Semi-automated enzymatic measurement of serum zinc concentration

Clinical Biochemistry 35 (2002) 41– 47

Semi-automated enzymatic measurement of serum zinc concentration Ozcan Erel*, Senel Avci Clinical Biochemistry Department, Medical Faculty of Harran University, Sanliurfa 63200, Turkey Received 2 October 2001; accepted 3 January 2002

Abstract Objective: To measure serum zinc concentration by means of carbonic anhydrase reactivation using an automated analyzer. Methods: The zinc content of carbonic anhydrase (CA), whose cofactor is zinc, was removed by dialysis against pyridine 2 to 6 dicarboxylic acid and a pure apoenzyme was obtained. Serum proteins were precipitated with trichloroacetic acid (TCA) solution and the supernatant fraction of the sample was used to determine the zinc concentration. The negative effects of the precipitant on CA activity in the assay were completely removed, reaction conditions for maximal CA activity were provided and the color of the product was enhanced and stabilized. P-nitrophenyl acetate was used as the substrate and the change of absorbance of p-nitrophenol which was produced was followed at 400 nm. The initial rate of the esterase activity of CA was measured by using an automated analyzer. Analytical performance characteristics of the assay were determined. The zinc concentrations in serum samples of healthy subjects and patients were measured. Results: The enzymatic assay is accurate, sensitive, specific and is not affected by other metals. There was excellent agreement with the results obtained using an atomic absorption spectrophotometer (AAS) (y ⫽ 0.98X ⫹ 0.18, r ⫽ 0.99). Serum zinc concentrations were found to be higher in patients with vivax malaria, and lower in patients with cutaneous leishmaniasis than in healthy subjects. Conclusion: The enzymatic method is suitable for semiautomated measurement of serum zinc concentration. © 2002 The Canadian Society of Clinical Chemists. All rights reserved. Keywords: AAS; Apocarbonic anhydrase; Automated analyzer; Carbonic anhydrase; Trichloroacetic acid; Zinc

1. Introduction The analysis of zinc in biologic media is of major importance in laboratories performing nutritional and clinical studies. Determination of zinc concentrations in biologic specimens is usually performed by atomic absorption spectrophotometry (AAS) [1] but the high instrument costs of this rather complicated technique have prevented its widespread use. Numerous enzymes are used in analytical chemistry because of their cofactor specificity [2]. The first step of the methods involves removal of the metal from the enzyme by a strong complexing agent. The apoenzyme formed is then brought into contact with the sample so that metal ions can be taken up and bound to the enzyme. Only enzyme molecules containing a metal ion will be active and the trace metal content can therefore be evaluated from a measurement of enzyme activity [3]. Carbonic anhydrase (CA; E 4.2.1.1), the first isolated metalloenzyme, contains zinc and it has a high turnover * Corresponding author. Tel.: ⫹90-414-314-9078; fax: ⫹90-414-3151181. E-mail address: [email protected] (O. Erel).

number for its substrates. The zinc ion is essential for the activity of CA and is bound to the enzyme with coordination bonds. By removing the zinc ion from the enzyme, apoCA, which has no enzymatic activity, is obtained. The enzyme regains its activity with the addition of zinc. Kobayashi et al. [4] determined zinc contents of water and fruit juices by means of the esterase activity of CA. Using apoCA Demir et al. [5] measured the zinc concentration in serum samples which had been pretreated with heat denaturation. In this study, we aimed to perform semiautomated enzymatic measurement of serum zinc concentration. 2. Materials and methods 2.1. Assay principle Zinc, which is the cofactor, is essential for the activity of carbonic anhydrase (CA). By removing the zinc ion from the enzyme, apocarbonic anhydrase (apoCA), which has no enzymatic activity, can be obtained. With the addition of zinc ions to apoCA, the enzyme regains enzymatic activity. Sera have high endogenous CA activity and the contained

0009-9120/02/$ – see front matter © 2002 The Canadian Society of Clinical Chemists. All rights reserved. PII: S 0 0 0 9 - 9 1 2 0 ( 0 2 ) 0 0 2 7 0 - 9

