Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 May 1.
Published in final edited form as: Dev Biol. 2017 Mar 15;425(1):70–84. doi: 10.1016/j.ydbio.2017.03.012

Paxillin Genes and Actomyosin Contractility Regulate Myotome Morphogenesis in Zebrafish

Andrew E Jacob 1, Jeffrey D Amack 1,*, Christopher E Turner 1,*
PMCID: PMC5495106  NIHMSID: NIHMS861477  PMID: 28315297

Abstract

Paxillin (Pxn) is a key adapter protein and signaling regulator at sites of cell-extracellular matrix (ECM) adhesion. Here, we investigated the role of Pxn during vertebrate development using the zebrafish embryo as a model system. We have characterized two Pxn genes, pxna and pxnb, in zebrafish that are maternally supplied and expressed in multiple tissues. Gene editing and antisense gene knockdown approaches were used to uncover Pxn functions during zebrafish development. While mutation of either pxna or pxnb alone did not cause gross embryonic phenotypes, double mutants lacking maternally supplied pxna or pxnb displayed defects in cardiovascular, axial, and skeletal muscle development. Transient knockdown of Pxn proteins resulted in similar defects. Irregular myotome shape and ECM composition were observed, suggesting an “inside-out” signaling role for Paxillin genes in the development of myotendinous junctions. Inhibiting non-muscle Myosin-II during somitogenesis altered the subcellular localization of Pxn protein and phenocopied pxn gene loss-of-function. This indicates that Paxillin genes are effectors of actomyosin contractility-driven morphogenesis of trunk musculature in zebrafish. Together, these results reveal new functions for Pxn during muscle development and provide novel genetic models to elucidate Pxn functions.

Keywords: Extracellular matrix, Genome editing, Myotendinous junction, Somite, Cell adhesion

Graphical abstract

graphic file with name nihms861477u1.jpg

Introduction

The extracellular environment plays a crucial role in the form and function of many tissues and organ systems. In particular, adhesion to the extracellular matrix (ECM), a complex network of many glycoproteins, is required for proper cell shape, size, survival, and differentiation (Frantz et al., 2010; Humphrey et al., 2014). One way by which cells adhere to the ECM is through protein complexes involving transmembrane Integrin receptors (Campbell and Humphries, 2011). Integrin binding to ECM recruits a plethora of intracellular proteins that act as hubs for signaling cascades and anchors for the actin cytoskeleton (Wozniak et al., 2004). The mechanisms by which these proteins act in order to influence cellular activity of diverse cell types in the contexts of embryonic development and disease are under intense investigation (Bokel and Brown, 2002; Huveneers et al., 2007; Ley et al., 2016).

The Integrin adhesome consists of over 150 proteins that serve to regulate nearly all aspects of intracellular activity (Winograd-Katz et al., 2014; Zaidel-Bar and Geiger, 2010). Paxillin is an adapter protein containing LIM domains and LD motifs that is one of the core members of these complexes (Deakin and Turner, 2008). Roles for Paxillin in regulating adhesion size, assembly, and cross-talk with growth factor signaling have been studied using a wide array of in vitro and in vivo model systems (Brown and Turner, 2004). In the context of embryonic development, invertebrate Paxillin proteins influence cytoskeletal function in skeletal muscle (Bataille et al., 2010; Warner et al., 2011; Yagi et al., 2001). In frog and zebrafish embryos, Paxillin proteins localize to sites of myofiber adhesion, herein called myotendinous junctions (MTJs) (Crawford et al., 2003; Turner et al., 1991), suggesting a role in these structures. In addition, Paxillin has been implicated in contractility of cardiac and smooth muscle (Hirth et al., 2016; Zhang et al., 2016). However, the functions of Paxillin during vertebrate embryogenesis remain poorly understood. This is due in part because Paxillin null mouse embryos are not viable. These mutants typically die due to poor vascularization and impaired morphogenesis of mesoderm-derived tissues by E9.5, precluding them from being used to investigate the functions of Paxillin during development of skeletal muscle and other organ systems (Hagel et al., 2002).

Zebrafish provide an alternative vertebrate model to study genetic Paxillin loss-of-function during embryonic development. Unlike other vertebrate clades, zebrafish and other Teleost fish possess two Paxillin genes, pxna and pxnb, as the result of a whole-genome duplication event prior to radiation of this lineage. Both Pxna and Pxnb are highly similar to the single human PXN protein. In addition, we recently described a splice isoform of pxnb, which we refer to as pxnb-ins, that has an insertion of three additional coding exons and may encode an ancestral version of the protein (Jacob et al., 2016). Characterization of the spatiotemporal expression patterns of pxna and pxnb during early zebrafish embryo development revealed that these transcripts are maternally supplied and have both unique and overlapping expression profiles (Jacob et al., 2016). For example, pxna mRNA transcription is restricted to the notochord, Kupffer’s vesicle, and developing somites, whereas pxnb mRNA is present more broadly at post-gastrulation stages. However, both pxna and pxnb mRNAs are expressed in myotomes by 24 hours post fertilization (hpf) and continuing through 48 hpf. Interestingly, pxnb mRNA is enriched in the developing heart at 48 hpf, while pxna mRNA was not detected by in situ hybridization in this organ. Conserved expression patterns for Paxillin genes between zebrafish and mammalian embryos in the notochord and somites suggest that these genes may regulate similar processes in vertebrate embryos (Crawford et al., 2003; Hagel et al., 2002; Jacob et al., 2016). The zebrafish also provides a model wherein unique functions of Paxillin gene paralogs between tissues can be explored.

Herein, we used CRISPR-Cas9 and antisense morpholino technologies to investigate functions of the zebrafish Paxillin orthologs pxna and pxnb during embryonic development. Loss of both maternal and zygotic expression of Paxillin proteins caused morphogenetic defects in several tissues, including the heart, notochord, and skeletal muscle. Since expression studies have found Paxillin to be enriched at MTJs (Crawford et al., 2003; Jacob et al., 2016), we focused on functions of Paxillin in myotome development. Axial myotomes in vertebrates develop from segmented precursor structures called somites. New somite segregation from unsegmented mesoderm is dependent on transmembrane signaling through the Eph/Ephrin pathway (Julich et al., 2009). Stability of the boundary between somites is then dependent on Integrin activation and ECM polymerization (Julich et al., 2015). Signaling downstream of Integrins initiates non-muscle Myosin-II (herein referred to as NMM-II) activity and actomyosin contractility, but functions for effector proteins involved in this activation and boundary stability are largely unknown. We have found that Paxillin proteins function downstream of contractility to regulate the composition of the ECM at somite boundaries. This function is critical for establishing and maintaining myotome shape and integrity. Our results suggest that Paxillin is a key regulator of mechanical force-dependent ECM remodeling in skeletal muscle and potentially several other tissues during vertebrate embryogenesis.

Materials and Methods

Zebrafish

The TAB strain of wild-type zebrafish was used in this study. In addition, transgenic Tg(actb2:myl12.1-mKate2)sny125 zebrafish that ubiquitously express a non-muscle Myosin-II Light Chain 12.1-mKate2 (herein called Myl12.1-mKate2) fusion protein (manuscript in preparation) were crossed with Tg(actb2:pxn-EGFP) fish that ubiquitously express a functional Pxna-EGFP fusion protein (Goody et al., 2010) to generate offspring expressing both transgenes. Embryos were raised at 28°C and staged according to (Kimmel et al., 1995). All animal experiments were approved by SUNY Upstate Medical University’s IACUC.

Live Imaging

At 16–18 hpf, embryos were anesthetized with Tricaine (Sigma Aldrich), positioned on in a petri dish with a glass bottom (MatTek), and immobilized with low melting agarose (Fisher Scientific). Imaging was performed using a Perkin Elmer spinning disc confocal microscope with a 40× objective lens and a Hamamatsu C9100-50 camera. A Z-series of 1 μm optical slices was taken every 6 minutes for ~1 hour.

Microinjections

Pulled glass needles were used to deliver 1nL of injection mixture into one-cell stage TAB zebrafish embryos. Morpholino oligonucleotides (MO) were obtained from GeneTools, LLC. Standard negative control MO (CCTCTTACCTCAGTTACAATTTATA) and pxna/b ATG-MO (GAAGAGCATCTAAATCGTCCATGTT) were used at 0.3 mM concentration. Control or pxna/b ATG-MO were co-injected with 0.3 mM p53 MO (Robu et al., 2007). 400 pg of GFP-Pxna mRNA was co-injected with the pxna/b ATG-MO and p53 MO mix for rescue experiments. mRNA for injection was synthesized in vitro using Ambion mMessage mMachine SP6 Polymerase kits and linearized pCS2+ plasmids as templates.

CRISPR-Cas9 Mutagenesis

Single guide RNA (sgRNA) design for pxna (NCBI RefSeq: NM_201588.1) and pxnb (Ensembl ID: ENSDARG00000060766) was performed using the online CHOPCHOP site (Montague et al., 2014). All sgRNAs were synthesized according to published protocols (Gagnon et al., 2014; Hwang et al., 2013). Cas9 pCS2+ plasmid (Gagnon et al., 2014) was used as a template for mRNA synthesis (Ambion, mMessage mMachine SP6 Polymerase kit). 250 pg of Cas9 mRNA + 10–50 pg sgRNA was injected into one-cell stage wild-type embryos. Single embryos were genotyped using PCR and restriction enzyme (New England Biolabs) digests (XhoI for pxna and BssSαI for pxnb) to verify the efficacy of injected sgRNAs. Injected embryos were raised to adulthood and crossed with wild-type fish to identify F0 germline founders and to generate F1 heterozygous embryos. Two mutant alleles for each genomic locus were used to establish stable single homozygous mutant lines (pxnasny6400, pxnasny6500, pxnbsny7100, pxnbsny7300). To generate a double mutant line, we first crossed pxnasny6500 and pxnbsny7100 homozygous adults to obtain double heterozygous animals. Adult double heterozygous fish were then crossed and their offspring were raised. Adult genotypes recovered from these crosses are summarized in Table 1.

Table 1.

Survival of zebrafish carrying pxn mutations to adulthood.

Observed Expected Mendelian Ratios for Het/Het Cross
pxna+/+; pxnb+/+ 16 8.625 0.06
pxna+/+;pxnbsny7100/+ 23 17.25 0.13
pxna+/+;pxnbsny7100/sny7100 10 8.625 0.06
pxnasny6500/+;pxnb+/+ 20 17.25 0.13
pxnasny6500/+;pxnbsny7100/+ 36 34.5 0.25
pxnasny6500/+;pxnbsny7100/sny7100 12 17.25 0.13
pxnasny6500/sny6500;pxnb+/+ 5 8.625 0.06
pxnasny6500/sny6500;pxnbsny7100/+ 16 17.25 0.13
pxnasny6500/sny6500;pxnbsny7100/sny7100 0 8.625 0.06
Total 138 138 1.00

Progeny from double heterozygous (pxnasny6500/+;pxnbsny7100/+) parents were raised to adulthood (4 months old) and then genotyped.

  • pxna sgRNA target: GGAGGGGAGATTACTCGAGGCGG

  • pxna forward primer: ACCTAACAGATGCTCTTCTCGC

  • pxna reverse primer: TCATCATTACTTCCTGAAGGGT

  • pxnb sgRNA target: GAGCTCCTCAAGTTCTCGTGTGG

  • pxnb forward primer: TTTATGTAAACAGACCCAGCCC

  • pxnb reverse primer: CCAAGACACAACGTCTCTTCAG

Western Blotting

Embryos were deyolked (Link et al., 2006), homogenized in 2× SDS-PAGE sample buffer, and boiled prior to running SDS-PAGE. Antibodies used were anti-Paxillin (clone 349) (BD Transduction) at 1:1000, anti-Actin (Millipore) at 1:1000, and anti-GFP (Santa Cruz) at 1:1000. Species-specific secondary antibodies conjugated to HRP (BioRad) were incubated at 1:10,000.

Immunofluorescence

Embryos were fixed and permeabilized using 4% paraformaldehyde (VWR) and 0.5% Triton-X100 (Sigma Aldrich) in phosphate buffered saline (PBS, Invitrogen) overnight at 4°C. After washing embryos with PBS+0.1% Tween-20 (Fisher Scientific), embryos were blocked for 4 hours at room temperature with 5% goat serum (or 5% BSA for Fibronectin staining) in PBS+0.1% Tween-20. Primary and secondary antibodies were diluted in block at 1:100 and incubated overnight while rocking at 4°C sequentially, with extensive washing in PBS+0.1% Tween-20 at room temperature after each incubation. Rhodamine-conjugated phalloidin (Life Tech) and DAPI (Life Tech) were incubated with secondary antibodies at 1:100 and 1:500 respectively. Embryos were then stored in 80% glycerol (Fisher Scientific) and mounted on glass slides for imaging. To analyze myotome shape, three MTJs above the posterior yolk extension (somites 17–19) were used for quantification in each embryo. Post-acquisition measurements and projections were made with Fiji (Schindelin et al., 2012) or Imaris (Bitplane) software.

Primary antibodies: Rabbit anti-Laminin-111 (Sigma, L9393), Mouse anti-Vinculin (Sigma, hVIN1), Rabbit anti-Fibronectin (Sigma, F3648), Mouse anti-Myosin Heavy Chain (DSHB, MF20), Mouse anti-ZO-1 (Invitrogen, 33–9100), Mouse anti-α-Actinin (Sigma, A7811)

Secondary antibodies: anti-Rabbit Alexa Fluor 488 (Life Tech), anti-Mouse Alexa Fluor 405 (Life Tech), anti-Rabbit Alexa Flour 568 (Life Tech), anti-Mouse Alexa Fluor 488 (Life Tech), anti-Mouse Alexa Fluor 561 (Life Tech)

Whole Mount mRNA in situ Hybridization

In situ visualization of cmlc2 mRNA at 30 hpf was performed as previously described (Gao et al., 2010). For genotyping, genomic DNA was extracted from individual embryos after staining using 50 mM NaOH and incubation at 95°C, then neutralized with 1M Tris HCl pH8.0 before use as a template for standard PCR (Lucigen, EconoTaq Green 2X) and restriction digests.

Drug Treatment

Blebbistatin (Sigma, 203390) was dissolved in 90% DMSO (Fisher Scientific) to generate a 50 mM stock solution. A final concentration of 50 μM (diluted in water) was used for treatments and 0.1% DMSO was used as a control treatment. Embryos at ~17 hpf were manually dechorionated and treated in a 24 well plate (USA Scientific, CC7682-7524). Control and blebbistatin treatments were performed at 28°C in the dark for ~4 hours. Embryos were fixed immediately or transferred to 100 mm petri dishes, washed twice, and allowed to grow post-treatment for 6 hours. Three to four MTJs in three fields of view for each embryo were examined for quantifications.

Statistics

Statistical analyses were performed using GraphPad Prism 6. Normality for each dataset was determined by D’Agostino’s and Pearson’s omnibus test. Significance was then determined by one-way ANOVA or Kruskal-Wallis test with Fisher’s LSD or Dunn’s post-hoc tests to compare each pxn mutant genotype with wild-type embryos. For individual comparisons, Student’s unpaired two-tailed t-test was used to determine significance and Welch’s correction was applied when appropriate.

Results

Generation of Paxillin Mutant Zebrafish

Disruption of the pxna or pxnb gene was carried out using the CRISPR/Cas9 endonuclease platform (Gagnon et al., 2014; Hwang et al., 2013). Microinjection of single guide RNAs (sgRNAs) targeting either pxna exon 2 or pxnb exon 6 (Fig. 1A) along with Cas9 mRNA into one-cell staged embryos produced multiple mutant alleles. Two germline founder alleles each for pxna and pxnb were used to generate stable single mutant lines. The pxnasny6400 and pxnasny6500 alleles harbor 22 base pair (bp) and 1 bp deletions respectively, and pxnbsny7100 has a 2 bp deletion with an A to C transition and pxnbsny7300 has a 2 bp deletion along with a CAC to ATT transition (Fig. 1B). Each of these mutations altered a restriction endonuclease site (Fig. 1B), which provides a simple and rapid method for genotyping (Fig. 1C). All mutant alleles for pxna and pxnb generate premature stop codons within the same exon to which the sgRNA was targeted and would be predicted to truncate Pxna and Pxnb proteins (Fig. 1D).

Fig. 1. CRISPR/Cas9 Targeting of Zebrafish pxna and pxnb Genes.

Fig. 1

(A) pxna and pxnb gene cDNA diagrams depicting exons coding for LD motifs, LIM domains, and the pxnb-ins insert region (Exon 7–9). sgRNA target exons are identified by arrowheads. (B) Wild-type (WT) pxna and pxnb DNA sequences aligned with the mutant alleles recovered as stable mutant lines. Nucleotides conserved between all sequences are colored yellow, while those conserved between two sequences are colored blue. Restriction enzyme sites used for genotyping are labeled. (C) Representative genotyping results show possible outcomes after PCR and restriction enzyme digest. Homozygous mutant PCR products are resistant to enzyme cleavage. (D) Predicted sequence alignments of WT pxna, WT pxnb, and mutant proteins. Amino acid numbers are indicated and * marks termination of translation of the mutant proteins.

Somewhat surprisingly, embryos that were homozygous for pxna or pxnb mutations (identified by genotyping) did not exhibit gross morphological defects through 4 days post-fertilization (dpf) (Fig. 2A–C) and survived to adulthood (Table 1). We next used adult homozygous single mutants to generate embryos that lacked both maternal and zygotic (MZ) pxna or pxnb expression. Both MZpxna and MZpxnb mutants were also morphologically normal at 4 dpf (Fig. 2D, E) and viable through adulthood. Since compensation of function between closely related genes in zebrafish has been previously observed (Rossi et al., 2015), we next generated double mutants to test the hypothesis that pxna and pxnb genes had compensatory roles in the single mutants. We found that double homozygous pxna;pxnb mutants were morphologically normal through the first 3 days of embryonic development but developed pericardial edemas by 4 days old, and none survived to adulthood (Fig. 2F, Table 1). These results suggested that either Paxilln proteins were dispensable for early zebrafish embryogenesis or that maternal Paxillin expression in the double mutants was sufficient to support normal development for at least 3 days.

Fig. 2. Double Paxillin Mutant Zebrafish Embryos Develop Gross Morphological Defects.

Fig. 2

(A–H) Gross morphology of wild-type (A), homozygous zygotic single pxn mutant (B, C), maternal and zygotic (MZ) single pxn mutant (D, E), zygotic double pxn mutant (F) and MZ double pxn mutant (G, H) embryos at 4 dpf. Defects in double mutants included pericardial edema (arrows in F, G, H), and kinked tail (arrowhead in G). (I) Western blotting of lysates from 4 dpf embryos revealed that double pxn mutants lacked full-length Pxna and Pxnb proteins, including a high molecular weight band possibly corresponding to the Pxnb-ins isoform.

To test the role of maternally supplied Paxillin proteins, we took advantage of viable zebrafish that were homozygous mutant for one pxn gene and heterozygous for the other. By crossing these fish we generated embryos that were homozygous zygotic mutant (Z) for both pxna and pxnb and lacked the maternal supply (M) of either pxna or pxnb (Table 2). We denote these embryos as either MZpxna;Zpxnb or Zpxna;MZpxnb and will refer to these embryos collectively as ‘MZ double mutants.’ Analysis of MZ double mutant embryos revealed gross developmental defects in multiple tissues (Fig. 2G, H). Genotyping verified that embryos with abnormal phenotypes were MZ double homozygous mutants. At 4 dpf, MZpxna;Zpxnb embryos had pericardial and cranial edema, shortened anterior-posterior length, and a posterior tail kink as compared to their morphologically normal MZpxna siblings (Fig. 2D, G). Similarly, Zpxna;MZpxnb embryos, at the same stage, had cardiac and cranial edemas. In contrast however, anterior-posterior length and tail morphology were similar between MZpxnb and Zpxna;MZpxnb siblings (Fig. 2E, H). These observations of gross morphology suggested that maternally supplied pxna plays a unique role in axial development and that maternally supplied pxnb alone cannot compensate for pxna loss.

Table 2.

Breeding strategy for generating pxna;pxnb double mutant embryos lacking maternal pxna or pxnb.

pxnasny6500/sny6500;pxnbsny7100/+ × pxnasny6500/sny6500;pxnbsny7100/+
Expected Offspring Genotype Ratios Designation
Herein
100% Homozgous Mutant pxna pxna
50% Heterozygous pxnb MZpxna
25% Homozygous Mutant pxnb MZpxna;Zpxnb
25% Homozygous Wild-Type pxnb MZpxna
pxnasny6500/+;pxnbsny7100/sny7100 × pxnasny6500/+;pxnbsny7100/sny7100
Expected Offspring Genotype Ratios Designation
Herein
100% Homozygous Mutant pxnb pxnb
50% Heterozygous pxna MZpxnb
25% Homozygous Mutant pxna Zpxna;MZpxnb
25% Homozgyous Wild-Type pxna MZpxnb

To determine the impact of the pxn mutations on Paxillin protein expression we performed Western blots using an antibody that detects multiple Paxillin family members, including zebrafish pxna and pxnb proteins (Jacob et al., 2016). Using lysates from 4 dpf embryos, a 60 kDa band corresponding to the predicted size of both Pxna and Pxnb was present in single mutant embryos but absent in MZ double mutants (Fig. 2I). This verified that both Pxna and Pxnb migrate at ~60 kDa. A diffuse 150 kDa band was detected in single MZpxna mutants, but not in homozygous mutant pxnb embryos (Fig. 2I), suggesting this band corresponds to the Pxnb-ins isoform (see Fig. 1A) that has a predicted weight of ~130 kDa (Jacob et al., 2016). These results indicate that MZ double mutants do not produce full-length Pxna or Pxnb proteins and that these recessive mutations are likely null alleles.

Developmental Defects in Multiple Tissues of Paxillin MZ Double Mutants

To better understand the developmental defects observed in MZ double mutants, we next analyzed embryos at earlier developmental stages. Firstly, all MZpxna;Zpxnb and Zpxna;MZpxnb embryos appeared morphologically normal through 30 hpf. By 48 hpf, however, multiple phenotypes were consistently observed in MZ double mutants. At this stage, MZpxna;Zpxnb embryos showed cranial swelling that was often accompanied by brain hemorrhaging, blood pooling, and pericardial edema (Fig. 3A). The Zpxna;MZpxnb embryos had milder phenotypes at 48 hpf, as these mutants did not show pericardial edema this early in development, although brain hemorrhaging and mild cranial hyperinflation was observed (Fig. 3A). Blood flow through the axial tail vasculature was impaired or absent in all MZ double mutants as compared to strong circulation in single mutant siblings (Movie 1, 2). All MZ double mutants also developed pericardial edema, but at different times during development. This defect was present in MZpxna;Zpxnb mutants at 48 hpf, whereas it was evident after 3 dpf in Zpxna;MZpxnb embryos (Fig. 3A).

Fig. 3. Maternal Paxillin Expression Differentially Regulates Cardiovascular and Notochord Development.

Fig. 3

(A) Gross morphological examination of the head and pericardial regions revealed that pxn double mutant embryos lacking maternal pxna expression (MZpxna;Zpxnb) exhibited swelling near the brain (arrowhead) and heart (arrow) at an earlier time point than those lacking maternal pxnb (Zpxna;MZpxnb). The number of MZ double mutants exhibiting the depicted phenotype over the total number of double mutants analyzed for each stage is shown. (B) 3D surface projections of Myosin immunostaining of the developing heart at 48 hpf or 3 dpf in MZpxna;Zpxnb and Zpxna;MZpxnb embryos. Disrupted cardiac morphology was observed in both pxn MZ double mutant genotypes. The number of hearts observed with the depicted phenotypes over the total number of embryos analyzed for each genotype is shown. Scale bars = 50 μm. (C) At 48 hpf, outer notochord sheath cells in the tail of the embryo (boxed region of embryo diagram) normally exhibit punctate ZO-1 distribution at cell-cell junctions in sagittal optical sections (arrows) as observed in MZpxna, MZpxnb and Zpxna:MZpxnb embryos. A portion of MZpxna;Zpxnb embryos exhibited sheath cell rounding and extrusion from the monolayer with disrupted ZO-1 distributions (arrowhead). The number of notochords observed with the depicted phenotypes over the total number of embryos analyzed for each genotype is shown. Scale bar = 10 μm.

To further investigate the circulation defects observed in MZ double mutants, we next analyzed heart development. MZ double mutant embryos displayed a range of defects in heart morphology. At 48 hpf, MZpxna;Zpxnb embryos had very thin, poorly ballooned heart chambers and incomplete looping (Fig. 3B). In contrast, Zpxna;MZpxnb mutant hearts showed defects at later stages (3 dpf) that typically included a smaller ventricle and incomplete looping as compared with normal siblings (Fig. 3B). These phenotypes are consistent with the Paxillin knockout mouse embryonic phenotype wherein mutant embryos develop small hearts (Hagel et al., 2002). To determine whether defective heart looping was due to altered left-right patterning of the heart, we analyzed the first asymmetric step of zebrafish heart morphogenesis (called ‘heart jogging’) that occurs at 24 hpf (Chen et al., 1997). During this phase of morphogenesis, heart tube morphology and asymmetric heart jogging were normal in both MZ double mutant embryos and MZ single mutant siblings (Fig. S1A), indicating that cardiac left-right patterning was not disrupted. Since zebrafish mutants for the adhesion protein Talin1 have disrupted cardiomyocyte sarcomere maintenance (Wu et al., 2015), we next tested whether loss of Paxillin had similar effects on cardiomyocyte organization. In contrast to Talin1 mutants, both MZpxna;Zpxnb and Zpxna;MZpxnb mutant hearts were contractile and exhibited normal sarcomeric α-Actinin distribution in cardiomyocytes at the stages when pericardial edemas have developed (Fig. S1B). These observations demonstrate that while early cardiac patterning and cardiomyocyte development proceeds normally, morphogenetic defects arise after the onset of heart beating in Paxillin MZ double mutants. Additionally, although MZ double mutant embryos of both genotypes exhibit cardiac defects, these results suggest that maternally supplied pxna and pxnb may play different roles in morphogenesis of the heart.

A phenotype that was specific for MZpxna;Zpxnb mutants was a shortened anterior-posterior axis length. The severity of this phenotype was variable, but mutant embryos with more pronounced axis shortening also had a sharp upward kink in the posterior region of the notochord. (Fig. 2G, arrowhead). To assess this phenotype in greater detail we examined the distribution of the cell-cell junction protein ZO-1 in posterior notochord sheath cells of 48 hpf embryos. Sheath cells in the posterior notochord of normal MZpxna embryos had ZO-1 distributed at cell-cell junctions (Fig. 3C, arrows), whereas these cells in many MZpxna;Zpxnb embryos had a more diffuse, cytoplasmic distribution of ZO-1 around the site of the kink (Fig. 3C). Rounded cells often appeared to be extruded from the sheath cell layer adjacent to the site of the upward kink (Fig. 3C, arrowhead). These defects were not observed in Zpxna;MZpxnb mutants that did not have axis length or tail kink defects (Fig. 3C). This analysis uncovers a role for Paxillin in notochord and axial development that is likely mediated by establishing or maintaining the integrity of cell adhesions.

In addition to axis elongation and notochord defects, development of trunk musculature was clearly perturbed in MZpxna;Zpxnb mutants at 4 dpf (Fig. 4A). This phenotype was reminiscent of previous work that has identified roles for ECM and adhesion proteins in zebrafish axial musculature development and function (Sztal et al., 2012). By 5 dpf, MZ double mutant embryos began to display patches of degeneration along the axial musculature and showed a striking amount of swelling in multiple tissues including the eye and gut, while their MZpxna or MZpxnb siblings remained healthy. Ultimately, necrosis in MZ double mutants became evident and the embryos died by 7 dpf. Altogether these data suggest that expression of one pxn gene in zebrafish is sufficient for viability and that maternally supplied pxna plays a unique role in notochord and skeletal muscle morphogenesis.

Fig. 4. Paxillin Mutants Exhibit Defects in Myotome Development.

Fig. 4

(A) Vinculin and Actin were visualized in 4 dpf embryos to show MTJs and myofibers respectively. Chevron angle (B) was measured for three consecutive MTJs and averaged for each embryo. MZpxna;Zpxnb embryos exhibited over-elongated myofibers passing through MTJs (white arrow) in some of their myotomes. Myofiber detachment and retraction (yellow arrowhead) was also observed. Zpxna;MZpxnb embryos had more mild MTJ breaching defects at this stage. Scale bar = 10 μm (B) MTJ chevron angles and disrupted MTJs were quantified at 4 dpf in wild-type embryos, Paxillin MZ single mutants, Paxillin MZ double mutants and wild-type embryos injected with either control MO or pxna/b ATG-MO. Ectopic expression of GFP-Pxna fusion protein partially rescued myotome defects in pxna/b ATG-MO injected embryos. Data points represent individual embryo means and error bars show standard deviations from three independent experiments. **** p < 0.001 as determined by ANOVA/Fisher’s LSD post-hoc test, # # p < 0.01 as determined by T-test, ns = not significant.

Maternal pxna plays a Key Role in Myotome Morphogenesis

The Pxna protein, along with several other Integrin adhesion-associated proteins, localize to specialized sites of adhesion called myotendinous junctions (MTJs) that form between myotomes during skeletal muscle development (Costa et al., 2008; Crawford et al., 2003; Postel et al., 2008). At 24 hpf, elongated myoblasts adhere to their basement membrane along the chevron-shaped MTJs (Snow et al., 2008). However, how Pxn proteins function at these junctions and the mechanisms that regulate Pxn localization to MTJs has remained unclear. In situ RNA hybridizations revealed that both pxna and pxnb expression is highly enriched in developing myotomes at 24 hpf and 48 hpf (Jacob et al., 2016). Thus, we focused on the role of Paxillin proteins in muscle development by analyzing Paxillin MZ double mutant embryos. Immunostaining for Vinculin and phalloidin staining for Actin filaments (F-Actin) was used to visualize myotomes in wild-type and Paxillin MZ double mutants that were easily distinguishable from single mutant siblings at 4 dpf (Fig. 4A). We found that both MZ single and double mutant embryos exhibited abnormally wide MTJ chevron angles at this stage compared with stage-matched wild-type embryos (Fig. 4B). Additionally, MZpxna;Zpxnb embryos exhibited muscle cell hyper-elongation through MTJs into adjacent myotome segments as well as muscle cell detachment from the MTJ (Fig 4A, arrow and arrowhead). These phenotypes observed at 4 dpf were suggestive of defects in earlier myotome morphogenesis (Goody et al., 2012). Examining myotomes at 48 hpf or 3 dpf, when Paxillin MZ double mutants could be phenotypically distinguished, revealed an earlier onset of the wider MTJ chevron angle and MTJ breaching phenotypes (Fig. S2A, B). Furthermore, large gaps between elongated myoblasts were frequently observed in MZpxna;Zpxnb embryos at 48 hpf (Fig. S2C, arrow). These MTJ angle and breaching phenotypes have been previously described in embryos with impaired Laminin organization and ECM adhesion at the MTJ (Goody et al., 2010; Kim and Ingham, 2009). Interestingly, the MTJ chevron angles between MZpxnb and Zpxna;MZpxnb embryos did not differ, while MZpxna;Zpxnb embryos had much wider MTJ chevron angles than MZpxna siblings at 4 dpf and earlier stages (Fig. 4B, Fig. S2B). These results indicate that maternally supplied pxna plays a major role in the regulation of myoblast elongation, adhesion, and MTJ development, which cannot be compensated for by maternally supplied pxnb alone.

To verify that the observed effects on myotome development were due to loss of pxn function, we used an antisense morpholino oligonucleotide (MO) that is complementary to the highly similar start codon regions of both pxna and pxnb (Fig. S3A). We predicted this ‘pxna/b ATG-MO’ would reduce maternal and zygotic expression of both genes. Indeed, injection of pxna/b ATG-MO into one-cell stage wild-type embryos resulted in robust knockdown of endogenous Pxn proteins at 3 dpf as revealed by Western blotting (Fig. S3B). Embryos injected with pxna/b ATG-MO developed myotome phenotypes that were similar to those observed in MZpxna;Zpxnb double mutants. At 3 dpf, pxna/b ATG-MO injected embryos had numerous MTJ breaches and wider MTJ chevron angles as compared to control MO injected embryos (Fig. 4B, Fig. S3C). Expression of a GFP-Pxna fusion protein (Jacob et al., 2016) attenuated the severity of the myotome phenotypes caused by pxna/b ATG-MO injection (Fig. 4B, Fig. S3C). The partial rescue demonstrated by re-expression of a Paxillin protein demonstrated that the observed phenotype was not likely due to off-target effects of MO injection. Taken together, these results indicate that acute reduction of pxn gene expression affects embryonic myotome development and that the myotome phenotypes observed in the genetic mutant embryos are indeed due to disrupted expression of zebrafish pxn genes.

Paxillin regulates ECM composition at MTJs

Laminin, an extracellular matrix component, is robustly localized to zebrafish MTJs at 24 hpf, where it functions to halt myoblast elongation during the transition from a somite with an epithelial border to a functional chevron-shaped myotome (Snow et al., 2008). Abnormally wide MTJ chevron angles and myoblast hyper-elongation phenotypes, as seen in MZpxna;Zpxnb embryos, have been observed in zebrafish embryos deficient for Laminin γ1 (Snow et al., 2008), the Laminin-binding α6-Integrin receptor (Goody et al., 2012), and enzymes required for proper Laminin basement membrane organization (Goody et al., 2010). Interestingly, poorly organized Laminin at the MTJ due to knockdown of Nrk2b, a kinase which generates NAD+, can be rescued by Pxna overexpression (Goody et al., 2010). Thus, we hypothesized that Paxillin plays a role in depositing or organizing Laminin at the MTJ. To test this, we used immunostaining to visualize and quantify Laminin accumulation at MTJs labeled by Vinculin immunostaining (Fig. 5A). Similar levels of Laminin relative to Vinculin at the MTJ were observed in wild-type and single MZpxna or MZpxnb mutant embryos at 24 hpf (Fig. 5B). In contrast, MZpxna;Zpxnb mutants showed a significant reduction in Laminin at the MTJ (Fig. 5A, B). Similarly, immunostaining of pxna/b ATG-MO injected embryos revealed that the amount of Laminin was reduced at the MTJ at 24 hpf compared with control MO injected embryos (Fig. 5A, B). Laminin intensity at the MTJs of Zpxna;MZpxnb embryos was comparable with that of wild-type and Paxillin MZ single mutant embryos, which is consistent with milder myotome phenotypes (e.g. MTJ angle and breaching defects) in these embryos. These findings indicated that Paxillin function is required for Laminin enrichment at developing MTJs.

Fig. 5. Paxillin Regulates ECM Composition at the MTJ.

Fig. 5

(A) Immunostaining for Laminin and Vinculin at MTJs at 24 hpf in wild-type, MZpxna, MZpxna;Zpxnb, MZpxnb, Zpxna;MZpxnb and control MO or pxna/b ATG-MO injected wild-type embryos. (B) Mean fluorescence intensity (MFI) ratios between Laminin and Vinculin revealed reduced Laminin accumulation at the MTJs of MZpxna;Zpxnb embryos and pxna/b ATG-MO injected embryos at this stage. Data points represent individual embryo means and error bars show standard deviations from three independent experiments. * p < 0.05 as determined by Kruskal-Wallis/Dunn’s post-hoc test, # # p < 0.01 as determined by T-test, ns = not significant. (C) Immunostaining for Fibronectin at MTJs at 32 hpf. Ectopic retention of Fibronectin at MTJs in the majority of MZpxna;Zpxnb embryos was observed (white arrowheads). The number of embryos from three independent experiments with the depicted MTJ phenotypes over the total number of embryos examined for each genotype is noted. (D) Immunostaining for Fibronectin at MTJs at 48 hpf. Similar to MZpxna embryos, Fibronectin gets downregulated at the MTJs in a majority of MZpxna;Zpxnb embryos. Some MZpxna;Zpxnb embryos with severe muscle degeneration exhibit persistent ectopic Fibronectin (white arrowhead). The number of embryos exhibiting each phenotype over the total number of embryos examined from three independent experiments is noted. Boxed areas show regions of interest visualized for each stage. Scale bars = 10 μm

During early somitogenesis stages the epithelial somite boundary ECM consists of mostly Fibronectin with comparatively less Laminin (Jenkins et al., 2016; Snow and Henry, 2009). During the transition to a chevron-shaped MTJ, myoblasts elongate anterio-posteriorly and a downregulation of Fibronectin at the boundary occurs adjacent to fast-twitch myofibers by 32 hpf (Jenkins et al., 2016). To determine whether Paxillin is involved in this ECM transition, we used immunostaining to assess Fibronectin localization in the developing myotome after 24 hpf. Visualizing Fibronectin in MZ single and double mutant embryos at two stages revealed that this downregulation is delayed in MZpxna;Zpxnb embryos. At 32 hpf, the anterior trunk myotomes of wild-type embryos lacked Fibronectin immunostaining at MTJs (Fig. 5C). Similarly, the majority of MZpxna and MZpxnb embryos (82% and 89%, respectively) lacked Fibronectin, whereas it was ectopically retained in most MZpxna;Zpxnb embryos (67%) (Fig. 5C). Surprisingly, while some Zpxna;MZpxnb embryos exhibited ectopic Fibronectin staining (25%), the frequency of this phenotype was attenuated compared to the MZpxna;Zpxnb genotype. Interestingly, although a majority of MZpxna;Zpxnb embryos showed reduced Fibronectin accumulation at the MTJ by 48 hpf (76%), a small number of embryos which exhibited myofiber detachment at this stage had ectopic accumulation of Fibronectin in the affected myotomes (24%) (Fig. 5D). These results suggest that Paxillin proteins function to modulate the extracellular environment during skeletal muscle development.

Paxillin is Recruited to the Developing MTJ After NMM-II

Integrin adhesion complexes serve as a physical linkage between the ECM and intracellular actin cytoskeleton of cells within tissues. Embryonic somite boundaries and MTJs serve similar roles during morphogenesis of the myotome. Paxillin proteins and NMM-II motor proteins that drive contractility of the actomyosin cytoskeleton have been found to localize to these complexes (Crawford et al., 2003; Sanger et al., 2009). However, the precise spatiotemporal dynamics of these components in complex assembly have not been elucidated. To investigate Paxillin dynamics in live embryos, we generated double transgenic zebrafish embryos that ubiquitously express Pxna-GFP (Goody et al., 2010) and NMM-II Light Chain (Myl12.1)-mKate2 fusion proteins. The Pxna-GFP and Myl12.1-mKate2 fusion proteins localize to myotome boundaries, which is consistent with the reported localization of endogenous Pxn and NMM-II proteins (Crawford et al., 2003; Sanger et al., 2009). Using live imaging, we examined newly forming somite boundaries in the posterior trunk of an 18 somite-stage embryo and found that Myl12.1-mKate2 was present at the boundary between non-segmented presomitic mesoderm and the newly formed somite (Fig. 6). Pxna-GFP was not present initially at this boundary but then accumulated as the boundary matured (~30 min after the enrichment of Myl12.1-mKate2), prior to the formation of the next new boundary (Fig. 6, Movie 3). This spatiotemporal pattern was repeated as new somite boundaries formed (Fig. 6, Movie 3). Together these observations revealed that adhesion complexes undergo a progressive maturation at newly forming somite boundaries and that Myosin II precedes the accumulation of Paxillin at these forming boundaries.

Fig. 6. Localization of NMM-II to Somite Boundaries Precedes Pxna Recruitment.

Fig. 6

Montage of live imaging of somite boundary formation at 18 hpf in the tail (boxed region of embryo diagram) of double transgenic embryos that express Myl12.1-mKate2 and Pxna-GFP fusion proteins. Arrowheads depict forming somite boundaries in each channel.

Paxillin recruitment to MTJs and Myotome Morphogenesis Requires Actomyosin Cytoskeletal Contractility

The transition into a functional myotome consists of myoblast elongation and adherence to the anterior and posterior somite boundaries (Henry et al., 2005; Snow et al., 2008). This event is coupled with a change of the boundary to a chevron shape and remodeling of the extracellular matrix at the MTJ (Snow and Henry, 2009). Localization of Paxillin to sites of adhesion has been demonstrated to be mechanosensitive in cultured cells (Schiller et al., 2011), but this has not yet been investigated in vivo. We hypothesized that NMM-II generated actomyosin contraction of the cytoskeleton may be required for Paxillin localization and MTJ maturation in the zebrafish embryo.

To test the role of cytoskeletal contractility in developing myotomes, a pharmacological inhibitor of NMM-II activity, blebbistatin, was used to reduce actomyosin contractility at specific stages during somite boundary-to-MTJ morphogenesis that have been previously described (Henry et al., 2005). Wild-type embryos treated with blebbistatin from the 16 somite stage to the 25 somite stage were either fixed immediately for protein localization analyses or washed thoroughly and allowed to develop without the drug for 6 hours (Fig. 7A). We found that Myl12.1-mKate2 still localized to MTJs after blebbistatin treatments, indicating these junctions remained intact. In contrast, Pxna-GFP distribution was diffuse rather than concentrated at MTJs as observed in control embryos (Fig. 7B). After the embryos were washed and allowed to develop to 28 hpf, Pxna-GFP localization to the MTJs was restored (Fig. 7C). This indicated that mis-localization of Paxillin in blebbistatin treated embryos was reversible and that Paxillin localization to MTJs is regulated by actomyosin cytoskeletal contractility. We observed that posterior somites that formed during the blebbistatin treatment period had an increased MTJ chevron angle, as well as multiple MTJ breaches (Fig. 7C, D). Thus, reduced actomyosin contractility—and loss of Paxillin MTJ localization—phenocopied myotome morphogenesis defects observed in Pxn loss-of-function mutants (see Fig. 4). Interestingly, more anterior somites that had already undergone the transition into a myotome at the time of treatment did not exhibit wide MTJ chevron angles, but disruptions in MTJs were sometimes present (Fig. 7D). These results pinpoint the developmental timing of cytoskeletal contractility and Paxillin recruitment to somite boundaries that is required for normal myotome formation.

Fig. 7. NMM-II Activity Shapes the Developing Myotome.

Fig. 7

(A) Diagram depicting the timing of pharmacological treatments with blebbistatin to inhibit NMM-II and the immunofluorescence (IF) staining of MTJ markers. Regions visualized by IF are boxed. (B) Embryos treated with blebbistatin formed MTJs marked by Myl12.1-mKate2, but Pxna-GFP MTJ localization was reduced as compared to control embryos at the 25 somite stage (C) Embryos treated with blebbistatin that were washed and allowed to develop without the drug recovered Pxna-GFP localization to MTJs by 28 hpf. However, wider chevron angle and MTJ gaps (yellow arrow) were observed. (D) Width of MTJ angle and gaps in MTJs were quantified in blebbistatin treated and control embryos at 28 hpf in the anterior region of the trunk, the posterior region of the trunk and the tail (see boxed region of embryo diagram). Data points represent individual embryo means and horizontal bars show means for each complete dataset from three independent experiments. **** p < 0.001, ns = not significant. Scale bars = 25 μm.

Discussion

Herein, we report overlapping and distinct developmental roles for two Paxillin orthologues, Pxna and Pxnb, in zebrafish. We found that simultaneous antisense knockdown or gene editing of both genes disrupted development of several tissues, including heart, notochord, and skeletal muscle. By focusing on the functions of Paxillin during muscle development, we elucidated a mechanism involving actomyosin contractility that recruits Paxillin to MTJs where it regulates the ECM environment that is critical for myotome morphogenesis. In addition to shedding new light on myotome development, the zebrafish pxn mutants reported here provide new animal models to further investigate the functions of Paxillin in adhesion signaling during embryonic morphogenesis.

Duplicated Zebrafish Paxillin Genes Have Unique Maternal Effects on Early Development

Early embryonic developmental defects were only observed when zygotic mutations in both pxna and pxnb were combined with loss of maternal expression of either gene. Quantification of MTJ chevron angles revealed subtle perturbations of myotome development in Paxillin MZ single mutants, but these alterations apparently do not have a significant impact on health since these animals become fertile adults. In contrast, Paxillin MZ double mutants had more severe developmental defects and did not survive beyond 7 dpf. Interestingly, Paxillin MZ double mutants had distinct phenotypic features (Table 3), highlighting a subset of differential functions for maternally-supplied zebrafish Paxillin proteins. For example, MZpxna;Zpxnb embryos developed a posterior tail kink and also exhibited more severe myotome defects compared with Zpxna;MZpxnb embryos. These observations provide evidence that maternally deposited pxna has a key function in the notochord and during early myotome morphogenesis which cannot be compensated for by maternally deposited pxnb. Expression of pxna transcript during somitogenesis is restricted to the notochord and somites, reflecting the phenotypes we observed (Jacob et al., 2016). Although we have detected maternal deposition of both pxna and pxnb mRNAs, antibodies that are specific for Pxna or Pxnb would be required to visualize protein expression during early development. Based on our genetic analysis, we predict Pxna is the key Paxillin protein that functions during early zebrafish embryogenesis.

Table 3.

Phenotypes Observed in MZ Double Mutant Embryos

MZpxna;Zpxnb Zpxna;MZpxnb
Pericardial Edema +++ ++
Poor Blood Flow ++ +++
Hemorrhage and Swelling in Brain +++ ++
Cardiac Morphogenesis Impaired +++ ++
Posterior Notochord Kink +++
Anterior-Posterior Axis Elongation +++
Impaired
MTJ Shape Abnormal +++ +
Myofiber Adhesion Defects +++ +

Double mutant pxna;pxnb embryos exhibited differential phenotypes dependent on maternal pxna or pxnb supply. Each phenotype was qualitatively scored based on age of onset and severity on a scale from – (not observed) or + (minor) to +++ (most severe).

How does Paxillin regulate Cardiovascular Development?

Zebrafish Paxillin MZ double mutants showed reduced or absent blood circulation, brain hemorrhaging and heart morphogenesis defects (Fig. 3). These phenotypes are consistent with phenotypes in Paxillin knockout mice, wherein embryonic lethality is caused by poor vasculogenesis and impaired heart development (Hagel et al., 2002). However, the mechanisms by which Pxn regulates cardiac and vascular development remain unclear. The zebrafish model developed herein provides a useful system for identifying roles for Paxillin proteins during cardiovascular development and future work will need to focus on mechanistic studies of Paxillin in the cardiovascular system. Interestingly, brain hemorrhaging defects observed in Paxillin MZ double mutants have also been observed in zebrafish mutants for a guanine nucleotide exchange factor, βPix (Liu et al., 2012), and one of its effectors, Pak2a (Buchner et al., 2007), and in embryos deficient in GIT1 (Liu et al., 2007). The GIT/PIX/PAK complex interacts with Paxillin in various mammalian cell types to spatially coordinate the activity of the Rho family GTPase Rac1 that promotes directional cell migration (Nayal et al., 2006; Turner et al., 1999; West et al., 2001). The same complex has also been implicated in endothelial cell focal adhesion formation and cell-cell barrier integrity (Birukova et al., 2007; Shikata et al., 2003). Functions for mammalian Paxillin in the recruitment of Rho GTPase activators and repressors have been extensively characterized in migrating cells (Brown and Turner, 2004). Thus, we speculate that the similar phenotypes observed in Paxillin MZ double mutant zebrafish and those with impaired Rho GTPase signaling reflect a role for Paxillin in regulating Rho GTPase activity during vascular development.

In addition to vascular defects, loss of Paxillin disrupted formation of the heart. Both MZpxna;Zpxnb and Zpxna;MZpxnb embryos have impaired heart development, but the visible onset of this phenotype differed between each genotype (Fig. 3). Double mutants lacking maternal pxna exhibited stringy, poorly looped hearts and pericardial edemas at 48 hpf, similar to zebrafish deficient in Rho GTPase signaling effectors (Dickover et al., 2014), while those without maternal pxnb developed visible cardiac defects by 3 dpf. Additionally, MZpxna;Zpxnb embryos had qualitatively more severe cardiac malformations than the majority of Zpxna;MZpxnb embryos. This variation in embryonic phenotype, dependent on the maternal contribution of either zebrafish pxn gene, suggests differential roles for each in this organ. One possibility is that maternal Pxna regulates cell shape changes and early bending of the heart tube, while Pxnb is required for later chamber morphogenesis or cardiac cell survival. Additionally, Paxillin MZ double mutant embryos had poor circulation throughout their vasculature. Since blood flow influences development of the heart (Dietrich et al., 2014), the loss of these mechanical forces may impact heart morphogenesis in the Paxillin MZ double mutants. Future studies are needed tease apart the interactions between the ECM, biophysical forces, Paxillin and Rho GTPase signaling during cardiac and vascular development.

Notochord Sheath Adhesion Requires Maternal Pxna in Pxn Double Mutants

Both mammalian Paxillin and zebrafish pxna expression are highly enriched in the embryonic notochord, a source of secreted morphogens and precursor to intervertebral discs (Corallo et al., 2013; Haga et al., 2009; Hagel et al., 2002). Although notochord morphogenesis has not been well characterized in Paxillin null mice, we observed disruption of notochord sheath cell adhesion and anterior-posterior axis elongation in MZpxna;Zpxnb zebrafish embryos. During normal development, a fibrous Collagen matrix envelops the notochord and acts as a rigid basement membrane substrate for the notochord sheath cells (Crawford et al., 2003). The organization of initially circumferential Collagen fibrils subsequently reorient longitudinally as inner notochord cell vacuoles expand, driving elongation of the anterior-posterior axis. We therefore speculate that Paxillin genes are required for adhesion to or remodeling of this matrix, since we observed relatively normal inner notochord cell inflation in MZpxna;Zpxnb embryos. More detailed ultrastructural description of the notochord ECM in Paxillin MZ double mutants will be required to test this model.

Functions for Paxillin during Skeletal Muscle Development

Our genetic approach to investigating the function of Paxillin during myotome morphogenesis suggests that an appropriate level of total Paxillin protein is required for MTJ shape change. While Paxillin MZ single mutants have subtle defects in MTJ shape, combined mutation of pxna and pxnb along with loss of maternally-deposited pxna exacerbates this phenotype. Although both pxna and pxnb are expressed in developing skeletal muscle, the distribution of each mRNA is distinct. Specifically, pxnb mRNAs are discretely localized to the MTJ while pxna mRNA is more diffuse along the length of elongated myoblasts (Jacob et al., 2016). Functional consequences of this differential distribution can be implied from the differential severity of myotome phenotypes observed in MZpxna;Zpxnb and Zpxna;MZpxnb embryos. A variety of defects including gaps between muscle cells, muscle cell detachment, and MTJ breaching were observed frequently in MZpxna;Zpxnb embryos, while Zpxna;MZpxnb embryos typically only had minor defects in MTJ chevron shape and myofiber adhesion later in development. It may be the case that distribution of pxna mRNA along the entire myofiber reflects a role for this gene in the adhesion of neighboring myofibers and to the MTJ. Conversely, local translation of mRNA would restrict Pxnb proteins to function at the MTJ alone. We speculate Pxnb is required to maintain MTJ shape and integrity. In contrast, our findings suggest that Pxna functions earlier in development, during the transition from a somite to a functional myotome to initially shape the MTJ. Strong expression of pxna mRNA in the youngest myotomes even at late stages of embryogenesis reflect this role (Jacob et al., 2016). Adhesion defects similar to Paxillin MZ double mutants in myotome morphogenesis have been well-documented in embryos lacking MTJ components such as ILK, a Paxillin-interacting protein, Dystroglycan, and Laminins (Bassett et al., 2003; Hall et al., 2007; Postel et al., 2008). Interestingly, defects in ILK and Dystroglycan mutant zebrafish develop after the onset of skeletal muscle contraction, implicating these proteins in maintaining adhesion in response to growing mechanical force through the myotome. We propose that Paxillin proteins, especially pxnb proteins, serve a similar role in later-stage mechanoregulation of myoblast adhesion.

Similar to MZpxna;Zpxnb embryos, obtusely-angled MTJs and over-elongated myoblasts were observed in zebrafish embryos deficient for Laminin adhesion (Goody et al., 2010; Goody et al., 2012; Snow et al., 2008). Our finding that loss of Pxn reduced Laminin accumulation at the embryonic MTJ (Fig. 5) is consistent with a previous finding that overexpression of Pxna can rescue myoblast elongation defects in zebrafish embryos with impaired Laminin organization at the MTJ (Goody et al., 2012), further supporting a role for Paxillin in “inside-out” Integrin signaling at the developing MTJ. MTJ composition and maturation is regulated through several mechanisms. The initial somite boundary is reinforced upon α5-Integrin activation and accumulation of extracellular Fibronectin bundles (Julich et al., 2015; Julich et al., 2009; Koshida et al., 2005). Similarly, accumulation and retention of Laminin at the MTJ is likely mediated via cell-surface receptors expressed by elongated myoblasts, such as α6- and α7-Integrins (Goody et al., 2012; Hamill et al., 2009). Differential Paxillin functions downstream of Fibronectin-binding and Laminin-binding Integrins have been described in cultured mammalian muscle cells (Sastry et al., 1999). In this system, overexpression of α5-Integrin, a receptor for Fibronectin, was associated with proliferation while α6-Integrin overexpression promoted differentiation and myoblast fusion (Sastry et al., 1999). The phenotype observed upon Paxillin overexpression was similar to that of α5-Integrin overexpression (Sastry et al., 1999). Interestingly however, introducing a non-phosphorylatable mutant of Paxillin into muscle cells did not affect muscle cell proliferation or differentiation. These results highlight that both Paxillin expression level and post-translational modifications impact myogenesis. While we did not observe gross defects in muscle cell differentiation in Paxillin MZ double mutant embryos, future work will be needed to characterize the dynamics of Paxillin phosphorylation during myogenesis in zebrafish.

Both Integrin “inside-out” and “outside-in” signaling may be regulated by Paxillin at MTJs. Localized accumulation of Integrin receptors in myoblasts may be regulated via Paxillin, which has been shown to influence the localization of vesicle trafficking machinery in other systems (Pignatelli et al., 2012; Spiczka and Yeaman, 2008). Recently, Paxillin has been implicated in Integrin activation to promote ECM binding and cell spreading (Theodosiou et al., 2016). Similar roles for Paxillin in Integrin activation at the developing MTJ may promote Laminin deposition. New tools to visualize Laminin-binding, Integrin localization and activity, as well as the dynamics of Laminin accumulation at the MTJ, will be necessary to investigate these possibilities. Furthermore, matrix metalloproteases (MMPs) are localized to and active at MTJs (Crawford and Pilgrim, 2005; Keow et al., 2012). In particular, clearing of Fibronectin from the ECM at MTJs is regulated by MMP11 (Jenkins et al., 2016). Zebrafish Paxillin genes may coordinate MMP localization and activity during the maturation of MTJs to allow a permissive extracellular environment for proper muscle morphogenesis. Surprisingly, a significant proportion of MZpxna;Zpxnb embryos have downregulated Fibronectin expression at MTJs by 48 hpf (Fig. 5D), suggesting a Paxillin or MMP11-independent mechanism for ECM remodeling at the MTJ. However, the small number of embryos which retain Fibronectin exhibit muscle cell detachment from the MTJ, further highlighting potential roles for Paxillin in modulating expression, localization, or activation of proper Integrin receptors in muscle cells.

Cytoskeletal Contractility Regulates Paxillin Localization and MTJ Development

Live imaging of somite boundary formation revealed that NMM-II is recruited prior to Pxna (Fig. 6; Movie 3). New boundaries form at the posterior interface of a somite and presomitic mesoderm via Eph/Ephrin signaling to activate α5-Integrin and repress the cell-cell adhesion protein Cdh2 (Julich et al., 2015; Julich et al., 2009). The activation of Integrins causes Fibronectin to polymerize and form a stable barrier. Integrin activation and subsequent ECM polymerization then activates further downstream adhesion signaling, likely including NMM-II activation, to maintain these boundaries. At a later phase, the somite boundary changes shape as elongated myoblasts adhere and continue their differentiation program. Coincident with this shape change is an accumulation of the ECM protein Laminin (Snow and Henry, 2009). Physical modeling has suggested that tonic contractions, independent of skeletal muscle movement, along the trunk of the embryo maintain the chevron shape of MTJs during this phase (Rost et al., 2014). Our live imaging observations suggest that Pxna is recruited to new somite boundaries after NMM-II is activated via α5-Integrin signaling, and that additional accumulation of Pxna occurs as elongated myoblasts adhere to the MTJ. A functional model to explain this observation involves Paxillin recruitment as a “checkpoint” for further ECM accumulation at the MTJ after the Fibronectin boundary has formed. Paxillin localization to focal adhesions is dependent on its LIM domains (Brown et al., 1996). Mechanical forces on the cytoskeleton influence the recruitment of many LIM domain proteins to sites of adhesion, possibly through physical unfolding of cryptic protein-protein interaction sites (Schiller et al., 2011). Future work will be needed to elucidate the biochemical signaling associated with recruiting Paxillin to the MTJ downstream of NMM-II activity.

Taking a pharmacological approach, we found that perturbation of cytoskeletal contractility during somitogenesis phenocopied the loss of Paxillin genes in terms of MTJ shape and integrity (Fig. 7). Interestingly, inhibition of contractility generated by NMM-II using blebbistatin disrupted recruitment of Pxna-GFP to the MTJ. This was similar to the reduced recruitment of many proteins possessing LIM domains, including Paxillin, Zyxin, and Migfilin to focal adhesions in mammalian cells treated with blebbistatin (Schiller et al., 2011). Treatment of zebrafish embryos with blebbistatin during mid-somitogenesis stages allowed us to affect myotomes acutely at different phases of their development (Henry et al., 2005) within individual embryos. Our findings suggest that NMM-II activity is required for normal myoblast elongation and boundary capture primarily during the early and transitory phases of somite-to-myotome maturation. Previous work has implicated the Paxillin binding partner, Git2, in NMM-II activation and regulation of cell shape during zebrafish epiboly (Yu et al., 2011). Interestingly, activation of NMM-II can promote Integrin activation and ECM fibril formation in newly formed somites (Julich et al., 2015). Tension generated by NMM-II can influence Integrin conformation and adhesion strengthening in a fibrosarcoma cell line (Friedland et al., 2009; Roca-Cusachs et al., 2013). The zebrafish MTJ provides an in vivo system to investigate the coordinated action of cytoskeletal mechanics, Integrin adhesion, and Paxillin signaling. Although direct roles for Rho GTPase involvement in MTJ adhesion and maturation have not been characterized in zebrafish, attenuated expression of Obscurin-a, or targeted disruption of its RhoGEF domain, results in disrupted MTJ shape and integrity (Raeker et al., 2010). We propose that Paxillin responds to mechanical stimuli on adhesion complexes to provide binding sites for recruitment of cytoskeletal regulators and signaling proteins to coordinate their activities to establish and maintain cell adhesion in the myotome and other tissues during zebrafish embryogenesis.

Altogether, this work has uncovered new roles for Paxillin in the vertebrate embryo. Analyses of mutant embryos indicated that Paxillin controls myotome morphogenesis by regulating ECM composition at the embryonic MTJ. Additionally, we have uncovered a role for actomyosin contractility that is critical for the localization of Paxillin to MTJs and the proper development of skeletal muscle. The generation of the pxn mutant zebrafish described here will enable future studies that further investigate the role of Paxillin in modulating adhesion signaling during vertebrate embryogenesis.

Supplementary Material

1

Fig. S1. Cardiac Left-Right Patterning and Sarcomere Assembly are Unaffected in Paxillin MZ Double Mutants. (A) Whole-mount mRNA in situ hybridization for heart-specific cmlc2 expression at the heart-jogging stage (30 hpf) revealed that asymmetric heart patterning is normal in Paxillin MZ double mutant embryos (indicated by direction of arrows). The number of embryos with proper heart-jogging direction over total number examined is indicated for each genotype. (B) Immunostaining for α-Actinin revealed that sarcomere formation is also normal in Paxillin MZ double mutant embryos. The number of embryos with normal Z-disc organization over total number examined is indicated for each genotype.

Fig. S2. Myotome and MTJ Defects Develop by 48 hpf in MZpxna;Zpxnb embryos. (A) Immunostaining for myofibers and MTJs at 48 hpf revealed that MZpxna;Zpxnb embryos exhibit large gaps and myoblast elongation through MTJs (white arrow). Scale bar = 25 μm (B) Quantification of MTJ chevron angle and MTJ breaching revealed that MTJ defects arise by 48 hpf in Paxillin MZ single and double mutant embryos. Injection of the pxna/b ATG-MO also resulted in a wider MTJ chevron angle compared to control MO injected embryos at an earlier stage of development. Data points represent individual embryo means and error bars show standard deviations from three independent experiments. **** p < 0.001, ** p < 0.005 determined by Kruskal-Wallis test with Dunn’s multiple comparisons post-hoc test, # # p < 0.01 determined by T-test, ns = not significant. (C) Immunostaining for Myosin revealed that a substantial number of MZpxna;Zpxnb embryos had large gaps between myofibers (arrow). The number of embryos exhibiting depicted phenotypes over the total number of embryos examined for three independent experiments is shown. Scale bar = 10 μm

Fig. S3. Validation of pxna/b ATG-MO Efficacy by Western Blotting and Rescue Experiments. (A) Targeting diagram for pxna/b ATG-MO. Sequence similarity between exon 1 (E1) and exon 2 (E2) of pxna and pxnb transcripts allows for translation blocking of both genes using one MO. Alignment shows MO target sequence compared with pxna and pxnb transcripts. Yellow highlights nucleotides shared between all sequences while blue highlights those shared between two sequences. (B) Western blotting with Paxillin antibody revealed robust knockdown of endogenous Pxn proteins in embryos injected with pxna/b ATG-MO and expression of exogenous GFP-Pxna fusion protein in embryos injected with pxna/b ATG-MO + GFP-Pxna mRNA at 3 dpf. (C) Pxn protein knockdown resulted in myotome defects which included wide MTJ chevron angles and MTJ breaches (arrows), these phenotypes were partially rescued by GFP-Pxna expression that localized to MTJs. Scale bar = 50 μm.

Movie 1. Blood Circulation in MZpxna and MZpxna;Zpxnb embryos at 48 hpf. 15 second time lapse of blood circulation through the tail region of representative MZpxna (top) and MZpxna;Zpxnb (bottom) embryos. MZpxna;Zpxnb embryos have reduced or absent circulation through tail vasculature.

Movie 2. Blood Circulation in MZpxnb and Zpxna;MZpxnb embryo at 3 dpf. 15 second time lapse of blood circulation through the tail region of MZpxnb (top) and Zpxna;MZpxnb (bottom) embryos. Zpxna;MZpxnb embryos had absent circulation at this stage.

Movie 3. Time-lapse imaging of Pxna-GFP and Myl12.1-mKate2 Dynamics during Somite Boundary Formation. Confocal images were taken every 6 minutes during the formation of new somite boundaries at 18 hpf in wild-type double transgenic embryos.

Download video file (37.6MB, avi)
2
Download video file (76.1MB, avi)
3
Download video file (91.6MB, avi)
4
5
6

Highlights.

  • Zebrafish paxillin genes regulate heart, notochord and skeletal muscle development.

  • Maternally supplied Paxillin proteins have non-redundant roles during development.

  • Loss of Paxillin alters extracellular matrix composition of myotendinous junctions.

  • Actomyosin cytoskeletal contractility is required for Paxillin localization to developing somite boundaries

Acknowledgments

We thank members of the Turner and Amack labs for helpful discussions, Sharleen Buel and Ashley Adler for technical assistance, and Clarissa Henry for sharing Tg(actb2:pxn-EGFP) fish.

Funding Sources

This work was supported by the National Institutes of Health (NIH R01 GM047607 to Christopher E. Turner and NIH R01 HL095690 to Jeffrey D. Amack).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Bassett DI, Bryson-Richardson RJ, Daggett DF, Gautier P, Keenan DG, Currie PD. Dystrophin is required for the formation of stable muscle attachments in the zebrafish embryo. Development. 2003;130:5851–5860. doi: 10.1242/dev.00799. [DOI] [PubMed] [Google Scholar]
  2. Bataille L, Delon I, Da Ponte JP, Brown NH, Jagla K. Downstream of identity genes: muscle-type-specific regulation of the fusion process. Developmental cell. 2010;19:317–328. doi: 10.1016/j.devcel.2010.07.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Birukova AA, Malyukova I, Poroyko V, Birukov KG. Paxillin-beta-catenin interactions are involved in Rac/Cdc42-mediated endothelial barrier-protective response to oxidized phospholipids. American journal of physiology. Lung cellular and molecular physiology. 2007;293:L199–211. doi: 10.1152/ajplung.00020.2007. [DOI] [PubMed] [Google Scholar]
  4. Bokel C, Brown NH. Integrins in development: moving on, responding to, and sticking to the extracellular matrix. Developmental cell. 2002;3:311–321. doi: 10.1016/s1534-5807(02)00265-4. [DOI] [PubMed] [Google Scholar]
  5. Brown MC, Perrotta JA, Turner CE. Identification of LIM3 as the principal determinant of paxillin focal adhesion localization and characterization of a novel motif on paxillin directing vinculin and focal adhesion kinase binding. The Journal of cell biology. 1996;135:1109–1123. doi: 10.1083/jcb.135.4.1109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Brown MC, Turner CE. Paxillin: adapting to change. Physiological reviews. 2004;84:1315–1339. doi: 10.1152/physrev.00002.2004. [DOI] [PubMed] [Google Scholar]
  7. Buchner DA, Su F, Yamaoka JS, Kamei M, Shavit JA, Barthel LK, McGee B, Amigo JD, Kim S, Hanosh AW, et al. pak2a mutations cause cerebral hemorrhage in redhead zebrafish. Proceedings of the National Academy of Sciences of the United States of America. 2007;104:13996–14001. doi: 10.1073/pnas.0700947104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Campbell ID, Humphries MJ. Integrin structure, activation, and interactions. Cold Spring Harbor perspectives in biology. 2011;3 doi: 10.1101/cshperspect.a004994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chen JN, van Eeden FJ, Warren KS, Chin A, Nusslein-Volhard C, Haffter P, Fishman MC. Left-right pattern of cardiac BMP4 may drive asymmetry of the heart in zebrafish. Development. 1997;124:4373–4382. doi: 10.1242/dev.124.21.4373. [DOI] [PubMed] [Google Scholar]
  10. Corallo D, Schiavinato A, Trapani V, Moro E, Argenton F, Bonaldo P. Emilin3 is required for notochord sheath integrity and interacts with Scube2 to regulate notochord-derived Hedgehog signals. Development. 2013;140:4594–4601. doi: 10.1242/dev.094078. [DOI] [PubMed] [Google Scholar]
  11. Costa ML, Escaleira RC, Jazenko F, Mermelstein CS. Cell adhesion in zebrafish myogenesis: distribution of intermediate filaments, microfilaments, intracellular adhesion structures and extracellular matrix. Cell motility and the cytoskeleton. 2008;65:801–815. doi: 10.1002/cm.20301. [DOI] [PubMed] [Google Scholar]
  12. Crawford BD, Henry CA, Clason TA, Becker AL, Hille MB. Activity and distribution of paxillin, focal adhesion kinase, and cadherin indicate cooperative roles during zebrafish morphogenesis. Molecular biology of the cell. 2003;14:3065–3081. doi: 10.1091/mbc.E02-08-0537. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Crawford BD, Pilgrim DB. Ontogeny and regulation of matrix metalloproteinase activity in the zebrafish embryo by in vitro and in vivo zymography. Developmental biology. 2005;286:405–414. doi: 10.1016/j.ydbio.2005.06.035. [DOI] [PubMed] [Google Scholar]
  14. Deakin NO, Turner CE. Paxillin comes of age. Journal of cell science. 2008;121:2435–2444. doi: 10.1242/jcs.018044. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Dickover M, Hegarty JM, Ly K, Lopez D, Yang H, Zhang R, Tedeschi N, Hsiai TK, Chi NC. The atypical Rho GTPase, RhoU, regulates cell-adhesion molecules during cardiac morphogenesis. Developmental biology. 2014;389:182–191. doi: 10.1016/j.ydbio.2014.02.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Dietrich AC, Lombardo VA, Veerkamp J, Priller F, Abdelilah-Seyfried S. Blood flow and Bmp signaling control endocardial chamber morphogenesis. Developmental cell. 2014;30:367–377. doi: 10.1016/j.devcel.2014.06.020. [DOI] [PubMed] [Google Scholar]
  17. Frantz C, Stewart KM, Weaver VM. The extracellular matrix at a glance. Journal of cell science. 2010;123:4195–4200. doi: 10.1242/jcs.023820. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Friedland JC, Lee MH, Boettiger D. Mechanically activated integrin switch controls alpha5beta1 function. Science. 2009;323:642–644. doi: 10.1126/science.1168441. [DOI] [PubMed] [Google Scholar]
  19. Gagnon JA, Valen E, Thyme SB, Huang P, Akhmetova L, Pauli A, Montague TG, Zimmerman S, Richter C, Schier AF. Efficient mutagenesis by Cas9 protein-mediated oligonucleotide insertion and large-scale assessment of single-guide RNAs. PloS one. 2014;9:e98186. doi: 10.1371/journal.pone.0098186. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gao C, Wang G, Amack JD, Mitchell DR. Oda16/Wdr69 is essential for axonemal dynein assembly and ciliary motility during zebrafish embryogenesis. Developmental dynamics: an official publication of the American Association of Anatomists. 2010;239:2190–2197. doi: 10.1002/dvdy.22355. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Goody MF, Kelly MW, Lessard KN, Khalil A, Henry CA. Nrk2b-mediated NAD+ production regulates cell adhesion and is required for muscle morphogenesis in vivo: Nrk2b and NAD+ in muscle morphogenesis. Developmental biology. 2010;344:809–826. doi: 10.1016/j.ydbio.2010.05.513. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Goody MF, Kelly MW, Reynolds CJ, Khalil A, Crawford BD, Henry CA. NAD+ biosynthesis ameliorates a zebrafish model of muscular dystrophy. PLoS biology. 2012;10:e1001409. doi: 10.1371/journal.pbio.1001409. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Haga Y, Dominique VJ, 3rd, Du SJ. Analyzing notochord segmentation and intervertebral disc formation using the twhh:gfp transgenic zebrafish model. Transgenic research. 2009;18:669–683. doi: 10.1007/s11248-009-9259-y. [DOI] [PubMed] [Google Scholar]
  24. Hagel M, George EL, Kim A, Tamimi R, Opitz SL, Turner CE, Imamoto A, Thomas SM. The adaptor protein paxillin is essential for normal development in the mouse and is a critical transducer of fibronectin signaling. Molecular and cellular biology. 2002;22:901–915. doi: 10.1128/MCB.22.3.901-915.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Hall TE, Bryson-Richardson RJ, Berger S, Jacoby AS, Cole NJ, Hollway GE, Berger J, Currie PD. The zebrafish candyfloss mutant implicates extracellular matrix adhesion failure in laminin alpha2-deficient congenital muscular dystrophy. Proceedings of the National Academy of Sciences of the United States of America. 2007;104:7092–7097. doi: 10.1073/pnas.0700942104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Hamill KJ, Kligys K, Hopkinson SB, Jones JC. Laminin deposition in the extracellular matrix: a complex picture emerges. Journal of cell science. 2009;122:4409–4417. doi: 10.1242/jcs.041095. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Henry CA, McNulty IM, Durst WA, Munchel SE, Amacher SL. Interactions between muscle fibers and segment boundaries in zebrafish. Developmental biology. 2005;287:346–360. doi: 10.1016/j.ydbio.2005.08.049. [DOI] [PubMed] [Google Scholar]
  28. Hirth S, Buhler A, Buhrdel JB, Rudeck S, Dahme T, Rottbauer W, Just S. Paxillin and Focal Adhesion Kinase (FAK) Regulate Cardiac Contractility in the Zebrafish Heart. PloS one. 2016;11:e0150323. doi: 10.1371/journal.pone.0150323. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Humphrey JD, Dufresne ER, Schwartz MA. Mechanotransduction and extracellular matrix homeostasis. Nature reviews. Molecular cell biology. 2014;15:802–812. doi: 10.1038/nrm3896. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Huveneers S, Truong H, Danen HJ. Integrins: signaling, disease, and therapy. International journal of radiation biology. 2007;83:743–751. doi: 10.1080/09553000701481808. [DOI] [PubMed] [Google Scholar]
  31. Hwang WY, Fu Y, Reyon D, Maeder ML, Tsai SQ, Sander JD, Peterson RT, Yeh JR, Joung JK. Efficient genome editing in zebrafish using a CRISPR-Cas system. Nature biotechnology. 2013;31:227–229. doi: 10.1038/nbt.2501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Jacob AE, Turner CE, Amack JD. Evolution and Expression of Paxillin Genes in Teleost Fish. PloS one. 2016;11:e0165266. doi: 10.1371/journal.pone.0165266. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Jenkins MH, Alrowaished SS, Goody MF, Crawford BD, Henry CA. Laminin and Matrix metalloproteinase 11 regulate Fibronectin levels in the zebrafish myotendinous junction. Skeletal muscle. 2016;6:18. doi: 10.1186/s13395-016-0089-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Julich D, Cobb G, Melo AM, McMillen P, Lawton AK, Mochrie SG, Rhoades E, Holley SA. Cross-Scale Integrin Regulation Organizes ECM and Tissue Topology. Developmental cell. 2015;34:33–44. doi: 10.1016/j.devcel.2015.05.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Julich D, Mould AP, Koper E, Holley SA. Control of extracellular matrix assembly along tissue boundaries via Integrin and Eph/Ephrin signaling. Development. 2009;136:2913–2921. doi: 10.1242/dev.038935. [DOI] [PubMed] [Google Scholar]
  36. Keow JY, Pond ED, Cisar JS, Cravatt BF, Crawford BD. Activity-based labeling of matrix metalloproteinases in living vertebrate embryos. PloS one. 2012;7:e43434. doi: 10.1371/journal.pone.0043434. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Kim HR, Ingham PW. The extracellular matrix protein TGFBI promotes myofibril bundling and muscle fibre growth in the zebrafish embryo. Developmental dynamics: an official publication of the American Association of Anatomists. 2009;238:56–65. doi: 10.1002/dvdy.21812. [DOI] [PubMed] [Google Scholar]
  38. Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, Schilling TF. Stages of embryonic development of the zebrafish. Developmental dynamics: an official publication of the American Association of Anatomists. 1995;203:253–310. doi: 10.1002/aja.1002030302. [DOI] [PubMed] [Google Scholar]
  39. Koshida S, Kishimoto Y, Ustumi H, Shimizu T, Furutani-Seiki M, Kondoh H, Takada S. Integrinalpha5-dependent fibronectin accumulation for maintenance of somite boundaries in zebrafish embryos. Developmental cell. 2005;8:587–598. doi: 10.1016/j.devcel.2005.03.006. [DOI] [PubMed] [Google Scholar]
  40. Ley K, Rivera-Nieves J, Sandborn WJ, Shattil S. Integrin-based therapeutics: biological basis, clinical use and new drugs. Nature reviews. Drug discovery. 2016;15:173–183. doi: 10.1038/nrd.2015.10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Link V, Shevchenko A, Heisenberg CP. Proteomics of early zebrafish embryos. BMC developmental biology. 2006;6:1. doi: 10.1186/1471-213X-6-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Liu J, Fraser SD, Faloon PW, Rollins EL, Vom Berg J, Starovic-Subota O, Laliberte AL, Chen JN, Serluca FC, Childs SJ. A betaPix Pak2a signaling pathway regulates cerebral vascular stability in zebrafish. Proceedings of the National Academy of Sciences of the United States of America. 2007;104:13990–13995. doi: 10.1073/pnas.0700825104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Liu J, Zeng L, Kennedy RM, Gruenig NM, Childs SJ. betaPix plays a dual role in cerebral vascular stability and angiogenesis, and interacts with integrin alphavbeta8. Developmental biology. 2012;363:95–105. doi: 10.1016/j.ydbio.2011.12.022. [DOI] [PubMed] [Google Scholar]
  44. Montague TG, Cruz JM, Gagnon JA, Church GM, Valen E. CHOPCHOP: a CRISPR/Cas9 and TALEN web tool for genome editing. Nucleic acids research. 2014;42:W401–407. doi: 10.1093/nar/gku410. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Nayal A, Webb DJ, Brown CM, Schaefer EM, Vicente-Manzanares M, Horwitz AR. Paxillin phosphorylation at Ser273 localizes a GIT1-PIX-PAK complex and regulates adhesion and protrusion dynamics. The Journal of cell biology. 2006;173:587–589. doi: 10.1083/jcb.200509075. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Pignatelli J, Jones MC, LaLonde DP, Turner CE. Beta2-adaptin binds actopaxin and regulates cell spreading, migration and matrix degradation. PloS one. 2012;7:e46228. doi: 10.1371/journal.pone.0046228. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Postel R, Vakeel P, Topczewski J, Knoll R, Bakkers J. Zebrafish integrin-linked kinase is required in skeletal muscles for strengthening the integrin-ECM adhesion complex. Developmental biology. 2008;318:92–101. doi: 10.1016/j.ydbio.2008.03.024. [DOI] [PubMed] [Google Scholar]
  48. Raeker MO, Bieniek AN, Ryan AS, Tsai HJ, Zahn KM, Russell MW. Targeted deletion of the zebrafish obscurin A RhoGEF domain affects heart, skeletal muscle and brain development. Developmental biology. 2010;337:432–443. doi: 10.1016/j.ydbio.2009.11.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Robu ME, Larson JD, Nasevicius A, Beiraghi S, Brenner C, Farber SA, Ekker SC. p53 activation by knockdown technologies. PLoS genetics. 2007;3:e78. doi: 10.1371/journal.pgen.0030078. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Roca-Cusachs P, del Rio A, Puklin-Faucher E, Gauthier NC, Biais N, Sheetz MP. Integrin-dependent force transmission to the extracellular matrix by alpha-actinin triggers adhesion maturation. Proceedings of the National Academy of Sciences of the United States of America. 2013;110:E1361–1370. doi: 10.1073/pnas.1220723110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Rossi A, Kontarakis Z, Gerri C, Nolte H, Holper S, Kruger M, Stainier DY. Genetic compensation induced by deleterious mutations but not gene knockdowns. Nature. 2015;524:230–233. doi: 10.1038/nature14580. [DOI] [PubMed] [Google Scholar]
  52. Rost F, Eugster C, Schroter C, Oates AC, Brusch L. Chevron formation of the zebrafish muscle segments. The Journal of experimental biology. 2014;217:3870–3882. doi: 10.1242/jeb.102202. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Sanger JW, Wang J, Holloway B, Du A, Sanger JM. Myofibrillogenesis in skeletal muscle cells in zebrafish. Cell motility and the cytoskeleton. 2009;66:556–566. doi: 10.1002/cm.20365. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Sastry SK, Lakonishok M, Wu S, Truong TQ, Huttenlocher A, Turner CE, Horwitz AF. Quantitative changes in integrin and focal adhesion signaling regulate myoblast cell cycle withdrawal. The Journal of cell biology. 1999;144:1295–1309. doi: 10.1083/jcb.144.6.1295. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Schiller HB, Friedel CC, Boulegue C, Fassler R. Quantitative proteomics of the integrin adhesome show a myosin II-dependent recruitment of LIM domain proteins. EMBO reports. 2011;12:259–266. doi: 10.1038/embor.2011.5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, et al. Fiji: an open-source platform for biological-image analysis. Nature methods. 2012;9:676–682. doi: 10.1038/nmeth.2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Shikata Y, Birukov KG, Garcia JG. S1P induces FA remodeling in human pulmonary endothelial cells: role of Rac, GIT1, FAK, and paxillin. J Appl Physiol (1985) 2003;94:1193–1203. doi: 10.1152/japplphysiol.00690.2002. [DOI] [PubMed] [Google Scholar]
  58. Snow CJ, Goody M, Kelly MW, Oster EC, Jones R, Khalil A, Henry CA. Time-lapse analysis and mathematical characterization elucidate novel mechanisms underlying muscle morphogenesis. PLoS genetics. 2008;4:e1000219. doi: 10.1371/journal.pgen.1000219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Snow CJ, Henry CA. Dynamic formation of microenvironments at the myotendinous junction correlates with muscle fiber morphogenesis in zebrafish. Gene expression patterns: GEP. 2009;9:37–42. doi: 10.1016/j.gep.2008.08.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Spiczka KS, Yeaman C. Ral-regulated interaction between Sec5 and paxillin targets Exocyst to focal complexes during cell migration. Journal of cell science. 2008;121:2880–2891. doi: 10.1242/jcs.031641. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Sztal TE, Sonntag C, Hall TE, Currie PD. Epistatic dissection of laminin-receptor interactions in dystrophic zebrafish muscle. Human molecular genetics. 2012;21:4718–4731. doi: 10.1093/hmg/dds312. [DOI] [PubMed] [Google Scholar]
  62. Theodosiou M, Widmaier M, Bottcher RT, Rognoni E, Veelders M, Bharadwaj M, Lambacher A, Austen K, Muller DJ, Zent R, et al. Kindlin-2 cooperates with talin to activate integrins and induces cell spreading by directly binding paxillin. eLife. 2016;5:e10130. doi: 10.7554/eLife.10130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Turner CE, Brown MC, Perrotta JA, Riedy MC, Nikolopoulos SN, McDonald AR, Bagrodia S, Thomas S, Leventhal PS. Paxillin LD4 motif binds PAK and PIX through a novel 95-kD ankyrin repeat, ARF-GAP protein: A role in cytoskeletal remodeling. The Journal of cell biology. 1999;145:851–863. doi: 10.1083/jcb.145.4.851. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Turner CE, Kramarcy N, Sealock R, Burridge K. Localization of paxillin, a focal adhesion protein, to smooth muscle dense plaques, and the myotendinous and neuromuscular junctions of skeletal muscle. Experimental cell research. 1991;192:651–655. doi: 10.1016/0014-4827(91)90090-h. [DOI] [PubMed] [Google Scholar]
  65. Warner A, Qadota H, Benian GM, Vogl AW, Moerman DG. The Caenorhabditis elegans paxillin orthologue, PXL-1, is required for pharyngeal muscle contraction and for viability. Molecular biology of the cell. 2011;22:2551–2563. doi: 10.1091/mbc.E10-12-0941. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. West KA, Zhang H, Brown MC, Nikolopoulos SN, Riedy MC, Horwitz AF, Turner CE. The LD4 motif of paxillin regulates cell spreading and motility through an interaction with paxillin kinase linker (PKL) The Journal of cell biology. 2001;154:161–176. doi: 10.1083/jcb.200101039. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Winograd-Katz SE, Fassler R, Geiger B, Legate KR. The integrin adhesome: from genes and proteins to human disease. Nature reviews. Molecular cell biology. 2014;15:273–288. doi: 10.1038/nrm3769. [DOI] [PubMed] [Google Scholar]
  68. Wozniak MA, Modzelewska K, Kwong L, Keely PJ. Focal adhesion regulation of cell behavior. Biochimica et biophysica acta. 2004;1692:103–119. doi: 10.1016/j.bbamcr.2004.04.007. [DOI] [PubMed] [Google Scholar]
  69. Wu Q, Zhang J, Koh W, Yu Q, Zhu X, Amsterdam A, Davis GE, Arnaout MA, Xiong JW. Talin1 is required for cardiac Z-disk stabilization and endothelial integrity in zebrafish. FASEB journal: official publication of the Federation of American Societies for Experimental Biology. 2015;29:4989–5005. doi: 10.1096/fj.15-273409. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Yagi R, Ishimaru S, Yano H, Gaul U, Hanafusa H, Sabe H. A novel muscle LIM-only protein is generated from the paxillin gene locus in Drosophila. EMBO reports. 2001;2:814–820. doi: 10.1093/embo-reports/kve178. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Yu JA, Foley FC, Amack JD, Turner CE. The cell adhesion-associated protein Git2 regulates morphogenetic movements during zebrafish embryonic development. Developmental biology. 2011;349:225–237. doi: 10.1016/j.ydbio.2010.10.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Zaidel-Bar R, Geiger B. The switchable integrin adhesome. Journal of cell science. 2010;123:1385–1388. doi: 10.1242/jcs.066183. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Zhang W, Huang Y, Gunst SJ. p21-Activated kinase (Pak) regulates airway smooth muscle contraction by regulating paxillin complexes that mediate actin polymerization. The Journal of physiology. 2016;594:4879–4900. doi: 10.1113/JP272132. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1

