Association between sperm mitochondrial DNA copy number and deletion rate and industrial air pollution dynamics

The effects of air pollution on men’s reproductive health can be monitored by evaluating semen quality and sperm DNA damage. We used real-time PCR to analyse the effects of air pollution on sperm mitochondrial DNA copy number (mtDNAcn) and deletion (mtDNAdel) rates in semen samples collected from 54 men in two seasons with different levels of industrial and traffic air pollution. MtDNAdel rates were significantly higher following the high exposure period and were positively correlated with mtDNAcn. However, we did not find any difference in mtDNAcn between the two seasons. MtDNAcn was positively correlated with the DNA fragmentation index and the rates of sperm with chromatin condensation defects, previously assessed by sperm chromatin structure assay, and negatively correlated with sperm concentration, progressive motility, viability, and normal morphology. This indicates that mtDNAcn is more closely associated with male fertility than mtDNAdel rates. In contrast, mtDNAdel might be a more sensitive biomarker of air pollution exposure in urban industrial environments.

and reproductive consequences of air pollution. MtDNA of semen samples collected from men following "high" (winter season) and "low" (summer season) exposure were analysed in association with previously obtained data on semen quality and sperm chromatin integrity 5 .

Material and methods
Samples. Semen samples were obtained from 54 healthy nonsmoking municipal policemen living and working in Ostrava (Czech Republic) who were exposed to air pollutants on a daily basis while patrolling in the city streets on their regular shifts. They spent 80% of their daily working time in both winter and summer patrolling on foot. The study group consisted of nonsmokers 40.4 ± 9.4 years old (range: 21-61 years). Their reproductive and general health and factors that might have affected their semen quality were assessed with a questionnaire. The obtained data did not show any exceptional conditions or health and reproductive issues. The subjects reported only moderate alcohol consumption, no drug abuse and no home exposure to chemical toxicants. All of them reported a mixed diet, including the consumption of meat, in both seasons. The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of the Institute of Experimental Medicine, Academy of Sciences of the Czech Republic (ASCR) in Prague (approval number: 2018/09). Written informed consent was obtained from all subjects involved in the study.
As spermatogenesis lasts 72-90 days, semen samples were collected in March and October 2019 following the period of high (winter) and low (summer) air pollution exposure, respectively. The concentrations of the main air pollutants in the two seasons were detected by fixed air pollution sensors run by the Czech Hydrometeorological Institute. Concentrations of PM 2.5 , PM 10 , and NO 2 were measured hourly by automatic emission monitoring, and the results of the one-hour measurements were processed to obtain daily averages. Benzene was measured as a 24-h sample every two weeks. A 24-h sample was evaluated for the B[a]P measurements every third day. Semen analysis. Semen samples were collected by masturbation after 2-7 days of sexual abstinence and allowed to liquefy at room temperature. Standard semen parameters (semen volume, sperm concentration, motility, morphology, and viability) were assessed in accordance with the World Health Organization (WHO) guidelines 30 . Briefly, sperm counts were determined using a Bürker chamber, sperm motility was evaluated under a light microscope at 200 × magnification, and sperm viability was assessed as the percentage of sperm without plasma membrane damage detected by staining with eosin-nigrosin. The percentage of morphologically normal sperm was determined by an evaluation of 200 sperm at 1000 × magnification after staining with a Diff-Quik rapid staining kit. Strict scoring criteria described by the WHO (2010) were applied. Sperm DNA damage was analysed by the Sperm Chromatin Structure Assay (SCSA) assessing the rates of sperm with fragmented DNA (DNA fragmentation index, DFI) and sperm with high-density staining (HDS, immature sperm) as previously described 31 . Then, the semen samples were aliquoted, cooled and transported to our laboratory.
Sample processing and DNA isolation. The semen aliquots intended for mtDNA analysis were processed using GuEX buffer within 24 h. Briefly, the samples were washed in PBS, and the sediment was processed using 400 µl of GuEX buffer (50 mM guanidine hydrochloride, 10.5 mM Tris pH 8.0, 10.5 mM NaCl, 10.5 mM EDTA pH 8.0, 1 mM NaOH, pH 8.0-8.5; Sigma, St. Louis, MO, USA) and 20 µl of proteinase K (20 mg/ml) (Qiagen, Hilden, Germany) for 15 min at 37 °C. After centrifugation at 6600 rpm for 10 min, the sediment was resuspended in 200 µl of PBS. The processed sperm samples were frozen until analysis. Sperm genomic DNA was isolated using the QIAamp DNA Blood Mini Kit (Qiagen, Hilden, Germany) with 30 µl of 1 M dithiothreitol (Sigma) and stored frozen.

