SIRT3 Activator Honokiol Inhibits Th17 Cell Differentiation and Alleviates Colitis

Abstract

Background

Honokiol (HKL), a natural extract of the bark of the magnolia tree and an activator of the mitochondrial protein sirtuin-3 (SIRT3), has been proposed to possess anti-inflammatory effects. This study investigated the inhibitory effects of HKL on T helper (Th) 17 cell differentiation in colitis.

Methods

Serum and biopsies from 20 participants with ulcerative colitis (UC) and 18 healthy volunteers were collected for the test of serum cytokines, flow cytometry analysis (FACS), and relative messenger RNA (mRNA) levels of T cell subsets, as well as the expression of SIRT3 and phosphorylated signal transducer and activator of transcription/retinoic acid-related orphan nuclear receptor γt (p-STAT3/RORγt) signal pathway in colon tissues. In vitro, naïve clusters of differentiation (CD) 4 + T cells isolated from the mouse spleen differentiated to subsets including Th1, Th2, Th17, and regulatory T (Treg) cells. Peripheral blood monocytes (PBMCs) from healthy volunteers were induced to the polarization of Th17 cells. After HKL treatment, changes in T cell subsets, related cytokines, and transcription factors were measured. The dextran sulfate sodium (DSS)-induced colitis and interleukin (IL)-10-deficient mice were intraperitoneally injected with HKL. These experiments were conducted to study the effect of HKL on the development, cytokines, and expression of signaling pathway proteins in colitis.

Results

Patients with UC had higher serum IL-17 and a higher proportion of Th17 differentiation in blood compared with healthy participants; while IL-10 level and the proportion of Treg cells were lower. Higher relative mRNA levels of RORγt and a lower SIRT3 expression in colon tissues were observed. In vitro, HKL had little effect on the differentiation of naïve CD4⁺ T cells to Th1, Th2, or Treg cells, but it downregulated IL-17 levels and the Th17 cell ratio in CD4⁺ T cells from the mouse spleen and human PBMCs under Th17 polarization. Even with a STAT3 activator, HKL still significantly inhibited IL-17 levels. In DSS–induced colitis mice and IL-10 deficient mice treated with HKL, the length of the colon, weight loss, disease activity index, and histopathological scores were improved, IL-17 and IL-21 levels, and the proportion of Th17 cells were decreased. Sirtuin-3 expression was increased, whereas STAT3 phosphorylation and RORγt expression were inhibited in the colon tissue of mice after HKL treatment.

Conclusions

Our study demonstrated that HKL could partially protect against colitis by regulating Th17 differentiation through activating SIRT3, leading to inhibition of the STAT3/RORγt signaling pathway. These results provide new insights into the protective effects of HKL against colitis and may facilitate the research of new drugs for inflammatory bowel disease.

Introduction

Inflammatory bowel disease (IBD) refers to chronic intestinal inflammatory diseases, including ulcerative colitis (UC) and Crohn’s Disease (CD).^1–3 An abnormal immune response is believed to be an essential cause of IBD, and abnormal initial clusters of differentiation (CD) 4⁺ T-cell polarization plays an important role in this process. In particular, the imbalance of T-helper (Th) 17/T-regulatory (Treg) cells is a critical factor for IBD,⁴ and many researchers have confirmed that correcting the disorder of Th17/Treg is an effective therapy for the prevention and treatment of IBD.⁵

Retinoic acid-related orphan nuclear receptor γt (RORγt) is a major transcription factor for developing Th17 cells.⁶ Signal transduction and activator of transcription 3 (STAT3), a member of the STAT family, can directly activate RORγt and induce Th17 cells. Thus, STAT3 is instrumental in regulating the Th17/Treg balance and releasing inflammation-related cytokines.⁷ It has been shown that STAT3 inhibitors can significantly inhibit the expression of RORγt and the differentiation of Th17 cells.⁸ Therefore, STAT3 is a key signal transducer downstream of the inflammatory response.⁹

Honokiol (HKL), one of the main active ingredients of Magnolia officinalis, has anti-inflammatory, anti-oxidative, anti-aging, and anti-tumor properties. It has been shown that HKL acts as a direct agonist for the mitochondrial protein SIRT3,¹⁰ a member of human sirtuins located in the mitochondrial matrix that exhibits nicotinamide adenine dinucleotide+-dependentdeacetylase activity.¹¹ Emerging evidence indicates the crucial role of mitochondrial function in aging and carcinogenesis.^12,13 Honokiol reduces lipopolysaccharide-induced lung inflammatory injury and human pulmonary microvascular endothelial cell apoptosis by regulating the mitochondrial pathway dependent on SIRT3.¹⁴

Although HKL has been shown to protect against inflammation in many diseases,^15,16 whether HKL plays a positive role in IBD remains unclear. Our study focused on inflammatory levels and associated protein expression in UC patients and healthy participants. And then we further clarified the effects of HKL on inflammatory cell models, dextran sulfate sodium (DSS)-induced mice, and interleukin (IL)-10-deficient mice. The potential role of HKL/SIRT3 and STAT3/RORγt signal pathways in this progress were also explored.

