**1. Introduction**

Over the last decades sugar consumption has risen dramatically and in most parts of the world sucrose—a disaccharide composed of glucose and fructose—is the most common sweetener [1]. Glucose is metabolized in every cell of the body and its metabolism is tightly regulated by insulin, whereas fructose is primarily metabolized in the liver promoting de novo lipogenesis [2]. High sucrose and/or high fructose consumption leads to persistent blood hyperglycemia and systemic disorders like metabolic syndrome (MetS) and diabetes type 2 [3].

There is evidence that the lung is a target organ for glucotoxicity-induced complications. MetS defined by hyperglycemia, hypertriglyceridemia, and hypertension among others is a risk factor for diabetes type 2, but also for lung function impairment and lung diseases like chronic obstructive pulmonary disease (COPD) and asthma [4]. In diabetic patients, a poor glycemic control leads to a progressive decline of lung function [5], and respiratory conditions including COPD and obstructive sleep apnea are frequent comorbidities of diabetes [6,7]. Moreover, high blood glucose levels were shown to affect fetal and postnatal lung development by generalized slowing of alveolar septal growth [8]. The underlying mechanistic links between hyperglycemia and lung dysfunction are not entirely understood. A better insight could help to develop novel therapeutic approaches and might identify hyperglycemia as an early risk factor for lung diseases.

Physical exercise was shown to be effective against a number of diseases including MetS and type 1 and type 2 diabetes, and to reduce blood glucose levels, triglycerides, blood pressure, or waist circumference [9,10]. Moreover, it also exerts beneficial effects on lung function [11]. Generally, exercise has an anti-inflammatory effect by induction of anti-inflammatory mediators such as IL-6, IL-1ra, TNF-R, and IL-10 [12]. In mice which were exposed to cigarette smoke for several weeks, physical exercise reduced oxidative stress in the lung and was therefore able to protect against emphysema development [13].

The present study tested the hypotheses, that: (i) prolonged excess dietary sucrose intake affects lung mechanics and structure in mice and that (ii) voluntary activity alleviates these sucrose-induced changes.

#### **2. Results**

#### *2.1. Body Weight and Blood Glucose Concentrations*

As already published by Schipke et al. [14], high sucrose diet (HSD)-fed mice had higher body weights (control diet (CD): 40.5 ± 3.6 g, HSD: 45.6 ± 2.0 g, *p* = 0.001) and elevated fasting blood glucose concentrations (CD: 8.4 ± 0.5 mmol/L, HSD: 9.8 ± 1.2 mmol/L, *p* < 0.001) in comparison to CD-fed mice after 30 weeks. Voluntary activity had no impact on the HSD-related body weight increase (CD-active (A): 39.2 ± 4.1 g, HSD-A: 43.4 ± 2.8 g, *p* = 0.045; HSD vs. HSD-A *p* = 0.176), but prevented hyperglycemia in the active HSD-group (CD-A: 8.0 ± 0.5 mmol/L, HSD-A: 8.3 ± 0.3 mmol/L, *p* = 0.373; HSD vs. HSD-A *p* < 0.001).

#### *2.2. Lung Mechanics*

HSD resulted in lower Elastance H (*p* = 0.015; Figure 1A) and higher static lung compliance (*p* = 0.012; Figure 1B) and inspiratory capacity (*p* = 0.043; Figure 1C) compared to CD. Voluntary activity did not influence the HSD-induced changes in lung mechanics; however, H was reduced in active CD-fed mice compared to non-active CD-fed mice (*p* = 0.03; Figure 1A).

**Figure 1.** Effects of high sucrose intake and voluntary activity on lung mechanics. Mice were fed a control diet (CD) or a high sucrose diet (HSD) and were left untreated or had access to running wheels (CD-A, HSD-A). Lung mechanics measurements were performed after 30 weeks. (**A**) Elastance H, (**B**) Static lung compliance, (**C**) Inspiratory capacity. Values are individual data points, with means indicated by horizontal lines. Data were compared by 2-Way ANOVA followed by Tukey test; *p*-values < 0.05 are indicated: \* *p* < 0.05, \*\* *p* < 0.01.

#### *2.3. Lung Structure*

Left lung volumes were higher in the HSD- as well as in the HSD-A-group compared to their respective controls (CD vs. HSD *p* = 0.039, CD-A vs. HSD-A *p* = 0.002, Figure 2A). This was due to an increase in the parenchyma volume (HSD vs. CD *p* = 0.043, HSD-A vs. CD-A *p* < 0.001, Figure 2B), whereas the non-parenchyma volume was not significantly altered (Figure 2C).

