Computational kinetic modeling informs on uORF buffering activity

From the Geballe and Subramaniam labs, Human Biology and Basic Sciences Divisions

Half of all human genes that are transcribed into mRNAs have one or more upstream open reading frame(s) (uORF) preceding the main ORF. Many of these uORFs reduce translation of the main ORF and when mutations in these regions occur, the dysregulation of the main ORF expression can cause disease including cancer. The mechanisms by which uORFs regulate the translation of the main ORF are many and include the following: uORF-mediated ribosome siphoning away from the main ORF, uORF-regulated ribosome re-initiation at the main ORF, and inefficient elongation or termination on the uORF that prevents ribosomes from reaching the main ORF start site. For many proposed models, validation is incomplete. Advances in computational modeling of translation kinetics has enabled the study of varied uORF and ribosome loading rates effects on translational output. To investigate the mechanisms of uORF-dependent regulation of main ORF translation, the Geballe and Subramaniam labs from Human Biology and Basic Sciences Divisions at Fred Hutchinson Cancer Center used an experimental system based on the cytomegalovirus UL4 mRNA containing a unique uORF to test newly derived computational kinetics models and design a framework for efficient simulation of translation. Their findings, led by graduate student Ty Bottorff, were recently published in PLoS Genetics.

“Upstream ORFs are a pervasive feature of human mRNAs, but the mechanism by which they regulate protein expression is still unclear,” stated Dr. Rasi Subramaniam. The cytomegalovirus UL4 mRNA contains three uORFs, one of which includes a diproline motif, or two side-by-side proline amino acids, that causes ribosomal stalling upstream of the main ORF. This resultant stalling “buffers” translation of the main ORF from variation in changes in ribosome loading and initiation, but the mechanism of how this is regulated is only partially understood. The researchers designed a reporter system to study the uORF2 of UL4 mRNA. They cloned the uORF2 into a dual-luciferase reporter system for efficient quantification of main ORF translation using a proxy luciferase signal readout. The uORF2 sequence was modified to include a stronger translation start site, no start codon, or removal of diproline motifs for comparison with wildtype sequence to assess changes in ribosomal stalling kinetics. Characterization of these reporter genes confirmed that both the start codon and diproline signature were required for translational repression of the main ORF.

Translational buffering can be regulated by stalling of ribosomes at specific sequences upstream of the main ORF. Increased retention of ribosomes on the uORFs results in a reduction of ribosome loading and less protein production from the main ORF.
Translational buffering can be regulated by stalling of ribosomes at specific sequences upstream of the main ORF. Increased retention of ribosomes on the uORFs results in a reduction of ribosome loading and less protein production from the main ORF. Image provided by Mr. Bottorff

Using this system, the researchers tested several computational kinetics models to identify which model best fit the observed kinetics of translation and ribosome loading. Those tested included models of ribosome dissociation from mRNA due to collision with stalled or elongating ribosomes or those queued behind stalled ribosomes and changes to ribosome re-initiation due to uORF features. Iterative computational kinetics modeling and experimental testing of the uORF2 UL4 luciferase reporter systems enabled selection of collision model as the best fit. In this kinetic model, the stalled ribosome creates a roadblock to scanning ribosomes and collision of the two knocks the scanning ribosome off the mRNA leading to reduced main ORF translation. “Our computational framework allows easy specification and efficient simulation of several previously proposed kinetic models of uORF regulation,” summarized the researchers.

One feature observed for uORFs it that they function as buffers to temper the outcome of increased or decreased translation of the main ORF. To determine if the computational kinetic models predicted for buffering could be validated experimentally, the researchers again returned to the uORF2 dual-luciferase reporter system. The uORF2 mutant lacking the diproline stall was used as the control ribosome loading rate. Buffering activity was inferred from luciferase expression under different conditions that altered the rate of ribosome loading and scanning of the uORF start site – addition of a stem-loop with variable GC content where higher GC content reduced ribosome scanning, induction of ER stress with drug treatment, and addition of synthetic uORF upstream of uORF2 to siphon ribosomes from uORF2. From these data, the researchers found that the luciferase indicator of main ORF translation negatively correlated with ribosome loading for each system. These findings support a model in which the wildtype uORF2 is buffering against reduced main ORF protein synthesis under low ribosome loading conditions. Therefore, by using this approach, “we can predict which models of uORF regulation allow buffering and which parameters are key for buffering in each model,” stated the researchers.

“Our work contributes to the translation field a general framework for dissecting uORF mechanisms through computational simulations and genetic perturbations,” stated Dr. Subramaniam. In the future, “we are interested in extending the computational modeling here to investigate, and visualize, more complex translational regulatory mechanisms: cap-tethered initiation, cellular stress-regulated elongation and re-initiation rates, among others,” concluded graduate student Mr. Bottorff.

The spotlighted research was funded by the National Institute of General Medical Sciences, National Science Foundation, and National Institute of Allergy and Infectious Diseases.

Fred Hutch/University of Washington/Seattle Children's Cancer Consortium members Adam Geballe and Rasi Subramaniam contributed to this work.

Bottorff TA, Park H, Geballe AP, Subramaniam AR. 2022. Translational buffering by ribosome stalling in upstream open reading frames. PLoS Genet. 18(10):e1010460.