Gatekeeper function for Short stop at the ring canals of the Drosophila ovary

Abstract

Growth of the Drosophila oocyte requires transport of cytoplasmic materials from the interconnected sister cells (nurse cells) through ring canals, the cytoplasmic bridges that remained open after incomplete germ cell division. Given the open nature of the ring canals, it is unclear how the direction of transport through the ring canal is controlled. In this work, we show that a single Drosophila spectraplakin Short stop (Shot) controls the direction of flow from nurse cells to the oocyte. Knockdown of shot changes the direction of transport through the ring canals from unidirectional (toward the oocyte) to bidirectional. After shot knockdown, the oocyte stops growing, resulting in a characteristic small oocyte phenotype. In agreement with this transport-directing function of Shot, we find that it is localized at the asymmetric actin baskets on the nurse cell side of the ring canals. In wild-type egg chambers, microtubules localized in the ring canals have uniform polarity (minus ends toward the oocyte), while in the absence of Shot, these microtubules have mixed polarity. Together, we propose that Shot functions as a gatekeeper directing transport from nurse cells to the oocyte via the organization of microtubule tracks to facilitate the transport driven by the minus-end-directed microtubule motor cytoplasmic dynein. VIDEO ABSTRACT.

Keywords: Drosophila; actin; cytoskeleton; dynein; intercellular transport; microtubules; molecular motor; oocyte; polarity; spectraplakin.

Figure 1.. Shot is required for Drosophila oocyte growth.

(A) A diagram of Shot domains and Shot crosslinking of microtubules and F-actin. Three independent shot-RNAi lines were used in this study, targeting ABD, Rod and EGC domains, respectively. shot^Rod-RNAi and shot^EGC-RNAi target all 22 isoforms of shot mRNA; shot^ABD-RNAi does not target 8 isoforms of shot mRNA lacking CH1 domain (RC, RH, RP, RX, RY, RAB, RAC, RAD). (B) A schematic illustration of Drosophila oogenesis in one ovariole and the shot-RNAi knockdown strategy to bypass the requirement of Shot in oocyte specification. Oocyte is shown in darker grey, while nurse cells are represented in lighter grey in the egg chambers. The mat αtub-Gal4^[V37] line starts the expression in stage 2–3 egg chambers, after the completion of oocyte specification. (C-E) Representative images of Rhodamine-conjugated phalloidin staining in control (mat αtub-Gal4^[V37]/+), shot^ABD-RNAi (mat αtub-Gal4^[V37]/UAS-shot^ABD-RNAi), and shot^EGC-RNAi (mat αtub-Gal4^[V37]/UAS-shot^EGC-RNAi). (F) Summary of phalloidin staining phenotypes in control, shot^ABD-RNAi and shot^EGC-RNAi. (G) Summary of GFP-Pav labeling and Orb staining phenotypes in control, shot^ABD-RNAi and shot^EGC-RNAi. (H-J’) Representative images of GFP-Pav labeling (H-J) and Orb staining (H’-J’) in control (ubi-GFP-Pav/+; mat αtub-Gal4^[V37]/+), shot^ABD-RNAi (ubi-GFP-Pav/+; mat αtub-Gal4^[V37]/UAS-shot^ABD-RNAi), and shot^EGC-RNAi (ubi-GFP-Pav/+; mat αtub-Gal4^[V37]/UAS-shot^EGC-RNAi). Oocytes are highlighted by orange arrowheads or brackets. Scale bars, 50 μm. See also Figure S1 and Videos S1–S2.

Figure 2.. Actin binding and microtubule interacting domains of Shot are essential for oocyte growth.

(A) Diagrams of the full-length Shot and truncated mutants. (B) Summary of Orb and phalloidin staining phenotypes in control (mat αtub-Gal4^[V37]/+), shot^Rod-RNAi (mat αtub-Gal4^[V37]/UASp-shot^Rod-RNAi), and shot^Rod-RNAi +shotARod-GFP (UASt-shot.L(A)ΔRod-GFP/+; mat αtub-Gal4^[V37]/UASp-shot^Rod-RNAi). (C) A schematic illustration of knockdown of wild-type Shot by shot-RNAi in shot truncated mutant heterozygous background. KD, knockdown. (D) Summary of Orb and phalloidin staining in listed phenotypes. Unlike one copy of shot^WT, one copy of shot^ΔABD or shot^ΔEGC is unable to drive normal oocyte growth.

