Adduct formation in phosphate buffer (shown here with GF)

We have recently reported about the formation of stable adducts of G-type nerve agents like Sarin, Soman and Cyclosarin with buffering compounds that are aminoalcohols like TRIS, TES or HEPES. The formation of the phosphorus diester adduts that are stable to hydrolysis is dependent on the concentrations of both agent and buffer compound. We have recommended to avoid these buffer compounds for analytical work and resort to buffer compounds like MES, MOPS and CHES that do not form adducts.

An other buffer compund that is widely employed in biological, biochemical and medical research, especially when pH is to be controlled at the physiological pH of 7.4, is inorganic phosphate. We were interested if the phosphate species present at this pH can react as nucleophiles with G-type nerve agents an form adducts as well. This is indeed the case. We were able to show that agents hydrolyse much more quickly in phosphate buffer than in MOPS buffer at the same pH. We were also able to to detect the formation of significant amounts of pyro-phosphate like adducts (phosphorylated methylphosphonates). These hydrolyzed slowly with a kinetic following pseudo-0th order. This resulted in a complex mixture of phosphorus containing species with changing concentrations over time. The molecular structures of these adducts were determined by 1D 1H–31P HSQC NMR and LC–ESI-MS/MS techniques. The rates of formation of the adducts is similar to the the rate of hydrolysis of the agents (leading to the primary hydrolysis products) and leads to the accumulation of significant amounts of the adducts within just a few minutes.

We propose the hydrogen phosphate ion to be the active nucleophile. The other dominating species at pH 7.4 is dihyrogen phosphate and is less nucleophilic. The observed pseudo-0th order hydrolysis kinetic of the adducts can be explained by the fact that only a very small amount is present as a neutral species at pH 7.4 , while the anionic species is protected from hydrolysis (like for example phosphorus diesters) and is present in large excess. This leads to a constant concentration of the neutral species for a long time and therefore results in a pseudo-0th order kinetic. For the competing hydrolysis reaction leading to the primary hydrolysis products we assume that the hydrogenphosphate dianion functions as a base and the reaction is base catalyzed, leading to rate increase compared to the rate determined in MOPS buffer at the same pH.

Formation of pyrophosphate-like adducts from nerve agents sarin, soman and cyclosarin in phosphate buffer: Implications for analytical and toxicological investigations.
Gäb J, John H, Blum MM.
Toxicol. Lett. 2011; 200(1):34-40.
http://dx.doi.org/10.1016/j.toxlet.2010.10.011

Abstract:
Phosphate buffer is frequently used in biological, biochemical and biomedical applications especially when pH is to be controlled around the physiological value of 7.4. One of the prerequisites of a buffer compound among good buffering capacity and pH stability over time is its non-reactivity with other con- stituents of the solution. This is especially important for quantitative analytical or toxicological assays. Previous work has identified a number of amino alcohol buffers like TRIS to react with G-type nerve agents sarin, soman and cyclosarin to form stable phosphonic diesters. In case of phosphate buffer we were able to confirm not only the rapid hydrolysis of these agents to the respective alkyl methylphosphonates but also the formation of substantial amounts of pyrophosphate-like adducts (phosphorylated methylphos- phonates), which very slowly hydrolyzed following zero-order kinetics. This led to a complex mixture of phosphorus containing species with changing concentrations over time. We identified the molecular structure of these buffer adducts using 1D 1H–31P HSQC NMR and LC–ESI-MS/MS techniques. Reaction rates of adduct formation are fast enough to compete with hydrolysis in aqueous solution and to yield substantial amounts of buffer adduct over the course of just a couple of minutes. Possible reaction mechanisms are discussed with respect to the formation and subsequent hydrolysis of the pyrophosphate-like compounds as well as the increased rate of hydrolysis of the nerve agent to the corresponding alkyl methylphosphonates. In summary, the use of phosphate buffer for the development of new assays with sarin, soman and cyclosarin is discouraged. Already existing protocols should be carefully reexamined on an individual basis.

Different reaction mechanisms discussed for DFPase

Knowledge about the reaction mechanism of the enzyme DFPase is a crucial prerequisite for successful directed protein engineering because the mechanism determines the orientation of the substrates in the binding pocket of the enzyme for catalytic turnover. Also residues important for the mechanism can be specifically optimized. For DFPase three different mechanisms were discussed in the past.

