Therapeutic Effect of HDAC5 Binding and Cell Penetrating Peptide for the Treatment of Inflammatory Bowel Disease

Abstract

Background:

Inflammatory bowel disease (IBD) is an incurable disease that negatively influences the quality of life of patients. Current and emerging therapies target proinflammatory cytokines and/or receptors to downregulate proinflammatory responses, but insufficient remission requires other therapeutic agents. Herein, we report that the synthetic anti-inflammatory peptide 15 (SAP15) is capable of cell penetration and anti-inflammatory activity in human macrophages.

Methods:

SAP15 was labeled with fluorescence and administered to human leukemia monocytic cells (THP-1) cells for cell penetration analysis. Using biolayer interferometry analysis, the binding affinity of SAP15 with histone deacetylase 5 (HDAC5) was measured. SAP15-treated THP-1 cells were analyzed by protein phosphorylation assay, flow cytometry, and enzyme-linked immunosorbent assay (ELISA). In addition, in vivo analysis of the therapeutic effect on IBD was observed in a dextran sulfate sodium (DSS)-induced model. Samples from SAP15-treated mice were analyzed at both the macroscopic and microscopic levels using ELISA, myeloperoxidase (MPO) assays, and histological evaluations.

Results:

SAP15 was internalized within the cytosol and nucleus of THP-1 cells and bound to the HDAC5 protein. SAP15-treated macrophages were assessed for protein phosphorylation and showed inhibited phosphorylation of HDAC5 and other immune-related proteins, which led to increased M2-like macrophage markers and decreased M1-like macrophage markers and tumor necrosis factor-α and interleukin-6 cytokine levels. The SAP15 treatment on IBD model showed significant recovery of colon length. Further histological analysis of colon demonstrated the therapeutic effect of SAP15 on mucosal layer. Moreover, proinflammatory cytokine levels and MPO activity from the plasma show that SAP15 is effective in reduced proinflammatory responses.

Conclusion:

These findings suggest that SAP15 is a novel peptide with a novel cell-penetrating peptide with anti-inflammatory property that can be used as a therapeutic agent for IBD and other inflammatory diseases.

Introduction

Inflammatory bowel disease (IBD) is an idiopathic disorder with chronic inflammation in the gastrointestinal tract. Ulcerative colitis (UC) and Crohn’s disease (CD) are the common types of IBD, and their incidence has increased greatly due to wide industrialization [1, 2]. Urbanization and industrialization of developing countries and people moving to the industrialized western countries alter many aspects of the environments, such as increased population, pollution and wastes, diet changes [3, 4]. These factors may affect composition and function of gut microbiota and intestinal homeostasis, which in effect, increase the IBD occurrences [5]. Patients diagnosed with IBD may have symptoms such as weight loss, bloody diarrhea, and abdominal pain. Various pathological characteristics, such as a reduction in the mucus layer, cell and microbiota infiltration in the submucosa, and an increased proinflammatory response, have been used to understand IBD mechanisms and identify potential targets, but no therapeutic agents have achieved complete remission [6]. Thus, the search for new IBD therapeutics is important for increasing quality of life.

Current IBD therapies range from anti-inflammatory drugs (5-aminosalicylic acids, corticosteroids, non-steroidal anti-inflammatory drugs) [7–9] to immunosuppressive drugs (azathioprine, cyclosporine, 6-mercaptopurine) [10–12] and biologics (infliximab, adalimumab) [13–16]. In particular, biological therapies have been extensively developed to achieve complete remission of IBD. Anti-tumor necrosis factor (TNF)-α, anti-interferon gamma (IFNγ), anti-interleukin (IL)-17A, and anti-IL-6 agents are current and developing therapeutics that block proinflammatory cytokines and/or inflammatory receptors, downregulating the proinflammatory response [17]. Because one-third of IBD patients are not responsive to anti-TNF agents, new therapies with different approaches are needed [18]. Other agents with different targets have been suggested as IBD therapeutics, such as antiadhesion agents, inflammatory cytokine modulators, and antitrafficking molecules [19]. Tofacitinib, a small molecule drug that targets the Janus kinase (JAK) family, and other JAK inhibitors have been promising agents for the clinical treatment of IBD patients [20, 21]. Although the Food and Drug Administration (FDA) has approved tofacitinib for UC therapy, targeting JAK has limitations of off-target specificity and long-term efficacy. Tofacitinib and other small molecules that target JAK are pan-JAK inhibitors with nonspecific selectivity for isoforms of JAKs [22]. In addition, JAK inhibition causes a reduction in effector cytokines that impair epithelial barrier function [23], which could be the possible reason for the ineffective response in patients with CD [24, 25]. Other therapeutic candidates, histone deacetylase inhibitors (HDACIs), are also considered as anti-inflammatory agents that targets HDAC and inhibit inflammatory mediator production in various immune- and inflammatory-related diseases, such as rheumatoid arthritis, IBD, and multiple sclerosis [26, 27].

