Anal. Chem. 2000, 72, 3142-3149
Optical Detection of Polycations via Polymer Film-Modified Microtiter Plates: Response Mechanism and Bioanalytical Applications Sheng Dai, Qingshan Ye, Enju Wang,‡ and Mark E. Meyerhoff*
Department of Chemistry, The University of Michigan, Ann Arbor, Michigan 48109-1055
Microtiter plate wells modified with thin (∼20 µm) polymeric films capable of optically sensing macromolecular protamine and other polycationic species are described. The plates are prepared by coating the bottom of each well of a conventional 96-well polypropylene plate with an adherent polymer film (a mixture of poly(vinyl chloride) and polyurethane) containing a lipophilic 2′,7′dichlorofluorescein derivative. Surprisingly, optical response toward polycations is shown to result from the extraction of the fluorescein derivative from the polymer film into a lyophobic colloidal phase at the sample/film interface. This new phase is likely composed of a micellular-type ion pair complex between the analyte polycation from aqueous sample phase and the deprotonated form of the fluorescein derivative. Accumulation of the deprotonated fluorescein species in this interfacial region induces an absorbance change measured at 540 nm. Optimized plates can be used to sense protamine concentrations in the range of 0-100 µg/mL in 10 min with little or no response to physiological levels of common cationic species (Na+, K+, Ca2+, etc.). The modified plates are shown to be useful as simple optical detectors for measuring heparin levels in plasma via titrations with protamine and for monitoring protease activities (trypsin and plasmin) that cleave polycationic peptides/proteins such as protamine into smaller peptide fragments that are not detected by the sensing films. Assays for “clot busting” plasminogen activators (streptokinase, urokinase, and tissue plasminogen activator) are also demonstrated using this relatively simple microtiter plate-based polycation detection system. Recently, it has been found that polymeric films (e.g., poly(vinyl chloride) (PVC), polyurethane (PU), and silicone rubber) doped with appropriate lipophilic ion exchangers can be employed for potentiometric detection of important polyanionic (heparin, polyphosphates, etc.) and polycationic (protamine, synthetic cationic polypeptides, etc.) species in samples as complex as whole blood.1-11 It has also been shown that certain chromoionophores * To whom correspondence should be addressed: (phone) (734)-763-5916; (fax) (734)-647-4865 (e-mail)
[email protected]. † Department of Chemistry, St. John’s University, 8000 Utopia Parkway, Jamaica, NY 11439. (1) Ma, S.; Yang, V. C.; Meyerhoff, M. E. Anal. Chem. 1992, 64, 694-697.
3142 Analytical Chemistry, Vol. 72, No. 14, July 15, 2000
can be incorporated into similar polymeric matrixes to yield thin films that respond optically with high selectivity toward polyions.12,13 For example, Wang et al.13 described a film containing a lipophilic fluorescein ester (2′,7′-dichlorofluorescein octadecyl ester (DCFOE)) that responds with high specificity toward the polycation protamine, an arginine-rich protein often used to reverse the anticoagulant activity of polyanionic heparin at the termination of coronary artery bypass surgery. Herein, we describe the unique optical response mechanism of this DCFOE-based sensing film toward polycations such as protamine and the adaptation of such films within a microtiter plate format for detection of heparin (via titration), as well as for monitoring given protease activities. The approach developed is similar to that described recently by Kim et al.,14 who provided preliminary results regarding the use of a previously reported polyanion (heparin) sensitive optical film12 within a 96-well microtiter plate detection format. In the case of polycation measurements, thin plasticized polymer films (15 wt % PVC/15 wt % PU) containing DCFOE are cast onto the bottom of each well of a conventional polypropylene microtiter plate. The composition and thickness of the films are optimized to yield significant and reproducible absorbance changes in response to protamine and other polycationic species. By monitoring the optical response of the 96 sensing wells simultaneously, a simple and convenient method is demonstrated for the titrimetric detection of heparin (2) Ma, S.; Yang, V. C.; Fu, B.; Meyerhoff, M. E. Anal. Chem. 1993, 65, 20782084. (3) Yun, J. H.; Ma, S.; Fu, B.; Yang, V. C.; Meyerhoff, M. E. Electroanalysis 1993, 5, 719-724. (4) Fu, B.; Bakker, E.; Yun, J. H.; Yang, V. C.; Meyerhoff, M. E. Anal. Chem. 1994, 66, 2250-2259. (5) Fu, B.; Yun, J. H.; Wang, E.; Yang, V. C.; Meyerhoff, M. E. Electroanalysis 1995, 7, 823-829. (6) Fu, B.; Bakker, E.; Yang, V. C.; Meyerhoff, M. E. Macromolecules 1995, 28, 5834-5840. (7) Yun, J. H.; Yang, V. C.; Meyerhoff, M. E. Anal. Biochem. 1995, 224, 212220. (8) Meyerhoff, M. E.; Yang, V. C.; Wahr, J. A.; Lee, L. M.; Yun, J. H.; Fu, B.; Bakker, E. Clin. Chem. 1995, 41, 1355-1356. (9) Meyerhoff, M. E.; Fu, B.; Bakker, E.; Yun, J. H.; Yang, V. C. Anal. Chem. 1996, 68, 168A-175A. (10) Han, I. S.; Ramamurthy, N.; Yun, J. H.; Schaller, U.; Meyerhoff, M. E.; Yang, V. C. FASEB J. 1996, 10, 1621-1626. (11) Esson, J. M.; Meyerhoff, M. E. Electroanalysis 1997, 9, 1325-1330. (12) Wang, E.; Meyerhoff, M. E.; Yang, V. C. Anal. Chem. 1995, 67, 522-527. (13) Wang, E.; Wang, G.; Ma, L.; Stivanello, C. M.; Lam, S.; Petal, H. Anal. Chim. Acta 1996, 334, 139-147. (14) Kim, S. B.; Cho, H. C.; Cha, G. S.; Nam, H. Anal. Chem. 1998, 70, 48604863. 10.1021/ac000060n CCC: $19.00
