The Aryl Hydrocarbon Receptor Mediates Sex Ratio Distortion in the Embryos Sired by TCDD-Exposed Male Mice

Abstract

Many reports describe an association between preconceptional paternal exposure to environmental chemicals, including the persistent organic pollutant 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) with an increased number of female offspring. We chronically treated wild-type C57BL/6 male mice with TCDD to investigate a role for the aryl hydrocarbon receptor (AHR) transcription factor. These mice had a 14% lower male:female sex ratio than control mice, which was not observed in TCDD-treated Ahr knock out mice. AHR target genes Cyp1a1 and Ahrr were upregulated in the liver and testis of WT mice and Ahr expression was higher in the epididymis (2-fold) and liver (18-fold) than in whole testis tissue. The AHR protein was localized to round spermatids, elongating spermatids, and Leydig cells in the testis of WT mice. These studies demonstrate AHR involvement in the sex ratio distortion of TCDD-exposed males and the need for evaluating the molecular and genetic mechanism of this process.

2. Introduction

The influence of a preconceptional paternal chemical exposure on offspring outcomes is a critical component of developmental toxicology that is often overlooked. Importantly, there are many chemicals that interact with the male reproductive system [1–4] and thereby have the potential to impact the development and future health status of offspring. One such example is the persistent organic pollutant, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Produced as a result of incomplete combustion (i.e., waste incineration, forest fires), as a contaminating by-product in chemical reactions or manufacturing processes, or as a consequence of industrial accidents, TCDD has been associated with a number of negative health effects in humans including the hallmark of TCDD exposure, a skin keratinizing condition called chloracne [5]. Furthermore, there is epidemiological evidence that men exposed to TCDD, occupationally or through environmental disasters, father more daughters than sons [6–10]. The most infamous example of TCDD-associated offspring sex ratio distortion occurred in 1976 following an industrial accident when an herbicide production plant exploded near Seveso, Italy, releasing a cloud of dioxin that blanketed nearby communities. Exposed men with the highest serum concentrations of TCDD (281–26400 ppt) had a female-biased offspring sex ratio of 0.383 (# males born/total number of children born; 95% CI 0.28–0.49) compared to the sex ratio in the unexposed population (0.557, 95% CI 0.49–0.61) [7]. The authors noted that the association was specific to the paternal serum concentration of TCDD and that the maternal concentration did not correlate with offspring sex ratio. Moreover, male occupational exposures to TCDD during the 1970’s and 1980’s in Austrian, Russian, and New Zealand chemical plants, were significantly associated with a female predominant offspring sex ratio [8–10].

A variety of animal studies have addressed the impact of TCDD on sex ratio following a maternal in utero exposure [11–13]; however Ishihara et al. were the first to focus on the paternal influence in a mouse model and to lay the groundwork for determining the underlying mechanism [14,15]. Adult, male ICR mice were treated with TCDD (2000 ng/kg initial, 400 ng/kg weekly) for 6 weeks after which they were mated with naïve female mice. The sex ratio was determined in the resultant offspring on postnatal day (PND) one or at the two-cell stage of development, and there was a significant reduction in both sex ratios compared to the control groups (PND 1: Control: 0.531 ± 0.017, TCDD: 0.462 ± 0.021; 2-cell: Control: 0.540 ± 0.0154, TCDD: 0.479 ± 0.020). While the impact of paternal TCDD exposure on a female-biased sex ratio was recapitulated in mice, the involvement of the aryl hydrocarbon receptor (AHR) was not examined. The AHR transcriptional signaling pathway is activated primarily in the livers of TCDD-exposed mice (i.e., activation of cytochrome p450 1A1, Cyp1a1 [16]); however, AHR is also expressed in the male reproductive tract of mice [17,18] suggesting that AHR in male-specific organs may play a role in TCDD-mediated sex ratio distortion. The purpose of this study was to verify the effect of paternal TCDD exposure on offspring sex ratio in a mouse model, and to characterize the involvement of the aryl hydrocarbon receptor (AHR) in wild-type mice as part of a long-term goal of determining the underlying mechanism of action.. Our findings demonstrate the involvement of the AHR in the biasing toward a greater ratio of female offspring from exposed males and provide evidence for gene expression changes in the exposed wild-type males.

3. Methods and Methods

3.1. Chemicals

All chemicals were obtained from Sigma-Aldrich (St. Louis, MO) unless otherwise noted.

