SCAR is a primary regulator of Arp2/3-dependent morphological events in Drosophila

Abstract

The Arp2/3 complex and its activators, Scar/WAVE and Wiskott-Aldrich Syndrome protein (WASp), promote actin polymerization in vitro and have been proposed to influence cell shape and motility in vivo. We demonstrate that the Drosophila Scar homologue, SCAR, localizes to actin-rich structures and is required for normal cell morphology in multiple cell types throughout development. In particular, SCAR function is essential for cytoplasmic organization in the blastoderm, axon development in the central nervous system, egg chamber structure during oogenesis, and adult eye morphology. Highly similar developmental requirements are found for subunits of the Arp2/3 complex. In the blastoderm, SCAR and Arp2/3 mutations result in a reduction in the amount of cortical filamentous actin and the disruption of dynamically regulated actin structures. Remarkably, the single Drosophila WASp homologue, Wasp, is largely dispensable for these numerous Arp2/3-dependent functions, whereas SCAR does not contribute to cell fate decisions in which Wasp and Arp2/3 play an essential role. These results identify SCAR as a major component of Arp2/3-dependent cell morphology during Drosophila development and demonstrate that the Arp2/3 complex can govern distinct cell biological events in response to SCAR and Wasp regulation.

Figure 1.

Molecular genetics of SCAR . (A) Genomic organization of the SCAR transcription unit. Boxes represent exons, and lines represent introns. The 253 nucleotide 5′ UTR and the 730 nucleotide longest 3′ UTR are shaded black. The site of the P-element insertion l(2)k13811 within the first exon and the extent of the Δ37 excision are indicated. Hatched boxes mark the ORFs of neighboring transcription units CG6105 and piwi. (B) Domain structure of Drosophila SCAR. This includes the regulatory SCAR homology domain (SHD), unique to this protein family, a proline-rich region (PR), and the COOH-terminal actin/Arp2/3-interacting WA domain shared by all WASp/Scar proteins. Amino acid residue numbers at domain boundaries and sequence homology percentages between Drosophila SCAR and human Scar/WAVE-1 are indicated. (C) Alignment of Scar protein sequences from Dictyostelium (Dict Scar), C. elegans (Ce Scar), Drosophila (Dm SCAR), and human (hScar/WAVE 1, 2, and 3). Homologies are boxed, and identities are boxed and shaded. The SCAR transcript was defined by sequencing of the LP11386, SD10808, and SD02991 ESTs (Rubin et al., 2000). LP11386 differs at amino acids L¹⁹⁴ and S¹⁹⁵. The SCAR transcript is described in the GadFly Genome Annotation Database (CG4636) and sequence data is available from GenBank/EMBL/DDBJ under accession no. AF247763.

Figure 1.

Molecular genetics of SCAR . (A) Genomic organization of the SCAR transcription unit. Boxes represent exons, and lines represent introns. The 253 nucleotide 5′ UTR and the 730 nucleotide longest 3′ UTR are shaded black. The site of the P-element insertion l(2)k13811 within the first exon and the extent of the Δ37 excision are indicated. Hatched boxes mark the ORFs of neighboring transcription units CG6105 and piwi. (B) Domain structure of Drosophila SCAR. This includes the regulatory SCAR homology domain (SHD), unique to this protein family, a proline-rich region (PR), and the COOH-terminal actin/Arp2/3-interacting WA domain shared by all WASp/Scar proteins. Amino acid residue numbers at domain boundaries and sequence homology percentages between Drosophila SCAR and human Scar/WAVE-1 are indicated. (C) Alignment of Scar protein sequences from Dictyostelium (Dict Scar), C. elegans (Ce Scar), Drosophila (Dm SCAR), and human (hScar/WAVE 1, 2, and 3). Homologies are boxed, and identities are boxed and shaded. The SCAR transcript was defined by sequencing of the LP11386, SD10808, and SD02991 ESTs (Rubin et al., 2000). LP11386 differs at amino acids L¹⁹⁴ and S¹⁹⁵. The SCAR transcript is described in the GadFly Genome Annotation Database (CG4636) and sequence data is available from GenBank/EMBL/DDBJ under accession no. AF247763.

Figure 2.

SCAR protein expression in embryos. Anti-SCAR polyclonal antibody (top row); filamentous actin labeled with phalloidin (bottom row). In wild-type embryos at cycle 12, SCAR protein colocalizes with actin in interphase (A, SCAR; G, actin) and metaphase (C, SCAR; I, actin). SCAR staining is reduced in cycle 12 SCAR ^mat mutant embryos at interphase (B) or metaphase (D), indicating that this antibody detects SCAR protein. SCAR ^mat blastoderm embryos were stained in the same tube as wild-type control embryos from oskar ¹⁶⁶ mutant mothers, identified by the absence of pole cells (Lehmann and Nusslein-Volhard, 1986). Note that actin structures appear disrupted in SCAR mutants (J); these defects are discussed below. In wild-type Oregon R embryos at stage 13, SCAR protein is enriched in growing axons (E) that also stain for filamentous actin (K). SCAR protein persists in later CNS axons (F) costained with actin (L). Bars, 10 μm.

