Membrane potential regulates Hedgehog signalling in the Drosophila wing imaginal disc

Abstract

While the membrane potential of cells has been shown to be patterned in some tissues, specific roles for membrane potential in regulating signalling pathways that function during development are still being established. In the Drosophila wing imaginal disc, Hedgehog (Hh) from posterior cells activates a signalling pathway in anterior cells near the boundary which is necessary for boundary maintenance. Here, we show that membrane potential is patterned in the wing disc. Anterior cells near the boundary, where Hh signalling is most active, are more depolarized than posterior cells across the boundary. Elevated expression of the ENaC channel Ripped Pocket (Rpk), observed in these anterior cells, requires Hh. Antagonizing Rpk reduces depolarization and Hh signal transduction. Using genetic and optogenetic manipulations, in both the wing disc and the salivary gland, we show that membrane depolarization promotes membrane localization of Smoothened and augments Hh signalling, independently of Patched. Thus, membrane depolarization and Hh-dependent signalling mutually reinforce each other in cells immediately anterior to the compartment boundary.

Keywords: Drosophila; Hedgehog; membrane potential; signaling.

Figure 1. Membrane potential is patterned in the third‐instar wing imaginal disc

1. A, B

   Live third‐instar discs incubated in DiBAC. DiBAC fluorescence is observed in the wing imaginal disc in both apical (A–A′) and basal (B–B′) optical sections. Increased fluorescence is observed in the centre of the pouch (white arrow, B′) and at the dorsoventral (D‐V) compartment boundary in the anterior compartment (yellow arrow, (B′), and inset, (D)).

2. C–E

   Comparison of DiBAC with the voltage‐insensitive dye FM4‐64. Incubation of live discs in FM4‐64 (E) shows more uniform fluorescence when compared to DiBAC (D). Quantitative comparison of fluorescence in the two white boxes in each panel is shown in (C). N = 7 discs for each treatment, data compared using an unpaired t‐test, ** indicates P < 0.001, error bars are standard deviations. The region boxed in yellow in (D) is shown at higher magnification to show DiBAC fluorescence at the D‐V compartment boundary.

3. F, G

   Incubation in ouabain results in brighter and more uniform DiBAC fluorescence.

4. H–H″

   Live wing discs expressing UAS‐RFP anterior to the compartment boundary, under the control of ptc‐Gal4, were incubated in DiBAC, showing that the stripe of increased DiBAC fluorescence coincides with the posterior edge of ptc expression. White arrowhead in (H) indicates the stripe in the wing pouch; yellow arrowheads indicate the stripe in the dorsal and ventral hinge.

5. I, I′

   Early L3 discs incubated in DiBAC. Patterned depolarization is evident throughout the third larval instar, with the stripe of increased fluorescence becoming narrower in more mature discs with developmental time (Compare I with A, bracket in I indicates width of increased DiBAC fluorescence).

Data information: Scale bars are 100 µm in all panels, except for (A′ and B′), where scale bars are 50 µm.

Figure 2. Expression of endogenous channels is patterned and contributes to depolarization anterior to the compartment boundary

1. A–A″

   Expression of Rpk is increased anterior to the A‐P compartment boundary and in two rows of cells flanking the D‐V compartment boundary in the A compartment.

2. B–B″

   Blockade of DEG/ENaC channels by incubation in amiloride abolishes the stripe of increased DiBAC fluorescence along the A‐P compartment boundary (white arrowhead). White boxes indicate regions used to calculate the average DiBAC fluorescence intensity ratio across the compartment boundary = 0.98, standard deviation = 0.05, n = 5 discs. Increased fluorescence at the D‐V boundary in the A compartment is still observable (yellow arrowhead). The same vehicle was used as in experiments with ouabain, and the control is in Fig 1F.

3. C, C′

   Amiloride incubation does not diminish Rpk expression. White arrowhead indicates approximate position of the A‐P compartment boundary, yellow arrowhead indicates approximate position of the D‐V compartment boundary.

4. D, E

   Expression of rpk‐RNAi using dpp‐Gal4 results in diminished DiBAC fluorescence along the A‐P compartment boundary (white arrowhead), while increased fluorescence at the D‐V boundary in the A compartment is still observable (yellow arrowhead).

Data information: Scale bars are 100 µm, except in (B″), where scale bar is 50 µm.

Figure EV1. rpk knockdown decreases DiBAC staining and Rpk expression at the A‐P compartment boundary

1. A

   Expression of rpk‐RNAi in the wing pouch decreases patterned DiBAC staining (white arrowhead).

2. B

   Expression of rpk‐RNAi in the wing pouch decreases anti‐Rpk staining anterior to the A‐P compartment boundary (white arrowhead).

3. C, D

   Control discs. Increased DiBAC staining and anti‐Rpk staining are visible in the centre of the wing pouch (white arrowheads).

