SEARCH

SEARCH BY CITATION

Keywords:

  • aspartic protease;
  • Wnt signaling pathway;
  • planar cell polarity;
  • Drosophila melanogaster

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Experimental Procedures
  6. Acknowledgments
  7. References

Background: Planar cell polarity (PCP) originally referred to the coordination of global organ axes and individual cell polarity within the plane of the epithelium. More recently, it has been accepted that pertinent PCP regulators play essential roles not only in epithelial sheets, but also in various rearranging cells. Results: We identified pepsinogen-like (pcl) as a new planar polarity gene, using Drosophila wing epidermis as a model. Pcl protein is predicted to belong to a family of aspartic proteases. When pcl mutant clones were observed in pupal wings, PCP was disturbed in both mutant and wild-type cells that were juxtaposed to the clone border. We examined levels of known PCP proteins in wing imaginal discs. The amount of the seven-pass transmembrane cadherin Flamingo (Fmi), one of the PCP “core group” members, was significantly decreased in mutant clones, whereas neither the amount of nor the polarized localization of Dachsous (Ds) at cell boundaries was affected. In addition to the PCP phenotype, the pcl mutation caused loss of wing margins. Intriguingly, this was most likely due to a dramatic decrease in the level of Wingless (Wg) protein, but not due to a decrease in the level of wg transcripts. Conclusions: Our results raise the possibility that Pcl regulates Wg expression post-transcriptionally, and PCP, by proteolytic cleavages. Developmental Dynamics 243:791–799, 2014. © 2014 Wiley Periodicals, Inc.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Experimental Procedures
  6. Acknowledgments
  7. References

In epithelia, cells are polarized along a fixed axis within the plane, which is critical for many organ functions. Underlying mechanisms of this planar cell polarity (PCP) have been best studied in the Drosophila wing, where epidermal cells somehow sense an organ axis, localize an assembly of actin filaments at the distal cell vertexes, and produce single wing hairs in pupae (Adler, 2002). It has been shown that evolutionary conserved regulators of PCP orchestrate a variety of collective cell behaviors, such as polarized protrusive cell activity, directional cell movement, and oriented cell division, so they are crucial for the normal development of both epithelial and non-epithelial tissues (Seifert and Mlodzik, 2007; Gray et al., 2011; Vichas and Zallen, 2011).

In spite of a number of molecular players identified, a long-standing question is how exactly individual cell polarity is coordinated with global organ axes. At the molecular level, this coordination is visible in the localization of the “core group” of the PCP regulators at selective plasma membrane domains, such as proximodistal cell boundaries in the Drosophila wing epidermis; and this core pathway plays an instructive role in the polarity establishment (Goodrich and Strutt, 2011). The core group includes the seven-pass transmembrane cadherin Flamingo/Starry night (Fmi/Stan) and Frizzled (Fz) (Usui et al., 1999; Chae et al., 1999; Strutt, 2001). The outstanding question above can now be rephrased as how the polarized core protein localization becomes aligned with organ axes, and what is the molecular identity of the polarizing cue.

Without the function of a distinct group of PCP regulators, the localization of the core proteins is misaligned with the proximodistal axis of the wing. This group includes atypical cadherins Fat (Ft) and Dachsous (Ds), and the Golgi kinase Four-jointed (Fj), which we refer to as the “Ft/Ds group” (Adler et al., 1998; Strutt and Strutt, 2002; Matakatsu and Blair, 2004; Ishikawa et al., 2008; Sharma and McNeill 2013). The Ft/Ds group can influence core protein localization, for example, by affecting the cell division axis and cell rearrangement (Ma et al., 2003; Aigouy et al., 2010), or by controlling the polarity of planar microtubules that are proposed to contribute to directionally biased transport of Fz (Shimada et al., 2006; Harumoto et al., 2010). The relationship between the Ft/Ds group and the core group has been a target of intense investigations (Casal et al., 2006; Thomas and Strutt, 2012; Brittle et al., 2012; Sagner et al., 2012; Blair, 2012).

