J Struct Funct Genomics (2014) 15:63–71 DOI 10.1007/s10969-014-9182-1
Structural basis of malaria parasite lysyl-tRNA synthetase inhibition by cladosporin Sameena Khan • Arvind Sharma • Hassan Belrhali Manickam Yogavel • Amit Sharma
•
Received: 30 April 2014 / Accepted: 24 May 2014 Ó Springer Science+Business Media Dordrecht 2014
Abstract Malaria parasites inevitably develop drug resistance to anti-malarials over time. Hence the immediacy for discovering new chemical scaffolds to include in combination malaria drug therapy. The desirable attributes of new chemotherapeutic agents currently include activity against both liver and blood stage malaria parasites. One such recently discovered compound called cladosporin abrogates parasite growth via inhibition of Plasmodium falciparum lysyl-tRNA synthetase (PfKRS), an enzyme central to protein translation. Here, we present crystal structure of ternary PfKRS-lysine-cladosporin (PfKRS-KC) complex that reveals cladosporin’s remarkable ability to mimic the natural substrate adenosine and thereby colonize PfKRS active site. The isocoumarin fragment of cladosporin sandwiches between critical adenine-recognizing residues while its pyran ring fits snugly in the ribose-recognizing cavity. PfKRS-K-C structure highlights ample space within PfKRS active site for further chemical derivatization of cladosporin. Such derivatives may be useful against additional human pathogens that retain high conservation in cladosporin chelating residues within their lysyl-tRNA synthetase. Keywords Cladosporin KRS Malaria X-ray crystal structure Inhibition
S. Khan A. Sharma M. Yogavel A. Sharma (&) Structural and Computational Biology Group, International Centre for Genetic Engineering and Biotechnology (ICGEB), New Delhi 110067, India e-mail:
[email protected] H. Belrhali European Molecular Biology Laboratory, 6 Rue Jules Horowitz, BP 181, 38042 Grenoble, France
Abbreviations aaRSs Aminoacyl-tRNA synthetases AMP-PNP Adenosine 50 -(b,c-imido)triphosphate ASU Asymmetric unit ATP Adenosine triphosphate CCDC Cambridge crystallographic data centre MST Microscale thermophoresis PDB Protein data bank PfKRS Plasmodium falciparum lysyl-tRNA synthetase PfKRS-K-C PfKRS-lysine-cladosporin RMSD Root-mean-square-deviation
Introduction The scourge of malaria continues with high morbidity and *0.5 million deaths annually (WHO, World Malaria Report 2013). Malaria eradication requires new drug targets and therapeutics with utility against both liver and blood stage parasites. Cladosporin has recently been discovered with necessary attributes that suggest its value as a base for next generation of anti-parasitics [1]. A comprehensive series of very elegant experiments have identified Plasmodium falciparum lysyl-tRNA synthetase (PfKRS) as the cellular target for cladosporin activity [1]. AminoacyltRNA synthetases (aaRSs) are ancient enzymes responsible for genetic code translation [2]. These ubiquitous enzymes attach amino acid onto cognate tRNA for protein translation in an aminoacylation reaction, and their structure– function attributes within malaria parasites are being actively dissected [1, 3–15]. Targeting parasitic aaRSs can provide an additional component in the present multi-drug
123
64
cocktail therapy against malaria [1, 3–15]. Discovery of cladosporin not only presents an opportunity to kill both liver and blood stage malaria parasites, but analysis of KRSs from other human malaria parasites like Plasmodium vivax suggests utility against them as well [1, 8]. Encouragingly, routes for isolation [16, 17] and complete synthetic production of cladosporin [18] are accessible, enabling grounds for derivatization of cladosporin to improve its drug-like properties. Further, availability of both malaria parasite and human KRS crystal structures [8, 19] has provided an architectural framework for understanding the specificity and selectivity displayed by cladosporin [1, 8]. In this vein, elucidation of the crystal structure of PfKRS-cladosporin complex seems vital to further drug development efforts via structure-based rational design strategies.
