Subcellular Dynamic Immunopatterning of Cytosolic Protein Complexes on Microstructured Polymer Substrates
Curation statements for this article:-
Curated by eLife
Evaluation Summary:
This manuscript describes a method of evaluating the steady state levels and kinetics of protein-protein interactions at the plasma membrane of living cells. This approach builds on previous work in this area, and with stronger validation and demonstration of biological applications, it may be complementary to other biochemical or imaging-based approaches to address important questions related to mechanisms of cell signaling.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 and Reviewer #2 agreed to share their name with the authors.)
This article has been Reviewed by the following groups
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
- Evaluated articles (eLife)
Abstract
Article activity feed
-
Evaluation Summary:
This manuscript describes a method of evaluating the steady state levels and kinetics of protein-protein interactions at the plasma membrane of living cells. This approach builds on previous work in this area, and with stronger validation and demonstration of biological applications, it may be complementary to other biochemical or imaging-based approaches to address important questions related to mechanisms of cell signaling.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 and Reviewer #2 agreed to share their name with the authors.)
-
Reviewer #1 (Public Review):
The authors have improved the patterned chemical functionalization of surfaces using large-area polydimethylsiloxane (PDMS)-based elastomeric stamps.They then proceed with investigating the interaction of Grb2 with EGFR by expressing an artificial transmembrane receptor construct that has Grb2 as bait (bait-PAR-Grb2). They could show that these constructs are indeed organized according to the extracellular antibody surface pattern. They then pattern EGFR by a micro-structured surface of anti-EGFR antibodies (ED) and observe the interaction with what is now the freely diffusing prey bait-PAR-Grb2 construct. After stimulation with the growth factor EGF they observe an increase of recruitment of bait-PAR-Grb2 above already basal levels in these EGFR receptor patterns.
Here, several problems occur. For one, …
Reviewer #1 (Public Review):
The authors have improved the patterned chemical functionalization of surfaces using large-area polydimethylsiloxane (PDMS)-based elastomeric stamps.They then proceed with investigating the interaction of Grb2 with EGFR by expressing an artificial transmembrane receptor construct that has Grb2 as bait (bait-PAR-Grb2). They could show that these constructs are indeed organized according to the extracellular antibody surface pattern. They then pattern EGFR by a micro-structured surface of anti-EGFR antibodies (ED) and observe the interaction with what is now the freely diffusing prey bait-PAR-Grb2 construct. After stimulation with the growth factor EGF they observe an increase of recruitment of bait-PAR-Grb2 above already basal levels in these EGFR receptor patterns.
Here, several problems occur. For one, there is no quantification of this enhanced recruitment, just exemplary images in contrast to some other experiments. Second and more important, since the now prey (bait-PAR-Grb2) construct is part of the construct that is inserted in the membrane both EGFR as well as Grb2 react in an approximately two-dimensional surface. The tremendous increase in effective concentration of the Grb2-prey can therefore not only force the interaction with EGFR in this second order reaction (possibly irrespective of phosphorylation, see below) but also enhance the effective kinase activity of EGFR towards these highly concentrated substrates. These could now recruit pTyr dependent SH2 domain containing signaling adapters or activities like Shc. For example, already the basal kinase activity of EGFR could thereby generate substantial recruitment without EGF-stimulus and the stimulus-dependent response could result in hyper-phosphorylation of Grb2. It is also possible that due to the enhanced concentration at the plasma membrane of bait-PAR-Grb2 there is an interaction of the TMD of the Bait-PAR-Grb2 with EGFR that patterns the construct. The reason why I allude to this, is because there is no interaction of the bait-PAR-Grb2 construct without linker with the freely diffusing Shc (which should not be a problem in terms of super-enhanced concentration driven interaction with the prey (Shc) because it is freely diffusing in the cytoplasm). This could point at a problem in terms of Grb2 hyper-phosphorylation by EGFR kinase activity in the construct. Indeed, when the authors increase the linker length between Grb2 and the artificial receptor, they can actually observe a recruitment of Shc to the Bait-Parc-Grb2 pattern (again not quantified). Since in this case, the bait-PAR-Grb2 was patterned, it could well be that EGFR is recruited to bait-PAR-Grb2 via SH2-pTyr interaction, which then hyper phosphorylates Grb2 to recruit Shc or Shc is simply recruited to pTyr on EGFR or both mechanisms apply. In general, any second order protein interaction or reaction (for example kinase) is a function of the effective concentration of its components, which is highly distorted in a second order reaction system that has both Grb2 and EGFR associated and diffusing in the 2D surface of the membrane. In case that the prey is freely diffusible and the bait is on the membrane, this does not apply.