The development of peptide-based drugs as novel therapeutics has gained attention in regenerative medicine and pharmaceutical research. Peptides, which have relatively lower immunogenicity than other biologics, exhibit key biological properties of binding selectivity and initiate specific signal transduction processes [28]. Various strategies to design and modify peptides have been applied in various targets [29, 30]. In IBD, peptides have also been shown to be therapeutic candidates by modulating inflammatory activities [31, 32]. We have also presented various peptides as therapeutic candidates in regenerative medicine [33–36]. Our previous study showed that SAP15, a 15-mer peptide derived from human β-defensin 3, has an anti-inflammatory effect on murine macrophages and exerts a therapeutic effect on a collagen-induced arthritis rat model [37]. Previous work identified that SAP15 peptide binds and interacts to HDAC5 protein in murine macrophage and inhibits proinflammatory activities. However, specific activity of SAP15 with HDAC5 proteins in human macrophages are not clearly understood. Additionally, the therapeutic potential of SAP15 was further applied in other chronic diseases such IBD.

In this study, the anti-inflammatory effects of the SAP15 peptide on human macrophage polarization and cytokine changes were investigated, specifically targeting to HDAC5 protein and inhibiting HDAC5 phosphorylation. Furthermore, the therapeutic effect of SAP15 on IBD was investigated in a dextran sulfate sodium (DSS)-induced mouse model, along with SAP15 localization to colon tissue.

Materials and methods

Peptide preparation

SAP15 was synthesized using Fmoc chemistry and solid phase peptide synthesis as previously described [38, 39]. The synthesized peptides were purified by reverse-phase high-performance liquid chromatography using a Vydac C18 column and a gradient of water/acetonitrile containing 0.1% TFA. The purity of the peptides was greater than 98%. For the SAP15 binding assay, biotin was manually conjugated to the SAP15 peptide by reacting the peptide with D(+)-biotin, hydroxybenzotriazole-anhydrous, and N,N′-diisopropylcarbodiimide (Daejung, Siheung, Korea) at an equivalent ratio of 3:4:4 for 24 h. For the biodistribution study of SAP15, Alexa Fluor 680 NHS Ester (AF680, Thermo Fisher Scientific, Waltham, MA, USA) was manually conjugated to the SAP15 peptide according to the manufacturer’s instructions with modifications. Peptides (10 mg/ml) were dissolved in 30% pyridine (v/v) and reacted with dye at molar ratios of approximately 0.7 for 15 min at 50 °C. The reacted peptides were lyophilized and stored until use.

Binding assay

All binding experiments were performed at 25 °C using the ForteBio Octet (Sartorius, Göttingen, Germany) platform. Streptavidin biosensor tips were first equilibrated for 10 min in phosphate-buffered saline (PBS, pH 7.4). A baseline was measured for 120 s followed by the immobilization of biotin-tagged SAP15 (1 µM) for 120 s, and then a second baseline was measured in PBS for 120 s. For affinity analysis, the diluted HDAC5 proteins were allowed to bind to the immobilized SAP15 to saturation for 120 s. The dissociation of the proteins was monitored for an additional 600 s. The binding curves were corrected by subtracting nonspecific binding, and the affinity constants were calculated by Octet Data Analysis software (Sartorius).

Cell lines and cell culture

Human leukemia monocytic cells (THP-1) were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA) for in vitro experiments. THP-1 cells were cultured in RPMI 1640 medium supplemented with 10% FBS, 1% antibiotic antimycotic solution, 1% GlutaMAX, and 0.1% beta-mercaptoethanol (Gibco, Grand Island, NY, USA). Cells were maintained at a cell density between 1 × 10⁵ and 1 × 10⁶ cells/ml in a humidified 37 °C incubator with 5% CO₂. For macrophage polarization, THP-1 monocytes (2 × 10⁵ cells/ml) were first differentiated into macrophages with 20 ng/ml phorbol 12-myristate-12 acetate (PMA, Sigma–Aldrich, St. Louis, MO, USA) [40] for 24 h, followed by 24 h of incubation in growth medium. Macrophages were then polarized to M1 macrophages by 100 ng/ml lipopolysaccharide (LPS, Sigma–Aldrich). SAP15 or interleukin-4 (IL-4, R&D Systems, Minneapolis, MN, USA) was administered with LPS for 48 h, and the efficacy was analyzed. For the therapeutic study, SAP15 or IL-4 were administered after LPS treatment for 24 h.

In vitro cellular internalization

THP-1 cells (1 × 10⁵ cells/ml) were seeded on glass chamber slides (Thermo Fisher Scientific). After 24 h of incubation with PMA, fresh complete medium containing 50 μM AF680-labeled SAP15 was added and incubated for 10 min at 37 °C in a humidified chamber with 5% CO₂. After the cells were washed with Dulbecco’s phosphate-buffered saline (DPBS), 4% paraformaldehyde was added and incubated for 30 min for fixation. Fixed cells were washed with DPBS and permeabilized with 0.2% Triton X-100 in DPBS. After being blocked with 2% BSA in DPBS for 1 h at room temperature, samples were stained with 4′,6-diamidino-2-phenylindole (DAPI, Thermo Fisher Scientific) and Alexa Fluor 546-labeled phalloidin. Sample slides were mounted and analyzed on a Carl Zeiss LSM700 confocal laser-scanning microscope controlled by ZENblack software (Carl Zeiss, Oberkochen, Baden-Württemberg, Germany).

