ISSN:1359-7337    

DOI:10.1039/AC99633FX001

出版商:RSC    

年代:1996

数据来源: RSC

ISSN:1359-7337    

DOI:10.1039/AC99633BX003

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, January 1996, Vol33 (5-8) 5 On-line lmmunoaffinity Chromatography- High-performance Liquid Chromatography- Mass Spectrometry for the Determination of Dexamet hasone Colin S. Creasers, Stephen J. Feely", Edward Houghtonb, Mark Seymourb and Philip Tealeb a Department of Chemistry and Physics, Nottingham Trent University, Clifton Lane, Nottingham, UK NGII 8NS Newmarket, SufSolk, UK CB8 7DT Horseracing Forensic Laboratory Limited, PO Box 1.5, Snailwell Road.The determination of the synthetic corticosteroid dexarnethasone in equine urine is described using coupled on-line immunoaffinity chromatography-high- performance liquid chromatography (IAC-HPLC) with detection by UV or mass spectrometry. The limit of detection for dexamethasone with UV detection was 30 pg 1-1 for a 10 ml urine sample.Atmospheric pressure chemical ionization mass spectrometry, with single ion monitoring, gave a limit of detection of 0.1 pg 1-1 for dexamethasone. The determination of dexamethasone by on-line IAC-HPLC-MS has been demonstrated in post-administration equine urine samples. The specificity that can be accomplished by the binding of analytes in a complex biological matrix to the antibodies of an immunoaffinity chromatography (IAC) stationary phase provides a very efficient sample clean-up.This specificity has been utilized for on-line sample preparation of biological samples prior to HPLC with UV detection.lJ Mass spectrometric detection in IAC-HPLC has also been reported, employing atmospheric pressure ionization for the determination of propranolol and LSD in urine, using a short trapping column to interface the IAC and HPLC,3 and continuous-flow fast atom bombardment ionization for the determination of diethylstilbestrol in the urine of rats and calve^.^ These reports show the potential of IAC-HPLC-MS for the determination of analytes at low levels in complex biological matrices.The detection of the abuse of high-potency synthetic corticosteroids in equine sports presents an analytical problem as administration at low levels and extensive metabolism results in low concentrations of drug/metabolites in biological fluids.The synthetic corticosteroids are used primarily for their anti- inflammatory effect in the horse and are prohibited substances under the Rules of Racing.Dexamethasone (9a-fluoro-1 lJ3, 17a, 21-trihydroxy-lba- methylpregna- 1,4-diene-3,2O-dione) is one of the major syn- thetic corticosteroids used in veterinary practice. It is relatively potent. Metabolic studies have shown the production of a number of phase I and phase I1 metabolites and rapid elimination of the parent Antibodies for dexametha- sone are readily available and thus it was an ideal candidate for study by on-line IAC.Techniques reported for the determination of dexamethasone and associated synthetic corticosteroids are numerous.*-*2 Radioimmunoassay (RIA) is used widely for screening equine urine samples, but lacks the selectivity to allow the unambigu- ous identification of the analyte because of cross-reaction with other related corticosteroids.In this paper we report a novel approach which provides the required selectivity and sensitivity for the detection of dexamethasone in equine urine by combining on-line IAC-HPLC with mass spectrometric de- tection. Experimental Materials Sepharose 4B CNBr-activated and Protein G stationary phases were obtained from Pharmacia (Uppsala, Sweden). Rabbit serum containing anti-dexamethasone antibody was provided by the Horseracing Forensic Laboratory (Newmarket, UK).Dexamethasone, sodium dihydrogen orthophosphate, sodium hydrogen orthophosphate, sodium acetate and sodium azide were purchased from the Aldrich (Gillingham, Dorset, UK). Propionic acid and methanol (Distol grade) were obtained from Fisons (Loughborough, Leicestershire, UK).Water was ob- tained from a MilliQ system (Millipore, Bedford, MA, USA). All eluants were filtered through 0.45 ym Millipore filters. Standard and Sample Preparation A stock solution of dexamethasone was prepared in methanol at a concentration of 1 mg ml-1. Further dilutions were prepared to give concentrations in the range 0.2-50 pg I-' for standard analysis and spiking experiments.Spiked and post-administra- tion urine samples (20 ml) were adjusted to pH 7.0 and centrifuged at 1500g for 10 min prior to analysis. The supernatant was removed and 10 ml applied to the IAC-HPLC system. Anti-dexamethasone IAC Column The crude rabbit serum samples were purified using a Protein G stationary phase (bed height 4 cm) packed in a Clo column (Phannacia, Uppsala, Sweden).The appropriate fractions were collected, pooled and bound to CNBr-activated Sepharose 4B by the method described by the man~facturer.~~ The prepared anti-dexamethasone stationary phase was packed into a C 10 column (Pharmacia, Uppsala, Sweden) to give a bed height of 5.4 cm. When not in use the IAC column was stored at 4 "C in 20 mmol 1-1 phosphate buffer containing 0.5 mol 1-1 NaCl + 0.2% sodium azide.Instrumentation IAC-HPLC-UV The IAC-HPLC-UV instrumentation [Fig. 1 (a)] consisted of a Waters, Model 6000A, HPLC pump (Bedford, MA, USA)6 Analytical Communications, January 1996, Vol33 (pump 1) which delivered mobile phase to the IAC column via a six-port injection valve (Rheodyne 7010) containing a 10 ml stainless steel injection loop (12.7 m X 1 mm id). The mobile phase flowed from the IAC column to V 1, a six-port switching valve (Rheodyne 7010) fitted with a 5 ml switching loop (6.4 m X 1 mm id).This enabled eluted sample fractions to be transferred from the low pressure IAC column to the high pressure HPLC column. Vl was connected to two Waters 501 HPLC pumps (Bedford, MA, USA) (pumps 2 and 3). Pump 2 was used to flush the IAC band from the switching loop onto the HPLC column in water and pump 3 delivered the mobile phase to chromatograph the concentrated sample.The analytical separation was carried out on a 250 X 4.6 mm id (5 pm) ODS column (Hichrom, Reading, Berkshire), and the effluent was transferred to a Waters Model 44 1 UV absorbance detector (A = 254 nm). IAC-HPLC-MS The instrumentation for IAC-HPLC-MS is shown schematic- ally in Fig.l(b). It differed from the IAC-HPLC-UV set-up at valve Vl. The sample was flushed from the switching loop with mobile phase delivered by a Waters 501 HPLC pump (Bedford, MA, USA) (pump 2) to a T-piece fed by a second Waters 501 HPLC pump (Bedford, MA, USA) (pump 3). The sample passed through a 5 ml mixing loop (6.4 m X 1 mm id) to V2, a (4 20 mmol 1-l Phosphate Injector buffer + 0.5 mol 1-l NaCl (10 ml loop) (PH 7.0) + 0.5 mol I-' NaCl Switching loop (5 mu Waste 0.05 rnol 1-l Acetate Injector buffer (pH 7.0) >p"mplE) l_, 50%MeOH 1 in 1 mol 1-l propionic acid IICl water* x, Switching loop (5 mi) MeOH -water gradient Fig.1 HPLC-MS system. Schematic diagram of (a) IAC-HPLC-UV system and (b) IAC- six-port valve (Rheodyne 7010), and on to the reversed-phase analytical column.A Hewlett-Packard 1050 LC delivery system (pump 4) delivered a methanol-water gradient to the column which was interfaced to a quadrupole mass spectrometer (VG Platform, Altringham, Cheshire) fitted with an atmospheric pressure chemical ionization (APCI) interface. Chromatographic Procedures IAC-HPLC-UV Mobile phase, 20 mmol 1-1 phosphate buffer and 0.5 mol 1-I NaC1, pH 7.0, was delivered to the IAC column at 1 ml min-* and a 10 ml sample was injected.The IAC column was flushed with mobile phase for 15 min, then with the switching valve (Vl) set to the load position, eluted with 1 mol 1-1 propionic acid with 0.5 mol 1-1 NaCl at 1 ml min-1. At 30 min V l was switched to the inject position and the chart recorder was started.The contents of the switching loop were flushed onto the reversed phase analytical column with water at 1 ml min-1. The sample was then eluted with 50% methanol-water (v/v) at 1 ml min-1. IAC-HPLC-MS Acetate buffer (0.05 mol 1-1, pH 7.0) was delivered to the IAC column at 1 ml min-1 and the sample (10 ml) was injected. The switching valve (Vl) was set to the load position and the IAC column was washed with buffer for 15 min; then eluting phase, 50% MeOH in 1 mol 1-1 propionic acid (v/v), was passed through the IAC column at 1 ml min-I.After 30 min V1 was switched to the inject position and the eluted band was transferred to the reversed phase analytical column in water (1 ml min-I) with an additional 2 ml min-1 of water (pump 3) via the 5 ml mixing loop.At 55 min the analytical column was switched on-line with pump 4 (Hewlett-Packard l050), which eluted the sample with a 0-100% water-methanol gradient (15 min) followed by 100% methanol at 1 ml min-1 for 5 min. Mass spectral detection was started 55 rnin after injection. Dex- amethasone was determined by selected ion monitoring of m/z 333 and 393 with a dwell time of 100 ms.Results and Discussion Initial investigations of IAC sample clean-up using anti- dexamethasone antibodies in a Sepharose column coupled directly to HPLC via a loop interface were carried out with UV detection. The dexamethasone was eluted from the IAC column with aqueous propionic acid. Experiments with tritium-labelled dexamethasone established that the analytes eluted from the IAC column in a 5 ml band and, by correct timing of valve switching, this band could first be flushed onto the switching loop and then onto the reversed-phase analytical column.The dexamethasone was focused at the head of the column due to its strong interaction with the ODS stationary phase in the presence of aqueous mobile phase. The focused band was then chromato- graphed isocratically with methanol-water mobile phase.The total recovery of dexamethasone for the IAC-HPLC system was 70%, and this procedure allowed the detection of dex- amethasone spiked in urine at a level of 30 pg 1-l (Fig. 2). The limit of detection was determined not by the minimum detectable amount of dexamethasone, but by the interfering peaks observed in the chromatogram which originated from the equine urine matrix.Non-selective adsorption by the IAC sepharose stationary phase was the most likely reason for the retention of matrix components by the IAC column. The IAC- HPLC-UV system established the potential of the coupling of the anti-dexamethasone IAC column with the HPLC column via a loop interface, but did not exhibit the required selectivity forAnalytical Communications, January 1996, Vol33 7 the analysis of post-administration equine urine samples for dexamethasone, due to the potency (and therefore low dose) of this synthetic corticosteroid.The combination of IAC-HPLC with mass spectrometry was investigated using an atmospheric pressure chemical ionization (APCI) interface. The mobile phase used to elute dex- amethasone from the IAC column was changed to 1 mol 1-l propionic acid in methanol-water for IAC-HPLC-MS, because of the incompatibility between the sodium chloride, used in the IAC eluting phase for the IAC-HPLC-UV system, and the APCI interface of the MS detector.The elution of dex- amethasone from the IAC column by the methanol-aqueous propionic acid mobile phase was determined using tritium- labelled dexamethasone as before and the dexamethasone was observed to elute in a 3 ml band between 11 and 14 min after the IAC column was switched to the eluting phase.However, the methanol in the IAC eluting phase presented a problem, since it prevented focusing of the dexamethasone at the head of the ODS column. In order to overcome this difficulty, a dilution stage was introduced which involved the addition of water to the IAC eluent in a 5 ml mixing loop ahead of the reversed phase analytical column [Fig.l(b)]. This gave the necessary focusing at the head of the HPLC column and the dexamethasone was (a) Absorbance at 254nm 0.031 0.02( 0.01( 0.0oc 90 Timelminutes (b) Absorbance at 254nrn 0.030 0.020 0.010 + H,O 0.30 90 Timdminutes Chromatograms from IAC-HPLC-UV analysis of (a) blank urine Fig. 2 and (b) 30 pg 1-1 dexamethasone spiked in urine. then chromatographed using a water-methanol gradient over 15 min. Under APCI conditions, dexamethasone showed a protonated molecular ion, [M + HI+, at m/z 393 and a prominent fragment ion, corresponding to loss of the C-17 side chain, at m/z 333.Selected ion monitoring (SIM) of the m/z 333 and 393 ions and the chromatographic retention time were used to confirm the presence of dexamethasone following IAC-HPLC clean-up. The IAC-HPLC-MS system gave a recovery of 75% for dexamethasone and had a limit of detection calculated at 0.1 pg 1-1 (signal: noise = 3 : 1). A good linear response, with a correlation coefficient of r = 0.995, was obtained for spiked urine samples in the range of 0.2-50 pg 1-1.Replicate injections of a spiked sample urine (1 pg 1 - 1 ) showed a relative standard deviation of 6.8%. The on-line combination of IAC-HPLC-MS eliminates the need for extraction and derivatization procedures which are necessary when using IAC followed by off-line GC-MS. This methodology is also simpler in terms of sample manipulation and, with less manual intervention, there are fewer stages where potential losses or contamination might occur. The IAC- HPLC-MS system has been used to analyse both equine samples spiked with dexamethasone and post-administration samples.Fig. 3(a) gives the selected ion chromatogram (m/z 333 + 393) from a sample spiked with dexamethasone (retention time, 14.7 min) at 5 pg 1 - 1 .Fig. 3(b) shows the chromatogram from a sample collected 8 h after a single intra-muscular injection (20 pg kg-1) of a Dextran preparation of dex- amethasone to a thoroughbred horse (retention time 14.8 min). The concentration of dexamethasone in this sample was measured as 4 pg 1-1. Conclusion On-line IAC-HPLC coupled with APCI-MS is shown to be an effective method for the confirmatory analysis of dexamethas- 50 c 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 (b) 14.80 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 Retention tirne/rnin Fig.3 Selected ion chromatogram (m/z 333 + m/z 393) of (a) a 5 pg 1-l dexamethasone-spiked urine sample, and (b) an 8-h post-administration urine sample.8 Analytical Communications, January 1996, Vol33 one in equine urine at concentrations down to 0.1 pg I-’. The use of APCI-MS, with its increased detector selectivity for dexamethasone compared with UV detection, allows the technique to be applied to the determination of post-adminis- tration samples.The authors acknowledge the Engineering and Physical Sci- ences Research Council (UK) and the Horserace Betting Levy Board for financial support.References Farjam, A., de Jong, G. J., Frei, R. W., Brinkman, U. A. Th., Haasnoot, W., Hamers, A. R. M., Schilt, R., and Huf, F. A., J . Chromatogr., 1988,452,419. Haasnoot, W., Ploum, M. E., Paulussen, R. J. A., Schilt, R., and Huf, F. A., J . Chromatogr., 1990, 519, 323. Rule, G. S., and Henion, J. D., J . Chromatogr., 1992, 582, 102.Davoli, E., Fanelli, R., and Bagnati, R., Anal. Chem., 1993, 65, 2679. Dumasia, M. C., Homer, M. W., Houghton, E., Moss, M. S., Chakraborty, J., and Marks, V., Biochem. Soc. Trans., 1976, 4, 119. Dumasia, M. C., Houghton, E., Moss, M. S., and Chakraborty, J., Proceedings of the 3rd Sympossium on Equine Medication Control, Lexington, Kentucky, 1979, p. 247. Dumasia, M. C . , Houghton, E., Moss, M.S., Chakraborty, J., and Marks, V., J . Steroid Biochern., 1986,25, 547. Chapman, D. I., Moss, M. S., and Whiteside, J., Vet. Rec., 1977,100, 447. Houghton, E., Dumasia; M. C., and Wellby, J. K., Biomed. Mass Spectrom., 1981, 8, 558. 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Houghton, E., Teale, P., and Dumasia, M. C., Analyst, 1984, 109, 273. Gaskell, S. J., and Sieckman, L., Clin.Chem. (Winston-Salem N . C.), 1986, 32, 536. Ishibashi, M., Takayama, H., Nakagawa, Y., and Harima, N., Chem. Pharm. Bull., 1988, 36, 845. O’Connor, P. J., Morgan, W. P., Raglin, J. S., Barksdale, C. M., and Kalin, N. H., Psychoneuroendocrinology, 1989, 14, 303. Maclean, C. J., Booth, C. W., Tattersall, T., and Few, J. D., Eur. J. Appl. Physiol., 1989, 58, 341. Park, J., Park, S., Lho, D., Choo, H. P., Chung, B., Yoon, C., Min, H., and Choi, M. J., J . Anal. Toxicol., 1990, 4, 66. Shibaski, H., Furata, T., Kasuya, Y., Okase, T., Katoh, T. Kogo, T., and Hirayama, T., Biomed. Environ. Mass Spectrom., 1990, 19, 225. Jap, B. K., Johnston, G. A. R., and Kazlauskas, R., J. Chromatogr., 1992,573, 183. Shibasaki, H., Furuta, T., and Kasuya, Y., J . Chromatogr., 1992,579, 193. Stanley, S. M. R., Wilhelmi, B. S., Rodgers, J. P., and Bertschinger, H., J . Chromatogr., 1993, 614, 77. Stanley, S. M. R., Wilhelmi, B. S., and Rodgers, J. P., J. Chromatogr., 1993,620, 250. Courtheyn, D., Vercammen, J., De Brabander, H., Vandenreyt, I., Batjoens, P., Vanoosthuyze, K., and Van Peteghem, C., Analyst, 1994,119,2557. Stanley, S. M. R., Wilhelmi, B. S., Rodgers, J. P., and Guthrie, A., Biol. Mass Spectrom., 1994, 23, 483. Affinity Chromatography, Principles and Methods, Product Informa- tion, Pharmacia, Uppsala, Sweden, 1988. Paper 5lO7542D Received November 20, 1995 Accepted November 29,1995

