|Home | About | Journals | Submit | Contact Us | Français|
Nerve agents are highly toxic organophosphorus compounds (OP) that are used as chemical warfare agents. Developing a catalytic bioscavenger to efficiently detoxify nerve agents in the bloodstream of affected individuals has been recognized as an attractive approach to prevent nerve agent toxicity. However, the search for nerve agent catalysts has been hindered by the lack of efficient direct assays for nerve agent hydrolysis. In addition, authentic nerve agents are restricted and access to use for experiments by the general research community is prohibited. Herein we report development of a method that combines use of novel nerve agent model compounds possessing a thiocholine leaving group that reacts with the fluorescent thio-detection probe, BES-Thio, to afford detection of sub-micromolar amounts of nerve agent model compounds hydrolysis products. The detection sensitivity of BES-Thio assay was approximately 10 times better than the Ellman assay. This developed method is useful as a direct, sensitive screening method for evaluating OP hydrolysis efficiency from catalytic cholinesterases. When the assay was assembled in the presence of oxime, OP-inhibited cholinesterases that were able to be reactivated by specific oxime showed oxime-assisted enzyme-mediated OP hydrolysis. Therefore, this method is also useful to screen oxime analogs to identify novel agents that can reactivate OP-inhibited cholinesterases or to screen various enzymes to identify pseudo-catalytic bioscavengers that can be readily reactivated by clinically approved oximes.
Nerve agents are highly toxic organophosphorus compounds (OP) that inhibit acetylcholinesterase (AChE). Serum cholinesterase, butyrylcholinesterase (BuChE), is a naturally occurring enzyme that scavenges stoichiometric amounts of nerve agents and protects AChE from low-level OP exposure. Developing a catalytic bioscavenger to efficiently detoxify nerve agents has been an attractive approach to prevent nerve agent toxicity [1, 2]. However, such effort has been hindered by the lack of efficient direct assays for nerve agent hydrolysis. In addition, the use of authentic nerve agents by the general research community is restricted.
The ground-breaking work for identifying nerve agent hydrolytic BuChE variants came from the report on the BuChE variant G117H . However, G117H BuChE hydrolyzes nerve agents slowly (e.g., turnover rate (kcat) for sarin (GB) is 0.004 min−1 ). Due to technical difficulties to study this low amount of nerve agent hydrolyzed, indirect assays involving multiple steps of enzyme manipulation (e.g., inhibition, gel filtration, titration) have been adopted . The procedures are time-consuming and cannot be easily adapted to a high throughput screening format.
We recently reported the chemical synthesis of a series of enantiomerically enriched nerve agent model compounds  by replacing the nerve agent leaving group (i.e., F−, CN− etc.) with thiocholine. These model compounds have been evaluated for cholinesterase inhibition potency and stereoselectivity as well as BuChE covalent adduct formation to validate their application as surrogates for authentic nerve agents [4, 5]. Hydrolysis of the nerve agent model compound liberates a thiol that can be efficiently trapped by a fluorescent thiaphile such as BES-Thio. Herein we report the development of an efficient fluorescent assay that is capable of direct detection of low level hydrolysis of nerve agent model compounds. This approach compliments strategies previously reported, where OP analogs with larger fluorescent leaving groups were used as substrates [6, 7]. The thiocholine leaving group for our analogs is the same as standard cholinesterase substrates acetylthiocholine and butyrylthiocholine (BTCh).
