Gene Expression Patterns 15 (2014) 67–79

Contents lists available at ScienceDirect

Gene Expression Patterns journal homepage: www.elsevier.com/locate/gep

Extensive nonmuscle expression and epithelial apicobasal localization of the Drosophila ALP/Enigma family protein, Zasp52 Beth Stronach ⇑ Department of Microbiology and Molecular Genetics, University of Pittsburgh School of Medicine, Pittsburgh, PA 15219, USA

a r t i c l e

i n f o

Article history: Received 10 March 2014 Received in revised form 5 May 2014 Accepted 8 May 2014 Available online 17 May 2014 Keywords: Epithelial polarity Integrins Drosophila Protein trap Actomyosin PDZ and LIM domain

a b s t r a c t This study describes the broad tissue distribution and subcellular localization of Drosophila Zasp52, which is related to the large family of ALP (a-actinin associated protein)/Enigma PDLIM (PDZ and LIM domain) proteins of vertebrates. Results demonstrate that ZCL423 is a protein trap insertion in the Zasp52 locus tagging multiple endogenous splice isoforms with GFP. While Zasp52 has been previously characterized in muscle tissues primarily, visualization of GFP fluorescence in Zasp52 protein trap lines revealed expression in many nonmuscle tissues including the central nervous system, secretory glands, and epithelial tissues constituting the embryonic epidermis, the somatic follicle cell layer encapsulating the germline during oogenesis, and imaginal disc precursors to the adult body. In epithelial cells, Zasp52 typically accumulated basally, adjacent to integrin adhesion sites, and apically along adherens junctions, particularly enriched near junctional vertices of multicellular interfaces. Also Zasp52 showed polarized accumulation at the leading edge of migrating cell populations and morphogenetic boundaries similarly enriched for myosin. As such, Zasp52 GFP protein traps may be useful molecular markers for dynamic epithelial rearrangements. Moreover, the pattern of Zasp52 expression within nonmuscle tissues reveals potential functional roles in cell–cell and cell–matrix adhesion, specifically at sites of increased actomyosin contractile tension. In these contexts, the investigation of Zasp52 may provide insights into the functions of numerous PDLIM proteins of the metazoan lineages. Ó 2014 The Author. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Members of the ALP1/Enigma family of PDLIM proteins are evolutionarily conserved adaptor molecules that associate with the actin cytoskeleton to couple structural integrity of the cell with localized regulation of their binding partners [recent reviews (Krcmery et al., 2010; te Velthuis and Bagowski, 2007; Zheng et al., 2010)]. PDLIM genes are so named because they encode proteins consisting of an N-terminal PDZ domain, which directly binds to the actin cross-linking protein a-actinin, followed by one to four C-terminal, zinc-coordinating, LIM domains, which interact with various signaling and transcriptional effectors (Krcmery et al., 2010; te Velthuis and Bagowski, 2007). Two subfamilies are discernable in vertebrates (Te Velthuis et al., 2007). The ALP gene subfamily, encoding proteins with a PDZ and single LIM domains, consists of PDLIM1/CLP36/Elfin, PDLIM2/Mystique/SLIM, PDLIM3/ALP, and ⇑ Address: Department of Microbiology and Molecular Genetics, University of Pittsburgh School of Medicine, 450 Technology Dr, Bridgeside Point 2, Ste 517, Pittsburgh, PA 15219, USA. Tel.: +1 (412) 648 7658; fax: +1 (412) 624 8997. E-mail address: [email protected] 1 Abbreviations used: ALP, a-actinin-associated LIM protein; GFP, green fluorescent protein; PDLIM, PDZ (Post-synaptic density/PSD-95, Discs large/Dlg, Zonula occludans/ZO-1) and LIM (Lineage/LIN-11, Islet/ISL1, Mechanosensory/MEC3) protein; ZASP, Z-band alternatively-spliced PDZ motif-containing protein.

PDLIM4/RIL. The Enigma gene subfamily, coding for proteins with a PDZ and three LIM domains, is comprised of PDLIM5/ENH, PDLIM6/LDB3/ZASP/Cypher, and PDLIM7/Enigma/LMP. Nematode and fly genomes contain a single locus that undergoes extensive alternative splicing to generate proteins resembling ALP and Enigma variants (Jani and Schock, 2007; Machuca-Tzili et al., 2006; McKeown et al., 2006). Phylogenetic analysis suggests that this gene architecture represents the ancestral state (Te Velthuis et al., 2007). Proteins derived from the alternative spliceforms of the invertebrate genes display diversity with respect to the presence of the PDZ domain and the number of encoded LIM domains (Benna et al., 2009; Katzemich et al., 2011; McKeown et al., 2006). Unifying features of the ALP/Enigma proteins include binding to a-actinin and co-localization with cellular actin-based structures such as stress fibers, adherens junctions, focal adhesions, and Zdiscs of myofibril sarcomeres (Boumber et al., 2007; Jani and Schock, 2007; Kiess et al., 1995; Loughran et al., 2005; Ohno et al., 2009; Tanaka et al., 2005; Torrado et al., 2004; Vallenius et al., 2000, 2004; Xia et al., 1997). Shuttling between the cytoskeleton and the nucleus has also been recognized as an important functional property of some family members (Krcmery et al., 2010). By virtue of these properties, ALP/Enigma proteins influence

http://dx.doi.org/10.1016/j.gep.2014.05.002 1567-133X/Ó 2014 The Author. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

68

B. Stronach / Gene Expression Patterns 15 (2014) 67–79

diverse cellular processes during development and disease, notably cancer. For example, sequestration of transcriptional mediators and kinases as well as modulation of actin assemblies underlie their roles in regulation of gene expression, cell signaling, adhesion, and motility. A growing number of PDLIM proteins have been shown to possess tumor suppressor activity, as their expression is suppressed by epigenetic silencing in several types of cancer cells, while restoration of expression mitigates various tumor phenotypes (Boumber et al., 2007; Qu et al., 2010). However, their roles appear to be context-dependent. For instance, recent studies indicate that PDLIM2/SLIM is highly expressed in invasive cancer cells and suppression correlates with loss of directional motility (Bowe et al., 2014; Loughran et al., 2005). While several ALP/Enigma proteins (e.g. CLP36, SLIM, RIL, Enigma) have broad expression patterns in nonmuscle tissues including neurons, bone, lung, colon, placenta, or leukocytes, ALP and ZASP (Z-band alternatively-spliced PDZ motif-containing protein) are predominantly expressed in muscle, located at the Z-line of sarcomeres (Faulkner et al., 1999; Zhou et al., 1999). Genetic analyses in worm, fly, fish, and mouse support a role for ZASP in maintenance of muscle architecture and function (Han and Beckerle, 2009; Huang et al., 2003; Katzemich et al., 2013; van der Meer et al., 2006; Zheng et al., 2009; Zhou et al., 2001). Consistent with animal models, mutations in human ZASP underlie myopathies of skeletal and cardiac musculature (Arimura et al., 2004; Selcen and Engel, 2005; Vatta et al., 2003). Invertebrate ALP and Enigma variants are derived from one genomic locus, alp-1 in C. elegans, and Zasp52 in Drosophila (Jani and Schock, 2007; McKeown et al., 2006). Their expression and functions have been investigated principally in muscle tissues where they participate in sarcomere organization; accumulation in nonmuscle epithelial tissues was also evident, though not investigated extensively (Delon and Brown, 2009; Han and Beckerle, 2009; Hudson et al., 2008; Jani and Schock, 2007; Katzemich et al., 2013, 2011; McKeown et al., 2006). Evidence demonstrates that Drosophila Zasp52 proteins physically interact with a-actinin and genetically interact with integrin adhesion receptors, reflecting a functional role in adhesion complex formation and activation (Bouaouina et al., 2012; Jani and Schock, 2007; Katzemich et al., 2013). Whether this is a primary function of Zasp52 proteins in developing epithelia remains an open question. Here, the characterization of GFP protein trap insertions in the Zasp52 gene demonstrates extensive expression of the fly ALP/Enigma proteins in nonmuscle cells and tissues. Enrichment of Zasp52 in basally located adhesion structures and apical zonula adherens of epithelial cells at morphogenetic boundaries suggests possible participation in contractile actomyosin dynamics mirroring its role in muscle sarcomere integrity. This study lays the foundation for future molecular genetic studies of ALP/Enigma functions outside of the musculature.