Aside from the Ft/Ds group, there has been a persistent candidate for the polarizing cue, the Wnt family, which acts in regulating development and also impacts diseases such as cancer (Sugimura and Li, 2010; Rao and Kuhl, 2010; Clevers and Nusse, 2012; Nusse and Varmus, 2012). Vertebrate Wnts could serve an instructive role, linking both cellular and organ polarity (Gao et al., 2011; Gray et al., 2011). In insects, Wingless (Wg), the Drosophila orthologue of Wnt1, was shown to be a morphogen that governs the dorsal-ventral patterning of the wing (Herranz and Milan, 2008), but it has been controversial whether Wg and other Drosophila Wnts provide the cue across the entire wing or not (Chen et al., 2008; Sagner et al., 2012; Wu et al., 2013).

To identify additional components that mediate PCP establishment, we conducted a mosaic screen of the X chromosome and isolated mutations that provoked drastic misorientation of wing hairs (Mouri et al., 2012). In this study, we focused on one intriguing mutation, which mislocalized Fmi and in addition down-regulated Wg protein. The causative gene is pepsinogen-like (pcl), whose product is highly homologous to members of the aspartic protease family including cathepsin D and E, pepsin, and beta-site APP-cleaving enzyme (BACE) (Dunn, 2002). These proteases show broad substrate specificities, and their activities are kept tightly in check to prevent uncontrolled proteolysis (Conus and Simon, 2010). Compared to the established roles of aspartic proteases in digestion and immunity, less is known about their contributions to developmental events. We discuss how the PCP phenotype and the down-regulation of Wg in the pcl mutant clones are related to each other.

Results and Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Experimental Procedures
  6. Acknowledgments
  7. References

A Mutation in pepsinogen-like (pcl) Results in a Planar Polarity Phenotype and Loss of the Wing Margin

To identify novel planar polarity genes, we performed a mosaic screen for X-chromosome mutations. We generated mosaic clones of about 3,000 lethal chromosomes and searched for the polarity phenotype in adult wings (see details in Mouri et al., 2012). We isolated 30 chromosomes that caused severe misorientation of wing hairs, and focused on one of them, no. 11166 (Fig. 1). In addition to the polarity defect, no. 11166 clones showed loss of wing margins (arrowheads in Fig. 1B). As described below, no. 11166 was mutated in pepsinogen-like (pcl)/CG13374 (McQuilton et al., 2012); thus, we designated this allele as pcl1 and hereafter refer to it as such.

image

Figure 1. A mutation in pcl results in a planar cell polarity phenotype and loss of the margin in the wing. A, B: Wild-type (A) and pcl1 mosaic (B) adult wings. Clones of pcl1 in the adult wing caused loss of the wing margin (arrowheads). In this and all subsequent figures, distal is to the right and anterior is at the top. C, D: Higher-power images of wild-type (C) and pcl1 mosaic wings (D) that are marked by red boxes in A and B, respectively. D: Wing hairs were misdirected posteriorly.

Download figure to PowerPoint

Next, we observed pcl1 homozygous clones in pupal wings 32 hr after pupalium formation (APF). A subpopulation of mutant cells along the distal clone border showed misorientation of prehairs (Fig. 2A–C; see left arrowhead in Fig. 2A); in contrast, mutant cells further inside the clone (e.g., near the left edge of Fig. 2A) did not. Intriguingly, neighboring wild-type cells that were located distal to the mutant clone also showed the misorientation (right arrowhead in Fig. 2A). These local cell-autonomous and non-cell-autonomous phenotypes were also revealed by mislocalization of Fmi at anterior-posterior cell boundaries in the clone (Fig. 2B and arrowheads in 2E) and in the adjacent wild-type cells (arrowheads in Fig. 2F), in contrast to the normal localization at distal cell boundaries (arrowheads in Fig. 2D). This mislocalization of Fmi was reminiscent of that in Ft/Ds group mutant clones (Ma et al., 2003; Strutt and Strutt, 2002). Clones with mutations in Ft/Ds group genes mislocalize Fmi proteins at the wrong cell boundaries (anterior-posterior boundaries) both cell autonomously and non-cell autonomously. This phenotype contrasts with mutant clones of core group genes, where Fmi no longer localizes tightly to particular cell boundaries (Usui et al., 1999).

