eLife Assessment

This study presents a useful investigation of the use of small, de novo-designed protein binding domains (mini-binders) against the Spike protein of SARS-CoV-2 and EGFR, as ligand binding domains on two classes of synthetic receptors, second-generation synNotch (SNIPR) and CAR. The methods and evidence supporting the focused claims are solid. This work will be of interest to synthetic biologists and cell engineers as a starting point to map out the rules for receptor engineering based on mini-binders and ultimately to advance them in biomedical applications.

Reviewer #2 (Public review):

Summary:

Weinberg et al. show that spike LCB minibinders can be used as the extracellular domain for SynNotch, SNIPR, and CAR. They evaluated their designs against cells expressing the target proteins and live virus.

Strengths:

This is a good fundamental demonstration of alternative use of the minibinder. The results are unsurprising but robust and solid in most cases.

Weaknesses:

The manuscript can benefit from better descriptions of the study's novelty. Given that LCB previously worked in SynNotch, what unexpected finding was uncovered by this study? It is well known that the extracellular domain of CAR is amendable to different types of binding domains (e.g., scFv, nanobody, DARPin, natural ligands). So, it is not surprising that a minibinder also works with CAR. We don't know if the minibinders are more or less likely to be compatible with CAR or SNIPR.

The demonstrations are all done using just 1 minibinder. It is hard to conclude that minibinders, as a unique class of protein binders, are generalizable in different contexts. All it can conclude is that this specific Spike minibinder can be used in synNotch, SNIPR, and CAR. The LCB3 minibinder seems to be much weaker.

The sensing of live viruses is interesting, but the output is very weak. It is difficult to imagine a utility for such a weak response.

Author response:

The following is the authors’ response to the original reviews.

In our initial submission, reviewers highlighted that the major limitations of our study were related to both the number of minibinders tested as well as the number of optimizations we evaluated for improving minibinder function. In this revision, we have focused on expanding the minibinders tested. To do so, we selected two previously published minibinders against the epidermal growth factor receptor (EGFR). Selection of EGFR as a target enabled us to evaluate two minibinders that bind at different sites, unlike the previously evaluated binders LCB1 and LCB3 which both bind the same interface on SARS-CoV-2 Spike. Further, using EGFR as a target enabled us to qualitatively compare the efficacy of minibinder-coupled chimeric antigen receptors against an existing anti-EGFR CAR. We believe the results here demonstrate broader generalizability of our approach across binding sites, targets, and minibinders. We hope this addition is sufficient to convince future would-be users of these tools to attempt synthetic receptor engineering using minibinders against their protein of choice.

Reviewers made comments about the presentation of flow data and the use of statistics throughout the manuscript. We did not modify how flow data are presented as the density plots we used are common throughout the field. We have opted to not include statistics – we believe that in the case of most of the experiments we show, our findings are obvious. In cases where statistics would be helpful for discerning whether subtle effects are real – for example, comparing the linker-based optimizations or comparing the anti-EGFR CARs – we believe that other experimental factors like construct expression are sufficient confounds that even in the presence of statistically significant effects we would be leading readers astray to make such claims about our data. As such, we have sought to limit the claims we make and hope that reviewers and audience agree we do not over interpret our data without statistical support.

On more minor points, both reviewers addressed the differences in Figure 5A and 5C, which we addressed in our figure legend and in the previous response to reviews is the result of these data originating from different time points of the same assay. Reviewer #2 believed we should be more staid in our comments about linker optimality, which we have addressed by changing the referenced line in the discussion. Otherwise, we have made no modifications to figures or text beyond the addition of new data.