The effect of environmental exposure to pyrethroids and DNA damage in human sperm

Abstract

The present study was designed to investigate whether environmental exposure to pyrethroids was associated with sperm DNA damage. Between January 2008 and April 2011 286 men under 45 years of age with a normal sperm concentration of 15–300 10⁶/ml [WHO 2010] were recruited from an infertility clinic in Lodz, Poland. Participants were interviewed and provided urine, saliva, and semen samples. The pyrethroids metabolites: 3-phenoxybenzoic acid (3PBA), cis-3-(2,2-dichlorovinyl)-2,2-dimethylcyclopropane carboxylic acid (CDCCA), trans-3-(2,2-dichlorovinyl)-2,2-dimethylcyclopropane carboxylic acid (TDCCA), and cis-2,2-dibromovinyl-2,2-dimethylcyclopropane-carboxylic acid (DBCA) were analyzed in the urine using a validated gas chromatography ion-tap mass spectrometry method. Sperm DNA damage was assessed using a flow cytometry based on sperm chromatin structure assay (SCSA). A positive association was observed between CDCCA >50th percentile and the percentage of medium DNA fragmentation index (M DFI) and percentage of immature sperms (HDS) (p = 0.04, p = 0.04 respectively). The level of 3PBA >50th percentile in urine was positively related to the percentage of high DNA fragmentation index (H DFI) (p = 0.03). The TDCCA, DBCA levels, and the sum of pyrethroid metabolites were not associated with any sperm DNA damage measures. Our results suggest that environmental pyrethroid exposure may affect sperm DNA damage measures index indicated the reproductive effects of pyrethroid exposure on adult men. In view of the importance of human reproductive health and the widespread usage of pyrethroids, it is important to further investigate these correlations.

• DNA damage
• environmental exposure to pyrethroids
• fertility
• sperm

Introduction

Pyrethroids are a group of synthetic (man-made) chemicals used as pesticides in a variety of locations including commercial, agricultural, homes communities, restaurants, hospitals, schools, and as a topical head lice treatment [ASTDR 2003; USEPA 2006a; 2006b]. Non-occupational exposure to pyrethroids is thought to occur mainly via residues in the diet and through inhalation or ingestion of contaminated household dust after indoor application.

Pyrethroids are chemicals that are very similar in structure to the pyrethrins (natural insecticides derived from the chrysanthemum plant), but are often more toxic to insects, as well as mammals, and last longer in the environment than pyrethrins. Many pyrethroids have also been linked to disruption of the endocrine system, which can adversely affect reproduction and sexual development, interfere with the immune system, and increase chances of breast cancer [Garey and Wolff 1998].

Despite the wide scale use and human exposure, relatively little is known about the human reproductive health effects of environmental exposure to pyrethroids, especially sperm DNA damage. Accumulating evidence suggests that sperm DNA damage may result in male infertility regardless of the number, motility, and morphology of spermatozoa [Agarwal and Said 2003; Guzick et al. 2001]. The natural causes of human sperm DNA damage remain to be determined although a number of human and animal studies have postulated the adverse effect of environmental chemicals on DNA damage [Ji et al. 1997]. These studies and others have demonstrated the potential for adverse effects of pyrethroids on semen parameters including motility, sperm motion parameters, and sperm concentration [Ji et al. 2011; Lifeng et al. 2006; Meeker et al. 2008; Perry et al. 2007; Toshima et al. 2012; Xia et al. 2008].

The relationship between environmental exposure to pyrethroids and DNA damage in human sperm has been previously investigated only in two studies [Ji et al. 2011; Meeker et al. 2008]. Meeker et al. [2008] noticed that among the comet assay measures, 3PBA and CDCCA were associated with increased sperm DNA damage, measured as percent DNA in the comet tail. Moreover a significant positive correlation between urinary 3PBA level and sperm DNA fragmentation was observed in the study among Chinese men environmentally exposed to pyrethroids [Ji et al. 2011].

