Automated enzymatic assay for measurement of lithium ions in human serum.
The most commonly used methods to detect serum lithium are ion-selective electrodes (ISEs) and flame emission photometry. ISE analyses rely on ion-specific electrodes. Ideally, each electrode possesses a unique ionselective property that allows it to respond to the desired ion. In practice, however, interference from other ions in the sample compromises the specificity of the detecting electrode, rendering the electrodes susceptible to false readings. The instrumentation for ISE analysis is relatively expensive, requires routine maintenance that is sometimes cumbersome and time-consuming, and demands that the operating technician have considerable skill and knowledge for accurate and consistent readings. Flame emission photometry relies on the principle that certain atoms, when energized by heat, become excited and emit a light of characteristic wavelength. Radiant energy produced by atoms in the flame is directly proportional to the number of atoms excited in the flames, which is directly proportional to the concentration of the substance of interest in the sample. Like ISE analysis, the instrumentation required for this method is complex and expensive. Moreover, flame emission photometry requires the use of combustible gas, necessitating expensive hazard prevention measures. More recently, a dye-based lithium assay has been developed and used in certain automated chemistry analyzers (5).
We recently developed an enzymatic lithium assay, which has been adapted to most automated clinical chemistry analyzers. In this assay, lithium is determined through a kinetic coupling system involving a lithium-sensitive enzyme, 3',5'-bisphosphate nucleotidase, from yeast. This enzyme is sensitive to lithium inhibition, with an [IC.sub.50] of 0.1 mmol/L. The enzyme is also sensitive to sodium inhibition, with a much higher [IC.sub.50] (>20 mmol/ L). The inhibitive effect of serum sodium ions is effectively masked by the sodium-specific binding reagent Kryptofix 221. Through enzymatic coupling, substrate adenosine 3',5'-bisphosphate (PAP) is converted to hypoxanthine by enzymatic reactions to generate uric acid and hydrogen peroxide ([H.sub.2][O.sub.2]). The generated [H.sub.2][O.sub.2] then reacts with N-ethyl-N-(2-hydroxy-3-sulfopropyl)-3-m-toluidine (EHSPT) and 4-aminoantipyrine (4-AA) to form a quinone dye with an absorbance maximum at 556 nm. The rate of dye formation is inversely proportional to the lithium concentration in serum:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
where PNP is purine nucleoside phosphorylase.
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
The assay is formulated into a lyophilized 2-reagent system with MES buffer (pH 6.0). The assay analyzes nonhemolyzed serum samples on a variety of automated analyzers in as little as 10 min.
The precision of the Diazyme enzymatic lithium assay was evaluated on a Cobas Mira analyzer according to Clinical and Laboratory Standards Institute (formerly NCCLS) guideline EP5-A. In the study, 2 specimens containing 1 and 2.3 mmol/L lithium, respectively, were tested in 2 runs per day with duplicates for a period of 20 working days. The results showed that within-run imprecision (CVs) was 4.7% for 1.0 mmol/L [Li.sup.+] and 3.3% for 2.3 mmol/L [Li.sup.+], respectively; and total imprecision (CVs) was 6.9% for 1 mmol/L and 5.5% for 2.3 mmol/L [Li.sup.+], respectively.
To demonstrate accuracy, the Diazyme lithium enzymatic assay was tested with individual serum samples and compared with both ISE and Thermo Trace colorimetric methods. Three concentrations of serum calibrators containing 0 mmol/L (low), 1 mmol/L (medium), and 3 mmol/L lithium (high) were prepared by adding 1 mol/L lithium acetate stock solution to the pooled lithium-free serum. After verification with a Trace lithium assay, the calibrators were sent out to a certified laboratory for confirmation.
The pooled serum and the individual patient serum samples used for this study were from a certified commercial source and were accompanied by an Institutional Review Board certificate of approval for all protocols, including informed consent, used to collect samples. To ensure that the lithium concentrations were distributed across the reportable dynamic range, some lithium serum samples (especially at lithium concentrations >2.5 mmol/L) were supplemented to the targeted concentrations with a stock solution of 1.0 mol/L lithium.
The Diazyme enzymatic lithium assay and ISE were compared on Cobas Mira [n = 56; lithium concentration, 0-3 mmol/L; y = 1.008x + 0.10 mmol/L ([r.sup.2] = 0.953; [S.sub.y|x] = 0.042 mmol/L)], Hitachi 717 [n = 67; lithium concentration, 0-3 mmol/L; y = 0.987x + 0.072 mmol/L ([r.sup.2] 0.962; [S.sub.y|x] = 0.19 mmol/L)], and Synchron CX-7 [n = 38; lithium concentration, 0-3 mmol/L; y = 0.980x + 0.028 mmol/L ([r.sup.2] = 0.989; [S.sub.y|x] = 0.019 mmol/L)] analyzers. The Diazyme lithium enzymatic assay and Trace lithium reagent were also compared [n = 61; lithium concentration, 0-3 mmol/L; y = 1.083x - 0.071 mmol/L ([r.sup.2] = 0.950; [S.sub.y|x] = 0.20 mmol/L)].
