Advancing dental biofilm models: the integral role of pH in predicting S. mutans colonization

Abstract

Mathematical models can provide insights into complex interactions and dynamics within microbial communities to complement and extend experimental laboratory approaches. For dental biofilms, they can give a basis for evaluating biofilm growth or the transition from health to disease. We have developed mathematical models to simulate the transition toward a cariogenic microbial biofilm, modeled as the overgrowth of Streptococcus mutans within a five-species dental community. This work builds on experimental data from a continuous flow reactor with hydroxyapatite coupons for biofilm growth, in a chemically defined medium with varying concentrations of glucose and lactic acid. The biofilms formed on the coupons were simulated using individual-based models (IbMs), with bacterial growth modeled using experimentally measured kinetic parameters. The IbM assumes that the maximum theoretical growth yield for biomass is dependent on the local concentration of reactants and products, while the growth rates were described using traditional Monod equations. We have simulated all the conditions studied experimentally, considering different initial relative abundance of the five species, and also different initial clustering in the biofilm. The simulation results only reproduced the experimental dominance of S. mutans at high glucose concentration after we considered the species-specific effect of pH on growth rates. This highlights the significance of the aciduric property of S. mutans in the development of caries. Our study demonstrates the potential of combining in vitro and in silico studies to gain a new understanding of the factors that influence dental biofilm dynamics. IMPORTANCE We have developed in silico models able to reproduce the relative abundance measured in vitro in the synthetic dental biofilm communities growing in a chemically defined medium. The advantage of this combination of in vitro and in silico models is that we can study the influence of one parameter at a time and aim for direct validation. Our work demonstrates the utility of individual-based models for simulating diverse conditions affecting dental biofilm scenarios, such as the frequency of glucose intake, sucrose pulsing, or integration of pathogenic or probiotic species. Although in silico models are reductionist approaches, they have the advantage of not being limited in the scenarios they can test by the ethical consideration of an in vivo system, thus significantly contributing to dental biofilm research.