image

Figure 2. The pcl1 mutation produces a non-cell autonomous effect. A–C: The 32-hr APF (after puparium formation) wing was stained with phalloidin (A), for Fmi (B), and for a clone marker GFP (blue in C). A homozygous mutant clone was recognized by the absence of the GFP marker, and wild-type cells that border the clone are indicated with yellow dots (A and B). Magenta arrowheads in A indicate misoriented prehairs of the wild-type cells (right) and in the mutant clone (left). Scale bar = 10 μm. D–F: Higher-power images of wild-type cells deep inside the clone (D), mutant cells close to the clone border (E), and wild-type cells close to the border (F) in (B). Both the mutant and wild-type cells close to the border mislocalized Fmi at anterior-posterior cell boundaries (arrowheads in D and F), in contrast to normal localization at proximal-distal boundaries of the wild-type cells (arrowheads in D).

Download figure to PowerPoint

pcl Is Required for Normal Planar Polarity and Wing-Margin Formation

Through genetic mapping and sequencing, we found mutations that resulted in two adjoining amino-acid substitutions in the pcl coding region, raising the possibility that pcl was the gene responsible for the planar polarity defect (Fig. 3A, see also Experimental Procedures section). The Pcl protein is predicted to belong to a family of aspartic proteases, and the substituted amino acids were located just N-terminal to a sequence, SSTY, which is well conserved among aspartic proteases.

image

Figure 3. pcl is the responsible gene. A: Representation of the domain structure of Pcl protein. The thick red bar indicates a well-conserved domain among aspartic proteases. Partial amino acid sequences of Pcl and three other aspartic proteases, human cathepsin E (hCathE), human pepsinogen 5 (hPGA5), and human cathepsin D (hCathD), are aligned below. Asterisks indicate conserved amino acids among these proteases. pcl1 had two mutations in the coding sequence of this domain (ACTCAA instead of AATAAA), which substitute amino acids NK with TQ.B, C: The misorientation phenotype of wing hairs in the pcl1 mutant was rescued by duplication Dp(1; 3)DC007. C: Higher-power image of a region that is marked by red box in B. D, E: The misorientation phenotype was partially rescued by UAS-pcl that was expressed by armadillo-GAL4. E: Higher-power image of a region that is marked by red box in D.

Download figure to PowerPoint

To verify that pcl was the responsible gene for the no. 11166 phenotypes, we performed a rescue experiment using two duplications (Venken et al., 2010) that partially overlapped each other. Dp(1; 3)DC007, which includes pcl, rescued the lethality and restored the wing phenotypes to normal (Fig. 3B and C), whereas Dp(1; 3)DC098 did not (data not shown). Because pcl is the only annotated gene that is inside DC007 and not contained on DC098, we concluded that pcl is most likely the responsible gene. We also performed a rescue experiment by expressing the pcl cDNA under the control of the armadillo-GAL4 or daughterless-GAL4. Both of the GAL4 lines rescued both lethality and the margin phenotype. We note, however, the PCP phenotype was only partially rescued, and a moderate polarity defect was still observed in the rescued animals (Fig. 3D and E).

pcl Regulates wg Expression in a Posttranscriptional Manner

It is known that the Wingless (Wg) and Notch pathways are required for formation of the wing margin (Neumann and Cohen, 1996; Herranz and Milan, 2008). The loss of wing margins elicited by the pcl mutation implied that pcl was necessary for either the Wg or the Notch pathway. To address which pathway and which step in either pathway was primarily affected by pcl1, we examined the expression levels of various proteins or markers in wing imaginal discs in late 3rd instar larvae that were populated by pcl1 mutant clones. As previously established, in the wild-type disc, Wg is expressed in the future wing margin (Couso et al., 1994; Micchelli et al., 1997). By contrast, the Wg signal was significantly decreased in the pcl mutant clones (Fig. 4A and A'). In some smaller mutant clones, the reduction of Wg signal was less obvious than in larger clones (Fig. 4B and B'), possibly due to a perdurance effect. The level of Cut protein was also lower in the mutant clones (Fig. 4C and C'). Because the cut gene is one of the direct targets in Wg-responsive cells and is essential for the margin formation, these results imply that the pcl1 mutation reduced the amount of Wg protein, which resulted in less cut expression in the signal-receiving cells, and ultimately the loss of the wing margin (Couso et al., 1994; Micchelli et al., 1997).

