Fine particulate matter (PM2.5) promotes heart oxidative stress (OS) and evokes anti-inflammatory responses observed by increased intracellular 70 kDa heat shock proteins (iHSP70). Furthermore, PM2.5 increases the levels of these proteins in extracellular fluids (eHSP70), which have proinflammatory roles. We investigated whether moderate and high intensity training under exposure to low levels of PM2.5 modifies heart OS and the eHSP70 to iHSP70 ratio (H-index), a biomarker of inflammatory status. Male mice (
Fine particulate matter (PM2.5) inhalation promotes cardiovascular injury by direct and/or indirect pathways related to oxidative stress (OS) [
The pro-oxidant state of an organism evokes the cell stress response, identified by increasing levels of stress proteins such as the 70 kDa heat shock proteins (HSP70). Intracellularly located HSP70 (iHSP70) maintain homeostasis, by avoiding the formation of toxic polypeptide aggregates that may trigger apoptosis or inflammation [
On the other hand, increasing eHSP70 levels are related to the intensity of acute exercise performed in humans [
Male (
The experimental design that characterizes the periodized exercise training and subchronic PM2.5 exposure protocol is summarized in Table
Experimental protocol of swimming exercise training in mice exposed to fine particulate matter (PM2.5).
Experimental groups | Periodized swimming exercise training protocols | Fine particulate matter (PM2.5) exposition protocol |
---|---|---|
CON | Mice maintained sedentary: 20 minutes in 30°C shallow water (2 cm), 5x/week, 12 weeks. | Mice received 10 |
MIT | Mice submitted to moderate intensity swimming training in 20 cm deep water (30°C), 5x/week, 12 weeks. Protocol started with 20 minutes of exercise. In the following weeks, the workload increased by additional weight attached in mice: in the 2nd week it increased to 1%, in the 3rd week to 2%, in the 4th week to 3%, and in the 5th week to 4%. In the following weeks, the session duration increased 10 min/week up to 60 minutes reached in the 9th week. These workload and duration remained unaltered until the 12th week. | Mice received 10 |
HIT | Mice submitted to high intensity swimming training in 20 cm deep water (30°C), 5x/week, 12 weeks. Protocol started with 20 minutes of exercise. In the following weeks, the workload increased by additional weight attached in mice: in the 2nd week it increased to 1%, in the 3rd week to 2%, in the 4th week to 3%, and in the 5th week to 4%. In the following weeks, the workload increased 1%/week up to 8% workload reached in the 9th week These workload and duration remained unaltered until the 12th week. | Mice received 10 |
PM2.5 | Mice maintained sedentary: 20 minutes in 30°C shallow water (2 cm), 5x/week, 12 weeks. | Mice received intranasal instillation of PM2.5 (5 |
MIT + PM2.5 | Mice submitted to moderate intensity swimming training in 20 cm deep water (30°C), 5x/week, 12 weeks. Protocol started with 20 minutes of exercise. In the following weeks, the workload increased by additional weight attached in mice: in the 2nd week it increased to 1%, in the 3rd week to 2%, in the 4th week to 3%, and in the 5th week to 4%. In the following weeks, the session duration increased 10 min/week up to 60 minutes reached in the 9th week. These workload and duration remained unaltered until the 12th week. | Mice received intranasal instillation of PM2.5 (5 |
HIT + PM2.5 | Mice submitted to high intensity swimming training in 20 cm deep water (30°C), 5x/week, 12 weeks. Protocol started with 20 minutes of exercise. In the following weeks, the workload increased by additional weight attached in mice: in the 2nd week it increased to 1%, in the 3rd week to 2%, in the 4th week to 3%, and in the 5th to 4%. In the following weeks, the workload increased 1%/week up to 8% workload reached in the 9th week These workload and duration remained unaltered until 12th week | Mice received intranasal instillation of PM2.5 (5 |
Mice were randomly divided into six treatment groups for 12 weeks: control (CON), moderate intensity exercise training (MIT), high intensity exercise training (HIT) (