ISSN:1359-7337    

DOI:10.1039/AC9963300005

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, January 1996, Vol33 (9-1 0) 9 Improved Method for the Determination of Ascorbic Acid in Beer by Using High-performance Liquid Chromatography With Electrochemical Detection David Madigan", Ian McMurrough" and Malcolm R. Smythb a Guinness Brewing Worldwide Research Centre, St. James's Gate, Dublin 8, Ireland a School of Chemical Sciences, Dublin City University, Dublin 9, Ireland A method for the determination of ascorbic acid in beer by using HPLC with dual electrode electrochemical detection is described.Ascorbic acid was separated by using ion suppression chromatography on a C18 column and detected by a dual-electrode system consisting of a porous graphite high efficiency electrode followed in series by a glassy carbon amperometric electrode. It was easily detectable at sub-ppm levels in beer.The use of dual electrode detection provided a constant measurement of both peak purity and electrode efficiency. Ascorbic acid (AA) is a powerful antioxidant and unlike many other antioxidants it has the ability to reduce molecular oxygen.' It is widely used as an antioxidant in the food and beverage industry, and some brewers add AA at 30-50 mg 1-l to beer in order to diminish residual oxygen after packaging. This is thought to delay the development of stale flavours in beer (believed to be a result of oxidative chemical reactions.2) The sensitivity of AA to oxygen, however, makes the analysis of the reduced form very difficult, particularly if the sample must undergo an extraction technique that permits oxidation. Therefore it is desirable to measure AA by rapid, sensitive methods which minimize contact of the sample with atmo- spheric oxygen.The standard AOAC method for AA determi- nation3 uses a titration based on the reduction of 2,6-di- chlorophenol indophenol. This method, however, is subject to numerous interferences in samples such as beers or fruit juices, which may contain substantial amounts of reducing substances such as flavanols, melanoidins or reductones.Added to this, the end-point of the titration is poorly defined, and this increases the possibilities for error in this method. The use of HPLC with UV absorbance detection is applicable to AA determination$" but is insufficiently sensitive for the determination of very low levels of AA.Sample preconcentration steps may be necessary, which may lead to the unwanted risk of absorption of oxygen by the sample. A suitable alternative to these methods is the use of electrochemical detection with HPLC. AA is very easily oxidized at a glassy carbon working electrode at low operating potentials, thereby providing a sensitive detection system. The ease of oxidation of AA means that potentially interfering analytes with higher oxidation potentials are not detected, so selectivity is high.While HPLC with electrochemical detection for AA determination has been widely documented,7-' 1 many reported methods suffer from either a lack of reproducibility or from losses of sensitivity due to fouling of the electrode with sample components. This has necessitated bracketing of samples and standards, which, as well as being troublesome to perform, greatly reduces the number of samples which may be analysed in a given time.Here, we present a method for the determination of AA in beer which provides excellent repeata- bility and long-term electrode stability, and which may also be applied to other beverages. Furthermore, the detection system used in this study possessed additional advantages over the more commonly used single-cell amperometric devices.The high efficiency (coulometric) electrode is claimed to be capable of oxidizing all of the sample which passes through it, due to the large electrode surface area. This should therefore provide added sensitivity and repeatability when compared to ampero- metric detectors, which, due to diffusion limitation, typically only oxidize 10% of the analyte.In addition, the response of the downstream amperometric electrode, which is operated at a higher potential, can be used as a valid indicator of peak purity and/or electrode poisoning. The ratio of current responses at the two electrodes is dependent on the shape of the hydrodynamic wave of the analyte, and also on the operating efficiency of the electrodes. Experimental Reagents and Instrumentation All reagents were of analytical-reagent grade.De-ionized water was prepared using an Elga Primamaxima purification System (Elgar, High Wycombe, Buckinghamshire). HPLC was per- formed on a Perkin-Elmer Integral 4000 HPLC system equipped with a photodiode-array detector (Perkin-Elmer, Beaconsfield, Buckinghamshire, UK).An ESA Analytical Coulochem I1 electrochemical detector (ESA Analytical, Hun- tingdon, Cambridgeshire, UK) fitted with a Model 5011 high sensitivity analytical cell was placed downstream of the photodiode-array detector. The cell contained a porous graphite high efficiency electrode placed upstream of a glassy carbon amperometric working electrode.The detector was operated under the following conditions for routine analysis, Channel 1 (high efficiency electrode): potential, +40 mV; output range, 10 pA; offset, +5%; and filter, 1 s. Channel 2 (amperometric electrode): potential, +350 mV; output range, 10 PA; offset, +5%; and filter, 1 s. The column used was a Waters Radial Compression Module (RCM) containing a cartridge ( 10 cm X 8 mm id) packed with 5 pm Resolve octadecyl stationary phase (Waters UK, Watford, Hertforshire, UK).This stationary phase was recommended for use at low pH in preference to other octadecyl phases available from Waters. The mobile phase was 50 mmol l-I potassium dihydrogen orthophosphate, 500 mg 1-1 Na2EDTA, adjusted to pH 3.0 with 10 moll-' orthophosphoric acid and supplied at a flow rate of 1 .O ml min- 1.The injection volume was 10 pl. Calibration Standard solutions of AA were prepared at 0.5-20 mg 1-1 in the mobile phase described above. These solutions were stored cold under N2 when not in use, and were prepared fresh daily. Before calibration of the HPLC system, 2 X 100 pl injections of a10 Analytical Communications, January 1996, Vol33 solution containing 1 g 1- AA were made, followed by 2 X 100 p1 injections of mobile phase.This removed residual oxidizing species from the system and improved repeatability. Sample Preparation Samples (50 ml) were de-gassed by helium sparging, avoiding any contact with atmospheric oxygen, and were mixed with an equal volume of mobile phase. A sub-sample (10 pl) of the resulting solution was injected immediately.Results and Discussion Sample Preparation Significant lossses of AA occurred during sample preparation if oxygen was not totally excluded. Even when very gentle agitation was used to release dissolved CO2, losses of 10-20% of the measurable AA were not uncommon. When helium de- gassing was used, however, repeatable results and high recoveries were obtained. Although this severely limits the number of samples which may be processed on a given day, it is an absolute requirement for accuracy.Separation Conditions The method of Bode and Rose12 was slightly modified. The buffer strength was reduced from 200 to 50 mmoll-1, in order to improve column lifetimes, and Na2EDTA was added to the mobile phase at 500 mg 1-1 to inhibit metal-ion-catalysed autoxidation of AA.By operating at pH 3.0 it was possible to separate AA, which has pK, values of 4.0 and 11.3,l by ion suppression HPLC, and thereby dispense with the requirement for corrosive ion-pairing reagents. The column used was very stable at this pH. Although a Waters Nova-Pak 4 pm stationary phase gave almost equally good resolution of AA, the Resolve packing material was chosen because of its long-term durability at low pH.Optimization of Detection Conditions A hydrodynamic voltammogram is illustrated in Fig. 1, in which the potential of the upstream high efficiency electrode was varied from -300 to +400 mV, while maintaining the downstream amperometric electrode at a constant potential of +350 mV.At the chosen combination of analytical potentials (+40 mV and +350 mV at the high efficiency and amperometric electrodes, respectively) the ratio of peak areas from the two electrodes remained constant within the linear range of the method. Provided that the electrode response is reproducible, any gross departure from this ratio in samples would be indicative either of the presence of interfering compounds or of electrode fouling.This is of particular importance when 2.5 Q, Excessive Background Current J 0 Coulometric Electrode (variable potential) a" 1.0 0 Arnperometric Electrode (fixed potential of +350 mV) I 0.01 ' I ' I ' I ' I ' I " " -300 -200 -100 0 100 200 300 400 Applied PotentiaVrnV vs Pd reference Fig. 1 Hydrodynamic voltammetry of ascorbic acid.analysing beer, which contains a wide variety of reducing substances.13 Method Performance The method was linear in the range 0.5-20 mg 1-1 AA, which was suitable for the applications described here. Note that at higher concentrations of AA the high efficiency electrode began to exhibit a non-linear response, whereas the amperometric electrode response remained linear. This is important both from the point of view of calibration and also if peak area ratioing is used as a measure of peak purity.The limit of detection was 0.2 mg 1-1 AA in beer for the method as described, although higher sensitivity could be obtained by varying the electrode potentials and injection volume. Twenty consecutive injections of a beer sample containing 10 mg 1-1 added AA, diluted with an equal volume of mobile phase, gave s, values in peak area of 1.3 and 1.4% at the coulometric and amperometric detectors, re- spectively, and in peak height of 1.2 and 0.9%, respectively.This demonstrated not only the repeatability of the analysis, but also the resistance of the electrodes to fouling by sample components. The inclusion of Na2EDTA was vital to maintain repeatability of the analysis.Eight consecutive injections, made over an 80 min period, of a beer sample without Na2EDTA added, showed a gradual decrease to 50% of the starting AA peak area. A similar effect was observed for a standard solution of AA prepared in water without Na2EDTA addition. Sample stability was found to improve with Na2EDTA present, and maximum stability was obtained through the addition of 0.5 g 1-l Na2EDTA to the mobile phase, and by using this mobile phase to dilute samples before analysis.The method was also used for the analysis of a limited number of non-alcoholic vitamin drinks and fruit juices. Conclusions The described method was shown to be suitable for the analysis of AA in beer, and should be adaptable to the analysis of other beverages or solid foodstuffs. This method offers distinct advantages over previously reported methods for analysis of AA in beer in terms of accuracy, freedom from interferences, and repeatability.Furthermore, use of the electrochemical detector allows the determination of low residual AA levels in beers that have deteriorated during prolonged storage. References 1 2 3 4 5 6 7 8 9 10 11 12 13 Wong, D.S. W., Mechanism and Theory in Food Chemistry, Van Nostrand Reinhold, New York, 1989, ch. 10, pp. 359-365. Irwin, A. J., Barker, R. L., and Pipasts, P., J . Am. SOC. Brew. Chem., 1991, 49, 140. Official Methods of Analysis of the Association of Official Analytical Chemists, AOAC, Arlington, VA, USA, 14th edn., 1984, sec. 43.064. Finley, J. W., and Duang, E., J . Chromatogr., 1981, 207, 449. Seiffert, B., Swaczyna, H.. and Schaefer, I., Dtsch. Lebensm. Rundsch., 1992, 88, 38. Nisperos-Carriedo, M., Buslig, B. S., and Shaw, P. E., J. Agric. Food Chem., 1992,40, 1127. Knudson, E. J., and Siebert, K. J., J . Am. Soc. Brew. Chem., 1987,45, 33. Moll, N., and Joly, J. P., J . Chromatogr., 1987, 405, 347. Iwase, H., and Ono, I., J. Chromatogr. A . , 1993, 654, 215. Leubolt, R., and Klein, H., J . Chromatogr., 1993, 640, 271. Felton, S . P., Grace, R., and Halver, J. E., J . Liq. Chromatogr., 1994, 17, 123. Rose, R. C., and Bode, A. M., Biochern. J., 1995,306, 101. Chapon, L., Louis, C., and Chapon, S., European Brewing Conven- tion, Proceedings of the 13th Congress, Estoril, 1971, Elsevier, Amsterdam, 1972, p. 307. Paper 510 7096A Received October 27, 1995 Accepted December 5, 1995