Materials and Methods

Subjects

Fresh biopsies of macroscopic mucosal and blood samples were collected from healthy participants (n = 18) and UC patients (n = 20) during colonoscopies at the Digestive Endoscopy Center of the Nanjing Drum Tower Hospital between November 2022 and January 2023. Clinicians diagnosed UC using an established criteria based on clinical, radiographic, endoscopic, and histological examinations. The demographic and clinical features of the healthy controls and patients were extracted from their medical records (Table 1). Both healthy participants and patients signed a consent form. The Ethics Committee of Medical Research of the Nanjing Drum Tower Hospital approved all experiments involving human samples.

Table 1.

The Demographic Data and Clinical Features of the Study Population.

Groups	HC (n=18)	Ulcerative Colitis (n=20)
		Mild UC (n = 7)	Moderate UC (n = 8)	Severe UC (n = 5)
Age (mean ± SD), yr	46.61 ± 2.52	33.29 ± 8.08a	41.38 ± 13.03b	48.80 ± 9.96c
Gender (male:female)	8:10	4:3	5:3	2:3
Extent of disease, n (%)	N/A
Proctitis		2(28.6)	4(50.0)	0(0)
Left-sided colitis		4(57.1)	2(25.0)	2(40.0)
Pancolitis		1(14.3)	2(25.0)	3(60.0)
TMMS, (mean ± SD)	N/A	4.00 ± 0.81	8.00 ± 1.30	11.60 ± 0.54

Disease activity and severity was assessed by the total modified Mayo score (TMMS) including endoscopy, patient-reported outcomes of stool frequency and rectal bleeding, and a physician global assessment. Active mild to severe disease is defined by TMMS (mild UC TMMS 3~5; moderate UC TMMS 6~10; severe UC TMMS 11~12).

Groups were compared using nonparametric analyses.

^a P = .07 compared with HC;

^b P = .80 compared with HC;

^c P= .99 compared with HC.

P < .05 was considered significant.

NA, not applicable.

Experimental Animals

Forty healthy SPF male C57BL/6J mice (8-10 weeks old) were purchased from the Experimental Animal Center of Nanjing Medical University. Twenty male C57BL/6-IL-10 knockout (KO) mice and twenty healthy SPF male C57BL/6J mice (7 weeks old) were purchased from GemPharmtech Co. Ltd.

Forty healthy SPF male C57BL/6J mice were averagely allocated into 4 groups: control, DSS-induced colitis, 2.5 mg/kg of HKL (at a dose of 2.5 mg/kg body weight), and 5 mg/kg of HKL (at a dose of 5 mg/kg body weight) groups. During the first 7 days, mice in the colitis group and 2 HKL groups drank the water with 3% DSS (MP Biomedicals, St. Ana, CA, USA). For the next 3 days, all mice resumed drinking normal water. Throughout the 10 days, the mice in HKL groups was intraperitoneally injected with HKL solution once daily.

Interleukin-10 knockout mice were randomly assigned to 2 groups, each containing 10 mice. Throughout the 10 days, all the mice drank normal water. Meanwhile, the mice in the IL-10 knockout group were intraperitoneally injected with HKL solution at a dose of 5 mg/kg body weight once daily.

All mice were anesthetized on the 11th day. Blood was sampled from the eyes of the mice. After centrifugation, the serum samples were collected for cytokine detection. The spleen was aseptically collected to detect T cell subsets. The length of colon was measured, and tissues were detached for pathological examination, immunofluorescence, and western blotting.