Exercise training protocols were applied to separate groups in three conditions: sedentary, moderate intensity training, and high intensity training, performed in the following groups: control (CON: remained sedentary), moderate intensity training (MIT: periodized swimming exercise training, for 60 min, 5x/wk, 4% overload), and high intensity training (HIT: periodized swimming exercise training, for 20 min, 5x/wk, 8% overload). Groups exposed to PM2.5 received daily nasotropic instillation of 5
All animals were accustomed to the water environment prior to exercise to avoid any stress response related to the new environment and situation. The adaptation period consisted of keeping the animals for 8 min in individual swimming pool chambers (10 cm × 10 cm × 30 cm) filled with water at 31 ± 1°C (20 cm depth), for three consecutive days, without any overload, to avoid stress behaviour (dive and freeze) during exercise training sessions. Afterwards, animals were randomly assigned to each training intensity protocol according to the exercise burden imposed by a progressive lead overload (until 4% or 8% of body weight) attached to the base of the tail and were then submitted to swimming exercise. Individual swimming pool chambers (20 cm deep) avoid jump and dive behaviour and allow energy expenditure higher than 3 METs [
The pollutant used in the experiment was PM2.5, collected in a polycarbonate filter through a gravimetric collector, on the terrace of the Faculty of Medicine, University of São Paulo (USP) in São Paulo, Brazil, as previously described [
Our protocol with 5
The biometric profile was monitored before the randomization, at the 4th, 8th, and 12th weeks. Body weight was checked with a semianalytical scale. Caudal venous lactate concentrations (~25
At the end of the 12 weeks of intervention, the animals were euthanized. Hearts were dissected, weighed, freeze-clamped in liquid nitrogen, and stored for further homogenization and analysis of antioxidant activity of the enzymes superoxide dismutase (SOD) and catalase (CAT) and lipid peroxidation levels. A portion of the tissues was homogenized in potassium phosphate buffer pH 7.4 containing the protease inhibitor PMSF (phenylmethylsulfonyl fluoride, 100
Homogenates were precipitated with 10% trichloroacetic acid (TCA), centrifuged, and incubated with thiobarbituric acid (1 : 1
SOD activity was determined by inhibition of autoxidation of pyrogallol [
In a quartz cuvette, 10
Animals were decapitated 48 h after the last exercise session, and whole blood was collected in EDTA-treated tubes. The samples were then immediately centrifuged (2000 ×g, at room temperature for 15 min) to obtain plasma samples. The HSP70 plasma concentration (eHSP70) was measured by using a high-sensitivity HSPA1A-specific HSP70 ELISA Kit (ENZO Life Sciences, EKS-715) in diluted (1 : 4) plasma samples as recommended. Absorbance was measured at 450 nm, and a standard curve constructed from known dilutions of recombinant 72 kDa heat shock protein (HSP72) to allow quantitative assessment of eHSP70 plasma concentration. Quantification was done using a microplate reader (Mindray MR-96A). The intra-assay coefficient of variation was identified as being <2%. It is expected that both the inducible HSPA1A and HSPA6 (HSP70B) forms as well as the cognate HSPA8 form of HSP70 (73 kDa heat shock proteins (HSP73)) should be delivered into the extracellular space of different cell types after appropriate stressful conditions. However, only HSPA1A ELISA kits have been sufficiently tested worldwide and proved to be of enough sensitivity (pg·mL−1 range) to detect minute HSP70 quantities in culture media and sera. Additionally, previous results [
After decapitation, blood was then immediately collected into heparinized (30 IU·mL−1 final volume) tubes (for metabolite measurements) or in disodium EDTA-treated tubes (2 mg·mL−1 final volume). Haematological parameters were investigated in EDTA samples in a Horiba ABX Micros 60 haematology analyser (for quantitative cell analysis).
