- Email: [email protected]
Delineating the reaction mechanism of reductase domains of Nonribosomal Peptide Synthetases from mycobacteria
Journal of Structural Biology 187 (2014) 207–214
Contents lists available at ScienceDirect
Journal of Structural Biology journal homepage: www.elsevier.com/locate/yjsbi
Delineating the reaction mechanism of reductase domains of Nonribosomal Peptide Synthetases from mycobacteria Asfarul S. Haque a,1, Ketan D. Patel a,1, Mandar V. Deshmukh a, Arush Chhabra b,2, Rajesh S. Gokhale b,c,d, Rajan Sankaranarayanan a,⇑ a
CSIR-Centre for Cellular and Molecular Biology, Hyderabad 500007, India National Institute of Immunology, New Delhi 110067, India CSIR-Institute of Genomics and Integrative Biology, Delhi 110007, India d Jawaharlal Nehru Centre for Advanced Scientific Research, Bangalore 560064, India b c
a r t i c l e
i n f o
Article history: Received 29 April 2014 Received in revised form 15 July 2014 Accepted 30 July 2014 Available online 7 August 2014 Keywords: Short-chain Dehydrogenase/Reductase (SDR) Sequential order bi–bi reaction mechanism Reductase domain NRPS Cation–p interaction
a b s t r a c t Substrate binding to enzymes often follows a precise order where catalysis is accomplished through programmed conformational changes. Short-chain dehydrogenase/reductase (SDR) enzymes follow sequential order ‘bi–bi’ reaction kinetics. The mechanistic study of a SDR homolog, reductase (R) domain, from multifunctional enzymes, e.g. Nonribosomal Peptide Synthetases (NRPSs) and Polyketide Synthases (PKSs) has revealed that it reductively releases 40 -phosphopantetheinyl arm-tethered peptidyl product. We report that the R-domains of NRPSs from Mycobacterium tuberculosis (RNRP) and Mycobacterium smegmatis (RGPL) do not strictly adhere to the obligatory mode of catalysis performed by SDRs, but instead can carry out reductive catalysis of substrate following random bi–bi reaction mechanism as deciphered by NMR and SAXS studies. The crucial conformational change associated with NADPH binding necessary to achieve catalytically competent conformation is also delineated by SAXS studies. Using ITC, we have demonstrated that mutation of catalytic tyrosine to phenylalanine in R-domains results in 3–4-fold decrease in affinity for NADPH and attribute this phenomenon to loss of the noncovalent cation–p interactions present between the tyrosine and nicotinamide ring. We propose that the adaptation to an alternative theme of bi–bi catalytic mechanism enables the R-domains to process the substrates transferred by upstream domains and maintain assembly-line enzymology. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction In most enzyme catalyzed reactions, conformational changes associated with substrate binding are necessary for product formation. A precise orientation of catalytic groups is achieved by cofactor binding and is often crucial to facilitate substrate binding as seen in cofactor dependent enzymatic reactions. In sequential order bi–bi reaction mechanism, the cofactor binds first followed by a substrate to form a product and a coproduct (Chang et al., 2007; Rudnick et al., 1991) while in random bi-bi reaction mechanism, the sequential binding is not strictly adhered to
⇑ Corresponding author. Address: Structural Biology Laboratory, CSIR-Centre for Cellular and Molecular Biology, Uppal Road, Hyderabad 500 007, India. Fax: +91 40 27160591. E-mail address: [email protected] (R. Sankaranarayanan). 1 Authors have contributed equally. 2 Present address: GlaxoSmithKline Consumer HealthCare Research & Development Centre, Gurgaon 122001, India. http://dx.doi.org/10.1016/j.jsb.2014.07.008 1047-8477/Ó 2014 Elsevier Inc. All rights reserved.
(Wang and Wu, 2007). In Ping Pong bi–bi reaction one substrate binds to the enzyme, transfers a group to it, leaves the enzyme followed by the second substrate binding to it and getting converted into product (Bretschneider et al., 2012). Short-chain dehydrogenases/reductases (SDRs) form a large, functionally heterogeneous family of enzymes and typically share residue identities between 15% and 30% (Jornvall et al., 1995). Despite their low amino acid sequence identities, the 3D structures share highly similar a/b folding patterns with a central b-sheet typical of the Rossmann fold for nucleotide binding. SDR family consists of enzymes such as oxidoreductases, lyases, isomerases etc. The majority of the SDRs are single-domain enzymes in which the cofactor binds in the N-terminal region and the substrate binds to the C-terminal region. The substrate-binding C-terminal region is highly variable amongst different SDR enzymes. These enzymes often follow sequential order bi–bi reaction mechanism in which the cofactor, NAD(P)H, first binds to the cofactor-binding pocket of the enzyme followed by binding of the substrate in the C-terminal region, leading to product and coproduct formation
208
A.S. Haque et al. / Journal of Structural Biology 187 (2014) 207–214
and subsequently their release (Giraud and Naismith, 2000; Kavanagh et al., 2008). The majority of the SDR enzymes are standalone enzymes that are biologically active as either a dimer or a tetramer. In the past decade, bioinformatic, biochemical and structural characterization has revealed that reductase (R) domains, which belong to SDR family of enzymes, are a constituent of multimodular enzymes, e.g. Nonribosomal Peptide Synthetase (NRPS), Polyketide Synthase (PKS), and are present at their C-terminus (Chhabra et al., 2012; Keating and Walsh, 1999). The catalytic role played by the R-domain as part of these multifunctional enzymes is to reductively release the product covalently linked to the 40 phosphopantetheine arm of the enzyme as either an aldehyde or an alcohol in an NAD(P)H-dependent manner (Fig. 1) (Chhabra et al., 2012; Read and Walsh, 2007). In mycobacterial species, the NRPSs having reductase domains are involved in the synthesis of mycobactin, secreted iron-chelating siderophores and glycopeptidolipid core which are implicated in virulence and sliding motility etc. (Gokhale et al., 2007; Martinez et al., 1999; Raman et al., 2006; Vats et al., 2012). The R-domain has a Rossmann fold at its N-terminal region which is typical of SDR enzymes and carries out oxidoreductive catalysis in an NADPH-dependent manner. Previous studies have revealed that RNRP and RGPL have a large substrate-binding pocket that contributes to their substrate promiscuity and consequently their ability to reduce acyl-CoAs, e.g. lauroyl-CoA, palmitoyl-CoA etc. in vitro (Chhabra et al., 2012). The substrate promiscuity demonstrated by RNRP and RGPL led us to carry out further studies using lauroyl-CoA as their substrate in the current study. In the present study, using Nuclear Magnetic Resonance (NMR) and Small-angle X-ray Scattering (SAXS) analyses we show that the offloading R-domain of multifunctional NRPS (Rv0101) from Mycobacterium tuberculosis (RNRP) and the Rdomain of mycobacterial peptide synthetase (mps2) from Mycobacterium smegmatis (RGPL) follow random bi–bi reaction mechanism unlike the majority of the SDR standalone enzymes that follow sequential order bi–bi mechanism, to carry out catalysis. Typically, SDRs have Tyr as a strictly invariant catalytic residue apart from other well-conserved residues, Ser/Thr-LysAsn (Oppermann et al., 2003). Here, using ITC, we show catalytic Tyr also plays an essential role in high affinity binding of NADPH and attribute this phenomenon to a noncovalent cation–p interaction present between tyrosine and nicotinamide.
