Paraquat reduces natural killer cell activity via metallothionein induction

Abstract

Paraquat (PQ), one of the most widely used herbicides, has been used for several decades in agriculture. Some studies suggest that PQ has effects on the immune system. Moreover, previous studies have shown that PQ imparted some immunosuppressive effects. In the present study, cytotoxicity assays using splenic NK cells from mice treated for 28 days with PQ (at 0.2, 1, and 5 mg/kg) were performed to determine whether PQ altered the function of NK cells. Given that PQ was expected to induce an immunosuppressive effect, it was hypothesized that a gene involved in cellular metal ion homeostasis, metallothionein-1 (MT-1), could play an important role in this outcome. This belief was based on the fact that MT1 encodes a protein responsible for zinc ion homeostasis, and that a reduction in free zinc ion levels impairs NK cell function. The results showed that PQ treatments led to increased MT expression in several organs (liver, kidneys, testes) and in splenocytes, caused a reduction of both free zinc ions in sera and in free intracellular zinc, and reduced the expression of GATA-3, a zinc-finger transcription factor important for maturation and activity of T-cells and NK cells. These results provide a basis for a new molecular mechanism to describe potential immunosuppressive effects of PQ in vivo.

• Immune response
• metallothionein
• NK cell
• paraquat
• pesticide

Introduction

Paraquat (PQ; trade name for 1,1-dimethyl-4,4-bipyridinium dichloride) is one of the most widely used non-systemic non-selective herbicides. It is classified as moderately toxic via the oral route (EPA toxicity Class II), and as slightly toxic by the dermal route (Class III) (Lee et al., 2009). In contrast, PQ is classified as highly toxic based on its inhalation toxicity (EPA Class I). The biochemical mechanism of PQ toxicity is well known. PQ will undergo redox cycling in vivo, being reduced by an electron donor (such as NADPH), before being oxidized by an electron receptor (such as dioxygen) to yield superoxide (Bus & Gibson, 1984).

Although there are several studies concerning PQ toxicity, few studies have focused on its immunotoxic potential. Some studies have shown that PQ presents an immunotoxic effect that caused reduction in the splenic and thymic weights, as well as in spleen cell content, in exposed hosts. PQ was also shown to cause a reduction in the proliferative response to lipopolysaccharide (LPS) by lymphocytes as well as a decrease in interferon (IFN)-γ production and phagocytic activity of monocytes (Riahi et al., 2010, 2011). In contrast, little is known regarding PQ effects on natural killer (NK) cells.

NK cells, a very important component of innate immunity, serve a main function in hosts by acting to eliminate virus-infected and tumor cells. NK cells are also involved in the regulation of the immune response by producing cytokines such as IFN-γ and tumor necrosis factor (TNF)-α, that can activate other cellular components of innate and adaptive immunity (di Santo, 2006; Gayoso et al., 2011). Given the importance of NK cells in the immune system, this study was designed to determine the effects of sub-acute PQ exposure on the immune system, with a particular focus on potential effects on the cytotoxic activity of NK cells.

Zinc, an essential metal for growth and development, is important for the proper function of many enzymes, as well as transcription and replication factors (Rink & Kirchner, 2000). It is also well established that zinc is necessary for normal immune system function; for example, a reduction in free zinc ion levels in the body is associated with impaired NK cell function (Ibs & Rink, 2003). GATA-3 (a member of the DNA-binding GATA family characterized by characteristic zinc-finger motifs (Ko & Engel, 1993)) is a transcription factor essential for maturation and activity of T-cells and NK cell-related immune function (Ferreira et al., 2005). Specifically, among T-cells, GATA-3 is important for T-helper (T_H)-type 2 cell differentiation as it induces secretion of T_H2 cytokines. The importance of GATA-3 for NK cells was illustrated in GATA-3-deficient mice whose NK cells had an immature phenotype and produced less IFN-γ than NK cells from wild-type mice (Samson et al., 2003).