Statistical analysis was developed first using two-way analysis of variance (2-way ANOVA) for the analysis of PM2.5 exposure, exercise training, and interaction effects. Post hoc multiple comparisons among groups were performed with Tukey’s test. All statistical analyses were performed using GraphPad Prism 6.0 for Windows. The level of significance was set to
The body weight profile of mice (Table
Body weight of mice exposed to fine particulate matter (PM2.5) submitted to 12 weeks of exercise training.
CON | PM2.5 | MIT | MIT + PM2.5 | HIT | HIT + PM2.5 | ANOVA ( |
|
---|---|---|---|---|---|---|---|
Before | 19.0 ± 3.0 | 18.5 ± 3.1 | 18.0 ± 1.8 | 17.6 ± 2.3 | 19.0 ± 1.7 | 19.5 ± 2.2 | 0.832 |
4th week | 24.6 ± 2.2† | 24.9 ± 2.1§ | 23.5 ± 2.7† | 23.1 ± 2.3† | 24.9 ± 0.5† | 22.2 ± 4.0 | 0.414 |
8th week | 27.3 ± 1.8§ | 28.1 ± 2.7§ | 24.7 ± 2.5 |
27.1 ± 3.0§ | 25.3 ± 0.9† | 25.2 ± 3.9† | 0.214 |
12th week | 28.0 ± 3.6§ | 28.7 ± 2.8§ | 26.7 ± 2.2 |
27.9 ± 2.9§ | 26.7 ± 2.5§ | 26.0 ± 3.3† | 0.650 |
Δ (12th week − before) | 8.9 ± 4.5 | 9.1 ± 4.4 | 8.6 ± 3.2 | 8.8 ± 3.7 | 8.1 ± 3.1 | 6.4 ± 4.7 | 0.896 |
Body weight (g) expressed as mean ± standard deviation. CON: control group, received 10
Animals were submitted to 12 weeks of periodized exercise training (5x/wk) at moderate or high intensity and were exposed to low levels of PM2.5 daily. Blood lactate concentration was measured to confirm different exercise intensities and evaluated whether PM2.5 exposure influences exercise response. In the 5th week of training, when all exercised groups reached 4% of workload (see exercise training protocols in Table
Blood lactate concentration of mice exposed to fine particulate matter (PM2.5) submitted to exercise training.
CON | PM2.5 | MIT | MIT + PM2.5 | HIT | HIT + PM2.5 | ANOVA ( |
|
---|---|---|---|---|---|---|---|
5th week | — | — | 4.54 ± 0.21 | 3.90 ± 0.33 | 4.17 ± 0.22 | 4.04 ± 0.19 | 0.299 |
9th week | — | — | 4.20 ± 0.20 | 4.00 ± 0.20 | 5.13 ± 0.86 |
5.40 ± 0.36 |
0.024 |
Blood lactate concentration (mmol·L−1) expressed as mean ± SEM. CON: control group, received 10
Fasting glycaemia of mice (Table
Fasting glycaemia of mice exposed to fine particulate matter (PM2.5) submitted to 12 weeks of exercise training.
CON | MIT | HIT | PM2.5 | MIT + PM2.5 | HIT + PM2.5 | ANOVA ( |
|
---|---|---|---|---|---|---|---|
Before | 120.3 ± 6.4 | 123.6 ± 23.0 | 127.2 ± 26.2 | 116.0 ± 22.4 | 104.8 ± 37.9 | 119.2 ± 27.4 | 0.789 |
4th week | 112.5 ± 19.5 | 96.2 ± 23.9 | 85.6 ± 13.7† | 104.8 ± 23.0 | 104.0 ± 14.8 | 99.2 ± 16.6 | 0.328 |
8th week | 102.5 ± 14.7 | 93.4 ± 14.7 | 96.0 ± 19.6 | 94.8 ± 15.2 | 96.2 ± 15.1 | 109.4 ± 22.0 | 0.652 |
12th week | 91.5 ± 21.3 | 74.0 ± 10.9§ | 80.7 ± 17.0† | 80.1 ± 8.8 |
77.2 ± 15.6 | 81.6 ± 4.0 |
0.458 |
Fasting glycaemia (mg·dL−1) expressed as mean ± standard deviation. CON: control group, received 10
Haematological parameters of mice exposed to fine particulate matter (PM2.5) submitted to 12 weeks of exercise training.
