The Drosophila FUS ortholog cabeza promotes adult founder myoblast selection by Xrp1-dependent regulation of FGF signaling.

The number of adult myofibers in Drosophila is determined by the number of founder myoblasts selected from a myoblast pool, a process governed by fibroblast growth factor (FGF) signaling. Here, we show that loss of cabeza (caz) function results in a reduced number of adult founder myoblasts, leading to a reduced number and misorientation of adult dorsal abdominal muscles. Genetic experiments revealed that loss of caz function in both adult myoblasts and neurons contributes to caz mutant muscle phenotypes. Selective overexpression of the FGF receptor Htl or the FGF receptor-specific signaling molecule Stumps in adult myoblasts partially rescued caz mutant muscle phenotypes, and Stumps levels were reduced in caz mutant founder myoblasts, indicating FGF pathway deregulation. In both adult myoblasts and neurons, caz mutant muscle phenotypes were mediated by increased expression levels of Xrp1, a DNA-binding protein involved in gene expression regulation. Xrp1-induced phenotypes were dependent on the DNA-binding capacity of its AT-hook motif, and increased Xrp1 levels in founder myoblasts reduced Stumps expression. Thus, control of Xrp1 expression by Caz is required for regulation of Stumps expression in founder myoblasts, resulting in correct founder myoblast selection.

Introduction

In Drosophila, two phases of somatic muscle development can be distinguished: during embryonic development, a set of muscles is generated that will mediate larval movement, while during metamorphosis, adult muscles are generated that allow for adult movement, including eclosion from the pupal case, locomotion and flight. In the first 24 h after puparium formation (APF), nearly all larval muscles are histolysed and removed by phagocytosis [1]. Larval motor neurons persist through metamorphosis, but their processes are substantially remodeled to innervate the adult muscles [2]. The latter are formed from undifferentiated myoblasts originating from mesodermal precursor cells that were set aside during embryonic development. These cells continue to express Twist in the late embryo and proliferate during larval life [3]. During early pupal stages, proliferation continues and myoblasts spread out by migrating along the peripheral nerves [1]. Around 24 h APF, a subset of adult myoblasts is selected to become founder myoblasts, while the remaining cells become fusion-competent myoblasts which later fuse with founder myoblasts to form the characteristic multinucleate adult myotubes [4].

Founder cell selection is accompanied by a decline in Twist expression and induction of dumbfounded (duf, also known as kirre) expression [4]. Intriguingly, while the selection of founder myoblasts during embryonic development involves Notch-mediated lateral inhibition [5], adult founder myoblast selection is mediated by FGF signaling and independent of Notch signaling [4, 6]. Indeed, manipulation of the expression level of the FGF receptor Heartless (Htl) in adult myoblasts modulates the number of adult founder myoblasts selected [6]. While Htl is expressed in both founder and fusion-competent myoblasts, Stumps (also known as Dof or Hbr), an FGF receptor-specific signaling molecule downstream of Htl, is selectively expressed in founder myoblasts from 24 h APF onwards [6]. Thus, restriction of Stumps expression to founder myoblasts may be a key event in founder myoblast selection. The molecular mechanism that mediates restricted Stumps expression in founder myoblasts is currently elusive.

Here, we identify a novel role for cabeza (caz) in adult muscle development. Caz is the single Drosophila orthologue of FUS, EWSR1 and TAF15, three highly homologous proteins that constitute the FET protein family in humans [7]. The FET proteins are DNA- and RNA-binding proteins involved in gene expression regulation, including transcription, mRNA splicing and mRNA subcellular localization [7]. Each of the three FET proteins has been implicated in the pathogenesis of the motor neurodegenerative disorder amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD) [8]. We previously reported that loss of caz function results in failure of pharate adult flies to eclose from the pupal case due to motor weakness. This is at least in part mediated by loss of neuronal caz function, as selective reintroduction of Caz in neurons was sufficient to rescue the caz mutant eclosion defect. Moreover, selective inactivation of caz in neurons was sufficient to induce an adult eclosion defect, albeit with a fraction of adult escaper flies that display dramatic motor performance deficits and reduced life span [9]. The fact that selective caz inactivation in neurons induces an adult eclosion defect that is not as severe as caz animals suggests that loss of caz function in other cell types may contribute to the caz mutant adult eclosion defect. We considered the adult abdominal muscles (Fig 1A) as a likely candidate, given their known role in mediating adult eclosion.

Fig 1

Loss of caz function deranges adult abdominal muscle development.

A, Schematic of adult abdominal muscle architecture in Drosophila. Dorsal, lateral and ventral abdominal muscles are indicated. B, A transgenic line, in which Myosin heavy chain is GFP-tagged (Mhc-GFP), allowed visualization of dorsal abdominal muscles (DAMs) at 96 h APF in wild type (WT), caz, and caz animals. Red arrowheads indicate misoriented DAMs. Scale bar: 50μm. C,D, Quantification of DAM number in abdominal segments 3 (A3) and 4 (A4) of caz (C) and caz (D) pupae at 96 h APF, as compared to WT. Mann Whitney test (C), t-test with Welch correction (D); ***p<0.0005; n = 10 WT versus 14 caz (A3), 11 WT versus 12 caz (A4), 13 WT versus 15 caz (A3), 8 WT versus 10 caz (A4). Average ± SEM. E,F, Percentage of misoriented DAMs per animal in segments A3 and A4 of caz (E) and caz (F) pupae at 96 h APF, as compared to WT. One sample Wilcoxon signed rank test; *p<0.05, **p<0.005; n = 10 WT versus 14 caz (A3), 11 WT versus 12 caz (A4), 13 WT versus 15 caz (A3), 8 WT versus 10 caz (A4). Average ± SEM.

We therefore investigated the role of caz function in adult abdominal muscle development. Loss of caz function resulted in a significant reduction of the number of adult founder myoblasts and dorsal abdominal muscles (DAMs), as well as DAM misorientation. Interestingly, loss of caz function in both adult myoblasts and neurons contributed to the deranged development of adult abdominal muscles. Increased expression of Htl or Stumps in adult myoblasts significantly rescued caz mutant muscle phenotypes, indicating that impaired FGF signaling contributes to these phenotypes. Consistently, Stumps levels were significantly reduced in caz mutant founder myoblasts. It was previously reported that neuronal dysfunction in caz mutants is mediated by increased expression of Xrp1, a DNA-binding protein involved in gene expression regulation [10]. Here, we could show that also the newly-discovered caz mutant muscle phenotypes are caused by increased Xrp1 expression levels in both adult myoblasts and neurons. Muscle phenotypes induced by increased Xrp1 expression were dependent on the DNA-binding capacity of its AT-hook motif, and Xrp1 overexpression suppressed Stumps expression in adult founder myoblasts. In summary, we uncovered a novel mechanism for regulation of Stumps expression in adult founder myoblasts: Caz controls Xrp1 expression, thus permitting increased Stumps levels in founder myoblasts and correct founder myoblast selection.

Results

Loss of caz function deranges adult abdominal muscle development

As proper abdominal muscle function is required for eclosion of adult Drosophila from the pupal case, we decided to investigate adult abdominal muscle development in two independent caz loss-of-function lines (caz and caz) [9]. We generated caz mutants in which the Myosin heavy chain (Mhc) gene is GFP-tagged to label muscle fibers [11-13]. In vivo confocal imaging of dorsal abdominal muscles (DAMs) 96 h APF revealed a reduced number of DAMs in both caz and caz (Fig 1B). Quantification of DAM number in abdominal hemisegments 3 and 4 revealed a significant reduction by ~30–40% in both body segments of the two caz mutant lines (Fig 1C and 1D). In addition, DAMs were frequently misoriented in caz mutant animals (arrowheads in Fig 1B; quantification in Fig 1E and 1F). These abdominal muscle defects are independent of Mhc-GFP expression, as immunostaining of late pupal filets for actin confirmed the profound muscle developmental defects, with both reduced number and misorientation of DAMs in caz mutants that did not carry additional transgenes (Fig 2A and 2B; S1A Fig). Furthermore, caz mutant muscle defects cannot be attributed to a delay in pupal development, as muscle defects were observed both in age-matched (100 h APF) and stage-matched pupae (based on outer appearance [14], S1A and S1B Fig). Importantly, muscle defects were not confined to DAMs, but also observed in other abdominal muscle groups, including ventral abdominal muscles (Fig 2N and 2R). Of note, the larval muscle pattern and number was not affected by loss of caz function (S2 Fig), indicating that caz mutant muscle defects were confined to adult muscle development, while embryonic muscle development occurred normally.

Fig 2

Motor neuron number and muscle innervation are not affected by loss of caz function.

A-H, Immunostaining for actin (A,B), Caz (C,D), and HRP (E,F) on abdominal filets of WT (A,C,E,G) and caz (B,D,F,H) pupae at 100 h APF showing DAM morphology and innervation. The merge of actin and HRP is shown in panels G,H. Scale bar: 50μm. I, Central motor neuron clusters in segments A2 to A7 in the ventral nerve cord of a third instar larva are delineated by a dashed red rectangle. J, Quantification of motor neuron number in segments A2 to A7 in WT versus caz third instar ventral nerve cord. Unpaired t-test; n = 17 WT versus 22 caz. Average ± SEM. K-R, Immunostaining for HRP (K,O), Dlg1 (L,P) and actin on abdominal filets of WT (K-N) and caz (O-R) pupae at 100 h APF showing ventral abdominal muscle morphology and innervation. Merges between HRP and Dlg1 (M,Q) and between HRP, Dlg1 and actin (N,R) are shown. Scale bar: 10μm. S, NMJ area relative to DAM area (shown as % of WT) in segments A3 and A4 of WT versus caz pupae at 96 h APF. Unpaired t-test; n = 18 WT versus 18 caz. Average ± SEM. T, Bouton number per DAM in segments A3 and A4 of WT versus caz pupae at 96 h APF. Unpaired t-test; n = 18 WT versus 19 caz. Average ± SEM.

These muscle morphological defects did not appear secondary to defective innervation by motor neurons, as the number of motor neurons in the third instar larval ventral cord was not altered in caz animals (Fig 2I and 2J). The fact that caz mutant larvae enter metamorphosis with a normal number of motor neurons is relevant for pupal development, as the larval motor neurons persist to adulthood, in contrast to the majority of larval muscles, which degenerate during metamorphosis [1, 2]. Furthermore, staining of abdominal filets of late pupae (100 h APF) for horseradish peroxidase (HRP), Discs large 1 (Dlg1) and actin revealed normal innervation of both dorsal and ventral abdominal muscles in caz knock-out animals (Fig 2A–2H and 2K–2R). This was confirmed by quantitative analysis of NMJ morphology on DAMs of WT and caz pupae 96 h APF, which revealed that both NMJ area and bouton number were not altered in caz (Fig 2S and 2T). Thus, loss of caz function induces muscle developmental defects, which are not attributable to defective motor innervation.

Reduced number of founder myoblasts in caz mutant animals

As the number of adult muscles, including DAMs, is dependent on the number of founder myoblasts that are selected during early pupal stages (~24 h APF), we investigated whether caz is expressed in founder myoblasts, and whether the number of founder myoblasts in the dorsal abdomen was reduced in caz mutants. To identify founder myoblasts, a duf-lacZ transgenic line was used in which β-galactosidase is selectively expressed in founder myoblasts under the control of the duf/kirre promoter [4]. Double immunostaining for Caz and β-galactosidase revealed that Caz is expressed in founder myoblasts (Fig 3A), as well as other cells, likely including fusion-competent myoblasts, consistent with the previously reported broad expression pattern of caz [9]. Importantly, the number of founder myoblasts was significantly reduced in caz mutant pupae as compared to wild type (WT) control (Fig 3B and 3C). Thus, the reduced number of DAMs in caz mutant late pupae is at least in part attributable to a reduced number of founder myoblasts 24 h APF.

Fig 3

Reduced founder myoblast number in caz mutants and rescue of caz mutant muscle phenotypes by adult myoblast-selective caz expression.

A, Immunostaining of duf-lacZ/Y transgenic pupal filets 24 h APF for β-galactosidase and Caz revealed Caz expression in founder myoblasts. Scale bar: 10μm. B, Immunostaining of founder myoblasts in a dorsal abdominal segment (A3 or A4) of 24 h APF duf-lacZ pupal filets, either WT (duf-lacZ/Y) or caz (caz,duf-lacZ/Y), for β-galactosidase (green) and Futsch (red). Scale bar: 20μm C, Quantification of the number of founder myoblasts in a dorsal abdominal segment (A3 or A4) of WT (duf-lacZ/Y) and caz (caz,duf-lacZ/Y) pupae at 24 h APF. Unpaired t-test; ***p<0.001; n = 8 WT versus 15 caz. Average ± SEM. D, Quantification of the number of founder myoblasts in a dorsal abdominal segment (A3 or A4) of 24 h APF caz pupae in which caz is selectively expressed in adult myoblasts (caz,duf-lacZ,1151-GAL4/Y;; UAS-caz/+) as compared to the relevant control genotypes (duf-lacZ,1151-GAL4/Y // duf-lacZ,1151-GAL4/Y;; UAS-caz/+ // caz,duf-lacZ,1151-GAL4/Y). Two-way ANOVA with Tukey’s multiple comparisons test; **p<0.001, ***p<0.001; n = 13 per genotype. Average ± SEM. E,F, Quantification of DAM number (E) and percentage of misoriented DAMs per animal (F) in segment A4 of 96 h APF caz pupae in which caz is selectively expressed in adult myoblasts (caz,1151-GAL4/Y; Mhc-GFP,his-RFP/+; UAS-caz/+) as compared to the relevant control genotypes (1151-GAL4/Y; Mhc-GFP,his-RFP/+ // 1151-GAL4/Y; Mhc-GFP,his-RFP/+; UAS-caz/+ // caz,1151-GAL4/Y; Mhc-GFP,his-RFP/+). Statistics (E): Brown-Forsythe and Welch ANOVA with Dunnett’s post test; **p<0.01, ***p<0.001; n = 8 per genotype. Average ± SEM. Statistics (F): One sample Wilcoxon signed rank test to compare all genotypes to control (set to 0), t-test with Welch’s correction to compare caz to caz;; UAS-caz/+; *p<0.05, **p<0.005, ***p<0.0001; n = 9 control, 8 UAS-caz, 10 caz, 18 caz; UAS-caz. G, Quantification of the number of founder myoblasts in a dorsal abdominal segment (A3 or A4) of 24 h APF pupae in which caz is selectively inactivated in myoblasts (caz,duf-lacZ,1151-GAL4/Y; UAS-FLP/+) as compared to the relevant control genotype (caz,duf-lacZ,1151-GAL4/Y). Unpaired t-test; ***p<0.0001; n = 18 caz, 25 caz; UAS-FLP. Average ± SEM. H,I, Quantification of DAM number (H) and percentage of misoriented DAMs per animal (I) in segment A4 of 96 h APF pupae in which caz is selectively inactivated in myoblasts (caz,1151-GAL4/Y; Mhc-GFP,his-RFP/UAS-FLP) as compared to the relevant control genotype (caz,1151-GAL4/Y; Mhc-GFP,his-RFP/+). Statistics (H): Mann Whitney test; ***p<0.0001; n = 23 caz, 22 caz; UAS-FLP. Average ± SEM. Statistics (I): One sample Wilcoxon test; *p<0.05; n = 23 caz, 22 caz; UAS-FLP. Average ± SEM.

Loss of caz function in adult myoblasts contributes to caz mutant muscle defects

Our findings suggested that the caz mutant muscle defects may be caused by loss of caz function in founder myoblasts. In line with this possibility, selective expression of Caz in adult myoblasts (1151-GAL4) of caz mutants rescued their reduced number of founder myoblasts to a level that was not statistically different from control genotypes (Fig 3D). Consistently, selective Caz expression in adult myoblasts also substantially rescued the reduced DAM number and DAM misorientation of caz pupae 96 h APF (Fig 3E and 3F). We further examined the effect of selective inactivation of caz in adult myoblasts by crossing animals that selectively express FLP recombinase in these cells (1151-GAL4>UAS-FLP) to conditional caz knock-out animals in which the caz gene is flanked by FRT sites (caz) [9]. Adult myoblast-selective caz inactivation resulted in a significantly reduced number of founder myoblasts at 24 h APF (Fig 3G) and a reduced DAM number and DAM misorientation at 96 h APF (Fig 3H and 3I; representative images in S3A Fig), albeit these muscle developmental defects appeared slightly less pronounced than in full-body caz knock-out animals. Together, these data show that Caz has a cell-autonomous function in adult myoblasts to promote founder myoblast selection and to ensure proper DAM number and orientation.

Neuron-selective loss of caz function causes muscle defects

The fact that (i) muscle phenotypes induced by adult myoblast-selective caz inactivation were not as severe as in full-body caz mutants, and (ii) adult myoblast-selective Caz reintroduction substantially but not fully rescued caz mutant muscle phenotypes, suggested that cell types other than muscle also contribute to the caz mutant muscle phenotypes. As caz function is known to be particularly important in neurons [9, 15], we evaluated whether loss of caz function in neurons could contribute to caz mutant muscle phenotypes. Consistent with this possibility, selective caz inactivation in neurons resulted in a significantly reduced number of founder myoblasts at 24 h APF (Fig 4A), and in a reduced DAM number and DAM misorientation at 96 h APF (Fig 4B and 4C; representative images in S3B Fig). We further evaluated the effect of neuron-selective Caz expression on caz mutant muscle phenotypes. Remarkably, even in a wild type genetic background, neuron-selective Caz overexpression significantly increased the number of DAMs at 96 h APF (Fig 4D). In caz pupae, neuron-selective Caz expression partially rescued the reduced DAM number (Fig 4D). Two-way ANOVA statistical analysis indicated that the DAM number was increased to a similar extent in the control and caz genetic background (the ‘interaction’ between caz genotype and Caz overexpression was not statistically significant, p = 0.086). DAM misorientation in caz pupae was substantially rescued by neuron-selective Caz expression (Fig 4E). In all, our data show that loss of caz function in both neurons and adult myoblasts contributes to caz mutant muscle phenotypes.

Fig 4

Neuron-selective loss of caz function causes muscle defects.

A, Quantification of the number of founder myoblasts in a dorsal abdominal segment (A3 or A4) of 24 h APF pupae in which caz is selectively inactivated in neurons (caz,duf-lacZ,elav-GAL4/Y; UAS-FLP/+) as compared to the relevant control (caz,duf-lacZ,elav-GAL4/Y). Unpaired t-test; **p<0.001, n = 14 caz versus n = 18 caz; UAS-FLP. Average ± SEM. B,C, Quantification of DAM number (B) and percentage of misoriented DAMs per animal (C) in segment A4 of 96 h APF pupae in which caz is selectively inactivated in neurons (caz,elav-GAL4/Y; Mhc-GFP,his-RFP/UAS-FLP) as compared to the relevant control genotypes (caz,elav-GAL4/Y; Mhc-GFP,his-RFP/+ // caz/Y; Mhc-GFP,his-RFP/UAS-FLP // elav-GAL4/Y; Mhc-GFP,his-RFP/UAS-FLP). Statistics (B): Kruskal-Wallis test with Dunn’s multiple comparisons test; *p<0.05, **p<0.005, ***p<0.0001. Statistics (C): one sample Wilcoxon signed rank test; **p<0.01; n = 14 caz, elav-GAL4, 14 caz; UAS-FLP, 14 elav-GAL4>UAS-FLP, 15 caz, elav-GAL4>UAS-FLP. Average ± SEM. D,E, Quantification of DAM number (D) and percentage of misoriented DAMs per animal (E) in segment A4 of 96 h APF caz pupae in which caz is selectively expressed in neurons (caz,elav-GAL4/Y; Mhc-GFP,his-RFP/+; UAS-caz/+) as compared to the relevant control genotypes (elav-GAL4/Y; Mhc-GFP,his-RFP/+ // elav-GAL4/Y; Mhc-GFP,his-RFP/+; UAS-caz/+ // caz,elav-GAL4/Y; Mhc-GFP,his-RFP/+). Statistics (D): two-way ANOVA with Tukey’s multiple comparisons test; ***p<0.0001; n = 21 per genotype. Statistics (E): one sample Wilcoxon signed rank test to compare all genotypes to control, Mann-Whitney test to compare caz to caz; UAS-caz; **p<0.01, ***p<0.0001; n = 21 control, 23 UAS-caz, 21 caz, 25 caz; UAS-caz. Average ± SEM.

Defective FGF signaling contributes to caz mutant muscle phenotypes

As the FGF signaling pathway has previously been implicated in adult founder myoblast selection [6], we evaluated whether defective FGF signaling might contribute to caz mutant muscle phenotypes. Interestingly, overexpression of the FGF receptor Heartless (Htl) in adult myoblasts (1151-GAL4) partially but significantly rescued the reduced DAM number and DAM misorientation in caz mutants (Fig 5A and 5B). Htl overexpression in an otherwise wild type background tended to increase DAM number, but this difference did not reach statistical significance. Importantly, two-way ANOVA revealed that the rescue of caz mutant muscle phenotypes by myoblast-selective Htl overexpression represents a genuine genetic interaction between caz and Htl, and is not merely an additive effect of Htl overexpression in a caz mutant background.

Fig 5

Defective FGF signaling contributes to caz mutant muscle phenotypes.

A,B, Quantification of DAM number (A) and percentage of misoriented DAMs per animal (B) in segments A3 and A4 of 96 h APF caz pupae in which Htl is selectively expressed in adult myoblasts (caz,1151-GAL4/Y; Mhc-GFP,his-RFP/+; UAS-htl/+) as compared to the relevant control genotypes (1151-GAL4/Y; Mhc-GFP,his-RFP/+ // 1151-GAL4/Y; Mhc-GFP,his-RFP/+; UAS-htl/+ // caz,1151-GAL4/Y; Mhc-GFP,his-RFP/+). Statistics (A): two-way ANOVA with Tukey’s multiple comparisons test; ***p<0.0001; n = 19 per genotype. Average ± SEM. Statistics (B): one sample Wilcoxon signed rank test to compare all genotypes to control, Mann-Whitney test to compare caz to caz; UAS-htl; *p<0.05, **p<0.01, ***p<0.001; n = 32 control, 37 UAS-htl, 33 caz, 19 caz; UAS-htl. Average ± SEM. C,D, Quantification of DAM number (C) and percentage of misoriented DAMs per animal (D) in segments A3 and A4 of 96 h APF caz pupae in which Stumps is selectively expressed in adult myoblasts (caz,1151-GAL4/Y; Mhc-GFP,his-RFP/UAS-stumps) as compared to the relevant control genotypes (1151-GAL4/Y; Mhc-GFP,his-RFP/+ // 1151-GAL4/Y; Mhc-GFP,his-RFP/UAS-stumps/+ // caz,1151-GAL4/Y; Mhc-GFP,his-RFP/+). Statistics (C): two-way ANOVA with Tukey’s multiple comparisons test; ***p<0.0001; n = 16 per genotype. Average ± SEM. Statistics (D): one sample Wilcoxon signed rank test to compare all genotypes to control, Mann-Whitney test to compare caz to caz; UAS-stumps for A3, t-test with Welch’s correction for A4; **p<0.005, ***p<0.001; n = 16 per genotype. Average ± SEM. E, Immunostaining of founder myoblasts in a dorsal abdominal segment (A3 or A4) of 24 h APF duf-lacZ pupal filets, either WT (duf-lacZ/Y) or caz (caz,duf-lacZ/Y), for β-galactosidase (green) and Stumps (red). Scale bar: 50μm. F, Quantification of Stumps staining intensity (% of WT) in founder myoblasts of caz (caz,duf-lacZ/Y) 24 h APF pupae relative to WT (duf-lacZ/Y). Unpaired t-test; *p<0.05; n = 9 WT, 10 caz. Average ± SEM.

To confirm the role of the FGF pathway in caz mutant muscle phenotypes, we selectively overexpressed Stumps, the FGF receptor-specific signal transduction molecule downstream of Htl, in adult myoblasts. This also significantly rescued the reduced DAM number and DAM misorientation (Fig 5C and 5D) in caz mutants. Together these data indicate that caz mutant muscle phenotypes are at least in part attributable to defective FGF signaling.

These findings raised the hypothesis that Caz may, directly or indirectly, regulate the expression level of FGF pathway components. As Stumps expression becomes restricted to founder myoblasts 24 h APF [6], suggesting a key role in founder myoblast selection, we evaluated Stumps expression levels in adult founder myoblasts by immunostaining, using a previously described specific anti-Stumps antibody [16]. This experiment revealed significantly reduced Stumps protein levels in caz founder myoblasts (Fig 5E and 5F). Thus, loss of caz function leads to reduced Stumps expression, which may explain the reduced founder myoblast number in caz mutant animals.

Increased Xrp1 levels mediate caz mutant muscle phenotypes

We have previously reported that neuronal dysfunction in caz mutants is mediated by increased levels of Xrp1, a DNA-binding protein involved in gene expression regulation [10]. Upregulation of Xrp1 by ~3-fold was not only found in caz mutant nervous system but also in larval body wall, which predominantly consists of muscles [10]. We therefore evaluated whether the newly discovered caz mutant muscle phenotypes are also mediated by increased Xrp1 levels. Consistent with increased Xrp1 expression as the key mediator of caz mutant phenotypes, heterozygosity for Xrp1 almost fully rescued the reduced DAM number (Fig 6A) and fully rescued DAM misorientation (S4A Fig) in caz mutant animals. Importantly, selective overexpression of Xrp1 in adult myoblasts (1151-GAL4) was sufficient to induce a significant reduction of DAM number (Fig 6B), as well as DAM misorientation (S4B Fig), and a substantial reduction in the number of founder myoblasts (Fig 6C). Thus, adult myoblast-selective Xrp1 overexpression phenocopied caz mutant muscle phenotypes. This was dependent on the DNA-binding capacity of Xrp1, as adult myoblast-selective overexpression of Xrp1 with a subtle mutation in its AT-hook DNA-binding domain did not induce muscle phenotypes (Fig 6B, S4B Fig). These data suggest that gene expression dysregulation induced by increased Xrp1 expression in adult myoblasts mediates caz mutant muscle phenotypes. Based on our finding that Stumps expression is reduced in caz mutant founder myoblasts (Fig 5E and 5F), we hypothesized that increased Xrp1 expression may suppress Stumps expression in adult founder myoblasts. Immunostaining for Stumps confirmed that Xrp1 overexpression in adult myoblasts of otherwise wild type animals was sufficient to significantly reduce Stumps levels in founder myoblasts (Fig 6D and 6E).

Fig 6

Increased Xrp1 levels mediate caz mutant muscle phenotypes.

A, Quantification of DAM number in segments A3 and A4 of 96 h APF caz pupae that are heterozygous for Xrp1 (caz/Y; Mhc-GFP,his-RFP/+; Xrp1/+) as compared to the relevant control genotypes (+/Y; Mhc-GFP,his-RFP/+ // +/Y; Mhc-GFP,his-RFP/+; Xrp1/+ // caz/Y; Mhc-GFP,his-RFP/+). Two-way ANOVA with Tukey’s multiple comparisons test; **p<0.01, ***p<0.0001; n = 15 per genotype. Average ± SEM. B, Quantification of DAM number in segments A3 and A4 of 96 h APF male pupae in which Xrp1 is selectively overexpressed in adult myoblasts (1151-GAL4), either as WT protein or with a subtle mutation that disrupts the DNA-binding capacity of the AT-hook motif (Mut), as compared to driver-only control. Brown-Forsythe and Welch ANOVA with Dunnett’s post test; ***p<0.001; n = 8 control, 23 UAS-Xrp1 WT, 20 UAS-Xrp1 Mut. Average ± SEM. C, Quantification of the number of founder myoblasts in 24 h APF male pupae in which Xrp1 is overexpressed in adult myoblasts (1151-GAL4). Mann Whitney test; ***p<0.001; n = 15 WT versus 19 UAS-Xrp1 WT. Average ± SEM. D, Immunostaining of founder myoblasts in a dorsal abdominal segment (A3 or A4) of 24 h APF duf-lacZ pupal filets, either WT (1151-GAL4,duf-lacZ/Y) or 1151-GAL4,duf-lacZ/Y;;UAS-Xrp1 WT/+, for β-galactosidase (green) and Stumps (red). Scale bar: 20μm. E, Quantification of Stumps staining intensity (% of WT) in founder myoblasts of 1151-GAL4,duf-lacZ/Y;; UAS-Xrp1 WT/+ 24 h APF pupae relative to WT (1151-GAL4,duf-lacZ/Y). Unpaired t-test; *p<0.05; n = 7 WT versus 8 UAS-Xrp1 WT. Average ± SEM. F,G, Quantification of DAM number (F) and percentage of misoriented DAMs per animal (G) in segments A3 and A4 of 96 h APF caz pupae in which Xrp1 is selectively knocked-down in neurons (caz,elav-GAL4/Y; Mhc-GFP,his-RFP/+; UAS-Xrp1-RNAi/+) as compared to the relevant control genotypes (elav-GAL4/Y; Mhc-GFP,his-RFP/+ // elav-GAL4/Y; Mhc-GFP,his-RFP/+; UAS-Xrp1-RNAi/+ // caz,elav-GAL4/Y; Mhc-GFP,his-RFP/+). Statistics (F): two-way ANOVA with Tukey’s multiple comparisons test; *p<0.05, **p<0.005, ***p<0.0001; n = 10 per genotype. Statistics (G): one sample Wilcoxon signed rank test to compare all genotypes to control, and caz to caz; UAS-Xrp1-RNAi, **p<0.005; n = 13 control, 15 UAS-Xrp1-RNAi, 10 caz, 16 caz; UAS-Xrp1-RNAi. Average ± SEM.

In addition to increased Xrp1 expression in adult myoblasts, increased Xrp1 levels in neurons may also contribute to caz mutant muscle phenotypes. Indeed, selective knock-down of Xrp1 in neurons (elav-GAL4) induced a partial but substantial rescue of caz muscle phenotypes, both DAM number and misorientation (Fig 6F and 6G). Thus, increased Xrp1 levels in both muscle and neurons mediate caz mutant muscle phenotypes.

Discussion

Cabeza (caz) is the single Drosophila orthologue of the human FET protein family, which consists of FUS, EWSR1 and TAF15 [7]. Caz was previously reported to have a particularly important function in neurons. Indeed, similar to full-body caz loss-of-function (LOF) mutants, selective caz inactivation in neurons induced motor deficits resulting in failure of the majority of adult flies to eclose from the pupal case, and adult escaper flies displayed dramatic climbing deficits and reduced life span [9]. Furthermore, neuron-selective expression of Caz or human FUS was sufficient to rescue the adult eclosion and motor performance defects of caz LOF mutants [9, 15]. Here, we show for the first time that caz mutants display adult abdominal muscle defects, including a reduced number and occasional misorientation of dorsal abdominal muscles (DAMs). The reduced DAM number correlated with a reduced number of adult founder myoblasts, and our data indicate that caz mutant adult muscle developmental defects are to an important extent attributable to loss of caz function in adult myoblasts. Indeed, similar to full-body caz LOF mutants, adult myoblast-selective inactivation of caz resulted in a reduced number of adult founder myoblasts and DAMs, as well as DAM misorientation. In addition, myoblast-selective Caz expression in a caz mutant background substantially rescued adult muscle developmental defects. Thus, caz function in adult muscle precursor cells is required for normal muscle development.

The normal muscle morphology and number in caz mutant third instar larvae indicated a selective defect in adult muscle development, with the reduced number of adult founder myoblasts as the earliest observable defect. While embryonic founder myoblast selection involves Notch-mediated lateral inhibition [5], FGF signaling through the FGF receptor Heartless (Htl) and its downstream signal transduction molecule Stumps/Dof/Hbr has been shown to mediate adult founder myoblast selection [6]. We therefore investigated the potential contribution of defective FGF signaling to caz mutant muscle phenotypes. We found that selective expression of Htl or Stumps in adult myoblasts significantly rescued the reduced DAM number and DAM misorientation in caz mutants. Consistent with this genetic evidence for a role of FGF signaling in caz mutant muscle phenotypes, immunostaining revealed that Stumps levels are significantly reduced in caz mutant founder myoblasts. Together, these data indicate that defective FGF signaling contributes to caz mutant muscle phenotypes.

Mallik et al. recently reported that neuronal dysfunction in caz loss-of-function mutants is mediated by increased expression of Xrp1 [10], a DNA-binding protein involved in protection against genotoxic stress [17], DNA damage repair [18], cell competition [19, 20] and intra- and inter-organ growth coordination [21]. Increased Xrp1 expression in caz mutant central nervous system led to gene expression dysregulation and neuronal dysfunction [10]. In this study, we show that increased Xrp1 levels also mediate the newly-discovered caz mutant muscle phenotypes, as heterozygosity for Xrp1 almost completely rescued caz mutant muscle phenotypes. Interestingly, selective overexpression of Xrp1 in adult myoblasts of otherwise WT animals was sufficient to induce a substantial reduction in the number of founder myoblasts and DAMs, and DAM misorientation. A subtle mutation in the Xrp1 AT-hook motif that abolishes its DNA-binding capacity completely prevented the induction of muscle phenotypes upon selective overexpression in adult myoblasts. This indicates that gene expression dysregulation as a consequence of increased Xrp1 levels mediates caz mutant muscle phenotypes. Intriguingly, in adult founder myoblasts, Stumps is a downstream target of Xrp1, as selective overexpression of Xrp1 in adult myoblasts of otherwise WT animals significantly decreased Stumps expression in founder myoblasts, as is the case in caz mutants. As Stumps expression becomes restricted to adult founder myoblasts at the time of founder myoblast selection (24 h APF) [6], suppression of Stumps by Xrp1 is likely a key molecular mechanism underlying the reduced adult founder myoblast number in caz mutants. Altogether, our data indicate that in adult founder myoblasts, loss of caz function leads to increased Xrp1 expression levels, which in turn suppresses Stumps expression, leading to a reduced number of founder myoblasts and consequently to a reduced DAM number.

Although our data indicate a role of caz in founder myoblast selection by controlling Xrp1 expression, which in turn regulates Stumps expression (and possibly other FGF pathway components), additional mechanisms may contribute to caz mutant muscle phenotypes. It is unlikely that defective myoblast fusion contributes to caz mutant muscle phenotypes, as the number of nuclei per DAM was not substantially changed in caz mutants at 96 h APF (S5 Fig). However, other processes that occur later during adult muscle development may be defective in caz mutants, including myoblast migration, attachment to tendons, or muscle degeneration may occur. Such mechanisms may underlie the frequent misorientation of DAMs that we observed in caz mutants, the mechanism of which remains to be elucidated.

Consistent with the previously recognized importance of caz function in neurons [9, 15], loss of caz function in neurons also contributed to caz mutant muscle phenotypes, despite the fact that the number of motor neuron cell bodies was not altered and that the remaining muscles in caz mutant animals were normally innervated. Indeed, neuron-selective inactivation of caz in an otherwise WT genetic background was sufficient to induce a reduced founder myoblast and DAM number, as well as DAM misorientation. Furthermore, neuron-selective reintroduction of Caz in a caz mutant background partially rescued the reduced DAM number and DAM misorientation. The neuronal contribution to caz mutant muscle phenotypes is also mediated by increased Xrp1 expression, as neuron-selective knock-down of Xrp1 partially rescued caz mutant muscle phenotypes. Thus, gene expression dysregulation in caz mutant neurons likely explains the non-cell-autonomous contribution to muscle phenotypes. Future studies are needed to identify the Xrp1 target genes in neurons that are responsible for caz mutant muscle defects. In line with the neuronal contribution to caz mutant muscle phenotypes, the development of adult muscles is known to be intricately linked to the development of adult innervation. Early during metamorphosis, adult myoblasts migrate along motor nerves [1] and denervation of adult abdominal hemisegments at the onset of metamorphosis results in thinner and less numerous DAMs [22]. Furthermore, innervation is absolutely required for the formation of the male specific muscles [22, 23].

As mutations in FUS cause aggressive familial forms of the motor neurodegenerative disease ALS [8, 24], our findings may be interesting in the context of ALS. Indeed, a possible contribution of the skeletal muscle to ALS pathogenesis has been suggested, but is controversial. On the one hand, selective expression of ALS-mutant SOD1 in skeletal muscle of mice resulted in progressive muscle atrophy, reduced muscle strength, alteration in the contractile apparatus, mitochondrial dysfunction, paresis and motor defects [25, 26]. On the other hand, selective reduction of mutant SOD1 expression in skeletal muscle of mutant SOD1 transgenic mice did not affect their disease course, neither did increasing muscle fiber number and diameter by follistatin expression [27, 28]. Intriguingly, a recent study reported that ALS-mutant FUS is intrinsically toxic to both motor neurons and muscle cells, and toxicity in muscle may be attributable to defective FUS-mediated transcriptional regulation of acetylcholine receptor subunit genes [29]. While this and other studies in FUS-ALS mouse models indicate that a gain-of-toxic-function of ALS-mutant FUS is required to cause motor neuron degeneration [30-32], a possible contribution of loss of nuclear FUS function to FUS-ALS pathogenesis cannot be excluded. Despite the fact that caz mutant muscle defects are developmental, our findings are consistent with a cell-autonomous function of the Drosophila FUS orthologue in muscle, and a possible role of the skeletal muscle in FUS-ALS pathogenesis.

Interestingly, FGF signaling has previously been implicated in ALS pathogenesis, through retrograde FGF signaling from muscle to motor neuron [33]. Beyond ALS, FGF signaling has also been implicated in spinal muscular atrophy (SMA) [34], a related motor neurodegenerative disease caused by loss of SMN1 function [35]. FUS also directly binds to SMN and associates with SMN complexes [36-39], and FUS knock-out or cytoplasmic mislocalization of ALS-mutant FUS induces cytoplasmic SMN mislocalization and a reduced number of nuclear SMN-containing Gems [31, 36, 38, 39]. Thus, our finding that defective FGF signaling contributes to caz mutant muscle phenotypes is consistent with dysregulation of the FGF pathway as a contributing factor to both FUS-ALS and SMA, further indicating that common molecular pathways may underlie FUS-ALS and SMA.

Materials and methods

Drosophila genetics

Flies were kept in a temperature-controlled incubator with 12 h on/off light cycle at 25°C. The X chromosome-inserted elav-GAL4 (Bloomington Drosophila Stock center (BDSC) stock number 458) was used for pan-neuronal expression of UAS transgenes. 1151-GAL4 was used for targeted expression in adult myoblast precursors [40] (kindly provided by K. VijayRaghavan). For motor neuron quantification, OK371-GAL4 was used to drive UAS-RedStinger. For in vivo imaging, elav-GAL4 and 1151-GAL4 drivers were recombined with caz, and recombinant females were crossed to Mhc-GFP, his-RFP/CyO [11-13] in order to generate an introgressed stock. To induce cell-type-specific caz inactivation, caz [9] was recombined with elav-GAL4 and 1151-GAL4; homozygous females were crossed to UAS-FLP (BDSC stock 4539) males and male offspring was used for the experiments. In order to evaluate the number of founder myoblasts at 24 h APF, 1151-GAL4 // caz // caz,1151-GAL4 // caz,1151-GAL4 // caz,elav-GAL4 were recombined with duf-lacZ [41]. The UAS-caz line used for rescue experiments was the UAS-flag-caz line described in [15]. The htl line used in this study was UAS-htl.ORF.3xHA (F000798; FlyORF). The UAS-stumps transgenic line was provided by M. Leptin [16]. The caz and caz lines [9], as well as Xrp1 and 5xUAS-Xrp1 lines were described [10] and the UAS-Xrp1-RNAi line used in this study was P[TRiP.HMS00053]attP2 (BDSC 34521).

In vivo imaging of dorsal abdominal muscles

To image dorsal abdominal muscles, appropriate crosses were set up, male larvae were selected and placed in a new vial containing standard Drosophila food supplemented with yeast paste. White prepupae were collected on wet Whatman filter paper in petri dishes and kept at 25°C for 96 h. After 96 h, pharate adults were dissected out from the pupal case and placed in an approximately 2 mm furrow of a plastic slide with a drop of 60% glycerol. Images of serial optical sections were acquired using a Zeiss LSM700 or a Leica SP8 laser scanning confocal microscope. For quantification, DAMs were manually counted using ImageJ/Fiji [42] by inspecting the acquired Z-stacks.

Histone-RFP (his-RFP) expression was used to quantify the number of nuclei that are present in the DAMs in abdominal segments A3 and A4.

Motor neuron quantification

The number of motor neuron cell bodies were quantified in third instar larval ventral nerve cords (VNC) of caz and WT controls. Motor neuron cell bodies were visualized by OK371-GAL4-driven UAS-RedStinger and the number of motor neuron cell bodies was counted in the central clusters in segments A2-A7 of the 3rd instar larval VNC.

Immunostainings

Immunostaining of pharate adults (96 h APF)

Open book preparations of pupal abdomens were fixed in 4% PFA in PBS for 30 minutes. Tissues were washed 3 x 10 min with PBS/0.2% Triton X-100 and blocked for 1 h at RT with 10% goat serum in PBS. Tissues were incubated overnight at 4°C with primary antibodies against Caz (mouse monoclonal clone 3F4; 1:30; [43]) or Dlg1 (mouse monoclonal clone 4F3; 1:200; Developmental Studies Hybridoma Bank). Secondary antibodies conjugated either with Alexa Fluor 488 or Alexa Fluor 568 (1:500; Molecular Probes), anti-HRP-549 (1:1000; Jackson ImmunoResearch) and Phalloidin-647 (1:20; Cell Signaling) were applied for 3 h at RT.

For quantification of neuromuscular endplate area we used ImageJ/Fiji in order to measure the endplate and muscle area in DAMs of abdominal segment A3 and A4, from maximum intensity projections of acquired confocal z-stacks. In case of multiple branches, the sum of the areas of individual branches was used as total endplate area. The endplate areas were then normalized to the respective muscle area.

Immunostaining of 24 h APF pupae

Open book preparations of abdomens were fixed in 4% PFA in PBS for 60 minutes. Tissues were washed 3 x 10 min with PBS/0.2% Triton X-100 and blocked for 1 h at RT in 10% goat serum in PBS.

For quantification of founder myoblasts and Caz immunostaining, tissues were incubated overnight at 4°C with the first primary antibody against β-galactosidase (mouse monoclonal 40-1a; 1:50; Developmental Studies Hybridoma Bank). Anti-mouse secondary antibody conjugated with Alexa Fluor 488 (1:500; Molecular Probes) was applied at RT for 1 h. Tissues were then washed 3 x 10 min in PBS/0.2% Triton X-100 and incubated overnight at 4°C with the second primary antibody against Caz or Futsch (mouse monoclonal clone 22C10; 1:100; Developmental Studies Hybridoma Bank). Anti-mouse secondary antibody conjugated with Alexa Fluor 568 (1:250; Molecular Probes) was applied at RT for 30 min. For imaging a Zeiss LSM700 or a Leica SP8 laser scanning confocal microscope was used.

For quantification of Stumps level, tissues were incubated overnight at 4°C with primary antibodies against β-galactosidase (mouse monoclonal 40-1a; 1:50; Developmental Studies Hybridoma Bank) and Stumps, a gift from M. Leptin (rabbit; 1:1000) [16]. Secondary antibodies conjugated either with Alexa Fluor 488 or Alexa Fluor 568 (1:500; Molecular Probes) were applied at RT for 1 h. As internal control, TetraSpeck Microspheres 4.0 μm (ThermoFisher) were applied on the slides while mounting the tissues. Images were acquired using a Leica SP8 laser scanning confocal microscope with a 40x/1.3 NA oil objective with the exact same settings for all the samples. Maximum intensity projections of acquired z-stacks were generated with ImageJ/Fiji. The mean cytoplasmic intensity was measured in these projections using a custom-made ImageJ/Fiji macro, upon thresholding and removing all background pixels’ intensities. Same approach was used for the quantification of beads’ intensity. For both Stumps staining and internal control, the mean intensity per pupa was calculated and the cytoplasmic Stumps staining intensity was normalized to the beads intensity.

Larval muscle quantification

To visualize larval muscles, elav-GAL4; Mhc-GFP and caz; Mhc-GFP larvae were dissected and briefly fixed in Bouin’s solution for 3 min, followed by 3 x 15-min washes with PBS/0.2% Triton X-100 and three 15-min washes in PBS, and mounted in Vectashield (Vector Laboratories). Samples were imaged using a Leica SP8 laser scanning confocal microscope. For quantification, larval muscles in A3 and A4 were counted using ImageJ/Fiji.

Statistical analysis

All results from analysis are presented as mean ± standard error of the mean (SEM) and differences were considered significant when p <0.05. Before analysis, a Robust regression and Outlier removal method (ROUT) was performed to detect all outliers. This nonlinear regression method fits a curve that is not influenced by outliers. Residuals of this fit were then analyzed by a test adapted from the False Discovery Rate approach, to identify any outliers. This analysis did not identify outliers, and therefore no data points were removed. Normality and homoscedasticity of all data was analyzed by a Shapiro–Wilk and Brown–Forsythe (F-test for t-tests) test, respectively. Subsequent statistical tests were only performed if all assumptions were met, except mild heteroscedasticity that was observed between experimental groups in Figs 3F, 5A, 5B, 6A, S3A and S3B(A3).

For comparison of normally distributed data of two groups, two-tailed unpaired Student’s t-test was used in combination with an F-test to confirm that the variances between groups were not significantly different. Non-parametric Mann–Whitney tests were performed if data were not normally distributed. An unpaired t-test with Welch’s correction was performed for data with heteroscedasticity.

Comparisons of data consisting of more than two groups, varying in a single factor, were performed using Kruskal–Wallis for not-normally distributed data with homogeneous variance, and using Brown-Forsythe with Welch’s correction for normally distributed data with heterogeneous variance, both using Dunn’s or Dunnett’s multiple comparisons post hoc test.

Comparisons of data consisting of more than two groups, varying in two factors, was performed using two-way ANOVA and subsequent Tukey’s multiple comparisons test for normally distributed data with homogeneous variance.

To analyze muscle misorientation, one sample two-tailed Wilcoxon signed rank test using the Pratt method was used to evaluate differences as compared to the control genotype (with a set value of 0).

Data were analyzed using GraphPad Prism v.8.0 and R (3.5.1).

Reduced DAM number in both age-matched and stage-matched caz pupae.

A,B, Quantification of DAM number in segments A3 and A4 of either age-matched (100 h APF, A) or stage-matched (P15: twitching legs, B) caz versus WT male pupal filets immunostained for actin. Unpaired t-test (A, B-A3) and t-test with Welch’s correction (B-A4); **p<0.01, ***p<0.001; n(A) = 4 WT versus 5 caz (A3), 5 WT versus 4 caz (A4); n(B) = 11 WT versus 10 caz (A3), 15 WT versus 8 caz (A4). Average ± SEM.

(TIF)

Click here for additional data file.

Loss of caz function does not affect larval muscle morphology.

A, A transgenic line in which the Myosin heavy chain gene is GFP-tagged (Mhc-GFP) allowed for visualization of muscles in abdominal segments A3 and A4 of WT (+/Y; Mhc-GFP,his-RFP/+) and caz (caz/Y; Mhc-GFP,his-RFP/+) third instar larvae. Scale bar: 100μm. B, Quantification of larval muscle number in segments A3 and A4. Mann-Whitney test; n = 6 per genotype. Average ± SEM.

(TIF)

Click here for additional data file.

DAM phenotypes induced by selective caz inactivation in adult myoblasts or neurons.

A, Representative images of DAMs in abdominal segments A3 and A4 of 96 h APF pupae in which caz was selectively inactivated in adult myoblasts (caz,1151-GAL4/Y; Mhc-GFP/UAS-FLP, right panel) as compared to the relevant control (caz,1151-GAL4/Y; Mhc-GFP/+, left panel). Scale bar: 100μm. B, Representative images of DAMs in abdominal segments A3 and A4 of 96 h APF pupae in which caz was selectively inactivated in neurons (caz,elav-GAL4/Y; Mhc-GFP/UAS-FLP, right panel) as compared to the relevant control (caz,elav-GAL4/Y; Mhc-GFP/+, left panel). Scale bar: 100μm.

(TIF)

Click here for additional data file.

Increased Xrp1 levels mediate caz mutant DAM misorientation.

A, Percentage of misoriented DAMs per animal in segments A3 and A4 of 96 h APF caz pupae that are heterozygous for Xrp1 (caz/Y; Mhc-GFP,his-RFP/+; Xrp1/+) as compared to the relevant control genotypes (+/Y; Mhc-GFP,his-RFP/+ // +/Y; Mhc-GFP,his-RFP/+; Xrp1/+ // caz/Y; Mhc-GFP,his-RFP/+). One sample Wilcoxon signed rank test to compare all genotypes to control, and caz to caz; Xrp1/+; ***p<0.0005; n = 15 per genotype. Average ± SEM. B, Percentage of misoriented DAMs per animal in segments A3 and A4 of 96 h APF male pupae that selectively overexpress Xrp1 in adult myoblasts (1151-GAL4), either as WT protein or with a subtle mutation that disrupts the DNA-binding capacity of the AT-hook motif (Mut), as compared to driver-only control. One sample Wilcoxon signed rank test to compare all genotypes to control and UAS-Xrp1 WT to UAS-Xrp1 Mut; **p<0.01, ***p<0.0001; n = 8 control, 23 UAS-Xrp1 WT, 20 UAS-Xrp1 Mut. Average ± SEM.

(TIF)

Click here for additional data file.

Loss of caz function does not preclude myoblast fusion.

The average number of nuclei per DAM was determined in abdominal segments A3 and A4 of 96 h APF WT, caz, and caz pupae that carried Mhc-GFP and his-RFP transgenes to visualize muscles and nuclei, respectively. Ordinary one-way ANOVA with Dunnett’s post test; *p<0.05; n = 7 per genotype. Average ± SEM.

(TIF)

Click here for additional data file.

Transfer Alert

This paper was transferred from another journal. As a result, its full editorial history (including decision letters, peer reviews and author responses) may not be present.

1 Nov 2019

Dear Dr Storkebaum,

Thank you very much for submitting your Research Article entitled 'The Drosophila FUS ortholog caz directs adult founder myoblast selection by Xrp1-dependent regulation of FGF signaling' to PLOS Genetics. Your manuscript was fully evaluated at the editorial level and by two independent peer reviewers. The reviewers thought that you had addressed an important question, but both raised some substantial concerns about what they consider as inconsistencies and overstatements that need to be amended. Based on the reviews, we will not be able to accept this version of the manuscript, but we would be willing to review again a much-revised version. We cannot, of course, promise publication at that time.

Should you decide to revise the manuscript for further consideration here, your revisions should address the specific points made by each reviewer. We will also require a detailed list of your responses to the review comments and a description of the changes you have made in the manuscript.

If you decide to revise the manuscript for further consideration at PLOS Genetics, please aim to resubmit within the next 60 days, unless it will take extra time to address the concerns of the reviewers, in which case we would appreciate an expected resubmission date by email to plosgenetics@plos.org.

If present, accompanying reviewer attachments are included with this email; please notify the journal office if any appear to be missing. They will also be available for download from the link below. You can use this link to log into the system when you are ready to submit a revised version, having first consulted our Submission Checklist.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see our guidelines.

Please be aware that our data availability policy requires that all numerical data underlying graphs or summary statistics are included with the submission, and you will need to provide this upon resubmission if not already present. In addition, we do not permit the inclusion of phrases such as "data not shown" or "unpublished results" in manuscripts. All points should be backed up by data provided with the submission.

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool.  PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org.

PLOS has incorporated Similarity Check, powered by iThenticate, into its journal-wide submission system in order to screen submitted content for originality before publication. Each PLOS journal undertakes screening on a proportion of submitted articles. You will be contacted if needed following the screening process.

To resubmit, use the link below and 'Revise Submission' in the 'Submissions Needing Revision' folder.

[LINK]

We are sorry that we cannot be more positive about your manuscript at this stage. Please do not hesitate to contact us if you have any concerns or questions.

Yours sincerely,

Claude Desplan

Associate Editor

PLOS Genetics

Gregory P. Copenhaver

Editor-in-Chief

PLOS Genetics

Reviewer's Responses to Questions

Comments to the Authors:

Please note here if the review is uploaded as an attachment.

Reviewer #1: The paper by Catinozzi et al. focuses on the contribution of the Drosophila FUS ortholog “Cabeza” (caz) to adult founder myoblast selection by Xrp-dependent regulation of FGF signaling. Caz appears to function in both motor neurons, as well as in dorsal abdominal muscles (DAMs) to promote proper DAM pattern and number. The authors investigated the role of cab in muscles, and discovered that similarly to the genetic connections described for cab in neurons, cab is acting upstream of the FGF receptor pathway and controls the levels of its downstream effector Stumps in the DAM founder cells. In cab mutants the levels of Stumps in these cells are lower resulting with reduced number of muscles, as well as with DAMs exhibiting aberrant pattern. Overexpression of Stumps or the FGF receptor Htl in myoblasts partially rescued these phenotype. Further experiments indicated that the caz phenotype is induced due to overexpression of Xrp1, a DNA binding protein involved in gene expression regulation, and that rescue of caz can be obtained by combining caz mutants with heterozygous Xrp1 alleles. In addition, a similar DAM phenotype (observed in caz) can be obtained by overexpression of Xrp1 in a wild type myoblasts, supporting the idea that caz represses Xrp1 levels. These results are interesting, the genetic experiments are very convincing and the images and statistics are appropriate.

1. A major concern lies in the conclusion of the authors regarding the specific activity of caz in muscle founder cells. FGF signaling is presumably required not only in founder cells but also in the myoblasts, for their migration, guidance, fusion and attachment. The broad expression of caz, and in particular the fact that the authors used for all their rescue experiments a GAL4 driver which is expressed in all myoblasts (and not a founder cell specific Gal4 driver) do not support a specific function of caz in founder cells selection, but rather in various FGF-dependent activities in the myoblasts.

