Click on the links below to access all the ArticlePlus for this article.
Please note that ArticlePlus files may launch a viewer application outside of your web browser.
There are indications that the sperm density among men from the Western world may have deteriorated during the past decades.1,2 It has been suggested that this trend is due to increasing fetal exposure to compounds with estrogenic or antiandrogenic effects, so-called “endocrine-disrupting compounds.” Exposure to endocrine-disrupting compounds during early development of the male gonad has been hypothesized not only to cause reduced sperm counts in the adulthood but also to increase the risk of testicular cancer and some congenital defects of male reproductive organs, such as cryptorchidism and hypospadias. This spectrum of abnormalities with a possible shared underlying cause has been called testicular dysgenesis syndrome.3
Phthalates are extensively used as plasticizers in household and consumer goods and in certain medical products.4 As a result of their extensive use and their moderate resistance to degradation, phthalates are distributed widely in the environment and often are found at low levels in food. Phthalates are on the list of suspected endocrine-disrupting compounds,5 based on animal studies showing that gestational, lactational, or puberatal exposure to di-n-butyl phthalate (DBP), butyl benzyl phthalate (BBzP), and di(2-ethylhexyl) phthalate (DEHP) causes testicular toxicity, resulting in effects consistent with an antiandrogenic mechanism, including hypospadia and reduced sperm production.6,7 There are also some data suggesting that exposure of adult rats to BBzP and DBP might cause reproductive hormone abnormalities,8 testicular atrophy, and reduced sperm production.7
Ingested phthalates are hydrolyzed rapidly to the biologically active monoesters.9 Thus, mono ethylhexyl phthaltale (MEHP) is a marker of DEHP exposure, mono benzyl phthalate (MBzP) is a marker of BBzP exposure, and mono butyl phthalate (MBP) is a marker, mainly of DBP but also of BBzP exposure (see Figure, available with the electronic version of this article at www.epidem.com).10 Mono ethyl phthalate (MEP) is a marker of diethyl phthalate (DEP). All monoesters may be metabolized to phthalic acid. Recent techniques using liquid chromatography coupled with tandem mass spectrometry (LC-MS-MS) have provided rapid methods for measuring phthalate esters in urine.11
High phthalate doses are needed to provoke male reproductive toxicity in rodents,12 and the relevance of these animal data for humans with much lower exposure levels is unclear. It was therefore of great interest that Duty et al13 recently reported a dose–response relationship between relatively low phthalate levels in urine and human semen parameters. This relationship raises the question whether humans are more sensitive to phthalate exposure than laboratory animals.
The aim of the present study was to assess in young men the association of phthalate metabolite levels in urine with semen function and reproductive hormone parameters. The study base consisted of young Swedish men from the general population undergoing a military conscript examination.
Approximately 95% of all Swedish men undergo a medical examination before military service. Only those men with serious chronic diseases are a priori excluded. Therefore, the group of men that undergo the conscript examination closely reflects the general population of young Swedish males. A total of 2255 men born 1979 to 1982 and living within 60 km of the city of Malmö in southern Sweden were asked in year 2000 (at the time of their conscription examination) to participate in a semen function study. Of these, 305 (14%) accepted the invitation.14 Urine sampling for phthalate analyses was not included initially in the study protocol. Therefore, the first 71 examined subjects did not take part in the study, leaving 234 men. Their median age at the time of examination was 18 years (range, 18–21), their median body mass index was 22.0 (range, 14.9–41.7) and the fraction of smokers was 28%. All subjects signed a consent form and the ethical committee at Lund University approved the study. Additional details regarding study methods are provided in supplementary material, available with the electronic version of this article.
