Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 8

Two-step Regulation of trachealess Ensures Tight Coupling of Cell Fate With Morphogenesis in the Drosophila Trachea

Affiliations

Two-step Regulation of trachealess Ensures Tight Coupling of Cell Fate With Morphogenesis in the Drosophila Trachea

Takefumi Kondo et al. Elife.

Abstract

During organogenesis, inductive signals cause cell differentiation and morphogenesis. However, how these phenomena are coordinated to form functional organs is poorly understood. Here, we show that cell differentiation of the Drosophila trachea is sequentially determined in two steps and that the second step is synchronous with the invagination of the epithelial sheet. The master gene trachealess is dispensable for the initiation of invagination, while it is essential for maintaining the invaginated structure, suggesting that tracheal morphogenesis and differentiation are separately induced. trachealess expression starts in bipotential tracheal/epidermal placode cells. After invagination, its expression is maintained in the invaginated cells but is extinguished in the remaining sheet cells. A trachealess cis-regulatory module that shows both tracheal enhancer activity and silencer activity in the surface epidermal sheet was identified. We propose that the coupling of trachealess expression with the invaginated structure ensures that only invaginated cells canalize robustly into the tracheal fate.

Keywords: D. melanogaster; Drosophila; cell differentiation; developmental biology; epithelial morphogenesis; gene expression; invagination; trachea.

