Chronic obstructive pulmonary disease (COPD) is a leading cause of mortality in the developed world. Smoking is an important risk factor for the disease, but genetics determine outcome and disease severity. A significant and broad challenge in establishing the causal molecular mechanism from genetic association data is the fact that a majority of COPD-associated variants map to non-coding regions of the human genome. One of the genes strongly associated with COPD is SERPINA1, which encodes the a-1 antitrypsin protein. The SERPINA1 gene is remarkably complex: It has eleven splice variants, all of which change the 5'-untranslated region (5'-UTR) without altering the sequence of the encoded protein. We found that translation efficiencies of the mRNAs varied by orders of magnitude due to the strengths of upstream RNA structure and of open reading frames (uORFs). uORFs are found in roughly 50% of human genes and tend to function to reduce translation of the downstream gene but, other than this observation, are poorly understood mechanistically. We have developed and parameterized a structure-based leaky scanning model of translation that considers alternative splicing, uORF Kozak sequence strength, the RNA structure at the initiation site of uORFs, and the efficiency of translation of the primary open reading frame. We propose in our first aim to define how RNA structure and alternative splicing control expression of a-1 antitrypsin and extend this approach to two other COPD-associated genes. In our second aim we will comprehensively characterize the RNA structural elements in the SERPINA1 mRNA 5'-UTRs and in two other COPD-related mRNAs that control translation efficiency, without and with uORFs. These experiments will establish accurate and broadly impactful frameworks to define the RNA structural features that modulate translation efficiency in 5'-UTRs and will refine ribosomal leaky scanning models to better predict tissue-specific expression of COPD-associated proteins.
In many cases, protein synthesis efficiency is governed by the amount and location of structure in the untranslated regions (UTRs) of an mRNA. In this project, we will generate functional and structural models for mRNAs related to chronic obstructive pulmonary disease (COPD). These models will establish accurate and broadly impactful computational frameworks to define the RNA structural features that modulate translation efficiency in 5'-UTRs and will create a resource for identifying causative variants in precision medicine.
Showing the most recent 10 out of 22 publications