CON | PM2.5 | MIT | MIT + PM2.5 | HIT | HIT + PM2.5 | ANOVA ( |
|
---|---|---|---|---|---|---|---|
WBC (103·mm−3) | 10.8 ± 2.8 | 6.8 ± 3.1 | 8.9 ± 3.3 | 7.2 ± 3.1 | 7.6 ± 1.2 | 6.6 ± 0.8 | 0.146 |
LYM 103·mm−3 (%LYM) | 8.96 ± 2.38 (83.0 ± 4.4) | 5.86 ± 2.87 (84.5 ± 1.5) | 7.56 ± 3.13 (83.4 ± 3.2) | 6.85 ± 2.68 (82.3 ± 2.4) | 6.51 ± 0.93 (84.5 ± 4.5) | 5.46 ± 0.81 (82.0 ± 3.0) | 0.214 |
MON 103·mm−3 (%MON) | 0.58 ± 0.19 (5.3 ± 0.8) | 0.31 ± 0.14 |
0.43 ± 0.12 (5.0 ± 1.2) | 0.47 ± 0.17 (5.1 ± 0.8) | 0.37 ± 0.06 (5.2 ± 0.9) | 0.35 ± 0.06 (5.1 ± 0.8) | 0.041 |
GRA 103·mm−3 (%GRA) | 1.34 ± 0.56 (12.5 ± 3.8) | 0.63 ± 0.31 |
0.97 ± 0.19 (11.6 ± 3.2) | 0.99 ± 0.30 (12.5 ± 2.5) | 0.71 ± 0.50 (10.2 ± 4.9) | 0.87 ± 0.15 (13.2 ± 3.1) | 0.047 |
RBC (106·mm−3) | 10.5 ± 1.1 | 7.92 ± 0.4 | 9.7 ± 1.3 | 8.94 ± 0.9 | 7.8 ± 1.7 | 8.31 ± 0.6 | 0.220 |
HGB (g·dL−1) | 17.4 ± 3.9 | 13.6 ± 1.5 | 15.8 ± 4.0 | 14.6 ± 2.9 | 13.3 ± 5.1 | 13.2 ± 2.2 | 0.286 |
HCT (%) | 40.2 ± 1.5 | 34.5 ± 1.5 | 39.8 ± 1.5 | 40.6 ± 4.5 | 37.1 ± 1.6 | 37.6 ± 2.9 | 0.291 |
PLT (103·mm−3) | 1436 ± 329 | 975 ± 372 | 1124 ± 364 | 1157 ± 248 | 1036 ± 575 | 906 ± 145 | 0.210 |
NEU/LYM ratio (103·mm−3) | 0.15 ± 0.05 | 0.11 ± 0.03 | 0.14 ± 0.04 | 0.15 ± 0.03 | 0.10 ± 0.06 | 0.16 ± 0.04 | 0.425 |
Haematological parameters expressed as mean ± SEM. CON: control group, received 10
We investigated heart lipid peroxidation levels and antioxidant enzyme activity to provide information about the effects of exercise training while exposed to low levels of PM2.5 on OS markers. Our data showed effects of PM2.5 exposure (
Effect of 12 weeks of exercise training combined with PM2.5 exposure on heart oxidative stress profile in mice. Lipid peroxidation levels (a), superoxide dismutase (b), and catalase activity (c) in the heart. CON = control, maintained sedentary. MIT = moderate intensity training. HIT = high intensity training. PM2.5 = exposed to PM2.5, maintained sedentary. MIT + PM2.5 = moderate intensity training exposed to PM2.5. HIT + PM2.5 = high intensity training exposed to PM2.5. PM2.5 groups received 5
Blood samples were collected to evaluate plasma eHSP70 levels as a proinflammatory biomarker. The levels of eHSP70 were not influenced by PM2.5 exposure (
Effect of 12 weeks of exercise training combined with PM2.5 exposure on plasma eHSP70, heart iHSP70, and extra-to-intracellular HSP70 ratio (H-index) levels in mice. Plasma eHSP70 (a), heart iHSP70 (b), and extra-to-intracellular HSP70 ratio (H-index) (c) levels. CON = control, maintained sedentary. MIT = moderate intensity training. HIT = high intensity training. PM2.5 = exposed to PM2.5, maintained sedentary. MIT + PM2.5 = moderate intensity training exposed to PM2.5. HIT + PM2.5 = high intensity training exposed to PM2.5. PM2.5 groups received 5
Summary of effects of 12 weeks of exercise training combined with low levels of PM2.5 exposure.
