This study describes the successful generation of a new chronic obstructive pulmonary disease (COPD) animal model by repeatedly exposing mice to high concentrations of ozone.
Chronic obstructive pulmonary disease (COPD) is characterized by persistent airflow limitation and lung parenchymal destruction. It has a very high incidence in aging populations. The current conventional therapies for COPD focus mainly on symptom-modifying drugs; thus, the development of new therapies is urgently needed. Qualified animal models of COPD could help to characterize the underlying mechanisms and can be used for new drug screening. Current COPD models, such as lipopolysaccharide (LPS) or the porcine pancreatic elastase (PPE)-induced emphysema model, generate COPD-like lesions in the lungs and airways but do not otherwise resemble the pathogenesis of human COPD. A cigarette smoke (CS)-induced model remains one of the most popular because it not only simulates COPD-like lesions in the respiratory system, but it is also based on one of the main hazardous materials that causes COPD in humans. However, the time-consuming and labor-intensive aspects of the CS-induced model dramatically limit its application in new drug screening. In this study, we successfully generated a new COPD model by exposing mice to high levels of ozone. This model demonstrated the following: 1) decreased forced expiratory volume 25, 50, and 75/forced vital capacity (FEV25/FVC, FEV50/FVC, and FEV75/FVC), indicating the deterioration of lung function; 2) enlarged lung alveoli, with lung parenchymal destruction; 3) reduced fatigue time and distance; and 4) increased inflammation. Taken together, these data demonstrate that the ozone exposure (OE) model is a reliable animal model that is similar to humans because ozone overexposure is one of the etiological factors of COPD. Additionally, it only took 6 – 8 weeks, based on our previous work, to create an OE model, whereas it requires 3 – 12 months to induce the cigarette smoke model, indicating that the OE model might be a good choice for COPD research.
It has been estimated that COPD, including emphysema and chronic bronchitis, might be the third leading cause of death in the world in 20201,2. The potential incidence of COPD in a population over 40 years old is estimated to be 12.7% in males and 8.3% in females within the next 40 years3. No medications are currently available to reverse the progressive deterioration in COPD patients4. Reliable animal models of COPD not only demand the imitation of the disease pathological process but also require a short generation period. Current COPD models, including the LPS or a PPE-induced model, can induce emphysema-like symptoms5,6. A single administration or a week-long challenge of LPS or PPE to mice or rats results in marked neutrophilia in the bronchoalveolar lavage fluid (BALF), increases pro-inflammatory mediators (e.g., TNF-α and IL-1β) in the BALF or serum, produces lung parenchymal destruction-enlarged air spaces, and limits airflow5,6,7,8,9,10. However, LPS or PPE are not causes of human COPD and thus do not mimic the pathological process11. A CS-induced model produced a persistent airflow limitation, lung parenchymal destruction, and reduced functional exercise capacity. However, a traditional CS protocol requires at least 3 months to generate a COPD model12,13,14,15. Thus, it is important to generate a new, more efficient animal model that meets the two requirements.
Recently, in addition to cigarette smoking, air pollution and occupational exposure have become more common causes of COPD16,17,18. Ozone, as one of the major pollutants (though not the major component of air pollution), can directly react with the respiratory tract and damage the lung tissue of both children and young adults19,20,21,22,23,24,25. Ozone, as well as other stimulators including LPS, PPE, and CS, are involved in a serious of biochemical pathways of pulmonary oxidative stress and DNA damage and are linked to the initiation and promotion of COPD26,27. Another factor is that the symptoms of some COPD patients deteriorate after being exposed to ozone, indicating that ozone can disrupt lung function18,28,29. Therefore, we generated a new COPD model by repeatedly exposing mice to high concentrations of ozone for 7 weeks; this resulted in airflow defects and lung parenchymal damage similar to those of previous investigations30,31,32. We extended the OE protocol to female mice in this study and successfully reproduced the emphysema observed in male mice in our previous studies30,31,32. Because COPD mortality has decreased in men but increased in women in many countries33, a COPD model in females is needed to study the mechanisms and to develop therapeutic methods for female COPD patients. The applicability of the OE model to all genders lends further support to its use as a COPD model.