In cells, MTs are low-molecular-weight metal-binding proteins with a high binding affinity for zinc (≈K_d = 1.4 × 10⁻¹³M). Intracellular zinc homeostasis is mainly regulated by MT, in part, via ion release secondary to reduction of constitutive thiol groups on the MT and the distribution of zinc during (rapid) inter-/intra-zinc finger cluster exchange events (Kagi & Schaffer, 1988; Maret, 2003; Mocchegiani et al., 2000, 2011). It is also known that the MT proteins are also highly inducible in response to oxidative stress (Thirumoorthy et al., 2007). Because PQ is known to induce reactive oxygen species (ROS) generation through activation of enzymes capable of initiating redox cycling of the herbicide (Castello et al., 2007; Miller et al., 2007), it would not be unexpected that this ROS generation would lead to increased MT expression levels. It would in turn follow that, with increased MT gene and protein expression, zinc sequestration in the cells would be increased (resulting in less availability of free intra-cellular zinc). As a result, the question could then arise ‘would any associated reduction in free zinc ion levels circulating in the body ultimately impact on the function of NK and other immune cell types?’

To better define mechanisms underlying potential PQ-induced suppressive effects in NK cells, we hypothesized this agent caused changes in expression of a gene involved in cellular metal ion homeostasis, i.e. MT-1, in PQ-exposed hosts. In turn, these changes would lead to alterations in levels of free zinc ions (in cells and in sera) that, in turn, would result in reduced expression of GATA-3 and, ultimately, changes in T-cell maturation and NK activity. To validate this, PQ-exposed mice (daily [6 days/week for 4 weeks]) were assessed for changes in splenic NK cell activities as well as in MT expression, zinc levels, and GATA-3 expression in their splenocytes.

Materials and methods

Chemicals

PQ dichloride salt was purchased from Sigma (St. Louis, MO). Phosphate-buffered saline (PBS), fetal bovine serum (FBS), and RPMI 1640 medium were purchased from Gibco (Grand Island, NY). Quantichrome™ zinc assay kit was bought from BioAssay Systems (Hayward, CA) and the CytoTox 96® Non-Radioactive Cytotoxicity Assay from Promega (Fitchburg, WV). SuperScript® III First-Strand Synthesis System and Trizol® were purchased from Invitrogen (Carlsbad, CA). iQ™ SYBR® Green Supermix was bought from BioRad (Hercules, CA). FluoZin™-3 was purchased from Molecular Probes (Eugene, OR). Rabbit IgG polyclonal anti-metallo-thionein antibodies were bought from Santa Cruz Biotechnology (Dallas, TX). All secondary antibodies were purchased from Cell Signaling (Beverly, MA). All antibodies for use in flow cytometry were obtained from eBioscience (San Diego, CA).

Experimental animals

Pathogen-free C57BL/6N mice (male, 8-weeks-of-age, 23 [±2] g) were purchased from ORIENT BIO Inc. (Seoul, Korea). The mice were acclimated for 1 week prior to the experiments. Mice were housed in a 12-h light/dark cycle in a controlled temperature (22 ± 2 °C) and humidity (50 ± 5%) environment and allowed ad libitum access to laboratory standard food and water. All procedures and animal treatments were performed in the clean laboratory animal room according to the Guidelines of Laboratory Animal Experimentation of Chung-Ang University. Appropriate Chung-Ang University Ethical Committees related to the use of laboratory animals approved in this work.

Doses and exposure schedules

For the experiments, mice were randomly allocated into four groups, with four animals per group. The mice were then treated intraperitoneally daily with suitable volumes of PQ solution (5 ml/kg) in order to receive 0, 0.2, 1.0, or 5.0 mg PQ/kg of for 4 weeks (6 days/week). At 24 h after the final treatment, blood samples were drawn from the portal vein of all mice, and then each host was euthanized by CO₂ asphyxiation and their spleen (as well as other organs, e.g., liver, testes, kidneys) removed for use in other assays (see below) or for isolation of splenocytes.

Isolation of splenocytes

At necropsy, each spleen was removed aseptically and a single-cell suspension was prepared. In brief, each isolated spleen was gently squeezed with the distal end of a syringe into a plate containing RBC lysis buffer (Sigma). The cells were then washed and cultured in RPMI 1640 medium (supplemented with 10% FBS). To separate non-adherent from adherent cells, the samples were then incubated on the plates for 2 h at 37 °C in a humidified atmosphere containing 5% CO₂. Thereafter, non-adherent cells were isolated and cell concentrations adjusted to 2 × 10⁶cells/ml with the same medium.