1. Results and discussion 1.1. ZCL423 is a GFP protein trap in the Zasp52 gene ZCL423 is a Drosophila P-element transposon line initially isolated in a screen for genomic insertions in which GFP sequences are spliced in frame to endogenous protein coding sequences (Morin et al., 2001). The resulting GFP-tagged proteins, or ‘protein traps’, provide a way to visualize not only the expression pattern of an endogenous gene but the cellular distribution of the encoded protein(s) as well. ZCL423 GFP was shown to decorate the leading edge (LE) of embryonic ectodermal cells during dorsal closure, allowing for striking visualization of the zippering process as it occurred in living embryos (Morin et al., 2001). The ZCL423 line has since been used as a fluorescent marker to screen for genes

required for proper dorsal closure (Jankovics et al., 2011), yet the identity of the tagged protein has not been determined. With a similar aim, I used the expression of ZCL423 GFP to follow the progress of embryonic dorsal closure and noted that its expression was temporally and spatially dynamic over the entire course of development. In particular, during mid-embryogenesis GFP accumulated in both the amnioserosa (AS) and dorsal-most ectodermal cells coincident with germband retraction and dorsal closure; strong enrichment at the LE of the dorsal epidermis was readily apparent (Fig. 1A). GFP was also detected in cephalic ectodermal cells (Fig. 1A) and at the ventral midline (not shown). With the onset of muscle differentiation, ZCL423 GFP became highly expressed in all mesodermal derivatives including somatic, heart, and visceral muscle (Fig. 1C and Ci). In live third instar larvae, GFP was visualized predominantly in sarcomeric repeats of the somatic musculature (Fig. 1D and Di). To determine the identity of the endogenous gene tagged by the ZCL423 insertion, flanking genomic DNA was captured by plasmid rescue, amplified by inverse PCR, cloned, and sequenced. A BLAST search of the Drosophila genome revealed that 350 nucleotides of DNA sequence were identical to a fragment encompassing the terminal exon of the annotated Zasp52-RK transcript (Fig. 2A). Thus, the transposon insertion site was 80 nucleotides upstream of exon 15 [numbering according to (Katzemich et al., 2011)], which encodes eight C-terminal amino acids of Zasp52-PK (Fig. 2A), followed by a stop codon and noncoding sequences. Based on the position of the insertion, it was conceivable that the GFP cassette exon could be spliced into all predicted Zasp52 variants (13 identified by Katzemich et al., 2011, 18 annotated by FlyBase except Zasp52-RP and -RO, which terminate upstream of the transposon) positioning GFP downstream of the first LIM domain in the encoded proteins (Fig. 2A). To address this possibility, Western immunoblot with an anti-GFP antibody was performed on lysates from ZCL423 embryos (prior to muscle differentiation to enrich epithelial isoforms), larvae, and adults (Fig. 2B). For comparison, lysates prepared from G00189, another Zasp52 GFP protein trap inserted 5-prime relative to ZCL423 (Jani and Schock, 2007; Morin et al., 2001) (Fig. 2A), were probed simultaneously. Multiple protein bands were detected in the lysates of both insertion lines suggesting that GFP was fused in frame with exons from multiple Zasp52 splice variants. The overall pattern of the tagged proteins from the two different lines was similar though not identical. The relative levels of the predominant isoforms differed in the lysates from ZCL423 and G00189 larvae. Moreover, a slight difference in the relative sizes of the major isoforms was also noted, though the reason for this difference is not known. To confirm the antibody specificity, two additional experiments were done. First, lysates from ZCL3111, another GFP protein trap line encoding integrin-linked kinase (Ilk::GFP), were probed with GFP antisera. A band at the expected size for an Ilk fusion protein was detected in each lysate (Fig. 2B), though at much reduced levels compared with the abundance of Zasp52 proteins. Second, the Zasp52 protein trap lysates were probed separately with a polyclonal antibody against Zasp52 (Jani and Schock, 2007) revealing a highly similar pattern of protein bands (Fig. 2B). Finally, immunofluorescence detection of endogenous Zasp52 proteins in embryos with the Zasp52 antisera revealed a pattern of expression and localization indistinguishable from ZCL423 GFP (Fig. 1B). Altogether, these results demonstrate the specificity of the antibodies, the presence of multiple tagged isoforms encoded by each protein trap line, and potential differences in proteins tagged at the distinct sites. To further confirm that the expression of GFP-tagged proteins in ZCL423 animals corresponded to Zasp52 isoforms, in vivo RNA interference was performed with a transgenic Zasp52 inverted repeat construct (Ni et al., 2008) to knockdown Zasp52 protein

B. Stronach / Gene Expression Patterns 15 (2014) 67–79

69

Fig. 1. ZCL423 GFP protein trap is expressed in dorsal tissues of the embryo and muscles. (A, C, Ci,) ZCL423 GFP fluorescence in live embryos. (A) Dorsolateral view of a stage 14 embryo during dorsal closure. GFP is prominently localized at the leading edge (LE) of the dorsal epidermis and in cells of the amnioserosa (AS). (B) Anti-Zasp52 antibody immunofluorescence in fixed, stage 14/15 embryo, dorsolateral view. (C) Striated pattern of GFP in the somatic musculature and muscle attachment sites of a late stage embryo, lateral view. (Ci) Deeper section of embryo showing striated GFP pattern in pharyngeal and visceral muscle surrounding the gut. (D) ZCL423 GFP fluorescence in the body wall muscles of live third instar larvae. Dorsal view (top), ventral view (bottom). (Di) Magnified view of dorsal body wall muscles showing GFP fluorescence in striated pattern. Scale bars show microns.

expression. Indeed, engrailed-Gal4-directed expression of UASZasp52-IR in the posterior domain of ZCL423 larval wing imaginal discs resulted in robust knockdown of GFP fluorescence (Fig. 2C) verifying that the ZCL423 protein trap targets the Zasp52 gene, henceforth referred to as Zasp52ZCL423. While homozygous Zasp52ZCL423 individuals are viable, partial deletion of the Zasp52 gene (Zasp52D56) results in recessive embryonic or early larval lethality (Jani and Schock, 2007). To determine whether the Zasp52ZCL423 insert causes a reduction of function, a complementation test was performed. Comparable numbers of heteroallelic Zasp52D56/Zasp52ZCL423 and control adults were recovered (Fig. 2D), suggesting that the GFP-tagged proteins encoded by Zasp52ZCL423 were functional for development, consistent with the homozygous viability of the stock. Given the extensive expression and requirements for Zasp52 in muscle, adult flies were tested for climbing ability to assess locomotory function. Although there was a small but significant difference in the time for Zasp52D56 heterozygotes to climb a designated distance compared with Zasp52ZCL423 heterozygotes or heteroallelic animals, the range of times from 4 to 6 s were consistent with published reports for normal flies. This contrasts with locomotor-impaired flies that require tens to hundreds of seconds to perform a similar climb (Palladino et al., 2002). Incidentally, a previous study showed that no locomoter defects were detected in larvae when Zasp52 isoforms were depleted from neuronal tissue using the neuron-specific Elav-Gal4 driver, but significant impairment was observed upon muscle-specific knockdown (Benna et al., 2009). Thus, these results suggest that the Zasp52ZCL423 transposon insertion does not cause a measurable loss of Zasp52 function under normal lab conditions.

1.2. Zasp52 is widely expressed in nonmuscle tissue throughout development 1.2.1. Expression in the larval central nervous system, secretory glands, and proventriculus The most obvious expression of Zasp52ZCL423 GFP in living larvae was in the body wall musculature. Yet upon dissection of larval tissues, lower intensity GFP expression was appreciated in nonmuscle tissues such as the central nervous system and imaginal discs, epithelial precursors to the adult body structures (Fig. 3). Thus, Zasp52 GFP expression was investigated in more detail and colocalized with molecular markers such as E-Cadherin, Armadillo (b-catenin), Integrin, and Actin, to highlight cell adhesive structures and polarity. In the larval central nervous system, Zasp52ZCL423 was detected in a few individual neurons in the brain and posterior ventral nerve cord (Fig. 3Aii, arrows). GFP was also specifically expressed in the neuroepithelial proliferative centers of the developing optic lobes, which have a more columnar appearance and accumulate other typical epithelial markers such as ECadherin (Fig. 3A, red lines) (Egger et al., 2007; Hayden et al., 2007). Restricted expression of GFP was observed in the neuroendocrine ring gland that is closely associated with the brain (Fig. 3A, yellow outline) (Adam et al., 2003; Siegmund and Korge, 2001). In particular, the distinct medial secretory cells comprising the corpus allatum expressed Zasp52ZCL423 GFP, which was enriched in focal contact-like puncta at the cell peripheries, and crescentshaped figures largely independent of, but often adjacent to, integrin receptors (Fig. 3C and D–Dii). Curiously, this component of the Zasp52ZCL423 expression pattern was not observed in the ring