3.2. Mice

All animal procedures were approved by the University of Pennsylvania Institutional Animal Care and Use Committee. Adult male and female C57BL/6J (WT) and B6.129-Ahr^tm1Bra/J (Ahr knock-out, KO) mice were purchased from Jackson Laboratory (Bar Harbor, ME) and allowed access to feed and water ad libitum. Male and female mice were mated to breed young male mice and begin treatment (corn oil (CO) or TCDD in CO) at 4 wks of age. Mouse genotypes (WT vs KO) were determined following preparation of DNA from mouse ear pieces with the DirectPCR Lysis Reagent (Mouse Tail, Viagen Biotech, Los Angeles, CA) and PCR using the protocol and primers recommended by The Jackson Laboratory (https://www2.jax.org/protocolsdb/f?p=116:5:0::NO:5:P5_MASTER_PROTOCOL_ID,P5_JRS_CODE:20775,002831). Beginning at 4 wks of age, male mice (n=5–9) were injected intraperitoneally (IP) with vehicle control (5 ml/kg CO) or TCDD (wk 1: 2000 ng/kg, wks 2–22: 400 ng/kg in CO vehicle) weekly for 2 or 22 wks. For the mice treated with TCDD for 22 wks, adult male mice were mated with female Hsd:NSA(CF-1) mice (Envigo, Huntingdon, Cambridgeshire, United Kingdom) every other week for a total of 10 matings. Embryos (gestation day (GD) 15.5) from pregnant CF-1 mice were collected, weighed, and frozen on dry ice. Following the 2 and 22-week treatments, livers and testes were collected, weighed, and snap-frozen or fixed in Bouin’s Solution. Cauda epididymal sperm were allowed to swim up into DMEM (Life Technologies, Carlsbad, CA) for 10 min at 37°C, 5% CO₂ after making multiple cuts in the cauda epididymis. Sperm were then processed for sex chromosome content by multiprobe fluorescent in situ hybridization (MP-FISH).

3.3. Mouse Embryo Genotypic Sex Determination

Embryo (GD 15.5) DNA was prepared using DirectPCR Lysis Reagent (Mouse Tail). As previously described (Prantner et al. 2016), embryo DNA was subjected to real-time quantitative PCR with one primer set that detects a gene on the X chromosome (Kdm5c) and a gene on the Y chromosome (Kdm5d) (Supplemental Table 1). Melt curve analysis revealed male embryos to have two peaks (Kdm5c and Kdm5d, 83°C and 81°C) and female embryos to have one peak (Kdm5c, 83°C). Sex ratios for offspring fathered by each mouse sire were calculated by dividing the total number of male embryos across all litters divided by the total number of embryos.

3.4. Multiprobe Fluorescence In Situ Hybridization

Cauda epididymal sperm in DMEM were centrifuged at 1000 rpm for 5 min, resuspended in 75 mM KCl, and incubated at 37°C for 30 min. The sperm were centrifuged at 1000 rpm for 5 minutes, resuspended in 75 mM KCl and diluted to 1,500,000 sperm/ml before being spread on a poly-l-lysine coated cover slip (Neuvitro, Vancouver, WA). Coverslips were stored at −20°C until MP-FISH was carried out. First, coverslips were incubated with 10 mM dithiothreitol (pH 8.0) on ice for 30 min, following which the coverslips were rinsed with MilliQ water before being fixed with methanol:acetic acid (3:1) 3 times for 2 min each. The sperm DNA was denatured with 70% formamide at 75°C for 5 min and dehydrated using a series of ethanol washes (70%, 85%, 100%; 2 min each). Probes specific for the murine X and Y chromosomes (Oxford Gene Technology Inc, Tarrytown, NY) were mixed, applied to the coverslips and denatured at 75°C for 5 min. Probe hybridization occurred overnight at 37°C. Following a series of washing steps, DAPI was applied to the coverslips and the slides were viewed and imaged using a Leica DM6000 Widefield fluorescence microscope (Buffalo Grove, IL) and Photometrics HQ2 high resolution monochrome CCD camera (Tucson, AZ) in the University of Pennsylvania Cell and Developmental Biology Microscopy Core Facility. The GFP, Cy3, and DAPI filters were used to image the X chromosome, Y chromosome, and total DNA, respectively. Using the 20X objective, 924–1115 sperm were counted using the ImageJ Cell Counter plugin for each treatment set and scored as X chromosome-bearing (GFP) or Y chromosome-bearing (Cy3) sperm.

