Mono-(2-ethylhexyl) Phthalate Increases Spermatocyte Mitochondrial Peroxiredoxin 3 and Cyclooxygenase 2


1230 York Ave, New York, NY 10021 (e-mail:


ABSTRACT: Mono-(2-ethylhexyl) phthalate (MEHP), the biologically active metabolite of the plasticizer di-(2-ethylhexyl) phthalate, is a member of a class of chemical compounds with known adverse effects on the male reproductive system. Recent studies showed that oxidative stress and mitochondrial dysfunction in germ cells may contribute to phthalate-induced disruption of spermatogenesis. To determine whether the redox-protein mitochondrial thioredoxin-dependent peroxidase, peroxiredoxin 3 (Prx3), may be a component of germ cell homeostasis mechanisms, this study first examined the physiologic relevance of Prx3 in the rodent testis by determining its cell-specific expression. Our findings show that prx3 mRNA is expressed in a developmental, cell-specific manner in rat Leydig cells, Sertoli cells, and germ cells; among mouse germ cells, prx3 expression was highest in spermatocytes, findings consistent with those in rat. In mouse meiotic spermatocytes, Prx3 was strikingly localized at the nuclear perimeter and cytoplasm, findings suggestive of a direct role for Prx3 in determining spermatocyte response to toxicants. To better define the mechanisms involved in male germ cell dysfunction following phthalate exposure, an immortalized mouse spermatocyte-derived germ cell line, GC-2spd(ts), was exposed to MEHP (24 hours; 100 and 200 μM). We determined whether Prx3 and cyclooxygenase-2 (COX-2), pivotal proteins involved in oxidative stress responses in spatially restricted subcellular domains, were affected. Mitochondrial Prx3 and mitochondrial and cytosolic COX-2 significantly increased following 200 μM MEHP treatment; proliferation was inhibited without inducing cell death. Using this germ cell model, the data suggest that changes in cellular oxidation-reduction (redox) homeostasis in the germline can accompany MEHP exposure, disrupting mitochondrial antioxidant defenses, despite absence of phthalate-induced apoptosis.

Consumer and industrial demand for polyvinyl chloride (PVC)–containing products has resulted in increased commercial synthesis of plasticizers, such as the commonly used phthalate, di-(2-ethylhexyl) phthalate (DEHP). Exposure to phthalates is ubiquitous—for example, in food containers, toys, baby bottles, intravenous tubing, blood storage bags, PVC flooring, and household dust (Graham, 1973; Rock et al, 1986; Bornehag et al, 2004). Phthalates can leach out of products, increasing human exposure levels of phthalates (Rock et al, 1986). Raising further health concerns, phthalates have been found in human breast milk (Main et al, 2006).

Phthalates are endocrine disruptors and peroxisome proliferators. Mono-(2-ethylhexyl) phthalate (MEHP), the biologically most active metabolite of DEHP, has deleterious effects on the male reproductive system, especially in neonatal and prepubertal males (Oishi, 1990; Li et al, 2000). Investigations over the last 3 decades indicate that Sertoli cells (SCs) and Leydig cells (LCs) represent the primary direct testicular targets for MEHP (Dostal et al, 1988; Heindel and Chapin, 1989; Thysen et al, 1990; Akingbemi et al, 2001; Mylchreest et al, 2002; Foster, 2005; Mahood et al, 2005; Ge et al, 2007) However, MEHP induces dramatic changes in germ cells, which until recently were thought to be mediated indirectly by somatic cell effects; these untoward effects include oxidative stress, mitochondrial dysfunction, cytochrome c release, and apoptosis (Richburg and Boekelheide, 1996; Kasahara et al, 2002). Induction of oxidative stress may represent a common mechanism in endocrine disruptor—mediated dysfunction, specific to certain testicular cells (Latchoumycandane et al, 2002). Recent evidence suggests that mitochondria are targets of phthalates. After administration of DEHP to rats, mitochondria isolated 6 hours later from the testis show reduced respiratory function (Oishi, 1990). In primary rat SCs treated with MEHP (24 hours), an increase in glycolysis, reduction of ATP levels, and a decrease in succinate dehydrogenase activity are observed (Chapin et al, 1988). Moreover, after in vivo or in vitro treatment with DEHP or MEHP, respectively, mitochondrial swelling in LCs has been observed (Jones et al, 1993). Collectively, these findings suggest that alterations in mitochondrial structure and function are cellular signatures of phthalate-induced testicular toxicity.

Given that mitochondria represent an intracellular target for MEHP and that increases in testicular reactive oxygen species (ROS) and oxidative stress follow MEHP exposure, we hypothesized that MEHP would affect levels of mitochondrial proteins involved in regulating cellular oxidation-reduction (redox) homeostasis.

Among oxidative stress—related genes altered in the testes of male rat fetuses exposed to phthalates, peroxiredoxin3 (Prx3) is reduced, as assessed using DNA microarray analysis (Liu et al, 2005). Increases in Prx3 protein are cytoprotective, maintaining mitochondrial integrity (Shibata et al, 2003; Matsushima et al, 2006); a significant reduction in Prx3 can potentially sensitize a cell to an apoptotic stimulus (Chang et al, 2004). Therefore, Prx3 may represent one target of MEHP-mediated oxidative stress in the male germline.

