2,3,7,8-Tetrachlorodibenzo- P -Dioxin (TCDD) or Diethylstilbestrol (DES) Cause Similar Hematopoietic Hypocellularity and Hepatocellular Changes in Murine Fetal Liver,but Differentially Affect Gene Expression

Abstract

TCDD and DES have immunotoxic effects, including selective diminution of T lymphocyte progenitors in the fetal liver. The histologic presentation of fetal liver after exposure to either chemical has not been described. Similarly, limited information exists regarding mechanisms by which TCDD or DES may alter fetal hematopoiesis. Treatment of pregnant C57BL/6 mice with either 10 μg/kg/day TCDD or 48 μg/kg/day DES on gestation days (gd) 14 and 16 led to increased fetal liver weight on gd 18. Moderate anisocytosis and anisokaryosis with increased cytoplasmic and nuclear sizes, and increased cytoplasmic basophilia were present within hepatocytes after TCDD or DES. Both chemicals also decreased the presence of hematopoietic cells, however megakaryocyte numbers were unaffected. In contrast to these similar outcomes, real time quantitative PCR using a preliminary panel of 4 genes suggested that the chemicals act through different gene targets. TCDD increased c-jun gene expression in fetal liver, and decreased p53 without alteration in bcl-2 expression, indicating possible pro-proliferative and antiapoptotic effects. DES decreased c-jun and bcl-2, without altering p53, suggesting a shift away from proliferation. Both agents decreased PKCα expression, which may suggest shared decreased phosphorylation of substrates required for normal cell cycle progression.

Introduction

Tetrachlorodibenzo-p-dioxin (TCDD) is an inadvertent by-product of chemical synthesis, manufacturing, and combustion. Human environmental, occupational, and accidental exposures to TCDD occur, and are of concern as this agent is a carcinogen, teratogen, endocrine disrupter, and potent immunotoxicant (reviewed by Mann, 1997; Birnbaum and Tuomisto, 2000). Diethylstilbestrol (DES) has been prescribed as a therapeutic agent and used as a model estrogen for environmental contaminants that act through the estrogen receptor (reviewed by Chodak et al., 2002; Bamigboye and Morris, 2003). It also has teratogenic and carcinogenic properties, and is an immunotoxicant and a well-established endocrine disrupter (Ahmed, 2000; Palanza et al., 2001; Veurink et al., 2005). Both DES and TCDD cross the murine placenta, and can be found in fetal tissues in readily measured quantities (Shah and McLachlan, 1976; McLachlan, 1979; Nau and Bass, 1981; Nau et al., 1986).

Both TCDD and DES share similarities in their hepatic effects as well. Enzyme induction (Reilly et al., 1991; Kushwaha et al., 1996; Vogel et al., 1997), hepatomegaly, and alterations in serum chemistry have been noted after administration of either compound (Mann, 1997; Barnes et al., 1983; Birnbaum and Tuomisto, 2000). Additionally, TCDD and DES are known to have myelotoxic effects and alterations in the hematopoietic compartments of laboratory animals have been described (Boorman et al., 1982).

In the mid- to end-gestation rodent fetus, the liver is the primary organ of hematopoiesis. The bone marrow and spleen assume this function perinatally, as the liver shifts toward increased metabolic function (Landreth and Dodson, 2004). It is further becoming apparent that the resident non-hematopoietic cells of the fetal liver (hepatocytes, satellite cells, oval cells) both support and contribute to the early regulation of hematopoiesis. For instance, Hackney et al. (2002) derived a supportive stem cell line from murine fetal liver, which maintained hematopoietic stem cells. These authors demonstrated that select gene products of the fetal liver cells were responsible for the nurturing and support of the hematopoietic cells, permitting self-renewal without differentiation. More recently, natural killer cell expansion and differentiation from purified hematopoietic stem cells (Sca-1⁺Lin⁻) were promoted by co-culturing with hepatocyte cell lines (Bordoni et al., 2004). These results suggest the fetal liver may indirectly influence hematopoiesis via production of specific cytokines and stimulating factors. As such, the impact of chemical agents on the fetal hepatocyte population, although until recently not considered in this regard, may impact the co-developing hematopoietic component.

Light microscopic morphology of fetal mouse liver remains essentially uncharacterized after exposure to myelotoxicants, including TCDD and DES, agents known to target this compartment. The present paper therefore presents a microscopic evaluation of fetal liver and hematopoietic cells after late-gestation exposure to TCDD or DES, and visually demonstrates targeting of each population by both chemicals. TCDD and DES bind different cellular receptors, the aromatic hydrocarbon receptor (AhR) and estrogen receptor (ER), respectively. However, activation of these receptors following ligand binding may affect shared genes that regulate progression through cell cycle, proliferation or apoptosis. Therefore, as an additional evaluation of possible shared mechanisms of developmental myelotoxicity and potential hepatotoxicity, expression of a preliminary panel of genes in fetal livers of the chemical-exposed mice was determined after both chemicals.

