Eliminate the Butariol

Document Sample
Eliminate the Butariol Powered By Docstoc
					CLIN. CHEM. 34/7,



Potassium Permanganate Used to Eliminate Interference with the Porter-Silber Reaction in the ButariolExtraction Method for Determinationof Urinary 17-Hydroxycorticosteroids
Takahiro Maeda,’ Hlroko Matsuzakl,1 and Talcamitsu Seklne We describe a simple, rapid, and less interference-susceptible method for using the Porter-Silber reaction for determination of urinary 17-hydroxycorticosteroids. The procedure is based on butanol extraction of the steroids after the urine sample is treated with potassium permanganate and sodium bisulfite. These treatments, together with an additional acidand alkali-washing of the extract, could eliminate most of the substances in urine that interfere with the Porter-Silber reactions. Values so obtained correlated well (r = 0.95) with those by Furuya’s method in which /3-glucuronidase is used. Our method may be useful and suitable for a screening test of adrenocortical and pituitary functions.
Addftlonal Keyphrases: urinary interferences . screening steroid hormones oxidizable

Reagents 3-Acetylphenythydrazine (APH) was obtained from Tokyo Kasei Industries Co., Tokyo, Japan. Cortisol (F), cortisone (E), tetrahydrocortisol (TifF), tetrahydrocortisone (THE), allo-THF, and /3-glucuronidase (EC, from bovine liver) were purchased from Sigma Chemical Co., St. Louis, MO. Amberlite XAD-2 (particle diameter, 0.3-1.0 mm) was from Organo Co., Tokyo, Japan. All other chemicals were of special analytical grade and were purchased from Wako Pure Chemical Industries Co., Osaka, Japan. Alkali washing solution. Dissolve 100 g of anhydrous Na2SO4 in 1 L of 0.1 mol/L NaOH. Acid washing solution. Dissolve 100 g of anhydrous Na2SO4 in 1 L of 0.05 moliL H2S04. Blank PS reagent. Dilute 660 mL of coned. H2S04 with 340 mL of distilled water. Color PS reagent. Dissolve 100 mg of APH in 100 mL of the blank PS reagent. The reagent is stable for one week at 0-4#{176}C. Procedures Isolation of crude

In 1950, Porter and Silber (1) reported a colorimetric phenylhydrazine method for urinary 17-hydroxycorticosteroids (17-OHCS).3 But the Porter-Silber (PS) reaction is susceptible to interference by urinary chromogens, drugs, and various metabolites in urine. Therefore, many methods have been proposed for removing these. Most of them, however, are more complicated and time consuming, e.g., Furuya’s method (2). On the other hand, in 1952, Reddy et al. (3) reported a method based on extraction of PS chromogens (PSC) with nbutanol, with no enzymatic hydrolysis. Therefore, this method was much simpler and faster as compared with those involving the enzyme. But many authors (4-6) have pointed out some definite disadvantages of the procedure in which hydrolysis of PSC conjugates is omitted. For example, the net absorbance from 17-OHCS was only a few percent of the total absorbance, resulting in less precision (4). Venning et al. (5) reported that n-butanol was a poor extraction solvent for PSC conjugates. Marks and Loftin (6) reported that the direct methods are less specific. These disadvantages come from the fact that n-butanol can extract interfering substances along with 17-OHCS from urine. We found that KMnO4 can oxidize these substances under weakly acidic conditions without appreciable loss of 17-OHCS.

PSC conjugates

from urine. The inter-

fering substances in about 4 L of a pooled urine were oxidized by adding KMnO4 until the color became purplish, and then NaHSO3 powder was added until the color disap-

Materials and Methods
Apparatus We used a Model 220 A spectrophotometer Tokyo, Japan).