Figure 3.

Defects in nuclear arrangement and morphology in SCAR and Arpc1 mutant embryos. (A–H) Surface views of syncytial embryos. Cortical nuclei are uniformly distributed in wild-type cycle 12 (A) and 13 (B) embryos. Nuclei exhibit abnormal spacing and morphology in SCAR ^mat cycle 12 (C) and 13 (D) embryos. Similar defects occur in Arpc1 ^mat cycle 12 (E) and 13 (F) embryos. Wsp ^mat cycle 12 (G) and 13 (H) embryos have wild-type nuclear spacing. (I–P) Cross-sections of syncytial embryos. Wild-type embryos at cycle 13 (I) and 14 (J) exhibit a uniform layer of surface nuclei and a subset of central yolk nuclei. In SCAR ^mat embryos, nuclei occasionally recede from the surface at cycle 13 (K) with a dramatic internal accumulation of nuclei by cycle 14 (L). Arpc1 ^mat embryos display mild nuclear disruption at cycle 13 (M) and severe internal accumulation of nuclei by cycle 14 (N). In contrast, Wsp ^mat cycle 13 (O) and 14 (P) embryos exhibit wild-type nuclear arrangement. Bars: (A–H) 10 μm; (I–P) 50 μm.

Figure 4.

Interphase actin structures in SCAR and Arpc1 mutant embryos. Organization of interphase actin caps (first and third columns) and corresponding nuclei (second and fourth columns) are shown. All embryos were fixed, stained, and imaged under identical conditions (except A, inset), allowing for direct comparison. In wild-type embryos (A and C), a cap of actin is present above each nucleus. Wild-type caps are dome shaped, leading to an apparent enrichment of actin at the edges in thick surface views (A and C). In a cycle 11 SCAR ^mat embryo (E), actin caps appear slightly smaller than in wild-type. SCAR ^mat actin caps are also flatter than wild-type, resembling a surface-most thinner section of a wild-type cap (A, inset). In a cycle 12 SCAR ^mat embryo (G), actin caps appear less discrete and are absent in regions above clustered nuclei. In Arpc1 ^mat embryos (I and K), actin is depleted above each nucleus (also Fig. 5 Q). Note that SCAR and Arpc1 consistently display lower levels of fluorescence than wild type. Bar, 10 μm.

Figure 5.

Metaphase actin furrows are defective in SCAR and Arpc1 embryos. Filamentous actin structures (phalloidin in green) and mitotic spindles (tubulin in red and DNA in blue) during cycles 11 (columns 1, 2, 5) and 12 (columns 3 and 4). Actin and spindle staining are superimposed in the insets (columns 2 and 4). All embryos are mitotic except E, K, and Q, which are in interphase and not stained for tubulin. Columns 1–4 are surface views, and column 5 shows cross-sections. In a mitotic wild-type cycle 11 (A and B) or 12 (C and D) embryo, actin is present in metaphase furrows between spindles. In cross-section, actin caps lie above each interphase nucleus (E, arrowheads), whereas actin furrows form transient invaginations between mitotic spindles (F, arrows). In a cycle 11 SCAR ^mat embryo (G and H), abnormal surface actin structures overlie individual spindles and fail to form metaphase furrows. In an occasional SCAR ^mat cycle 12 embryo (I and J), a partial metaphase furrow network forms. Cross-sections of SCAR ^mat embryos show actin caps above each interphase nucleus (K, arrowheads) and aberrant actin structures above each mitotic spindle (L, arrows). In an Arpc1 ^mat embryo (M and N), actin does not form metaphase furrows and is depleted in the region above each spindle. In contrast, Wsp ^mat embryos form normal metaphase furrows (O and P). Cross-sections of Arpc1 ^mat embryos demonstrate actin depletion above interphase nuclei (Q, arrowheads) and mitotic spindles (R, arrows). Bar, 10 μm.

Figure 6.

Reduced filamentous actin in SCAR and Arpc1 mutant embryos. Fluorescence intensity is indicated in arbitrary units on the y axis. Error bars represent the standard error of the mean and depict the variability between embryos. Embryos were staged according to nuclear morphology (Foe et al., 1993, 2000) and grouped according to syncytial division number and cell cycle stage. To allow for simultaneous staining, embryos derived from oskar mutant mothers (Lehmann and Nusslein-Volhard, 1986) served as controls. Such embryos, in which cortical actin structures form normally, are readily distinguished from SCAR ^mat and Arpc1 ^mat embryos by the absence of pole cells. (A) SCAR ^mat embryos display comparable actin levels to oskar controls at the onset of cycle 11 followed by a reduction in filamentous actin during later syncytial divisions. Control embryos in anaphase and telophase display less actin than at other points in the cell cycle, consistent with previous observations (Foe et al., 2000). 66 SCAR ^mat and 67 oskar embryos were scored. (B) Arpc1 ^mat embryos exhibit a more extreme reduction in actin levels throughout syncytial divisions 11–13. 62 Arpc1 ^mat and 54 oskar embryos were scored. Cell cycle–dependent fluctuations in filamentous actin levels persist in SCAR ^mat and Arpc1 ^mat. Similar results were obtained when SCAR and Arpc1 were compared with wild-type Oregon R embryos processed in separate tubes. Combining syncytial stages 10–13 (except ana/telophase), SCAR exhibited 56 ± 8% of the wild type 100 ± 6% actin levels (n = 19 SCAR ^mat, 32 Oregon R embryos) and Arpc1 exhibited 17 ± 4% of the wild type 100 ± 7% actin levels (n = 17 Arpc1^mat, 24 Oregon R embryos).