Andrologic Examination and Sampling Procedures
The andrologic examination included scrotal ultrasound in supine position by use of a 7.5-MHz transducer connected to an Aloka 900 SSE scanner (Aloka, Tokyo, Japan). Each testis was investigated in 2 projections, and the volume was calculated as length × width × depth × 0.52. The volumes of the 2 testes were added. Immediately thereafter the subjects delivered semen, urine, and venous blood samples. The semen sample was provided by masturbation into a wide-mouthed plastic cup. The urine grab samples was collected in plastic cups and stored in polyethylene tubes until analysis. Blood samples were analyzed for 2,2′,4,4′,5,5′-hexachlorobiphenyl (CB-153). These results have been presented previously.15
Semen and Hormone Analyses
Semen concentration and motility were analyzed according to the World Health Organization's 1999 recommendations16 and as described by Richthoff et al14 Sperm motility also was assessed by CRISMAS computer-aided sperm motility analyzer (IHMedical A/S, Copenhagen, Denmark).17 The Sperm Chromatin Structure Assay (SCSA) was performed as previously described with minor modification.18,19 Neutral α-glucosidase,20 zinc,20,21 prostate-specific antigen,20 and fructose20,22 in seminal plasma were assessed as described previously. Follicle-stimulating hormone, luteinizing hormone, sexual hormone binding globulin, testosterone, and estradiol in serum were measured using an automated fluorescence detection system (Autodelfia, Wallac Oy, Turku, Finland). Inhibin B in serum was assessed as previously described.23 (Further description of these analyses can be obtained from the electronic version of this article at www.epidem.com).
Analysis of Phthalate Metabolites
Aliquots of 2 mL of urine were combined with 500 ng of tetra-deuterium labeled phthalic acid, used as an internal standard for all analyzed compounds plus 2.5 μL of glucoronidas/arylsulfatase (helix pomatia) from Boehringer Mannheim (Mannheim, Germany) and 0.5 mL of 1 M ammonium acetate (pH 6.5). The samples were incubated at 37°C overnight, then 0.5 mL of 6 M NaCl and 0.1 mL of 6 M HCl were added and the samples were extracted by 1 portion of 6 mL of ethyl acetate. The organic phase was evaporated and the residue dissolved in 0.2 mL of water containing 0.5% acetic acid.
We analyzed samples by LC-MS-MS using a Perkin Elmer Series 200 LC system with autosampler (Applied Biosystems, Norfolk, CT), coupled to an API 3000 LC-MS-MS (Applied Biosystems/MDS-SCIEX, Toronto, Canada). The column used was a Genesis C18 (50 × 2.1 mm), with a particle size of 4 μm (Jones, Lakewood, CO). Mobile phases were (A) 0.5% acetic acid in water and (B) 0.5% acetic acid in methanol. A gradient from 70% A to 100% B was applied in 5 minutes and then kept at 100% B for 2 minutes. The column was reconditioned at 70% A for 3 minutes. The temperature in the ion source was 350°C, the ion spay voltage −3000 volts, and the declustering potential −26 volts. Analyses were performed in the negative ion multiple reaction monitoring mode and the mass fragments used were for phthalic acid m/z 164.7/76.9 (collision energy [CE] −23 volts), for MEP m/z 192.7/77.0 (CE −28 volts), for MBP m/z 221.1/71.1 (CE −22 volts), for MBzP m/z 254.9/105.1 (CE −22 volts), for MEHP m/z 276.9/134.1 (CE −24 volts), and for deuterium labeled phthalic acid m/z 168.8/81.0 (CE −25 volts). Standards were prepared by addition of different amounts of MEP, MBP, MBzP, MEHP, and phthalic acid to normal urine. The peak area ratios between the analytes and the internal standards were used for determination. The urine sampling procedure was tested by use of purified water, and no measurable contamination of the phthalate metabolites was found during this procedure. The detection limits were 15, 30, 15, 7, and 15 ng/mL for phthalic acid, MEP, MBP, MBzP, and MEHP, respectively. The precisions were obtained from comparison of duplicate analysis on different days and found to be 10% (at 50 ng/mL; n = 138) for phthalic acid; 47% (at 300 ng/mL; n = 50) for MEP; 12% (at 50 ng/mL; n = 126) for MBP; 15% (at 15 ng/mL; n = 156) for MBzP; and 23% (at 40 ng/mL; n = 66) for MEHP.
Concentrations of Phthalate Metabolites
The concentrations of phthalate metabolites are shown in Table 1. The highest levels were found for MEP. Low levels were found for MBzP and MEHP, whereas the levels for phthalic acid and MBP were intermediate. There was a wide variation among individuals; for example, 1 subject had 0.25 mg MEP/mL urine, indicating an extremely high exposure to DEP. This subject had also 4.5 μg/mL of phthalic acid in the urine, which also was very high. The monoesters and phthalic acid are rather hydrophilic compounds at urinary pH and should therefore preferably be corrected for urinary concentration. Table 1 also provides the creatinine-corrected levels of the metabolites. There were close associations between creatinine-adjusted and nonadjusted values, with Spearman’s rank correlation coefficients between 0.78 and 0.98.