When the first X-ray structure of DFPase was published in 2001 (Scharff et al., Structure 9 (2001) 493-502) residue H287 was found to be part of the enzyme’s binding pocket. Mutant H287N only retained minimal residual activity and it was therefore concluded that H287 is activating a water molecule for nucleophilic attack on the substrate’s phosphorus atom. The substrate itself is activated by coordination to the catalytic calcium ion via the phosphoryl oxygen (top scheme in the figure). As mutants like H287F ratain almost full catalytic activity this mechanism was refuted.

It was alternatively proposed that the calcium ion in the catalytic binding site of DFPase activates a directly coordinated water molecule (resulting in a coordinated hydroxide species). This hydroxide ion would then act as the nucleophile to attack the phosphorus atom of the substrate that is also coordinated to the calcium ()middle scheme in the figure). This mechanism was refuted based on the neutron diffraction structure of DFPase that clearly reveals the identity of the coordinated water as a water molecule and not as hydroxide. It is important to mention in this context that the neutron data (and the respective X-ray data for joint refinement) were recorded at room temperature, which is the relevant temperature for catalytic activity. The neutron structure is however compatible to a third mechanism, which was proposed based on isotope labeling, mutational studies and the structure of a protein-inhibitor complex.

Special issue of Acta D

This mechanism (Blum et al., JACS 128 (2006) 12750-12757) identifies the calcium coordinating residue D229 as the active nucleophile. Als an intermediate an instable high-energy phospho-enzyme species is generated, which is subsequently hydrolyzed by water, regenerating the enzyme and releasing the product (bottom scheme in the figure).

The results of the neutron diffraction experiments with DFPase as well as the results of mutational and kinetic studies were now related to each other in the journal Acta Cryst. D. The article is part of a special issue with the title “Neutrons in Biology“. Even though all articles of the issue are worth reading one article is especially recommended: Benno P. Schoenborn, who published the first neutron diffraction structure of a protein (myoglobin) at the end of the 1960s, offers a fascinating overview over more than fourty years of history of the use of neutron in biomolecular research in his article “A history of neutrons in biology: the development of neutron protein crystallography at BNL and LANL“.

Neutron structure and mechanistic studies of diisopropyl fluorophosphatase (DFPase).
Blum MM, Tomanicek S, John H, Hanson L, Rüterjans H, Schoenborn BP, Langan P, Chen JC.
Acta Crystallogr D Biol Crystallogr. 2010; 66(11):1131-1138.
http://dx.doi.org/10.1107/S0907444910034013

Article Reprint for Download (PDF)

Abstract:
Diisopropyl fluorophosphatase (DFPase) is a calcium-dependent phosphotriesterase that acts on a variety of highly toxic organophosphorus compounds that act as inhibitors of acetylcholinesterase. The mechanism of DFPase has been probed using a variety of methods, including isotopic labelling, which demonstrated the presence of a phosphoenzyme intermediate in the reaction mechanism. In order to further elucidate the mechanism of DFPase and to ascertain the protonation states of the residues and solvent molecules in the active site, the neutron structure of DFPase was solved at 2.2 Å resolution. The proposed nucleophile Asp229 is deprotonated, while the active-site solvent molecule W33 was identified as water and not hydroxide. These data support a mechanism involving direct nucleophilic attack by Asp229 on the substrate and rule out a mechanism involving metal-assisted water activation. These data also allowed for the re-engineering of DFPase through rational design to bind and productively orient the more toxic S stereoisomers of the nerve agents sarin and cyclosarin, creating a modified enzyme with enhanced overall activity and significantly increased detoxification properties.

