Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 39 (5), 560-571

Long Oskar Controls Mitochondrial Inheritance in Drosophila Melanogaster


Long Oskar Controls Mitochondrial Inheritance in Drosophila Melanogaster

Thomas Ryan Hurd et al. Dev Cell.


Inherited mtDNA mutations cause severe human disease. In most species, mitochondria are inherited maternally through mechanisms that are poorly understood. Genes that specifically control the inheritance of mitochondria in the germline are unknown. Here, we show that the long isoform of the protein Oskar regulates the maternal inheritance of mitochondria in Drosophila melanogaster. We show that, during oogenesis, mitochondria accumulate at the oocyte posterior, concurrent with the bulk streaming and churning of the oocyte cytoplasm. Long Oskar traps and maintains mitochondria at the posterior at the site of primordial germ cell (PGC) formation through an actin-dependent mechanism. Mutating long oskar strongly reduces the number of mtDNA molecules inherited by PGCs. Therefore, Long Oskar ensures germline transmission of mitochondria to the next generation. These results provide molecular insight into how mitochondria are passed from mother to offspring, as well as how they are positioned and asymmetrically partitioned within polarized cells.

Keywords: actin; cytoskeleton; germ cells; germ plasm; localization; maternal inheritance; mitochondria; mitochondrial inheritance; mtDNA; oskar.