Phosphorylation array

To screen phosphorylated proteins in macrophages, differentiated THP-1 cells were cultured with LPS alone or LPS plus SAP15 for 24 h. Cells were lysed with RIPA lysis buffer (Thermo Fisher Scientific) containing Halt protease inhibitor cocktail (Thermo Fisher Scientific) and a phosphatase inhibitor cocktail (Sigma–Aldrich) for 30 min, and proteins were isolated by centrifugation. The protein concentration was quantified using a bicinchoninic acid assay kit (Thermo Fisher Scientific). Fifty micrograms of protein sample was analyzed with the NF-κB signaling phospho-specific antibody array (Full Moon Biosystems, Sunnyvale, CA, USA). Slides were scanned using a GenePix 4100A scanner (Molecular Devices, San Jose, CA, USA), and each intensity was quantified by GenePix 7.0 Software (Molecular Devices). The phosphorylation ratio was calculated using the following equation: phosphorylation ratio = (phosphorylated experimental/ unphosphorylated experimental)/(phosphorylated control/unphosphorylated control). Protein information was annotated using UniProt DB. The results were evaluated using ExDEGA 2.0 (eBiogen, Seoul, Korea).

Flow cytometry

Cell-specific antigens on macrophages were evaluated by flow cytometry. Differentiated macrophages were treated with LPS and/or peptide for 48 h at 37 °C, dissociated by TrypLE™ Express (Gibco), washed with DPBS, and resuspended in 5% BSA at 4 °C. Cells were stained with PE anti-human CD80 and APC anti-human CD206 (Biolegend, San Diego, CA, USA) antibodies for 30 min at 4 °C. After being washed with DPBS, the stained cells were analyzed by flow cytometry with a FACSCalibur (BD Biosciences, Franklin Lakes, NJ, USA).

DSS-induced IBD mouse model

The mouse study was conducted according to protocols approved by the Seoul National University Institutional Animal Care and Use Committee (IACUC) guidelines for the care and use of laboratory animals (Approval number # SNU-190121-1). All animals in this experiment were ICR mice aged 6–8 weeks that were obtained from Orient Bio Inc. (Seongnam, Korea) and maintained in the Division of Laboratory Animal Resources animal facility throughout the experimental period. DSS-induced IBD mouse model ICR mice were fed 5% (w/v) DSS in their drinking water for 7 days. The animals were randomized into four groups as follows: (a) Control group (Normal); (b) 5% DSS-induced colitis group (Defect); 10 mg/kg anti-TNF-α antibody + DSS group as a positive control (anti-TNF-α); and 20 mg/kg SAP15 + DSS group (SAP15). Drugs were intraperitoneal (IP) administered to DSS mice once per day from Day 0. On Day 7, the mice were euthanized, and colons were isolated and assessed for length. Colons and blood were then collected for histology and cytokine assays.

Cytokine assay

Cell culture supernatants were collected after 48 h of LPS and peptide treatment. For the in vivo assay, blood samples were collected in heparin-coated tubes (BD Biosciences, San Jose, CA, USA) and centrifuged to isolate plasma. For the cytokine assays (human TNF-α, human IL-6, mouse TNF-α, mouse IL-6), ELISA kits (R&D Systems) were used according to the procedure recommended by manufacturer. The resulting optical density of each well was measured with a Multiskan microplate spectrophotometer (Thermo Fisher Scientific), and the concentration was quantified in pg/ml.

Colon assessments (MPO, H&E)

Colon tissues were assessed by myeloperoxidase (MPO) assays and histological evaluations. For the MPO assay, colon tissue was prepared by weighing and homogenizing each sample. MPO activity was measured by an MPO colorimetric activity assay kit (Sigma–Aldrich) according to the manufacturer’s instructions. For histological evaluation, each colon was prepared as a Swiss roll and fixed overnight with 4% paraformaldehyde, followed by tissue embedding in paraffin. Then, 5 μm-thick sections were stained with hematoxylin and eosin (H&E) for histopathological imaging. The severity of colitis was scored blindly by 3 individuals based on the scale shown in Table 1.

Table 1.

Severity of colitis grading

Score	Histologic features
0	Normal crypt architecture
1	Crypt shortening, loss of basal one-third crypt, lamina propria prominent with minimal inflammatory changes
2	Crypt shortening, loss of basal two-third crypt, thinning of epithelium with mild inflammatory changes
3	Loss of entire crypt with intact epithelial layer and moderate inflammatory changes
4	Loss of entire crypt along with the epithelial layer (erosion) with severe inflammatory changes

Statistical analysis

Statistical analysis was performed using GraphPad Prism (GraphPad Software, San Diego, CA, USA). All data shown are representative of 3 or more experiments with similar results. Quantitative data are expressed as the mean ± standard deviation. The results were analyzed by one-way analysis of variance (ANOVA), followed by the Tukey’s multiple comparison test. A p value less than 0.05 was considered significant.