42

O. Erel, S. Avci / Clinical Biochemistry 35 (2002) 41– 47

zinc is bound to albumin. Before the enzymatic assay, all internal CA activity of sera must be removed and the bound zinc must be freed. By the precipitation of serum proteins, endogenous CA is denaturated and zinc ions are freed. During assay the zinc ion which is a cofactor is bound to apoCA and holoenzyme occurs. Using an appropriate neutralizing solution, the negative effects of the protein precipitant remnant, which remains in the supernatant fraction of the sample, on CA activity in the assay is neutralized and the reaction medium for maximal enzymatic activity of the holocarbonic anhydrase is provided. The rate of CA activity is related to the amount of zinc in the specimen. The sequence of the reactions in our method is shown below. Carbonic anhydrase (CA) without zinc is apocarbonic anhydrase (apoCA). Carbonic anhydrase 共CA兲

removal of zinc O ¡

apocarbonic ahydrase (ApoCA)

(1)

Serum sample ⫹ TCA

(2)

¡

The supernatant

TRIS buffer ⫹ the supernatant ⫹ substrate ⫹ApoCA ¡ CA ⫹ product

(3)

The initial rate of product formation is linearly related to the esterase activity of the CA, and the rate is dependent on the zinc content of the supernatant of the serum sample. The zinc concentration of sera can be estimated by following p-nitrophenol absorbance using an automated analyzer. 2.2. Apparatus A Cecil 3000 spectrophotometer with a temperaturecontrolled cuvette holder (Cecil), a Hitachi 911 automated analyzer (Roche), and a Varian SpectrAA 250 Plus atomic absorption spectrophotometer (Varian) were used. 2.2.1. Atomic absorption spectrophotometry A Varian SpectrAA 250 Plus atomic absorption spectrophotometer (Varian) was used. The instrument settings were made according to the standard procedure recommended by the manufacturer. The wavelength was 213.8 nm, with a deuterium background correction and the burner gas mixture was air-acetylene. 2.3. Chemicals and reagents Pyridine 2,6-dicarboxylic acid was obtained from Merck Co., other chemicals and bovine carbonic anhydrase enzyme were purchased from Sigma Chemical Co. All chemicals were ultra pure grade, and type I reagent grade deionized water was used. All volumetric flasks were washed thoroughly with 1% nitric acid and rinsed thoroughly with the deionized water.

2.3.1. Dialysis solution Buffered-zinc chelator solution: 12.5 g (0.075 mol) of pyridine 2,6-dicarboxylic acid was dissolved in 950 mL deionized water and 24 g (0.2 mol) of NaH2PO4 was added to this solution, and the solution was adjusted to pH 7.4 by adding 1 N NaOH solution, and lastly the volume was completed to 1000 mL. This solution was used to remove zinc ions from CA. 2.3.2. Precipitant solution Trichloroacetic acid (TCA) solution: 200 g of TCA was dissolved in 800 mL deionized water. The final volume was completed to 1000 mL (20% w/v). Diluted 5% TCA solution was prepared from the stock TCA solution. 2.3.3. Assay buffer Tris(hydroxymethyl)aminomethane (TRIS) buffer 0.5 mol/L pH 8.3. 60.6 g (0.5 mol) of tris[hydroxymetyl]aminometane was dissolved in 900 mL deionized water and this solution was adjusted to pH 8.3 by adding 1 N HCl solution. The final volume was completed to 1000 mL. This buffer was used as Reagent 1 in the assay. 2.3.4. Substrate solution The p-nitrophenyl acetate 0.0275 g (0.1518 mmol) was dissolved in 1 mL of acetone then the final volume was completed to 50 mL by slowly mixing with deionized water. The final substrate concentration was 3.0 mmol/L. Since the substrate solution has a basal autocatalytic activity, it had to be prepared daily. 2.3.5. Stock and working zinc standards Commercial zinc atomic absorption standard solution (1000 ␮g/mL, Sigma Co.) was used. Working standards were prepared from the stock standard solution by diluting with dilute HCl solution. 2.3.6. Preparation of apocarbonic anhydrase from carbonic anhydrase For the preparation of apocarbonic anhydrase (apoCA), 100 mg of the commercial bovine carbonic anhydrase (CA) was dissolved in 10 mL of phosphate buffer and placed into dialysis tubing. The enzyme was dialyzed against bufferedzinc chelator solution for 12 h. Then the chelator solution was discharged, deionized water was replaced and changed three times during last 12 h period at 4°C. In the first step, all zinc elements were removed from the enzyme and in the second step the chelator remnants were removed from the dialysis system. Consequently, 100% pure apoCA was obtained [5– 6]. A part of the apoenzyme was dissolved in the TRIS buffer and used as Reagent 2 in the assay. The apoCA solution is stable for two months at 4°C and it is also stable for at least six months at ⫺80°C. 2.4. Blood samples Blood samples of healthy subjects from a blood bank, of patients with vivax malaria from the Malaria Eradication

O. Erel, S. Avci / Clinical Biochemistry 35 (2002) 41– 47

43

Fig. 1. Absorbance spectrums of the substrate (p-nitrophenyl acetate) and the product (p-nitrophenol).

