Quantification of Prothrombin in Human Plasma Amplified by

Feb 10, 2012 - Autocatalytic Reaction. Ana Virel, Laura Saa, and Valeri Pavlov*. CIC biomaGUNE, Parque Tecnológico de San Sebastián, Paseo Miramón ...
4 downloads 0 Views 3MB Size
Article pubs.acs.org/ac

Quantification of Prothrombin in Human Plasma Amplified by Autocatalytic Reaction Ana Virel, Laura Saa, and Valeri Pavlov* CIC biomaGUNE, Parque Tecnológico de San Sebastián, Paseo Miramón 182, 20009, San Sebastián, Spain S Supporting Information *

ABSTRACT: By site directed mutagenesis, we have produced recombinant mutants of human and mouse prethrombin-2 which are able to convert themselves autocatalytically into α-thrombin. We also have created a new method to amplify the signal of bioanalytical assays based on the autocatalytic activation of these mutated proenzymes. The activation of the mutants by active α-thrombin triggers an autocatalytic reaction which leads to more active thrombin resulting in the amplification of the readout signal. Addition of mutated mouse prethrombin-2 into the conventional assay for prothrombin level in human plasma, employing ecarin and the fluorogenic substrate, resulted in improvement of the detection limit by 2 orders of magnitude.

S

detection limit of enzyme assays, there are still no commercially available kits based on natural autocatalytic proenzymes. For development of our autocatalytic assays, we have chosen the enzyme α-thrombin, a serine protease that selectively cleaves two Arg-Gly bonds in fibrinogen.14 α-Thrombin is a crucial enzyme involved in the blood coagulation cascade and is also related to the inflammatory response and tissue repair.15 Tests for thrombin16−21 are used to evaluate the rate of blood coagulation; therefore, it is important to develop sensitive methods to monitor prethrombin and thrombin activities. αThrombin is used in surgery as a very potent topical hemostat when surgical ligation of bleeding fails. It helps also barbers, boxers, and soldiers in hemostasis of cuts and wounds. The employment of α-thrombin in hospitals dates back to the 1940s. Currently, thrombin is utilized in more than 1 000 000 patients in the U.S. each year.22 The main source of α-thrombin is pooled human plasma collected at licensed collection centers from donors. The plasma is treated by complicated filtration and separation steps including vapor heat treatment, ultrafiltration, solvent-detergent treatment, and lyophilization. However, no procedure is completely effective against viral particles derived from human plasma. Therefore, there is no 100% assurance that α-thrombin obtained from donor blood is free of pathogens.23 Alternatively, recombinant α-thrombin can be produced that is devoid of the risks of blood-borne viral particles. Thrombin is naturally produced in the form of a precursor called prothrombin which can be cleaved by factor Xa or by ecarin24 at two places, an Arg-Thr and Arg-Ile bond. When prothrombin is only cleaved at the first factor Xa site, the outcome is

ignal amplification of a biorecognition event is of great importance in analytical chemistry. Enzymes are commonly used in different bioanalytical assays for detection and amplification of signal. They are employed in quantification of glucose,1 H2O2,2 pesticides,3 cholesterol,4 and ethanol5 and are the basis for enzyme linked immunosorbent assays (ELISAs).6 The effectiveness of enzyme-based bioanalytical methods depends on the affinity and selectivity of the enzyme for its substrate and also on the rate of the enzymatic reaction. Enzyme amplification techniques7 have been used to improve the sensitivity of several bioanalytical assays. The principle of enzyme amplification is based on the presence of a primary enzyme label which creates a triggering component for a second system that will generate a large amplified signal.8 We have previously reported the double amplification cascade in which ecarin, a protease found in snake poison, converted prothrombin to α-thrombin to digest an artificial fluorogenic substrate.9 Prothrombin is one of the proteolytic enzymes. These enzymes are normally synthesized in their inactive form, known as proenzymes or zymogens.10 When the product of the proenzyme cleavage reaction catalyzes the same reaction, the process is called an autocatalytic activation.11 Some examples of natural autocatalytic enzymes are trypsinogen, pepsinogen, or the blood coagulation factor XII.12 The autocatalytic behavior of these enzymes could be applied for analytical purposes. The signal obtained by a low concentration of an enzyme can be considerably amplified by means of an autocatalytic reaction simply by adding an excess of its zymogen. Unfortunately, the use of zymogens in bioanalysis is limited since all known natural autocatalytic proenzymes are unstable in vitro, their preparations from their natural sources contain the corresponding active enzymes,13 and they normally have low affinity for their proenzymes. To the best of our knowledge, despite the obvious belief that the use of autocatalytic proenzymes could significantly improve the © 2012 American Chemical Society

