Stochastic modeling of the quorum-sensing system in the streptococcus mutans

A detailed stochastic model of the quorum-sensing (QS) system of S. mutans is presented. Bacteria use QS to activate specific functions based on their population density. Using the competence stimulating peptide (CSP), the QS system of S. mutans controls the vital functions of biofilm formation, acid tolerance, and competence induction. A clear understanding of the genetics of QS may offer new methods of defence against dental caries, and new strategies to deter the spread of antibiotic resistance genes. To further our understanding, a stochastic model of the S. mutans CSP system is presented that incorporates existing structural knowledge and new quantitative experimental results. Previous work has qualitatively delineated the genes and the mechanisms responsible for CSP production and competence induction. In the present work, realtime reverse transcriptase polymerase chain reactions were used to determine the expression profile of the CSP genes in response to exogenous CSP. A typical expression profile of an up-regulated gene consists of three phases. Initially, the transcript level remains unchanged for 5 minutes. It then ramps up to its maximal two-fold induction by the 10-15 minutes point. Finally in the shutdown phase, the transcript level is decreased to its basal level in 5 minutes. This phase may include a sub-basal refractory period. It is unclear how interactions in the up-regulation pathway lead to the long initial activation delay. By using the expression profiles to calibrate the model and obtain rate parameters of some of the constituent reactions, the space of unknown parameters may be restricted so that these may be explored through simulation with alternative parameter values. This work demonstrates the range of viable solutions. In addition, a hypothetical CSP shutdown mechanism has been incorporated into the model for evaluation. This uses a gene induced during the ramp up period to negatively regulate the initially up-regulated genes. This mechanism is found to be particularly sensitive to the shutdown gene's effective transcript initiation rate, and to the binding efficiency of the gene products in competition with others to the same promoter. This work demonstrates the capability of stochastic modeling to identify gap in the knowledge of pathway structure, such as the activation delay, and to effectively investigate alternative mechanistic hypotheses.