Results

Characteristics of the cell-penetrating and HDAC5 binding activity of SAP15

The cell-penetrating property of SAP15 was observed in human macrophages. SAP15 was administered to differentiated THP-1 cells for 10 min and analyzed by confocal microscopy. The results showed that SAP15 penetrated the cells and was located in the cytoplasm and nucleus. Compared to the control peptide, we observed that more SAP15 was internalized in THP-1 cells (Fig. 1A). Then, the colocalization of the cell line with SAP15 was quantitatively measured. The colocalization coefficients of SAP15 with the channels representing the nucleus and cytoplasm showed mean values of 52.45 ± 9.55% and 42.5 ± 15.56%, respectively (Fig. 1B). Almost all SAP15 penetrated THP-1 cells and localized to the nucleus and cytoplasm, verifying its internalization.

Fig. 1.

Cellular localization of AF680-labeled SAP15 and control peptide in THP-1 cells. A AF680 (red) was observed in the SAP15 only group, and the peptide was internalized throughout the cytosol and nucleus. F-actin and nuclei are shown in green and blue, respectively. Scale bar = 20 μm. B Measurement of the colocalization coefficient using ZENblack software. The threshold value was set based on the histogram, and the results are quantified in the graph. The data (mean ± SD) were analyzed by ANOVA, followed by the Tukey’s multiple comparison test. Statistical significance was set at **p < 0.01, ***p < 0.001

To analyze the binding characteristics of SAP15 to HDAC5, biolayer interferometry (BLI) was used to measure the binding kinetics between the peptide and HDAC5 (Fig. 2A). The association rate constant (k_a) of SAP15 was 2.88 × 10⁵  M⁻¹sec⁻¹, which was ten times higher than that of the mismatched peptide (k_a of 1.34 × 10⁴ M⁻¹sec⁻¹). Moreover, the dissociation rate constant (k_d) of SAP15 (1.0 × 10^–7 sec⁻¹) was 100 times lower than that of the mismatched peptide (6.59 × 10^–5 sec⁻¹), suggesting that SAP15 detached from HDAC5 at a much slower rate. Each value was used to calculate the equilibrium dissociation constant (K_D = k_d/k_a), and values of 0.347 pM and 4.9 nM were determined for SAP15 and the mismatched peptide, respectively. The affinity of SAP15 was 10,000-fold higher than that of the mismatched peptide, showing strong binding between SAP15 and HDAC5 and a low dissociation rate.

Fig. 2.

Binding affinity of SAP15 for the HDAC5 protein. A The BLI binding sensorgram of SAP15 for HDAC5 was analyzed by immobilizing biotinylated SAP15 or control peptide on streptavidin biosensor tips and incubating the peptides with a range of HDAC5 protein concentrations. B Each binding kinetic of SAP15 or control peptide on HDAC5 was measured by global fitting and calculation in a 1:1 binding model using Octet Data Analysis software

The SAP15 peptide modulates inflammation-related protein phosphorylation.

Assessment of protein phosphorylation by SAP15 was then performed via phospho-array assays (Fig. 3A). Differentiated THP-1 cells were stimulated with 100 ng/ml LPS to observe changes in phosphorylation in the inflammatory state, and LPS-induced cells were also treated with 100 μM SAP15 (LPS + SAP15) to compare the protein profile between untreated macrophages (NT) and inflamed macrophages. A set of proteins from each group was subjected to a detailed analysis by hierarchical cluster form, and the results are presented as a heatmap (Fig. 3B). The heatmap showing the expression between LPS and NT group showed that the protein expression was inversely correlated, and LPS treatment altered the expression of proteins that promote inflammation. However, the LPS + SAP15 group exhibited a reversal in the protein profiles of macrophages treated with LPS only. Most of the increased proteins induced by LPS were reduced by SAP15 treatment, and vice versa. The dendrogram of the sample tree shows that LPS + SAP15 was more closely related to NT, indicating that SAP15 maintains the imbalance initiated by LPS. The phosphorylation of normalized proteins in the NT, LPS, and LPS + SAP15 groups was directly compared to each other and is represented in a heatmap (Fig. 3C). The level of phosphorylated HDAC5 at serine 498 was increased in the LPS group compared to the NT group (2.657-fold), as were the levels of other proinflammatory proteins, such as NF-κB-p65. In contrast, LPS + SAP15 notably reduced the phosphorylation of HDAC5 to 0.478-fold that of the LPS group. Western blot analysis also confirmed a reduction in the phosphorylation of HDAC5 by SAP15 in LPS-induced macrophages (Fig. 3D).

Fig. 3.