image

Figure 4. pcl regulates Wingless (Wg) expression. A–B': Wing imaginal discs of 3rd instar larvae were stained for Wg (magenta in the merged images in A and B), Cut (magenta in C), and a clone marker GFP (green in A–F). Mutant clones of pcl were marked by the loss of GFP. Expression of Wg or Cut was reduced in the mutant clones (arrowheads in A' and C'), but it was less obvious in clones in B'. D–F': Expression of vestigial-lacZ (vg-lacZ) (magenta in D), wg-lacZ (magenta in E), and Wntless (Wls; magenta in F) along the dorsal-ventral boundary was not altered in the mutant clones. Scale bar = 20 μm.

Download figure to PowerPoint

How, then, does the pcl mutation abrogate expression of wg? In the wing disc, transcription of the wg gene is induced by Notch signaling (Rulifson and Blair, 1995). Thus, we examined the effect of the pcl mutation on Notch signaling by using vg-lacZ (Neumann and Cohen, 1996) and wg-lacZ (Kassis et al., 1992) reporter genes as readouts. Intriguingly, we could not detect any change in either the vg-lacZ or the wg-lacZ signal in pcl mutant clones (Fig. 4D–E'). These results suggested that Notch signaling and its downstream events, including transcription of wg, were not affected by the pcl mutation. Therefore, the reduction of Wg protein was possibly due to a defect(s) in post-transcriptional regulation, such as at the level of translation, intracellular or extracellular degradation, and/or secretion.

In contrast to the large number of studies on the signaling pathway in Wnt signal-responsive cells, our knowledge is still limited about how Wnt proteins such as Wg are processed and secreted. Enzymes and secretory proteins that are dedicated to Wnt signals have been identified (Herr et al., 2012). Nonetheless, only a few proteins are known to regulate the secretion of Wnt protein: Porcupine (Porc) palmitoylates Wnt at the ER, whereas Wntless/Evenness Interrupted (Wls/Evi) facilitates the secretion of Wnt protein from the Golgi to the extracellular space (Kadowaki et al., 1996; Banziger et al., 2006; Bartscherer et al., 2006). Wg proteins accumulate in Wg-producing cells that are mutant for porc or wls (van den Heuvel et al., 1993; Banziger et al., 2006; Bartscherer et al., 2006), which contrasts with the disappearance of Wg protein in pcl clones. We suspected that pcl might regulate Wg proteins by modulating the level of Porc or Wls. However, when we immunostained Wls protein, we saw no difference between pcl mutant and wild-type cells (Fig. 4F and F').

The pcl Mutation Does Not Affect the Expression of four-jointed and dachsous

Because the mislocalization of Fmi in pcl1 mutant clones resembled that in fj or ds mutant clones as described above, it could be that pcl was functionally associated with these genes. To test this possibility, we observed the signal of a fj-lacZ reporter gene in pcl1 mutant clones. We did not detect significant changes in the signal between pcl mutant and wild-type cells (Fig. 5A–A''). We also immunostained pupal wings with anti-Dachsous (Ds) antibody, and did not detect any alteration of the Ds level between mutant clones and adjacent WT clones in the wing pouch (Fig. 5B–B').