S. Khan et al.
known small molecule geometry (CCDC-872667) [18] using Sketcher, which is part of the CCP4 Suite [24]. The inhibitor was modeled into the electron density and protein model was manually improved using COOT [25]. The final model has good geometry and 94 % residues are in favoured regions of the MolProbity Ramachandran plot [26]. Statistics for data collection and structure refinement are given in Table 1. All structural superimposition and preparation of figures was done using UCSF Chimera [27], LIGPLOT [28] and PyMOL (http://www.pymol.org). The atomic coordinates and structural factors have been deposited into protein data bank with accession number 4PG3. Thermal shift assay
PfKRS was produced in accordance with methods published recently [8]. Highly pure PfKRS was stored in 50 mM Tris–HCl pH 8.0, 200 mM NaCl, 10 mM bMe at 193 K. Before crystallization, 500 lM cladosporin and 2 mM L-Lysine were added to PfKRS and mixture was incubated at 277 K for 10 min. Crystals of PfKRS-K-C complex were obtained at 293 K by the hanging-drop vapour-diffusion method using 1 ll PfKRS-cladosporin and 1 ll crystallization well solution with mother liquor [0.1 M Bis-Tris pH 6.5, 2 % (v/v) Tascimate pH 6.0, 20 % (w/v) PEG 3350]. Plate-shaped crystals appeared within 10 days.
Protein melting curve assays was performed as reported earlier [29]. PfKRS was diluted in buffer containing 20 mM Tris, pH 8.0, 200 mM NaCl, 5 mM MgCl2 and 19 SYPRO orange dye (Life Technologies). Three different ligands at 20 lM each ATP (Sigma), AMP-PNP (Sigma) and cladosporin (gifted by Bart Staker, SSGCID) were added to PfKRS (2 lM) and incubated at room temperature for 10 min. PfKRS-apo and PfKRS-ligands were heated from 20 to 96 °C at a rate of 1 °C/min and fluorescence signals were monitored by StepOnePlus quantitative realtime PCR system (Life Technologies). Each curve was an average of three measurements and data were analyzed using Thermal shift software (Life technologies). Cladosporin alone and ATP alone in assay buffers, along with no protein controls were used and flat lines were observed for these fluorescence readings at all temperatures. Boltzmann Tm (melting temperature) and derivative Tm were found to have similar values and latter was used for analyses.
Data collection and structure determination
Microscale thermophoresis binding assay
Single crystal was transferred to mother liquor supplemented with 20 % glycerol and cryo-cooled with liquid nitrogen at 100 K. X-ray diffraction data were collected using synchrotron radiation facility at the BM14 beamline (ESRF, Grenoble, France). A total of 300 images were collected in 0.5° oscillation steps with 12 s exposure per frame. The diffraction images were processed and scaled with HKL2000 program suite [20]. Structure was solved by molecular replacement (MR) method with the program Phaser-MR [21] from PHENIX program suite [22] using apo-PfKRS structure (PDB: 4H02) [8] as the template. There are two dimers of PfKRS-K-C complex in the asymmetric unit with a solvent content of *55 %. The model was initially refined using PHENIX [22] and completed with REFMAC5 [23]. The X-ray refinement restraint parameters were generated for cladosporin molecule with
A NanoTemper Monolith Instrument (NT.115) was used for measuring thermophoresis [30]. Purified PfKRS was first labeled with NT-647 dye. AMP-PNP (0.5 mM–3.8 nM) and cladosporin (2.5 lM–0.03 nM) were titrated to a constant amount of labeled PfKRS (10 nM) in 50 mM Tris (pH 7.6), 150 mM NaCl, 10 mM MgCl2, 0.05 % Tween-20 buffer and 1 mM L-Lys. The samples were incubated at room temperature for 10 min and were loaded into MST hydrophilic treated glass capillaries and analysis was performed using 50 % LED power and 20 % MST at 30 °C. KD values were calculated using Nano Temper software.
Materials and methods Protein expression, purification and crystallization
123
Molecular modeling Structural alignment of the X-ray crystal structures of HsKRS-K-ATP (PDB code 3BJU) [19] and PfKRS-K-C
Structural basis of malaria parasite
65
Table 1 X-ray data and refinement statistics PfKRS-K-C Crystal parameters Space group Cell dimensions ˚) a, b, c (A a, b, c (°)
P 21 21 2
homology models were built for these KRSs using Prime software (Schro¨dinger, LLC, New York, NY, 2012). All molecular models were superimposed onto PfKRS-K-C and analyzed thereafter. The active site cavity was generated using Fpocket server [31].