70

B. Stronach / Gene Expression Patterns 15 (2014) 67–79

Fig. 2. The ZCL423 GFP protein trap resides in the Zasp52 locus. (A) Screenshot from the GBrowse viewer in FlyBase of the Zasp52 locus on chromosome 2 encoding alternative splice forms. Two GFP protein trap insertions are indicated by the inverted red triangles. Grey highlighting shows the flanking DNA captured by plasmid rescue of the ZCL423 insertion, which falls just upstream of the last exon of the RK transcript encoding eight C-terminal amino acids. Schematic of full-length Zasp52 protein showing PDZ, ZM (Zasp motif), and LIM domains relative to two GFP protein trap insertions. (B) GFP-tagged proteins are detected by Western immunoblot with anti-GFP (left panel) or antiZasp antibodies (right panel) in embryonic (E), larval (L), or adult (A) lysates from Zasp52 protein trap lines or ZCL3111 encoding Ilk::GFP. Molecular weight markers are indicated. (C) RNAi knockdown of Zasp52 depletes ZCL423 GFP fluorescence in the posterior, Engrailed (En) domain of the larval wing imaginal disc (bottom panel). Scale bar is 30 lm. (D) Graphs of results from complementation and climbing tests. Zasp52ZCL423 complements Zasp52D56 loss of function. Bars show the mean time ± SD.

gland of Zasp52G00189 larvae (Fig. S1), although both lines were positive for Zasp52 in the trachea associated with the ring gland. In addition, a subset of cells in the larval lymph gland showed weak GFP fluorescence. Other tissues found in association with the dissected central nervous system revealed additional fluorescence from the Zasp52 protein traps. For instance, a strong GFP signal was associated with the proventriculus, a folded valve-like structure derived from cells of the foregut/midgut boundary of the digestive tract (Johansen et al., 2003; Lechner et al., 2007). The periodic sarcomeric structure of the visceral muscle enveloping the proventriculus was evident, along with apical junctional localization within the ectodermal and endodermal epithelium (Fig. 3B, open arrowheads). Expression was also present in the salivary glands, which are secretory tubular

epithelia, but not in the attached fatbody tissue (Fig. 4). Confocal sections at various positions along the apical/basal axis of the cells showed GFP colocalization with integrin on basal and lateral cell surfaces (Fig. 4A and B), but GFP also specifically highlighted basal membrane infoldings (Fig. 4A, cyan arrowheads) (Tjota et al., 2011), the apical membrane, and adherens junctions, where integrin localization was minimal. While Zasp52 GFP was not particularly enriched at septate junctions (Fig. 4C, red arrowheads), it was not excluded from this region of the membrane to the same extent as integrin. 1.2.2. Expression of Zasp52 in the larval imaginal disc epithelia Zasp52 was localized both apically and basally in polarized epithelial cells comprising the imaginal discs (Figs. 3, 5, 6 and, S1). For

B. Stronach / Gene Expression Patterns 15 (2014) 67–79

71

Fig. 3. Zasp52ZCL423 GFP is expressed in nonmuscle tissues of a dissected larva. (A and B) Single confocal tissue sections labeled with DAPI for nuclei (blue), anti-E-cadherin (red, single channel in Ai, Bi), and Zasp52ZCL423 GFP (green, single channel Aii, Bii). (A–Aii) Red lines mark the inner proliferative center (ipc) of the optic lobe in the brain. The yellow line encircles the ring gland (rg). White arrows point to several GFP-positive neurons in the brain and central nervous system (cns). Tissues associated with the cns are imaginal discs (id). (B–Bii) Open arrowheads point to GFP localization in the muscle and epithelial tissue of the proventriculus (pv). (C and D) Colocalization in the larval endocrine ring gland of bPS-integrin (red, single channel in Di), Zasp52ZCL423 (green, single channel in Dii), and nuclei (DAPI in blue). (C) Boxed area surrounds the medial corpus allatum. (D) Magnification of a single confocal section of the region boxed in panel C. Arrow points to circular figure positive for Zasp52 protein. Arrowhead indicates GFP signal adjacent to integrin signal. GFP is notably absent from surrounding cells compared to integrin. Scale bars are in microns.

instance in the eye-antennal disc, Zasp52 GFP signal was noted clearly in apical (ap) junctions of cells in the morphogenetic furrow (mf) and opposing peripodial membrane (Fig. 5A, aii and C). At the basal (ba) surface of the eye disc, GFP was enriched in puncta arranged like stitches, alternating with slightly larger punctate accumulations of bPS-integrin receptors marking focal adhesionlike structures (Fig. 5B, Bi, and Bii). Zasp52 GFP-expressing hemocytes were also observed in association with the basal surface of the eye disc epithelium (Fig. 5A, ai, and aii), consistent with reported expression of Zasp52 in cultured hemocyte-like cell lines (Jani and Schock, 2007). Closer examination of the developing larval eye disc revealed strong Zasp52 GFP expression in cells anterior to and within the morphogenetic furrow (Fig. 5A and C). Posterior to the furrow, as cellular preclusters of the differentiating ommatidia emerge (Fig. 5C, open arrows, and D–F), GFP expression diminished but was still detectable in the photoreceptor clusters and surrounding interommatidial cells (Fig. 5D–F). E-Cadherin protein highlights the recruitment of the initial photoreceptors, R8, R2 and R5, and becomes enriched at the plasma membrane interfaces between these cells (Fig. 5C, Ci, circle) (Mirkovic and Mlodzik, 2006). Another adherens junction protein, Armadillo (Arm or b-catenin), labels all cells with particularly intense staining at the photoreceptor cell interfaces (Fig. 5D and E). Comparison of Zasp52 GFP with these markers revealed coexpression within the developing photoreceptors, but unlike Arm, Zasp52 was not uniformly associated with the zonula adherens. Instead, it appeared conspicuously cytoplasmic in R2 and R5 (Fig. 5D and E), largely absent from R8 in the given image plane (Fig. 5E), but enriched in puncta at many junctional vertices where multiple cell membranes meet (Fig. 5D and E). Zasp52 GFP was also present in R3 and R4, the next photoreceptor cells differentiating in the cluster (Fig. 5D–F). Intercellular interactions between R3 and R4 are involved in the planar polarized rotation of the ommatidial clusters relative to the equator (eq) of the eye disc, and one key protein that is required for this process is the atypical Cadherin, Flamingo (Fmi) (Das et al., 2002; Ho et al., 2010). Simultaneous visualization of Zasp52 GFP with Fmi, which becomes enriched in R3/R4 cells, affirmed the presence of Zasp52 in these cells (Fig. 5F, arrowheads).

Similar to the eye-antennal disc, in the wandering third instar wing imaginal disc epithelium, Zasp52 GFP was localized prominently in both basal and apical domains (Fig. 6). Basally, the GFP signal appeared like stitches along threads, alternating with bPSintegrin receptor foci identifying focal contact-like structures (Fig. 6A–Aiii). Compared with IlkZCL3111, encoding integrin-linked kinase::GFP, whose signal was strongly codistributed with integrin (Fig. 6E–Eiii), the alternating signals of integrin and Zasp52 were distinct. Apically, Zasp52 GFP was detected in the epithelial zonula adherens, but with less uniform distribution than E-Cad (Fig. 6B– Biii and Di–Diii). Also, GFP fluorescence was strongly enriched in cell contacts flanking the evolving dorsal–ventral and anterior– posterior compartment boundaries (Fig. 6Bi, Biii, Di and Diii). Indeed, in cells positioned at these boundaries, Zasp52 appeared to accumulate preferentially in puncta aligned with the axis, abutting or coincident with vertices formed at the interface of three or more cells (Fig 6Bi–Biii, arrowheads). Accumulation at junctional vertices was also readily observed as a circle of dots in a field of cells containing a mitotic cell bounded by up to eight neighbors (Fig. 6C, circle). These distributions were also seen in the wing discs of Zasp52G00189 larvae (Fig. S1). Whether this intriguing localization of Zasp52 reflects a requirement for, or simply results from boundary formation, awaits further functional studies. 1.2.3. Somatic expression during oogenesis The somatically-derived follicle cells (FCs) that surround the germline tissue during oogenesis in the female constitute another epithelial tissue that expresses high levels of Zasp52 protein (Fig. 7). As individual egg chambers (i.e. follicles) grow and mature, several populations of FCs become morphologically and functionally distinct, including the squamous nurse cell FCs anteriorly, the columnar oocyte FCs posteriorly, and the invasive border cell (BC) cluster that migrates collectively between the germline nurse cells toward the anterior boundary of the oocyte. Zasp52 GFP was observed in differentiating follicle cells from the earliest time of their origin in the germarium (Fig. 8D) and remained in all of these FC populations. Notably, in the BC cluster, Zasp52 GFP was enriched at the tip of cellular processes extending from the leading cells as they invaded the germline (Fig. 7A–Aii, cyan arrowhead).

72

B. Stronach / Gene Expression Patterns 15 (2014) 67–79

Fig. 4. Colocalization of Zasp52ZCL423 and bPS-integrin in the larval salivary gland. (A–C) Single confocal sections at basal (A), lateral (B), and apical, luminal (C) positions. Individual channels are labeled beneath. (Ai–Ci) bPS-integrin. (Aii–Cii) Zasp52ZCL423 GFP. Proteins colocalize on basolateral membranes. Attached fatbody (fb) expresses only integrin, not Zasp52. Cyan arrowheads indicate basal membrane folds enriched for Zasp52. Yellow arrows highlight apical junctions. Red arrowheads mark septate junctions where integrin is excluded. Scale bar is 10 lm.

