QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 9 May 2006
doi: 10.1242/jcs.02933

This Article Summary Full Text Full Text (PDF) Supplementary Material Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gatt, M. K. Articles by Glover, D. M. Search for Related Content PubMed PubMed Citation Articles by Gatt, M. K. Articles by Glover, D. M.

The Drosophila phosphatidylinositol transfer protein encoded by vibrator is essential to maintain cleavage-furrow ingression in cytokinesis

Cancer Research UK Cell Cycle Genetics Research Group, Department of Genetics, University of Cambridge, Downing Street, Cambridge, CB2 3EH, UK

View larger version (126K): [in a new window]   Fig. 1. Vibrator mutants show a failure of cytokinesis during male meiosis. (A) Phase-contrast image of a wild-type cyst of 64 onion-stage spermatids showing the expected 1:1 ratio of nuclei (arrow) to nebenkern (arrowhead). (B) Phase-contrast image of a spermatid cyst from a vibrator mutant (vibS110416 / Df(3R)Dl-BX12). Many spermatids have up to four nuclei and a nebenkern of increased size. This particular cyst has 12 cells with the expected 1:1 ratio of nuclei to nebenkern (red arrow), 4 cells with a 2:1 ratio (red arrowhead), 4 cell with a 3:1 ratio (yellow arrow) and 7 cells with a 4:1 ratio (yellow arrowhead). (Note that there are four additional cells outside the frame of this image.) (C) Early elongating vibrator (vibS110416 / Df(3R)Dl-BX12) spermatids that have 1, 2 or 3 nuclei (arrow, arrowhead and asterisk, respectively) associated with the elongating flagellum (stained with anti- tubulin antibody). DNA is stained red. Bars, 10 µm. (D) Schematic image, showing the meiotic divisions of one primary spermatocyte in a cyst of 16 cells; adapted from Fuller (Fuller, 1998) with permission.

View larger version (33K): [in a new window]   Fig. 2. Vibrator alleles and their relative strengths. (A) Schematic representation of P-element insertions within the vibrator gene. The gene spans a region of 8908 bp and is composed of nine exons and eight introns. The predicted gene model (shown in blue) comprises 1178 bp, the open reading frame (red) encodes a protein of 272 aa. The position of P-element insertions is represented by inverted triangles. Sequencing indicated that the insertion vibj7A3 (P-element insertions `A' in panels A and B) lay within the second intron of the gene; the vibj5A6 insertion (P-element insertions `D' in panel A) was within the first intron; the insertions vibS110416 (P-element insertions `B' in panels A-D), vibEP513 (P-element insertions `C' in panels A and B) and vibS045002 (P-element insertions `E' in panels A and 2B) all lay within the first exon. The insertion vibEP651 (P-element insertions `F' in panels A and B) was 222 bp upstream. Chromosomes carrying vibEP513 and vibEP651 had second site mutations, causing early embryonic lethality (Table 1). An additional P-element-insertion site in vibS045002 at 77B;78A was uncovered by Df(3L)ri-79c (77B;78A) (Deak et al., 1997). Work presented here suggests that vibS110416 carries a second site mutation (see footnote of Table 3). Consequently, most analyses were preformed on hemizygotes against the deficiency Df(3R)Dl-BX12. The lethal stage of hemizygous alleles (see Results) indicated the allelic series: vibj7A3>vibS110416>vibEP513 vibj5A6>vibS045002>vibEP651. (B) Levels of vibrator protein are diminished in P-element-mediated mutants. Anti-vibrator antibody recognises a band at an expected molecular mass of 35 kDa in western blots. Lanes 1-4, vibrator protein levels in hemizygous and homozygous vibj7A3 second instar larvae, compared with the balanced stock. The extracts analysed in lanes 1 and 2 or 3 and 4 are from two or four larvae, respectively. Lanes 5-8, extracts of early pupae of the indicated genotypes, representing a series of alleles, all as transheterozygotes with vibS110416. Protein levels reflect mutant strength in the allelic series and are lowest in vibj7A3, at intermediate levels in vibEP513 and vibS045002, and highest in vibEP651. Actin is shown as a loading control. Df stands for the deficiency (Df(3R)Dl-BX12); TM6C is a balancer chromosome. (C) Protein levels are dramatically reduced in vibS110416 / Df(3R)Dl-BX12 testes. Protein extracts of three pairs of wild-type or mutant third instar larval testes are loaded in each lane. A band at the expected mass of 35 kDa is present in Oregon R testes but is absent in the hemizygous mutant. Actin is shown as loading control. (D) Protein levels show little diminution in homozygous vibS110416 larval neuroblasts. Protein extracts of Oregon R or mutant larval neuroblasts were loaded in each lane. Levels of vibrator protein at 35 kDa are not significantly different between mutant and wild-type. -tubulin is shown as the loading control.