The present study adds to the previous human sperm DNA damage studies of non-occupational pyrethroid exposure more statistical power, by including additional pyrethroid metabolites in the analysis (four different pyrethroid metabolites: CDDCA, TDDCA, 3PBA, and DBCA), by expanding outcome measures to include the assessment of DNA fragmentation index (DFI), medium (M DFI) and high DNA fragmentation index (H DFI), and high DNA stainability (percentage of immature sperms, HDS), and by assessing sperm chromatin structure among a large number of subjects. To our knowledge this is the first study to assess the environmental exposure to pyrethroids evaluating four different pyrethroid metabolites, DFI (medium and high), and HDS. The present study was designed to investigate whether environmental exposure to pyrethroids was associated with sperm DNA damage measures: DFI, M DFI, H DFI, and HDS.

Results

Study population

The study population consisted of 286 men who attended infertility clinics for diagnostic purposes. The mean age of men participating in this study was 32.2 years. Most had higher (42.7%) or secondary (38.1%) education, while about 19% had only vocational education (Table 1). A total of 68% of the men examined had a BMI ≥25 kg/m². Most of the participants were nonsmokers (73%) (Table 1). Past diseases which may have had an impact on semen quality (e.g., mumps, cryptorchidism, testes surgery, testes trauma) were reported by 13% of participants. The abstinence period before the semen analysis was about 5 days (Table 1). The duration of the couple’s infertility lasted from 1 to 2 years was 39.2% and from 2–3 years, 33.9%. As summarized in Table 1, most of the men in the study consumed 1–3 alcohol drinks per week (51%).

Table 1. Characteristics of the study population.

Characteristics	N (%)
Education
Primary and vocational	55 (19.23)
Secondary	109 (38.11)
Higher	122 (42.66)
Smoking determined by cotinine level
No	210 (73.43)
Yes	76 (26.57)
BMI (kg/m^2)
<25	91 (31.82)
≥25	195 (68.18)
Mean (SD)	27.2 ± 3.7
Median (min–max)	27.3 (18.3–39.5)
Duration of couple’s infertility (years)
1--2	112 (39.16)
2--3	97 (33.92)
4--5	39 (13.63)
>5	38 (13.29)
Past diseases, which may have impact on semen quality*
No	248 (86.71)
Yes	38 (13.29)
Abstinence [days]
<3	14 (4.90)
3--7	212 (74.13)
>7	17 (5.94)
Missing data	43 (15.03)
Mean (SD)	5.0 ± 2.3
Median (min–max)	5.0 (0.0–20.0)
Age (years)
Mean (SD)	32.22 (4.45)
Median (min–max)	31.70 (22.01–44.26)
Alcohol use
None or <1 drink/week	94 (32.87)
1–3 drinks/week	146 (51.05)
Everyday	46 (16.08)

BMI: body mass index; *Past diseases which may have impact on semen quality - mumps, cryptorchidism, testes surgery, testes trauma

DNA damage and main semen parameters among study participants

Table 2 presents the sperm characteristics and sperm DNA damage among the study subjects. The semen quality among the study participants were in the normal range of the WHO [2010] semen quality indicators. The mean semen concentration was 53.6 million (10⁶)/ml (SD 52.4;median 33.6) (Table 2). The mean percentage of motile sperm cells was 57 (SD 19.8) (median 54.0). The percentage of abnormal morphology was 47% (SD 19.7) (median 46.0). The mean percentage of DNA fragmentation index was 16.2% (SD 11.0%) (median 13.3%), M DFI 8.5% (SD 7.5%) (median 6.5%), H DFI 7.9% (SD 6.6%) (median 6.0%), and HDS was 8.8% (SD 4.3%) (median 8.3%) (Table 2).

Table 2. Sperm DNA damage and sperm quality among study participants.