297.9, 59.0, 141.5 90, 90, 90
Results
˚ 3/Da) Matthews coefficient (A
2.7
Solvent content (%)
54.8
Cladosporin binds with high affinity to PfKRS
Data collection ˚) Wavelength (A
0.953725
˚) Resolution (A
30.0–2.70 (2.75–2.70)
No. of unique reflections Rsym (%)
69,825 (3,406) 0.09 (0.624)
The discovery of cladosporin’s activity against both liver and blood stage malaria parasites, along with experimental dissection of its exclusive targeting to PfKRS is a seminal achievement in anti-malarial target validation/drug development [1]. The isocoumarin fragment based cladosporin scaffold is a natural product and fungal secondary metabolite found in diverse fungi including in Aspergillus flavus and Cladosporium cladosporioides [16, 17]. More encouragingly, complete synthetic production of cladosporin [18] has recently been accomplished making it amenable for generating a whole battery of scaffolds that can potentially be fine-tuned to improve drug-like properties. Cladosporin is composed of a THP ring (2,6-disubstituted tetrahydropyran) fused to isocoumarin moiety [(S)3,4-dihydro-6,8-dihydroxyisochromen-1-one (a d-valerolactone coupled to 1,3-dihydroxybenzene, Fig. 1a]. In order to interrogate relative binding affinities of cladosporin, ATP or AMP-PNP (a non-hydrolysable analog adenosine 50 -(b,c-imido) triphosphate) for PfKRS, we performed thermal shift and microscale thermophoresis (MST) assays using highly purified enzyme. The thermal melting profile of PfKRS is only slightly altered by either ATP or AMPPNP with a shift 0.5 ± 0.2 °C at 20 lM substrate concentration suggesting moderate increment in enzyme stability upon ligand engagement (Fig. 1b). In contrast, addition of cladosporin (again at 20 lM) shifted PfKRS melting curve by 16 ± 0.4 °C indicating both higher affinity and greater thermal stability of the PfKRScladosporin complex (Fig. 1b). In a separate experiment, an increase in thermal stability of PfKRS as a function of increasing concentrations of cladosporin was also observed. These data suggest very stable and high affinity complex formation between PfKRS and cladosporin. Further, MST binding assays were conducted to determine binding constants of cladosporin and AMPPNP with purified PfKRS enzyme. These experiments revealed KD value of *4.2 ± 0.045 lM for PfKRS-AMPPNP complex (Fig. 1c), and a *300 fold tighter binding with cladosporin at KD of 14 ± 1.4 nM (Fig. 1c). These data suggest very high specificity interaction between PfKRS and cladosporin, and highlights its unusual affinity for PfKRS active site. These binding affinity measurements are generally
Mean redundancy
6.2 (6.3)
Overall completeness (%)
99.9 (100.0)
Mean I/r(I)
18.2 (2.7)
Refinement residuals ˚) Resolution (A
30.0–2.70 (2.75–2.70)
No. of unique reflections
67,498
No. of reflections in Rfree set
5,853
Rfree (%)
24.9 (33.5)
Rwork (%)
18.3 (25.9)
Completeness (%)
96.5 (78.7)
Model quality ˚) RMSD bond lengths (A
0.008
RMSD bond angles (8)
1.267
Molprobity Ramachandran distribution Most favored (%) Allowed (%)
93.3 6.4
Disallowed (%)
1.3
Model contents Protomers in ASU
4
Protein residues
1,944
Ligands
Clado and L-Lys
No. of protein atoms
15,736
No. of ligand atoms (Clado/L-Lys)#
84/40
No. of water molecules
204
PDB accession code
4PG3
Entries in parentheses report data from the limiting resolution shell. Data reduction and refinement statistics come from HKL2000 [20] and PHENIX [22] respectively. The abbreviations RMSD and ASU stand for root-mean-square deviation and asymmetric unit respectively #
Clado = Cladosporin; L-Lys = L-Lysine
using Chimera [27] reveal high degree of similarity. HsKRS X-ray structure showed highest amino acid sequence conservation with KRSs from A. flavus and C. sphaerospermum, T. vivax, T. congolense, T. cruzi, S. mansoni and L. loa. Therefore, HsKRS was selected as the template and