View larger version (133K): [in a new window]   Fig. 3. Localisation of vibrator protein in spermatocytes. An antibody raised against vibrator recombinant protein was used in immunolocalisation studies. Microtubules are shown in green, vibrator in red (also shown in monochrome), DNA is in blue. (A) Vibrator localises to membranous structures in primary spermatocytes including the nuclear and plasma membranes. (B) At anaphase, vibrator decorates the region occupied by the central spindle microtubules. This region contains many membranous structures including mitochondria. (C) At the onion stage of early spermatid development vibrator associates with the mitochondrial derivative (nebenkern). Bars, 10 µm.

View larger version (99K): [in a new window]   Fig. 4. Cleavage-ring components localise normally in early cytokinesis and are misplaced later in cytokinesis in vibrator mutants. In all panels microtubules are stained green, DNA is blue. Red staining indicates pavarotti, anillin or peanut as indicated. (A,D) Pavarotti. In wild-type spermatocytes (inset in A), pavarotti becomes positioned to a tight band at the midzone of the interdigitating central spindle microtubules at late anaphase and/or telophase (arrowhead). It is also maintained on the ring canals that persist throughout male meiosis (white arrow). Central spindle microtubules appear to be formed correctly in vibrator primary spermatocytes and pavarotti localises correctly within a tight band on the midzone (arrowhead). However, in some cells the central spindle microtubules appear either not to have constricted or to have partially collapsed, and pavarotti is similarly misplaced (yellow arrow). In vibrator secondary spermatocytes in telophase of meiosis II (D), the failure of cytokinesis in the previous meiotic division has resulted in the formation of two secondary spindles within a common membrane. Bipolarity is nevertheless established with two nuclei in each polar position. Pavarotti localises to the mid-zone of this common central spindle. (B,E) Anillin. In wild-type primary spermatocytes (inset in B), anillin concentrates to a tight ring-like structure (arrow) during telophase and is also persistent in ring canals (arrowhead). In vibrator primary spermatocytes, anillin forms a continuous band across the central spindle region of late anaphase (arrow) and can be seen localised as tight rings (arrowhead) in telophase cells (B). In meiosis II, spindles form around two nuclei within a common membrane in vibrator mutants indicating that cytokinesis failed in the previous meiotic division of these cells (E). This panel also shows an example of ring-like canals within the common cell membrane (arrow). These persistent ring-like canals manifest as disorganised, enlarged rings or as scattered or streaked concentrations of anillin (arrowhead). Notice the enlarged diameter of the ring canals compared with those from previous mitotic divisions (compare arrow with asterisk). (C,F) Peanut. Peanut localisation in a ring at the midzone of wild-type primary spermatocytes (inset in C). In most vibrator primary spermatocytes in which the central spindle region has formed correctly, peanut is concentrated as a tight ring structure at the midzone (C). In one cell, peanut is absent from the central spindle midzone, even though the central spindle microtubules appear to have formed correctly (arrow). Cells in which the central spindle microtubules are disorganised have poor peanut localisation (arrowheads). Tetranucleate vibrator cells at telophase of meiosis II (F) in which peanut is present in a single major ring (arrows). Bars, 10 µm.

