**4. Materials and Methods**

#### *4.1. Animals and Study Design*

All animal experiments were approved by the Local Institutional Animal Care and Research Advisory committee and permitted by the Lower Saxony State Office for Consumer Protection and Food Safety (Reference number 33.14-42502-04-13/1244, approval date 11.09.2013). Male C57BL/6N mice were purchased from Charles River (Sulzfeld, Germany) at the age of five weeks. After one week of acclimatization in the local housing facility, animals were randomly assigned to four groups. Animals were housed separately under 12 h light and 12 h dark cycle and were fed ad libitum either a CD with a carbohydrate:protein:fat ratio of 70:20:10 kcal% containing 7% sucrose (D12450J, Research diets, New Brunswick, NJ, USA) [14] or a HSD with a carbohydrate:protein:fat ratio of 70:20:10 kcal% containing 35% sucrose (D12450B, Research diets) [14]. Some animals had free access to running wheels for voluntary activity resulting in four experimental groups: (1) CD (n = 10), (2) CD-A (n = 9), (3) HSD (n = 9) and (4) HSD-A (n = 7).

For all animals used in this study, basic data (body weight, energy intake, running distance, circulating plasma lipid levels and glucose homeostasis) were recently published in another context and compared with additional experimental groups by 3-Way ANOVA [14]. Therefore, body weights and fasting blood glucose concentrations of the mice in this study were subjected to the statistical

analysis used here (2-Way ANOVA and Tukey post-hoc test) and means and p-values are reported in the results section of this paper.

#### *4.2. Lung Mechanics*

After 30 weeks, mice were anesthetized with Ketamine (100 mg/kg body weight; Dr. Graeub AG, Bern, Switzerland) and Xylazin (5 mg/kg body weight; Rompun®, Bayer, Leverkusen, Germany) via intraperitoneal injection. Afterwards mice were tracheostomized and mechanically ventilated using the Flexivent small animal ventilator (SCIREQ, Montreal, QC, Canada) with a frequency of 150/min and a tidal volume of 10 mL/kg body weight. To prevent spontaneous breathing, 0.8 mg/kg body weight pancuronium bromide (Actavis®, Inresa Arztneimittel GmbH, Freiburg, Germany) was injected intraperitoneally. Three different mechanical parameters were assessed: the tissue elastance (H), the static lung compliance (CST) and the inspiratory capacity (IC) as described elsewhere [41,42]. In brief, elastance H was assessed using the broadband forced oscillation technique at a positive end-expiratory pressure (PEEP) of 3 cm H2O. Quasi-static pressure volume loops were measured to calculate CST according to the Salazar-Knowls equation and IC was determined by deep inflation of the lung at a pressure of 30 cm H2O.

#### *4.3. Lung Fixation, Sampling, and Processing*

Right lung lobes were ligated and the left lung was fixed via tracheal instillation at a hydrostatic pressure of 20 cm H2O using 1.5% paraformaldehyde (Sigma-Aldrich, St. Louis, MO, USA) and 1.5% glutaraldehyde (Merck, Darmstadt, Germany) in 0.15 M HEPES buffer (Sigma-Aldrich). Right lung lobes were snap frozen and stored at −80 ◦C. The left lung was kept in fixative for at least 24 h. The left lung volume was determined by fluid displacement (Principle of Archimedes) [43]. Afterwards, systematic uniform random sampling (SURS) was performed [44] and tissue slices were randomly allocated to light microscopy (LM) or transmission electron microscopy (TEM) analysis.

Slices for LM analysis were embedded in Technovit 7100 (Heraeus Kulzer, Wehrheim, Germany) as described previously [45]. In brief, tissue slices were osmicated for 2 h followed by an overnight incubation in a half saturated aqueous solution of uranyl acetate. After dehydration with ascending acetone concentrations, samples were embedded in Technovit 7100, 1.5 μm thick sections were cut and stained with toluidine blue.

Slices for TEM analysis were randomly subsampled into 1 mm<sup>3</sup> tissue blocks and embedded in epoxy resin (Epon®, Serva, Heidelberg, Germany) as described previously [45]. In brief, tissue blocks were post-fixed with osmium tetroxide followed by an overnight en bloc incubation with uranyl acetate. After dehydration with ascending acetone concentrations, samples were embedded in epoxy resin and 60 nm ultrathin sections were cut for analysis.

### *4.4. Stereological Analysis—Light Microscopy*

The following parameters were determined by design-based stereology at the LM level: Volume densities and total volumes of the parenchyma and non-parenchyma, the alveolar septa and the alveolar and ductal airspace as well as the surface density and the total surface area of the alveolar septa. Three to four tissue slices per animal were digitized using a slide scanner (Axio Scan.Z1, ZEISS, Oberkochen, Germany) and analyzed using the newCast software (Visiopharm, Hørsholm, Denmark). The analyst was blinded for experimental groups throughout the analysis.

