For decades, the central dogma of molecular biology operated under a strict "coding" versus "non-coding" binary. We believed that protein synthesis required a linear messenger RNA (mRNA) equipped with a 5' cap and a 3' poly(A) tail to recruit the ribosome. Anything else—introns, antisense transcripts, and especially circular RNAs (circRNAs)—was dismissed as "transcriptional noise" or splicing errors.
However, a growing body of evidence suggests that the mammalian proteome is far more expansive than our standard gene models suggest. By utilizing non-canonical mechanisms like Internal Ribosome Entry Sites (IRES) and alternative start codons, the cell translates a "hidden proteome" from sequences once thought to be silent.
The Circular RNA Paradox
Circular RNAs are perhaps the most striking example of this hidden coding potential. Because they are covalently closed loops, they lack the 5' cap that the eukaryotic translation machinery typically uses to recognize and bind to a transcript. Despite this, Zi-Hao Zhang and colleagues have shown that circRNAs are not just molecular sponges for miRNAs, but active templates for protein synthesis [The function and mechanisms of action of circular RNAs in Urologic Cancer].
The intuition for how a circular transcript can be translated lies in the Internal Ribosome Entry Site (IRES). An IRES is a complex RNA secondary structure that acts as a physical "landing pad" for the ribosome, allowing it to bypass the need for a 5' end. As Jun Zheng demonstrated with circFOXK2, these IRES-driven peptides can directly modulate metabolic enzymes, driving the Warburg effect in hepatocellular carcinoma [CircFOXK2 Functions as a Protein Encode and a microRNA Sponge to Promote Hepatocellular Carcinoma Progression].
Bypassing the Scanning Machinery
Standard translation involves a "scanning" process where the 40S ribosomal subunit moves along the 5' untranslated region (UTR) until it finds an AUG start codon. This process is easily blocked by cellular stress, chemotherapy, or complex RNA folds like G-quadruplexes.
To survive these conditions, cells employ "backdoor" mechanisms:
- IRES-Mediated Entry: During stress, when cap-dependent translation is inhibited, trans-acting factors bind to IRES motifs to keep protein production running. Wenqing Gao found that the Aryl Hydrocarbon Receptor (AHR) uses this mechanism to maintain production during paclitaxel treatment, contributing to drug resistance [Paclitaxel-mediated human aryl hydrocarbon receptor mRNA translation].
- Ribosomal Tethering: Some transcripts use a "stop-start" or termination-reinitiation strategy. Sawsan Napthine and Michael Powell have characterized how specific sequences can base-pair directly with the 18S rRNA of the ribosome. This physical tether prevents the ribosome from dissociating after it finishes one protein, allowing it to immediately "hop" to a nearby start codon for a second protein [Characterization of the termination–reinitiation strategy employed in the expression of influenza B virus BM2 protein].
- Chemical Modifications: Modifications like N6-methyladenosine (m6A) can also recruit ribosomes directly to internal sites, bypassing the 5' cap requirement entirely [The Mechanism and Clinical Significance of Circular RNAs in Hepatocellular Carcinoma].
Mapping the Dark Proteome
If these non-canonical mechanisms are widespread, our current maps of the proteome are incomplete. Sudhakaran Prabakaran and Martin Hemberg used mass spectrometry to identify hundreds of peptides translated from "non-coding" regions like introns and pseudogenes, many of which are regulated by neuronal activity [Quantitative profiling of peptides from RNAs classified as noncoding].
Recent large-scale efforts by the TransCODE consortium, led by researchers like Sonia Chothani and Jorge Ruiz-Orera, have expanded the human catalog of open reading frames (ORFs) by fourfold. Their work reveals that translation frequently initiates at near-cognate codons (like CUG or GUG) rather than the standard AUG, particularly for microproteins shorter than 16 codons [An expanded reference catalog of translated open reading frames for biomedical research].
Why It Works: Structural Control
The "why" behind these mechanisms often comes down to RNA architecture. RNA is not a limp string; it is a highly structured molecule.
- Barriers as Regulators: Conserved stem-loops or G-quadruplexes in the 5' UTR act as physical roadblocks. Joanna Kovalski showed that regulatory proteins like RBM42 act as "architects" that remodel these barriers to allow translation of oncogenes like MYC [Functional screen for mediators of onco-mRNA translation specificity].
- Stop Codon Readthrough: In some cases, the ribosome ignores a "stop" signal. Irwin Jungreis found that downstream stem-loops can stall the ribosome just long enough to allow a "near-cognate" tRNA to slip in, extending the protein into a longer, functional isoform [Evidence of abundant stop codon readthrough in Drosophila and other metazoa].
By moving beyond the 5' cap, the cell gains a layer of translational control that is independent of global protein synthesis levels. This allows for the production of essential survival factors during the very stress conditions—hypoxia, nutrient deprivation, or chemotherapy—that shut down the rest of the cell's machinery [The Role of Translation Initiation Regulation in Tumorigenesis, Progression and Drug Resistance].