| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Department of Laboratory Medicine, Shinshu University Hospital, Matsumoto, Japan;2 Clinical Laboratory Center, The University of Tokyo Hospital, Tokyo, Japan;3 Department of Laboratory Medicine, Shinshu University School of Medicine, Matsumoto, Japan;4 Shino-test Corporation, Kanagawa, Japan;5 Asahikasei Pharma Corporation, Shizuoka, Japan;
aaddress correspondence to this author at: Department of Laboratory Medicine, Shinshu University, Hospital, 3-1-1 Asashi, Matsumoto 390-8621, Japan; fax 81-263-34-5316, e-mail yamauchi{at}hsp.md.shinshu-u.ac.jp
Various methods have been used to measure calcium in body fluids. Atomic absorption spectrophotometry (AAS) is the most reliable (1), but it requires special instrumentation. The most widely used method involves colorimetric detection of calcium complexes by o-cresolphthalein complexone (o-CPC) (2)(3) or arsenazo. With the o-CPC method, reagent stability and recoveries at low concentrations are poor, and magnesium interferes in the reaction (4). Newer methods using o-CPC (5)(6)(7) or other colorimetric agents (8)(9)(10) are not entirely satisfactory.
Various enzymatic methods have been described, including methods using porcine pancreatic 
-amylase (EC 3.2.1.1) (11), phospholipase D (PL-D; EC 3.1.4.4) (12)(13), and urea amidolyase (14). The first two are based on activation of enzymes by calcium, whereas the third is based on inhibition of the enzyme by calcium. The -amylase method is reportedly inaccurate for patients with hyperamylasemia (11), and the other 2 methods each require 2 reaction steps (12). In this report, we describe a new, simple, specific enzymatic assay based on activation of PL-D. We investigated the assay characteristics and its suitability for use in routine laboratory tests.
We obtained 126 serum samples from patients admitted to Shinshu University Hospital after receiving informed consent from the patients and approval by our institutional ethics committee.
We obtained PL-D from Streptomyces chromofuscus [ (15); Asahi Kasei Pharma], bis(p-nitrophenyl) phosphate (BPNPP) from Kanto Chemical Co., Good’s buffer from Doujin Laboratories, and SRM 915 and 909a from NIST. The reagent sets for the o-CPC and -amylase methods were from Serotec Co. Ltd. and Ono Pharmaceutical Co. Ltd., respectively. Other reagents were analytical grade (Wako Pure Chemical).
The new method is based on increased PL-D-catalyzed hydrolysis of BPNPP by calcium ions, as follows. The p-nitrophenol released by the reaction is detected at 405 nm (Scheme 1
).
|
The assay is a 2-point fixed-rate assay performed at 37 °C; for our experiments we used a Hitachi 7170 analyzer. Briefly, 9.5 µL of sample was mixed with 160 µL of solution containing 750 U/L PL-D and 0.275 mmol/L calcium acetate in 60 mmol/L Good’s buffer (pH 7.5). After incubating the mixture for 5 min at 37 °C, we added 160 µL of reagent containing 1.5 mmol/L BPNPP in 60 mmol/L Good’s buffer (pH 7.5) and measured the absorbance at time points 23–34 (6.76–10.00 min). We performed the o-CPC and -amylase methods on the same analyzer according to the manufacturer’s instructions.
For the AAS method (1), we used a 0.1-mL sample on a Hitachi Z-5000 Atomic Absorption Spectrophotometer (flame-type analyzer) equipped with an autosampler and a multirange micropipetting device (Gilson Inc.) with triple rinsing of the inside of the tip. We increased sample dilution from 50- to 75-fold to decrease viscosity and increase linearity.
To prepare calibrators, we dissolved NIST SRM 915 (10.010 g) in 55 mL of 2 mol/L hydrochloric acid and brought the volume to 1000 mL with water. We diluted this stock solution with a diluent containing 140 mmol/L NaCl and 5 mmol/L KCl to prepare calibrators (1.5, 2.5, and 3.5 mmol/L). We assayed the calibrators included with each reagent set by the AAS method and calculated correction factors for the calibration of each method. We performed linear regression analyses with the Medical Communication Program (Ver. 5 for Windows 95; Sysmex Co. Ltd.). Details of the method optimization experiments are shown in the Data Supplement that accompanies the online version of this Technical Brief at http://www.clinchem.org/content/vol51/issue6/.
To inhibit the activity of endogenous phosphodiesterase, we used an anionic detergent at a final concentration of 2.5 g/L. To increase the linear portion of the calibration curve, we added sodium sulfate and tartaric acid (as inhibitors of the PL-D reaction) to reagents 1 and 2, respectively. We decided on final tartaric acid and sodium sulfate concentrations of 37.5 and 200 mmol/L, respectively.
