Andrographolide and pterostilbene inhibit adipocyte differentiation by downregulating PPARγ through different regulators

Abstract Adipogenesis involves commitment of stem cells and their differentiation into mature adipocytes. It is tightly regulated by hormones, nutrients and adipokines. Many natural compounds are being tested for their anti-adipogenic activity which can be attributed to apoptosis induction in adipocytes, blocking adipocyte differentiation, or inhibiting intracellular triglyceride synthesis and accumulation. In this study, we have determined molecular mechanism of two phytocompounds: andrographolide (AN) and pterostilbene (PT) during differentiation of the human MSCs into adipocyte. Interestingly, AN upregulates miR27a, whereas, PT upregulated SIRT1 which inhibits the expression of PPARγ. Thus, our results clearly demonstrate that both AN and PT inhibited adipogenesis by blocking a surge of reactive oxygen species (ROS) during differentiation and inhibiting expression of crucial transcription factors like SREBP1c and PPARγ. Graphical Abstract


Introduction
During adipogenesis, stem cells are committed to adipose cell lineage and are differentiated into lipid droplet containing mature adipocytes (Geloen et al. 1989). It is a multistep process, where gene expression is regulated through a cascade of transcription factors (TFs) and cell cycle proteins leading to adipocyte differentiation and development (Stephens 2012). Peroxisome proliferator-activated receptor gamma (PPARc) and CCAAT enhancer-binding protein beta (CEBPb) are key TFs participating in adipogenic differentiation. These are expressed in very early stages during adipogenesis and induce expression of each other to enhance adipogenesis. Differentiation inducers like insulin and glucocorticoids activate expression of CEBPb early in the differentiation, which in turn activates PPARc. PPARc is considered as a master regulator of adipogenesis and is regulated through many positive and negative regulators. Downstream targets of these TFs are genes involved in fatty acid synthesis like fatty acid synthase (FAS), fatty acid metabolism like lipoprotein lipase (LPL), fatty acid storage like fatty acid binding proteins (FABPs) and few other genes such as glucose transporter-4 (Glut4), AdipoQ etc (Frith and Genever 2008). Accumulating evidence has suggested that in addition to these TFs, regulation of redox state is also important for adipocyte differentiation and certain amount of reactive oxygen species (ROS) are necessary for both initiation of adipocyte lineage and commitment of precursor stem cells (Wang and Hai 2015). Further, scavenging of ROS has also been shown to inhibit adipogenesis (Liu et al. 2012).
Many anti-obesity drugs used currently are known to have unwanted side effects such as headache and blood pressure abnormalities. Efforts are therefore being made to investigate different natural phytocompounds for checking anti-adipogenic activity (Kim et al. 2018;John and Arockiasamy 2020;Gire et al. 2021). In this study, we have investigated anti-adipogenic activity of andrographolide (AN) and pterostilbene (PT) using differentiating hMSCs.
AN is a crystalline, natural labdane bicyclic diterpenoid lactone obtained from the plant Andrographis paniculate (Lim et al. 2012). Extract of A. paniculata and AN are commonly used in traditional medicine in India and have several beneficial properties (Zeng et al. 2022). PT is a methylated analogue of resveratrol and is a principal compound found in the bark of Pterocarpus marsupium. PT is also known to have diverse pharmacological activities (McCormack and McFadden 2013). Both AN and PT are present in the ayurvedic formulations used to treat obesity. In this study, we observed that both AN and PT inhibited adipogenesis during the differentiation of hMSCs into adipocytes by down-regulating master regulator PPARc through different regulators.

Effect of AN and PT on cell viability
In case of AN treated cells, viability was significantly reduced only at 20 mM and remained unaffected at 0, 1, 5 and 10 mM ( Figure S1A). Similarly, in case of PT viability reduced significantly at 10 mM and was not affected at lower concentrations (1 and 5 mM) used ( Figure S1A). Cell viability was also checked for 10 mM AN and 5 mM PT concentrations for 21 days of differentiation and it was not found to be significantly affected ( Figure S1A). Thus, 10 mM AN and 5 mM PT concentrations were used for further experiments.

Effect of AN and PT on differentiation of hMSCs to adipocytes
In control (untreated) differentiating hMSCs, oil droplets were clearly seen on day 6 which further increased significantly on day 21. In differentiating hMSCs, treated with AN or PT significant reduction in oil red O staining was observed both on day 6 and 21 of differentiation ( Figure 1A). Lipid accumulation was also observed to be significantly decreased on day 6 and 21 in cells treated with AN or PT compared to control untreated cells (Figures 1B, 1C).
Oil droplets formed during adipocyte differentiation are due to triglycerides. Therefore, in hMSCs differentiated in presence of AN or PT triglycerides were quantitated on the last day of differentiation. Compared to control differentiated hMSCs, percent triglyceride content reduced significantly in hMSCs differentiated in presence of AN or PT ( Figure 1D).
Previously in differentiating 3T3-L1 cells, AN (Jin et al. 2012) or PT (Hsu et al. 2012) have been shown to inhibit triglyceride accumulation. 3T3-L1 cell line is a well-established pre-adipose cell line developed from murine Swiss 3T3 cells and has often been used to study adipocyte differentiation. In this study, we have used hMSCs which are human stem cell population and can be differentiated in different cell lineages and offer a better model to study differentiation.

