tpHusion: An efficient tool for clonal pH determination in Drosophila

Genetically encoded pH indicators (GEpHI) have emerged as important tools for investigating intracellular pH (pHi) dynamics in Drosophila. However, most of the indicators are based on the Gal4/UAS binary expression system. Here, we report the generation of a ubiquitously-expressed GEpHI. The fusion protein of super ecliptic pHluorin and FusionRed was cloned under the tubulin promoter (tpHusion) to drive it independently of the Gal4/UAS system. The function of tpHusion was validated in various tissues from different developmental stages of Drosophila. Differences in pHi were also indicated correctly in fixed tissues. Finally, we describe the use of tpHusion for comparative analysis of pHi in manipulated clones and the surrounding cells in epithelial tissues. Our findings establish tpHusion as a robust tool for studying pHi in Drosophila.

(1) The figures and figure legends can be better annotated to highlight different regions of the tissue. Dashed lines, arrowheads, or brackets can be used to indicate various regions within the tissue. For example, on the eye discs, a dashed line along the morphogenetic furrow can more clearly show the amf versus the pmf. Similarly, arrowheads pointing out the FSCs versus the follicle cells allows for better interpretation of the distribution of the tpHusion signal in the ovarioles. The wing disc and the larval brain images can also benefit from outlining and labeling the different compartments.
The suggested annotations have been introduced in Figures 2 and 3. Arrows and arrowheads indicate follicle cells and FSCs, a dashed line labels the morphogenetic furrow, the pouch and notum have been clearly separated, and brackets have been used to denote the different regions of the larval brain. Corresponding information has also been included in the associated figure legends.
(2) The dashed squares in the figures are difficult to see. Replace them with solid squares of a different color or represent another way. It would be helpful to have the squares on the images in the single channels too.
The color of the dashed square has been replaced with yellow, and the squares are also added to the single channel images.
(3) Explain the calibration bar a bit more -specifically the colors and the associated numbers.
The calibration bar represents the relative ratio of SEpHluorin and FusionRed intensities within a tissue from low (blue) to high (red). The limits are set to clearly denote the differences over the entire tissue area. The above information is now included in the Materials and Methods section.

(4) Increase the magnification of the ovarioles in figure 2. The magnification of the ovarioles differs between figure 2 and figure 3. At the higher magnification used in figure 3, it is easier to see the cells and the tpHusion signal.
The magnification of ovarioles in Figure 2 has been increased (similar to Figure 3). The corresponding changes to the scale are indicated in the figure legend.
(5) Include low magnification images of the single channels of figure 4 for an unbiased view of the distribution of the tpHusion signal across clones throughout the entire wing disc.
We excluded the low magnification images because the membrane localization and distribution of tpHusion is difficult to view at lower magnification. For reference, a panel of discs containing LacZ, Ras V12 -overexpressing and Tsc1-knockdown clones is shown in Figure R1. If the reviewer prefers, we can include a similar figure for the different conditions tested in Figure 4 as a supplementary figure.
(6) The authors state that the detection of intracellular pH in fixed tissue is similar to that observed in live tissue. However, there appears to be a significant reduction of tpHusion signal in the images of the fixed eye discs. Based on the images presented, the tpHusion signal in the fixed larval brain appears expanded and more intense. Finally, the tpHusion signal in fixed ovarioles seems shifted with more intense FusionRed signal compared to SEpHluorin signal in fixed compared to live tissues. Discuss these differences.
The comparison has been made between the SEpHluorin to FusionRed intensities ratio of different regions. The absolute pHi values cannot be determined for the fixed tissues, so a direct comparison with live tissues is not possible. However, the variability in pHi of different regions within a tissue is conserved based on ratio of intensities. The corresponding statement has also been made clearer in the text.
The requested changes have been made in the manuscript.

Reviewer #2: Gupta et al. describe the development of tpHusion, a membrane associated cytosolic pH sensor. The authors describe their strategy, which relies on expressing the sensor under the control of the tubulin promoter in order to obviate the necessity of the GAL4/UAS system. The utility of the strategy was validated by measurements of pH in different tissues during development, and in both fixed as well as live tissues.
The manuscript is technically sound, and the authors' conclusions are backed by their data. This tool will certainly be useful to researchers interested in studying cytosolic changes during development. There remains, however, one minor concern that could be addressed by additional experiments as detailed below. tpHusion is tagged to the membrane using an HRAS-sequence. Though this is a reasonable strategy to attach the sensor to the PM, there is a concern regarding localization of the sensor to specific domains at the PM. To address the reviewer's concerns about variations in pHi with cholesterol levels in the plasma membrane, we tested the ratio of SEpHluorin and FusionRed intensities after incubation of live wing discs with methyl-β-cyclodextrin (MβCD), as suggested. Starting with 10 mM MβCD concentration (Dason and Charlton, 2014), we did not observe any change in the SEpHluorin and FusionRed intensities or their ratio ( Figure R2) up to 50 min after MβCD incubation. Upon incubation with higher concentrations, no change was observed with 25 mM MβCD, whereas a significant disruption of tissue architecture was observed at 50 mM MβCD after 30 min. Due to the lack of a positive control to test for loss of cholesterol levels at lower concentrations of MβCD, we decided not to include the data in the manuscript. However, we do agree that the tpHusion reporter has to be used with caution to compare pHi in cells with different cholesterol levels. Further work will be required to establish the behavior of tpHusion in dependence of membrane cholesterol levels but we feel that this is beyond the scope of our present manuscript. Figure R2. Effect of MβCD on tpHusion membrane localization. SEpHluorin (SEpH), FusionRed (FR) and ratiometric images of wing pouches incubated in indicated concentrations of MβCD for 10-50 minutes. n > 7 larvae. Data are represented as mean ± standard deviation. Scale bar = 50 µm.