following document is presented for informational purposes only!
Mutogenesis vol.6 no.2 pp.0000-0000, 1990
Interpretation of results with the 8-azaguanine resistance system in Salmonella typhimurium: no evidence for direct acting mutagenesis by 15-oxosteviol, a possible metabolite of steviol.
Emily Procinska, Bryn A. Bridges and James R. Hanson
School of Molecular Sciences and MRC Cell Mutation Unit, University of Sussex, Falmer, Brighton, UK
15-oxosteviol, postulated to be the mutagenic metabolite of steviol, was observed to be non-mutagenic in preliminary experiments using a number of different systems. Repetition of the original experiment in Salmonella TM677 failed to show any significant induction of 8-azaguanine resistant mutants by 15-oxosteviol even when the number of bacteria tested was greatly increased. Examination of the earlier positive result showed that it could not be justified from the data and revealed a commonly applied way of mishandling data obtained with the TM677 system.
Steviol (ent-13-hydroxykaur-16-on-19-oic acid) is the aglycono of stevioside, a sweet-tasting component of Stevia Rebaudiana used as a sugar substitute in Japan and other parts of the world. Stevioside is rapidly converted to steviol when incubated with gut microflora from rats (Wingard, 1980). Pezzuto et al. (1985) reported that steviol induced mutations to 8-azaguanine resistance in Salmonella typhinurium TM677 in the presence of induced rat liver S9 fraction. The metabolic pathways by which steviol might exert its mutagenicity have been further studied (Pezzuto et al., 1986) and it has been reported (Pezzuto, 1986) that 15-oxosteviol, a possible metabolite of steviol, is a direct acting mutagen in TM677. It could thus be the mutagenically active metabolite of steviol.
In the course of experiments designed to develop a non-mutagenic derivative of steviol we examined the mutagenicity of 15-oxosteviol in Escherichia Coll. 15-oxosteviol was prepared by Dr. Bras H. De Oliveira by hydroxylation of steviol with t-butylhydroperoxide and selenium dioxide to form ent-13, 15b-dihydroxykaur-16-en-19-ole acid. The latter was then oxidized with chromium trioxide in pyridine to give ent-13-hydroxy-15-oxokaur-16-en-19-ole acid (15-oxosteviol)(Avent et al., 1990). We used mutation to Trp+ in strain WP2 levrA and its pKM101-containing derivative CM891. This system has been shown to be comparable to the Ames Salmonella strains in its ability to detect mutagens and carcinogens (Bridges et al., 1981). We also used mutation to rifamplein resistance, a forward mutation system which, although not used routinely for screening, is widely used in research laboratories as a system generally responsive to a wide variety of mutagens. We also carried out experiments with the histidine-independence assay in Salmonella strains TA98 and TA100 using the treat and plate protocol. In a series of preliminary experiments we saw no evidence of mutagenicity in any of these systems.
We therefore turned to the Salmonella TM677 system used by Pezzuto et al. Which selects for mutants resistant to 8-azaguanine using methodology developed by Skopek et al. (1978a,b). This is a 'treat and plate' method in which expression of newly induced mutations is achieved by a limited but unknown amount of growth in the selective agar plates (Skopek et al., 1978a). A similar situation exists with the reverse mutation systems to histidine- or tryptophan-Independence commonly employed with Salmonella and E.coli strains. Evidence for positive mutagenicity can be accepted if there is a significant increase in the number of mutant colonies per plate.
It is common with the TM677 system to calculate the mutant fraction as:
M / V x D (1)
where M is the number of colonies (mutants) on selective agar plates, V is the number of colonies (viables) on unselective agar plates and D is the dilution factor between the number of bacteria plated on the two types of plate, assuming equal volumes are added to all plates. We contend, however, that this formula is invalid since a significant proportion of the number of mutant colonies arise spontaneously during growth on the selective plates. The number of these 'plate' mutants is determined only by amounts of growth permitted on the selective plate and not (over a wide range) by the number of bacteria initially plated. If one uses this invalid formula any toxic agent can be made to appear mutagenic. By reducing the number of viable bacteria but not reducing the number of plate mutants a spurious elevation in 'mutant fraction' can be obtained. In this way it can be shown that distilled water is 'mutagenic', as pointed out by Green and Muriel (1976).