123
66
S. Khan et al.
Fig. 1 Chemical structure and binding affinity of cladosporin. a Chemical structures of cladosporin and adenosine with constituent chemical fragments. The explicit hydrogen atoms in cladosporin are highlighted in blue. b Thermal melting profile of apo-PfKRS and PfKRS bound to ATP, AMPPNP and cladosporin. The plot shows derivative Tm of PfKRS against fluorescence (arbitrary units) in y-axis and temperature in x-axis. c Analysis of AMPPNP and cladosporin interactions with PfKRS by microscale thermophoresis (MST). Unlabeled ligands AMPPNP (0.5 mM to 3.8 nM) and cladosporin (2.5 lM to 0.03 nM) were titrated into a fixed concentration of labeled PfKRS (10 nM)
consistent with those reported in cell based assays against malaria parasite lines (IC50 *\100 nM) [1]. These data together indicate a high affinity interaction between PfKRS and cladosporin (relative to natural substrate) that results in a structurally stable complex capable of imparting significantly higher PfKRS resistance to thermal denaturing.
123
Structure determination of PfKRS-lysine-cladosporin ternary complex To investigate the structural basis of inhibition by cladosporin, we produced recombinant PfKRS [8] and incubated the enzyme with L-Lys and cladosporin in co-crystallization
Structural basis of malaria parasite
67
molecules that form two homodimers where each subunit contains the canonical N-terminal anticodon binding and an attached C-terminal catalytic domain (Fig. 2a). From the experimental electron density maps, it was immediately clear that extra electron densities were present in the ATP and amino acid binding pockets in each of the four PfKRS protomers. These were unambiguously assigned to cladosporin and L-Lysine, and the atomic structure of the PfKRSK-C was thence built. Interestingly, the cladosporin structure built into the experimental electron density maps fits the known X-ray structure of cladosporin (CCDC-872667) [18] with correct stereochemistry for all asymmetric carbons (Fig. 2b). The location of the bound lysine in PfKRS-K-C is similar to that in structure of HsKRS-K-ATP complex (Fig. 2b) [19]. Superposition of PfKRS-K-C structure onto apo-PfKRS (PDB code 4H02) [8] shows significant struc˚ ) in four regions (residues 294–299, tural changes ([1.5 A 330–340, 380–390 and 400–460) of the protomer. In PfKRS˚ ) are observed in loop residues K-C, large deviations ([2.5 A 330–340 where this loop contributes cladosporin-interacting residues Glu332 and Asn339 (Fig. 2b). The loop residues 282–288 are ordered in PfKRS-K-C such that Gly284 contributes a hydrogen bond to bound lysine (Fig. 2b). In both apo-PfKRS and PfKRS-K-C structures, residues 515–540 were disordered. However, well-defined electron density was observed for a disulfide bond (Cys517–Cys540) in PfKRS-K-C and not in apo-PfKRS (Fig. 2c). These structural observations imply several subtle conformational changes that occur upon cladosporin engagement with PfKRS. Cladosporin mimics adenosine binding
Fig. 2 Structure of PfKRS-K-C ternary complex and interactions. a The biological dimer of PfKRS-K-C is shown with bound cladosporin (yellow) and L-Lys (blue). b A simulated annealing mFo-DFc omit map (2.5 r level) displaying bound cladosporin and LLysine in PfKRS-K-C. c Ordering of loop constituting residues 280–290 (green) in PfKRS-K-C structure that is disordered in apoPfKRS. Another protein segment is disordered in both bound and apo PfKRSs, but is ordered in HsKRS (shown in blue). The disulfide bond in PfKRS-K-C (Cys517–Cys540) is also shown (right panel) and this region is highly disordered in apo-PfKRS. The simulated annealing mFo-DFc omit map was computed by removing Cys517 and Cys540 atoms and here is contoured at 2.5 r level
experiments. This resulted in single, high quality PfKRS-K˚ resolution (Table 1). We C crystals that diffracted to 2.7 A solved the PfKRS-K-C structure by molecular replacement using apo-PfKRS as a template (PDB: 4H02) [8]. The asymmetric unit in PfKRS-K-C crystals contains four PfKRS
ATP mimicry by cladosporin as a basis for its inhibitory action against PfKRS has been proposed based on biochemical data [1]. Consistent with this, our elucidation of PfKRS-K-C structure reveals cladosporin housed in the canonical ATP binding pocket of this class II aaRS (Fig. 3a). The isocoumarin moiety of cladosporin is buried in a pocket mainly composed of residues Arg330, Glu332, Thr337, His338, Asn339, Phe342, Arg559 and Ile570. In addition to hydrogen bonds between two waters, Glu332, Asn339 and the hydroxyl O atoms of dihydroxybenzene ring, the isocoumarin moiety is mainly stabilized by hydrophobic contacts—in particular by stacking with the adenine-recognizing Phe342 (Fig. 3a). The interactions contributed by guanidine group of Arg330 and Arg559 further stabilize the d-valerolactone ring of isocoumarin (Fig. 3a). In addition, edge-to-face interactions contributed by His338 accommodate the dihydroxybenzene ring (Fig. 3a). The THP ring is accommodated in ribose-recognizing subpocket that is of dual character (hydrophilic and hydrophobic residues) built by PfKRS residues
