The cells that exhibited uniform expression showed robust activation after SJN 2511 ALK inhibitor ionophore stimulation, whereas the AG cells showed significantly decreased levels of response signal, most likely owing to the unfavorable aggregation of Camui��- mGsR. We also carried out the same set of experiments with mCherry-based Camui�� mutants, and found that the response variability of Camui��4m-mGmC was smaller than that of Camui��- mGmC, similar to the results for the sREACh-based Camui��. However, the sREACh CX-4945 version of Camui�� showed a greater response signal compared to the mCherry version, consistent with the previously reported results showing the greater maturation efficiency of sREACh compared with mCherry. Among the identified four mutations, F394L, A430T, and I434T face the binding regions of an adjacent association domain. This fact raises the possibility that these mutations may inhibit proper association between the CaMKII subunits. Therefore, to test if oligomerization was hindered, we co-transfected HeLa cells with Camui��4m-mGsR and CaMKII�� and immunoprecipitated the proteins using an anti-GFP antibody. Western blotting revealed the presence of two bands corresponding to the molecular weights of Camui��4m-mGsR and CaMKII��. In addition, the ratio of the band intensities of Camui��4m-mGsR and CaMKII�� was similar to that of Camui��-mGsR and CaMKII��, suggesting that Camui��4m-mGsR is incorporated into CaMKII�� in a similar way to Camui��. However, when CaMKII��4m and Camui��4m-mGsR were co-expressed and immunoprecipitated, the interaction between CaMKII��4m and Camui��4m-mGsR was dramatically decreased, suggesting that the association domain with four mutations does not form a homo-oligomer. Consistent results were obtained by monitoring the inter-molecular FRET in living HeLa cells using 2pFLIM. These facts could explain that the reason for the smaller basal fluorescence lifetime of Camui�� in Figs. 6 and 8 may be due to the inter-molecular FRET between Camui�� sensors, further increasing basal FRET, because Camui�� association domain forms tight oligomer compared with association domain with four mutations. Next, to identify the mutation that hinders the oligomeric formation, we carried out native-PAGE analysis and found that F394L and A434T hinder oligomer formation. Furthermore, we tested the Ca2+/calmodulin-dependent autophosphorylation capabilities of Camui��4m-mGsR in HeLa cells by western blotting with anti-phospho antibody. After stimulation of the cells with ionophore, Camui��4m was shown to autophosphorylate in a similar manner to the wild-type Camui��-mGsR, suggesting that Camui��4m-mGsR is activated in a similar way. Since CaMKII�� is specifically expressed in neurons, Camui��4m-mGsR will be used in neurons for monitoring CaMKII�� activity. Therefore, we tested if Camui��4m-mGsR expression alters the dendritic spine density of hippocampal neurons, compared with Camui��-mGsR, and found that there was no significant difference between Camui��-mGsR and Camui��4m-mGsR, showing that there is no unfavorable effect on the neuronal morphology. Furthermore, we monitored Camui��4m-mGsR activity during spine enlargement upon local glutamate uncaging, and found that both activity and spine volume changes are similar to those of Camui��-mGsR. In contrast, Camui��4m-mGsR with T286A mutation which abolishes autophosphorylation showed the decreased activation and spine volume change, consistent with the previously reported result with Camui��. Taken together, these results indicate that Camui��4m-mGsR may be useful for monitoring CaMKII activity. We have demonstrated here that molecular evolution of the association domain in the Camui�� FRET sensor successfully improves its expression pattern and minimizes response variability.