© 2000 American Chemical Society Published on Web 05/23/2000
in plasma. It is also shown that these same polycation-sensitive microtiter plates can be used to monitor specific protease enzymatic activities (e.g., trypsin and plasmin). This is accomplished by monitoring the decrease in optical response toward given polycationic substrates as they are cleaved into smaller cationic fragments that are not detected by the optical sensing film. Such a protease assay scheme is further extended to the measurement of plasminogen activator (PA)-type thrombolytic agents (streptokinase (SK), urokinase (u-PA) and tissue-type plasminogen activator (t-PA)) in plasma by monitoring the generation of plasmin from a fixed amount of plasminogen via use of protamine as the substrate for plasmin. EXPERIMENTAL SECTION Materials. Bis(2-ethylhexyl) sebacate (DOS), 2′,7′-dichlorofluorescein, 1-iodooctadecane, tetrahydrofuran (THF), and high molecular weight PVC were obtained from Fluka (Ronkonkoma, NY). Tecoflex polyurethane (PU, SG-80A) was obtained from Thermedics (Woburn, MA). Protamine sulfate (from herring, grade III, and from salmon, grade X), poly(L-lysine) hydrobromide (MW ) 1000-4000 and 4000-15 000), trypsin, plasma (sheep and human), human plasmin, human plasminogen (0.45 unit), solid heparin (sodium salt from porcine intestine mucosa, 169 USP units/mg), and N-(2-hydroxyethyl)piperazine-N′-2-ethanesulfonic acid (HEPES) were purchased from Sigma Chemical Co. (St. Louis, MO). t-PA and t-PA stimulator were from Chromogenix (Molnsal, Sweden), while streptokinase and urokinase were purchased from Abbott (Abbott Park, IL). Standard solutions and buffers were prepared with reverse osmosis-deionized water. The buffer solutions used for most assays were 50 mM HEPES, pH 7.4, or 50 mM Tris-HCl, pH 7.4, unless noted otherwise. The synthesis of the chromoionophore DCFOE has been described elsewhere.13 Preparation of Microtiter Plates with Optical Sensing Films. Polycation-responsive films were formulated with 30 wt % polymer (15-18 wt % PVC and 15-12 wt % PU), 69 wt % DOS, and 1 wt % chromoionophore (DCFOE). The film composition, especially the ratio of PVC to DOS, was carefully optimized to obtain a rapid and reproducible optical response toward protamine. The casting solutions were prepared by dissolving a total amount of 200 mg of film components in 2 mL of freshly distilled THF. These solutions were then uniformly dispensed (10 µL/well) into each U-bottomed microwell of polypropylene plates (model M4404, Sigma Chemical Co.), using micropipets (Finnpipet, Labsystem, Helsı´nkı´, Finland.). The plates were air-dried in a dust-free vessel for 1 day prior to use. Polymeric films were also cast on glass slides via spin coating to obtain complete UV/visible absorption spectra of the polymer coating under varying solution conditions. Apparatus. The absorbance spectra of the various film formulations were recorded with a UV/visible double-beam spectrophotometer (DU series 600, Beckman, Fullerton, CA). Microtiter plate-based optical measurements were performed using a MRXII precision microplate reader (Dynex Technologies, Chantilly, VA). All quantitative measurements were made at 540 nm. Prior to exposure to a given polycation test solution, all the wells were filled with HEPES buffer pH 7.4 (250 µL) until the polymer film-coated wells had a stable absorbance value (∼20 min). Unless otherwise stated, all dose responses for protamine standards were obtained using grade III protamine. Absorbance
values for each well of the modified plates were determined after delivering a fixed volume (200 µL) of various protamine standard solutions into the microtiter plate wells coated with the optical sensing films. The absorbance was recorded at predetermined time intervals. Optical response of the films toward small cations was measured using corresponding chloride salt solutions buffered to pH 7.4. All measurements and preequilibrations were carried out under ambient conditions (22-24 °C). Response Mechanism Study. The diffusion of DCFOE from the bulk of the polymer film toward the surface was investigated by optical microscopy with a charged-coupled device (CCD) detector. The DCFOE optical membrane was cast onto a specially designed flow-through thin-layer cell20 that was placed on the specimen stage of an optical microscope. The film itself had an O-ring shape with about 6 mm i.d., 7 mm o.d., and 0.1 mm in thickness. By continuously flowing protamine solutions (0.2 mL/ min) controlled by an LC pump (model 200 LC, Scientific System Inc.), the outer rims of the film were brought in contact with a high concentration of protamine (10 mg/mL) continuously. A segment of the polymer film ring was studied at room temperature under the microscope (Olympus CX 40, Olympus America, Inc., Melville, NY) by illuminating it with a halogen lamp. The appropriate wavelengths of