123
68
S. Khan et al.
Fig. 3 Interactions of cladosporin, adenosine and lysine molecules in KRSs. a The isocoumarin moiety of cladosporin stacks between Ph342 and Arg559 in PfKRS. In addition, His338 provides edge-toface interaction with hydroxybenzene ring of isocoumarin moiety. The THP ring points towards L-Lys binding pocket and its methyl group faces Ser344. b Superposition of bound cladosporin molecules (a–d yellow) on to its small molecule crystal structure (CCDC-872667 green) [18]. c Superposition of PfKRS-K-C (blue) and HsKRS-KATP (green) shows identical positioning of cladosporin and adenosine within KRS actives sites. The bound cladosporin (yellow) in
PfKRS-K-C and adenosine portion (green) of ATP in HsKRS-K-ATP are shown as stick models. d Bound lysine molecule in PfKRS-K-C active site and hydrogen bonding interactions are shown as dashed lines. e Surface representation of the active site cavity in PfKRS with bound cladosporin (yellow) and lysine (green) molecules shown as sticks. The inner surface colors blue, grey and orange represent the charged, polar and hydrophobic regions of PfKRS. The white arrows point to areas within the active site which form extended cavities with potential for exploitation in cladosporin derivatization
Arg330, Phe342, Ser344, Glu500, Val501, Leu502, Ans503, Gly554, Lue555 and Glu556 (Fig. 3a). The THP and d-valerolactone rings adopt chair and half-chair conformations respectively which are nearly identical to the conformation found in X-ray structure of cladosporin (Fig. 3b, CCDC-872667) [18]. The average root-mean square deviation (RMSD) between the four bound cladosporin’s non-H atoms and cladosporin crystal structure are ˚ (Fig. 3b). The superposition in range of *0.13 to 0.23 A
of catalytic domains from PfKRS-K-C and HsKRS-K-ATP ternary complexes reveals striking congruence in structural positioning of cladosporin and adenosine moieties respectively (Fig. 3c). The cladosporin’s d-valerolactone ring superimposes on the five-membered imidazole ring of adenosine (Fig. 3c). The hydroxyl O1 atom of cladosporin’s dihydroxybenzene and the double bonded O4 of dvalerolactone ring reside where N1 and N3 atoms of adenine moiety do in HsKRS-K-ATP (Fig. 3c). The O4 atom
123
Structural basis of malaria parasite
interacts with protein atoms (main-chain N atoms of Asp588 and Arg559, and side-chain Od1 atom of Asp558) via a water molecule which is similar to that of N3 atom in adenine ring. The hydroxyl O1 atom accepts hydrogen atom from main-chain N atom of Asn339 in PfKRS-K-C that is similar to that of N1 atom in adenine ring (Fig. 3c). The other hydroxyl O2 of cladosporin’s dihydroxybenzene atom points towards the side-chain O atom of Glu332 that is similar to the amino group N6 in adenine moiety in HsKRS-K-ATP (Fig. 3c). Remarkably, cladosporin’s THP ring mimics the ribose moiety of adenosine. Once again, orientation of ribose in HsKRS-K-ATP and THP in PKRSK-C along with location of atoms O5 (THP ring) and O40 (sugar ring) are strikingly similar (Fig. 3c). Hence, both the isocoumarin and the THP fragments of cladosporin possess remarkable ability to mimic respectively adenine and ribose moieties of the natural substrate adenosine (within ATP) resulting in a scaffold capable of high potency competitive inhibition of PfKRS. RMSDs between non-H atoms of lysine molecules ˚. (A–D) in PfKRS-K-C ranges between *0.24 and 0.75 A The bound lysine conformation and interactions with protein residues between ternary complexes