Effect of chronic exposure to fine particulate matter on cardiac tissue of NZBWF1 mice

Epidemiological and toxicological studies have shown that inhalation of particulate matter (PM) is associated with development of cardiovascular diseases. Long‐term exposure to PM may increase the risk of cardiovascular events and reduce life expectancy. Systemic lupus erythematosus (SLE) is a chronic inflammatory disease, autoimmune in nature, that is characterized by the production of autoantibodies that affects several organs, including the heart. Air pollution ‐ which can be caused by several different factors ‐ may be one of the most important points both at the onset and the natural history of SLE. Therefore this study aims to investigate whether exposure to air pollution promotes increased inflammation and cardiac remodelling in animals predisposed to SLE. Female NZBWF1 mice were exposed to an environmental particle concentrator. Aspects related to cardiac remodelling, inflammation and apoptosis were analysed in the myocardium. Body weight gain, cardiac trophism by heart/body weight ratio, relative area of cardiomyocytes and the fibrotic area of cardiac tissue were evaluated during the exposure period. Animals exposed to PM2.5 showed increased area of cardiomyocytes, and area of fibrosis; in addition, we observed an increase in IL‐1 and C3 in the cardiac tissue, demonstrating increased inflammation. We suggest that air pollution is capable of promoting cardiac remodelling and increased inflammation in animals predisposed to SLE.

human health. PM encompasses dust, smoke and all sorts of solid and liquid materials that remain suspended in the atmosphere due to their small size. 2 PM is classified according to its size, and can be divided into: inhalable (≤ 10 μm; PM10), fine (≤2.5 μm; PM2.5) and ultrafine (≤0.1 μm; PM0.1) particles.
The smaller the size, the greater the chance of causing health problems, since it can reach the entire respiratory tract and lodge in pulmonary alveoli. 3 Thus, high concentrations of PM, especially fine (PM2.5) and ultrafine (PM0.1) particles, represent a major risk in terms of increased morbidity and mortality related to respiratory and cardiovascular diseases. [4][5][6][7] Air pollution is known to be an environmental factor that plays a key role in the development of heart disease, since pollutants can act by reducing the heart's capacity, and can thus make people susceptible to arrhythmia and cardiac arrest. 8 Investigations focusing on the cardiovascular system demonstrate that exposure of mice to PM2.5 promotes cardiac abnormalities 9 such as coronary artery fibrosis, elastosis 10 and systolic/diastolic dysfunction. 11 There is also evidence of decreased function of cardiomyocytes isolated from the exposed mice and increased expression of molecular markers of cardiac hypertrophy. In addition, among other pollutants, fine particulate matter stands out in association with cardiac arrhythmias. 12 SLE is considered as a complex and multifactorial disease with diverse clinical manifestations including joint inflammation and neuronal, renal, cutaneous, gastrointestinal and haematological alterations. Also, SLE patients may present increased risk for cardiovascular diseases. The main cardiovascular manifestations in SLE are coronary artery disease, acute myocardial infarction and peripheral vascular disease. 13 Also, autoimmune damage to vascular tissue increases the risk of atherosclerosis. 14 A systematic review composed of 28 studies showed that the risk for the development of cardiovascular diseases in lupus patients is two-fold higher in comparison to healthy subjects and it is one of the main causes of death among SLE patients. 15 Besides genetic factors, different environmental factors are being investigated as triggers for the development or aggravation of SLE in predisposed individuals. Evidence also demonstrate that exposure to inhalable particles increases airway inflammation as well as pulmonary and systemic inflammation, especially in patients with SLE. 16 However, there is a lack of studies that focus on cardiovascular outcomes, especially those regarding the exposure to air pollution in individuals predisposed to SLE.
Hypertension has a high prevalence among SLE patients, [17][18][19][20][21][22][23] however, the mechanisms are not elucidated. Factors such as age, sex, ethnics and obesity combined with immune alterations, inflammation, impairments of renin-angiotensin system and side effects of drugs, and environmental factors, such as air pollution may contribute to this association. [24][25][26][27] Renal function is also important for the long-term control of arterial pressure. However, studies that investigated the association between renal alterations and cardiovascular diseases are controversial. For example, Ward and Studenski 28 and Petrin et al 29 verified that hypertension occurs independently from lupus nephritis and more recently Shaharir and colleagues 23 observed that in a cohort of SLE patients, 53% were hypertensive, but did not present nephritis. Nevertheless, impairments in the glomerular filtration rate (GFR) and in the renal plasma flow are present in about 50% of SLE patients. 30,31 Therefore, given the increase in incidence of SLE which cannot be explained only by genetic factors, and knowing that air pollution appears as a possible factor associated with the onset of the disease, and even as a factor for aggravation, we propose in this study an experimental investigation of the effects of air pollution on the cardiovascular system of animals predisposed to lupus.

