a Department of Chemistry, University of Loughborough, Loughborough, Leicestershire LE11 3TU, U.K.
b Department of Chemistry, Imperial College of Science, Technology and Medicine, London SW7 2AY, U.K.
Summary: Theoretical methods using the PM3/COSMO continuum solvation model have been used to model the trend in the reduction potentials of a series of indolequinone bioreductive anticancer agents.
Keywords: bioreductive/indolequinone/reduction potential
Introduction
Mitomycin C (MMC) 1, a clinically useful antitumour antibiotic, is the archetypical quinone bioreductive alkylating agent (Carter & Crooke, 1979; Franck & Tomasz, 1990; Remers & Dorr, 1988). Reductive activation of MMC, and related indolequinones such as EO9 2 (Oostveen & Speckamp, 1987), leads to the formation of highly electrophilic intermediates which are capable of alkylating DNA, and therefore, given the importance of the initial reduction step, it is not surprising that the reductive activation of quinones has been widely studied (Andrews et al., 1986; Driebergen et al., 1990; Driebergen et al., 1993; Driebergen et al., 1992; Maliepaard et al., 1993; Maliepaard et al., 1992; Rao et al., 1977; Wardman, 1990; Workman & Walton, 1990). We have recently described the synthesis of a series of indolequinones 3-13 based on the novel cyclopropamitosene ring system (Cotterill et al., 1994a). Preliminary biological results have shown that these compounds do indeed function as bioreductive cytotoxic agents, and that, in common with other mitosenes (Pan & Gonzalez, 1990), the toxicity is redox related (Cotterill et al., 1994b). However, since the cyclopropamitosenes are only available by fairly lengthy chemical synthesis, we were interested in developing a theoretical method (Mallik & Datta, 1994; Rzepa & Suñer, 1993) for the prediction of redox potentials of indolequinones in aqueous solutions to assist in the design of further analogues, and we now report the results of this study.
Methods
Calculations were performed on the cyclopropamitosenes 3-13, and the related indolequinones 14-18 together with their respective hydroquinones, as well as mitomycin C (MMC) 1 as a reference. Theoretical calculations were carried out at the restricted Hartree-Fock level (RHF) using the PM3 semi-empirical SCF-MO method, as implemented in the MOPAC 93 program. The solvation model used was the COSMO (conductor screening) model (Klamt & Shuurmann, 1993), where the dielectric screening energy and derivative terms for a solute charge distribution derived from an atom rather than cavity centred distributed multipole analysis involving dipole and quadrupole terms allow an efficient geometry optimisation. All structures were optimized using the keywords PRECISE, EF, GNORM=0.5, EPS=78.4 (relative permitivity) and NSPA=60 (surface segments per atom), followed by a vibrational analysis, calculation of thermodynamic properties and hence correction of ΔH(aq) to ΔG298(aq). For every structure, a prior molecular mechanics study of the side-chains conformations was carried out using the Macromodel program (V3.5a), with MM3 as the selected force field. The calculated two electron electrode potentials of the compounds 3-18 were obtained by using equation (1):
ΔG(tot)= ΔG[Q'(aq)] + ΔG[QH2(aq)] - ΔG[Q'H2(aq)] - ΔG[Q(aq)]
(1)
ΔG(tot)= - nF(EQ-EQ')
Results
The results of the calculations for the 16 indolequinones, together with the electrochemically measured one-electron reduction potential (Cotterill et al., 1994b), are shown in Tables I and II.
Table I. Measured and calculated reduction potentials of cyclopropamitosenes
R R' X Eredox /V ΔΔG kcal Eredox /V
vs. Fc mol-1 (PM3)
3 CH2OCONH2 Me aziridinyl -1.355 -20.81 -1.579
4 CH2OCONH2 H aziridinyl -1.360 -20.96 -1.575
5 CH2OCONH2 Me MeO -1.370 -18.63 -1.626
6 CH2OCONH2 H 2-methylaziridin -1.371 -19.86 -1.599
yl
7 CH2OCOCH3 H MeO -1.380 -19.51 -1.607
8 H H aziridinyl -1.384 -19.39 -1.609
9 CH2OCONH2 H MeO -1.395 -17.86 -1.643
10 H H 2-methylaziridin -1.401 -17.66 -1.647
yl
11 H H MeO -1.403 -17.16 -1.658
12 CH2OCONH2 H pyrrolidinyl -1.572 -15.28 -1.699
13 CH2OCONH2 Me pyrrolidinyl -1.588 -15.03 -1.704
Eredox (+/-0.010 V) calculated as (Epc + Epa)/2 from 100
mVs-1 cyclic voltammograms
Epc = cathodic peak potential
Epa = anodic peak potential
ΔΔG = ΔG(hydroquinone) - ΔG(quinone)
Eredox (PM3) using MMC 1 as a reference.