PfKRS-K-C and HsKRS-K-ATP are expectedly similar (Fig. 3d). We propose that derivatization of cladosporin can be achieved as it is evident from the lysine binding groove within PfKRS active site that extension of the THP ring into structural spaces that extend into the lysine binding pocket is both structurally and chemically feasible (Fig. 3e). Selectivity determinants for cladosporin Recently, the biochemical basis for cladosporin selectivity for PfKRS over HsKRS was attributed to two key residues in the ATP binding pocket at positions Val328 and Ser344 (based on PfKRS numbering from apo-KRS) [1]. In particular, it was proposed that a larger side chain at either of Val328 or Ser344 positions (as in human KRS where these are Gln and Thr respectively) likely discourages cladosporin engagement [1]. To further address this in context of PfKRS-K-C structure, we first undertook a comparison of KRS catalytic domain residues between P. falciparum, humans and cladosporin-producing fungi A. flavus and C. sphaerospermum (Fig. 4a). As expected, but nonetheless noteworthy, the fungal KRS active sites contain cladosporin-unfriendly residues (Gln and Thr in A. flavus; Val and Ile in C. sphaerospermum) for the two key positions of Val328 and Ser344 (Fig. 4a). A molecular surface representation of architectural details from PfKRS-K-C, apoPfKRS and HsKRS-K-ATP structures puts these data in structural context at atomic resolution (Fig. 4b–e). Firstly, the apo-PfKRS structure reveals significant alternations in the rotameric conformations of two key adenine-
69
recognition residues Phe342 and Arg559 when bound to cladosporin (Fig. 4b). Secondly, the spatial constraints that dictate cladosporin binding in PfKRS-K-C are manifest via a specific rotameric conformation of residue Ser344, where its side chain can allow optimal fitting of THP ring into PfKRS active site spaces (Fig. 4c). It is feasible that a smaller side chain in place of Ser344 (say alanine) may still allow cladosporin binding. Superposition of PfKRS-KC and HsKRS-K-ATP structures reveals potential steric clashes between methyl group of THP and Ser344’s equivalent residue in human KRS Thr337 (Fig. 4d), reaffirming centrality of Ser344 (or a smaller side chain residue like alanine) for cladosporin accommodation. Indeed, in C. sphaerospermum KRS, Ser344 is substituted by Ile236 and that may result in a short contact distance between methyl group of THP and Ile236 side-chain (Fig. 4e), possibly leading to poorer engagement. Hence, a clear distinction within KRS active sites can be drawn between sequences (and their respective organisms) that are likely to be cladosporin-accommodating (i.e. retain Val328 and Ser344 positions) versus those which may interact poorly with cladosporin (i.e. contain Gln and Thr or Val and Ile respectively). In a reversal of this logic, we sought and discovered several human and animal pathogens where KRS active sites retain identity of the critical cladosporinaccommodating residues Val328 and Ser344, besides displaying high conservation in other parts of their KRS active sites (Fig. 4a). These parasites include pathogens that cause trypanosomiasis (including Chagas disease), intestinal schistosomiasis and loaiasis (respective infectious agents: T. cruzi, T. vivax, T. congolense, S. mansoni and L. loa). Hence, it is likely that cladosporin-based derivative compounds can be of much wider applicability as agents that can specifically inhibit pathogen but not human lysyltRNA synthetase.
