Exposure to cypermethrin and mancozeb alters the expression profile of THBS1, SPP1, FEZ1 and GPNMB in human peripheral blood mononuclear cells

Abstract

The complex immune system displays a coordinated transcriptional response to xenobiotic exposure by altering expression of designated transcription factors that, in turn, trigger immune responses. Despite the identification of several transcription factors that contribute to regulatory response, very little is known about the specific role of factors that are triggered due to exposure to obnoxious pesticides. Here, for the first time, alterations in human peripheral blood lymphocyte expression of transcriptional factors – thrombospondin-1 (THBS-1), secretory phospho-protein-1 (SPP-1), glycoprotein non-metastatic-β (GPNMB) and fasciculation and elongation factor ζ-1 (FEZ-1), due to in vitro exposure to the crop protection chemicals cypermethrin and mancozeb are reported. Results revealed significant changes in expression profiles due to mancozeb exposure, supporting its immune dysfunction potential; in contrast, cypermethrin exposure did not cause significant changes. Based on these effects on gene expression across the doses tested, it was likely key components of immune mechanisms such as proliferation, cell adhesion, apoptosis and cell activation in human PBMC were affected. Although these data are from in vitro experiments, the results point out the potential role for changes in these factors in the etiology of defective T-cell immune function seen in humans occupationally exposed to crop protection chemicals like mancozeb. These studies suggest the involvement of transcription factors in regulation of pesticide-induced immune dysfunction; these studies also represent a novel approach for identifying potential immune-related dysfunctions due to exposure to pesticides. Further studies are needed to better understand the functional significance of these in vitro findings.

Keywords:

• Cypermethrin
• in vitro immunotoxicity
• mancozeb
• pesticides
• transcriptomic profile

Introduction

Pesticide-induced immune dysfunction is a major concern, eventually having an impact on life-threatening immune-related diseases, such as cancer, hypersensitivity and autoimmunity (Corsini et al. 2008). Nevertheless, epidemiological associations between exposure and the myriad complex immune-related functions in a host are often challenging and there are often no sound conclusions reached about causal relationships between pesticide exposure and such effects (Colosio et al. 2013). Apart from standard paradigms that form the basis of common assays of immunotoxicity, gene–environment interactions also might contribute to immune dysfunction in response to toxicants, including pesticides. Thus, information on expression profile changes in specific transcription factors due to pesticide exposure might help improve the overall scientific knowledge about, and risk assessment strategies related to, these agents (Corvi et al. 2006).

Molecular genetics has a pivotal role in the response to xenobiotics by the immune system of man and other vertebrates. Global gene expression studies in the draining lymph nodes after a single topical exposure to the contact allergen dinitrofluorobenzene (DNFB) had shown significant changes on GlyCAM-1 (glycosylation-dependent cell adhesion molecule 1), guanylate binding protein 2 and onzin (Betts et al. 2003). Similarly, occupational exposure to benzene identified CXCL16, ZNF331, JUN and PF4, as potential biomarkers of early responses to benzene in peripheral blood mononuclear cells (PBMC) of six exposed-control pairs (Forrest et al. 2005). In vitro exposure to TBTO (bis-[tri-n-butyltin] oxide) by the Jurkat human T-lymphocyte cell line (Katika et al. 2011) and mouse thymocytes (van Kol et al. 2012) showed significant changes in Atf4, Atf6, Hspa5, Park7 (Dj-1) and Atox1, that are involved in endoplasmic reticulum stress, and in expression of Bax and Bcl2l11 (Bim) associated with apoptosis.

Several studies have successfully demonstrated expression changes and their casual relation with representative responses in identifying immunotoxicity potentials of various agents (Zeytun et al. 2002; Kinser et al. 2004; Baken et al. 2006, 2007a,b, 2008). Similarly, gene expression profiling studies conducted in tissues/cells such as the liver, spleen, thymus, lymph nodes and peripheral blood lymphocytes also proved useful in predictive risk assessment (Bulera et al. 2001; Thomas et al. 2001; Waring et al. 2001; Hamadeh et al. 2002a,b; Steiner et al. 2004). Therefore, gene expression profiling is a promising tool in toxicity assessment and can be useful for developing new biomarkers (van Leeuwen et al. 2005, 2008; Baken et al. 2007a,b; Hochstenbach et al. 2010).

