USA). The authors also thank Laura Jean Liermann for the ICP-AES analysis at the Laboratory for Isotopes and Metals in the Environment at the Pennsylvania State University (University Park, PA, USA).

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Aravind, L.; Koonin, E.V. The HD domain defines a new superfamily of metal-dependent phosphohydrolases. Trends Biochem. Sci. 1998, 23, 469–472. [Google Scholar] [CrossRef] [PubMed]
  2. Langton, M.; Sun, S.; Ueda, C.; Markey, M.; Chen, J.; Paddy, I.; Jiang, P.; Chin, N.; Milne, A.; Pandelia, M.-E. The HD-Domain Metalloprotein Superfamily: An Apparent Common Protein Scaffold with Diverse Chemistries. Catalysts 2020, 10, 1191. [Google Scholar] [CrossRef] [PubMed]
  3. Minazzato, G.; Gasparrini, M.; Amici, A.; Cianci, M.; Mazzola, F.; Orsomando, G.; Sorci, L.; Raffaelli, N. Functional Charac-terization of COG1713 (YqeK) as a Novel Diadenosine Tetraphosphate Hydrolase Family. J. Bacteriol. 2020, 202, 10–1128. [Google Scholar] [CrossRef] [PubMed]
  4. Ferguson, F.; McLennan, A.G.; Urbaniak, M.D.; Jones, N.J.; Copeland, N.A. Re-evaluation of Diadenosine Tetraphosphate (Ap4A) From a Stress Metabolite to Bona Fide Secondary Messenger. Front. Mol. Biosci. 2020, 7, 606807. [Google Scholar] [CrossRef] [PubMed]
  5. Ji, X.; Zou, J.; Peng, H.; Stolle, A.-S.; Xie, R.; Zhang, H.; Peng, B.; Mekalanos, J.J.; Zheng, J. Alarmone Ap4A is elevated by aminoglycoside antibiotics and enhances their bactericidal activity. Proc. Natl. Acad. Sci. USA 2019, 116, 9578–9585. [Google Scholar] [CrossRef]
  6. Monds Russell, D.; Newell Peter, D.; Wagner Jeffrey, C.; Schwartzman Julia, A.; Lu, W.; Rabinowitz Joshua, D.; O’Toole George, A. Di-Adenosine Tetraphosphate (Ap4A) Metabolism Impacts Biofilm Formation by Pseudomonas fluorescens via Modulation of c-di-GMP-Dependent Pathways. J. Bacteriol. 2010, 192, 3011–3023. [Google Scholar] [CrossRef]
  7. Rapaport, E.; Zamecnik, P.C. Presence of diadenosine 5′,5′′′-P1, P4-tetraphosphate (Ap4A) in mamalian cells in levels varying widely with proliferative activity of the tissue: A possible positive “pleiotypic activator”. Proc. Natl. Acad. Sci. USA 1976, 73, 3984–3988. [Google Scholar] [CrossRef]
  8. Bochner, B.R.; Lee, P.C.; Wilson, S.W.; Cutler, C.W.; Ames, B.N. AppppA and related adenylylated nucleotides are synthesized as a consequence of oxidation stress. Cell 1984, 37, 225–232. [Google Scholar] [CrossRef]
  9. Ismail, T.M.; Hart, C.; McLennan, A.G. Regulation of Dinucleoside Polyphosphate Pools by the YgdP and ApaH Hydrolases Is Essential for the Ability of Salmonella enterica serovar Typhimurium to Invade Cultured Mammalian Cells. J. Biol. Chem. 2003, 278, 32602–32607. [Google Scholar] [CrossRef]
  10. Kimura, Y.; Tanaka, C.; Sasaki, K.; Sasaki, M. High concentrations of intracellular Ap4A and/or Ap5A in developing Myxococcus xanthus cells inhibit sporulation. Microbiology 2017, 163, 86–93. [Google Scholar] [CrossRef]
  11. Luciano, D.J.; Belasco, J.G. Np 4 A alarmones function in bacteria as precursors to RNA caps. Proc. Natl. Acad. Sci. USA 2020, 117, 3560–3567. [Google Scholar] [CrossRef] [PubMed]
  12. František Potužník, J.; Nešuta, O.; Škríba, A.; Volenikova, B.; Mititelu, M.B.; Mancini, F.; Serianni, V.; Fernandez, H.; Spustová, K.; Trylčová, J. Diadenosine tetraphosphate (Ap4A) serves as a 5′ RNA cap in mammalian cells. Angew. Chem. 2024, 136, e202314951. [Google Scholar] [CrossRef]
  13. Despotović, D.; Brandis, A.; Savidor, A.; Levin, Y.; Fumagalli, L.; Tawfik, D.S. Diadenosine tetraphosphate (Ap4A)—An E. coli alarmone or a damage metabolite? FEBS J. 2017, 284, 2194–2215. [Google Scholar] [CrossRef] [PubMed]
