In vitro exposure to benzo[a]pyrene damages the developing mouse ovary

Abstract

Females are born with a finite number of oocytes, collectively termed the ovarian reserve, established within the developing fetal ovary. Consequently, maternal exposure to reproductive toxicants can have harmful effects on the future fertility of her unborn female fetus. The chemical benzo[a]pyrene (B[a]P) is a prominent component of cigarette smoke. Despite it being a known ovotoxicant, around 8% of women in Europe smoke during pregnancy. The purpose of this research was to examine the effect of B[a]P on the developing ovary, using the mouse as a model and with experiments carried out in vitro. B[a]P-exposure to the fetal ovary prior to follicle formation reduced the number of germ cells and subsequently, the number of healthy primordial follicles, by up to 76%; however, while proliferation of germ cells was not affected, the germ cells contained higher levels of DNA double-strand breaks. Exposure to B[a]P also affected the proportion of oocytes progressing through prophase I of meiosis. B[a]P exposure to neonatal mouse ovaries, after follicle formation, resulted in an 85% reduction in the number of healthy follicles, with a corresponding increase in apoptotic cell death and reduction in somatic cell proliferation. Although there was a trend towards a higher level of oxidative stress in B[a]P-exposed ovaries, this was not statistically significant; likewise, the antioxidant melatonin failed to protect against the B[a]P-induced ovarian damage. Together, the results here demonstrate that B[a]P-exposure damages the developing ovary, both before and shortly after follicle formation, an effect that could lead to a subsequent decrease in fertility.

Figure 1

Comparison of timings of developmental events between the mouse and human ovary. Migrating PGCs invade the developing female gonad, after which time the germ cells continue to proliferate until they enter into prophase I of the first meiotic division. Oogonia undergo early stages of prophase I, at which point germ cell nests have begun to form. Subsequently, the germ cell nests begin to break down, with follicles forming around the time of birth in the mouse, but during the second trimester of the human pregnancy. From the point of PMF formation, some PMFs will gradually begin to initiate follicle growth. Follicle growth starts only after birth in the mouse ovary, but from the second and third trimesters in the human ovary. Black boxes indicate culture periods used in the current study with yellow areas indicating the duration of B[a]P administration, and white areas indicating the period of time where ovaries were kept in untreated medium: (A) E12.5 ovaries cultured for 24 h with B[a]P or DMSO, (B) E13.5 ovaries cultured for 24 h with B[a]P or DMSO, (C) E13.5 ovaries cultured for 12 days with B[a]P or DMSO added to the media for the first 6 days of culture only and left in the untreated medium for a further 6 days and (D) PND4 ovaries cultured for 6 days with B[a]P or DMSO. DPC, days post-coitum; WPC, weeks post-conception.

Figure 2

B[a]P affects the number and distribution of healthy follicles in the embryonic ovary. (Ai–Aiv) Photomicrographs of haematoxylin and eosin stained sections from embryonic mouse ovaries cultured with increasing concentration of B[a]P. Insert in the top right-hand corner shows representative images of PMF, TRNS and PRMRY follicles. Histograms show (B) total numbers and (C) distribution of the developmental stage of healthy follicles across different B[a]P exposure groups. Scale bars: 50 µm. Bars denote mean + s.e.m.; n = 7. Stars denote significant differences relative to control (**P < 0.01).

Figure 3

B[a]P reduces the number of germ cells in the fetal ovary but does not affect the proliferation of germ cells. (A) Representative photomicrographs of cultured E13.5 embryonic mouse ovaries cultured for 24 h under either (Ai) control conditions, or (Aii) with 1 µg/mL B[a]P, showing non-proliferating germ cells (DDX4-positive; green) and proliferating germ cells (DDX4-positive/BrdU-positive; green and pink). White arrowheads show examples of non-proliferating germ cells; asterisks show examples of proliferating germ cells. Histograms show total number of germ cells in ovaries cultured from (B) E12.5 ovaries and (D) E13.5 ovaries, as well as the percentage of germ cells undergoing proliferation during culture of (C) E12.5 ovaries and (E) E13.5 ovaries. Scale bars represent 25 µm. Bars denote mean + s.e.m.; n = 6 for both E12.5 and E13.5 treatment groups; n = 6 for the E13.5–E14.5 control group and n = 7 for the E13.5–E14.5 B[a]P-treated group. Stars denote significant differences relative to control (**P < 0.01).

