Molecular interactions between piperine and peroxisome proliferator-activated receptor gamma ligand-binding domain revealed using co-crystallization studies

Piperine has been investigated for a diverse array of biological effects, including a potential role in modulating peroxisome proliferator-activated receptor gamma (PPARγ), a nuclear receptor that plays a pivotal role in regulating lipid and glucose metabolism. This study conducted a comprehensive co-crystallo­graphic analysis of the complex of piperine with the PPARγ ligand-binding domain (PPARγ-LBD), with the objective of elucidating the precise binding interactions of piperine. The co-crystal structure revealed that piperine binds within the ligand-binding pocket of PPARγ-LBD via hydrogen-bonding and hydrophobic interactions with residues of the ligand-binding site. Notably, in contrast to conventional full agonists, piperine does not directly stabilize helix H12. This could contribute to the comparatively weaker agonistic activity of piperine. The results of this study also suggest that piperine binding facilitates a role as a partial agonist or even an antagonist under certain physiological conditions. Collectively, these findings contribute to a greater understanding of the manner in which piperine modulates PPARγ function and its potential as a therapeutic candidate for the treatment of metabolic disorders. Given its natural origin and relatively minimal side-effect profile, piperine and its derivatives could be promising alternatives to synthetic PPARγ modulators such as thiazolidinediones, which have significant side effects.

1. Introduction

Piperine, an alkaloid found in Piper nigrum (black pepper) and P. longum (long pepper), has a long history in traditional medicine due to its diverse biological activities. Recent research has highlighted its antioxidant, anti-inflammatory, anticancer and antidiabetic properties (Imran et al., 2022; Yadav et al., 2023). Notably, piperine is distinguished by its capacity to augment the bioavailability of other pharmaceutical agents by modulating the activity of enzymes such as the cytochrome P450 superfamily and P-glycoprotein (Ashokkumar et al., 2021; Alves et al., 2022).

Of particular interest is the molecular target of peroxisome proliferator-activated receptor gamma (PPARγ), given its role in regulating lipid metabolism, glucose homeostasis and adipogenesis. PPARγ agonists, such as thiazolidinediones (TZDs), are effective insulin sensitizers but are associated with adverse effects, including weight gain and cardiovascular risks. Thus, natural PPARγ modulators such as piperine may offer therapeutic benefits with fewer side effects (Mirza et al., 2019; Cheng et al., 2019). However, conflicting reports exist regarding the effects of piperine on PPARγ. Some studies suggest that piperine functions as a partial agonist, enhancing insulin sensitivity and promoting glucose uptake, while others indicate that it acts as an antagonist, inhibiting adipogenesis and lipid accumulation (Park et al., 2012; Kharbanda et al., 2016; Yan et al., 2019; Ma et al., 2017). The dual nature of the activity of piperine against PPARγ poses an intriguing challenge to researchers, as further investigation will be needed to fully elucidate its mechanism of action.

Molecular-docking studies have predicted that piperine interacts with key residues within the PPARγ ligand-binding domain (PPARγ-LBD) through hydrogen-bonding and hydrophobic interactions (Francis et al., 2024; Dhadke et al., 2024; Kharbanda et al., 2016). However, discrepancies among computational methods highlight the need for experimental validation, such as by X-ray crystallography. In this study, we co-crystallized piperine with PPARγ-LBD and performed an X-ray crystallographic analysis.