Plot of C-α RMSD values between d-DFPase and h-DFPase

Protein structures solved by neutron diffraction have the significant advantage compared to X-ray structures that hydrogen atoms are clearly observable in the nuclear density maps. This allows the determination of protonation states in amino acid side-chains and the orientation of solvent molecules (especially water) in space. This is of special importance for the elucidation of enzyme mechanisms. The disadvantages of neutron diffraction with proteins exist in the small number of powerful neutron sources and dedicated instruments worldwide. Also – neutron flux even at the most powerful sources is small compared to photon flux at X-ray sources. Therefore large protein crystals are required and data collection times can easily be in the region of several weeks. Another problem is the signal to noise ratio. Hydrogen atoms in the crystal are major contributors to this problems because the large incoherent scattering cross section of hydrogen. High values for this incoherent scattering cross section lead to diffuse scattering and negatively influence the signal to noise ratio. Deuterium on the other hand displays a significantly smaller value (2.05 b for deuterium compared with 80.27 b for normal hydrogen; 1 b = 100 fm²). To overcome this problem, crystals grown with hydrogenous protein are normally soaked with deuterated mother liquor (or brought in contact via gas diffusion).

Cover of Acta F with DFPase

Labile hydrogens (e.g. those of water in the solvent of in acidic or basic functional groups in the protein) are exchanged with deuterium. Non-labile hydrogens like those in aliphatic or aromatic C-H bonds are no exchanged. Therefore a significant number of hydrogen atoms remain in the protein. Such a partially exchanged crystal was used for the already published neutron structure of DFPase.

To achieve full deuteration of the protein it has to be grown in fully deuterated media. We now report the X-ray structure of fully deuterated DFPase in a new publication in the journal Acta Cryst. F. The structure (solved at room temperature at a resolution of 2.1 Å) shows that full deuteration leads to practically no changes in the protein structure. But even though a very large crystal of d-DFPase was grown (> 2mm³) it did not diffract neutrons beyond very low resolution. An explanation for this unexpected result can be found either in the differences in data aquisition (cross section of the neutron beam compared to the X-ray beam used) or can be based on crystallographic parameters. The scattering characteristics of successful neutron experiments and associated X-ray data are presented in tabulated form and can serve as a guidance for future neutron experiments.

X-ray structure of perdeuterated diisopropyl fluorophosphatase (DFPase): Perdeuteration of proteins for neutron diffraction.
Blum MM, Tomanicek S, John H, Hanson L, Rüterjans H, Schoenborn BP, Langan P, Chen JC.
Acta Crystallogr. Sect. F Struct. Biol. Cryst. Commun. 2010; 66(4):379-385.
http://dx.doi.org/10.1107/S1744309110004318

 Article Reprint for Download (PDF)

Abstract:
The signal-to-noise ratio is one of the limiting factors in neutron macromolecular crystallography. Protein perdeuteration, which replaces all H atoms with deuterium, is a method of improving the signal-to-noise ratio of neutron crystallography experiments by reducing the incoherent scattering of the hydrogen isotope. Detailed analyses of perdeuterated and hydrogenated structures are necessary in order to evaluate the utility of perdeuterated crystals for neutron diffraction studies. The room-temperature X-ray structure of perdeuterated diisopropyl fluorophosphatase (DFPase) is reported at 2.1 Å resolution. Comparison with an independently refined hydrogenated room-temperature structure of DFPase revealed no major systematic differences, although the crystals of perdeuterated DFPase did not diffract neutrons. The lack of diffraction is examined with respect to data-collection and crystallographic parameters. The diffraction characteristics of successful neutron structure determinations are presented as a guideline for future neutron diffraction studies of macromolecules. X-ray diffraction to beyond 2.0 Å resolution appears to be a strong predictor of successful neutron structures.

Catalytic calcium binding-site of DFPase

The enzyme DFpase is structurally well characterized. An atomic resolution X-ray structure and a neutron structure are available. In addition to this a number of DFPase mutants were generated and for some of them X-ray structures were determined. What was missing up to now was a systematic investigation of the calcium binding-site of DFPase with respect to catalytic activity and the ability for metal binding. A summary of existing data and three new X-ray structures of mutants changing residues in the calcium binding-site were now published in a paper in the journal Chemico-Biological Interactions. We also deduce rules for activity and metal biding.

A comparison with the structurally related proteins Paraoxonase (PON1), Drug Resistance Protein 35 (Drp35) from S. aureus or the Gluconolactonase XC5397 from Xanthomonas campestris reveal calcium binding sites with highly similar topology but in part different enzymatic activities (PON1 is also a phosphotriesterase but the native substrates of PON1, Drp35 und XC5397 seem to be lactones). The possible applicability of the rules deduced for DFPase to the other proteins is discussed.