BES-Thio (2,4-dinitrobenzenesulfonyl, 2′,7′-dimethylfluorescein) was synthesized following procedures described previously . The nerve agent model compounds, Sp-O-isopropyl S-(2-trimethylammoniumethyl) methylphosphonothioate iodide (SpGBC) and Sp-O-cyclohexyl S-(2-trimethylammoniumethyl) methylphosphonothioate iodide (SpGFC), were synthesized as previously described [4, 9]. BTCh, 5,5′-dithiobis(2-nitrobenzoic acid) (DTNB), pralidoxime chloride (2-PAM), and recombinant AChE were purchased from Sigma (St Louis, MO). Ecothiophate (ETP), highly purified human BuChE (serving as ELISA standards), and anti-BuChE polyclonal antibodies were gifts from Dr. Oksana Lockridge (University of Nebraska Medical Center, Omaha, NE). HI-6 was a gift from Dr. Douglas Cerasoli (USAMRICD, APG, MD). Recombinant WT and G117H BuChE used in the assay were expressed in CHO cells. Culture medium containing the secreted BuChE enzyme was concentrated using 50 kDa Amicon ultra centrifugal filters (Millipore, Bedford, MA) according to the manufacturer’s protocol. The concentrated enzyme was washed 3 times with Tris-HCl buffer (40 mM, pH 7.4 at 25°C). Enzyme activity of the concentrated enzyme was determined by the Ellman assay . Enzyme concentration was determined using ELISA. All other reagents and buffers used in this work were of analytical grade and commercially available.
A thiocholine stock solution was prepared by hydrolyzing 5 mM BTCh with 0.5 M NaOH for 1 h followed by HCl neutralization. The concentration of thiocholine produced was confirmed with an Ellman assay. The thiocholine solution was serially diluted in 20 mM Tris-HCl buffer (pH 7.4 at 25 °C). BES-Thio stock solution (5 mM) was prepared in DMSO, and then diluted to 50 μM with isopropanol as a working solution. For the calibration curve of the BES-Thio assay, same volume of thiocholine (0.025–12.8 μM) and BES-Thio stock solution in isopropanol was mixed in a flat bottom 96-well black plate (Greiner Bio-one, Frickenhausen, Germany) to achieve a final BES-thio concentration of 25 μM. The fluorescence signal was monitored for 20 min and data were collected every 1 min with a Victor2 plate reader (Perkin Elmer, Waltham, MA) at an Ex/Em of 485nm/535nm. The relative fluorescent units (RFU) observed were plotted versus time after BES-Thio addition. The rate of fluorescent signal development (RFU/min) was calculated from the slope at the linear range of fluorescence development. Both the RFU at specific time points and RFU/min were plotted versus thiocholine concentration to construct standard curves. For comparison, the thiocholine calibration curve using the Ellman’s reagent was done in parallel. The same amount of thiocholine (0.8 to 31.8 μM) and 100 μM DTNB was mixed in clear bottom 96-well plate (Greiner Bio-one). The absorbance was measured at 405 nm on the same plate reader.
For all the BES-Thio assays reported, OP hydrolysis incubations were terminated by addition of BES-Thio to achieve a final concentration of 25 μM and bring the final isopropanol amount to 50%. The fluorescence signal was monitored for 20 min and data were collected every 1 min as described above. A thiocholine standard series was prepared in parallel for all assays on the same assay plate as the samples. The slope of the fluorescence signal versus time plot for calibration standards was calculated and used to construct thiocholine calibration curves. Accordingly, the concentration of thiocholine was determined using these calibration curves for each hydrolysis incubation.
To evaluate the time-dependent hydrolysis of ETP by G117H BuChE, samples comprising of 0.43 μM G117H and 0.4 mM ETP were incubated in 20 mM Tris-HCl buffer (pH 7.4) in a 96-well plate format for various lengths of time (i.e., 5, 10, 20, 30, 45, 60, 90, 120, and 240 min) at room temperature. The formation of thiocholine from each incubation was determined with the BES-Thio method described above. The background fluorescence contributed to the overall signal from samples containing no substrate was subtracted.
To evaluate substrate-dependent ETP hydrolysis by G117H, 0.43 μM of G117H was incubated with ETP at various concentrations (i.e., 20, 40, 80, 140, 400, and 1000 μM) in 20 mM Tris-HCl buffer (pH 7.4) at room temperature for 1 h. Thiocholine production from each incubation was determined with the BES-Thio method described above. A plot of thiocholine producing rate versus ETP concentration afforded a hyperbolic curve that was fitted to the Michaelis-Menten equation to determine Km and kcat.