Modern trends in genomic research have made it possible to generate a hypothesis based on available data that saves the time and resources. Indeed, these approaches address the major goal of identifying key genes and their products that show significant changes in expression profile in a diseased state or due to chemical exposure (Hamadeh et al. 2002a,b; Zhang et al. 2002). This approach is also a promising tool for the generation and testing of toxicity hypotheses (Donald et al. 2002; Zhang et al. 2002) or the identification of perturbed pathways (Borisy et al. 2003) and, thus, plays a crucial role in risk assessment.

Immune functions are complex responses coordinated among various cell types by several transcription factors such as thrombospondin-1 (THBS-1), secretory phospho-protein-1 (SPP-1), glycoprotein non-metastatic-β (GPNMB) and fasciculation and elongation factor ζ-1 (FEZ-1) that have recently been identified as highly and differentially expressed in human lymphocytes upon exposure to exogenous chemicals (Hochstenbach et al. 2010). These key transcription factors (alone or in combination) play significant roles in several key immune functions such as cell adhesion, proliferation, motility, survival (Kyriakides and MacLauchlan 2009; Zhang et al. 2013) and mediating the clearance of apoptotic cells by phagocytes (Tabib et al. 2009) and cell activation (Poole et al. 2013). Similarly, important roles in cytokine production, promotion of cell survival by regulating apoptosis (Wang and Denhardt 2008), neutrophil migration (Apte et al. 2005) and mast cell migration/degranulation (Nagasaka et al. 2008) have also been noted for these factors.

Cypermethrin and mancozeb are widely used in agriculture, households and industry due to their “low” toxicity in mammals and short environmental persistence (Hayes and Laws 1990; Lorgue and Lechenet 1996; Descotes 2004; Hasan 2010). Nevertheless, studies have noted immunomodulatory effects from exposure to cypermethrin (Dési et al. 1985, 1986; WHO 1989; Liu et al. 2006) or mancozeb (Colosio et al. 1996; Corsini et al. 2005; Mandarapu and Prakhya 2015). Accordingly, the present in vitro study using human PBMC cultures was undertaken to assess immunomodulatory potentials of the pesticides based on changes induced in expression of some key transcriptional factors (i.e. THBS1, SPP1, GPNMB, and FEZ1) upon exposure to the pesticides at non-cytotoxic concentrations. To provide relative context about the potential utility of examining these specific factors here, analyses were performed in parallel using both a known immunotoxicant (i.e. benzo(a)pyrene; positive control) and non-immunotoxicant (i.e. urethane; negative control).

Materials and methods

Chemicals

Cypermethrin (> 99%), mancozeb (> 95%), benzo(a)pyrene (> 96%), and urethane (> 99%) were purchased from Sigma (St. Louis, MO). Stock solutions of these chemicals were made in dimethyl sulfoxide (DMSO; Sigma) and stored at −80 °C until further dilution in culture medium at desired concentration(s). The final concentrations of DMSO never exceeded 0.1%.

Peripheral blood mononuclear cells

Upon obtaining informed consent, peripheral blood was collected from individual male volunteers (26–35 years-of-age). The 3^rd Institutional Ethics Committee of the International Institute of Biotechnology and Toxicology approved all experiments. Each blood sample was collected into a heparinized tube and mononuclear cells subsequently isolated using Ficoll-Hypaque (ρ = 1.077 g/ml) density gradient centrifugation (400 × g, 30 min). The buffy coat containing mono-nuclear cells was isolated, transferred to a fresh centrifuge tube and washed twice with phosphate-buffered saline (PBS, pH 7.4), using ≈3 vol of collected buffy coat each time. The final cell pellet containing peripheral blood mononuclear cells (PBMC) was re-suspended to a final level of 1–2 × 10⁶ cells/ml in RPMI 1640 medium supplemented with L-glutamine, 10% heat-inactivated fetal bovine serum (FBS), 100 U penicillin/ml and 0.1 mg streptomycin/ml (all Gibco, Paisley, UK).