Structural characterization of the catalytic calcium binding site in diisopropyl fluorophosphatase (DFPase) and comparison with related β-propeller enzymes.
Blum MM, Chen JC.
Chem. Biol. Interact. 2010; 187(1-3):373-379.
http://dx.doi.org/10.1016/j.cbi.2010.02.043

Abstract:
The calcium-dependent phosphotriesterase diisopropyl fluorophosphatase (DFPase) from the squid Loligo vulgaris efficiently hydrolyzes a wide range of organophosphorus nerve agents. The two calcium ions within DFPase play essential roles for its function. The lower affinity calcium ion located at the bottom of the active site participates in the reaction mechanism, while the high affinity calcium in the center of the protein maintains structural integrity of the enzyme. The activity and structures of three DFPase variants targeting the catalytic calcium-binding site are reported (D121E, N120D/N175D/D229N, and E21Q/N120D/N175D/D229N), and the effect of these mutations on the overall structural dynamics of DFPase is examined using molecular dynamics simulations. While D229 is crucial for enzymatic activity, E21 is essential for calcium binding. Although at least two negatively charged side chains are required for calcium binding, the addition of a third charge significantly lowers the activity. Furthermore, the arrangement of these charges in the binding site is important for enzymatic activity. These results, together with earlier mutational, structural, and kinetic studies, show a highly evolved calcium-binding environment, with a specific electrostatic topology crucial for activity. A number of structural homologues of DFPase have been recently identified, including a chimeric variant of Paraoxonase 1 (PON1), drug resistance protein 35 (Drp35) from Staphylococcus aureus and the gluconolactonase XC5397 from Xanthomonas campestris. Surprisingly, despite low sequence identity, these proteins share remarkably similar calcium-binding environments to DFPase.

Mechanism of adduct formation

Buffer compounds are used to maintain a certain pH in solution. Besides buffer capacity and pH stability over prolonged times the non-reactivity with other components of the solution is an important factor for buffer selection. In a new publication in the Journal of Chrmatography B we report the formation of stable adducts of nerve agents like Sarin, Soman or Cyclosain with common and widely employed buffer compounds like TRIS, TES or HEPES. The reaction proceeds in competition with spontaneous hydrolysis in aqueous solution and yields can reach up to 40% based on the organophosphate/phosphonate present. Highest yields are achieved with high buffer concentrations and high pH.

Using a recently presented NMR method to monitor the degradation of nerve agents by enzymes in buffered solution we were able to observe the formation of new, phosphorus containing and stable compounds. Using LC-ESI-MS/MS we were able to show that the compounds are adducts of the buffer and the nerve agent. Buffer compounds like TES or TRIS can act both as nitrogen nucleophiles (via the amino group) and as oxygen nucleophiles (via the oxygen atoms of the hydroxyl groups). The identification of the adducts as phosphodiesters (“O-adducts”) was finally achieved by NMR spectroscopy.

As a potential reaction mechanism we propose that the amino group of the buffer acts as an intramolecular proton acceptor, which can accept a proton from a hydroxyl group of the buffer. This increases the nucleophilicity of the hydroxyl oxygen atom attacking the phosphorus atom of the warfare agent leading to the formation of a phosphodiester (with organophosphonates like Sarin, Soman and Cyclosarin). As alternative buffer compounds for work with nerve agents we propose MOPS (pK = 7.2), CHES (pK = 9.3) und MES (pK = 6.15). These compounds do not contain a combination of hydroxyl and amino groups and do not show any adducts formation in solution.

Stable adducts of nerve agents sarin, soman and cyclosarin with TRIS, TES and related buffer compounds–Characterization by LC-ESI-MS/MS and NMR and implications for analytical chemistry.
Gäb J, John H, Melzer M, Blum MM.
J. Chromatogr. B 2010; 878(17-18):1382-1390.
http://dx.doi.org/10.1016/j.jchromb.2010.01.043

