Arabidopsis SCARs function interchangeably to meet actin-related protein 2/3 activation thresholds during morphogenesis

Abstract

During polarized growth and tissue morphogenesis, cells must reorganize their cytoplasm and change shape in response to growth signals. Dynamic polymerization of actin filaments is one cellular component of polarized growth, and the actin-related protein 2/3 (ARP2/3) complex is an important actin filament nucleator in plants. ARP2/3 alone is inactive, and the Arabidopsis thaliana WAVE complex translates Rho-family small GTPase signals into an ARP2/3 activation response. The SCAR subunit of the WAVE complex is the primary activator of ARP2/3, and plant and vertebrate SCARs are encoded by a small gene family. However, it is unclear if SCAR isoforms function interchangeably or if they have unique properties that customize WAVE complex functions. We used the Arabidopsis distorted group mutants and an integrated analysis of SCAR gene and protein functions to address this question directly. Genetic results indicate that each of the four SCARs functions in the context of the WAVE-ARP2/3 pathway and together they define the lone mechanism for ARP2/3 activation. Genetic interactions among the scar mutants and transgene complementation studies show that the activators function interchangeably to meet the threshold for ARP2/3 activation in the cell. Interestingly, double, triple, and quadruple mutant analyses indicate that individual SCAR genes vary in their relative importance depending on the cell type, tissue, or organ that is analyzed. Differences among SCARs in mRNA levels and the biochemical efficiency of ARP2/3 activation may explain the functional contributions of individual genes.

Figure 1.

SCAR Protein Domains and Physical Maps of the Arabidopsis scar T-DNA Insertion Alleles. (A) Domain organization of a typical plant SCAR family protein, Arabidopsis SCAR4. (B) to (D) Molecular characterization of SCAR T-DNA insertion alleles. The presence of SHD and WA encoding regions of scar transcripts was assayed using RT-PCR (right panels). Detection of the GAPC gene served as a positive control. Wild-type (Col-0) cDNA and a no-RT reaction served as the positive and negative controls, respectively. The position of the primers used for characterizing T-DNA insertions and the transcription of SHD and WA domains are marked on the physical map by arrows. Red boxes indicate SHD-encoding exons, blue boxes indicate exons encoding the WA domain, gray boxes indicate the regions that encode the nonconserved region in the SCAR proteins, and white boxes indicate the untranslated regions. The inverted triangles label the position of the T-DNA insertions. (B) Analysis of scar1-t1 (salk_017554) transcripts. (C) Analysis of scar3-t1 (salk_036994) transcripts. (D) Analysis of scar4-t2 (salk_022766) transcripts.

Figure 2.

Distance-Based Gene Tree of the SHD and WA Domains Manually Fused Together for Arabidopsis, M. truncatula, P. trichocarpa, O. sativa, and P. patens. We designated SCARs within the Arabidopsis SCAR2/4 clade as the A clade and those within the SCAR1/3 clade as the B clade. The numbering scheme for the SCAR genes is not intended to reflect ancestral relationships. With the exception of Arabidopsis SCARL that contains only an SHD domain, SCAR-like proteins from other species that did not contain both a SHD and WA domain were not included in this tree. The resulting tree had a minimum evolution score of 567.3. Corresponding accession numbers for each gene can be found under accession numbers and the corresponding Nexus alignment as Supplemental Data Set 1 online.

Figure 3.

SCAR Genes Function Redundantly during Morphogenesis and in Dark-Grown Seedlings. (A) to (D) Scanning electron micrograph of mature trichomes. Bars = 50 μm. (A) The mature wild-type (Col-0) trichome has three highly elongated branches. (B) The mature scar2 (dis3-1) trichome with a typical weak distorted phenotype. (C) The mature scar4 (scar4-t2) trichome has three highly elongated branches. (D) The mature scar2;scar4 double mutant trichome has a very strong swollen and distorted phenotype. (E) to (H) Shoots of 7-DAG dark-grown seedlings of wild-type (E), scar2 (F), scar1;scar3 (G), and arpc2 (dis2) (H). (I) to (L) Wide-field fluorescence images of fields of cotyledon epidermal pavement cells of wild-type (I), scar2;scar4 (J), scar1 scar2;scar3;scar4 (K), and arpc2 (dis2) (L). (M) to (P) Digitally processed images of individual pavement cells from (I) to (L) overlayed with their calculated skeletons.