2. Materials and methods 2.1. Mutagenesis RNRP was cloned from 6166 bp to 7676 bp of M. tuberculosis gene Nrp, and RGPL was cloned from 6301 bp to 7539 bp of M. smegmatis gene mps2 (Chhabra et al., 2012). Cloning was achieved using pET-28(b) vector (Novagen) as N-terminal 6xHis expressing proteins. Site-directed mutagenesis of tyrosine to phenylalanine residue in RNRP and RGPL genes was performed using QuickChange site-directed mutagenesis kit (Stratagene). Primers were as follows (mutated nucleotides are underlined): RGPL: Y227F 50 -GCCAACGG CTTCGGGAATTCCAAGTGGGCAGGCG-30 and 50 -CGCCTGCCCACTTGGAATTCCCGAAGCCGTTGGC-30 ; RNRP: Y228F 50 -GCTGGCGGCTTCGG CACCAGCAAGTGGGCCGGT-30 and 50 -ACCGGCCCACTTGCTGGTGCC GAAGCCGCCAGC-30 . Confirmation of the mutagenesis was done by DNA sequencing. 2.2. Protein expression and purification R-domains were expressed as N-terminal 6xHis-tag proteins in Escherichia coli strain BL21 (DE3). Cultures were grown in sterile Luria–Bertani medium containing 50 lg/ml kanamycin at 37 °C with 200 rpm shaking conditions. Cells were induced with 0.5 mM IPTG at 0.6 O.D. at 600 nm and grown further at 22 °C for 12–14 h. Cells were harvested and 5 g of pellet was resuspended in 100 ml of lysis buffer containing 50 mM Tris–HCl pH 8.0 and 150 mM NaCl. Lysis of the cells was done by sonication at an amplitude of 22 for 30 min with 3 s on/off pulses. Lysate was mixed with 1% streptomycin sulfate for 30 min at 4 °C and centrifuged at 10,000 rpm and supernatant was taken further for purification. All purification and concentration steps were performed at 4 °C. Supernatant was loaded onto Ni-NTA Sepharose column preequilibrated with lysis buffer. The bound protein was washed with 5 column volumes of wash buffer 1 (10 mM imidazole, 50 mM Tris–HCl pH 8.0, 150 mM NaCl). Second wash was performed by 10 column volumes of wash buffer 2 (50 mM Tris–HCl pH 8.0, 500 mM NaCl). Elution was carried out with 10 column volumes of a gradient of 20–250 mM imidazole in lysis buffer. Fractions containing the protein were analyzed by SDS–PAGE, pooled together and concentrated up to 10 mg/ml using 30-kDa Amicon centrifugal filter devices (Millipore). The concentrated sample was loaded onto Sephacryl S200 gel filtration column pre-equilibrated with buffer 20 mM Tris–HCl pH 8.0, 20 mM NaCl and then eluted. Fractions containing proteins were concentrated using 30-kDa Amicon centrifugal filter devices (Millipore) and used for Isothermal Titration Calorimetry (ITC) and Small-angle X-ray Scattering (SAXS) experiments. For Nuclear Magnetic Resonance (NMR) experiments, uniformly 15 N-labeled protein samples were prepared by expressing in E. coli grown in modified M9 minimal medium containing 15NH4Cl as the sole nitrogen source. Purification procedures for 15N-labeled proteins were identical as for the unlabeled proteins except for buffer having Tris–HCl at a pH of 7. 2.3. Isothermal Titration Calorimetry (ITC) experiments
Fig.1. Two domain architecture of R-domain: the crystal structure of RNRP (PDB: 4DQV) solved previously reveals a two domain arrangement of the R-domain with NADPH-binding domain (Blue) and substrate-binding domain (brown) with cofactor (NADPH) and substrate (Lauroyl CoA) modeled here. It performs reductive release of the substrate in the form of an alcohol. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Titration experiments for RNRP and RGPL and their corresponding tyr-phe mutants were performed at 25 °C with VP-ITC calorimeter (Microcal, Northampton, MA). Samples containing 2 mM NADPH were prepared in buffer 20 mM Tris–HCl pH 8.0, 20 mM NaCl and loaded onto the syringe. The reaction cell contained 1.6 ml of 96 lM R-domain with a buffer composition of 20 mM Tris–HCl pH 8.0, 20 mM NaCl. Titrations performed with 20 injections of 8 ll NADPH solution for 5 s each at 2 min intervals using
A.S. Haque et al. / Journal of Structural Biology 187 (2014) 207–214
rotating-stirrer syringe. The titration data were analyzed using ORIGIN data analysis software 5.0 (Microcal Software, Northampton, MA). 2.4. Nuclear Magnetic Resonance (NMR) The NMR experiments were performed at 25 °C on a Bruker 600 MHz spectrometer equipped with a triple resonance cryogenically cooled probe by using protein solutions that contained 20 mM Tris–HCl pH 7.0, 20 mM NaCl. 2D 15N–1H Transverse Relaxation-Optimized Spectroscopy (TROSY-HSQC) spectra were recorded with uniformly 15N-labeled RNRP and RGPL and their mutants (95% protein solution/5% D2O, pH 7.0) with 1024 256 data points. The titrations of the R-domains and their mutants with either NADPH (cofactor) or lauroyl CoA (substrate) was performed with a molar ratio of 1:5 and 1:10 which amounted to complete saturation of the enzyme with these ligands until no further perturbations in chemical shifts could be observed. The 2D data was processed using Topspin (v 2.1) and analyzed with Sparky3 (Goddard and Kneller, 2008) on a remote workstation. 2.5. Small Angle X-ray Scattering (SAXS) data collection and processing Precise concentration of the protein samples at 280 nm was determined using NanoDrop 1000 spectrophotometer. The monodispersity of each sample was verified before SAXS data collection using dynamic light scattering technique. Data collection for RNRP and RGPL and their mutants were performed at a concentration of 10 mg/ml and 4 mg/ml, respectively, by addition of 1:10 ratio with (1) NADPH, (2) lauroyl-CoA, (3) NADPH followed by lauroyl-CoA and (4) lauroyl-CoA followed by NADPH with an exposure time of 1 h at 297 K. Corresponding buffers for each of these samples were exposed for the same duration, at the same temperature and data were collected. The sample size in each of these cases was 35 ll. The equipment used for SAXS data collection was S3-micropix SAXS system from Hecus XRS. The beam delivery system was from Xenocs mounted on a sealed Cu tube X-ray generator. The azimuthal averaging of raw data was done using FIT2D program. For further data analyses, programs from the ATSAS 2.5.2 software suite was used (Konarev et al., 2006).
209
Likewise, the RGPL of mps2 from M. smegmatis yielded similar profile to that of RNRP under identical experimental conditions (data not shown). Thus, NMR-based titration studies suggest that the R-domains which are a constituent of megasynthetases follow random ‘bi–bi’ mechanism of action. Interestingly, the resultant chemical shift perturbations in both cases are similar, suggesting that although lauroyl-CoA can bind to the R-domains independently of the presence of NADPH, the final catalytically competent conformation is attained only when both are bound to the enzyme. Previous studies have suggested that conformational changes brought about by binding of NADPH to RNRP and RGPL from NRPSs are essential for these enzymes to carry out nonprocessive reductive catalysis (Chhabra et al., 2012). In the wake of deviation from sequential order ‘bi–bi’ reaction mechanism as shown using 2D 15 N–1H TROSY-HSQC binding studies and substrate promiscuity demonstrated by RNRP and RGPL, we set out to find the role played by NADPH and lauroyl-CoA in driving the enzyme towards catalytically feasible conformation change, if any, associated with them. To this end, SAXS studies were performed with both RNRP and RGPL. 3.2. NADPH-mediated conformational change is a key step to attain catalytically competent conformation SAXS analysis revealed that there was a decrease in both Rg and Dmax upon binding of NADPH to the enzyme at first and further decrease in these parameters was brought about upon binding of lauroyl-CoA due to enhanced structural rearrangements. However, there was negligible change in Rg and Dmax upon binding of lauroylCoA in absence of NADPH (Fig. 3a and b). The data strongly suggest that a large conformational change is associated with NADPH binding while lauroyl-CoA binding brings subtle structural changes when it occupies the large substrate-binding pocket. Similarly, in the case of RGPL, the magnitude of domain motion is significantly large when NADPH binds to it as compared with the binding of lauroyl-CoA. Therefore, SAXS studies clearly demonstrate that the binding of NADPH plays a key role in the enzymes’ attaining a catalyticallycompetent conformation. 3.3. Structural analyses of SDR family of oxidoreductases in complex with NADPH reveal noncovalent interactions between catalytic tyrosine and nicotinamide of NADPH
3. Results and discussion 3.1. Chemical shift perturbations obtained by 15N–1H TROSY-HSQC NMR experiments are suggestive of random bi–bi reaction mechanism To ascertain the binding of lauroyl-CoA to RNRP and RGPL, both in absence and in presence of NADPH, NMR-based binding studies were performed. The 15N–1H TROSY-HSQC NMR experiments revealed significant changes in the chemical shifts of resonances of RNRP upon titration with NADPH. Following saturation of RNRP with NADPH, titration with lauroyl-CoA led to additional and unidirectional chemical shift perturbations in the same residues (Fig. 2a). Interestingly, when the titration order was reversed, same residues in the RNRP experienced weak chemical shift perturbations in the presence of lauroyl-CoA alone. These perturbations enhanced significantly upon addition of NADPH (Fig. 2b). Similar chemical shift perturbations were obtained when identical experiments were carried out with the Y228F mutant of RNRP, suggesting a similar mode of binding of NADPH and lauroyl-CoA (Fig. 2c and d). The 15N–1H TROSY-HSQC spectra acquired in both cases suggest that the presence of NADPH is not necessary for the substrate to bind to the enzyme as often suggested for SDR family of proteins which follow sequential order ‘bi–bi’ reaction mechanism.