Induction of cyclooxygenase-2 (COX-2), arachidonic acid utilization, and prostaglandin production are critical regulators in response to cellular redox status and extracellular proapoptotic conditions (Feng et al, 1995; Jiang et al, 2004). Pertinent to this study in a germline model and further attesting to the overall biologic significance of this mechanism, a recent study shows that inhibition of COX-2 augments hydrogen peroxide—induced apoptosis in mouse embryonic stem cells (Liou et al, 2007).

Therefore, we investigated the effect(s) of short-term MEHP exposure on cellular Prx3 and COX-2 levels in male germ cells using the SV-40 immortalized mouse spermatocyte-derived cell line model, GC-2spd(ts) (Hofmann et al, 1994; Wolkowicz et al, 1996). Our findings indicate that MEHP disrupts spermatocyte cellular redox mechanisms.

Materials and Methods

Tissue Processing, Cell Isolation, and Purification

Animals were housed in standard lighting (12 h light, 12 h dark) with food and water allowed ad libitum in facilities approved by the American Association for the Accreditation of Laboratory Animal Care. Procedures involving the use of animals strictly followed the Guidelines for Care and Use of Laboratory Animals set forth by the National Institutes of Health, and protocols received Institutional Animal Care and Use Committee approval.

For reverse transcriptase—polymerase chain reaction (RT-PCR) and real-time quantitative RT-PCR (Q-PCR) analyses, rat SCs were prepared as in Jenab and Morris, 1998; and Morris et al, 1988; LCs were prepared as in Kanzaki and Morris, 1998, 1999; and germ cells were prepared as in Jenab and Morris, 1998. Total RNA was extracted from isolated testicular cells in TRIzol reagent (Invitrogen Life Technologies Inc, Grand Island, New York), according to the manufacturer's instructions. RNA concentration was measured using a Lambda35 UV/Vis Spectrophotometer (Perkin-Elmer Corp, Norwalk, Connecticut).

For bioimaging studies, whole testes were obtained from 7-week-old and 1-year-old male C57BL/6 mice, and cells were isolated and prepared as previously described (Page et al, 1998). Briefly, dissected testicles were placed in a freshly prepared solution of 2% formaldehyde in 1× calcium- and magnesium-free Dulbecco phosphate-buffered saline, pH 7.1 (PBS) with 0.05% Triton-X for 5–10 minutes at room temperature. Seminiferous tubules were liberated by removal of the tunica albuginea and dispersion into fixative. Tissue (3–5 mm) was placed in a small droplet of fixative on a glass microscope slide rinsed in 100% ethanol with 1 drop of concentrated hydrochloric acid. Tubules were minced to release cells, covered with a 22×22-mm cover slip, tapped gently with a pencil eraser to squash cells, and immersed into liquid nitrogen until the bubbling stopped. After freezing, the cover slip was removed, and the slide was immediately placed in PBS for three 5-minute washes and allowed to air dry. Dry slides were immediately used for immunolabeling or stored at −80°C.

RT-PCR Analysis

Total RNA (4 μg) was reverse transcribed for 15 minutes at 42°C in a mixture containing 5 mM MgCl2, 1× PCR buffer II, 4 mM each of deoxy-NTP, 1 U/μL ribonuclease inhibitor, and 2.5 mM random hexamers. Samples were denatured at 99°C for 5 minutes. PCR was performed using 2 μL of each RT product as a template. The following primers were used: Prx3 forward primer, 5′-GCTGAGTCTCGACGACTTTAAGGG-3′; Prx3 reverse primer, 5′-CTTGATCGT AGGGGACTCTGGTGT-3′; S16 ribosomal gene forward primer, 5′-TCCGCTGCAGTCCGTTCAAGTCTT-3′; and S16 ribosomal gene reverse primer, 5′-GCCAAACTTCTTGGATTCGCAGCG-3′. AmpliTaq DNA polymerase (Applied Biosystems, Foster City, California) was used. The PCR mixture (25 μL) contained 2 mM MgCl2, 1× PCR buffer II, and each primer at 0.2 μM. Amplification was performed in a programmable thermal controller (model PTC-100; MJ Research Inc, Watertown, Massachusetts). The samples were first denatured at 95°C for 2 minutes, followed by 30 PCR cycles; the temperature profile was 95°C (30 seconds), 60°C (30 seconds), and 72°C (1.5 minutes). After the last cycle, additional extension incubation at 72°C (7 minutes) was performed. After amplification, PCR products were separated on a NuPAGE-Novex polyacrylamide gel (4%–20% Tris/boric acid/EDTA; Invitrogen). The bands were visualized by ultraviolet fluorescence after staining with ethidium bromide (1 μg/mL) for 15 minutes, and the image was recorded using a computer-assisted camera (Eastman Kodak Co, Rochester, New York).


Using a standard curve method of analysis, fluorescence-monitored Q-PCR assays were conducted to quantitatively determine the levels of Prx3 mRNA in each sample. For each experimental sample, 18S ribosomal RNA was used as an endogenous control to normalize the Prx3 data. Reactions (25 μL total volume) were set up in quadruplicate in optical quality 96-well reaction plates containing 1× Power SYBR Green PCR Master Mix (Applied Biosystems) and 200 nM primers for Prx3, 1× Q-PCR Master Mix Plus (Eurogentec, Philadelphia, Pennsylvania), and VIC dye-labeled probe (Applied Biosystems) for 18S ribosomal RNA. The experimental cDNA sample, 2 μL of a sixfold dilution of the RT product, was subsequently added to each well. Q-PCR was performed using an Applied Biosystems model 7700 Sequence Detection System. The temperature profile for the reactions was 50°C (2 minutes), 95°C (10 minutes), and 40 cycles of 95°C (15 seconds) and 60°C (1 minute). Using the manufacturer's software, a threshold above noise was chosen, and the cycle number (CT) at which fluorescence exceeded the threshold was determined. For each real-time RT-PCR assay, a standard curve was generated using cDNA corresponding to the control and based on 7 twofold serial dilutions (“control” cDNA in water). The mean CT value for each cDNA sample was expressed as an arbitrary value relative to the standard curve after linear regression analysis. Experimental samples were diluted sixfold for comparison with the standard curve. A no-template control, in duplicate, was performed for each reaction. Prx3 data were normalized using the corresponding 18S values, and results are expressed as arbitrary units relative to control, set as a value of 1.