| Animals and experimental protocol
Twenty mice of the NZBWF1 strain (acquired from the Jackson Laboratory -https://www.jax.org), which develop SLE spontaneously, were used in an animal model to access the influence of long-term exposure to inhaled fine particles on health of these animals. The effects on the renal sytem were published previously. 32 In this study, the formalin-fixed heart tissue of the NZBWF1 mice was used, to evaluate specifically the influence of air pollution on SLE in the heart, using selected markers.
Young animals were used, exposed or not to pollution (PM2.5) during the 4-month period (n = 10/group). Exposiure started at 3 months of age and the groups were divided into filtered air lupus (LFA) and polluted air lupus (LPA).

| Exposure to air pollution
Animal exposure to PM2.5 was performed in the Harvard Ambient Particle Concentrator (HAPC) which concentrates ambient levels of PM2.5 by a factor of 20. 33 For example, if the ambient level of PM2.5 is 10 μg/m 3 , HAPC concentrates PM2.5 to 200 μg/m 3 . Thus, to achieve the accumulated dose of 600 μg/m 3 (equivalent to the accumulated dose that a human is exposed daily living in a city as Sao Paulo), the daily period of exposure varied according to the ambient PM2,5 level of each day. The time of exposure was calculated daily, based on the ambient PM2.5 level of each day. The formula used to determine the time of permanency into the exposure chambers was the following: The equipment is located in the Faculty of Medicine of the University of Sao Paulo, where there is high traffic of cars. The HAPC has the ability to concentrate the particulate material present around it and also has a filter where only 2.5 μm sized particles are captured and transferred to a chamber where the animals of the exposed groups are placed. In addition, the equipment also has a chamber in which animals from control group are placed and exposed to filtered air.

| Euthanasia and tissue collection
After a period of 4 months of exposure, animals were euthanized by overdose of inhalational anaesthetic isoflurane (Cristália, Brazil). After euthanasia, blood was collected, centrifuged, and the serum collected and frozen in an ultrafreezer for analysis of anti-DNA antibodies. The heart was collected and immersed in a potassium chloride solution to arrest in diastole and used for histopathological analysis and immunohistochemical assay. Cardiac trophism was assessed by the ratio of total heart weight to mouse body weight.

| Processing and histological analysis of the heart
The collected material underwent a process of fixation, dehydration, diaphanization and embedding in paraffin. After this histological sections were obtained (thickness 5 μm) and common and silanized slides were used for further analysis. The sections were stained with haematoxylin and eosin for the analysis of the cardiomyocyte area and with picro-sirius red staining for the analysis of the fibrotic area. Subsequently, the slides were scanned with the slide scanner and viewed through the Pannoramic Viewer software (3DHistech, USA). For both analyses, fifteen photos were captured per animal (×400 magnification), and the ImageJ software (Media Cybernetics) was used to perform the tests. For fibrosis analysis, the area marked in red was expressed as a percentage in relation to the non-fibrotic area of each image taken. For analysis of cardiomyocyte area, 50 cells in cross section and central nucleus were quantified from each animal.