The driving force for RNRP and RGPL to acquire catalyticallycompetent conformation in the presence of NADPH prompted structural analyses of known crystal structures of SDR family of enzymes bound to NAD(P)H. Protein Data Bank was used to acquire the known crystal structures of SDR family of proteins and cutoff of resolutions higher than 2 Å was used to short-list the structures. PDBeMOTIF was used to ascertain the interactions between nicotinamide and tyrosine. A total of 121 structures were segregated as a result of this, out of which 84 having NADPH or NAD in similar orientation were used for final structural analysis. A close probing of these structures revealed that the catalytic tyrosine is present in close proximity to the nicotinamide ring with the –OH moiety of the tyrosine residue facing it (Fig. 4). An average distance of 3.97 (±0.47) Å between the centre of the nicotinamide ring and –OH of the tyrosine is highly indicative of an interaction between them and possibly this tyrosine would play a role in the binding and stabilization of NADPH. In a previous study, NADPH-bound crystal structure of b-ketoacyl-acyl carrier protein reductase (FabG) having a single point mutation of catalytic tyrosine to phenylalanine revealed that the electron density for the nicotinamide moiety was conspicuous by its absence which was suggestive of impaired NADPH binding to the enzyme. It was subsequently supported by biochemical studies as determined
210
A.S. Haque et al. / Journal of Structural Biology 187 (2014) 207–214
Fig.2. Zoomed in illustrations from 2D 15N–1H TROSY-HSQC of RNRP representing chemical shift perturbations upon titration with NADPH and L.CoA. (a) RNRP (red) resonances experience perturbations in the chemical shifts upon titration with NADPH (green) followed by L.CoA (blue). (b) Chemical shift perturbations as observed on reversing the order of titration to L.CoA (blue) followed by NADPH (green) with RNRP (red). (c) and (d) correspond to titrations as followed in (a) and (b), respectively, with Y228F mutant of RNRP. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
by intrinsic NADPH fluorescence (Chhabra et al., 2012; Price et al., 2004). These findings suggest that the catalytic tyrosine not only acts as proton donor during catalysis, but it is also required for high-affinity NADPH binding to the enzyme. However, the actual impact of the mutation on the binding affinity of this enzyme for NADPH was not known. We set out to quantify the binding affinity of R-domains for NADPH using ITC and the contribution of tyrosine towards this binding. 3.4. Impaired binding of NADPH to Y228F mutant form of RNRP and Y227F of RGPL ITC experiments were performed to assess the binding affinity of NADPH with both the wild type RNRP and RGPL (Fig. 5a and c) and their corresponding tyr-phe mutants (Fig. 5b and d). In each of the cases, it was found that there is a 3-fold and a 4-fold decrease in the binding affinities of the tyr-phe mutant R-domains towards NADPH when compared with their wild type enzyme (Fig. 5e). Interestingly, in the case of Y227F mutant of RGPL the
reaction was transformed from exothermic to endothermic by mutation of the catalytic tyrosine alone. In the case of wild type and mutant RNRP, the ITC experiments revealed that the binding of the NADPH was enthalpy driven, though there was a decreased binding affinity towards NADPH. The binding of NADPH to wild type RGPL was also enthalpy driven but its mutant binds to NADPH in an entropy-driven manner. Possibly, the hydrophobic environment provided by the mutation of tyrosine to phenylalanine in Y227F RGPL was further compounded by hydrophobic residues present in the neighboring region where nicotinamide binds. We performed the structure-based sequence alignment of the amino acid residues present in the loop that is involved in NADPH binding in RNRP and RGPL that shows VNA-FPY and VNSVLPY, respectively, in case of M. tuberculosis and M. smegmatis. The variations in amino acid residues including extra valine could possibly provide hydrophobic environment at the nicotinamide-binding site resulting in NADPH binding in entropy-driven manner. The ITC experiment along with the structural analysis suggests the presence of an interaction between the catalytic tyrosine and
A.S. Haque et al. / Journal of Structural Biology 187 (2014) 207–214
211
Fig.3. The SAXS data with Rg and Dmax changes upon binding of NADPH and L.CoA. (a) The table suggests that there is a subtle change in Rg or Dmax upon addition of L.CoA alone while the same is not true when it is added in the presence of NADPH. (b) The pair-distance distribution curve of different titrations shows that virtually there is no domain motion associated with only L.CoA titration against RNRP.
3.5. Cation–p interaction between tyrosine and nicotinamide of NADPH
Fig.4. The superimposed catalytic tyrosines and NADPH from known crystal structures is suggestive of the presence of a noncovalent interaction between them.
the nicotinamide ring of NADPH which is crucial for the enzyme to obtain catalytically competent conformation.
The noncovalent interactions of various types, e.g. salt bridges, hydrogen bonds, van der Waals forces etc., are crucial for catalytic activity and structural integrity of biological macromolecules. One of the seldom-reported noncovalent interactions in biological macromolecules is cation–p interaction which takes place between aromatic rings and positively charged amino acid residues (Gallivan and Dougherty, 1999; Ma and Dougherty, 1997). The magnitude of this interaction varies with the nature of the cation and the aromatic system. For example, Li+, Na+, K+, Rb+ interact in decreasing order with the aromatic ring of benzene (Ma and Dougherty, 1997). Apart from acting as a binding site for the cations, these aromatic rings can compete with the highly favorable solvation of an ion provided by an aqueous medium. Unlike the usual cation-binding structures like Asp and Glu, in the case of cation–p interactions these are aromatic amino acids Trp, Tyr and Phe, while the side chains of positively charged Arg and Lys act as cation. The cations interact with the delocalized p electron density of the aromatic ring of Trp, Tyr and Phe. In the present study, it has been demonstrated, using ITC, that there is a 3–4-fold decrease in the binding affinity of NADPH upon mutation of highly conserved catalytic residue Tyr to Phe in R-domains of multimodular NRPSs in mycobacterial species. Here, we suggest that either the cation–p or the dipole–p interaction plays a role in the binding of nicotinamide moiety of NADPH to these enzymes wherein the Tyr performs the function of a cation while the nicotinamide ring
212
A.S. Haque et al. / Journal of Structural Biology 187 (2014) 207–214
Fig.5. Impaired NADPH binding upon mutation of catalytic Tyr to Phe. (a–d) Binding study of wild type and tyr-phe mutant R-domains with NADPH (e) 3–4-fold decrease in binding affinities were found upon mutation of catalytic tyrosine to phenylalanine.
A.S. Haque et al. / Journal of Structural Biology 187 (2014) 207–214
213
Fig.6. The schematic representation of the proposed random bi–bi reaction performed by the mycobacterial R-domains that act as a constituent of the NRPSs. The representative value of Rg as observed from SAXS experiments performed with RNRP.
acts as a p system (Dougherty, 1996; Gupta and Durani, 2011). The study shows that the Tyr plays a crucial role in NADPH binding apart from acting as a proton donor during catalysis in tyrosinedependent oxidoreductases from SDR family of proteins.