	Percentage distribution of sperm characteristics
Sperm quality	25%	50%	75%	95%	Max	Mean	SD
Main semen parameters:
Concentration (mln/ml)	22.5	33.6	85.77	125	360.0	53.6	52.4	286
Motility (%)	46	54.0	66	83	99.0	57.4	19.8	286
Sperm with abnormal morphology [%]	31.22	46.0	69.51	75.25	96.0	47.2	19.7	286
Chromatin:
DFI (%)	8.39	13.3	19.82	35.66	71.2	16.2	11.0	286
M DFI (%)	4.23	6.5	9.56	21.13	60.5	8.5	7.5	286
H DFI (%)	3.46	6.0	10.16	19.73	44.5	7.9	6.6	286
HDS (%)	5.98	8.3	10.90	16.94	30.7	8.8	4.3	286

DFI: DNA fragmentation index; M DFI: medium DNA fragmentation index; H DFI: high DNA fragmentation index; HDS: high DNA stainability index

Pyrethroid metabolite levels

Table 3 summarizes the unadjusted and adjusted urinary pyrethroid metabolite concentrations. The pyrethroid metabolite with the highest geometric mean concentration was 3PBA (0.17 µg/l, 0.16 µg/g creatinine), followed by TDCCA (0.16 µg/l, 0.15 µg/g creatinine), CDCCA (0.12 µg/l, 0.11 µg/g creatinine), and DBCA (0.05 µg/l, 0.04 µg/g creatinine) (Table 3). The sum of pyrethroid metabolites (CDCCA + TDCCA + DBCA + 3PBA) was 0.51 µg/l, 0.50 µg/g creatinine. Pyrethroid metabolite groups for CDCCA and TDCCA were moderately correlated with each other (Kendall correlation coefficient = 0.63) and with 3PBA (Kendall correlation coefficient = 0.52 and 0.62, respectively). No correlations were found between CDCCA, TDCCA, 3 PBA, and DBCA (Kendall correlation coefficient = 0.13, 0.13, 0.28, respectively). The percentage of men with measurable levels of 3PBA was 71.86%, TDCCA 65.51%, CDCCA 57.97%, and DBCA 16.81%.

Table 3. Pyrethroid metabolite levels in urine.

	Percentage distribution of pyrethroid metabolities
Pyrethroid metabolite	25%	50%	75%	95%	Max	Mean ± SD	Geometric mean	N
Unadjusted (µg/l)
CDCCA	0.05	0.12	0.16	0.46	3.27	0.25 ± 0.28	0.12	286
TDCCA	0.05	0.14	0.23	0.60	9.86	0.33 ± 0.15	0.16	286
3PBA	0.05	0.16	0.23	0.51	4.24	0.30 ± 0.16	0.17	286
DBCA	0.05	0.05	0.05	0.26	3.43	0.31 ± 0.51	0.05	286
∑pyrethroids	0.11	0.32	0.88	4.76	22.33	0.22 ± 0.29	0.51	286
CR-adjusted (µg/g creatinine)
CDCCA	0.03	0.11	0.14	0.42	3.18	0.22 ± 0.24	0.11	286
TDCCA	0.03	0.14	0.20	0.56	9.83	0.35 ± 0.11	0.15	286
3PBA	0.05	0.13	0.19	0.46	4.23	0.29 ± 0.11	0.16	286
DBCA	0.03	0.03	0.04	0.24	3.37	0.28 ± 0.49	0.04	286
∑pyrethroids	0.10	0.30	0.85	4.72	22.23	0.20 ± 0.27	0.50	286

SD: standard deviation; n-number of participants; CDCCA: cis-3-(2,2-dichlorovinyl)-2,2-dimethylcyclopropane carboxylic acid; TDCCA: trans-3-(2,2-dichlorovinyl)-2,2-dimethylcyclopropane carboxylic acid; 3PBA: 3-phenoxybenzoic acid; DBCA: cis-2,2-dibromovinyl-2,2-dimethylcyclopropane-1- carboxylic acid; ∑pyrethroids: CDCCA + TDCCA + DBCA + 3PBA

The association between pyrethroids metabolites concentrations and sperm DNA damage

A positive association was observed between CDCCA >50th and the percentage of medium DFI and the percentage of immature sperm (p = 0.04, p = 0.04, respectively). The level of 3PBA >50th in urine was positively related to the percentage of high DFI (p = 0.03). The TDCCA, DBCA levels and the sum of pyrethroid metabolites were not associated with any sperm DNA damage measures (Table 4). The results were adjusted for age, smoking, alcohol, past diseases, and sexual abstinence.

Table 4. Pyrethroids metabolites concentration in urine and sperm DNA damage measures.