Data information: Scale bars are 100 µm.

Figure 3. Patterned Rpk and ATPα expression and membrane depolarization require Hh signalling

1. A–A‴

   Immunostaining of discs expressing ptc>RFP with antibodies to Rpk and ATPα showing elevated levels of both proteins anterior to the compartment boundary. White arrowheads indicate approximate position of the A‐P compartment boundary.

2. B–C′

   L3 discs that are temperature‐sensitive for hh following shift to the restrictive temperature for 12 h show loss of increased expression of Rpk (B, B′) and ATPα (C, C′) anterior to the compartment boundary. Controls shown in Fig EV2. White arrowheads indicate approximate position of the A‐P compartment boundary.

3. D–E‴

   Clones of cells expressing an activated form of the transcription factor Ci show elevated Rpk and ATPα expression in the anterior compartment (white box) as well as the posterior compartment (yellow arrowhead). (E–E‴) A single clone in the A compartment is shown at higher magnification.

4. F–F′

   A clone of cells in the anterior compartment expressing ci3M following incubation in DiBAC.

5. G–G′

   A clone of cells in the anterior compartment expressing ptc‐RNAi following incubation in DiBAC. The clones also express RFP.

Data information: All scale bars are 100 µm, except for (B′), (C′) and (F–G′) where scale bars are 50 µm and (E–E‴) where scale bars are 25 µm.

Figure EV2. Patterned Rpk and ATPα expression and membrane depolarization require Hh signalling

1. A–A″

   Immunostaining of discs heterozygous for the hh^AC allele and upshifted to 30°C for 12 h with antibodies to Rpk and ATPα showing elevated levels of both proteins anterior to the A‐P compartment boundary. All scale bars are 100 µm.

Figure EV3. Hh pathway activation results in depolarization

1. A–A′

   Clones of cells expressing ci3m accumulate more DiBAC relative to surrounding tissue in both the anterior (white arrow) and posterior (yellow arrow) compartments.

2. B–B′

   Clones of cells expressing ptc‐RNAi accumulate more DiBAC relative to surrounding tissue in the anterior compartment (white arrow), but not the posterior compartment (yellow arrow).

3. C

   The ratio of DiBAC fluorescence in ci3m‐expressing clones to equally‐sized regions of tissue not expressing ci3m. N = 10 clones from three imaginal discs. Individual data points are shown, as well as a box plot showing descriptive statistics. Bounds of the box show lower and upper quartiles, bar within the box shows the median of the data, and whiskers show minimum and maximum bounds of data.

Data information: All scale bars are 100 µm.

Figure EV4. Manipulating levels of the ion channel Rpk impacts both Hh and Wg signal transduction

1. A, B

   Expression patterns of ap‐Gal4 (A) and rn‐Gal4 (B).

2. C, D

   Immunostaining of Smo protein (red), full‐length Ci (light blue) (C–C″) and Ptc protein (D) in discs expressing ATPα^RNAi in the dorsal compartment. Brackets indicate tissue expressing RNAi and control tissue.

3. E–F″

   Immunostaining of Smo protein (red) and full‐length Ci (light blue) in (E–E″) control discs and (F–F″) discs expressing rpk^RNAi for 48 h before dissection.

4. G–H′

   Immunostaining of Ptc protein in (G, G′) control discs and (H, H′) discs expressing rpk^RNAi for 48 h before dissection.

5. I–J″

   Immunostaining for Wg protein in (I–I″) control discs and (J–J″) discs expressing rpk^RNAi for 48 h before dissection.

6. K–L′

   Immunostaining for Cut protein in (K, K′) control discs and (L, L′), discs expressing rpk^RNAi for 48 h before dissection.

Data information: All scale bars are 100 µm except (D), where scale bar is 50 µm.

Figure 4. The ENaC channel Rpk is required for Hh signal transduction

1. A–A‴

   Immunostaining of Smo protein (red) and full‐length Ci (light blue) in discs expressing en‐Venus. Smo is expressed at higher levels in the P compartment and in a stripe 5–10 cells wide just anterior to the A‐P compartment boundary. White arrowhead indicates cells just anterior to the A‐P compartment boundary.

2. B, C

   Effect of expressing an RNAi against the ENaC channel rpk in the dorsal compartment of the disc using ap‐Gal4. Knockdown of Rpk in the dorsal compartment results in decreased accumulation of the Hh signal transducer Smo (B, B′) and the activator form of Ci (B, B″). The expression of ptc, visualized using an anti‐Ptc antibody (C), a downstream target gene of Hh signalling, is also diminished. The rpk‐RNAi line BL:39053 is used in all panels shown, but the phenotype was validated with a second RNAi line (BL:25847).

Data information: All scale bars are 50 µm, except for (A) and (B), where scale bars are 100 µm.