3.5. RNA Isolation and Quantitative Real-Time PCR

Total RNA was isolated from frozen liver, testis, and cauda epididymis samples using TRI Reagent and 1-bromo-3-chloropropane extraction procedure. Briefly, samples were disrupted in TRI Reagent using a handheld homogenizer, following which 1-bromo-3-chloro-propane was added to the sample and incubated for 10 min at room temperature. Samples were centrifuged at 12,000 × g for 15 min at 4°C and the top layer containing the RNA was then precipitated and washed with 2-propanol and 75% ethanol. The resultant RNA was reconstituted with Molecular Biology Grade Water (Corning, Corning, NY). DNase digestion and RNA clean up were performed using a DNase I Set (Zymo Research, Irvine, CA) and RNeasy Mini Kit (Qiagen, Germantown, MD) according to the manufacturers’ instructions. A Nanodrop Spectrophotometer (Fisher Scientific, Waltham, MA) was used to determine total RNA concentration and purity (260/280). Using 500 ng RNA, cDNA was generated with the High Capacity cDNA Reverse Transcription Kit (Life Technologies) and a Techne TC-5000 thermal cycler (Burlington, NJ). Real-time qPCR was performed using PowerUp™ SYBR™ Green Master Mix (Life Technologies), cDNA, gene-specific forward and reverse primers (Integrated DNA Technologies, Inc., Coralville, IA; Supplemental Table 1), and a QuantStudio 7 Flex Real Time PCR System (Life Technologies). Delta delta cycle threshold (ΔΔCt) values were generated from Ct values by comparison to β-actin (Actb) as a reference gene and the WT CO-exposed mice.

3.6. Immunohistochemistry

Tissue fixed in Bouin’s Solution was processed, embedded in paraffin, and cut into 5 μm sections. Samples were deparaffinized and rehydrated using xylene and a graded series of ethanol washes (100%, 95%, 80%, 70%, 50%, water). Antigen retrieval was performed by boiling sodium citrate buffer (10 mM sodium citrate, 0.05% Tween-20, pH 6.0), following which endogenous peroxidases were quenched with 3% hydrogen peroxide in methanol. Sections were blocked with 10% normal goat serum (Vector Laboratories, Burlingame, CA) in 1X PBS for 1 h. The AHR primary antibody (Abcam Ab84833, Cambridge, UK) or IgG control (Vector Laboratories) were diluted to 1:100 in 10% goat serum and incubated with the deparaffinized sections at 4°C overnight. Following three washing steps (5 min, 0.5% Tween-20/PBS, 5 min 0.1% Tween-20/PBS, 5 min PBS), anti-rabbit secondary antibody (1:100, VectaStain Elite Kit, Vector Laboratories) was incubated with the sections for 1 h at room temperature. Following three washing steps (5 min, 0.5% Tween-20/PBS, 5 min 0.1% Tween-20/PBS, 5 min PBS) Avidin (Reagent A) and biotinylated horseradish peroxidase (Reagent B) were mixed and added to the sections according to the manufacturer’s instruction (VectaStain Elite Kit, Vector Laboratories). After washing in Tween-PBS, the peroxidase substrate solution was added to the samples for 1 min during which the brown stain developed. Slides were counterstained with hematoxylin for 30 s and coverslips were applied before analysis. A Nikon Eclipse TE-2000 inverted microscope (Nikon Instruments, Melville, NY), CFW1610C Digital Firewire Camera (Scion, Frederick, MD), and NIH ImageJ imaging software (https://imagej.nih.gov/ij/) were used to capture images.

3.7. Statistical Analysis

Data are presented as medians with 95% confidence interval and analyzed using GraphPad Prism 8.0 (GraphPad Software Inc., La Jolla, CA). Sample size per treatment group varying from 5 to 9 male mice. Mann Whitney test was used to compare median values of two treatment groups with nominal p<0.025 as statistically significant to account two tests for WT and KO mice respectively. Kruskal-Wallis test with Dunn’s multiple comparison test was used to compare median values of >2 treatment groups; adjusted p<0.05 is considered statistically significant.

