Targeted delivery of Chinese herb pair-based berberine/tannin acid self-assemblies for the treatment of ulcerative colitis

Abstract

Introduction: Ulcerative colitis (UC) is a chronic recurrent idiopathic disease characterized by damage to the colonic epithelial barrier and disruption of inflammatory homeostasis. At present, there is no curative therapy for UC, and the development of effective and low-cost therapies is strongly advocated.

Objectives: Multiple lines of evidence support that tannic acid (TA) and berberine (BBR), two active ingredients derived from Chinese herb pair (Rhei Radix et Rhizoma and Coptidis Rhizoma), have promising therapeutic effects on colonic inflammation. This study aims to develop a targeted delivery system based on BBR/TA-based self-assemblies for the treatment of UC.

Methods: TA and BBR self-assemblies were optimized, and hyaluronic acid (HA) was coated to achieve targeted colon delivery via HA-cluster of differentiation 44 (CD44) interactions. The system was systematically characterized and dextran sodium sulfate (DSS)-induced mouse colitis model was further used to investigate the biodistribution behavior, effect and mechanism of the natural system.

Results: TA and BBR could self-assemble into stable particles (TB) and HA-coated TB (HTB) further increased cellular uptake and accumulation in inflamed colon lesions. Treatment of HTB inhibited pro-inflammatory cytokine levels, restored expression of tight junction-associated proteins and recovered gut microbiome alteration, thereby exerting anti-inflammatory effects against DSS-induced acute colitis.

Conclusion: Our targeted strategy may provide a convenient and powerful platform for UC and reveal new modes of application of herbal combinations.

Keywords: Berberine; Drug delivery; Gut microbiome; Tannin acid; Ulcerative colitis.

Graphical abstract

Fig. 1

Preparation and characterization of TB. (A) Chemical structures of TA and TB. (B) Size distribution and morphology of TB. Data are expressed as mean ± SD (n = 3). (C) ITC raw data of TB. (D) Fitted curves and thermodynamic parameters of the titration of TB. (E) Infrared radiation profiles of TB and their monomers. (F) UV–Vis spectroscopy of TB and their monomers. (G) XRD profiles of TB and their monomers.

Fig. 2

Preparation and characterization of HTB. (A) Schematic diagram of preparing HTB. (B) Size distribution and morphology of HTB. (C) Zeta-potential distribution of HTB and TB. (D) Stability of size distribution of TB and HTB. Data are expressed as mean ± SD (n = 3).

Fig. 3

Cellular uptake of TB and HTB. (A) Qualitative analysis and (C) quantitative measurement of TB or HTB in Raw264.7 cells with or without the pretreatment of 1 mg/mL of HA for 4 h. (B) Qualitative analysis and (D) quantitative measurement of TB or HTB in NCM460 cells with or without the pretreatment of HA for 4 h. Data are expressed as mean ± SD (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 4

In vivo adhesion of TB and HTB. (A) In vivo fluorescence imaging of orally delivered IR780-TB and IR780-HTB in colitic and healthy mice. (B) Representative fluorescence images of distal colons separated from colitic and healthy mice after administration for 24 h. (C) Region of interest (ROI) fluorescence intensity of the fluorescence images of distal colons measured with Living Image 4.5 software. Data are expressed as mean ± SD (n = 6). *P < 0.05 and **P < 0.01.

Fig. 5

In vivo therapeutic effects of HTB. (A) C57BL/6 mice were supplied with water or 3% DSS in drinking water for 11 d. On days 3, 5, 7 and 9, mice were orally administered with TB (BBR at 20 mg/kg), HTB (BBR at 20 mg/kg) or 5-ASA (100 mg/kg). (B) Curve chart changes of mice body weight across time among the five intervention means, normalized to the percentage of the initial body weight. (C) Curve chart variations of in vivo DAI scores of mice. (D) Histogram statistics of colon length with different intervention means. (E) Images of mouse colons of different groups. (F) H&E staining of colonic sections after the five different treatments. Scale bar was 200 μm. Data are expressed as mean ± SD (n = 6). *P < 0.05, **P < 0.01 and ***P < 0.001.

Fig. 6

Oxidative stress markers and colonic inflammatory cytokine changes after HTB treatment. (A-B) Histogram analysis of oxidative stress markers (H₂O₂ and MDA) changes in serum among the five formulations. (C) Histogram analysis of serum MPO expression. (D-F) Histogram analysis of murine colonic inflammatory cytokines (IL-6, IL-1β and TNF-α) changes. Data are presented as mean ± SD (n = 6). *P < 0.05, **P < 0.01 and ***P < 0.001.

Fig. 7

HTB regulates gut barrier function and apoptosis in colitis mice. (A) The expression of ZO-1 protein in colon by IHC assay. Scale bar was 200 μm (B) Colon tissues treated as shown were processed by the immunofluorescence assay, followed by visualization by confocal microscopy. Scale bar was 100 μm. Data are expressed as mean ± SD (n = 6).

Fig. 8

HTB partially recovered DSS-mediated gut microbiome alteration. (A-C) Alpha diversity index of observed species, Shannon and chao1. (D) PcoA analysis at OTU level (displaying top 15 features). (E) Stacked area plot of phylum abundance cross groups. (F-I) Relative abundance of individual phylum. (J) LEfSe analysis based on genus level (top 15 features ranked from most significance). (K) LEfSe analysis based on species level. (L-M) Genus abundance of Lactobacillus and Akkermansia bacteria. (N-O) Species abundance of lactobacillus bacteria. Data are expressed as mean ± SD (n = 4 or 5). *P < 0.05, **P < 0.01 and ***P < 0.001.