Henning Sloth Pedersen

Exposure to perfluorinated compounds and human semen quality in arctic and European populations

BACKGROUND Perfluorinated compounds (PFCs) have been suspected to adversely affect human reproductive health. The aim of this study was to investigate the associations between PFC exposure and male semen quality. METHODS PFCs were measured in serum from 588 partners of pregnant women from Greenland, Poland and Ukraine who provided a semen sample, using liquid chromatography tandem mass spectrometry. Perfluorooctane sulfonate (PFOS), perfluorooctanoic acid (PFOA), perfluorohexane sulfonic acid (PFHxS) and perfluorononanoic acid (PFNA) could be detected in >97% of the samples. The associations between levels of these compounds and semen volume, sperm concentration, total sperm count, motility and morphology were assessed. RESULTS Across countries, sperm concentration, total sperm count and semen volume were not consistently associated with PFOS, PFOA, PFHxS or PFNA levels. The proportion of morphologically normal cells was 35% lower [95% confidence interval (CI): 4-66%) for the third tertile of PFOS exposure as compared with the first. A similar reduction was found in relation to increasing PFHxS levels. At the third PFOA exposure tertile, the percentage of motile spermatozoa was 19% (95% CI: 1 to 39%) higher than in the first. CONCLUSIONS The most robust finding in the present study was the negative associations between PFOS exposure and sperm morphology suggesting adverse effects of PFOS on semen quality, possibly due to interference with the endocrine activity or sperm membrane function. It cannot be excluded that this association and the positive association between PFOA and semen motility, which was not consistent across countries, might represent a chance finding due to the multiple statistical tests being performed.

Menstrual cycle characteristics in fertile women from Greenland, Poland and Ukraine exposed to perfluorinated chemicals: a cross-sectional study

STUDY QUESTION Does perfluorooctane sulfonate (PFOS) and perfluorooctanate (PFOA) exposure disrupt the menstrual cyclicity? SUMMARY ANSWER The female reproductive system may be sensitive to PFOA exposure, with longer menstrual cycle length at higher exposure. WHAT IS KNOWN ALREADY PFOS and PFOA are persistent man-made chemicals. Experimental animal studies suggest they are reproductive toxicants but epidemiological findings are inconsistent. STUDY DESIGN, SIZE, DURATION A cross-sectional study including 1623 pregnant women from the INUENDO cohort enrolled during antenatal care visits between June 2002 and May 2004 in Greenland, Poland and Ukraine. PARTICIPANTS/MATERIALS, SETTING, METHODS Information on menstrual cycle characteristics was obtained by questionnaires together with a blood sample from each pregnant woman. Serum concentrations of PFOS and PFOA were measured by liquid chromatography tandem mass spectrometry. Multiple imputations were performed to account for missing data. The association between PFOS/PFOA and menstrual cycle length (short cycle: ≤24 days, long cycle: ≥32 days) and irregularities (≥7 days in difference between cycles) was analyzed using logistic regression with tertiles of exposure. Estimates are given as adjusted odds ratios (ORs) with 95% confidence intervals (CIs). MAIN RESULTS AND THE ROLE OF CHANCE Higher exposure levels of PFOA were associated with longer menstrual cycles in pooled estimates of all three countries. Compared with women in the lowest exposure tertile, the adjusted OR of long cycles was 1.8 (95% CI: 1.0; 3.3) among women in the highest tertile of PFOA exposure. No significant associations were observed between PFOS exposure and menstrual cycle characteristics. However, we observed a tendency toward more irregular cycles with higher exposure to PFOS [OR 1.7 (95% CI: 0.8; 3.5)]. The overall response rate was 45.3% with considerable variation between countries (91.3% in Greenland, 69.1% in Poland and 26.3% in Ukraine). LIMITATIONS, REASONS FOR CAUTION Possible limitations in our study include varying participation rates across countries; a selected study group overrepresenting the most fertile part of the population; retrospective information on menstrual cycle characteristics; the determination of cut-points for all three outcome variables; and lacking information on some determinants of menstrual cycle characteristics, such as stress, physical activity, chronic diseases and gynecological disorders, thus confounding cannot be excluded. WIDER IMPLICATIONS OF THE FINDINGS The generalizability of the study results is restricted to fertile women who manage to conceive and women who do not use oral contraceptives when getting pregnant or within 2 months before getting pregnant. To our knowledge only one previous epidemiological study has addressed the possible association between perfluorinated chemical exposure and menstrual disturbances. Though pointing toward different disturbances in cyclicity, both studies suggest that exposure to PFOA may affect the female reproductive function. This study contributes to the limited knowledge on effects of exposure to PFOA and PFOS on female reproductive function and suggests that the female reproductive system may be affected by environmental exposure to PFOA. STUDY FUNDING/COMPETING INTEREST(S) Supported by a scholarship from Aarhus University Research Foundation. The collection of questionnaire data and blood samples was part of the INUENDO project supported by The European Commission (Contract no. QLK4-CT-2001-00 202), www.inuendo.dk. The Ukrainian part of the study was possible by a grant from INTAS (project 012 2205). Determination of PFOA and PFOS in serum was part of the CLEAR study (www.inuendo.dk/clear) supported by the European Commissions 7th Framework Program (FP7-ENV-2008-1-226217). No conflict of interest declared.