  14. Oka, M.; Takegawa, K.; Kimura, Y. Enzymatic characterization of a class II lysyl-tRNA synthetase, LysS, from Myxococcus xanthus. Arch. Biochem. Biophys. 2015, 579, 33–39. [Google Scholar] [CrossRef] [PubMed]
  15. Wright, M.; Boonyalai, N.; Tanner, J.A.; Hindley, A.D.; Miller, A.D. The duality of LysU, a catalyst for both Ap4A and Ap3A formation. FEBS J. 2006, 273, 3534–3544. [Google Scholar] [CrossRef]
  16. Hansen, S.; Lewis, K.; Vulić, M. Role of Global Regulators and Nucleotide Metabolism in Antibiotic Tolerance in Escherichia coli. Antimicrob. Agents Chemother. 2008, 52, 2718–2726. [Google Scholar] [CrossRef]
  17. Zheng, T.; Jing, M.; Gong, T.; Yan, J.; Zeng, J.; Li, Y. Deletion of the yqeK gene leads to the accumulation of Ap4A and reduced biofilm formation in Streptococcus mutans. Mol. Oral Microbiol. 2021, 37, 9–21. [Google Scholar] [CrossRef]
  18. Branda, S.S.; González-Pastor, J.E.; Dervyn, E.; Ehrlich, S.D.; Losick, R.; Kolter, R. Genes Involved in Formation of Structured Multicellular Communities by Bacillus subtilis. J. Bacteriol. 2004, 186, 3970–3979. [Google Scholar] [CrossRef]
  19. McLennan, A.G. The Nudix hydrolase superfamily. Cell. Mol. Life Sci. 2006, 63, 123–143. [Google Scholar] [CrossRef]
  20. McLennan, A.G. Dinucleoside polyphosphates—Friend or foe? Pharmacol. Therap. 2000, 87, 73–89. [Google Scholar] [CrossRef]
  21. Guranowski, A. Specific and nonspecific enzymes involved in the catabolism of mononucleoside and dinucleoside polyphosphates. Pharmacol. Ther. 2000, 87, 117–139. [Google Scholar] [CrossRef] [PubMed]
  22. Sasaki, M.; Takegawa, K.; Kimura, Y. Enzymatic characteristics of an ApaH-like phosphatase, PrpA, and a diadenosine tetraphosphate hydrolase, ApaH, from Myxococcus xanthus. FEBS Lett. 2014, 588, 3395–3402. [Google Scholar] [CrossRef] [PubMed]
  23. Jin, Y.; Ke, J.; Zheng, P.; Zhang, H.; Zhu, Z.; Niu, L. Structural and biochemical characterization of a nucleotide hydrolase from Streptococcus pneumonia. Structure 2024, 32, 197–1207.e4. [Google Scholar] [CrossRef] [PubMed]
  24. Plateau, P.; Fromant, M.; Brevet, A.; Gesquiere, A.; Blanquet, S. Catabolism of bis(5′-nucleosidyl) oligophosphates in Escherichia coli: Metal requirements and substrate specificity of homogeneous diadenosine-5′,5′′′-P1,P4-tetraphosphate pyrophosphohydrolase. Biochemistry 1985, 24, 914–922. [Google Scholar] [CrossRef] [PubMed]
  25. He, P.; Deng, C.; Liu, B.; Zeng, L.; Zhao, W.; Zhang, Y.; Jiang, X.; Guo, X.; Qin, J. Characterization of a bifunctional enzyme with (p)ppGpp-hydrolase/synthase activity in Leptospira interrogans. FEMS Microbiol. Lett. 2013, 348, 133–142. [Google Scholar] [CrossRef]
  26. Mechold, U.; Murphy, H.; Brown, L.; Cashel, M. Intramolecular Regulation of the Opposing (p)ppGpp Catalytic Activities of RelSeq, the Rel/Spo Enzyme from Streptococcus equisimilis. J. Bacteriol. 2002, 184, 2878–2888. [Google Scholar] [CrossRef]
  27. Jeon, Y.J.; Park, S.C.; Song, W.S.; Kim, O.-H.; Oh, B.-C.; Yoon, S.-I. Structural and biochemical characterization of bacterial YpgQ protein reveals a metal-dependent nucleotide pyrophosphohydrolase. J. Struct. Biol. 2016, 195, 113–122. [Google Scholar] [CrossRef]
  28. Sun, S.; He, Z.; Jiang, P.; Baral, R.; Pandelia, M.-E. Metal Dependence and Functional Diversity of Type I Cas3 Nucleases. Biochemistry 2022, 61, 327–338. [Google Scholar] [CrossRef]
  29. Sun, S.; Wang, R.; Pandelia, M.-E. Vibrio cholerae V-cGAP3 Is an HD-GYP Phosphodiesterase with a Metal Tunable Substrate Selectivity. Biochemistry 2022, 61, 1801–1809. [Google Scholar] [CrossRef]