Figure 4

B[a]P increases DNA double-strand breaks within germ cells of the fetal ovary but does not significantly increase apoptosis levels. Representative images of E12.5 ovaries cultured for 24 h under either (Ai, Ci) control conditions or (Aii, Cii) with 1 µg/mL B[a]P. (Ai, Aii) Ovary sections show γH2AX positive oogonia (green, pink) and γH2AX negative oogonia (green). White arrowheads show examples of γH2AX negative oogonia; asterisks show examples of γH2AX positive oogonia. (B) Histogram shows the percentage of oogonia counted that were γH2AX positive. (Ci, Cii) Ovary sections show DDX4 localisation (green) and cells expressing CC3 (pink). (D) Histogram shows the percentage of tissue section area positive for CC3. To measure the area of CC3 expression. (E) The fluorophore area for CC3 was measured using ImageJ/Fiji software, as a percentage of the total area of the section. Scale bars represent 50 µm. Bars denote mean + s.e.m.; n = 5 for all. Stars denote significant differences relative to control (**P < 0.01).

Figure 5

B[a]P increases the proportion of oocytes at pre-pachytene stages of prophase I. (A–B) Representative images of E13.5 ovaries cultured for 72 h under either (Ai) control conditions or (Aii) with 1 µg/mL B[a]P and stained for SYCP3 (green) in order to assess the synaptonemal complex (SC). Oocytes were categorised into different stages of prophase I depending on the stage of SC formation. Oocytes were classed as pre-pachytene (leptotene or zygotene) when the SC was seen to be assembling (Bi), whereas oocytes with fully synapsed SC, were categorised as pachytene stage oocytes (Bii). Finally, if the SC could be seen to be disassembling, but still present, oocytes were classed as being at the diplotene stage (Biii). Histogram shows the percentage of oocytes at pre-pachytene (leptotene/zygotene), pachytene or diplotene stages in control vs B[a]P-treated ovaries (C). Scale bars represent 10 µm in (A) and 5 µm in B(i) and (ii). Bars denote mean + s.e.m.; n = 6. Star denotes significant differences relative to control (*P < 0.05).

Figure 6

Neonatal ovaries cultured with B[a]P contain fewer healthy follicles. (A) Representative histological sections of neonatal ovaries cultured for 6 days in either (Ai) control conditions, or with increasing concentrations of B[a]P: (Aii) 0.01 µg/mL, (Aiii) 0.1 µg/mL or (Aiv) 1 µg/mL. Histograms show (B) the number and (C) the distribution of healthy follicles across different B[a]P treatment groups. Scale bars: 50 µm. Bars denote mean + s.e.m.; n = 8 for control group, n = 7 for 0.01 µg/mL 10 and 0. 1 µg/mL group and n = 9 for 1 µg/mL group. Stars denote significant differences relative to control (*P < 0.01, **P < 0.01, ***P < 0.001).

Figure 7

Apoptosis is increased, while somatic cell proliferation is decreased, following exposure to B[a]P. (A) Representative photomicrographs of cultured neonatal (Ai) control and (Aii) B[a]P treated tissues showing the localisation of CC3 (green) and BrdU (red), counterstained with DAPI (blue). Insets show higher magnification images of ovarian follicles. (B) Each fluorophore area (except for DAPI) was measured as a percentage of total area of the section using ImageJ/Fijij software. (C) Histograms show the percentage of the tissue section that is BrdU positive and (D) the percentage of tissue section that is positive for CC3. Green arrows show examples of cells expressing CC3 and red arrowheads show the localisation of BrdU within cells. Scale bars denote 50 µm, 20 µm in insets (Ai, ii) or 25 µm for images showing analysis of signal area (Bi-iii). Bars denote mean + s.e.m.; n = 8 for both groups. Stars denote significant differences relative to control (***P < 0.001).

Figure 8

B[a]P does not significantly affect ovarian MDA expression. (A) Representative photomicrographs of (Ai) control and (Aii, Aiii) B[a]P-treated cultured ovaries showing immunohistochemical localisation of MDA (red), DDX4 (green) and DAPI (blue). Due to the large variation of MDA signalling observed in B[a]P ovaries, representative images from ovaries with low (Aii, Bii) and high (Aiii, Biii) levels of MDA expression have been shown. (B) Shows higher power (100×) magnification images of (Bi) control and (Bii, Biii) B[a]P-treated ovaries. (C) Histogram shows the percentage of tissue area positive for MDA for both groups. (D) Histogram shows the intensity of MDA expression per tissue area for both groups. Scale bars denote 50 µm, or 10 µm in the high-power images. Bars denote mean + s.e.m.; n = 9 for both groups.

Figure 9

Melatonin does not protect the ovary from B[a]P-induced depletion of healthy follicles. (A–D) Representative photomicrographs of (A) control, (B) B[a]P-treated, (C) MT-treated and (D) B[a]P + MT-treated ovaries, stained with haematoxylin and eosin. Histograms show the (E) total number of healthy follicles and (F) distribution of healthy follicles in ovaries following exposure to B[a]P alone (blue), MT alone (green) or both B[a]P and MT (purple). Bars denote mean + s.e.m.; n = 6 for all experimental conditions. Columns with different letters are statistically significant to each other (P < 0.05).