Sónia
Rocha
a,
Inês
Santos
a,
M.
Luísa Corvo
b,
Eduarda
Fernandes
*a and
Marisa
Freitas
*a
aLAQV, REQUIMTE, Laboratory of Applied Chemistry, Department of Chemical Sciences, Faculty of Pharmacy, University of Porto, 4050-313 Porto, Portugal. E-mail: marisafreitas@ff.up.pt; egracas@ff.up.pt; Tel: +351 220428664 Tel: +351 220428675
bResearch Institute for Medicines, Faculdade de Farmácia, Universidade de Lisboa, 1649-003 Lisbon, Portugal
First published on 12th June 2025
In type 2 diabetes mellitus (DM), there is a combination of impaired insulin secretion and resistance in the target tissues. In the case of the liver, these events lead to decreased insulin effectiveness and increased glucagon levels, resulting in an imbalance that promotes excessive hepatic gluconeogenesis and glycogenolysis, contributing to hyperglycemia. Effective management of hyperglycemia and insulin resistance is crucial, underscoring the need for innovative liver-specific interventions. Polyphenols, renowned for their diverse biological activities, have emerged as promising candidates to treat type 2 DM. Based on a literature review spanning the last decade, this comprehensive systematic review thoroughly evaluates the effectiveness of polyphenols in targeting hepatic pathways for managing type 2 DM. The focus will be on assessing how polyphenols affect key targets, including protein tyrosine phosphatase 1B (PTP1B), the glucagon receptor, glucokinase, glycogen phosphorylase, and fructose 1,6-bisphosphatase. While there has been considerable attention on polyphenols as PTP1B inhibitors, studies on their impact on other targets have been comparatively limited. Notably, there is a lack of studies exploring polyphenols as glucagon receptor antagonists. Among polyphenols, flavonoids exhibit significant potential across diverse pathways, with hydroxy groups playing a pivotal role in their biological activities. However, further research, especially in cellular and animal models, is warranted to thoroughly validate their efficacy.
Various pharmacological interventions for DM aim to lower blood glucose levels. However, current therapeutic approaches have inherent limitations and adverse effects, including hypoglycemia, gastrointestinal disturbances, urinary tract infections, weight gain, and cardiovascular risk. These challenges underscore the need for novel treatment strategies to improve DM management.2 Given the liver's crucial role in glucose regulation, liver-targeted therapies are essential for glycemic control. However, the limited development of treatments directly addressing hepatic dysfunction reflects a prevailing challenge in current therapeutic strategies.3,4
A diverse range of bioactive compounds derived from plant-based sources, including vegetables, fruits, and edible leaves, has exhibited promising pharmacological properties, with polyphenols standing out as particularly significant.5–9 Despite the growing interest in the pharmacological potential of polyphenols, a comprehensive review of their effects on hepatic targets in the context of DM is still lacking. A comprehensive review of the existing literature on polyphenolic structures that have been investigated is crucial for identifying research gaps and potential therapeutic applications. This review aims to bridge this knowledge gap by summarizing current findings on polyphenols and their modulation of liver-related targets involved in glucose homeostasis and insulin resistance. Specifically, it explores polyphenols as inhibitors of protein tyrosine phosphatase 1B (PTP1B), glucagon receptor antagonists, glucokinase (GK) activators, glycogen phosphorylase (GP) inhibitors, and fructose 1,6-bisphosphatase (FBPase) inhibitors.