ISSN:1359-7337    

DOI:10.1039/AC9963300009

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, January 1996, Vol33 ( I 1-13) 11 Ion Chromatography Coupled With Inductively-coupled Argon Plasma Mass Spectrometry: Multielement Speciation as Well as On-line Matrix Separation Technique Jorg Feldmann" Institute of Environmental Analytical Chemistry, University of Essen, 0-45127 Essen, Germany Coupling of an ion chromatograph used for anion chromatography with an inductively coupled argon plasma mass spectrometer is used for the simultaneous determination of common anions and anions of heavy elements without any matrix separation techniques.Up to 0.1 % sodium chloride does not interfere significantly with arsenic determination. In addition to its use for matrix separation, the method can be used as an element speciation technique, which is shown by the analysis of urine to determine arsenic.Most of methylated organometallic compounds in liquids are rather unstable complexes, which can easily be destroyed by changing the conditions. During analytical separation by HPLC the pre-existing equilibrium may be shifted. When determining low concentrations, the analytical methods include enrichment,' which shifts the equilibrium in the sample and may alter the species.A method without clean-up or enrichment steps is desirable, especially when speciation analysis is being carried out. Screening for metal and metalloid compounds in liquids requires an extremely sensitive multielement detection method, such as inductively coupled argon plasma mass spectrometry (ICP-MS). The limiting features associated with ICP-MS include loss of molecular information as well as the occurrence of polyatomic ion interferences, usually in the region below 90 u.If the chloride content of a sample is high enough (e.g., sea- water or urine), because of the interference of the polyatomic ion 75(40Ar35Cl+) with 75As, determination of arsenic will be unattainable. The fact that arsenic is a mono-isotopic element prevents the use of a different isotope to resolve the interference problem.In the environment, many elements occur in different species of different toxicity, mobility and lipophilicity . Trivalent arsenic is more toxic than pentavalent arsenic. Inorganic and methylated arsenic species have been determined in sea-water, rivers2 and human urine.3 Most of the samples contain a high concentration of chloride and sulfur anions, which react with the polyatomic ions mentioned above.Matrix separation techniques for ICP-MS analysis are described extensively in the literature. It is well known that chloride interference as well as an increase of the sensitivity is achieved by hydride generation techniques with cryogenic trapping prior to ICP-MS analysis.In the inductively coupled argon plasma itself, the formation of argon chloride ions is also * Present address: Department of Chemistry, 2036 Main Mall, The University of British Columbia, Vancouver, BC, V6T 1Z1, Canada. E-mail:joerg@chem.ubc.ca influenced by the content of nitrogen in the plasma gas. Branch et al.4 and Ford et al.5 have shown that the addition of nitrogen to the carrier gas reduced the value of the polyatomic ion 75(ArC1+) to a negligible value with a chloride content of up to 1%, because the formation of 49(NCl+) is preferred.Ion chromatographic,6 as well as HPLC,7 techniques can be used for the speciation of arsenic (As"' or As"). Also, the anions of arsenic, dimethylarsinic acid (DMA) and monomethyl- arsonic acid (MMA) can be separated.If the retention times of the chloride species are different from the retention times of the arsenic species, no interference will be observed in inductively coupled argon plasma mass spectrometry. The aim of this investigation is to demonstrate the technical procedure for coupling ion chromatography to inductively coupled plasma mass spectrometry as a multielement-multi- species technique by using urine as an environmental sample with high chloride concentrations.Experimental The ICP-MS used was a VG Plasmaquad PQ2+ (VG Elemental, Winsford, Cheshire, UK), which is designed with a nebulization system consisting of a water-cooled Scott spray chamber and a de Galan V-groove nebulizer. The ion chromatograph (IC) used was a Dionex DX500 (Dionex, Sunnyvale, CA, USA), which contains a conductivity detection system. The AS4A-SC column (id 4 mm, 23 cm length) with an AG4A-SC precolumn (id 4 mm, 3.5 cm length) was used in combination with a Dionex suppression technique. As a reagent blank, doubly distilled water was used.An isocratic pump, applied to just one mobile phase, was used for the separation described. Following detection in the conductivity cell, the whole sample is transported to the nebulizer without any peristaltic pump.The flow rate was optimized according to the separation of arsenic and chloride species. This resulted in a very high solution uptake rate of the nebulizer, which led to the use of a low spray chamber temperature to decrease the liquid uptake. Table 1 shows the instrumental parameters of the IC and ICP-MS.Results and Discussions For multielement detection the following isotopes are meas- ured: 3lP, 48(SO+), 5I(ClO+), 75As, 77Se, 81Br, 82Se, I2iSb and 1271. The combination of the two different detectors gives information such as is shown in Fig. 1. The multielement- multispecies chromatogram of a standard sample is shown, which contains the non-metallic anions phosphate, bromide, chloride, nitrate and sulfate, as well as arsenite, arsenate, selenite and selenate.The most common anions that are difficult to detect by ICP- MS can be detected by conductivity, these being chloride,12 Analytical Communications, January 1996, Vol33 nitrate, sulfide, phosphate, bromide and iodine. The anions can be detected in the mg 1-1 level.Although the detection of phosphate is also possible up to 0.1 mg 1-l by using ICP-MS, a high background on m/z 31 (300000 counts per second) probably occurs by forming 15N160. The concentrations of the heavy element anions as well as of the organometallic anions, are mostly at the pg 1-1 level in environmental samples (geothermal waters, urine). Nevertheless, those anions are detectable by ICP-MS in a liquid sample.Thus, metalloid anions such as arsenite, arsenate, MMA, DMA, selenite and selenate, as well as phosphate, nitrate, sulfate and chloride, are detectable simultaneously in one sample without any enrich- ment steps. Table 2 shows the data concerning the linearity of the determination of arsenite and arsenate in the standard solution (doubly distilled water, < 100 nS). There is a correlation coefficient of r2 = 0.9900 for 10 and 100 ng ml-1 and a detection limit of about 3.7 ng ml-l for arsenite, and 2.2 Table 1 The operating parameters of the IC-ICP-MS coupling ICP-MS Power Coolant flow (argon) Auxiliary flow (argon) Injection flow (argon) Sample uptake Nebulizer Spray chamber Sample tube/PTA Sampler, skimmer Interface pressure Analysis mode Quadrupole pressure IC Precolumn Analyser column Detector Reagent Sample loop Mobile phase A Mobile phase B VG Plasmaquad PQ2 Turbo Plus 1350 W 13 1 min-l 1.3 1 min-I 0.900 1 min-1 2.00 ml min-1 de Galan Scott, cooled at +4 “C 1/16” Ni (orifice diameter, 1 mm, 0.7 mm) 2.3 mbar TRA, 1 s time slice in peak hopping, 1 channel per u 2.8 X 10-6mbar Dionex DX-500 AG4A-SC (id 4 mm, I 3.5 cm) AS4A-SC (id 4 mm, I 2 3 cm) Anionic suppression, conductivity cell, ED 40 Doubly distilled water, < 100 nS 0.020 ml 1.44 mol 1-’ sodium carbonate, 5 mmol 1-I sodium tetraborate 1.36 mmol sodium hydrogencarbonate I h c .- C 2 L .- v) a, c 5 = f 2 MMA 3 AsO,* 4 Se0,2- 5 Se0,2- 6 Br- 7 I- 8 CI- 9 Br- 10 NO,- I 1 P0,3-/Se0,2- 13 AsO,> 14 I- 12 so:- ng ml-’ for arsenate, respectively.Thus, the method shows adequate sensitivity and the requirements for environmental studies. Fig. 1 shows that chloride is well separated from arsenite and arsenate, which results in a matrix elimination on mass 75. By varying the chloride concentration from 0.001 to 0.5% as sodium chloride, the determination of the arsenite (30 ng ml-l) and arsenate (1 8 ng ml-l) is tested for interference.The method can interpret spiking/recovery experiments with an artificial sample with known chloride content. The determination of the arsenite and arsenate concentration by ICP-MS depends on the chloride concentration, as is shown in Fig. 2. The spiked concentrations are set at 100%. Although an interference appears to start above 0.01% for arsenite, the arsenite concentration is determined within a confidence level of 99.6% by varying the chloride concentrations up to 0.1 %.No significant dependence is seen for arsenate up to a chloride concentration of 0.1%. Above this level neither a good separation nor the determination of arsenite and arsenate is sufficient, and a dilution of the samples is necessary. The ion extraction from the plasma into the interface of the ICP-MS is stable up to concentration of 0.1% chloride.A signal depression of 50% is found at a concentration of 0.5% chloride. Obviously, the determination of arsenite is more influenced by the chloride concentration than by arsenate, because of the similar retention time to that of chloride. The matrix separation and speciation technique is not limited to inorganic species; organometallic anions can also be separated.MMA and DMA are interesting species in respect of the biochemical transformation of arsenic in mammals or humans. In particular, the urine of an uncontaminated person contains both species at the ng ml-l level. The separation of arsenite, arsenate, MMA and DMA is performed by a standard solution of monomethylarsenic Table 2 Parameters for the quantification of arsenite and arsenate Calibration Species equation Detection limits Reproducibility* Arsenite y = 816x - 52t 3.7 ng ml-1 (30) 6.3% (7 replicates) Arsenate y = 719x + 164t 2.2 ng ml-I (30) 2.5% (7 replicates) r2 = 0.9900 r2 = 0.9977 * Based on 18 ng ml- I arsenate and 30 ng ml-l arsenite without any t x, Concentration in the sample; y, peak area (blank subtracted).chloride matrix. ([I c I . I I , I I * I I C 1 5 10 50 loo 500 lo00 5000 .- .- 0 I hsenate Fig. 1 Determination of a standard solution, which contains the common anions (chloride, bromide, iodine, nitrate and phosphate) at ppm level as well as heavy element anions (selenite, selenate, arsenite, arsenate, MMA, DMA) at the ppb level.IC-ICP-MS allows the simultaneous detection of over-all conductivity and the element-specific detection of inductively coupled plasma mass spectrometry. Log c(NaCl)lmg I-’ + Fig. 2 Solutions, 30 ng ml-I of arsenite and 18 ng ml-l of arsenate, spiked to different chloride concentrations. The error bars show the variation of the non-chloride solution (99.6% confidence level) and can be used to show the variation of the recovery.Analytical Communications, January 1996, Vol33 13 dibromide, dimethylarsonic bromide (both obtained from Alfa Chemicals, Bologna, Italy) and sodium arsenite and arsenate (both obtained from Kraft, Duisburg, Germany).The mobile phase of HC03-/C032- is too high in ionic strength to allow for a good separation of arsenite and DMA since these species elute into the dead volume within the first 2 min.Thus, a mobile phase lower in ionic strength can be used (e.g., 5 mmol 1-l sodium tetraborate) for the separation of arsenite and DMA. However, the other arsenic compounds remain on the column. Only gradient or step elutions can separate all of the arsenic species, as is shown elsewhere. Fig.3 shows a comparison of monoelement speciation of arsenic in a standard solution with that in urine. The urine from 15 test persons between the ages of 23 and 39, who worked in a laboratory, were sampled. All of the samples contained MMA and DMA/arsenite as the main species; no arsenate was detected. Two samples show an unknown peak of an arsenic species, which is plotted in Fig.3 (at mlz = 75). No interference from chloride is detected on mass 77. This indicates that no interference in taking place at mass 75, which is the mass of arsenic, and no further dilution of the sample is necessary. However, the high solids content in urine results in a high memory effect, and 2 min of washing between samples is I Urine I .- I Standard 0 200 300 40 Retention tirne/s Fig.3 Comparison of the standard solution of arsenite (10 ng ml-l), arsenate (0.18 pg ml-I), MMA (0.1 pg ml-l) and DMA (30 ng ml-l), which is detected on mass 75, with chromatograms of one urine sample, which detected arsenic on mass 75, and also the chloride interference on 51. recommended. The chloride concentration of the urine of 15 test persons varied between 0.1 and 0.5%.For reliable arsenite and DMA determination a dilution of 1 : 1 of urine samples is necessary to exclude the interference which can be monitored simultaneously on mass 77 for 40Ar37Cl. However, the concentrations of arsenite and arsenate in undiluted samples are determined without any interference calculation. Arsenate was found in concentrations under the detection limits in the urine samples from the test persons.The concentrations of total arsenic (DMA, MMA and arsenite) vary between the detection limits and 15 ng ml-1. Conclusions Preliminary results demonstrate the analytical potential of the simultaneous use of two different detectors. Ion chromato- graphy is used as a speciation technique as well as a matrix separation method, which can be used as the introduction system for ICP-MS.Only by use of a dilution step without any clean-up procedure can urine be measured directly. Further investigations for the quantification of environmental samples with reference standards are necessary. The study of equilib- rium changes involving unstable anions, especially for heavy elements, during ion chromatography, is recommended and planned for future work. The author thanks the Alexander-von-Humboldt Foundation for financial support, and Alfred Vitalis Hirner for the opportunity to carry out these experiments. References 1 2 3 4 5 6 7 Van Elteren, J. T., Gruter, G. J. M., Das, H. A., and Brinkman, U. A. Th., Int. J. Environ. Anal. Chern., 1990, 43,41. Andreae, M. O., Anal. Chern., 1977, 49, 820. Braman, R. S., and Foreback, C. C., Science (Washington, D.C.), 1973,182, 1247. Branch, S., Ebdon, L., Ford, M., Foulkes, M., and O’Neill, P.,J. Anal. At. Spectrorn., 1991, 6, 151. Ford, M., Ebdon, L., and Hill, S. J., Anal. Proc., 1992, 29, 104. Han, H.-B., Liu, Y.-B., Mou, S.-F., and Ni, Z.-M., J. Anal. At. Spectrorn., 1993, 8, 1085. Demesmay, C., Olle, M., and Porthault, M., Fresenius J . Anal. Chern., 1994,348,205. Paper 51068966 Received October 18, I995 Accepted November 13, I995