We also determined the linearity of the Diazyme lithium enzymatic assay. A serum sample containing 0 mmol/L Li' was supplemented with lithium acetate stock solution to a concentration of 3.0 mmol/L [Li.sup.+]. The serum sample containing 3.0 mmol/L was then serially diluted with a serum sample containing 0 mmol/L lithium to obtain final lithium concentrations of 0, 0.5,1.0,1.5, 2.0, 2.5, and 3.0 mmol/L. The serum samples prepared as described were tested in triplicate with the Diazyme enzymatic lithium assay on the Cobas Mira. The result demonstrated that the assay was linear at least up to 3.0 mmol/L lithium (y = 0.9857x + 0.0548 mmol/L; [r.sup.2] = 0.9863).
To determine the extent of interference from other cations and substances typically present in serum, we supplemented 1 mL of serum containing 1.0 mmol/L lithium with various concentrations of substances, and then assayed the sera (6 replicates) with the Diazyme enzymatic lithium assay on the Hitachi 917. The results are shown in Table 1.
In summary, we have developed an enzymatic coupling assay for quantitative measurement of lithium in nonhemolyzed human sera and have developed applications for commonly used automated chemistry analyzers. The assay uses 2 reagents, and applications have been developed for testing human serum specimens on the Cobas Mira, Synchron CX-7, and Hitachi 717. The within-run CV was <4.7%, and the total CV was <6.9%. The study testing human sera with lithium concentrations ranging from 0 to 3 mmol/L demonstrated good correlation with both a commercially available ISE method and a colorimetric method on various automated analyzers. The assay was linear up to 3.0 mmol/L. No interference was detected from the following substances at the indicated concentrations: sodium, 200 mmol/L; ammonium, 0.5 mmol/L; calcium, 4.0 mmol/L; magnesium, 2.0 mmol/L; ascorbic acid, 5.0 mmol/L; zinc, 0.25 mmol/L; iron, 0.25 mmol/L; copper, 0.25 mmol/L; potassium, 10 mmol/L; triglycerides, 2.82 mmol/L (250 mg/dL); and bilirubin, 770 [micro]mol/L (45 mg/dL).
(1.) Murguia JR, Belles JM, Serrano R. A salt-sensitive 3'(2'),5'-bis phosphate nuclectidase involved in sulfate activation. Science 1995;267:232-4.
(2.) Moyer TP, Pippenger CE. Therapeutic drug monitoring. In: Burtis CA, Ashwood ER, eds. Tietz textbook of clinical chemistry, 2nd ed. Philadelphia: WB Saunders, 1994:1094-154.
(3.) Amdisen A. Serum lithium determinations for clinical use. Scand J Clin Lab Invest 1967;20:104-8.
(4.) Wachtel MS, Paulson R, Plese C. Creation and verification of reference intervals. Lab Med 1995;26:593-7.
(5.) Rumbelow B, Peake M. Performance of a novel spectrophotometric lithium assay on a routine biochemistry analyzer. Ann Clin Biochem 2001;38: 684-6.
DOI : 10.1373/clinchem.2005.053439
Chao Don, Olga Aleshin, Abhijit Datta, and Chong Yuan * (Diazyme Laboratories Division, General Atomics, 3550 General Atomics Ct., San Diego, CA 92121; * author for correspondence: fax 858455-4754, e-mail www.diazyme.com)
Table 1. Interference by cations and substances typically found in serum. Interferent Concentration Deviation, % N[H.sub.4.sup.+] 0.5 mmol/L +0.8 [P.sub.I] 1.5 mmol/L -1.6 [Ca.sup.2+] 5.0 mmol/L +4.7 [Mg.sup.2+] 2.0 mmol/L +6.2 [Na.sup.+] 200 mmol/L +3.9 [K.sup.+] 10 mmol/L -0.8 [Cu.sup.2+] 0.25 mmol/L -0.8 [Fe.sup.3+] 0.25 mmol/L +0.8 [Zn.sup.2+] 0.25 mmol/L +0.8 Triglycerides 2.82 mmol/L (250 mg/dL) -1.2 Ascorbic acid 5.0 mmol/L +0.8 Bilirubin 770 mol/L (45 mg/dL) +6.5 (a) The interference study was carried out by adding various concentrations of the substances examined to individual patient serum samples and then testing with the Diazyme enzymatic lithium assay (6 replicates) on a Hitachi 917 analyzer. The cations and substances listed produced a <10% deviation from the result for 1.0 mmol/L lithium at the concentrations given.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Abstracts of Oak Ridge Posters|
|Author:||Don, Chao; Aleshin, Olga; Datta, Abhijit; Yuan, Chong|
|Date:||Oct 1, 2005|
|Previous Article:||Development of a novel enzymatic cycling assay for total homocysteine.|
|Next Article:||Development of an automated enzymatic assay for the determination of glycated serum protein in human serum.|