Abstract:
Buffering compounds like TRIS are frequently used in chemical, biochemical and biomedical applications to control pH in solution. One of the prerequisites of a buffer compound, in addition to sufficient buffering capacity and pH stability over time, is its non-reactivity with other constituents of the solution. This is especially important in the field of analytical chemistry where analytes are to be determined quantitatively. Investigating the enzymatic hydrolysis of G-type nerve agents sarin, soman and cyclosarin in buffered solution we have identified stable buffer adducts of TRIS, TES and other buffer compounds with the nerve agents. We identified the molecular structure of these adducts as phosphonic diesters using 1D 1H-31P HSQC NMR and LC-ESI-MS/MS techniques. Reaction rates with TRIS and TES are fast enough to compete with spontaneous hydrolysis in aqueous solution and to yield substantial amounts (up to 20-40%) of buffer adduct over the course of several hours. A reaction mechanism is proposed in which the amino function of the buffer serves as an intramolecular proton acceptor rendering the buffer hydroxyl groups nucleophilic enough for attack on the phosphorus atom of the agents. Results show that similar buffer adducts are formed with a range of hydroxyl and amino function containing buffers including TES, BES, TRIS, BIS-TRIS, BIS-TRIS propane, Tricine, Bicine, HEPES and triethanol amine. It is recommended to use alternative buffers like MOPS, MES and CHES when working with G-type nerve agents especially at higher concentrations and over prolonged times.

A new publication in Analytical and Bioanalytical Chemistry describes the use of 1H-31P HSQC NMR spectroscopy to monitor of the degradation of highly toxic organophosphorus compounds by the enzyme DFPase. The method can be used for methylphosphonates, a group of compounds including nerve agents sarin (GB), soman (GD), cyclosarin (GF) and also VX. The limit of quantitation (LOQ) of the method is around 100 μM when using a 400 MHz NMR spectrometer. The work is founded on previous results from Koskela et al. (Koskela et al., Anal. Chem. 2007; 79:9098-9106) who were able to use the method to detect agents and hydrolysis products in comlex decontamination fluids. We were now able to show that the procedure works not only in the static case but also for dynamic reaction monitoring.

The method is of special relevance for reaction monitoring in complex media. We were able to show that monitoring is also possible in multi-phase systems by using a biodiesel based bicontinuous microemulsion as a model system. Other methods like pH-stat titration or the use of fluoride sensitive electrodes regularly fail in these complex fluids.

Monitoring the hydrolysis of toxic organophosphonate nerve agents in aqueous buffer and in bicontinuous microemulsions by use of disopropyl fluorophosphatase (DFPase) with 1H-31P HSQC NMR spectroscopy.
Gäb J, Melzer M, Kehe K, Wellert S, Hellweg T, Blum MM.
Anal. Bioanal. Chem. 2009;396(3):1213-1221.
http://dx.doi.org/10.1007/s00216-009-3299-2

Abstract:
The enzyme diisopropyl fluorophosphatase (DFPase, EC 3.1.8.2) from the squid Loligo vulgaris effectively catalyzes the hydrolysis of diisopropyl fluorophosphate (DFP) and a number of organophosphorus nerve agents, including sarin, soman, cyclosarin, and tabun. Until now, determination of kinetic data has been achieved by use of techniques such as pH-stat titration, ion-selective electrodes, and a recently introduced method based on in situ Fourier-transform infrared (FTIR) spectroscopy. We report the use of 1D 1H-31P HSQC NMR spectroscopy as a new method for real-time quantification of the hydrolysis of toxic organophosphonates by DFPase. The method is demonstrated for the agents sarin (GB), soman (GD), and cyclosarin (GD) but can also be used for V-type nerve agents, for example VX. Besides buffered aqueous solutions the method was used to determine enzymatic activities in a biodiesel-based bicontinuous microemulsion that serves as an example of complex decontamination media, for which other established techniques often fail. The method is non-invasive and requires only limited manual handling of small volumes of liquid (700 μl), which adds to work safety when handling highly toxic organophosphorus compounds. Limits of detection are slightly below 100 μM on a 400 MHz spectrometer with 16 FIDs added for a single time frame. The method is not restricted to DFPase but can be used with other phosphotriesterases, for example paraxonase (PON), and even reactive chemicals, for example oximes and other nucleophiles, as long as the reaction components are compatible with the NMR experiment.