Figure 5. Membrane potential modulates Smo membrane abundance and downstream Hh signal transduction in larval salivary glands

1. A–Q

   Manipulation of V_mem in salivary glands dissected from third‐instar larvae. (A–C) Localization of Smo protein (blue) following expression of 71B‐Gal4 (A), 71B‐Gal4 and UAS‐NaChBac (B) or 71B‐Gal4 and UAS‐Kir2.1 (C). Expression of the sodium channel NaChBac increases membrane localization of Smo while expression of the potassium channel Kir2.1 reduces membrane localization and increases cytosolic Smo (B, C). (D) Quantification of the ratio of membrane Smo fluorescence to cytosol fluorescence. N = 7 salivary glands per genotype. Data were compared using ANOVA followed by Tukey′s test for significance (** indicates P < 0.01, *** indicates P < 0.001), error bars are standard deviations. (E, F) Expression of the Hh target gene ptc is visualized using anti‐Ptc immunostaining. Compared to 71B‐Gal4 (E), Ptc expression is increased in glands expressing 71B‐Gal4 and UAS‐NaChBac (F). (G‐I) Effect of expression and activation of the depolarizing channelrhodopsin ChR2 on Smo protein membrane abundance. After dissection, glands were kept in either dark conditions (G) or exposed to 15 min of blue light (H). (I) Quantification of membrane fluorescence; N = 13 glands, data compared using an unpaired t‐test (** indicates P < 0.01), error bars are standard deviations. (J–L) Effect of expression and activation of the Cl⁻‐selective channelrhodopsin ChloC. An overall reduction in Smo immunostaining was observed following channel activation (K), as compared to control glands kept in the dark (J). (L) Quantification of membrane fluorescence; N = 10 glands, data were compared using an unpaired t‐test (*** indicates P < 0.001), error bars are standard deviations. (M–Q) Time course of change in Smo localization. Salivary glands expressing the channelrhodopsin ChR2 were subjected to increasing intervals of activating light, then fixed and stained for Smo protein (M–P″). (Q) Quantification of membrane‐associated fluorescence; N = 7 glands/timepoint, data were compared using ANOVA followed by Tukey′s test for significance (** indicates P < 0.01), error bars are standard deviations. Scale bars are 50 µm, except for (A–C), where scale bars are 25 µm.

Figure EV5. Optogenetic manipulation of V_mem in Drosophila salivary glands

1. A–G′

   Salivary glands expressing the channelrhodopsin ReaChR were subjected to 15 min in activating light, then fixed and stained for Smo protein (A–C). N = 9 glands, data were compared using an unpaired t‐test (***P < 0.001), error bars are standard deviations. Membrane abundance of the beta integrin subunit Mys and the integrin‐associated protein Talin are not altered upon ChR2 activation (D–G′). Scale bars are 50 µm in all panels.

Figure 6. V_mem regulates Smo membrane localization independently of Ptc in the wing imaginal disc

1. A–B″

   Wing discs expressing the channelrhodopsin ChR2 were subjected to 0 min or 25 min of activating light in culture, then fixed and stained for Smo protein. In (B″), the yellow arrowhead indicates dorsoventral compartment boundary, and the white arrowhead indicates anteroposterior compartment boundary.

2. C

   Quantification of the ratio of mean Smo fluorescence intensity in 25 µm² regions in the dorsal anterior compartment compared to the ventral posterior compartment. N = 5 for 0 min discs, n = 9 for 25 min discs, data were compared using an unpaired t‐test (***P < 0.001), error bars are standard deviations. Identical analyses were carried out comparing dorsal posterior fluorescence to ventral posterior fluorescence (P = 0.132), dorsal posterior fluorescence to dorsal anterior fluorescence (P = 0.12) and ventral posterior fluorescence to ventral anterior fluorescence (P = 0.36).

3. D–E′

   Discs expressing the red light‐activated channelrhodopsin ReaChR under control of ap‐Gal4 were raised on a 12‐h light/dark cycle, dissected and stained for Smo protein (D, D′) or Ptc protein (E, E′).

Data information: Scale bars are 100 µm in all panels except (B″) and (E′), where scale bars are 50 µm.

Figure 7. Membrane potential modulates Hh signalling at the anteroposterior compartment boundary

Cells in the posterior compartment (blue) produce the short‐range morphogen Hedgehog, which is received by cells immediately anterior to the compartment boundary, in the anterior compartment (dark green). Hh signal releases Ptc‐mediated inhibition of Smo, allowing for the stabilization of the transcriptional co‐effector Ci, and increased expression of the ENaC channel Rpk, which depolarizes cells immediately anterior to the compartment boundary. Depolarization results in increased membrane abundance of the Hh activator Smo and increased transduction of Hh signal. This, in turn, increases expression of Rpk, reinforcing high levels of Hh signal transduction in the region immediately anterior to the compartment boundary.