Also, similar to other epithelia described here, GFP was detected in both apical and basal locations within the cuboidal and columnar FCs, though it is noteworthy that the apicobasal axis is inverted relative to that of the primary embryonic epithelium. The apical domain is juxtaposed with the germline, and discrete apical foci of GFP correspond to adherens junctions (Fig. 7A–Aii, yellow arrow). Basal enrichment of GFP mirrors the planar polarization of the basal actin cytoskeleton (Fig. 7B–Bii) and basement membrane surrounding the egg chambers (Gutzeit et al., 1991). This was evident in stage 10 egg chambers where actin stress fiber bundles are oriented perpendicular to the long axis of the follicle and terminate in massive focal adhesions (Fig. 7B, Bi inset). These structures contain integrin receptors and many other focal adhesion proteins including a-actinin and Zasp52 (Delon and Brown, 2009). Zasp52 GFP also decorated the stress fibers (Fig. 7Bii inset, yellow arrowheads) but did not appreciably colocalize with basolateral cortical actin in the columnar FCs at this stage (Fig. 7B– Bii, inset, cyan arrowhead). However at later stages, actin and GFP did overlap around the circumference of the flattened FCs, presumably within zonula adherens (Fig. 7C–Cii), and basally, Zasp52 GFP continued to highlight peripheral adhesions. Finally, Zasp52 GFP was observed in striated muscle tissue surrounding the egg chamber assembly lines (Fig. 7B, m) as shown for Zasp52G00189 GFP (Hudson et al., 2008).

1.2.4. Colocalization with myosin light chain In the embryo, larval imaginal discs, and somatic follicle cells of the developing egg chambers, evidence has been presented of polarized Zasp52 GFP accumulation in morphogenetic tissues, motile cells, and at compartment boundaries. Many of the same locations become enriched for actin and myosin, serving the dual purposes of architectural integrity and contractility. To determine the relative positions of myosin and Zasp52 in nonmuscle tissues, ZCL423 GFP and the nonmuscle myosin light chain, Spaghetti squash (Sqh), fused to the mCherry fluorescent protein (Martin et al., 2009), were visualized simultaneously (Fig. 8). In most cases, such as at the LE of the embryonic dorsal epidermis, the photoreceptor preclusters emerging from the morphogenetic furrow of the eye imaginal disc, and the axial compartment boundaries of the wing disc, Zasp52 GFP and Sqh::mCherry were found in discontinuous foci, associated with each other but typically alternating, rather than coincident (Fig. 8A–C). This pattern was reminiscent of the alternating stitch-like localization of Zasp52 and integrin in the basal adhesions of the imaginal discs (Figs. 5 and 6). This pattern of periodic enrichment was less obvious in follicle cell populations of the developing egg chambers, however it is worth noting that Zasp52 and Sqh proteins were both present in the follicle cell precursors at the earliest point in the germarium where they encapsulate the germline, in region 2b (Fig. 8D), and they are both

B. Stronach / Gene Expression Patterns 15 (2014) 67–79

73

Fig. 5. Zasp52ZCL423 is distributed both apically and basally in the larval eye imaginal disc. (A) Single confocal section of the eye disc triple-labeled for nuclei (DAPI, blue), bPSintegrin (red), and Zasp52ZCL423 GFP (green). Boxed regions are magnified in the corresponding panels. (ai and aii) Cross-section of cells in the morphogenetic furrow showing basal (ba) or apical (ap) enrichment of integrin (ai) or Zasp52 (aii), respectively. (B) Magnified view of confocal Z-axis projection of the basal surface of the eye disc showing focal adhesion-like structures marked by integrin puncta (Bi) interspersed with Zasp52 puncta (Bii). (C) Z-projection of tangential sections through the apical surface of cells in the morphogenetic furrow (mf) and emerging photoreceptor preclusters posterior to the furrow (open arrows). Circles surround a single cluster of differentiating photoreceptors cells. Arrowheads point to junction between R3/R4 cells. Anterior–posterior and equatorial–lateral axes are indicated. Nuclei (DAPI, blue), E-Cadherin (red, single channel in Ci), and Zasp52ZCL423 GFP (green, single channel in Cii). (D and E) High magnification view of two photoreceptor clusters (rows 2 and 3 (D), rows 3 and 4 (E)) with R cells indicated. Nuclei (blue), Armadillo (red, single channel indicated) and Zasp52ZCL423 GFP (green, single channel in grey). (F) Magnified view of photoreceptor clusters from rows 4 and 5 (numbers mark R cells) stained for Flamingo (red, single channel indicated), Zasp52ZCL423 GFP (green, single channel far right), and DAPI (blue nuclei). Arrowheads point to membrane interface between R3/R4 as ommatidial rotation is underway. Scale bars indicate microns.

enriched in centripetal migrating FCs and the BC cluster (not shown). Altogether, this limited tissue analysis demonstrates that although Zasp52 proteins are not necessarily coexpressed or localized precisely with myosin in nonmuscle tissues, they are often organized and deployed concurrently with actin and myosin in morphogenetic tissues. This paper describes the widespread tissue distribution and subcellular localization of the Drosophila ALP/Enigma family member, Zasp52, outside of the musculature. Identification of a GFP protein trap line, ZCL423, in the Zasp52 locus, provided a fluorescent tool to visualize endogenous isoforms in many tissues and at different life stages. Comparison of ZCL423 distribution with an extant Zasp52 protein trap (Zasp52G00189) revealed nearly identical expression and localization patterns, with the exception of the ring gland, suggesting that the patterns are largely representative of the endogenous locus. Yet, differences observed by Western immunoblot in the suite of proteins encoded by the two lines must reflect the difference in insertion site within the locus. Whether the minor disparities in protein isoform size and levels are due to unique tagged splice forms, the position of specific internal promoters,

posttranslational effects, or some other mechanism, will be a topic for future investigation. The distribution of Zasp52 at the tissue and cellular levels may have functional implications for cell adhesion and morphogenesis. In epithelial cells of the imaginal discs, for example, basal Zasp52 GFP localization was prominent, tracking linearly with fibers of the cytoskeleton or extracellular matrix, although not coincident with bPS-integrin (schematized in Fig. 9). Back of the envelope accounting of GFP puncta per cell points to approximately 1–2 foci of basal Zasp52 accumulation in each cell in the eye and wing disc. This is borne out visually in XZ or YZ slices through confocal stacks of whole discs. The implication of Zasp52 distribution and its submicron separation from bPS-integrin foci is that their physical and functional interaction may also be separable in these tissues. Curiously, knockdown of Zasp52 protein levels upon expression of Zasp52-IR in the developing discs had no effect on adult morphology. This result differs from focal adhesion proteins such as Talin or members of the IPP (ILK, PINCH, Parvin) complex, which tightly colocalize with integrin (e.g. Fig. 6) and whose clonal loss or knockdown in the wing disc causes wing blisters, reflecting their

74

B. Stronach / Gene Expression Patterns 15 (2014) 67–79

Fig. 6. Zasp52ZCL423 is enriched basally, and apically along compartment boundaries in the larval wing imaginal disc. (A) Two slice confocal Z-projection of the basal surface of a disc centered on the wing pouch. Boxed area is magnified in subsequent panels (Ai–Aiii). Triple-labeling shows nuclei (DAPI, blue), bPS-integrin (red, single channel in Aii), and Zasp52ZCL423 GFP (green, single channel in Aiii). Positions of XZ and YZ slices through the entire disc are shown with apical surfaces facing inward. (Ai–Aiii) Arrowheads indicate Zasp52 protein adjacent to integrin signal highlighting focal adhesion-like structures. (B–D) Tangential sections at apical junctions of wing disc cells labeled for nuclei (DAPI, blue), E-Cadherin (red, single channel in Bii, Dii), and Zasp52ZCL423 GFP (green, single channel in Biii, Diii). Boxed regions in (B) refer to relative positions of magnified images in subsequent panels. (Bi–Biii) Intersection of anterior–posterior (AP) and dorsal–ventral (DV) axes, as indicated. Zasp52 puncta (arrowheads) are enriched around the vertices of neighboring cells flanking the DV compartment boundary in a late third instar disc, whereas E-Cad distribution is more uniform. (C) Field of cells centered on the DV compartment boundary with dividing cell encircled by punctate Zasp52 accumulations at junctional vertices. (Di–Diii) AP compartment boundary (arrowheads) where flanking cells are more square-shaped and Zasp52 becomes enriched at the level of E-Cad-positive adherens junctions. (E) Confocal Z-projection of the basal surface of a wing disc centered on the wing pouch. Boxed region is magnified in panels Ei–Eiii. Nuclei are shown in blue (DAPI). bPS-integrin (red, single channel in Eii) and IlkZCL3111 GFP (green, single channel in Eiii) are colocalized in focal adhesion-like structures (arrowheads). Scale bars indicate microns.

