Formaldehyde exposure induces differentiation of regulatory T cells via the NFAT-mediated T cell receptor signalling pathway in Yucatan minipigs

The use of minipigs (Sus scrofa) as a platform for toxicological and pharmacological research is well established. In the present study, we investigated the effect of formaldehyde (FA) exposure on helper T cell-mediated splenic immune responses in Yucatan minipigs. The minipigs were exposed to different inhaled concentrations of FA (0, 2.16, 4.62, or 10.48 mg/m3) for a period of 2 weeks. Immune responses elicited by exposure to FA were determined by assessing physiological parameters, mRNA expression, and cytokine production. Additionally, the distribution of helper T cells and regulatory T (Treg) cells and expression of NFAT families, which are well-known T cell receptor signalling proteins associated with regulatory T cell development, were evaluated. Exposure to FA suppressed the expression of genes associated with Th1 and Th2 cells in minipigs in a concentration-dependent manner. The subsequent production of cytokines also declined post-FA exposure. Furthermore, exposure to FA induced the differentiation of CD4+ Foxp3+ Treg cells with divergent expression levels of NFAT1 and NFAT2. These results indicated that exposure to FA increased the Treg cell population via the NFAT-mediated T cell receptor signalling pathway, leading to suppression of effector T cell activity with a decline in T cell-related cytokine production.


Results
FA exposure does not induce a change in body or organ weight. Exposure to FA did not cause any differences in the body weights (p > 0.918) of or any weight gain (p > 0.444) in the minipigs in the FA-exposed group (Fig. 1a,d). In the control group exposed to clean air, the body weights increased from 10.1 ± 1.6 kg to 11.7 ± 1.6 kg. All animals in the FA-exposed group exhibited similar changes in body weight with time as those in the control group. We also assessed the effect of FA exposure on the relative organ weight in minipigs in the following manner.
There were no significant changes in the relative weights of organs, including the lung, spleen, and thymus [lung (control group: 161.0 ± 24.0%, 2.16 mg/m 3  FA exposure causes a decline in IFN-γ, TNF-α, and IL-4 production. To investigate the effect of FA exposure on the immune system, we assessed the expression of genes associated with Th1 (IFN-γ, TNF-α) and Th2 (IL-4) cells and the subsequent production of their cytokines. Exposure to FA resulted in a decline in IFN-γ, TNF-α, and IL-4 mRNA levels in a concentration-dependent manner, with significant suppression of IL-4 expression observed at all assessed FA levels. Additionally, gene expression was significantly lower in the animals exposed to 10.48 ± 0.64 mg/m 3 FA than in the control group animals (Fig. 2a-c). A similar expression pattern was observed in the production of Th1-and Th2-related cytokines; however, the production of IL-4 (p = 0.02) and TNF-α (p = 0.02) was significantly lower in the FA exposure groups than in the control group ( Fig. 2d-f). FA exposure results in increased NFAT1 expression and decreased NFAT2 expression. Nuclear factor of activated T cells (NFAT) families are well-known T cell receptor (TCR) signalling proteins important for the regulation of Treg cell development 42 . To investigate the molecular mechanism of action of Treg cells in T Figure 1. Body and organ weights in Yucatan minipigs. Changes in body weight (a) and weight gain (d) of minipigs exposed to FA. Relative weights of organs, including the lung (b), spleen (c), and thymus (e) were calculated. Data are presented as the mean ± SD (n = 2 minipigs/group). FA formaldehyde.

Figure 2.
Helper T cell-related mRNA expression and cytokine production by FA exposure. Splenocytes isolated from FA-exposed minipigs were cultured with 2.5 μg/mL Concanavalin A (Con A) for 72 h. The mRNA expression of genes associated with Th1 (IFN-γ, TNF-α) and Th2 (IL-4) was evaluated by qRT-PCR (a-c). Target gene expression was normalised to GAPDH expression and the expression is presented as fold change relative to the control group. Production of cytokines associated with Th1 (IFN-γ, TNF-α) and Th2 (IL-4) was evaluated (d-f). Data are presented as the mean ± SD from two independent experiments (n = 2 minipigs/group). *p < 0.05, **p < 0.01. FA formaldehyde. www.nature.com/scientificreports/ cell-related cytokine suppression due to FA exposure, we evaluated the protein expression of NFAT1 and NFAT2.
Notably, NFAT1 expression in the 4.62 mg/m 3 FA exposure group was nearly 1.28-fold higher than that in the control group. In contrast, NFAT2 expression in the FA exposure groups was decreased by approximately 0.94fold compared with that in the control group (Fig. 4).