Bioimaging and Immunofluorescence

Mouse testes from young (7-week-old) and older adult (1-year-old) mice were used to obtain “squash preparations” of testicular cells. For bioimaging and immunofluorescence, the previously air-dried slides were placed in a 0.04% PhotoFlo solution (Eastman Kodak Co) for 2 minutes, drained, and allowed to air dry. Once-dried slides were hydrated in 1× antibody dilution buffer (ADB) containing 1% normal donkey serum (Sigma-Aldrich Corp, St Louis, Missouri), 0.3% BSA, and 0.005% Triton-X (in PBS) for 30 minutes at room temperature. Slides were incubated overnight with 60 mL of the primary antibody cocktail, which consisted of Synaptonemal Complex Protein 3 (SCP3; 1:100; Novus, Littleton, Colorado), a standard marker of early meiotic cells, and Prx3 (Santa Cruz Biotechnology, Santa Cruz, California) (1:100) in 1× ADB at 37°C in a humid chamber. Following the primary antibody incubation, slides were washed once in 1× ADB for 20 minutes, then a second wash in 1× ABD at 4°C for at least 5 hours. Slides were then incubated for 1 hour with 60 mL of the secondary antibody cocktail (fluorescein donkey anti-rabbit and rhodamine donkey anti-goat [1:100 in 1× ADB; Jackson ImmunoResearch, West Grove, Pennsylvania]) at 37°C in a humid chamber. Slides were washed 3 times in 1× PBS for 10 minutes each and counterstained with 4,6-diamidino-2-phenylindole dihydrochloride (DAPI) in Antifade (BioRad Laboratories, Hercules, California), and a cover slip was applied. Cells were located using a wide-field fluorescence microscope (Zeiss, Oberkochen, Germany), and images were collected and analyzed with the MetaVue acquisition software (Universal Imaging, Downingtown, Pennsylvania) using a Hamamatsu Orca ER B/W digital camera.


MEHP (TCI America, Portland, Oregon) was dissolved in DMSO to a concentration of 400 mM and stored in single-use aliquots (−20°C).

Cell Culture and MEHP Treatment

The GC-2spd(ts) cell line, herein abbreviated as GC-2, was purchased from the American Type Culture Collection (Manassas, Virginia). This cell line expresses a temperature-sensitive mutant of the mouse p53 tumor suppressor gene, encoding an Ala135 to Val135 change that affects protein folding. Cells were cultured at 37°C, at which they express both wild-type and mutant p53 (Hofmann et al, 1994; Wolkowicz et al, 1996), and were maintained in log-phase growth in flasks (75 cm2; BD Falcon, Bedford, Massachusetts) at 5% CO2 with Dulbecco modified Eagle medium (Invitrogen) containing 10% FBS (JRH Biosciences, Lenexa, Kansas) and 4 μg/mL Gentamycin (Invitrogen).

Cells were seeded in replicate culture plates (as indicated below) 24 hours prior to dosing. MEHP was added to a final concentration of either 100 or 200 μM MEHP. Cells were treated with either 0.05% or 0.1% DMSO as the matched vehicle control, as indicated. Following treatment, cells were maintained for 24 hours until harvested.

Cell Proliferation

Cells were seeded at a density of 2 × 105 cells/well in 6-multiwell plates (Becton Dickinson, Franklin Lakes, New Jersey), and harvested using trypsin (0.25%; Invitrogen). Cell number was determined using a hemocytometer, and viability was assessed by trypan blue exclusion. For cell cycle studies, cells were harvested by trypsinization, rinsed, an aliquot counted, and the remaining cells fixed in 70% ethanol, then stored at −20°C until use. On the day of the flow cytometric assay, cells were washed twice (PBS, 1% FBS), labeled with Isotype or Ki67 FITC (BD Pharmingen, San Diego, California), and incubated for 30 minutes at room temperature in the dark. Cells then were washed once, samples were aspirated, and cells were resuspended in PI Master Mix (propidium iodide, RNAase, and PBS 1% Triton-X). The cells were incubated in the dark (30 minutes, room temperature) and then analyzed using a FACS Calibur (BD Biosciences, San Jose, California).

Preparation of Whole-Cell Lysates

Cells were seeded at a density of 1 × 106 in 75-cm2 flasks, rinsed in ice-cold PBS, and harvested with a 25-cm cell scraper (Sarstedt Inc, Newton, North Carolina). Whole-cell lysates were obtained (Ishikawa et al, 2005). Protein concentration was determined with BSA as a standard using the Bio-Rad protein assay reagent (Bio-Rad Laboratories). Absorbance was read at 590 nm in a Lambda35 UV/Vis Spectrophotometer (Perkin-Elmer Corp).