During the 10 days, the disease activity index (DAI) was observed. The grading criteria of DAI showed in Supplementary Table 1. The final DAI score was the average of the 3 sections. The corresponding scores of these parameters were evaluated. Two pathologists blindly assigned the colon pathology scores according to the Neurath Scoring criteria.¹⁷

Cell Isolation, Differentiation

Naïve CD4⁺ T cells were isolated from male C57BL/6J mouse spleens by magnetic beads (STEMCELL, Canada) ex vivo and cultured in medium (RPMI 1640 [CORNING, USA] mixed with 10% fetal calf serum [Biological Industries, Israel], 100 μg/mL of streptomycin and 100 μg/mL of penicillin). Naïve CD4⁺ T cells with a concentration of 2 × 10⁶ cells/mL were stimulated on coated plates (10 μg/mL of anti-CD3 and 4 μg/mL of anti-CD28, eBioscience) and cultured under the following conditions: Th1 cells: IL-12 (10 ng/mL, Novoprotein), anti-IL-4 (2500 ng/mL; eBioscience), and IL-2 (10ng/mL; Peprotech); Th2: IL-4 (20ng/mL, Peprotech), anti-IFN-γ (5000 ng/mL, eBioscience), and IL-2 (10 ng/mL; Peprotech); Th17: TGF-β1 (1 ng/mL, Novoprotein), IL-6 (10 ng/mL, Peprotech), anti-IL-4 (5000 ng/mL, eBioscience), and anti-IFN-γ (5000 ng/mL, eBioscience); Treg: TGF-β1 (1 ng/mL;Novoprotein), IL-2 (10 ng/mL, Peprotech), anti-IL-4 (5000 ng/mL, eBioscience), and anti-IFN-γ (5000 ng/mL; eBioscience). Sirtuin-3 activator HKL (MCE, USA) was soluted in dimethyl sulfoxide (DMSO), with the concentration of 0 (only with 0.1%DMSO), 2 μM, and 5 μM being added into the wells in induced groups. The T cells were cultured with different substances for 5 days.

Isolated PBMCs from healthy participants and UC patients were prepared for flow cytometry analysis (FACS). Some PBMCs from healthy participants were cultured in the medium mentioned previously (10 μg/mL of anti-CD3 and 4 μg/mL of anti-CD28; eBioscience, USA) and cultured under the Th17-polarization conditions: TGF-β (3 ng/mL, Peprotech, USA), IL-6 (40 ng/mL, R&D), IL-23 (30 ng/mL, R&D), TNF-α (20 ng/mL, PeproTech), and IL-1β (10 ng/mL, PeproTech). In the first PBMC experiment, HKL (MCE, USA) with 0 (0.1% DMSO), 2 μM, and 10 μM was added into wells in Th17-induced groups. In the second PBMC experiment, besides 0.1% DMSO, 10 μM of HKL and STAT3 activator Colivelin TFA 18 μM (MCE, USA) were separately or combinedly added into the wells in the Th17-induced groups.

Flow Cytometry Analysis and Enzyme-linked Immunosorbent Assay

Flow cytometry analysis of T cells was performed with routinely 1 × 10⁶ cells per sample. For Th1, Th2, and Th17, human PBMCs were labeled with FITC-CD4, PE-IFN-γ, APC-IL-4, and BV421-IL-17A, for Treg, FITC-CD4, APC-CD25, and PE-Foxp3. For Th1, Th2 and Th17, in vitro induced mouse CD4⁺ T cells were labeled with FITC-CD4, PerPE-INF-γ, APC-IL-4, and PE-IL-17A. Tregs were labeled with FITC-CD4, APC-CD25, and PE-Foxp3. Dead cells were labeled with the Fixable Viability Dye eFluor 506. In vitro-induced human PBMCs were labeled with PerCP-CD8 and APC-IL-17A to identify Th17 cells. Apoptosis was also detected via FACS using an Annexin V FITC Apoptosis Detection Kit. Mouse spleenocytes were labeled with FITC-CD4, PerCP-IFN-γ, APC-IL-4, PE-IL-17A, BV605-CD25, and BV421-Foxp3. Before FACS, a cell activation cocktail (Biolegend, USA) was added to the wells for Th1, Th2, and Th17 differentiation and continued to culture cells at 37°C for 6 hours in a 5% CO₂ culture incubator. All antibodies used for fluorescence-activated FACS were purchased from eBioscience.

Cytokines (IFN-γ, IL-4, IL-17A, IL-10, and IL-21) were detected using a commercial enzyme-linked immunosorbent assay kit (eBioscience). All steps followed the instructions.

Immunofluorescence

The 1-cm mouse colon tissue was placed on a cooling table, embedded with optimal cutting temperature compound, and was completely solidified. Slices with a thickness of 8 μm were cut and placed on slides. The intestinal tissue sections were sealed, hybridized with primary antibody, hybridized with secondary antibody, stained with nuclear antibody, dropped with a little anti-quench agent, immediately observed under a fluorescence microscope, photographed, and saved. Antibodies included SIRT3 (2627s, CST), STAT3 (4904s, CST), p-STAT3 (Tyr705, 9145s, CST), ROR-γ (ab207082, Abcam), β-actin (Bioworld, St. Louis Park, MN, USA), horseradish peroxidase (HRP) conjugated antimouse antibodies (7076, CST), HRP-conjugated antirabbit antibodies (7074s, CST), and donkey antigoat IgG-HRP (sc-2020, Santa Cruz, Dallas, TX, USA).

Western Blot Analysis

Human or mouse colon tissue or cell sediment was added to an appropriate amount of Radio Immunoprecipitation Assay (RIPA) strong lysis buffer (Biosharp, Hefei, China) and placed on ice for 30 minutes. After thorough grinding, the supernatant was centrifuged at four 4°C at 13800 g for 15 minutes to collect. The bicinchoninicacid (BCA) protein concentration test instructions determined the number of prepared cells in each group. The cells were incubated for 30 minutes. The sample buffer was reduced 5 times and then heated at 100°C for 5 minutes to denature the protein. After electrophoresis, wet transfer, sealing, primary antibody incubation, secondary antibody incubation, film exposure, fixing, etc., film scanning results were analyzed. The proteins were visualized with ECL western blotting reagents (Millipore). The antibodies used were the same as those used for immunofluorescence analysis. Quantitative protein expression analysis was performed using ImageJ 1.50i software (National Institutes of Health, USA).

Quantitative Real-time Polymerase Chain Reaction

Total RNA was extracted with TRIzol reagent (Invitrogen, Carlsbad, CA, USA) from cells or colon tissues. The cDNA was synthesized from 1 μg of total RNA using a reverse transcription kit (Takara Bio Inc., Japan). The PrimeScript RT Master Mix Perfect Real-Time Kit (Takara Bio Inc., Japan) was used for real-time polymerase chain reaction (PCR), and reactions were performed using an ABI PRISM 7300 Sequence Detection System (Applied Biosystems, USA). The polymerase chain reaction was performed as follows: denaturation for 10 minutes at 95°C, followed by 40 cycles at 95°C for 15 seconds and 60°C for 1 minute. The relative expression of mRNA was analyzed using the 2^−ΔΔCt method. The primer sequences for mouse and human transcription factors and cytokines associated with Th1, Th2,Th17, or Treg cells are listed in Supplementary Tables 2 and 3. The primer sequences of human transcription factors of SIRT3, STAT3, and RORγt are listed in Supplementary Table 2.

Statistical Analysis

Data analyses were performed using GraphPad Prism (version 8.0; La Jolla, CA, USA). Student’s t test was used for comparisons between 2 groups, and 1-way analysis of variance (ANOVA) was used for assessing differences between all experimental groups. All data are presented as mean ± SEM. A value of P < .05 was considered statistically significant. In all figures, “*” represents P < .05, whereas “**” represents P < .01.

Results

Th17/Treg Imbalance and Lower Colonic SIRT3 Expression in Patients With UC

An imbalance in T cell subsets is one of the important reasons in the progression of IBD.¹⁸ Different T cells produce different cytokines that cause differences in function, and each subgroup possesses lineage-specific transcription factors.¹⁹ We first compared serum relative cytokines, Th1, Th2, Th17, and Treg proportions, and transcription levels in patients and healthy controls to identify T cell differentiation. Consistent with previous reports,²⁰ increased IL-17A-producing or IL-17A-expressing Th17 cells and decreased IL-10-producing or Foxp3-expressing Tregs were observed in UC patients compared with healthy participants (Figure 1A, 1B). In patients’ tissues, Th17-related RORγt expression increased, whereas Treg-related FoxP3 mRNA expression decreased (Figure 1C). The difference in the Th1 subset was not very significant, especially serum IFN-γ level and the expression of Tbx21 in the colon. A decreased IL-4-producing and increased IL-4-expressing Th2 cells were found in patients with UC, but Th2-related Gata3 mRNA expression was raised in the colon (Supplementary Figure 1A-1C).

Figure 1.

The percentage of Th17 and Treg cells, cells-related cytokines, and transcription factors in healthy controls and UC patients. A, ELISA measured serum levels of IL-17A and IL-10 in healthy controls and patients with UC. B, Flow cytometry figures (left) and statistical analysis (right) of Th17 and Treg cells in human PMBCs in 2 groups. C, Relative mRNA expression of RORγt, ΙL-17A, and Foxp3 in human colon tissues by qPCR. D, Relative mRNA expression of SIRT3 in colons. Ε, The protein expression of SIRT3, STAT3, p-STAT3 and RORγt in colons. Data are expressed as mean ± SEM. Asterisks indicate significant differences (*P < .05; **P < .01). Abbreviations: HC, Healthy controls (n = 18)；UC, ulcerative colitis (n = 20)；HC-1, HC-2, HC-3, healthy controls (n = 6)；UC-Mi, patients with mild UC (n = 7)；UC-Mo, patients with moderate UC (n = 8)；UC-S, patients with severe UC (n = 5).

As SIRT3 is downregulated in IBD, we measured SIRT3 transcriptional levels and protein expression in colonic tissues. Both SIRT3 transcription and protein levels in UC patients were down-regulated compared with healthy volunteers (Figure 1D, 1E).

Furthermore, in colonic tissues of UC patients, SIRT3 expression was lower in the moderate group than that in the mild group, and it was the lowest in the severe group. Protein expression levels of STAT3, p-STAT3, and RORγt were consistent and opposite to SIRT3 in each group (Figure 1E).

HKL Inhibited Th17 Differentiation in CD4⁺ T cells in Vitro

Honokiol is a direct agonist of SIRT3 and has anti-inflammatory function.¹² To determine whether HKL regulates T cell differentiation, we treated CD4⁺T cells under different T subsets’ polarizing conditions²¹ in the presence or absence of HKL with 2 doses. The results showed that HKL had no significant effect on the differentiation of Th1 and Th2 cells or the levels of the corresponding transcription factors (Supplementary Figure 2A, 2B). Under Th17-polarizing conditions, HKL had the strongest inhibition of Th17 and Th17 transcription factor RORγt in cell culture (Figure 2A), whereas it did not increase the dead ratio of cells (Figure 2B). Furthermore, under Treg-polarizing conditions, only the transcription factor Foxp3 increased (Figure 2C). The protein expressions of the pathway showed that SIRT3 protein expression was downregulated under Th17-polarizing conditions but upregulated in the presence of HKL. Oppositely, protein expressions of STAT3, p-STAT3, and RORγt were upregulated under Th17-polarizing conditions but downregulated in the presence of HKL in a dose-dependent manner (Figure 2D).

Figure 2.

The differentiations of Th17 and Treg cells in CD4⁺T cells under conditions of corresponding polarization in vitro. A, Flow cytometry figures (up) and statistical analysis (down) of Th17 and Treg cells in CD4⁺T cells in 5 groups. B, Flow cytometry figures (left), cells colonies photos and statistical analysis (right) of the live cells in 5 groups. C, Relative mRNA expression of RORγt and Foxp3 in CD4⁺T cells by qPCR. D, The protein expression of SIRT3, STAT3, p-STAT3 and RORγt. Data are expressed as mean ± SEM (n = 6). Asterisks indicate significant differences (*P < .05; **P < .01). Ctrl-, Control group without polarization; Ctrl+, Control group with polarization; DMSO, The medium group of HKL; HKL 2μM, Honokiol group with the dose of 2 μM; HKL 5 μM, Honokiol group with the dose of 5 μM.

HKL Inhibited IL-17 Production and the Differentiation of Th17 in Human PBMCs in Vitro

To determine whether HKL regulates Thl7 cell differentiation in human PBMCs, we treated PBMCs with Th17-polarization conditions in the presence or absence of HKL. Since no remarkable difference was observed in the results of CD4⁺ T cells in vitro with the doses of 5 μM and 2 μM of HKL, we increased the dose of 5 μM to 10 μM in human PBMCs. Honokiol treatment decreased the proportion of Th17 (Figure 3A) in a dose-dependent manner. Honokiol treatment did not increase T-cell apoptosis (Figure 3B). Consistent with these results, HKL downregulated IL-17A in the supernatant and at the mRNA levels. Simultaneously, SIRT3 mRNA expression was upregulated (Figure 3C).

Figure 3.

The differentiations of Th17 cells in PBMCs under the condition of Th17-polarization in vitro. A, Flow cytometry figures (left) and statistical analysis (right) of Th17 in PBMCs in four groups. B, Flow cytometry figures (left) and statistical analysis (right) of apoptosis in four groups. C, The concentration of IL-17A in supernate by ELISA and the relative mRNA expression of IL-17A and SIRT3 in PBMCs by qPCR. D, The concentration of IL-17A in supernate and relative mRNA expression of IL-17A in PBMCs added with TFA. Data are expressed as mean ± SEM (n = 6). Asterisks indicate significant differences (*P < .05; **P < .01). Ctrl, Control group without Τh17-polarization; DMSO, The medium group of HKL with Τh17-polarization; HKL, Honikiol group with the dose of 5 μM; TFA, STAT3 activator Colivelin TFA group with the dose of 18 μM; HKL + TFA, Honikiol + Colivelin TFA group.

Colivelin TFA, an effective activator of STAT3, dramatically increased p-STAT3 protein expression in BV-2 cells.²² In this part, PBMCs were added with equal volume DMSO, 10 μM of HKL, STAT3 activator Colivelin TFA 18μM, or HKL and Colivelin TFA, respectively, under the condition of Th17 cell differentiation for 5 days of culture. Both the concentration of IL-17A in supernate and the relative mRNA expression in PBMCs were downregulated (Figure 3D). Their levels remained low, even when Colivelin TFA activated STAT3.

HKL Ameliorated DSS-induced Colitis in Mice by Activating SIRT3 and Inhibiting STAT3/RORγt

The DSS colitis model is one of the most commonly used mouse models for experimentally induced colitis.²³ All groups except the control group received 3% DSS and the corresponding treatments. Each group’s clinical signs and symptoms were examined (Figure 4A-D). Honokiol inhibited the development of colitis, as evidenced by reduced body weight loss, milder diarrhea, almost no bloody stools, and longer colons than those in untreated mice. A histological analysis was performed (Figure 4E). Compared with untreated mice, HKL-treated mice showed mild inflammatory cell infiltration, significant relief from structural destruction, and relatively low edema levels. Dextran sulfate sodium–induced mice injected with the dose of 5 mg/kg HKL showed a more substantial improvement than those treated with 2.5 mg/kg of HKL (Figure 4E). Honokiol treatment inhibited the Th17-secreting cytokines IL-21 and IL-17 (Figure 4F). Honokiol consistently decreased the percentage of Th17 cells with no effect on the differentiation of Th1 or Th2 (Figure 4G; Supplementary Figure 3A, 3B). Conversely, HKL mildly increased the ratio of Treg cells in the spleen of DSS-induced mice (Figure 4G). The tendency of serum cytokine and transcription factor levels in colon tissues was similar to that of the differentiation of each T cell subset (Figure 4F, 4G; Supplementary Figure 3A, 3B)

Figure 4.

Honokiol exerts anti-inflammatory effects in mice of DSS-induced colitis. A, body weight changes. B, DAI score. C, colon figures. D, colon length. E, Representative images of mouse colonic mucosa (left) and colon Neurath scores were evaluated (right). Upper to lower panel magnifications are ×100, ×200, and ×400, respectively. F, Τhe concentration of IL-17A and IL-21 in mouse serum. G, Flow cytometry figures (left) and statistical analysis (right) of Th17 and Treg cells in each group of the mouse spleens. H, Immunofluorescence of SIRT3, STAT3, p-STAT3, RORγt in each group of the mouse spleens. I, Protein expression figures (left) and statistical analysis (right) of SIRT3, STAT3, p-STAT3, and RORγt by western blot in each group of the mouse colons. Data are expressed as mean ± SEM (n = 10). Asterisks indicate significant differences (*P < .05; **P < .01). Abbreviations: Ctrl, control group; DSS, mice group of DSS-induced colitis; HKL 5 mg/kg, Honikiol group with the dose of 5 mg/kg; HKL 2.5mg/kg, Honikiol group with the dose of 2.5 mg/kg.

Retinoic acid-related orphan nuclear receptor γt is a master transcription factor for the differentiation and development of Th17 cells.⁸ We then investigated whether HKL could inhibit RORγt to suppress Th17 cells. We checked SIRT3, STAT3, p-STAT3, and RORγt expression by immunofluorescence and western blot in the mice treated with or without HKL upon DSS insults. As shown in Figure 4H and 4I, HKL treatment inhibited RORγt expression. STAT3 is critical in the induction of RORγt. To determine whether HKL inhibits RORγt by inhibiting STAT3 activation, we also measured STAT3 activation, which decreased in HKL-treated mice. Since HKL is an activator of SIRT3, which directly deacetylates STAT3 to inhibit its activity, we confirmed that HKL treatment activated SIRT3. The trend of immunofluorescence intensity of SIRT3, STAT3, p-STAT3, and RORγt was consistent with the trend of protein expression measured by western blot.

HKL Ameliorated Colitis in IL-10 Knockout Mice

Interleukin-10-deficient mice spontaneously develop intestinal inflammation, which has many similarities to IBD.²⁴ In this part, mice in the IL-10 KO + HKL group received the treatment at 5 mg/kg of HKL. Each group’s clinical signs and symptoms were examined (Figure 5A–D). The HKL group showed less body weight loss, milder diarrhea, almost no bloody stools, and a longer colon than untreated mice. A histological analysis was performed (Figure 5E). Compared with untreated mice, although the colonic mucosal structure of IL-10 KO mice was relatively intact, inflammatory cell infiltration was prominent. Honokiol-treated mice showed mild inflammatory cell infiltration with normal colonic mucosal structure. The IL-10 KO mice treated with HKL exhibited reduced serum levels of IL-21 and IL-17 (Figure 5F). The ratio of Th17 cells in the IL-10 KO mice treated with HKL was lower (Figure 5G). Lastly, we measured pathway protein expressions in colonic tissue from mice. Interestingly, in 2 groups of wild-type mice, SIRT3 protein expression was obviously higher, and p-STAT3 protein expression was extremely lower in the presence of HKL than in absence of HKL. Meanwhile, in 2 groups of IL-10 KO mice, SIRT3 protein expression was a little higher in the IL-10 KO + HKL group. Particularly, protein expressions of STAT3, p-STAT3, and RORγt were downregulated in the IL-10 KO + HKL group compared with the IL-10 KO group (Figure 5H).

Figure 5.

Honokiol exerts anti-inflammatory effects by reducing Th17 cells in IL-10 knockout mice. A, Body weight changes. B, DAI score. C, colon figures. D, colon length. E, Representative images of mouse colonic mucosa (left) and colon Neurath scores were evaluated (right). Upper to lower panel magnifications are ×100, ×200, and ×400, respectively. F, Τhe concentration of IL-17A and IL-21 in mouse serum. G, Flow cytometry figures (left) and statistical analysis (right) of Th17 cells in each group of the mouse spleens. H, Protein expression of SIRT3, STAT3, p-STAT3 and RORγt by western blot in each group of the mouse colons. Data are expressed as mean ± SEM (n = 10). Asterisks indicate significant differences (*P < .05; **P < .01). Abbreviations: Ctrl, control group; IL-10 KO, IL-10 knockout mice group; HKL, Honikiol group with the dose of 5 mg/kg.

Discussion

However, the pathogenesis of IBD is complex and multifactorial. In recent years, it has been reported that the intestinal mucosal immune abnormality plays a leading role in the occurrence, development, and prognosis of IBD.^25,26 Previous studies have confirmed the influence of the imbalance in Th17 and Treg cells.^27,28 In the first part of the study, we also found that UC patients had a higher ratio of Th17 cells and a lower ratio of Tregs in PBMCs than healthy controls, and the corresponding levels of cytokines were consistent. The transcription factor RORγt of Th17 cells in the colon tissue of patients with UC was upregulated. Foxp3, the transcription factor of Treg cells, showed the opposite trend in colon tissue. This phenomenon is thought to be caused by the transfer of Tregs to the colon tissue as anti-inflammatory cells. The Th1/Th2 cells are also shown to be associated with IBD, but our result showed that IFN-γ-producing Th1 and cytokine levels are inconsistent with transcription factors in colon tissue, similar to IL-4-producing Th2 cells. In colon tissues of UC patients, we also found that SIRT3 protein expression was gradually reduced in the mild, moderate, and severe groups. The protein expression tendencies of STAT3, p-STAT3, and RORγt were opposite to SIRT3. These results piqued our interest in further study of the relationship among SIRT3, STAT3/RORγt, and T cells differentiation.

Sirtuin-3, the major mitochondrial deacetylase, has been proven to be involved in whole mitochondrial metabolism. Sirtuin-3 disorders also induce a variety of various diseases. Human and mouse SIRT3 mRNA is widely expressed in multiple tissues and organs.²⁹ We found the expressions of SIRT3 gene mRNA and protein levels were lower in the colon tissues of patients with UC, which prompted us to investigate whether lower expression of SIRT3 is associated with an imbalance of Th17/Treg cells.

Honokiol, a major active compound isolated from the bark, roots, and stems of traditional herbs and various Magnolia species, is a SIRT3 activator.^29,30 These plants have been used against multiple inflammatory and neuronal diseases,^31,32 and HKL has shown to have various beneficial effects, including neuroprotective, anticancer, and anti-inflammatory activities.^33,34 In this study, we demonstrated that HKL inhibited the differentiation of Th17 cells and mildly upregulated the differentiation of Treg cells in mouse CD4⁺ T cells and human PBMCs under different T cell polarization conditions with no toxic effects.

Especially in the case of HKL, both downregulated differentiation in Th17 cells and the transcription factor RORγt in human PBMCs in a dose-dependent manner were observed. However, no effect was observed on the differentiation of Th1 and Th2 cells.

Interestingly, HKL treatment significantly improved weight loss in mice with DSS-induced colitis in vivo. The DAI and endoscopic scores also improved significantly. Furthermore, consistent with other research,³⁵ Th17 cells were decreased, and the levels of pro-inflammatory factors IL-21 and IL-17 in the serum were also reduced considerably. Collectively, these results suggest that HKL improves colitis by inhibiting Th17 cells. Honokiol did not affect Th1 or Th2 cell differentiation in mice with DSS-induced colitis. In contrast, the inhibitory effect on Th17 cells was more evident than the stimulatory effect on Treg cells, similar to the results observed for CD4⁺ T cells in vitro. It is possible that HKL can also affect the differentiation of Tregs during anti-inflammation, which will be further studied in the future. To confirm the effect of HKL on Th17, IL-10 KO mice were analyzed further. The knockout of IL-10 in mice is a well-established model of colitis.³⁶ Deficiency of the downregulatory cytokine IL-10 results in chronic intestinal inflammation. Tregs in IL-10 KO mice did not produce IL-10; therefore, the effect of HKL on Tregs could be excluded in this model of colitis. This was consistent with our hypothesis that HKL would ameliorate IL-10 KO mice colitis. These results indicate that HKL improved colitis by inhibiting Th17 cells.

How does HKL activate SIRT3 to regulate Th17 cell differentiation? A recent study found that HKL plays a protective role in renal ischemia-reperfusion injury by suppressing oxidative stress, inducible nitric oxide synthase, inflammation, and STAT3 expression in rats.^37,38 Sirtuin-3 is thought to act upstream of STAT3 in acute coronary artery syndrome.³⁹ It has been shown that SIRT3 directly binds to and deacetylates STAT3 to inhibit its activity in the research of angiotensin II-induced cardiac fibrosis.⁴⁰ Retinoic acid-related orphan nuclear receptor γt is a key transcription factor that regulates Th17 cell differentiation and is governed by STAT3.⁴¹ Thus, we speculated that HKL might also inhibit STAT3 phosphorylation and the expression of RORγt by upregulating SIRT3 in vitro and in vivo. Honokiol inhibited Th17 cell differentiation and alleviated the inflammatory response both in CD4⁺ T cells and human PBMCs in vitro, and the same results were shown in 2 mouse models of colitis. Honokiol activated SIRT3; meanwhile, total STAT3, phosphorylated STAT3, and RORγt decreased significantly. This indicated that activation of SIRT3 may have some negative correlation with STAT3, p-STAT3, and RORγt levels.

In summary, our study demonstrated that HKL could inhibit colitis by regulating Th17 cell differentiation, mediated by activating SIRT3 and inhibiting STAT3 activation and RORγt expression. The change of the Th17 cell ratio affects cytokines, IL-17, and IL-21, ameliorating colitis lesions (Figure 6). Thereby, this study provides a novel target for the development of therapeutics in treating IBD.

Figure 6.

Schematic model of HKL’s anti-inflammatory effects on colitis in IBD. Honokiol activates SIRT3. The activation of SIRT3 leads to the downregulation of the phosphorylation of STAT3, which inducts RORγt expression, thus reducing inflammatory Th17 cell activation. The change of the Th17 cell ratio affects cytokines, IL-17, and IL-21, ameliorating colitis lesions.

Author Contributions

X.C., Z.G., Y.L.: Design and conduction of the study, patients’ recruitment, data collection, and writing up of the first draft of the paper.

T.Y., C.P., L.Z., X.L., D.Z.: Support of animal and cell experiments, analysis of the data.

F.Z., X.Z., M.Z., C.Y.: Study design, scientific advice, supervision, and drafting of the manuscript.

Funding

This work was supported by funding from the National Natural Science Foundation of China (No.81970487，No.82170551).

Conflicts of Interest

The authors have no conflicts of interest to declare.

Data Availability

Please contact the corresponding author (Chenggong Yu, chenggong_yu@nju.edu.cn) with a reasonable request.