Figure 1
Figure 1. mtDNA copy number increases during oogenesis and drastically decreases upon PGC formation in wild type controls
(A) Distinct mtDNA FISH foci are observed in 0 to 1 hour old syncytial embryos. (B) Quantitative mtDNA FISH can accurately determine the mtDNA copy number in embryonic PGCs in situ. mtDNA was measured in PGCs from 1 to 3 hour old control embryos by mtDNA FISH. mtDNA copy number was determined in FACS-sorted PGCs from 0 to 11 hour old embryos by qPCR. Data are the mean +/− SEM. (C) Cartoon of D. melanogaster germline development. (D, E and F) mtDNA in a 2 to 3 hour old embryo (D), a L3 larval gonad (E) and an adult ovariole (F) as visualized using mtDNA FISH. Samples were immunostained with α-Vasa that detects the germline-specific protein Vasa. (G) mtDNA copy number throughout germline development. mtDNA copy number was determined in PGCs, GSCs, egg chambers and eggs using mtDNA FISH. mtDNA copy number in PGCs from 2 to 4 and 9 to 11 hour old embryos was determined by FACS-sorting and qPCR. Data are the means at least three independent replicates +/− SEM. See also Figure S1.
Figure 2
Figure 2. Mitochondria are enriched at the embryo posterior and accumulate there during stage 10 of oogenesis. Arrows indicate posteriorly enriched mitochondria
(A) mtlrRNA and mitochondria were visualized using FISH and mito-EYFP to detect the mtlrRNA and mitochondria, respectively. See also Figure S2. (B) Mitochondria accumulate posteriorly starting at stage 10 of oogenesis and persist there until embryogenesis. mito-EYFP labeled mitochondria were visualized live using a multi-photon microscope. Mitochondrial accumulation (EYFP fluorescence) is shown using an ICA lookup table which ranges from blue (background) to yellow and white (high accumulation). See also Movie S1. (C) Percentage of stage 9 to 14 egg chambers and eggs with posteriorly enriched mitochondria.
Figure 3
Figure 3. Cytoplasmic streaming is required to localize mitochondria to the oocyte posterior
(A) Cartoon of cytoplasmic streaming. (B and C) mito-EYFP tagged mitochondria undergo cytoplasmic streaming in stage 10/11 egg chambers but not in those treated with colcemid. Each panel represents time projections acquired over 0, 40 or 80 seconds (s). EYFP, white. (D) mito-EYFP tagged mitochondria do not accumulate at the posterior of stage 10 egg chambers treated with colcemid (0 min) and aged 100 minutes. See also Movie S2. (E) The speed of mito-EYFP tagged mitochondria in stage 10/11 egg chambers treated with and without colcemid. Data are the mean +/− SEM. (F) Percentage of egg chambers treated with or without colcemid at stage 10 and aged 100 minutes that had posteriorly enriched mitochondria.
Figure 4
Figure 4. Long Oskar and its N-terminal domain are both necessary and sufficient to trap and retain mitochondria at the cortex. Arrows indicate mitochondrial accumulation
(A) Loss of oskar, but not tudor nor vasa, prevents posterior mitochondrial accumulation. 0 to 1 hour old embryos from wild-type, oskar null mutant (oskA87/Df(3R)p-XT103), tudor null mutant (tudortux46/Df(tudor)), or Vasa knockdown (nosgal4::VP16, mito-EYFP/uas-vasa RNAi) mothers were immunostained with α-ATP synthase α. See also Figure S3. (B) Loss of long, but not short, oskar prevents posterior mitochondrial accumulation. 0 to 1 hour old embryos from wild-type mothers, oskar null mutant mothers (oskA87/Df(3R)p-XT103), or oskar null mutant mothers transgenically expressing either Long Oskar (short osk−/−; oskM139L (go M2-12)/+; oskA87/Df(3R)p-XT103) or Short Oskar (long osk−/− ;oskM1L(go M1-7)/+;oskA87/Df(3R)p-XT103) were immunostained with α-ATP synthase α and α-Oskar. (C) The N-terminal domain is sufficient to trap mitochondria at the anterior when it is expressed there. Left, 0 to 1 hour old embryos from wild-type mothers were immunostained with α-ATP synthase α and α-Oskar. Right, 0 to 1 hour old embryos from mothers expressing the N-terminus of Long Oskar fused to mCherry, FLAG, and HA at the anterior using the bcd 3’UTR (N-term osk-bcd 3’UTR: uas-osk N-term::mCherry::3xFLAGHA/+; nosgal4::VP16, mito-EYFP/+) were imaged with EYFP (mitochondria) and mCherry (N-term. Oskar). See also Figure S4.
Figure 5
Figure 5. Long Oskar interacts with the actin cytoskeleton
(A) Silver stained polyacrylamide gel of proteins co-immunoprecipitated with Long Oskar, Short Oskar or the N-terminal domain of Long Oskar (N-term). Proteins were co-immunoprecipitated from 0 to 2 hour old embryos from mothers expressing Long Oskar (;uas-long oskar::mCherry::3xFLAG HA/+; nosgal4::VP16, mito-EYFP/+), Short Oskar (;uas-short oskar::mCherry::3xFLAG HA/+; nosgal4::VP16, mito-EYFP/+) or N-term (;uas-osk N-term::mCherry::3xFLAG HA/+; nosgal4::VP16, mito-EYFP/+). Baits are indicated by red boxes as determined by apparent molecular weight (MW). (B) Ten most abundant proteins enriched in Long and N-term Oskar, but not Short Oskar, co-immunoprecipitations. Protein abundance (Abd.) was normalized to bait abundance. Long/Short and N-term/Short, were calculated by dividing normalized protein abundances. See also Tables S1 and S2. (C) Gene ontology enrichment analysis of proteins that co-immunoprecipitated with Long Oskar and its N-terminal domain, but not Short Oskar. (D–G) F-actin in fixed S2R+ cells expressing either Long Oskar (act-gal4 uas-long oskar::mCherry::3xFLAG HA) (D to F) or Short Oskar (act-gal4 uas-short oskar::mCherry::3xFLAG HA) (G). Images are of single sections (D, F and G) or maximum intensity projections (E). F-actin was visualized with Alexa Fluor® 488 phalloidin (red), Oskar with mCherry (green) and DNA with DAPI (blue). The white dashed circles indicate the Oskar expressing cells.
Figure 6
Figure 6. Posterior trapped mitochondria require the actin cytoskeleton
(A to C) Treatment of stage 12 egg chambers with the actin polymerization inhibitor cytochalsin D (cyto. D) (C), but not the microtubule depolymerizing agent colcemid (colc.) (B) or vehicle control (A), caused mito-EYFP mitochondria to detach from the posterior. Images were taken 0, 1 and 2 hours after the addition of drug. White arrows and dashed shapes indicate posteriorly enriched mitochondria. (D) Percentage of stage 12 colc., cyto. D, or vehicle only (control) treated egg chambers that had discernable posteriorly enriched mitochondria. (E) In the absence of TmII, anteriorly expressed Oskar is not able to retain mitochondria at the embryo anterior. 0 to 1 hour old embryos from mothers transgenically expressing Oskar at the anterior in the presence (P{ry+, osk-bcd}42/+;TmIIgs1/+) or absence (P{ry+, osk-bcd}42/+;TmIIgs1/ TmIIgs) of TmII were immunostained with α-ATP synthase α and α-Oskar. Arrows indicate anteriorly accumulated mitochondria. (F) Percentage of 0 to 1 hours old embryos with posteriorly and/or anteriorly enriched mitochondria. See also Fiugre S5.
Figure 7
Figure 7. Most mitochondria require Long Oskar to enter PGCs
(A) Cartoon of PGC formation. (B) Mitochondrial targeted PAGFP (mito-PAGFP) just prior to photoactivation (left), immediately after photoactivation (middle) and during PGC formation (right). Dashed red box indicates the area of photoactivation. See also Movie S3. (C and D) mtDNA FISH of embryos just after PGC formation from control (C) and long oskar mutant (D) mothers. PGCs were stained with α-Vasa and are delineated by the white dashed shape. See also Figure S6. (E) The number of mtDNA copies per PGC as determined by quantitative mtDNA FISH. Data are the mean +/− SEM. For long osk/+, 99 PGCs from 5 embryos were counted. For long osk−/−, 70 PGCs from 4 embryos were counted. ** p < 0.01 (independent two-tailed Student’s t-test). See also Figure S7. (F) The number of PGCs in cycle 14 embryos from control and long oskar null mutant mothers. Data are the mean +/− SEM. 20 embyros were counted for each genotype. ** p < 0.001 (independent two-tailed Student’s t-test).

Similar articles

See all similar articles

Cited by 17 PubMed Central articles

See all "Cited by" articles

LinkOut - more resources