  30. Sun, S.; Pandelia, M.-E. HD-[HD-GYP] Phosphodiesterases: Activities and Evolutionary Diversification within the HD-GYP Family. Biochemistry 2020, 59, 2340–2350. [Google Scholar] [CrossRef]
  31. Mulepati, S.; Bailey, S. Structural and Biochemical Analysis of Nuclease Domain of Clustered Regularly Interspaced Short Palindromic Repeat (CRISPR)-associated Protein 3 (Cas3). J. Biol. Chem. 2011, 286, 31896–31903. [Google Scholar] [CrossRef] [PubMed]
  32. Gerber, E.; Bernard, R.; Castang, S.; Chabot, N.; Coze, F.; Dreux-Zigha, A.; Hauser, E.; Hivin, P.; Joseph, P.; Lazarelli, C.; et al. Deinococcus as new chassis for industrial biotechnology: Biology, physiology and tools. J. Appl. Microbiol. 2015, 119, 1–10. [Google Scholar] [CrossRef] [PubMed]
  33. Hoiczyk, E.; Hansel, A. Cyanobacterial cell walls: News from an unusual prokaryotic envelope. J. Bacteriol. 2000, 182, 1191–1199. [Google Scholar] [CrossRef] [PubMed]
  34. Beck, J.L.; De Jersey, J.; Zerner, B.; Hendrich, M.P.; Debrunner, P.G. Properties of the Fe(II)-Fe(III) derivative of red kidney bean purple phosphatase. Evidence for a binuclear zinc-iron center in the native enzyme. J. Am. Chem. Soc. 1988, 110, 3317–3318. [Google Scholar] [CrossRef]
  35. Durmus, A.; Eicken, C.; Sift, B.H.; Kratel, A.; Kappl, R.; Hüttermann, J.; Krebs, B. The active site of purple acid phosphatase from sweet potatoes (Ipomoea batatas). Eur. J. Biochem. 1999, 260, 709–716. [Google Scholar] [CrossRef]
  36. Merkx, M.; Averill, B.A. The Activity of Oxidized Bovine Spleen Purple Acid Phosphatase Is Due to an Fe(III)Zn(II) ‘Impurity’. Biochemistry 1998, 37, 11223–11231. [Google Scholar] [CrossRef]
  37. Reiter, T.A.; Rusnak, F. Electrochemical Studies of the Mono-Fe, Fe−Zn, and Fe−Fe Metalloisoforms of Bacteriophage λ Protein Phosphatase. Biochemistry 2004, 43, 782–790. [Google Scholar] [CrossRef]
  38. Davis, J.C.; A Averill, B. Evidence for a spin-coupled binuclear iron unit at the active site of the purple acid phosphatase from beef spleen. Proc. Natl. Acad. Sci. USA 1982, 79, 4623–4627. [Google Scholar] [CrossRef]
  39. Wörsdörfer, B.; Lingaraju, M.; Yennawar, N.H.; Boal, A.K.; Krebs, C.; Bollinger, J.M.; Pandelia, M.-E. Organophosphonate-degrading PhnZ reveals an emerging family of HD domain mixed-valent diiron oxygenases. Proc. Natl. Acad. Sci. USA 2013, 110, 18874–18879. [Google Scholar] [CrossRef]
  40. Xing, G.; Barr, E.W.; Diao, Y.; Hoffart, L.M.; Prabhu, K.S.; Arner, R.J.; Reddy, C.C.; Krebs, C.; Bollinger, J.M., Jr. Oxygen ac-tivation by a mixed-valent, diiron(II/III) cluster in the glycol cleavage reaction catalyzed by myo-inositol oxygenase. Biochemistry 2006, 45, 5402–5412. [Google Scholar] [CrossRef]
  41. Bollinger, J.M.; Diao, Y.; Matthews, M.L.; Xing, G.; Krebs, C. myo-Inositol oxygenase: A radical new pathway for O2and C–H activation at a nonheme diiron cluster. Dalton Trans. 2008, 6, 905–914. [Google Scholar] [CrossRef] [PubMed]
  42. Lynch, J.; Juarez-Garcia, C.; Münck, E.; Que, L., Jr. Mössbauer and EPR studies of the binuclear iron center in ribonucleotide reductase from Escherichia coli: A new iron-to-protein stoichiometry. J. Biol. Chem. 1989, 264, 8091–8096. [Google Scholar] [CrossRef] [PubMed]
  43. Langton, M.; Appell, M.; Koob, J.; Pandelia, M.-E. Domain Fusion of Two Oxygenases Affords Organophosphonate Degradation in Pathogenic Fungi. Biochemistry 2022, 61, 956–962. [Google Scholar] [CrossRef] [PubMed]
  44. Chen, J.; Calderone, L.A.; Pan, L.; Quist, T.; Pandelia, M.-E. The Fe and Zn cofactor dilemma. Biochim. Biophys. Acta-Proteins Proteom. 2023, 1871, 140931. [Google Scholar] [CrossRef] [PubMed]
  45. Olczak, M.; Morawiecka, B.; Watorek, W. Plant purple acid phosphatases—Genes, structures and biological function. Acta Biochim. Pol. 2003, 50, 1245–1256. [Google Scholar] [CrossRef] [PubMed]