4. Conclusion The SDR family of enzymes is often known to follow sequential order bi–bi reaction mechanism. These enzymes exist in either dimeric or tetrameric form and carry out catalysis in a standalone manner. In the present study, the offloading R-domains from mycobacterial species that are a part of multifunctional enzymes, e.g. NRPS and PKS, and belong to SDR family are shown to perform random bi–bi reaction mechanism (Fig. 6). In most of the NRPSs, the offloading of the substrate is performed by a dedicated thioesterase (TE) domain present at their C-terminus (Du and Lou, 2010). It hydrolyses the 40 -phosphopantetheine arm-tethered substrate brought to it by the thiolation (T) domain to release the product, which is attached posttranslationally to the T-domain (Lambalot et al., 1996). The cellular pool contains copious amounts of not only free CoA but also various acyl-CoA derivatives e.g. acetyl-CoA which can get attached to the thiolation domain and block the thio-template based assembly line enzymology performed by NRPSs (Quadri et al., 1998). The cell expresses TEII enzyme that removes the spuriously attached acetyl-CoA and resumes the enzymatic reactions (Schwarzer et al., 2002; Yeh et al., 2004). The R-domains of the NRPSs from mycobacterial species have one of the largest C-terminal substrate-binding domains spanning approximately 150 amino acids with unique hydrophobic pocket to accommodate large acyl-CoAs and there are possibilities that acyl-CoAs may occupy the substrate-binding pocket, thus blocking the reactions. The R-domains, by virtue of random bi-bi reactions, are able to release the unwanted acetyl-CoA from their substrate-binding pocket and resume the assembly line enzymology. The decrease in binding affinity of NADPH in tyr-phe mutant of R-domains and structural analysis reveal that there is indeed an interaction between the catalytic tyrosine of R-domains and the nicotinamide ring of NADPH suggestive of cation–p interaction.
We propose that as the products of NRPSs are crucial for the survival and proliferation of these pathogens, and following sequential order reaction might block thiotemplate-based assembly line enzymology which would be detrimental to their survival. The current study provides insights regarding the alternative form of a reaction mechanism obtained by R-domain as a constituent the multifunctional enzymes e.g. NRPSs and PKSs where this domain performs the task of offloading the ‘product-in-waiting’. This acquired catalytic competency by virtue of random bi–bi reaction mechanism provides leverage to the multifunctional enzymes to process the bound substrate in a cofactor-dependent manner irrespective of their order of binding to the enzyme that is crucial to maintain thiotemplate-based assembly line enzymology.
References Bretschneider, T., Zocher, G., Unger, M., Scherlach, K., Stehle, T., Hertweck, C., 2012. A ketosynthase homolog uses malonyl units to form esters in cervimycin biosynthesis. Nat. Chem. Biol. 8, 154–161. Chang, Y.H., Chuang, L.Y., Hwang, C.C., 2007. Mechanism of proton transfer in the 3alpha-hydroxysteroid dehydrogenase/carbonyl reductase from Comamonas testosteroni. J. Biol. Chem. 282, 34306–34314. Chhabra, A., Haque, A.S., Pal, R.K., Goyal, A., Rai, R., Joshi, S., Panjikar, S., Pasha, S., Sankaranarayanan, R., Gokhale, R.S., 2012. Nonprocessive [2 + 2]e off-loading reductase domains from mycobacterial nonribosomal peptide synthetases. Proc. Natl. Acad. Sci. U.S.A. 109, 5681–5686. Dougherty, D.A., 1996. Cation–pi interactions in chemistry and biology: a new view of benzene, Phe, Tyr, and Trp. Science 271, 163–168. Du, L., Lou, L., 2010. PKS and NRPS release mechanisms. Nat. Prod. Rep. 27, 255–278. Gallivan, J.P., Dougherty, D.A., 1999. Cation–pi interactions in structural biology. Proc. Natl. Acad. Sci. U.S.A. 96, 9459–9464. Giraud, M.F., Naismith, J.H., 2000. The rhamnose pathway. Curr. Opin. Struct. Biol. 10, 687–696. Goddard, T.D., Kneller, D.G., 2008. SPARKY 3. University of California, San Francisco, http://www.cgl.ucsf.edu/home/sparky. Gokhale, R.S., Saxena, P., Chopra, T., Mohanty, D., 2007. Versatile polyketide enzymatic machinery for the biosynthesis of complex mycobacterial lipids. Nat. Prod. Rep. 24, 267–277. Gupta, P., Durani, S., 2011. Aromatic interactions at atom-to-atom contact and just beyond: a case study of protein interactions of NAD(+)/NADP(+). Int. J. Biol. Macromol. 49, 999–1006. Jornvall, H., Persson, B., Krook, M., Atrian, S., Gonzalez-Duarte, R., Jeffery, J., Ghosh, D., 1995. Short-chain dehydrogenases/reductases (SDR). Biochemistry 34, 6003–6013. Kavanagh, K.L., Jornvall, H., Persson, B., Oppermann, U., 2008. Medium- and shortchain dehydrogenase/reductase gene and protein families: the SDR
214
A.S. Haque et al. / Journal of Structural Biology 187 (2014) 207–214
superfamily: functional and structural diversity within a family of metabolic and regulatory enzymes. Cell. Mol. Life Sci. 65, 3895–3906. Keating, T.A., Walsh, C.T., 1999. Initiation, elongation, and termination strategies in polyketide and polypeptide antibiotic biosynthesis. Curr. Opin. Chem. Biol. 3, 598–606. Konarev, P.V., Volkov, V.V., Sokolova, A.V., Koch, M.H.J., Svergun, D.I., 2006. ATSAS 2.1, a program package for Small-angle scattering data analysis. J. Appl. Crystallogr. 39, 277–286. Lambalot, R.H., Gehring, A.M., Flugel, R.S., Zuber, P., LaCelle, M., Marahiel, M.A., Reid, R., Khosla, C., Walsh, C.T., 1996. A new enzyme superfamily – the phosphopantetheinyl transferases. Chem. Biol. 3, 923–936. Ma, J.C., Dougherty, D.A., 1997. The Cation minus sign pi interaction. Chem. Rev. 97, 1303–1324. Martinez, A., Torello, S., Kolter, R., 1999. Sliding motility in mycobacteria. J. Bacteriol. 181, 7331–7338. Oppermann, U., Filling, C., Hult, M., Shafqat, N., Wu, X., Lindh, M., Shafqat, J., Nordling, E., Kallberg, Y., Persson, B., Jornvall, H., 2003. Short-chain dehydrogenases/reductases (SDR): the 2002 update. Chem. Biol. Interact. 143–144, 247–253. Price, A.C., Zhang, Y.M., Rock, C.O., White, S.W., 2004. Cofactor-induced conformational rearrangements establish a catalytically competent active site and a proton relay conduit in FabG. Structure 12, 417–428. Quadri, L.E., Weinreb, P.H., Lei, M., Nakano, M.M., Zuber, P., Walsh, C.T., 1998. Characterization of Sfp, a Bacillus subtilis phosphopantetheinyl transferase for peptidyl carrier protein domains in peptide synthetases. Biochemistry 37, 1585–1595.
Raman, S., Puyang, X., Cheng, T.Y., Young, D.C., Moody, D.B., Husson, R.N., 2006. Mycobacterium tuberculosis SigM positively regulates Esx secreted protein and nonribosomal peptide synthetase genes and down regulates virulenceassociated surface lipid synthesis. J. Bacteriol. 188, 8460–8468. Read, J.A., Walsh, C.T., 2007. The lyngbyatoxin biosynthetic assembly line: chain release by four-electron reduction of a dipeptidyl thioester to the corresponding alcohol. J. Am. Chem. Soc. 129, 15762–15763. Rudnick, D.A., McWherter, C.A., Rocque, W.J., Lennon, P.J., Getman, D.P., Gordon, J.I., 1991. Kinetic and structural evidence for a sequential ordered Bi Bi mechanism of catalysis by Saccharomyces cerevisiae myristoyl-CoA: protein Nmyristoyltransferase. J. Biol. Chem. 266, 9732–9739. Schwarzer, D., Mootz, H.D., Linne, U., Marahiel, M.A., 2002. Regeneration of misprimed nonribosomal peptide synthetases by type II thioesterases. Proc. Natl. Acad. Sci. U.S.A. 99, 14083–14088. Vats, A., Singh, A.K., Mukherjee, R., Chopra, T., Ravindran, M.S., Mohanty, D., Chatterji, D., Reyrat, J.M., Gokhale, R.S., 2012. Retrobiosynthetic approach delineates the biosynthetic pathway and the structure of the acyl chain of mycobacterial glycopeptidolipids. J. Biol. Chem. 287, 30677–30687. Wang, Z.X., Wu, J.W., 2007. The complete pathway for ERK2-catalyzed reaction. Evidence for an iso random Bi Bi mechanism. J. Biol. Chem. 282, 27678–27684. Yeh, E., Kohli, R.M., Bruner, S.D., Walsh, C.T., 2004. Type II thioesterase restores activity of a NRPS module stalled with an aminoacyl-S-enzyme that cannot be elongated. ChemBioChem 5, 1290–1293.