the protonated and deprotonated forms of the chromoionophore DCFOE were selected via the appropriate interference filters. A CCD imaging camera (AT200, Photometrics Ltd., Tucson, AZ) was used to record the microscope images. Titrations of Heparin Using Film-Modified Microtiter Plates. Heparin measurements via titration with protamine were carried out by adding aliquots (10 µL) of different protamine standard solutions (100-1000 µg/mL) to a series of separate plastic tubes containing a fixed volume (190 µL) of a given heparin solution (or plasma containing heparin). After incubating for a short time (3-5 min), 200-µL aliquots of the protamine-heparin solutions were transferred into a series of microtiter plate wells coated with the optical sensing films to detect the unbound protamine. As before, the absorbance changes were measured at 540 nm and plotted vs the mass of protamine added to each test sample of heparin. The protamine response of the films in plasma samples was determined by using sheep plasma. To ensure a constant pH 7.4, a 100-µL aliquot of a 10-fold concentrated phosphate buffer (PBS; 0.5 M, pH 7.4) was added to a 900-µL aliquot of plasma. The microtiter plate-based optical sensing films were first preconditioned with buffered plasma before the absorbance changes were measured for samples containing varying levels of protamine (either to determine response toward protamine or for titrations of heparin in plasma samples). Trypsin and Plasmin Assays Using Protamine as Substrate. Trypsin assays in HEPES buffer were performed by adding 10 µL of a 1.6 mg/mL protamine solution to 190 µL of working (15) Cohen, E. J.; Camerlengo, L. J.; Dearing, J. P. J. Extracorporeal Tech. 1980, 12, 139-141. (16) Teien, A. N.; Lie, M.; Abildgaard, U. Thromb. Res. 1976, 8, 413-416. (17) Ramamurthy, N.; Baliga, N.; Wahr, J. A.; Schaller, U.; Yang, V. C.; Meyerhoff, M. E. Clin. Chem. 1998, 44, 606-613. (18) Ong, E. B.; Johnson, A. J. Anal. Biochem. 1976, 75, 569-582. (19) Badr, I. H. A.; Ramamurthy, N.; Yang, V. C.; Meyerhoff, M. E. Anal. Biochem. 1997, 250, 74-81. (20) Schneider, B.; Zwickl, T.; Federer, B.; Lindner, E.; Pretsch, E. Anal. Chem. 1996, 68, 4342-4350
Analytical Chemistry, Vol. 72, No. 14, July 15, 2000
3143
buffer solution containing different amounts of trypsin (0-100 units/mL). After 10 min of incubation, 200 µL of the total trypsinprotamine reaction mixture was added to microtiter plate wells coated with the polycation sensing films. The unreacted protamine in the solution was monitored by recording the absorbance change of the wells at 540 nm after 15 min. Similarly, the assays for plasmin were performed by adding 10 µL of a 1.6 mg/mL protamine solution to 190 µL of working buffer solution containing different amounts of plasmin (0-100 units/mL). After 15 min of incubation at 37 °C, 200 µL of the plasmin-protamine reaction mixture was added to microtiter plate wells coated with the polycation sensing film, and the unreacted protamine in the solution was determined by monitoring the absorbance change. Plasminogen Activator Assays. PA activities in plasma samples can be followed optically by monitoring the degradation of protamine by the plasmin generated due to PA activation of plasminogen. The PA assays of SK, u-PA, and t-PA using microtiter plate-based optical sensing film detection were performed in a final volume of 200 µL. The SK and u-PA assays were carried out by adding 20 µL of the sample mixture to 180 µL of HEPES working buffer and transferring the solution to polymer film-coated microwells. The sample reaction mixture (20 µL total volume) contained 10 µL of human plasma spiked with a given plasminogen activator activity and 10 µL of a substrate solution containing protamine and plasminogen. The substrate solution (10 µL) used in these experiments contained 16 µg of protamine and 0.02 unit of human plasminogen in a 50 mM Tris-HCl, pH 7.4, working buffer. The reaction mixture was incubated at 37 °C for 15 min. After the reaction mixture was added to the 180 µL of buffer, the diluted reaction mixture was transferred into the optical filmcoated microwells. The unreacted protamine was detected by the absorbance changes of the films after 10 min. For the t-PA assay, the 20-µL reaction mixture contained 10 µL of plasma spiked with different amounts of t-PA and 10 µL of substrate solution containing 16 µg of protamine, 0.02 unit of plasminogen, and 0.8 µg of PA stimulator. After a fixed incubation time (15-30 min) at 37 °C, the reaction mixture was diluted (1: 10) with HEPES buffer and transferred into microtiter wells coated with the polycation sensing films. The absorbance changes were recorded to monitor unreacted protamine, and these absorbance values were inversely proportional to the t-PA activities in the plasma sample. RESULTS AND DISCUSSION Principle. The synthesized lipophilic fluorescein chromophore contains an acidic phenol group which becomes negatively charged upon deprotonation. The optical response of DCFOEdoped polymeric films toward polycations results from deprotonation of the DCFOE due to its cooperative ion pairing interaction with the analyte polycation and yields a significant absorbance change at 536 nm.13 In previous work with films doped with DCFOE, it was assumed that this ion pairing reaction takes place within the organic phase of the film, in much the same way that both optical12 and electrochemical1-3 responses are observed for polyanionic heparin using films doped with the exchanger tridodecylmethylammonium ion (TDMA) and an additional chromoionophore for optical sensing.4,5 However, preliminary experiments using optical microscopy to view the interfacial regions of 3144 Analytical Chemistry, Vol. 72, No. 14, July 15, 2000
