University of Rochester Rochester, New York, United States
Purpose: Understanding gene expression opens new avenues for novel diagnostic and therapeutic interventions. Messenger RNA (mRNA) is tightly regulated to meet biological needs, especially during initiation. Briefly, during translation initiation, the 40S small ribosomal subunit is recruited by the eIF4F preinitiation complex (PIC), which then scans the highly structured 5'untranslated region (5'UTR) to search for the start codon (i.e. AUG). The 5'UTR is embellished with numerous forms of regulatory elements, one of which is upstream open reading frames (uORFs), which are small peptide-encoding sequences that reside within ~50% of mRNA in humans. Upon their translation, the main open reading frame (mORF; also referred to as protein-coding sequence) is suppressed due to the premature consumption of translation factors whilst translating these uORFs. Work by other groups show the importance of uORFs in regulating multiple biological processes and disease progression. Considering the large number of uORFs in human transcriptome, the mechanistic regulation of most individual uORFs remains elusive. Since the 40S ribosomal subunit lacks any helicase activity and given the highly structured nature of 5'UTR, we hypothesize that these structures guide the 40S to initiate at uORFs. More work is needed to in this area since there is a dearth of knowledge concerning the RNA structure-function relationship of the majority of 5'UTR. Here we elucidated the relationship between uORFs and RNA structures in artificial and human 5’UTR of GATA4, which is an essential cardiac transcription factor for the development and development of cardiac hypertrophy, among others. We used this understanding to develop novel antisense oligonucleotides (ASOs) that could increase or decrease protein levels in cells or used to treat heart disease mouse models. Methods: We used firefly (FLuc) luciferase assays consisting of artificial 5'UTR-FLuc DNA constructs encoding for an RNA hairpin structure spaced in various intervals relative to a start codons. FLuc readings were used to measure changes in protein expression. We further characterized the 5'UTR of GATA4 to serve as a biologically relevant example for such regulation. The RNA Structure was determined using RNA Selective 2' Hydroxyl Acylation analyzed by Primer Extension (SHAPE), in which the RNA is treated via 2-methylnicotinic acid imidazolide (NAI), which selectively acylates unstructured regions of the RNA. These modifications serve as roadblocks when the RNA is reverse transcribed. Using an RNA sequencing gel, we could determine the positions of these modifications at a single nucleotide resolution, and thus determine the RNA structure. The elucidated structure was used to inform 5'UTR-FLuc reporter assays, similar to the ones described above, in which the RNA structures via mismatches, allowing us to understand the importance of a given dsRNA on uORF activity. We also generalized these findings by performing the same FLuc reporter assays on the 5'UTR of MFN and DHTKD1, which are predicted to possess dsRNA elements downstream of their uORFs. From these findings, we designed GATA4 5'UTR-specific ASOs, that could either disrupt dsRNA from the formation (Class 1) or artificially form a bimolecular dsRNA (Class 2). The ASOs were transfected into embryonic stem cell-derived cardiomyocytes (CM) that are wild-type (WT) or lacking the uORF (ΔuORF) to test their specificity. We also tested Class 2 as an anti-hypertrophy drug in mice that have undergone trans-aortic constriction, which induces cardiac hypertrophy. Echocardiography was to monitor cardiac function, histology to assess cardiac remodeling, and biochemical assays to measure GATA4 protein levels. Results: We found that the presence of a double-stranded RNA (dsRNA) structure downstream of a uORF start codon promotes uORF-mediated suppression of the mORF, especially when they are spaced by 2-8 nucleotides (nts) apart. The 5’UTR of the GATA4 mRNA contained a 9-nt dsRNA region downstream of the GATA4 uORF (Figure 1). This dsRNA promotes its activity, reducing mORF translation. Abolishing this dsRNA via MM mutations alleviates the uORF activity and enhances mORF translation. Likewise, MM in their predicted dsRNA within DHTKF1 and MFN1 show recapitulates this mechanism. We found that 5'UTR dsRNA is a promising target for ASOs. GATA4 Class 1 ASOs could disrupt 5'UTR dsRNA, inhibit the uORF, and promote mORF translation, which leads to CM hypertrophy. In contrast, ASOs designed to base-pair downstream of uORF (Class 2) enhance translation of the uORFs and suppress protein levels, leading to CM atrophy (Figure 2). These effects were only observed in only WT, but not ΔuORF CM, suggesting uORF-specific effects. To our surprise, we also found that the former ASO caused cellular hypertrophy, while the latter caused cellular atrophy. Our mouse experiment found that Class 2 ASOs repress GATA4 protein levels and elicit an anti-hypertrophic response (Figure 3) and improve cardiac function. Conclusion: From this work, we found that RNA secondary structures can act as cis-acting regulatory elements that regulate mRNA translation possibly through the regulation of uORFs. This knowledge inspired the development of novel ASOs that can serve as a tool for modulating protein levels in cells for various applications. References: Akazawa, H., and Komuro, I. (2003). Roles of cardiac transcription factors in cardiac hypertrophy. Circ Res 92, 1079-1088.
Calvo, S.E., Pagliarini, D.J., and Mootha, V.K. (2009). Upstream open reading frames cause widespread reduction of protein expression and are polymorphic among humans. Proc Natl Acad Sci U S A 106, 7507-7512.
Leppek, K., Das, R., and Barna, M. (2018). Functional 5' UTR mRNA structures in eukaryotic translation regulation and how to find them. Nat Rev Mol Cell Biol 19, 158-174.
Acknowledgments: This work was supported in part by the National Institutes of Health [R01 HL132899 and R01HL147954 to Peng Yao] and T32 GM068411 to Omar Hedaya. Right, sequencing gel showing every single nucleotide of the GATA4 5'UTR as well as SHAPE products, which are the reverse transcriptase interruptions due to the chemical acylation of 2-methylnicotinic acid imidazolide. Left, predicted structure based on RNA SHAPE gel.
ASO Class 1 leads to CM hypertrophy while ASO Class 2 leads to CM atrophy
ASO Class 2 renders mouse heart resistant to hypertrophy following trans-aortic constriction surgery.