Center of Sanliurfa, and of patients with cutaneous leishmaniasis from the Leishmania Eradication Center of Sanliurfa in Turkey were obtained. Individuals fasted for 12 h before sample collection. All participants gave informed consent for the present study. Obtained venous blood samples were collected into tubes without any addition of anticoagulants, and serum was separated from cells by centrifugation at 1500 g for 10 min. Serum samples were stored at ⫺80°C until study. Equal volumes of the thawed serum and TCA solution (5% w/v) were mixed, vortexed and centrifuged. The zinc determination of the serum sample was performed from its supernatant fraction. 2.5. Statistical analyses ANOVA, Student’s t-test, correlation analyses, and linear regression analyses were performed by using SPSS for Windows, Release 9.5 computer program (SPSS Inc.). Deming regression analyses and difference analyses were performed by using Method Validator Computer Program, Version 1.1.9.0. (Metz-France, 1999).

3. Results 3.1. Optimization studies 3.1.1. The preparation of pure apocarbonic anhydrase and reactivation of the enzyme While the prepared apoCA was incubated with the substrate no change of absorbance was observed, and when the

zinc standard solution was added to the incubation medium the enzymatic activity appeared. Thus, we decided that the prepared apoCA was 100% pure and it could be used as an apoenzyme in the assay. A little esterase activity was shown when dialyses duration was shortened to less than 12 h. 3.1.2. Absorbance spectrums of the substrate and the product To estimate CA activity, we determined the spectral analyses of the substrate and the product (Fig. 1). The best appropriate wavelength to be followed was found. As seen in Fig. 1 the molar absorption coefficient of the product was the highest level at 400 nm. Thus the change of absorbance was followed at 400 nm wavelength. 3.1.3. The choice of the best appropriate protein precipitation method for semiautomated enzymatic zinc measurement of sera We tested almost all kinds of chemical protein precipitation methods, such as cold absolute ethanol, metaphosphoric acid 6% (w/v), phosphotungstic acid 5% (w/v) and trichloroacetic acid solutions 5% (w/v). We precipitated serum proteins using each of the precipitants. While we incubated the supernatants which were obtained by using different precipitants with apoCA and the substrate, we were not able to obtain any CA activity, in any one. We determined that the supernatants still contained the used precipitant remnants which denaturated (apo)CA in the assay. Therefore no enzymatic activity appeared. So, we had to completely neutralize and remove the negative effects of the used precipitant remnants in the supernatant fraction of

44

O. Erel, S. Avci / Clinical Biochemistry 35 (2002) 41– 47

Fig. 2. Absorbance spectrums of the product (p-nitrophenol) in TRIS buffers 500 mmol/L at various pH values.

samples before adding apoCA. We had to also provide an appropriate reaction medium for the maximal activity of CA in the assay. For these reasons, we tried to neutralize the used precipitants. Using more concentrated and alkaline buffers we could completely neutralize the TCA effect on the assay. Consequently, among the used precipitants we were able to neutralize the negative effects only of TCA using TRIS and phosphate buffers at different concentrations and at various pH values. 3.1.4. Type, molarity and pH of buffer to be used in the assay We tested different concentrations of TRIS and phosphate buffers to neutralize the TCA effect. We determined that the best appropriate buffer to neutralize the negative effects of TCA solution was the TRIS buffer at 0.5 mol/L concentration. At lower buffer concentrations the neutralization capacity was weak and at higher concentrations the autocatalyses of the substrate was higher. In the tested phosphate buffers, the autocatalytic activity of the substrate was higher than that found in the TRIS buffers. We also determined that the best appropriate pH value of the buffer for the maximal absorbance point of the product was 8.3. (Fig. 2).