  46. Funhoff, E.G.; Bollen, M.; Averill, B.A. The Fe(III)Zn(II) form of recombinant human purple acid phosphatase is not activated by proteolysis. J. Inorg. Biochem. 2005, 99, 521–529. [Google Scholar] [CrossRef]
  47. Goldstone, D.C.; Ennis-Adeniran, V.; Hedden, J.J.; Groom, H.C.T.; Rice, G.I.; Christodoulou, E.; Walker, P.A.; Kelly, G.; Haire, L.F.; Yap, M.W.; et al. HIV-1 restriction factor SAMHD1 is a deoxynucleoside triphosphate triphosphohydrolase. Nature 2011, 480, 379–382. [Google Scholar] [CrossRef]
  48. Vorontsov, I.I.; Minasov, G.; Kiryukhina, O.; Brunzelle, J.S.; Shuvalova, L.; Anderson, W.F. Characterization of the deoxynu-cleotide triphosphate triphosphohydrolase (dNTPase) activity of the EF1143 protein from Enterococcus faecalis and crystal structure of the activator-substrate complex. J. Biol. Chem. 2011, 286, 33158–33166. [Google Scholar] [CrossRef]
  49. Bridwell-Rabb, J.; Kang, G.; Zhong, A.; Liu, H.-W.; Drennan, C.L. An HD domain phosphohydrolase active site tailored for oxetanocin-A biosynthesis. Proc. Natl. Acad. Sci. USA 2016, 113, 13750–13755. [Google Scholar] [CrossRef]
  50. Davidson, J.F.; Whyte, B.; Bissinger, P.H.; Schiestl, R.H. Oxidative stress is involved in heat-induced cell death in Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 1996, 93, 5116–5121. [Google Scholar] [CrossRef]
  51. Marcén, M.; Ruiz, V.; Serrano, M.J.; Condón, S.; Mañas, P. Oxidative stress in E. coli cells upon exposure to heat treatments. Int. J. Food Microbiol. 2017, 241, 198–205. [Google Scholar] [CrossRef] [PubMed]
  52. Morano, K.A.; Grant, C.M.; Moye-Rowley, W.S. The Response to Heat Shock and Oxidative Stress in Saccharomyces cerevisiae. Genetics 2012, 190, 1157–1195. [Google Scholar] [CrossRef] [PubMed]
  53. Kohanski, M.A.; Dwyer, D.J.; Hayete, B.; Lawrence, C.A.; Collins, J.J. A Common Mechanism of Cellular Death Induced by Bactericidal Antibiotics. Cell 2007, 130, 797–810. [Google Scholar] [CrossRef] [PubMed]
  54. Dwyer, D.J.; Belenky, P.A.; Yang, J.H.; MacDonald, I.C.; Martell, J.D.; Takahashi, N.; Chan, C.T.; Lobritz, M.A.; Braff, D.; Schwarz, E.G.; et al. Antibiotics induce redox-related physiological alterations as part of their lethality. Proc. Natl. Acad. Sci. USA 2014, 111, E2100–E2109. [Google Scholar] [CrossRef] [PubMed]
  55. Dukan, S.; Farewell, A.; Ballesteros, M.; Taddei, F.; Radman, M.; Nyström, T. Protein oxidation in response to increased transcriptional or translational errors. Proc. Natl. Acad. Sci. USA 2000, 97, 5746–5749. [Google Scholar] [CrossRef]
  56. Ding, H.; Hidalgo, E.; Demple, B. The Redox State of the [2Fe-2S] Clusters in SoxR Protein Regulates Its Activity as a Transcription Factor. J. Biol. Chem. 1996, 271, 33173–33175. [Google Scholar] [CrossRef]
  57. Traoré, D.A.K.; El Ghazouani, A.; Jacquamet, L.; Borel, F.; Ferrer, J.-L.; Lascoux, D.; Ravanat, J.-L.; Jaquinod, M.; Blondin, G.; Caux-Thang, C.; et al. Structural and functional characterization of 2-oxo-histidine in oxidized PerR protein. Nat. Chem. Biol. 2008, 5, 53–59. [Google Scholar] [CrossRef]
  58. Fuangthong, M.; Herbig, A.F.; Bsat, N.; Helmann, J.D. Regulation of the Bacillus subtilis fur and perR genes by PerR: Not all members of the PerR regulon are peroxide inducible. J. Bacteriol. 2002, 184, 3276–3286. [Google Scholar] [CrossRef]
  59. Porcheron, G.; Garénaux, A.; Proulx, J.; Sabri, M.; Dozois, C.M. Iron, copper, zinc, and manganese transport and regulation in pathogenic Enterobacteria: Correlations between strains, site of infection and the relative importance of the different metal transport systems for virulence. Front. Cell. Infect. Microbiol. 2013, 3, 90. [Google Scholar] [CrossRef]