Figure 7.

CNS axon morphology is disrupted by mutations in SCAR, Wasp, and the Arp2/3 complex components Arp3 and Arpc1 . CNS axons were visualized with the axon-specific BP102 antibody. (A–H) Ventral view of stage 14/15 embryos; three to four segments in each panel. (A) In wild-type embryos, CNS axons travel in two longitudinal bundles and two commissural bundles in each segment. (B) In a SCAR ^k13811 homozygous mutant embryo, CNS axon morphology appears wild type. SCAR ^Δ37 homozygotes were also wild type (unpublished data). (C) In an embryo with reduced maternal and zygotic SCAR ^k13811 function (SCAR ^mat/zyg), gaps appear in both longitudinal and commissural bundles. (D) In an embryo completely lacking maternal and zygotic Wsp function (Wsp ^mat/zyg), axon bundles appear thicker and defasciculated; in one segment, axons collapse at the midline. (E) In an Arp3/Df(3L)pbl-X1 embryo, breaks in the commissural bundles are observed. (F) In an Arpc1 ^Q25sd/Df(2L)b84a9; Arp3/+ embryo, commissural axons are severely reduced. (G) In a SCAR ^k13811 /SCAR ^k13811; Wsp ¹ /+ embryo, there is an apparent depletion of axons in both longitudinal and commissural bundles. (H) In a SCAR ^k13811/+; Wsp ¹/Df embryo, breaks occur in both longitudinal and commissural axon bundles, and axons are medially displaced. (I and J) Quantitation of CNS axon defects in single and double mutant embryos. The y axis indicates the percentage of embryos with axon defects. n, number of embryos scored (at least 8 segments per embryo). Embryos were scored as mutant if more than half of the segments were defective. Asterisks indicate single mutants that are significantly different from wild-type controls (p < 0.001, chi-square test) and double mutants that are significantly different from wild-type and single mutant or heterozygote controls (p < 0.001, chi-square test). Bar, 10 μm.

Figure 8.

Abnormal oogenesis in SCAR and Arpc1 mutants. (A–D) Single egg chambers stained to reveal nuclear arrangement (green, visualized with OliGreen) and nurse cell membranes (red, visualized with phalloidin). (A) Nurse cell (nc) nuclei in a late stage wild-type egg chamber are enclosed in individual cells separated by actin-rich membranes (oo, oocyte). (B) In contrast, a SCAR ^Δ37 mutant egg chamber displays a characteristic multinucleate phenotype (arrows). (C) A similar deterioration of egg chamber structure occurs in Arpc1 ^W108R germline clones. (D) Wild-type appearance of a late stage egg chamber from a Wsp ³ germline clone. (E–H) Ring canals visualized with the Hts-RC antibody (Robinson et al., 1994). Stage 10A (E and F) and 10B (G and H) ring canals are shown. In contrast to the wild-type structures (E and G), SCAR ^Δ37 ring canals (F and H) are considerably smaller in size and often occluded. Bars: (A–D) 100 μm; (E–H) 10 μm.

Figure 9.

Distinct head bristle patterns and eye morphologies of Wsp , Arpc1 , and SCAR mutants. Scanning electron microscope images of head cuticle (A–D) and enlarged portions of compound eyes (E–H). Nearly all bristles present on the dorsal aspect of the head and the interommatidial bristles of the eye in wild type (A) are missing in a Wsp ¹/Df(3R)3450 mutant (B). Extensive bristle loss and a rough-eye phenotype are apparent in an Arpc1 ^Q25sd mosaic head (C). The bristle pattern of a SCAR ^Δ37 mosaic (D) appears wild type, whereas the eye facet arrangement is abnormal. The enlargement in E shows a highly symmetrical organization of ommatidia and interommatidial bristles in a wild-type eye. A Wsp ¹/Df(3R)3450 eye (F) exhibits a normal ommatidial array and a pronounced lack of bristles. The abnormal eye facet pattern of a Arpc1 ^Q25sd mosaic eye (G) is characterized by irregularly shaped ommatidia, craters of missing lens material, and a bristle-loss phenotype. A SCAR ^Δ37 mosaic eye (H) shows similar morphological defects, but a considerable number of interommatidial bristles are present. Bars: (A–D) 100 μm; (E–H) 25 μm.