Preparation of Mitochondrial-Enriched and Cytosolic Fractions

Cells were seeded at a density of 3 × 106 in 150-cm2 flasks (Corning Costar Corp, Cambridge, Massachusetts). After rinsing with ice-cold PBS, cells were harvested using a 38-cm cell scraper (Sarstedt) with PBS (5 mL) twice, pooled, pelleted by centrifugation at 650 × g (5 minutes, room temperature), suspended in 1 mL PBS, and centrifuged at 960 × g (5 minutes; 4°C; microcentrifuge [model 5417R; Eppendorf, Westbury, New York]). The wash was repeated and pellets resuspended in 1 mL of sucrose mitochondrial isolation buffer (SMIB; 0.25 M sucrose, 20 mM HEPES, 10 mM KCl, 1.5 mM NaCl, 1 mM EDTA, 1 mM EGTA, pH 7.5, containing 1 mM dithiothreitol, 1 mM sodium orthovanadate, 0.1 mM phenylmethylsulfonylfluoride, 2 μg/mL aprotinin, 2 μg/mL pepstatin, and 2 μg/mL leupeptin) and incubated on ice (10 minutes). Cells were lysed by 25 up-and-down passes through a 27.5-gauge G needle fitted to a 1-mL syringe (Becton Dickinson). Homogenates were centrifuged at 1020 × g (10 minutes), supernatants centrifuged at 1020 × g (10 minutes) to remove unbroken cells and nuclei, and then centrifuged at 15 000 × g (10 minutes) to pellet mitochondria. The postmitochondrial supernatant was taken as the “cytosolic” fraction. The mitochondrial-enriched pellet was suspended in 200 μL of SMIB and centrifuged at 12 200 × g (10 minutes); this wash was repeated once. The final mitochondrial-enriched pellet was suspended in SMIB, and samples were stored at −80°C. Protein concentration was determined using Bio-Rad protein assay reagent; absorbance was read at 590 nm using an MRX plate reader with Revelation software, version 3.2 (DYNEX Technologies Inc, Chantilly, Virginia).

Western Blot Analysis

Whole-cell lysates (15 μg), mitochondrial-enriched extracts (15 μg), or cytosolic extracts (15 μg) were subjected under reducing conditions to sodium dodecyl sulfate—polyacrylamide gel electrophoresis using 10% Tris/Bis NuPAGE gels (Invitrogen) and were transferred using the Bis/Tris transfer system (Invitrogen) to a 0.45-μm nitrocellulose membrane (Schleicher & Schuell, Keene, New Hampshire). Membranes were hybridized sequentially with rabbit anti-mouse Prx3 antiserum (1:2000; a gift from Dr Chi V. Dang, Johns Hopkins University, Baltimore, Maryland), rabbit polyclonal anti—COX-2 antibody (1:3000; Cayman Chemical Co, Ann Arbor, Michigan), mouse monoclonal anti—voltage-dependent anion channel (VDAC) antibody (1:1000; clone 89–173/016; EMD Biosciences, San Diego, California), and mouse monoclonal anti–β-actin antibody (1:12 000; Sigma-Aldrich). Blots were developed with the ECL Western blotting system (Amersham Biosciences, Arlington Heights, Illinois) and exposed to X-ray film (Eastman Kodak). Densitometric analysis was performed using NIH ImageJ, version 1.33u ( Signals were normalized to VDAC (mitochondrial-enriched) or β-actin (whole-cell or cytosolic) protein lysates.

Statistical Analysis

Densitometric analyses are expressed in arbitrary units. All results are the mean ± SEM derived from the number of individual different experiments (3–7), unless noted. For cell line studies, the vehicle “control” was used for normalization to a value of “one” for each experiment. For cell-specific expression studies, data were normalized as indicated in the figure legend. Statistical analyses to determine significant differences used 1-way ANOVA analysis with Dunnett multiple comparison test. A P value ≤ .05 was considered significant.


Expression of Prx3 in the Rodent Testis

Although it is well established that the testicular somatic cells, LCs and SCs, are primary targets of phthalates, recent studies have shown that germ cells undergo dramatic changes in response to MEHP. Kasahara et al (2002) showed that MEHP induces oxidative stress and mitochondrial dysfunction in the rat testis; isolated germ cells, but not SCs, undergo oxidative stress and apoptosis after exposure to MEHP. Therefore, we hypothesized that a mitochondrial redox protein, Prx3, could be a potential molecular target for MEHP-induced oxidative stress in the testis, including the germ cells.

RT-PCR analyses were performed to determine the physiologic expression of Prx3 in the rodent testis. Since rat and mouse models are used primarily to investigate the effects of phthalates on male reproductive biology, primers that recognize both rat and mouse Prx3 mRNA transcripts in freshly isolated testicular cells were used in this study. Our studies indicate that prx3 mRNA is expressed in spermatogonia, pachytene spermatocytes, and round spermatids of both rat and mouse testis (Figure 1). Interestingly, pachytene spermatocytes from both rat and mouse testis show the highest expression levels of prx3 mRNA comparatively among these germ cells (Figure 1A).

Figure 1.

. Prx3 mRNA expression in rat and mouse testicular germ cells. Total RNA was extracted from freshly isolated rat (left) and mouse (right) germ cells. RT-PCR analyses were performed using primers that recognize both rat and mouse prx3 mRNA. (A) Densitometry data are expressed in arbitrary units relative to the whole testis, set as 1. Error bars represent standard deviation. (B) A representative image of the PCR products. T indicates whole testis; Spg, spermatogonia; PS, pachytene spermatocytes; RSd, round spermatids; -RT, no reverse transcriptase (negative control); S16, S16 ribosomal RNA used for normalization of the individual sample PCR products.