Figure 4.

The Trichome Actin Cytoskeleton Defects of the scar2;scar4 and the scar1 scar2;scar3;scar4 Mutants Are Indistinguishable from Strong wave and arp2/3 Mutants. The left panels are the maximum projections of confocal image stacks of whole-mount trichomes probed with phalloidin. Boxes in the left panels show regions highlighted in adjacent panels to the right. The right panels are the projections of branch midplanes that contain core cytoplasmic bundles. Bars = 5 μm. (A) and (B) F-actin organization in a wild-type stage 4 trichome showing actin bundles that are aligned with the long axis of trichome branch elongation. (C) and (D) F-actin organization in a scar2;scar4 stage 4 trichome showing disorganized actin filaments or bundles. (E) and (F) F-actin labeling of a similarly staged scar1 scar2;scar3;scar4 quadruple mutant. (G) and (H) F-actin labeling of a brk1 stage 4 trichome. (I) and (J) F-actin organization in an arpc2 (dis2) stage 4 trichome.

Figure 5.

SCAR3 Can Substitute for SCAR2 during Trichome Morphogenesis. Full-length SCAR2 and SCAR3 cDNAs were transformed into the scar2 background. Trichome morphogeneis was assayed in expanding leaves. (A) Wild-type leaf with highly elongated trichomes. (B) The expanding leaf of a scar2 mutant with distorted trichomes. (C) The expanding leaf of a scar2 mutant transformed with 35S:SCAR2. (D) The expanding leaf of a scar2 mutant transformed with 35S:SCAR3. Both wild-type trichomes (arrows) and trichomes with a weak distorted phenotype (arrowheads) are present.

Figure 6.

Quantitative Analysis of SCAR Transcript Levels in Wild-Type and scar2 Trichomes. Steady state mRNA levels of SCAR genes were quantified in trichomes purified from wild-type and scar2 plants (see Methods). mRNA levels were quantified by qRT-PCR, and SCAR expression was normalized to the GAPC (AT3G04120) gene. The normalized transcript values represent the means ± se of triplicate measurements as described in Methods. Groups with an identical superscripted letter were not significantly different according to an ANOVA Tukey multiple comparison test (P < 0.01). The scar2 (dis3-1) allele causes premature transcriptional termination after the SHD-encoding exons (Basu et al., 2005).

Figure 7.

ARP2/3 Activation Efficiency of SCAR2, SCAR3, and SCAR4 WA Domains. (A) Kinetic analysis of actin polymerization in the presence of ARP2/3 was monitored with the indicated GST-tagged activator protein, 3 μM actin (5% pyrene-labeled), and 10 nM bovine ARP2/3 complex. (B) Comparison of the concentrations of barbed ends generated at half-maximal polymerization in the presence of varying amounts of GST-SCAR2-WA, GST-SCAR3-WA, and GST-SCAR4-WA. SCAR2-WA's end concentration plot typically saturates at 50 nM SCAR2-WA at 10 × 10⁻⁴ μM ends and then decreases at higher concentrations. SCAR3-WA's end concentration plot saturates at 1 μM SCAR3-WA at ∼20 × 10⁻⁴ μM ends and then decreases by ∼25% at 3 μM SCAR3WA. SCAR4-WA's end concentration plot saturates at 1.5 μM SCAR4-WA at ∼7 × 10⁻⁴ μM ends. (C) Coomassie blue–stained SDS-PAGE gel showing molecular size standards (lane 1), GST-SCAR2-WA (lane 2), GST-SCAR3-WA (lane 3), and GST-SCAR4-WA (lane 4).