  60. Payne, S.M.; Neilands, I.B. Iron and Virulence in the Family Enterobacteriaceae. CRC Crit. Rev. Microbiol. 1988, 16, 81–111. [Google Scholar] [CrossRef]
  61. Caza, M.; Kronstad, J.W. Shared and distinct mechanisms of iron acquisition by bacterial and fungal pathogens of humans. Front. Cell. Infect. Microbiol. 2013, 3, 80. [Google Scholar] [CrossRef] [PubMed]
  62. Zallot, R.; Oberg, N.; Gerlt, J.A. The EFI Web Resource for Genomic Enzymology Tools: Leveraging Protein, Genome, and Metagenome Databases to Discover Novel Enzymes and Metabolic Pathways. Biochemistry 2019, 58, 4169–4182. [Google Scholar] [CrossRef] [PubMed]
  63. Katoh, K.; Rozewicki, J.; Yamada, K.D. MAFFT online service: Multiple sequence alignment, interactive sequence choice and visualization. Briefings Bioinform. 2019, 20, 1160–1166. [Google Scholar] [CrossRef] [PubMed]
  64. Hoang, D.T.; Chernomor, O.; Von Haeseler, A.; Minh, B.Q.; Vinh, L.S. UFBoot2: Improving the Ultrafast Bootstrap Approximation. Mol. Biol. Evol. 2018, 35, 518–522. [Google Scholar] [CrossRef] [PubMed]
  65. Trifinopoulos, J.; Nguyen, L.-T.; von Haeseler, A.; Minh, B.Q. W-IQ-TREE: A fast online phylogenetic tool for maximum likelihood analysis. Nucleic Acids Res. 2016, 44, W232–W235. [Google Scholar] [CrossRef]
  66. Leslie Dutton, P. [23] Redox potentiometry: Determination of midpoint potentials of oxidation-reduction components of bio-logical electron-transfer systems. In Methods Enzymol; Academic Press: Cambridge, MA, USA, 1978; Volume 54, pp. 411–435. [Google Scholar]
  67. Schenk, G.; Boutchard, C.L.; Carrington, L.E.; Noble, C.J.; Moubaraki, B.; Murray, K.S.; de Jersey, J.; Hanson, G.R.; Hamilton, S. A Purple Acid Phosphatase from Sweet Potato Contains an Antiferromagnetically Coupled Binuclear Fe-Mn Center. J. Biol. Chem. 2001, 276, 19084–19088. [Google Scholar] [CrossRef]
Catalysts 14 00652 sch001
Scheme 1. Reaction catalyzed by YqeK.
Scheme 1. Reaction catalyzed by YqeK.
Catalysts 14 00652 sch001
Catalysts 14 00652 g001
Figure 1. Occurrence and phylogenetic distribution of YqeK hydrolases. (A) Sequence similarity network (SSN) of YqeK. The SSN contains 8028 unique sequences found with nodes colored based on taxonomic order. The cluster collecting cyanobacterial sequences is shown separately. Clusters containing fewer than 4 sequences were omitted. (B) Maximum likelihood phylogenetic tree of YqeK generated with the IQ-tree software (v2.3.2) with the uncharacterized HD-domain hydrolase (pdb: 2PQ7) used as outgroup. The branches have been labeled according to the genus. The scale bar represents the number of substitutions per site. The bootstrap values are shown for the major branches. Ca YqeK is part of the blue clade (Clostridium), while Bh YqeK is found in the pink clade (Bacillus).
Figure 1. Occurrence and phylogenetic distribution of YqeK hydrolases. (A) Sequence similarity network (SSN) of YqeK. The SSN contains 8028 unique sequences found with nodes colored based on taxonomic order. The cluster collecting cyanobacterial sequences is shown separately. Clusters containing fewer than 4 sequences were omitted. (B) Maximum likelihood phylogenetic tree of YqeK generated with the IQ-tree software (v2.3.2) with the uncharacterized HD-domain hydrolase (pdb: 2PQ7) used as outgroup. The branches have been labeled according to the genus. The scale bar represents the number of substitutions per site. The bootstrap values are shown for the major branches. Ca YqeK is part of the blue clade (Clostridium), while Bh YqeK is found in the pink clade (Bacillus).