Many previous studies to determine phthalate effects on the testis have focused on those affecting LCs and SCs; therefore, we investigated prx3 mRNA expression in these somatic cells as well. Highly purified, macrophage-free populations of SCs and LCs isolated from rat testis were analyzed for prx3 expression. Each of these testicular somatic cells express prx3 mRNA (Figure 2A, B, and C). Interestingly, there were somatic cell—specific, age-related differences in prx3 expression. To better evaluate these differences, Q-PCR analyses were employed using the immature SC prx3 mRNA: 18S ribosomal RNA level, set as a unit of “1,” for comparison (Figure 2B and C). In the somatic cells of the rodent testes examined, the adult LC shows the greatest abundance of prx3 mRNA, a 40-fold higher prx3 level compared with that of immature SCs (Figure 2C). Strikingly, the adult LC level of prx3 mRNA is 25-fold higher than that of the progenitor LC, which is in turn 1.6-fold greater than that of the immature SC (Figure 2B and C). Adult SCs had a 20% higher level of prx3 mRNA than immature SCs. Relative maturational levels of prx3 were in each case significantly higher for Leydig than Sertoli, and in the fully differentiated, comparable cell of the adult.

Figure 2.

. Developmental prx3 mRNA expression in rat Leydig and Sertoli cells. (A) RT-PCR and (B, C) Q-PCR analyses were performed using total RNA extracted from freshly isolated and highly purified rat Sertoli (immature, 95%, and mature, 90% purity) and Leydig cells (adult, ≥98% and progenitor, 90%-93% purity) to determine prx3 mRNA expression. (A) A representative gel image of the RT-PCR cDNA products. (B, C) Q-PCR values for prx3 mRNA and 18S are expressed as a ratio, normalized as arbitrary units relative to the immature Sertoli cell level, which is set as “1”. All samples (n = 4) were analyzed simultaneously in the same prx3 mRNA and 18S assays. Error bars represent SEM, *p < .05; **p < .01 indicate significance differences of the means from that of the immature SCs using ANOVA with Dunnett multiple comparison, n = 4. Note the difference in scales, Panel B vs Panel C.

Pachytene spermatocytes show the highest levels of prx3 among the male germ cells examined. Therefore, we next determined Prx3 in mouse meiotic spermatocytes. For the immunofluorescent detection of Prx3, we used cells obtained from squash preparations of whole testes obtained from 7-week-old and 1-year-old adult mice (Figure 3). Pachytene spermatocytes were identified using an antibody against SCP3, the linear structures representing the meiosis-specific synaptonemal complexes that tether duplicated homologous chromosomes for recombination (Figure 3A and B). As illustrated, pachytene spermatocytes were abundant in the mouse testes of both young adult and older males (Figure 3A, inset). Prx3 was strikingly localized at the perimeter of the nucleus and in the cytoplasm of the meiotic germ cells. These bioimaging observations were also confirmed using a second antibody against Prx3 (data not shown).

Figure 3.

. Prx3 is located in the perinuclear and cytoplasm of male mouse pachytene cells. Meiotic spermatocytes isolated from young adult mouse testes (7 weeks, postnatal) were prepared using a squash technique to maintain the three-dimensional structure of the cells. Isolated spermatocytes were immunostained for Prx3 (FITC, green) and SCP3 (Texas Red, red); cell nuclei were counterstained with DAPI (blue); Panel A shows the merged image. Illustrated here are meiotic cells at various stages of pachytene, as demonstrated by the SCP3 (red) linear structures. (A, B) Prx3 is localized with the nuclear membrane, as indicated by strong signals immediately adjacent to the nucleus and abundant staining in the cytoplasm but not within the nucleus itself (A-C); see scattered punctate signals over the nucleus (asterisk in C) and DAPI counterstaining. Similar localization patterns were observed in older adult male testes (1 year, postnatal; A, inset). Bioimages were captured at ×40 magnification (A-C); inset (A) ×100.

MEHP Inhibits GC-2 Cell Proliferation in the Absence of Cell Death or Cell Cycle Changes

To date, the effects of phthalates on male reproductive biology are largely thought to be due to alterations in SC and LC signaling, gene expression, and cellular function. In this paradigm, MEHP exposure subsequently and indirectly affects male germ cells through such changes in supporting somatic cell function. However, more recent studies bring into question whether there are direct effects on developing germ cells, effects that may not require initial changes in somatic cell function. For example, Ono et al (2004) showed that DEHP metabolites were found in germ cells of the rat testis 6 hours and 24 hours following DEHP treatment. As such, male germ cells represent a novel target for DEHP metabolites, such as MEHP. MEHP may induce testicular dysfunction by directly acting on germ cells in addition to inducing alternations in LC and SC function. Therefore, we used a mouse spermatocyte-derived cell line to investigate whether the mode of action of MEHP in inducing testicular dystrophy would involve a direct effect on the male germ cell.