Received: November 25, 2011 Accepted: February 10, 2012 Published: February 10, 2012 2380

dx.doi.org/10.1021/ac203138y | Anal. Chem. 2012, 84, 2380−2387

Analytical Chemistry

Article

(Stratagene) and PCR nucleotide mix (Promega). Amplification primers were 5′-cgcggatccaccaccgatgcggagttcc-3′ (forward) and 5′-ccgctcgagctatccaaattgatcaatgac-3′ (reverse) containing BamHI and XhoI restriction sites, respectively. The PCR product was run on an agarose gel, purified with QIAquick gel extraction kit (Qiagen) and digested with BamHI and XhoI (Takara). Afterward, the fragment was purified using the QIAquick PCR purification kit (Qiagen) and ligated into the pET-TEV expression vector, a pET-19b vector modified by introducing a TEV protease cleavage site (generously provided by Dr. Lars Backman) giving pET-TEV-prethr-2 plasmid. This plasmid was used as a template to mutate the prethrombin-2 gene. Site-specific mutations were introduced using the following primers, 5′-cttcttgactcttacatagtcccgcgcggcgtggagggctgggac-3′ (forward) and 5′-gtcccagccctccacgccgcgcgggactatgtaagagtcaagaag-3′ (reverse), with the QuikChange II Site -Directed Mutagenesis Kit (Stratagene). The correctness of the mutant sequence was verified by DNA sequencing analysis (MWG Germany). Cloning and Site Specific Mutations of Human Prethrombin-2. The vector pOTB7 containing the full length human prothrombin cDNA was obtained from Geneservice (U.K.). The prethrombin-2 fragment was amplified from the same vector by PCR using PfuTurbo (Stratagene) and PCR nucleotide mix (Promega). Amplification primers were 5′ACGCGTCGACACCGCCACCAGTGAG-3′ (forward) and 5′-CCCAAGCTTCTACTCTCCAAACTGATC-3′ (reverse) containing SalI and HindIII restriction sites, respectively. The PCR product was run on an agarose gel, purified with QIAquick gel extraction kit (Qiagen) and digested with SalI and HindIII (Takara). Afterward, the fragment was purified using the QIAquick PCR purification kit (Qiagen) and ligated into the pQE9 expression vector (Qiagen), giving pQE9hprethr-2 plasmid. This plasmid was used as a template to mutate the human prethrombin-2 gene. Site-specific mutations were introduced using the following primers: 5′-cctggaatcctacatcgtcccgcgcggtgtggagggctcggatg-3′ (forward) and 5′-catccgagccctccacaccgcgcgggacgatgtaggattccagg-3′ (reverse) with the QuikChange II Site -Directed Mutagenesis Kit (Stratagene). The correctness of the mutant sequence was verified by DNA sequencing analysis (MWG Germany). Expression and Purification of Mutant Prethrombin-2. Escherichia coli BL21 (DE3) cells were transformed by heat shock with the plasmid containing the mutant mouse prethrombin-2 gene. Cells were grown at 37 °C in Luria− Bertani media supplemented with 100 μg/mL ampicillin to reach an OD600 ≈ 0.7. Protein expression was induced by adding 0.1 mM isopropyl thio-β-D-galactoside. Cells were grown for 4 h at 37 °C and harvested by centrifugation. Protein purification from inclusion bodies was performed as described in Soejima et al.31 with some modifications. After protein refolding, the sample was dialyzed against 50 mM Tris-HCl, pH 7.6, 150 mM NaCl, during 2 days at 4 °C without stirring. The dialyzed sample was centrifuged and the resulting supernatant was filtered. The pass-through fraction was loaded on a nickel HisTrap HP affinity column (GE Healthcare) controlled by an Ä KTA purifier equipment (GE Healthcare). Protein was eluted with a linear gradient from 0 to 500 mM of imidazole in 50 mM Tris-HCl, pH 7.6, 150 mM NaCl. Protein purity was investigated by 12% SDS-PAGE. For the purification of the human mutant prethrombin-2, E. coli XL1-blue cells were transformed by heat shock with the plasmid containing the mutant human prethrombin-2 gene.