The most valid way of calculating the induced mutant fraction requires a knowledge of the numbers of pre-existing mutants and of the numbers of 'plate' mutants (see, e.g. the formula presented by Sedgwick and Bridges, 1972). Where it is not possible to measure directly the numbers of pre-existing mutants, or where these can be assumed to be very low, then their contribution may be neglected and a valid though not totally accurate formula for the induced mutation frequency may be used (cf. Green and Muriel, 1976).
(Mt - Mg x D) / Vt
where Mt and Mg are the number of colonies (mutants) on selective agar plates for treated and untreated bacteria, respectively, Vt is the number of colonies arising from treated bacteria on unselective agar plates and D is the dilution factor between the number of treated bacteria plated on the two types of plate, assuming equal volumes are added to all plates. Using this formula there must be an increase in the absolute number of mutants per plate after treatment if the induced mutation frequency is to have a finite value.
In Pezzuto (1986) the data for 15-oxosteviol are given as mutant fractions calculated according to (1). In Table 1 we have recalculated these data as the number of mutanta per plate. We find there is no statistically significant dose-dependent increase in the number of mutants per plate as determined by a regression analysis with the GLIM statistical programme (Royal Statistical Society). The increase in 'mutant frequency' may therefore be attributed simply to a reduction in the number of viable bacteria plated.
The possibility of such spurious positive results arising as a consequence of toxicity seems to be inherent in the published protocol since the number of cells plated is of the order of 10 to the 6th power per plate, in our view too low. In Figure 1 we show the results of three experiments in which varying numbers of untreated bacteria were plated on 8-azaguanine selective plates. The number of mutants apperaing per plate is not proportional to the number of bacteria plated. Below -10 to the 6th power bacteria per plate the number of mutants per plate is more or less constant' these mutants are presumably 'plate' mutants that have arisen during residual growth on the plate. As the number of bacteria added to each plate is increased the number of mutants scored per plate rises, gradually at first, and then more steeply as mutants presumably pre-existing in the population are increasingly present among the number of cells plates.
Two lines have been fitted as examples assuming slightly different values for the number of 'plate' mutants and the number of pre-existing mutants (these would in any case be expected to vary from one experiment to another). In Figure 2 we have plotted 'mutant frequencies' from one of these experiments calculated according to (1). With this formula, the lower the number of viable cells plated, the higher the 'mutant fraction', as discussed above. It is probable that the pre-existing mutant frequency in this experiment was -3 x 10 to the -5th power and this should have led to a plateau at this value at the higher plating densities (see dotted curve). This was not observed, a fact which may indicate that with inocula above -2 x 10 to the 7th power bacteria per plate not all preexisting mutants are able to form visible colonies. The departure of the high density points from the dotted curve in figure 1 is another reflection of this effect.
Two conclusions are apparent from this. Firstly, the relation between the number of cells plated and the number of mutants subsequently appearing on the plate is complex and (as discussed above) does not permit the valid calculation of mutant fractions. Secondly, it would be unwise to accept any result with this system as indicating mutagenicity that did not show an absolute increase in the number of mutant colonies per plate so that one could use formula (2).
It is generally considered desirable that a sensitive mutation assay should be able to detect an increase in mutant frequency equal to the spontaneous frequency. If that is accepted, then it is necessary to treat and to plate at least 5 x 10 to the 5th power cells to stand a reasonable chance of reaching this sensitivity.
