Spectrofluorimetric determination of guanethidine sulphate, guanoxan sulphate and amiloride hydrochloride in tablets and in biological fluids using 9,10-phenanthraquinone

0039-9140/92 SW0 + 0.00

Tafonfa, Vol. 39, No. IO, pp. 13694375, 1992

Printed in Great Britain. All rights reserved

Copyright 6 1992 Pergamon Press Ltd

SPECTROFLUORIMETRIC DETERMINATION OF GUANETHIDINE SULPHATE, GUANOXAN SULPHATE AND AMILORIDE HYDROCHLORIDE IN TABLETS AND IN BIOLOGICAL FLUIDS USING 9,10-PHENANTHRAQUINONE MOHAMED H. AFIDEL-HAY, SI~~REENM. GALAL, MONA M. BEDAIR, AZZA A. GAZY and ABDEL AZIZ M. WAHBI Faculty of Pharmacy, University of Alexandria, Pharmaceutical Analytical Chemistry Department, Alexandria, 21521, Egypt (Received 27 May 1991. Revised 9 March 1992. Accepted 23 March 1992)

Summary-A highly sensitive spectrofluorimetric method for the determination of some drugs of the monosubstituted guanidine derivatives in laboratory made tablets, in spiked human serum and in urine samples is presented. The method is based on the reaction of guanethidine sulphate (I), guanoxan sulphate (II) and amiloride hydrochloride (III) with 9,10-phenanthraquinone (IV) to give highly fluorescent derivatives. The linearity ranges were found to be 0.06-0.96 pg/ml for (I) and (II) and 0.04428 pg/ml for (III), with relative standard deviation less than 2%. Mean percentage recoveries for tablets were found to be 99.9 f 1.3, 100.5 f 1.1 and 100.0 f 1.6 for I, II and III, respectively. For I and III the results are highly correlated with the B.P. methods. Using the synchronous fluorimetry, differentiation between I and II was possible. Chloroform, dichloromethane and ethyl acetate have been used to extract I, II and III, respectively from serum and urine at basic pH, followed by applying the proposed fluorimetric method. Percentage recoveries were found to be 95.7-102.2%. The limit of detection is 0.04 rg/ml for I and II and 0.02 pg/ml for III.

The fluorescence reaction of 9,1 O-phenanthraquinone as a very sensitive test for arginine and arginine-containing peptides was first introduced by Itano and Yamada.’ The mechanism of the fluorescence reaction of benzylguanidine as monosubstituted guanidines with 9, lo-phenanthraquinone was investigated.* Biguanides reacted similarly giving highly fluorescent products.3 Guanethidine and guanoxan sulphates have been widely used as antihypertensive agents. They exhibit marked neuron blocking effects in the treatment of moderate and severe hypertension. Several methods have been reported for their determination, including spectrophotometry,4 oxidimetry,s complexometry,6 colorimetry,7-‘0 fluorimetry,“-‘3 and highchromatography.14’6 performance liquid Amiloride hydrochloride is used as potassiumsparing diuretic in edema, renal and liver disorders. It interferes with the process of cationic exchange in the distal tubule by blocking the resorption of sodium ion and the secretion of potassium ion. It has been

determined calorimetrically using sodium methylbenzothiazolin hydranitroprusside,” zone”’ and N,N-dimethyl-phenylenediamine.‘9 Second derivative spectrophotometric determination of amiloride hydrochloride in a two component mixture was described.20 Methylguanidine and guanidine in physiological fluids have been determined by high-performance liquid chromatography with fluorescence detection*’ and determination** using their reaction with phenanthrene-9,10-dione. Guanethidine and guanoxan have been determined in biological fluids by gas chromatography after selective extraction procedures.23 The official methods” for the assay of the above-selected drugs are, (i) a calorimetric procedure using sodium nitroprusside and potassium hexacyanoferrate for guanethidine sulphate and (ii) a spectrophotometric measurement at 363 nm for amiloride hydrochloride using tributylorthophosphate as a solvent. The present work deals with the determination of guanethidine sulphate, guanoxan sulphate and amiloride hydrochloride using the

1369

MOHAMED H. ABDEL-HAY ef al.

