2.3. Ovarian histology

To determine the effect of TPP and DPP exposure on the developing ovary, morphological evaluations were performed on ovaries at 19 days of age. Ovaries were collected and fixed in 10% neutral formalin and stained with hematoxylin and eosin according to standard laboratory procedure. Three sections of both ovaries from each mice (six replicates per group) were analyzed for the presence and number of multi-oocyte follicles (MOFs) as previously reported (Jefferson et al., 2009). The MOFs were determined because previous studies showed that exposure of fetal or newborn animals to endocrine disruptors and phytoestrogens, such as diethylstilbestrol (Iguchi et al., 1990), isoflavone (Takashima-Sasaki et al., 2006), genistein (Jefferson et al., 2002), and bisphenol A (Rodriguez et al., 2010) were associated with the increased MOFs numbers, and the increased MOFs numbers were thought to be generated by alterations of nest breakdown and/or follicle assembly.

2.5. Hormone analysis

Hormone analysis of estradiol levels was performed in serum collected at 12 weeks from all the treatment groups with an ELISA kit (Elabscience Biotechnology Co., Ltd) according to the manufacturers' instructions. Studies were performed with mice in the proestrus phase, determined by vaginal smears, and vaginal smears were performed before mice were killed. Mice in the proestrus phase were selected due to the sufficient numbers for all the experiments.

2.6. Measurement of body weight and oral glucose tolerance test (OGTT)

Mice were weighed weekly, starting at 21 days (3 weeks) until 12 weeks of age. After exposure to TPP or DPP for 10 weeks, six mice per group were fasted overnight and a d-glucose (2 g/kg) oral gavage was administered. Blood samples were collected from the tail vein before and 15, 30, 45, 75, and 105 min after delivery of the glucose load. Blood glucose levels were measured using a glucose meter.

2.7. 1H-NMR-based metabolomics experiment

Sample preparation Serum samples were prepared by mixing 200 μL of serum with 300 μL of saline solution and 100 μL of D2O containing 5 mM formate. After centrifugation at 14,000×g for 10 min, 550 μL of the supernatant were transferred into a 5-mm NMR tube for NMR analysis.

1H-NMR analysis1H NMR spectra of serum were recorded on a BRUKER AVANCE Ⅲ 600 spectrometer equipped with an inverse cryogenic probe. All one-dimensional 1H NMR spectra were acquired at 300 K using the water-presaturated Carr-Purcell-Meiboom-Gill (CPMG) spin-echo pulse sequence (recycle delay-90°-(τ-180°-τ) n-acquisition). A total of 64 transients were collected into 32 K data points using a spectral width of 20 ppm, a relaxation delay of 4 s, and an acquisition time of 2.72 s. Magnetic field homogeneity was optimized using the first sample in every batch and assessed for each sample by measuring the signal-to-noise (S/N) ratio and the full width at half-maximum (fwhm) of internal standard peak, formate, characterized by good signal to noise ratio and appropriate line width ∼1–1.5 Hz. Samples for which the formate line width greater than 1.5 Hz were re-extracted and data re-acquired to the meet the standard or the samples were excluded in the statistical analysis. The variations in NMR profiles due to the extraction protocol (reproducibility) were assessed by comparing the differences among 1D 1H NMR spectra from extracting aliquots of the same sample on ten different occasions, and the variations due to the NMR measurement (repeatability) was assessed by evaluating NMR data from one sample on ten different occasions. The variances in the scores of Principal component analysis (PCA) models constructed using these NMR data was assessed to determine the reproducibility and repeatability of the profiles. Under these above conditions, the reproducibility and repeatability was reliable (data not shown).

An exponential window function was used with a line broadening factor of 1 Hz on all acquired free induction decays prior to Fourier transformation. Spectra were then manually phase- and baseline-corrected, and referenced to glucose (δ 5.23). The spectral region of δ0.50–9.00 was integrated into regions with an equal width of 0.01 ppm using MestRenova 6.1.0. Regions δ 4.5–5.1 containing residue water resonance due to imperfect water saturation and regions δ 8.4–8.6 referring to formate were excluded prior to data normalization. The remaining integrated areas were normalized to the total spectral area to facilitate comparisons among the samples.

Multivariate analysis Multivariate data analysis was performed with SIMCA-P+ software (version 13.0, Umetrics, Sweden). Principal component analysis (PCA) and orthogonal projection to latent structures with discriminant analysis (OPLS-DA), were performed on the binned NMR data to identify the intrinsic clusters and model the treatment-related differences. A 7-fold cross-validation method was used to validate the OPLS-DA models, and the quality of the model was described by R2X and Q2 values. S-plots, loading plots, and contribution plots were used to determine bins that best define group separation. The subset of bins corresponding to metabolites with a variable influence on projections (VIP) >1.0 were selected and further used for semi-quantification analysis. These semi-quantitative methods, using formate as internal standards, were able to determine the relative concentrations of metabolites.

2.8. Statistical analysis

All the experimental values were expressed as a mean ± standard error of the mean. Graphical illustrations were performed using GraphPad Prism version 6.0 (GraphPad Software, Inc., USA), and statistical analyses were performed using SPSS 19.0 (IBM, USA). The significance of variables among the different groups was determined by one-way ANOVA and a Tukey's post-hoc test (P < 0.05). The normality of the variables was confirmed by the Shapiro–Wilk test and homogeneity of variance by the Levene's test. A non-parametric Kruskal–Wallis with Dunnett's post-hoc test was used when data did not meet the parametric assumptions."

Male and Female ICR Mice