NOTE: The OE model has been generated and used in previously reported research30,31,32. All animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of Shanghai Jiaotong University.
1. Mice
2. Ozone or Air Exposure
3. Micro-computed Tomography
4. Treadmill Test
5. Pulmonary Function Measurements
6. BALF Collection
7. Cardiac Blood Sampling
8. Lung Morphometric Analysis
Examples of 3D µCT images of each group are displayed in Figure 1a. The ozone-exposed mice had a significantly larger total lung volume (Figure 1a and b) and LAA% (Figure 1c) than did the air-exposed control mice. The lung volume and LAA% remained elevated after six weeks of ozone exposure31,32. Increased lung volume and LAA% represent the emphysema phenotype. Examples of lung alveolar enlargement in Figure 2a illustrate the emphysema formation. An increase in the mean linear intercept (Lm) was observed in ozone-exposed mice (Figure 2b), which confirmed that lung parenchymal destruction occurred after ozone exposure.
Lung function was measured by the airflow rate parameters, shown as FEV25/FVC, FEV50/FVC, and FEV75/FVC. The results showed that all of the parameters that decreased in ozone-exposed mice (Figure 3a-c) were consistent with the typical lung functional defects in COPD patients4,37. To further evaluate the OE model, we carried out an exercise tolerance test that substituted for the 6-min walk test (6MWT)38,39, which is commonly used to assess changes in functional exercise capacity in COPD patients. The ozone exposure significantly decreased the fatigue time and fatigue distance (Figure 4a and b).
To address the pathogenesis of COPD in the OE model, the macrophages and neutrophils were counted from the BALFs of the mice; pro-inflammatory cytokines (IL-1β and TNF-α) and anti-inflammatory cytokines (IL-10) were detected in the mouse sera. The ozone-exposed mice showed a significant increase in inflammatory cells, including macrophages and neutrophils (Figure 5a-c), as well as a significant decrease in IL-10 and an increase in IL-1β and TNF-α (Figure 6a-c). All of the data demonstrated that the OE model recapitulated human COPD-like symptoms.
Figure 1.Ozone exposure increased the lung volume and LAA% detected by µCT.(a) Representative 3D images showing the lungs of air- or ozone-exposed mice at 7 weeks after the respective exposure. (b) Individual total lung volumes of the two groups were extracted from the 3D images for statistics. Ozone-exposed mice showed a significant increase in total lung volume. (c) Individual and mean LAA% of the two groups. Ozone-exposed mice showed a significant increase in LAA%. The red color indicates LAA (voxels with densities of 2,550-2,700 Hounsfield units). The trachea and bronchi shown in red in this figure were removed to calculate the lung LAA. The data are presented as the mean ± S.E.M. **P<0.01, ***P<0.001. Please click here to view a larger version of this figure.
Figure 2. Ozone exposure increased the Lm. (a) Representative micrographs of lung alveolar spaces in HE-stained sections of air-exposed or ozone-exposed mice. (b) Individual values of Lm were quantified from the lung sections of the two groups for statistics. An increase in Lm was observed in ozone-exposed mice. The data are presented as the mean ± S.E.M. **P<0.01. Please click here to view a larger version of this figure.
Figure 3. Ozone exposure decreased pulmonary function. (a-c) For both the air-exposed and the ozone-exposed mice, the individual FEV in the first 25, 50, and 75 ms of fast expiration (FEV25, FEV50, and FEV75, respectively), as well as the FVC, were all recorded. The percentages of FEV25, FEV50, and FEV75 to FVC were calculated separately. FEV25/FVC, FEV50/FVC, and FEV75/FVC all decreased significantly in ozone-exposed mice. The data are presented as the mean ± S.E.M. *P<0.05, **P<0.01. Please click here to view a larger version of this figure.