Figure 1. Optical microscope study of DCFOE diffusion from polymer film in contact with 10 mg/mL protamine solution (0.05 M Tris-HCl, pH 7.4) after varying time periods. Light source was filtered with 536-nm band-pass filter to show absorbance (dark area) due to the deprotonated DCFOE. (A) 4 h; (B) 24 h.
the DCFOE-doped films suggested a somewhat different picture regarding the location of the ion-pairing chemistry. Using an experimental imaging arrangement similar to that reported previously by Schneider et al.,20 a more detailed analysis of the optical response of the films toward protamine and polycations was carried out. Initially, both interfaces of the ringshaped film were conditioned by flowing Tris buffer (0.05 M, pH 7.4) through the system. Images of the bulk film and the interfaces at 535 nm, using an appropriate interference filter (535 ( 10 nm), were taken as a measure of concentration of the deprotonated DCFOE species in each region of the image field. At zero time (t ) 0), the buffer bathing the outer side of the ring was changed to a protamine solution (10 mg/mL in Tris buffer). To follow the expected ion-exchange process (initially thought to be protamine exchanging for protons within the film), images were taken after given equilibration times. As shown in Figure 1, there is no increase in absorbance at 535 nm (dark field) at any location within the polymer film doped with DCFOE, even after 24 h of continuous flow of protamine solution. This suggests that there is no deprotonated DCFOE within the polymer phase. Large absorbance changes at 535 nm corresponding to the deprotonated form of DCFOE were observed in a new phase, which grows with time at the interface of polymer film and the aqueous phase (shown in Figure 1 after 4 and 24 h). No such changes were noted in blank experiments in which protamine-free buffer was equilibrated with the films for similar time periods. This new interfacial phase can actually be physically
removed by scraping the surface of the film with a spatula, etc. In addition, the concentration of neutral DCFOE (viewable, but with much lower absorptivity at 535 nm) within the side of the polymer film exposed to the protamine solution was found to decrease with time (depletion concentration gradient over some 50 µm distance after 24 h), indicating loss of DCFOE from the sample side of the organic polymer phase (see Figure 1B). Further experiments were performed to prove that DCFOE can partition out of an organic phase to form a third phase at the interface of an aqueous sample containing protamine. In one such experiment, 0.2 mg of DCFOE was first dissolved in 2 mL of hexane, and then 2 mL of protamine solution (20 mg/mL in Tris buffer, pH 7.4) was added to the DCFOE hexane solution. The two-phase mixture was shaken vigorously and allowed to separate. No color change was noted in aqueous phase, and a decrease in absorbance at 480 nm was observed in the hexane phase, indicating a loss of neutral DCFOE from the organic layer. However, a relatively intense pink-red color appeared in a third phase formed between the hexane and aqueous phase, indicating the presence of the deprotonated form of DCFOE in this newly formed layer. No such interfacial layer was observed in the absence of protamine for these bulk-phase extraction experiments. The results described above clearly suggest that DCFOE within the organic polymer films employed for this work is capable of diffusing out of the organic phase to bind, in deprotonated form, with protamine diffusing to the surface of the film from the aqueous sample phase. This process is illustrated schematically in Figure 2 for the specific microtiter plate arrangement described herein. The result is the formation of a unique lyophobic colloidal phase at the polymer film/sample interface. This phase appears to be composed of a stable ion pair between protamine and deprotonated DCFOE. Indeed, recent experiments in this laboratory have quantitated the cooperative binding of water-soluble anionic surfactants with polycationic species, including protamine,21 and others have observed similar strong cooperative binding between charged surfactants and various polyion structures.22-24 It is believed, based on light-scattering experiments,25,26 that such interactions form micellular complexes with the polyion wrapped around the outer charged surface of the surfactant micelle, stabilizing such structures, even below the critical micelle concentration (cmc) value for the given ionic surfactant. Since, in the present system, deprotonated DCFOE is not completely soluble in aqueous solution, the protamine serves as the stabilizing species for formation of this lyophobic colloidal phase at the interface (Figure 2), and the amount of complex formed after a fixed period of time is directly proportional to the concentration of polycation in the sample. More detailed studies of this unique optical response mechanism, including the development of an appropriate mathematical model that predicts the (21) Esson, J. M.; Ramamurthy, N.; Meyerhoff, M. E. Anal. Chim. Acta 2000, 404, 83-94. (22) Attwood, D.; Florence, A. T. Surfactant Systems. Their Chemistry, Pharmacy and Biology; Chapman and Hall: London, 1984. (23) Eicke, H.