The fraction of individuals whose exposure concentrations were below detection limits ranged from 8% to 63% for the various exposure variables. Individuals with exposure concentrations below detection limits were given a fixed value below the detection limits. The phthalate parameters were then divided into quartiles. We calculated mean difference (with 95% confidence intervals [CIs]) between the lowest and the highest quartiles of phthalate metabolites in urine for each marker of reproductive function. For MEHP, the individuals who had concentrations below the detection limit (63%) were compared with the 18% who had the highest concentrations. If the mean difference was in the direction of the hypothesis and the CI did not include 0, we tested whether there was a dose–response relation by including the intermediate categories (as dummy variables) in linear regression models. Abstinence time (≤48, 49–72, 73–96, and ≥97 hours) and smoking (yes or no) were then considered as potential confounders. The reason for the categorization of abstinence time was a possible nonlinear relation with some of the outcome variables, for instance, semen volume. These potential confounders were included, one at a time, together with the exposure variable in the models. None of the adjusted exposure-effect estimates differed at least 15% from the unadjusted ones, and so the potential confounders were not kept in the model. Model assumptions were checked by means of residuals.
Details regarding semen and sperm parameters and the hormone levels are presented in a supplementary table, available with the electronic version of this article. Table 2 shows the mean differences between the lowest and the highest quartiles of phthalate metabolites for the various markers of reproductive function. For MBP, MBzP, and MEHP we saw no apparent associations with any of the reproductive biomarkers. Subjects within the highest quartile for MEP had larger testis volume (mean difference, 3.2 mL; 0.4–6.0), but fewer CASA motile sperm (8.8%; 0.8–17), more CASA immotile sperm (8.9%; 0.3–18), and lower luteinizing hormone values (0.7 IU/L; 0.1–1.2). The levels in the intermediate groups were very similar to those in the lowest exposed group, suggesting a possible threshold effect (Table 3).
The results for phthalic acid were opposite what had been hypothesized; the quartile with the highest phthalic acid levels had larger testis volume (mean difference, 3.8 mL; 0.9–6.8), more motile sperm (9.4%; 3.7–15), and fewer immotile sperm (7.0%; 2.3–11.8) than the lowest exposed quartile.
As a biomarker of PCB exposure, CB-153 had been analyzed previously in serum samples from the examined subjects.15 We could therefore assess whether there were any interactions between exposure to phthalates and PCB, another endocrine disruptor, by including CB-153 and 1 phthalate variable at the time in linear regression models. The model also included an interaction term (CB 153 × phthalate exposure). Despite a large number of comparisons, we did not find any convincing evidence that a mixture of these compounds negatively affected male reproductive functions.
The results showed some weak associations between phthalate biomarkers in urine and reproductive function biomarkers. This pattern may reflect a hazard for male reproduction from phthalate exposure, but the results also may be spurious. Our findings have to be considered in light of the available toxicologic and epidemiologic data, as well as the methodological strengths and limitations of our study.
The urinary concentrations of phthalate metabolites in our study were the same order of magnitude (with the exception of higher MEP values) as a previous U.S. study that assessed the association between phthalate exposure and semen function parameters.4 Despite the high levels of MEP, this compound was not detected in all individuals due to the high detection limit for MEP. Levels of urinary phthalate metabolites were somewhat higher in subjects from the general population in Germany.24 As in the U.S. and German studies, we found large interindividual variations in urine concentrations of the various monoesters. The sampling procedure was confirmed as negative for contamination and cannot explain the variation. Furthermore, the imprecision in the analytical method was not large enough to cause this variation, despite the fact that the precision for MEP was as low as 47% at a concentration 10 times the detection limit.
We assume that the 234 young men in this study are fairly representative for this age group of men in southern Sweden with respect to semen function, despite the fact that only 14% of eligible subjects agreed to deliver a semen sample. It is unlikely that these 18- to 21-year-old men had much knowledge about their reproductive capability. The low participation rate should therefore not imply selection bias with respect to fertility. A recent Danish conscript study with a correspondingly low participation rate concluded that the participants were representative for the whole group of conscripts, based on the levels of reproductive hormones in participants and nonparticipants in the semen study.25 In principle, there could also be a selection bias with respect to the exposure, but it is difficult to see the mechanism for such selection. A selection bias due to chance cannot be totally excluded.