Phosphorylation analysis of SAP15 in THP-1 macrophages. A Representative phospho-array results of the key molecules in signaling pathways for each group. B Normalized protein intensities are visualized as a heatmap and clustered using the maximum distance method. C Heatmap showing the normalized ratio of the phosphorylated to unphosphorylated forms. D Western blot analysis of p-HDAC5 and HDAC5 in THP-1 cells treated with LPS and SAP15 were imaged and quantified. The data (mean ± SD) were analyzed by ANOVA, followed by the Tukey’s multiple comparison test. Statistical significance was set at ns, not significant, **p < 0.01 vs. SAP15⁻/LPS⁻ (negative control)

SAP15 reduces macrophage polarization and cytokine release

THP-1 polarization into proinflammatory M1 macrophages is a widely used model of the inflammatory response [41, 42]. Differentiated THP-1 cells were induced with LPS for 48 h, and surface markers for M1 and M2 macrophages were stained with anti-CD80 and anti-CD206 antibodies, respectively, for flow cytometry (Fig. 4A). IL-4 was administered M2 macrophages. The results showed that surface levels of CD80 were present in the LPS group, whereas the IL-4 group showed only CD206. The effect of SAP15 was analyzed in the context of LPS and showed a decrease in CD80-positive cells and an increase in CD206-positive cells. Quantification of the results also indicated that 100 μM SAP15 reduced CD80⁻positive cells to 25.4% and increased CD206-positive cells to 19.47% compared to those in the LPS-treated group (Fig. 4B, C). The effectiveness of SAP15 can be compared to IL-4 treatment in that IL-4 co-treatment from LPS-treated cells insignificantly decreased M1-like macrophages to 6.47% and increase M2-like macrophages to 3.40%. M1- or M2-like macrophages were also examined by comparing the ratio of M1/M2 macrophages; the ratio in the LPS-treated group was 44, LPS/IL-4 co-treatment was 24.7, and SAP15 lowered it to 3.2 (Fig. 4D). The effect of SAP15 was also assessed by comparing proinflammatory cytokines secreted by macrophages. Increased TNF-α and IL-6 cytokine levels in the LPS group were decreased upon SAP15 treatment (Fig. 4E, F). These results show that LPS-induced proinflammatory effects on human monocytes were inhibited by SAP15, as confirmed by the modulation of macrophage polarization and cytokine release.

Fig. 4.

SAP15 modulates monocyte-derived macrophage polarization. THP-1 macrophages were stimulated with LPS and treated with 50 and 100 μM SAP15. IL-4 (25 ng/ml) was used as a positive control. A CD80 (M1) and CD206 (M2) protein expression on macrophages was analyzed by flow cytometry. The analyzed cells were quantified as B M1-like CD80 + cells, C M2-like CD206 + cells, and D the M1/M2 ratio. SAP15-treated macrophages were also analyzed to determine the secreted levels of the cytokines E TNF-α and F IL-6, quantified in pg/ml. The data (mean ± SD) were analyzed by ANOVA, followed by the Tukey’s multiple comparison test. Statistical significance was set at *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (vs. LPS only)

SAP15 was further evaluated for its therapeutic effect by stimulating macrophages with LPS prior to SAP15 induction. LPS was administered to macrophages for 24 h, and SAP15 or IL-4 was administered for an additional 24 h. Macrophage polarity showed significant changes; SAP15 decreased CD80-positive cells from 96.3 ± 0.32% to 93.5 ± 0.25% and increased CD206-positive cells from 1.36 ± 0.085% to 3 ± 0.52% while IL-4 marginally changed CD80-positive cells and CD206-positive cells to 95.82 ± 0.170 and 1.59 ± 0.156, respectively (Fig. 5A-C). Additionally, the M1/M2 ratio significantly decreased from 71 to 31.7, as SAP15 was effective after LPS treatment (Fig. 5D). Similarly, treatment with SAP15 resulted in low proinflammatory cytokine levels when cells were treated with LPS. TNF-α and IL-6 levels were reduced to 362.1 and 25.56 pg/ml, respectively (Fig. 5E, F).

Fig. 5.

SAP15 modulates monocyte-derived macrophage polarization after LPS induction. THP-1 macrophages were stimulated with LPS for 24 h. Then, the cells were treated with 100 μM SAP15 and IL-4 (25 ng/ml) for another 24 h to analyze the therapeutic effect. A CD80 (M1) and CD206 (M2) protein expression on macrophages was analyzed by flow cytometry. The analyzed cells were quantified as B M1-like CD80 + and C M2-like CD206 + cells and the D M1/M2 ratio. SAP15-treated macrophages were also analyzed to determine the secreted levels of the cytokines E TNF-α and F IL-6, quantified in pg/ml. The data (mean ± SD) were analyzed by ANOVA, followed by the Tukey’s multiple comparison test. Statistical significance was set at **p < 0.01, ***p < 0.001, ****p < 0.0001 (vs. LPS only)