requirement in integrin-based adhesion (Brown et al., 2002; Clark et al., 2003; Vakaloglou et al., 2012; Zervas et al., 2001). Thus, despite the functional requirements for Zasp52 in the assembly of integrin adhesion sites in cultured cells and muscles, where they localize coincidently (Jani and Schock, 2007), it remains to be determined whether basal Zasp52 proteins in epithelial cells are required for integrin adhesion complex formation or function. Moreover, the localization of Zasp52 in zonula adherens and junctional vertices of neighboring epithelial cells (Fig. 9) suggests potential additional roles in cell–cell adhesion, independent of integrin. Indeed, an intriguing connection can be drawn between localization of Zasp52 at Z-lines in the sarcomeres of muscle, anchoring microfilaments against myosin contraction, and its enrichment at epithelial compartment boundaries, where evidence suggests that myosin-based cortical tension underlies in part the lineage restriction (Landsberg et al., 2009; Major and Irvine, 2006). At the DV compartment boundary in the late third instar wing disc for instance, Zasp52 GFP localization is reminiscent of Myosin II and

Actin staining (Major and Irvine, 2005, 2006), but colocalization with the myosin light chain, Sqh, revealed that the two signals are interspersed, recalling the distribution of Zasp52 and muscle myosin at the Z- and M-lines of myofibrils respectively. The LE of the dorsal epidermis in the embryo is another example of a tissue boundary with massive accumulation of actin and myosin under tension (Franke et al., 2005; Kiehart et al., 2000; Young et al., 1993), in which Zasp52 becomes highly enriched. The mechanical coupling of contraction from individual cells, through their adherens junctions, to neighbors all across the LE fortifies a supracellular actomyosin purse string to coordinate cell lengthening in the DV axis and uniform progression of the epithelial front toward the dorsal midline (Jacinto et al., 2002; Young et al., 1993). Compromising cell–cell or cell–ECM junctions at the LE impairs dorsal closure, underscoring the importance of adhesion complexes in anchoring the forces generated by contractile tension (Gorfinkiel and Arias, 2007; Homem and Peifer, 2008; Narasimha and Brown, 2004). In this context, the periodicity of Zasp52, alternating with the Sqh ‘bars-on-a-string’ distribution (Franke et al., 2005),

B. Stronach / Gene Expression Patterns 15 (2014) 67–79

75

Fig. 7. Zasp52ZCL423 expression in the somatic follicle cell populations during oogenesis. (A–C) Single confocal sections of egg chambers at different developmental stages. Fluorescent labels show nuclei (DAPI, blue), actin (phalloidin, red, single channel in Ai–Ci), and Zasp52ZCL423 GFP (green, single channel in Aii–Cii). Zasp52ZCL423 GFP is absent from the germline including the oocyte (oo), but present in encircling follicle cell layer with apposed apical (ap) membrane domain and basal (ba) domain facing outward. Anterior–posterior (AP) axis is shown. (A) Longitudinal cross-section shows that Zasp52 GFP is prominent basally, but decorates apical adherens junctions (arrow). Delaminating border cell (BC) follicle cell cluster invades the germline by sending out a leading edge process, enriched for actin and Zasp52 GFP (cyan arrowhead). (B) Tangential section of follicle cells over the oocyte during a stage when basal actin filaments become circumferentially planar polarized, perpendicular to the long axis of the egg. Inset is magnified view of several follicle cells. Zasp52 decorates actin stress fibers (yellow arrowheads), which terminate in focal adhesions, strongly enriched for Zasp52. Note that Zasp52 does not overlap lateral cortical actin (cyan arrowhead). Associated striated muscle (m) also expresses Zasp52. (C) Tangential section of flattened follicle cell layer of an elongated egg chamber. Actin and Zasp52 colocalize at apical junctions, but only Zasp52 remains at the periphery of cells on basal surface. Scale bar is 20 lm.

resembles the myofibril ultrastructure, but on a multicellular scale (Fig. 9). A sarcomeric organization of nonmuscle myosin at the apical junctional region of epithelial cells is widespread in many murine tissues (Ebrahim et al., 2013). Moreover, the periodicity of myosin is matched across junctions of adjacent cells to integrate tensional forces throughout the tissue (Ebrahim et al., 2013). In light of these observations, I speculate that Zasp52 in flies, and presumably the broader mammalian ALP/Enigma family members, may be enriched at particular nonmuscle cell or tissue sites as structural reinforcement to withstand increased actomyosin contractile tension along morphogenetic boundaries. Zasp52 may be recruited to its sites of localization via multiple mechanisms. The PDZ domain can bind directly to a-actinin, a prominent actin bundling protein, however genetic evidence indicates that at least in muscle tissue, Zasp52 is upstream of a-actinin

recruitment at Z-discs (Jani and Schock, 2007; Katzemich et al., 2013). Zasp52 isoforms also encode variable numbers of LIM domains. Recent proteomic studies have demonstrated that nonmuscle LIM proteins including PDLIM1, 2, 4, 5, and 7, are recruited to maturing focal adhesions under contractile tension, and that their recruitment is sensitive to myosin II inhibition (Schiller et al., 2011). These results along with in vitro and cell-based studies have led to the proposal that LIM domains function as cytoskeletal mechanosensors (Schiller and Fassler, 2013). Mechanistically, contractile tension may alter the conformation of certain proteins to expose additional LIM domain binding sites, allowing the strengthening of protein network connections comprising the ‘integrin adhesome’ (Wolfenson et al., 2013). Another possibility is that within the maturing focal adhesion, actin filament bundling and continued fortification of the adhesome into a highly stratified

76

B. Stronach / Gene Expression Patterns 15 (2014) 67–79

Fig. 8. Comparison of cellular distributions of Zasp52ZCL423 GFP and nonmuscle myosin light chain. (A) Dorsal view at the completion of embryonic dorsal closure. The epidermal leading edges are juxtaposed at the dorsal midline revealing the alternating ‘bars on a string’ pattern of Sqh (red, magnified in Ai) and Zasp52 (green, magnified in Aii). (B) Confocal Z-projection of the larval eye-antennal disc. Boxed region of morphogenetic furr ow magnified in subsequent panels (Bi–Biii). Nuclei (DAPI, blue), Sqh (red, single channel in Bii), and Zasp52 (green, single channel in Biii). Interdigitating Sqh and Zasp52 puncta in lines and arcs highlight newly forming photoreceptor preclusters emerging posterior to the furrow (arrows). (C) Confocal section of larval wing imaginal disc at the level of the adherens junctions, with the same fluorescent markers. Sqh (red, and in Cii) and Zasp52 (green, and in Ciii). Boxed region of pouch shows intersection of DV and AP compartment boundaries, magnified in subsequent panels. Yellow arrows indicate enrichment of Zasp52 at AP boundary, typically interspersed with Sqh. Distribution is also apparent around the circumference of dividing cells (cyan arrows). (D) Zasp52 expression (green, and in Dii) in the differentiating follicle cells (fc) of the germarium region 2b at the earliest stages of oogenesis and later. Surrounding muscle (m) is also highly enriched for Zasp52. In contrast, Sqh (red, and in Di) is absent from muscle, but present in the germline (gl), in addition to its expression in follicle cells. Scale bars show microns.

B. Stronach / Gene Expression Patterns 15 (2014) 67–79

77

Nevertheless, the fluorescent Zasp52 protein traps described here add to the tremendous toolbox to visualize and interrogate epithelial cell biology in morphogenetic contexts. 2. Experimental procedures 2.1. Fly stocks w1118 was used as a wildtype control line. Zasp52D56 is a homozygous lethal deletion within the locus (Jani and Schock, 2007). Zasp52JF01133 encodes a long hairpin inverted repeat (IR) construct developed by TRiP in the Valium1 vector, containing 10 copies of the UAS site (Ni et al., 2008). Protein trap lines used were Zasp52ZCL423 [P(PTT-G?)Zasp52ZCL423], Zasp52G00189 [P(PTT-GB) Zasp52G00189], and IlkZCL3111 [P(PTT-GB)IlkZCL3111] (Morin et al., 2001). Sqh-Sqh::mCherry is a red fluorescent protein fusion with nonmuscle myosin light chain (Martin et al., 2009). Engrailed-Gal4 (En-G4) was used to express Gal4-responsive UAS-transgenes in the posterior compartment of embryonic segments and imaginal discs. 2.2. Molecular biology Fig. 9. Summary of proposed Zasp52 distribution in epithelia versus muscle. Component proteins are labeled at right. Zasp52 (green diamonds) is detected apically in adherens junctions enriched at junctional vertices where it may function to anchor actomyosin contractility along the zonula adherens similar to its role at Zlines of muscle sarcomeres. At the basal membrane domain of epithelial cells, Zasp52 accumulates in puncta tracking cytoskeleton or extracellular matrix (ECM) fibers. Resembling ‘stitches of thread’, Zasp52 puncta are distinct from and adjacent to integrin receptor foci, which colocalize with the Ilk-PINCH-Parvin (IPP) complex of the focal adhesion.