Cytotoxicity assessment

Preliminary cytotoxicity studies were performed using PBMC from two donors/test agents to assess biovariance. PBMC (10⁵ cells/well, 24-well plate) were exposed for 24 h to serial doses of cypermethrin or mancozeb in the presence of a metabolic activator (rat liver S9 fraction; Moltox, Boone, NC). Immediately before use, 10% S9-mix containing 15% S9 fraction was added to the reaction medium. Based on trypan blue (Gibco) dye exclusion; doses that caused > 10% cytotoxicity were excluded from further analysis.

Culture set-up/PBMC exposure

To assess the biological variance, assays were performed using PBMC from three donors with each chemical separately. Cultures of PBMC (10⁵ cells/well, 24-well plate) were exposed to non-cytotoxic concentrations of cypermethrin (7.50, 15 or 30 μM) or mancozeb (1.87, 3.75 or 7.50 μM) or benzo(a)pyrene (2.5, 5 or 10 μM; positive control) or urethane (250, 500 or 1000 μM; negative control) or to medium containing solvent (DMSO) or medium alone for 6 h in the presence of freshly-prepared S9 mix at 37 °C under 5% CO₂ and 95% humidity. At the end of the exposure, cells are washed with fresh culture medium and then RNA was isolated.

RNA isolation and qRT-PCR

Total RNA from cells was isolated using RNeasy Mini kit (Qiagen, Hilden, Germany) according to manufacturer protocols. Real-time quantitative PCR assay was performed with a Verso SYBR green 1-step qRT-PCR kit (ThermoFisher Scientific, Chennai, India) using Qiagen RotorGene Q plus with a 1-step protocol. In brief, 30 min at 50 °C (one cycle) for reverse transcription, 10 min at 95 °C to fully activate the Taq DNA polymerase (one cycle), 30 s in 35–40 cycles of 95 °C to denaturate DNA, 1 min in 60 °C followed by 30 s at 72 °C to anneal and extend the primers (data acquisition). Specificity of amplification was controlled by melt-curve analysis (dissociation program) and gel electrophoresis. The amplified product was incubated at 95 °C for 1 min, ramping down to 55 °C at a rate of 2 °C/s. The melt curve analysis was begun at 55 °C and ended at 95 °C by increasing 0.5 °C/cycle. RT-PCR results were analyzed with the ΔΔ threshold cycle method, using β-actin as an internal standard to normalize mRNA amounts.

The primer sequences used were: β-actin: 5′-CCTGGCACCCAGCACAAT-3′ (forward) and 5′-GCC-GATCCACACGGAGTACT-3′ (reverse); THBS1: 5′-CATGCCACGGCCAACAA-3′ (forward) and 5′-TGGCCCAGGTAGTTGCACTT-3′ (reverse); SPP1: 5′-CAGCCACAAGCA-GTCCAGATTATA-3′ (forward) and 5′-CCTGACTATCAATCACATCGGAAT-3′ (reverse); GPNMB: 5′-TCTAAGATCATGTTCCAAGCTAACTGA-3′ (forward) and 5′-GGCTTGGGCCT-GTTATTGTTC-3′ (reverse); FEZ1: 5′-ACAAACATTCTCTTTGCCATGAAG-3′ (forward) and 5′-GTTAGGTAGGGCAGAGCACTTTTAA-3′ (reverse). All primers were obtained from VBC Biotech (Vienna, Austria).