Detection of NK cell cytotoxic activity

All procedures were conducted using a CytoTox 96® Non-Radioactive Cytotoxicity Assay kit (Promega) following manufacturer protocols. The target cells were a murine lymphoma cell line (YAC-1) purchased from the Korean Cell Line Bank (Seoul). This line was chosen as it is known to be a sensitive target for the cytotoxic activity of mouse NK cells. YAC-1 cells were maintained in continuous culture throughout the study. The non-adherent lymphocytes isolated as outlined above were directly used as NK effector cells. These cells were washed and re-suspended in RPMI-1640 medium (supplemented with 5% FBS) prior to the start of the assay. For the assay, NK cells (non-adherent splenocytes at 2 × 10⁶cells/ml) together with YAC-1 cells (at 2 × 10⁵cells/ml) were centrifuged at 450 × g for 4 min in wells of a U-bottom 96-well culture plate (Effector:Target ratio = 10:1) in a total volume of 100 μl/well. In parallel wells, 100 μl of NK cells (E), 100 μl of YAC-1 cells (Target cell maximum LDH release control), 100 μl of YAC-1 cells (Target cell spontaneous control), and 100 μl RPMI only were also plated and then centrifuged to serve as controls and blank, respectively. All samples were plated in triplicate. After 4 h of incubation at 37 °C in a humidified incubator with 5% CO₂, kit-provided lysis solution (10 μl) was added to the target cell maximum LDH release control wells 45 min prior to conducting the assay; all other wells received PBS. After incubating a further 45 min at 37 °C, the plate was again centrifuged (450 × g, 4 min), and then 50 μl supernatant from each well was transferred to a corresponding well in a flat-bottom 96-well assay plate. To each well, 50 μl LDH substrate mix was then added and the plate was incubated 30 min at room temperature (protected from light). Then, 50 μl stop solution was added to each well and the absorbance in each well was recorded at 490 nm in a plate reader (TECAN, Männedorf, Switzerland). NK cell cytotoxic activity (%) was computed as 100 × [experimental well – effector only well – target spontaneous well]/[target maximum well – target spontaneous well]. Data were ultimately presented as the mean of triplicate samples (±SEM).

RNA isolation, cDNA synthesis, and quantitative real-time PCR amplification

The organs (liver, testes, kidneys) from each mouse isolated at necropsy were snap frozen and stored at −80 °C until used in these protocols. From each sample, RNA was extracted using Trizol reagent. For quantitative real-time PCR, 0.1 μg total RNA was reverse transcribed using a SuperScript III First-Strand Synthesis System for RT-PCR. Real-time PCR was then performed (in triplicate) using a MyiQ single-Color Real-Time PCR Detection System and an iQ SYBR Green Supermix (Bio-Rad). Relative expression values were all normalized to GAPDH levels. The primer sequences used here were: GAPDH, forward (5′-TTCACCACCATGGAGAAGGC-3′) and reverse (5′-GGCATGGACTGTGGTCATGA-3′); Mt1, forward (5′-ACGTGCTGTGCC-TGATGTGACGAACAG-3′) and reverse (5′-TAGACTCAAACAGGCTTTTATTATTAACG-3′). The thermal cycling conditions used were: pre-denaturation at 95 °C for 5 min, denaturation at 95 °C for 10 s, annealing at 56 °C for 15 s, and extension at 72 °C for 20 s. For the splenocytes, 5 × 10⁶ non-adherent lymphocytes were used for quantification of intracellular mRNA levels followed the same procedures. Relative expression was again normalized to GAPDH levels. Primer sequences used were: GAPDH, forward (5′-TTCACCACCATGG AGAAGGC-3′) and reverse (5′-GGCATGGACTGTGGTCATGA-3′); Gata-3, forward (5′-CTACGGTGCAGAGGTATCC-3′) and reverse (5′-GATGGACGTCTTGGAGA-AGG-3′). The thermal cycling conditions used were: pre-denaturation at 95 °C for 5 min, denaturation at 95 °C for 10 s, annealing at 60 °C for 15 s, and extension at 72 °C for 20 s.