ISSN:1359-7337    

DOI:10.1039/AC9963300011

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, January 1996, VoE33 (1 5-1 7) 15 Comparison of Supercritical, Subcritical, Hot, Pressurized and Cold Solvent Extraction of Four Drugs From Rodent Food John R. Williams", E. David Morgan"* and Brian Lawb a Department of Chemistry, Keele University, Keele, StafSordshire, UK ST5 5BG Macclesfeld, Cheshire, UK SKI 0 4TG Toxicology Resources Group, Zeneca Pharmaceuticals, Alderley Park, Propranolol, tamoxifen and two experimental drugs (all basic but of differing pK, and polarity) were extracted from rodent food by supercritical carbon dioxide-methanol more efficiently than by a more polar, subcritical mixture of carbon dioxide and methanol.Hot, pressurized methanol alone was as efficient as the supercritical fluid. The efficiency of these methods was similar to conventional solvent extraction with acidified methanol, but conventional solvent extraction gave the best precision, as measured by the relative standard deviation.In recent times, interest has grown in alternatives to conven- tional solvent extraction for the isolation and concentration of organic substances from a variety of matrices. The drive to find alternatives to solvent extraction are several-fold and include the wish to; avoid using and disposing of large volumes of flammable or chlorinated solvents; reduce manipulation (which is costly in manpower and a potential source of errors); and achieve shorter extraction times.Supercritical fluid extraction (SFE)1-5 has come to be the most prominent alternative to solvent extraction, but other methods such as microwave- assisted extractionCl0 and accelerated solvent extraction' are being offered as alternatives.To date, only one direct com- parison has been made between these modem approaches to extraction in which Soxhlet, microwave and SFE were compared for the recovery of polycyclic aromatic hydrocarbons from highly contaminated soils.I2 The highest total amount of polycyclic aromatic hydrocarbons was extracted using SFE, followed by microwave-assisted extraction, and then Soxhlet extraction; but the conditions and solvent were not optimized for the Soxhlet method.Methods of the extraction of pharmaceutical compounds from medicated rodent diet is of interest. Polar drugs can be difficult to extract efficiently using supercritical carbon dioxide, and in normal circumstances, it is necessary to add a polar solvent (e.g., methanol, ether, acetonitrile) to the carbon dioxide to increase its polarity and solvent properties. It is, nevertheless, logical to ask if this is the most efficient extraction medium.What if the fluid is made more polar by adding a greater proportion of the conventional solvent, and the mixture drops below its critical point? Will that be more efficient for the purpose, and how would these compare with simply using a liquid such as methanol either at room temperature and pressure or at high temperature and pressure? Experiments that attempt to answer these questions are described for a group of four pharmaceutical compounds (Fig. 1) that have been incorporated into rodent diet.* To whom correspondence should be addressed. Experimental Apparatus and Reagents SFE, subcritical fluid extraction (SubFE) and hot, pressurized solvent extraction were carried out with a manual, single- sample extractor. The extractor consisted of a Philips Analytical (Cambridge, UK) LC-XP3 pump to deliver liquid carbon dioxide, a Waters (Milford, MA, USA) M-45 pump to contribute methanol, a Philips Analytical Series 104 Chromato- graph oven to regulate temperature and a stainless-steel extraction cell (3 cm X 4.6 mm id) made from high- performance liquid chromatography (HPLC) column parts to contain the sample.Pressure in the system was maintained with a silica capillary (25 cm X 50 pm id; Composite Metal Services, Worcester, UK). The HPLC consisted of a Philips Analytical LC-XP3 pump to deliver the mobile phase, a Rheodyne (Cotati, CA, USA) 7125 injection valve, a Philips Analytical LC/UV detector and a Shimadzu (Kyoto, Japan) C-R3A integrator.Propranolol (as the hydrochloride), tamoxifen (as the base), ZM 95527, ZM 169369 (two candidate drugs without generic names) and the rodent food were obtained from Zeneca Pharmaceuticals (Macclesfield, UK) and used without further purification. The carbon dioxide was industrial grade (98.5%) from BOC (Guildford, UK).The methanol was HPLC grade from Prolabo (Paris, France) and trifluoroacetic acid (99%) was from Avocado (Lancaster, UK). Ammonium formate (analyt- ical-reagent grade) was purchased from Fisons (Loughborough, UK) . General Procedure Sample Preparation The drugs and rodent food were blended together at concen- trations of 5 mg g-1 with a Turbula mixer (Stanmore, UK) for I A I Ropranolol " - O q Tamoxifen ZM 169369 Fig.1 Structures of the four basic compounds used in this study.16 Analytical Communications, January 1996, Vol33 10 min. The resultant mixture was further ground with a pestle and mortar to aid sample uniformity.Supercritical Fluid Extraction Carbon dioxide-methanol (85 + 15 v/v or 83 + 17 mol%) was pumped (2 cm3 min-l) through the extractor at 70 "C and 17.25 MPa. The sample size was 0.3 g. The extracted drugs were depressurized into a glass tube (15 cm X 2.5 cm) containing 5.0 cm3 of methanol. When extraction was complete, the collector tube was emptied and rinsed with methanol and the total volume of liquid adjusted to 25.0 cm3 (with methanol) ready for analysis.Subcritical Fluid Extraction The same procedure and conditions were used as for SFE, except that the extracting fluid was carbon dioxide-methanol (15 + 85 v/v) and flow rates of 1 cm3 min-l and 2 cm3 min-1 were used. Solvent Extraction Hot, pressurized solvent extraction A similar procedure to that used for SFE was followed, except that methanol was used at 1 cm3 min-' at 55 "C and 100 "C, and a flow rate of 2 cm3 min- was used at 70 "C.No methanol was necessary in the glass collector tube at the start of the extraction. Cold solvent extraction For cold (unpressurized) solvent extraction two methods were used, method A was a similar procedure to SFE but the temperature was at 20 "C and no silica capillary or collection solvent was used.For the second, method B, the drugs-food sample (0.3 g) was placed in a sample tube [poly(propylene) 15 cm X 1.5 cm] and the extraction solvent added [5.0 cm3 of methanol or methanol-aqueous 1 mol dm-3 trifluoroacetic acid (96.2 + 3.8 v/v)]. Sample and solvent were then thoroughly mixed by vortexing for 1.5 min, followed by the separation of the solvent from insoluble matter by centrifugation (5 min at 1500 rpm) before analysis.HPLC Quantitation was by isocratic HPLC using a Spherisorb 5 SCX strong cation exchange column (10 cm X 4.6 mm id; Phase Separations, Deeside, UK) and a mobile phase of 0.02 mol dm-3 ammonium formate in methanol-water (80 + 20 v/v) at pH 2.45 (adjusted by addition of trifluoroacetic acid). The flow rate was 1 cm3 min-l and the ultraviolet detector was set at 270 nm.Results and Discussion Propranolol, tamoxifen, ZM 95527 and ZM 169369 represent a range of basicity (as measured by pK,) and polarity (as measured by log P) and these values are given in Table 1. Tamoxifen is the least polar drug. ZM 95527 is the most polar drug and is also the least basic of the drugs.Propranolol is the most basic of the four drugs. There appeared to be a correlation between recovery of the drugs from rodent food and drug basicity when methanol (with and without aqueous tri- fluoroacetic acid) at 20°C was used in three methods as the extraction medium. Within this group, recovery appeared to increase with increasing basicity.However, the correlation did not extend to other similar basic drugs (unpublished results). For the other extraction methods, there was no correlation between either recovery and basicity or recovery and polarity (see Table 1). Some of the extractions were carried out for 3 h. The plots of percentage recovered versus time for propranolol show an initial rapid rise in recovery, followed by a slowing down in the extraction rate leading to a plateau (Fig.2). Similar curves were obtained for the other three drugs. From the results for recovery after 10 min (Table l), a good impression of the relative efficiency of recovery after 3 h is achieved. Recoveries by the different extraction methods were com- pared using the paired t-test13 with a confidence interval of 95%.Hot, pressurized methanol at 2 cm3 min-l and 70 "C was more efficient than SFE (with methanol-modified carbon dioxide) for the recovery of ZM 95527, but not significantly different for the recovery of propranolol and ZM 169369. It was less efficient than SFE for the recovery of tamoxifen. Although one might have expected a subcritical mixture of methanol and carbon dioxide at 2 cm3 min-I and 70 "C to be intermediate between SFE and hot, pressurized methanol, it was, in fact, slightly less efficient.The subcritical method was also relatively insensitive to flow rate; recoveries were similar at 1 and 2 cm3 min-1. Precision, as indicated by the relative standard deviation (Table l), was extremely good for recovery by a routine methanol-aqueous acid procedure (not optimized for this group of compounds).Furthermore, precision was also good for the Table 1 Comparison of SFE and SubFE with the use of hot, pressurized and cold solvent extraction of propranolol, tamoxifen, ZM 95527 and ZM 169369 from rodent food. The results obtained after extraction for 10 rnin are means of six determinations and s, values (%) are shown in brackets. Methanol was pumped through the sample in method A and vortex-mixed in method B.Recovery (%) Extraction fluid Supercritical Subcritical Subcritical Hot, press. MeOH Hot, press. MeOH Hot, press. MeOH Method A MeOH Method B MeOH Method B aq. TFA C02-MeOH (85 + 15) C02-MeOH (1 5 + 85) C02-MeOH (15 + 85) Flow1 cm3 min-1 2 2 1 2 1 1 2 - - Propranolol Temperature1 pK,, 9.42; "C log P , 3.56 70 88.4 (12.9) 70 78.2 (15.2) 70 73.2 (25.5) 70 91.0 (8.1) 55 38.1 (35.8) 100 43.2 (37.0) 20 92.1 (1.8) 20 88.8 (3.9) 20 95.7 (1.4) Tamoxifen pK,, 8.57; log P , 6.63 86.2 (8.1) 70.8 (15.5) 67.1 (24.2) 82.3 (7.8) 35.9 (34.8) 40.4 (39.7) 80.3 (0.7) 77.7 (3.0) 84.5 (0.8) ZM 95527 pK,, 7.93; log P , 1.07 84.5 (12.1) 78.6 (16.7) 69.4 (26.8) 92.1 (9.0) 37.9 (30.1) 36.0 (30.5) 59.6 (3.8) 54.6 (6.9) 72.2 (1.0) ZM 169369 pK,, 8.23; log P , 4.50 81.2 (1 2.2) 68.1 (14.5) 65.8 (24.5) 78.9 (7.6) 37.7 (32.8) 42.1 (41.3) 74.4 ( 1.4) 69.6 (3.8) 76.5 (2.7)Analytical Communications, January 1996, Vol33 17 other two methanol-based solvent extraction methods con- ducted at 20°C.The precision was poor for supercritical and subcritical conditions and hot, pressurized solvent extraction. However, for these latter methods, the precision seemed higher the more efficient the recovery.Hot, pressurized solvent extraction at a low flow rate gave poor extraction and very poor precision. As expected, recovery of all four drugs from rodent food was significantly better when following method A than method B (methanol) and precision was also better (see Table 1).Aqueous acidified methanol produced significantly better recovery of all four drugs than methanol alone. Method A is directly comparable (same extractor, flow rate and extraction time) with the methods of SFE, SubFE and hot, pressurized methanol. Recovery of the drugs by SFE was significantly better than by cold, unpressurized methanol (except for propranolol where the reverse was true) but precision with SFE was poorer (Table 1).In contrast, recovery of the drugs by cold, unpressurized methanol was significantly better than by SubFE (except for recovery of ZM 95527 where the reverse was true) but again precision was superior for the cold, unpressurized solvent. Only the recovery of ZM 95527 and ZM 169369 were significantly better for hot, pressurized methanol than cold, unpressurized methanol; there was no significant difference between the two methods for recovery of propranolol and tamoxifen.In general, precision with cold, unpressurized methanol was better than with hot, pressurized solvent. Conclusions Recovering basic drugs from cereal-based food is not as simple as recovering a compound from an inert matrix.The variable '"1 I 0 100 200 Ti rn e/m i n Fig. 2 Plot of propranolol recovery versus time for several extraction fluids. 1, MeOH, 55 OC, 1 ml min-1; 2, MeOH, 100 "C, 1 ml min-I; 3, C02-MeOH (15 + 85), 1 mi min-1; 4, C02-MeOH (15 + 85), 2 ml min-1; 5, C02-SMeOH (85 + 15), 2 ml min-1; 6, MeOH, 70 OC, 2 ml min-1; and 7, MeOH, 20 OC, 2 ml min-1. results suggest there is some intereaction between the com- pounds and the food.Conditions may be quite different in the extraction of less polar compounds, e.g., extraction of poly- cyclic aromatic hydrocarbons 14-16 or polychlorinated bi- phenyls16.17 from soil or sediment. Newer methods of extraction deserve careful testing and comparison. However, in this case and under these conditions, they appear to offer no clear advantage.For two of these compounds, the supercritical fluid method offered the highest extraction and it was second highest for another. Hot, pressurized methanol may appear hazardous, but the solvent was cold when it emerged from the silica capillary after depressurizing. J.R.W. thanks The Royal Society of Chemistry for the award of a SAC industrial studentship. References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Analytical Supercritical Fluid Chromatography and Extraction, ed.Lee, M. L., and Markides, K. E., Chromatography Conferences, Provo, Utah, 1990. Supercritical Fluid Extraction and Its Use in Chromatographic Sample Preparation, ed. Westwood, S. A., Blackie, London, 1992. Analysis With Supercritical Fluids: Extraction and Chromatography, ed.Wenclawiak, B., Springer-Verlag, London, 1992. Hyphenated Techniques in Supercritical Fluid Chromatography and Extraction, ed. Jinno, K., Elsevier, Amsterdam, 1992. Luque de Castro, M. D., Valcircel, M., and Tena, M. T., Analytical Supercritical Fluid Extraction, Springer-Verlag, London, 1994. Ganzler, K., Salg6, A., and Valko, K., J . Chromatogr., 1986, 371, 299. Neilson, R. C., J . Liq. Chromatogr., 1991, 14, 503. Onuska, F. I., and Terry, K. A., Chromatographia, 1993,36, 191. Lopez-Avila, V., Young, R., and Beckert, W. F., Anal. Chenz., 1994, 66, 1097. Barnabas, I. J., Dean, J. R., FowIis, I. A., and Owen, S. P., Analyst, 1995, 120, 1897. Richter, B. E., Ezzell, J. L., Felix, D., Roberts, K. A., and Later, D. W., Int. Lab., 1995, May, 18. Dean, J. R., Barnabas, I. J., and Fowlis, I. A., Anal. Proc., 1995,32, 305. Miller, J. C., and Miller, J. N., Statistics for Analytical Chemistry, Ellis Horwood, Chichester, 1989, ch. 3, pp. 58-59. Lee, H.-B., Peart, T. E., Hong-You, R. L., and Gere, D. R., J . Chromatogr. A , 1993, 653, 83. Reimer, G., and Suarez, A., J . Chromatogr. A , 1995, 699, 253. Langenfeld, J. J., Hawthorne, S. B., Miller, D. J., and Pawliszyn, J., Anal. Chem., 1993, 65, 338. Bowadt, S., Johannsson, B.. Wunderli, S., Zennegg, M., de Alencastro, L. F., and Grandjean, D., Anal. Chem., 1995, 67, 2424. Paper 5l05804J Received September I , I995 Accepted November 10,1995