The present method for determination of phthalate metabolites is technically simpler than the method described by Blount et al,11 with a somewhat higher limit of detection and a somewhat lower precision. A low precision might result in a nondifferential misclassification of exposure that could blur true associations.
In the present study we used glucoronidas/arylsulfatase from helix pomatia for deconjugation of glucoronic acid. Blount el al4,11 reported that this enzyme might have lipase activity that could hydrolyze the phthalate diesters to monoesters. However, we found no differences in the results obtained for any of the metabolites when using either glucoronidas/arylsulfatase from helix pomatia or glucoronidas from Escherichia coli. Similarly, there was no increase in the concentrations of phthalic acid when the hydrolysis was omitted. Thus, the enzymes do not appear to hydrolyze the monoesters to phthalic acid.
Each of the 5 exposure markers was studied in association with 20 effect markers. Thus, we should expect a certain number of statistically significant associations to occur by chance. We did not find many more associations than expected. The results have therefore to be considered with respect to the strength, biologic plausibility, and consistency of the observed associations. All the observed differences between subjects by exposure were rather modest. Animal studies have shown that DBP, BBzP, and DEHP may impair male reproductive functions,6,7 and it was therefore reasonable to primarily hypothesize that their metabolites, MPB, MBzP, and MEHP (rather than MEP and the more unspecific end product phthalic acid) would be associated with male reproductive function in humans. However, such a clear-cut pattern was not evident. MEP was the only phthalate biomarker negatively associated with sperm motility.
Our results were not very consistent with those from the U.S. study. Those authors found negative associations with sperm concentration for MBP and MBzP but not for MEHP.13 They found a negative association between MBP and sperm motility, whereas we saw an association with motility only for MEP. However, the higher levels of MEP in our study could have provided more power to detect an association. Furthermore, the U.S. researchers13 analyzed their data in a somewhat-different way, using logistic regression models. When we applied logistic models to the analysis of sperm concentration (<20 million/mL, >20 million/mL) and sperm motility (<50%, >50%), the patterns of our results were unchanged and were still not very consistent with those from the US study.
In the U.S. study, there was a tendency of increased DNA damage in sperm (assessed by the neutral Comet assay) for subjects with the highest urinary levels of MEP.26 We, however, found no association between MEP (or any other phthalate biomarker) and sperm chromatin integrity using the SCSA. Both the SCSA and the neutral Comet method assess double-stranded DNA breaks; however, there are not enough data to estimate the degree of concordance between the results from these 2 techniques.
Our study is based on men from the general population and not from subfertile couples, as in the U.S. study.13,26 Men with poor spermatogenesis might be more prone to adverse effects of environmental toxicants. Moreover, all subjects in the present study were 18- to 21-year-old whereas in the studies by Duty et al13,26 the median age was higher and the age range wider. It is not known whether susceptibility to phthalates could be dependent on age. Furthermore, there might be a genetic difference in susceptibility to adverse effects of phthalates between a Swedish and an American population.
Animal data show that phthalates early in life are harmful to later male reproductive development,6,7 but limited data also support the possibility that adult rodents are susceptible.7,8 This supports the study of phthalate exposures in human adults and their possible associations with reproductive function. A drawback of the exposure biomarker is that the half-lives of the mono-phthalates are only about 12 hours, and there are no data on how well the concentrations of phthalate metabolites in spot samples of urine estimate the internal phthalate dose during the preceding months, which is the relevant period for spermatogenesis.
In conclusion, we found some weak associations between phthalate exposure and biomarkers of male reproductive functions. Our findings are, however, not consistent with those from a recent U.S. study on males from subfertile couples.13,26 Phthalates are ubiquitous environmental contaminants with endocrine-disrupting potential. Even though the results from the recent U.S. study and the present one are inconsistent, both suggested that exposure levels occurring in the general population might be hazardous. From a population health perspective, it is vital that further research establish the extent to which phthalates represent a real threat to male reproductive function in humans.
We thank H. Ottosson, C. Anderberg, K. Jepson, and A. Bremer for skillful technical assistance.
