Spatial-temporal targeting of lung-specific mesenchyme by a Tbx4 enhancer

Abstract

Background: Reciprocal interactions between lung mesenchymal and epithelial cells play essential roles in lung organogenesis and homeostasis. Although the molecular markers and related animal models that target lung epithelial cells are relatively well studied, molecular markers of lung mesenchymal cells and the genetic tools to target and/or manipulate gene expression in a lung mesenchyme-specific manner are not available, which becomes a critical barrier to the study of lung mesenchymal biology and the related pulmonary diseases.

Results: We have identified a mouse Tbx4 gene enhancer that contains conserved DNA sequences across many vertebrate species with lung or lung-like gas exchange organ. We then generate a mouse line to express rtTA/LacZ under the control of the Tbx4 lung enhancer, and therefore a Tet-On inducible transgenic system to target lung mesenchymal cells at different developmental stages. By combining a Tbx4-rtTA driven Tet-On inducible Cre expression mouse line with a Cre reporter mouse line, the spatial-temporal patterns of Tbx4 lung enhancer targeted lung mesenchymal cells were defined. Pulmonary endothelial cells and vascular smooth muscle cells were targeted by the Tbx4-rtTA driver line prior to E11.5 and E15.5, respectively, while other subtypes of lung mesenchymal cells including airway smooth muscle cells, fibroblasts, pericytes could be targeted during the entire developmental stage.

Conclusions: Developmental lung mesenchymal cells can be specifically marked by Tbx4 lung enhancer activity. With our newly created Tbx4 lung enhancer-driven Tet-On inducible system, lung mesenchymal cells can be specifically and differentially targeted in vivo for the first time by controlling the doxycycline induction time window. This novel system provides a unique tool to study lung mesenchymal cell lineages and gene functions in lung mesenchymal development, injury repair, and regeneration in mice.

Figure 1

Analysis of the 5.5 kb genomic DNA sequences of the potential mouse Tbx4 lung enhancer. (A) Placental mammal basewise conservation scores determined by phyloP program [11]. Positive scores are assigned to the sites with conserved DNA sequences among the studied vertebrate species, which include mouse, rat, kangaroo rat, naked mole rat, guinea pig, squirrel, rabbit, pika, human, chimp, gorilla, orangutan, gibbon, rhesus, baboon, marmoset, squirrel monkey, tarsier, mouse lemur, bushbaby, treeshrew, pig, alpaca, dolphin, sheep, cow, cat, dog, panda, horse, microbat, megabat, hedgehog, shrew, elephant, rock hyrax, tenrecs, manatee, armadillo, sloth, opossum, Tasmanian devil, wallaby, platypus, turkey, chicken, zebra finch, budgerigar, lizard, painted turtle, xenopus tropicalis, coelacanth, tetraodon, fugu, nile tilapia, stickleback, medaka, Atlantic cod, zebrafish, lamprey. (B) The related sequence alignments of the above vertebrates to mouse Tbx4 genomic DNA sequence by Multiz program [12]. Most sequence alignments were omitted due to limited space. Identical nucleotide sequences are indicated by vertical black bar.

Figure 2

Generation of an embryonic/fetal lung-specific Tbx4-rtTA transgenic mouse line. (A) Schematic diagram of transgenic DNA construct. The position of the DNA probe for Southern blot is indicated. (B) Screen of transgenic founder lines by genomic DNA PCR. (C) Verification of transgenic lines by Southern blot. Genomic DNA was digested by XhoI and SalI prior to separation by gel electrophoresis. (D) Expression of rtTA-IRES-LacZ transgene in E13.5 lung was verified by X-gal staining (blue). (E) Cre-mediated mGFP expression (green) was detected in the lung of the triple transgenic mice (Tbx4-rtTA/TetO-Cre/mT-mG), but not in the double transgenic control mice (TetO-Cre/mT-mG), in which floxed-mTomato expression (red) was not affected. Dox induction was started from E6.5, and E13.5 mouse embryos were isolated and visualized under a fluorescence dissecting microscope. (F) Sagittal frozen section of E10.5 embryo of the triple transgenic mice with Dox induction from E6.5 to E10.5. Right panel shows the lung bud structure under high magnification. (G) Fluorescence microscopic examination of tissue frozen sections from E18.5 triple transgenic mice (Tbx4-rtTA/TetO-Cre/mT-mG), in which Dox induction was initiated from E6.5. mGFP was detected only in lung and tracheal mesenchyme, but not in other tissues. Dox, doxycycline; E, embryonic day.