-F.; Kvita, P. In Reverse Micelles: Biological and Technological Relevance of Amiphilic Structures in Apolar Media; Luisi, L. P., Straub, B. E., Eds.; Plenum Press: New York, 1984. (24) Zana, R. Surfactant Solutions; Marcel Dekker: New York, 1987. (25) Xia, J.; Dubin, P. L.; Kim, Y. J. Phys. Chem. 1992, 96, 6805-6811. (26) Li, Y.; Dubin, P. L.; Dautzenberg, H.; Luck, U.; Hartmann, J.; Tuzar, Z. Macromolecules 1995, 28, 6795-6798.
Figure 2. Representation of process involved in the formation of lyophobic micelle structure of deprotonated DCFOE and protamine at film/sample interface within the microtiter plate optical sensing arrangements.
polyion concentration-dependent response, are currently in progress and will be the focus of a future report. It should be noted the protamine-induced extraction of DCFOE from the organic phase of the polymer film can be reversed by acidifying the sample phase to break up the interfacial colloidal complex between protamine and micelles of deprotonated DCFOE (by forcing the protonation of anionic DCFOE). In fact, adding 0.02 M HCl to the microtiter plate wells after the film-modified plates have been used to sense protamine, and equilibrating the plates for 5 h, enables the DCFOE to be almost completely backextracted into the polymer film. The plates can then be reused for sensing protamine or other polycations with nearly equal sensitivity as fresh film-modified plates. However, use of the highthroughput microtiter plate sensing platform described here makes such regeneration unnecessary, since the plates are relatively inexpensive and could simply be disposed of after the single use of each of the 96 wells. Response of Film-Modified Microtiter Plates toward Polycations. Polymer films composed of PVC and PU doped with DCFOE exhibit an initial absorbance maximum at 480 nm at pH 7.4 owing to the presence of the DCFOE in protonated form completely within the organic phase of such films. In the presence of protamine and other polycations, the absorbance maximum Analytical Chemistry, Vol. 72, No. 14, July 15, 2000
3145
Figure 3. Microtiter plate-based optical film response with time toward varying protamine concentrations: (9) 10, ([) 20, (b) 50, (2) 100, (4) 200, (×) 400, (O) 1000, and (0) 10 000 µg/mL. Film absorbance measured at 540 nm; buffer, 0.05M HEPES, pH 7.4; sample volume, 200 µL. The results are the means of three measurements.
shifts to 536 nm with a shoulder at 510 nm, and it is attributed to the accumulation of deprotonated DCFOE in the colloidal phase that forms at the film/sample interface (see above). For the microtiter plate-based optical sensing, a significant absorbance change (using a 540-nm filter of plate reader) of each well coated with the DCFOE-doped polymer film occurs when solutions of protamine or other polycations are added to the wells. Figure 3 shows the typical absorbance changes as a function of time of film-modified microtiter wells exposed to eight different concentrations of protamine (200-µL sample volume, unstirred). As can be seen, the absorbance change is faster as the protamine concentration in the buffer solution increases. With very low protamine concentrations (below 1 µg/mL), the absorbance changes very little, even after a 24-h period, since very few of the total DCFOE species are extracted in deprotonated form at the interface. At higher protamine concentrations (above 200 µg/mL), the polymer film reaches full equilibrium within 5 h, and the absorbance does not change with longer incubation times. The data suggest that once a certain amount of protamine has extracted most of the DCFOE from the thin organic film coated on the bottom of each well to form the colloidal micelle complex, the protamine replaces nearly all of the protons of DCFOE and the absorbance of the microtiter wells stops changing (i.e., saturation). However, the range of polycation sensing can be increased merely by increasing the concentration of the DCFOE reagent within the sensing layer or the thickness of the film (data not shown). (Note: Images shown in Figure 1 were taken with films having a cross section of 1 mm, much thicker than the films cast in the wells of the microtiter plates. Thus, the amount of total DCFOE available per unit surface area of film is much greater in the image experiment, and this is why the colloid phase continues to grow over 24 h.) 3146 Analytical Chemistry, Vol. 72, No. 14, July 15, 2000
Figure 4. Typical protamine calibration curve for optimized microtiter plate-based optical film. Film absorbance, 540 nm; response time, 10 min; 0.05 HEPES pH 7.4; sample volume, 200 µL. The results are the means and standard deviations of three separate measurements (mean ( SD).