Materials and Methods

Animal Model

Eight-week-old C57BL/6 timed pregnant female mice were purchased from Harlan Sprague–Dawley (Indianapolis, IN), and picked up at their local facility on the morning of gestation day (gd) 14. Mice were arbitrarily assigned to control, TCDD, or DES treatment groups. Insufficient time-pregnant mice were available from the supplier on any single experimental day to produce the target group size of 5–6 pregnant mice. Typically, 5 to 6 pregnant mice were obtained for each experiment, and results pooled across experimental days until the desired group size was obtained for evaluations. All animal care and treatment protocols were reviewed and approved by the Virginia Tech Institutional Animal Care and Use Committee (IACUC), prior to their initiation. Virginia Tech’s IACUC ensures that research involving animals is in compliance with the National Research Council / Institute of Laboratory Animal Resources (NRC/ILAR) “Guide for the Care and Use of Laboratory Animals” (http://www.nap.edu/readingroom/books/labrats/).

Chemical Exposure

TCDD (AccuStandard, Inc., New Haven, CT) was dissolved in corn oil (Sigma Aldrich, St. Louis, MO) to a concentration of 1.8 μg/mL (for the 10 μg/kg exposure) and administered in a volume of 120–200 μL by oral gavage. DES (Sigma Aldrich) was dissolved in corn oil to a concentration of 14.5 μg/mL (48 μg/kg exposure) and administered in a volume of 120–200 μL by oral gavage. Doses were administered on gd 14 and 16, based on the body weight of the pregnant dam. Control dams were administered comparable volumes of plain corn oil by oral gavage on the same dates.

Tissue Collection for Macroscopic and Microscopic Examination

On gd 18, pregnant dams were euthanized by cervical dislocation and fetal mice collected. Total fetal weight per litter as well as total fetal number per litter were recorded. Fetal livers were removed and placed in preweighed culture dishes containing 2 mL RPMI (Roswell Park Memorial Institute) culture media (Sigma Aldrich). Dishes were re-weighed (Mettler Toledo PB303, Carlton Scale, Roanoke, VA), and then stored in the refrigerator until livers were collected from all fetal mice. Two livers from each litter were arbitrarily removed from culture dishes and submersed in 10% neutral-buffered formalin for a minimum of 24 hours, for later light microscopic evaluation.

Histopathology

After fixation, fetal livers were embedded in paraffin, sectioned at 6 μm, and stained with hematoxylin and eosin. Livers were then evaluated microscopically. Sections were assessed for alterations in architecture or cellularity at 500× and 1000× magnification.

Livers were assessed for relative proportions of hematopoietic cells to hepatocytes. A minimum of 4 representative fields per slide was used to count hematocytes and hematopoitic cells, at 1000× magnification. Hematopoietic cells were also assessed for indications of morphologic changes associated with apoptosis. Hepatocyte morphology was evaluated for alterations in nuclear morphology and size, nuclear to cytoplasm ratios, and cytoplasm volume and quality. The alterations were noted of the treated fetal livers compared to controls.

cDNA Synthesis

Four control and four TCDD- or DES-treated dams were used for gene expression experiments. A minimum of 2 livers from each litter was used for RNA isolation. RNA was extracted from the fetal livers using Trizol (Gibco, Rockville, MD) according to manufacturer protocol, and then dissolved in DNAse- and RNAse-free distilled water (Gibco). Reverse transcription of the RNA was performed using 2 μg of total RNA and the Promega Reverse Transcription System (Madison, WI) according to the manufacturer protocol.

Quantitative PCR

A 100 ng RNA-equivalent was used for each reaction in quantitative PCR. TaqMan Universal PCR Master Mix (Applied Biosystems, Foster City, CA) and pre-developed primer/probes (Qiagen, Valencia, CA) specific to c-jun, bcl-2, p53, and PKCα were used according to manufacturer protocol with the ABI 7700 (Applied Biosystems) to quantitate the expression of genes relative to controls. 18s rRNA (Applied Biosystems) was used as the internal reference.

Statistics

The 2-tailed Student’s t-test was applied to weight data, litter size data, and numerically evaluated light microscopic observations (scored), considering p < 0.05 as significant. Data means described as different in this report were significantly different. Some observations, such as alterations of cellular morphology, were not enumerated. Such findings were assessed as being present or absent and thus described.