MtDNA copy number and deletion analysis by real-time PCR. MtDNAcn and mtDNAdel rates
were assessed using real-time PCR 18 . A single copy nuclear locus (beta-2 microglobulin, β2M), the invariable mtDNA MinorArc, and the mtDNA MajorArc comprising most of the known deletion sites (https:// www. mitom ap. org) were targeted in three separate reactions using primers displayed in Table 1.
All real-time PCR assays were performed in triplicate with 10 µl containing 1 × SYTO-9 Master Mix (Top-Bio, Prague, Czech Republic), 0.3 μM primers and 10 ng of genomic DNA. Real-time PCR was performed using the following conditions: 95 °C for 4 min and 40 cycles of 94 °C for 60 s, 55 °C for 30 s and 72 °C for 45 s on the CFX96 Touch Real-time PCR Detection System (BioRad, Hercules, CA, USA). The melting curves were assessed at 55-95 °C. The amplification efficiencies evaluated using six-point standard curves were 95% for β2M and MinorArc and 93% for MajorArc.

Results
In 2019, the first and third quarterly average concentrations of the pollutants in the whole territory of Ostrava were 32. The results of the semen and chromatin integrity (sperm chromatin structure assay; SCSA) analyses are shown in Supplementary Table S1. There were no significant differences in semen quality between the spring and autumn sampling periods, except sperm motility, which was significantly lower in autumn. The DFI and rates of HDS sperm were significantly higher in spring, following the high air pollution period 5 .
The results of the mtDNA analysis are displayed in Fig. 1 and Supplementary Table S1. We did not detect any significant difference in mtDNAcn between the spring and autumn semen collection (P = 0.247). MtDNAdel rates were significantly higher in spring, following the high exposure period (P = 0.049).
As displayed in Table 2, mtDNAcn was significantly negatively correlated with sperm concentration, progressive motility, and normal morphology rates in both collection seasons and was significantly negatively correlated with sperm viability in spring. Additionally, mtDNAcn was significantly positively correlated with the mtDNAdel rate in spring and with the DFI and HDS rates in autumn. We did not detect any correlation between mtDNAdel rates and semen parameters or chromatin integrity. Neither mtDNAcn nor mtDNAdel rates showed any significant association with age.