4. Results

4.1. AHR is required for the offspring sex ratio distortion in TCDD-treated sires.

We first set out to determine whether wild-type (WT) and Ahr KO male mice exposed to TCDD exhibit an altered offspring sex ratio when mated with naïve female mice. WT C57BL/6 (n=9, 4 weeks of age) and Ahr KO C57BL/6 (n=6–7, 4 weeks of age) males were treated weekly with either TCDD (wk 1: 2000 ng/kg, wks 2–22: 400 ng/kg in CO vehicle) or corn oil vehicle (5 ml/kg CO) over the course of 22 weeks and each were mated with a naïve Hsd:NSA (CF-1) female mouse every other week beginning 2 weeks following the initial TCDD treatment. Male mouse body weights and testis:body weight ratios did not vary between treatment groups (Supplemental Table 2). Liver:body weight was 25–30% greater in WT mice as compared to KO mice, which is in line with previous findings that Ahr KO mice have a reduced liver weight (Schmidt et al. 1996). Pregnancy success, percentage of resorptions, and embryo weight did not vary between treatment groups for both WT and KO mice, respectively. While litter size was 25% lower in the WT TCDD mice compared to the WT CO group, the difference was not statistically significant (Table 1).

Table 1.

Fertility parameters and embryo weights

	WTa		KOa
	Corn Oil	TCDD	Corn Oil	TCDD
Pregnancy success (# pregnancies/# matings)	0.50 (0.38–0.62)	0.50 (0.34–0.53)	0.52 (0.32–0.65)	0.50 (0.28–0.63)
Litter size (# embryos per litter)	9.0 (6.9–10.2)	6.9 (5.5–8.6)	11.9 (8.79–13.7)	11.2 (9.42–15.1)
Resorptions (%)	3.92 (2.14–5.24)	0.0 (−0.17–3.5)	4.94 (1.59–10.7)	3.57 (0.865–7.58)
Embryo weight (g)	0.252 (0.206–0.307)	0.268 (0.182–0.347)	0.296 (0.214–0.425)	0.287 (0.234–0.335)

^aAll data presented as median (95% CI)

The sex of each embryo recovered from the pregnant dams was determined by the procedure of Prantner et al. (2016). The median sex ratio from embryos sired by WT males mice treated with TCDD was 0.448 (95% CI 0.390–0.522), significantly lower than the median of 0.517 (95% CI 0.484–0.613) sired by WT CO-treated mice (Mann-Whitney test, p=0.0179), approximately by 14%. This effect was not observed in the offspring of the Ahr KO mice (Mann-Whitney test, p=0.876) with a median of 0.550 (95% CI 0.461–0.638) versus 0.559 (95% CI 0.532–0.625) from CO-treated and TCDD-treated mice, respectively (Figure 1). These results demonstrate that the female-biased offspring sex ratio observed in TCDD-treated WT male mice is dependent on AHR.

Figure 1. Mouse embryo sex ratio following TCDD exposure.

To identify genotypic male (XY) and female (XX) GD 15.5 embryos, genes specific to the X (Kdm5c) and Y (Kdm5d) chromosomes were detected by qPCR. For each adult male mouse, the overall sex ratio was determined by dividing the number of male offspring by the total number of all offspring across 2–9 litters. Data are presented as box and whisker plots. The line represents the median, while the dots represent individual male mice. The box represents 50% of the data and the whiskers are those data in the 25^th and 75^th quartiles (n=6–9 male mice). *Statistically significant differences (p<0.025) compared with WT CO.

4.2. TCDD treatment of males does not alter the ratio of X or Y chromosome-bearing sperm.

To determine the effect of TCDD on the ratio of X and Y chromosome-bearing sperm, cauda epididymal sperm were collected from all four groups (WT CO, WT TCDD, KO CO, KO TCDD) after 22 wk of treatment and genotyped for X or Y chromosomes using MP-FISH. The percentage of X chromosome-bearing cauda epididymal sperm was near 50% and did not vary between treatment groups for both WT and KO mice, respectively (all p>0.025; Table 2, Figure 2) suggesting that TCDD may alter the functionality of the X/Y sperm or the viability of male embryos.

Table 2.

Summary of MP-FISH detection of X chromosomes in mouse sperm

	WT		KO
	Corn Oil	TCDD	Corn Oil	TCDD
Sample size (n)	6	6	5	5
Total sperm counteda	1013 (945.1–1076)	1057 (814.9–1142)	1022 (869.8–1095)	1005 (875.9–1080)
Percent X sperma	50.1 (48.7–52.4)	50.1 (48.7–51.2)	50.6 (49.5–53.2)	51.2 (50.2–52.4)

^aData represent median (95% CI)

Figure 2. Determination of mouse sperm sex chromosome content by MP-FISH.