Contents lists available at ScienceDirect
Journal of Structural Biology journal homepage: www.elsevier.com/locate/yjsbi
Delineating the reaction mechanism of reductase domains of Nonribosomal Peptide Synthetases from mycobacteria Asfarul S. Haque a,1, Ketan D. Patel a,1, Mandar V. Deshmukh a, Arush Chhabra b,2, Rajesh S. Gokhale b,c,d, Rajan Sankaranarayanan a,⇑ a
CSIR-Centre for Cellular and Molecular Biology, Hyderabad 500007, India National Institute of Immunology, New Delhi 110067, India CSIR-Institute of Genomics and Integrative Biology, Delhi 110007, India d Jawaharlal Nehru Centre for Advanced Scientific Research, Bangalore 560064, India b c
a r t i c l e
i n f o
Article history: Received 29 April 2014 Received in revised form 15 July 2014 Accepted 30 July 2014 Available online 7 August 2014 Keywords: Short-chain Dehydrogenase/Reductase (SDR) Sequential order bi–bi reaction mechanism Reductase domain NRPS Cation–p interaction
a b s t r a c t Substrate binding to enzymes often follows a precise order where catalysis is accomplished through programmed conformational changes. Short-chain dehydrogenase/reductase (SDR) enzymes follow sequential order ‘bi–bi’ reaction kinetics. The mechanistic study of a SDR homolog, reductase (R) domain, from multifunctional enzymes, e.g. Nonribosomal Peptide Synthetases (NRPSs) and Polyketide Synthases (PKSs) has revealed that it reductively releases 40 -phosphopantetheinyl arm-tethered peptidyl product. We report that the R-domains of NRPSs from Mycobacterium tuberculosis (RNRP) and Mycobacterium smegmatis (RGPL) do not strictly adhere to the obligatory mode of catalysis performed by SDRs, but instead can carry out reductive catalysis of substrate following random bi–bi reaction mechanism as deciphered by NMR and SAXS studies. The crucial conformational change associated with NADPH binding necessary to achieve catalytically competent conformation is also delineated by SAXS studies. Using ITC, we have demonstrated that mutation of catalytic tyrosine to phenylalanine in R-domains results in 3–4-fold decrease in affinity for NADPH and attribute this phenomenon to loss of the noncovalent cation–p interactions present between the tyrosine and nicotinamide ring. We propose that the adaptation to an alternative theme of bi–bi catalytic mechanism enables the R-domains to process the substrates transferred by upstream domains and maintain assembly-line enzymology. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction In most enzyme catalyzed reactions, conformational changes associated with substrate binding are necessary for product formation. A precise orientation of catalytic groups is achieved by cofactor binding and is often crucial to facilitate substrate binding as seen in cofactor dependent enzymatic reactions. In sequential order bi–bi reaction mechanism, the cofactor binds first followed by a substrate to form a product and a coproduct (Chang et al., 2007; Rudnick et al., 1991) while in random bi-bi reaction mechanism, the sequential binding is not strictly adhered to
⇑ Corresponding author. Address: Structural Biology Laboratory, CSIR-Centre for Cellular and Molecular Biology, Uppal Road, Hyderabad 500 007, India. Fax: +91 40 27160591. E-mail address: [email protected] (R. Sankaranarayanan). 1 Authors have contributed equally. 2 Present address: GlaxoSmithKline Consumer HealthCare Research & Development Centre, Gurgaon 122001, India. http://dx.doi.org/10.1016/j.jsb.2014.07.008 1047-8477/Ó 2014 Elsevier Inc. All rights reserved.
(Wang and Wu, 2007). In Ping Pong bi–bi reaction one substrate binds to the enzyme, transfers a group to it, leaves the enzyme followed by the second substrate binding to it and getting converted into product (Bretschneider et al., 2012). Short-chain dehydrogenases/reductases (SDRs) form a large, functionally heterogeneous family of enzymes and typically share residue identities between 15% and 30% (Jornvall et al., 1995). Despite their low amino acid sequence identities, the 3D structures share highly similar a/b folding patterns with a central b-sheet typical of the Rossmann fold for nucleotide binding. SDR family consists of enzymes such as oxidoreductases, lyases, isomerases etc. The majority of the SDRs are single-domain enzymes in which the cofactor binds in the N-terminal region and the substrate binds to the C-terminal region. The substrate-binding C-terminal region is highly variable amongst different SDR enzymes. These enzymes often follow sequential order bi–bi reaction mechanism in which the cofactor, NAD(P)H, first binds to the cofactor-binding pocket of the enzyme followed by binding of the substrate in the C-terminal region, leading to product and coproduct formation
208
A.S. Haque et al. / Journal of Structural Biology 187 (2014) 207–214
and subsequently their release (Giraud and Naismith, 2000; Kavanagh et al., 2008). The majority of the SDR enzymes are standalone enzymes that are biologically active as either a dimer or a tetramer. In the past decade, bioinformatic, biochemical and structural characterization has revealed that reductase (R) domains, which belong to SDR family of enzymes, are a constituent of multimodular enzymes, e.g. Nonribosomal Peptide Synthetase (NRPS), Polyketide Synthase (PKS), and are present at their C-terminus (Chhabra et al., 2012; Keating and Walsh, 1999). The catalytic role played by the R-domain as part of these multifunctional enzymes is to reductively release the product covalently linked to the 40 phosphopantetheine arm of the enzyme as either an aldehyde or an alcohol in an NAD(P)H-dependent manner (Fig. 1) (Chhabra et al., 2012; Read and Walsh, 2007). In mycobacterial species, the NRPSs having reductase domains are involved in the synthesis of mycobactin, secreted iron-chelating siderophores and glycopeptidolipid core which are implicated in virulence and sliding motility etc. (Gokhale et al., 2007; Martinez et al., 1999; Raman et al., 2006; Vats et al., 2012). The R-domain has a Rossmann fold at its N-terminal region which is typical of SDR enzymes and carries out oxidoreductive catalysis in an NADPH-dependent manner. Previous studies have revealed that RNRP and RGPL have a large substrate-binding pocket that contributes to their substrate promiscuity and consequently their ability to reduce acyl-CoAs, e.g. lauroyl-CoA, palmitoyl-CoA etc. in vitro (Chhabra et al., 2012). The substrate promiscuity demonstrated by RNRP and RGPL led us to carry out further studies using lauroyl-CoA as their substrate in the current study. In the present study, using Nuclear Magnetic Resonance (NMR) and Small-angle X-ray Scattering (SAXS) analyses we show that the offloading R-domain of multifunctional NRPS (Rv0101) from Mycobacterium tuberculosis (RNRP) and the Rdomain of mycobacterial peptide synthetase (mps2) from Mycobacterium smegmatis (RGPL) follow random bi–bi reaction mechanism unlike the majority of the SDR standalone enzymes that follow sequential order bi–bi mechanism, to carry out catalysis. Typically, SDRs have Tyr as a strictly invariant catalytic residue apart from other well-conserved residues, Ser/Thr-LysAsn (Oppermann et al., 2003). Here, using ITC, we show catalytic Tyr also plays an essential role in high affinity binding of NADPH and attribute this phenomenon to a noncovalent cation–p interaction present between tyrosine and nicotinamide.