	CDCCA			TDCCA			3PBA				DBCA			∑pyrethroids
	coef	95%CI	p	coef	95%CI	p	coef	95%CI	p		coef	95%CI	p	coef	95%CI	p
DFI
≤50th	−0.05	−0.14–0.24	0.59	−0.07	−0.12–0.26	0.47	−0.07	−0.27–0.12	0.47	≤LOD	−0.07	−0.13–0.27	0.51	−0.02	−0.05–0.22	0.21
>50th	−0.05	−0.15–0.24	0.63	−0.06	−0.13–0.26	0.53	−0.09	−0.29–0.12	0.41	>LOD	−0.08	−0.13–0.28	0.47	−0.01	−0.05–0.13	0.25
M DFI
≤50th	0.03	−0.15–0.22	0.72	0.14	−0.01–0.35	0.07	0.02	−0.17–0.21	0.84	≤LOD	0.11	−0.09–0.31	0.28	0.06	−0.10–0.18	0.44
>50th	1.10	1.04–1.21	0.04	0.17	−0.01–0.36	0.07	0.13	−0.20–0.20	0.85	>LOD	0.13	−0.07–0.33	0.22	0.13	0.01–1.22	0.13
H DFI
≤50th	0.10	−0.14–0.34	0.40	0.07	−0.31–0.18	0.60	0.09	−0.34–1.16	0.48	≤LOD	0.03	−0.29–0.22	0.81	0.08	−0.08–1.09	0.42
>50th	0.11	−0.14–0.35	0.40	0.08	−0.33–0.17	0.52	1.07	1.04–1.52	0.03	>LOD	0.05	−0.31–0.22	0.73	0.15	0.01–0.32	0.22
HDS
≤50th	0.11	−0.25–0.13	0.13	0.02	−0.12–0.16	0.76	0.04	−0.15–0.15	0.85	≤LOD	0.04	−0.19–0.11	0.62	0.02	−0.09–0.08	0.77
>50th	1.09	1.04–1.55	0.04	0.02	−0.12–0.16	0.77	1.05	−022–1.14	0.22	>LOD	0.05	−0.20–0.10	0.53	0.03	−0.09–0.08	0.76

Statistically significant at the level 0.05; coef: β coefficient; multivariate model adjusted for: sexual abstinence, age, smoking, alcohol consumption, and past diseases; CDCCA: cis-3-(2,2-dichlorovinyl)-2,2-dimethylcyclopropane carboxylic acid; TDCCA: trans-3-(2,2-dichlorovinyl)-2,2-dimethylcyclopropane carboxylic acid; 3PBA: 3-phenoxybenzoic acid; DBCA: cis-2,2-dibromovinyl-2,2-dimethylcyclopropane-carboxylic acid; DFI: DNA fragmentation index; M DFI: medium DNA fragmentation index; H DFI: high DNA fragmentation index; HDS: high DNA stainability index; ∑pyrethroids: CDCCA + TDCCA + DBCA + 3PBA

Bold values-statistically significant values

Discussion

In the present study, we found evidence for a relationship between urinary metabolites of pyrethroid insecticides and DNA damage in sperm. These findings may be of concern due to the increased use of pyrethroid pesticides resulting in widespread exposure among the general population. The results of the study suggest that urinary metabolites of pyrethroid insecticides were positively associated with the percentage of M DFI, immature sperm (CDCCA), and the percentage of H DFI (3PBA). Whereas the TDCCA, DBCA levels and the sum of pyrethroid metabolities were not associated with any sperm DNA damage measures.

For pyrethroid exposure, animal data was consistent with adverse effects on semen volume, sperm concentration, and motility [El-Demerdash et al. 2004; Yousef et al. 2003]. Pyrethroids may affect estrogen receptors in Sertoli cells [Taylor et al. 2010] and have antiandrogenic effects by antagonizing the androgen receptor and affecting seminal vesicle weight [Zhang et al. 2008].

Two studies have thus far been aimed at detecting the relationships between internal exposure levels of pyrethroid metabolites and human sperm DNA damage in the general population [Ji et al. 2011; Meeker et al. 2008]. The results were in line with our study regarding the urinary level of 3PBA, CDCCA, and DNA damage. Meeker et al. [2008] noticed that among the comet assay measures, 3PBA and CDCCA were associated with increased sperm DNA damage. In comparison Ji and co-workers observed a significant positive correlation between urinary 3PBA level and the level of sperm DNA fragmentation [Ji et al. 2011].