Figure 3.. Shot controls directionality of cargo transport from the nurse cells to the oocyte.

(A-C) Golgi transport in the nurse cell-oocyte ring canals in control (A) and in shot-RNAi (B). Golgi are labeled with RFP-tagged human galactosyltransferase (GalT) (RFP-Golgi). (C) Quantification of Golgi transport directions in control and in shot-RNAi. Chi-square test between control and shot-RNAi, p-value < 0.00001 (****). (D-F) Staufen RNP transport in the nurse cell-oocyte ring canals in control (D) and in shot-RNAi (E). Staufen RPNs are labeled with RFP-tagged Staufen (RFP-Staufen). (F) Quantification of Staufen transport directions in control and in shot-RNAi. Chi-square test between control and shot-RNAi, p-value < 0.00001 (****). (H-J) Mitochondria transport in the nurse cell-oocyte ring canals in control (H) and in shot-RNAi (I). Mitochondria are labeled with Mito-MoxMaple3 (red channel, after global photoconversion). (J) Quantification of total mitochondria fluorescence intensity (mean ± 95% confidence interval) in control (N=18) and in shot-RNAi (N=22) oocytes. Unpaired t test with Welch’s correction between control and shot-RNAi, p-value < 0.0001 (****). (L-N) Transport of lipid droplets in the nurse cell-oocyte ring canals in control (L) and in shot-RNAi (M). Lipid droplets are labeled with GFP-tagged lipid droplet domain of Drosophila protein Klar (GFP-LD). (N) Quantification of lipid droplet total fluorescence intensity and average fluorescence intensity (inset) (mean ± 95% confidence interval) in control (N=33) and in shot-RNAi (N=28) oocytes. Unpaired t test with Welch’s correction of total fluorescence intensity of GFP-LD between control and shot-RNAi, p-value < 0.0001 (****); Unpaired t test with Welch’s correction of average fluorescence intensity of GFP-LD between control and shot-RNAi, p-value < 0.0001 (****). Left side: the nurse cells; right side, the oocyte; small oocytes in shot-RNAi are pointed with the white arrowheads; ring canals are labeled with either GFP-Pav (A-B, D-E, H-I) or F-Tractin-tdTomato (L-M); inverted kymographs were created along a ~3.7 μm-width line from the nurse cell (N, left side) to the oocyte (O, right side) through the ring canals (marked as capped lines underneath in the kymographs) in the white dashed box areas; scale bars, 50 μm. See also Videos S3–S7.

Figure 4.. Shot is localized at the asymmetric actin fibers of the nurse cell-oocyte ring canals.

