All mice used in this study have been previously reported (Zhang et al., 2013; Hrycaj et al., 2018a). Mice were treated with 2 mg/ml Doxycycline (DOT Sci., #DSD43020) (in water with 2.5 mg sucrose added per ml) at the age of 8 weeks for 3 days or 2 weeks. Mice were euthanized and perfused with phosphate buffered saline (PBS) via the right ventricle. Lungs were isolated, inflated and fixed as previously reported (Hrycaj et al., 2018b). The left lung lobes were vacuum embedded in paraffin; the right superior lung lobes were embedded in OCT (Fisher Sci., #23730571); the right middle and inferior lobes were digested for fibroblast isolation; the right accessory lobes were used for protein or RNA extraction. All experiments were performed following protocols approved by the Institutional Animal Care and Use Committee (IACUC) guidelines at the University of Michigan or the University of Wisconsin-Madison.

Adult left lung lobes were fixed in 4% paraformaldehyde (PFA) in PBS overnight at 4°C then transferred to absolute MeOH through MeOH/PBS dilution series: 25, 50, 75 and 100% MeOH. Tissues then were incubated in Dent’s bleach (MeOH:DMSO:30% H₂O₂ = 4:1:1) for 2 h at room temperature to remove any coloration, and transferred to absolute MeOH for imaging on a Leica MZ125 dissecting microscope. PFA (Sigma-Aldrich, #P6148), MeOH (Sigma-Aldrich, #179337), DMSO (Sigma-Aldrich, #D2650), H₂O₂ (Sigma-Aldrich, #323381).

Hematoxylin (Fisher Sci., #SH30500D) and Eosin (Fisher Sci., #SE23) staining was performed on 7 μm paraffin lung sections. Images were captured on a Nikon Ds-Fi3 camera. Mean alveolar chord length (MACL) measurements were taken using the grid function on ImageJ 2.0, as previously described (Sajjan et al., 2009).

Analyses were performed as previously described (Hrycaj et al., 2018b). Briefly, 10 week old control and Tbx4rtTA; TetOCre; Hox5a^fa^fbbcc mice (treated with Dox from 8 to 10 weeks) were anesthetized prior to the insertion of a tracheal tube. Mechanical breathing measurements were performed at baseline to examine changes in lung function.

Paraffin sections were deparaffinized in xylenes and rehydrated in an ethanol series prior to antigen retrieval in 10 mM Sodium Citrate buffer. Cryosections were washed in PBS to remove excess OCT. Sections were blocked in 5% Normal Donkey Serum (Sigma-Aldrich, #566460) and incubated in primary antibodies in 4^oC overnight. A complete list of primary antibodies and the dilution used is provided in Supplementary Table 1. Sections were rinsed, incubated in secondary antibodies at room temperature for 2 h, with a 10 min 1 μg/ml DAPI (Thermo Sci., #62248) incubation at room temperature and mounted using ProLong Gold mountant (Fisher Sci., #P36930). Images were captured on a Nikon Eclipse Ti-U camera or on a Keyence BZ-X810 fluorescence microscope.

Total RNA was extracted from the right accessory lobe of wildtype and mutant mice using the RNeasy Mini Kit (Qiagen, #74104) and dissolved in 32 μl of DNase/RNase-free deionized water. cDNA was synthesized from 1 μg RNA using the iScript Reverse Transcription Supermix (Bio-Rad, #1708841). Quantitative real-time PCR was conducted using 2 × SYBR Green qPCR Master Mix (Fisher Sci., #4309155) on a StepOnePlus™ Real-Time PCR System Machine (Fisher Sci., #4376600). Threshold cycles (Ct) in target gene expression were calculated and compared to Ct values of house-keeping gene β-actin. Primers for quantitative reverse transcription PCR (RT-qPCR) are listed in Supplementary Table 2.

Lung right accessory lobes were lysed in radioimmunoprecipitation assay buffer (50 mM Tris⋅HCl, pH 7.2, 150 mM NaCl, 0.1% Triton X-100, 1% sodium deoxycholate, 5 mM EDTA) containing Complete Mini Protease Inhibitor Mixture (Sigma-Aldrich, #11873580001), and extracts were cleared by centrifugation at 20,000× g for 30 min at 4°C. Total protein content was assessed using the Pierce BCA protein assay kit (Thermo Sci. #23227) and analyzed by SDS/PAGE after boiling in Laemmli sample buffer (Bio-Rad, #1610737). Proteins were transferred to low-fluorescence polyvinylidene fluoride (Cytiva, #10600022), blocked in 3% BSA with sodium azide, and probed with primary antibodies as indicated in Supplementary Table 1. Fluorescence signals were detected on an Azure imaging system or a LI-COR Odyssey Fc imaging system. Western blot densitometry analysis was quantified using ImageJ 2.0.