Figure 3

Dynamic expression profile of Tbx4-rtTA-mediated Tet-On targeting system shown by the triple transgenic reporter (Tbx4-rtTA/TetO-Cre/mT-mG). (A) Tbx4-rtTA-mediated Tet-On system targeted lung mesenchyme was initiated immediately after lung morphogenesis at E9.5. Dox induction was given at different time windows of early gestation, and the sagittal frozen sections of E13.5 embryos were used to detect mGFP (green) and mTomato (red) expression. Nuclei were counterstained with DAPI (blue). Lung tissue (L) was marked on the left side of the dotted line, and vertebrae (V) were located on the right side. (B) Efficiencies of cell targeting by the Tbx4-rtTA mediated Tet-On inducible system at different prenatal and postnatal stages. The time windows of Dox induction are indicated above each panel, and the ages of the examined lung specimens are specified inside each panel. DAPI, 4',6-diamidino-2-phenylindole; Dox, doxycycline; E, embryonic day.

Figure 4

Differential targeting of lung smooth muscle cells by the Tbx4 lung enhancer at different gestational ages. (A)Tbx4 lung enhancer-mediated Tet-On induction activity was only detected in cytokeratin (Cyt)-negative cells, shown by GFP and Cyt co-immunostaining. In addition, not all GFP-positive cells were lacZ-positive, shown by GFP and LacZ co-immunostaining. The lung specimens were taken from the E18.5 triple transgenic fetuses with Dox induction time windows indicated above the panels. (B-D) Co-immunostaining of GFP, LacZ and SMA for the lungs of E18.5 (B), E15.5 (C), and E18.5 (D) triple transgenic fetuses with different Dox inductions indicated above the panels. a: airway; v: vasculature. Dox, doxycycline; SMA, α-smooth muscle actin.

Figure 5

Differential targeting of lung vascular endothelial cells by the Tbx4 lung enhancer at different gestational ages. (A) In E18.5 lung of the triple transgenic mice with Dox induction from E6.5, PECAM-1-positive endothelial cells were GFP positive, but LacZ negative. (B-C) in the E15.5 lung of the triple transgenic mice with different Dox induction time windows as indicated above the panel, vascular endothelial cells were GFP-positive if Dox induction was given prior to E10.5 (B), but LacZ expression was already negative. Consistently, Dox induction from E11.5 to E15.5 was not able to target vascular endothelial cells, shown by double negative staining for GFP and LacZ (C). V: vasculature. Dox, doxycycline; E, embryonic day; PECAM-1, platelet endothelial cell adhesion molecule.

Figure 6

Both myofibroblasts (SMA positive) and lipofibroblasts (ADRP positive) were marked by the Tbx4 lung enhancer driven Tet-On system. E18.5 lung tissue sections of triple transgenic mice (Tbx4-rtTA/TetO-Cre/mT-mG) with different Dox induction during early or late gestation as indicated were co-stained with mGFP/ADRP (A) or mGFP/SMA (B). Cell nuclei, counter-stained with DAPI (blue), were not included in the merged panels. ADRP, adipophilin; DAPI, 4',6-diamidino-2-phenylindole; Dox, doxycycline; E, embryonic day; SMA, α-smooth muscle actin.

Figure 7

NG2-positive cells were targeted by the Tbx4-lung enhancer driven Tet-On system. Lung tissue sections of the triple transgenic mice (Tbx4-rtTA/TetO-Cre/mT-mG) with different Dox induction as indicated were co-immunostained by GFP, NG2, and SMA or PECAM-1. (A) In the E18.5 lung of the triple transgenic mice with Dox induction from E6.5 to E18.5, NG2-positive cells, including SMA-positive vasculature smooth muscle cells and SMA-negative pericytes, were all marked by GFP expression, suggesting that these cells were targeted by the Tet-On system in fetuses. (B-C) With Dox induction at either mid-gestation (E11.5 to E15.5) or late gestation (E15.5 to E18.5), most NG-2 positive cells of the triple transgenic mouse lungs at E15.5 (B) or E18.5 (C) were positive for GFP. These NG2-positive cells were adjacent to PECAM-1 positive endothelial cells. Dox, doxycycline; E, embryonic day; PECAM-1, platelet endothelial cell adhesion molecule; SMA, α-smooth muscle actin.