Sex chromosome content of caudal epididymal sperm was assessed using fluorescent probes specific for the X (green) and Y (red) chromosome. Sperm were counterstained with DAPI (blue). Images are representative from each treatment group (scale bar: 50 μm).

4.3. TCDD activates the AHR signaling pathway in wild-type but not Ahr KO testes and livers.

We then sought to demonstrate the activation of the AHR signaling pathway in the testes and livers of male mice by treating WT and Ahr KO mice with TCDD or vehicle for two weeks (Supplemental Table 3). We found that the AHR signaling pathway was activated in the livers and testes of WT mice treated with TCDD for two weeks as the AHR target genes aryl hydrocarbon receptor repressor (Ahrr) and Cyp1a1 were 2–3000 fold higher in the WT TCDD mice as compared to the WT CO (Figure 3A), demonstrating that the AHR signaling pathway is activated in the livers and testes of WT male mice following TCDD treatment. The same phenomenon did not occur in the KO mice, however KO mice treated with TCDD exhibited 20% reduction in Ahrr expression in the testis compared to KO mice treated with vehicle, but this was not observed in the liver.

Figure 3. Liver and testis gene expression analysis.

(A) Expression of hepatic mRNAs from the AHR signaling pathway (Ahrr, Cyp1a1) were assessed by qPCR. (B) Expression of testis mRNAs from the AHR signaling pathway (Ahrr, Cyp1a1) sex ratio determining genes (Slx, Sly), and sperm motility gene (Akap4) were assessed by qPCR. All expression values were normalized to vehicle treated WT mice and the Actb housekeeping gene. All data are presented as box and whisker plots. The line represents the median, while the dots represent individual male mice. The box represents 50% of the data and the whiskers are those data in the 25th and 75th quartiles (n=5–6). *Statistically significant differences (p<0.025) compared to WT CO. ^#Statistically significant differences (p<0.025) compared to KO CO.

4.4. TCDD does not differentially affect expression of candidate genes for offspring sex ratio distortion.

To begin to tease out the underlying mechanism by which TCDD exposure leads to a female-biased sex ratio, we assessed the expression of candidate germ cell-specific genes involved in sex ratio determination and sperm function in mice. In males where the relative levels of Sycp-like-X-linked, Slx, and Sycp-like-Y-linked, Sly, mRNAs have been experimentally manipulated, the ratios of Slx:Sly mRNA influence the sex ratios of offspring [19]. Furthermore, Sly contains a dioxin (xenobiotic) response element and, therefore, may be a target for TCDD [20–22]. Another gene, Akap4, encodes the major fibrous sheath protein and is expressed from the X chromosome [23,24]; its protein product, AKAP4, is essential for male fertility [25]. Despite the AHR pathway being activated in the testis, TCDD did not have an impact on the levels of Slx, Sly, and Akap4 in mice treated for two weeks (Figure 3B). These genes and 49 other candidate genes specific to germ cells or sex hormone signaling were also measured in mice treated with TCDD for 22 weeks and there was no effect of the chemical exposure (Supplemental Figure 1).

4.5. Ahr/AHR are expressed in somatic cells and spermatids of the testis.

To gain a better understanding of Ahr/AHR expression patterns in naïve WT mice, we assessed the localization of AHR protein in the testis by immunohistochemistry and levels of the Ahr mRNA in the testis, cauda epididymis, and liver. We observed the AHR protein to be uniform throughout the adult WT testis, localized to somatic (Leydig cells) and haploid germ cell (round spermatids, elongating spermatids) populations (Figure 4). The localization of the AHR protein was associated with the developing acrosome of round and elongating spermatids and, possibly, the nucleus of elongating spermatids. While AHR protein was present in the testis, Ahr mRNA levels were 2- and 18-fold higher in the cauda epididymis and liver, respectively (Figure 5).

Figure 4. Localization of AHR in mouse testis.

Immunohistochemical detection of AHR in testis of male mice. Paraffin-embedded sections (5 μm) were stained using the AHR antibody (scale bar: 50 μm). AHR staining was localized to elongating spermatids (E), Leydig cells (L), and round spermatids (R). An IgG isotype control antibody was used as a control and showed no brown staining.

Figure 5. Ahr mRNA expression.