We have carried out four such experiments with 15-oxosteviol generally using the methodology of Skopek et al. (1978a) but plating many more cells per plate. The results presented in Table II show no increase in the number of mutants per plate and thus no evidence for any direct acting mutagenicity of 15-oxosteviol.
In conclusion, a previous report that 15-oxosteviol induces mutations to 8-azaguanine resistance in S.typhimurium TM677 in the absence os S9 fraction is shown to be a consequence of mishandling of data. Repetition of the experiment did not reveal any evidence of mutagenicity in TM677 nor was any evidence found of induction of mutations in any other system. 15-oxosteviol is therefore unlikely to be the active metabolite responsible for the mutagenicity of steviol. Since the evidence for the mutagenicity of steviol itself rests on results obtained with the TM677 system we have also reexamined those data as reported in Pezzuto et al. (1986). In contrast to the results with 15-oxosteviol we calculate that there is a real increase in the number of 8-azaguanine resistant colonies per plate as the concentration of steviol is increased. However, as pointed out by Pezzuto (1986), the mutagenicity of steviol increases as the amount added to the solution increases up to 10 mg/ml whereas the concentration remaining in solution saturates at ~ 2 mg/ml. Such a result suggests that it might be worth exploring the possibility that the mutagenicity of steviol (as in the experiments of Pezzuto et al., 1986) is due to an impurity.
Bridges, B.A. et al. (1981) Summary report on the performance of bacterial mutation assays. In de Serres, F.J. and Ashby, J. (eds), Evaluation of Short-Term Tests for Carcinogens. Report of the International Cottaborative Program.
Pezzuto J. (1986) Chemisty, metabolism and biological activity of steviol (ent-13-hyroxykaur-16-en-19-oic acid), the aglycone of stevioside. In attar-ur-Rabman and Le Quesne. P.W. (eds), New Trends in Natural Products Chemistry 1986. Elsevler, Amsterdam, pp. 371-385.
Pezzuto, J., Compadre, C., Swanson, S., Dhammika, N., Nanayakkara, N and Kinghorn, A. (1985) Metabolically Activated steviol, the aglycone of stevioside is mutagenic. Proc. Natl. Acad. Sci. USA 82, 2478-2482.
Puzzuto, J. Nanayakkara, N., Compadre, C., Swanson, S., Kinghorn, A., Guenthner, T., and Lam, L. (1986) Characterization of bacterial mutagenicity mediated by 13-hydroxy-enl-kaurenoic acid (steviol) and several structurally related derivatives and evaluation of potential to induce glutathione S-transferse in mice. Mutat. Res., 169, 93-103.
Sedgwick, S.G. and Bridges, B.A. (1972) Survival, mutation and capacity to repair single-strand DNA breaks after gamma-irradiation in differnt Exr strains of Escherichla colt. Mal. Gen. Clanet., 119, 93-102.
Skopek, T., Liber, H., Krolewski, J., and Thilly, W. (1978a) Quantitative forward mutation assay in Salmonella typhimurium using 8-azaguanine as a genetic marker, Proc. Natl. Acad. Sci. USA, 75, 410-419.
Skopek, T., Liber, H., Kaden, D., and Thilly, W. (1978b) Relative sensitivities of forward and reverse mutation assays in Salmonella Typhimurium, Proc. Natl. Acad. Sci. USA, 75, 4465-4469.
Wingard, R., Brown, J., Enderlin, F., Dale, J., Hale, R. And Seltz, C. (1980) Intestinal degradation and absorption of the glycoside sweetners stevioside and rebaudioside. Expertentla, 36, 519-520.
Received on August 3, 1990; accepted on October 10, 1990.
Table 1. 8-azaguanine resistant mutant
colonies per plate calculated to have arisen (column4) as a function of
exposure to increasing concentrations of 15-oxosteviol (columns 1, 2 and
3 are taken from table 6 of Pezzuto 1986)
Secure Ordering Ordering
Of Great Stevia Products Through Our