prethrombin-2, an inactive single-chain precursor that has the same size as thrombin. Cleavage of prethrombin-2 at the second factor Xa site leads to the formation of a two-chain protein, which corresponds to the active form α-thrombin.25 Activation of prethrombin-2 to α-thrombin occurs through internal rearrangement of the initial peptide chain upon cleavage of the Arg-Ile bond with factor Xa or by ecarin. The cleaved shorter peptide section does not leave a thrombin macromolecule but stays linked with the longer peptide sequence via S−S bonds. The molecular weight of α-thrombin is the same as that of prethrombin-2. Currently, it is impossible to express directly recombinant active α-thrombin from the corresponding fragment of the prothrombin gene because the resulting protein will be always inactive prethrombin-2. Nevertheless, it is possible to express prethrombin-2 and activate it to α-thrombin. The preparation of recombinant prethrombin-2 and its activation to α-thrombin by ecarin was described in the literature elsewhere.26 Human prethrombin-2 was expressed in mouse myeloma cells. Lysis of cells and purification of lysate by fast protein liquid chromatography (FPLC) yields prethrombin-2 which is treated with ecarin. The reaction mixture is purified from ecarin by affinity chromatography using a benzamidine-sepharose gel. The main drawback of this procedure is the necessity to employ extremely hazardous ecarin, the primary reagent in the venom of the saw-scaled viper, Echis carinatus,27 to activate thrombin. Ecarin should be removed from recombinant α-thrombin in order to avoid nondesirable lethal thrombosis in patients. The removal stage significantly diminishes the yield and increases the costs of the α-thrombin preparation. Still there is no 100% assurance that recombinant α-thrombin is not contaminated with ecarin. This makes the large scale industrial production of recombinant human α-thrombin even more risky than its generation from human plasma. In theory, factor Xa can be used instead of lethal ecarin but it requires for its optimal operation a complex with factor V, platelet phospholipids, and calcium,28 which also must be separated from α-thrombin; consequently, its practical usage in the large scale production of α-thrombin is problematic. Here, we report a new methodology for production of mutant recombinant prethrombin-2 which is able to convert itself autocatalytically into fully functional α-thrombin in the absence of ecarin or factor Xa. This methodology has preparative and analytical applications. Our approach consists of changing the cleavage site of a recombinant proenzyme by site directed mutagenesis to obtain a stable protein that can be cleaved by the corresponding active enzyme. We have cloned the gene of the shortest precursor of thrombin without protease activity which corresponds to prethrombin-2. Further, we have changed the factor Xa cleavage site in prethrombin-2 to a thrombin cleavage site and produced a stable artificial selfreplicating enzyme. We have observed that in the presence of a small amount of thrombin, the mutant is activated triggering an enzyme-amplification reaction. This new concept of signal amplification using self-replicating enzymes can be applied to improve the thrombin detection limit of other existing methods.29,30



EXPERIMENTAL SECTION Cloning and Site Specific Mutations of Mouse Prethrombin-2. The vector pCMV-SPORT6 containing the full length mouse prothrombin cDNA was obtained from Geneservice (U.K.). The prethrombin-2 fragment was amplified from the same vector by PCR using PfuTurbo 2381

dx.doi.org/10.1021/ac203138y | Anal. Chem. 2012, 84, 2380−2387

Analytical Chemistry

Article

Cells were grown at 37 °C in Luria−Bertani media supplemented with 100 μg/mL ampicillin to reach an OD600 ≈ 0.6. Protein expression was induced by adding 1 mM isopropyl thio-β-D-galactoside. Cells were grown for 3 h at 37 °C and harvested by centrifugation. Protein purification from inclusion bodies was performed as described before for the mouse mutant prethrombin-2. After protein refolding, the sample was dialyzed against 50 mM Tris-HCl, pH 7.0, at 4 °C without stirring. The dialyzed sample was centrifuged, and the resulting supernatant was filtered. The pass-through fraction was loaded on a HiTrap HeparinHP affinity column (GE Healthcare) controlled by an Ä KTA purifier equipment (GE Healthcare). Protein was eluted with a linear gradient from 0 to 2 M NaCl in 50 mM Tris-HCl, pH 7.0. Protein purity was also investigated by 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Western Blot. Proteins were separated by 12% SDS-PAGE and electroblotted onto a nitrocellulose membrane. The membrane was then incubated for 1 h in phosphate-buffered saline (PBS), containing 0.05% Tween 20, and 10% defatted milk. Afterward, the membrane was washed with PBS containing 0.05% Tween 20 and incubated with anti-His mouse monoclonal antibodies (GE Healthcare) for 1 h. Then, the membrane was washed with PBS containing 0.05% Tween 20 and further incubated with secondary rabbit antibodies antimouse IgG conjugated to alkaline phosphatase (SigmaAldrich) for 1 h. After washing with PBS containing 0.05% Tween 20, protein bands were developed with 5-bromo-4chloro-3-indolyl phosphate/nitro blue tetrazolium solution (Fluka). Thrombin Activity Assays. Thrombin activity assays were performed in a Varioskan Flash microplate reader (Thermo Scientific) using black microwell plates at room temperature. Samples (100 μL final volume) were incubated in 50 mM TrisHCl, pH 9.0, 150 mM NaCl with 4 × 10−6 M (p-tosyl-Gly-ProArg)2-rhodamine 110 (Invitrogen), and the fluorescence of the resulting solution was monitored using λexc = 498 nm and λem = 521 nm. Mutant Prethrombin-2 Self-Replication Assays. Mouse mutant prethrombin-2 was cleaved using the Thrombin CleanCleave Kit (Sigma-Aldrich) as described in the manufacturer’s directions. The protein solution was then filtered through a 0.22 μm filter to eliminate any possible remaining α-thrombin modified-beads from the kit. To investigate the autocatalytic reaction, 5.38 × 10−9 M cleaved mouse mutant prethrombin-2 was mixed with and without 2.3 × 10−7 M untreated mouse mutant prethrombin-2. Then, the evolution of the fluorescence intensity was monitored using λexc = 498 nm and λem = 521 nm. Detection of Thrombin. Different concentrations of human α-thrombin (Sigma-Aldrich) were mixed with and without untreated 5.3 × 10−7 M mouse mutant prethrombin-2. The reactions were carried out in 50 mM Tris-HCl, pH 9.0, 150 mM NaCl with 4 × 10−6 M (p-tosyl-Gly-Pro-Arg)2-rhodamine 110 (Invitrogen), and the fluorescence of the resulting solution was monitored using λexc = 498 nm and λem = 521 nm. The final volume of the mixtures was 100 μL. Detection of Prothrombin in Human Plasma. Varying concentrations of pooled human plasma (Sigma-Aldrich) were mixed with 6 × 10−8 M ecarin (Sigma-Aldrich) with or without 1.5 × 10−6 M mouse mutant prethrombin-2. The reactions were carried out in 50 mM Tris-HCl, pH 9.0, 150 mM NaCl with 4 × 10−6 M (p-tosyl-Gly-Pro-Arg)2-rhodamine 110