Quantitative PCR results were analyzed by the Virginia Tech Statistical Consulting Center using the Dunnett’s t-test to compare the treatment groups to the control group. The cycle threshold (CT) method described by Livak and Schmittgen (2001) was employed for evaluating gene expression results, with target expression first normalized to 18s rRNA within each sample (ΔCT), then compared between TCDD-, DES-, and vehicle-treated groups (ΔΔCT).

Results

Fetal Number and Weights

No significant changes were seen in the fetal number per litter between the control and treatment groups (Table 1) No differences in pooled fetal weights or individual fetal weights were present after TCDD treatment. The pooled fetal liver and individual fetal liver weights were significantly increased in the TCDD-treated fetuses. In the DES-treated group, the pooled fetal weights and individual fetal weights were significantly decreased relative to corn oil controls. However, the pooled and individual fetal liver weights were not significantly different in the DES treatment group relative to controls. Ratios of pooled liver weights to pooled fetal weights, expressed in percentages, were increased with both treatment groups relative to the controls.

Histopathology

In corn oil-treated controls, the fetal liver displayed clearly visible central veins but portal triads were inconspicuous as were sinusoidal spaces and individual plates of hepatocytes. A rare bile duct, suggesting the presence of a triad region, lined by plump cuboidal epithelial cells having oval centrally located nuclei and prominent single nucleolus, was also present (Figure 1A). Hepatocytes were uniform in appearance, and contained discrete large intracytoplasmic vacuoles displacing their oval vesicular nuclei to the cell margins. A small amount of amorphous granular eosinophilic intracytoplamsic material was noted. Multifocal and diffuse extramedullary hematopoiesis consisting of erythroid and myeloid precursors was present, with occasional megakaryocytes scattered throughout the hepatic parenchyma. Within these foci of hematopoiesis were a population of small mononuclear cells with deeply staining nuclei, a compact chromatin pattern, and a scant rim of basophilic cytoplasm. These cells were compatible with hematopoietic stem cells, hematopoietic progenitor cells, or small lymphoid cells. Low numbers of predominantly mature red blood cells and rare metarubricytes were present in the sinusoidal spaces and within central veins.

TCDD-treated fetal livers showed a decrease in hepatocyte vacuolation and a concomitant increase in eosinophilic cytoplasmic staining (Figure 1B). Hepatocytes in TCDD-treated livers displayed mild to moderate anisocytosis and anisokaryosis, with increased nuclear and cytoplasmic size, increased prominence of nucleoli, and increased cytoplasmic basophilia. Megakaryocytes did not appear to be altered in number or morphology by treatment. The relative presence of small mononuclear cells was diminished (Table 2), suggesting decreased hematopoiesis following TCDD treatment.

DES-treated fetal livers resembled the TCDD-treated livers in that hepatocellular vacuolation was diminished, eosinophilic cytoplasmic staining was more prominent, anisocytosis and anisokaryosis of the hepatocytes were increased over controls, and small mononuclear cells were relatively diminished; however these effects were to a lesser degree following DES treatment, as compared to TCDD livers (Figure 1C).

Quantitative PCR Analysis

Fetal livers from dams treated with TCDD demonstrated increased expression of c-jun and decreased expression of p53 relative to fetal livers from corn oil-treated pregnant dams (Figure 2a). TCDD treatment also resulted in a modest decrease in PKCα expression relative to controls. Fetal livers from dams treated with DES demonstrated modest but significant decreases in the expression of bcl-2, PKCα, and c-jun relative to corn-oil treated dams (Figure 2b). DES had little effect on the expression of p53.

Discussion

Prior studies have examined the myelotoxicity of a number of chemicals, including TCDD and DES. Sakai et al. (2003) reported that treatment with TCDD diminished the long-term reconstituting activity of hematopoietic stem cells in C57BL/6 mice, even though the numbers of these CD34⁻, c-kit⁺, Sca-1⁺ cells increased after a single oral administration of TCDD. In vitro, TCDD and DES diminished total cellularity as well as pluripotent stem cells and granulocyte-macrophage progenitors in clonal bone marrow suspensions (Boorman et al., 1982). These reports suggest a similar pattern of hematopoietic progenitor cell targeting by TCDD and DES.