Discussion
Minipigs, which bear close similarity to humans, are a useful non-rodent animal model for toxicology research studies. This is the first study on minipigs addressing the issue of FA exposure-induced immune modulation.
Many studies have reported that FA exposure affects helper T cell-related immune responses. However, the role of Treg cells in FA exposure-induced immune responses is still not well understood.  www.nature.com/scientificreports/ FA is a ubiquitous environmental pollutant and its ingestion by inhalation constitutes occupational and environmental health hazards 47 . In the present study, we report the suppressive effects of FA exposure on splenic immune responses, determined by evaluating the helper T cell and Treg cell populations and by assessing the expression of immune-related factors including mRNAs, cytokines, and proteins in Yucatan minipigs. We administered FA at a concentration of 2.16 mg/m 3 in this study based on the no-observed-adverse-effect-concentration (for mouse: 2.46 mg/m 3 , for monkey: 1.23 mg/m 3 ) assigned by the Organisation for Economic Co-operation and Development Screening Information Dataset (OECD SIDS) 48 . The current short-term exposure limit for FA in the United States is 2.46 mg/m 349 . We also assessed the effect of FA at a concentration of 4.62 mg/m 3 , as it was the lowest-observed-adverse-effect l concentration (for mouse: 5.04 mg/m 3 , for monkey: 3.69-7.38 mg/m 3 ) as determined by the OECD SIDS 48 . Finally, a concentration of 10.48 mg/m 3 FA was used as the optimal high concentration to continuously control the minipig inhalation system. A previous animal study using these concentrations reported that FA exposure impaired the function and differentiation of natural killer cells 50 . Other studies have also shown that FA exposure affects immune responses, including helper T cells and lung inflammation-related up-and down-regulation of gene and protein expression 32,51,52 . Additionally, the mouse FA concentrations used in several studies were converted into minipig FA concentrations using Alexander's formula 53 ; these results and the common ratio of 2 are reflected in our exposure concentrations (Park et al. 46 27 : 0.5, 3 mg/ m 3 → 1.14, 6.86 mg/m 3 ; Jung et al. 32 and Kim et al. 50 ; 5, 10 ppm → 5.64, 11.38 mg/m 3 ). Thus, based on these studies and their results, we exposed minipigs to 2.16, 4.62, and 10.48 mg/m 3 at 2 h/day for 2 weeks (5 days a week).
To investigate the effect of FA exposure on physiological parameters, the body weights and relative organ weights (including the lung, spleen, and thymus) of the Yucatan minipigs exposed to FA were assessed. Our data showed that exposure to FA caused no difference in body weight, weight gain, and the relative weight of various organs. These findings are consistent with those of previous studies 30,32 . Additionally, to examine the effect of FA exposure on airway inflammation, the total and differential cell counts of macrophages, eosinophils, neutrophils, and lymphocytes in the bronchoalveolar lavage fluid were determined and histopathological analyses were conducted (data not shown). No significant differences were observed between the total and differential cell counts of the FA-exposed and control groups. These findings are consistent with those of previous studies showing that exposure to FA does not induce significant differences in the counts of various inflammatory cell types in bronchoalveolar lavage fluid 30,46 . However, in the histopathological analysis, infiltration of inflammatory cells and degeneration of the bronchial epithelium were noted to be increased in the 10.48 mg/m 3 FA exposure group. These observations suggested that the FA concentrations used in this study caused minimal airway inflammation but did not induce direct lung injury. Thus, in the present study, we investigated changes in the splenic immune response at FA concentrations that did not directly promote lung injury.
The spleen is a highly organised lymphoid organ and is important for innate and adaptive immune responses 55 . In the spleen, the proper differentiation and development of different subsets of effector T cells (Th1, Th2) and Treg cells are initiated in the presence of lineage-specific effector cytokines during T cell activation 56 . Recent human and animal studies have reported that FA exposure adversely affects the immune system by altering the population of different types of T cells as well as the production of helper T cell-related cytokines 29,[33][34][35]46 . Thus, we investigated the potential effect of FA exposure on splenic immune responses by evaluating the expression of helper T cell-related mRNAs and cytokines in splenocytes that were activated by Concanavalin A, which induces the mitogenic activity of T cells and increases the synthesis of cellular products. Our results demonstrate that FA exposure suppressed the expression of all helper T cell-related genes in a concentration-dependent manner, while IL-4 expression was significantly decreased at all inhaled concentrations. These findings are consistent with the results of our previous study, which showed the suppression of Th-1, Th-2, and Th-17 cell-related splenic cytokine production and mRNA expression due to FA exposure in a concentration-dependent manner 46 . Furthermore, Wei et al. reported that levels of helper T cell-related cytokines were suppressed in FA-exposed C57BL/6 mice 29 . Recent studies have revealed that FA exposure suppressed Th1-and Th2-related cytokines in rodent models with ovalbumin sensitisation, thereby resulting in a decrease in airway inflammation and bronchial hyperresponsiveness 27,30 . These results indicate that FA exposure suppresses effector T cell activity, inducing decreased T cell-related mRNA expression and cytokine production.