The electrode potential of MMC is -1.421 V (vs. ferrocene).
In this paper: 1 cal = 4.184 J; F= 96485 C mol-1
Table II. Measured and calculated reduction potentials of pyrrolo[1,2-a]indolequinones
R X Eredox /V ΔΔG kcal Eredox /V
vs. Fc mol-1 (PM3)
14 CH2OCONH2 aziridinyl -1.385 -20.49 -1.586
15 CH2OCOCH3 MeO -1.387 -18.44 -1.630
16 CH2OCONH2 2-methylaziridinyl -1.395 -19.48 -1.607
17 H aziridinyl -1.398 -17.61 -1.648
18 H MeO -1.412 -12.91 -1.750
Eredox (+/-0.010 V) calculated as (Epc + Epa)/2 from 100
mVs-1 cyclic voltammograms
Epc = cathodic peak potential
Epa = anodic peak potential
ΔΔG = ΔG(hydroquinone) - ΔG(quinone)
Eredox (PM3) using MMC 1 as a reference
The electrode potential of MMC is -1.421 V (vs. ferrocene).
In this paper: 1 cal = 4.184 J; F= 96485 C mol-1
Discussion
We initially applied our solvation model for the two-electron reduction to the 16 indolequinones to obtain relative electrode potentials, normalised to the value for MMC itself. This procedure had resulted in good correlations with known two electron potentials measured for a series of substituted benzoquinone derivatives in aqueous solution (Rzepa and Suñer, 1993). We found there was a reasonable correlation between the experimental one-electron potentials measured for these systems and the calculated two electron electrode potentials. Not surprisingly, the substituent effect on the two electron potential is calculated to be much larger than that for the measured one electron potential. More importantly, the trends observed for measured and calculated quantities show a good congruence. We also note that the significant conformational mobility of the side-chains, as well as the side chain hydrogen bonding present when R = CH2OCONH2 may play a role in moderating the properties of these systems. Our calculations were for a single minimum energy conformation, whereas the experimental results relate to the mixed Boltzmann population of different conformers. A further source of error may occur in the dispersion contributions to the total energy of solvation, which are not currently included in the COSMO treatment (Rzepa and Suñer, 1993). Thus differences in the size of the side chains may lead to additional errors.
In an effort to establish a better model for the reduction, we also investigated calculating the one electron reductions, which would result in open shell radical anion species. Unfortunately, the convergence behaviour of the COSMO/Unrestricted Hartree Fock algorithm is very significantly slower for open shell species than for closed shell species, and in some case convergence difficulties are experienced. This is associated with the very strict criteria required for geometry optimisation necessary to perform a FORCE calculation in this mode. Typically, a single system that does converge can take > 36 hours of CPU time on a modern workstation. Because of these practical difficulties, we were not able to evaluate the calculated one electron reduction model for the series of compounds, although in principle this could be done for smaller systems where the limitations of computer time are not so serious.
We nevertheless believe that two electron reduction models for quinone species do give reasonable correlations with those measured for bioreductive indolequinones, and although the method has some limitations, we believe it will be useful in the design of further analogues of these important anticancer agents.
Acknowledgements
We thank the Cancer Research Campaign for their generous support of our research programme (C.J.M.), and the DGICYT for a postdoctoral studentship (G.A.S.).
References
ANDREWS, P. A., PAN, S.-S. & BACHUR, N. R. (1986). Electrochemical reductive activation of mitomycin C. J. Am. Chem. Soc., 108, 4158-4166.
CARTER, S. K. & CROOKE, S. T. (1979). Mitomycin C; Current Status and New Developments . New York: Academic Press.
COTTERILL, A. S., HARTOPP, P., JONES, G. B., MOODY, C. J., NORTON, C. L., O'SULLIVAN, N. & SWANN, E. (1994a). Cyclopropamitosenes, Novel Bioreductive Anticancer Agents. Synthesis of 7-Methoxycyclopropamitosene and Related Indolequinones. Tetrahedron, 50, 7657-7674.
COTTERILL, A. S., MOODY, C. J., MORTIMER, R. J., NORTON, C. L., O'SULLIVAN, N., STEPHENS, M. A., STRADIOTTO, N. R., STRATFORD, I. J. & SWANN, E. (1994b). Cyclopropamitosenes, Novel Bioreductive Anticancer Agents. Synthesis, Electrochemistry and Biological Activity of 7-Substituted Cyclopropamitosenes and Related Indolequinones. Journal of Medicinal Chemistry, 37, 3834-3843.