2. Materials and methods 2.1. Mutagenesis RNRP was cloned from 6166 bp to 7676 bp of M. tuberculosis gene Nrp, and RGPL was cloned from 6301 bp to 7539 bp of M. smegmatis gene mps2 (Chhabra et al., 2012). Cloning was achieved using pET-28(b) vector (Novagen) as N-terminal 6xHis expressing proteins. Site-directed mutagenesis of tyrosine to phenylalanine residue in RNRP and RGPL genes was performed using QuickChange site-directed mutagenesis kit (Stratagene). Primers were as follows (mutated nucleotides are underlined): RGPL: Y227F 50 -GCCAACGG CTTCGGGAATTCCAAGTGGGCAGGCG-30 and 50 -CGCCTGCCCACTTGGAATTCCCGAAGCCGTTGGC-30 ; RNRP: Y228F 50 -GCTGGCGGCTTCGG CACCAGCAAGTGGGCCGGT-30 and 50 -ACCGGCCCACTTGCTGGTGCC GAAGCCGCCAGC-30 . Confirmation of the mutagenesis was done by DNA sequencing. 2.2. Protein expression and purification R-domains were expressed as N-terminal 6xHis-tag proteins in Escherichia coli strain BL21 (DE3). Cultures were grown in sterile Luria–Bertani medium containing 50 lg/ml kanamycin at 37 °C with 200 rpm shaking conditions. Cells were induced with 0.5 mM IPTG at 0.6 O.D. at 600 nm and grown further at 22 °C for 12–14 h. Cells were harvested and 5 g of pellet was resuspended in 100 ml of lysis buffer containing 50 mM Tris–HCl pH 8.0 and 150 mM NaCl. Lysis of the cells was done by sonication at an amplitude of 22 for 30 min with 3 s on/off pulses. Lysate was mixed with 1% streptomycin sulfate for 30 min at 4 °C and centrifuged at 10,000 rpm and supernatant was taken further for purification. All purification and concentration steps were performed at 4 °C. Supernatant was loaded onto Ni-NTA Sepharose column preequilibrated with lysis buffer. The bound protein was washed with 5 column volumes of wash buffer 1 (10 mM imidazole, 50 mM Tris–HCl pH 8.0, 150 mM NaCl). Second wash was performed by 10 column volumes of wash buffer 2 (50 mM Tris–HCl pH 8.0, 500 mM NaCl). Elution was carried out with 10 column volumes of a gradient of 20–250 mM imidazole in lysis buffer. Fractions containing the protein were analyzed by SDS–PAGE, pooled together and concentrated up to 10 mg/ml using 30-kDa Amicon centrifugal filter devices (Millipore). The concentrated sample was loaded onto Sephacryl S200 gel filtration column pre-equilibrated with buffer 20 mM Tris–HCl pH 8.0, 20 mM NaCl and then eluted. Fractions containing proteins were concentrated using 30-kDa Amicon centrifugal filter devices (Millipore) and used for Isothermal Titration Calorimetry (ITC) and Small-angle X-ray Scattering (SAXS) experiments. For Nuclear Magnetic Resonance (NMR) experiments, uniformly 15 N-labeled protein samples were prepared by expressing in E. coli grown in modified M9 minimal medium containing 15NH4Cl as the sole nitrogen source. Purification procedures for 15N-labeled proteins were identical as for the unlabeled proteins except for buffer having Tris–HCl at a pH of 7. 2.3. Isothermal Titration Calorimetry (ITC) experiments
Fig.1. Two domain architecture of R-domain: the crystal structure of RNRP (PDB: 4DQV) solved previously reveals a two domain arrangement of the R-domain with NADPH-binding domain (Blue) and substrate-binding domain (brown) with cofactor (NADPH) and substrate (Lauroyl CoA) modeled here. It performs reductive release of the substrate in the form of an alcohol. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Titration experiments for RNRP and RGPL and their corresponding tyr-phe mutants were performed at 25 °C with VP-ITC calorimeter (Microcal, Northampton, MA). Samples containing 2 mM NADPH were prepared in buffer 20 mM Tris–HCl pH 8.0, 20 mM NaCl and loaded onto the syringe. The reaction cell contained 1.6 ml of 96 lM R-domain with a buffer composition of 20 mM Tris–HCl pH 8.0, 20 mM NaCl. Titrations performed with 20 injections of 8 ll NADPH solution for 5 s each at 2 min intervals using
A.S. Haque et al. / Journal of Structural Biology 187 (2014) 207–214
rotating-stirrer syringe. The titration data were analyzed using ORIGIN data analysis software 5.0 (Microcal Software, Northampton, MA). 2.4. Nuclear Magnetic Resonance (NMR) The NMR experiments were performed at 25 °C on a Bruker 600 MHz spectrometer equipped with a triple resonance cryogenically cooled probe by using protein solutions that contained 20 mM Tris–HCl pH 7.0, 20 mM NaCl. 2D 15N–1H Transverse Relaxation-Optimized Spectroscopy (TROSY-HSQC) spectra were recorded with uniformly 15N-labeled RNRP and RGPL and their mutants (95% protein solution/5% D2O, pH 7.0) with 1024 256 data points. The titrations of the R-domains and their mutants with either NADPH (cofactor) or lauroyl CoA (substrate) was performed with a molar ratio of 1:5 and 1:10 which amounted to complete saturation of the enzyme with these ligands until no further perturbations in chemical shifts could be observed. The 2D data was processed using Topspin (v 2.1) and analyzed with Sparky3 (Goddard and Kneller, 2008) on a remote workstation. 2.5. Small Angle X-ray Scattering (SAXS) data collection and processing Precise concentration of the protein samples at 280 nm was determined using NanoDrop 1000 spectrophotometer. The monodispersity of each sample was verified before SAXS data collection using dynamic light scattering technique. Data collection for RNRP and RGPL and their mutants were performed at a concentration of 10 mg/ml and 4 mg/ml, respectively, by addition of 1:10 ratio with (1) NADPH, (2) lauroyl-CoA, (3) NADPH followed by lauroyl-CoA and (4) lauroyl-CoA followed by NADPH with an exposure time of 1 h at 297 K. Corresponding buffers for each of these samples were exposed for the same duration, at the same temperature and data were collected. The sample size in each of these cases was 35 ll. The equipment used for SAXS data collection was S3-micropix SAXS system from Hecus XRS. The beam delivery system was from Xenocs mounted on a sealed Cu tube X-ray generator. The azimuthal averaging of raw data was done using FIT2D program. For further data analyses, programs from the ATSAS 2.5.2 software suite was used (Konarev et al., 2006).
209
Likewise, the RGPL of mps2 from M. smegmatis yielded similar profile to that of RNRP under identical experimental conditions (data not shown). Thus, NMR-based titration studies suggest that the R-domains which are a constituent of megasynthetases follow random ‘bi–bi’ mechanism of action. Interestingly, the resultant chemical shift perturbations in both cases are similar, suggesting that although lauroyl-CoA can bind to the R-domains independently of the presence of NADPH, the final catalytically competent conformation is attained only when both are bound to the enzyme. Previous studies have suggested that conformational changes brought about by binding of NADPH to RNRP and RGPL from NRPSs are essential for these enzymes to carry out nonprocessive reductive catalysis (Chhabra et al., 2012). In the wake of deviation from sequential order ‘bi–bi’ reaction mechanism as shown using 2D 15 N–1H TROSY-HSQC binding studies and substrate promiscuity demonstrated by RNRP and RGPL, we set out to find the role played by NADPH and lauroyl-CoA in driving the enzyme towards catalytically feasible conformation change, if any, associated with them. To this end, SAXS studies were performed with both RNRP and RGPL. 3.2. NADPH-mediated conformational change is a key step to attain catalytically competent conformation SAXS analysis revealed that there was a decrease in both Rg and Dmax upon binding of NADPH to the enzyme at first and further decrease in these parameters was brought about upon binding of lauroyl-CoA due to enhanced structural rearrangements. However, there was negligible change in Rg and Dmax upon binding of lauroylCoA in absence of NADPH (Fig. 3a and b). The data strongly suggest that a large conformational change is associated with NADPH binding while lauroyl-CoA binding brings subtle structural changes when it occupies the large substrate-binding pocket. Similarly, in the case of RGPL, the magnitude of domain motion is significantly large when NADPH binds to it as compared with the binding of lauroyl-CoA. Therefore, SAXS studies clearly demonstrate that the binding of NADPH plays a key role in the enzymes’ attaining a catalyticallycompetent conformation. 3.3. Structural analyses of SDR family of oxidoreductases in complex with NADPH reveal noncovalent interactions between catalytic tyrosine and nicotinamide of NADPH
3. Results and discussion 3.1. Chemical shift perturbations obtained by 15N–1H TROSY-HSQC NMR experiments are suggestive of random bi–bi reaction mechanism To ascertain the binding of lauroyl-CoA to RNRP and RGPL, both in absence and in presence of NADPH, NMR-based binding studies were performed. The 15N–1H TROSY-HSQC NMR experiments revealed significant changes in the chemical shifts of resonances of RNRP upon titration with NADPH. Following saturation of RNRP with NADPH, titration with lauroyl-CoA led to additional and unidirectional chemical shift perturbations in the same residues (Fig. 2a). Interestingly, when the titration order was reversed, same residues in the RNRP experienced weak chemical shift perturbations in the presence of lauroyl-CoA alone. These perturbations enhanced significantly upon addition of NADPH (Fig. 2b). Similar chemical shift perturbations were obtained when identical experiments were carried out with the Y228F mutant of RNRP, suggesting a similar mode of binding of NADPH and lauroyl-CoA (Fig. 2c and d). The 15N–1H TROSY-HSQC spectra acquired in both cases suggest that the presence of NADPH is not necessary for the substrate to bind to the enzyme as often suggested for SDR family of proteins which follow sequential order ‘bi–bi’ reaction mechanism.