ISSN:1359-7337    

DOI:10.1039/AC9963300015

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, January 1996, Vol33 ( I 9-20) 19 Determination of Poly(ethy1ene Glycol)-400 in Urine by Fourier Transform-Infrared Spectrometry Lale Ersoy, Sedef Atmaca, Serap Saklik and Sedat Imre Faculty of Pharmacy, Department of Analytical Chemistry, University of Istanbul, 34452 Beyazit, Istanbul, Turkey A simple and rapid method for the analysis of poly(ethy1ene glycol)-400 (PEG-400) (used in the study of intestinal physiology) in human urine is described using Fourier transform-infrared spectrometry.The quantitative measurements were carried out by using the absorbance value of the C-0-C band (wavenumber 1100 cm-1). The lower limit of determination of PEG-400 in human urine was 0.8 mg cm-3. Analytical recovery of PEG-400 added to urine controls was 99.6%. The results of the analyses of one urine sample were compared with those obtained by using high-performance liquid chromatography.Limitations and advantages of the two systems are discussed. Poly(ethy1ene glyco1)s (PEGs) have been widely used diagnos- tically to determine intestinal permeability1 as they are not metabolized by colonic bacteria and their toxicity is low.* PEGs and their derivatives are also extensively used in various products including drugs, foods, cosmetics and detergents.Several gas chromatographic (GC) and high-performance liquid chromatographic (HPLC) methods have been developed to determine PEGs in biological fluids, especially in urine. Liquid chromatographic analyses of PEG were performed without derivatization by using a refractive index detector34 or after derivatization by using an ultraviolet (UV) detector.'^^ Gas chromatographic methods also require derivatization of PEG prior to anal~sis.9~10 Kumar has developed a Fourier transform- infrared spectrometric (FT-IR) method for the determination of PEG-400 in high density polyethylene resin. The sample was directly melt-pressed into very thin plaques and the integrated absorbance (1000-1 170 cm-1) per unit thickness was measured.1 1 In this paper an FT-IR method for estimating urinary concentration of PEG-400 is described. The results of one urine sample analysed by both FT-IR and HPLC were compared statistically. Experimental Materials and Reagents PEG-400 was purchased from Sigma (St. Louis, MO, USA). Other chemicals and solvents were obtained from Merck (Darmstadt, Germany).All reagents used were of analytical- reagent grade. Apparatus For IR measurements a Perkin-Elmer 1600 FT-IR spectrometer with a lithium tantalate detector was used. The spectra were recorded by using fixed pathlength (0.1 mm) NaCl liquid sampling cells. For each determination, 64 scans were added at a 4 cm-1 resolution.For HPLC, a model 6000A solvent delivery system, a U6K universal injector and a Model 440 absorbance detector (Waters, Milford, MA, USA) at 280 nm was used. The detector was connected to a strip-chart recorder (Linear 355). The column used was a 10 pm Bondpak C18 300 mm X 3.9 mm (Waters). The mobile phase was methanol-water (95 + 5 ) at a flow rate of 0.8 cm3 min-I. Stock and Standard Solutions PEG-400 (500 mg) was accurately weighed into a 10 cm3 calibrated flask and made up to the mark with trichloromethane. A series of standard solutions containing 1-7 mg cm-3 of PEG- 400 were prepared by pipetting 0.2-1.4 cm3 of stock solution into a 10 cm3 calibrated flask and adding trichloromethane.The IR absorbance spectra of these solutions were obtained to establish a calibration graph.For recovery studies, urine samples containing known amounts of PEG-400 were also prepared. For this purpose, appropriate volumes of standard solutions were transferred into test-tubes. After the evaporation of the solvent, 2 cm3 of drug-free urine was added. To obtain a calibration graph for the HPLC method, 1.2-6 cm3 of PEG-400 solution (1 mg cm-3) were evaporated and derivatized as described.Extraction PEG-400 was extracted from urine samples using a slightly modified previously reported method.12 After 2 g of ammonium sulfate was added into 2 cm3 of urine sample and mixed for 20 s, PEG-400 was extracted with 5 cm3 of dichloromethane for 2 min on a vortex mixer. The mixture was then centrifuged at 3000g for 3 min. After the aqueous phase was discarded and the organic phase was dried over anhydrous Na2S04, 3 cm3 of dichloromethane phase was transferred into another tube. The solvent was evaporated under nitrogen at 50 "C.The residue was then either dissolved in 1 cm3 trichloromethane for FT-IR analysis or derivatized with benzoyl chloride for HPLC analysis. FT-IR Trichloromethane and solutions containing PEG-400 (standard solutions or extraction residue in trichloromethane) were injected into the NaCl cell, respectively.After the transmittance spectrum was recorded and converted into absorbance units, the chloroform spectrum was automatically subtracted from that of the PEG-400 solution. The absorbance value of the analytical band at 1100 cm-1 was recorded for quantitative measure- ments.HPLC For comparison, Kinahan and Smith's HPLC method7 was used. PEG-400 in urine samples was analysed under the20 Analytical Communications, January 1996, Vol33 modified chromatographic conditions after esterification. PEG- 400 extracted from urine was esterified with benzoyl chloride in pyridine as previously described? After extraction with di- chloromethane, the evaporation residue of the organic phase dried over anhydrous Na2S04 was dissolved in 1 cm3 of mobile phase and 10 mm3 of this solution was injected into the column. The peak heights at 4.5 min were measured to determine the PEG-400 concentration in the samples.Application of the Method Following oral administration of 5.6 g of PEG-400 in 200 cm3 of water, a urine sample was collected after 6 h.The specimen was mixed thoroughly and its total volume measured. About a 50 cm3 aliquot of this was stored at -20 "C until analysis. Results and Discussion Urine samples were analysed by FT-IR spectrometry after extraction to assay PEG-400 which was consumed to determine g 0.0000 _I 1200 1100 1000 1200 1100 1000 Wavenumberkm-' Fig. 1 (A) FT-IR original absorbance spectra and (B) difference absorbance spectra of PEG-400 standard solutions (a, 1.0; b, 2.0; c, 3.0; d, 4.0; e, 5.0; f, 6.0 and g, 7.0 mg ~ m - ~ ) Table 1 Analysis of PEG-400 in urine (n = 3) Amount Amount found Mean added/ mean/ recovery Sample no.mg ~ m - ~ mg cm-3 sr (%) (%I 1 1.503 1.502 2.92 99.9 2 3.007 2.997 2.16 99.7 3 4.5 10 4.47 1 1.89 99.1 Table 2 Determination of PEG-400 in urine Statistical value FT-IR HPLC X 2.92 2.96 S 0.08 1 0.077 s r (%I 2.77 2.60 n 5 5 t-test of significance F-test of significance t = 0.80 F = 1.11 t = 2.31 (p = 0.05) F = 6.39 (p = 0.05) the intestinal permeability.Since PEG-400 shows an intense C- 0-C ether linkage band at 1100 cm-1,11 the absorbance value of this band was used for quantitative measurements. The absorbance spectra of PEG-400 solutions are shown in Fig.1(A). To measure PEG-400 at 1100 cm-1 is impossible owing to the strong absorption of trichloromethane at this wav- enumber. In order to resolve the relatively weaker C-0-C absorption band from the much stronger trichloromethane band, the spectral subtraction approach was used. The trichloro- methane subtracted spectra of the standard solution are shown in Fig.l(B). The relationship between absorbance units of the standard solutions at 1100 cm-1 and concentrations was linear for the range 1-7 mg cm-3. The regression equation was A = 0.0214~ -0.0021 (r = 0.9993). The limit of determination was 0.8 mg for 1 cm3 of urine sample. The recovery of PEG-400 from urine was determined by analysing the added urine samples at three different concentrations.The average recovery was 99.6% (Table 1) i.e., it can be extracted quantitatively from urine and there is no interference in urine. When using HPLC, the concentration range of 0.4-2.0 mg cm-3 was studied. In this range the relationship between concentration of PEG-400 and peak height was linear, A = 1.065~ - 0.04 (r = 0.9995). The determination limit of this method was 0.02 mg of PEG-400 for 1 cm3 of urine sample.Ten urine samples collected after 6 h were analysed by both methods. Average percentage urinary excretions of the administered dose of 5.6 g were 26.2 and 24.9% using FT-IR and HPLC, respectively. These results were found to be related ( r = 0.9965). Furthermore, one unknown urine sample was analysed by both FT-IR and HPLC to compare the results statistically.There is no significant difference between mean values and standard deviations (Table 2). Although the limit of determi- nation by FT-IR is higher than that of HPLC, it is acceptable for medicinal research studies. However, the determination of PEG-400 by FT-IR without any derivatization after extraction is faster and simpler.It is also possible to analyse urine samples directly without extraction. This would be particularly useful in routine analysis. References 1 2 3 4 5 6 7 8 9 10 11 12 Chadwick, V. S., Phillips, S. F., and Hofmann, A. F. , Gastro- enterology, 1977, 73, 241. F.D.A. Drug Bull., 1982, 12, 25. Young, G. 0. , Ruttenberg, D., and Wright, J. P., Clin. Chem. (Winston-Salem, N.C.), 1990, 36, 1800. Trathnigg, B, Thamer, D., Yan, X., and Kinugasa, S., J. Liq. Chromatogr., 1993, 16, 2439. Ryan, C. M., Yarmush, M. L., and Tompkins, R. G., J. Pharm. Sci., 1992, 81, 350. Delahunty, T., and Hollander, D., Clin. Chem. (Winston-Salem, N.C.), 1986, 32, 351. Kinahan, I. M., and Smyth, M. R., J. Chromatogr., 1991, 565, 297. Murphy, R., Selden, A. C., Fisher, M., Fagan, E. A., and Chadwick, V. S. J. Chromatogr., 1981, 211, 160. Bouska, J. B., and Phillips, S. F., J. Chromatogr., 1980, 183, 72. Sivakumaran, T., Jenkins, R. T., Walker, W. H. C., and Goodacre, R. L., Clin. Chem. (Winston-Salem, N.C.), 1982, 28, 2452. Kumar, T., Analyst, 1990, 115, 1597. Schwertner, H. A., Patterson, W. R.,Cissik, J. H., and Wilson, K. W., J. Chromatogr., 1992, 578, 297. Paper 510621 40 Received September 20, I995 Accepted November 17, I995