Effect of phytocompounds on ROS during adipocyte differentiation
Both AN and PT are reported to have strong antioxidant activity (Acharya and Ghaskadbi 2013;Mittal et al. 2016). Therefore, to check whether the inhibition of differentiation is due to decrease in ROS, intracellular ROS was measured using DCFHDA dye from day 0 to 7 of differentiation. In control untreated hMSCs, significant peak of ROS was observed on day 3 and 4. However, in case of cells treated with AN or PT, this sharp increase in ROS was not detected ( Figure 1E).
ROS is known to be produced endogenously during adipogenesis which acts as a crucial mediator for adipogenic differentiation. Adipogenic hormonal cocktail used in culture is known to increase ROS production and thereby enhance the process of differentiation (Wang 2017). Importance of these ROS is evidenced by the fact that treatment with antioxidant such as N acetyl cysteine (Guo et al. 2006) or knocking down an enzyme NOX-4 responsible for generation of ROS results in inhibition (Pinney and Emerson 1989) of adipocyte differentiation.

Expression of genes important for adipogenesis during differentiation of hMSCs
Expression of important genes for adipogenesis such as PPARc, CEBPb and SREBP1c was checked by qRT-PCR. In hMSCs, differentiating in presence of AN, expression of PPARc ( Figure S2A) and its positive regulators SREBP1c ( Figure S2B) and CEBPb ( Figure  S2C) was found to be significantly reduced compared to control. PPARc is additionally negatively regulated by miR-27a (Kim et al. 2010). In hMSCs, differentiating in presence of AN, expression of miR-27a was indeed found to be significantly increased concomitant with significant decrease in its negative regulator SATB2 ( Figure S2D, E) (Gong et al. 2014). Expression of glut4, a marker of mature adipocyte (Sakaue et al. 1998) was also significantly decreased in hMSCs differentiated in the presence of AN ( Figure S2F).
Like in AN, in presence of PT, expression of PPARc ( Figure S3A), and its positive regulators SREBP1c ( Figure S3B) and CEBP-b ( Figure S3C) and their target gene glut4 ( Figure S3F) was found to be significantly reduced compared to control untreated cells. However, among negative regulators, SIRT1, was found to be significantly decreased in only PT treated cells as compared to control ( Figure S3D). In order to determine the regulation of SREBP1c by AN and PT, expression level of FOXO1, a regulator of SREBP1c was determined. FOXO1 was observed to be upregulated only in PT treatment ( Figure S3E).
Thus, expression of two positive regulators namely CEBPb and SREBP1c were found to be significantly affected by both AN and PT. CEBPb activates PPARc transcription, whereas SREBP1c increases PPARc activity by production of endogenous ligands (Kim et al. 1998). In case of PT, decrease in SREBP1c mRNA was through increase in FOXO1 known to inhibit SREBP1c transcription. While in case of AN, SREBP1c was downregulated possibly through Protein Kinase A (PKA). We have already demonstrated in our earlier work that AN activates PKA through Adenosine A 2A receptor (Mittal et al. 2016). Thus, both AN and PT significantly reduced the expression of SREBP1c but through different regulators.
PPARc is additionally known to get negatively regulated by many miRNAs which inhibit translation by binding to UTRs. In differentiating hMSCs treated with AN, miR-27a was found to be upregulated along with downregulation of its negative regulator SATB2. SATB2 is also known to be inhibited by miR-31 whose transcription is activated by RUNX2, which has been shown to be upregulated by AN (Phunikom et al. 2021) Interestingly, treatment with PT did not change expression of either miR-27a, or SATB2.
Significant increase in SIRT1, other negative regulator of PPARc, was observed only in cells treated with PT (Guo et al. 2016). SIRT1 deacetylates PPARc and inactivates it. In 3T3L1 cells, during adipocyte differentiation PT has been shown to attenuate expression of CEBPa, PPARc (Hsu et al. 2012) and inhibit mitotic clonal expansion via upregulation of HO-1 (Seo et al. 2017). However, in hMSCs, PT inhibited PPARc by upregulating FOXO1, SIRT1 and downregulating SREBP1c.

Experimental
See supplementary material.

Conclusion
Both AN or PT inhibited adipogenesis in differentiating hMSCs by downregulating PPARc, a master regulator of adipogenesis differently through its positive and negative regulators.

Disclosure statement
No potential conflict of interest was reported by the authors.

Funding
This study was supported financially by "Assistance by Savitribai Phule Pune University for Project based innovative research (Aspire) and RUSA-2.0. SG acknowledges University Grant Commission, Center for advanced studies grant (UGC-CAS; UGC-018 (151)) and Department of Science and Technology for promotion of University Research and Scientific Excellence (DST-PURSE; GOI-670). SDK, GSB is financially supported by CSIR-SRF (Council of Scientific and Industrial Research-GoI).