Expression of Ahr mRNA in the testis, cauda epididymis, and liver was assessed by qPCR. All expression values were normalized to testis Ahr mRNA expression and the Actb housekeeping gene. All data are presented as box and whisker plots. The line represents the median, while the dots represent individual male mice. The box represents 50% of the data and the whiskers are those data in the 25th and 75th quartiles (n=3–5). *Statistically significant differences (p<0.05) compared to testis Ahr.

5. Discussion

5.1. Significance of our findings

The goals of this study were to verify and establish the mouse as an appropriate model for assessing the sex ratio distortion observed for offspring that were sired by males exposed to TCDD and, further, to characterize the involvement of the AHR signaling pathway as a prelude to determining the cellular and molecular mechanisms underlying this phenomenon. These aims are important because an increasing number of reports around the world describe a primarily female offspring sex ratio fathered by males which have been exposed to dioxins and some other environmental toxicants. Furthermore, these studies may shed light on those populations where sex ratio distortion is observed, yet the responsible chemical insult is still to be determined [6,8–10,26–30]. Related findings are exemplified by the United States and Japanese populations at large that have experienced a greater proportion of female births than had been historically recorded for each population between the 1940’s and 1960’s [31–33]. While it is unlikely that a single mechanism explains how sex ratio distortion occurs within each population, uncovering the molecular underpinnings of the relationship between paternal TCDD exposure and offspring sex ratio may aid in the interpretation of epidemiological data and provide opportunities for environmental chemical exposure intervention.

5.2. Reproducibility in a mouse model

While much of the available data regarding paternal TCDD-mediated offspring sex ratio distortion is epidemiological and associative, one other research group has drawn a causative link using a mouse model. In outbred ICR male mice, Ishihara et al. [14,15] found that following a 6-wk exposure (7–12 wks of age) to TCDD (2000 ng/kg loading dose, 400 ng/kg weekly dose) by oral gavage, there was a significant reduction in the primary (2-cell stage) and secondary (postnatal day 1) sex ratio of the offspring by 11–13%. We sought to more closely model the exposure pattern observed in Seveso, Italy, and thereby differed from those of Ishihara et al. in many ways [14,15]. First, the Ishihara et al. team utilized the outbred ICR strain of mice, which are derived from CD-1 mice that we have determined carry the less sensitive Ahr d allele (Bircsak, Gerton unpublished). In our study, we employed inbred C57BL/6 males with the Ahr^b1 allele, which codes for an AHR protein with a high affinity for TCDD [34]. The TCDD doses we administered (wk 1: 2000 ng/kg, wks 2–22: 400 ng/kg) were chosen based on those used in Ishihara et al., 2007, 2010 to maintain a constant body burden of TCDD over time. However, our dosing methods were dissimilar as Ishihara et al. used exposure by oral gavage and we used intraperitoneal (IP) injection. While we did not measure the TCDD concentration present in the blood, it is plausible that the mice in this study experienced higher TCDD blood concentrations following IP administration due to the initial bypassing of portal circulation/hepatic metabolism. We began TCDD treatment at 4 weeks of age during which the male reproductive system of the mouse is undergoing dynamic, pubertal development [35]. This is important because the men in Seveso, Italy who experienced the lowest sex ratio (0.38) were first exposed to TCDD before the age of 19 [29]. Further, we treated the male mice for an extended period of time (22 weeks) as the TCDD persisted in the Seveso area environment following the disaster for many years; the dioxin has a half-life in soil of approximately 9.1 years [36]. This suggests that the population was continually exposed to the chemical, potentially affecting multiple pregnancies over the course of a father’s reproductive life span. Finally, we collected embryos prior to parturition to avoid the loss of pups during birth and/or as a result of maternal cannibalism, which could skew the sex ratio independent of TCDD exposure. Despite these differences in study design, we observed a female sex ratio bias of a similar magnitude to that of the Ishihara et al. [14,15] studies following TCDD treatment (Table 3). All together, these data suggest the effect of paternal TCDD exposure on offspring sex ratio is robust and can be recapitulated in inbred as well as outbred mouse models. Understanding the sex ratio at fertilization (primary) and birth (secondary) as well as the ratio of X and Y chromosome-bearing sperm is crucial to identifying the cause of offspring sex ratio distortion. Our findings on secondary sex ratio and the unaltered percentage of X chromosome-bearing sperm in TCDD-exposed sires is in line with the studies by Ishihara et al [14,15]. While we did not assess primary sex ratio, Ishihara et al. reported a female predominant primary sex ratio in preimplantation embryos conceived with sperm from TCDD-exposed sires. When interpreted together and because the sex ratio distortion was observed soon after fertilization, these findings suggest that X and Y chromosome-bearing sperm are functionally different. In preliminary studies we found that cauda epididymal sperm motility of the entire sperm population (X and Y) did not vary between all 4 treatment groups in this study (WT CO, WT TCDD, KO CO, KO TCDD) (data not shown). Future experiments should confirm the effect of TCDD on primary sex ratio and assess the motility and fertilization success of separated X and Y chromosome-bearing sperm populations to identify deficits in Y chromosome-bearing sperm function (motility, acrosomal exocytosis, capacitation) and/or improvement in X chromosome-bearing sperm function following paternal TCDD treatment.