Hydrolysis of SpGBC and SpGFC by G117H BuChE was examined by incubating 0.4 mM nerve agent model compound with 0.43 μM G117H in 20 mM Tris-HCl buffer (pH 7.4) at room temperature for 4 h. Thiocholine production from each incubation was determined with the BES-Thio method described above. For comparison, single turnover of SpGBC and SpGFC by wild type BuChE was examined in parallel using the same assay.
The oxime-assisted hydrolysis of SpGBC by AChE or BuChE was examined using 2-PAM or HI-6, respectively. Enzyme (0.25 μM AChE or BuChE), 0.1 mM SpGBC and 0.5 mM oxime (2-PAM or HI-6) were incubated in 20 mM Tris-HCl buffer solution (pH 7.4). At certain time points (i.e., 10, 30, 60, and 120 min), thiocholine production from each incubation was determined with the BES-Thio method described above.
All linear and non-linear regression analyses were done using Graphpad Prism Programs (Version 3.00, Graphpad, Inc., La Jolla, CA) to obtain the best fit values and standard errors. Z′ values were calculated to examine if the signal was reliably above background using a method previously reported .
BES-Thio has previously been reported to show detection of glutathione, cysteine, and other free thiols in the picomole range . For results described herein, detection of hydrolysis of nerve agent model substrates including SpGFC, SpGBC, and ETP (Figure 1) required monitoring the release of the thiocholine leaving group by forming a fluorescent product with BES-Thio (Figure 1). We first examined the chemical stability of thiocholine in the assay format described above (data not shown). Although not indefinitely stable, preparing fresh stock solutions of thiocholine and including parallel standard curves for all assays afforded reliable data. Because different free thiols have different reaction rates with BES-Thio , the incubation of thiocholine with BES-Thio was initially investigated to evaluate overall assay sensitivity. As shown in Figure 2A, the reaction of BES-Thio with thiocholine produced a linear increase in fluorescent signal for the initial few minutes and the slope of the fluorescent signal correlated with the concentration of thiocholine in each incubation.
RFU at 1 and 10 min were plotted versus thiocholine concentration (Figure 2B) and showed thiocholine-dependent fluorescence intensity. Z′ values were calculated from the signal at specific thiocholine concentration over background fluorescence levels . Z′>0.5 was set as the threshold for the determination of the assay lower detection limit. The lower detection limit improved from 12.8 μM to 1.6 μM when the fluorescent signal was recorded at 10 min instead of 1 min after BES-Thio addition (Figure 2B). Linear regression analysis of the initial time-dependent increase in fluorescence intensity (i.e., 0–7 min) afforded the slope (i.e., RFU/min) as an indicator of fluorescent product formation rate (Figure 2A). A plot of each slope versus the thiocholine concentration showed an improved concentration-dependent linearity at the low thiocholine concentration range (i.e., R2=0.995 for a thiocholine concentration between 0.2–12.8 μM) (Figure 2B). Using this method, the accuracy and precision of the assay improved as reflected in an increase in Z′ values (i.e., Z′ of 0.80–0.89 for slopes based analysis versus a Z′ value of 0.54–0.79 for a single time point based analysis (i.e.,10-min RFU) for thiocholine concentrations above 1.6 μM). The assay sensitivity was also improved to 0.4 μM (Z′=0.55) when calibration curves were generated based on slope. A comparison of standard curves obtained using the Ellman’s reagent showed that the BES-Thio assay had ~10-fold greater detection sensitivity than the Ellman assay (data not shown). Thereafter, freshly prepared standard calibration curves were generated based on fluorescent product formation rate, RFU/min, for all the following assays and provided a sensitive means to quantify the hydrolysis of the target OP compounds with the BES-Thio method.
To explore the utility of the BES-Thio assay for OP hydrolysis detection, time-dependent ETP hydrolysis by catalytic BuChE variants G117H was examined. A plot of thiocholine formation versus incubation time showed a linear correlation during the first hour of incubation (Figure 3A). The rate of product formation became non-linear afterward, possibly due to a combination of factors including progressive aging of the ETP phosphonylated enzyme and/or loss of enzyme activity due to enzyme instability associated with extended incubation times. Substrate-dependent ETP hydrolysis was evaluated using a 1 h incubation period and a G117H BuChE concentration of 0.43 μM. A fit of the product formation data to the Michaelis-Menten equation afforded a kcat of 0.36 min−1 and a Km of 42.7 μM (Figure 3B). These values are slightly lower than the reported value of kcat 0.75 min−1 and Km of 73 μM using purified G117H BuChE at 3 μM to examine ETP hydrolysis using a traditional Ellman assay . The overall catalytic efficiency kcat/Km for ETP hydrolysis by G117H BuChE was comparable to the reported value (8.6 versus 10.3 mM−1min−1) .
The direct hydrolysis of GF and GB model compounds SpGFC and SpGBC by G117H BuChE was examined (Figure 4A). After 4 h incubation, thiocholine produced from SpGFC and SpGBC was 1.3 and 1.0 μM respectively, corresponding to 2–3 turnovers for each substrate. When wild type BuChE was examined under the same assay conditions, a stoichiometric amount of thiocholine was formed from an incubation of 0.43 μM WT BuChE with SpGFC and SpGBC, producing 0.45 and 0.42 μM thiocholine, respectively. Because the rate of SpGFC and SpGBC hydrolysis catalyzed by G117H was ~100-fold lower than the hydrolysis of ETP, the specific activity of enzyme-mediated hydrolysis could not be reliably estimated from these assays. However, based on the reported hydrolysis rate, a 4 h incubation of GB with G117H resulted in an estimated single turn-over (i.e., 0.24 h−1 × 4 h = 0.96)  plus the initial stoichiometric turnover based on phosphylation, the resultant estimated 2 turnover for G117H-mediated GB hydrolysis was in good agreement with the results reported herein for SpGBC. Despite the limitations on the determination of specific activity for OP hydrolysis by G117H, the direct BES-Thio assay was useful to verify whether multiple turnover (i.e., hydrolysis) for OP substrates could occur for cholinesterase variants and provide a reliable means of ranking OP hydrolysis rates for multiple enzymes or enzyme variants in a high throughput format.
We also extended the analysis to oxime-assisted hydrolysis of SpGBC by wild type AChE or BuChE. Traditionally, oxime-reactivation of OP-inhibited cholinesterase has been assayed through detection of regenerated cholinesterase activity following complete enzyme inhibition by excess OP inhibitors and then removal of excess OP inhibitors through gel filtration. The BES-Thio assay system described above was used to directly examine HI-6-and 2-PAM-mediated SpGBC hydrolysis in a single step by both human wild type AChE and wild type BuChE, respectively. As shown in Figure 4B, HI-6 assisted hydrolysis of SpGBC by WT AChE was most prominent. 2-PAM assisted SpGBC hydrolysis by WT AChE and BuChE and HI-6 assisted hydrolysis by WT BuChE was detectably above background but relatively low. This finding is in good agreement with the previously reported oxime-mediated reactivation of sarin-inactivated AChE and BuChE .
The method reported herein involves the combined application of novel nerve agent model compounds possessing a thiocholine leaving group that can be combined with the fluorescent thio-detection probe, BES-Thio, to afford direct detection of sub micromolar hydrolysis of nerve agent model compounds. The method is useful as a sensitive high-throughput screening method for evaluating OP hydrolysis efficiency from cholinesterase variants or other OP catalytic enzymes. The assay can be used to efficiently identify oximes that can reactivate cholinesterases to afford net enzyme-mediated OP hydrolysis. The oxime-assisted assay can also be used for functional screening of enzyme variants readily reactivated by clinically approved oximes, thereby identifying pseudo-catalytic bioscavenger variants.
We thank Dr. Cynthia Gilley for assistance with the chemical synthesis of the nerve agent model compounds. We thank Dr. Oksana Lockridge (University of Nebraska Medical Center, Omaha, NE) and Dr. Douglas Cerasoli (USAMRICD, APG, MD) for providing essential reagents and helpful comments. This work was supported in part by the National Institute of Health grants U54NS058183 Project 2 to JZ and U01NS058038 to JRC.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.