Treg cells actively suppress pathological and physiological immune responses, which contribute to the maintenance of immunological self-tolerance and immune homeostasis 57 . The suppressive functions of Treg cells can be grouped into four modes of action: (1) suppression mediated by the cytokines IL-10, IL-35, and TGF-β; (2) suppression by cytolysis mediated by granzyme A or B; (3) suppression by metabolic disruption mediated by high-affinity CD25 and cyclic AMP; and (4) suppression by targeting dendritic cells through LAG3 and CTLA4 58 . Hence, to determine whether FA exposure suppresses immune responses via Treg cells, we evaluated the population of helper T cells and Treg cells and evaluated the changes in their signalling pathways. Our results show that exposure to FA caused no difference in the population percentage of CD4 + helper T cells in minipigs. However, exposure to FA significantly increased the population size of splenic CD4 + Foxp3 + Treg cells. Thus, our findings are consistent with those of prior studies on FA-exposed human and rodent models 32,46,59,60 .
Recent studies have indicated that the NFAT-mediated TCR signalling pathway contributes to the induction of Foxp3 expression, which controls the differentiation and function of Treg cells 42 . NFAT proteins are activated by cell surface receptors that are coupled to Ca 2+ mobilisation 61 . The increased levels of cytosolic calcium are bound by calmodulin, which in turn activates calcineurin, a calcium-and calmodulin-dependent serine/threonine protein phosphatase 62 . NFAT proteins are dephosphorylated by activated calcineurin, resulting in nuclear translocation of these proteins and the induction of NFAT-mediated gene transcription 63 . Recent in vivo and in vitro studies have revealed that NFAT1 plays a crucial role in the suppressive function of Treg cells [64][65][66][67] , along with enhancing and maintaining stable Foxp3 expression 62,68-71 . In contrast, NFAT2 induces the activation of effector T cells and the production of effector cytokines in the immune system 72 www.nature.com/scientificreports/ resulted in an increase in NFAT1 expression and Treg cell population size in minipigs, coupled with a decline in IL-4 production. In contrast, FA exposure precipitated no notable difference in NFAT2 expression. These results indicate that FA exposure activated the NFAT-mediated TCR signalling pathway with divergent expression of NFAT1 and NFAT2, leading to an increase in the population of Treg cells. These events may have subsequently induced an immunosuppressive microenvironment along with inhibition of effector T cell activity.
Owing to their close sequence homology with humans, minipigs are considered a useful non-rodent animal model platform for conducting toxicology research. In this study, we evaluated the effects of FA exposure on splenic immune responses in Yucatan minipigs. Our results revealed that exposure to FA increased the differentiation of Treg cells via the NFAT-mediated TCR signalling pathway with divergent expression of NFAT1 and NFAT2, resulting in the suppression of effector T cell activity with decreased production of T cell-related cytokines. Although some studies have reported that FA exposure may provoke or exacerbate Th2-type responses in murine and human models, other studies have found that FA exposure does not aggravate allergic responsiveness and that FA exposure reduces the development of allergic lung inflammation. The differences in species and strains of animals, concentrations and durations of FA exposure, and experimental protocols result in disparate immune responses being observed in response to FA exposure. Therefore, further studies under various conditions (28 days or 90 days for long-term studies; with administration of low and high doses of FA) are necessary to determine the impact of FA exposure on the immune systems. In conclusion, our findings provide insight into the molecular mechanisms underlying the FA exposure-induced development of an immunosuppressive microenvironment, characterised by an increased Foxp3 + Treg cell population. Development of such an immunosuppressive microenvironment may potentially result in detrimental health effects, such as increasing host susceptibility to opportunistic infections and the progression of cancer.

Methods
Animals. Six-month-old male Yucatan minipigs (Optipharm Inc., Chenongju, Korea) were used in this study. Each minipig was placed in an individual pen in an animal room with controlled temperature (22 ± 2 °C) and humidity (50 ± 5%), a 12 h light/dark cycle, and positive pressure with HEPA-filtered air. The animals were fed sterilised food pellets (Farm Story Dodram B&F, Seoul, Korea) and sterilised tap water ad libitum and were acclimatised for 3 weeks prior to commencement of FA exposure. All experimental procedures carried out in this study were reviewed and approved by the Institutional Animal Care and Use Committee of the Korea Institute of Toxicology (IACUC #1908-0297). All procedures were performed in accordance with the relevant guidelines and regulations, including compliance with ARRIVE guidelines.