structure might arrange binding sites with precise periodicity, facilitating recruitment of proteins with a serial arrangement of LIM repeats (Kadrmas and Beckerle, 2004; Kanchanawong et al., 2010; Petit et al., 2003; Schiller and Fassler, 2013). The extensive nonmuscle expression of Zasp52 may serve as an important alternative context to investigate the functional properties of nonmuscle ALP/Enigma family members related to vertebrate PDLIM1, 2, 4, 5, and 7 proteins. For example, Zasp52 expression was observed in distinct neurons and neuroepithelial proliferative centers of the larval brain. In that context, it is provocative to consider whether Drosophila might hold insights relevant to mammalian nervous system function, given that psychiatric disorders and neural tumors have been linked to polymorphisms or altered expression of PDLIM5/ENH (Herrick et al., 2010; Horiuchi et al., 2013; Lasorella and Iavarone, 2006; Wong et al., 2012; Zain et al., 2013). In conclusion, characterization of GFP-bearing protein traps inserted into the Zasp52 locus has allowed examination of the extensive pattern of Zasp52 expression in multiple cell types besides the musculature. A description of the subcellular distribution of Zasp52 protein isoforms raises the possibility of roles, not only in epithelial integrin-based cell adhesion, but also in strengthening tissues under the contractile forces of apical nonmuscle actomyosin networks. Notably, localization of Zasp52 GFP did not reveal a nuclear pool of protein in any cell types examined. Either nuclear-localized Zasp52 is transient and difficult to detect or alternative spliceforms that are not tagged by GFP may be found in the nuclear compartment. Alternatively, the nuclear functions of vertebrate ALP/Enigma family members were acquired after the ancestral split with the invertebrate lineage. Whether Zasp52 has essential functions outside of the musculature will require additional experiments using tissue-specific reagents.

Plasmid rescue and inverse PCR experiments were based on the protocol from the Berkeley Drosophila Genome Project modified by Roger Hoskins of the Drosophila Gene Disruption Project (http://flypush.imgen.bcm.tmc.edu/pscreen). The primers, EY.3.F and EY.3.R, were used for amplification of purified, HinP1 I-digested, and ligated, ZCL423 genomic DNA. Product DNA was purified and cloned into the pGEM-T vector (Promega Corp). After bacterial transformation, insert DNA from several individual colonies was further amplified and sequenced (Genewiz Inc). 2.3. Tissue staining, antibodies, and fluorescence detection Immunofluorescence staining procedures for embryos, third instar larval tissues, and ovaries from >3-day old mated females were standard (Patel, 1994). Reagents used for tissue staining include rabbit anti-Zasp52 FL at 1:400 (Jani and Schock, 2007), mouse anti-bPS-integrin at 1:50–1:200 (CF.6G11 DSHB, (Brower et al., 1984)), rat anti-DE-Cadherin at 1:50 (DCAD2, DSHB, (Oda et al., 1994)), mouse anti-Flamingo at 1:20 (#74, DSHB, (Usui et al., 1999)), mouse anti-Armadillo at 1:50 (N2 7A1, DSHB, (Peifer et al., 1994)), TexasRed-Phalloidin (1:500), TexasRedsecondary antibody conjugates at 1:200 (Jackson ImmunoResearch Laboratories), and Vectashield mounting medium with DAPI (Vector Labs). Fluorescent proteins in fixed samples were imaged without antibody probes. Live embryos were dechorionated in diluted bleach, washed, and mounted in 1:1 mix of halocarbon oils 27 and 700 for microscopy. Images were obtained by laserscanning confocal microscopy using an Olympus FV1000 Fluoview system on an IX81 compound inverted microscope and assembled using ImageJ and Adobe Photoshop software. Contrast was adjusted for maximal clarity. Images of live larvae were obtained with NIS-Elements software using a Nikon DS-Fi1 digital camera mounted on a Nikon SMZ1500 stereomicroscope. 2.4. Western blot Protein lysates were made by homogenizing collections of embryos laid overnight at 22 °C, or 8–12 wandering third instar larvae or adults in RIPA buffer supplemented with protease inhibitors. 25–30 lg of total protein per sample was separated by SDS– PAGE. After Western blotting, duplicate membranes were probed with either mouse monoclonal anti-GFP at 1:1000 (B-2, sc-9996,

78

B. Stronach / Gene Expression Patterns 15 (2014) 67–79

Santa Cruz,) or rabbit anti-Zasp52 FL at 1:625 (Jani and Schock, 2007). 2.5. Complementation and climbing assays Zasp52D56/CyO females were crossed to either w1118 or Zasp52ZCL423 males and raised at 25 °C. Resulting progeny from three vials were counted. Climbing ability was determined as follows: Five to eight groups of ten flies per genotype were tested in four repetitions of their climbing ability, measured by tapping flies down to the bottom of a graduated cylinder and recording the time it took for 50% of the flies to cross a line marked at 12 cm from the bottom (Palladino et al., 2002). Average climbing time in seconds ±SD was analyzed by Student’s t-test. Acknowledgements Several antibodies were acquired from the Developmental Studies Hybridoma Bank (DSHB) developed under the auspices of the NICHD. TRiP at Harvard Medical School (NIH/NIGMS R01GM084947) provided the transgenic Zasp52 RNAi fly stock. Thanks to Frieder Schöck for providing the Zasp52 mutant and antibody, and to Jeff Hildebrand for helpful comments on the manuscript. I am indebted to John Malta for his successful plasmid rescue of ZCL423 on the first attempt. I also heartily thank Ashley Lennox, Thomas Dohle, Julia Barker, Anthony Gak, and Rebecca Garlena for their assistance with crosses, stock maintenance, and tissue staining. A. Lennox took the confocal images in Fig. 7. Funding was provided by start-up contributions from the University of Pittsburgh School of Arts and Sciences and School of Medicine. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.gep.2014.05.002.

References Adam, G., Perrimon, N., Noselli, S., 2003. The retinoic-like juvenile hormone controls the looping of left–right asymmetric organs in Drosophila. Development 130, 2397–2406. Arimura, T., Hayashi, T., Terada, H., Lee, S.Y., Zhou, Q., Takahashi, M., Ueda, K., Nouchi, T., Hohda, S., Shibutani, M., Hirose, M., Chen, J., Park, J.E., Yasunami, M., Hayashi, H., Kimura, A., 2004. A Cypher/ZASP mutation associated with dilated cardiomyopathy alters the binding affinity to protein kinase C. J. Biol. Chem. 279, 6746–6752. Benna, C., Peron, S., Rizzo, G., Faulkner, G., Megighian, A., Perini, G., Tognon, G., Valle, G., Reggiani, C., Costa, R., Zordan, M.A., 2009. Post-transcriptional silencing of the Drosophila homolog of human ZASP: a molecular and functional analysis. Cell Tissue Res. 337, 463–476. Bouaouina, M., Jani, K., Long, J.Y., Czerniecki, S., Morse, E.M., Ellis, S.J., Tanentzapf, G., Schock, F., Calderwood, D.A., 2012. Zasp regulates integrin activation. J. Cell Sci. 125, 5647–5657. Boumber, Y.A., Kondo, Y., Chen, X., Shen, L., Gharibyan, V., Konishi, K., Estey, E., Kantarjian, H., Garcia-Manero, G., Issa, J.P., 2007. RIL, a LIM gene on 5q31, is silenced by methylation in cancer and sensitizes cancer cells to apoptosis. Cancer Res. 67, 1997–2005. Bowe, R.A., Cox, O.T., Ayllon, V., Tresse, E., Healy, N.C., Edmunds, S.J., Huigsloot, M., O’Connor, R., 2014. PDLIM2 regulates transcription factor activity in epithelialto-mesenchymal transition via the COP9 signalosome. Mol. Biol. Cell 25, 184– 195. Brower, D.L., Wilcox, M., Piovant, M., Smith, R.J., Reger, L.A., 1984. Related cellsurface antigens expressed with positional specificity in Drosophila imaginal discs. Proc. Natl. Acad. Sci. U.S.A. 81, 7485–7489. Brown, N.H., Gregory, S.L., Rickoll, W.L., Fessler, L.I., Prout, M., White, R.A., Fristrom, J.W., 2002. Talin is essential for integrin function in Drosophila. Dev. Cell 3, 569–579. Clark, K.A., McGrail, M., Beckerle, M.C., 2003. Analysis of PINCH function in Drosophila demonstrates its requirement in integrin-dependent cellular processes. Development 130, 2611–2621. Das, G., Reynolds-Kenneally, J., Mlodzik, M., 2002. The atypical cadherin Flamingo links Frizzled and Notch signaling in planar polarity establishment in the Drosophila eye. Dev. Cell 2, 655–666.

