Propionic acid counteracts the inflammation of human subcutaneous adipose tissue: a new avenue for drug development

Abstract

Adipose tissue is a primary site of obesity-induced inflammation, which has been emerging as an important contributor to obesity associated disorders. The factors influencing adipose tissue-induced inflammation and the resulting pathophysiological events remain poorly understood. However, dietary fiber consumptions appear to be protective. Short-chain fatty acids such as propionic acid (PA) are the principal products of the dietary fiber fermentation by microbiota. Therefore, we aim to investigate the influence of PA on inflammation, lipogenesis and glucose uptake markers from human subcutaneous adipose tissue (SAT). We showed that the treatment of SAT with PA resulted in a significant downregulation of inflammatory parameters (e.g. TNF-α and IP-10) and macrophage markers (e.g. CD163 and MMP-9). The expression levels of PA receptors (i.e. G protein coupled receptor-41 and -43) in human primary adipocytes were very low in comparison with SAT and macrophages. Upon PA treatment, no anti-inflammatory effect was observed in human adipocytes. PA significantly upregulated the expression of lipoprotein lipase (LPL), sterol regulatory-element-binding protein-1c (SREBP-1c) and glucose transporter 4 (GLUT-4), which are associated with lipogenesis and glucose uptake. We also showed that the observed anti-inflammatory effects of PA on SAT were partly mediated by Gi/o protein coupled receptor. Our data suggests that PA anti-inflammatory effects on SAT are mediated partly via Gi/o proteins, leading to the improved expression of factors associated with lipogenesis and glucose uptake. These responses appeared to be not mediated by adipocytes; but most probably by macrophages. The current study provides new knowledge, which can be used as a potential new avenue for drug development in preventing obesity-related inflammation and metabolic disorders in future. Graphical abstract Schematic presentation of study flow and the components of the investigation. In this study the effect of propionic acid (PA) on inflammation investigated in human subcutaneous adipose tissue (SAT), human primary adipocytes and the expression of a few hallmark inflammatory components produced by SAT and human adipocytes.

Keywords: Gi/o proteins coupled receptors; IP-10 and TNF-alpha, macrophage; Inflammation; Propionic acid; Subcutaneous adipose tissue.

Graphical abstract

Schematic presentation of study flow and the components of the investigation. In this study the effect of propionic acid (PA) on inflammation investigated in human subcutaneous adipose tissue (SAT), human primary adipocytes and the expression of a few hallmark inflammatory components produced by SAT and human adipocytes.

Fig. 1

The influence of Propionic Acid (PA) on the secretion of chemokines and cytokines by subcutaneous adipose tissue (SAT). SAT explants of each subject were incubated in triplicate with or without 3 mM PA for 24 h. Secreted quantities of chemokines and cytokine in the media were determined by multiplex-ELISA. Results are expressed as relative quantities (RQ) and compared to the control (without PA; C). (N = 10). **, P < 0.01, ***, P < 0.001 vs. control (without PA; C). M, mean

Fig. 2

The effect of Propionic Acid (PA) on the secretion of chemokines and cytokines by human adipocytes. Adipocytes incubated in triplicate with or without 3 mM PA for 24 h. Secreted quantities of chemokines and cytokine in the media were determined by multiplex-ELISA. Only RANTES and MCP-1 were detected and the rest weren’t. Results are expressed as relative quantities (RQ) compared to the control (without PA; C). *, P < 0.05 vs. control (without PA; C). M, mean

Fig. 3

The effect of propionic Acid (PA) on the mRNA expression of adipose tissue macrophage (ATM) related markers. Human subcutaneous adipose tissue (SAT) explants of each subject were incubated in triplicate with or without 3 mM PA for 24 h. PA downregulated all of the ATM markers, i.e. CD16, CD163 and MMP-9. The mRNA expression levels were determined by RT-PCR and expressed as relative quantities (RQ) and compared to the control (without PA; C). (N = 10) *, P < 0.05 vs. control (without PA; C). M, mean

Fig. 4

Comparing mRNA expression of adipose tissue macrophage (AMT) markers produced in omental adipose tissue (OAT), SAT and adipocytes. mRNA was isolated from untreated explants and adipocytes in triplicate. mRNA expression levels were determined by RT-PCR and expressed as relative quantities (RQ) compared to the control (OAT). (N = 10). *, P < 0.05, ***, P < 0.001 vs. OAT. M, mean

Fig. 5

Comparing mRNA levels of GPCR41 and 43 produced in human subcutaneous adipose tissue (SAT), adipocytes and macrophages. mRNA was isolated from untreated explants and adipocytes in triplicate. mRNA expression levels were determined by RT-PCR and shown as relative quantities (RQ) compared to the control (SAT). (N = 10). *, P < 0.05 vs. SAT. M, mean

Fig. 6

The role of Propionic Acid (PA) on the mRNA expression of lipogenesis and insulin sensitivity related markers. Human subcutaneous adipose tissue (SAT) explants of each subject were incubated in triplicate with or without 3 mM PA for 24 h. PA upregulated the expression of lipoprotein lipase (LPL), SREBP-1c and GLUT-4. The mRNA expression levels were determined by RT-PCR and shown as relative quantities (RQ) compared to the control (without PA; C). (N = 10). *, P < 0.05, **, P < 0.01, ***, P < 0.001 vs. control (without PA; C). M, mean

Fig. 7

Involvement of G protein coupled receptor(s). The involvement of Gi/o coupled receptors in mediating Propionic Acid (PA) effects on adipokines expression was determined by blocking the Gi /o signaling pathway with pertussis toxin (PTX). Human subcutaneous adipose tissue (SAT) explants of each subject were incubated in triplicate with PTX (100 ng / ml) for 2 h before incubation for 24 h with or without 3 mM PA. Protein (TNF-α and IP-10) and mRNA expression (CD163 and MMP-9) levels were determined using ELISA and RT-PCR respectively. Results were depicted as relative quantities compared with controls (without PA). (N = 5). **, P < 0.01, ***, P < 0.001 vs. control (without PA; C). M, mean