RNA yield was assessed spectrophotometrically and quality determined and considered sufficient by having absorbance values (OD 260/280) that fell between 1.9–2.1. RNA from control samples served as a calibrator (reference standard). Each PCR analysis was run in parallel with a known positive template control (PTC) containing quantified RNA and a known negative template control (NTC) containing water to test the reaction solution.

Data analysis

The fluorescence data acquired on the RotorGene Q plus was analyzed using Rotor Gene Q2plex software (v.1.74; Qiagen) using standard curves specific for each transcription factor (THBS1, SPP1, GPNMB and FEZ1), including the housekeeping β-actin gene and recorded as Ct values (2^−ΔCT and 2^−ΔΔCT). Relative expression changes were determined based on Ct values obtained from the samples for each transcription factor and for the reference β-actin gene using the Relative Expression Software Tool (REST 2009; v2.0.13; Qiagen) which includes normalization and pair-wise fixed reallocation randomization. The program makes no assumptions about distribution of observations in populations and is more flexible than non-parametric tests based on ranks (i.e. Mann–Whitney, Kruskal-Wallis, etc.) and does not suffer a reduction in power relative to parametric tests (i.e. t-tests, ANOVA, etc.) (Pfaffl et al. 2002) with a p = 0.05. Quantification results were expressed in terms of fold-change with respect to the corresponding control samples (calibrator). Different sample groups were obtained by clustering using Euclidean Distance as the distance metric and complete linkage clustering procedure using GenEx software (v6.0; multiD Analyses, Göteborg, Sweden) (Figure 1).

Figure 1. Heat-map analysis. Different sample groups were obtained by clustering using Euclidean distance as the distance metric and complete linkage clustering procedure using GenEx software (Version 6.0).

Results

Quality control of samples

Sample quality and the assay procedures were assessed at different stages. Prior to chemical exposure cell viability was checked and those with > 90% were allowed for exposure. Extracted RNA from the cells was assessed for its quantity and integrity. The average quantity of RNA present in the samples was 5.2 (4.1–6.3) μg and the spectrophotometric readings for quality were between 1.9–2.1. Melt-curve analysis after PCR cycles confirmed primer specificity and identity; all samples met these criteria. The efficiency of the real-time PCR experiment derived from the standard curve was 0.98–1.00, with slopes of −3.38 to −3.31 and linear regressions (R²) of 0.98–1.00 among all the samples. NTC (no template controls) showed no amplification in any run validating the experiments.

Effect on thrombospondin-1 (THBS-1)

Mancozeb and benzopyrene (positive control) caused a dramatic effect on THBS1 expression profiles. At lower concentrations, expression was up-regulated with respect to the reference gene ACTB (Figure 2); however, at high doses, there was a down-regulation. Nevertheless, when expression profiles were assessed together, there appeared to be a concentration-related decrease indicative of a treatment-related effect. In contrast, cypermethrin and urethane (negative control) did not cause significant changes in expression profiles related to treatment.

Figure 2. Thrombospondin-1 (THBS1). Expression profile in exposed PBMC cultures. (A) Cypermethrin. (B) Mancozeb. (C) Urethane. (D) Benzopyrene.

Effect on secretary phospho protein-1 (SPP-1)

Mancozeb and benzopyrene (positive control) had shown a concentration-related decrease in expression profiles of SPP1 with corresponding endogenous reference gene (ACTB), similar to THBS1 (Figure 3). On the other hand, cypermethrin caused a decreased expression profile only at high concentration. Urethane (negative control) exposure did not cause significant changes that could be attributable to treatment at any concentration.

Figure 3. Secretary phosphoprotein-1 (SPP1). Expression profile in exposed PBMC cultures. (A) Cypermethrin. (B) Mancozeb. (C) Urethane. (D) Benzopyrene.