ISSN:1359-7337    

DOI:10.1039/AC9963300019

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, January 1996, Vol33 (21-22) 21 Rapid Fluorescence Enzyme Linked lmmunosorbent Assay for Subtilisin Ioana Nitescuayb Frederick J. Rowell" and Robert H. Cummingb a,h North East Biotechnology Centre, School of Health Sciences, University of Sunderland, Sunderland, UK SR2 7EE University of Teesside, Middlesbrough, Cleveland, UK TSI 3BA North East Biotechnology Centre, School of Science and Technology, A simple and rapid ELISA with a fluorescent end-point has been developed for the measurement of subtilisin.The ELISA can be applied to the analysis of enzyme in solutions that have been eluted from filters following the capture of air samples in the workplace. The assay has adequate precision and shows no interference from other enzymes tested.A comparison has been made between this assay and a similar ELISA method using spectrophotometric detection. The use of fluorescence detection reduces the over-all assay time from 60 to 25 min and reduces the concentrations of antibody reagents required in the assay. The assay described is quicker and more sensitive than the current methods in use. During the production of many biologically active products, there is a chance that plant operators will be exposed to the product in the form of an aerosol or dust particles.Such exposure can lead to plant personnel developing allergies to the product. Subtilisin is an example of a typical serine protease enzyme used in the detergent industry. In common with other proteases it is a respiratory sensitizer and long- and short-term occupational exposure limits of 60 ng 117-3 for periods of 8 h and 15 min, respectively, have been set for workers who may inhale these proteases in the work place.2 At present, factory air is passed through a filter in a Galley sampler operating at 650 1 min-1 and the dust collected on the filter is eluted and assayed for the protease.There are two methods currently employed to monitor protease in the collected dust.Both of them are based on detection of the enzyme by measuring its catalytic activity .3,4 Measurements performed with a Casella Model A personal monitor (SKC, Blandford Forum, Dorset, UK), operating at 2 1 min-1, gave higher results than for the air samples monitored with Galley samplers, and owing to its lower sampling rate, half the samples collected were not detected by the enzymic method used.4 The use of the Casella device as a personal sampler offers the advantage of giving a better indication of individual worker exposure and hence development of more sensitive analytical methods are required to enable its use in this context.An extremely sensitive assay based on enzyme activity, which uses p-nitroanilide substrates has been proposed by the detergent industries, but the assay is not rapid and requires expensive synthetic substrate^.^ Ideally the assay should also be simple and sufficiently rapid to enable analysis of material eluted from filters to be .performed immediately at the site of sampling rather than the current post- sampling analysis which takes place at remote laboratories.We report on the use of fluorescence detection in a subtilisin- specific ELISA and its impact on the assay's performance compared with the use of a spectrophotometric end-point in this assay. Experimental Apparatus The assays were performed using black Dynatech (Billinghurst, West Sussex, UK) Immulon 4 microtitre plates. Absorbance measurements were recorded on a Dynatech MR 7000 microtitre- plate reader at 405 nm.Fluorescence measurements were recorded on a Dynatech Fluorolite 1000 fluorescence microtitre- plate reader at A,, = 340-360 nm and A, = 440-460 nm. Reagents All reagents were purchased from Sigma (Poole, Dorset, UK) unless otherwise stated. The buffer salts were of analytical- reagent grade (Merck, Poole, Dorset, UK). Polyclonal anti- bodies were generated in rabbits by using Subtilisin Carlsberg type 8 : bacterial.6 The antisera from rabbits A and B were mixed (1 + 1 v/v), affinity-purified and used for assay development .6 Coating buffer, pH 9.6.This was prepared from 2.93 g of sodium hydrogen carbonate, 2.5 g of sodium carbonate and 0.2 g of sodium azide, made up to 1 1 with distilled water.Substrate buffer (PNPP), pH 9.8. This was prepared from 105 g of diethanolamine, 0.1 g of magnesium chloride made up to 1 1 with distilled water. Phosphate buffered saline with 0.05% Tween 20 (PBST), pH 7.4 (assay buffer). This was prepared from 36 g of sodium chloride, 2.15 g of potassium dihydrogen phosphate, 7.4 g of disodium hydrogen phosphate, 2.5 ml of Tween 20, and 0.5 g of sodium azide, made up to 5 1 with distilled water.TRZS buffer, pH 7.0. This was prepared from 0.05 mol 1-' TRIS-HC1, 0.01 mol -1 calcium chloride, made up to 1 1 with distilled water. Ethanolamine bufSer, pH 8.6. This was prepared from 12.2 g of ethanolamine, made up to 1 1 with distilled water. Optimization of the Fluorescence Assay Wells of the black Immulon 4 microtitre plates were coated by passive adsorption with solutions of subtilisin in coating buffer (pH 9.6) over the range 1-10 pg ml-1.After standing at 4 "C overnight, the plates were washed three times with distilled water, dried at 35 "C and immediately sealed and stored, at room temperature, in the dark until used. The optimum concentration was 10 pg ml-1. The optimum dilution of affinity-purified rabbit antiserum and goat anti-rabbit IgG labelled with alkaline phosphatase was determined by in- cubating dilutions of 1 : 100-1 : 1000 of the former and 1 : 100-1 : 2500 of the latter.Respective dilutions of 1 : 1000 and 1 : 2500 produced assays of the maximum sensitivity. ELISA for Subtilisin Using Fluorescence Detection Equal volumes of 1 : 1000 affinity-purified antiserum and 1 : 2500 goat anti-rabbit IgG labelled with alkaline phosphatase, both diluted in PBST were incubated for at least 1 h at room temperature prior to the assay.Aliquots of sample in PBST (50 1-11), or freshly prepared subtilisin standards ranging from 0.1 pg ml-l to 100 pg ml-l (prepared in 50 pl of PBST) were added to the wells of the coated plates, followed by the pre-incubated antibodies (100 pl).After incubation for 15 min, the wells were washed three times with distilled water. Finally, aliquots of 4-methylumbelliferyl phosphate (50 1-11 of 0.2 g 1-1 solution in substrate buffer, pH 9.8) were added to each well and the fluorescence was measured after 10 min. The standard curve for22 Analytical Communications, January 1996, Vol 33 fluorescence detection was steepest at 10 ng ml-1-5 pg ml-1 of subtilisin corresponding to a drop of 1500 fluoresence units.ELISA for Subtilisin Using Spectrophotometic Detection Equal volumes of 1 : 100 affinity-purified rabbit antiserum and 1 : 250 goat anti-rabbit IgG labelled with alkaline phosphatase, both diluted in PBST were pre-incubated for at least 1 h at room temperature prior to the assay.Aliquots of sample in PBST (50 pl) of freshly prepared standards of subtilisin ranging from 0.1 pg ml-l to 100 pg ml-1 (prepared in 50 p1 of PBST) were added to the wells of the coated plates followed by the pre-incubated antibodies (100 pl). After incubation for 30 min, wells were washed three times with excess amounts of distilled water. Finally aliquots of 4-nitrophenyl phosphate (50 pl of a 2 g 1-1 solution in substrate buffer) were added to each well and the absorbance measured at 405 nm, when a pronounced yellow colour developed at room temperature.The resulting standard curve for a development time of 30 min is shown in Fig. 1, which also shows the corresponding standard curve for the fluorescence assay. The steepest part of the former standard curve was between 10 ng ml-1 and 5 pg ml-1 of subtilisin, corresponding to a fall of 0.5 absorbance units.Precision and Sensitivity Within-assay precision (s,) for both assays was determined by performing six assays with a subtilisin sample of 10, 100 and 1000 ng ml-1 on one plate. Between-assay s, was assessed using the 10 ng ml-1 sample in six separate assays. Both ELISAs were performed with subtilisin standards and six sample blanks (assay buffer).The mean and standard deviation (s) of the blanks were used to determine the LOD of subtilisin from the standard curve at the 95% confidence level (i.e., the concentration of subtilisin corresponding to zero fluorescence or absorbance minus 2s of the zero). Specificity Stock solution of active subtilisin in PBST buffer (100 pg ml- I ) was inactivated by overnight incubation at 80 "C and then used in the rapid fluorescence and spectrophotometric assays as described above.The assay was also performed in the presence of esperase, collagenase and a-amylase, over the concentration range 10.0 ng ml-1-100 pg ml-1. The assay was also run with subtilisin standards in an assay buffer containing 1% non- enzyme detergent.Results and Discussion Precision and Sensitivity For the fluorescence assay, the within-assay s, was 6.67,7.92 and 5.95% for 10,100 and 1000 ng ml- 1 subtilisin, respectively. The between-assay s, was 6.43% for 10 ng ml-1 subtilisin. For the spectrophotometric assay, the within-assay s, was 3.35, 1.8 and 7.42% for 10, 100 and 1000 ng ml-1 subtilisin, respectively.The between-assay s, was 13% for 10 ng ml- *. Spe ciji c ity No significant interference was detected with inactivated subtilisin in either the fluorescent or the spectrophotometric assays. A possible explanation is that proteases are subject to autodegradation and molecules which are partially unfolded due to heat treatment are vulnerable to autolysis.7 No significant cross-reaction was observed for esperase, collagenase and amylase in either assay.The presence of 0.1% non-enzyme detergent in the assay buffer did not significantly change the shape of the subtilisin standard curve for either assay. Conclusions Although both assays displayed the same LOD, a high background signal is observed in the fluorescence assay due to The LOD of subtilisin was 1 pg ml-I for both assays.100 I 90 80 70 6 60 2 50 .p 40 (I) 30 20 10 0 h - 0 0.001 0.1 10 1000 100000 Su bti I is in con cent rat ionhg ml-' Fig. 1 trophotometric (.) detection. Comparison of rapid ELISA using fluorescence (0) and spec- the instability of the substrate (4-methylumbellifery1 phos- phate). Thus, the percentage displacement from the blank signal is reduced in comparison with spectrophotometric detection.The sensitivity of the fluorescence assay at 50% signal displacement was better than the sensitivity of the spectrophoto- metric assay (see Fig. 1). The over-all time for the assay is improved by using fluorescence detection instead of spec- trophotometric detection from 60 to 25 min. The current chromogenic assays,s,' which take 20 min per sample are therefore considerably slower than either of these ELISA methods, especially since use of 96-well microtitre plates enables batch processing of up to 40 samples in duplicate per plate.The amount of affinity-purified antibodies required to prepare the single reagent for the fluorescence assay is one tenth of the quantity used in the spectrophotometric assay, making the assay more cost-effective to operate.Similar results in terms of precision and specificity were obtained. Heat-inactivated sub- tilisin did not cross-react in either assay. The relative potencies of active and inactive forms of subtilisin as respiratory sensitizing agents are currently unknown. The sensitivity and precision of both ELISA methods combined with their speed commend them for use in monitoring the workplace atmosphere as part of an occupational health and safety monitoring programme.The authors thank the DTI Chemical and Biotechnological Division for financial support, and Dynatech Laboratories for the use of the Fluorolite fluorescence plate reader. References Koochaki, Z., Cumming, R. H., Rowell, F. J., and Stewart, I. W., Process Biochem., 1995, 30, 589. Health and Safety Executive, Occupational Exposure Limits, 1994, EH 40/94, HMSO, London. Dunn, E., and Brotherton, R., Analyst, 1971, 96, 159. Bruce, C. F., Dunn, E., Brotherton, R., Davies, D. R., Hall, F., and Potts, M., Ann. Occup. Hyg., 1978, 21, 1. Rothgeb, T. M., Goodlander, B. D., Garrison, P. H., and Smith, L. A., J. Am. Oil Chem. Soc., 1988,65, 806. Rowell, F. J., Cumming, R. H., and Nitescu, I., Anal. Chim. Actu, 1995, in the press. Gallagher, T., Bryan, P., and Gilliland, G., Proteins: Structure, Function, and Genetics, 1993, 16, 205. The Standing Committee on Enzymatic Washing Products, Fifth Report, 1991, The Soap and Detergent Industry Association, Cumming, R. H., Rowell, F. J., and Nitescu, I., Proceedings of the Eighth Conference of the Aerosol Society, July 1994, York. p. 1-65. Paper 5107452E Received November 14, I995 Accepted December 13, 1995

