Stochastic Models of –1 Programmed Ribosomal Frameshifting

Abstract

Many viruses can produce multiple proteins from a single mRNA sequence by encoding the proteins in overlapping genes. One mechanism that causes the ribosomes of infected cells to decode both genes is –1 programmed ribosomal frameshifting. In this process, structural elements of the viral mRNA signal the ribosome to shift reading frames at a specific point. Although –1 frameshifting has been recognized since 1985, the mechanism is not well understood. I have developed a stochastic model of mRNA translation that includes the possibility of a –1 frameshift at any codon. The transition probabilities between states of the model are based on the energetics of local molecular interactions. The model reproduces observed translation rates as well as both the location and efficiency of frameshift events in the HIV-1 gag-pol sequence. In this work, the model is used to predict changes in the frameshift efficiency due to mutations in the viral mRNA sequence or variations in relative tRNA abundances. The model is sensitive to the size of the translating ribosome and to assumptions about the unfolding pathway of the stimulatory structure. As knowledge in the field of RNA structure prediction grows, that knowledge can be incorporated into the model developed here to make improved predictions. The single-ribosome translation model has been extended to polysomes by including initiation and termination rates and an exclusion principle, and allowing the stimulatory structure to refold on an appropriate timescale. The predicted frameshift efficiency for a given mRNA can be tuned by varying the ribosome density on the mRNA. This finding affects the interpretation of frameshift efficiencies measured in the lab. In the parameter regime where translation is initiation-limited, the frameshift efficiency also depends on the structure refolding rate, which determines the availability of the downstream structure for stimulating –1 frameshifts. Furthermore, there is a trade-off between frameshift efficiency and protein synthesis rate.