| Immunohistochemical evaluation of inflammatory markers, endothelial growth and apoptosis
For immunohistochemical assay, the tissue was kept at a temperature of 94°C for 20 minutes in sodium citrate buffer (pH 6) for antigen retrieval. The tissue was cooled to room temperature and subsequently endogenous peroxidase was blocked using a 3% hydrogen peroxide solution in methanol. Then, blocking of nonspecific binding was performed with bovine serum albumin (BSA) 2%. After, the slides were incubated with the primary antibodies, anti-IL17 (ab214588; 1:100), anti-ascular endothelial growth factor (VEGF) (sc-7269; 1:500), anti-caspase3 (ab4051; 1:250), anti-IL1 (ab205924; 1:100) and anti-C3 (ab200999; 1:250) in a humid chamber at 4°C overnight and the following day, after washing the primary antibody, slides were incubated with peroxidase labelled secondary antibody. The labeling was revealed by the chromogen diaminobenzidine (DAB) resulting in a brown colour. Slides were counterstained with Harris haematoxylin and after finalization, the slides were scanned with a slide scanner and viewed through Panoramic Viewer software (3DHistech, USA) where fifteen photos were captured per animal (×400 magnification), and the ImageJ software (Media Cybernetics) used to perform the analysis.

| Anti-DNA quantification
For quantification of anti-DNA antibodies an indirect immunofluorescence technique was used with a NOVA Lite® dsDNA Crithidia luciliae kit (Inova Diagnostics). For this blood was centrifuged and serum collected, following the manufacturer's protocol. Goat anti-Mouse IgG (H + L) Cross-Adsorbed Secondary Antibody, Alexa Fluor 488 (Invitrogen, A-11001) was utilized at 1:100 for conjugation.

| Statistical analysis
All variables were analysed in the GraphPad Prism 6 software (©2017 GraphPad Software). The normality of distributions was calculated from the Shapiro-Wilk test. For comparison between two distribution variables Student's T test was applied (relative area of cardiomyocytes, C3 and IL17A). For comparison of two variables with non-normal distribution, the nonparametric Mann-Whitney test was applied (heart weight/body weight, relative area of fibrosis, anti-DNA, Casp3, IL1β and VEGF). Weight evolution data were analysed using the two-way ANOVA test, followed by Sidak's post hoc test. For all tests a level of 5% was considered as significant.

| Evolution of body weight
To identify whether the pollution could influence on body weight, the mice were monitored to assess gains in both LFA and LPA groups during the 4 months of exposure to air pollution. Here, we demonstrate that there was no difference in body weight between the groups (Figure 1).

| Heart weight
In parallel with body weight heart weight was analysed to check for possible cardiac atrophy or hypertrophy. At the end of 4 months of exposure, animals were euthanized and the heart was collected weighed and normalized body weight. There was no significnat difference between the LFA and LPA groups. (Figure 2).

| Analysis of the relative area of cardiomyocytes
As fan index of cardiac remodelling, we measured the relative area of individual cells. Hypertrophy of cardiomyocytes was observed, as judged by the increase in area of cardiomyocyte in the LPA group compared to the control LFA group (p = 0.0005) (Figure 3).

| Analysis of cardiac fibrosis
To evaluate collagen deposition in cardiac tissue, the relative area of fibrosis of animals in the LFA and LPA group were quantified. When comparing the groups, we observed an increase of collagen deposition in the group exposed to pollution (p = 0.0001) (Figure 4).

| Immunohistochemistry analysis
In the immunohistochemical assays analysis of inflammatory markers, endothelial growth and apoptosis area was performed No significant difference was observed between groups when the VEGF marker was analysed ( Figure 5). Although there was an apparent increase in expression of IL-17 ( Figure 6) and caspase 3 (Figure 7) in the LPA group, the results were not statistically signficiant. IL-1β, was increased in the LPA group in comparison with the LFA group (p = 0.0286) (Figure 8).
Likewise analysis of C3 complement deposition showed, an increase in the LPA compared to the LFA group (p = 0.0079) (Figure 9).

| Analysis of Anti-DNA antibodies
To verify the influence of pollution on SLE disease activity , the production of anti-DNA antibodies was analysed using a semi-quantitative method. There was no statistical difference between the two groups ( Figure 10).