**Figure 2.** Effects of high sucrose intake and voluntary activity on lung and parenchyma volumes. Mice were fed a control diet (CD) or a high sucrose diet (HSD) and were left untreated or had access to running wheels (CD-A, HSD-A) for 30 weeks. (**A**) Volume of the left lung; (**B**) Volume of left lung parenchyma, (**C**) Volume of left lung non-parenchyma. Values are individual data points, with means indicated by horizontal lines. Data were compared by 2-Way ANOVA followed by Tukey test; *p*-values < 0.05 are indicated: \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001.

#### *2.4. Parenchyma Composition*

The parenchymal composition differed between inactive and active HSD-groups (Figure 3A,E). HSD alone induced a higher septal volume (CD vs. HSD *p* = 0.001, Figure 3B) and surface area (CD vs. HSD *p* < 0.001, Figure 3C) compared to controls. In contrast, the combination of activity and HSD resulted in a higher airspace volume (CD-A vs. HSD-A *p* < 0.001, Figure 3F), which was due to increases in both ductal (CD-A vs. HSD-A *p* = 0.001; Figure 3G) and alveolar (CD-A vs. HSD-A *p* = 0.002, Figure 3H) airspace, accompanied by a higher septal surface area (CD-A vs. HSD-A *p* < 0.001, Figure 3C). The thickness of alveolar septa did not differ significantly among the groups (Figure 3D).

**Figure 3.** Effects of high sucrose intake and voluntary activity on parenchyma composition. Mice were fed a control diet (CD) or a high sucrose diet (HSD) and were left untreated or had access to running wheels (CD-A, HSD-A) for 30 weeks. (**A**) Septal volume density, (**B**) Septal volume, (**C**) Septal surface area; (**D**) Septal thickness; (**E**) Airspace volume density; (**F**) Airspace volume; (**G**) Ductal airspace volume; (**H**) Alveolar airspace volume. Values are individual data points, with means indicated by horizontal lines. Data were compared by 2-Way ANOVA followed by Tukey test; *p*-values < 0.05 are indicated: \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001.

#### *2.5. Septal Composition*

The increased septal volume in HSD-fed animals was accompanied by volume increases of endothelial cells (CD vs. HSD *p* < 0.001, Figure 4A) and the capillary lumen (CD vs. HSD *p* = 0.003, Figure 4B) compared to CD. Similarly, the absolute volume of interstitial cells (mainly fibroblasts) showed a strong tendency to higher levels (CD vs. HSD *p* = 0.053, Figure 4C). In contrast, the epithelial cell volume was not significantly changed in response to HSD (Figure 4D). Voluntary activity in combination with HSD resulted in elevated absolute volumes of interstitial cells compared to CD-A (*p* = 0.005, Figure 4C). The volume of lipid droplets within interstitial cells was not significantly affected by diet or activity (Figure 4 E).

**Figure 4.** Effects of high sucrose intake and voluntary activity on septal composition. Mice were fed a control diet (CD) or a high sucrose diet (HSD) and were left untreated or had access to running wheels (CD-A, HSD-A) for 30 weeks. (**A**) Volume of septal endothelial cells; (**B**) Volume of septal capillary lumen; (**C**) Volume of septal interstitial cells; (**D**) Volume of septal epithelial cells; (**E**) Volume of lipid droplets within septal interstitial cells. Values are individual data points, with means indicated by horizontal lines. Data were compared by 2-Way ANOVA followed by Tukey test; *p*-values < 0.05 are indicated: \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001.

#### *2.6. Extracellular Matrix Composition*

Compared to CD, the absolute volumes of the extracellular matrix (ECM) (defined as all non-cellular spaces of the septum which includes proteoglycans, collagen fibers and elastic fibers among others; CD vs. HSD *p* = 0.001, Figure 5A) and of collagen fibers within the ECM (CD vs. HSD *p* = 0.009, Figure 5B) were significantly increased in HSD-fed animals. This was consistent with the higher septal volume in these animals. In contrast, the volume of elastic fibers was unchanged in the HSD-group (Figure 5C) and the ratio of elastic-to-collagen fiber volumes was reduced in HSD compared to CD (*p* = 0.003, Figure 5D).

**Figure 5.** Effects of high sucrose intake and voluntary activity on extracellular matrix composition. Mice were fed a control diet (CD) or a high sucrose diet (HSD) and were left untreated or had access to running wheels (CD-A, HSD-A) for 30 weeks. (**A**) Volume of septal extracellular matrix; (**B**) Volume of septal collagen fibers; (**C**) Volume of septal elastic fibers; (**D**) Ratio of the elastic fiber volume to the collagen fiber volume. Values are individual data points, with means indicated by horizontal lines. Data were compared by 2-Way ANOVA followed by Tukey test; *p*-values < 0.05 are indicated: \*\* *p* < 0.01.