At pH 7.4, with a fixed equilibration time of 10 min and measurement using a sample volume of 200 µL, an almost linear dose-response curve can be obtained at low protamine concentrations (100 µg/mL), the absorbance reaches a maximum value, corresponding to the complete extraction of the DCFOE from the thin film at a given film thickness (∼20 µm) and DCFOE concentration (1 wt %). Significant optical responses of film-modified wells can also be obtained toward other synthetic polycationic (arginine-rich) peptides (data not shown). In contrast, when films are exposed to buffered test solutions containing tri- or tetraarginine peptides (R-R-R and R-R-R-R, respectively), little optical response toward these much smaller polycationic species is observed (see Figure 5). These data suggest that, as expected, such smaller polycation structures do not stabilize the formation of the micellular complexes of deprotonated DCFOE at the interface of the films. While the microtiter wells coated with films containing DCFOE respond to low levels of appropriately sized polycations, the optical response to simple cations, such as Na+, K+, NH4+, Li+, Mg2+, and Ca2+ in biological samples, is negligible (no response to a 0.1 M concentration of these species in pH 7.4 buffer). This suggests that these polymer film-modified microtiter wells provide adequate selectivity for carrying out polycation measurements in complex physiological samples without a significant increase in the background absorbance of the wells (see below). Heparin Titrations with Protamine. Heparin is a complex, highly negatively charged polysaccharide with an average molecular weight of 15 000. Currently, available heparin assays are based on the measurement of plasma or whole blood clotting time or the detection of the proteolytic activity of certain clotting factors using synthetic substrates (e.g., anti-Xa chromogenic assay).15,16 Although these methods are well accepted by clinicians, they are indirect and yield results that are not necessarily relevant to the total therapeutic quantity of heparin present in the blood. Recently
Figure 5. Response of microtiter plate-based optical sensing films toward various polycations: (O) Arg-Arg-Arg; (4) Arg-Arg-Arg-Arg; (0) protamine. Absorbance wavelength, 540 nm; 0.05 M HEPES, pH 7.4; response time, 15 min; sample volume, 200 µL. The results are the means of three measurements.
Ramamurthy et al.17 reported that polycation-sensitive membrane electrodes could be used to monitor heparin concentrations in whole blood via protamine titrations. Interestingly, such a protamine titrimetric method can also be applied using the polymer film-modified microtiter plate optical sensing arrangement described herein. Indeed, Figure 6A illustrates the results for the optical titration of varying heparin levels in HEPES buffer, pH 7.4. These titrations were carried out by adding fixed volumes of a given heparin solution to a series of tubes containing known, increasing amounts of protamine. After a short incubation period (3 min), the heparin-protamine solution mixtures were transferred into the microtiter plate and read after a 15-min incubation period. The end points for heparin levels of g1 unit/mL can clearly be distinguished from the blank (no heparin) response of the modified wells. Extrapolation of the rising optical response line to the x-axis provides the amount of protamine required to reach the end point for each titration. As can be seen, the end points correspond to ∼10 µg/mL of protamine for each 1 unit/mL increment of heparin. This is in good agreement with the known stoichiometry for the binding of protamine with porcine heparin.17 Significant optical response to protamine can also be obtained in concentrated plasma samples. However, in this case, the plasma must be well buffered to ensure that the activity of protons in the sample phase remains constant in accordance with the requirements for optical sensing with pH-dependent chromoionophores (i.e., high pH values will deprotonate DCFOE in the film phase and enhance extraction into the aqueous test phase, yielding a protamine-like optical response). This was accomplished by mixing sheep plasma (as a model) at a ratio (v/v) of 9 parts:1 part of a 10× concentrated PBS buffer, pH 7.4. In preliminary experiments, it was found that the response toward protamine in such plasma was somewhat less than that in buffer solution, possibly due to the higher viscosity of the solution (and thus slower mass transfer of protamine to the surface of the film) as well as the binding of
Figure 6. Titration of heparin with protamine using microtiter platebased optical film detection: (A) in HEPES buffer and (B) in sheep plasma. Absorbance wavelength, 540 nm; response time, 15 min; sample volume, 200 µL. The results are the means and standard deviations of three separate measurements (mean ( SD).