image

Figure 5. The pcl1 mutation does not significantly affect fj and ds expression. A–A'': Mutant clones of pcl in the wing disc were marked by the loss of GFP (green in A, and A″). Expression of fj-lacZ (magenta in A, and A') was not significantly altered in the clones (arrowhead). B–C': pcl mosaic clones were stained for Dachsous (Ds) (magenta in B and C, and B' and C') and for the clone marker GFP (green in B and C). A pcl mosaic clone in a 24-hr APF wing (B,B') and those in the dorsal hinge region in the larval wing disc (C and C'). The nematic orders in the individual cells are overlaid on the image of Ds signals (yellow bars in C'). Distal is to the bottom of the panel. Scale bar = 20 μm in A–B', 5 μm in C and C'. D–F: Quantifications of the amount of Ds (D) and its polarized localization (E and F) in the wild-type cells (WT; N=140) and pcl mutant cells (pcl; N=96). D, E: Box-and-whisker plots depicting the signal intensity of Ds at cell boundaries (D) and the magnitudes of nematic order for individual cells (E). The box plots show median (line), top, bottom (whiskers), after removing outliers, and 25th and 75th percentile (boxes). The signal intensity of Ds is indicated by artificial unit per pixel (D) and the magnitude of nematic order (length of each yellow bar in C') is indicated by artificial unit (E). P > 0.05, Wilcoxon rank sum test. NS: not significant. F: The axis distributions of the PCP nematic are shown by rose diagrams in a point symmetry manner. Each diagram is composed of 24 bins of 15° each, with an approximate direction of the dorsal/ventral compartment boundary (the presumptive wing margin) set to 0°, and concentric circles are drawn with 5% increments between them. The distributions in the two genotyped cells are not significantly different (P > 0.05, Mardia-Watson-Wheeler test). We performed the quantifications of total four discs that had mutant clones in the dorsal hinge regions, and the data of one of the four are shown in C–F, which is similar to that of the remaining three discs (data not shown).

Download figure to PowerPoint

We further examined quantitatively whether the amount of Ds and its polarized localization at cell boundaries were affected or not. For this purpose, we focused on mutant clones that reached the dorsal hinge region where Ds is localized in a polarized fashion along the proximal-distal axis (Brittle et al., 2012), and compared “PCP nematic” (Aigouy et al., 2010) in the wild-type cells with that of the mutant cells (Fig. 5C–F, see also Experimental Procedures section). We could not detect significant differences in the amount (Fig. 5D) or the magnitude (Fig. 5E) of the axis (Fig. 5F) of the PCP nematic between the wild-type and mutant cells.

These results suggest that pcl did not regulate the expression level of fj or the amount and the polarized localization of Ds; however, it does not necessarily exclude possible functional interactions between Pcl and these proteins. For example, the Ds ectodomain is subject to endoproteolytic processing and this is modulated by Ft (Ambegaonkar et al., 2012), implying a possible involvement of the hypothetical molecular activity of Pcl.

Fmi Protein Is Less Abundant in pcl Mutant Clones in Larval Imaginal Discs

Although the loss of wing margins was most likely a direct consequence of the decrease in Wg as described above, how did the pcl mutation give rise to the defect in PCP? It has been recently reported that Wg, together with DWnt4, plays a role in PCP by providing a long-range directional cue to cells (Wu et al., 2013). However, mutant clones that remove wg, dWnt4, and two other dWnts genes cause misorientation of wing hairs along the margin (Wu et al., 2013), and adult wings comprised predominantly of wg mutant cells show a mild hair misorientation only at the distal margin (Sagner et al., 2012). In contrast, the aberrant PCP phenotype along pcl mutant clone borders was seen when the clones were generated either along the margin or inside the wing blade (Figs. 1B,D, 2A–C). So a decrease in the amount of Wg in the pcl clones may not be a sole cause of the PCP phenotype.