Our results indicated that summed metabolite levels were not better associated with sperm DNA damage than individual metabolites. Although there have been no reports of animal studies comparing the effects of different pyrethroid insecticides on male reproductive functions, an in vitro study demonstrated different sex hormone receptor activities for different pyrethroid insecticides [Du et al. 2010], thus indicating that different pyrethroid insecticides have different reproductive effects. Additionally, the levels of DBCA and TCDDA were not related to any sperm DNA damage measures. Data on endocrine or reproductive effects specific to trans-permethrin or TDCCA found that cis-permethrin is associated with higher acute toxicity than trans-permethrin [ATSDR 2003]. No effect of DBCA on sperm DNA measures can be associated with the fact that the level of this metabolite in urine was very low and was detected only in 16.8% of samples.

The possible mechanism of pyrethroid exposure on sperm DNA damage. Pyrethroids, due to their hydrophobic nature and small molecular size, may be able to pass through the blood testis barrier and reach the sperm nucleus. Perhaps within the nucleus, pyrethroids bind to DNA through the reactive groups of its acid moiety, leading to destabilization as well as unwinding of the DNA. This could be a possible explanation for the genotoxicity [Saxena et al. 2005]. Indirect effects are also possible because pyrethroids have been shown to induce oxidative stress through the generation of reactive oxygen species (ROS) in experimental models [Giray et al. 2001; Kale et al. 1999]. It has been demonstrated that ROS is an important cause of DNA damage, which could lead to single-strand breaks and mutation [Aitken and De Iuliis 2010]. The clinical significance of sperm DNA damage lies in its association not only with natural conception rates, but also with assisted reproduction success rates [Horak et al. 2007; Zini and Libman 2006]. This might have a serious consequence on developmental outcome of the newborn [Ji et al. 1997].

The distribution of pyrethroids metabolite levels in the present study were comparable to those sampled from males in the general US population, as reported in the fourth report of the National Health and Nutrition Examination Study (NHANES) [CDC 2009], as well as in other studies [Meeker et al. 2008; Young et al. 2013]. Unadjusted CDCCA and TDCCA concentrations were similar in the current study (0.12 μg/l and 0.16 μg/l). Unadjusted median for 3PBA were lower for the current study (0.17 μg/l, respectively) compared to 0.30 μg/l in the NHANES fourth report [CDC 2009].

There are limitations to the use of non-persistent pesticide exposure biomarkers. The metabolites measured in this study were not specific to a single pesticide. In addition, on the one hand, pyrethroids and other non-persistent pesticides are rapidly metabolized and excreted, so metabolite concentrations in urine reflect exposure over hours or days preceding sample collection. On the other hand, consistent individual time-activity patterns may lead to stable metabolite concentrations over long periods of time [Meeker et al. 2008]. As spermatogenesis occurs over approximately two months [Heller and Clermont 1963] previous studies by Meeker et al. [2005] and Young et al. [2013] showed that a single urine sample can be used as a predictive measure of levels of pyrethroid pesticides metabolites over a three month average.

The men in this study were members of subfertile couples seeking infertility evaluation. However, all men in our study had normal semen parameters based on WHO [2010] classification. Although they may differ from men in the general population, there is currently no evidence showing that they would differ in ways that would alter their response to pyrethroids. Thus, our results may apply to general population samples as well. Participants were heterogeneous in their semen profiles and as indicated had normal semen parameters.

Our study had several methodological strengths. Although two studies have explored the association between pyrethroid exposure and male reproductive function, none had previously assessed the percentage of medium and high DNA fragmentation index and the percentage of immature sperm. Additionally this is the first study to assess the level of four different pyrethroid metabolites and sperm DNA damage measures. A detailed questionnaire that included demographics, medical, and lifestyle risk factors among study participants allowed us to control for confounding effects in the statistical models. Also, the relative homogeneity of the study participants (educated, white) helped reduce the chance that our findings resulted from unmeasured health, behavioral, or exposure factors. This homogeneity increased the internal validity of our study, but limited the generalization of study findings to more diverse populations [Young et al. 2008]. An additional strength was that in this study smoking status was verified using the level of cotinine in saliva.