1370

condensation reaction of the guanidino group with 9,10-phenanthraquinone with the purpose of increasing sensitivity and selectivity. In urine and serum these drugs were first extracted according to the GC methodz3 followed by applying the proposed fluorimetric method. EXPERIMENTAL

Apparatus

All fluorimetric measurements were performed on a Perkin-Elmer Model 650-10s fluorescence spectrophotometer equipped with 1.O x 1.O cm quartz cells, a 150-W xenon lamp, excitation and emission grating monochromators and a Perkin-Elmer Model 56 recorder. The spectra were measured with the normal instrument gain and minimum ratio mode. The sensitivity ranges used were 0.1 and 0.3 according to the concentration level. Reagents and samples

All reagents were of AnalaR grade, and were checked before use for the presence of fluorescent contaminants. 9,10-Phenanthraquinone reagent, 2 pgg/ml in dimethylformamide, Gold label, free of anthraquinone, Aldrich Chemicals Co., Inc. The solution was kept in a refrigerator. Fresh solutions were prepared every week. Authentic samples of guanethidine sulphate (Ciba-Geigy, Basle, Switzerland), guanoxan sulphate (Pfizer, Kent, England) and amiloride hydrochloride (Kahira Pharmaceuticals & Chemical Industries Co., Cairo, Egypt) were kindly donated by the manufacturer and were used without further purification. Biological samples. Serum and urine specimens are collected from adult healthy volunteers who are not under medical treatment. The samples were first analysed by the proposed method to ensure that there is no interference from endogenous compounds. Fluorimetric procedure Preparation of calibration graphs.

Prepare a solution containing 50 mg of either guanethidine sulphate, guanoxan sulphate or amiloride hydrochloride in 50 ml of distilled water. Place in a 25-ml calibrated flask, an appropriate volume of the drug solution (6-24 ~1) for guanethidine sulphate and guanoxan sulphate and (2-8 ~1) for amiloride hydrochloride so that the final content is

between 6-24 pg for I and II and 2-8 pg for III. Add 2.5 ml of 9,10-phenanthraquinone reagent (2 pg/ml) and 0.25 ml of 1M sodium hydroxide solution. Allow the mixture to stand at room temperature for 45 min, then add 0.25 ml of concentrated hydrochloric acid. Make up to volume with distilled water and measure the fluorescence intensity. Correct the observed fluorescence by subtracting the fluorescence intensity measured using the same procedure on a reagent blank. Make the fluorimetric measurements using an excitation wavelength of 310 nm for I and II and 360 nm for III and an emission wavelength of 395 nm for I and II and 420 nm for III (Figs. 1 and 2). of Guanethidine sulphate, Determination Guanoxan sulphate and Amiloride hydrochloride in laboratory prepared tablets. Weigh and pow-

der 20 tablets. Weigh accurately a quantity of the powdered tablets equivalent to about 50 mg of each of the three drugs. Extract with three lo-ml portions of distilled water and filter (using Whatman No. 1) the combined extracts into a 50-ml calibrated flask. Complete to volume with distilled water. Apply the above procedure, using 20 ~1 of the final tablet solution for I and II and 5 ~1 of the final tablet solution for III beginning at “Add 2.5 ml of 9, lo-phenanthraquinone . . .“.

I

(F)

(al

(b)

so -

z

r

4o

r

3

.c g 3o-

: 8 3 LL20 -

IO -

I 260

420

340

500

X(nm)

Fig. 1. (a) Excitation and (b) emission spectra of 9,10phenanthraquinone derivatives of (-) guanethidine (0.60 pg/ml) and (---) guanoxan (0.36 &ml). (A,, = 310 nm, & = 395 nm).

Spectrofluorimetric

determination of guanidine derivatives

(F) (a)

(b)

300 A (nm)

Fig. 2. (a) Excitation and (b) emission spectra of 9,10phenanthraquinone derivatives of amiloride (0.16 p&l) (A, = 36Onm, &,, = 420 nm).