To determine whether short-term MEHP exposure directly affects proliferation, GC-2 cells were treated with MEHP for 24 hours. Compared with matched vehicle-treated control cells, those exposed to 100 or 200 μM MEHP were significantly reduced in numbers by 20% and 40%, respectively (Figure 4). Cell viability, as assessed by trypan blue exclusion, was unaffected by MEHP treatment (Figure 4, inset), indicating no induction of cell death. No change in adherence or anoikis (death by detachment) was noted. Cell viability data are consistent with our cell cycle studies indicating that there were no changes in cell cycle distributions (data not shown). Our findings indicate that MEHP (doses up to 200 μM) did not induce cell death or cell cycle arrest in GC-2 cells.

Figure 4.

. Effect of MEHP exposure on GC-2 cell proliferation and viability. Cells treated with 0.1% DMSO (vehicle control) or 100 or 200 μM MEHP (24 hours) were harvested and counted, and viability was assessed as indicated. Data are expressed in arbitrary units relative to vehicle control, set as 1. Error bars represent SEM, *P < .01, indicates significance differences from vehicle by ANOVA with Dunnett multiple comparison, n = 4. Inset: viability assessed by trypan blue exclusion. Open bar indicates unstained cells (% total cells); closed bar, trypan-stained cells (% total cells). The 100 and 200 indicate MEHP (μM).

MEHP Increases Mitochondrial Prx3 in GC-2 Cells

Since pachytene spermatocytes express both Prx3 mRNA and protein, we focused on spermatocyte Prx3 as a novel target for the direct action of phthalates on germ cells. We investigated the potential role of Prx3 in affording spermatocytes protection from environmental stressors by using a Prx3-positive mouse spermatocyte-derived cell line to further delineate the mechanistic role of peroxiredoxins in male germ cells.

To determine the mechanisms involved in the response of GC-2 cells to MEHP, total and mitochondrial protein levels of the Prx3 antioxidant protein were evaluated. MEHP treatment (24 hours) did not alter total Prx3 protein levels (Figure 5A). Next, the mitochondrial-enriched fraction was isolated from MEHP-treated GC-2 cells. MEHP (200 μM) significantly increased mitochondrial Prx3 levels (∼50%; Figure 5B), whereas increases at 100 μM MEHP were not significant (∼40%; Figure 5B). These data show that mitochondrial Prx3 is increased in response to MEHP.

Figure 5.

. MEHP exposure and cellular and mitochondrial Prx3 levels. (A) Whole-cell lysates (15 μg) or (B) mitochondria-enriched lysates (15 μg) were isolated from cells treated with 0.05% DMSO (vehicle) or 100 or 200 μM MEHP (24 hours). Sequential Western blot analyses; signals normalized with (A) β-actin and (B) VDAC. Average vehicle values were (A) 0.868 (B) 0.634. Data are expressed as arbitrary units relative to vehicle, set as 1. Error bars indicate SEM, * P < .05 indicates significance differences from vehicle (DMSO) by ANOVA with Dunnett multiple comparison, (A) n = 3 and (B) n = 7. Representative Western blot is shown beneath corresponding bar graph.

Both MEHP (Ledwith et al, 1997) and oxidative stress (Feng et al, 1995; Li et al, 2002) have been reported to regulate COX-2 levels. Therefore, we next investigated whether MEHP induces COX-2 in GC-2 spermatocytes.

MEHP Induces COX-2 Expression in GC-2 Cells

In situ hybridization studies to determine cellular expression in the testis of normal adult rats showed basal levels of COX-2 mRNA in spermatogonia, spermatocytes, and SCs (Winnall et al, 2007). Previous studies from our laboratory show low, basal expression of COX-2 in rat LCs and SCs (Walch and Morris, 2002; Ishikawa et al, 2005). The current study shows that COX-2 is constitutively expressed in the SV-40-transformed GC-2 cell line, similar to those heightened levels shown in several immortalized cancer-derived and non—cancer-derived cell lines (Lee et al, 2002; Liou et al, 2005; Richardson et al, 2005). MEHP (100 or 200 μM; 24 hours) did not significantly further increase these steady-state cellular COX-2 levels (Figure 6A).

Figure 6.

. MEHP exposure and COX-2 induction. (A) Whole-cell lysates (15 μg), (B) mitochondrial (15 μg), or (C) cytosolic (15 μg) proteins were isolated from cells treated with 0.05% DMSO (vehicle) or 100 or 200 μM MEHP (24 hours). Sequential Western blot analyses; signals are normalized to (A, C) β-actin or (B) VDAC, respectively. For each experiment, the average vehicle values were (A) 0.862, (B) 0.868, and (C) 0.638. Data are expressed as arbitrary units relative to vehicle, set as 1. Error bars indicate SEM, * P < .01 indicates significance differences from vehicle by ANOVA with Dunnett multiple comparison. A representative Western blot is shown beneath. (A) n = 3, (B, C) n = 4 each.

Increases in constitutive mitochondrial COX-2 levels have been observed in breast, hepatic, endothelial, and colon cancer cell lines; such increases correlate with increased resistance and reduced sensitivity to proapoptotic stimuli (Liou et al, 2005). We observed no increase in cell death in GC-2 treated with either 100 or 200 μM MEHP for 24 hours (Figure 4, inset).

Because COX-2 is predominately located in the nuclear envelope as well as endoplasmic reticulum, substantive subcellular domain changes could be masked in whole-cell lysates. Therefore, cellular fractionation was performed to determine whether subcellular COX-2 levels may be affected following acute MEHP exposure. Mitochondrial COX-2 significantly increased following 200 μM MEHP treatment (∼40%; Figure 6B). Subsequently, the postmitochondrial supernatant containing lighter-weight membrane-bound organelles and cytosolic proteins was collected as the cytosolic fraction and analyzed by Western blot for COX-2. Additionally, cytosolic COX-2 levels significantly increased following 200 μM MEHP (60%; Figure 6C). However, MEHP did not affect nuclear COX-2 levels (data not shown). Increases in mitochondrial and cytosolic COX-2 levels are significant and contribute to the nonsignificant increase observed in the overall cellular COX-2 levels (Figure 6A). Taken together, the data are indicative of increased mitochondrial COX-2 levels in GC-2 cells in response to MEHP treatment.