In conclusion, our results suggest that environmental pyrethroid exposure may affect sperm DNA damage measures and have reproductive effects on adult men. In view of the importance of human reproductive health and the widespread usage of pyrethroids, it is important to further investigate these correlations.

Materials and Methods

Study population

Study subjects were a subset of 286 men from a parent study of 344 men assessing the impact of environmental, lifestyle, and occupational exposure on semen quality. In the parent study men aged under 45 years of age (range: 22.7–44.8 y) who attended an infertility clinic in Lodz, Poland for diagnostic purposes with normal semen concentration of 15–300 10⁶/ml [WHO 2010] between 2008–2011 from the study “Environmental factors and male infertility” were eligible for inclusion. The Nofer Institute of Occupational Medicine Bioethical Committee Board had approved the study (Resolution No 9/2007 (04.06.2007)) and written informed consent was obtained from all subjects before their participation. Information about socio-demographic characteristics, occupational and lifestyle factors, and medical history was obtained during an interview. Full details of the parent study have been described elsewhere [Jurewicz et al. 2014]. Among 334 men who agreed to participate a subset of 286 (83.14%) men were eligible for this substudy (sufficient semen and urine samples). Approximately 11% of study subjects declared occupational exposure to pesticides, but none of them were occupationally exposed to pyrethroids.

Semen analysis

Main semen parameters analysis

All men provided a semen sample. Ejaculate was obtained by masturbation into a sterile a standard plastic container after a period of sexual abstinence about mean 5 d as a part of fertility investigation. Semen analysis was performed after 30 min of liquefaction at 37°C. The samples were examined by trained laboratory technicians. The semen analysis included determination of ejaculate volume, sperm concentration, and sperm motility according to WHO guidelines [WHO 1999]. Sperm morphology was quantified using strict Kruger criteria [Kruger et al. 1988]. The semen smears were air-dried, fixed, and stained according to Papanicolaou [Menkveld et al. 1990]. Results were reported as the percent of abnormal forms among the spermatozoa examined.

Sperm chromatin structure assay

The assessment of the sperm chromatin structure assay (SCSA) was performed using flow cytometry [ASRM 2006]. SCSA data resolved three different cell populations: 1) % sperm without DNA fragmentation, 2) % sperm with DNA fragmentation (DFI), and 3) high DNA stainability index – percentage of immature sperms (HDS). The cells with abnormal chromatin structure (i.e., fragmented DNA) showed a distinct shift of alpha t parameter value (alpha-t = red/(red + green) fluorescence). The DFI was calculated according to the formula:

DFI = M DFI + H DFI.

The medium and high DFI inform us about the degree of chromatin fragmentation. In SCSA the spermatozoa with lack of chromatin compaction (‘immature’ spermatozoa) have higher acridine orange stainability than those with normally condensed chromatin increasing green fluorescence. The % of sperm with high DNA stainability (% HDS) related to retained nuclear histones consistent with immature sperm [Evenson 2013]. High DNA stainability (% HDS) was calculated on the basis of the percentage of sperm with high levels of green fluorescence, which are thought to represent immature spermatozoa with incomplete chromatin condensation [Evenson et al. 2002]. Results were reported as the percentage of DFI, M DFI (part of sperm with M DFI), H DFI (part of sperm with H DFI), and as the percentage of HDS.

Urine sample collection and analysis

Urinary metabolites of pyrethroids were measured in spot urine samples provided by each subject. Specimens were frozen at −20°C and shipped to the Department of Toxicology, Medical University of Gdańsk where the pyrethroid metabolites: 3-phenoxybenzoic acid (3PBA), cis-3-(2,2-dichlorovinyl)-2,2-dimethylcyclopropane carboxylic acid (CDCCA), trans-3-(2,2-dichlorovinyl)-2,2-dimethylcyclopropane carboxylic acid (TDCCA), and cis-2,2-dibromovinyl-2,2-dimethylcyclopropane carboxylic acid (DBCA) were measured. The laboratory successfully participated in an external quality assurance program for pyrethroid metabolites in urine, The German External Quality Assessment Scheme for Toxicological Analyses in Biological Materials. Metabolites were determined using a validated gas chromatography – mass spectrometry method (GC-MS) as described previously [Wielgomas et al. 2013].