1. Giwercman A, Carlsen E, Keiding N, et al. Evidence for increasing incidence of abnormalities of the human testis: a review. Environ Health Perspect Suppl
2. Swan SH, Elkin EP, Fenster L. The question of declining sperm density revisited: an analysis of 101 studies published 1934–1996. Environ Health Perspect
3. Skakkebaek NE, Rajpert-De Meyts E, Main KM. Testicular dysgenesis syndrome: an increasingly common developmental disorder with environmental aspects. Hum Reprod
4. Blount BC, Silva MJ, Caudill SP, et al. Levels of seven urinary phthalate metabolites in a human reference population. Environ Health Perspect
5. Toppari J, Larsen JC, Christiansen P, et al. Male reproductive health and environmental xenoestrogens. Environ Health Perspect Suppl
6. Gray LEJr, Ostby J, Furr J, et al. Perinatal exposure to the phthalates DEHP, BBP, and DINP, but not DEP, DMP, or DOTP, alters sexual differentiation of the male rat. Toxicol Sci
7. Foster PM, Mylchreest E, Gaido KW, et al. Effects of phthalate esters on the developing reproductive tract of male rats. Hum Reprod Update
8. Nagao T, Ohta R, Marumo H, et al. Effect of butyl benzyl phthalate in Sprague-Dawley rats after gavage administration: a two-generation reproductive study. Reprod Toxicol
9. Li LH, Jester WFJr, Orth JM. Effects of relatively low levels of mono-(2-ethylhexyl) phthalate on cocultured Sertoli cells and gonocytes from neonatal rats. Toxicol Appl Pharmacol
10. Anderson WA, Castle L, Scotter MJ, et al. A biomarker approach to measuring human dietary exposure to certain phthalate diesters. Food Addit Contam
11. Blount BC, Milgram KE, Silva MJ, et al. Quantitative detection of eight phthalate metabolites in human urine using HPLC-APCI-MS/MS. Anal Chem
12. Mylchreest E, Sar M, Wallace DG, et al. Fetal testosterone insufficiency and abnormal proliferation of Leydig cells and gonocytes in rats exposed to di(n-butyl) phthalate. Reprod Toxicol
13. Duty SM, Silva MJ, Barr DB, et al. Phthalate exposure and human semen parameters. Epidemiology
14. Richthoff J, Rylander L, Hagmar L, et al. Higher sperm counts in Southern Sweden compared with Denmark. Hum Reprod
15. Richthoff J, Rylander L, Jönsson BAG, et al. Serum levels of 2,2′,4,4′,5,5′-hexachlorobiphenyl (CB-153) in relation to markers of reproductive function in young males from the general Swedish population. Environ Health Perspect
16. World Health Organization. WHO Laboratory Manual for the Examination of Human Semen and Sperm–Cervical Mucus Interaction
. Cambridge: Cambridge University Press; 1999.
17. Larsen L, Scheike T, Jensen TK, et al. Computer-assisted semen analysis parameters as predictors for fertility of the general population. Hum Reprod
18. Evenson DP, Jost A. Sperm chromatin structure assay is useful for fertility assessment. Methods Cell Sci
19. Richthoff J, Spano M, Giwercman YL, et al. 2002. The impact of testicular and accessory sex gland function on sperm chromatin integrity as assessed by sperm chromatin structure assay (SCSA). Hum Reprod
20. Elzanaty S, Richthoff J, Malm J, et al. The impact of epididymal and accessory sex gland function on sperm motility. Hum Reprod
21. Makino T, Saito M, Horiguchi D, et al. A highly sensitive colorimetric determination of serum zinc using water-soluble pyridylazo dye. Clin Chim Acta
22. Wetterauer U, Heite HJ. Eine empfehlenswerte Methode zur gleichzeitigen enzymatischen Bestimmung von Citrat und Fruktose im Seminalplasma. Akt Dermatol
23. Groome NP, Illingworth PJ, O'Brien M, et al. Measurement of dimeric inhibin B throughout the human menstrual cycle. J Clin Endocrinol Metab
24. Koch HM, Rossbach B, Drexler H, et al. Internal exposure of the general population to DEHP and other phthalates—determination of secondary and primary phthalate monoester metabolites in urine. Environ Res
25. Andersen AG, Jensen TK, Carlsen E, et al. High frequency of sub-optimal semen quality among young men of the general Danish population. Hum Reprod
© 2005 Lippincott Williams & Wilkins, Inc.
26. Duty SM, Singh NP, Silva MJ, et al. The relationship between environmental exposure to phthalates and DNA damage in human sperm using the neutral comet assay. Environ Health Perspect