The driving force for RNRP and RGPL to acquire catalyticallycompetent conformation in the presence of NADPH prompted structural analyses of known crystal structures of SDR family of enzymes bound to NAD(P)H. Protein Data Bank was used to acquire the known crystal structures of SDR family of proteins and cutoff of resolutions higher than 2 Å was used to short-list the structures. PDBeMOTIF was used to ascertain the interactions between nicotinamide and tyrosine. A total of 121 structures were segregated as a result of this, out of which 84 having NADPH or NAD in similar orientation were used for final structural analysis. A close probing of these structures revealed that the catalytic tyrosine is present in close proximity to the nicotinamide ring with the –OH moiety of the tyrosine residue facing it (Fig. 4). An average distance of 3.97 (±0.47) Å between the centre of the nicotinamide ring and –OH of the tyrosine is highly indicative of an interaction between them and possibly this tyrosine would play a role in the binding and stabilization of NADPH. In a previous study, NADPH-bound crystal structure of b-ketoacyl-acyl carrier protein reductase (FabG) having a single point mutation of catalytic tyrosine to phenylalanine revealed that the electron density for the nicotinamide moiety was conspicuous by its absence which was suggestive of impaired NADPH binding to the enzyme. It was subsequently supported by biochemical studies as determined
210
A.S. Haque et al. / Journal of Structural Biology 187 (2014) 207–214
Fig.2. Zoomed in illustrations from 2D 15N–1H TROSY-HSQC of RNRP representing chemical shift perturbations upon titration with NADPH and L.CoA. (a) RNRP (red) resonances experience perturbations in the chemical shifts upon titration with NADPH (green) followed by L.CoA (blue). (b) Chemical shift perturbations as observed on reversing the order of titration to L.CoA (blue) followed by NADPH (green) with RNRP (red). (c) and (d) correspond to titrations as followed in (a) and (b), respectively, with Y228F mutant of RNRP. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
by intrinsic NADPH fluorescence (Chhabra et al., 2012; Price et al., 2004). These findings suggest that the catalytic tyrosine not only acts as proton donor during catalysis, but it is also required for high-affinity NADPH binding to the enzyme. However, the actual impact of the mutation on the binding affinity of this enzyme for NADPH was not known. We set out to quantify the binding affinity of R-domains for NADPH using ITC and the contribution of tyrosine towards this binding. 3.4. Impaired binding of NADPH to Y228F mutant form of RNRP and Y227F of RGPL ITC experiments were performed to assess the binding affinity of NADPH with both the wild type RNRP and RGPL (Fig. 5a and c) and their corresponding tyr-phe mutants (Fig. 5b and d). In each of the cases, it was found that there is a 3-fold and a 4-fold decrease in the binding affinities of the tyr-phe mutant R-domains towards NADPH when compared with their wild type enzyme (Fig. 5e). Interestingly, in the case of Y227F mutant of RGPL the
reaction was transformed from exothermic to endothermic by mutation of the catalytic tyrosine alone. In the case of wild type and mutant RNRP, the ITC experiments revealed that the binding of the NADPH was enthalpy driven, though there was a decreased binding affinity towards NADPH. The binding of NADPH to wild type RGPL was also enthalpy driven but its mutant binds to NADPH in an entropy-driven manner. Possibly, the hydrophobic environment provided by the mutation of tyrosine to phenylalanine in Y227F RGPL was further compounded by hydrophobic residues present in the neighboring region where nicotinamide binds. We performed the structure-based sequence alignment of the amino acid residues present in the loop that is involved in NADPH binding in RNRP and RGPL that shows VNA-FPY and VNSVLPY, respectively, in case of M. tuberculosis and M. smegmatis. The variations in amino acid residues including extra valine could possibly provide hydrophobic environment at the nicotinamide-binding site resulting in NADPH binding in entropy-driven manner. The ITC experiment along with the structural analysis suggests the presence of an interaction between the catalytic tyrosine and
A.S. Haque et al. / Journal of Structural Biology 187 (2014) 207–214
211
Fig.3. The SAXS data with Rg and Dmax changes upon binding of NADPH and L.CoA. (a) The table suggests that there is a subtle change in Rg or Dmax upon addition of L.CoA alone while the same is not true when it is added in the presence of NADPH. (b) The pair-distance distribution curve of different titrations shows that virtually there is no domain motion associated with only L.CoA titration against RNRP.
3.5. Cation–p interaction between tyrosine and nicotinamide of NADPH
Fig.4. The superimposed catalytic tyrosines and NADPH from known crystal structures is suggestive of the presence of a noncovalent interaction between them.
the nicotinamide ring of NADPH which is crucial for the enzyme to obtain catalytically competent conformation.
The noncovalent interactions of various types, e.g. salt bridges, hydrogen bonds, van der Waals forces etc., are crucial for catalytic activity and structural integrity of biological macromolecules. One of the seldom-reported noncovalent interactions in biological macromolecules is cation–p interaction which takes place between aromatic rings and positively charged amino acid residues (Gallivan and Dougherty, 1999; Ma and Dougherty, 1997). The magnitude of this interaction varies with the nature of the cation and the aromatic system. For example, Li+, Na+, K+, Rb+ interact in decreasing order with the aromatic ring of benzene (Ma and Dougherty, 1997). Apart from acting as a binding site for the cations, these aromatic rings can compete with the highly favorable solvation of an ion provided by an aqueous medium. Unlike the usual cation-binding structures like Asp and Glu, in the case of cation–p interactions these are aromatic amino acids Trp, Tyr and Phe, while the side chains of positively charged Arg and Lys act as cation. The cations interact with the delocalized p electron density of the aromatic ring of Trp, Tyr and Phe. In the present study, it has been demonstrated, using ITC, that there is a 3–4-fold decrease in the binding affinity of NADPH upon mutation of highly conserved catalytic residue Tyr to Phe in R-domains of multimodular NRPSs in mycobacterial species. Here, we suggest that either the cation–p or the dipole–p interaction plays a role in the binding of nicotinamide moiety of NADPH to these enzymes wherein the Tyr performs the function of a cation while the nicotinamide ring
212
A.S. Haque et al. / Journal of Structural Biology 187 (2014) 207–214
Fig.5. Impaired NADPH binding upon mutation of catalytic Tyr to Phe. (a–d) Binding study of wild type and tyr-phe mutant R-domains with NADPH (e) 3–4-fold decrease in binding affinities were found upon mutation of catalytic tyrosine to phenylalanine.
A.S. Haque et al. / Journal of Structural Biology 187 (2014) 207–214
213
Fig.6. The schematic representation of the proposed random bi–bi reaction performed by the mycobacterial R-domains that act as a constituent of the NRPSs. The representative value of Rg as observed from SAXS experiments performed with RNRP.
acts as a p system (Dougherty, 1996; Gupta and Durani, 2011). The study shows that the Tyr plays a crucial role in NADPH binding apart from acting as a proton donor during catalysis in tyrosinedependent oxidoreductases from SDR family of proteins.
