Abstract
The specification of tissue identity during embryonic development requires precise spatiotemporal coordination of gene expression. Many transcription factors required for the development of organs have been identified and their expression patterns are known; however, the mechanisms through which they coordinate gene expression in time remain poorly understood. Here we show that hormone-induced transcription factor Blimp-1 participates in the temporal coordination of tubulogenesis in Drosophila melanogaster by regulating the expression of many genes involved in tube maturation. In particular, we demonstrate that Blimp-1 regulates the expression of genes involved in chitin deposition and F-actin organization. We show that Blimp-1 is involved in the temporal control of lumen maturation by regulating the beginning of chitin deposition. We also report that Blimp-1 represses a variety of genes involved in tracheal maturation. Finally, we reveal that the kinase Btk29A serves as a link between Blimp-1 transcriptional repression and apical extra-cellular matrix organization.
Introduction
Specialized cellular functions and cell lineage fates are usually regulated by only a few key instructive transcription factors required to activate or repress specific patterns of gene expression. During the development of multicellular organisms, these events must be precisely timed. As a result, organogenesis requires an accurate spatio-temporal regulation of gene expression over extended periods. While many transcription factors required for the development of organs have been identified and their expression pinpointed spatially, the mechanisms through which they coordinate downstream gene expression in time remain poorly understood. Here we addressed part of this question during the development of the Drosophila melanogaster tracheal system, a model used to study epithelial organ development. The tracheal system of D. melanogaster is formed by a network of epithelial tubes that requires tight temporal regulation of gene expression. Tracheal tube maturation involves the timely and spatially regulated deposition of a chitinous apical Extracellular Matrix (aECM), a process that is governed by downstream effectors of the mid-embryonic ecdysone hormone pulse (Chavoshiet al. 2010). One of these ecdysone response genes is the D. melanogaster B-Lymphocyte Inducing Maturation Protein-1 (Blimp-1) (Nget al. 2006; Chavoshiet al. 2010). Blimp-1 is the homolog of human Prdm1 (Positive regulatory domain containing 1) (Huang 1994). Blimp- 1/PRDM1 is a zinc finger transcriptional repressor that belongs to the Prdm gene family and it was originally identified as a silencer of β-interferon gene expression (Keller and Maniatis 1991). Prdm family members contain a conserved N-terminal domain, known as a positive regulatory domain (PR domain). This domain has been associated with the SET methyltransferase domain, which is important for the regulation of chromatin-mediated gene expression (Hohenauer and Moore 2012). In addition, Prdm family proteins contain multiple zinc fingers that mediate sequence-specific DNA binding and protein-protein interactions (Turneret al. 1994). Prdm family members modulate key cellular processes, including cell fate, and the aberrant function of some members may lead to malignant transformation (Foget al. 2012). During embryonic development, Blimp-1 controls a plethora of cell-fate decisions in many organisms (Bikoffet al. 2009; Hohenauer and Moore 2012). In D. melanogaster, Blimp- 1 serves as an ecdysone-inducible gene that regulates ftz-f1 in pupal stages (Agawaet al. 2007). By acting as a transcriptional repressor, Blimp-1 prevents the premature expression of ftz-f1, thereby influencing the temporal regulation of events that are crucial for insect development. The expression level and stability of Blimp-1 is critical for the precise timing of pupariation (Akagiet al. 2016).
Blimp-1 exerts a function in tracheal system morphogenesis during embryonic development (Nget al. 2006; ÖztÜrk-Çolaket al. 2016). However, the question remains as to how this transcription factor regulates tube maturation events downstream of the hormone ecdysone. Here we studied the role of Blimp-1 in the transcriptional regulation of the regulation of tracheal tube maturation in D. melanogaster. We found that Blimp-1 is a transcriptional repressor of many genes involved in tracheal development and that its levels are critical for the precise timing of luminal maturation and the final stages of tubulogenesis in the embryo. Our results indicate that Blimp-1, working downstream of ecdysone, acts as a link of hormone action during tube maturation in organogenesis.