Western blot analysis

Samples of the organ tissues were homogenized in 25 mM HEPES (pH 7.5) containing 5 mM EDTA, 2 mM DTT, 0.1% CHAPS, and protease inhibitor cocktail tablets (Roche, Woerden, the Netherlands). Homogenates were centrifuged at 12,500 × g at 4 °C for 20 min. Protein levels in the cytosolic extracts were determined using a Pierce BCA protein assay (Thermo Fisher, Rockford, IL) with bovine serum albumin as standard. Thereafter, equal amounts of whole protein extracts (50 μg) were resolved over a 15% acrylamide gel by SDS-PAGE electrophoresis, and the proteins were then electrotransferred onto polyvinylidene fluoride (PVDF) membranes. After blocking in a solution of 5% [w/v] low-fat milk powder in TBST (Tris-buffered saline [pH 7.4] containing 0.1% Tween-20) for 2 h at room temperature, the membranes were incubated with specific primary antibodies: rabbit polyclonal anti-MT1 (1:200 dilution) overnight at 4 °C. The filters were then washed 3-times for 10 min (using TBST) and then incubated in blocking solution containing horseradish peroxidase (HRP)-labeled secondary antibody anti-rabbit (1:1000 dilution) for 2 h at room temperature. To verify equal loading, blots were also probed with anti-HSP60 antibody after these analyses. Immunodetection was performed using an Amersham™ ECL™ Western blot detection system (GE Healthcare, Buckinghamshire, UK). Densitometric analysis was then done and all values normalized to HSP60 control values. All signals were analyzed and quantified by Image J (NIH Image, Bethesda, MD).

Determination of serum zinc ion concentrations

Zinc ion levels in serum samples derived from blood collected at necropsy were determined using a QuantiChrom Zinc Assay kit, according to manufacturer protocols. This colorimetric assay is based on zinc binding to a chromogen that then is quantified at 425 nm in the microplate reader.

Flow cytometry

Aliquots (2 × 10⁶ cells/host) of non-adherent lymphocytes were washed in PBS (5 min, 2000 rpm) and then incubated with 0.5 μl Mouse BD F_c Block for 5 min (to block F_c-mediated adherence of antibodies) prior to staining with specific antibodies. The cells were then stained (simultaneously) for 30 min at room temperature in the dark for surface antigens (CD3, NK1.1, and CD122). Thereafter, the cells were washed free of unbound antibody and re-suspended in PBS for analysis in a BD FACSCalibur™ flow cytometer equipped with Cell Quest Pro® software (Becton Dickinson [BD] Immunocytometry System, San Jose, CA). Fluorescence data from a minimum of 20,000 NK cells (CD3⁻, NK1.1⁺, and CD122⁺) were acquired for each sample. The percentages of splenic NK cells presented (see Figure 1) represent the mean from four mice.

Figure 1. Suppressive effects of PQ on NK cells. (A) NK cell activity was determined by LDH assay. Splenocytes were incubated with YAC-1 cells at 10:1 E/T ratio for 4 h. Results are presented in percentages compared to vehicle-treated control cell activities. (B) Percentages of NK cells (CD122⁺ and NK1.1⁺) determined by flow cytometry. CD3⁻ cells were gated first and then the percentages of NK (CD3⁻, CD122⁺, and NK1.1⁺) cells were determined by NK1.1 vs CD122 dot plot. Data represent the mean of three independent experiments. Bar represents mean ± SEM. n = 4 mice/group, *p < 0.05, **p < 0.01 vs vehicle-treated control.

Quantification of free intracellular zinc in splenic NK cells

Non-adherent lymphocytes were placed at 2 × 10⁶ cells/well in a 6-well plate and then FluoZinTM-3 AM ester (dissolved 1:1 in Pluronic® F-127) was added to a final concentration of 1 μM. The cells were incubated at 37 °C for 30 min, washed in PBS (5 min, 2000 rpm), and subsequently incubated with 0.5 μl Mouse BD F_c Block for 5 min prior to antibody staining. The cells were then stained (simultaneously) for surface antigens (CD3, NK1.1) for 30 min at room temperature in the dark, then washed free of unbound antibody and re-suspended in PBS for analysis in the FACSCalibur system. Fluorescence data from a minimum of 20,000 NK cells (CD3⁻NK1.1⁺) were acquired. All results were reported as mean fluorescence intensity (MFI).