Figure 4. Ozone exposure decreased fatigue time and fatigue distance. (a) Individual running times for air- or ozone-exposed mice were recorded. Ozone-exposed mice showed a significant decrease in fatigue time. (b) Individual running distances of the two groups were recorded. Ozone-exposed mice exhibited a significant decrease in fatigue distance. The data are presented as the mean ± S.E.M. *P<0.05. Please click here to view a larger version of this figure.
Figure 5. Inflammatory cells counted from BALF. (a) Individual and mean numbers of total cells (TOTAL) in air- or ozone-exposed mice. (b) Individual and mean numbers of macrophages (MAC) in the two groups. (c) Individual and mean numbers of neutrophils (NEU) in the two groups. The data are presented as the mean ± S.E.M. *P<0.05, **P<0.01. Please click here to view a larger version of this figure.
Figure 6. Inflammatory and anti-inflammatory cytokine detection in serum. (a) Individual and mean values of the amount of IL-10 in air- or ozone-exposed mice. (b) Individual and mean values of the amount of IL-1β in the two groups. (c) Individual and mean values of the amount of TNF-α in the two groups. The data are presented as the mean ± S.E.M. *P<0.05. Please click here to view a larger version of this figure.
In this study, we present a reliable method for generating a new COPD model. Compared to other models (i.e., LPS or PPE models), this OE model recapitulates the pathological process of COPD patients. Because cigarette smoke is the main hazardous material that causes COPD in human patients40, the CS model remains the most popular COPD model41,42. However, the CS model requires a 3- to 12-month R&D period for new drugs. Compared to the CS model, the current OE model reduces the generation period to 6-8 weeks. We have observed emphysema in male mice after 6 weeks of exposure to ozone in our previous studies30,31,32. In this study, we applied the OE protocol to female mice for 7 weeks and successfully generated a female OE model. Because it has been reported that COPD mortality has decreased in men but increased in women in some countries33, it is necessary to study the pathogenic mechanisms and to develop cure approaches for female COPD patients using a female COPD model. We know that the abovementioned COPD models (i.e., LPS, PPE, and CS) can work in both male and female animals43. The aim of this work was to propose an additional COPD model that both recapitulates the pathological process of COPD patients and requires a very short generation-period.
The key step for generating this model was exposing the mice to ozone (at a level of 2.5 ppm) two times per week (once every 3 days) for 6-8 weeks (in this study, we used 7 weeks, although we did not try dose escalation). By controlling the critical parameters of exposure frequency and ozone concentration, we successfully reproduced emphysema in male C57BL/6 mice31,32, male BALB/c mice30, and female BALB/c mice (the current study). The emphysema phenotype resulted from ozone exposure, with the airflow limitation and lung parenchymal destruction similar to the changes seen in COPD patients44,45.
There are still limitations to this study. For example, because the pathogenesis of human COPD is related to the anatomy of the lungs, an ideal COPD model should be generated in animals that have a pulmonary anatomical structure similar to that of humans.Compared to large animals, small animals possess even less extensive airway branching than humans46. We must admit that it would be more meaningful to generate a COPD model in large animals. However, it is very difficult to establish large-scale models in animals. Another limitation of this model is its clinical relevance. Although it is certain that ozone can react with the respiratory tract and damage lung tissue19,22, and although there is evidence that the symptoms of some COPD patients deteriorate after exposure to ozone28, ozone is not the main cause of COPD in patients. However, we are still proposing and using this OE model because both ozone and CS cause damage to the respiratory system by inducing inflammation and lead to oxidative stress26,27. Thus, a new drug that can cure COPD in the OE model can assumably also work in the CS model and therefore potentially be developed for COPD patients.