Determination of the selected guanidino drugs in urine and serum. Adjust the pH of urine to 10.0 and of serum to 7.0. Place 5 ml of serum or urine in 50-ml stoppered shaking tube. Add a 6.0-ml aliquot of I and II and 2.0 ml of III. Add 20 ml of toluene in the case of urine and 20 ml diethyl ether in the case of serum, shake the tube for 10 min and centrifuge. Reject the organic layer. Add 0.5 ml of 50% sodium hydroxide, extract for 20 min using 30 ml of each of chloroform for I, dichloromethane for II and ethyl acetate for III. Centrifuge and remove the upper aqueous layer. Transfer the organic phase to another shaking tube containing 10 ml of 0.W hydrochloric acid, shake the tube for 10 min and centrifuge. Transfer the aqueous layer to a loo-ml standard flask and complete to the mark with distilled water. Pipette 0.1-0.2 ml of this solution into a 25-ml standard flask, follow exactly as in “Calibration graphs above beginning at “Add 2.5 ml of 9,10-phenanthraquinone . . .“. Calculate the concentration from a calibration graph prepared similarly using aqueous solutions in place of the urine and serum.

to give an intermediate compound which can easily undergo hydrolysis in acid medium to give a fluorescent product and aldehyde.2 Similarly, the reaction between guanethidine (I) and guanoxan (II) with IV can be explained to follow scheme I in the present work. The excitation and emission wavelengths of the fluorescent product (VI) were found to occur at 310 and 395 nm, respectivelyi (Fig. 1). These findings differ slightly from the previously reported wavelengths using simple guanidines.’ The reaction of 9,10-phenanthraquinone (IV) with amiloride (III) in alkaline medium is suggested to give another intermediate substance VII containing amide linkage. The latter (VII) cannot undergo hydrolysis in acid medium at room temperature. The excitation (360 nm) and emission (420 nm) wavelengths of the product (Fig. 2) were found to be completely different from those of compound VI. Moreover, the reaction product with amiloride was found to be 10 times more sensitive than VI (Table 1). Accordingly, we suggest that a further condensation of VII with another molecule of IV would occur to give a highly fluorescent compound VIII as shown in scheme 2. Factors affecting the jluorimetric procedure Influence of reagent concentration and reaction time. The effect of 9,10-phenanthraquinone concentration and reaction time selected for the recommended method was studied by carrying out the fluorimetric procedure using 0.72 pg/ml

(P)

R==

(PI)

(I)

-CHi_N 3

RESULTS AND DISCUSSION R=

9,10-Phenanthraquinone (IV) reacts with guanidino compounds in an alkaline medium

1371

l;n

tn, (Scheme 1)

MOHAMEDH. ALIDEL-HAY et al.

1372

Table 1. Analytical data for guanethidine sulphate, guanoxan sulphate and amiloride hydrochloride using the proposed fluorimetric method Wavelength, am

Linear? regression

Concentration range rralml

Sensitivity setting

a

b

Corr. Coeff. r

RSD*, %

Compound

1.,r

L

Guanethidine sulphate

310

395

0.24-O.% 0.06-0.36

0.1 0.3

-0.36 -3.83

90.27 255.40

0.9997 0.9998

1.30 0.96

Guanoxan sulphate

310

395

0.24-0.84 0.06-0.36

0.1 0.3

0.58 -5.18

102.19 310.00

0.9995 0.9999

1.06 1.33

Amiloride hydrochloride

360

420

0.12-0.28 0.04-0.24

0.1 0.3

- 1.76 -5.39

373.00 1079.50

0.9992 0.9990

1.70 1.60

*Relative standard deviation (six replicate determinations). TRelative digital units, where (a) is the intercept and (b) is the slope.