Materials and Methods
D. melanogaster strains and genetics
All D. melanogaster strains were raised at 25°C under standard conditions. Mutant chromosomes were balanced over LacZ or GFP-labelled balancer chromosomes. Overexpression and rescue experiments were carried out either with btl-GAL4 (kindly provided by M. Affolter) or AbdB-GAL4 (kindly provided by E. Sánchez-Herrero) drivers at 25°C or 29°C. y1w118 (used as wild-type), Blimp-1KG09531, and UAS-srcGFP are described in FlyBase; UAS- Blimp-1 (ÖztÜrk-Çolaket al. 2016); Btk29Ak00206 and UAS-Btk29A (kindly provided by M. Strigini).
Embryo staging and synchronization
Embryos were staged following (Campos-Ortega and Hartenstein 1985). To study temporal chitin deposition, cages were set at 25°C for 2 h, and embryos were then allowed to develop for 18 h, 20 h, 22 h or 24 h at 18°C in order to obtain early and late stages 12, and early and late 13, respectively. Blimp-1 mutant embryos were compared to Blimp-1 heterozygotes after staining with CBP and 2A12. Heterozygote embryos were differentiated from homozygous mutant embryos by the presence or absence of a β-Gal-expressing balancer.
Synthesis of pri/tal RNA probes for in-situ hybridization
The pri/tal RNA probes were synthesized using a PCR-based technique. The pri gene region (524bp, covering all coding Open Reading Frames (ORFs) of the gene) was defined, and the forward (5’TAATACGACTCACTATAGGTTTTTGGTCAATACACGGCA3’) and reverse (5’AATTAACCCTCACTAAAGGAGTTTGTGGATAAGGCACGG3’) primers were designed accordingly so that the PCR product carried the two RNA promoters T3 and T7. The gene region of interest, flanked by the T3 and T7 sequences, was amplified from previously isolated genomic DNA via PCR under standard PCR conditions. The newly synthesized RNA was then purified by precipitation, re-suspended in hybridization buffer, and stored at -20°C.
Immunohistochemistry, image acquisition and processing
Standard protocols for immunostaining were applied. The following antibodies were used: rat anti-DE-cad (DCAD2, DSHB); rabbit anti-GFP (Molecular Probes); anti-Gasp mAb2A12 (DSHB); guinea pig anti-Blimp-1(S. Roy); anti-expansion and anti-rebuf (from M. Llimargas); anti-knk (from A. Uv) anti Btk29A (from M. Strigini); anti aPKC (Santa Cruz Biotechnology); and chicken anti-β-gal (Cappel). Biotinylated or Cy3-, Cy2- and Cy5-conjugated secondary antibodies (Jackson ImmunoResearch) were used at 1:300. Chitin was visualised with Fluostain (Sigma) at 1 μg/ml or CBP (Chitin Binding Probe, our own, made according to NEB protocols). Confocal images of fixed embryos were obtained either with a Leica TCS-SPE, a Leica TCS- SP2, or a Leica TCS-SP5 system. Images were processed using Fiji and assembled using Photoshop. 3D cell shape reconstructions were done using Imaris software.
Fluorescent in-situ hybridization
Freshly fixed embryos were washed and kept at 56°C in Hybridization Buffer for 3 h for pre-hybridization. In the last 10 min of pre-hybridization, probes (1:100 in Hybridization Buffer) were prepared for hybridization. The probes were hybridized with the embryos at 56°C overnight. The next day the embryos were washed and incubated in POD-conjugated anti-Dig (in PBT) for 1 h. The fluorescent signal was developed by the addition of Cy3 Amplification Reagent (1:100) diluted in TSA Amplification Diluent and incubation at room temperature in the dark for 10 min. Finally, the embryos were either mounted in Fluoromount medium or subjected to antibody staining.
In silico analysis of Blimp-1 binding sites
The Blimp-1 position weight matrix was taken from the reported binding sequences in (Kuo and Calame 2004). We extracted sequences 2000 bp up- and down-stream from all annotated isoforms in the D. melanogaster genome using the biomaRt database (Smedleyet al. 2015). All computations were performed within the R statistical framework (http://www.R-project.org). The Matscan software (Blancoet al. 2006) was used to find putative binding sites for the Blimp-1 position weight matrix in the aforementioned regions. For each binding site, we computed the mean conservation score of the corresponding positions following the D. melanogaster related species’ conservation track of the USCS browser (Fujitaet al. 2011). For genes with multiple transcription start sites (TSSs), non-redundant binding candidates for all TSSs were reported. Each gene was assigned with the maximum Matscan score of the corresponding binding sites after filtering by conservation score.