Table 3.

Summary of offspring sex ratio following paternal TCDD exposure in this and other studies

Study	Offspring age	Offspring sex ratio
		Control	TCDD
Bircsak et al. 2020a	GD 15.5	0.517 (95% CI 0.484–0.613)	0.448 (95% CI 0.390–0.522)
Ishihara et al., 2010b	2-cell	0.540 ± 0.0154	0.479 ± 0.020
Ishihara et al., 2006b	PND 1	0.531 ± 0.017	0.462 ± 0.021

^aData presented as median (95% confidence interval)

^bData presented as mean ± standard deviation [14,15]

GD: Gestational day

PND: Postnatal day

5.3. TCDD operates through the AHR to alter the offspring sex ratio of exposed males

For the first time, we demonstrate that when male mice lacking AHR (KO) are exposed to TCDD, the offspring sex ratio is near normal (median=0.559, 95% CI 0.532–0.625) and 20% higher than treated sires expressing AHR (median=0.448, 95% CI 0.390–0.522) confirming the involvement of AHR in TCDD-mediated offspring sex ratio distortion of male mice. This is in line with other studies that have described the dependence of polyaromatic hydrocarbon (i.e., TCDD, benzo[a]pyrene) on the AHR signaling pathway to elicit biological effects including tumor occurrence, Cyp1a1 activation in the liver and skin [37], and thymic atrophy [38].

Recently, You et al. [39] explored the effects of TCDD on sperm function by evaluating changes on mouse sperm motility (%), viability (%) and capacitation status of spermatozoa following in vitro exposure of these cells to TCDD. These authors concluded that short-term, direct exposure of sperm to TCDD decreases sperm motility and viability, as well as increasing acrosome reaction rates. They also reported indirect evidence that exposed Y spermatozoa survive for less time than X spermatozoa at high concentrations of TCDD and the decreased sex ratio of embryos is associated with the short lifespan of Y spermatozoa; in other words, direct TCDD exposure affects the fertility of Y spermatozoa more than X spermatozoa. This study has some methodological challenges with regard to assigning mobility, viability, and capacitation status properties to individual sperm that then must be karyotyped for X and Y chromosomes. In addition, the authors do not provide a mechanism to explain how treatment of sperm with TCDD, a chemical working through a transcription factor (AHR), can directly and differentially affect the function of Y chromosome-bearing sperm cells that are transcriptionally inactive.

5.4. TCDD alters gene expression in the testes of wild-type but not Ahr KO male mice.

To elucidate the molecular mechanism by which offspring sex ratio is distorted by TCDD exposure of male mice, we measured the expression of several testis-specific genes that we hypothesized may play a role in sex ratio determination in WT mice following TCDD exposure. Slx and Sly are mouse X and Y chromosome-linked multicopy genes that have antagonistic effects on sex chromosome-specific gene expression [40]. Genetic knock-down of Slx or Sly in males results in a male (0.60) or female predominant (0.478) offspring sex ratio, respectively [19,40]. Despite Sly containing a dioxin response element [20–22], the gene was not upregulated in the testis of WT mice treated with TCDD (Figure 3B). We also assessed the expression of Akap4, an Xchromosome-linked gene that codes for a major fibrous sheath protein in the sperm flagellum [23,24], that is essential for fertility [25] and plays a role in sperm motility of both X and Y chromosome-bearing sperm [41]. TCDD did not impact Akap4 expression in WT male mice. Interpretation of these data is aided by also examining the expression of known AHR target genes Ahrr and Cyp1a1 in the testis and liver of male mice. While Ahrr and Cyp1a1 were upregulated approximately 2-fold in the testis of WT-mice, both target genes were upregulated to a much greater extent in the liver (10–1500 fold higher), a difference explained by the abundant expression of AHR in the liver as compared to the restricted cellular expression in the testis (Figure 5). AHR protein was localized by immunohistochemistry to post-meiotic germ cells (round spermatid and elongating spermatids) and Leydig cells. More specifically, AHR was expressed in the acrosomal region of round and elongating spermatids, in agreement with Hansen et al. [18]. These investigators also reported AHR expression in the flagellum of the sperm, a finding not corroborated by our studies. Curiously in our studies, Ahr KO mice treated with TCDD exhibited approximately 20% lower Ahrr gene expression than vehicle-treated KO mice, which did not occur in the liver and suggests that TCDD may operate through an AHR-independent route to influence Ahrr expression. In the future, analysis of isolated, specific cell types (e.g., round spermatids) may amplify the observed effect of TCDD on known target genes and reveal novel effects on other genes.