Delon, I., Brown, N.H., 2009. The integrin adhesion complex changes its composition and function during morphogenesis of an epithelium. J. Cell Sci. 122, 4363– 4374. Ebrahim, S., Fujita, T., Millis, B.A., Kozin, E., Ma, X., Kawamoto, S., Baird, M.A., Davidson, M., Yonemura, S., Hisa, Y., Conti, M.A., Adelstein, R.S., Sakaguchi, H., Kachar, B., 2013. NMII forms a contractile transcellular sarcomeric network to regulate apical cell junctions and tissue geometry. Curr. Biol. 23, 731–736. Egger, B., Boone, J.Q., Stevens, N.R., Brand, A.H., Doe, C.Q., 2007. Regulation of spindle orientation and neural stem cell fate in the Drosophila optic lobe. Neural Dev. 2, 1. Faulkner, G., Pallavicini, A., Formentin, E., Comelli, A., Ievolella, C., Trevisan, S., Bortoletto, G., Scannapieco, P., Salamon, M., Mouly, V., Valle, G., Lanfranchi, G., 1999. ZASP: a new Z-band alternatively spliced PDZ-motif protein. J. Cell Biol. 146, 465–475. Franke, J.D., Montague, R.A., Kiehart, D.P., 2005. Nonmuscle myosin II generates forces that transmit tension and drive contraction in multiple tissues during dorsal closure. Curr. Biol. 15, 2208–2221. Gorfinkiel, N., Arias, A.M., 2007. Requirements for adherens junction components in the interaction between epithelial tissues during dorsal closure in Drosophila. J. Cell Sci. 120, 3289–3298. Gutzeit, H.O., Eberhardt, W., Gratwohl, E., 1991. Laminin and basement membraneassociated microfilaments in wild-type and mutant Drosophila ovarian follicles. J. Cell Sci. 100 (Pt 4), 781–788. Han, H.F., Beckerle, M.C., 2009. The ALP-Enigma protein ALP-1 functions in actin filament organization to promote muscle structural integrity in Caenorhabditis elegans. Mol. Biol. Cell 20, 2361–2370. Hayden, M.A., Akong, K., Peifer, M., 2007. Novel roles for APC family members and Wingless/Wnt signaling during Drosophila brain development. Dev. Biol. 305, 358–376. Herrick, S., Evers, D.M., Lee, J.Y., Udagawa, N., Pak, D.T., 2010. Postsynaptic PDLIM5/ Enigma homolog binds SPAR and causes dendritic spine shrinkage. Mol. Cell Neurosci. 43, 188–200. Ho, Y.H., Lien, M.T., Lin, C.M., Wei, S.Y., Chang, L.H., Hsu, J.C., 2010. Echinoid regulates Flamingo endocytosis to control ommatidial rotation in the Drosophila eye. Development 137, 745–754. Homem, C.C., Peifer, M., 2008. Diaphanous regulates myosin and adherens junctions to control cell contractility and protrusive behavior during morphogenesis. Development 135, 1005–1018. Horiuchi, Y., Ishikawa, M., Kaito, N., Iijima, Y., Tanabe, Y., Ishiguro, H., Arinami, T., 2013. Experimental evidence for the involvement of PDLIM5 in mood disorders in hetero knockout mice. PLoS ONE 8, e59320. Huang, C., Zhou, Q., Liang, P., Hollander, M.S., Sheikh, F., Li, X., Greaser, M., Shelton, G.D., Evans, S., Chen, J., 2003. Characterization and in vivo functional analysis of splice variants of cypher. J. Biol. Chem. 278, 7360–7365. Hudson, A.M., Petrella, L.N., Tanaka, A.J., Cooley, L., 2008. Mononuclear muscle cells in Drosophila ovaries revealed by GFP protein traps. Dev. Biol. 314, 329–340. Jacinto, A., Wood, W., Woolner, S., Hiley, C., Turner, L., Wilson, C., Martinez-Arias, A., Martin, P., 2002. Dynamic analysis of actin cable function during Drosophila dorsal closure. Curr. Biol. 12, 1245–1250. Jani, K., Schock, F., 2007. Zasp is required for the assembly of functional integrin adhesion sites. J. Cell Biol. 179, 1583–1597. Jankovics, F., Henn, L., Bujna, A., Vilmos, P., Kiss, N., Erdelyi, M., 2011. A functional genomic screen combined with time-lapse microscopy uncovers a novel set of genes involved in dorsal closure of Drosophila embryos. PLoS ONE 6, e22229. Johansen, K.A., Green, R.B., Iwaki, D.D., Hernandez, J.B., Lengyel, J.A., 2003. The DrmBowl-Lin relief-of-repression hierarchy controls fore- and hindgut patterning and morphogenesis. Mech. Dev. 120, 1139–1151. Kadrmas, J.L., Beckerle, M.C., 2004. The LIM domain: from the cytoskeleton to the nucleus. Nat. Rev. Mol. Cell Biol. 5, 920–931. Kanchanawong, P., Shtengel, G., Pasapera, A.M., Ramko, E.B., Davidson, M.W., Hess, H.F., Waterman, C.M., 2010. Nanoscale architecture of integrin-based cell adhesions. Nature 468, 580–584. Katzemich, A., Liao, K.A., Czerniecki, S., Schock, F., 2013. Alp/Enigma family proteins cooperate in Z-disc formation and myofibril assembly. PLoS Genet. 9, e1003342. Katzemich, A., Long, J.Y., Jani, K., Lee, B.R., Schock, F., 2011. Muscle type-specific expression of Zasp52 isoforms in Drosophila. Gene Expr. Patterns 11, 484–490. Kiehart, D.P., Galbraith, C.G., Edwards, K.A., Rickoll, W.L., Montague, R.A., 2000. Multiple forces contribute to cell sheet morphogenesis for dorsal closure in Drosophila. J. Cell Biol. 149, 471–490. Kiess, M., Scharm, B., Aguzzi, A., Hajnal, A., Klemenz, R., Schwarte-Waldhoff, I., Schafer, R., 1995. Expression of ril, a novel LIM domain gene, is down-regulated in Hras-transformed cells and restored in phenotypic revertants. Oncogene 10, 61–68. Krcmery, J., Camarata, T., Kulisz, A., Simon, H.G., 2010. Nucleocytoplasmic functions of the PDZ-LIM protein family: new insights into organ development. BioEssays 32, 100–108. Landsberg, K.P., Farhadifar, R., Ranft, J., Umetsu, D., Widmann, T.J., Bittig, T., Said, A., Julicher, F., Dahmann, C., 2009. Increased cell bond tension governs cell sorting at the Drosophila anteroposterior compartment boundary. Curr. Biol. 19, 1950– 1955. Lasorella, A., Iavarone, A., 2006. The protein ENH is a cytoplasmic sequestration factor for Id2 in normal and tumor cells from the nervous system. Proc. Natl. Acad. Sci. U.S.A. 103, 4976–4981. Lechner, H., Josten, F., Fuss, B., Bauer, R., Hoch, M., 2007. Cross regulation of intercellular gap junction communication and paracrine signaling pathways during organogenesis in Drosophila. Dev. Biol. 310, 23–34.

