Polyoxypregnane Aryl Esters as Non-Substrate P-gp Modulators

Drug-resistant tumor cells survive chemotherapy when P-glycoprotein continuously expels structurally unrelated agents before intracellular drug levels can reach a lethal threshold. That problem has lingered in cancer pharmacology for decades, not because the transporter is obscure, but because the chemistry required to modulate it without adding fresh toxicity has been difficult to secure. Multidrug resistance is especially frustrating in this setting: one membrane pump, fueled by ATP hydrolysis, can lower the intracellular exposure of many anticancer agents at once, leaving recurrence, metastasis, and treatment failure tied to the same transport machinery.  Early inhibitors entered the field with recognizable pharmacology but poor tolerability. Later candidates improved selectivity, though clinical use still met practical limits. A transporter that handles many substrates and shifts drug disposition so broadly does not invite easy intervention. Any useful modulator has to interfere with efflux strongly enough to matter while avoiding the burden of becoming a toxic partner in combination therapy.

Natural products continue to draw attention for exactly this reason. They offer structural motifs that medicinal chemistry did not invent and often preserve patterns of oxygenation and stereochemistry that create unusual protein contacts.   In a recent research paper published in Pharmaceuticals (Basel), Miss Yujia Guo, Miss Huiwen Wu, Mr. Taorui Wu, Professor Xiaoling Shen and Professor Yingjie Hu from the Science and Technology Innovation Center at Guangzhou University of Chinese Medicine, developed and characterized two polyoxypregnane aryl esters linked to multidrug-resistance reversal: a natural cinnamoylated pregnane, compound 1, and a new semisynthetic 3-O-nicotinate analogue, compound 1a. They established that both molecules reverse P-gp-mediated resistance in HepG2/Dox cells without lowering P-gp expression, and they tied that activity to direct transporter binding and inhibition of efflux function. They also showed that each compound behaves as a non-substrate in the Caco-2 transport model, which separates these agents from simple competitive cargo molecules.

The investigators isolated compound 1 from Metaplexis japonica as a naturally occurring pentahydroxylated pregnane bearing a cinnamate at C-12, then prepared compound 1a by installing a nicotinate at C-3 while retaining the cinnamate unit. The team built 1a on the premise that aromatic acyl substitution and nicotinyl esterification had already shown value in related systems, so the semisynthetic step tested a structure–activity hypothesis from the outset. Spectroscopic assignment established 1 as 3β,12β,14β,17β,20(S)-pentahydroxy-5α-pregnan-12β-O-(E)-cinnamate and 1a as its 3β-O-nicotinate analogue, giving the biology a defined chemical base.

The authors then moved directly to the resistant-cell phenotype and used HepG2/Dox cells which expressed much more P-gp than parental HepG2 cells and displayed strong resistance to doxorubicin, paclitaxel, and vinblastine. They found that both compounds reduced resistance at 5 and 10 µM without measurable cytotoxicity of their own at those concentrations. At 10 µM, compound 1 increased paclitaxel and vinblastine sensitivity by 118-fold and 198-fold, while 1a raised paclitaxel sensitivity to 335-fold and vinblastine sensitivity to 140-fold. Doxorubicin reversal was more modest, which is useful to see because it keeps the chemistry from being flattened into a one-number story. The research team then asked whether paclitaxel regained biological force once efflux pressure was relieved. Paclitaxel alone at 500 nM failed to drive appreciable apoptosis in HepG2/Dox cells, yet combining paclitaxel with either compound at 10 µM increased apoptosis and raised caspase-9 expression without changing caspase-8 in the same direction. That pattern places the regained drug response closer to mitochondrial apoptotic signaling than to a broad shift in death-receptor biology.

The researchers observed no meaningful change in P-gp protein expression after treatment with 1 or 1a, alone or with paclitaxel, yet they recorded clear retention of Rhodamine 123 inside HepG2/Dox cells. The study examined substrate status next through a Caco-2 monolayer assay, and both compounds showed efflux ratios below 2, with 0.8 for 1 and 0.9 for 1a, while digoxin behaved as a strong positive control. Docking and surface plasmon resonance completed the picture. Compound 1 and 1a bound P-gp with calculated energies of −8.2 and −8.4 kcal/mol, and direct binding constants of 5.5 and 3.7 µM, respectively. The two ligands contacted different residue sets, which is a quiet but important detail; similar phenotypes do not require identical binding geography.

What gives the study of Professor Yingjie Hu and colleagues important value is that these two polyoxypregnane aryl esters reversed resistance in a hepatoma model, and also the work distinguished transporter modulation from transporter suppression. When a compound restores drug response without lowering P-gp expression, the pharmacology shifts toward immediate interference with substrate handling, and that opens a cleaner route for combination dosing. A slower transcriptional or translational effect can still matter, of course, but it carries extra uncertainty about timing and off-target adaptation. It is worth noting the chemistry of the innovative compounds: compound 1 is a natural product isolated from the plant, while 1a adds a nicotinate group through semisynthesis. That modest structural change did not produce a trivial increment. It shifted reversal potency, binding strength, and permeability behavior in ways that remind us how sensitive P-gp modulation can be to local ester patterning on a steroidal scaffold. Compound 1a bound P-gp somewhat more tightly by SPR and docking, yet compound 1 showed much higher apparent permeability in the Caco-2 assay.  Binding strength and permeability did not move in parallel for the two analogues, which makes the comparison scientifically useful for future optimization.

There is a second important conclusion which is that much of the multidrug-resistance field has been shaped by canonical inhibitor programs that begin from synthetic chemical series and push toward transporter blockade. The work of Professor Yingjie Hu and team keeps natural-product logic alive, though in a more disciplined form than simple extract screening. The investigators started from a family history of polyoxypregnane activity, changed one ester feature with intent, and then linked cellular pharmacology to direct binding data. That sequence creates a useful template for future work on plant-derived MDR modulators: isolate, edit sparingly, test in resistant cells, and verify contact with the transporter itself. The study establishes a strong cell-based and biophysical foundation for these polyoxypregnane aryl esters as P-gp modulators. On that basis, compounds of this type could be considered as adjuvant leads for paclitaxel- or vinblastine-based treatment settings in which P-gp-driven efflux contributes to failure. Their non-substrate behavior in the Caco-2 model also keeps open the possibility that such molecules may prove useful in strategies aimed at improving the oral performance of P-gp substrate drugs.