Effect on glycoprotein non-metastatic-β (GPNMB)

Mancozeb and benzopyrene (positive control) caused expression changes in GPNMB that were similar to THBS1. At lower concentrations, expression showed up-regulation with respect to the reference gene (ACTB) and at high doses there was a down-regulation. Nevertheless, when expression profiles were assessed together from all three experiments, there was a concentration-related decrease – thus indicating treatment-related effects (Figure 4). Cypermethrin and urethane (negative control) did not cause any significant changes in expression profile that could be related to treatment.

Figure 4. Glycoprotein non-metastatic B (GPNMB). Expression profile in exposed PBMC cultures. (A) Cypermethrin. (B) Mancozeb. (C) Urethane. (D) Benzopyrene.

Effect on fasciculation and elongation factor ζ-1 (FEZ-1)

At lower concentrations, both mancozeb and benzopyrene (positive control) did not cause any expression changes (Figure 5). However, at intermediate and high doses, FEZ-1 expression was down-regulated with respect to the endogenous ACTB reference gene. In contrast, cypermethrin caused up-regulation of FEZ-1 expression in a concentration-related fashion. Urethane (negative control) did not cause any expression changes.

Figure 5. Fasciculation and elongation factor ζ-1 (FEZ1). Expression profile in exposed PBMC cultures. (A) Cypermethrin. (B) Mancozeb. (C) Urethane. (D) Benzopyrene.

Biological interpretation

An attempt was made to identify various biological functions related to the immune system that might be affected by the observed expression changes among the factors analyzed. This approach used gene ontology-based weighing for identifying the biological functions related to the selected query genes (THBS1, SPP1, FEZ1, GPNMB) in association with 10 more closely related genes with/without 10 attributes using GeneMania software (Figure 6). The query genes were involved in several biological functions directly or in combination with other genes that are related to the immune system. Interestingly, several of these biological functions were closely associated with xenobiotic exposure, such as: chemotaxis of leukocytes, granulocytes, and neutrophils; migration of leukocytes, granulocytes, and neutrophils; regulation of cell adhesion; positive regulation of cell motility; regulation of leukocyte activation; macrophage differentiation; positive regulation of T-cell/lymphocyte activation; positive regulation of cell migration; platelet activation; and apoptotic cell clearance.

Figure 6. Functional network – GeneMania.

THBS1 in combination with other genes was involved in most of the functions, followed by SPP1 > FEZ1 > GPNMB. Therefore, any down-regulated expression due to mancozeb or benzopyrene exposure might indicate likely immunomodulatory effects at the doses tested.

Discussion

The present study analyzed the expression profiles of key transcriptional factors THBS1, SPP1, GPNMB and FEZ1 in human PBMC cultures after exposure to cypermethrin, mancozeb, benzopyrene or urethane. These factors might operate as key modulators of immune function under basic conditions or in the case of xenobiotic exposure and could be helpful in identifying the candidate marker genes of toxicity. This study showed that these transcription factors expressed a balanced profile due to exposure of chemicals and their changes were correlatable with the known immunomodulatory properties of the two test pesticides, mancozeb and benzopyrene (positive control). To our knowledge, this is the first study to examine the expression profiles of these transcription factors in cultured human PBMC after exposures to cypermethrin or mancozeb.

Adhesion and migration are two key mechanisms in regulation of lymphocyte behavior and function and are sensitive to signals delivered through antigen recognition. THBS1 is one of the key transcriptional factors in regulating T-cell adhesion and migration (Forslöw et al. 2007), inflammation and tissue damage or stress (Kyriakides and MacLauchlan 2009; Zhang et al. 2013). The predicted biological functions based on gene ontology also indicate the role of THBS1 in several immunological functions. In the present study, the observed down-regulation of THBS1 due to mancozeb or benzopyrene (positive control) exposure (in dose-related fashion) suggests that THBS1 could be a promising candidate biomarker in assessing inducible changes in immune function.

