Methods for emission spectrochemical analysis

Abstract

Blood glucose is measured using enzymatic conversion of /3-o-glucose to o-gluconic acid and hydrogen peroxide in an immobilized glucose oxidase (EC 1.1.3.4) column. The per-oxide subsequently reacts with a mixed luminol-ferricya-nide reagent to produce chemiluminescene, proportional to jS-o-glucose concentration. The method is linear between 10-8 and 10~4Af glucose, and correlates well with standard methods for glucose determination. With prior adsorption of uric acid, the chemiluminescent technique may be used for urine glucose analysis. The system may also be applied to the analysis of hydrogen peroxide in the 10~8 to 10-5M range. Estimation of true blood glucose has been hampered by the relative nonspecificity of most analytical techniques. Glucose analysis based on the inherent specifity of an enzy-matic reaction has provided the most accurate means for obtaining blood glucose concentration. The glucose oxidase method, originally described by Keston (1), is the most commonly employed enzymatic technique for routine blood glucose analysis. The method is based on the following reaction sequence: glucose oxi dase ß-glucose + 02-* d-gluconic acid + H202 peroxidase Ho02 + chromogenic oxygen acceptor-chromogen in which the chromogen most frequently is o-dianisidine or o-toluidine. The first reaction is highly specific for glucose (2); however, the second reaction is subject to several interferences. These include reducing substances, such as biliru-bin, ascorbic acid, uric acid, and drug metabolites, which may depress results by either competing with the chromo-gen for peroxide or by reducing the chromogen (3). Negative error may also be observed if the pH is too acidic for the enzymatic reactions. Under these conditions, peroxi-dase is inhibited by fluoride and chloride ions which may be present in the reaction media as serum preservatives (4). Glucose estimation may be pH dependent if the pH of the final solution remains above four, since the absorption maximum of the oxidized chromogen is pH dependent above this value (5). To circumvent many of the interferences associated with the peroxidase-coupled glucose oxidase method, the present technique monitors hydrogen peroxide concentration using the chemiluminescence of luminol (5-amino-2,3-dihy-drophthalazine-l,4-dione). In the presence of certain metals , peroxide reacts with luminol in basic media to form an excited aminophthalate anion, which returns to ground state by the emission of a photon (6, 7). In the glucose oxi-dase-luminol coupled reaction sequence, the amount of light emitted is proportional to ß-D-glucose concentration. Generally, metal ions possessing oxidation states requiring a one-electron transfer are capable of promoting the chemiluminescent reaction between peroxide and luminol in water (8). These have included Fe(II)-containing compounds , such as hemin (9, 1.0) and hematogen (11). Cop-per(II) (12-16), as well as mixed Cu(II)-persulfate (17) and Cu(II)-hemin (18) solutions, have also been employed in the luminol reaction. Cobalt(II) (19), Fe(III) (20), Fe(CN)63_ (21, 22) and SbCle-(23) have been cited as re-agents capable of producing chemiluminescence in the presence of luminol and hydrogen peroxide. The present paper summarizes the chemiluminescent response promoted by several of the above metals, observed during attempts to establish a procedure for peroxide analysis based on the luminol reaction. It further describes the adaptation and development of this analysis for the determination of blood glucose. A preliminary communication of this work has appeared (24). Other workers have independently reported a similar method (25). EXPERIMENTAL Apparatus. The chemiluminescence produced by the oxidation of luminol is followed in a continuous flow system using the apparatus shown in Figure 1. The system uses three 50-ml plastic syringes , containing luminol dissolved in 0.1 H3BO3-KOH buffer, KaFetCNle, or another metal, and an aqueous background of 0.004M acetate buffer. The syringes are driven by a Harvard Model 600-2-200 infusion pump, .capable of maintaining uniform flow against back pressures greater than 250 psi produced by the enzyme column. Solutions from the ferricyanide and luminol syringes are joined by a glass Y-tube containing a platinum coil to enhance mixing. A platinum gauze plug is located father down the flow line for the same purpose. Samples are introduced into the acetate flow line by a Chroma-tronix SV-8031 sample injection valve. Either the acetate background or the same slug flows into the glucose oxidase column. The column itself is a 16-cm Pyrex tube with an i.d. of 4 mm; 2 cm from each end of the column, the i.d. is decreased to 3 mm to accommodate Chromatronix column fittings. As the glucose sample enters the column, hydrogen peroxide generated and carried in the column effluent to the cell where it reacts with the luminol-ferricya-nide reagent. Nitrogen gas is bubbled through the cell to ensure uniform mixing. The chemiluminescence produced by the luminol-peroxide-fer-ricyanide reaction is detected by an RCA 1P21 photomultiplier