Volumes were determined by point counting [44]. For estimating the parenchyma (V(par,lung)) and non-parenchyma volume (V(nonpar,lung)), random fields of view were provided by the Visiopharm software at a sampling fraction of 50% and an objective lens magnification of 5 x. The test grid consisted of 24 points and points hitting the structure of interest (parenchyma and non-parenchyma) as well as points hitting the reference space (lung tissue) were counted. Volume densities (VV) were calculated by dividing the sum of the points hitting the structure of interest by the sum of the points hitting

the reference space (Equation (1)) and total volumes (V) were calculated by multiplying the volume density by the total volume of the reference space (Equation (2)):

$$\text{PV}\_{\text{V}}(\text{struct/ref}) = \sum \text{P}(\text{struct}) / \sum \text{P}(\text{ref}) \tag{1}$$

$$\text{V(struct,ref)} = \text{V}\_{\text{V}} \text{ (struct/ref)} \times \text{V(lung)}\tag{2}$$

For estimating the septal volume (V(sept,par)) and the airspace volume (V(air,par)) within the parenchyma, fields of view obtained with a sampling fraction of 5% and an objective lens magnification of 20 x were analyzed using the same test grid as above. Volumes were calculated as shown in Equations (1) and (2) with septum or airspace representing the structure of interest and the parenchyma representing the reference space.

For estimating the alveolar (V(alvair,par)) and ductal (V(ductair,par)) airspace volumes within the parenchyma, random images were obtained at a sampling fraction of 5–10% and an objective lens magnification of 10 x using the same test system as above. Volumes were calculated as shown in Equations (1) and (2) with the alveolar or ductal airspace representing the structure of interest and the parenchyma representing the reference space.

For estimating the septal surface area within the parenchyma (S(sept,par)), random images obtained at a sampling fraction of 5% and an objective lens magnification of 20 x were analyzed. The test grid consisted of two lines and four points with a known length per point (l/p = 34.14 μm). All intersections of the test lines with the septal surface and all points hitting the parenchyma were counted [43] and the surface area was calculated as shown in Equations (3) and (4):

$$\text{S}\_{\text{V}}(\text{sept/par}) = 2 \times \sum \text{I}(\text{sept}) / (\text{l/p} \times \sum \text{P}(\text{par})),\tag{3}$$

$$\text{S(set,par)} = \text{S}\_{\text{V}} \text{ (set/par)} \times \text{V(par,lung)}.\tag{4}$$

The septal thickness (τ(sept)) was calculated as shown in Equation (5) [44].

$$\tau(\text{sept}) = 2 \times \text{V} \sqrt{\text{sept} \langle \text{par} \rangle} \text{S} \sqrt{\text{sept} \langle \text{par} \rangle} \tag{5}$$

#### *4.5. Stereological Analysis—Transmission Electron Microscopy*

The following parameters were estimated by design-based stereology at the TEM level: volume densities and total volumes of epithelial cells, endothelial cells, interstitial cells, the capillary lumen and the extracellular matrix (ECM; defined as all non-cellular spaces of the septum which includes proteoglycans, water, collagen and elastic fibers among others). Moreover, volume densities and total volumes of collagen fibers and elastic fibers within the septal ECM and of lipid droplets within interstitial cells were quantified in a separate analysis. Three tissue blocks per animal were analyzed. At least 90 random images per animal were taken according to SURS standards with a Morgagni 268 microscope (FEI, Eindhoven, Netherlands) at a primary magnification of 14,000 x. Images were analyzed using the STEPanizer stereology online tool [46] by an analyst blinded for experimental groups.

Volume densities were estimated by point counting as described above. For estimation of epithelium, endothelium, capillary lumen, interstitial cells and ECM, the test grid consisted of 16 points and for estimation of collagen fibers, elastic fibers and lipid droplets within interstitial cells the test grid consisted of 400 points. Points hitting epithelial cells (P(epi)), endothelial cells (P(endo)), capillary lumen (P(caplum)), interstitial cells (P(intcell)), ECM (P(ECM)), elastic fibers (P(elast)), collagen fibers (P(coll)) and lipid droplets (P(LD-intcell)) and points hitting the reference space (septum; P(sept)) were counted. Volume densities and total volumes were calculated as described in Equations (1) and (2).

#### *4.6. Scoring of Structural Elastic Fiber Appearance*

The structural appearance of elastic fibers was assessed by an analyst who was blinded for experimental groups. At least 90 random images at a primary magnification of 14,000 x per animal were analyzed and elastic fibers were assigned to one of three groups: (i) loosely arranged, more amorphous components, (ii) intermediate, (iii) densely packed, more electron dense, fibrillar structures (Figure 6A).

#### *4.7. Protein Islolation And Western Blot*

Lung samples were homogenized with a tissue lyser (Qiagen, Hilden, Germany), proteins were isolated using the NucleoSpin®RNA/Protein Kit (Macherey-Nagel, Düren, Germany) and protein concentration was measured with a protein quantification assay (Macherey-Nagel).

20 μg proteins per lane were loaded, fractionated by SDS-PAGE (polyacrylamide gel concentrations in Table 1) and transferred to PVDF membranes (Bio-Rad, Hercules, CA, USA). Membranes were blocked (blocking conditions in Table 1), incubated with the primary antibody, washed, incubated with the secondary antibody (antibody details in Table 1), washed and developed using the Clarity Max™ Western ECL Blotting Substrate (Bio-Rad) and the ChemiDocTM Touch Imager (Bio-Rad). Protein bands intensities were assessed with the Image LabTM Software (Bio-Rad), normalized according to the loading control and expressed as percentage of the CD mean of the respective membrane.


**Table 1.** Gel concentrations, blocking conditions, and antibodies used for Western Blot analysis.