‘Central Clinical Laboratories, Tokyo Metropolitan Police Hospital, 2-10-41 Fujiini, Chiyoda-ku, Tokyo 102, Japan. 2Department of Biochemistry, Juntendo University School of Medicine, Hongo, Bunkyo-ku, Tokyo ii3, Japan. Present address: Department of Biochemistry, Saitama Medical School, 38 Moroyama, Irurna-gun, Saitama 350-04, Japan. 3Nonstandard abbreviations: i7-OHCS, i7-hydroxycorticosteroids; PS, Porter-Silber, PSC, PS chromogen; APH, /3-acetylphenylhydrazine; E, cortisone; F, cortisol; THE, tetrahydrocortisone; THF, tetrahydrocortisol; THS, tetrahydro-ii-deoxycortisol. Received December 3, 1987; accepted March 16, 1988. 1392

peared. About 500 mL of Amberlite XAD-2 was added to the urine and filtered out. The resin remaining on the filter was washed twice with de-ionized water, and then PSC were eluted from the resin with two portions of methanol. After the methanol was removed under reduced pressure, the residue was dissolved in de-ionized water and washed three times with methylene chloride. The aqueous layer contained conjugated PSC. The PSC conjugates were assayed by highperformance liquid chromatography (HPLC) as described by Kawasaki et al. (7). Determination of urinary 17-OHCS. Exact determination requires a double-blank test to eliminate possible colored products originating from both APH solution (reagent blank) and urine (sample blank). Three milliliters of a urine sample, the standard solution, or water for the reagent blank was mixed with 1.0 mL of acetate buffer (2 moJJL, pH 5.0), 0.4 g of anhydrous Na2SO4 was dissolved in the mixture, and then 0.3 molfL KMnO4 was added until the mixture remained purplish (2-8 drops). About 1 g of NaHSO3 was then added. 17-OHCS in the solution were extracted with 3.0 inL of n-butanol, with shaking for at least 10 mix, and the tubes were centrifuged at 2500 rpm for 5 mm. Each supernatant butanol layer was washed with 2 mL of the alkali-washing solution and then with 1 mL of the acid-washing solution, each with 1 mm of shaking, and 1.5 mL of the blank PS reagent and of the color PS reagent were added for each 1.0 mL of the extract. The reaction mixtures were vigorously shaken for 1 mm, incubated at 60#{176}C20 mm, and then cooled in a water bath for for 5 mm. First, the absorbance obtained with the PS color reagent was determined at 410 nm vs the reagent blank solution, water. Second, the value was corrected against the

CLINICALCHEMISTRY, Vol. 34, No. 7, 1988

corresponding value obtained with the blank PS reagent (sample blank). The original method was modified in the following three ways for estimating its effect on eliminating interfering substances from urine: (a) without KMnO4 oxidation and NaHSO3 reduction (Reddy’s method), (b) without the oxidation alone (modified Reddy’s method), and (c) with an additional step of washing a 3.0-mL urine sample with 3 mL of ethyl acetate three times, shaking for 5 mm, before proceeding with our method (our modified method). Furuya’s method in which /3-glucuronidase is used (2) and its modified method without NaHSO3 treatment (modified Furuya’s method) were performed at the same time, for comparison. To compare these six methods under the same conditions, we added 1.5 mL of our PS reagent to 1.0 mL of n-butanol or ethanol that contained PSC extracted from 1.0 mL of urine by each method.


a a a


Stability of free 17-OHCS to KMnO4 oxidation. Compounds E, F, and their tetrahydro derivatives (THE, TifF, and allo-THF) were subjected to oxidation with 2 mmolfL KMnO4. As can be seen in Figure 1, E and F were immediately oxidized, while THE, THF, and allo-THF were quite stable at 20#{176}C considerably so at 60#{176}C. and Because urine usually contains much oxidizable material, we used a KMnO4 concentration as high as 0.3 mol/L. Under similar conditions, recovery tests of the free steroids (20 mg/L) in three different types of urine-normal, patient, and pregnancy-gave almost the same results as those shown in Figure 1 (data not shown). Elimination of PS-reaction-positive and -negative interfering compounds by KMnO4 oxidation. For examining the effect of KMnO4 on eliminating the interfering substances, we studied the n-butanol extracts obtained by the six methods (see Materials and Methods) spectrophotometrically in the wavelength range 350-700 nm. Figure 2 shows PS spectra of urines (A, B, and C). Reddy’s method shows thatA