| DISCUSSION
In the present study, we investigated the effects of exposure to air pollution on cardiac morphology of lupusprone NZBWF1 female mice. The exposure to PM2.5 was performed during a period of 4 months to evaluate the long-term effects of chronic exposure to air pollution. We started the exposure when animals were 3 months-old, the expected age for beginning of clinical manifestations of SLE in this model. 34 Our results corroborate pre-existing data showing that exposure to air pollution may aggravate or be a risk factor for SLE and that cardiac alterations may be accelerated due to this exposure. In our study, the daily dose of PM2.5 was estimated as 600 μg/m 3 , referred to the estimated daily dose to which citizens of big cities are exposed. 35,36 This value of 600 μg/m 3 corresponds to the accumulated dose of PM2.5 that a person is exposed to over a period of 24hs living in a city such as Sao Paulo. This dose was determined based on the formula provided by the Environmental Protection Agency 37 .
In our previous study, in which lupus-prone NZBWF1 female mice were exposed to air pollution, we observed decreased survival, increased circulating neutrophils, accelerated onset of proteinuria and increased kidney weight with expansion of the cortex. 32 Here, our aim was to evaluate morphological outcomes in the cardiovascular system using the same model. In our study, we evaluated only lupus-prone female mice because in humans SLE is more prevalent in young adult women than men (ratio of 3:1), and this gender difference proportion is even higher in children (ratio of 9:1). 38,39 Although there are studies demonstrating that air pollution contributes to an increase in body weight and a predisposition to metabolic syndrome, [40][41][42] showing a positive correlation of exposure to air pollution with increased body weight, 42,43 our results indicate absence of significant difference between the LPA group exposed and the LFA group. Some studies have shown a protective effect of oestrogen in females, resulting in resistance to weight gain, 44 which may explain why we found no increase in body mass in the exposed group. Regarding heart weight, evaluation of cardiac trophism was performed by dividing heart weight by body weight (normalization). We observed no differences between groups.
The cardiomyocyte remodelling, including hypertrophy, and features such as apoptosis and fibrosis were observed and may explain the increased risk of cardiovascular outcomes. 45 Also, the quantification of the area composed of collagen fibres demonstrated an increase in the percentage of cardiac fibrosis in animals exposed to air pollution for 4 months, corroborating previous data that demonstrate the influence of air pollution in the deposition of collagen and suggestive alterations in cardiac function. 11,35 Our results did not show increased heart weight but did show an increased surface area of cardioomyocytes in animals exposed to air pollution. Thus air pollution is able to promote hypertrophy of cardiomyocytes. This result is similar to other published studies that observed similar alterations in rodents, 35,46 and correlated with the worsening of the heart in SLE animal models -for example as shown by Jain et al 47 who also observed myocyte damage together with fibrosis deposition.
Immunohistochemistry was used to evaluate local response of inflammatory markers, cell death and growth factors in the cardiac tissue. To evaluate the dynamics of endothelial growth we used the marker VEGF which has an important role in the development of new vessels and in the increased vascular permeability. 48 In addition, serum levels of VEGF were used as an independent predictor of SLE activity. However, there are studies showing similar alterations in SLE patients. 49,50 Our results did not demonstrate significant differences in VEGF marker when comparing the two groups, diverging from studies that showed increased blood levels of VEGF associated with acute exposure to PM2.5; however, the animal model and the dose and time of exposure were different from those used in this study. 51,52 One of the inflammatory markers analysed was interleukin-17A (IL-17A), an important pro-inflammatory molecule, which is involved in recruitment of neutrophils, responses against extracellular pathogens and induction of inflammation, 53 and it is also associated with autoimmune diseases. 54 Studies have shown that IL-17 blocking can decrease SLE manifestations, 55 suggesting that IL-17 is an important molecule that needs to be investigated in the present study. Our results did demonstrate a tendency for increased expression of IL-17A in cardiac tissue, 56 in line with previous studies, also performed with animals, which suggest that this marker is involved in SLE but there was no statistically significant differences between the LPA and LFA groups.
