Chem. Res. Toxicol. 1993,6, 310-312
310
The Importance of Hydrophobicity in the Mutagenicity of Methanesulfonic Acid Esters with Salmonella typhimurium TA100 Asim Kumar Debnath and Corwin Hansch* D e p a r t m e n t of Chemistry, P o m o n a College, Claremont, California 91711 Received N o v e m b e r 3, 1992
The analysis of the mutagenic activity of a series of 15methanesulfonate esters with Salmonella typhimurium TAlOO shows that it can be correlated with the following equation: log TAlOO = 1.10 log P + 0.73 log MMI - 2.53, where MMI is the relative rate of reaction of the esters with N-methyl-2-mercaptoimidazole and P is the octanol/water partition coefficient. The results are compared with a number of other structure-activity studies on mutagenesis by avariety of types of chemicals.
Introduction We have been engaged in the formulation of quantitative structure-activity relationships (QSAR)' for the mutagenic activity of a variety of organic compounds in various bacterial systems (1-7). Despite the interest in correlating mutagenicity and chemical structure, relatively little work outside of our laboratory has been concerned with investigating the importance of the hydrophobicity of chemicals in their mutagenic potency. Frierson et al. have reviewed the early work on the QSAR of mutagenesis (8). In our approach to the SAR of chemicals acting on various biochemical and biological systems we have made the basic assumption that the three most important properties of a set of congeners producing a standard response in a fixed time interval are hydrophobic, electronic, and steric. Other properties such as hydrogen bonding or dipole moments are less important or are covered by the other parameters. The following generalized approach has been found to be successful. log A = f(1og P) + f(e1ect.) + f(steric) + const (1) Equation 1suggests that activity ( A ) is some function of hydrophobic (P),electronic, and steric properties of a set of bioactive compounds (9). The puzzle is to find out how these properties affect activity. This approach to biological QSAR has achieved significant success during the past 30 years (10). The results from this approach applied to mutagenicity are summarized for the role of hydrophobicity in Table I. Here are listed the coefficients (h)with the log P terms for a variety of different compounds causing mutagenicity in a variety of different bacterial test systems. Some classes of compounds require microsomal (S9) activation and some do not, yet all contain a hydrophobic term and except for one example the value of h is near 1. In this report, we consider the mutagenicity of a set of methanesulfonate esters which do not require metabolic activation (S9).
Methodology For this study we have used reported mutagenic activity for a set of methanesulfonic acid esters acting on Salmonella 1 Abbreviations: QSAR (quantitative structure-activity relationship); NBP (4-(pnitrobemyl)pyridine);MMI (N-methyl-2-mercapbimidazole); rev (revertants).
0893-228x/93/2706-0310$04.00/0
typhimurium TA100. Eder and K u t t reported mutagenic activity in terms of rev/Hmol (11). We have converted their values to revinmol for easier comparison with our other studies. Experimental log P values for the esters are not available; hence log P was calculated using the CLOGP program version 3.54 (12). T h e method has been shown to be reliable for a family of closely related compounds (3-413). For the present data set, where there are relatively large substituents around the ester linkage, shielding of t h e 0 may elevate experimental values somewhat over calculated log P (13). We expect t h a t if measured values were used, our correlations would be a bit better. Eder and Kutt (11)used four experimental models to evaluate the stereoelectronic factors for t h e interaction of t h e esters with nucleophiles: N-methylmercaptoimidazole (MMI),I-(p-nitrobenzy1)pyridine (NBP), trifluoroacetate, and water. It was hoped t h a t one or more of these would model the nucleophilic reaction of the esters with DNA. We have used the MMI rate constants and log Pvalues (Table 11)t o derive QSAR 3. T h e other models did not yield satisfactory results. Table I11 contains results on three compounds whose MMI values were not precisely defined.