Discussion
Air pollution in the Ostrava industrial region consists of particulate matter, as well as gaseous pollutants. In particular, the concentrations of polycyclic aromatic hydrocarbons (PAHs) regularly exceed acceptable thresholds 21 . In 2019, the Czech Hydrometeorological Institute in Ostrava reported quarterly average concentration of B[a]P emissions in winter (4.6 ng/m 3 ) that were four times higher than the yearly national limit of 1.0 ng/m 3 (Czech law No. 369/2016).
Exposure to industrial air pollution is a known risk factor for male reproduction. It can affect sperm concentration, motility, morphology, sperm chromatin integrity and DNA methylation 1-7 . Additionally, sperm mtDNA A B www.nature.com/scientificreports/ is susceptible to induced changes; sperm mtDNAcn and mtDNAdel rates are emerging biomarkers of environmental exposure 8,9,32 . Although the responsiveness of mtDNA biomarkers to environmental exposure has previously been reported, the data are still contradictory. Higher mtDNAcn was previously detected in the peripheral blood of workers chronically exposed to PAHs 33 . In contrast, other authors observed decreased mtDNAcn in peripheral blood in association with exposure to PAHs 34,35 . Increased blood mtDNAcn was also found associated with traffic-related air pollution 36,37 and exposure to benzene 38 . However, decreased blood mtDNAcn was detected in individuals exposed to traffic-generated particulate matter in another study 39 . Moreover, regarding sperm mtDNA, decreased sperm mtDNAcn was reported after exposure to PAHs 32 , but other studies found a positive association of sperm mtDNAcn with air pollution and urinary monocarboxy-isononyl phthalate concentrations 8,9 . No significant relationships between air pollution exposure and sperm mtDNAcn and mtDNA integrity were reported by other authors 40 . Similarly, no correlation was detected between sperm MtDNA integrity and exposure to PAHs or phtalates 8,32 . In this context, also the dynamics and time course of mtDNA changes require further exploration. However, such studies, including evaluations of mtDNA characteristics in repeated samples from the same subjects, are lacking.
In the current preliminary study, we analysed successive semen samples collected from a group of healthy men to identify potential differences in sperm mtDNAcn and mtDNAdel rates between two seasons that had different levels of industrial and traffic air pollution. Our method of repeated sampling of the same men (instead of comparing different study groups) allowed us to minimize any effects of internal and lifestyle factors, including smoking, diet and consumption of supplements, as well as specific gene interactions on sperm mtDNA. Additionally, we enrolled only nonsmokers in this study, and the study group was homogeneous regarding profession and time spent outdoors, moderate alcohol consumption, and absence of drug abuse or additional exposure to chemical toxicants. According to the questionnaires, the subjects did not exhibit any change in diet, lifestyle or exposure in the period covered in this study.
The sperm mtDNAcn values did not significantly differ between the two seasons. However, we detected significantly higher mtDNAdel rates following high air pollution exposure during winter. MtDNAcn was positively correlated with mtDNAdel rates in spring, negatively correlated with most semen parameters, and positively correlated with the sperm chromatin fragmentation (DFI) and HDS rates. These findings are in agreement with those of other papers reporting significantly elevated mtDNAcn in infertile men with abnormal semen parameters 21,41 . Higher sperm mtDNAcn is associated with a lower probability of achieving pregnancy within 12 months and a longer time to pregnancy 24 . Additionally, increased mtDNAdel rates were previously reported to be associated with a decline in sperm motility and fertility 20,21 . Nevertheless, such a correlation was not found in our preliminary study.
It is not surprising that the quality of sperm mitochondria and mtDNA are closely related to fertility. Mitochondria produce the energy required for sperm motility, which is one of the main features that characterize fertile sperm. However, mitochondrial biochemical activity can result in extensive ROS formation associated with increased oxidative stress and its consequences. ROS generated by mitochondrial lipid peroxidation play an important role in sperm physiology and function by inducing sperm hyperactivation, capacitation, the acrosome reaction and binding to the zona pellucida 42 . Nevertheless, increased ROS formation and oxidative stress negatively influence sperm chromatin condensation during maturation, impair sperm motility and viability, and induce DNA damage [42][43][44][45][46][47] . This and the previously reported association between mtDNAcn and ROS levels 45 can explain the correlations between mtDNAcn and semen parameters, abnormal chromatin condensation (HDS rate), nuclear DNA fragmentation (DFI), and mtDNAdel rates in our study. Considering the relative stability of mtDNAcn in the two analysed seasons, the observed increase in mtDNAdel rates in the spring can be attributed to the seasonal increase in oxidative stress resulting from high air pollution exposure in winter. However, we cannot exclude the possible roles of other factors. For example, the effects of ambient and scrotal temperature, daylight length and associated hormonal levels must be further investigated. Such factors probably play a role in the previously described natural seasonal changes in semen quality, which can be characterized by improved Table 2. Spearman's correlation of mtDNAcn and mtDNAdel rate with semen parameters and sperm chromatin characteristics following periods of high and low air pollution, adjusted for age. **Correlation is significant at the 0.01 level. *Correlation is significant at the 0.05 level. www.nature.com/scientificreports/ semen parameters in winter and spring followed by lower sperm concentrations and progressive motility rates in summer and autumn, the latter of which was observed also in this study [48][49][50][51] .

Conclusions
In this preliminary study, we showed an association between high sperm mtDNAcn and lower semen parameters, and with the sperm chromatin and mtDNA damage. Nevertheless, mtDNAcn remained constant between the two sampling periods; thus, mtDNAdel rate appears to be a more sensitive biomarker of seasonal changes in air pollution exposure in urban industrial environments.

Data availability
Data on mtDNA is contained within this article. Data on the sperm quality, DFI and HDS rates used for the correlation analysis in this study were previously published 5 , and their summary is available in Supplementary  Table S1.