2. Fig 2 – how the authors prove that the innervation pattern is normal? In my view the innervation pattern is different from that of wild type. It might be due to the aberrant muscle pattern, but this should be better analyzed and quantified.

3. Fig 3B -Duf-lacZ might persist longer than the actual protein, and it appears that the image in Fig 3B shows myotubes rather than founder cells.

4. Discussion – is too lengthy. I think the last part discussing speculations about ALS and SMA should be shorten.

Reviewer #2: The manuscript by Catinozzi et al presents functional analysis of Drosophila FUS ortholog caz during adult muscle development. Authors demonstrate that caz acts via Xrp1 and regulates FGF signaling known to trigger adult founder myoblast specification. They analyse, in different genetic contexts, number of Duf-positive dorsal adult founders and number of derived from these founders dorsal abdominal muscles (DAMs). They conclude that caz influences DAMs development both cell autonomously and via its role in motor neurons. Authors draw then parallels between their findings and contribution of muscle defects to pathogenesis of ALS and SMA neurodegenerative diseases in which FUS and FGF are involved. Overall, this study identifies caz as a new player in adult Drosophila myogenesis and points to its role in coordinated development of adult muscles and motor neurons.

It is a valuable study but it carries are several inconsistencies and overstatements that need to be amended.

1. Authors need to clarify what are the genotypes tested in the case GAL4 drivers are used (Figs. 3, 5, 6). For example in Fig. 3D is « control » corresponding to 1151-GAL4 homozygous or to 1151-GAL4/+ ? Full genotype descriptions for all GAL4-involving contexts is required.

2. To determine contribution of adult myoblasts versus neurons to casKO phenotypes authors have to use the same genetic contexts in rescue experiments. In Fig. 3D casKO alone reduces DAM number to less than 10 wereas in Fig. 4C casKO in the context of UAS-cas leads to a milder reduction of DAM number (more than 10). This makes rescue experiments incompatible for comparison. It also indicates that UAS-cas transgene could be leaky….. Did authors test this ?

3. Number of misoriented DAMs par animal in casKO is about 10% in Fig. 1E but more than 50% in Fig.3E – this needs to be clarified !

4. Representative views of phenotypes for conditional cazKO in muscles and in motor neurons need to be shown (at least in a supplementary fig).

5. Is conditional caz knockout in motor neurons affecting number of founders ? This experiment could help in defining wherther caz expressed in motor neurons influences specification of adult founders

6. Measurement of intensity of stumps signal is not convincing. (Fig. 5C,D) Intensity of stumps signal needs to be referenced to an internal control signal that is not changing in casKO context and ideally also to the background signal. In Fig. 5C in casKO duf-lacZ seems reduced compared to wt. It is thus not a good reference.

Authors have to provide more details about how measurements of intensity was performed using ImageJ.

7. In discussion authors claim :

« ….. our data indicate that caz mutant adult muscle developmental defects are to a major extent attributable to loss of caz function in adult myoblasts. Indeed, similar to full-body caz LOF mutants, adult myoblast-selective inactivation of caz resulted in a reduced number of DAMs, as well as their misorientation…. »

This appears to be overstatement considering that conditional casKO in muscle precursors (Fig.3F) leads to loss of DAMs that is similar to that of conditional casKO in motor neurons (Fig.4A).

8. « directs » in title is not appropriate as it suggests that without caz founders are not selected. It could be replace by « promotes »

Minor points :

1.Misorientation of some of DAMs, as discussed by authors, could indicate affected capacity to interact with tendon cells, another role of caz. Testing expression of betaPS Integrin could help in defining whether misoriented DAMs keep capacities to attach.

2. Did authors test cazRNAi knockdowns with different drivers, for example in flight muscles?

**********

Have all data underlying the figures and results presented in the manuscript been provided?

Large-scale datasets should be made available via a public repository as described in the PLOS Genetics data availability policy, and numerical data that underlies graphs or summary statistics should be provided in spreadsheet form as supporting information.

Reviewer #1: Yes

Reviewer #2: Yes

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

20 Mar 2020

Dear Dr Storkebaum,

We are pleased to inform you that your manuscript entitled "The Drosophila FUS ortholog cabeza promotes adult founder myoblast selection by Xrp1-dependent regulation of FGF signaling" has been editorially accepted for publication in PLOS Genetics. Congratulations!

Before your submission can be formally accepted and sent to production you will need to complete our formatting changes, which you will receive in a follow up email. Please be aware that it may take several days for you to receive this email; during this time no action is required by you. Please note: the accept date on your published article will reflect the date of this provisional accept, but your manuscript will not be scheduled for publication until the required changes have been made.

Once your paper is formally accepted, an uncorrected proof of your manuscript will be published online ahead of the final version, unless you’ve already opted out via the online submission form. If, for any reason, you do not want an earlier version of your manuscript published online or are unsure if you have already indicated as such, please let the journal staff know immediately at plosgenetics@plos.org.

In the meantime, please log into Editorial Manager at https://www.editorialmanager.com/pgenetics/, click the "Update My Information" link at the top of the page, and update your user information to ensure an efficient production and billing process. Note that PLOS requires an ORCID iD for all corresponding authors. Therefore, please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field.  This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager.

If you have a press-related query, or would like to know about one way to make your underlying data available (as you will be aware, this is required for publication), please see the end of this email. If your institution or institutions have a press office, please notify them about your upcoming article at this point, to enable them to help maximise its impact. Inform journal staff as soon as possible if you are preparing a press release for your article and need a publication date.

Thank you again for supporting open-access publishing; we are looking forward to publishing your work in PLOS Genetics!

Yours sincerely,

Claude Desplan

Associate Editor

PLOS Genetics

Gregory P. Copenhaver

Editor-in-Chief

PLOS Genetics

www.plosgenetics.org

Twitter: @PLOSGenetics

----------------------------------------------------

Comments from the reviewers (if applicable):

Reviewer's Responses to Questions

Comments to the Authors:

Please note here if the review is uploaded as an attachment.

Reviewer #1: The authors fully addressed my comments.

Reviewer #2: Authors made several amendments and revisions in the text and in the figures, which clarified all points raised in my comments. The current version of the manuscript appears appropriate for publication in PLoS Genet

**********

Have all data underlying the figures and results presented in the manuscript been provided?

Large-scale datasets should be made available via a public repository as described in the

PLOS Genetics

data availability policy, and numerical data that underlies graphs or summary statistics should be provided in spreadsheet form as supporting information.

Reviewer #1: None

Reviewer #2: Yes

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review?

For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

----------------------------------------------------

Data Deposition

If you have submitted a Research Article or Front Matter that has associated data that are not suitable for deposition in a subject-specific public repository (such as GenBank or ArrayExpress), one way to make that data available is to deposit it in the Dryad Digital Repository. As you may recall, we ask all authors to agree to make data available; this is one way to achieve that. A full list of recommended repositories can be found on our website.

The following link will take you to the Dryad record for your article, so you won't have to re‐enter its bibliographic information, and can upload your files directly:

http://datadryad.org/submit?journalID=pgenetics&manu=PGENETICS-D-19-01711R1

More information about depositing data in Dryad is available at http://www.datadryad.org/depositing. If you experience any difficulties in submitting your data, please contact help@datadryad.org for support.

Additionally, please be aware that our data availability policy requires that all numerical data underlying display items are included with the submission, and you will need to provide this before we can formally accept your manuscript, if not already present.

----------------------------------------------------

Press Queries

If you or your institution will be preparing press materials for this manuscript, or if you need to know your paper's publication date for media purposes, please inform the journal staff as soon as possible so that your submission can be scheduled accordingly. Your manuscript will remain under a strict press embargo until the publication date and time. This means an early version of your manuscript will not be published ahead of your final version. PLOS Genetics may also choose to issue a press release for your article. If there's anything the journal should know or you'd like more information, please get in touch via plosgenetics@plos.org.