Conflict of interest statement

TK, SH No competing interests declared

Figures

Figure 1.
Figure 1.. trh is essential for maintaining the invaginated tracheal structures.
(A) Schematic of the tracheal morphogenesis process. For clarity, only apical surfaces are shown. (B, C) Live imaging of tracheal invagination in a control embryo (B) and a trh mutant (C). Red arrowheads: apical constriction forming a tracheal pit. Yellow circle: a mitotic cell associated with accelerated invagination, distinguished by condensed histone. Red arrows: transient invagination and return to epidermis in a trh1 mutant. Gray arrows: segmental groove, which is not a tracheal structure. Par-6::GFP indicates the apical cell side, and His2Av::mRFP indicates chromosomes. Time point zero is set to the onset of germband retraction. (D) Activity of R14E10-GAL4 in a control embryo monitored using UAS-mCD8RFP. (E–G) Activity of R14E10-GAL4 in trhA14-3/B16-11 mutant embryos monitored using UAS-mCD8RFP. Green: DE-cad, Magenta: mCD8RFP driven by R14E10-GAL4. Cells expressing RFP initiated invagination at stage 10 (E), and invaginated structures formed within the RFP-positive cell cluster at stage 12 (F). However, these invaginated structures were not observed at stage 16, and RFP-positive cells were observed in the surface epidermis (G). Scale bars, 10 μm.
Figure 1—figure supplement 1.
Figure 1—figure supplement 1.. Activation pattern of EGFR signaling before invagination.
Activation pattern of ERK (di-phosphorylated ERK) in a control, a trhA14-3 mutant, and a rhodel1 mutant embryo at stage 10 before invagination. Before invagination, ERK was phosphorylated in the tracheal placodes of the control and trh mutant but not in those of the rho mutant. tracheal placodes were labeled by using trh66-lacZ.
Figure 1—figure supplement 2.
Figure 1—figure supplement 2.. Characterization of TALEN-induced trh mutants.
(A) Predicted amino acid sequences of trhA14-3 and trhB16-11 TALEN mutant proteins. Each allele has a frameshift mutation on the most upstream common exon among all isoforms. Both frameshift mutations lead to premature translational termination and the loss of the PAS A, PAS B and transcription activation domains. The trhA14-3 allele also loses the bHLH domain. The changes in the predicted amino acid sequences of the Trh-PA isoform are shown as an example. (B) No Trh protein expression was detected in trhA14-3/B16-11 mutant embryos. β-gal driven by R14E10-GAL4 labels would-be tracheal cells.
Figure 1—figure supplement 3.
Figure 1—figure supplement 3.. The trh mutant phenotype was not rescued by btl overexpression or the inhibition of apoptosis.
(A–D) Phenotypes resulting from gene overexpression in the tracheal placodes of trhA14-3/B16-11 mutants. Each gene was overexpressed using R14E10-GAL4. (A) trh overexpression, (B) btl:GFP overexpression, (C) p35 overexpression. Scale bars, 10 μm.
Figure 2.
Figure 2.. Trh expression is maintained only in invaginated tracheal cells.
(A, B) Trh expression in a tracheal placode of a control (rhodel1 bnlP1 /+) (a) and a rhodel1 bnlP1 mutant (B) embryo at stage 10 before invagination. (C–E) Trh and β-gal expression in a control embryo (C), a rhodel1 bnlP1 mutant (D), and a rhodel1 bnlP1 mutant with p35 overexpression (E) at stage 14 after invagination. β-gal and p35 expression were driven by R14E10-GAL4. (F) Schematic of the tracheal branching pattern after invagination. (G) Upper: Boxplot of cell numbers. Initial Trh+: the number of Trh-expressing cells before invagination, trh-Off: the number of cells expressing β-gal driven by R14E10-GAL4 in the epidermis (Trh-negative) after invagination (stage 13–14), trh-On: the number of invaginated tracheal cells expressing Trh after invagination (stage 13–14), sum: the sum of trh-Off and trh-On. **: Exact Wilcoxon-Mann-Whitney Test, p=0.001331, *: Steel-Dwass test, p<0.001 (for trh-Off, control vs rho bnl: p=5.3 × 10−9, control vs rho bnl +p35: p=8.3 × 10−10, rho bnl vs rho bnl +p35: p=3.9 × 10−8; for trh-On, control vs rho bnl: p=5.28 × 10−10, control vs rho bnl +p35: p=8.4 × 10−10, rho bnl vs rho bnl +p35: p=2.8 × 10−10; for sum, control vs rho bnl: p=3.1 × 10−10, control vs rho bnl +p35: p=2.0 × 10−4, rho bnl vs rho bnl +p35: p=3.2 × 10−9). Lower: mean numbers of trh-On cells and trh-Off cells at stages 13–14 after invagination. (H) Schematic of the dynamics of Trh expression during invagination. Scale bars, 10 μm.
Figure 2—figure supplement 1.
Figure 2—figure supplement 1.. Only a few invaginated cells maintain Trh expression in rho CycA bnl mutants.
(A) Trh expression in the tracheal cells of a rhodel1 CycAC8LR1 bnlP1 mutant at stage 10 before invagination and stage 14 after invagination. Although Trh expression was almost normal at stage 10, only a few invaginated cells were Trh-positive after invagination. Scale bars, (stage 10) 50 μm, (stage 14) 10 µm. (B) Boxplot showing the numbers of Trh+ cells in each placode (tr4, 5 and 6) at stage 10 before invagination and stage 14 after invagination.
Figure 3.
Figure 3.. R15F10 reproduces the tubule-restricted Trh pattern.
(A) Genomic positions of R15F01, other enhancers, and the insertion site of 1-eve-1 at the trh locus (B–D) Enhancer activity of R15F01 monitored using R15F01-GAL4 with UAS-mCD8RFP in a control (B’), rhodel1 bnlP1 mutant (C), and rhodel1 CycAC8LR1 btldeltaOh10 mutant (D) embryo. β-gal indicates the expression from a lacZ enhancer trap line for trh, 1-eve-1 (B). (E) Phenotype in adult cuticle pigmentation. The R15F01-GAL4 fly had a lighter body color than the R14E10-GAL4 fly, indicating that R15F01 represses an adjacent mini-yellow gene. (F–M) Enhancer activity of R15F01 monitored using the direct lacZ reporter in control embryos and in several invagination mutants: rhodel1 CycAC8LR1 (H), rhodel1 bnlP1 (I), CycAC8LR1 bnlP1 (J), rhodel1 CycAC8LR1 bnlP1 (K), Df(1)BSC352 (deficient in all upd1, 2 and 3), arrowhead: a segment with invaginated trachea, arrow: a segment without trachea (L), and trhA14-3/B16-11 R14E10-GAL4 > UAS-flag::trh[S665D] (M). Scale bars, 10 μm.
Figure 3—figure supplement 1.
Figure 3—figure supplement 1.. Activity of trh enhancers in tracheal cells and surrounding epidermal cells.
(A) The trh47 activity in control and rhodel1 bnlP1 mutant embryos. In both cases, cells activating the trh47 element were observed in both the trachea and epidermis at stage 14 after invagination. Scale bars, 10 μm. (B) The trh67 activity in control and rhodel1 bnlP1 mutant embryos. In both cases, the trh67 element was activated in tracheal cells in a sporadic manner. Scale bars, 10 μm.
Figure 3—figure supplement 2.
Figure 3—figure supplement 2.. Activity of R14E10 and R15F01 in rho bnl and trh mutants.
(A) Schematic of the GAL4- or lacZ transgene structure after integration into the attP2 or attP40 site. (B) β-gal reporter expression in a R14E10-lacZ strain at stage 10 (before invagination). (C) Boxplot of the number of β-gal positive cells in each placode (tr4, 5 and 6). n = 15. (D–E) Enhancer activity of R15F01 monitored using the direct lacZ reporter at stage 10 (before invagination): in a rhodel1 bnlP1 mutant (d) and a trhA14-3 mutant (e). (F) Enhancer activity of R15F01 monitored using the direct lacZ reporter at stage 15 in a trhA14-3 mutant. (G) Boxplot of the number of β-gal positive cells in each placode (tr4, 5 and 6) at stage 10 (before invagination). n = 18 (control), n = 15 (rho bnl), n = 15 (trh). *: Steel-Dwass test, p<0.01 (control vs rho bnl: p=0.0027, control vs trh: p=5.8 × 10−6, rho bnl vs trh: p=5.0 × 10−4. (H) Boxplot of the number of β-gal positive cells in each placode (tr4, 5 and 6) at stage 15 in trh mutants. n = 15.
Figure 3—figure supplement 3.
Figure 3—figure supplement 3.. trh-overexpression by using R15F01-GAL4 rescued the trh mutant phenotype in maintaining invaginated structures.
trh-OE using R14E10-GAL4 rescued both tracheal invagination and the maintenance of invaginated structures of trh mutants. On the other hand, trh-OE using R15F01-GAL4 did not rescue the invagination phenotype at stage 12, which may be due to the delayed onset of R15F01 and/or a smaller number of R15F01-positive cells, but allowed cells to keep invaginated structures later.
Figure 4.
Figure 4.. The R15F10 CRM contains multiple tracheal enhancers and epidermal silencers.
(A) Summary of the domain mapping of R15F01. ± indicates sporadic expression. (B–E) Activities of R15F01 (full length), D1, D7, and del4 monitored using GAL4 and UAS-mCD8RFP. D7 activated RFP expression in both tracheal and surrounding epidermal cells (arrows in E). D1 induced RFP expression in a sporadic manner. (F–I) Activities of D1 and D7 monitored using the direct lacZ reporter at stage 10 (before invagination), stage 11 (during invagination), and stage 15 (after invagination). D7 showed enhancer activity in the tracheal placodes before invagination, but D1 activated the reporter in tracheal cells after invagination. Arrows in I indicate epidermal expression of β-gal. Scale bars, 10 μm.
Figure 4—figure supplement 1.
Figure 4—figure supplement 1.. Enhancer and silencer activities of truncated R15F01 elements.
(A) Adult body colors of each transgenic strain. (B–G) β-gal reporter expression in D2, D3, D4, D5, D6, and D8-lacZ strains at stage 14 or 15 after invagination. (H–K) β-gal reporter expression in del1, del2, del3, and del4-lacZ strains at stage 10 (before invagination), stage 11 (during invagination), and stage 14 (after invagination). del1-lacZ did not show β-gal expression before invagination and showed sporadic tracheal expression after invagination. Scale bars, 10 μm.
Figure 5.
Figure 5.. The R15F01 epidermal silencers counteract a heterologous epidermal enhancer.
(A) Schematic of chimeric reporters with svb-E6B and R15F01 fragments (B, C) Reporter β-gal expression (magenta or fire) from E6B-lacZ and E6B-[D3-6]-lacZ integrated at the attP40 site. The epidermal β-gal expression in E6B-[D3-6]-lacZ was significantly weaker than that in E6B-lacZ. (D–F) Reporter β-gal expression (magenta or fire) from E6B-lacZ, E6B-[D1-2]-lacZ, and E6B-[del1]-lacZ integrated at the ZH-51c site. The epidermal β-gal expression in E6B-[D1-2]-lacZ and E6B-[del1]-lacZ was significantly weaker than that in E6B-lacZ, while both showed tracheal reporter expression. (G–I) Reporter GFP expression (green or fire) in (G) arm-GAL4 >3×UAS-GAP43GFP, (H) arm-GAL4 >3 × UAS-R15F01-GAP43GFP, and (I) 3×UAS-GAP43GFP only. Both 3×UAS-GAP43GFP and 3×UAS-R15F01-GAP43GFP transgenes were integrated at the attP2 site. The epidermal GFP expression (the surface of embryos) in arm-GAL4 >3×UAS-R15F01-GAP43GFP was significantly weaker than that of 3×UAS-GAP43GFP, while hindgut GFP expression was detectable. Embryos possessing only 3×UAS-GAP43GFP showed tracheal GFP expression but not hindgut expression. Arrows indicate the hindgut.