Experimental groups. The minipigs were randomly divided into four groups (n = 2 minipigs/group): control group, 2.16 mg/m 3 FA exposure group, 4.62 mg/m 3 FA exposure group, and 10.48 mg/m 3 FA exposure group. The control minipigs were not exposed to FA. Minipigs in the treatment groups were exposed to FA for a period of 2 weeks at 2 h/day for 5 days a week through a minipig mask-type inhalation system (Fig. 5a,b). Body weights were measured on days 2, 6, 9, and 13 prior to FA exposure. Terminal body weight was measured 24 h after the last FA exposure. The pigs were sedated using 0.5 mg/kg midazolam and 5 mg/kg ketamine intramuscularly and were euthanised under isoflurane anaesthesia. All samples were collected for subsequent analysis. FA exposure. FA was administered as a methanol-free ultrapure 10% FA solution (Polysciences Inc., Warrington, PA, USA) using the minipig mask-type inhalation system. The FA was diluted with clean air using a multi-neck flask to achieve the desired FA concentrations and delivered through the minipig mask (Fig. 5a). The FA in the mask was sampled using a Top Solid DNPH cartridge (Top-Trading Co., Seoul, Korea) and was monitored hourly by high-performance liquid chromatography with ultraviolet detection (HPLC-UV). All of the minipig groups were exposed to 0 (control), 2.16 ± 0.16 (mean ± SD), 4.62 ± 0.36, or 10.48 ± 0.64 mg/m 3 FA (Fig. 5c,d).
For flow cytometric analysis, the splenic single cells were resuspended in Flow Cytometry Staining Buffer (eBioscience Inc., San Diego, CA, USA) prior to analysis. Cytokine production of spleen cell cultures. The splenocyte culture medium, obtained 72 h after primary splenocyte culture, was assessed for the cytokines IL-4, IFN-γ, and TNF-α using the ProcartaPlex™ immunoassay kit (eBioscience) and the Luminex 200™ system (Luminex Corporation, Austin, TX, USA) according to the manufacturer's instructions. Cytokine concentrations were quantified by evaluating the fluorescent signal of analyte-specific capture beads and analysed using ProcartaPlex Analyst 1.0 (eBioscience). All standards and samples were measured in duplicate.
Flow cytometric analysis. The single spleen cells isolated from FA-exposed minipigs were washed with Flow Cytometry Staining Buffer (eBioscience) and stained with PerCP-Cy™5.5-conjugated anti-CD4 monoclonal antibody (BD Biosciences, San Jose, CA, USA) for 60 min on ice. For intracellular staining, the spleen cells were permeabilised with a Fixation/Permeabilization solution (eBioscience) and stained with an FITC-conjugated anti-Foxp3 monoclonal antibody (eBioscience) for 60 min at 20 °C. All samples were run on a CytoFlex flow cytometer (Thermo Scientific). Data of 50,000 events were collected and analysed using CytoExpert 2.3 software (Thermo Scientific).
Western blot analysis. The spleen tissue was processed in RIPA buffer (Pierce Biotechnology, Rockford, IL, USA) to generate denatured protein lysate, which was quantified using the Pierce BCA Protein Assay kit (Thermo Scientific). Protein lysates (30 µg) were mixed with 4 × Laemmli sample buffer (Bio-Rad Laboratories, Hercules, CA, USA), separated by 8% SDS-PAGE, and transferred onto polyvinylidene difluoride (PVDF) membranes (Millipore, Billerica, MA, USA). The membranes were blocked using 5% bovine serum albumin in Tris-buffered saline with 0.1% Tween 20 (TBS-T) for 60 min at 20-24 °C The membranes were then incubated Figure 5. Schematic of the minipig mask-type inhalation system for FA exposure, experimental design, and FA concentrations for assessing the effect of FA exposure on minipigs. (a) FA was generated using a gas bubbler and diluted with clean air using a multi-neck flask. (b) All minipigs received mask training for 2 h per day before FA exposure. Minipigs in the treatment groups were exposed to FA for a period of 2 weeks at 2 h/day, 5 days a week using the minipig mask-type inhalation system. The (c) mean and (d) daily concentrations of FA exposure were monitored using a Top Solid DNPH cartridge and HPLC system. The groups of Yucatan minipigs were exposed to 0 mg/m 3 (control), 2.16 ± 0.16 mg/m 3 , 4.62 ± 0.36 mg/m 3 , or 10.48 ± 0.64 mg/m 3 FA for 2 weeks (5 days/week) at 2 h/day. Data are presented as the mean ± SD. FA formaldehyde.