Following perfusion with PBS and lung isolation, the trachea and proximal airway were removed, and the right middle and right inferior lung lobes were minced and digested in 2 mg/ml Collagenase I (Gibco, #17100017) and 3 mg/ml Dispase (Gibco, #17105041) for 2 h at 37°C. This cell suspension was incubated with red blood cell lysis buffer on ice and filtered through a 100 μm nylon cell strainer (Fisherbrand, #22363549). Cells were then washed twice in fresh DMEM/F12 media (Gibco, #11320033). Cells were plated in 6-well tissue culture plates using DMEM/F12 media supplemented with 10% FBS (Gibco, #10437010) and 1% 10,000 U/ml Penicillin-Streptomycin (Gibco, #15140122). Media was changed every third day.

Assays were performed as previously described (Hrycaj et al., 2018a). Non-tissue culture treated polystyrene 96-well flat bottom microtiter plates (Denville, # T1097) were coated and incubated with bovine plasma fibronectin (Sigma-Aldrich, #F1141) at 20 μg/ml for 1 h at 37°C. Plates were then blocked with 100 μl/well of 1% BSA (Sigma-Aldrich, #A7906) in serum-free DMEM/F12 (Gibco, #11320033) for 30 min at 37°C. Fibroblasts were seeded at 10,000 cells per individual well. Plates were centrifuged (top side up) at 10 × g for 5 min to reduce the variability inherent in the settling of cells onto the plate surface and were incubated for 1 h at 37°C with 5% CO₂. Non-adherent cells were removed by centrifugation (top side down) at 48 × g for 5 min. Adherent cells remaining on the plate were fixed and stained with 1% formaldehyde (Fisher Sci., #BP228-100), 0.5% crystal violet (Sigma-Aldrich, #C6158), 20% MeOH followed by PBS washes. Individual wells were imaged on a Leica MZ125 dissecting microscope and manually quantified using ImageJ 2.0 cell counter function.

GraphPad Prism software (version 8.4.3 for macOS) was used to perform unpaired Student’s t-test analysis, with P-values less than 0.05 considered significant and P-values greater than or equal to 0.05 considered not significant.

To investigate whether Hox5 plays a role in the adult lung, we used our previously described Hoxa5 conditional allele combined with the lung mesenchyme-specific Tbx4rtTA; TetOCre driver and enacted deletion with Doxycycline (Dox) beginning at 8 weeks of age in the presence of Hoxb5/Hoxc5 loss-of-function (Hoxa5^LoxP/LoxP; Hoxb5^–/–; Hoxc5^–/–; Tbx4rtTA; TetOCre +) (referred to as Hox5 adult conditional mutants or as Tbx4rtTA; TetOCre; Hox5a^fa^fbbcc throughout) (Hrycaj et al., 2018a). Examination of the lungs 2 weeks after the initiation of Hox5 deletion (at 10 weeks of age) resulted in significantly expanded distal airspaces that are clearly visible in the whole mount lung tissue after inflation (Supplementary Figure 1). This is supported by histological examination of sections and morphological measurements of mean alveolar chord length (MACL). Tbx4rtTA; TetOCre; Hox5a^fa^fbbcc Dox-treated mutant animals demonstrate a ∼45% increase in MACL 2-weeks post-deletion and thinned septal walls compared to wild-type animals (Figures 1A,A’,C,C’,D). Additionally, adult mice carrying only the Hox5bbcc alleles in the absence of Cre only developed a moderate phenotype (∼20% increase in MACL) (Figures 1B,B’,D), demonstrating that conditional, loss of Hoxa5 function contributes directly to the adult phenotype.

Measurements of pulmonary function revealed changes in the Hox5 adult 10-week-old conditional mutants (Dox treated from 8 to 10 weeks) consistent with an emphysema-like phenotype. Lung compliance was increased as measured by increased chord compliance (Cchord), compliance at zero pressure (Cp0) and peak compliance (Cpk), with changes accompanied by a significant decrease in elastance (Figures 2A–D). Concomitantly, Hox5 adult conditional mutants exhibit increased lung volumes, demonstrated by significant increases in inspiratory capacity (IC), vital capacity (VC), forced vital capacity (FVC), and forced expiratory volume (FEV) (Figures 2E–H).

To examine whether changes occurred in distal lung cell composition, we examined epithelial, fibroblast, and endothelial cell populations in the distal lung after mesenchymal deletion of Hoxa5. We observed no significant changes in the morphology or distribution in T1α + alveolar epithelial type I cells (AECI) or Surfactant Protein C (SPC) + alveolar epithelial type II cells (AECII) (Figures 3A–C). We also observed no changes in the number or relative distribution of Platelet-derived growth factor receptor alpha (PDGFRα) + fibroblasts (Figures 3D–F), Adipocyte differentiation-related protein (ADRP) + lipofibroblasts (Figures 3G–I), or Platelet endothelial cell adhesion molecule (PECAM) + endothelial cells (Figures 3J,K) in the distal lung of the Hox5 mutants. Additionally, no changes in mRNA levels of these cell markers was observed using RT-qPCR (Supplementary Figures 2A–E).

