1991, VOL. 21,


6, 751-754

Separation of hydroxyquinolines by high-performance liquid chromatography I. NAGIEL-OSTASZEWSKI, M. T. VAVREK and J. H. WEISBURGER American Health Foundation, Valhalla, New York 10595-1599, USA

Xenobiotica Downloaded from informahealthcare.com by RMIT University on 09/05/14 For personal use only.

Received I2 April 1990; accepted 3 January 1991 1. Hydroxyquinolines are important chemicals for pharmaceutical and cosmetic use. This report describes an improved method for separating these chemicals, by utilizing a high-performance liquid chromatography technique.

2. 8-Hydroxyquinoline is an excellent chelating agent. Previous separation procedures failed because of this property. An important requirement for the success of the method described is to provide a metal-free environment for the separation of hydrowyquinolines by replacing metal tubing with Polyplex and polyether-ether ketone (PEEK) tubing and utilizing a metal-free Hamilton PRP-1 column.

Introduction 8-Hydroxyquinoline has important uses as an antimicrobial drug. It is a bacteriostatic additive in cosmetics and shampoos and has a number of other industrial applications. Because it is an excellent chelating agent, it is also of interest in analytical chemistry. It is important to have sensitive specific analytical techniques for 8-hydroxyquinoline and related compounds to determine their purity when used as intermediates in chemical reactions, or for their identification and quantitative determination in formulations into which they have been incorporated for a specific application. Also, studies on the metabolism and mode of action of quinoline, hydroxyquinolines and related chemicals and drugs require specific techniques for their separation and hence identification through instrumental analysis. Wojtowicz (1984) discovered, and we confirmed, that standard reverse phase liquid chromatographic techniques were not suitable for 8-hydroxyquinoline and similar chemicals with chelating properties. T h e problem appeared to be an irreversible binding of this chemical to the metal components of the chromatographic columns. This investigator therefore selected a phenyl-silica column together with the mobile phase containing NiCl,, and satisfactory resolution was reported (Wojtowicz 1984). In connection with other work involving chelating metabolites of the structurally related carcinogen containing a quinoline ring, similar problems were encountered (Luks et al. 1989). For this reason, we developed a novel method, eliminating the metal tubing and using a metal-free column, which lead to a successful separation and analysis of 8-hydroxyquinoline and related compounds.

Materials and methods Chemicals Quinoline, 8-hydroxyquinoline (8-01I-quinoline) and 5-OH-quinoline were purchased from Aldrich Chemical Company (Milwaukee, WI, USA). 8-OH-quinoline glucuronide was purchased from Sigma Chemical Company (St Louis, MO, USA). All other chemicals and solvents were of analytical or higher grade. 0049-8254/91 $3.00 6 1991 Taylor & Francis Ltd.


I. Nagiel-Ostaszewski et al.

Xenobiotica Downloaded from informahealthcare.com by RMIT University on 09/05/14 For personal use only.

Equipment and instrumentation Spectrophotometric analysis was done on a Beckman Model D U - 7 spectrophotometer (Beckman Instruments Inc., Scientific Instruments Division, Fullerton, CA, USA). H.p.1.c. analysis was performed using two Waters Model 510 pumps (Millipore, Waters Division, Milford, MA, USA), Model 680 automated gradient controller, Model 810 Maxima data control system, a Waters Model 440 UV detector with fixed wavelength at 254nm, and a Rheodyne Model 7125 injector, with Polyplex tubing, 1 / 1 6 i n ( 1 5 9 m m) overall diameter x 0.010 in (0.254 mm) int. diam. on the low pressure side, and P E E K tubing, 1/16 in overall dism. x 0.010 in int. diam. on the high pressure side (Alltech Associates, Inc. Deerfield, IL,, USA). T h e column was custom packed Hamilton PRP-1 material in a metal-free P E E K casing (4.6 m m x 250mm; Alltech Associates, Inc.). Spectrnphntometry Stock solution of individual standards of 8-OH quinoline glucuronide, 5-OH-quinoline, quinoline and 8-OH-quinoline were prepared fresh each day in methanol to a concentration of 1 mg/ml. Prior to spectral identification of each compound, an aliquot of each stock solution was diluted to obtain 1.2.5 and 10pg/ml final concentration. Dilutions were made in a mixture of 0.01 M acetate buffer p H 5.0 (A), and acetonitrile (B), in a proportion equivalent to that corresponding to the proportion of mobile phase at the elution time for each compound. For 8-OH-quinoline glucuronide, the proportion was 55%A to 45%B, for 5-OH-quinoline SO%A to 50%B, for quinoline 37-63%B, for 8-OH-quinoline 32%A to 68%B, respectively. Spectral idcntification of each diluted compound was carried out by determining its absorbance from 200 to 400nm. Quantitative analysis on the standard curves was carried out by reading the absorbance of each compound in the concentration range of l-lOpg/ml at 240 nm, 2 5 4 n m and 3 0 0 n m wavelengths for 8-OH-quinoline and 5-OH-quinoline, 254 n m for quinoline and 240 n m for 8-OH-quinoline. These wavelengths were peak absorbance points for each chemical. H.p.1.c. At the beginning of each day, the column was perfused with 0.01% E D T A solution for 30 min, following which initial condition for gradient chromatography were established. T h e column was allowed to equilibrate for an additional 30min prior to running standards. T h e conditions for separation of standards were as follows: initial 95%A to 5%B, where A=0.01 M acetate buffer, p H 5.0, and I3 = acetonitrile, was run isocratically from t =0 to t = 3 min; from t = 3 min to t = 30 min, a Waters gradient curve no. 5 was used, ending at a ratio of 10%A to 90XB. T o determine the retention time of each compound, each standard was run individually. 'I'he eluting fraction corresponding to each peak was collected and identified for its integrity in the 2 0 W 0 0 n m spectral region, as compared to the chemical compound not injected into the h.p.1.c. column. T h e mixture of compounds was prepared from stock solutions, and analysed by h.p.1.c. T h e concentration of each compound in a recovered fraction was determined by reading the absorbance of each peak fraction at 240, 254 and 300 nm wavelengths. Quantitative recovery of each compound was determined from a standard curve corresponding to each compound.

Results Quinoline, 5-OH-quinoline, 8-OH-quinoline and 8-OH-quinoline glucuronide were separately injected into the system in order to determine the mobility of each of the chemicals. Under these conditions the reproducible elution time, in the order cited above, was 17.6, 13.1, 20.9 and 10.8 min, respectively. Injection of the mixture of the four chemicals showed them to be clearly separated. Importantly, the chelating agent 8-OH-quinoline eluted quantitatively (figure 1). In the initial runs with a new PRP-1 column the U.V. spectrum of 8-OHquinoline recovered from the column showed a displacement to longer wavelength with a main peak appearing at 257 nm and a longer wavelength peak appearing at 385. In contrast, pure 8-OH-quinoline not injected into h.p.1.c. has a sharp peak at 243 mn and the longer wavelength peak at 3 18 mn. T h e entire system was, therefore, rinsed with a 0.01% EDTA solution, under the assumption that the reason for the bathochromic shift was metal ion binding effects. Upon this preparation of the column, followed by injection of 8-OH-quinoline, the standard retention time for 8-OH-quinoline, 20.9 min, remained unchanged but the U.V. spectrum of the compound was now identical to that of the pure compound not injected into h.p.1.c. IJnder these conditions, injection of 4-20 p g gave recoveries approximating 100% (table 1).


Improved h.p.1.c. of hydroxyquinolines


Xenobiotica Downloaded from informahealthcare.com by RMIT University on 09/05/14 For personal use only.


-0 P


0 Figure 1 .


10 15 20 Minutes


Separation of quinolines by h.p.1.c.

8-Hydroxyquinoline glucuronide (peak at 10.8 min), 15.0 p g ; 5-hydroxyquinoline (1 3.1 min), 6.25 p g ; quinoline (17,6min), 12.5 pg; and 8-hydroxyquinoline ( 2 0 9 min), 17.5 pg.

Table 1 .

Percentage recoveries of individual components of mixtures from h.p.1.c. Chemical Quinoline 5-OH-quinoline 8-OH-quinoline 8-OH-quinoline glucuronide

% Recoveries (x SD)? 106f5 104+5 110+ 1 99+8

t Mean & SD of four individual determinations. Discussion 8-Hydroxyquinoline and related compounds are important pharmaceuticals and intermediates in industry. T h e drug has displayed antiproliferation, antineoplastic effects, in part through an action enhancing redifferentiation (Nordenberg et al. 1990). 8-Hydroxyquinoline is also a useful analytical agent with powerful chelating properties. It is these chelating properties that thus far prevented separation and isolation of this compound from mixtures through the application of h.p.1.c. One study utilized nickel chloride as part of the elution system and permitted recovery of this chelating agent, presumably as the nickel compound (Wojtowicz 1984). However, no method existed to recover the pure compound. Application of a


Improved h.p.1.c. of hydroxyquinolines

polymer column PRP-1 in a totally metal-free system has overcome the problem of the isolation of this and related compounds. As we noted, there may be trace amounts of metal ions a s part of contaminants in the mobile phase or possibly on the column. It is therefore desirable, and indeed essential, as part of the initial rinsing and preparation procedures of the analytical system to include a wash with E D T A in order to fully avoid that problem. Thus, we have applied a new type of adsorbent material, together with a fully non-metallic tubing assembly that now permits the routine separation and isolation of 8-hydroxyquinoline and related quinoline compounds by efficient, rapid and quantitative h.p.1.c. methods.

Xenobiotica Downloaded from informahealthcare.com by RMIT University on 09/05/14 For personal use only.

Acknowledgements This research was supported in part by a PHS-NIH grant CA-42381 from the National Cancer Institute. We are indebted to Ms Julie Howard and Ms Nadine Kjellberg for editorial assistance.

References IJJKS, H. J., SPRATT, T. E., VAVREK, M. T., ROLAND, S. F., and WEISBURGER, J. H., 1989, Identification of sulfate and glucuronic acid conjugates of the 5-hydroxy derivative as major metabolites of 2amino-3-methylimidazo[4,5-f]quinolinein rats. Cancer Res., 47, 4407441 1. NORDENBERG, J., NOVOCRODSKY, A,, BEERY, E., PATIA,M., WASSERMAN, L., and WARSHAWSKY, A,, 1990, Antiproliferative effects and phenotypic alterations induced by 8-hydroxyquinoline in melanoma cell lines. Eur. J . Comer, 26,905-907. Wojwwicz, E. J., 1984, Reverse-phase high-performance liquid chromatographic determination of halogenated 8-hydroxyquinoline compounds in pharmaceuticals and bulk drugs. J . Pharm. Sci., 73, 1430-1433.

Separation of hydroxyquinolines by high-performance liquid chromatography.

1. Hydroxyquinolines are important chemicals for pharmaceutical and cosmetic use. This report describes an improved method for separating these chemic...
226KB Sizes 0 Downloads 0 Views