Catalysts 14 00652 g001
Catalysts 14 00652 g002
Figure 2. Characterization of Bh and Ca YqeK isolated from LB. (A) 15% SDS-PAGE of Ca and Bh YqeK show a band at ~21 kDa. (B) Representative size exclusion chromatogram of YqeK, eluting at the expected molecular weight for a homodimer. (C) Crystal structures of Bh YqeK (left, pdb: 2O08) and Ca YqeK (right, pdb: 3CCG) with a dGDP and phosphate bound, respectively. Heteroatoms are colored as follows: nitrogen (blue), oxygen (red), and phosphorus (orange). Fe ions are shown as orange spheres. (D) High-spin region of the as-isolated Bh YqeK, (E) low-spin region of the as-isolated Bh YqeK, (F) high-spin region of the as-isolated Ca YqeK, and (G) low-spin region of the as-isolated Ca YqeK. Experimental conditions: temperature = 10 K, microwave frequency = 9.36 GHz, modulation amplitude = 1 mT, and microwave power = 0.2 mW. (H) HPLC chromatograms of Bh YqeK and (I) of Ca YqeK monitoring Ap4A hydrolysis and ADP formation. Conditions are as-isolated under ambient conditions (black trace) and treated with 1 mM H2O2 (orange trace) and 2 mM (green trace). The product ADP elutes at ~16 min, while the substrate Ap4A elutes at ~22 min. These experiments are an average of eight replicates, and the average representative chromatograms are shown.
Figure 2. Characterization of Bh and Ca YqeK isolated from LB. (A) 15% SDS-PAGE of Ca and Bh YqeK show a band at ~21 kDa. (B) Representative size exclusion chromatogram of YqeK, eluting at the expected molecular weight for a homodimer. (C) Crystal structures of Bh YqeK (left, pdb: 2O08) and Ca YqeK (right, pdb: 3CCG) with a dGDP and phosphate bound, respectively. Heteroatoms are colored as follows: nitrogen (blue), oxygen (red), and phosphorus (orange). Fe ions are shown as orange spheres. (D) High-spin region of the as-isolated Bh YqeK, (E) low-spin region of the as-isolated Bh YqeK, (F) high-spin region of the as-isolated Ca YqeK, and (G) low-spin region of the as-isolated Ca YqeK. Experimental conditions: temperature = 10 K, microwave frequency = 9.36 GHz, modulation amplitude = 1 mT, and microwave power = 0.2 mW. (H) HPLC chromatograms of Bh YqeK and (I) of Ca YqeK monitoring Ap4A hydrolysis and ADP formation. Conditions are as-isolated under ambient conditions (black trace) and treated with 1 mM H2O2 (orange trace) and 2 mM (green trace). The product ADP elutes at ~16 min, while the substrate Ap4A elutes at ~22 min. These experiments are an average of eight replicates, and the average representative chromatograms are shown.
Catalysts 14 00652 g002
Catalysts 14 00652 g003
Figure 3. CW X-Band EPR of the M9-Fe forms of Bh and Ca YqeK. (A) High-spin region of the CW X-Band EPR spectra recorded on the as-isolated M9-Fe Bh YqeK (top) and Ca YqeK (bottom), (B) low-spin region of the X-Band EPR spectra of the ascorbate-reduced M9-Fe Bh YqeK as isolated (top) or chelated and reconstituted (bottom), (C) low-spin region of the X-Band EPR spectra of the ascorbate-reduced M9-Fe Ca YqeK as isolated (top) or chelated and reconstituted (bottom). (D) CW EPR spectra of the M9-Fe Bh YqeK incubated with a series of (d)/NTPs. The asterisk denotes a small amount of paramagnetic MnII. Experimental conditions: temperature = 10 K, microwave frequency = 9.36 GHz, modulation amplitude = 1 mT, and microwave power = 0.2 mW.
Figure 3. CW X-Band EPR of the M9-Fe forms of Bh and Ca YqeK. (A) High-spin region of the CW X-Band EPR spectra recorded on the as-isolated M9-Fe Bh YqeK (top) and Ca YqeK (bottom), (B) low-spin region of the X-Band EPR spectra of the ascorbate-reduced M9-Fe Bh YqeK as isolated (top) or chelated and reconstituted (bottom), (C) low-spin region of the X-Band EPR spectra of the ascorbate-reduced M9-Fe Ca YqeK as isolated (top) or chelated and reconstituted (bottom). (D) CW EPR spectra of the M9-Fe Bh YqeK incubated with a series of (d)/NTPs. The asterisk denotes a small amount of paramagnetic MnII. Experimental conditions: temperature = 10 K, microwave frequency = 9.36 GHz, modulation amplitude = 1 mT, and microwave power = 0.2 mW.