Fig. S1. Cardiac Left-Right Patterning and Sarcomere Assembly are Unaffected in Paxillin MZ Double Mutants. (A) Whole-mount mRNA in situ hybridization for heart-specific cmlc2 expression at the heart-jogging stage (30 hpf) revealed that asymmetric heart patterning is normal in Paxillin MZ double mutant embryos (indicated by direction of arrows). The number of embryos with proper heart-jogging direction over total number examined is indicated for each genotype. (B) Immunostaining for α-Actinin revealed that sarcomere formation is also normal in Paxillin MZ double mutant embryos. The number of embryos with normal Z-disc organization over total number examined is indicated for each genotype.

Fig. S2. Myotome and MTJ Defects Develop by 48 hpf in MZpxna;Zpxnb embryos. (A) Immunostaining for myofibers and MTJs at 48 hpf revealed that MZpxna;Zpxnb embryos exhibit large gaps and myoblast elongation through MTJs (white arrow). Scale bar = 25 μm (B) Quantification of MTJ chevron angle and MTJ breaching revealed that MTJ defects arise by 48 hpf in Paxillin MZ single and double mutant embryos. Injection of the pxna/b ATG-MO also resulted in a wider MTJ chevron angle compared to control MO injected embryos at an earlier stage of development. Data points represent individual embryo means and error bars show standard deviations from three independent experiments. **** p < 0.001, ** p < 0.005 determined by Kruskal-Wallis test with Dunn’s multiple comparisons post-hoc test, # # p < 0.01 determined by T-test, ns = not significant. (C) Immunostaining for Myosin revealed that a substantial number of MZpxna;Zpxnb embryos had large gaps between myofibers (arrow). The number of embryos exhibiting depicted phenotypes over the total number of embryos examined for three independent experiments is shown. Scale bar = 10 μm

Fig. S3. Validation of pxna/b ATG-MO Efficacy by Western Blotting and Rescue Experiments. (A) Targeting diagram for pxna/b ATG-MO. Sequence similarity between exon 1 (E1) and exon 2 (E2) of pxna and pxnb transcripts allows for translation blocking of both genes using one MO. Alignment shows MO target sequence compared with pxna and pxnb transcripts. Yellow highlights nucleotides shared between all sequences while blue highlights those shared between two sequences. (B) Western blotting with Paxillin antibody revealed robust knockdown of endogenous Pxn proteins in embryos injected with pxna/b ATG-MO and expression of exogenous GFP-Pxna fusion protein in embryos injected with pxna/b ATG-MO + GFP-Pxna mRNA at 3 dpf. (C) Pxn protein knockdown resulted in myotome defects which included wide MTJ chevron angles and MTJ breaches (arrows), these phenotypes were partially rescued by GFP-Pxna expression that localized to MTJs. Scale bar = 50 μm.

Movie 1. Blood Circulation in MZpxna and MZpxna;Zpxnb embryos at 48 hpf. 15 second time lapse of blood circulation through the tail region of representative MZpxna (top) and MZpxna;Zpxnb (bottom) embryos. MZpxna;Zpxnb embryos have reduced or absent circulation through tail vasculature.

Movie 2. Blood Circulation in MZpxnb and Zpxna;MZpxnb embryo at 3 dpf. 15 second time lapse of blood circulation through the tail region of MZpxnb (top) and Zpxna;MZpxnb (bottom) embryos. Zpxna;MZpxnb embryos had absent circulation at this stage.

Movie 3. Time-lapse imaging of Pxna-GFP and Myl12.1-mKate2 Dynamics during Somite Boundary Formation. Confocal images were taken every 6 minutes during the formation of new somite boundaries at 18 hpf in wild-type double transgenic embryos.

Download video file (37.6MB, avi)
2
Download video file (76.1MB, avi)
3
Download video file (91.6MB, avi)
4
5
6

RESOURCES