3.2. Assay procedure 3.2.1. Automated measurement After manual spectrophotometric optimization processes, the method was applied to an automated analyzer, Hitachi 911. The assay format of the test is below. Reagent 1 volume 200 ␮L (Reagent 1: TRIS buffer 500 mmol/L pH 8.3) Sample volume 20 ␮L (Sample: the supernatant) Reagent 2 volume 20 ␮L (Reagent 2 (the substrate): p-nitrophenyl acetate 3 mmol/L) Reagent 3 volume 10 ␮L (Reagent 3: apoCA 3.1 mg protein of apoCA/mL) Wavelength 415 nm (the nearest wavelength to 400 nm in this autoanalyzer) Reading Point Two point, the first absorbance reading point at 30 s after the addition of the last reagent (R3), and the second absorbance reading point at 210 s after the addition of the R3 Calibration type Linear Reaction monitoring of the assay is shown in Fig. 3.

O. Erel, S. Avci / Clinical Biochemistry 35 (2002) 41– 47

45

Fig. 3. Time courses for overall reaction rates with various concentrations of zinc ion.

3.3. Analytical performance characteristics 3.3.1. Kinetics of reaction and linearity We examined the kinetics of the overall reaction and the linearity of the assay. There was a lag phase period less than 5 s after the addition of apoCA (Reagent 3). The first reading point was at 30 s after addition of Reagent 3. All of the reading points were performed after the lag phase period. The last reading point was at 210 s after addition of Reagent 3. The upper limit of the linearity was 38.25 ␮mol/L (250 ␮g/dL). In the regression analyses, the r value was 0.999, the slope was 3.27 ⫻ 10⫺2 (Sy/x ⫽ 0.001, p ⬍ 0.00001), the intercept was 0.1 (Sy/x ⫽ 0.001, p ⬍ 0.00001). In addition, by shortening the last reading points and/or by decreasing the sample volume ratio, the upper limit of the linearity can also be increased. Absorbance traces of the standards which are 0, 7.65, 15.30, 38.25 and 76.25 ␮mol/L are given in Fig. 3. Deionized water with and without TCA were used as zero standards. As seen in the figure the absorbance traces of water with and without TCA precipitant are not different. So, all negative effects of TCA on the assay were completely neutralized and removed. 3.3.2. Detection limit The detection limit of the method was determined by evaluating the zero calibrator 10 times. The detection limit, defined as the mean zinc concentration of the zero calibrator ⫹ 3 SD, was 0.32 ␮mol/L.

3.3.3. Precision To determine the precision of the enzymatic method we assayed three levels of sera. The sera pool which had the lowest zinc level was from the patients with cutaneous leishmaniasis and the sera pool which had the highest level was from the patients with vivax malaria and the sera pool with a normal level was from the healthy subjects. Withinand between-batch precision calculated for each of the three sera pools are shown in Table 1. 3.3.4. Analytical sensitivity As the slope of the calibration line, the analytical sensitivity was found to be 8.6 ⫻ 10⫺4 Absorbance/Amount, [A ⫻ (␮mol/L)⫺1]. 3.3.5. Analytical recovery We examined the recovery of serum zinc by enzymatic method by adding zinc standard to two pooled serum specimens and assaying. The analytical recoveries of added zinc were 97 to 102%. 3.3.6. Analytical quality Commercial quality control preparations from Bio-Rad Co. (Bio-Rad, Lypochek, Assayed Chemistry Control (Human) Level 1 (14051) and Level 2 (14052) were used. While the given mean value of level 1 was 11.3 and the acceptable range was 9.01 to 13.5 ␮mol/L, we measured it to be 10.83 ␮mol/L, and the mean value of level 2 was

46

O. Erel, S. Avci / Clinical Biochemistry 35 (2002) 41– 47 Table 2 Serum zinc concentrations of patients with vivax malaria, cutaneous leishmaniasis and healthy subjects

Table 1 Within- and between-batch precision data of automated enzymatic measurement method

Within-batch measurement High Medium Low Between-batch days High Medium Low

Number

Mean (␮mol/L)

CV %

20 20 20

23.17 15.15 8.44

3.8 2.4 4.1

20 20 20

23.25 15.45 8.59

3.9 2.7 3.8

given as 8.45, the acceptable range was 6.76 to 10.1, and we found it to be 8.76 ␮mol/L using the enzymatic method. 3.3.7. Specificity and interference Specificity of the assay for zinc was evaluated by adding Fe⫹2, Cu⫹2, Pb⫹2 and Mn⫹2 metals at 1.0 mmol/L total final concentration to apoCA. No reactivation was observed in any one. 3.3.8. Method comparison Zinc concentrations of serum samples were measured by semiautomated enzymatic measurement method and AAS.