Wave Isnthni.
Fig. 2. PSC spectra of three patients’ unnes containing interfenng substances reacting with sulfuric acid or the PS color reagents Administered drugs: patient A. splronolacne (75 mg/day) and tumeemide (40 mg/day); a proecillaridin (1.5 mg/day) and nlfedipine (30 mg/day); and C. halopeddol (10 mg/day).Methods: 1, Reddy’s; Z modified Reddy’s; and 3 our method. Spectra: reaction with the color PS reagent; b, with the blank PS reagent;and c net (a minus . The anowe show the peak of 17-OHCS









10 Time 0 (sin) *








Fig. 1. Effect of KMnO4on the free 17-OHCS
The compounds, 20 mg/i.., were incubated with 2 mmol/t. KMnO4at pH 5.0 (a) at 20#{176}C;at 60 ‘C. i, E; 0, F; A, THE; S, THF; #{149}, (b) alIo-THF

and C contain some compound(s) reacting strongly with the blank PS reagent to the same degree as with the color PS reagent (PS-reaction negative, a = b), while B contains some compound(s) showing a peak at 560 nm (PS-reaction positive, c in B-i). Addition of the NaHSO3 reduction procedure to Reddy’s method (modified Reddy’s) markedly improved the elimination of the PS-reaction-negative compound(s) in A, but was not so effective in C. The latter sample was made worse by the modification (c in C-2). As seen in the bottom row, however, our method combining KMnO4 oxidation and NaHSO3 reduction was successful in removing almost completely both PS-reaction-negative and -positive interfering substances from urine. Figure 3 shows PS spectra of a urine (D) from a patient being medicated with other drugs, which contained interfering substances negative to the PS reaction with a sharp absorption peak at 360 nm and positive ones with a peak at 560 nm. Both substances could not be removed by any of the methods except our modified method. Accuracy and precision. THF and crude PSC conjugates were determined by our method. The calibration curves were linear between 0 and 50 mg/L when expressed as TifF. The accuracy was determined by measuring analytical recovery of known amounts of PSC conjugates added to a 20 mg/L solution of THF and to two samples of urine. Recoveries ranged from 96.1 to 108.9%. The within-assay CV was 3.18% (mean ± SD = 12.52 ± 0.40 mg/L, n = 20) when about 10 mg of PS conjugates was added per liter to a pooled specimen of urine. The betweenCLINICALCHEMISTRY, Vol. 34, No. 7, 1988 1393


1.4 1.2 1.0 0.8 0.6 0.4

0 S.

0.2 0 1.2 l.0 0.8 0.6 0.4 0.2 0 -0.2

Wave Fig.

length (flfll)

3. PSC spectra


of a patient’s urine 0 containing interfering reacting to yield color with APH and sulfuric acid of the PS

Administered drugs: carbamazepine (400 mg/day), dihydroergotamine mesilate (3 mg/day) and azulene sodhim sulfonate (4.5 mg/day). Methods: 1, Reddy’s; modthed Reddy’s: 3, ours: 4, our modified; 5, modified Furuya’s; and 6, Furuya’s. a,b,c: see the legend for Figure 2

assay CV for the same specimen was 4.14% (mean ± SD = 12.86 ± 0.53 mg/L, n = 15 in one month). Reliability was assessed by comparing results for 40 urine samples with those obtained by Furuya’s method (2) and Reddy’s method. The values obtained by our and Furuya’s methods agreed well, with a correlation coefficient (r) of 0.9525. The r value was 0.70 for Reddy’s method (n = 40), or 0.91 when we excluded results for a sample that showed an extremely high value only by Reddy’s method. The means of the PSC values (mg/L) and the sample blank absorbances were, respectively, as follows: our method, 7.80 and 0.100; Reddy’s method, 9.36 and 0.539 (ii = 39); and Furuya’s method, 8.78 and 0.054. The dexamethasone suppression test and the metyrapone test were performed on a patient with suspected Cushing’s syndrome (Table 1). 17-OHCS in his urine were measured by the three methods and found to be in good agreement for both tests, although Reddy’s method gave higher values.