4. Conclusion The SDR family of enzymes is often known to follow sequential order bi–bi reaction mechanism. These enzymes exist in either dimeric or tetrameric form and carry out catalysis in a standalone manner. In the present study, the offloading R-domains from mycobacterial species that are a part of multifunctional enzymes, e.g. NRPS and PKS, and belong to SDR family are shown to perform random bi–bi reaction mechanism (Fig. 6). In most of the NRPSs, the offloading of the substrate is performed by a dedicated thioesterase (TE) domain present at their C-terminus (Du and Lou, 2010). It hydrolyses the 40 -phosphopantetheine arm-tethered substrate brought to it by the thiolation (T) domain to release the product, which is attached posttranslationally to the T-domain (Lambalot et al., 1996). The cellular pool contains copious amounts of not only free CoA but also various acyl-CoA derivatives e.g. acetyl-CoA which can get attached to the thiolation domain and block the thio-template based assembly line enzymology performed by NRPSs (Quadri et al., 1998). The cell expresses TEII enzyme that removes the spuriously attached acetyl-CoA and resumes the enzymatic reactions (Schwarzer et al., 2002; Yeh et al., 2004). The R-domains of the NRPSs from mycobacterial species have one of the largest C-terminal substrate-binding domains spanning approximately 150 amino acids with unique hydrophobic pocket to accommodate large acyl-CoAs and there are possibilities that acyl-CoAs may occupy the substrate-binding pocket, thus blocking the reactions. The R-domains, by virtue of random bi-bi reactions, are able to release the unwanted acetyl-CoA from their substrate-binding pocket and resume the assembly line enzymology. The decrease in binding affinity of NADPH in tyr-phe mutant of R-domains and structural analysis reveal that there is indeed an interaction between the catalytic tyrosine of R-domains and the nicotinamide ring of NADPH suggestive of cation–p interaction.
We propose that as the products of NRPSs are crucial for the survival and proliferation of these pathogens, and following sequential order reaction might block thiotemplate-based assembly line enzymology which would be detrimental to their survival. The current study provides insights regarding the alternative form of a reaction mechanism obtained by R-domain as a constituent the multifunctional enzymes e.g. NRPSs and PKSs where this domain performs the task of offloading the ‘product-in-waiting’. This acquired catalytic competency by virtue of random bi–bi reaction mechanism provides leverage to the multifunctional enzymes to process the bound substrate in a cofactor-dependent manner irrespective of their order of binding to the enzyme that is crucial to maintain thiotemplate-based assembly line enzymology.
References Bretschneider, T., Zocher, G., Unger, M., Scherlach, K., Stehle, T., Hertweck, C., 2012. A ketosynthase homolog uses malonyl units to form esters in cervimycin biosynthesis. Nat. Chem. Biol. 8, 154–161. Chang, Y.H., Chuang, L.Y., Hwang, C.C., 2007. Mechanism of proton transfer in the 3alpha-hydroxysteroid dehydrogenase/carbonyl reductase from Comamonas testosteroni. J. Biol. Chem. 282, 34306–34314. Chhabra, A., Haque, A.S., Pal, R.K., Goyal, A., Rai, R., Joshi, S., Panjikar, S., Pasha, S., Sankaranarayanan, R., Gokhale, R.S., 2012. Nonprocessive [2 + 2]e off-loading reductase domains from mycobacterial nonribosomal peptide synthetases. Proc. Natl. Acad. Sci. U.S.A. 109, 5681–5686. Dougherty, D.A., 1996. Cation–pi interactions in chemistry and biology: a new view of benzene, Phe, Tyr, and Trp. Science 271, 163–168. Du, L., Lou, L., 2010. PKS and NRPS release mechanisms. Nat. Prod. Rep. 27, 255–278. Gallivan, J.P., Dougherty, D.A., 1999. Cation–pi interactions in structural biology. Proc. Natl. Acad. Sci. U.S.A. 96, 9459–9464. Giraud, M.F., Naismith, J.H., 2000. The rhamnose pathway. Curr. Opin. Struct. Biol. 10, 687–696. Goddard, T.D., Kneller, D.G., 2008. SPARKY 3. University of California, San Francisco, http://www.cgl.ucsf.edu/home/sparky. Gokhale, R.S., Saxena, P., Chopra, T., Mohanty, D., 2007. Versatile polyketide enzymatic machinery for the biosynthesis of complex mycobacterial lipids. Nat. Prod. Rep. 24, 267–277. Gupta, P., Durani, S., 2011. Aromatic interactions at atom-to-atom contact and just beyond: a case study of protein interactions of NAD(+)/NADP(+). Int. J. Biol. Macromol. 49, 999–1006. Jornvall, H., Persson, B., Krook, M., Atrian, S., Gonzalez-Duarte, R., Jeffery, J., Ghosh, D., 1995. Short-chain dehydrogenases/reductases (SDR). Biochemistry 34, 6003–6013. Kavanagh, K.L., Jornvall, H., Persson, B., Oppermann, U., 2008. Medium- and shortchain dehydrogenase/reductase gene and protein families: the SDR
214
A.S. Haque et al. / Journal of Structural Biology 187 (2014) 207–214
superfamily: functional and structural diversity within a family of metabolic and regulatory enzymes. Cell. Mol. Life Sci. 65, 3895–3906. Keating, T.A., Walsh, C.T., 1999. Initiation, elongation, and termination strategies in polyketide and polypeptide antibiotic biosynthesis. Curr. Opin. Chem. Biol. 3, 598–606. Konarev, P.V., Volkov, V.V., Sokolova, A.V., Koch, M.H.J., Svergun, D.I., 2006. ATSAS 2.1, a program package for Small-angle scattering data analysis. J. Appl. Crystallogr. 39, 277–286. Lambalot, R.H., Gehring, A.M., Flugel, R.S., Zuber, P., LaCelle, M., Marahiel, M.A., Reid, R., Khosla, C., Walsh, C.T., 1996. A new enzyme superfamily – the phosphopantetheinyl transferases. Chem. Biol. 3, 923–936. Ma, J.C., Dougherty, D.A., 1997. The Cation minus sign pi interaction. Chem. Rev. 97, 1303–1324. Martinez, A., Torello, S., Kolter, R., 1999. Sliding motility in mycobacteria. J. Bacteriol. 181, 7331–7338. Oppermann, U., Filling, C., Hult, M., Shafqat, N., Wu, X., Lindh, M., Shafqat, J., Nordling, E., Kallberg, Y., Persson, B., Jornvall, H., 2003. Short-chain dehydrogenases/reductases (SDR): the 2002 update. Chem. Biol. Interact. 143–144, 247–253. Price, A.C., Zhang, Y.M., Rock, C.O., White, S.W., 2004. Cofactor-induced conformational rearrangements establish a catalytically competent active site and a proton relay conduit in FabG. Structure 12, 417–428. Quadri, L.E., Weinreb, P.H., Lei, M., Nakano, M.M., Zuber, P., Walsh, C.T., 1998. Characterization of Sfp, a Bacillus subtilis phosphopantetheinyl transferase for peptidyl carrier protein domains in peptide synthetases. Biochemistry 37, 1585–1595.
Raman, S., Puyang, X., Cheng, T.Y., Young, D.C., Moody, D.B., Husson, R.N., 2006. Mycobacterium tuberculosis SigM positively regulates Esx secreted protein and nonribosomal peptide synthetase genes and down regulates virulenceassociated surface lipid synthesis. J. Bacteriol. 188, 8460–8468. Read, J.A., Walsh, C.T., 2007. The lyngbyatoxin biosynthetic assembly line: chain release by four-electron reduction of a dipeptidyl thioester to the corresponding alcohol. J. Am. Chem. Soc. 129, 15762–15763. Rudnick, D.A., McWherter, C.A., Rocque, W.J., Lennon, P.J., Getman, D.P., Gordon, J.I., 1991. Kinetic and structural evidence for a sequential ordered Bi Bi mechanism of catalysis by Saccharomyces cerevisiae myristoyl-CoA: protein Nmyristoyltransferase. J. Biol. Chem. 266, 9732–9739. Schwarzer, D., Mootz, H.D., Linne, U., Marahiel, M.A., 2002. Regeneration of misprimed nonribosomal peptide synthetases by type II thioesterases. Proc. Natl. Acad. Sci. U.S.A. 99, 14083–14088. Vats, A., Singh, A.K., Mukherjee, R., Chopra, T., Ravindran, M.S., Mohanty, D., Chatterji, D., Reyrat, J.M., Gokhale, R.S., 2012. Retrobiosynthetic approach delineates the biosynthetic pathway and the structure of the acyl chain of mycobacterial glycopeptidolipids. J. Biol. Chem. 287, 30677–30687. Wang, Z.X., Wu, J.W., 2007. The complete pathway for ERK2-catalyzed reaction. Evidence for an iso random Bi Bi mechanism. J. Biol. Chem. 282, 27678–27684. Yeh, E., Kohli, R.M., Bruner, S.D., Walsh, C.T., 2004. Type II thioesterase restores activity of a NRPS module stalled with an aminoacyl-S-enzyme that cannot be elongated. ChemBioChem 5, 1290–1293.