We also analysed the interleukin IL-1β, a proinflammatory cytokine produced mainly by macrophages, monocytes and dendritic cells. 57,58 In our results, we observed increased IL-1β in mice exposed to PM2.5 when compared to control group. These data corroborate with previous studies where exposure to air pollution was directly associated with increased serum and molecular levels of IL-1β 59,60 As for the association of IL-1β with SLE ,Shin et al. observed that among several proteins IL-1β plays a crucial role in the pathogenesis of juvenile SLE. 61 C3 complement is an inflammatory mediator that is used as an important marker of SLE physiopathology. When activated, it can cause cytolysis, leukocyte recruitment and inflammation in the tissues where it is deposited. 62 We performed the analysis of C3 expression and observed increased deposition of C3 in the myocardium of animals that were exposed to air pollution. This outcome corroborates data found in the literature that shows increased deposition of immune complexes in different tissues, including cardiac tissue, when the disease is active. 63,64 We also evaluated apoptosis in the cardiac tissue of animals exposed to PM2.5 compared to controls. Apoptosis is a mechanism of programmed cell death, which in physiological conditions is a system of maintenance of cells and replacement of tissues. 65 Pathological apoptosis, in the other hand, can be triggered by harmful factors that cause cell injury. For example, PM2.5 was evidenced as an inducter of apoptosis in epithelial bronchial cells. 66 Caspase-3 is a widelly used marker in immunohistochemistry for analysis of apoptosis. 67 Serum levels of caspase-3 and caspase-9 are increased in SLE patients when compared to healthy subjects. 68 As mentioned, in the present study we analysed the expression of caspase-3 in cardiac tissue and our results demonstrated a tendency to increase although not statistically significant. Caspase-3 was evaluated because intrinsic cell death is necessary for the development of cardiomyocyte hypertrophy. 69 This outcome corroborates results described previously [70][71][72] where increased expression of caspase-3 in animals exposed to air pollution in other models.
The evaluation of cardiac impairment in animal models of spontaneous SLE (BXSB, MRL/lpr and NZBWF1) shows that all lineages develop lesions in coronary vessels, myocardial strokes characterized by cardiomyocyte necrosis and focal haemorrhage, infiltration of leucocytes and macrophages, and formation of cicatricial tissue. In our study we could not observe these alterations, probably due to the fact that our animals were younger and, therefore, the alterations observed could be regarded as an initial stage. Deposits of IgG and C3 complement are observed in the atrial and ventricular vessels of all lupus-prone lineages and, therefore, immune complexes deposition may have an important role in the subjacent pathogenesis. 73,74 So another important characteristic marker of SLE is the production of anti-DNA antibodies that is directly associated with disease activity, being used by many studies as a parameter for diagnosis. 75,76 In this way, we performed an assay to detect these antibodies in our animals to verify the plausible interference of the exposure to PM2.5 in autoantibody production. We were expecting to obtain increased levels of anti-DNA antibodies. However, we did not observe statistical differences between exposed and control groups. Both groups produced anti-DNA antibodies probably confirming the SLE prone phenotype in NZBW animals so that significant differences were more difficult to demonstrate.
In summary, in this present study, we observed a strong influence of exposure to PM2.5 on expression of inflammatory markers in the myocardium of lupus-prone animals, and increased surface area of cardiomyocytes and collagen deposition. As limitations in this study, we had a lack of material to carry out biomolecular experiments; also we could not perform an evaluation of cardiac function, although morphological alterations in heart were observed showing potential to promote such dysfunctions. However, the mechanisms involved are not completely elucidated and therefore further studies with genetic and molecular tools are necessary for a better understanding of interaction between PM2.5 and cardiac manifestations in SLE.

| CONCLUSIONS
This work demonstrates that exposure to fine particulate matter is capable of promoting cardiomyocyte hypertrophy and increase collagen deposition in animals that develop SLE spontaneously, as well as increase immunoexpression of IL-1β and C3, associating these markers with inflammation and SLE activity, respectively. These data suggest that PM2.5 is capable of intensifying disease activity and worsening associated cardiovascular effects.