Secretary phosphoprotein-1 (SPP1, also called osteopontin) has both inflammatory and anti-inflammatory actions, the capability to modify gene expression and an ability to promote migration of monocytes/macrophages (Denhardt et al. 2001). The expression changes reported here showed an up-regulation due to cypermethrin exposure and a down-regulation due to mancozeb (or benzopyrene [positive control]) in a dose-related fashion. The network predictions based on biological functions also indicated a role for SPP1 in modulating the immune functions such as T-cell chemotaxis, adhesion and proliferation as a result of toxicant exposure. Thus, SPP1 could be another sensitive marker for assessing immunomodulation due to toxicant/pesticide exposure.

GPNMB is expressed on antigen-presenting cells (APC) that attenuate T-cell activation by binding to Syndecan-4 (SD-4) on activated T-cells (Chung and Pyo 2005; Chung et al. 2007a,b; Tomihari et al. 2010). This factor is also expressed constitutively at high levels by many other APC subsets and at lower levels by some non-lymphoid cells (Weterman et al. 1995; Shikano et al. 2001). Key immunological functions mediated by GPNMB include macrophage differentiation and serving as a negative regulator of macrophage inflammatory responses (Ripoll et al. 2007). In the current study, decreased expression of GPNMB due to mancozeb or benzopyrene (positive control) exposure was observed. Ultimately, these changes could manifest as an inhibition of macrophage differentiation after exposure to these chemicals.

Unlike the other three factors, the role of FEZ1 is not clearly defined in immune-related functions. Nevertheless, FEZ1 was shown to have a differential expression in a recent study conducted in human bone marrow stromal cells (BMSC; Ren et al. 2011). It has been suggested FEZ1 may actually play a role in cell adhesion (Bloom and Horvitz 1997) and interactions with various intracellular partners, such as signaling, motor and structural proteins, to function as an anti-viral factor (Maturana et al. 2010). Here, FEZ1 expression increased due to cypermethrin and decreased due to mancozeb or benzopyrene (positive control). However, the analyses here were unable to define any immune system-related biological functions that were mediated through FEZ1 alone or in combination with other genes, based on network predictions using Genemania.

Down-regulation of some of these transcription factors due to mancozeb or benzopyrene (positive control) exposure was likely a basis for previous observations showing there were significant changes in proliferation and T_H1/T_H2 cytokine (TNFα, IFNγ, IL-2, IL-4, IL-6, IL-10) secretion by human peripheral blood lymphocytes (Mandarapu et al. 2014). Further, the functional significance of these altered gene expressions may contribute to a multitude of immune cell-related functions/processes, including chemotaxis, migration, cell adhesion and motility, lymphocyte activation, immune cell migration, macrophage differentiation, platelet activation and apoptotic cell clearance/cell pathways/processes, due to mancozeb or benzopyrene (positive control) exposure.

As genomic-based approaches are gaining more attention in toxicity assessment, the current approach using selected transcription factors based on various methods might help in improving the current testing strategies. Further, the approach here reflects the potential utility of including these novel protocols in assessing changes in immune function that reflect toxicities of these specific and other commonly-encountered pesticides/environmental agents.

Conclusions

The present in vitro study demonstrated the utility of transcription factors in enhancing researchers’ understanding of potential immunomodulatory effects due to pesticide exposure. Furthermore, the results obtained also indicated there was now also an opportunity for understanding the functional immunotoxicologic significance of alterations in gene expressions in PMBC due to pesticide exposure. We wish to note for the readers, however, that a fuller validation of the outcomes here requires an even more robust approach. While one might view the current work as mainly preliminary, we strongly believe the current experimental set-up with three different donors was sufficient enough to generate an understanding of the utility of the approach. This then paved the way for future investigations with a far more expansive pool of donors, etc. to more fully understand the functional immunotoxicologic significance of alterations in gene expressions in PMBC due to pesticide exposure.

Acknowledgements

This work is a part of the PhD Thesis of Rajesh Mandarapu, approved by the University of Madras, and was undertaken at the International Institute of Biotechnology and Toxicology (IIBAT). The clinical support of Dr Srivatsa Prakhya (SRL, Indore, India) is appreciated.

Declaration of interest

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