This study focuses on understanding how the pathological mutation L50R in R1beta subunits alters the properties of the protein. The study can be considered as a partner article to the study published earlier this year in Brain that utilised a mouse cell line carrying the mutation, and focused on aggregation. The work presented here takes a different approach, including experiments with fibroblasts derived from a human patient, and comprehensive characterisation of the binding kinetics and cAMP activation properties of PKA holoenzymes including the mutated R1b subunits. Overall this is an interesting study that extends our understanding of the pathology of the L50R mutation. Specific points and questions are as follows:

1. Regarding first paragraph on page 6, it looks like the first half of CNB-A is the least variant region of all. Does this reflect its key gatekeeper role in dictating release of C subunits from R1?

2. Regarding the CD data: is the material in soluble aggregates for the mutant? Perhaps dimerization is required for alpha-helix formation since one might not expect reduction from 87 to 43 % helicity with disruption of just the helix containing position 50

3. For Fig 1E, it was not immediately obvious that the top row was distinct from the lower three. Some additional labelling would help the reader here. Very stiking and interesting effect of the mutation though.

4. For figure 2B, the normalisation is potentially a bit misleading as makes it look like there is more monomer in the mutant vs WT. Would it not be better to show the data without normalising the total to 1 for either RI variant?

5. For figure 2, in general should the L50R samples be labelled WT/L50R since they are taken from a heterozygous individual? This would explain the difference in appearance of e.g. Figure 1D and Figure 2E.

6. For figure 4D, the numbers in the table don’t seem to match the plot ie L50R appears to achieve 50% activation at a lower cAMP concentration than WT.

7. For Figure 5: it would be helpful to cross reference these peaks to a molecular weight standard series. Was a SEC-MALS instrument not available? It’s not clear where the void volume of the column is in the figure.

8. Regarding the transcriptome analysis. Would one expect to detect multiple axon genes in fibroblasts? The transcriptome analysis focuses on classes of genes rather than individual genes. Were there specific genes that jumped out? Furthermore, focusing on components of the cAMP signalling pathway (PKA subunits, PDEs, cyclases, AKAPs), were there any notable changes for the genes corresponding to these proteins?

The authors have rigorously responded to my comments.