To remove the negative interference of EDTA in the determination of serum calcium concentrations, we added nickel ions [nickel(II) acetate tetrahydrate] to reagent 1. A final concentration of 0.4 mmol/L or higher was necessary to prevent the influence of EDTA completely. The concentration of nickel ion had no effect on either dilution linearity or assay sensitivity. We decided on a final nickel ion concentration of 0.5 mmol/L.
The absorbance increased linearly above time point 17, and we decided to use the change in absorbance between time points 23 and 34 to calculate calcium concentrations. Ca2+ was added to reagent 1 to improve the linearity at low concentrations. The calibration curve was linear from 0 to 7.0 mmol/L. When we used serial dilutions of SRM 915 in a pooled serum, the assay was linear from 0 to 6.0 mmol/L.
The results of the reproducibility study are summarized in Table 1
. When we assayed SRM 915 solutions and samples of SRM 909a to which 0.252 or 1.26 mmol/L calcium had been added, the recoveries were 97.0% and 99.5%, respectively.
|
The mean (SD) values for SRM 909a-I (labeled value, 2.225 mmol/L) and SRM 909a-II (labeled value, 3.540 mmol/L) obtained from triplicate determinations in 5 separate experiments were 2.218 (0.001) and 3.542 (0.002) mmol/L, respectively.
We observed no interference when we added Mg2+, Fe2+, Fe3+, Cu2+, and Zn2+ at maximum concentrations of 205.7, 89.5, 53.7, 31.5, and 30.6 µmol/L, respectively (the calcium values obtained being 98.9%–100.3% of the expected values). Unconjugated bilirubin up to 196 mg/L did not interfere, nor did conjugated bilirubin up to 205 mg/L, hemoglobin up to 4.6 g/L, turbidity up to 2000 hormadin turbidity units, or EDTA up to 1.00 g/L.
We compared the new method with the AAS method (Fig. 1A
) and the o-CPC method (Fig. 1B
), and the -amylase method with the AAS method (Fig. 1C
). The absolute value of the y-intercept obtained with the new enzymatic method was the smallest among the 3 methods. Moreover, the slope for the new method was closer to unity than were the slopes of the other methods.
|
We assessed the stability of the reagents stored for 2 months at 5 °C and room temperature, using 3 serum samples. The measured values, as a percentage of the initial values, were 99%–101% at 5 °C and 96%–98% at room temperature.
Various enzymatic assay methods for measuring calcium have been developed (11)(12)(13) in attempts to overcome the problems experienced with the o-CPC method, the most widely used conventional method. PL-D is specifically activated by calcium, not by magnesium, which can affect the values obtained with the o-CPC method. Although a workable method using PL-D was developed, the actual method involved coupled enzyme reactions using both PL-D and choline oxidase (12)(13). The proposed method is a simple assay system consisting of a 1-step reaction with BPNPP used as the substrate for PL-D.
We carefully devised the composition of the reagent mixture to establish the new method. It was necessary to first inhibit the endogenous phosphodiesterase because the absorbance obtained for the serum blank is increased by the reaction of phosphodiesterase with BPNPP. The addition of anionic detergent seemed to suppress this influence efficiently. Partial inhibition of PL-D was also necessary to make the assay adequately quantitative. The Km of PL-D for calcium is comparatively low, 7.5 x 10–5 mol/L (16), and the absorbance did not increase in a calcium-concentration–dependent manner. We therefore added tartaric acid and sodium sulfate to the reagent mixture to inhibit the PL-D activity. Tartaric acid effectively inhibited PL-D activity by its chelating action, and sodium sulfate acted as a competitive inhibitor. These agents made the Km value higher, and under those conditions the absorbance increased in proportion to the calcium concentration. Consequently, assay linearity was improved. We also added nickel ion to the reagent mixture to eliminate the influence of the chelating action of EDTA. Because the chelating stability constant of EDTA for nickel ions (18.56) (17) is much larger than that for calcium (10.96) (18), nickel ions could be completely substituted for calcium ions, in terms of chelation by EDTA.
The o-CPC method does not measure albumin-bound calcium completely and is insufficient in terms of stoichiometry because the affinity of calcium for albumin is greater than that of calcium for o-CPC. This problem is made apparent by the lower recoveries at low calcium concentrations. In our study, this influence was revealed by the y-intercepts obtained in the correlation between the o-CPC and the AAS methods (–0.161 mmol/L). The affinity of calcium for PL-D is much greater than that of calcium for albumin; therefore, the reagents in the new enzymatic method should react with albumin-bound calcium. Indeed, the y-intercept (–0.023 mmol/L) was improved vs that seen for the o-CPC method.
Reagent stability is also a major problem with the o-CPC method, the stability being adversely affected by CO2 absorption (8). The reagent used in the new enzymatic method was stable when stored for 2 months at 5 °C and room temperature.
In conclusion, we have developed a new, simple enzymatic assay for the measurement of serum calcium. The method has good precision, is specific for calcium, being free from influences by metal ions and EDTA, and may be suitable for clinical laboratory tests.
References
-amylase. Clin Chem 1994;40:781-784.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