of I, 0.60 pg/ml of II and 0.12 pg/ml of III. Changing the concentration of 9, lo-phenanthraquinone over the range 0.04-0.32 pg/ml showed that the optimum concentration to be used is 0.20 pg/ml. Maximum fluorescence intensity for all three was developed within 45 min and remained stable for 60 min at room temperature. Effect of temperature, sodium hydroxide and hydrochloric acid concentrations. An experiment

has been designed to study the effect of carrying out the reaction at 20” (room temperature), 40” and 60”. It was found that as the temperature increases, the fluorescence decreases with loss of sensitivity. Accordingly, the reaction was maintained at room temperature. The effect of sodium hydroxide and hydrochloric acid concentrations on the fluorescence intensity has been carried out separately. It was found that for maximum fluorescence intensity associated with a minimum blank reading, 1M sodium hydroxide and IOM hydrochloric acid solutions should be used. Sensitivity, precision, limits of detection and

accuracy of the procedure. Calibration graphs were constructed from five points covering the concentration ranges 0.24-0.96 pg/ml for I and II and 0.04-0.24 pg/ml for III using a sensitivity setting of 0.1. For the analysis in urine and serum, the concentration ranges for the linearities were prepared to be 0.06-0.36 pg/ml for I and II and 0.02-0.10 pg/ml for III using sensitivity setting at 0.3. Regression analysis indicated a linear relationship between fluorescence intensity and concentration (Table 1). The correlation coefficients were found to range between 0.9990 and 0.9997. Six replicate determinations at different concentration levels were carried out to test the precision of the method. The relative standard deviations were found to be less than 2%, indicating excellent reproducibility of the method. The limits of detection, where the fluorescence of the sample reading was double the blank reading, were found to be 0.04 pg/ml for I and II and 0.02 pg/ml for III. The accuracy of the procedure was tested by assaying the selected drugs in laboratory prepared tablets (prepared to contain 10 mg, 10 mg

(Scheme 2)

Spectrofluorimetric

determination of guanidine derivatives

1373

Table 2. Determination of guanethidine sulphate, guanoxan sulphate and amiloride hydrochloride in prepared tablet using the proposed fluorimetric method and 05cial methods Percentage of expected*

Drug Guanethidine sulphate

Fluorimetric method

Official method

100.20 f 1.18

100.30 + 0.81 tt=0.17 F$ = 2.12

Guanoxan sulphate

100.30 f 1.24

100.20 * 0.75 tt = 0.17 Fj = 2.75

Amiloride hvdrochloride

100.30 f 0.96

100.10 f 1.28

I

rt=0.31 Ff = 1.78 *Average of six determinations It standard deviation. tTheoretica1 value : t = 2.23 at the 95% confidence level. #Theoretical value : F = 5.05 at the 95% confidence level.

and 5 mg per tablet, respectively) and in spiked human urine and serum. The mean percentage recoveries were found to be around 100.00% (mean of j-replicates) for tablets (Table 2) and between 95.7-102.2% (mean of j-replicates) for urine and serum (Tables 3 and 4). The official methods (B.P. 1988) have been applied to assay the laboratory prepared tablets and the results were found to be in good agreement with the proposed fluorimetric method. The calculated t- and F-values did not exceed the theoretical values, indicating that there is no significant difference between the mean recoveries obtained by either method (Table 2). Stability indicating assay. The fluorimetric method has been tested as a stability indicating assay of the studied guanidines. The

effect of heating at 40, 50, 60 and 70” with 4M sodium hydroxide solution with subsequent fluorimetric determination has been carried out. Compounds I and II did not show any degradation indicating good stability. Rate of degradation of amiloride (III) in 4M sodium hydroxide solution at 40-70” using the fluorimetric method was found to be first order (Fig. 3). A typical Arrhenius plot was obtained for III with a degradation rate constant 0.000195 min-’ and tlj2 of 59.26 hr. Degradation of III has been reported” to occur in the guanidine side chain, This is in good agreement with the fact that the proposed fluorimetric method can be considered a stability indicating assay. The official B.P. method has been proven to be non-specific when applied to III in 4M