Indices of methylation in sperm DNA from fertile men differ between distinct geographical regions

STUDY QUESTION Which are the main determinants, if any, of sperm DNA methylation levels? SUMMARY ANSWER Geographical region resulted associated with the sperm methylation status assessed on genome-wide repetitive sequences. WHAT IS KNOWN ALREADY DNA methylation level, assessed on repetitive sequences from peripheral blood lymphocyte, can vary with age, gender, alcohol consumption and white blood cell counts. STUDY DESIGN, SIZE, DURATION A cross-sectional study. Individual data were collected from 269 young healthy men of proven fertility living in three geographical regions: Inuits from Greenland, Caucasians from Warsaw (Poland) and Kharkiv (Ukraine). Semen samples were collected between May 2002 and February 2004 and aliquots were immediately frozen. PARTICIPANTS/MATERIALS, SETTING, METHODS We estimated sperm DNA global methylation level (DGML) in two ways. First DNA methylation in repetitive DNA sequences (LINE-1, Satα and Alu) was quantified by PCR pyrosequencing after bisulfite conversion and second by flow cytometry (FCM) using fluorescently labeled monoclonal antibodies anti-5-methylcytosine. We analyzed whether personal characteristics and habits, body mass index, semen quality parameters, sperm chromatin integrity, biomarkers of accessory gland function and the plasma concentration of reproductive hormones were associated with sperm DNA methylation levels in men. Associations were evaluated by analysis of variance and linear regression analyses. MAIN RESULTS AND THE ROLE OF CHANCE The geographical location emerged as the main determinant when using the methylation level in repetitive sequences. FCM DGML results were not associated with those from repetitive sequence analysis. No other consistent associations between methylation markers and the assessed variables were identified across countries. LIMITATIONS, REASONS FOR CAUTION The methods used are only surrogates of the actual sperm methylome and the methylation levels at individual specific loci were not explored. WIDER IMPLICATIONS OF THE FINDINGS Sperm DGML is relatively independent from semen quality parameters and is a new candidate biomarker for epidemiological studies of the impact of environmental contaminants on male fertility. STUDY FUNDING/COMPETING INTERESTS The study is part of the project CLEAR (Climate change, Environmental contaminants and Reproductive health) supported by the European Commission 7th framework program, contract no: FP7-ENV-2008-1-226217. No competing interest is declared.

Non-linear association between androgen receptor CAG and GGN repeat lengths and reproductive parameters in fertile European and Inuit men

Recently the dogma that there is an inverse linear association between androgen receptor (AR) CAG and GGN polymorphisms and receptor activity has been challenged. We analysed the pattern of association between 21 male reproductive phenotypes and AR CAG/GGN repeat lengths in 557 proven-fertile men. A linear association was only found between sperm DNA fragmentation index (DFI) and CAG length, and between inhibin B and GGN length. Men with longer CAG then the reference (22-24), had higher oestradiol levels, whereas men with shorter CAG stretches had a higher DFI and a higher proportion of Fas-positive germ cells. Subjects with either short or long CAG had increased seminal levels of prostate-specific antigen and neutral α-glucosidase activity. Compared to men with the median GGN length of 23, those with shorter GGN repeats had higher levels of inhibin B, higher proportions of normal and progressive sperm, and a higher fraction of Fas-positive sperm, while men with longer GGN had higher oestradiol levels. These data indicate that at least for some markers of male reproductive function the association with CAG or GGN repeat length is curvilinear.