some protamine to other plasma proteins. Nonetheless, response curves were generally linear in the range of 20-120 µg/mL protamine using a 15-min incubation period to observe the optical response. Typical optical titrations of sheep plasma samples spiked with varying levels of heparin are shown in Figure 6B. Again, the shift in the response to added protamine corresponds well with the known stoichiometry for the protamine-heparin reaction.17 Heparin levels of g1 unit/mL can clearly be discerned. Protease Assays Using Film-Modified Microtiter Plates. The polymer film-modified microtiter plates can also serve as useful detectors in the design of novel optical enzyme assays. Indeed, specific protease enzymes can be detected by observing a decrease in the optical response to a given polycationic substrate as such substrates are degraded into smaller fragments with less number of positive charges. The resulting smaller fragments yield little or no optical response when in contact with the microtiter plate-based sensing film, analogous to the behavior observed for tri- and tetraarginine (see Figure 5). This same principle has been Analytical Chemistry, Vol. 72, No. 14, July 15, 2000
3147
Figure 7. Assays of protease activities using microtiter plate-based optical polycation sensing film detection for (0) trypsin and (b) plasmin. Absorbance wavelength, 540 nm; protamine concentration, 80 µg/mL; preincubation time, 10 min; temperature, 37 °C; response time, 15 min; sample volume, 200 µL. The results are the means and standard deviations of three separate measurements (mean ( SD).
used previously to devise several protease assays based on electrochemical polycation detection. 7,10,19 Protamine is an excellent substrate for both trypsin7 and plasmin18 because these proteases cleave at the many arginine residues of this protein, yielding a variety of cationic fragments. As shown in Figure 7, when a solution of 80 µg/mL protamine is incubated for 15 min with varying levels of trypsin or plasmin, the optical response of the film-modified microtiter wells decreases upon subsequent incubation of the reaction mixture in the wells for 10 min. The range of detectable protease activities can be varied by simply changing the amount of protamine or the preincubation time of the protease with the protamine. As shown in Figure 7, using the conditions outlined above, the lower limit of detection for trypsin is 0.1 unit/mL, while for plasmin, it is 0.01 unit/mL. Such responses are reproducible ((4% (SD)) and can likely be utilized to detect a wide range of other proteases, perhaps using specially designed synthetic polycationic substrates with specific cleavage sites spaced within an arginine-rich sequence. Such an approach has already been used successfully with analogous electrochemical polyion sensors for monitoring chymotrypsin (placing phenylalanines into an arginine-rich peptide) and renin (placing leucine-leucine links within the peptide) activities.10 Plasminogen Activators Assays. PAs function in the fibrinolytic pathway of the hemostasis system, where they convert the zymogen substrate plasminogen (PLG) to plasmin, a proteolytic enzyme that acts to lyse clots. Pathologically, fibrin clots formed within coronary arteries hinder blood flow and reduce the oxygen supply to the cardiac tissue, potentially causing necrosis of myocardial cells. Since recombinant forms of PAs are now available commercially, PAs with “clot-busting” capability are now used routinely in emergency situations to treat patients experiencing the early stages of heart attacks and strokes. With its increased 3148 Analytical Chemistry, Vol. 72, No. 14, July 15, 2000
Figure 8. Assays of plasminogen activator assays using microtiter plate-based optical polycation sensing film detection for (0) streptokinase and (b) urokinase. Absorbance wavelength, 540 nm; plasminogen concentration, 0.1 unit/mL; protamine concentration, 80 µg/mL; incubation time, 15 min; temperature, 37 °C; response time, 10 min; sample volume, 200 µL. The results are the means and standard deviations of four separate measurements (mean ( SD).
application, there is a need to monitor PA activity levels in blood during thrombolytic therapy. The primary thrombolytic agents associated with such treatment are streptokinase, urokinase, and tissue-type plasminogen activator. At present, the activity of PAs in clinical samples are measured by antigenic or activity assays.27-29 Since protamine is known to be an excellent substrate for plasmin, the plasminogen activator activities in plasma samples can be followed conveniently by monitoring the degradation of protamine by the plasmin produced due to PA activation of plasminogen. Chang et al.30 have demonstrated that sensitive assays for PAs can be developed using protamine as substrate and detection by a polycation-sensitive membrane electrode. The same basic principle can be employed to devise PA assays using the microtiter plate-based optical sensing arrangement. SK and u-PA assays were performed by incubating varying amounts of SK or u-PA with a fixed volume (10 µL) of substrate solution containing 16 µg of protamine and 0.02 unit of human plasminogen for 15 min at 37 °C. After incubation, the reaction mixture was diluted 10 times with HEPES buffer and transferred onto optical membrane-coated microwells. The unreacted protamine was detected by the absorbance changes of the films after 10 min. Calibration plots for plasma SK and u-PA were constructed by graphing the average absorbance change of microtiter platebased optical membrane vs the PA activity (IU/mL) spiked into the plasma (shown in Figure 8). After only a 15-min incubation between the plasma sample and protamine (600 µg/mL), the (27) Verheijen, J. H.; Mullaart, E.; Chang, G. T. G.; Kluft, C.; Wijngaards, G. Thromb. Haemostasis 1982, 48, 266-269. (28) Hayashi, S.; Yamada, K. Thrombo. Res. 1981, 22, 573-578. (29) Bergmeyer, H. U. Methods of Enzymatic Analysis, 3rd ed.; verlag c hemie: Weinheim, Germany, 1983; Vol. V (Enzyme 3). (30) Chang, L.; Meyerhoff, M. E.; Yang, V. C. Anal. Biochem. 1999, 276, 8-12.