Sedentary + low levels of PM2.5 | Moderate intensity exercise training + low levels of PM2.5 | High intensity exercise training + low levels of PM2.5 | Observed effects | |
---|---|---|---|---|
Heart oxidative stress | — | Increase |
Increase |
|
eHSP70/iHSP70 (H-index) inflammatory state | — | — | Decrease |
|
Leucocyte count (granulocytes and monocytes) | Decrease## | — | — | ##Compared to sedentary animals not exposed to PM2.5 |
Exercise training can promote countless benefits to the body with highlighted effects on the cardiovascular system. The volume and intensity employed in exercise prescription are crucial to better results. Exercise clearly provides health benefits to humans that subsist even with short sessions beneath the recommended intensity and duration. However, there is no definitive “dose” of exercise that may positively impact the cardiorespiratory system, especially if exercise is performed under the influence of an inadequate air pollution environment. Thus, we investigated whether a “good dose” of exercise can be changed to a “risk dose” if it is practised while exposed to low levels of environmental air pollution. In this case, the population may not interpret a risk and consider an urban area a safe environment in which to walk or run. The major findings of our study were the following: (a) exercise training, even at low levels of environmental air pollution, may predispose to subclinical heart risk (OS); (b) heart OS mediated by particle exposure can be increased by exercise training; (c) training intensity influences OS per se and HSP balance; and (d) high intensity training can promote an anti-inflammatory stress protein profile despite environmental pollution.
First, it is important to highlight that the majority of experimental studies regarding adverse effects of PM2.5 on OS parameters (and others) are conducted with animals under rest conditions and use higher levels of particle exposure. Usually, OS is observed in experimental designs that expose mice or rats to high levels of concentrated particles [
Our study evidenced chronicity effects of interaction between exercise and PM2.5 exposure since our data represents 12 weeks of continuous exposure and exercise training. Also, biological samples (heart and blood) were extracted 48 h after the last exercise session and PM2.5 administration to avoid acute effects on OS and stress protein levels. A previous study showed that intratracheal particle administration (500
It may be questioned whether the oxidative effects observed in HIT and HIT + PM2.5 were induced by higher total energy expended during exercise than in MIT and MIT + PM2.5. Also, it may be questioned whether the oxidative effects observed in MIT + PM2.5 (in comparison to MIT) were due to distinct exercise performance. Measurements of blood lactate during exercise provide information concerning the energy necessary for execution effort in the animals of our study. A maximum lactate steady state concentration is postulated between 3% and 6% of workload in 92% of mice [
In our study, exercise training did not avoid the oxidative effects of pollution. Exercise performed before environmental pollution exposure can provide some protection by reducing minute ventilation and frequency of breaths at rest [
Exercise intensity has also been studied in terms of stress protein response. Historically, studies have been dedicated to iHSP70 analysis in cardiac or skeletal muscle after exhaustive animal protocols [
While iHSP70 represent the intracellular stress response, eHSP70 are considered molecules with immunomodulatory functions, such as chaperokines that can stimulate innate and adaptive immunity [