Catalysts 14 00652 g003
Catalysts 14 00652 g004
Figure 4. 57Fe Mössbauer spectra of the as-isolated (top), ascorbate-reduced (middle), and dithionite-reduced (bottom) M9-Fe Bh YqeK. The black vertical lines correspond to the experimental spectra. (top) The green lines correspond to total fits, and the solid black, dotted, and solid gray lines correspond to the constituent doublets. (middle) The green lines correspond to the total simulation of the FeIII–FeII subspectrum (60% of the total Fe in the sample), and the solid blue and solid orange lines correspond to the simulations of the FeII and FeIII subsites of the mixed-valent site. (bottom) The green lines correspond to total fits, and the solid black, dotted lines correspond to the constituent doublets. The spectra were recorded at 4.2 K and in the presence of a small magnetic field of 53 mT applied parallel to the direction of the γ-beam.
Figure 4. 57Fe Mössbauer spectra of the as-isolated (top), ascorbate-reduced (middle), and dithionite-reduced (bottom) M9-Fe Bh YqeK. The black vertical lines correspond to the experimental spectra. (top) The green lines correspond to total fits, and the solid black, dotted, and solid gray lines correspond to the constituent doublets. (middle) The green lines correspond to the total simulation of the FeIII–FeII subspectrum (60% of the total Fe in the sample), and the solid blue and solid orange lines correspond to the simulations of the FeII and FeIII subsites of the mixed-valent site. (bottom) The green lines correspond to total fits, and the solid black, dotted lines correspond to the constituent doublets. The spectra were recorded at 4.2 K and in the presence of a small magnetic field of 53 mT applied parallel to the direction of the γ-beam.
Catalysts 14 00652 g004
Catalysts 14 00652 g005
Figure 5. Ap4A hydrolysis by M9-Fe Ca and Bh YqeK under different redox conditions. (A) HPLC chromatogram of M9-Fe Ca YqeK and (B) M9-Fe Bh YqeK under ambient conditions (“as is”, black), treatment with 1 mM H2O2 (orange trace), treatment with 2 mM sodium dithionite (green trace). The product ADP elutes at ~16 min, while the substrate Ap4A elutes at ~22 min. These experiments are an average of 8 replicates, and the average representative chromatograms are shown. (C) Product formation over time for the M9-Fe Ca YqeK and (D) M9-Fe Bh YqeK reduced with 1 μM ascorbate (blue) or 2 mM sodium dithionite (green). Other experimental conditions: [YqeK] = 25 nM, [Ap4A] = 300 μM. The results are an average of ten independent replicates.
Figure 5. Ap4A hydrolysis by M9-Fe Ca and Bh YqeK under different redox conditions. (A) HPLC chromatogram of M9-Fe Ca YqeK and (B) M9-Fe Bh YqeK under ambient conditions (“as is”, black), treatment with 1 mM H2O2 (orange trace), treatment with 2 mM sodium dithionite (green trace). The product ADP elutes at ~16 min, while the substrate Ap4A elutes at ~22 min. These experiments are an average of 8 replicates, and the average representative chromatograms are shown. (C) Product formation over time for the M9-Fe Ca YqeK and (D) M9-Fe Bh YqeK reduced with 1 μM ascorbate (blue) or 2 mM sodium dithionite (green). Other experimental conditions: [YqeK] = 25 nM, [Ap4A] = 300 μM. The results are an average of ten independent replicates.
Catalysts 14 00652 g005
Catalysts 14 00652 g006
Figure 6. Redox titrations of M9-Fe Ca and Bh YqeK. (A) EPR spectra of M9-Fe Bh YqeK as a function of measured potential. (B) EPR spectra of M9-Fe Ca YqeK as a function of measured potential. (C) Plotted EPR intensities against measured potential for Bh YqeK. The data were fitted to the Nernst equation. Intensities of Component 1 (blue circles) and Component 2 (red triangles) were plotted and fitted independently. (D) Plotted EPR intensities against measured potential for Ca YqeK. Experimental conditions: temperature = 10 K, microwave frequency = 9.36 GHz, modulation amplitude = 1 mT, and microwave power = 0.2 mW.
