4. Discussion
To investigate effects of MQL on allergic immune responses, we used an OVA-induced allergic mouse model, and showed that MQL suppresses Th2-related immune responses such as serum IgE production and Th2 cytokine production by splenocytes. In our previous studies, we revealed that MQL was one of active compounds in Rosae multiflorae fructus extract, which can ameliorate asthma, rhinitis, and food allergies21,40. Furthermore, Nguyen et al . reported that Rosae multiflorae fructus has an anti-allergic effect by suppressing antigen-specific T cell activation and proliferation in vivo 41. Our results in vivo study also suggest that MQL suppresses splenocyte proliferation and IL-2 production, thereby inhibiting Th2-related immune responses. Therefore, we focused on CD4+ T cells and investigated the anti-allergic mechanisms of MQL.
Th2-dominant responses, showing increased Th2 (IL-4, IL-5, and IL-13) cytokines and IgE production, are characteristic immune responses in patients with allergic diseases such as food allergy, asthma, and AD42–44. We previously reported some strategies to treat allergic diseases45–47. The first involves regulating the Th1/Th2 immune balance by suppressing excessively induced Th2-mediated immune responses or increasing Th1-related immune responses48–50. However, here we confirmed that MQL treatment does not increase the Th1-associated cytokine IFN-γ. Thus, we focused on T cell proliferation as reducing T cell clonal expansion and differentiation represents an alternative way to manage allergic diseases2. Mechanisms for suppressing T cell proliferation are typically inducing regulatory T cells (Treg) or up-regulating HO-1 expression. When we investigated the effects of MQL on Treg induction, MQL treatment did not increase CD4+Foxp3+ T cell populations (data not shown). Therefore, in the present study, we investigated the effects of MQL on Th2-dominant allergic immune responses and mechanisms involved in HO-1 expression.
HO-1 is known to have various physiological functions including a protective effect against oxidative stress12,13. In particular, it was reported that up-regulation of HO-1 in CD4+ T cells suppresses their proliferation by inhibiting IL-2 production7. Similarly, our results showed that MQL suppresses CD4+ T cell proliferation and IL-2 production induced by anti-CD3/CD28 antibody stimulation, while HO-1 expression was increased. Based on these results, we hypothesize that MQL may be able to suppress the CD4+ T cell-mediated immune responses by inhibiting proliferation of CD4+ T cells via up-regulation of HO-1. Furthermore, we speculate that HO-1 is a key regulator of CD4+ T cell proliferation.
Up-stream signals of HO-1 have been reported by many previous studies. Notably, Nrf2, which is normally repressed by Keap1, translocates to the nucleus upon activation and binds to ARE sequences of the HO-1 gene. Especially, about the regulation of HO-1 on cell proliferation, activation of MAPKs (ERK, JNK, and p38) as up-stream signals of Nfr2 plays as factors, which are able to regulate cell proliferation. However, the specific MAPKs involved depend on the administered compounds. For example, it was reported that quercetin enhances Nrf2/HO-1 activity via both ERK and p38 signaling pathways51,52 and suppresses T cell proliferation by enhancing HO-1 expression20. Quercetin-3-O-β- D-glucuronopyranoside, also known as isoquercitrin, can also up-regulate HO-1 expression via ERK pathways53. Although in a different cell type, in a recent study, Lee et al. revealed that MQL up-regulates HO-1 expression via the Nrf2 pathway in human hepatoma cells (HepG2) but does not activate nuclear factor-kappa B (NF-κB)54. In this study, we found that MQL activated Nrf2-HO-1 signals through phosphorylation of ERK in CD4+ T cells. In experiments using MAPK inhibitors, we confirmed that U0126 (ERK inhibitor) most significantly suppressed HO-1 expression, and MQL strongly increased ERK phosphorylation in CD4+ T cells. Although treatment with SB20350 (p38 inhibitor) suppressed HO-1 expression, it was relatively weak compared with the effect of U0126. Moreover, p38 phosphorylation was not detected following MQL treatment. Therefore, our results demonstrate that ERK activation is upstream of MQL-induced HO-1 expression in CD4+ T cells.
Ras is a small GTP-binding protein with three isoforms (H-Ras, K-Ras, and N-Ras), and controls growth via ERK activation and other intracellular signaling pathways55. Membrane-bound Ras (GTP-bound active form) can activate Raf by recruiting Raf and promoting the formation of B-Raf /C-Raf complexes or homodimers. Activated Raf protein can then induce phosphorylation of ERK1/234,56. Previous studies have reported that HO-1 expression can be up-regulated by activating the Ras-Raf-ERK signaling pathway57,58. Based on these findings, we investigated whether MQL-induced HO-1 up-regulation was associated with C-Raf phosphorylation. Our results showed that MQL could induce phosphorylation of C-Raf in CD4+ T cells, indicating that C-Raf could be a potential up-stream signal of the ERK-Nrf2 signaling pathway.
ROS such as superoxide anion radicals (·O2-), hydroxyl radicals (·OH), and hydrogen peroxide (H2O2) are constantly produced as a byproduct of mitochondrial oxidative metabolism. Excessively generated ROS, however, can induce oxidative stress, which underlies various diseases, while normal ROS levels can regulate cell growth by modulating proliferation and differentiation. Specifically, ROS activate cellular pathways such as MAPK, JAK/STAT, Nrf2, and NF-κB signals59–61. Furthermore, intracellular ROS are known to induce HO-1 expression62,63. Consistent with our results, it was reported that intracellular ROS activates the Raf-MEK-ERK signaling pathways64 and could induce ERK1/2 phosphorylation by activating Ras39,65. Importantly, flavonoids are antioxidant secondary phenolic metabolites naturally produced in fruits and vegetables, and are able to against oxidative stress66. However, the flavonoids can also increase intracellular ROS levels as prooxidants. For example, McNally et al , revealed that curcumin increases HO-1 expression via ROS generation67. In addition, it has been reported that quercetin can also act as a prooxidant capable of producing intracellular ROS37,68. In this study, we found that MQL induces ROS production in CD4+ T cells, which enhances HO-1 expression via Raf-ERK activation. Consequently, this leads to suppression of IL-2 production and CD4+ T cell proliferation.
TMA as a hapten has been known to induce allergic AD by evoking Th2-dominant immune responses in mice26,69–71. Haptens act as antigens by generating hapten-protein complexes through conjugation with self-proteins then activate adaptive immune responses when recognized by APCs such as DCs71. Langerhans cells, which are DCs that reside in the epidermis, induce naïve CD4+ T cell differentiation by presenting antigens after migrating into T cell-rich areas, such as the paracortex of the DLNs72. To characterize the effects of MQL on a TMA-induced AD-like mouse model, we administered MQL either orally or topically, since both oral and topical therapies are effective for treating AD. Considering that this was a disease model, we used a higher dosage of MQL (4 and 10 mg/kg) than in the OVA-induced Th2-dominant mouse model (MQL, 2 and 4 mg/kg). Our results revealed that AD symptoms such as ear swelling, tissue infiltration of inflammatory cells, and IgE production were ameliorated by MQL in both topical and oral treatments. Interestingly, serum IgE levels were strongly reduced by oral administration of MQL compared with topical administration, possibly because of IgE production involves a systemic immune response. However, in DLNs, which represent peripheral immune responses, IL-13 levels induced by TMA were dramatically reduced by topical treatment of MQL compared with oral administration. These results indicate that MQL may be effective as an oral or topical drug depending on the internal or external causes of atopic dermatitis. Further anti-proliferative effects of MQL were seen in allergic mouse models as Th2 (IL-13) and Th1 (IFN-γ) cytokines decreased, but not IL-12 (cytokine derived from APC). Fig. 6 shows that IFN-γ is suppressed by MQL topical treatment, but not IL-12, while oral administration of MQL did not affect IFN-γ production but increased IL-12 production. More evidently, IFN-γ/IL-12 ratios showed that MQL treatment decreases IFN-γ production compared with the TMA alone group as follows: Ratio of IFN-γ/IL-12 in the naïve group (oral, 1.3; topical, 1.7), TMA-induced AD group (oral, 2.4; topical, 8.9), MQL 4 group (oral, 2.3; topical, 6.5), and MQL 10 group (oral, 2.2; topical, 5.3). Therefore, we believe that MQL has the potential to be developed as a therapeutic agent for AD.
In conclusion, we demonstrated that MQL suppressed Th2-related immune responses by reducing CD4+ T cell proliferation via up-regulation of HO-1 expression. Mechanistically, MQL increased HO-1 expression via activation of the Raf-ERK-Nrf2 pathway and by generating ROS in CD4+ T cells. Furthermore, we verified that these effects of MQL on CD4+ T cells leads to alleviated allergic diseases such as an atopic dermatitis in vivo . It is possible that MQL may provide clinical benefits on other allergic and CD4+ T cell-mediated diseases. The application MQL on various diseases remains the focus of further studies.