2. Materials and methods

2.1. Expression and purification

The human PPARγ-LBD (amino acids 204–477) was expressed using a modified pET-30a vector with an N-terminal 6×His-tag cleavable by TEV protease. Escherichia coli Rosetta 2 (DE3) cells were freshly transformed with the plasmid and cultured in four flasks containing 0.75 l 2×TY medium with 34 µg ml⁻¹ kanamycin and 50 µg ml⁻¹ chloramphenicol at 37°C to an optical density at 600 nm of 1.0. Protein synthesis was then induced by the addition of 0.5 mM isopropyl β-D-1-thiogalactopyranoside and the cultures were incubated for an additional 18 h at 20°C. The cells were harvested and resuspended in lysis buffer (20 mM Tris–HCl pH 8.0, 100 mM NaCl, 1 mM TCEP, 0.5 mM EDTA, 1× protease-inhibitor cocktail). The cells were then lysed by sonication and the soluble fraction was isolated by centrifugation (18 000g for 20 min). The supernatant was applied onto cOmplete His-Tag Purification Resin (Roche), which was then thoroughly washed with wash buffer (20 mM Tris–HCl pH 8.0, 100 mM NaCl, 1 mM TCEP, 0.5 mM EDTA, 5 mM imidazole). The human PPARγ-LBD was eluted with elution buffer (20 mM Tris–HCl pH 8.0, 100 mM NaCl, 1 mM TCEP, 0.5 mM EDTA, 250 mM imidazole). TEV protease was added to the eluate and the mixture was dialyzed overnight at room temperature against buffer (20 mM Tris–HCl pH 8.0, 1 mM TCEP, 0.5 mM EDTA). The cleaved protein was then passed through cOmplete His-Tag Purification Resin and the flowthrough was loaded onto a Resource Q (6 ml) column (GE Healthcare) equilibrated with buffer (20 mM Tris–HCl pH 8.0, 1 mM TCEP, 0.5 mM EDTA). The column was eluted using a 0–0.5 M gradient of NaCl in starting buffer. Fractions of the eluate were concentrated and loaded onto a Superdex 75 (24 ml) gel-filtration column equilibrated with buffer (20 mM Tris–HCl pH 8.0,100 mM NaCl, 1 mM TCEP, 0.5 mM EDTA). The protein was concentrated using a Vivaspin 20 (Sartorius) and centrifuged at 3000g at 4°C until the desired concentration (8 mg ml⁻¹) was reached. The protein was not frozen and was stored at 4°C until use.

2.2. Crystallization and data collection

Crystals were obtained by co-crystallization with piperine. Co-crystallization was performed by vapor diffusion at room temperature using a hanging drop prepared by mixing 1 µl protein solution (8 mg ml⁻¹ in 20 mM Tris–HCl pH 8.0, 1 mM TCEP, 0.5 mM EDTA) with 0.25 mM ligand in 1 µl reservoir solution (0.7 M sodium citrate, 0.1 M Tris pH 7.0) under a nitrogen atmosphere to minimize piperine isomerization. To further prevent isomerization, the mixture was stored in the dark, and prismatic crystals appeared after a few days (Table 1). The crystals were flash-cooled in liquid nitrogen after fast soaking in a cryoprotectant (LV Cryo Oil; MiTeGen, New York, USA). Diffraction data sets were collected on beamline BL-5A of the Photon Factory Advanced Ring (PF-AR) at the High-Energy Accelerator Research Organization (KEK), Tsukuba, Japan. Reflections were recorded with an oscillation range per image of 1.0°. Data were indexed, integrated and scaled using MOSFLM (Leslie, 2006) and the CCP4 suite of programs (Agirre et al., 2023). Structures were solved using molecular replacement and then rebuilt and refined using Coot (Emsley & Cowtan, 2004) and REFMAC (Murshudov et al., 1997) (Tables 2 and 3). Modeling of the ligand was guided at various stages using simulated-annealing omit maps and simulated-annealing composite omit maps (calculated using CCP4i). Coordinate data for the structures were deposited in the Protein Data Bank under accession code 9l3t.

Table 1
Crystallization

Method	Hanging drop
Plate type	24-well crystallization plate for hanging drops
Temperature (K)	300
Protein concentration (mg ml^−1)	8
Buffer composition of protein solution	20 mM Tris–HCl, 0.5 mM EDTA pH 7.4
Composition of reservoir solution	0.7 M sodium citrate, 0.1 M Tris pH 8.0
Volume and ratio of drop	2 µl (1:1 protein:reservoir)
Volume of reservoir (µl)	500

Table 2
Data collection and processing

Values in parentheses are for the outer shell.

Diffraction source	BL-5A, PF-AR
Wavelength (Å)	1.0000
Temperature (K)	100
Detector	Dectris PILATUS3 S6M
Crystal-to-detector distance (mm)	474.61
Rotation range per image (°)	1
Total rotation range (°)	180
Exposure time per image (s)	2.00
Space group	C2
a, b, c (Å)	93.43, 62.02, 119.49
α, β, γ (°)	90, 102.09, 90
Mosaicity (°)	0.89
Resolution range (Å)	45.68–2.36 (2.45–2.36)
Total No. of reflections	85928 (8900)
No. of unique reflections	27580 (2863)
Completeness (%)	99.6 (99.5)
Multiplicity	3.1 (3.1)
〈I/σ(I)〉	9.2 (2.6)
Rr.i.m.	0.058 (0.344)
Rp.i.m.	0.040 (0.236)
CC1/2	0.995 (0.936)
Overall B factor from Wilson plot (Å^2)	50.005

Table 3
Structure solution and refinement

Values in parentheses are for the outer shell.