Figure 6. Redox titrations of M9-Fe Ca and Bh YqeK. (A) EPR spectra of M9-Fe Bh YqeK as a function of measured potential. (B) EPR spectra of M9-Fe Ca YqeK as a function of measured potential. (C) Plotted EPR intensities against measured potential for Bh YqeK. The data were fitted to the Nernst equation. Intensities of Component 1 (blue circles) and Component 2 (red triangles) were plotted and fitted independently. (D) Plotted EPR intensities against measured potential for Ca YqeK. Experimental conditions: temperature = 10 K, microwave frequency = 9.36 GHz, modulation amplitude = 1 mT, and microwave power = 0.2 mW.
Catalysts 14 00652 g006
Catalysts 14 00652 g007
Figure 7. Role of conserved positively charged residues in the YqeK active site. (A) Crystal structures of Bh YqeK (top, pdb: 2O08) and Ca YqeK (bottom, pdb: 3CCG) labeled with key conserved residues. (B) EPR spectra of wild type (WT) Bh YqeK and the alanine substituted Bh YqeK variants R18A (green), R133A (blue), and K54A (yellow). (C) HPLC chromatogram for WT and the three variants monitoring Ap4A hydrolysis with 15 nM protein concentration. (D) HPLC chromatogram for WT and the three variants monitoring Ap4A hydrolysis with 1 μM protein concentration. Experimental conditions for EPR: [YqeK] = 500 μM, [ascorbate] = 1 mM, temperature = 10 K, microwave frequency = 9.36 GHz, modulation amplitude = 1 mT, and microwave power = 0.2 mW.
Figure 7. Role of conserved positively charged residues in the YqeK active site. (A) Crystal structures of Bh YqeK (top, pdb: 2O08) and Ca YqeK (bottom, pdb: 3CCG) labeled with key conserved residues. (B) EPR spectra of wild type (WT) Bh YqeK and the alanine substituted Bh YqeK variants R18A (green), R133A (blue), and K54A (yellow). (C) HPLC chromatogram for WT and the three variants monitoring Ap4A hydrolysis with 15 nM protein concentration. (D) HPLC chromatogram for WT and the three variants monitoring Ap4A hydrolysis with 1 μM protein concentration. Experimental conditions for EPR: [YqeK] = 500 μM, [ascorbate] = 1 mM, temperature = 10 K, microwave frequency = 9.36 GHz, modulation amplitude = 1 mT, and microwave power = 0.2 mW.
Catalysts 14 00652 g007
Table 1. Metal content of Bh and Ca YqeK obtained from ICP-AES. The results are the averages of three independent protein preparations. LB corresponds to cells grown in LB media and M9-Fe corresponds to cells grown in minimal media with Fe supplementation.
Table 1. Metal content of Bh and Ca YqeK obtained from ICP-AES. The results are the averages of three independent protein preparations. LB corresponds to cells grown in LB media and M9-Fe corresponds to cells grown in minimal media with Fe supplementation.
Metal (mol/mol Protein Monomer)
FeCoMnNiZn
Bh YqeKLB0.58 ± 0.150.01 ± 0.010.06 ± 0.010.04 ± 0.020.40 ± 0.17
M9-Fe1.31 ± 0.320.01 ± 0.010.03 ± 0.040.05 ± 0.050.10 ± 0.08
Ca YqeKLB0.49 ± 0.230.03 ± 0.040.05 ± 0.020.05 ± 0.02 0.13 ± 0.03
M9-Fe1.5 ± 0.060.01 ± 0.010.01 ± 0.010.06 ± 0.040.07 ± 0.06
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Ueda, C.; Chin, N.; Yang, Q.; Pan, L.; Ponniah, R.; Pandelia, M.-E. Diverse YqeK Diadenosine Tetraphosphate Hydrolases Control Biofilm Formation in an Iron-Dependent Manner. Catalysts 2024, 14, 652. https://doi.org/10.3390/catal14090652

AMA Style

Ueda C, Chin N, Yang Q, Pan L, Ponniah R, Pandelia M-E. Diverse YqeK Diadenosine Tetraphosphate Hydrolases Control Biofilm Formation in an Iron-Dependent Manner. Catalysts. 2024; 14(9):652. https://doi.org/10.3390/catal14090652

Chicago/Turabian Style

Ueda, Chie, Natalie Chin, Qianyi Yang, Luying Pan, Rheann Ponniah, and Maria-Eirini Pandelia. 2024. "Diverse YqeK Diadenosine Tetraphosphate Hydrolases Control Biofilm Formation in an Iron-Dependent Manner" Catalysts 14, no. 9: 652. https://doi.org/10.3390/catal14090652

APA Style

Ueda, C., Chin, N., Yang, Q., Pan, L., Ponniah, R., & Pandelia, M.-E. (2024). Diverse YqeK Diadenosine Tetraphosphate Hydrolases Control Biofilm Formation in an Iron-Dependent Manner. Catalysts, 14(9), 652. https://doi.org/10.3390/catal14090652

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop