A previous study isolated a collection of maternal mutants defective in blastoderm cellularization and/or germ-band extension (Pimenta-Marques, 2008). Complementation group 7 contained two different mutant alleles with similar defects in blastoderm cellularization. Through deficiency mapping and a candidate gene approach we concluded that both were allelic to the uncharacterized coding gene CG6197 (Flybase). To confirm the mutants’ identity, their zygotic lethality, female sterility (germ-line clones), and blastoderm cellularization defects (maternal mutant embryos) were rescued using a genomic fragment construct that contained a wild-type copy of CG6197. Both isolated alleles of CG6197 showed identical phenotypes: maternal mutant embryos (hereafter referred to as mutant embryos) showed normal syncytial nuclear divisions but subsequently failed to elongate the cortical nuclei, which became mislocalized during blastoderm cellularization. The blastoderm cellularization phenotype was remarkably similar to that described for kugelkern/charleston mutant embryos. Based on the observed phenotypes, the corresponding gene fandango was named, after the Iberian folk dance (Guilgur, 2014).

fandango encodes the Drosophila ortholog of yeast SYF1 (synthetic lethal with cdc41) and human XAB2 (XPA binding protein 2). These proteins were described as subunits of the NTC/Prp19 complexes, which are important for spliceosome stabilization and activation (Chan, 2003; Chang, 2009; Hogg, 2010). Fandango protein has multiple tetratricopeptide repeat (TPR) motifs, which is a protein–protein interaction module. Sequencing both alleles of fandango (fand¹ and fand2) revealed distinct mutations within the fandango open reading frame (ORF). fand¹ contained a missense point mutation in a highly conserved residue within TPR domain VII (from an alanine to a valine; A401V), whereas fand² contained a microdeletion of 18 nucleotides within TPR domain VI, which deleted six conserved amino acids from position 355 to 360. In total protein extracts, both fand¹ and fand² mutant embryos showed a significant reduction in Fandango protein levels compared to control. fandango mRNA levels, analyzed by real-time qPCR, were similar between control and fand¹ mutant embryos, suggesting that the mutation did not alter the stability of the encoding pre-mRNA (Guilgur, 2014).

As noted above fandango maternal mutant embryos and kugelkern (kuk) mutant embryos showed remarkably similar blastoderm cellularization defects. Since fandango encodes a protein whose yeast and human orthologs are required for efficient spliceosome activity, it was hypothesized that Fandango was required for splicing of kuk transcripts. kuk encodes two different transcripts, which vary in intron size. Both transcripts are predicted to encode the same protein. Analysis of publicly available modENCODE transcriptome datasets suggested that the large kuk transcript was maternally expressed, whereas the small kuk transcript was only expressed zygotically. Through RT-PCR analysis we confirmed that similarly to control maternal genes (nanos and oskar) the large kuk transcript was maternally expressed (being present in unfertilized eggs), whereas the small kuk transcript was exclusively zygotically expressed (being present only in fertilized eggs) as the case of well-known early zygotic genes (even-skipped and Krüppel) (Guilgur, 2014).

To investigate by RT-PCR whether Fandango was required for splicing of kuk pre-mRNAs, specific sets of primers (exon–exon, e–e; intron–exon, i–e) were designed for each kuk transcript, taking advantage of a longer 3'UTR in the small kuk transcript. Surprisingly, whereas fandango embryos showed significant splicing defects of the small zygotic kuk transcript, the large maternal kuk transcript was correctly spliced. Splicing defects were fully rescued by a genomic fragment construct that contained a wild type copy of fandango. The differential requirement of Fandango for splicing of kuk transcripts prompted an investigation of more than 20 other maternal and early zygotic genes. RT-PCR analysis of fandango embryos invariably showed splicing defects of early zygotic but not maternally encoded transcripts. High-throughput transcriptome sequencing (RNAseq) confirmed that splicing of early zygotic but not maternally encoded gene products was affected in fandango embryos. Maternal transcripts, whose intron size was equivalent to those observed in early zygotic transcripts, were unaffected, which showed that Fandango was not specifically rate limiting for splicing of small introns. Comparison analysis of 5' and 3' splice site consensus sequences between maternal and zygotic pre-mRNA transcripts showed no significant differences and the two populations of transcripts displayed a similarly heterogeneous exon–intron structure. RT-PCR analysis of maternally encoded transcripts from wild-type and fandango mutant ovaries (germ-line clones) also failed to detect splicing defects. This suggested that the absence of splicing defects of maternally encoded transcripts in fandango embryos was not due to specific degradation of unspliced transcripts during oogenesis (Guilgur, 2014).

The differential requirement of Fandango for splicing of early zygotic encoded transcripts is fully consistent with the observation that maternally controlled oogenesis, primordial germ-cell formation, and syncytial nuclear divisions were normal in fandango mutants, whereas the first detectable phenotype only occurred during zygotically controlled blastoderm cellularization. Despite the fact that clonal analysis of the female germ line for both alleles of fandango showed normal oogenesis and egg laying, Fandango protein levels were significantly reduced in the mutant ovaries (germ-line clones). fandango embryos also failed to initiate germ-band extension after blastoderm cellularization. It was previously shown that anterior–posterior (A–P) patterning is required for germ-band extension. Consistently, fandango embryos showed A–P patterning defects in the early zygotic pair-rule gene even-skipped (Guilgur, 2014).

Endogenous Fandango and Prp19 physically interacted in the early embryo. Moreover, both endogenous Fandango and Prp19 physically interacted with endogenous ISY1 and CDC5L, confirming that Fandango is a bona fide subunit of Drosophila NTC/Prp19 complexes. Immunoprecipitation of Myc-tagged Fandango and Myc-tagged Prp19 from embryonic protein extracts also identified an identical group of interacting proteins. Whereas Myc-Fandango mostly interacted with the NTC/Prp19-related complex subunits, Myc-Prp19 interacted principally with the NTC/Prp19 complex subunits. This illustrated that, as in humans, distinct but interacting NTC/Prp19 complexes exist in Drosophila, in agreement with the recent suggestion that a remarkable degree of conservation of distinct splicing complexes exists among metazoans (Guilgur, 2014).

The differential requirements of Fandango for pre-mRNA splicing of maternal and early zygotic transcripts potentially suggest distinct interactions between Fandango and other splicing proteins during oogenesis and early embryonic development. Nevertheless, immunoprecipitation of Myc-Fandango specifically expressed in the female germ line during oogenesis and in the early embryo identified a virtually identical group of interacting proteins: mostly subunits of the NTC/Prp19-related complex, and to a lesser extent, subunits of the NTC/Prp19 complex. These results showed that Fandango physically interacts with a similar group of splicing proteins during oogenesis and in the early embryo (Guilgur, 2014).

To better understand the splicing defects observed in fandango embryos, whether the integrity of NTC/Prp19 complexes was affected in this mutant was examined. Size-exclusion chromatography showed detectable changes in the integrity of NTC/Prp19 complexes in fandango embryos, with a significant reduction in the levels of ISY1 protein. ISY1 is a NTC/Prp19-related complex subunit. The loss of integrity of the ISY1-positive ∼600–800 kDa NTC/Prp19 complex and concomitant reduction in the stability of some of their subunits, most likely impaired efficient activation of the spliceosome and were likely explanations for the splicing defects observed in fandango embryos. In agreement with the suboptimal spliceosome activation hypothesis, intron retention was the main splicing defect of early zygotic transcripts in fandango embryos (Guilgur, 2014).

Levels of ISY1 were similarly affected in fandango mutants during oogenesis and in the early embryo (Figure 3C), suggesting this decrease did not explain the differential requirements of Fandango for splicing of early zygotic and maternally encoded transcripts. Mutant clonal analysis of a stronger allele of fandango (nonsense mutation), showed a complete loss of the female germ line in adult ovaries. This demonstrated that the two isolated alleles of fandango are hypomorphic and suggested that Fandango was required, albeit at lower levels, for splicing of maternal transcripts. It is concluded it is unlikely that a differential expression and/or association of core components of the spliceosome could potentially explain the differential requirements for Fandango between oogenesis and the early embryo. The most likely explanation is that Fandango is quantitatively (but not qualitatively) differentially required during early embryonic development (Guilgur, 2014).

Transcriptional elongation can affect co-transcriptional splicing. It was recently shown that Syf1, the yeast ortholog of Fandango, is also important for RNApol II transcriptional activity, therefore it was decided to investigate transcription in fandango embryos. Three intronless early zygotic genes (nullo, snail, and scute) and two early zygotic genes with introns (even-skipped and tailless) were selected for further analysis by real-time qPCR. During mid/late-syncytial blastoderm, no significant differences in transcript abundance were observed between control and fandango, whereas embryos mutant for grapes showed the expected reduction of transcript levels (Guilgur, 2014).

During transcriptional elongation, RNApol II is specifically phosphorylated on the Ser2 residue of its carboxy-terminal domain (CTD). In agreement with the onset of early zygotic transcription, a significant increase was observed in RNApol II CTD Ser2 phosphorylation as the embryo developed from early/mid-syncytial blastoderm (stage A), into mid/late-syncytial blastoderm (stage B), and blastoderm cellularization (interphase 14) (stage C). Both control and fandango embryos showed a similar increase in global levels of RNApol II CTD Ser2 phosphorylation . As transcriptional regulation during interphase 14 (stage C) relies on correct expression of early zygotic genes and degradation of many maternal RNAs (MZT), it is concluded that transcriptional changes at this stage were most likely a consequence of the widespread defects occurring during mid/late-syncytial blastoderm. Altogether, it is concluded that the observed reduction in Fandango levels affects mainly its splicing function (Guilgur, 2014).

To investigate if the differential requirement of Fandango for splicing of early zygotic and maternally encoded transcripts potentially resulted from distinct transcript sequences, an early zygotic kuk transcript (kuk-LacZ) was generated under the control of an UAS/Gal4 inducible promoter, where the open reading frame (ORF) was replaced by LacZ . As expected, when this construct was expressed zygotically, it was correctly spliced in control but not in fandango embryos. In contrast, splicing of the kuk-LacZ construct occurred normally in both control and fandango mutants when it was expressed maternally. Since maternal expression of an early zygotic transcript, in a fandango mutant background, was enough to suppress its splicing defects, it is concluded that the differential requirement of Fandango for splicing of early zygotic transcripts was most likely due to the developmental context of gene expression and not a particularity in the early zygotic pre-mRNA sequences. Consistently, no differences were detected related to intron size, splice sites consensus, and exon–intron structure between maternal and zygotic transcripts (Guilgur, 2014).

fandango mutants showed a significant reduction in Fandango and ISY1 protein levels, which most likely impaired efficient activation of the spliceosome. Since mitosis inhibits splicing, pre-mRNA splicing of early zygotic transcripts needs to be highly efficient for these genes to be correctly expressed. This suggests the existence of a developmental pre-requisite for highly efficient splicing, so that a suboptimal activation of the spliceosome would specifically impair pre-mRNA splicing of early zygotic but not maternal transcripts. Wild-type embryos already showed a detectable amount of intron retention in early zygotic transcripts, which was dramatically exacerbated in fandango embryos (Guilgur, 2014).

It was hypothesized that regardless of transcript size, there was also a constraint on pre-mRNA splicing of early zygotic transcripts in wild-type embryos. A gene was generated where the 5'UTR sequence including the intron of the small zygotic kuk transcript was quadruplicate to test this hypothesis. Quadruplicate introns were linked by in-frame LacZ coding sequences, and the entire construct (4x intron kuk-LacZ) was under the control either of an endogenous early zygotic minimal promoter (nullo-4x intron kuk-LacZ, ∼2.5 Kb) or an inducible UAS/Gal4 promoter (UAS-4x intron kuk-LacZ, ∼2.5 Kb). The total size of the encoded pre-mRNAs was comparable to many other endogenous early zygotic genes (e.g., kugelkern, runt, kruppel). As a control, point mutations were introduced in the splice sites of these constructs to generate comparable intronless transcripts (no intron kuk-LacZ) (Guilgur, 2014).

Only the first intron (int1) of the 4x intron kuk-LacZ construct was correctly spliced when it was zygotically expressed in wild-type embryos under the control of an endogenous early zygotic minimal promoter. Likewise, when the 4x intron kuk-LacZ construct was early zygotically expressed under the control of the inducible promoter UAS/Gal4 there were similar splicing defects (intron retention) (Guilgur, 2014).

Measurement of in vivo kinetics of mRNA splicing showed that half-lives for splicing reactions are <1 min for the first intron, but 2–8 min for both second and third introns. Hence, splicing of two or more introns requires more time than transcription and becomes rate limiting. Consistent with the hypothesis of a temporal constraint on pre-mRNA splicing, when the 4x intron kuk-LacZ construct was zygotically expressed, the splicing defects of the firstly transcribed 5'-localized introns (Int1 and Int2) were significantly weaker than those observed in the later transcribed 3'-localized introns (Int3 and Int4). Importantly, maternal expression of this construct was sufficient to significantly suppress its splicing defects. Real-time qPCR analysis showed that these constructs were equivalently zygotically and maternally expressed. This suggested that splicing did not quantitatively impair early zygotic transcription, which was consistent with the observation that the rates of transcriptional elongation proceed independently of splicing (Guilgur, 2014).

This study showed that in wild-type embryos pre-mRNA splicing imposed significant constraints on early zygotic expression, which is a likely explanation why most early zygotic genes are intronless. Although a moderate decrease in the length of syncytial blastoderm interphases (seen in grapes mutant embryos was not sufficient to induce splicing defects in otherwise wild-type embryos, it was hypothesized that avoidance of pre-mRNA splicing during early zygotic expression is a consequence of the extremely short interphases and frequent mitotic cycles. Similarly to Drosophila, mosquito Aedes aegypti and the zebrafish Danio rerio early zygotic transcripts are frequently intronless when compared with the rest of the transcriptome. This suggests that highly proliferative tissues need coordination between cell cycle and gene architecture for correct gene expression and avoidance of abnormally processed transcripts (Guilgur, 2014).

The results highlight cell cycle constraints during early embryonic development as a force capable of driving changes in gene architecture of multicellular organisms. In unicellular organisms intron paucity correlates with a bias toward the 5' ends, whereas introns from multicellular genomes are evenly distributed throughout the genes. This suggests that similar constraints on gene architecture are also likely to exist in yeast and other fast-dividing single cell eukaryotes (Guilgur, 2014).

The way splicing efficiency might be regulated through changes in constitutive spliceosome factors and how it might influence differential gene expression is a new area of interest. This study has presented experimental evidence supporting the hypothesis of a requirement for highly efficient pre-mRNA splicing during early embryonic development. Since the NTC/Prp19 complexes are known to be important for efficient spliceosome activation, and the mutant alleles specifically impaired pre-mRNA splicing of early zygotic but not maternally encoded transcripts, it is proposed that overall requirements for splicing efficiency are likely to vary during development, being the NTC/Prp19 complexes a key modulator of spliceosome activation rates. In agreement with this hypothesis, Prp19 expression varies during neuronal differentiation (Guilgur, 2014).

In plants it was recently shown that removal of retained introns regulates translation in rapidly developing gametophytes. In Drosophila, a sub-population of early zygotic transcripts with introns similarly showed some degree of intron retention in wild-type embryos. The current results also suggest that the pre-requisite for highly efficient splicing during early embryonic development is paradoxically also likely to play an important regulatory role in the expression of a subset of early zygotic transcripts, which further supports the possibility that modulation of spliceosome activation per se is important for differential regulation of gene expression during development (Guilgur, 2014).

Morphogenesis requires the dynamic regulation of gene expression, including transcription, mRNA maturation and translation. Dysfunction of the general mRNA splicing machinery can cause surprisingly specific cellular phenotypes, but the basis for these effects is not clear. This study shows that the Drosophila faint sausage (fas; renamed fandango) locus, implicated in epithelial morphogenesis and previously reported to encode a secreted immunoglobulin domain protein, in fact encodes a subunit of the spliceosome-activating Prp19 complex, which is essential for efficient pre-mRNA splicing. Loss of zygotic fas function globally impairs the efficiency of splicing, and is associated with widespread retention of introns in mRNAs and dramatic changes in gene expression. Surprisingly, despite these general effects, zygotic fas mutants show specific defects in tracheal cell migration during mid-embryogenesis when maternally supplied splicing factors have declined. The study proposes that tracheal branching, which relies on dynamic changes in gene expression, is particularly sensitive for efficient spliceosome function. These results reveal an entry point to study requirements of the splicing machinery during organogenesis and provide a better understanding of disease phenotypes associated with mutations in general splicing factors (Sauerwald, 2017).