ISSN:1359-7337    

DOI:10.1039/AC9963300021

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, January 1996, Vol33 (23-25) 23 Direct Non-enzymic Amperometric Glucose Sensor Based on a Novel Glucose Selective Membrane ~ Yazid Benmakroha", Ian Christie and Pankaj Vadgama Department of Clinical Biochemistry, University of Manchester, Hope Hospital, Eccles Old Road, Salford, UK M6 8HD A selective amperometric glucose sensor is described, based on direct electrochemical oxidation without recourse to voltage or time resolution techniques. In the presence of ionic and also uncharged interferents, selectivity is provided by a novel membrane composite of sulfonated polyether-ether sulfonepolyether sulfone copolymer and PVC plasticized with isopropyl myristate.Ratios for equimolar glucose : interferent are 220 : 1 for ascorbate, urate and ethanol and 210 : 1 for acetaminophen.Linear response extends to at least 600 mmol 1-1, encompassing clinical use and agrofood applications. Many of the strategies explored for the electrochemical measurement of glucose have been enzyme based but the enzymic devices are inherently subject to constraints of enzyme kinetics: notably, there is a finite linear range. Most direct glucose sensors have been based on oxidation of the aldehyde group.Work on the fuel cell principle' and also voltammetry has been moderately successful. Marincic et aL2 found complex, potential dependent mechanisms of parallel or sequential reactions but inhibition was n0ted;3.~ in further work, pulsed voltammetric techniquess improved the selectivity. The use of membrane barriers has been discussed6 and examples of M , 500 cut off are in use; however, the effects of H+, hydrogencarbonate and amino acids are unlikely to be elimi- nated.Measurements from 2-20 mmol 1-l were achieved by the compensated net charge (cnc) and the cnc ratio technique^.^.^ A further technique9 was based on potential jumps from 1600 mV, oxidizing the Pt surface, to 100 mV to discharge the electrode capacitance and then to the working potential of 400 mV, current integration giving a measure of glucose concentration.Impedance experiments have also been undertaken to improve the performance of the potential jump method.9 Bockris et a1.1" have postulated a mechanism for alkaline oxidation of glucose whereby the carbohydrate repeatedly cleaves one carbon atom and is simultaneously oxidized to a terminal-COOH, giving for total completion C6H12O6 + 6H20 + 6C02 + 24H+ 24e- Over-all charge transfer numbers ( n ) of 2-20 have been noted for RDE sweeps of -0.2 to +0.2 V, confirming oxidation beyond gluconic acid where n = 2.Over time n decreases, indicating passivation and a change in electrode processes taking place. Recent work has highlighted the role of the electrode material in the oxidation of glucose; ruthenium dioxide-carbon paste electrodes have exhibited catalytic be- haviour.11 * Present address: Department of Biomedical Engineering and Medical Physics, North Staffordshire Hospital Centre, Thornburrows Drive, Hartshill, Stoke on Trent, UK, ST4 7QB.Membranes used to exclude interferents from amperometric sensors have commonly exploited the anionic nature of clinical interferents.Plasticized PVC excluded charged species by a differential partitioning mechanism, * whereas sulfonated poly- ether-ether sulfonated polyether-ether sulfone/polyether sul- fone (SPEES-PES) carries a charged sulfonate group. l3314 This work describes the use of SPEES-PES and plasticized PVC mixtures for the fabrication of membranes permselective to glucose.l 5 Materials and Methods All assays were conducted in phosphate-chloride buffer with sample solutions at pH 7.4 unless indicated. Glucose, urate, ascorbate and acetaminophen were reagent grade, and were made up in buffered stock solution and diluted in the electrochemical cell to the required concentration. Stock glucose solutions were left overnight prior to use.The 'ethanol' was 95% ethyl alcohol and 5% isopropyl alcohol by volume. Undiluted human blood and serum (collected in fluoride- oxalate tubes) was obtained from the Department of Clinical Biochemistry Laboratory, Hope Hospital. Cuprophan dialysis membranes were removed from a hemodialysis cartridge (Gambro, Lund, Sweden). Other mem- branes were solvent cast in a covered Petri dish.The PVC solution contained 0.06 g of PVC (M, 200 000, Merck, Poole, UK) and 0.15 ml isopropyl myristate (IPM) plasticizer dissolved in 5 ml of tetrahydrofuran. The SPEES-PES solution contained 0.4 g of SPEES-PES (a gift from ICI Chemicals and Polymers, Runcorn, UK), dissolved in 7.5 ml of dimethylforma- mide and 2.5 ml of methoxyethanol. SPEES-PES-PVC mem- branes were cast from a mixture of 3 ml SPEES-PES solution and 3 ml of PVC solution.15 The electrochemical cell and potentiostat have been de- scribed earlier.16 In the case of stability studies in serum and blood, hydrogen peroxide measurements were made following periods of specified exposure of the sensor to undiluted human serum and blood.Results Glucose is oxidized electrochemically as pH is raised, particu- larly above 10 (Fig.1). Under these conditions on a Pt anode (+0.65 V versus Ag/AgCl) no plateau region of pH indepen- dence was observed. When the applied polarizing voltage was varied (Fig. 2) using pH 13.5 for optimal response, a sigmoidal potential dependence was seen, and though there was no definite plateau, +0.65 V versus Ag/AgCI was adopted as a standard.Responses to glucose in the range of interest in clinical monitoring (0-30 mmol 1-1) are shown in Fig. 3 (+0.65 V versus Ag/AgCl, pH 13.4). Glucose did not permeate PVC, but there are, however, responses when SPEES-PES is employed, which are amplified when the mixed membrane of SPEES-24 Analytical Communications, January 1996, Vol33 200 160 p 120 .u, PES-PVC is used, such that the response with the mixed membrane is not intermediate between that of either polymer alone but significantly enhanced ( X 3 at 20 mmol 1-1 of glucose) over SPEES-PES alone. Response with the mixed membrane is also linear over a greater glucose range than with SPEES-PES alone, up to at least 600 mmol 1- l , i.e., - 10% m/v glucose (y = 0.16~ - 1.2, I^ = 0.99).For all levels of glucose down to 5 mmol 1-1, current intensity was sufficient for assay purposes, but thinner membranes, or more precise tailoring of membrane composition, offer a means of signal enhancement at low concentrations. Partitioning of the buffer was maintained for prolonged periods with no significant signal drift during 8 h periods of use on repeated occasions.The selectivity of the membranes was investigated (Table 1) for commonly encountered interferents in clinical analyses. Ascorbate and urate signals are eliminated using the SPEES- . 40 / 7 8 9 10 1 1 12 13 14 PH Responses of a Cuprophan covered platinum electrode to 10 Fig. 1 mmol 1-1 glucose at varying pH (+0.65 V versus Ag/AgCl). 300 p 200 g . v) d 100 0 0 0.2 0.4 0.6 0.8 1 .o Polarising voltageN Fig.2 Responses of a Cuprophan covered platinum electrode to 10 mmol 1-1 glucose in pH 13.5 buffer at varying polarizing voltages (versus Ag/AgC1) h Al 0 5 10 15 20 25 30 [GI ucose]/mmol I-' Fig. 3 Responses to glucose at an electrode covered by PVC (a), SPEES- PES (A) and SPEES-PES-PVC (0) membranes, overlying a Cuprophan membrane bathed in pH 13.4 buffer (+0.65 V versus Ag/AgCl).PES-PVC membrane, as is the response to acetaminophen, a common problem even with other selective membrane materials including PVC12 and celluose acetate.16 Ethanol only shows any interference at higher concentrations. The biocompatibility of the membranes was studied by comparing responses to 2 mmol 1-1 H202 at membrane- protected non-enzyme electrodes used at +0.65 V versus Ag/ AgCl during exposure to serum and blood, with an underlying Cuprophan membrane included with the test membrane (inter- nal electrolyte pH 7.4).When a selective membrane was not used, over 60% signal attentuation occurred over a 200 min interval. Unplasticized PVC gave a final 85% remaining signal, SPEES-PES 90% and plasticized PVC 92%. The membrane with the least reduction in signal, however, was the mixed SPEES-PES-PVC membrane with a 95% retention of response (90% when blood was used).A general pattern of early rapid ( 10 min) decrease in signal was followed by an extended period of relative stability. Discussion Glucose signals at high pH are attributed to deprotonation of the slightly ionic 0-H bond on the C-1 of glucose, causing ring rupture.17 It appears likely that the involvement of an enolate ion,18 formed under alkaline conditions, is implicated in the increase in glucose oxidation at high pH.Glucose exclusion by PVC is possibly related to the polar OH groups of glucose. SPEES-PES permeability may be due to its polar SO3- groups interacting with the OH groups of glucose or enabling a degree of hydration14 such that partitioning of solute into the membrane is favoured.The likely explanation for the high responses with the composite membrane is a fine balance between the non-polar character of PVC-IPM constituents and polar SPEES-PES.15 The extended linearity with this sensor is attributed to the avoidance of enzyme saturation and the restriction of glucose flux through the membrane.The elimina- tion of both aromatic (acetaminophen) and charged (urate, ascorbate) interference is remarkable, and it appears fine tuning of permeability towards glucose is accompanied by exclusion of species of both charged and hydrophobic character. Also, ethanol interference is low, in part due to the intrinsically low electrochemical activity of the molecule, but also reduced by a factor of approximately 10 compared to a plasticized PVC covered electrode (Table 1).The discrimination between polar, uncharged species merits future study. Electrochemical oxidation of certain interferents produces a passivating film on the electrode surface: small molecule passivation of a polarized electrode surface by diffusible solutes has been reported by us.19 The selectivity shown by the highly perm-selective membranes results in the exclusion of such passivating species, not achieved by Cuprophan.At the outer Table 1 Responses (nA) to pH 7.4 solutions with electrode covered by Cuprophan dialysis membrane and overlaid polymer membranes (+0.65 V versus Ag/AgCl). Concen- tratiord SPEES- SPEES- mmol 1-1 PVC PES-PVC PES pH 7.4 electrolyte Ascorbate 10 <0.1 <0.1 0.30 Urate 10 <0.1 <0.1 0.25 Acetaminophen 5 335 <0.1 <0.1 pH 13.4 electrolyte - 2.0 Glucose 10 Ascorbate 1 <0.1 Ethanol 20 - - - Acetaminophen 1 - <0.1 - - < 0.2 -Analytical Communications, January 1996, Vol33 25 sensor surface the plasticizer component of the membrane could possibly account for non-adhesion of protein and cells during exposure to blood, by presenting a quasi-fluid surface, similar to that of a plasticized PVC membrane.The slight superiority in non-fouling shown for the mixed polymer membrane requires further elucidation. The surface heterogeneous polarity with hydrophobic (PVC) and charged (SPEES) domains could be a further factor in reducing surface fouling. Potential amphiphilic protein adherents may therefore encounter an interface that diminishes surface denaturation, reducing the tendency for further adhesion of proteins.Conclusions A sensor for glucose has been developed based on direct electrochemical oxidation of glucose at a platinum anode. The selectivity required has been imparted by the use of a novel membrane composite, in marked contrast to conventional enzyme glucose sensors in which the enzyme reaction product is selectively permeable through an inner membrane.The perm- selectivity properties shown are not predictable from those of either constituent membrane, and further study is being undertaken to establish the parameters involved, particularly the balance between polar and non-polar constituents.The work has opened up a route to more selective amperometric detection and sophisticated membrane materials that can be tailored to offer new avenues of sensor design. The authors thank SERC for financial support for this work. The SPEES-PES polymer was a gift from ICI Chemicals and Polymers (Runcorn, UK). References 1 Yao, S. J., ‘Chemistry and Potential Methods for In Vivo Glucose Sensing’ in Bioinstrumentation and Biosensors, ed. Wise, D.L., Marcel Dekker, NY, USA, 1991, p. 229. Marincic, L., Soeldner, J. S., Colton, C. K., Giner, J., and Morris, S., J . Electrochem. Soc., 1979, 126, 43. 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Marincic, L., Soeldner, J. S., Giner, J., and Colton, C. K., J . Electrochem. Soc., 1979, 126, 1687. Gough, D. A., Anderson, F.C., Giner, J., Colton, C. K., and Soeldner, J. S., Anal. Chem., 1978, 50, 941. Lewandowski, J. J., Malchesky, P. S., Moorman, M. J., Nalecz, M., and Nose, Y., IEEE Trans. Biomed. Eng., 1986, 33, 147. Yao, S. J., Chan, L.-T., Wolfson, S. K., Krupper, M. A., and Zhou, H. F., IEEE Trans. Biomed. Eng., 1986, 33, 139. Lemer, H., Giner, J., Soeldner, J. S., and Colton, C. K., An.N.Y. Acad. Sci., 1984, 428, 263. Sarangapani, S., Giner, J., Soeldner, J. S., Colton, C. K., Picha, G., Mayhan, K. G., and Drake, R. F., ‘Electrocatalytic Glucose Sensor’ in Implantable Glucose Sensors-The State of the Art (Hormone and Metabolic Research Supplement Series, Vol. 20), ed. Pfeiffer, E. P., and Kerner, W., Georg Thieme Verlag, Stuttgart, Germany, 1988, von Lucadou, I., Luft, G., Preidel, W., and Richter, G.J., ‘The Electrocatalytic Glucose Sensor’ in Implantable Glucose Sensors- The State of the Art (Hormone and Metabolic Research Supplement Series, Vol. 20), eds. Pfeiffer, E. P., and Kemer, W., Georg Thieme Verlag, Stuttgart, Germany, 1988, p. 41. Bockris, J. O’M., Piersma, B. J., and Gileadi, E., Electrochim. Actu, 1964, 9, 1329. Lyons, M. E. G., Fitzgerald, C. A., and Smyth, M. R., Analyst, 1994, 119, 855. Christie, I. M., Treloar, P. H., and Vadgama, P., Anal. Chim. Actu, 1992, 269, 65. Vadgama, P., Spoors, J., Tang, L. X., and Battersby, C., Biomed. Biochim. Acta, 1989, 48, 935. Bunn, A,, and Rose, J. B., Polymer, 1993, 34, 1 1 14. Benmakroha, Y., Christie, I., Desai, M., and Vadgama, P., Analyst, in the press. Koochaki, Z., Christie, I., and Vadgama, P., J . Memb. Sci., 199 1, 57, 83. Delahey, P., and Strassner, J. E., J . Am. Chem. Soc., 1952, 74, 893. Solomons, T. W. G., Fundamentals of Organic chemistry, Wiley, New York, USA, 1994, vol. 4, p. 904. Higson, S. P. J., Desai, M. A., Ghosh, S., Christie, I., and Vadgama, P., J . Chem. Soc., Faruday Trans., 1993, 89, 2847. p. 43. Paper 5107604H Received July 17, I995 Accepted November 21,1995

