Tracing Information Flow from Erk to Target Gene Induction Reveals Mechanisms of Dynamic and Combinatorial Control

Abstract

Summary Cell signaling networks coordinate specific patterns of protein expression in response to external cues. Yet the logic by which signaling pathway activity determines the eventual abundance of target proteins is complex and poorly understood. Here, we describe an approach for simultaneously controlling the Ras/Erk pathway while monitoring a target gene’s transcription and protein accumulation in single live cells. We apply our approach to dissect how Erk activity is decoded by immediate-early genes (IEGs). We find that IEG transcription decodes Erk dynamics through a shared band-pass filtering circuit: repeated Erk pulses transcribe IEGs more efficiently than sustained Erk inputs. However, despite highly similar transcriptional responses, each IEG exhibits dramatically different protein-level accumulation, demonstrating a high degree of post-transcriptional regulation by combinations of multiple pathways. Our results demonstrate that the Ras/Erk pathway is decoded both by dynamic filters and logic gates to shape target gene responses in a context-specific manner.

Optogenetic rescue of a developmental patterning mutant

“…This complete molecular sufficiency is non-obvious: RTKs activate many intracellular pathways that are bypassed by OptoSOS stimulation (10), some of which have been suggested to play roles in early Drosophila embryogenesis (23). Nevertheless, our result is consistent with reports that RTK-driven gene expression (and thus cell fate determination) is mediated primarily by the Ras pathway (24), and that activating Ras pathway mutations are genetic suppressors of Tor partial loss-of-function alleles (25).…”

“…These observations immediately raise an open challenge: we still lack a clear picture of the signal interpretation circuits that respond to different profiles of developmental Erk signaling. Future work combining optogenetic control with high-resolution transcriptional studies (27,31) to track the localized expression of classic downstream genes (e.g. Tll, Hkb) could be crucial for closing this gap.…”

Oncogenic mutant RAS signaling activity is rescaled by the ERK/MAPK pathway

“…Conversely, gene products subject to rapid degradation kinetics such as c-Fos and Egr-1 would be only weakly elevated in RAS mutants, compared to the large changes in expression driven by sporadic wild type activity. Products under negative regulation, such as those that degrade rapidly even with extended activity (Wilson et al, 2017), may actually be suppressed by the chronic ERK activity. Thus, while enhanced ERK kinase activity as an indicator of early RAS mutant cells is difficult to detect without live-cell measurements, the resulting expression profile -particularly the ratio between long-term and short-term responsive genes -may be more informative.…”

Engineering combinatorial and dynamic decoders using synthetic immediate-early genes

“…Second, a quantitative understanding of signal decoding requires the ability to monitor mRNA-and protein-level responses over time in each individual SynIEG-expressing cell. Building off of a strategy we recently developed for endogenous target genes (14), we included a YFP-24xMS2 tag in each SynIEG. In this system, instantaneous transcription can be visualized as a bright nuclear spot when the nascent MS2 RNA loops are bound to the fluorescent RNA-binding protein MCP-mCherry, and protein accumulation can be monitored by YFP fluorescence.…”

“…In this system, instantaneous transcription can be visualized as a bright nuclear spot when the nascent MS2 RNA loops are bound to the fluorescent RNA-binding protein MCP-mCherry, and protein accumulation can be monitored by YFP fluorescence. As a first test case, we combined a 2-kb region upstream of the FOS transcriptional start site that contains its promoter and canonical upstream regulatory elements (15,16), the FOS 5'UTR, the coding sequence for monomeric super-folder YFP (msfYFP), 24xMS2 RNA stem-loops, and the TUBA1B 3'UTR in a single lentiviral vector, which we named fos-tubulin for its 5' and 3' elements, respectively ( Figure 2B), and introduced it into a clonal NIH3T3 cell line already expressing MCP-mCherry and H2B-iRFP as a nuclear marker (the "chassis" cell line; see Methods) (14).…”