mRNA splicing is required to process nearly all eukaryotic transcripts. As splicing can be rate-limiting for efficient gene expression, reduced splicing efficiency can perturb many cellular processes. Although defects in RNA processing have been associated with pleiotropic phenotypes during development, inactivation of splicing factors can cause surprisingly specific cellular defects. Mutations in core components of the spliceosome are associated with human diseases, including retinitis pigmentosa and spinal muscular atrophy, and can contribute to carcinogenesis. However, the basis for these specific phenotypes is not clear (Sauerwald, 2017).

Morphogenesis requires dynamically regulated gene expression programs to coordinate cell movements and differentiation. Tracheal branching morphogenesis in Drosophila is guided by the dynamic expression of the fibroblast growth factor (FGF) branchless (Bnl) in cells surrounding the tracheal primordia. Bnl activates the FGF receptor (FGFR) breathless (Btl) on tracheal cells, which move as a cohort towards the Bnl source. This study shows that mutations in the faint sausage (fas) gene specifically affect branch outgrowth and cellular rearrangements during tracheal morphogenesis. We found that fas encodes a subunit of the spliceosome-activating Prp19 complex (Prp19C; also known as NineTeen complex, NTC; Chanarat, 2013), contrary to an earlier report that fas encodes a secreted immunoglobulin (Ig) domain protein (Lekven, 1998). Lack of zygotic Prp19C function broadly impairs the efficiency of mRNA splicing and leads to extensive changes in global gene expression. These findings suggest that tracheal branching morphogenesis is particularly sensitive to efficient spliceosome function, thus providing an entry point to investigate the requirements of splicing during organogenesis (Sauerwald, 2017).

This study describes the requirement of the NTC/Prp19C subunit Fas for tracheal branching morphogenesis. Although mRNA splicing is generally required for transcript maturation, this study found that embryos lacking zygotic fas function display surprisingly specific organogenesis defects. First, it was shown that the NTC/Prp19C subunit CG6197 is required for tracheal branching. The fas locus, previously reported to encode the secreted immunoglobulin domain protein CG17716 (Lekven, 1998), in fact encodes the NTC/Prp19C subunit CG6197 also known as fand (Guilgur, 2014). Second, we show that loss of zygotic fas function results in widespread perturbation of splicing, consistent with previous work demonstrating an essential requirement of maternal CG6197/fand function for efficient splicing during early embryogenesis (Guilgur, 2014; Martinho, 2015). Abnormal transcript processing in fas mutants manifests predominantly in intron retention, accompanied by changes in transcript abundance. Third, it was shown that compromised FGF and EGF signalling may contribute to the tracheal branching defects in fas mutants. The requirement of Fas in non-tracheal cells and the results of epistatic analysis suggest that fas function is essential for the activation of the Btl FGFR (Sauerwald, 2017).

The late onset and the specific nature of tracheal and CNS (Lekven, 1998) defects in zygotic fas embryos was surprising, given that pre-mRNA processing is generally perturbed in the mutants. However, maternally provided gene products, including Fas protein itself, are likely to allow for largely normal development during early embryogenesis. As maternal Fas protein decays over time, Fas levels appear to become limiting at the onset of tracheal morphogenesis during mid-embryogenesis. Perturbed pre-mRNA processing in fas mutants is accompanied by substantial changes in the levels of many transcripts. Nonsense-mediated mRNA decay is expected to degrade a large fraction of mis-spliced transcripts. In addition, indirect effects on transcriptional regulation are likely to influence gene expression in fas mutants. Of note, the abnormally processed transcripts in fas embryos include CG17716 mRNA, consistent with the finding of Lekven (1998) that CG17716 protein was undetectable in fas embryos (Sauerwald, 2017).