The effects of SAP on DSS-induced colitis mice

To determine the therapeutic effect of SAP15 on IBD, acute DSS-induced colitis mice were used for an in vivo study. While the mice were provided DSS in their drinking water, daily injections of SAP15 and anti-TNF-α were performed. Although the results were insignificant, the body weight of SAP15 group have maintained to 90%, while defect group have reduced to 85% from Day 0 (Fig. 6A). DSS-treated mice showed relatively short colon lengths (66 ± 8.87 mm) compared with those of normal mice (99.83 ± 13.88 mm), and SAP15 (86 ± 6.83 mm) significantly increased the colon lengths of IBD mice compared to those of normal mice (Fig. 6B). H&E staining showed microscopic changes among each group (Fig. 6C). To observe the entire colon, the Swiss roll technique was used for evaluations. Normal mice exhibited intestinal crypts that were properly lined with absorptive cells, goblet cells, and lamina propria. In the defect group, most of the intestinal structures were damaged, especially in the mucosa layer, the epithelial layer was fragmented and crypts were disturbed. Without any crypt structure, goblet cells and the lamina propria were lost, and inflammatory cells infiltrated the mucosa and submucosa layer. On the other hand, SAP15-treated IBD mice maintained colon morphology similar to that of normal mice. Representative images showed clear crypt formation, with distinct goblet cells and lamina propria in the mucosa. Inflammatory cell infiltration appeared to be decreased, while absorptive cells in the epithelial layer were arranged normally. The histological characteristics of the SAP15 group were compared with those of the anti-TNF-α group, which is a currently used medication for IBD, and SAP15 robustly reduced the severity of IBD in the overall colon, while the anti-TNF-α group partially regenerated the colon. Histopathological differences in each group were also measured via severity grading. The normal group and defect group each scored 0 and 10, respectively, and the SAP15 group showed a relative score of 1, which confirmed the healing effect of SAP15. Proinflammatory cytokines in plasma were also measured by ELISA (Fig. 6D, E). The concentration of TNF-α in the defect group was 624.69 ± 166.74 pg/ml, which was a 4.49-fold increase compared with that in the normal group. SAP15 administration reduced TNF-α concentrations to 136.21 ± 25.72 pg/ml, which was lower than the current treatment with anti-TNF-α (176.07 ± 40.74 pg/ml). Similarly, IL-6 levels in serum also showed that DSS-induced IL-6 levels (72.33 ± 46.68 pg/ml) significantly decreased after SAP15 was injected (15.59 ± 11.45 pg/ml) while IL-6 levels in the presence of anti-TNF-α decreased to 28.51 ± 7.86 pg/ml. MPO activity, another important factor indicating the infiltration of neutrophils, was measured in colon tissue (Fig. 6F). The defect group showed an increase in MPO activity (8.92 ± 0.89 units/mg), while the normal group averaged only 2.10 ± 0.9 units/mg activity. MPO activity, however, was significantly decreased in the colons in the SAP15-treated group (3.98 ± 1.9 units/mg), which correlated with the histological analysis. Thus, SAP15 was highly effective in reducing the severity of inflammatory responses in the colon caused by IBD.

Fig. 6.

SAP15 alleviates DSS-induced colitis in mice, as determined by serum levels, MPO activity, and macroscopic and microscopic changes. A. Body weight changes following DSS induction were plotted as the percentage from day 0. B Representative images of mouse colons at 7 days post-exposure to DSS in water and IP administration of SAP15 and anti-TNF-α as a control, and the quantification. C H&E staining and histopathological scoring of each treatment group were analyzed. Scale bar = 100 μm. Serum levels of D TNF-α and E IL-6 and F MPO activities in colon tissue lysates were analyzed. The data (mean ± SD) were analyzed by ANOVA, followed by the Tukey’s multiple comparison test. Statistical significance was set at *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (vs. defect group)

Discussion

The graphical schematic of the effect of SAP15 on IBD is shown in Fig. 7. Phenotypical changes in IBD include a reduced mucosal layer, dysfunction of the epithelial barrier, and infiltration of microbiota into the lamina propria. Endotoxins such as lipopolysaccharide (LPS) from the microbiota stimulate immune reactions by interacting with macrophages to initiate inflammatory responses. The inflammatory response of macrophages occurs by the phosphorylation of HDAC5, which is unable to inhibit histone acetylation associated with inflammatory gene expression. Proinflammatory gene mediate M1 macrophage polarization and inflammatory cytokine production and promote excessive immune reactions. The functions of SAP15 include (1) tissue/cell penetration that transports peptides intracellularly and (2) HDAC5 protein binding, which inhibits HDAC5 phosphorylation. SAP15-bound HDAC5 can bind to histones, and proinflammatory genes cannot be expressed, which will switch macrophage polarization from M1 to M2, facilitating an anti-inflammatory response and reduce proinflammatory cytokine levels.

Fig. 7.