The reducing forms of 17-OHCS, their main forms in were quite stable against KMnO4 oxidation under our conditions, whether in their free or conjugated form (Figure 1). On the other hand, their non-reduced forms, labile to the oxidation (Figure 1), are rarely excreted in urine in more than trace amounts. In addition, results after dexamethasone suppression and metyrapone tests, as obtained by the three methods, agreed well with each other (Fable 1), indicating that the KMnO4-NaHSO3 treatment rarely converts dexamethasone to reactive substances and THS to inactive ones for the PS reaction. Therefore, there is practically no problem with our method, even though compounds E and F cannot be measured. We used KMnO4 as an oxidant in the hope of selectively eliminating urinary chromogens and interfering substances with no loss of 17-OHCS, in addition to the technical advantage that the end point of the oxidation reaction is visible. The KMnO4-NaHSO3 treatment was found to be excellent for eliminating interfering substances reacting with the color and blank PS reagents (Figure 2). We found an exceptional case, a patient being treated with carbamazepine, in which our procedures could remove only half of the interfering substances (the PS reaction was negative, with the absorption around 400-420 nm, D-3 in Figure 3). Values for 17-OHCS as determined by our method agreed well with those by Furuya’s method, but not by Reddy’s method. The discrepancy may come mainly from the latter’s high blank absorbance. We must point out that some samples contained a large amount of PS-reaction negativeand -positive interfering substances that could not be removed by Reddy’s method (see A-I and B-i in the upper row of Figure 2, respectively). Our method is much simpler and faster than the other methods with or without enzyme hydrolysis. Increased sensitivity of the method owing to minimized blank absorbance makes possible use of the procedure on a much smaller scale, say in a 10-mL tube, and analytical recovery of the conjugate of PSC is complete with a single extraction with n-butanol. These improvod procedures save much time and make it possible to omit overnight enzymatic hydrolysis; at the same time, they are more accurate and precise. Therefore, our method with KMnO4-NaHSO3 treatment and alkali-acid washing lends confidence to the butanol extraction method, countering criticisms of it (4-6), and makes it suitable for routine work.

We are indebted to Laboratory of Medical Science of Mitsubishiyuka Co., Ltd., Tokyo, for urine samples containing interfering substances and for the HPLC investigation. References 1. Porter CC, Silber RH. A quantative color reaction for cortisone

Table 1. 17-OHCS Measured after the Dexamethasone Suppresslon and Metyrapone Tests
Amounts of 17-OHCS, mg/day Dexamethasono Methods Control
lit (day)

Metyrapon. Control 4.0 4.3
3.7 let 2nd 3rd

Ours Reddy’s

5.4 6.4

1.7 1.9

Furuya’s 4.7 Control: the day before administrationof the drugs.

7.1 7.7 7.5

22.9 28.0

7.0 9.5




CLINICALCHEMISTRY, Vol. 34, No. 7, 1988

and related
1950;185:201-7. 2.


J Biol


Furuya E. The effect of sodium bisulfite on the removal of drugs and their metabolites interfering with the Porter-Silber reaction in the determination of urinary 17-hydroxycorticosteroids. Am J Clin Pathol 1975;64:315-26. 3. Reddy WJ, Jenkins D, Thorn GW. Estimation of 17-hydroxycorticosteroids in urine. Metabolism 1952;1:511-27. 4. Schopman W, Hiss LG, Vies J, Hinlzen DAVML. Some experiences with a modification of the method of Reddy, Jenkins and Thorn for quantitative determination of 17,21-dihydroxy-20-keto-

steroids in urine. Acta Endocrinol 1958;28:153-62. 5. Venning EH, Dyrenfurth I, Kazmin VE. Hydrolysis and extraction of corticoids and 17-ketosteroids from body fluids. Rec Prog Horm Res 1953;8:27-.50.
6. Marks hi, Leftin JH. A note of caution on the lack of specificity of the Porter-Silber reaction for 17,21-dihydroxy-20-kethsteroids. J Clin Endocrinol Metab 1954;14:1263-5. 7. Kawasaki T, Maeda M, Tsuji A. Determination of 17-hydroxycorticosteroids in urine by fluorescence high performance liquid chromatography using DNS-hydrazine as a pre-column labeling reagent. J Chromatogr 1982232:1-11.

CLINICALCHEMISTRY, Vol. 34, No. 7, 1988 1395

Shared By: