Thrombosis Research. Vol. 74, No. 4. pp. 419-42&1994 CaatiPht 0 1994 Elscvier sm ud &tin the USA. AU rights resewed 0049-3848/94 $6.00 + .oo
INTERACTION OF MITOXANTRONE WITH HEPARIN AND ITS APPLICATION TO THE QUANTITATION OF HEPARIN Jean-Yves Follezou Hdpital de la Pitie Salpetriere, Service de Radiotherapie, 47, bd de I’Hopital 75634 Paris Cedex 13, France.
(Received 1 November 1993 by Editor C. Sofia; revised/accepted 28 February 1994)
ABSTRACT The interaction of heparin of high molecular weight with mitoxantrone, an anthraquinone derivative, had been studied by spectrophotometry. Heparin links the mitoxantrone and the binding sites are propably the anionic groups of the mucopolysaccharide since the linking is displaced by Na+. This interaction of mitoxantrone with heparin could provides a sensitive and simple method for quantitation of heparin in non proteic medium and could appear of clinical relevance in some circumstances.
Mitoxantrone (1,4-dihydroxy-5,8-bis[2-[(2-hydroxyethyl)-amino] ethyl] amino-g, lo-anthracenedione) is an anthraquinone derivative synthesized in an effort to identify a potent DNA-reactive cytotoxic agent devoid of cardiotoxicity. It is now used in the treatment of breast cancer (1) and lymphomas (2). The purpose of this paper was to study the interaction of mitoxantrone with heparins, sulfate-mucopolysaccharides present in many tissues. Mitoxantrone is a positively charged molecule, which must easily interact with heparin, like other cationic dyes (3,4). The interest of this study is at least triple: 1) it can provide useful informations concerning the pharmacology and some aspects of the mechanism of the antitumoral action of mitoxantrone; Key words: heparin, mitoxantrone
interaction indications (if they after extravasation for the quantitation
QUANTITATION OF HEPARIN
Vol. 74, No. 4
of these two drugs may lead to suggest therapeutic counter might be used simultaneously) or indications (injection of heparin or over dosage of mitoxantrone); 3) it can provide a simple method of heparin.
MATERIAL AND METHODS Mitoxanthrone dihydrochloride was purchased from Lederle, Gosport (U.K.). Dry heparin was from Roche (Neuilly, France). It is an heparin sodium salt extracted from porcine intestinal mucosa, (15,000 MW) with the following composition (expressed as per cent of the dry matter): S 11 .O, C 23.0, N 2.2, H 2.7, Na 11.5. Reactions were performed in bisdistilled water and in phosphate buffered saline (PBS), O.O2M, pH 7,0. The measurements were performed spectrophotometrically at 660 nm (E= 18,300 in H20 and E = 10,800 in PBS, for pure mitoxantrone at 1O-4- 1O-5 M). The anti-Xa activity of heparin, expressed as international units (IU), was determined by the amidolytic method using a chromogenic substrate (5) with Stachrom Heparin reagents (purchased from Stago, Paris, France). It is identical to the Roche Lab. titration with a precision of 2 per cent (measured on 10 vials titred to 100 IU each). The unit-weight-molar equivalences are: 165,000 IU = 1 g and 1 IU = 4.04 lo-1OM.
RESULTS The spectrum of absorbance of mitoxantrone in the visible (in H20) is characterized by two peaks (respectively at 607 and 660 nm) and a shoulder at 565 nm(Fig.1). The interaction between heparin and mitoxantrone produced modifications of this spectrum consisting in an hypochromicity and a small translation to 658 nm of the 660 nm peak (Fig. 1). In the presence of mitoxantrone, there is a linear relationship between heparin concentration and the decrease of absorbance (Ar). However, according to the mitoxantrone concentration, the relation between the variation of Ar (at 660 nm) and the concentration of heparin varies. As seen on the figure 2, at low doses of mitoxantrone (0.18 mM/I), there is a relationship between the decrease of Ar and the concentration of heparin in a range of 0 to 1 IU/ml of heparin. When the doses of mitoxantrone are increasing, the linear relationship is extended (to 5 IU/ml of heparin for a concentration of mitoxantrone of 60 mM/I). Similar results are obtained when the discriminations are performed just after the contact of reagents or 3 to 6 hours latter. Thus, in this medium, it is easy to quantify relatively small amounts of heparin by means of its interaction with mitoxantrone. If the complex heparin-mitoxantrone follows the law of Beer-Lambert we have (for an optical course of 1 cm): Ar = EMCU + EMHCb (1) where EM is the molar absorbance of pure mitoxantrone, &MH that of the complex and Cu and Cb are respectively the concentrations of unbound and bound mitoxantrone.
Vol. 74, No. 4
FIG.1 Absorption spectra of mitoxantrone in Hz0 (45 mM/I) (A) and of heparin-mitoxantrone complexes for 0.25, 0.50 and 1.O IU/ml of heparin (6, C and D).
w i o,o
FIG.2 Absorbance (at 660 nm) of the heparin-mitoxantrone complexes (Ar) as a function of the concentration of heparin, for 4 concentrations of mitoxantrone (expressed in mM/I): 60 (O--O), 45 (W-H), 25 (O--O) and 18 (O--O).
Vol. 74, No. 4
Co being the total concentration of mitoxantrone involved in the reaction: Co=Cu+Cb(2) From equations 1 and 2: Ar = EMCO - Cb (EM-EMH) (3) The absorbance of the total concentration of mitoxantrone is Ao = Q&O, then: Ao-Ar = Cb (EM-EMH) (4) If we call
AA the difference
mitoxantrone and the mixture (AA= Ao - Ar) and A& the difference between the molar absorbance of these two solutions (A&=EM-&MH), the concentration of bound mitoxantrone is: Cb = AA/A& (5) The Figure 2 shows that Ar becomes constant when the concentration of heparin increases. We could admit that, at this condition, all the mitoxantrone molecules bound to heparin and, therefore, Cb = Co. From (5):
Amax being at the plateau phase. On the Figure 3, we have plotted Amax against Co. The result is clearly a straight line and its slope is A&. The calculation give A& = 13,500 and then, &MH= 4,800.
A Amax I,0 -
FIG.3 Difference between the absorbance of the pure solution of mitoxantrone and the heparin-mitoxantrone complexes at the plateau phases of the Figure 2 (Amax) as a function of the concentration of pure mitoxantrone (all the reactions are performed in
Vol. 74, No. 4
cB M X 181/t 50-
FIG.4 Binding curves representing the concentration of bound mitoxantrone (Cb) as a function of the concentration of heparin ([HI) for 4 concentrations of mitoxantrone (expressed in mM/I): 60 (O--O), 45 (U--m), 25 (D-0) and 18 (O--O).
The Figure 4 shows the variation of Cb as a function of the concentration of heparin. The linear initial portion of these curves correspond to the complete saturation of available binding sites of heparin. Te slope of these straight lines decreases when the concentration of mitoxantrone increases. Thus, the calculated number of binding sites per molecule of heparin, in H20, falls down from 44.1 (ti.1) for 18 mM/I of mitoxantrone to 25.9 (kO.9) for 60 mM/I of mitoxantrone. According to the sulphur content of high molecular weight heparin used in this study, the mean number of sulfhydril groups per molecule is approximately 165, corresponding to 55 disaccharides, each bearing 3 anionic sites. The mean number of molecules of mitoxantrone linked per disaccharide falls down from 0.77 to 0.50 in the range of concentration of mitoxantrone used. The results are similar when the reactions were performed in PBS, but all the measured values were lower than in H20 (from 0.69 to 0.20). The Figure 5 shows the effect of NaCl on the binding of mitoxantrone to heparin. lt appears that the increasing of the concentration of NaCl in the medium produces a decreasing of the binding capacity of heparin. The final pH had been measured in each medium in the range of concentrations of mitoxantrone tested, for 1 IU/ml of heparin. It vanes from 2.94 to 3.65 in H20, and it is constant at 6.90 in PBS.
Vol. 74, No. 4
0 VW (M/I)
FIG.5 Absorbance of the heparin-mitoxantrone concentration of NaCl ([Na]) in the medium
(Ar) as a function
DISCUSSION Mitoxantrone binds to heparin in distilled water and in PBS and the reaction is easily quantifiable by spectrophotometry. Since the mitoxantrone is positively charged, the linking likely arises on the anionic sulfhydril groups of heparin. The displacement of the linking by NaCl confirms this hypothesis, the Na+ makes a competition the mitoxantrone on the anionic sites. Similarly, the lower amount of mitoxantrone fixed by heparin in PBS may be explained in pan by the Na+ content of this buffer. Otherwise, the reaction of the ions phosphate with mitoxantrone may inhibited the linking of the drug to heparin and explained the hypochromicity observed in PBS with a pure solution of the drug (e is 10,800 in PBS vs 18,300 in H20). This fact must be put together with the inhibition by the ions phosphate of some effects of adriamycin (6) and should be.of interest in the comprehension of some aspects of the resistance of tumors to these classes of drugs since amounts of phosphate are liberated during the tumor lysis induced by chemotherapy. At low concentrations of heparin, in the presence of an excess of mitoxantrone, all the available sites of heparin might be linked. Then, the slope of the initial portion of the linking curves (Fig. 4) should have been identical whatever the concentration of
Vol. 74, No. 4
QUANTITATION OF HEPARIN
mitoxantrone greater than its equivalent in potentially free sites of heparin. It is that was observed by Menozzi and Arcamone (7) with adriamycin, in PBS, O.O16M, pH 7.2. However our results show that, for the same concentration of heparin, the number of bound sites decreased when the concentration of mitoxantrone is increasing. This phenomenon was unrelated to the pH because it was observed at low and variable pH (in H20) as well as at physiological and constant pH (in PBS). It is easily explained by the fact that the commercially available solutions of mitoxantrone and heparin used in the experiments contain ions Na+ which compete the mitoxantrone on anionic sites of heparin with a higher affinity. At low concentrations of heparin we can admit that all its anionic sites are linked by a cation (Na or mitoxantrone). Thus, when the concentration of cations increases, the number of Na+ linked grows and that of molecules of MTX decreases, proportionally to the ratio of the constants of affinity of these two ligands. The interaction of mitoxantrone with heparin provides a non specific but simple and sensitive method for quantitation of heparin in non-proteic medium. This process is faster and cheaper than the classical dosage of heparin based on the inhibition of factor X activation. Such a method may be interesting in industrial applications.lt is also convenient for the quantitation of low molecular weight heparins (data not shown), may be easily miniaturized (in microplates) and will be of clinical interest if it is adjustable to plasma or serum (under investigation)
1. STUART-HARRIS, R.C., BOZEK, T., PAVLIDIS, N.A., SMITH I.E. Mitoxantrone: an new active agent in the treatment of advanced breast cancer. Cancer Chemother. Pharmacol., 72, 1-4, 1984. 2. NISSEN, N.I., HANSEN, S.W. High activity of daily-schedule mitoxantrone in newly diagnosed low-grade non-Hodgkin lymphoma: a 5-year follow-up. Semin. Oncol., 17 (suppl. 7U), 20-23, 1990. 3. YOUNG, M.D., PHILLIPS, G.O., BALAZS, E.A. Polyanions and their complexes. I. Thermodynamic studies of heparin-azur A complexes in solution. Biochem. Biophys. Acta 147, 374-381, 1967. 4. STONE, A.L., BRADLEY, D.F. Aggregation of cationic dyes on acid polysaccharides. I. Spectophotometric titration with acridine orange and other metachromatic dyes. Biochem. Biophys. Acta, 748, 172-l 92, 1967. 5. TEIEN, A.N., LIE, M. Evaluation of amidolytic Heparin assay method : Increased sensitivity by adding purified Antithrombin III. Thromb. Res., IO, 399-410, 1977. 6. FOLLEZOU, J.Y., BIZON, M., GICQUEL, J. Potentialisation par le serum de I’hemolyse induite in vitro par I’adriamycine, chez I’Homme. C. R. Acad. SC. Paris, 296, 25-28, 1983. 7. MENOZZI, M., ARCAMONE, F. Binding of adriamycin to sulfated mucopolysaccharides. Biochem. Biophys. Res. Commun.,80, 313-318, 1978.