Graphical scheme showing the effect of SAP15 on inflammatory bowel disease (IBD). Weakened epithelial cells and a reduced mucous layer allow microbes to infiltrate the lamina propria, inducing various immune responses. Gut microbes induce macrophage polarization toward the M1 state, and these cells release inflammatory cytokines to promote helper T cells. SAP15 penetrates the macrophages nucleus and inhibits HDAC5 phosphorylation by binding to the protein. Unphosphorylated HDAC5 inhibits the acetylation of histones and proinflammatory gene transcription, which shifts M1 macrophages to M2 macrophages and reduces proinflammatory cytokines, facilitating macrophage polarization homeostasis in the intestine

Human β-defensin 3 is a promising therapeutic agent that regulates the antimicrobial effect of various pathogens and drug-resistant microbes [43]. With effective antimicrobial activity and low cytotoxic activity in eukaryotic cells, human β-defensin 3 and its analogs have been widely studied in various immune responses, infections, and wound healing [44, 45]. We developed a synthetic peptide called SAP15 that is derived from human β-defensin 3, and we evaluated its biological significance in CPP and anti-inflammatory regulation in a collagen-induced arthritis (CIA) rat model [37]. To ensure the activities of SAP15 in human cells, we further examined human macrophage penetration and anti-inflammation by binding to HDAC5. Moreover, a study was carried out on the application of SAP15 in an IBD model, another immune-mediated inflammatory disease, to reduce inflammatory activity and facilitate tissue recovery.

For proper activity of SAP15, effective peptide delivery to human macrophages is necessary. Many cell-penetrating peptides (CPPs) have been introduced for their effective cell internalization, but the use of CPPs in the clinic is still limited due to their variances in cargo proteins, concentration, cell type, and toxicity [46–48]. As other CPPs have cationic properties that enable them to cross the plasma membrane into the cytosol, SAP15 (GKCSTRGRKCCRRKK) also contains high levels of basic amino acids such as arginine and lysine residues. The SAP15 property as a CPP was evaluated using murine macrophage cells (RAW264.7) in a previous study [37] and human macrophages (THP-1), as shown in Fig. 1A. Localization of SAP15 showed that SAP15 completely infiltrated the membrane within a short time. At higher resolutions, we identified that SAP15 was expressed in both the cytosol and nucleus, suggesting that SAP15 can be internalized into both the cytosol and nucleus of human macrophages.

Another property of SAP15 is enhanced binding to HDAC5, which was confirmed by surface plasmon resonance (SPR) analysis. SPR analysis is a well-established label-free method that is used to detect interactions between molecules [49, 50]. Because SPR method alone cannot demonstrate accurate binding affinity of a molecule, another label-free assay called BLI method was used to obtain a reliable binding kinetic constant comparable to that of the SPR method [51–53]. We used BLI analysis to verify the binding affinity of SAP15 for HDAC5, in contrast to the control peptide. In Fig. 2B, SAP15 showed high binding affinity to HDAC5 with a KD value of 0.347 pM, while the KD value of the control peptide was 10,000-fold higher, which correlates with previous SPR results. Interestingly, high ka and low kd values of SAP15 were measured via the BLI method compared to the SPR method, and it appears that SAP15 has high binding and low disassociation kinetics to HDAC5, as shown by BLI. The specific binding site of HDAC5 and SAP15 was not determined in this study, but negatively charged HDAC5 has an isoelectric point of 5.83, and cationic SAP15 has firm binding potential under physiological conditions [37].

SAP15-treated human macrophage cells were thoroughly analyzed for the high-throughput screening of protein expressions. Heatmap analysis of the protein intensities in THP-1 macrophages showed that proteins in the LPS group were inversely expressed compared to those in the NT group, demonstrating altered protein expression induced by LPS. However, when SAP15 was administered with LPS, protein expression shifted from that in the LPS group only. Most of the proteins that were reduced by LPS were increased by SAP15, and vice versa. These protein clusters in the LPS, LPS + SAP15, and NT groups were compared for correlation, and the LPS + SAP15 group was closely related to the NT group but not the LPS-treated group. This result suggests that proinflammatory macrophages are shifting back to anti-inflammatory phenotype. Furthermore, the ratio of phosphorylated proteins to unphosphorylated proteins showed that several proinflammatory proteins were dephosphorylated by SAP15, including HDAC5. A previous study also reported that SAP15 treatment downregulated phosphorylated HDAC5. Under inflammatory conditions, HDAC5 is phosphorylated at the serine 498 site, creating docking sites for 14–3–3 to form a complex with HDAC5 and proceed with nuclear export. When SAP15 was introduced to macrophages, SAP15 binding to HDAC5 interfered with the phosphorylation sites and prevented HDAC5 from nuclear export. HDAC5 prevented the transcription of proinflammatory genes and cytokines, and the inflammatory response was alleviated.

After examining the signaling pathway via cell internalization and the binding of HDAC5, the effect of SAP15 on macrophage polarization and cytokines was analyzed. Macrophages exhibit high phenotypic plasticity, and this plasticity enables them to adapt and respond to various environmental stimuli, facilitating tissue homeostasis and inflammation [54]. Polarized macrophages are classified into either M1 or M2 subtypes depending on their activation states. Generally, M1 macrophages are proinflammatory, while M2 macrophages are anti-inflammatory, and the relevant proteins are expressed and secreted [55]. Macrophages are activated into an M1-like state by LPS, which is characterized by CD80 surface markers and increased expression of proinflammatory cytokines such as TNF-α or IL-6 [56]. In contrast, M2-like macrophages are induced by IL-4 and are defined by CD206 expression and reduced proinflammatory cytokines [57, 58]. We stimulated differentiated THP-1 macrophages with LPS and/or IL-4 and characterized their surface markers. As shown in Fig. 4A, we observed that only CD80 and CD206 were expressed in LPS- and IL-4-induced macrophages, respectively. Cotreatment with LPS and IL-4 was also performed, and a slight increase in CD206 + cells was observed. When SAP15 was administered with LPS, however, a significant increase in CD206 + cells and a decrease in CD80 + cells were observed, which indicated macrophage polarization from the non-polar state toward an anti-inflammatory phenotype. Although SAP15 highly reduces proinflammatory activities via regulating macrophages, complete inhibition of M1 polarization is unlikely to occur, as there are networks of several signaling molecules, including HDAC5, that regulate macrophage polarization. The effect of SAP15 has potential to enhance its activity by combination therapy with other immunosuppressants. Consistent with surface marker expression, LPS-induced increases in proinflammatory cytokines were reduced to one third by SAP15. SAP15 penetrated cells and showed binding affinity to HDAC5, dephosphorylating HDAC5 and reducing proinflammatory macrophages and cytokines. For the post-treatment of SAP15 after macrophage polarization, the peptides seem to trigger trans-polarization of polarized macrophage cells, although changes were minimal. The trans-polarization of SAP15 can be further investigated.

After the assessment of SAP15 efficacy in vitro, we further analyzed the therapeutic effect of SAP15 in a DSS-induced colitis mouse model, which is a widely used IBD model, at the macroscopic, microscopic, and molecular levels. Anti-TNF-α was used as a positive control, as most of the FDA-approved IBD therapeutics are mostly anti-TNF-α agents. Macroscopic changes of SAP15 treatment was compared with the defective and anti-TNF-α groups. Although the changes in body weight were observed to be insignificant, SAP15 treatment for 7 days showed 5% recovery from defect group. This result can suggest that SAP15 treatment has healing effect of inflamed tissue by regaining body weights. In addition, histological analysis also indicated that fewer immune-related cells were present in the colon after SAP15 treatment, which was more effective than anti-TNF-α treatment. This result was supported by the MPO activity assay results, which showed that SAP15 significantly reduced MPO activity, comparable to that of anti-TNF-α. MPO is a key marker for inflamed tissue in that it is secreted from neutrophils and macrophages in inflammatory conditions [59, 60]. A reduced number of infiltrated immune cells mediated by SAP15 resulted in decreased MPO activity in the colon. SAP15 also reduced the proinflammatory cytokines TNF-α and IL-6 in serum, which were increased by DSS, and these results correlate with the previous results of SAP15. Moreover, histological imaging indicated that SAP15 seemed to repair crypts and the epithelium in colon tissue, while anti-TNF-α did not. As few studies have indicated that human β-defensin 3 is a potential therapeutic treatment for IBD due to its antimicrobial activities [43, 61, 62], SAP15, which is derived from human β-defensin 3, might have additional antimicrobial activities or other activities that are similar to human β-defensin 3, which will be further investigated in future studies.

In our study, we expanded the research on SAP15 as a potential anti-inflammatory agent. The SAP15 characteristics of cell penetration, HDAC5 binding, and mode of action were further evaluated in human macrophage cells and showed the efficacy of this peptide. As a result, SAP15-mediated polarization of human macrophages shifted from a proinflammatory state to an anti-inflammatory state, with reduced proinflammatory cytokine release. Furthermore, an in vivo IBD model showed localized SAP15 that alleviated inflamed colon tissue.

In conclusion, the novel CPP SAP15 induces anti-inflammatory effects via the regulation of HDAC5 phosphorylation in human macrophages. Furthermore, SAP15 treatment ameliorates changes in colon length, the mucosa layer, and proinflammatory cytokines in DSS-induced colitis mice by effectively delivering peptides to colon tissue. Taken together, these data suggest that SAP15 is an effective peptide drug for IBD and other inflammatory diseases.

Authors contribution

Conceptualization and draft supervision of project: JYL, YSP, CPC, and YJP. Overall experiments and preparation of the original draft: DK Writing the manuscript: DK, JYL, and YJP. Biological experiments: DWL, and GY. Peptide synthesis: EKJ, MSC, HCL. All authors have read and agreed to the published version of the manuscript.

Declarations

Conflict of interest

The authors have no financial conflicts of interest.

Ethical statement

The animal studies were performed after receiving approval of the Institutional Animal Care and Use Committee (IACUC) in Seoul National University (IACUC Approval No. SNU-190121-1).