The structural appearance of septal elastic fibers differed between animals. They appeared either loosely arranged, densely packed, or showed an intermediate phenotype (Figure 6A). Scoring revealed that elastic fibers of HSD-fed animals resembled the loose phenotype in contrast to the more densely packed elastic fibers in control mice (Figure 6B). Voluntary activity partly changed this categorization. Regarding collagen fibers, there was no morphological difference between the groups.

Next, the parenchymal protein expression of elastin and fibrillin as main components of elastic fibers, and of collagen I and collagen III as most common collagen types within the lung parenchyma (4) was assessed. The elastin expression was markedly increased to about 500% of control levels in response to HSD (CD vs. HSD *p* < 0.001, Figure 6C). Fibrillin protein expression was increased to 200% of control levels (CD vs. HSD *p* < 0.001, Figure 6D), whereas collagen I and III levels were similar to CD (Figure 6E,F). Voluntary activity significantly alleviated the HSD-induced changes in elastin (HSD vs. HSD-A *p* = 0.006, Figure 6C) and fibrillin (HSD vs. HSD-A *p* = 0.014, Figure 6D) expression.

**Figure 6.** Effects of high sucrose intake and voluntary activity on composition of elastic fibers and collagen fibers. Mice were fed a control diet (CD) or a high sucrose diet (HSD) and were left untreated or had access to running wheels (CD-A, HSD-A) for 30 weeks. (**A**) Representative electron microscopical images showing loose, intermediate or dense structural composition of elastic fibers. E, elastic fibers; C, collagen fibers. (**B**) Scoring of elastic fiber appearance. (**C**) Elastin protein expression; (**D**) Fibrillin-1 protein expression; (**E**) Collagen I protein expression; (**F**) Collagen III protein expression. (**C**–**F**) left: PVDF membranes with protein bands used for quantification, molecular marker bands are indicated; right: protein band intensity signals normalized to the respective loading control and expressed as percentage of the CD mean value of the respective membrane; values are individual data points, with means indicated by horizontal lines. Data were compared by 2-Way ANOVA followed by Tukey test; *p*-values < 0.05 are indicated: \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001.

#### **3. Discussion**

Excess dietary sucrose intake for 30 weeks resulted in hyperglycemia and lung mechanics alterations indicating reduced elasticity, higher septal volumes, and elastic fiber remodeling. Voluntary activity prevented hyperglycemia and alleviated ECM alterations in HSD-fed mice. Moreover, the parenchymal airspace volume, but not the septal volume, was increased in the HSD-A-group.

Within the lung, the ECM significantly influences the elastic properties [15]. The ECM is mainly composed of elastic fibers, collagen fibers, and proteoglycans. The molecular components are synthesized by interstitial fibroblasts and released into the interstitium, where they assemble into their fibrillar structure. Collagen and elastic fibers cooperate to create a stable, but also elastic septal architecture [15]. In HSD-fed mice, the collagen fiber volume was higher in accordance with the septal volume increase. Moreover, the protein amount of collagen I and III (as the main collagen types of the lung) within 20 μg total parenchymal protein was similar to control levels. This indicates that the amount and the composition of septal collagen fibers in HSD-fed mice are comparable to control conditions. In contrast, the volume of elastic fibers was similar between HSD- and CD-fed mice despite the higher septal volume in the HSD-group, and the ratio of elastic-to-collagen fiber volumes was reduced. Additionally, the structural appearance of the elastic fibers was different in HSD-fed mice. They appeared loosely arranged with more amorphous material and less electron dense structures in contrast to the more densely packed elastic fibers in the control group. Analysis of the parenchymal protein composition revealed that in response to HSD, the amount of elastin was markedly increased to on average 500% of control levels. Also, the fibrillin amount was higher in HSD-fed mice, although this reached only about 200% of control levels. Thus, the prolonged high dietary sucrose intake influenced the expression of the main elastic fiber components elastin and fibrillin. The altered elastin-to-fibrillin ratio might have caused a divergent composition and thus altered structural appearance of the elastic fibers. Since elastic fibers are mainly responsible for the pulmonary elastic recoil during expiration, this finding could at least contribute to the lower lung elasticity we observed upon HSD feeding.