This study report an example of strikingly specific developmental defects associated with a lack of efficient splicing. How can a general perturbation of splicing lead to specific phenotypes? First, tissue-specific expression of some splicing factors, including Prp19, might account for tissue-dependent differences in splicing efficiency. Second, pre-mRNAs contain diverse auxiliary cis-acting regulatory elements that are recognized by a multitude of splicing factors. Consequently, different introns may show distinct sensitivities towards the lack of a given splicing factor. In addition, intron number, transcript abundance and transcript stability may render some RNAs more prone to accumulating splicing errors than others. Finally, dynamic cellular processes, such as cellularization and tracheal branching, which involve rapid regulation of gene expression, may depend more acutely on efficient mRNA processing. Consistent with this idea, highly expressed and rapidly regulated genes tend to have only few and short introns (Sauerwald, 2017).

These findings allow definition of the most sensitive gene expression events required for proper organogenesis. Characterizing NTC/Prp19C function and regulation in different organs could therefore contribute to a better understanding of how differential gene expression is regulated during organogenesis (Sauerwald, 2017).

One of several striking aspects of the phenotype resulting from loss of fas function provides evidence that sensory neurons of the peripheral nervous system (PNS) do not move normally out of the epidermal layer; this results in defective sensillum differentiation. Further, neurons of the central nervous system (CNS) do not migrate to their proper locations during the phase of germ band retraction and CNS condensation, which leads to gross abnormalities in cell shape and position; subsequently, patterning defects in axonal pathfinding occur (Lekven, 1998).

The Drosophila CNS develops from an invariant population of progenitor cells (neuroblasts) that segregate from the ectoderm and divide in a reproducible pattern to form a multilayer of neurons. At the approximate time when neurons start differentiating (stages 12-14), the CNS primordium undergoes a dramatic compaction. First, the germband (including the CNS primordium) retracts [Images], leading to a more than 50% reduction in CNS length. After germband retraction (stages 13-16) the CNS shortens further, until it measures only about 30% of its original, maximally extended length. The reduction in length of the CNS is mainly compensated for by an increase in the length of the dorsoventral axis and a higher cell-packing density, both for neuroblasts and neurons. Thus, cells that in the early embryo are arranged one behind the other, come to lie closer or even above and/or beside one another in the CNS of mature embryos. The CNS primordium of fas null mutant embryos develops normally until the late extended germ band stage; then, towards the end of germ band retraction (stage 12), defects in CNS morphogenesis and the expression of differentiation markers become apparent. Thus, markers for subsets of neuroblasts and their progeny (seven up, even-skipped, engrailed) do not show detectable abnormalities in the neuroblast map of early fas mutant embryos. However, the shortening and thickening of the CNS that normally take place during stages 13-16 fails to occur. At the same time, the fas mutant CNS remains flattened (along the dorsoventral axis) and widened (along the mediolateral axis). There does not appear in fas mutant embryos an obviously decreased number of neurons or glial cells, as visualized with antibodies against the Elav and Repo proteins, respectively. Also, markers for specific subsets of neurons are expressed in fas mutants (e.g., Even skipped, expressed in the aCC/pCC pair of neurons, and Engrailed, expressed in the progeny of the MNB neuroblast). However, the repositioning of neurons and glia that normally takes place during CNS condensation fails to occur. For example, longitudinal glial cells (LGCs) remain in segmental clusters, separated by gaps; they also stay at a more lateral level than their wild-type counterparts. Midline cells, such as the MNB progeny, do not form the regular, dense clusters typical for the wild-type embryo. Thus, the correct specification of NBs does not require fas, but the later morphogenetic movements of nerve cord condensation and consequent cell positioning do require fas. In summary, loss of faint sausage is associated with severe defects in tissue organization and cell movement in the epidermis and CNS. These defects, which first become manifest in the stages during and after germ band retraction, with the cell rearrangements that accompany this phase of development, result in generally abnormal shape and position of neurons. Subsequently, patterning defects in axonal pathfinding occur, possibly as a secondary consequence of the abnormal cell body positions, since Fas is not expressed on the surface of the axons themselves (Lekven, 1998).