Table 3. Precision and relative recovery in the determination of guanethidine sulphate, guanoxan sulphate and amiloride hydrochloride in spiked human urine Guanethidine sulphate Added

teglml 0.06 0.12 0.18 0.24 0.36 0.48 0.60 0.72 0.84 0.96

% Recovery*

Guanoxan sulphate Added

% Recovery*

(c.v.%)*-

&ml

(C.V.%)i

95.7 (1.46) 99.7 (0.31) 100.7 (0.20) 101.3 (0.09) 99.3 (0.10) 100.4 (0.17) 101.2 (0.16) 99.9 (0.09) 100.9 (0.08) 100.9 (0.10)

0.12 0.18 0.24 0.36 0.48 0.60 0.72 0.84 -

*Mean the five experiments.

0.06

99.3 (0.63) 99.9 (0.26) 100.1 (0.11) 100.2 (0.12) 97.0 (0.05) 99.3 (0.12) 101.4 (0.07) 98.7 (0.12) 99.6 (0.11) --

Amiloride hydrochloride Added

Pgglml 0.02 0.04 0.06 0.08 0.10 0.12 0.16 0.20 0.24 0.28

% Recovery*

(C.V.%)i 96.3 (0.82) 100.7 (0.19) 101.7 (0.09) 101.0 (0.07) 101.9 (0.08) 100.6 (0.19) 97.6 (0.10) 102.2 (0.07) 99.9 (0.06) 99.2 (0.05)

1374

MOW

H.

ABDEL-HAY et

al.

Table 4. Precision and relative recovery in the determination of guanethidine sulphate, guanoxan sulphate and amiloride hydrochloride in spiked human plasma Guanethidine sulphate Added &?glml 0.06 0.12 0.18 0.24 0.36 0.48 0.60 0.72 0.84 0.96

% Recovery* (c.v.%)* 95.1 97.8 98.2 98.6 98.0 97.9 97.8 98.4 98.6 98.4

(1.06) (0.17) (0.13) (0.09) (0.10) (0.14) (0.18) (0.14) (0.12) (0.10)

Guanoxan sulphate % Recovery* (c.v.)*

Added &ml 8::

94.4 97.1 98.1 98.9 98.4 97.9 98.9 98.2 98.9 -

0.18 0.24 0.36 0.48 8:: 0.84 -

(0.66) (0.18) (0.17) (0.08) (0.05) (0.19) (0.09) (0.08) (0.07) -

Amiloride hydrochloride Added pgglml 0.02 0.04 0.06 0.08 0.10 0.12 0.16 0.20 0.24 0.28

% Recovery* (c.v.%)* 92.4 97.0 98.4 98.6 97.2 98.1 98.2 99.3 98.0 98.2

(0.68) (0.15) (0.15) (0.13) (0.09) (0.24) (0.10) (0.10) (0.06) (0.08)

*Mean of five experiments.

sodium hydroxide solution solution of 1 mg %w/v of degradation was found to analytical wavelength used. Differentiation between

at 40-70”. Thus a III after complete read 0.406 at the guanethidine

and

1.6

60

120

160

Time (mm)

Fig. 3. Apparent first order degradation of amiloride hydrochloride in 4M sodium hydroxide solution at different temperatures. (C, = initial concentration, 0.16 pg/ml, C, = remaining concentration in pg/ml after time, I).

guanoxan has been carried out using synchronous fluorimetry. 26*27Thus, by scanning the synchronous fluorescence of the reaction products at a starting excitation wavelength 220 nm with A1 = 40 nm (i.e., emission wavelength at 260 nm), II was found to exhibit a Gaussian band at 280, 320 nm (Fig. 4), whereas I did not show such a band. In conclusion, the proposed spectrofluorimettic method, being simple, accurate, precise and highly sensitive is suitable for determination of the investigated drugs in dosage forms and in different biological fluids. In particular, the method is much simpler in technique than the GC method previously reported for biological fluids.23 Furthermore the method can be used for the determination of amiloride hydrochloride in the presence of its degradation products, an advantage over the official method, and for the differentiation between guanethidine sulphate and guanoxan sulphate utilizing the synchronous spectrofluorimetry. Acknowledgemenrs-The authors wish to thank Ciba-Geigy Co., for providing guanethidine sulphate, Plixer Co. for providing guanoxan sulphate and Kahira Co. for providing amiloride hydrochloride as pure authentic samples. The authors also thank the Alexander von Humboldt foundation in Germany for the donation of the spectrofluorimeter.