(Invitrogen), and the fluorescence of the resulting solution was monitored using λexc = 498 nm and λem = 521 nm. The final volume of the mixtures was 100 μL. To generate the calibration curve using the standard addition method, different concentrations of commercial human prothrombin (Apollo Scientific) were added to human plasma. Samples of 47 pL were taken, and the prothrombin concentration was measured as described above.



RESULTS AND DISCUSSION Cloning and Site Specific Mutations. We report the preparation of an autocatalytic recombinant enzyme and its application to signal amplification in bioassays. As an example, we have cloned and mutated the gene of prethrombin-2, the smallest precursor of thrombin without proteolytic activity. Prethrombin-2 has only one factor Xa cleavage site, which contains an Arg-Ile bond. Rupture of this bond results in the formation of catalytically active α-thrombin. Our mutation strategy relies on converting the factor Xa cleavage site into the thrombin cleavage site by site directed mutagenesis to create a self-replicative protease. Thrombin selectively cleaves Arg-Gly bonds in fibrinogen, but it can also cleave other polypeptides. Studies on the thrombin cleavage site from 30 different polypeptides revealed that the optimum cleavage site has the structure of P4-P3-Pro-Arg-P1′-P2′, where P4 and P3 are hydrophobic amino acids and P1′and P2′ are nonacidic amino acids. It was also observed that polypeptides which contained Gly at P1′ were especially susceptible to thrombin cleavage.14 Taking these studies into consideration and trying to make the fewest modifications possible in order to avoid possible structural changes, we proceeded to change the cleavage site of the FXa for a thrombin one. The sequences corresponding to the mouse and human prethrombin-2 gene were cloned into the corresponding expression vectors which carry an Nterminal His-Tag. Primers were designed to introduce five single mutations by site-directed mutagenesis in order to create the mouse and human mutant prethrombin-2. As a result, the FXa cleavage site present in wild type (WT) mouse and human prethrombin-2, corresponding to residues IDGRIV, was changed to IVPRGV which corresponds to a thrombin cleavage site. Figure 1 shows the alignment of the FXa and thrombin

Figure 1. Sequence alignment of the cleavage site of wild type and mouse and human mutant prethrombin-2.

cleavage sites of mouse and human mutant prethrombin-2 and WT, respectively. Changed amino acids are highlighted in red. Purification of Mouse and Human Mutant Prethrombin-2. After protein expression, mutant prethrombin-2 was obtained in the form of inactive and insoluble inclusion bodies. A protocol to solubilize and refold prethrombin-2 described by Soejima et al.31 was followed to obtain soluble protein. The refolded protein was dialyzed against 50 mM Tris-HCl, pH 7.6, 2382

dx.doi.org/10.1021/ac203138y | Anal. Chem. 2012, 84, 2380−2387

Analytical Chemistry

Article

150 mM NaCl and subjected to affinity chromatography. Electrophoretic analysis of mouse purified protein showed a single band corresponding to the expected size (Figure 2A).

Figure 2. Analysis of the purified mouse mutant prethrombin-2. (A) 12% SDS-PAGE stained with Coomassie Brilliant Blue. (B) Nitrocellulose membrane probed with anti-His-tag mouse monoclonal antibodies. Lane 1, molecular mass references; lane 2, 0.3 μg of mutant prethrombin-2.

Figure 3. Evolution of fluorescence intensity in samples containing (a) 2.2 × 10−12 M human thrombin and 2.1 × 10−7 M human mutant prethrombin-2, (b) 2.2 × 10−12 M human thrombin and 2.1 × 10−7 M mouse mutant prethrombin-2, (c) 2.2 × 10−12 M human thrombin, (d) 2.1 × 10−7 M human mutant prethrombin-2, (e) 2.1 × 10−7 M mouse mutant prethrombin-2.

Protein identity was confirmed by Western blot using antibodies against His-tagged polypeptides (Figure 2B). Similar results were obtained for human mutated prethrombin-2 (Supporting Information, Figure S2). Auto Catalytic Generation of α-Thrombin from Mutant Prethrombin-2. In order to verify whether human thrombin could cleave the mutants, human and mouse mutated prethrombin-2 were incubated with and without human αthrombin in the presence of the commercially available fluorogenic substrate (p-tosyl-Gly-Pro-Arg)2-rhodamine 110 which upon cleavage by α-thrombin yields fluorescent rhodamine 110. We observed that in the presence of human thrombin, the protease activity increased considerably with time (Figure 3, curves a and b), compared to the samples where only thrombin was added (curve c). On the other hand, in the absence of thrombin, both mutants did not present enzymatic activity (curves d and e), this being a sign of the stability of the mutants under our experimental conditions. These results clearly indicate that thrombin is able to cut both mutant prethrombin-2 to generate an active enzyme with thrombin activity. Scheme 1 depicts the operational mechanism of our system. Exogenous α-thrombin activates the mutant prethrombin-2 converting it into endogenous α-thrombin, which in turn cleaves other macromolecules of prethrombin-2 and the fluorogenic substrate in the course of this autocatalytic reaction leading to the growth in enzymatic activity. We wanted to demonstrate that this growth in protease activity is not only a result of the cleavage of the mutated prethrombin-2 by exogenous human α-thrombin, and that in fact, generated endogenous α-thrombin is also able to cut mutated prethrombin-2, triggering a self-replicative cascade. For this purpose, we first cleaved mouse mutated prethrombin2 with a special agarose containing immobilized α-thrombin. Subsequently, uncleaved mutant was incubated in the presence of a mouse cleaved mutant to see whether it produced an increase in the fluorescence signal upon interaction with the

fluorogenic substrate. Samples containing only cleaved or uncleaved mutant were also included as controls. The enzymatic activity of α-thrombin in the samples was monitored with the fluorogenic substrate. The results depicted in Figure 4 show that in the presence of a small amount of the cleaved mutant and excess of the untreated mutant (curve a), the fluorescence signal increased considerably compared with the sample where only the cleaved mutant was included (curve b). The mouse mutant treated with the buffer solution used to wash α-thrombin on agarose demonstrated no activity (curve c). Consequently, growth in the protease activity was not caused by exogenous α-thrombin that could be detached from the agarose. This experiment confirms the ability of endogenous α-thrombin derived from mutated prethrombin-2 to participate in the autocatalytic reaction. Detection of Human α-Thrombin. We studied the effect of autocatalytic amplification with mutated prethrombin-2 on the sensitivity of detection of human α-thrombin in a buffer solution. Different concentrations of commercially available purified human α-thrombin were mixed with the fluorogenic substrate in the presence or absence of mouse mutated prethrombin-2. Figure 5A shows evolution of the fluorescence intensities for varying concentrations of human α-thrombin mixed with the fluorogenic substrate in the presence (curves a− h) and absence (curves i−p) of the mutant. The fluorescence intensities demonstrated exponential growth in the presence of the mutant, pointing to the build up of α-thrombin in the course of the autocatalytic reaction. In such a case, calibration plots of the first derivative of fluorescence intensity with respect to time dF/dt, representing the rate of reaction after 2 h, versus analyte concentration (Figure 5B) are more informative than the conventional plots of fluorescence intensity versus analyte concentration.32 The results show that the signal obtained at a given concentration of thrombin is amplified in the presence of the mutant. On the basis of the obtained calibration curves, we 2383

dx.doi.org/10.1021/ac203138y | Anal. Chem. 2012, 84, 2380−2387

Analytical Chemistry

Article

Scheme 1. Autocatalytic Assay of Mutant Prethrombin-2

with ecarin followed by detection of the resulting α-thrombin using enzymatic substrates.29 We studied the effect of added ecarin on the performance of the assay based on the fluorogenic thrombin substrate in the presence of fixed plasma volumes of 47 pL per well (Figure 1S in the Supporting Information). One can see that the dF/dt vs ecarin concentration plots of the systems operating with mutant prethrombin-2 (curve a) and without it (curve b) approached their plateaus starting from 50 nM of ecarin. Therefore, we used 60 nM ecarin concentration for the subsequent experiments. In order to evaluate the amplification power of our system, varying volumes of human plasma were mixed with ecarin and the fluorogenic substrate in the presence or absence of mouse mutant prethrombin-2 in microplate wells. The evolution of the fluorescence intensity was monitored with the fluorometer. The time course of the enzymatic reaction clearly demonstrates exponential growth in the fluorescence intensity when mouse mutant prethrombin-2 was present in the reaction mixture (Figure 6 A, curves a−f). The exponential shape of the obtained curves is consistent with the proposed exponential amplification depicted in Scheme 1. In the absence of any mutated prethrombin-2, the rate of enzymatic reaction was not so high (Figure 6 A, curves g−l). The detection limits of amplified (7.7 pL, S/N = 3, RSD = 12%, n = 3) and nonamplified (710 pL, S/N = 3, RSD = 6%, n = 3) assays were calculated according to the calibration plots (Figure 6 B) depicting dF/dt at a fixed time of 60 min versus volume of human plasma per microplate well. Taking into consideration that human plasma usually contains 90 μg/mL of prothrombin,29 the assay with autocatalytic amplification allowed detection as low as 0.693 pg of prothrombin per microplate well. The conventional nonamplified assay allowed quantification as low as 63.9 pg of the analyte per well. Thus, employment of an autocatalytic amplification cascade allows diminishing the volume of human plasma needed for the prothrombin assay by 2 orders of magnitude. This is a significant improvement, taking into consideration that the concentration of the fluorogenic substrate was previously optimized to maximize the rate of its decomposition

Figure 4. Time courses of mouse mutant prethrombin-2 activation and self-replication: (a) 5.38 × 10−9 M cleaved mutant and 2.3 × 10−7 M untreated mutant, (b) 5.38 × 10−9 M cleaved mutant, and (c) 2.3 × 10−7 M untreated mutant.

calculated the detection limit of the systems operating without mutated prethrombin-2 (0.488 pM, S/N = 3, n = 3) and with the mutant (10 fM, S/N = 3, n = 3). The employment of the autocatalytic amplification cascade leads to improvement in the detection limit by 50 times. In the absence of thrombin, either with mouse mutant or without it (first points of curves a and b, respectively) there is no increase in the fluorescence signal, indicating that both the mouse mutant and the artificial substrate are stable during the time of the experiment. Detection of Prothrombin in Human Plasma. Thrombin in human blood exists in the form of prothrombin. Therefore, the previously published assays for quantification of α-thrombin/prothrombin in blood rely on treatment of plasma 2384

dx.doi.org/10.1021/ac203138y | Anal. Chem. 2012, 84, 2380−2387

Analytical Chemistry

Article

Figure 5. (A) Evolution of fluorescence intensity in the presence (curves a−h) of 5.3 × 10−7 M mouse mutant prethrombin-2 or its absence (curves i−p). The samples contained different concentrations of human thrombin: (a and i) 5 × 10−12 M, (b and j) 3.75 × 10−12 M, (c and k) 2.5 × 10−12 M, (d and l) 1.5 × 10−12 M, (e and m) 7.5 × 10−13 M, (f and n) 5 × 10−13 M, (g and o) 2.5 × 10−13 M, (h and p) 0 M. (B) Calibration curve of αthrombin in the presence (curve a) and absence (curve b) of mouse mutant prethrombin-2.

Figure 6. (A) Evolution of fluorescence intensity in human plasma samples in the presence of 1.5 × 10−6 M mouse mutant prethrombin-2 (curves a−f) or its absence (curves g−l). The samples contained 6 × 10−8 M ecarin and different volumes of human plasma: (a and g) 2.6 nL, (b and h) 2 nL, (c and i) 1.3 nL, (d and j) 0.67 nL, (e and k) 0.33 nL, (f and l) 0 nL. (B) Calibration curve of human plasma in the presence (curve a) or absence (curve b) of mouse mutant prethrombin-2.

by α-thrombin. This autocatalytic signal amplification proved to be even more efficient in quantification of prothrombin in plasma than in detection α-thrombin, probably, due to higher stability of prothrombin in comparison with that of α-thrombin under the experimental conditions. We applied the standard addition method to detect the concentration of human prothrombin in 47 pL of human plasma per well. The known amounts of human prothrombin were added to plasma samples to obtain the calibration curve shown in Figure 7A. We plotted the data with the concentration of standard added in the x-axis and the first derivative of fluorescence intensity with respect to time dF/dt

in the y-axis. We performed the linear regression analysis to calculate the intercept of the calibration with the x-axis showing the content of prothrombin in plasma without added standard. The dilution factor was considered to calculate the concentration of prothrombin in plasma. The intercept shows a concentration of prothrombin 4.33 pg per well corresponding to 92 μg of prothrombin per milliliter of plasma sample (curve a). This correlates well with the previously reported prothrombin concentrations in human plasma between 90 and 100 μg per mL.29 The same assay performed in the absence of mutated prethrombin-2 did not allow us to detect prothrombin in 47 pL of human plasma per well (curve b). 2385

dx.doi.org/10.1021/ac203138y | Anal. Chem. 2012, 84, 2380−2387

Analytical Chemistry

Article

Figure 7. Representative calibration curves of prothrombin spiked to human plasma in the presence (curves a) or absence (curves b) of 1.5 × 10−6 M mouse mutant prethrombin-2: (A) quantification of prothrombin in human plasma with the method of standard addition and (B) calibration with high concentration of prothrombin spiked to human plasma.

ecarin or factor Xa. This general concept can be applied to the large scale production of human α-thrombin without dangerous contaminations. In theory, this methodology can be applied to proteases which are synthesized naturally in the form of an inactive zymogen. The recombinant stable autocatalytic enzyme permits us to introduce a new concept for the amplification of the readout signal of a bioanalytical assay. Our method would allow significant improvement of the sensitivity of any commercial available tests for α-thrombin and/or prothrombin by only adding mutant prethrombin-2 into the sample.

In order to evaluate the linear range of our method, we performed another calibration with high concentrations of human prothrombin spiked into plasma (Figure 7B). The calibration curve represents the plot of prothrombin concentration in plasma against dF/dt obtained in the presence of mutated prethrombin-2, showing the linear range up to 270 pg per well, corresponding to 5.7 mg of prothrombin spiked per milliliter of plasma (curve a). The curve b demonstrates the same plot in the absence of mutated prethrombin-2 showing the linear range up to 130 pg per well, corresponding to 2.7 mg of prothrombin per milliliter of plasma. In order to check out the reproducibility of the method, we obtained three independent calibration curves in which each point was triply replicated. Considering the definition of sensitivity as the slope of the linear section of a calibration curve, the assay performed in the presence of mutant prethrombin-2 revealed the medium slope of 1.32 × 10−4 (units/s)(pg/well)−1 with a RSD of 10%, the assay performed without mutant prethrombin-2 revealed the medium slope of 1.52 × 10−6 (units/s)(pg/well)−1 with a RSD of 8%. Thus, the employment of mutated prethrombin-2 leads to improvement in sensitivity by 2 orders of magnitude. In studies with spiked plasma in the presence of mouse mutant prethrombin-2, recovery of 103% was obtained at 38 pg of prothrombin per well, and 97% was obtained at 6 pg per well. The versatility of our assay would allow us in principle to improve the detection limit of commercially available tests for prothrombin in plasma. Moreover, we believe that the principle of our method can be further applied to amplify the activity of proteases which exist in the form of an inactive precursors activated by another protease. For instance, blood coagulation factors such as FXI activated by FXIIa, FIX activated by FXIa, FX activated by FIXa, and FV activated by thrombin.



ASSOCIATED CONTENT

S Supporting Information *

Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*Phone: +34943005308. Fax: +34943005314. E-mail: [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS V.P. acknowledges the contract Ramon y Cajal from the Spanish Ministry of Science and Innovation. This work was supported by the Spanish Ministry of Science and Innovation (Project BIO2008-04856).





REFERENCES

(1) Zhu, L.; Yang, R.; Zhai, J.; Tian, C. Biosens. Bioelectron. 2007, 23, 528−535. (2) Nakamura, H.; Mogi, Y.; Akimoto, T.; Naemura, K.; Kato, T.; Yano, K.; Karube, I. Biosens. Bioelectron. 2008, 24, 455−460. (3) Virel, A.; Saa, L.; Pavlov, V. Anal. Chem. 2009, 81, 268−272. (4) Pollegioni, L.; Piubelli, L.; Molla, G. FEBS J. 2009, 276, 6857− 6870.