5.5. What is the mechanism for this effect?

This report establishes the mouse as an excellent model for examining the mechanism involved in biasing the sex ratios of offspring conceived by fathers exposed to TCDD. The fact that Ahr KO males do not exhibit an altered sex ratio constitutes proof that TCDD works through this receptor and, most likely, involves a difference between gene expression in germ cells of control and exposed sires. We expected the inbred strain of mice used in our study (C57BL/6J) to exhibit a greater effect compared to the studies of Ishihara et al. [14,15], who used outbred ICR mice expressing a less sensitive allele, Ahr^d (Bircsak, Gerton, unpublished), compared to the C57BL/6J strain, which carries the Ahr^b1 allele [42]. That a more extreme offspring sex ratio was not observed in our studies may reflect the complex nature of spermatogenesis, AHR function within male germ cells, and, potentially, not AHR-dependent responses to TCDD as evidence by the effect of TCDD on Ahrr mRNA levels in the Ahr KO mouse testis (Fig 3B).

Curiously, we noted that the sizes of litters resulting from TCDD WT males were smaller in comparison to litters from the control WT mice. On the other hand, litters from KO males were larger than WT regardless of whether they had been treated with TCDD or not (Table 1) The significance of these findings are not clear but suggest that AHR make play a role in regulating litter size. In WT males, the endogenous levels of the activated AHR transcription factor may limit the litter sizes but when the AHR activity is up-regulated by TCDD, the litter sizes decreased. On the contrary, when AHR is absent in the KO males, the litter sizes are larger but TCDD had no effect. Since the dams used in this study were cycling naturally, the impact on litter size is due to an effect on the male, suggesting AHR may be controlling some feature influencing the efficacy of the sperm or perhaps some other feature of semen such as seminal plasma proteins or non-coding RNAs.

The actual cellular and molecular mechanisms that would yield a greater likelihood of female offspring following paternal exposure to TCDD (or other environmental toxicants) are still not clear. The skewing of the sex ratio at conception [15] suggests that X chromosome-bearing sperm somehow gain a competitive advantage over Y chromosome-bearing sperm following exposure of adolescent or adult males to TCDD. The functional non-equivalence of X and Y sperm contravenes the generalized concept that genotypically haploid spermatids are phenotypically diploid as a consequence of the sharing of messenger RNAs and/or proteins between conjoined spermatids through the intercellular bridges formed by incomplete cytokinesis following mitosis and meiosis [43,44]. Gene expression differences between X and Y chromosome-bearing spermatids due to genes, such as Slx, Slxl1, and Sly [19,40,45], or endocrine disruption actions of TCDD [46–48] do not explain the sex ratio alteration at the cellular level. Further studies are warranted on this problem, with a focus on defining how and why paternal exposure to TCDD alters the functional characteristics of X and Y chromosome-bearing sperm.

Highlights.

• C57BL/6 male mice chronically treated with 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) sired more female than male pups.

• These mice had a 14% lower male:female offspring sex ratio than untreated sires.

• The offspring sex ratio distortion was not observed in TCDD-treated Ahr knock-out mice.

• AHR target genes were upregulated in the liver and testis of wild-type mice but not Ahr knock-out mice.

• The AHR protein was localized to round spermatids, elongating spermatids, and Leydig cells in the testis of WT mice.

Abbreviations:

AHR

aryl hydrocarbon receptor

CI

confidence interval

GD

gestation day

KO

knock out

MP-FISH

multi-probe fluorescence in situ hybridization

PND

postnatal day

TCDD

2,3,7,8-tetrachlorodibenzo-p-dioxin

WT

wild-type