Serum Zn, ␮mol/L

Healthy Subjects n ⫽ 94

Vivax Malaria n ⫽ 57

Cutaneous Leishmaniasis n ⫽ 61

15.1 ⫾ 2.4

16.9 ⫾ 2.8a

13.2 ⫾ 1.9ab

Mean ⫾ 1 SD, ap ⬍ 0.05 vs. healthy pesons, bp ⬍ 0.05 vs. vivax malaria.

There was no significant difference between the results (p ⫽ 0.106) (Fig. 4A) and the correlation coefficient was r ⫽ 0.99, p ⬍ 0.001 n ⫽ 104 and according to Deming regression, the slope was 1.007 and 0.95 confidence interval was 0.987 to 1.027 and the intercept was ⫺0.006 and 0.95 confidence interval was ⫺0.268 to 0.255 (Fig. 4B). 3.3.9. Reference interval To determine the reference interval for serum zinc, we assayed serum specimens from 112 healthy individuals using the enzymatic method. The reference range for serum zinc concentration was 10.1–20.1 ␮mol/L. 3.3.10. The results for healthy subjects and patients Serum zinc concentration was found to be lower in patients with cutaneous leishmaniasis and was found to be higher in patients with vivax malaria than in healthy subjects according to the ANOVA test (Table 2).

4. Discussion

Fig. 4. (A) Difference plot for serum zinc, with mean values as abscissa and the difference between enzymatic method and AAS as ordinate. (B) Relationship between automated enzymatic measurement method and atomic absorption spectrophotometry.

Metalloenzymes must have their cofactor, which is a metal ion, for their activity. With the removal of their cofactors, they lose their enzymatic activity and they are converted to apoenzyme. If they regain the metal element, they produce enzymatic activity again. They show selectivity, specificity and sensitivity for metal elements. These characteristics made their usage possible in trace element determination. Among metalloenzymes CA provides a big advantage for analytical sensitivity because it has a high turn over number for its substrates. For example, its turnover number for bicarbonate is 400,000 kcat (s⫺1). Sera have high endogenous CA activity and also a large amount of zinc is bound to albumin. These obstacles are the main difficulties in using CA the determination of serum zinc concentration. Indeed, only a few studies have been reported in the literature. Kobayashi et al. [4] measured zinc concentration of water and fruit juices using (apo)carbonic anhydrase. In their study, three percent amount of CA could not converted to apoCA, therefore a high baseline absorbance appeared in the assay. We achieved a 100% removal of zinc cofactor from the enzyme. Demir et al. [5] used CA to measure zinc concentration of heated serum samples. Before the assay, they heated

O. Erel, S. Avci / Clinical Biochemistry 35 (2002) 41– 47

serum samples to inactivate the CA in serum and to release zinc from proteins. Donangelo and Chang [7] incubated apoalkaline phosphatase with human serum and rat plasma without pretreatment. Thus, they were able to estimate total zinc content only for the dynamic fraction of serum (plasma). On the other hand, more than one zinc atom is required to reactivate alkaline phosphatase and aminopeptidase enzymes from the zinc free state. The relationship between zinc content and recovery of activity is complicated, and not linear [4]. Usage of CA is preferable to alkaline phosphatase and aminopeptidase for zinc determination. Serum proteins must be precipitated before enzymatic zinc assay. In the previous study, Demir et al. denaturated serum proteins by physical methods. They heated serum samples in a boiling bath for 1 h in capped tubes to inactivate endogenous CA and to release zinc from carrier protein albumin and other protein molecules. The heat denaturation process is not appropriate and not practical for routine usage in clinical laboratories. On the other hand, the heat denaturation method leads to bad analytical performance characteristics, such as unacceptable precision values. For these reasons we aimed to precipitate serum proteins using an easy and practical chemical method. We tested almost all chemical protein precipitation methods to precipitate serum proteins. Among the tested precipitants TCA solution was the best appropriate precipitant solution because the negative effects on the assay could be completely neutralized and removed. The negative effects of other precipitants on the assay were not able to be completely removed. On the other hand, we determined that the best appropriate buffer, which is able to completely neutralize the negative effects of TCA in supernatant and permits maximal catalytic activity of CA, is TRIS buffer at 0.5 mol/L pH 8.3. In accordance with the reported studies, which use pnitropenyl acetate as substrate and p-nitrophenol as the product, the changes in the absorbance were followed at 348 nm [4 –5]. However, we determined that the absorbance value at 400 nm is higher than that at 348 nm, as seen in Fig. 1. Thus the changes in absorbance were followed at a wavelength of 400 nm instead of 348 nm. We also showed that the maximal absorbance of p-nitrophenol was obtained in the range between pH 7.7 and pH 9.1 (Fig. 2). The interference effect of other metals create the main problem in colorimetric zinc measurement methods [8 –9]. Enzymatic determination has big advantages because it shows specificity for its cofactor. We supplemented Fe⫹2, Cu⫹2, Mn⫹2 and Pb⫹2 metal ions to apoCA but no reactivation of the enzyme was observed. It has been well documented that Fe⫹2, Cu⫹2, Mn⫹2, Cd⫹2, Pb⫹2, Ba⫹2, Be⫹2, Hg⫹2, Ca⫹2 and Mg⫹2 cations have no interference effect on the re-activity of CA, but only Co⫹2 cation at a concentration equal to that of zinc ion increases the enzymatic