View larger version (124K): [in a new window]   Fig. 5. Actin localisation in vibrator spermatocytes. In wild-type primary spermatocytes, actin appears as a compact ring at the spindle midzone (arrow, A) and remains localised in the constricted contractile ring, which forms during cytokinesis (arrow, C). In comparison, actin has a broad and discontinuous localisation in vibrator primary spermatocytes at telophase (arrow, B). Binucleate cells at prophase of meiosis II occur in vibrator secondary spermatocytes that have failed the previous cytokinesis. These cells show a disorganised accumulation of actin within the plasma membrane (arrow, D). Dispersed foci of actin are also present throughout the cell. Microtubules, green; actin, red; DNA, blue. Bars, 10 µm.

View larger version (108K): [in a new window]   Fig. 6. Furrow ingression in wild-type spermatocytes. (A) DIC images and (B) corresponding images from a time-lapse series of a wild-type primary spermatocyte expressing GFP-tagged ß-tubulin (Inoue et al., 2004). Fluorescence shown is the maximum intensity projection and the DIC image is from the central-most focal plane. Time is shown in minutes relative to anaphase onset (0). During metaphase I the nucleus is surrounded by a double-nuclear membrane, three to five layers of double parafusorial membranes and mitochondria which lie just outside, although parallel, to this membrane system (Fuller, 1993) that can be seen as parallel contrasting structures by DIC optics. As the cell enters anaphase the spindle elongates, central spindle microtubules form and the cleavage furrow is initiated (8 minutes, arrows). The furrow ingresses, and compresses the central spindle microtubules. A ring canal, detected in this cell at 19 minutes (arrow), encircles the parafusorial membranes (19 minutes, arrowheads). Dissolution of the aligned parafusorial membranes occurs on average 4 minutes before cleavage (23 minutes), which occurs at 28 minutes as assessed by the appearance of GFP-ß-tubulin at the cell periphery and by the complete breakdown of the parafusorial membranes seen by DIC. The central spindle microtubules are compacted to a maximal point at around 19 minutes and appear to gradually `degrade' as cleavage progresses (19-36 minutes GFP panels). See supplementary material, Movie 1. Bar, 10 µm.

View larger version (150K): [in a new window]   Fig. 7. Loss of vibrator protein affects furrow stability. (A) DIC images and (B) selected corresponding frames from time-lapse imaging of a vibS110416 / Df(3R)Dl-BX12 mutant primary spermatocyte expressing GFP-tagged ß-tubulin. Time (in minutes) is shown relative to anaphase onset (0). In this cell, furrow initiation occurred 12 minutes after anaphase onset. Although the central spindle microtubules form correctly they fail to become fully compressed (38 minutes). The cell remains in this state until 51 minutes after anaphase onset when there is a rapid dissociation of the cleavage-furrow membrane from the central spindle region within 1 minute (notice the disappearance of the membrane at 52 minutes, arrow). Dissolution of the parafusorial membranes does not occur in this cell (38-62 minutes). See supplementary material, Movie 2. Bar, 10 µm.

View larger version (84K): [in a new window]   Fig. 8. Comparison of furrow ingression and cleavage in wild type and mutant vibrator primary spermatocytes. (A,B) Selected images from a time-lapse series of wild-type (Fig. 6, panel A) and vibS110416 / Df(3R)Dl-BX12 (Fig. 7, panel B) primary spermatocytes. (C) Graphic shows measurements of furrow ingression (distance on one focal plane between opposing furrows observed by DIC) and width of the central spindle (cs) (measured from fluorescence of GFP-labelled microtubules). The vibrator mutant shows a delayed furrow initiation after anaphase onset, a slower rate of furrow ingression and a sudden relaxation of the ingressing membrane (52 minutes in B and C). Furthermore, its central spindle does not become tightly compacted when compared with the wild-type spermatocyte. Successful cytokinesis occurred in the control cell at 28 minutes (A, see blue curve in C). Anaphase onset occurred at zero minutes in C. Time in minutes. Bar, 10 µm.