To explore the basis of the pcl PCP phenotype, we examined levels of PCP core proteins in pcl mutant clones in both pupal wings and larval discs. We found that Fmi at cell boundaries was significantly decreased, but not totally eliminated, in pcl mutant clones in 3rd instar larval discs (Fig. 6A–C) and this decrease in the apical Fmi was not associated with relocalization of Fmi at the basal level in the same cells (data not shown). In some small clones, down-regulation of Fmi was less obvious. In contrast, expression levels of a basolateral marker Discs large (Dlg), the Drosophila β-catenin Armadillo (Arm), one of the PCP core-group members Dishevelled (Dsh), or DE-cadherin was not altered in mutant clones (Fig. 6D–E'; data of Dsh and DE-cadherin were not shown). This specific reduction of Fmi in pcl mutant clones in larval imaginal discs is puzzling, because the amount of Fmi was not apparently decreased (although it was mislocalized) in pupal wings as described above (Fig. 2B and E). It has been shown that cell-boundary localization of Fmi and other core-group proteins is already polarized at a late larval stage (Classen et al., 2005; Sagner et al., 2012). It remains to be studied whether the decrease in the Fmi amount in discs is at least one cause of the clone-border selective PCP phenotype in pupal wings.

image

Figure 6. Fmi is less abundant in pcl mutant clones. pcl mutant clones in wing discs were stained for Fmi (magenta in A–B'), Discs large (Dlg; magenta in D, D'), or Armadillo (Arm; magenta in E, E'). The mutant clone in E and E' was located in the future notum where the reduction of Wg signaling had less of an effect on the expression of Arm, compared to the wing margin. pcl mutant clones were recognized by the absence of the GFP marker (A–E). B, B': Higher-power image of A and A', respectively. C: a and b indicate each end of the line. A plot of fluorescent intensities of Fmi (magenta) and GFP (green) along the line in A that spans a pcl mutant clone and its adjacent wild-type clone. Scale bar = 10 μm.

Download figure to PowerPoint

Conclusion: Pcl Acts in Both PCP and Wg Signaling

In this study, we reported that both of the two well-known developmental mechanisms, PCP formation and Wnt signaling, require pcl encoding a putative aspartic protease. Aspartic protease family proteins play extracellular and intracellular roles; for example, pepsinogen digests foods in the stomach lumen, whereas cathepsins function in adaptive immunity in lysosomes (Conus and Simon, 2010). Further studies will clarify where in the Drosophila wing epidermis and where in the cell Pcl is required, whether Pcl indeed possesses an aspartic protease activity or not, and if so, what are its endogenous substrates. Considering that aspartic proteases have broad substrate specificity (Conus and Simon, 2010), Pcl may indirectly control the activity and/or stability of Wg and Fmi through proteolytic cleavages of distinct substrates.

Experimental Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Experimental Procedures
  6. Acknowledgments
  7. References

Mutant Screening

Mutation screening was done as described previously (Mouri et al., 2012). Briefly, mutations were induced in w FRT19A/Y males, and the mutagenized chromosomes were balanced. Mutant clones were induced and phenotypes were analyzed in the adult wing. pcl1/pcl1 was early larval lethal.

Fly Strains and Genetics

Control strains used were y w. Transgene UAS-pcl was expressed by using the GAL4/UAS system (Brand and Perrimon, 1993). Driver strains employed were armadillo (arm)-GAL4 and daughterless (ds)-GAL4, both of which were obtained from the Bloomington Drosophila Stock Center. Duplication and deficiency strains used in mapping (described below) were provided by the Drosophila Genetic Resource Center (DGRC). Other stocks were vg-lacZ (Neumann and Cohen, 1996), wg-lacZ (Kassis et al., 1992), fj-lacZ (Brodsky and Steller, 1996), Dp (1; 3)DC007 and Dp(1; 3)DC098 (Bloomington Drosophila Stock Center). All fly embryos, larvae, pupae, and adults were reared at 25°C unless described otherwise. Exact genotypes of individual animals used in figures are as follows:

Fig. 1.

(A and C) y w

(B and D) w pcl1 FRT19A/w FRT19A; vg-GAL4 UAS-FLP/+

Figs. 2 and 6.

w pcl1 FRT19A/y w ubi-GFP FRT19A; vg-GAL4 UAS-FLP/+

Fig. 3.

(B and C) w pcl1 FRT19A/Y; Dp(1; 3)DC007/+

(D and E) w pcl1 FRT19A/Y; UAS-pcl/arm-GAL4

Fig. 4.