Results QSAR 2 and 3 have been derived using the data in Table 11. log TAlOO = 0.27(*0.38) log MMI - 1.15(*0.54) (2) n= 15, r = 0.386, s = 0.815, F1,13 = 2.27 log TAlOO = 0.73(&0.24) log MMI + l.lO(f0.37) log P - 2.53(*0.54) (3) n = 15, r = 0.900, s = 0.402, F,,,z = 41.59, F,,,, = 25.5 In these expressions, TAlOO is the mutagenicity in S. t y p h i m u r i u m TAlOO expressed as revlnmol, P is the octanollwater partition coefficient, MMI is the relative rate of reaction of the sulfonate esters with MMI, n represents the number of data points, r is the correlation coefficient, s is the standard deviation, F is the test value of the significance of each additional term starting with log MMI, and the figures in parentheses are for construction of the 95% confidence limits. In this stepwise development of eq 3, it is seen that log MMI gives the best single-variable equation although eq 2 is not very significant according to the F test. Equation 3 is highly significant, showing the important role of relative hydrophobicity. In the derivation of eqs 2 and 3, two data points 0 1993 American Chemical Society
Hydrophobicity in Mutagenesis
Chem. Res. Toxicol., Vol. 6, No. 3, 1993 311
Table I. Coefficients with Log P ( h )for Various Sets of Compounds Acting in Various Bacterial Mutagenicity Tests no. of compounds 188 117 88 67 21 15 21 12 40 20 30 15
type of compounds aromatic and heteroaromatic nitro compounds aromatic and heteroaromatic nitro compounds aromatic and heteroaromatic amines (+S9) aromatic and heteroaromatic amines (+S9) XCsHdN=NN(R)(CH3) (+S9) aromatic nitro compounds quinolines (+S9) XCeH&H2N(CH3)N=O (+S9) nitrofurans nitrofurans N-methyl-2-aminobenzimidazoles (+S9) CHBSOZOR
test TA98 TAlOO TA98 TAlOO TA92 E . colib TAlOO TA1535 E . coli TAlOO TA98 TAlOO
h 0.65 1.10 1.08 0.92 0.97 1.07 1.14 0.92 1.00 1.15 0.96 1.10
ref 3 3 7 7 2 6 5 3 a a a this study
A. K. Debnath and C. Hansch, unpublished results. Escherichia coli. Table 11. Parameters Used To Derive Equations 2 and 3 for the Mutagenicity of CH3S020R loa (revhmol) R obsd" calcdb A l o n P lonMMIc methyl -0.84 -0.88 0.04 -0.48 3.00 3-methyl-2-butyl -1.05 -0.96 -0.09 1.29 0.20 2-phenylpropyl -0.46 -0.55 0.09 1.87 -0.12 0.11 3-pentyl -0.43 -0.54 1.42 0.59 1-phenyl-2-butyl -0.18 -0.06 -0.12 2.31 -0.10 1-phenyl-2-propyl -0.23 -0.36 0.13 1.78 0.28 cyclopropylmethyl -0.38 -0.53 0.15 0.50 2.00 1,3-dichloro-2-propyl -2.32 -2.50 0.18 0.49 -0.70 0.41 0.16 0.25 0.10 3.56 allyl 2-methylpropyl -1.58 -1.32 -0.26 0.98 0.17 2-phenylethyl -0.52 -0.19 -0.33 1.47 0.99 cyclohexyl -0.83 -1.32 0.49 1.55 -0.70 ethyl -2.55 -1.92 -0.63 0.05 0.76 2-chloroethyl -1.25 -1.93 0.68 0.11 0.65 n-propyl -2.05 -1.36 -0.69 0.58 0.73 2-butyld 0.14 -0.97 1.11 0.89 0.80 2-propyld 0.42 -2.13 2.55 0.36 0.00 ~
" From Eder and Kutt (11). Calculated via eq 3. Relative reactivity withN-methylmercaptoimidazole. These data points were not used in the derivation of eqs 2 and 3. Table 111. Parameters Used To Calculate MMI for Three CH3S020R Analogs Whose MMI Values Were Determined To Be