Another problem is that the presented models are not unambiguous nor provide any insight what they mean for EGFR signaling beyond what was already known for quite a while now. In that sense, none of the findings on constant SH3 mediated (Grb2-SOS or Grb2-Gab) interactions or pTyr conditional SH2 mediated interactions (Shc, Grb2, p85) are surprising or provide new information about EGFR signaling. What would have been interesting is if and how the pTyr-SH2 mediated complexes are distributed between the phosphorylated receptor and phosphorylated adapters in a EGFR density dependent manner and what impact this would have had on signaling. This would of course require constructs with mutations on pTyr sites and the monitoring of interactions with multiple Bait-PARs of wt and mutant proteins, all in the same cell. In this respect, the major point that could lift the methodology presented in the paper in relation to what was previously done, is the patterning of N different Bait-Parcs to simultaneously monitor multiple different interactions during receptor-induced signaling in the same cell. The modular design principle of bait-PARCs enables this simultaneous protein-interaction measurements and thereby allows for the direct analysis of relations between proteins in interconnected signaling networks. This is necessary to for example unravel mechanisms of signaling via distributed processes, signaling crosstalk and off-target effects of drugs, which are the rule more than the exception and otherwise blurred by cell-to-cell variance in protein expression levels and internal reaction states. This also applies to the presented studies on small molecule PPI disruptors that yielded nice results but no new significant insights.
-
Reviewer #2 (Public Review):
The approach is based on plating cells on a substrate containing micropatterned antibodies to the extracellular domain of a transmembrane bait protein, resulting in areas of high concentration of the intracellular portion of the bait, surrounded by areas of lower concentration. Fluorescently labeled prey proteins are co-expressed in the same cell, and association of bait and prey is assessed (via TIRF microscopy) by the colocalization of the prey with areas of high local bait concentration. As a test system, the authors chose the Grb2 adaptor as a bait, which can interact with other signaling proteins in the EGFR pathway via its SH2 domain or its two SH3 domains. In addition to being able to assess whether pairwise interactions are dependent on EGFR activation (and thus tyrosine phosphorylation) or are …
Reviewer #2 (Public Review):
The approach is based on plating cells on a substrate containing micropatterned antibodies to the extracellular domain of a transmembrane bait protein, resulting in areas of high concentration of the intracellular portion of the bait, surrounded by areas of lower concentration. Fluorescently labeled prey proteins are co-expressed in the same cell, and association of bait and prey is assessed (via TIRF microscopy) by the colocalization of the prey with areas of high local bait concentration. As a test system, the authors chose the Grb2 adaptor as a bait, which can interact with other signaling proteins in the EGFR pathway via its SH2 domain or its two SH3 domains. In addition to being able to assess whether pairwise interactions are dependent on EGFR activation (and thus tyrosine phosphorylation) or are constitutive, further experiments tested the ability of cell-penetrating compounds to inhibit bait-prey interactions, and used FRAP to assess the kinetics of prey association.
Strengths: The beauty of this approach is that experiments can be performed in living cells in real time, using relatively simple and straightforward imaging approaches. Most other approaches to address similar questions either require cell lysis, or more sophisticated imaging modalities. In particular, the ability to examine both on-rates (via FRAP) or off-rates (using chemical inhibitors) provides valuable information that is often difficult to obtain in living cells. Thus in principle this has the potential be a real workhorse method for dissecting the dynamics of signaling events at the membrane.
Weaknesses: Despite the potential advantages of the approach, there are a number of serious weaknesses. Both bait and prey are extensively modified (by addition fusion to fluorescent proteins, and in the case of the bait by covalent tethering to the transmembrane fusion). High level expression of both bait and prey can potentially distort the kinetics of interactions and/or compete with endogenous interactions. Specifically in this case, membrane localization of Grb2 is expected to lead to constitutive activation of the Ras and possibly PI3K pathways, which is likely to induce feedback inhibition in cells expressing these constructs.
Another concern is that the signaling interactions studied are potentially much more complicated than the simple (and in some cases non-canonical) models presented here. Grb2 SH2 domain can interact directly with phosphorylated EGFR or with phosphorylated Shc (which is recruited to and phosphorylated by activated EGFR); PI3K can be activated by recruitment to phosphorylated Gab1 or EGFR, or by activated Ras; EGFR activation not only provides binding sites for SH2 domains on the receptor itself, but also phosphorylates Shc, Gab1, and many other proteins. So the recruitment of a prey protein to Grb2 bait upon EGF treatment can be due to any number of molecular interactions, greatly complicating interpretation of results.
Finally, the interpretation of the inhibitor experiments in Fig. 6 is complicated because of lack of information on the precise mechanisms involved-actinomycin D is not a well characterized signaling inhibitor, and the cyclic SH2 ligand seems to be acting to inhibit SH3-mediated interactions instead of SH2-mediated interactions as expected. These inhibitor data are quite surprising and difficult to rationalize, and thus in my mind do not provide solid evidence of the usefulness of the system for studying inhibitors.
-
-