Data analysis

All data were expressed as mean ± SE. Differences between groups were analyzed using a one-way analysis of variance ANOVA) with subsequent Turkey’s tests. p-values <0.05 were considered statistically significant.

Results

Immunosuppressive effect of PQ on splenic NK cells

Assays were performed on isolated splenocytes from 16 mice (four mice/group) that were treated daily with IP injection for 28 days with 0, 0.2, 1.0, or 5.0 mg/kg PQ to assess any PQ-induced immunosuppressive effects on NK cells (Figure 1). As shown in Figure 1a, levels of cytotoxic activity were significantly and dose-relatedly lower in the splenic NK cells from PQ-treated hosts compared to that by cells from the vehicle-treated controls. Using flow cytometry to assess populations of NK cells (CD3⁻, NK1.1⁺, and CD122⁺), populations of NK cells were also decreased significantly and dose-relatedly by PQ treatments (Figure 1b).

Induction of MT in organs of PQ-treated mice

It has been reported that MT plays a pivotal role in metal-related cell homeostasis because of its high affinity for metals, in particular zinc (K_d = 1.4 × 10¹³). In several studies, PQ induced oxidative stress that would increase MT expression (Mussi & Calcaterra, 2010). Induction of MT could reduce free zinc ion levels, resulting in reduced NK cell activity. It was hypothesized that PQ had immunosuppressive effects on NK cells via this mechanism. To confirm this, MT expression was quantified by Western blot and RT-PCR in liver, kidney, and testes that were known to have high MT levels. As shown in Figure 2, PQ treatment increased both the mRNA and protein levels of MT in a dose-related manner in all three organs. Induction of MT by PQ treatment was most significant in the liver, but relatively less so in the testes.

Figure 2. Dose effect of PQ on MT levels. Levels found in the (A) liver, (B) kidney, and (C) testes. Organs were removed from mice after 28 days of treatment with PQ. Protein synthesis (left panel) and mRNA expression (right panel) were examined, respectively, by Western blot and quantitative real-time PCR. Densitometry was performed after normalization to control (HSP60). Mt1 mRNA levels were normalized to GAPDH expression in each sample. Data represent the mean of three independent experiments. Bar represents mean ± SEM. n = 4 mice/group; *p < 0.05, **p < 0.01 vs vehicle-treated control.

Free zinc ion levels in serum of PQ-treated mice

Free zinc ion levels in the serum of all mice were analyzed to identify if available free zinc was affected by the change in MT expression. The results shown in Figure 3 indicate that levels of free zinc ion in serum were significantly and dose-relatedly lower in the serum of the PQ-treated hosts compared to those in the serum of vehicle-treated control mice. Changes were significant at doses of 1.0 (p < 0.05) and 5.0 mg PQ/kg (p < 0.01). In the 5.0 mg PQ/kg group, levels were reduced by 20% compared to those in the serum of the control mice.

Figure 3. Concentration of free zinc ions in serum. Levels were assessed in blood collected after treatment with PQ for 28 days in vivo. Zinc ion level in serum of PQ treated mice was determined. Data represent the mean of three independent experiments. Bar represents mean ± SEM. n = 4 mice/group, *p < 0.05, **p < 0.01 vs vehicle-treated control.

MT expression in splenocytes

To verify whether PQ treatment could directly induce MT expression in splenic NK cells, MT mRNA in splenocytes recovered from all the mice was quantified by RT-PCR. As shown in Figure 4, MT mRNA expression was significantly increased in a dose-related manner in PQ-treated mice. These changes were significant (p < 0.01) at doses of 1.0 and 5.0 mg PQ/kg. The values in these two groups were 3.0- and 3.6-fold greater than in the cells from the control mice.