DRIEBERGEN, R. J., HOLTHUIS, J. J. M., BLAUW, J. S., KELDER, S. J. P., VERBOOM, W., REINHOUDT, D. N. & LINDEN, W. E. V. D. (1990). Electrochemistry of potential bioreductive alkylating quinones. Part 2. Electrochemical properties of 2,5-bis(1-aziridinyl)- 3,6-bis(ethoxycarbonylamino)-1,4-benzoquinone and some model compounds. Anal. Chim. Acta, 234, 285-307.
DRIEBERGEN, R. J., MORET, E. E., JANSSEN, L. H. M., BEIJNEN, J. H., HOLTHUIS, J. J. M., KELDER, S. J. P., VERBOOM, W., REINHOUDT, D. N. & LELIEVELD, P. (1993). Electrochemistry of potentially bioreductive alkylating quinones. Part 4. Qualitative and quantitative structure-activity relationships of aziridinylquinones. Recl. Trav. Chim. Pays- Bas, 112, 174-185.
DRIEBERGEN, R. J., MORET, E. E., JANSSEN, L. M. H., BLAUW, J. S., HOLTHUIS, J. J. M., KELDER, S. J. P., VERBOOM, W., REINHOUDT, D. N. & LINDEN, W. E. V. D. (1992). Electrochemistry of potentially bioreductive alkylating quinones. Part 3. Quantitative structure-electrochemistry relationships of aziridinylquinones. Analytica Chimica Acta, 257, 257-273.
FRANCK, R. W. & TOMASZ, M. (1990). The chemistry of mitomycins. In D. E. V. Wilman (Eds.), Chemistry of Antitumor Agents (pp. 379-393). Glasgow: Blackie and Son Ltd.
KLAMT, A. & SHUURMANN, G. (1993) COSMO: A New Approach to Dielectric Screening in Solvents with Explicit Expressions for the Screening Energy and its Gradient. J. Chem. Soc., Perkin Trans. 2, 799.
MACROMODEL V3.5a. Copyright Columbia University, New York, USA, 1992.
MALIEPAARD, M., MOL, N. J. D., JANSSEN, L. H. M., HOOGVLIET, J. C., NEUT, W. V. D., VERBOOM, W. & REINHOUDT, D. N. (1993). Reductive Activation of Potential Antitumor Mitosene Compounds. Journal of Medicinal Chemistry, 36, 2091-2097.
MALIEPAARD, M., MOL, N. J. D., JANSSEN, L. M. H., NEUT, W. V. D., VERBOOM, W. & REINHOUDT, D. N. (1992). Role of lipophilicity in the in vitro antitumor activity of a series of new mitosene compounds. Anti-Cancer Drug Design, 7, 415-425.
MALLIK, B. & DATTA, S. N. (1994). Semi-empirical quantum mechanical treatment of the standard reduction potentials of quinone and plastoquinone in water. International Journal of Quantum Chemistry, 52, 629-649.
MOPAC 93.00: J. J. P. Stewart, Fujitsu Limited, Tokyo, Japan, 1993. Available from Quantum Chemistry Program Exchange, University of Indiana, Bloomington, USA.
OOSTVEEN, E. A. & SPECKAMP, W. N. (1987). Mitomycin analogs. I. Indolequinones as (potential) bis alkylating agents. Tetrahedron, 43, 255-262.
PAN, S.-S. & GONZALEZ, H. (1990). Mitomycin antibiotic reductive potential and related pharmacological activities. Mol. Pharmacol., 37, 966-970.
RAO, G. M., BEGLEITER, A., LOWN, J. W. & PLAMBECK, J. A. (1977). Electrochemical studies of antitumor antibiotics II. Polarographic and cyclic voltammetric studies of Mitomycin C. Journal of the Electrochemical Society, 124, 199-202.
REMERS, W. A. & DORR, R. T. (1988). Chemistry, biology, and therapeutics of the mitomycins. In S. W. Pelletier (Eds.), Alkaloids: Chemical and Biological Perspectives (pp. 1-74). New York: Wiley.
RZEPA, H. S. & SUÑER, G. A. (1993). Theoretical calculationsof benzoquinone redox potentials using the COSMO continuum solvation model. Journal of the Chemical Society, Chemical Communications, 1743-1744.
WARDMAN, P. (1990). Reductive activation of quinones: redox properties and thiol reactivity. Free Rad. Res. Commun., 8, 219-230.
WORKMAN, P. & WALTON, M. I. (1990). Enzyme-directed Bioreductive Drug Development. In G. E. Adams et al. (Eds.), Selective Activation of Drugs by Redox Processes (pp. 173-191). New York: Plenum Press.