Similar articles

See all similar articles

References

    1. Andrew DJ, Ewald AJ. Morphogenesis of epithelial tubes: insights into tube formation, elongation, and elaboration. Developmental Biology. 2010;341:34–55. doi: 10.1016/j.ydbio.2009.09.024. - DOI - PMC - PubMed
    1. Bischof J, Maeda RK, Hediger M, Karch F, Basler K. An optimized transgenesis system for Drosophila using germ-line-specific phiC31 integrases. PNAS. 2007;104:3312–3317. doi: 10.1073/pnas.0611511104. - DOI - PMC - PubMed
    1. Brodu V, Baffet AD, Le Droguen P-M, Casanova J, Guichet A. A developmentally regulated Two-Step process generates a noncentrosomal microtubule network in Drosophila tracheal cells. Developmental Cell. 2010;18:790–801. doi: 10.1016/j.devcel.2010.03.015. - DOI - PubMed
    1. Brodu V, Casanova J. The RhoGAP crossveinless-c links trachealess and EGFR signaling to cell shape remodeling in Drosophila tracheal invagination. Genes & Development. 2006;20:1817–1828. doi: 10.1101/gad.375706. - DOI - PMC - PubMed
    1. Brown S, Hu N, Hombría JC-G. Identification of the first invertebrate interleukin JAK/STAT receptor, the Drosophila gene domeless. Current Biology. 2001;11:1700–1705. doi: 10.1016/S0960-9822(01)00524-3. - DOI - PubMed

Publication types

LinkOut - more resources

Feedback