Briefly, 100 µl of IS solution (2-PBA, 0.06 µg/ml of acetonitrile) was added to 3 ml of thawed urine into a 10 ml screw-top glass tube along with 0.6 ml concentrated hydrochloric acid. Acidic deconjugation of metabolites was performed at 95°C in the oven for 90 min.

Analytes were isolated from hydrolyzed urine by liquid-liquid extraction using two 4 ml portions of hexane. Extracts were combined, partitioned with 0.5 ml of 0.1 M NaOH and after centrifugation, the hexane layer was discarded. Aqueous phase was acidified with 0.1 ml of concentrated HCl and extracted with 2 ml of hexane. Organic phase was carefully collected and evaporated to dryness under the gentle stream of nitrogen at 45°C. Derivatization of analytes was done with 10 µl 1,1,1,3,3,3-hexafluoroisopropanol, 20 µl of N,N′-diisopropylcarbodiimide, and 250 µl of hexane during 10 min at room temperature. Following derivatization, 1 ml of 5% K₂CO₃ was added and the mixture was vigorously shaken to remove excess of reagents. Phases were separated by centrifugation and the hexane layer was transferred to an autosampler vial. Two microliters of the final extract were analyzed using a Varian GC-450 gas chromatograph equipped with a 220-MS ion-trap mass spectrometer working in the selected ion storage (SIS) mode. The limit of detection for all metabolites was 0.1 ng/ml with between day imprecision of 4.6–8.3%.

Statistical analysis

Descriptive statistics on subject demographics were calculated, along with the distributions of urinary pyrethroid metabolite concentrations, and sperm DNA damage measures. Bivariate analysis was conducted between all sperm DNA damage, pyrethroid metabolite, and demographic variables to investigate differences between distributions or categories and the potential for confounding. Differences were tested statistically using parametric or non-parametric methods where appropriate. Proportion of samples below the limit of detection (LOD) were as follow: 3PBA was 28.41%, for TDCCA 34.49%, for CDCCA 42.03%, and for DBCA 83.19%. Creatinine adjusted pyrethroids metabolites: 3PBA, TDCCA, and CDCCA were categorized into below or equal to the median for each metabolite and greater than the median. Owing to the high proportion of samples below the LOD in the case of DBCA these creatinine adjusted pyrethroid metabolite concentrations were categorized as above and below LOD. Additionally, the sum of pyrethroids was presented and consisted of specific metabolities. Zero was used in the calculation of these summed variables for values below the LOD, prior to categorizing into groups. The following covariates were evaluated as potential confounders: sexual abstinence (days), age (years), smoking (yes/no), past diseases (yes/no), alcohol consumption (none or <1 drink/w, 1–3 drinks/w, 4–7 drinks/w). The relationship between urinary pyrethroid metabolite categories and sperm DNA damage was evaluated using multiple linear regression. DNA damage measures were log transformed. Kendall correlation coefficient was used to assess the correlations between pyrethroids metabolities. The R 2.15.1 statistical program was used to analyze data [R Core Team 2013].

Acknowledgment

The excellent technical assistance of Ms. Teresa Pawłowska is greatly acknowledged.

Declaration interest

This study was performed under the project “Epidemiology of reproductive hazards - multicenter study in Poland” supported by the National Center for Research and Development in Poland, from grant no. PBZ-MEiN-/8/2//2006; contract no. K140/P01/2007/1.2.1.2, and partly funded by grant ST-5 from the Medical University of Gdańsk. The authors declare no conflict of interest.

Author contributions

Study concept, design, and data interpretation: JJ, WH, MR; Data analysis: WS; Drafted the manuscript: JJ; Responsible for recruitment of men to the study and analysis of main semen parameters: MR, PR; Responsible for analysis of urinary level of metabolites of pyrethroids: BW, MP; Responsible for DNA damage analysis: MB. All authors provided substantial intellectual contributions and approved the final version of manuscript.