Results
Blimp-1 modulates tracheal tube size and apical ECM formation
Blimp-1, an ecdysone response gene (Becksteadet al. 2005; Chavoshiet al. 2010) (Supplementary Fig. S1), encodes the D. melanogaster homolog of the transcriptional factor B-lymphocyte-inducing maturation protein gene, whose mutants have been reported to have misshapen trachea with severe defects in taenidia (Nget al. 2006; ÖztÜrk-Çolaket al. 2016). Detailed expression analysis of Blimp-1 protein in tracheal cells showed that expression is not detectable until embryonic stage 12 and is then observed until stage 15 (Fig.1 A-D). At this stage, Blimp-1 protein levels started decreasing, first in dorsal trunk (DT) cells and by stage 16 the protein was no longer detectable in any tracheal cell. In addition to its tracheal expression, Blimp-1 was also detected in the epidermal, midgut and hindgut cells throughout these stages (Fig.1 A-D). In Blimp-1 mutant stage 16 embryos, we detected lower levels of chitin, inflated tracheal tubes and disorganized taenidial ridges (ÖztÜrk-Çolaket al. 2016) (Fig.1 E-H).
The tube shape of Blimp-1 mutant embryos was altered, DTs showing a smaller diameter at the fusion points than in the rest of the tubes (Fig.1 H). To better study whether this phenotype was due to constrictions at fusion points and/or dilations along the entire length of the tubes, we measured the tube diameter at fusion points of tracheal metameres Tr7-Tr8 and Tr8-Tr9 and the largest tube diameter at the Tr8 metamere between fusion points. Blimp-1 mutant embryos had a significantly (p-value: 4.4E-07, by Student’s T-test, n=10) larger tube diameter between fusion points than wild-type (wt) DTs (n=15), while at the fusion points the tube diameter was similar to the wt (Fig.1 I I). These results suggest that Blimp-1 is required to maintain the luminal structures from over-expanding in between fusion points.
Since tube expansion is related to the apical cell surface, we next examined cell shape in the trachea of Blimp-1 mutant embryos. We labelled the apical cell junctions in Blimp-1 mutants and found that the apical cell shape differed to that of wt cells. In Blimp-1 mutants, the longest cell axis appeared to be perpendicular to the tube axis, while in the wt it was parallel (Fig.1 J-M). Most of the Blimp-1 mutant cells had a similarly reduced apical cell surface area, thereby suggesting a uniform organization throughout the tube (except for the fusion cells, which already had a distinct apical cell shape that did not seem to be affected by the loss of function of Blimp-1) (Fig. 1 K). To have a better idea of cell organization in Blimp-1 mutant trachea, we traced tracheal cell surfaces and compared them to those of the wt. We observed altered cell shapes in the former. In this regard, the mutant trachea showed cells that were less elongated and more “square-shaped” than wt ones (Fig.1 L, M).
Taken together, these observations reveal that Blimp-1 affects various stages of tube maturation, from chitin deposition to tube expansion, as well cellular morphology.
Blimp-1 modulates the timing of chitin deposition
In wt embryos, a matrix composed of chitin and proteins such as Gasp, accumulates in the tracheal lumen. This matrix plays a key role in the regulation of tube length and diameter expansion (Moussianet al. 2005; Tonninget al. 2005; Moussianet al. 2015). Due to the striking tube size phenotype of Blimp-1 mutants, and because mutations in genes involved in chitin biogenesis and assembly result in irregular diametric expansion leading to locally constricted and dilated tubes, we examined the deposition of these markers from stage 12 (Fig. 2). In the wt, early chitin deposition began at stage 13, when a chitinous filament started to be deposited inside the DT just prior to tube expansion (Moussianet al. 2015). In parallel, Gasp was detected from stage 13, but at this stage it was mainly cytoplasmatic. None of these markers were detected at earlier stages. At stage 13, chitin began to be deposited in the lumen of all branches, starting from the DT (Moussianet al. 2015), and a mature pattern was achieved at stage 17, when taenidial ridges are fully formed (TiklovÁet al. 2013; ÖztÜrk-Çolaket al. 2016). Blimp-1 mutant embryos showed chitin deposition as early as stage 12 (Fig. 2 B, C). The pattern of chitin deposition in mutants at stage 13 was similar to that of the wt (Fig. 2 D, F). However, at later stages, the trachea of Blimp-1 mutants showed lower levels of chitin than the wt (Fig. 2 G) (ÖztÜrk-Çolaket al. 2016). Gasp localization was lower in Blimp-1 mutants throughout embryogenesis, especially in the DT during later stages (Fig. 2 K), a pattern resembling that of embryos mutant for genes involved mutants involved in chitin synthesis and organization (AraÚjoet al. 2005). We have previously shown that when we overexpressed Blimp-1 in the posterior part of the embryo using Abd-BGAL4 to create tracheal DTs with distinct cellular compositions (FÖrsteret al. 2009), we could detect lower levels of chitin in the posterior domain expressing higher levels of Blimp-1 (öztürk-Qolaket al. 2016). To further analyse the influence of Blimp-1 in timely chitin deposition, we used the same experimental conditions and found that chitin was hardly detected in the posterior part of the trachea (Fig. 2 L) (Ozturk-Colaket al. 2016). However, further examination with higher laser power and gain showed that indeed both the chitin filament and taenidia were present throughout the trachea, although they displayed much lower levels in the posterior metameres, resembling chitin levels at earlier stages of development (Fig. 2 M). Blimp-1 overexpression lead to a tubular structure that regarding chitin composition and organization was apparently younger than the same tube with normal levels of Blimp-1 protein. Thus, Blimp-1 seems to regulate the timing of the beginning of chitin deposition by repressing target proteins. When Blimp-1 levels are high, chitin deposition is delayed, whereas when they are low, chitin deposition begins earlier in the tracheal tubes.
The expression pattern of tarsaless/polished rice is altered in Blimp-1 mutants
Mutants for tarsal-less (tal), also known as polished rice (pri), are affected in both F-actin and taenidia organization and we analysed them in parallel to Blimp-1 mutants (ÖztÜrk-Çolaket al. 2016). Like Blimp-1, the tracheal expression of tal/pri in wild-type embryos is regulated by ecdysone (Chanut-Delalandeet al. 2014)(Supplementary Fig. S1) and started in tracheal cells at stage 12 (Supplementary Fig. S2). At early stages this expression was restricted to DT cells but was later on observed uniformly in all tracheal cells until stage 15 (Supplementary Fig. S2). From this stage, tal/pri expression gradually decreased in all DT cells, except fusion cells, and at stage 16, coinciding with the time Blimp-1 disappears from tracheal cells, it became higher in fusion cells than in the rest of the DT (Supplementary Fig. S2 and Fig. 2 O). tal/pri expression in Blimp-1 mutants at stage 14 was not as uniform as in the wt, as it was higher in some cells than in others (Fig. 2 P). To verify whether the cells with higher tal/pri expression were indeed fusion cells, we double labelled the embryos with the fusion cell marker dysfusion (dys) (Jiang and Crews 2003). High levels of both tal/pri and dys were detected in the same cells, thereby confirming that these were indeed fusion cells (Fig. 2 Q). These results indicate that tal/pri expression at stage 14 Blimp-1 embryos resembles expression of tal/pri in stage 16 wt embryos, suggesting that Blimp-1 regulates the timing of the onset of tal/pri differential expression, from embryonic stage 16. We hypothesize that Blimp-1 might be involved in the differential regulation of tal/pri expression in tracheal cells. Blimp-1 could play a role in repressing tal/pri expression at high levels in fusion cells and, from stage 16, when Blimp-1 is no longer present in tracheal cells, this repression would no longer be exerted leading to higher levels of expression in fusion cells.
Blimp-1 regulates a variety of genes involved in tube maturation
To further study the influence of Blimp-1 in tracheal development, we searched in silico for Blimp-1 binding sites in the promoter regions of all D. melanogaster genes using the Matscan software (Blancoet al. 2006) and the reported position weight matrix corresponding to Blimp-1 (Kuo and Calame 2004). We found 3949 genes with at least one putative binding site (binding score larger than 75% of maximum value) within 2000 bases of their annotated TSSs. We prioritized candidate positions on the basis of binding score and evolutionary conservation from the UCSC Drosophila-related species track (Fujitaet al. 2011) [Supplementary material T1]. In particular, we found that Blimp-1 can potentially regulate its own transcription and also the transcription of a variety of genes involved in tracheal tube development. We asked whether Blimp-1 regulation of tracheal genes was enriched in relation to all genes in the fly genome. The transcription factor Trachealess (Trh) is among the first genes to be expressed in the cells that will form the trachea. In the absence of Trh, tracheal cells fail to invaginate to form tubes and remain at the embryo surface (Isaac and Andrew 1996; Wilket al. 1996). It has been shown that expression of nearly every tracheal gene requires trh (Chunget al. 2011). Thus, we wondered how many genes regulated by Blimp-1 were also downstream of Trh and, therefore, bonafide tracheal genes. To test this we combined published tracheal gene sets (Chunget al. 2011; Hammondset al. 2013) and checked for enrichment among the genes with predicted Blimp-1 binding sites. We found that for a conservation threshold of 2 and Matscan scores larger than 75% tracheal genes are significantly enriched (fisher-test OR=1.60, p-value<0.001). In order to show robustness against the choice of Matscan score threshold we repeated the Fisher test for values ranging from 75% to 90%. We found that the odds ratio was kept relatively constant until 84%, dropping to non-significant values for the remaining thresholds (Fig. 3 B, C and Supplementary material T3).
We then selected a shorter list of genes reported to be involved in tube maturation stages (Table 1).
Clearly, this is not an exhaustive list as it only includes the genes detected using these restrictive parameters. Thus, for example, this analysis did not allow us to detect any Blimp-1 binding sites in the tal/pri region. However, due to the changes in tal/pri expression in Blimp-1 mutants, we further studied the tal/pri region under less restrictive conditions. We detected four binding sites with Matscan scores above the 72 percentile and conservation scores of 1.
Blimp-1 regulates the levels of Exp, Reb, aPKC, Knk and Btk29A
In order to further analyze the relationship between Blimp-1 and tracheal maturation, we compared the levels of five key proteins, Exp, Reb, aPKC, Knk and Btk29A, in tube maturation in the heterozygous and homozygous Blimp-1 mutant embryos. Exp and Reb are atypical Smad-like proteins that regulate tube size in the tracheal system by promoting chitin deposition (Moussianet al. 2015). aPKC is a serine/threonine protein kinase required for apico-basal cell polarity and a member of the Par complex. It has been shown to be involved in the orientation of actin rings and taenidial ridges in larval stages of tube maturation (Hosonoet al. 2015). Knk is a GPI anchored protein needed for chitin organization and the regulation of tracheal tube diameter (Moussianet al. 2006). Btk29A (also known as Tec29A) is the only member of the Tec family of kinases in Drosophila, and it is expressed in many developmental stages of the fly. In the tracheal system, Btk29A is involved in spiracular chamber invagination, as well as in tracheal cuticle patterning (Matuseket al. 2006; Tsikalaet al. 2014). Mutants and overexpression conditions for each of these genes showed phenotypes that can be correlated to the Blimp-1 tracheal maturation phenotypes. We therefore hypothesized that Blimp-1 acts as a transcriptional repressor of these genes during tube maturation (Matuseket al. 2006; Moussianet al. 2006; Hosonoet al. 2015; Moussianet al. 2015).
At early tube maturation stages, Blimp-1 mutants displayed higher levels of Exp, aPKC, Knk and Btk29A than same-stage wt embryos (Fig. 4), thereby suggesting that Blimp-1 represses the expression of these proteins, as hypothesized. Of the four proteins studied, Btk29A levels showed a more pronounced difference between heterozygous and homozygous Blimp-1 mutant embryos. Btk29A levels at stage 14 were hardly detectable in the tracheal system of wt embryos (Fig 4 G’)(Tsikalaet al. 2014), whereas in mutant tracheal cells they were detected at much higher levels (Fig 4 H’). Taken together with our previous in silico approach, our findings suggest that Blimp-1 directly regulates the levels of these five proteins in tracheal cells during tube maturation, working as a repressor during early tracheal developmental stages.
During lumen maturation, chitin deposition requires the activity of Krotzkopf verkehrt (Kkv) together with Exp and Reb (Moussianet al. 2015). However, we could not find any evidence that Blimp-1 regulates kkv expression. Due to the earlier chitin deposition observed in Blimp-1 mutant embryos (Fig.2), we wondered if Reb expression was also detected earlier, consequently leading to earlier chitin deposition. Indeed, we observed that in Blimp-1 homozygous embryos, Reb expression was already detected at stage 12, in contrast to wildtype and Blimp-1 heterozygous embryos where it starts being detected at stage 13 (Fig. 4 I-L) (Moussianet al. 2015). Thus, the earlier chitin deposition detected in Blimp-1 embryos may be triggered by this earlier expression of Reb together with the higher levels of Exp in tracheal cells.
Btk29A works downstream of Blimp-1 to regulate luminal aECM organization
What is the role of Btk29A in tube maturation downstream of Blimp-1? At late embryonic stages, a strong Btk29A mutant allele displays disorganized apical F-actin bundles and taenidial ridges (Matuseket al. 2006; Ozturk-Qolaket al. 2016), with heterogeneous patterns of F-actin bundling in the same trachea, showing stretches of perpendicular bundles followed by stretches of parallel bundles (Ozturk-Qolaket al. 2016). In addition, overexpression of full-length Btk29A in all tracheal cells gave rise to an expansion phenotype similar to Blimp-1 mutants at stage 16 (Fig. 5 C, compare to 1 H). Thus, and due to our hypothesis of Blimp-1 being a repressor of Btk29A expression, we examined whether derepression of Btk29A might partially account for the Blimp-1 mutant phenotype. To do so, we used Btk29AK00206, a hypomorphic allele in which low mRNA levels are still detected in the embryonic tracheal system (Tsikalaet al. 2014). In this hypomorphic allele, there were no apparent differences in chitin deposition between heterozygous and homozygous Btk29A mutant embryos at stage 16 (Fig. 5 A-B). At late stage 17, Btk29A K00206 homozygous embryos showed mild DT phenotypes compared to heterozygous embryos, such as a very mild DT expansion phenotype (Fig. 5 D, E). However, there were no detectable taenidial ridge orientation phenotypes in stage 17 Btk29AK00206 mutants, which showed parallel taenidial ridges perpendicular to the tube length as in the wild-type (Fig. 5 E INSET?). We then combined the Btk29AK00206 mutation with Blimp-1 and observed that the incomplete removal of embryonic Btk29A could partially rescue the taenidial ridge orientation phenotypes in most of the embryos examined (64%, n=11). These double Btk29AK00206; Blimp-1 mutants showed heterogeneous patterns of taenidial ridge organization in the same trachea, showing stretches of perpendicular bundles followed by stretches of parallel ones (45% of embryos, Fig. 5 G, H, arrow in G), as well as diagonal ridges (19% of embryos, Fig. 5 H, arrow). We also observed that Btk29A; Blimp-1 double mutant DTs showed a milder tube expansion phenotype than Blimp-1 mutants (Fig. 5 F-H). Taken together, these results indicate that part of the Blimp-1 phenotype can be attributed to excess Btk29A.
Discussion
Here we found that Blimp-1 regulates multiple tracheal targets, thus acting as a key gene in tracheal development (Table 1 and Fig. 6A). Blimp-1 is an ecdysone response gene (Becksteadet al. 2005; Chavoshiet al. 2010) (Fig. S1) and therefore a link between the hormonal signal and the timing of tracheal tube maturation in both embryos and larvae. We show that Blimp-1 regulates the expression of many genes required for tube maturation. Interestingly, in silico, we detected four Blimp-1 binding sites in Blimp-1 regulatory sequences using the parameters described, which suggest that Blimp-1 may regulate its own expression. This is in agreement with recent data showing that Blimp-1/PRDM1 is also able to regulate its own expression in mammals (Mitaniet al. 2017). Self-regulation of expression is consistent with the feedback loops in which Blimp-1/PRDM1 participates and also with its role in regulating many developmental processes (Gong and Malek 2007; Bikoffet al. 2009). We also found Blimp-1 binding sites in the region of Tramtrack (Ttk), another transcription factor involved in many features of tube maturation (AraÚjoet al. 2007).