47

activity by 50%. However, Co⫹2 activation requires a several hour lag phase [4,10]. So, the interference effect of Co⫹2 cation can be disregarded. Using this semiautomated enzymatic measurement method, we measured serum zinc concentrations of healthy subjects and patients. Serum zinc concentration was found to be higher in patients with vivax malaria than in healthy subjects as seen in Table 2. Serum zinc concentrations decrease in infectious disease as in our other group’s results, which belong to the patients with cutaneous leishmaniasis. The decrease in serum zinc concentration is a defense mechanism of the host against infection, like the changes in iron metabolism [11]. Slight hemolyses occurs in malaria because of plasmodium infection. Erythrocytes have ten times the zinc concentration of plasma. The increased serum zinc concentration of our patients with malaria may be due to hemolyses and may also reflect the degree of hemolyses. In summary, semiautomated enzymatic zinc measurement method is accurate, sensitive and is not affected by other metals. There was no significant difference between the results of our method and the reference AAS method. There was also a most important correlation and excellent agreement between the results of both methods. We conclude that the enzymatic measurement method is suitable for use with automated analyzers for zinc determination of serum. On the other hand, we thought that apocarbonic anhydrase preparate will have a commercial product potential.

References [1] Milne DB. Trace Elements. In: Burtis CA, Ashwood ER, editors. Tietz Textbook of Clinical Chemistry (3rd ed.). Philadelphia: W.B. Saunders Company, 1999. p. 1037– 41. [2] Taylor RP, James T. Enzymatic determination of sodium and chloride in sweat. Clin Biochem 1996;29:33–7. [3] Fujita T, Hamasaki H, Furukata C, Nonobe M. New enzymatic assay of iron in serum. Clin Chem 1994;40(5):763–7. [4] Kobayashi K, Fujiwara K, Haraguchi H, Fuwa K. Determination of ultratrace zinc by enzymatic activity of carbonic anhydrase. Bull Chem Soc Jpn 1979;52(7):1932– 6. [5] Demir N, Kufrevioglu OI, Keha EE, Bakan E. An enzymatic method for zinc determination in serum. BioFactors 1993;4(2):129 –32. [6] Hunt JB, Rhee M, Storm CB. A rapid, and convenient preparation of apocarbonic anhydrase. Anal Biochem 1977;79:614 –7. [7] Donangelo CM, Chang GW. An enzymatic assay for available zinc in plasma and serum. Clin Chim Acta 1981;113(2):201– 6. [8] Homsher R, Zak B. Spectrophotometric investigation of sensitive complexing agents for the determination of zinc in serum. Clin Chem 1985;31(8):1310 –3. [9] Makino T. A sensitive, direct colorimetric assay of serum zinc using nitro-PAPS, and microwell plates. Clin Chim Acta 1991;197:209 –20. [10] Lindskog S, Malmstrom BG. Metal binding and catalytic activity in bovine carbonic anhydrase. J Biol Chem 1962;237(4):1129 –37. [11] Galloway P, McMillan D, Sattar N. Effect of the inflammatory response on trace element and vitamin status. Ann Clin Biochem 2000;37:289 –97.