Resolution range (Å)	45.721–2.360 (2.421–2.360)
Completeness (%)	99.4 (99.2)
No. of reflections, working set	26196 (1901)
No. of reflections, test set	1374 (99)
Final Rwork	0.201 (0.255)
Final Rfree	0.255 (0.285)
No. of non-H atoms
Protein	4062
Ligand	21
Water	36
R.m.s. deviations
Bond lengths (Å)	0.010
Angles (°)	2.200
Average B factors (Å^2)
Protein chain A/chain B	67.7/70.0
Ramachandran plot
Most favored (%)	98
Allowed (%)	2
Disallowed (%)	0

3. Results and discussion

Co-crystallization of the hPPARγ-LBD with piperine (Fig. 1) was successful and the resulting complex was subsequently analyzed using X-ray crystallography (PDB entry 9l3t). The analysis was conducted using the region of hPPARγ-LBD encompassing amino-acid residues 204–477. As no co-crystal structures of compounds structurally similar to piperine have been reported, the crystal structure of the piperine–PPARγ-LBD complex was compared with the following representative crystal structures. The overall structure exhibited a high degree of alignment with PPARγ-LBD structures with no bound ligands (PDB entry 3prg; Uppenberg et al., 1999), bound to the full agonist rosiglitazone (PDB entry 2prg; Nolte et al., 1998), to the partial agonist nTZDpa (PDB entry 2q5s; Bruning et al., 2007) and to the antagonist MMT-160 (PDB entry 7wox; Yoshizawa et al., 2022) (Fig. 2). The root-mean-square deviation values in comparison with the respective crystal structures were 0.454, 0.360, 0.257 and 0.218 Å. As also observed in PDB entries 2prg, 2q5s and 7wox, PDB entry 9l3t also contained two protein molecules in the asymmetric unit. Molecule A displayed an active conformation comparable to that observed in the crystal structure of hPPARγ-LBD complexed with the full agonist rosiglitazone (Liberato et al., 2012), whereas molecule B exhibited an alternative conformation with helix H12 stabilized in a different position. Piperine was bound to the active conformation of the hPPARγ-LBD.

Figure 1 Structure of piperine.

Figure 2 Superposition of the backbones and ligands of PDB entries 9l3t (green), 3prg (gray), 2prg (magenta), 2q5s (cyan) and 7wox (yellow).

The degree of stability imparted by piperine, rosiglitazone, nTZDpa and MMT160 was then compared using the normalized B factor (B′; Yuan et al., 2003; Parthasarathy & Murthy, 1997; Fig. 3). In piperine-bound hPPARγ-LBD, molecule A demonstrated a stabilizing effect within the Val455–Leu469 loop, which encompasses regions of helices H11 and H12. Differences in B′ values were observed in regions encompassing amino acids 247–249 and 345–349, indicating that β-strand 1 (amino acids 247–249) and β-strand 4 (amino acids 345–349) were moderately stabilized by piperine, whereas β-strand 2/3 (amino acids 338–341) exhibited only a minimal change. It has been hypothesized that when β-strand 1 is unstable it adopts a loop-like conformation, facilitating the phosphorylation of PPARγ by CDK5 (Ribeiro Filho et al., 2019). However, if a ligand stabilizes β-strand 1 this conformational change may be prevented, reducing CK5 accessibility to the phosphorylation site and ultimately inhibiting the phosphorylation of PPARγ. It is notable that the ability of piperine to stabilize this region could influence the susceptibility of Ser245 (in PPARγ1 numbering; Ser273 in PPARγ2 numbering) to phosphorylation, thus suggesting a potential mechanism by which piperine modulates PPARγ activation through structural stabilization, thereby contributing to its biological effects.

Figure 3 Comparison of normalized C^α-atom B factors of PDB entries 9l3t (green), 3prg (gray), 2prg (magenta), 2q5s (cyan) and 7wox (yellow).