Furthermore, we observed that Blimp-1 modulates the timing of the expression of Reb and Exp, two genes involved in the genetic programme triggering timely chitin deposition (Moussianet al. 2015). Untimely chitin deposition was shown to disturb tube maturation, thereby demonstrating that this process has to be tightly regulated during tracheal development. Tracheal overexpression of reb leads to earlier chitin deposition in all branches from stage 13 and sometimes chitin appearance at stage 12 ((Moussianet al. 2015) and M. Llimargas personal communication). Accordingly, our results show that Reb is expressed earlier in Blimp-1 embryos (Fig. 4 J, L). This agrees with the early chitin deposition phenotypes observed in Blimp-1 mutants (Fig.2 B, C). Furthermore, Blimp-1 also modulated knk expression during tube maturation stages. Knk is involved in directing chitin assembly in the trachea (Moussianet al. 2006) and correct amounts of Knk at specific times during metamorphosis are important for correct wing cuticle differentiation and function (Liet al. 2017). Taken together, these in silico and in vivo results indicate that Blimp-1 is a transcription factor that acts downstream of ecdysone and that it is involved in the correct timing of chitin synthesis and deposition during embryonic development.
We also found Blimp-1 binding sites in the aPKC coding region. aPKC is involved in the junction anisotropies that orient both actin rings and taenidial ridges in the lumen of tracheal tubes (Hosonoet al. 2015). In Blimp-1 mutants, both actin rings and taenidial ridges are either undetectable or misoriented (Ôztürk-Çolaket al. 2016)—observations that are consistent with changes in junction anisotropy.
We previously showed that Blimp-1 regulates chitin deposition levels and architecture and that subsequently the chitin aECM feeds back on the cellular architecture by stabilizing F-actin bundling and cell shape via the modulation of Src42A phosphorylation levels (ÖztÜrk-Çolaket al. 2016). However, in this report, we provided no link between the chitinous aECM and Src42A. Btk29A mutant larvae have an aECM phenotype, which may be the result of their actin bundle phenotype (Matuseket al. 2006; ÖztÜrk-Çolaket al. 2016). Here, we found that Btk29A removal can partially rescue the Blimp-1 taenidial orientation and expansion phenotype. In view of these results, we propose that the contribution of Btk29A can be added to the feedback model for the generation of supracellular taenidia put forward in Ôzturk-Çolak et al. (Ôztürk-Çolaket al. 2016). We now add on our previous model by hypothesizing that Blimp-1 acts as a link between the aECM and cells by regulating the levels of Btk29A (Fig. 6 B). Btk29A and Src42A, together with the formin DAAM, have been shown to regulate the actin cytoskeleton (Matuseket al. 2006). In agreement with our results, we speculate that Btk29A might phosphorylate Src42A and that this phosphorylation event could be modulated by Blimp-1 and DAAM.
To conclude, our results indicate that Blimp-1 is a key player in the regulation of tracheal tube maturation and, consequently, in the feedback mechanism involved in the generation of supracellular taenidia.
Acknowledgements
We thank M. Llimargas, E. Sanchez-Herrero, S. Roy, H. Ueda and the Bloomington Stock Center for flies and reagents. Thanks also go to the IRB Barcelona Biostatistics and Bioinformatics Unit. We acknowledge L. Bardia, A. Lladó and J. Colombelli from the IRB Barcelona Advanced Digital Microscopy Facility for help and advice with confocal microscopy and software and E. Fuentes and N. Martin for technical assistance. S.J.A. was a Ramon y Cajal Researcher (RYC-2007-00417); A.O. was the recipient of an IRB-La Caixa fellowship. This work was supported by grants from the Generalitat de Catalunya and the Spanish Ministerio de Ciencia e Innovación (BFU2009-07629).