We analyzed several key ECM components in Hox5 adult, conditional mutant distal lungs administered Dox from 8 to 10 weeks. Immunohistochemical immunofluorescence (IHC-IF) analyses with subsequent ImageJ quantification of pixel intensity demonstrated ECM phenotypes strikingly similar to those previously described with postnatal deletion (Hrycaj et al., 2018a). We observed no changes in the basement membrane component laminin, interstitial collagen 3 or fibronectin in the Hox5 adult conditional mutant lungs compared to wild-type lungs (Figures 4A–G). However, we observed a substantial decrease in elastin staining (∼42% of wild-type levels) and disruption of the integrity of the elastin network in Hox5 adult conditional mutant lungs (Figures 4H–J). This phenotype was particularly surprising as it indicates rapid loss and/or destruction of the elastin matrix that was fully established normally at 8-weeks of age and is considered very stable (Shapiro et al., 1991). RT-qPCR shows no differences of elastin mRNA level between controls and mutants, which indicates the changes in elastin structural integrity are likely an indirect consequence of other disruptions in the distal lung (Supplementary Figure 2F).

Previous studies have demonstrated that leukocytes and macrophages secrete matrix metalloproteinases (MMPs), including neutrophil elastase (MMP12), that are capable of degrading the extracellular matrix during lung development and homeostasis (Gibbs et al., 1999; Grumelli et al., 2004; Zeng et al., 2014; Morrell et al., 2020). Further, in a study using Hox5 compound null alleles (Hoxa5^{+ ⁣/−}; Hoxb5^–/–; Hoxc5^{+ ⁣/−}), mice present with an increased Th2 cells response and exacerbated lung tissue pathology in asthma models (Ptaschinski et al., 2017). In order to determine whether an increase in either leukocytes or macrophages could account for the alveolar enlargement and the decrease in elastin in our Hox5 conditional mutant lungs, we performed IHC-IF and quantifications for Cluster of differentiation 45 (CD45, also known as the leukocyte antigen) (Figures 5A–C and Supplementary Figures 3A,B), the macrophage marker F4/80 (Figures 5D–F and Supplementary Figures 3C,D), and Cluster of differentiation 68 (CD68), which labels tissue-resident macrophages (Figures 5G–I and Supplementary Figures 3E,F) immediately following Hox5 deletion (3 days post-Dox treatment initiated at 8 weeks of age). We observed a significant increase in F4/80+ and CD68+ macrophages number in Hox5 mutant distal lungs (Figures 5F,I), consistent with an increased inflammatory response following Hox5 mesenchymal deletion.

Consistent with the influx of macrophages, we observed a significant increase of MMP12, a major elastase reported to be secreted by activated lung macrophages (McGarry Houghton, 2015), in Hox5 conditional mutant lungs compared to controls 3 days after Doxycycline induction (Figures 5J–L). MMPs are shown to target both ECM components and adhesion receptors to alter cell behaviors in lung and other cancer cells (Yu and Stamenkovic, 2000; Rolli et al., 2003; Zeng et al., 2014). A previous study also suggests enhanced production of MMPs in Hoxa5^–/– juvenile mice lungs (Mandeville et al., 2006). However, by 14 days after the initiation of deletion, there was no observable differences either in the number or location of leukocytes, macrophages (Figures 5M–U) or in the expression of MMP12 (Figures 5V–X and Supplementary Figure 4) between Hox5 mutants and control lungs.

Together, these data are consistent with an acute inflammatory response in Hox5 adult mutant distal lungs after conditional Hox5 deletion in the lung fibroblasts.

Our previous work examining lung fibroblasts from postnatal Hox5 conditional mutants demonstrated that these cells exhibit reduced adhesion to fibronectin compared to controls (Hrycaj et al., 2018a). To examine whether Hox5 plays a similar role in regulating adhesion of lung fibroblasts in response to deletion at adult stages (8–10 weeks of Dox-induced deletion), we cultured primary lung cells harvested from adult control and Hox5 conditional mutant animals on fibronectin-coated plates. In vitro adhesion assays demonstrated that Hox5 adult mutant fibroblasts exhibit a ∼60% decrease in their ability to adhere to fibronectin (Figure 6A). Consistent with what we reported after postnatal loss of Hox5 function, there is a dramatic decrease in the protein expression of integrin α5, an important component of fibroblast adhesion, in mutants compared to controls (Figure 6B).