ISSN:1359-7337    

DOI:10.1039/AC9963300023

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, January 1996, Vol33 (27-30) 27 Use of Surfactant-modified Cellulose Acetate for a High-linearity and pH-resistant Glucose Electrode Andrew Mainesa, Andrea Cambiasob, Luca Delfinob, Giovanni Verreschib, Ian Christie" and Pankaj Vadgamaa a Department of Medicine (Clinical Biochemistry), University of Manchester, Hope Hospital, Salford, UK M6 8HD. E-mail: amaines@fsl .ho.man.ac.uk Dipartimento di Ingegneria Biofisica ed Elettronica, Universita di Genova, Via Opera Pia I IA, 16145 Genova, Italy Cellulose acetate modified by incorporation of the surfactant Tween-80 is used as an outer membrane in a glucose oxidase based enzyme electrode.The membrane acts as a glucose diffusion restricting barrier, allowing extension of response linearity to 0.4 mol I-l, with sensitivity of 2.5 nA 1 mmol-1.Response times of < 100 s can be achieved by use of initial slope ( I h ) measurements. The membrane also provides protection for the enzyme from low pH deactivation for at least 10 min, allowing for re-use in low-pH samples such as citrus fruit. Amperometric enzyme electrodes for the measurement of glucose have been the most extensively studied type of biosensor.1,2 Glucose biosensors are generally based on the enzyme glucose oxidase, which catalyses the oxidation of p-D- glucose by molecular oxygen producing gluconic acid and hydrogen p e r ~ x i d e . ~ glucose P-D-glucose + O2 + H20 - D-gluconic acid + H202 oxidase ( 1) The increase in hydrogen peroxide concentration and the decrease in oxygen concentration are proportional to the glucose concentration and either may be detected ampero- metri~ally.~ Detection of hydrogen peroxide is made by electrochemical oxidation at an anode.5 The major driving force behind this research remains the need for clinical glucose monitoring6 For clinical use, the basic specifications are relatively restricted; the glucose concentra- tion range is S40 mmol 1-1,7 the pH range is 7.0-7.5 and the ionic strength is =300 mmol 1-1.This demands a relatively small extension of the apparent enzyme Michaelis constant, K,, from that of free glucose oxidase (4.2 mmol 1-1)g to obtain a linear or near-linear calibration. This has been mostly achieved by the use of diffusion-restricting membranes to limit glucose flux, so avoiding enzyme saturation.Barrier membranes for clinical use have included low porosity membranes, either unmodified,9,10 or silane-treated, diamond-like carbon-coated12 or lipid treated.9 Other mem- brane materials include polyamide coated cellulose1 3 and polyurethane. l 4 , I These membranes extended substrate linear- ity to a maximum of 0.15 mol 1-1.13 In foodstuffs, notably fruit juices, glucose concentration may be up to 0.5 mol1-1.'6 Amine et a1.10 reported the measurement of glucose up to 1 mol 1-1 using a buffered flow injection system, but in this instance inevitably the device was not in equilibrium with the analyte.Other workers have measured high concentrations of glucose in banana17 and molasses18 but only after dilution. Such systems are not suitable for direct and in the field use on unmodified food samples, which moreover may have variable pH and ionic strength.The values of pH may differ significantly from the enzyme optimum (5.6 for glucose oxidase8), for example in citrus fruits the pH can be 2-3. Amine et a1.10 found immobilized glucose oxidase activity to decrease significantly below pH 4. For measurement without sample pre- treatment, therefore, the enzyme must be protected from pH deactivation.The present work explored a novel route to construct a membrane suitable for high-concentration glucose measure- ment in low, variable pH samples as a precursor to the development of reagentless devices for glucose in citrus fruit juices. Experimental Chemicals Glucose oxidase (GOD, EC 1.1.3.4) from Aspergillus niger [75% protein, 180 U mg-1 (1 U = 16.67 nkat) solid] and albumin (Bovine, Fraction V powder, 98-99% albumin) were from Sigma (Poole, UK).Alpha-D-glucose, NaZEDTA, hydro- chloric acid, cellulose acetate (39.7% acetyl content), acetone (99.9 + % HPLC grade) and Tween-80 (polyoxyethylene 20 sorbitan monooleate) were from Aldrich (Gillingham, UK). Sodium dihydrogenphosphate, disodium hydrogenphosphate, sodium chloride, sodium hydroxide, glutaraldehyde (25 % solution, electron microscopy grade) and aluminium oxide were from Merck (Poole, UK).Buffer A buffer comprising 15.6 mmol 1--l NaH2P04.H20, 52.8 mmol 1-1 Na2HP04, 51.3 mmol 1-1 NaCl and 2.1 mmol 1-l Na2EDTA was prepared in distilled water and adjusted to the required pH with HCl or NaOH. All solutions of glucose were made up with this buffer at the required pH and enzyme/ albumin solutions were made made up with this buffer at pH 7.4.Membranes Solvent cast polymer membranes were formed in 90 mm diameter glass Petri dishes, covered to retard solvent evapora- tion, rotated slowly to distribute the solution evenly and left at room temperature for completion of solvent evaporation (24 h for acetone, 72 h for tetrahydrofuran).Surfactant-modified cellulose acetate membranes (CA/ Tween) were cast from 5 ml of 5% m/v CA in acetone containing 1 % v/v Tween-80 non-ionic surfactant. Standard Cuprophan dialysis membranes used simply as support membranes for the enzyme laminates were obtained from a hemodialysis cartridge (Gambro, Lund, Sweden).28 Analytical Communications, January 1996, Vol33 Enzyme Laminate Fabrication A composite solution of GOD (5400 U ml- l ) and albumin (0.1 g ml-1) was prepared in buffer solution.GOD-albumin solution (10 y1) and 5 y1 of glutaraldehyde (5% v/v in buffer) were mixed rapidly and placed on a 1 cm2 piece of Cuprophan dialysis membrane. A further 1 cm* piece of Cuprophan was placed on top, and glass plates used to compress the enzyme so that the enzyme was evenly distributed.The cross-linked enzyme/membrane laminate was allowed to air-dry for 10 min. The Cuprophan laminate was used in all experiments, with or without an additional outer membrane. A 'blank' laminate was also constructed using an equivalent mass of albumin alone. Results and Discussion An outer Cuprophan dialysis membrane alone gave response linearity to approximately 3 mmol 1-1, though with high sensitivity (450 nA 1 mmol-l).The response was rapid, with a plateau current, representing steady-state conditions, attained after approximately 100 s (see Fig. 3, later). Cuprophan is a cuprammonium processed, regenerated cellulosic film of = 35 ym thickness and is normally used for excluding proteins (relative molecular mass cut-off 350-600 Da);19 it presents a negligible barrier to the diffusion of smaller molecules.The addition of an outer CA/Tween membrane increases linearity to at least 0.4 mol l-l, with a reduction of sensitivity to approximately 2.5 nA mmol 1-1 (Fig. 1). The response was slowed considerably, with an approximate plateau approached at 5-10 min.The assay time could be shortened, by measure- ment of initial slope (Ilt), to less than 100 s, with equivalent linearity to fixed time response (Fig. 1). Cellulose acetate has been frequently used to form mem- branes of the reverse osmosis type in order to effect selectivity, primarily for hydrogen peroxide detection at enzyme elec- trodes; its selective properties are based on charge and relative molecular mass.20 Non-ionized species permeate more rapidly than cations while anions are rejected, but the dense unmodified cellulose acetate membrane causes a great general reduction in flux for a range of species.21 Modification by incorporation of surfactant can increase the permeability of cellulose acetate to species such as hydrogen peroxide and paracetam0122 as well as improving mechanical properties.23 In the present investigation, incorporation of the surfactant Tween-80 into cellulose acetate produced a membrane that was sufficiently permeable to glucose to allow sensitive detection whilst restricting glucose diffusion to the enzyme layer such that linearity could be greatly increased.The mechanism by which general permeability can be increased by surfactant incorporation is unclear.The polyoxyethylenesorbitan moiety of the Tween-80 molecule would be expected to form abundant dipole4ipole interactions with the cellulose acetate polymer and therefore be relatively strongly retained in the membrane. Surfactant incorporation may alter hydrophobic/hydrophilic properties, and the high molecular mass Tween-80 molecule may also influence cellu- lose acetate spacing, possibly 'opening up' the dense poly- mer.The response linearity achieved with CA/Tween is signifi- cantly greater than that previously reported using unmodi- fied9.10 and modified11T12v24 porous and non-porous13-15 outer membranes and would allow measurement without dilution in a range of fruits. Exposure of enzyme electrodes to environments with pH values differing significantly from the enzyme pH optimum may lead to de-activation of the enzyme component.lO Fig.2 Apparatus and Electrode Assembly The amperometric cell (Rank Brothers, Bottisham, UK) consisted of a central 2 mm diameter platinum working electrode and outer concentric 12 mm diameter, 1 mm wide silver ring (Ag/AgCl) as a combined counter and reference electrode.l1 Before use, electrodes were polished with wet and then dry aluminium oxide powder.The electrodes were then covered with a small volume of buffer and the Cuprophan enzyme laminate with or without an additional outer surfactant modified cellulose acetate membrane was laid over the electrodes. The working electrode was polarized at +650 mV (versus Ag/AgCl) for the oxidation of enzymically generated H202, using a potentiostat (Chemistry Workshops, University of Newcastle, UK) with an output to a chart recorder (Lloyd Instruments, Fareham, UK) for recording of the current-time response.Glucose Response Baseline current in buffer (0.5-5 nA) was attained before measurement. A volume of 1 ml of the required concentration of glucose solution (left for 24 h to mutarotate before use) was added to the sample chamber and the response was monitored.Between exposures the sample chamber was rinsed three times with buffer and allowed to recondition in the buffer for 30 min. pH 7.4 calibration The enzyme laminates with and without an outer CAnween membrane were conditioned in pH 7.4 buffer and exposed to a range of glucose concentrations at pH 7.4, with the response recorded as the current at a fixed time of 2 min for Cuprophan and 10 min Cuprophan + CA/Tween.Varying pH calibration The enzyme laminates were conditioned and exposed to glucose solutions at the required pH, with the response recorded as the current at 2 min. A blank laminate was also exposed to pH 2.4. Current-time response with varying pH The enzyme laminate with and without an outer CAflween membrane was exposed to 100 mmol 1-1 and 1 mmol 1-l glucose solution, respectively, at the required pH, and the current response recorded over a 10 min period.Owing to the rapid response for the enzyme laminate alone, the response was recorded at a rate of 10 s-1 using an AT-MIO-16 data acquisition board (National Instruments, Austin, Texas, USA) in a personal computer (PC 466/VL, Elonex) controlled by a Labwindows (National Instruments) based 'in-house' software package.The much slower response with the CAflween covered laminate was recorded manually. 1200 1 7'5 p 1000 z \ -- 800 c 600 v) 400 a, 11I 200 6.0 - 2. E 4.5 ur Q 3.0 5 c. -0 "1. 1.5 I I I ' 0.0 0 100 200 300 400 Glucose concentration/mmol I-' Fig.1 Glucose calibration using Cuprophan GOD laminate with outer Tween-80 modified cellulose acetate membrane. @, Response at 10 min; ., initial slope responseAnalytical Communications, January 1996, Vol33 29 demonstrates the reduction in sensitivity due to continued exposure of the unprotected enzyme laminate to low pH values typical of those in citrus fruit (pH 3.4-2.4).The laminate retains some activity (1-2% of pH 7.4 activity) at pH 2.4, compared to the blank laminate. This activity was very low but stable, and activity could be restored by exposure to pH 7.4 buffer, demonstrating reversibility of the low pH de-activation of GOD. Fig. 3 shows the effect of a step exposure to low pH following pre-conditioning at pH 7.4.For pH 7.4,3.4 and 2.4 the response rate is equivalent for the first 20 s. Subsequently, the rate slows 1 o4 p 1000 $ 100 g $ 1 -.. .- E - 10 0.1 0.0 1.5 3.0 4.5 6.0 7.5 Glucose concentration/rnrnol I-’ Fig. 2 Effect of continual exposure of Cuprophan glucose oxidase laminate to varying pH. 10 min fixed time response. a, pH 7.4; ., pH 3.4; A , pH 2.4; V, pH 2.4 (blank-albumin only) 400 350 300 < 5 250 8 s 200 a 150 100 50 0 0 100 200 300 400 500 600 Tirne/s Fig.3 Effect of step exposure of Cuprophan GOD laminate to varying pH (1 mmol 1-1 glucose) plotted as response during initial 10 min. Data acquisition rate, 10 s-l 250 300 t 0 1 2 3 4 5 6 7 8 9 10 Tirne/rn in Fig. 4 Effect of step exposure of Cuprophan GOD laminate plus outer cellulose acetatenween-80 membrane to varying pH (100 mmol 1- glucose) plotted as response during initial 10 min.Data plotted as mean of 3 runs, standard deviation as y error bar. W, pH 7.4; A , pH 3.4; a, pH 2.4 at pH 3.4 to reach a plateau below that for pH 7.4, and at pH 2.4 a peak is reached at approximately 60 s, followed by a rapid decrease towards very low activity from 300 s onwards. However, by use of the outer CARween membrane the enzyme was protected from pH deactivation at least over the first 10 min of exposure to low pH (Fig.4), allowing sufficient time for pH-independent measurement and re-use. For example, the response at three successive exposures to pH 2.4 at 3 min expressed as a percentage of the first exposure was 100,97 and 101%. There was no indication of a significant loss of activity with repeated exposure to low pH.Desai et a122 found that Tween-80 could be incorporated into cellulose acetate without unduly compromising rejection of ascorbate at the relatively low concentrations typical of blood. In the present investigation, the direct electrode response for ascorbate at a much higher concentration typical of citrus fruit (2.8 mmol 1-1) was 1-2% of the response to 100 mmol 1-l glucose (typical citrus) at pH 7.4, therefore selectivity was sufficiently maintained.However, the charge-based selectivity of cellulose acetate varies with pH and interferent pKa20 and loses effectiveness at low pH as acidic interferents such as ascorbate become uncharged. The relative responses to 2.8 mmol 1-1 ascorbate at pH 7.4, 3.4 and 2.4 using the outer CA/ Tween membrane were 1 .O : 2.9 : 9.8.The development of pH- independent ascorbate rejecting membranes is ongoing. The use of Tween-80 modified cellulose acetate as an outer membrane in GOD-based enzyme electrodes therefore allows direct glucose measurement in samples such as fruit with a greatly extended glucose concentration range and provides protection for the enzyme from low-pH de-activation such that re-use is feasible in the varied pH conditions encountered in foodstuffs.The authors acknowledge financial support from the European Commission Standards, Measurements and Testing Programme (Contract MAT 1 -CT93-0034). References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Vadgama, P., J. Membrane Sci., 1990, 50, 141.Wilson, G. S., in Biosensors: Fundamentals and Applications, ed. Turner, A. P. F., Karube, I., and Wilson, G. S., Oxford University Press, New York, 1987, p. 165. Vaidya, R., and Wilkins, E., Med. Eng. Phys., 1994, 16, 416. Lowry, J. P., McAteer, K., El Altrash, S. S., Duff, A., and O’Neill, R. D., Anal. Chem., 1994,66, 1754. Koochaki, Z., Higson, S. P. J., Mutlu, M., and Vadgama, P.M., J . Membrane Sci., 1993, 76, 261. Velho, G. D., Reach, G., and Thevenot, D. R., in Biosensors: Fundamentals and Applications, ed. Turner, A. P. F., Karube, I., and Wilson, G. S., Oxford University Press, New York, 1987, p. 390. Churchouse, S. J., Mullen, W. H., Keedy, F. H., Battersby, C . M., and Vadgama, P. M., Anal. Proc., 1986, 23, 146. Schonberg, D., and Stephen, D., Enzyme Handbook Vol.P l a s s I .13-1.974xidoreductases, Springer-Verlag, Berlin, 1994. Tang, L. X., Koochaki, Z. B., and Vadgama, P. M., Anal. Chim. Acta, 1990,232,357. Amine, A., Patriarche, G. J., Marrazza, G., and Mascini, M., Anal. Chim. Acta, 1991, 242, 91. Mullen, W. H., Keedy, F. H., Churchouse, S. J., and Vadgama, P. M., Anal. Chim. Acta, 1986, 183, 59. Higson, S.P. J., and Vadgama, P. M., Anal. Chim. Acta, 1993, 271, 125. Eisele, S., Arnmon, H. P. T., Kindevater, R., Grobe, A., and Gopel, W., Biosens. Bioelectron., 1994, 9, 119. Shichiri, M., Kawamori, R., Goriya, Y., Yamasaki, Y., Nomura, M., Hakui, N., and Abe, H., Diabetologia, 1983, 24, 179. Vadgama, P., Spoors, J., Tang, L. X., and Battersby, C., Biomed. Biochim. Acta, 1989, 48, 935. Belitz, H. D., and Grosch, W., Food Chemistry, Springer-Verlag, Heidelberg, 1987. D’Costa, E. J., Dillon, M., Hodgson, F. J. A., and Quantick, P. C., Analyst, 1988, 113, 225.30 Analytical Communications, January 1996, Vol33 18 Bradley, J., Kidd, A. J., Anderson, P. A., Dear, A. M., Ashby, R. E., and Turner, A. P. F., Analyst, 1989, 114, 375. 19 Gutch, C. F., Ann. Rev. Biophys. Bioeng., 1975, 4,405. 20 Christie, I. M., and Vadgama, P. M., Chem. Sens. Technol., 1994,5, 131. 21 Koochaki, Z., Christie, I., and Vadgama, P., J . Membrane Sci., 1991, 57, 83. Paper YO754 7E 22 Desai, M. A., Ghosh, S., Crump, P. W., Benmakroha, Y., and Received November 20,199.5 Vadgama, P. M., Scand. J . Clin. Lab. Invest., 1993, 53, 53. Accepted December 1.5, 1995 23 Ghosh, S., Desai, M., Christie, I., Vadgama, P., Polymer Mater.: Sci. Eng., 1994, 70, 184. 24 Higson, S . P. J., Desai, M., Koochaki, Z., and Vadgama, P. M., Anal. Chim. Acta, 1993, 276, 335.

ISSN:1359-7337    

DOI:10.1039/AC9963300027

出版商:RSC    

年代:1996

数据来源: RSC