Optogenetic control of Protein Kinase C-epsilon activity reveals its intrinsic signaling properties with spatiotemporal resolution

Reviewer #1 (Public review):

Summary:

Optogenetic tools enable very precise spatiotemporal control of the signaling pathway. The authors developed an optimized light-regulated PKC epsilon, Opto-PKCepsilon using AlphaFold for rational design. Interactome and phosphoproteome studies of light-activated Opto-PKCepsilon confirmed a high similarity of interaction partners to PMA-stimulated wild-type PKCepsilon and high specificity for PKCepsilon substrates. Light-dependent recruitment of Opto-PKCepsilon to the plasma membrane revealed the specific phosphorylation of the insulin receptor at Thr 1160 and recruitment to mitochondria the phosphorylation of the complex I subunit NDUFS4 correlating with reduced spare respiratory capacity, respectively. The interactome and phosphoproteome studies confirm the functionality of Opto-PKCepsilon.

Strengths:

AlphaFold simulations enable the design of an optimized Opto-PKCepsilon with respect to dark-light activity. Opto-PKCepsilon is a versatile tool to study the function of PKCepsilon in a precisely controlled manner.

Weaknesses:

Light-controlled PCKepsilon was recently reported by Gada et al. (2022). Ong et al. developed an optimized Opto-PKCepsilon and presented in their manuscript the potential of this tool for controlling signaling pathways. However, some data have to be improved and appropriate controls are still missing for some experiments.

Major comments:

(1) The group of proteins detected as phosphorylated PKC substrates (phospho-Ser PKC substrate antibody) induced by Opto-PKCepsilon varies significantly between Figure 1 C and Figure 2 C. Have the authors any explanation for this? Do both figures show similar areas of the membrane? The size marker indicates that this is not the case.

(2) The ratio of endogenous and exogeneous PCKepsilon is quite different in the experiments shown in Figure 1 C and Figure 2 C. What is the reason for this effect?

(3) In addition to the overall phosphorylation of PKC substrates, the PKCepsilon mutants should be tested for phosphorylation of a known PKCepsilon substrate. The phosphorylation of the insulin receptor at Thr 1160 by Opto-PKCepsilon (see Figure 6) is very convincing and would provide clearer results for comparing the mutants.

(4) The quality of the fluorescence images shown in Figure 5 is poor and should be improved. In addition, a MitoTracker dye for mitochondria labeling should be included to confirm the mitochondrial localization of Opto-PKCepsilon.

(5) Figure S6 shows a light experiment in the absence of insulin, as stated in the headline of the figure legend and in the main text. Does this mean that Figure 6B shows an experiment in which the cells were exposed to light in the presence of insulin? If so, this should be mentioned in the legend of the figure and in the main text. What influence does insulin have on IR phosphorylation at Thr 1160?

(6) The signal of NDUSF4 phosphorylation induced by Opto-PKCepsilon is weak in the experiment shown in Figure 7E. What about the effect of shorter and longer exposure times? How many times was this experiment repeated?

Reviewer #2 (Public review):

Summary:

The authors developed an optogenetic tool (Opto-PKCε) and demonstrated spatiotemporal control of optoPKCε at different subcellular compartments such as the plasma membrane or mitochondria. Signaling outcomes of optoPKCε were characterized by phosphoproteomics and biochemical analysis of downstream signaling effectors.

Strengths:

(1) Conventional strategy to activate PKC often involves activation of multiple downstream signaling pathways. This work showcases an alternative strategy that could help dissect the effect of specific PKC-elicited signaling outcomes.

(2) The differential phosphoproteomic analysis of PKC substrates between PMA stimulation and optoPKCε activation is insightful. A follow-up question is whether co-transfection of CIBN-GFP-CaaX and optoPKCε increases the pool of substrate compared to optoPKCε only, or optoPKCε activation at the plasma membrane is more effective in phosphorylating its substrates?

(3) The finding that PKC activation at the plasma membrane is required for insulin receptor activation is interesting. Why does Thr1160 phosphorylation lead to a reduction of Thr1158/1162/1163? Does "insulin-stimulated" imply that insulin was administrated in the culture during optogenetic stimulation? Also, did the author observe any insulin receptor endocytosis upon optoPKCε activation?

Weaknesses:

(1) When citing the previous work on optogenetics, the reviewer believes a broader scope of papers (reviews) and recent research articles should be cited, especially those that used similar strategies, i.e., membrane translocation followed by oligomerization (of cryptochrome), as reported in this work.

(2) In terms of molecular modeling, how would the author enable AlphaFold3 structure prediction of activated optoPKCε (or the blue-light stimulated state of cryptochrome)? Current methods only describe that "To generate models of the monomer, an amino acid sequence corresponding to Opto-PKCɛ, 2 ATPs and 1 FAD were used as input whereas for the tetramer, copies of Opto-PKCɛ, 8 ATPs and 4 FADs were used as input" (likely missing "four" between "tetramer" and "copies"). However, simply putting four monomers would not ensure that each monomer is in the "activated" state, which involves excitation of the FAD cofactor and likely conformational changes in cryptochrome.

(3) It would be helpful if the authors could help interpret some results. For example, Figure S1: Was the puncta of mCherry-PKCε on the plasma membrane or within the cytosol? Also, why does optoPKCε only work when PKCε is fused at the C-terminus? When screening for the optoPKCε system with the largest light-to-dark contrast, the AGC domain was truncated. What is the physiological function of AGC? Does AGC removal limit PKC's access to its endogenous substrates?