Since iHSP70 are anti-inflammatory proteins whose expression is highly depressed in chronic inflammatory conditions, eHSP70 operate in an opposing fashion. In fact, the extra-to-intracellular HSP70 ratio (H-index), measured in different tissues and cell types in relation to plasma or culture media, has started to be ascribed as a novel and overall index of the immunoinflammatory status of an individual [
Recently, one study showed that exercise training can decrease eHSP70 levels [
Finally, to avoid possible interpretation mistakes in eHSP70 levels and H-index, we evaluated fasting glycaemia, body weight, and haematological parameters. Since obesity, type two diabetes mellitus, and inflammation [
Our study has some limitations in terms of variables that could allow more detail about cardiovascular function and inflammatory and oxidative markers. Also, it can be pointed that larger sample sizes could possibly show that the activities of heart antioxidant enzymes CAT (
It is important to know that the sum of evidence from epidemiological, clinical, and experimental studies strongly indicates that exercise performance [
Together, these previous studies and our data indicate that acute and chronic exercise exposed to PM2.5, even at low doses of pollution, may be a risk for heart OS. Also, the intensity of exercise training is a critical parameter for exercise prescription under PM2.5 exposure to avoid adverse environmental pollution effects on an organism. Moderate exercise training under exposure to low levels of PM2.5 induces heart OS but does not modify eHSP70 to iHSP70 ratio (H-index) balance. High intensity exercise training promotes anti-inflammatory profile despite exposure to low levels of PM2.5.
Catalase
70 kDa family of heat shock proteins
Extracellular HSP70
Intracellular HSP70
eHSP70 to iHSP70 ratio
Oxidative stress
Fine particulate matter
Superoxide dismutase
Trichloroacetic acid.
The authors declare that they do not have competing financial interests.
Aline Sfalcin Mai, Analu Bender dos Santos, Renan Daniel Bueno Basso, and Lucas Machado Sulzbacher performed biometric and metabolic profiles. Analu Bender dos Santos and Lílian Corrêa Costa Beber performed experiments on OS parameters. Matias Nunes Frizzo and Renan Daniel Bueno Basso performed haematological analyses. Analu Bender dos Santos and Thiago Gomes Heck performed eHSP70 analyses. Lílian Corrêa Costa Beber and Lucas Machado Sulzbacher performed iHSP70 immunodetection. H-index was calculated by Thiago Gomes Heck. Statistical analyses were performed by Thiago Gomes Heck. Aline Sfalcin Mai, Thiago Gomes Heck, Pauline Brendler Goettems-Fiorin, and Mirna Stela Ludwig cowrote the paper. Thiago Gomes Heck and Claudia Ramos Rhoden designed the study, provided experimental advice, and helped with manuscript revision. All the authors had final approval of the submitted and published versions.
The authors are grateful to Professor Paulo Ivo Homem de Bittencourt Jr (UFRGS) for critical and insightful discussions during the preparation of the present manuscript. The authors would like to thank M. Oara, M.M. Sulzbacher, G. Wildner, and Y.H. Donato for their technical support. This work was supported by UNIJUÍ and by grants from the FAPERGS (PqG-2013-FAPERGS Process no. 002106-2551/13-5 and ARD/PPP-2014-FAPERGS Process no. 16/2551-00001196-6 to Thiago Gomes Heck), from the CNPq (Process no. 407329/2016-1 to Thiago Gomes Heck), and from the CAPES (PGCI-CAPES Process no. 88887.141981/2017-00). Analu Bender dos Santos and Lílian Corrêa Costa Beber were recipients of scholarships from CAPES.