Investigations in vivo and in vitro from multiple laboratories over the last 25 years have established that LCs and SCs are the primary targets of phthalate action in the testis (Gray and Butterworth, 1980; Foster et al, 1983, 2001; Sjoberg et al, 1986; Sharpe, 2001). Based on several studies, the major mode of action of MEHP on male germ cells is believed to be indirect (ie, through SC and LC dysfunction). However, DEHP metabolites have been found in germ cells, findings consistent with direct effects (Ono et al, 2004). Furthermore, at levels that induced apoptosis in germ cells and resulted in atrophy of the testis, DEHP treatment provoked oxidative stress, as measured by increases in ROS in subsequently isolated spermatocytes, but not SCs (Kasahara et al, 2002). Thus, the mitochondrion and redox homeostasis in male germ cells may represent a new target of phthalates, both organelle and stress-response mechanism directly and/or indirectly affected by MEHP exposure. Herein, this study investigated whether a mitochondrial redox protein, Prx3, is expressed in the rodent testis, with specific focus on developing male germ cells and the spermatocyte. Spermatogonia, pachytene spermatocytes, and round spermatids differentially express prx3. The differential expression of Prx3 between mitotic, meiotic, and postmeiotic germ cells may afford a defense against environmental stressors to some developmental stages while rendering others more susceptible to those stresses.

As both LCs and SCs have been shown to be primary targets of phthalate action, we determined prx3 expression in the somatic cells of the rat testis. Our findings of differential expression for Prx3 between LCs and SCs, as well as between their immature relative to their differentiated cell types, may have physiologic implications during development. For example, expression in a given cell may provide a protective mechanism with which to respond to such a stress, or its relative deficiency could underlie vulnerability or sensitivity to low toxicant levels at various stages of maturation, perhaps as an underlying contributory factor to the differential effects of phthalates on neonatal and adult testis.

Both analyses by RT-PCR and bioimaging show that pachytene spermatocytes express Prx3. Therefore, we focused on Prx3 in the spermatocyte as a novel target for the direct action of phthalates on the germ cell. Since exposure to MEHP (50–200 μM) and not DEHP induces oxidative stress in germ cells but not SCs in cultures of testicular cells isolated from DEHP-treated rats (Kasahara et al, 2002), the Prx3-positive, spermatocyte-derived germ cell line GC-2 was employed to model the effects of short-term MEHP exposure on cellular redox parameters in the spermatocyte. Our findings that MEHP does not induce cell death in GC-2 cells are in agreement with a recent study that showed that exposure to MEHP (200 μM; 24 hours) does not induce apoptosis in GC-2 cells (Chandrasekaran et al, 2006). However, we did observe a reduction in cell number after 24-hour exposure to both 100 and 200 μM MEHP without changes in cell cycle parameters. Similar inhibitory effects of phthalates on cell growth have been reported using other nontesticular cell lines. For example, dimethoxyethyl phthalate inhibits growth of mouse fibroblast cells (Dillingham and Autian, 1973). The present findings are consistent with a MEHP effect on cell proliferation.

Here we report that 24-hour exposure to MEHP increases mitochondrial Prx3. Oxidative stress and mitochondrial insult can increase levels of Prx3. For instance, the bovine homolog of Prx3 increases in cow aortic endothelial cells after 24-hour exposure to various oxidative stresses and 12-hour exposure to actinomycin A, a mitochondrial respiratory chain inhibitor (Araki et al, 1999). Increases in the antiapoptotic Prx3 in the GC-2 cells are consistent with the findings that 24-hour exposure to MEHP does not induce apoptosis. Our data suggest that the ability to induce Prx3 is protective, since reducing Prx3 sensitizes cells to oxidative stress and apoptosis. For example, acute myeloid leukemia cells showing heightened sensitivity to oxidative stress have an approximately threefold decrease in Prx3 expression (Oh et al, 2004). Prx3 levels are downregulated in motor neuron disease, a disease characterized by oxidative stress (Wood-Allum et al, 2006). Both acute and chronic oxidative injuries can lead to reduced Prx3 levels (Hattori et al, 2003; Wood-Allum et al, 2006). Furthermore, depleting Prx3 not only increases intracellular levels of ROS in unstimulated HeLa cells, but also increases ROS generation and subsequent apoptosis induced by staurosporine (mitochondrial-dependent pathway) and by tumor necrosis factor-α (death receptor—mediated pathway; Chang et al, 2004). Prx3 knockout mice have reduced body weight and are more susceptible to lipopolysaccharide-induced oxidative stress than their wild-type littermates (Li et al, 2007). Taken together, these data indicate that the loss of Prx3 results in diminished protective responsiveness and increased susceptibility to oxidative stressors. Given the role of Prx3 in maintaining cellular redox and mitochondrial homeostasis, as well as preventing mitochondrial-dependent apoptosis, the increases in mitochondrial Prx3 steady-state levels seen in GC-2 cells reflect changes in cellular redox homeostasis and/or inhibition of apoptosis following short-term MEHP exposure. By extension, our studies suggest that individual germ cell sensitivity to untoward effects of phthalates may be based in part on Prx3 levels.