B. Stronach / Gene Expression Patterns 15 (2014) 67–79 Loughran, G., Healy, N.C., Kiely, P.A., Huigsloot, M., Kedersha, N.L., O’Connor, R., 2005. Mystique is a new insulin-like growth factor-I-regulated PDZ-LIM domain protein that promotes cell attachment and migration and suppresses Anchorage-independent growth. Mol. Biol. Cell 16, 1811–1822. Machuca-Tzili, L., Thorpe, H., Robinson, T.E., Sewry, C., Brook, J.D., 2006. Flies deficient in Muscleblind protein model features of myotonic dystrophy with altered splice forms of Z-band associated transcripts. Hum. Genet. 120, 487–499. Major, R.J., Irvine, K.D., 2005. Influence of Notch on dorsoventral compartmentalization and actin organization in the Drosophila wing. Development 132, 3823–3833. Major, R.J., Irvine, K.D., 2006. Localization and requirement for Myosin II at the dorsal–ventral compartment boundary of the Drosophila wing. Dev. Dyn. 235, 3051–3058. Martin, A.C., Kaschube, M., Wieschaus, E.F., 2009. Pulsed contractions of an actin– myosin network drive apical constriction. Nature 457, 495–499. McKeown, C.R., Han, H.F., Beckerle, M.C., 2006. Molecular characterization of the Caenorhabditis elegans ALP/Enigma gene alp-1. Dev. Dyn. 235, 530–538. Mirkovic, I., Mlodzik, M., 2006. Cooperative activities of drosophila DE-cadherin and DN-cadherin regulate the cell motility process of ommatidial rotation. Development 133, 3283–3293. Morin, X., Daneman, R., Zavortink, M., Chia, W., 2001. A protein trap strategy to detect GFP-tagged proteins expressed from their endogenous loci in Drosophila. Proc. Natl. Acad. Sci. U.S.A. 98, 15050–15055. Narasimha, M., Brown, N.H., 2004. Novel functions for integrins in epithelial morphogenesis. Curr. Biol. 14, 381–385. Ni, J.Q., Markstein, M., Binari, R., Pfeiffer, B., Liu, L.P., Villalta, C., Booker, M., Perkins, L., Perrimon, N., 2008. Vector and parameters for targeted transgenic RNA interference in Drosophila melanogaster. Nat. Methods 5, 49–51. Oda, H., Uemura, T., Harada, Y., Iwai, Y., Takeichi, M., 1994. A Drosophila homolog of cadherin associated with armadillo and essential for embryonic cell–cell adhesion. Dev. Biol. 165, 716–726. Ohno, K., Kato, H., Funahashi, S., Hasegawa, T., Sato, K., 2009. Characterization of CLP36/Elfin/PDLIM1 in the nervous system. J. Neurochem. 111, 790–800. Palladino, M.J., Hadley, T.J., Ganetzky, B., 2002. Temperature-sensitive paralytic mutants are enriched for those causing neurodegeneration in Drosophila. Genetics 161, 1197–1208. Patel, N.H., 1994. Imaging neuronal subsets and other cell types in whole-mount Drosophila embryos and larvae using antibody probes. In: Goldstein, L.S.B., Fyrberg, E.A. (Eds.), Drosophila melanogaster: Practical Uses in Cell and Molecular Biology. Academic Press, San Diego, pp. 446–487. Peifer, M., Sweeton, D., Casey, M., Wieschaus, E., 1994. Wingless signal and Zestewhite 3 kinase trigger opposing changes in the intracellular distribution of Armadillo. Development 120, 369–380. Petit, M.M., Meulemans, S.M., Van de Ven, W.J., 2003. The focal adhesion and nuclear targeting capacity of the LIM-containing lipoma-preferred partner (LPP) protein. J. Biol. Chem. 278, 2157–2168. Qu, Z., Fu, J., Yan, P., Hu, J., Cheng, S.Y., Xiao, G., 2010. Epigenetic repression of PDZLIM domain-containing protein 2: implications for the biology and treatment of breast cancer. J. Biol. Chem. 285, 11786–11792. Schiller, H.B., Fassler, R., 2013. Mechanosensitivity and compositional dynamics of cell-matrix adhesions. EMBO Rep. 14, 509–519. Schiller, H.B., Friedel, C.C., Boulegue, C., Fassler, R., 2011. Quantitative proteomics of the integrin adhesome show a myosin II-dependent recruitment of LIM domain proteins. EMBO Rep. 12, 259–266. Selcen, D., Engel, A.G., 2005. Mutations in ZASP define a novel form of muscular dystrophy in humans. Ann. Neurol. 57, 269–276. Siegmund, T., Korge, G., 2001. Innervation of the ring gland of Drosophila melanogaster. J. Comp. Neurol. 431, 481–491. Tanaka, T., Soriano, M.A., Grusby, M.J., 2005. SLIM is a nuclear ubiquitin E3 ligase that negatively regulates STAT signaling. Immunity 22, 729–736. te Velthuis, A.J., Bagowski, C.P., 2007. PDZ and LIM domain-encoding genes: molecular interactions and their role in development. ScientificWorldJournal 7, 1470–1492.

79

Te Velthuis, A.J., Isogai, T., Gerrits, L., Bagowski, C.P., 2007. Insights into the molecular evolution of the PDZ/LIM family and identification of a novel conserved protein motif. PLoS ONE 2, e189. Tjota, M., Lee, S.K., Wu, J., Williams, J.A., Khanna, M.R., Thomas, G.H., 2011. Annexin B9 binds to beta(H)-spectrin and is required for multivesicular body function in Drosophila. J. Cell Sci. 124, 2914–2926. Torrado, M., Senatorov, V.V., Trivedi, R., Fariss, R.N., Tomarev, S.I., 2004. Pdlim2, a novel PDZ-LIM domain protein, interacts with alpha-actinins and filamin A. Invest. Ophthalmol. Vis. Sci. 45, 3955–3963. Usui, T., Shima, Y., Shimada, Y., Hirano, S., Burgess, R.W., Schwarz, T.L., Takeichi, M., Uemura, T., 1999. Flamingo, a seven-pass transmembrane cadherin, regulates planar cell polarity under the control of Frizzled. Cell 98, 585–595. Vakaloglou, K.M., Chountala, M., Zervas, C.G., 2012. Functional analysis of parvin and different modes of IPP-complex assembly at integrin sites during Drosophila development. J. Cell Sci. 125, 3221–3232. Vallenius, T., Luukko, K., Makela, T.P., 2000. CLP-36 PDZ-LIM protein associates with nonmuscle alpha-actinin-1 and alpha-actinin-4. J. Biol. Chem. 275, 11100– 11105. Vallenius, T., Scharm, B., Vesikansa, A., Luukko, K., Schafer, R., Makela, T.P., 2004. The PDZ-LIM protein RIL modulates actin stress fiber turnover and enhances the association of alpha-actinin with F-actin. Exp. Cell Res. 293, 117–128. van der Meer, D.L., Marques, I.J., Leito, J.T., Besser, J., Bakkers, J., Schoonheere, E., Bagowski, C.P., 2006. Zebrafish cypher is important for somite formation and heart development. Dev. Biol. 299, 356–372. Vatta, M., Mohapatra, B., Jimenez, S., Sanchez, X., Faulkner, G., Perles, Z., Sinagra, G., Lin, J.H., Vu, T.M., Zhou, Q., Bowles, K.R., Di Lenarda, A., Schimmenti, L., Fox, M., Chrisco, M.A., Murphy, R.T., McKenna, W., Elliott, P., Bowles, N.E., Chen, J., Valle, G., Towbin, J.A., 2003. Mutations in Cypher/ZASP in patients with dilated cardiomyopathy and left ventricular non-compaction. J. Am. Coll. Cardiol. 42, 2014–2027. Wolfenson, H., Lavelin, I., Geiger, B., 2013. Dynamic regulation of the structure and functions of integrin adhesions. Dev. Cell 24, 447–458. Wong, M.L., Dong, C., Andreev, V., Arcos-Burgos, M., Licinio, J., 2012. Prediction of susceptibility to major depression by a model of interactions of multiple functional genetic variants and environmental factors. Mol. Psychiatry 17, 624– 633. Xia, H., Winokur, S.T., Kuo, W.L., Altherr, M.R., Bredt, D.S., 1997. Actinin-associated LIM protein: identification of a domain interaction between PDZ and spectrinlike repeat motifs. J. Cell Biol. 139, 507–515. Young, P.E., Richman, A.M., Ketchum, A.S., Kiehart, D.P., 1993. Morphogenesis in Drosophila requires nonmuscle myosin heavy chain function. Genes Dev 7, 29– 41. Zain, M.A., Roffeei, S.N., Zainal, N.Z., Kanagasundram, S., Mohamed, Z., 2013. Nonsynonymous polymorphisms of the PDLIM5 gene association with the occurrence of both bipolar disorder and schizophrenia. Psychiatr. Genet. 23, 258–261. Zervas, C.G., Gregory, S.L., Brown, N.H., 2001. Drosophila integrin-linked kinase is required at sites of integrin adhesion to link the cytoskeleton to the plasma membrane. J. Cell Biol. 152, 1007–1018. Zheng, M., Cheng, H., Banerjee, I., Chen, J., 2010. ALP/Enigma PDZ-LIM domain proteins in the heart. J Mol Cell Biol 2, 96–102. Zheng, M., Cheng, H., Li, X., Zhang, J., Cui, L., Ouyang, K., Han, L., Zhao, T., Gu, Y., Dalton, N.D., Bang, M.L., Peterson, K.L., Chen, J., 2009. Cardiac-specific ablation of Cypher leads to a severe form of dilated cardiomyopathy with premature death. Hum. Mol. Genet. 18, 701–713. Zhou, Q., Chu, P.H., Huang, C., Cheng, C.F., Martone, M.E., Knoll, G., Shelton, G.D., Evans, S., Chen, J., 2001. Ablation of Cypher, a PDZ-LIM domain Z-line protein, causes a severe form of congenital myopathy. J. Cell Biol. 155, 605– 612. Zhou, Q., Ruiz-Lozano, P., Martone, M.E., Chen, J., 1999. Cypher, a striated musclerestricted PDZ and LIM domain-containing protein, binds to alpha-actinin-2 and protein kinase C. J. Biol. Chem. 274, 19807–19813.