REFERENCES

Wavelength

(nm)

Fig. 4. Synchronous fluorescence spectra of 9,10-phenanthraquinone derivatives of 0.48 pg/ml guanethidine sulphate (-) and 0.48 pg/ml guanoxan sulphate (---).

H. A. Itano and S. Yamada, Anal. Blochem., 1972, 48, 483. S. Tanabe and T. Sakaguchi, Chem. Pharm. Bull., 1978. 2& 337. Idem, ibid., 1978, 26, 423. R. D. Kadyrova, L. T. Ikramov and E. T. Tegisbaev, Farmatsiya (Moscow), 1988, 37,44.

Spectrofluorimetric

determination of guanidine derivatives

5. I. Ganescu, G. Brinzar and C. Verhelyi, Chem. Anal. (Warsaw), 1984, 29, 549. 6. E. Zollner, Acta Pharm. Hung., 1978, 48, 76. 7. G. R. Rao and S. Raghuveer, Indian J. Pharm. Sci., 1980, 42, 141. 8. I. Ganescu, I. Papa and M. Preda, Pharmazie, 1985, 40,495. 9. S. Tanabe, T. Oya and T. Sakaguchi, Chem. Pharm. Bull., 1975, 23, 1657. 10. M. A. Arustamyan, L. E. Zel’tser, D. Kh. Yunusov and N. Suleimanova, Zh. Anafit. Khim., 1983, 38, 129. Il. M. Kai, T. Miura, K. Kohashi and Y. Ohkura, Chem. Pharm. Bull., 1981, B, I1 15. 12. M. A. Pamiak. G. Lange and T. Viswanatha, J. Biochem. Biophys. Methods, 1983, 7, 267. 13. S. Takeichi, T. Shinzo, Y. Hisako, M. Tomoko and S. Akemi, Yakugaku Zasshi, 1977, 97, 1053. 14. Y. Kobayashi, H. Kubo and T. Kinoshita, Anal. Sci., 1987, 3, 363. 15. Y. Hiraga and T. Kinoshita, J. Chromatogr., 1985, 342, 269.

1375

16. M. Kai, T. Miyazaki and Y. Ohkura, J. Chromatog., 1984, 311, 257. 17. J. Vachek, Cesk. Farm., 1985, 34, 226. 18. C. S. P. &try, T. N. V. Prasad, B. S. Sastry and E. V. Rao, Analysr, 1988, 113, 255. 19. C. S. P. &try, M. V. Suryanarayana and A. S. R. P. Tipimeni, Talanta, 1989, 36, 491. 20. M. Parissi-Poulou, V. Reizopoulou, M. Koupparis and P. Macheras, I~JI. J. Pharm., 1989, 51, 169. 21. Y. Yukio, S. Akira, M. Tadatomi, M. Kenji and 0. Kazuhiro, J. Chromatog., 1979, 162, 23. 22. S. Higashidate, T. Maekubo, M. Saito, M. Senda and T. Hoshino, Bunseki Kagaku, 1984, 33, 366. 23. J. H. Hengstmann, F. C. Falkner, J. Throk Watson and J. Oates, Anal. Chem., 1974, 46, 34. 24. The British Pharmacopoeia, HMSO, London,l980. 25. K. Florey, Analytical Profiles of Drug Substances, 1986, 15, 26. 26. J. B. F. Lloyd and I. W. Evett, Anal. Chem., 1977, 49, 1710. 27. T. Vo-Dinh, ibid., 1978, SO, 396.