Figure 4. Effect of PQ treatment on expression of Mt1 mRNA in splenocytes. Mt1 mRNA expression in splenocytes of PQ treated mice was examined by quantitative real-time PCR. Mt1 mRNA levels were normalized with GAPDH expression in each sample. Data represent the mean of three independent experiments. Bar represents mean ± SEM. n = 4 mice/group, *p < 0.05, **p < 0.01 vs vehicle-treated control.

Quantification of free intracellular zinc in splenic NK cells

Intracellular free zinc ion levels in splenic NK cells (CD3⁻ NK1.1⁺) were quantified by flow cytometry to verify if there were reductions in zinc levels subsequent in association with any increased MT expression. The results in Figure 5 illustrate that free zinc ion levels decreased dose-relatedly and significantly at doses of 1.0 (p < 0.05) and 5.0 mg PQ/kg (p < 0.01). In the 5.0 mg PQ/kg group, intracellular free zinc ion levels were reduced by >50% compared to in control mice cells.

Figure 5. Free intracellular zinc ion levels in splenic NK cells. Levels were assessed in cells isolated from hosts following treatment with PQ for 28 days. Intracellular zinc levels in NK cells were determined by flow cytometry using a FluoZinTM-3 AM ester. Data shown are mean of three independent experiments. Bar represents mean ± SEM. n = 4 mice/group, *p < 0.05, **p < 0.01 v. vehicle-treated control.

GATA-3 expression

Several transcription factors can be affected by zinc, including one that can affect NK cell function. As it was hypothesized that changes in GATA-3, an essential transcription factor for maturation and activity of NK cells, were related to decreases in NK cell function in PQ treated mice, GATA-3 mRNA expression levels in splenocytes of all mice were analyzed. The results in Figure 6 indicate that PQ treatment causes strong reduction of GATA-3 expression. By even the dose of 0.2 mg PQ/kg, the reduction was ≈70%. Values for the expression in cells from mice that received the 1.0 and 5.0 mg PQ/kg regimens were significantly lower compared to that in the cells from the control mice, with values now 86 and 93% lower.

Figure 6. GATA-3 expression in splenocytes. Expression was assessed in cells isolated from hosts following treatment with PQ for 28 days. Splenocyte GATA-3 mRNA expression was examined by quantitative real-time PCR. GATA-3 mRNA levels were normalized to that of GAPDH expression in each sample. Data represent the mean of three independent experiments. Bar represents mean ± SEM. n = 4 mice/group, *p < 0.05, **p < 0.01 vs vehicle-treated control.

Discussion

PQ is a widely used herbicide, but little is known regarding its influence on the immune system. The main target organ for PQ toxicity is the lung, due to PQ accumulation through the highly developed polyamine uptake system and its capacity to generate a redox cycle (Dinis-Oliveira et al., 2007). Several studies have reported that PQ also affects the liver, kidney, heart, and central nervous system (see Mohammadi-Karakani et al., 2006). Several studies have reported on the immunomodulatory (i.e., immunosuppressive) effects of PQ (Caroleo et al., 1996). Sub-acute exposure to PQ suppressed splenocytes proliferative response to PHA and cytokine production of IFN-γ, IL-4 in Balb/c mice (Riahi et al., 2010). Furthermore, sub-acute exposure to PQ induced decreases in spleen cellularity and splenic CD49b cell levels (Riahi et al., 2011). Acute exposure of PQ at concentrations equivalent to ADI induced up-regulation of IL-17 cytokine family in Balb/c mice (Hassuneh et al., 2012). PQ administration also altered the innate immune systems such as suppressed phagocytic activity of circulating macrophages/granulocyte following PQ exposure in mice (Riahi et al., 2011). In addition, PQ exposure caused irregular change of IgM content and decreased C3 content in fish (Ma et al., 2014).

Residues of PQ have been reportedly found in the immune and hematopoietic systems, such as the bone marrow, spleen and thymus, of PQ-treated rats (Nagao et al., 1994). More recently, a report showed that PQ had suppressive effects on the phagocytic activity of granulocytes and monocytes (Riahi et al., 2011). Thus, while it is clear that PQ can enter the immune system and impart suppressive effects, little is known about the specific mechanisms involved.

