CD1 mice (Charles River, Calco, Italy) were housed in a thermostatically maintained animal house (T=22°C±2°C; relative humidity 55%±5%) with a 12 hours light cycle, free food access (Italiana Mangimi) and water ad libitum. Mating was arranged by caging females with one male of proven fertility overnight. The morning with evident vaginal plugs was considered day 0 of gestation (E0). Pregnant mice were euthanized by CO₂ inhalation and uteri were removed and kept in ice-cold phosphate-buffered saline (PBS) 0.1 M pH 7.4 for dissection and explant of conceptuses at different developmental stages (from E12 to E18). Since respiratory system, liver and pancreas begin to appear in their peculiar morphology at E12 [17], this developmental stage was selected as earliest time point for the subsequent described analyses. The animal protocol was approved by the Ministry of Health – Department for Veterinary Public Health, Nutrition and Food Safety Committee. Animals were treated humanely and in compliance with procedures that alleviate suffering.

Samples were processed as previously reported [14, 18, 19]. Eexcept E12 embryos, processed as a whole, mouse conceptuses were subdivided in head, upper limb and lower limb portions, 4% paraformaldehyde immersion-fixed, dehydrated with ascending grades of ethanol, cleared in xylene, and paraffin embedded. Tissue sections (4 µm thick) were stained with hematoxylin-eosin (H&E) and Masson’s trichrome for histological examination [20].

To evaluate TLR4 and TLR7 expression, immunohistochemistry was performed as previously described [18]. Antigen retrieval was performed on deparaffinized sections by autoclave (6 minutes at 120°C in Na citrate buffer 0.01 M pH 6). Sections were incubated with a solution of 0.3% H²O² in PBS for endogenous peroxidase activity quenching. Non-specific binding site saturation was performed for 30 minutes with a solution of 0.05 M Tris-HCl, 0.15 M NaCl, 0.1% gelatin, 0.5% ovalbumin, 0.05% Tween-20, and 0.2% fish gelatin (all from Merck). Sections were then incubated with polyclonal rabbit anti-mouse TLR4 (dilution 1:50; overnight at 4°C; Novus Biologicals) or rabbit anti-mouse TLR7 (dilution 1:100; 1 hour at 37°C; Imgenex). Goat anti-rabbit antibody (dilution 1:100; 1 hour at room temperature, Sigma-Aldrich), rabbit peroxidase anti-peroxidase antibody (PAP; DAKO, dilution 1:100; 1 hour at room temperature) were utilized as secondary antibodies. Liquid DAB substrate chromogen system (DAKO; Agilent) were used to develop the reaction. To evaluate the distribution and localization of neurons and nerve fibers, β-tubulin (mouse anti mouse-β3-tubulin, dilution 1:200; 1 hour at 37°C; Santa Cruz Biotechnology) immunohistochemistry was performed. Slides were then dehydrated through a series of alcohol solutions in ascending concentrations, counterstained with hematoxylin, mounted with Entellan (Merck). Samples were observed under a Nikon Eclipse 80i microscope (Nikon) equipped with a digital Nikon DS-L1 camera. The omission of the primary antibody, the sequential omission of the secondary antibody and PAP, or the incubation with DAB alone to exclude the possibility of non-suppressed endogenous peroxidases served as internal controls. Immunohistochemical experiments were performed on at least ten paraffin sections for each sample.

The observation of sections obtained from mouse conceptuses at different developmental stages confirmed the morphological preservation of samples and showed the regions considered for TLR4 and TLR7 immunohistochemical analysis of the respiratory system, liver and pancreas. (Supplementary Figs. 1, 2).

During the development of the respiratory system, TLR4 immunostaining begins to be detectable at E17 in the cytoplasm of cells belonging to the tracheal pseudostratified epithelium (Fig. 1A, asterisk) and in tracheal smooth muscle cells (Fig. 1A, arrows). At the same time-point, TLR4 immunoreactivity was also found in intrapulmonary bronchi (Fig. 1B, C, arrows) and in small ganglia located in the wall of intrapulmonary bronchi (Fig. 1C, thick arrow and Fig. 1D, arrow). Moreover, TLR4 positive cells were observed in bronchioles and in nerve fibers running in the pulmonary parenchyma (data not shown). At E18 stage, TLR4 expression remains unchanged (data not shown).

A completely different spatiotemporal pattern of expression was observed when analysing the liver. Starting from E12, TLR4 appeared to be localized in cells of hepatoblast derivation, particularly abundant at the periphery of the organ (Fig. 2A). Cells differentiated from erythroblast progenitors were found TLR4 negative, while the cytoplasm of differentiating megakaryocytes was found TLR4 positive (Fig. 2A, arrow in the inset). TLR4 distribution pattern was maintained at E14 (Fig. 2B, C), while at E17 its expression is weaker (Fig. 2D).

From E12, TLR4 appeared in cells of the ventral and dorsal pancreatic primordial buds (Fig. 3A). At this embryonic stage, TLR4 showed cytoplasmatic localization in almost all the cells of primordial buds while, at E14, it can be possible to observe this TLR in the apical area of cells in the developing exocrine acini (Fig. 3B, arrows). At E17, TLR4 localization remained unaltered (Fig. 3C, arrows).

TLR7 was detectable in the developing respiratory system starting from E14, especially localized in some nerve fibers running in the lung parenchyma (Fig. 4A, arrows). At E17, TLR7 appeared also in small ganglia and nerve fibers in the smooth muscle cells of the tracheal wall (Fig. 4B, thick arrow in the inset and arrows, respectively), pulmonary parenchyma and bronchioles (Fig. 4C, thick arrows). At E18, in addition to the same structures mentioned before, TLR7 immunoreactivity was also observed in small ganglia and in nervous fibers of the smooth muscle of main bronchi (Fig. 4D).

In the developing liver, TLR7 expression pattern was similar to that observed for TLR4 at every considered embryological stage, but always exhibiting a weaker intensity. At E12, TLR7 positivity was found stronger in megakaryocytes than in cells of hepatoblast derivation (Fig. 5A, arrows in the inset indicate megakaryocytes). Starting from E14, TLR7 expression appeared also in developing hepatocytes and cholangiocytes, especially localized close to the centrilobular veins (Fig. 5B, C). Similar to TLR4, TLR7 staining became weaker by E17 (Fig. 5D).

In developing pancreas, TLR4 and TLR7 expression was detected in cells of the ventral and dorsal pancreatic primordial buds at E12 (Fig. 6A). At this embryonic stage, TLR7 was evident in the perinuclear region of a few number of cells (Fig. 6A). At E14, TLR7 labelling extended to nearly all cells of the developing pancreas, appearing more evident in the region of ducts and endocrine parenchyma, such as the islets of Langerhans than in the region of exocrine acini (Fig. 6B, arrow and dotted line, respectively). At E17, TLR7 immunoreactivity remained clearly visible only in the cells forming the excretory ducts and islets of Langerhans, showing a perinuclear and cytoplasmic localization (Fig. 6C, arrows and dotted line, respectively).