Elastic fibers are complex structures and the largest ECM component of the lung. Their extracellular assembly is directed by the fibroblast and requires the coordinated expression of tropoelastin and microfibril components as well as enzymes essential for elastin cross-linking [16]. In other organs like the heart, hyperglycemia-induced effects on fibroblasts are well studied with upregulation of collagen expression as one main effect [17–20]. In contrast, little is known about glucotoxic effects on pulmonary fibroblasts. Glucose-stimulation of fetal rat lung explants results in greater lipid inclusions within fibroblasts [21], contradicting the unchanged lipid droplet volumes in fibroblasts of HSD-fed mice found in this study. This might be due to differences in age (fetal vs. adult) and/or experimental design (lung explant vs. living organism). In cultured human lung fibroblasts, insulin stimulates collagen synthesis [22], however, direct glucose-related effects on lung fibroblasts are not well studied. Also endothelial cells influence pulmonary ECM composition, either via nitric oxide production or by endothelial-mesenchymal transition, and may therefore contribute to the sucrose-induced changes observed in this study [23].

A functional correlation between hyperglycemia and elastic fiber remodeling could add important insight into the complex and not well understood association between diabetes and pulmonary dysfunction. Abnormal elastic fiber assembly and integrity as a result of genetic mutations is associated with an increased susceptibility for lung diseases [24] and changes in content and composition of ECM components significantly contribute to pathogenesis and progression of asthma, COPD, idiopathic pulmonary fibrosis, pulmonary arterial hypertension, and lung cancer [15,25]. On the other hand, type 1 diabetes mellitus leads to a decrease of lung elasticity [26,27]. Moreover, a poor glycemic control indicated by high HbA1c concentrations in diabetic patients results in a progressive decline in lung function [5] and diabetic individuals are at higher risk of COPD, asthma, pulmonary fibrosis, and pneumonia [28]. While diabetic microangiopathy in the lung is one explanation for these alterations, also structural changes are considered to play a major role [29,30]. In diabetic rats, the relative amounts of collagen, elastin, and basal laminae in the septum are increased [31] and diabetes induction by streptozotocin, in 3 week old rats, results in increased collagen and elastin [32]. Another study examining normally fed and undernourished diabetic rats concludes that experimental diabetes affects lung connective tissue metabolism and breakdown and thereby leads to increases in lung collagen and elastin [33]. In humans, fasting plasma glucose concentrations of 6.1–6.9 mmol/L without or with impaired glucose clearance indicate a prediabetic state, whereas a fasting plasma glucose concentration equal to or higher than 7 mmol/L is defined as diabetes [34]. This is not directly transferable to mice, as plasma glucose concentrations of 5–8 mmol/L are reported for mice under control conditions [14,35]. Severe diabetes upon streptozotocin injection is reflected by blood glucose concentrations greater than 14 mmol/L [35]. The HSD-fed mice examined here had blood glucose concentrations around 10 mmol/L, which was significantly higher than CD-fed mice (8.5 mmol/L), but below the severe diabetes conditions of other studies [35]. Moreover, glucose tolerance measured by an oral glucose tolerance test was unimpaired in the HSD-group [14], pointing to a prediabetic or early diabetic state. Thus, elastic fiber remodeling seems to be an early hyperglycemia-induced pulmonary alteration that might prime the lung for injury or disease. Moreover, the mouse model used in this study is suitable to monitor for early hyperglycemia-induced pulmonary alterations and to test therapeutic options against disease progression.

One intervention strategy already known to be effective against many metabolic disorders is physical exercise [9]. Here, voluntary activity prevented hyperglycemia in HSD-fed mice [14] in line with others showing improved blood sugar control due to physical exercise in prediabetic or diabetic individuals [36,37]. In active, HSD-fed mice the parenchymal airspace volume, but not the septal volume, was increased and the absolute volume of interstitial cells was higher compared to CD-A. Moreover, the septal amount of elastic fibers, the elastic-to-collagen fiber ratio and the protein expression of elastic fiber components was similar to control levels. This indicates that voluntary activity alleviated the HSD-induced septal remodeling processes, possibly by prevention of hyperglycemia. Although this did not result in normalization of lung mechanics, the increase in airspace volume instead of septal volume in HSD-A may point to a beneficial effect for elasticity. It was shown before that physical activity affects expression and activity of matrix metalloprotease- 2 (MMP-2) and MMP-9 and their inhibitors TIMP-1 and TIMP-2 in human muscle and tendon tissue [38,39] which are involved in degradation and stabilization of ECM components. In line with that, physical exercise has a positive effect on ECM remodeling in patients with diabetes type 2 by influencing expression of MMP-2 and its tissue inhibitor TIMP-2 [40].
