Preview

PDF, English Download (7MB) | Terms of use

Abstract

The association of proteins into functional oligomeric complexes is crucial for nearly all cellular processes. Despite rapid progress in characterizing the structure of native assemblies, the underlying mechanisms that guide faithful complex formation in the crowded cellular environment are understood only superficially. To secure efficient complex biogenesis and limit the exposure of aggregation-prone intermediates, many proteins assemble co-translationally, via interaction of a fully synthetized and a nascent protein subunit (co-post assembly). Here, we explore the prevalence and the mechanistic principles of a putative co-translational assembly mechanism, which involves the direct interaction of nascent subunits emerging from proximal ribosomes (co-co assembly). To obtain direct evidence of this putative assembly mode, we apply a newly developed method based on Ribosome Profiling, named Disome Selective Profiling (DiSP), which allows to monitor the conversion of single ribosomes to nascent chain connected ribosome pairs across the proteome with high resolution. We use this approach to analyse co-co assembly in two human cell lines and demonstrate that it constitutes a general mechanism inside cells that is employed by hundreds of high confidence and thousands of low confidence candidates, comprising 11 to 32% of all complex subunits. Analysing the features of the co-co assembly proteome, we reveal that this mechanism guides formation of mostly homomeric complexes and typically relies on interaction of N-terminal nascent chain segments. We further identify five dimerization domains mediating the majority of co-co interactions, which are either partially or completely exposed at the onset of nascent chain dimerization, implying different folding and assembly mechanisms. The detectable fraction of each candidate’s nascent chains that co-co assemble is in median 40% and in some cases exceeds 90%, suggesting that this co-translational assembly path may be employed as the main route for complex formation. To gain deeper insights into the mechanistic basis of co-co assembly, we took a series of experimental approaches that distinguish between interactions of nascent chains emerging from the same or different polysomes (termed assembly in cis and in trans, respectively). These experiments could not support a model of assembly in trans. Conversely, we find indications supporting a cis assembly model for nuclear lamin C, one of our high confidence candidates. This mechanism provides a simple explanation for the remarkable specificity of lamin homodimer formation in vivo, where splice variants with largely overlapping sequences do not mix. We propose that assembly in cis more generally secures specific homomer formation of isoforms and structurally-related proteins which are highly prone to promiscuous interactions inside cells. In conclusion, this study provides a global annotation of nascent chain interactions across the human proteome and elucidates the basic principles of this widespread assembly pathway. Our findings raise a number of fundamental questions concerning the mechanisms ensuring high-fidelity protein biogenesis, including the implications of co-co assembly on polysome structure, the possible consequences of co-co assembly failure, the inter-dependence with co-translational folding and the synchronization and coordination with translation kinetics.