(A–C', F, and F') w pcl1 FRT19A/y w ubi-GFP FRT19A; vg-GAL4 UAS-FLP/+

(D and D') w pcl1 FRT19A/y w ubi-GFP FRT19A; vg-GAL4 UAS-FLP/vg-lacZ

(E and E') w pcl1 FRT19A/y w ubi-GFP FRT19A; vg-GAL4 UAS-FLP/wg-lacZ

Fig. 5.

(A) w pcl1 FRT19A/y w ubi-GFP FRT19A; vg-GAL4 UAS-FLP/fj-lacZ

(B-C') w pcl1 FRT19A/y w ubi-GFP FRT19A; vg-GAL4 UAS-FLP/+</UL>

Mapping

To identify the affected genes in isolated mutants, no. 11166/FM7 flies were mated with four deficiency and duplication stocks (DGRC stock number 108921, 108145, 108138, and 106068). Taking together the information of lethality and deficiency/duplication points in individual lines, we narrowed the genomic region of the responsible gene of stock 11166 to 200 kb, which included 3 genes (CG32816, l(1)sc, and pcl). We sequenced coding regions of these 3 genes in the female genome of no. 11166, and found two substitution mutations in the pcl gene.

Molecular Biology

A cDNA fragment encoding the pcl gene was cloned into a pUAST-based plasmid containing a UAS promoter. To generate transgenic flies, this construct, pUAST-pcl, was microinjected into fly embryos carrying the attP2-site, and integrated into the site by phiC31-mediated site-specific recombination (Bateman et al., 2006; Bischof et al., 2007).

Immunohistochemistry

Wing imaginal discs in wandering 3rd instar larvae or pupal wings were fixed and used for immunohistochemistry. Primary antibodies used were rabbit anti-GFP (Molecular Probes, Eugene, OR), mouse anti-Fmi no. 74 and rat anti-Fmi (Usui et al., 1999), rabbit anti-Dlg (Woods and Bryant, 1991), rabbit anti-Ds (Strutt and Strutt, 2002), and rabbit anti-Wls (Port et al., 2008). Mouse anti-Arm N2 7A1, mouse anti-Wg 4D4, and mouse anti-Cut 2B10 were obtained from the Developmental Studies Hybridoma Bank (DSHB) at the University of Iowa. Secondary antibodies were purchased from Molecular Probes and Jackson ImmunoResearch (West Grove, PA). Immunostaining and acquisition of confocal images were done as described previously (Mouri et al., 2012), and fluorescent intensity was measured by using Image J software (NIH).

Quantification of Ds Distribution

We followed the quantification method (nematic order), which is essentially described by Aigouy et al. (2010), to determine the magnitude and axis of nematic order for a single cell as the “PCP nematic” for that cell. We calculated the average value of the intensity for I(r, θ) for each θ section with a 5° range (e.g., 2.5–7.5°, 7.5–12.5°, etc.), where r is the distance between the pixel and the center of the cell and θ is the angle indicating the position of the pixel.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Experimental Procedures
  6. Acknowledgments
  7. References

DGRC at Kyoto Institute of Technology, the Bloomington Stock Center, and the Developmental Studies Hybridoma Bank at the University of Iowa provided the fly stocks and the antibodies. We thank Paul Adler for his detailed advice for our mutant screening and Yukiko Miyake, Kumiko Shimizu, Jyunko Mizukoshi, Kanae Oki, and Mayumi Futamata for their technical assistance. We are also grateful to Konrad Basler, David Strutt, Dan Woods, Stephen DiNardo, Shinichi Yanagawa, Shigeo Hayashi, Ritsuko Takada, Shinji Takada, and Uemura lab members for providing other reagents, and protocols, and discussing the phenotypes. We thank James Hejna for his efforts to polish the manuscript. This work was supported by a CREST grant to T.U. from JST and a Research Grant in the Natural Sciences to T.U. from Mitsubishi Foundation. M. K. and S. H. are recipients of fellowships from the Japan Society for the Promotion of Science.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Experimental Procedures
  6. Acknowledgments
  7. References