Prx3 is not only important in maintaining both mitochondrial and cellular redox homeostasis, but also influences cell growth. In this study, increases in steady-state levels of endogenous Prx3 accompany the reduction in GC-2 cell number after short-term exposure to MEHP. In a previous study, Prx3 overexpression in mouse WEHI7.2 thymoma cells slowed cell proliferation without alteration in apoptosis relative to the vector control (Nonn et al, 2003). However, another study reported that stable transfection with Prx3 antisense DNA of the Rat1a fibroblast cell line overexpressing the transcription factor c-Myc (R1a-myc) and a human breast cancer epithelial cell line (MCF7/ADR) increased doubling times compared with vector controls (Wonsey et al, 2002). These reports demonstrate the ability of Prx3 to affect cell growth, and they suggest collectively that the effects of Prx3 on cell growth are cell type dependent. The slowed growth rate observed in GC-2 cells in response to MEHP could be, in part, a consequence of the effects of increased mitochondrial Prx3.

Given the effect of MEHP on Prx3 in GC-2 cells, we investigated whether short-term exposure to MEHP would affect steady-state levels of COX-2. Various stimuli can alter redox homeostasis and induce COX-2 (Feng et al, 1995; Kiritoshi et al, 2003). COX-2 protects mouse embryonic stem cells from oxidative stress—induced apoptosis (Liou et al, 2007). MEHP has been shown to potently induce COX-2 in an immortalized mouse hepatocyte cell line (Ledwith et al, 1997). First, we found COX-2 to be constitutively expressed in GC-2 cells, consistent with basal levels of expression in several immortalized cancer-derived and non—cancer-derived cell lines (Lee et al, 2002; Liou et al, 2005; Richardson et al, 2005), unlike primary rat SCs and human fibroblasts, which have low to nondetectable basal COX-2 levels (Ishikawa et al, 2005; Liou et al, 2005). In situ hybridization studies to determine cellular expression in the testis of normal adult rats showed basal levels of COX-2 mRNA in spermatogonia, spermatocytes, and SCs (Winnall et al, 2007). The constitutive expression of COX-2 observed in the GC-2 cell model is reminiscent of that in normal testicular physiology. Second, COX-2 is detected in the mitochondrial-enriched fractions of GC-2 cells, and the mitochondrial levels increase following 24 hours of MEHP exposure, changes that may afford resistance to MEHP-induced apoptosis. In agreement with these findings, in several cancer cell lines, COX-2 localizes to mitochondria and confers resistance to apoptosis induced by oxidative stress (Liou et al, 2005). Third, the effect of MEHP exposure overall is an increase in COX-2 in GC-2 spermatocytes.

Induction of COX-2 or stabilization of its protein may partially represent underlying mechanisms responsible for the slowed growth of GC-2 cells observed. For example, overexpression of COX-2 induces cell cycle arrest by a prostanoid-independent mechanism in various immortalized cell lines, including human umbilical endothelial vein cell—derived ECV-304, mouse fibroblast NIH3T3, African green monkey kidney fibroblast-like COS7, and human embryonic kidney HEK293, as well as in primary bovine microvascular endothelial cells (Trifan et al, 1999). Interestingly, 100 μM MEHP, a dose that potently induces COX-2 protein levels, does not significantly affect prostaglandin E2 synthesis in immortalized mouse hepatocytes (Ledwith et al, 1997). Whether COX-2 effects on GC-2 cell growth involve the activation of a specific prostaglandin cascade or subsequent peroxidase activity remains to be determined in ongoing studies.

Rat testis removed 6–24 hours after administration of DEHP (2 g/kg) showed increased ROS, as measured by superoxide and hydrogen peroxide generation. Exposure to doses up to 50–200 μM MEHP but not DEHP for 30 minutes increases ROS generation in primary rat germ cells but not SCs obtained from DEHP-treated rats (Kasahara et al, 2002). Our present study identified spermatocyte Prx3 and COX-2 as potential cellular MEHP sensors and indicates that increased steady-state levels of both Prx3 and COX-2 could result from early redox signaling events following MEHP exposure.

In summary, Prx3 is differentially expressed in mouse and rat testicular cells. Under normal physiologic conditions in the rodent testes, the expression of both stressor responders Prx3 and COX-2 in spermatocytes validates the utilization of this mouse spermatocyte cell line model to further identify the effects of MEHP on cellular redox homeostatic mechanisms. The present study identified 2 direct germ cell phthalate responses; that is, increases in 2 redox-sensitive proteins, mitochondrial Prx3 and COX-2. Short-term exposure to MEHP negatively affects cell proliferation, an effect accompanied by increases in Prx3 and COX-2, despite the absence of cell death. Further understanding of cellular redox status following transient exposure of the male germline to MEHP will facilitate our understanding of the male reproductive health risks of chronic, low level, and/or long-term exposure to phthalates in our environment.


The cell handling and molecular biology expertise provided by Lyann Mitchell and KeumSil Hwang are greatly appreciated. Flow cytometric analysis by Catherine Rapelje is acknowledged, as is the use of the Population Council's Cell Biology and Flow Cytometry Facility.


  1. Supported by R01HD039024 (P.L.M.). Fellowship support provided in part by NRSA ES013008 (T.M.O.) and the F.M. Kirby Foundation (P.W.B.).