The microtiter plate-based polycation detection system was also applied for the assay of t-PA. In the t-PA assay, a PA stimulator (fibrin fragment) must be added to the reaction mixture in addition to PLG and protamine to enhance the activation of this PA species.31 After a given incubation time (15-30 min) at 37 °C, the reaction mixture was diluted (1:10) with HEPES buffer and transferred onto microtiter wells coated with the polycation sensing film. The absorbance change was recorded to detect unreacted protamine, and it is inversely proportional to the concentration of t-PA in the plasma sample (Figure 9). The t-PA assay was further optimized by varying the concentrations of plasminogen and stimulator in substrate solution. As shown in Figure 9A, detection limits toward t-PA shift to lower levels when the PLG concentration is increased from 0.2 to 0.3 unit. With longer incubation time (30 min), the microtiter plate-based assays can detect t-PA at even lower levels, down to 50 unit/mL in plasma (Figure 9B), which approaches the basal levels of this PA in blood.32
Figure 9. Tissue-type plasminogen activator assays using microtiter plate-based optical polycation sensing film detection. (A) Different plasminogen concentrations: (0) 0.2 and (b) 0.3 unit/mL. Absorbance wavelength, 540 nm; protamine concentration, 80 µg/mL; preincubation time, 15 min; temperature, 37 °C; response time, 10 min; sample volume, 200 µL; (B) Different incubation time: (0) 15 and (b) 30 min. Protamine concentration, 80 µg/mL, plasminogen concentration, 0.2 unit/mL; temperature, 37 °C; response time, 10 min; sample volume, 200 µL. The results are the means and standard deviations of four separate measurements (mean ( SD).
decrease in absorbance response is quite significant and is proportional to PA activity present in the plasma sample. Using a PLG level of 1 unit/mL, both PAs can easily be detected in their normal therapeutic range (150-1500 IU/mL). (31) Mukherjee, M.; Sembhi, K.; Kakkar, V. V. Blood Coagulation Fibrinolysis 1996, 7, 491-496. (32) Lijnen, H. R.; Collen, D. Thromb. Haemostasis 1991, 66, 88-110. (33) Wang, K.; Seiler, K.; Rusterholz, B.; Simon, W. Analyst 1992, 117, 57-60. (34) Wang, E.; Ohashi, K.; Kamata, S. Chem. Lett. 1992, 939-942. (35) Seiler, K.; Simon, W. Anal. Chim. Acta 1992, 266, 73-87. (36) Lerchi, M.; Bakker, E.; Rusterholtz, B.; Simon, W. Anal. Chem 1992, 64, 1534-1540. (37) Lerchi, M.; Reitter, E.; Simon, W.; Pretsch, E. Anal. Chem 1994, 66, 17131717.
CONCLUSIONS The experimental results presented here clearly demonstrate that microtiter plates modified with thin, DCFOE-doped polymer films, respond to polycations via an unusual response mechanism compared to previously reported optical sensing films for small ions,33-37 as well as an earlier optical polyanion sensor for heparin.12 The response is now ascribed to the formation of a colloidal phase at the film/sample interface, resulting from extraction of DCFOE in deprotonated form to create a micellular ion pair complex with the polycation species. In the microtiter plate arrangement, it is shown that such films can be used to detect protamine (and other large polycation structures) at concentrations ranging from 5 to 100 µg/mL. Response to these polycation species occurs in both buffer and plasma. The polycation response is reproducible and achieved over a reasonable time period of equilibration (10 min). It has further been shown that such a detection mode can be utilized to quantitate heparin levels in plasma via protamine titrations and monitor specific protease activities (trypsin and plasmin). The ability to use such a technology to quantitate low levels of plasmin activity has also enabled the same microtiter plate detection system to be used in the design of novel assays for monitoring specific plasminogen activators (e.g., urokinase and streptokinase) at therapeutic levels in plasma samples. Since it adopts the standard 96-well microplate format (or even, in principle, plates with a larger number of wells) without the need for additional reagents for the detection of polycations such as protamine, this methodology may be attractive for carrying out a wide array of biomedically important assays at a low cost with high sample throughput. ACKNOWLEDGMENT The authors thank Professor Michael D. Morris, Department of Chemistry, The University of Michigan, for his help in setting up the optical microscope imaging system used in this work. We gratefully acknowledge the National Institutes of Health (Grant GM28882) and Medtronic Inc. for financial support of this work. Received for review January 19, 2000. Accepted April 19, 2000. AC000060N Analytical Chemistry, Vol. 72, No. 14, July 15, 2000
3149