In order to facilitate a comparison of the ligand-binding regions in PDB entries 2prg, 2q5s, 7wox and 9l3t, an enlarged view is presented in Fig. 4. Additionally, an electron-density map and structural model of piperine bound to molecule A are shown in Fig. 5. This figure highlights the amino-acid side chains and water molecules involved in polar interactions with each ligand. The positioning of piperine is of particular interest, with its pyridine ring in close proximity to helix H2′ and its dioxolane ring situated between helix H3 and β-strand 2/3, with no contact with helix H12. This indicates an absence of direct interaction with helix H12, a characteristic that is also exhibited by MMT-160 and nTZDpa. Conformational changes in the amino-acid residues of PDB enty 9l3t within the PPARγ ligand-binding pocket (LBP) were not observed in comparison to PDB entry 3prg. A water molecule within the LBP formed hydrogen bonds to both piperine and Ser342, thus contributing to the stabilization of the orientation of piperine within the pocket. Furthermore, hydrophobic interactions were observed between piperine and several residues, including Ile262, Ile249, Leu330, Ile341 and Met364. However, no additional direct interactions between piperine and other amino-acid residues were identified. The distance between Ile249 (β-strand 1) and Ile281 (helix H3) was measured as an indicator of the spatial relationship between β-strand 1 and helix H3. The distances obtained for PDB entries 3prg, 2prg, 2q5s, 7wox and 9l3t were 12.1, 11.7, 11.6, 11.5 and 11.3 Å, respectively. The distances were measured using the PyMOL `distance' function, which calculates atomic distances between selected atoms. It is noteworthy that PDB entry 9l3t exhibited a slightly shorter distance compared with the other structures.

Figure 4 Detailed examination of the ligand-binding site. The amino acids and waters that are in polar contact with the ligand are shown for PDB entries 9l3t (green), 2prg (magenta), 2q5s (cyan) and 7wox (yellow).

Figure 5 The electron density in the piperine-binding site. Electron density (blue mesh) is shown from a simulated-annealing composite omit 2Fo − Fc map (contoured at 2.5σ) of piperine, which was calculated as an omit map with the ligand model deleted. The map was generated using CCP4i and was visualized in PyMOL.

It is widely acknowledged that PPARγ agonists play a pivotal role in maintaining the active conformation of the receptor via direct interactions with helix H12. Previous in silico modeling studies hypothesized that piperine interacts directly with helix H12. However, analysis of our co-crystal structure demonstrated that piperine is accommodated in the Ω-pocket and does not directly interact with helix H12. The closest proximity between piperine and Tyr473 on helix H12 is 11.2 Å. The structural arrangement observed in this study, in which the majority of partial agonists bind to the Ω-pocket, supports the hypothesis that piperine acts as a partial agonist. The indirect stabilization of upstream regions is likely to confer mild agonistic activity, a characteristic that has been observed with other ligands that do not interact directly with helix H12 (Montanari et al., 2020). Furthermore, the stabilization of β-strand 1 by piperine may promote PPARγ transcriptional activation by inhibiting receptor phosphorylation. These findings suggest that piperine functions as an inhibitor of phosphorylation, thereby indirectly enhancing receptor activation via stabilization of β-strand 1 (Frkic et al., 2021; Laghezza et al., 2018).

4. Conclusion

This study represents a significant advancement in our understanding of the molecular mechanisms through which piperine exerts regulatory effects on PPARγ. Successful co-crystallization of the piperine–hPPARγ-LBD complex and subsequent X-ray crystallographic analysis provided useful structural details that revealed the precise binding orientation of piperine within the hPPARγ-LBD. These structural insights indicate that piperine engages in both hydrophobic and hydrogen-bond interactions with key residues, while notably maintaining distance from helix H12, a characteristic commonly associated with partial agonists. This indirect stabilization of upstream regions, particularly β-strand 1, may affect the phosphorylation status of Ser245, which could in turn contribute to the modulation of PPARγ activity by piperine via indirect enhancement of transcriptional activation.

Additionally, piperine is known to undergo cis–trans isomerization upon light exposure, a factor that may influence its binding mode and functional activity. This structural flexibility could account for the variability in its agonistic and antagonistic effects, emphasizing the need for further investigation into its isomeric forms to clarify its biological activity.

The findings of the present study enhance our comprehension of piperine as a natural modulator of PPARγ and establish a robust structural foundation for prospective drug-design efforts. Given its dual agonist/antagonist potential and its natural origin, studies of piperine represent a promising avenue for the ultimate development of safer alternatives to synthetic PPARγ modulators, which are often associated with adverse effects. This study also provides a rationale for the design of piperine derivatives or synthetic analogs exhibiting optimal efficacy and safety profiles and offers valuable insights that could support the development of effective therapies for metabolic and inflammatory diseases, including diabetes, obesity and related conditions. The insights gained here advance our comprehension of the dual-function modulation of PPARγ by piperine and pave the way for the development of safer, more effective therapeutic strategies targeting complex metabolic disorders.