The present pregnant mice were dosed with TCDD or DES at levels that cause approximately 50% reduction in term fetal thymus weights (Besteman et al., 2005a, 2005b), and alterations in fetal liver that could be detected via light microscopy were examined. Grossly, neither the TCDD nor DES livers differed from controls except for their increase in size relative to the fetus. This previously not reported finding of relative fetal hepatic enlargement following either chemical exposure appears consistent with reports of hepatic hyperplasia, hypertrophy, and hepatomegaly in adult mice (Birmbaum et al., 1990; Chauhan et al., 1991; Reilly et al., 1991; Sanchez et al., 2003). The fetal hepatic enlargement also suggests hepatocyte enzyme induction and metabolic response capacity are at least partially established in day 18 fetal mouse for both TCDD and DES. This and collective fetal hepatocyte effects noted in the present experimental mice, including decreased vacuolization and increased cytoplasmic staining, raise questions about potential contributions to myelotoxicity as a consequence of altered hepatocellular support of hematopoiesis, beyond direct targeting of progenitor cells. Interestingly, no signs of inflammation were found within fetal livers after TCDD or DES, nor was there any indication of necrosis.

The most noteworthy change in hematopoiesis detected microscopically was the relative diminution of small mononuclear cells following TCDD or DES exposure. This visual evidence of cell loss may correlate with previous flow cytometric studies that found reduced percentages of TdT+ cells (T-lymphocyte precursors) in fetal mouse liver after TCDD or DES (Silverstone et al., 1992; Holladay et al., 1993). The architectural organization of the liver was also affected by the overall reduced hematopoiesis, as the foci of stem and progenitor cells were of lower cellularity and were more widely scattered. Megakaryopoiesis appeared to be unaffected by either agent.

Preliminary gene expression analysis was conducted to explore possible shared mechanisms by which TCDD or DES may impact fetal liver and hematopoiesis, with bcl-2, p53, c-jun, and PKCα selected for study. Relative expression of p53 and bcl-2 is important for regulating both proliferation and cell death, especially in development (Sharova et al., 2000). Orelio et al. (2004) found that bcl-2 overexpression resulted in increased hematopoietic stem cell activity. Likewise, p53 plays a part in normal hematopoietic development as its deletion or mutation is often associated with myelodysplastic disease (Lai et al., 1995; Kurotaki et al., 2000) and inhibition of terminal differentiation of myeloid progenitors (Soddu et al., 1996).

The c-jun proto-oncogene and transcription factor is essential for normal murine embryonic survival and development (Hilberg et al., 1993; Hart et al., 2003). Further, tightly regulated c-jun expression is necessary for normal hepatic development (Eferl et al., 1999; Behrens et al., 2002) and has a role in hematopoietic differentiation, proliferation and apoptosis (Szabo et al., 1991; Mouthon et al., 1992; Shimizu et al., 1996; Liebermann et al., 1998). PKC α signaling plays a part in differentiation of hematopoietic progenitor cells (Rossi et al., 1996; Pierce et al., 1998; Darley et al., 2002; Myklebust et al., 2002). The expression of PKCα increases to near adult levels in the fetal rodent liver as parturition approaches (Gruppuso, 1990).

Given the known similar effects of TCDD and DES in fetal thymus and liver, we anticipated these agents might affect similar gene targets downstream to their respective receptor binding. However, TCDD increased c-jun expression in fetal liver, while DES decreased c-jun expression. In hematopoietic cells, c-jun acts as a positive modulator of apoptosis (Liebermann et al., 1998). Previously, we found that increased apoptotic precursor T cells (thymocytes) in TCDD-treated fetal mouse thymus were readily evident via light microscopy (Besteman et al., 2005b). However, enhanced apoptosis of hematopoietic cells was not histologically observed in the present fetal livers. In this regard, bcl-2 expression was increased by TCDD while p53 was markedly decreased, suggesting a shift away from cell death and toward proliferation, and an effect that could override a pro-apoptotic signal from c-jun (Sharova et al., 2000).

In DES-treated fetal livers, p53 was unchanged while bcl-2 significantly decreased, shifting the ratio of these genes toward p53 that might signal enhanced cell death. However, increased apoptotic hematopoietic cells were again not visually evident. Decreased c-jun expression by DES, and decreased PKC α by TCDD and DES, may correlate with reduced differentiation and proliferation of hematopoietic cells. This reduction in proliferation could relate to diminished hematopoietic cell numbers independent of cell death. For all these gene expression results, it must be remembered that fetal hepatocytes and fetal hematopoietic cells were pooled for these analyses, and may show different patterns of effect on expression of the selected genes. As such, additional experiments that separate fetal livers into hepatocyte-enriched and hematopoietic cell-enriched populations may be a valuable extension of the present experiments.

In summary, previous studies have suggested a similar profile of hematopoietic effect in fetal thymus and liver after TCDD or DES. Consistent with these reports, these agents produced similar histopathological changes in the metabolic and hematopoietic compartments of the late-gestation fetal liver. However, different gene targets in fetal hepatocytes or hematopoietic cells may underlie the hematopoietic cell depletion caused by both agents.