INSL3 in amniotic fluid was measured using a semicompetitive time-resolved fluorometric immunoassay (TRFIA) slightly modified from that described in detail by Anand-Ivell et al.
INSL3 levels are highly dependent on gestational age at amniocentesis with a peak around week 15 (Anand-Ivell et al.
INSL3 was measured in amniotic fluid samples from 475 pregnancies (74% of cryptorchidism cases, 84% of hypospadias cases, and 71% of controls).
The association between PFOS exposure, testosterone, and INSL3 was similar in the case and control groups as indicated by p-values of product interaction terms in the regression models (Table 2) and estimates (Table 3).
PFOS concentrations measured in amniotic fluid were associated with higher steroid hormone levels and lower INSL3 in the combined study population but were not associated with cryptorchidism or hypospadias (270 and 75 cases, respectively, compared with 300 controls).
To our knowledge, the present study is the first human study evaluating potential association between fetal exposure to PFOS and biomarkers of human fetal steroid and INSL3 levels.
The inverse association between prenatal concentrations of PFOS and INSL3 may be consistent with a direct effect on fetal Leydig cells, because, based on the current knowledge, INSL3 is produced only by male fetal Leydig cells, whereas steroid hormones are also produced by the fetal adrenal gland (Anand-Ivell and Ivell 2014).
In utero exposure of rats to di(n-butyl) phthalate, a compound with antiandrogenic properties, caused reduced INSL3 expression in fetal Leydig cells and an increase in cryptorchidism, and INSL3 has been suggested as an endogenous marker of endocrine disruption (Anand-Ivell and Ivell 2014).
We measured PFOS, steroid hormones, and INSL3 in amniotic fluid for several reasons.
However, PFOS, testosterone, and INSL3 levels were only weakly correlated with year of amniocentesis (r-values of 0.
Associations of PFOS with INSL3 and steroid hormone concentrations in amnionic fluid suggest that prenatal PFOS exposure may have affected fetal Leydig cell function in our study population.