(A-A’) Rhodamine-conjugated phalloidin staining shows asymmetric actin fibers (the white dashed box) at the ring canal (ring canal inner rim is labeled with GFP-Pavarotti) on the nurse cell side, not on the oocyte side. (B) Quantification of the length of actin fibers on the nurse cell side. The lengths of the four longest actin fibers were measured for each ring canal (59 ring canals from 15 control egg chambers). The average actin fiber length on the nurse cell side is 12.0 ± 0.7 μm (mean ± 95% confidence interval). (C) Asymmetric actin fibers, labeled with TagRFP-tagged LifeAct, are seen at all four ring canals connecting nurse cells and the oocyte in a live sample. See also Video S8. (D-D’) A representative image of Shot antibody staining in a TagRFP-LifeAct-expressing egg chamber. Shot is localized at the asymmetric actin fibers on the nurse cell side of the ring canal, but it is not concentrated in the F-actin core of the ring canal inner rim. (E) Schematic illustrations of a stage 8 Drosophila egg chamber and an interconnected 16-cell germline cyst, including 1 oocyte (cell 1) and 15 nurse cells (cells 2–16, numbered according to the order of cell divisions). Ring canals are categorized depends on their relative distance to the oocyte [62] and are color-coded: (1) nurse cell-oocyte ring canals, directly connected to the oocyte, green, “O”; (2) posterior nurse cell-nurse cell ring canal, having one nurse cell between this ring canal and the oocyte, orange, “P”; (3) middle nurse cell-nurse cell ring canal, having two nurse cells between this ring canal and the oocyte, magenta, “M”; (4) anterior nurse cell-nurse cell ring canal, having three nurse cells between this ring canal and the oocyte, blue, “A”. (F) The asymmetry of actin fibers is quantified as the ratio of LifeAct-TagRFP fluorescence signal on the anterior side to the signal on the posterior side of the ring canals (see more details in Materials and Methods). Numbers of ring canals for each type: anterior nurse cell-nurse cell ring canals (A), N=17; middle nurse cell-nurse cell ring canals (M), N=46; posterior nurse cell-nurse cell ring canals, N=55; nurse cell- oocyte ring canals (O), N=70. Unpaired t tests with Welch’s correction were performed in following groups: “O” and “P”, p<0.0001 (****); “O” and “M”, p<0.0001 (****); “O” and “A”, p<0.0001 (****); “P” and “M”, p=0.0024 (**); “P” and “A”, p<0.0001 (****); “M” and “A”, p=0.0143 (*). (G) A representative image of a stage 8 egg chamber expressing LifeAct-TagRFP. Four types of ring canals are highlighted in colored boxes with zoom-in images below. N, nurse cell; scale bars, 10 μm (A, C, D-D’) and 50 μm (G). See also Figure S2 and Video S8.

Figure 5.. Shot controls microtubule polarity in the nurse cell-oocyte ring canals.

(A-B) Stable microtubule organization is not affected by shot knockdown. (A) In control, stable microtubules are localized at the ring canals between the nurse cell and the oocyte (A’) and between two nurse cells (A”). (B) Knockdown of shot does not change stable microtubule distribution at the ring canals between the nurse cells and the oocyte (B’) and between two nurse cells (B”). Stable microtubules are labeled by overexpressed GFP-tagged Patronin, and ring canals are labeled with Rho-Phalloidin staining. Scale bars, 50 μm. (C-E) Knockdown of shot results in a mixed orientation of microtubules in the ring canals. (C) EBI-GFP-labeled microtubule +end comets in the ring canal (labeled by GFP-Pav) connecting a nurse cell and an oocyte in control. (C’) A color-coded hyperstack of the EB1 comet movement of (C). (C”) A kymograph of EB1 comet movement at the ring canal (the white dashed box in C) in control. (D) EBI-GFP-labeled microtubule +end comets at the ring canal (labeled by GFP-Pav) connecting two nurse cells and an oocyte in shot-RNAi. (D’) A color-coded hyperstack of the EB1 comet movement of (D). (D”) A kymograph of EB1 comet movement at the ring canal (the white dashed box in D) in shot-RNAi. (E) Quantification of the fraction of EB1 comets moving through the ring canals from the oocyte towards the nurse cells in control and in shot-RNAi. Control, N=30; shot^EGC-RNAi, N=22; unpaired t test with Welch’s correction between control and shot-RNAi, p-value < 0.0001(****). (F) Quantification of EB1 comet numbers in the nurse cell-oocyte ring canals in control and shot-RNAi. Control, N=30; shot^EGC-RNAi, N=22; unpaired t test with Welch’s correction between control and shot-RNAi, p-value= 0.0004 (***). (G-G”) A representative image of dual labeling of microtubules (by FITC-conjugated tubulin antibody) and F-actin (by Rho-Phalloidin) at a nurse cell-oocyte ring canal. Alignment of microtubules on asymmetric actin filaments are highlighted by small white arrowheads. (C-D and G-G’) N, nurse cell; O, oocyte; scale bars, 10μm. See also Figure S2 and Videos S9–S10.

Figure 6.. Shot is a gatekeeper at the ring canal for Drosophila oocyte growth.

Shot controls microtubule orientation in the ring canal, via regulating the interaction between EBI/microtubule plus-ends and asymmetric actin fibers on the nurse cell side. Therefore, Shot is essential for directing dynein-dependent transport of various cargoes (including Golgi units, osk/Staufen RNPs, mitochondria, and lipid droplets) from the nurse cells to the oocyte. See also Figure S3 and Video S11.