CONCLUSIONS We have cloned and purified mouse and human mutants of prethrombin-2, which contain a thrombin cleavage site instead of a factor Xa cleavage site. The stable mutant is able to convert itself autocatalytically into active α-thrombin in the absence of 2386

dx.doi.org/10.1021/ac203138y | Anal. Chem. 2012, 84, 2380−2387

Analytical Chemistry

Article

(5) Caseli, L.; Perinotto, A. C.; Viitala, T.; Zucolotto, V.; Oliveira, O. N. Langmuir 2009, 25, 3057−3061. (6) Lequin, R. M. Clin. Chem. 2005, 51, 2415−2418. (7) Bates, D. L. Trends Biotechnol. 1987, 5, 204−209. (8) Stanley, C. J.; Ellis, D. H.; Bates, D. L.; Johannsson, A. J. Pharm. Biomed. Anal. 1987, 5, 811−820. (9) Shlyahovsky, B.; Pavlov, V.; Kaganovsky, L.; Willner, I. Angew. Chem., Int. Ed. 2006, 45, 4815−4819. (10) Khan, A. R.; James, M. N. Protein Sci. 1998, 7, 815−836. (11) Gadgil, C. J.; Kulkarni, B. D. AIChE J. 2009, 55, 556−562. (12) Wu, J.-W.; Wu, Y.; Wang, Z.-X. Eur. J. Biochem. 2001, 268, 1547−1553. (13) Barrett, D. Biochem. J. 1970, 117, 57−59. (14) Chang, J. Y. Eur. J. Biochem. 1985, 151, 217−224. (15) Goldsack, N. R.; Chambers, R. C.; Dabbagh, K.; Laurent, G. J. Int. J. Biochem. Cell Biol. 1998, 30, 641−646. (16) Wang, Y.; Bao, L.; Liu, Z.; Pang, D. W. Anal. Chem. 2011, 83, 8130−8137. (17) Pavlov, V.; Xiao, Y.; Shlyahovsky, B.; Willner, I. J. Am. Chem. Soc. 2004, 126, 11768−11769. (18) Hecht, A.; Kumar, A. A.; Kopelman, R. Anal. Chem. 2011, 83, 7123−7128. (19) Iliuk, A. B.; Hu, L.; Tao, W. A. Anal. Chem. 2011, 83, 4440− 4452. (20) Li, F.; Li, J.; Wang, C.; Zhang, J.; Li, X. F.; Le, X. C. Anal. Chem. 2011, 83, 6464−6467. (21) Swain, M. D.; Octain, J.; Benson, D. E. Bioconjugate Chem. 2008, 19, 2520−2526. (22) Lawson, J. H. Semin. Thromb. Hemost. 2006, 32 (Suppl 1), 98− 110. (23) Heffernan, J. K.; Ponce, R. A.; Zuckerman, L. A.; Volpone, J. P.; Visich, J.; Giste, E. E.; Jenkins, N.; Boster, D.; Pederson, S.; Knitter, G.; Palmer, T.; Wills, M.; Early, R. J.; Rogge, M. C. Regul. Toxicol. Pharmacol. 2007, 47, 48−58. (24) Doyle, M. F.; Mann, K. G. J. Biol. Chem. 1990, 265, 10693− 10701. (25) DiBella, E. E.; Maurer, M. C.; Scheraga, H. A. J. Biol. Chem. 1995, 270, 163−169. (26) Yonemura, H.; Imamura, T.; Soejima, K.; Nakahara, Y.; Morikawa, W.; Ushio, Y.; Kamachi, Y.; Nakatake, H.; Sugawara, K.; Nakagaki, T.; Nozaki, C. J. Biochem. 2004, 135, 577−582. (27) Potzsch, B.; Hund, S.; Madlener, K.; Unkrig, C.; MullerBerghaus, G. Thromb. Res. 1997, 86, 373−383. (28) Fujikawa, K.; Davie, E. W. Methods Enzymol. 1976, 45, 89−95. (29) Rob, J. A.; Tollefsen, S.; Helgeland, L. Anal. Biochem. 1997, 245, 222−225. (30) Wang, X.; Dong, P.; Yun, W.; Xu, Y.; He, P.; Fang, Y. Biosens. Bioelectron. 2009, 24, 3288−3292. (31) Soejima, K.; Mimura, N.; Yonemura, H.; Nakatake, H.; Imamura, T.; Nozaki, C. J. Biochem. 2001, 130, 269−277. (32) Saa, L.; Virel, A.; Sanchez-Lopez, J.; Pavlov, V. Chem.Eur. J. 2010, 16, 6187−6192.

2387

dx.doi.org/10.1021/ac203138y | Anal. Chem. 2012, 84, 2380−2387