The study reported here showed, for the first time, that PQ-mediated immunosuppressive effects in splenic NK cells were associated with increased MT expression, reduced free intracellular zinc levels and changes in GATA-3 expression. As shown in the present study, MT expression and protein synthesis were both increased by PQ treatment. MT are a class of ubiquitously occurring low molecular mass, cysteine- and metal-rich proteins containing sulfur-based metal clusters formed with Zn(II), Cd(II), and Cu(I) ions. MT-1–MT-3 are secreted, suggesting that they may play different biological roles in the intracellular and extracellular space. Recent reports established that these isoforms play an important protective role in brain injury and metal-linked neurodegenerative diseases (Vasak & Meloni, 2011). Although their primary roles have not been identified and remain elusive (Coyle et al., 2002), the MT have a range of functions, including toxic metal detoxification and protection of cells against oxidative stress. With regard to its role in metal ion homeostasis, MT can bind to seven zinc ions with very high affinity and act as a zinc regulator (Sutherland & Stillman, 2011). Thus, although induction of MT by PQ can cause zinc deficiency in the body (Habel et al., 2013), it remains to be determined if such changes in MT levels can be directly/indirectly related to the immunosuppressive effect of PQ and, if so, it should be determined if this is also related to changes in zinc metal ion availability to host immune cells.

Along those lines, the experiments here showed that PQ treatment did cause reduced serum and intracellular levels of free zinc ions. It is clear that zinc is essential for normal functions of the immune system and zinc availability has a significant impact on immune cells, including T-cells, B-cells, and especially NK cells (Borrego et al., 2002; Haase & Rink, 2013; Ozturk et al., 1994). Thus, it is plausible to consider that the decreased zinc levels reported in this study could be a major reason underlying the observed decreases in NK cell cytotoxicity after host PQ treatment.

Even if the change in zinc levels is key to the changes in NK function noted here, the precise mechanism underlying MT induction by PQ is still unclear. Some studies reported that MT could act as a reactive oxygen species (ROS) scavenger (Choi, 2003). To see if one mechanism for the PQ-induced MT expression in NK cells was ROS-related, DCF-DA assays were performed to measure NK cell ROS levels, taking into consideration that PQ can act as a ROS generator (Mussi & Calcaterra, 2010). As expected, ROS levels were significantly increased in the NK cells of PQ treated mice (albeit in non-dose-related manners). In addition, GATA-3 expression in the splenic NK cells was altered following PQ administration. In this study, we showed, for the first time, that PQ could reduce splenocyte GATA-3 expression. As noted above, GATA-3 (a zinc finger transcription factor) is an essential transcription factor for maturation and activity of T-cells and NK cell-related immunity (Ferreira et al., 2005; Schwenger et al., 2005). In terms of NK cell function, GATA-3 promotes NK cell maturation, IFN-γ production, and NKG2A expression (Marusina et al., 2005). A report indicated that, in GATA-3-deficient mice, NK cells present an immature phenotype and produce less IFN-γ than NK cells from wild-type mice (Samson et al., 2003). Considering that GATA-3 expression is dependent on zinc (Muzzioli et al., 2007), it is possible that the PQ-induced reduction in zinc levels observed here caused the decrease in GATA-3 expression.

Synthetically, limited quantities of PQ still may affect the cytotoxic function, one of the principal protective functions in immunity. Therefore, there is a need to more widely progress in the investigation of this and other kinds of environmental contaminants and their effects on important immunological activities.

Conclusions

In summary, we demonstrated the immunosuppressive effect of PQ on NK cells and its underlying mechanisms. We confirmed our hypothesis that PQ induces MT and reduces zinc ion levels. Reduced zinc ion levels result in decreased GATA-3 expression in NK cells, causing a reduction in cytotoxic activity. Further studies are needed to address how GATA-3 affects NK cell functions. We will also need to investigate if zinc supplementation might help prevent some of the immunosuppressive effects of PQ.

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

This research was supported by the Chung-Ang University Research Scholarship Grants in 2012 and the Public Welfare & Safety Research Program through the National Research Foundation of Korea (NRF), as funded by the Ministry of Science, ICT and Future Planning [NRF-2010-0020844].