Discussion Malaria remains a killer disease whose total eradication using cocktail drug therapy and other prevention methods is now the generally accepted endgame for public health organizations worldwide. However, inevitable development of drug resistance in malaria parasites is an expected clinical hurdle, and hence efforts directed at identification and validation of novel drug targets along with new inhibitor chemical classes are a key feature of many research programs. Targets including kinases, lipid and histone binding parasite proteins have previously been explored in some detail [32–35]. The protein translation machinery in cells is reliant on many factors including the tRNA synthetase family of enzymes, which together govern transmission of genetic information to proteins [2]. The
123
70
S. Khan et al.
Fig. 4 Active site and cladosporin interacting residues within KRSs. a A portion of sequence alignment of PfKRS with other KRSs. In this, two key residues (and their equivalents) that likely endow sensitivity to cladosporin are highlighted in red boxes. KRS sequences from potentially poor cladosporin binders like C. sphaerospermum and A. flavus are shown. Conservation of Val328 and Ser344 (PfKRS numbering) positions in KRSs from pathogens (in green) hints at possible utility of cladosporin against their KRSs. b Modeled cladosporin (yellow) in apoPfKRS form showing open conformations of Phe342 and Arg559. Protein residues are shown in stick representations with a transparent overlay of the corresponding Van der Waal’s surface. c The bound cladosporin in PfKRS-K-C is sandwiched between Phe342 and Arg559 (closed conformation). d Cladosporin molecule was placed in ATP binding pocket of HsKRS based on superposition of PfKRS-K-C onto HsKRS-K-ATP. The methyl group of THP ring faces towards Thr337 with potential atomic clash distance shown in dashed line. e The modeled active site architecture of C. sphaerospermum KRS showing bulkier side-chain of Ile236 that may diminish cladosporin engagement
36 member malaria parasite tRNA synthetase ensemble has been focus of intense research lately [1, 3–15], and in this stream the recent discovery and validation of cladosporin has been a seminal event [1], both in terms of its utility against multiple parasitic stages and its specificity for P. falciparum lysyl-tRNA synthetase [1]. With a wider view, cladosporin not only provides a new scaffold for antimalarial drug development, but also sets the stage for similar studies centered on the exploitation of abundant chemical wealth possessed by natural products in conjunction with phenotypic screening. Our present work investigates the structural basis for exclusivity displayed by cladosporin for targeting the malaria parasite KRS and not the human enzyme counterpart.
123
Using direct PfKRS-ligand binding experiments, we show that cladosporin confers significant thermal stability to PfKRS and binds to it with high affinity. Cladosporin’s mode of action is indeed based on competitive inhibition of the adenosine moiety, and its engagement with the PfKRS causes significant local perturbations in active site side chains leading to chemically robust steric and stereochemical accommodation. However, cladosporin docking is poor in human KRS, likely by virtue of discrimination residues that line KRS active site. Our sequence-based exploration of other pathogen KRSs that manage to retain cladosporin-favouring residues within their active sites suggests much wider utility of this privileged scaffold in drug development efforts. Hence, besides malaria parasites,
Structural basis of malaria parasite
pathogens like T. cruzi, T. vivax, T. congolense, S. mansoni and L. loa display high conservation in cladosporin engagement residues, and thereby could be equally targeted. Finally, a giant step in derivative chemistry centered on the cladosporin scaffold has been taken in light of the recent experimental demonstration of cladosporin synthesis [18]. The ADME properties of cladosporin need to be refined for its therapeutic applicability, and in this vein our data clearly provide a structural rationale for branching or extension from the THP moiety into the lysine-binding pocket within KRS active sites. In sum, this work directs focus on exploitation of cladosporin-based leads for generation of specific and effective therapeutics against human and animal pathogens. Acknowledgments The authors thank Bart Staker, Seattle Structural Genomics Center for Infectious Disease (SSGCID), for supplying cladosporin. This research was supported by Department of Biotechnology, Government of India OSRP Grant PR6303 to AS.