The enzyme Diisopropyl fluorophosphatase (DFPase) catalyses the hydrolysis of the toxic organophosphorus compound Diisopropyl fluorophosphat (DFP) and a range of highly toxic nerve agents of the so called G-series. This class of nerve agents includes compounds like tabun (GA), sarin, (GB), soamn (GD) and cyclosarin (GF). DFP does not contain any stereocenters and is achiral but all mentioned G-type nerve agents have four different substitutents at the central phosphorus atom. Therefore this phosphorus atom is asymmetric and forms a stereocenter. GA, GB, GF are chiral and exist as pair of enantiomers, which relate to each other like left hand and right hand (GD is a more complicated case as there is an additional stereocenter in one of the side chains).

There is no difference between the enantiomers in chemical reactions that take place in an achiral environment. An example for this is the aqueous hydrolysis of the agents at high pH. But when interacting with other chiral molecules like proteins they diplay distinct differences. The main “target” for the nerve agents is the enzyme acteylcholinesterase (AChE), which is inhibited by the nerve agents by the formation of a stable covalent adduct. One of the two enantiomers shows a significantly higher inhibitory power compared to the other, which is only mildly inhibiting or even non-inhibiting. A similar effect is seen when hydrolyzing (and detoxifying) the agents using DFPase as an enzymatic catalyst. Wildtype DFPase prefers the less toxic enantiomer. This is especially problematic for medical application (in vivo or topical) as rapid reduction of toxicity is of special importance. A potential strategy to reverse enantioselectivity of DFPase is to alter the active site of the enzyme by mutagenesis.

Model of the phosphoenzyme intermediate
of DFPase for the substrate sarin

Due to the legal restrictions for handling highly toxic compounds* the standard approach for mutagenesis using evolutionary methods and high-throuput screening of mutant libraries could not be used. Instead we turned to rational protein design. One should note that reversing enantioselectivity of an enzyme (and keeping activity) is not a trivial task and the number of publications dealing with this issues is still limited. Detailed knowledge about the DFPase structure and the reaction mechanism, which is though to proceed via a phosphoenzyme intermediate, formed an invaluable base for planning the mutations.

The constructed mutants finally displayed the desired reversed enantioselectivity without loosing enzymatic activity. The mutants rather showed enhanced activities (up to 8-fold higher in case of GB). For an optimal combination of selectivity , activity and substrate affinity the mutants should be mixed with wildtype DFPase. Using such mixtures total hydrolysis of the agents can be achieved 4-fold faster even at low substrate concentrations (where substrate affinity is an issue) with an even faster reduction in toxicity as in the first phase of the reaction the more toxic enantiomer is hydrolyzed faster.

Structures of the mutants were deposited in the PDB under accession codes 3HLI and 3HLH.

* Work with nerve agents was conducted in accordance with the Chemical Weapons Convention (CWC) at the Bundeswehr Institute for Pharmacology and Toxicology,

Reversed Enantioselectivity of Diisopropyl Fluorophosphatase against Organophosphorus Nerve Agents by Rational Design.
Melzer M, Chen JC, Heidenreich A, Gäb J, Koller M, Kehe K, Blum MM.
J. Am. Chem. Soc. 2009;131(47):17226-17232.
http://dx.doi.org/10.1021/ja905444g

Research Highlight in Nature Chemistry:
http://dx.doi.org/110.1038/nchem.487

Abstract:
Diisopropyl fluorophosphatase (DFPase) from Loligo vulgaris is an efficient and robust biocatalyst for the hydrolysis of a range of highly toxic organophosphorus compounds including the nerve agents sarin, soman, and cyclosarin. In contrast to the substrate diisopropyl fluorophosphate (DFP) the nerve agents possess an asymmetric phosphorus atom, which leads to pairs of enantiomers that display markedly different toxicities. Wild-type DFPase prefers the less toxic stereoisomers of the substrates which leads to slower detoxification despite rapid hydrolysis. Enzyme engineering efforts based on rational design yielded two quadruple enzyme mutants with reversed enantioselectivity and overall enhanced activity against tested nerve agents. The reversed stereochemical preference is explained through modeling studies and the crystal structures of the two mutants. Using the engineered mutants in combination with wild-type DFPase leads to significantly enhanced activity and detoxification, which is especially important for personal decontamination. Our findings may also be of relevance for the structurally related enzyme human paraoxonase (PON), which is of considerable interest as a potential catalytic in vivo scavenger in case of organophosphorus poisoning.