Applications of the OE model are not restricted to deciphering the molecular and cellular mechanisms of COPD. Our two recent papers also investigated the efficacy of N-acetylcysteine (NAC)31 and NaHS32 (an exogenous donor of H2S) at treating COPD via the exogenous administration of the two drugs to the OE model. In the first study, we found a reversal of airway hyper-responsiveness and a reduction of airway smooth muscle mass after the administration of NAC. These effects might indicate the basis for potential clinical benefits of NAC in COPD patients31. In the second study of the OE model, we found that the administration of exogenous NaHS reversed lung inflammation and partially reversed the features of emphysema. Thus, with this OE model, we demonstrated in a preliminary study that NaHS could be developed as a potential drug candidate for COPD patients32. Therefore, the OE model has potential applications for both mechanism research and drug screening for COPD.
The authors have nothing to disclose.
The authors would like to express gratitude to Mr. Boyin Qin (Shanghai Public Health Clinical Center) for the technical assistance with the µCT evaluation in this protocol.
BALB/c mice | Slac Laboratory Animal,Shanghai, China | N/A | 7-to-9-week-old female BALB/c mice were used in this study. |
Individual ventilated cages | Suhang, Shanghai, China | Model Number: MU64S7 | The cages were used for housing mice in the animal facility. |
Sealing perspex-box | Suhang, Shanghai, China | N/A | The box was used to contain the ozone generator. Mice were exposed to ozone within the box. |
Electric generator | Sander Ozoniser, Uetze-Eltze, Germany | Model 500 | The device was used for generating ozone. |
Ozone probe | ATi Technologies, Ashton-U-Lyne, Greater Manchester, UK | Ozone 300 | The device was used for monitoring and controlling the generation of ozone. |
Pelltobarbitalum natricum | Sigma, St. Louis, MO, USA | P3761 | Mice were anesthetized by intraperitoneal injection of pelltobarbitalum natricum. |
Micro-Computed Tomography | GE Healthcare, London, ON, Canada | RS0800639-0075 | This device was used for acquiring images of the lung. |
Micro-view 2.01 ABA software | GE Healthcare, London, ON, Canada | Micro-view 2.01 | This device was used for reconstruct the lung and analyze volume, LAA of the lung. |
Treadmill machine | Duanshi, Hangzhou, Zhejiang, China | DSPT-208 | This machine was usd for fatigue test. |
Body plethysmograph | eSpira™ Forced Manoeuvres System, EMMS, Edinburgh, UK | Forced Manoeuvres System | This device was used to test spirometry pulmonary function. |
Ventilator | eSpira™ Forced Manoeuvres System, EMMS, Edinburgh, UK | Forced Manoeuvres System | This device was used to test spirometry pulmonary function. |
Slide spinner centrifuge | Denville Scientific, Holliston, MA, USA | C1183 | It was used to spin BALF cells onto slides. |
Wright Staining | Hanhong, Shanghai, China | RE04000054 | It was used to staining macrophages, neutrophils in the suspended BALF. |
Hemocytometer | Hausser Scientific, Horsham, PA, USA | 4000 | It was used to count cells. |
IL-1β | Abcam, Cambridge, MA, USA | ab100704 | They were used to test the respective factors in serum. |
IL-10 | Abcam, Cambridge, MA, USA | ab46103 | They were used to test the respective factors in serum. |
TNF-α | Abcam, Cambridge, MA, USA | ab100747 | They were used to test the respective factors in serum. |
Paraformaldehyde | Sigma, St. Louis, MO, USA | P6148 | The lung was inflated by 4% paraformaldehyde. |
Paraffin | Hualing, Shanghai, China | 56# | It was used to embed the lung. |
Rotary Microtome | Leica, Wetzlar, Hesse, Germany | RM2255 | It was used for sectioning the lung. |
Hgaematoxylin and Eosin (H&E) staining solution | Solarbio, Beijing, China | G1120 | H&E staining was done for morphometric analysis. |
Upright bright field microscope | Olympus, Center Valley, PA, USA | CX41 | It was used to image the H&E staining slides. |
Adobe Photoshop 12 | Adobe, San Jose, CA, USA | Adobe Photoshop 12 | It was used to count the number of alveoli on the H&E stained images. |
GraphPad prism 5 | Graphpad Software Inc., San Diego, CA | GraphPad prism 5 | It was used for data analysis and production of figures. |