References 1. Hoepfner D, McNamara CW, Lim CS, Studer C, Riedl R et al (2012) Cell Host Microbe 11:654–663 2. Ibba M, Soll D (2000) Annu Rev Biochem 69:617–650 3. Bhatt TK, Kapil C, Khan S, Jairajpuri MA, Sharma V et al (2009) BMC Genomics 10:644 4. Bhatt TK, Khan S, Dwivedi VP, Banday MM, Sharma A et al (2011) Nat Commun 2:530 5. Istvan ES, Dharia NV, Bopp SE, Gluzman I, Winzeler EA et al (2011) Proc Natl Acad Sci USA 108:1627–1632 6. Jackson KE, Habib S, Frugier M, Hoen R, Khan S et al (2011) Trends Parasitol 27:467–476 7. Khan S, Sharma A, Jamwal A, Sharma V, Pole AK et al (2011) Sci Rep 1:188 8. Khan S, Garg A, Camacho N, Van Rooyen J, Kumar Pole A et al (2013) Acta Crystallogr D Biol Crystallogr 69:785–795 9. Khan S, Garg A, Sharma A, Camacho N, Picchioni D et al (2013) PLoS One 8:e66224 10. Azcarate IG, Marin-Garcia P, Camacho N, Perez-Benavente S, Puyet A et al (2013) Br J Pharmacol 169:645–658 11. Hoen R, Novoa EM, Lopez A, Camacho N, Cubells L et al (2013) ChemBioChem 14:499–509
71 12. Filisetti D, Theobald-Dietrich A, Mahmoudi N, Rudinger-Thirion J, Candolfi E et al (2013) J Biol Chem 288:36361–36371 13. Koh CY, Kim JE, Napoli AJ, Verlinde CL, Fan E et al (2013) Mol Biochem Parasitol 189:26–32 14. Mailu BM, Ramasamay G, Mudeppa DG, Li L, Lindner SE, Peterson MJ et al (2013) J Biol Chem 288:32539–32552 15. Pham JS, Dawson KL, Jackson KE, Lim EE, Pasaje CF et al (2014) Int J Parasitol Drugs Drug Resist 4:1–13 16. Cattel L, Grove JF, Shaw D (1973) J Chem Soc Perkin 1(21):2626–2629 17. Jacyno JM, Harwood JS, Cutler HG, Lee MK (1993) J Nat Prod 56:1397–1401 18. Zheng H, Zhao C, Fang B, Jing P, Yang J et al (2012) J Org Chem 77:5656–5663 19. Guo M, Ignatov M, Musier-Forsyth K, Schimmel P, Yang XL (2008) Proc Natl Acad Sci USA 105:2331–2336 20. Otwinowski Z, Minor W (1997) Methods Enzymol 276:307–326 21. McCoy AJ, Grosse-Kunstleve RW, Adams PD, Winn MD, Storoni LC, Read RJ (2007) J Appl Crystallogr 40:658–674 22. Adams PD, Afonine PV, Bunkoczi G, Chen VB, Davis IW et al (2010) Acta Crystallogr D Biol Crystallogr 66:213–221 23. Murshudov GN, Skubak P, Lebedev AA, Pannu NS, Steiner RA et al (2011) Acta Crystallogr D Biol Crystallogr 67:355–367 24. Winn MD, Ballard CC, Cowtan KD, Dodson EJ, Emsley P et al (2011) Overview of the CCP4 suite and current developments. Acta Crystallogr D Biol Crystallogr 67:235–242 25. Emsley P, Lohkamp B, Scott WG, Cowtan K (2010) Features and development of coot. Acta Crystallogr D Biol Crystallogr 66:486–501 26. Chen VB, Arendall WB 3rd, Headd JJ, Keedy DA, Immormino RM et al (2010) Acta Crystallogr D Biol Crystallogr 66:12–21 27. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM et al (2004) J Comput Chem 25:1605–1612 28. Wallace AC, Laskowski RA, Thornton JM (1995) Protein Eng 8:127–134 29. Niesen FH, Berglund H, Vedadi M (2007) Nat Protoc 2:2212–2221 30. Wienken CJ, Baaske P, Rothbauer U, Braun D, Duhr S (2010) Nat Commun 1:100 31. Le Guilloux V, Schmidtke P, Tuffery P (2009) BMC Bioinform 10:168 32. Doerig C, Baker D, Billker O, Blackman MJ, Chitnis C et al (2009) Parasite 16:169–182 33. Hora R, Bridges DJ, Craig A, Sharma A (2009) J Biol Chem 284:6260–6269 34. Sharma A, Yogavel M, Akhouri RR, Gill J, Sharma A (2008) J Biol Chem 283:24077–24088 35. Gill J, Yogavel M, Kumar A, Belrhali H, Jain SK et al (2009) J Biol Chem 284:10076–10087
123