DM manifests clinically as hyperglycemia and can be classified, according to the Standards of Medical Care in Diabetes (2022) of the American Diabetes Association, into type 1 DM, type 2 DM, specific types of DM due to other causes, and gestational DM.16 Among all individuals with DM, 10–15% have type 1 DM,17 while over 90% have type 2 DM, making it the most prevalent form of the condition.12
In type 1 DM, hyperglycemia develops as the consequence of the loss of the pancreatic islet β cells. Two forms of type 1 DM have been described, type 1A (autoimmune) and type 1B (idiopathic) DM. Type 1B is far less common and its pathogenesis remains unclear. Type 1A represents around 70–90% of patients, and displays evidence of an autoimmune response against pancreatic β cells.17 Despite the exact mechanisms remaining unclear, recent studies indicate that the β cell response is more complex than being just a passive target. The classic view considers that β cell loss is mediated by autoimmune mechanisms, where autoreactive T cells erroneously destroy healthy β cells. However, new insights have been recently proposed, considering β cells as key contributors to the disease. β Cells rapidly respond to glucose variations to maintain normal glucose levels. However, this constant demand for insulin release results in endoplasmic reticulum stress and accumulation of misfolded proteins, creating vulnerability and β cell exposure to the immune system.18–20 These new insights into β cell vulnerability and type 1 DM aetiology are creating novel opportunities for treatment.18
Type 2 DM is characterized by the occurrence of insulin resistance in insulin-sensitive tissues and impaired insulin secretion due to progressive dysfunction of β cells.12 However, the interplay between these two defects remains undefined in terms of primary cause-and-effect. The prevailing paradigm defends the idea that insulin resistance precedes impaired insulin secretion.21 In this case, to offset the imbalance due to insulin resistance, β cells increase insulin secretion, leading to hyperinsulinemia. This chronic adaptation of β cells, together with environmental and genetic factors, results in β cell malfunction and a progressive decline of insulin secretion. Subsequently, hyperglycemia and type 2 DM develops when the β cells are incapable of compensating for this imbalance.12,21 The less cited paradigm proposes an opposing viewpoint, suggesting that the primary defect leading to the development of type 2 DM is insulin hypersecretion, or hyperinsulinemia. This condition ultimately results in insulin resistance and eventual β cell failure. However, it is important to note that these paradigms are not fully understood.21–23
![]() | ||
Fig. 1 Schematic representation of glucose metabolism during the postprandial and fasting states in skeletal muscle, adipose tissue, and liver. |
The liver represents one of the most important organs for glucose metabolism. Hepatocytes express several enzymes with actions depending on glucose concentrations. In the postprandial state, blood glucose enters hepatocytes by the glucose transporter type 2 (GLUT2), an energy-independent membrane-bound transporter (Fig. 2).27 Once inside the hepatocyte, glucose is phosphorylated to glucose 6-phosphate by GK, lowering intracellular glucose concentrations and increasing glucose uptake. Glucose transporters are not able to transport glucose 6-phosphate, so it remains retained inside the hepatocyte. Glucose 6-phosphate may be further metabolized in glycolysis, a critical ten-step process to generate energy. Also, in the postprandial state, glucose 6-phosphate is used to synthesize glycogen by glycogen synthase, the process known as glycogenesis (Fig. 2).28 The accumulation of glycogen, which consists of polymerized glucose, during the postprandial state in the liver, is an essential storage form of glucose, which can be utilized during fasting conditions.29 In the fasting state, hepatic glucose production accounts for nearly 90% of endogenous glucose production,30 providing glucose for other tissues such as the brain and muscle. Initially, hepatic glucose production starts with the release of glucose from stored glycogen, a process known as glycogenolysis (Fig. 2), with GP playing a major role. During prolonged periods of starvation, liver glycogen stores are depleted, and de novo glucose synthesis is initiated through gluconeogenesis. The primary source of new glucose is generated from non-carbohydrate precursors such as lactate, glycerol, and amino acids.29 These two pathways, gluconeogenesis and glycogenolysis, lead to an increase in glucose release into the bloodstream, with a consequent uptake of glucose by the peripheral tissues (Fig. 2).3
Typically, glucose levels within the normal range fall between 70 to 90 mg per dL. Type 2 DM develops when insulin levels are insufficient to counteract insulin resistance.12 In individuals with type 2 DM, the hepatic glucose production is increased and the suppression of glucose metabolism by insulin is impaired both in the fasted and postprandial states. However, DM is not solely characterized by insulin deficiency. Elevated levels of glucagon are characteristic of type 2 DM patients, leading to increased gluconeogenesis and glycogenolysis, contributing to hyperglycemia.3,12 Additionally, individuals with insulin resistance often exhibit enhanced de novo lipogenesis, resulting in fat accumulation in the liver and increased secretion of triglycerides, leading to elevated blood lipid levels.31 Type 2 DM frequently coexists with non-alcoholic fatty liver disease (NAFLD), now recognized as metabolic dysfunction-associated fatty liver disease (MAFLD). MAFLD encompasses steatosis (non-alcoholic fatty liver, NAFL) and non-alcoholic steatohepatitis (NASH), with increasing hepatic fibrosis, which can progress to cirrhosis, liver cancer, end-stage liver disease, and death.32,33 The global prevalence of MAFLD in patients with type 2 DM is estimated to be around 56%, posing an increased risk of adverse hepatic and extra-hepatic clinical outcomes.34
Prolonged hyperglycemia in DM is associated with micro- and macrovascular complications, such as damage to the kidneys, blood vessels and eyes. Treating hyperglycemia is of utmost importance in type 2 DM patients, with the liver being an important target organ for glucose homeostasis. Therefore, drugs that can target glucose metabolism or glucose uptake in the liver have the potential to improve hyperglycemia.3 The increased hepatic glucose output, a hallmark of liver dysregulation in type 2 DM, is not directly addressed by the currently prescribed antidiabetic medications, except for metformin.35 This biguanide compound derivative, unlike most commercial drugs, is derived from a natural product used in herbal medicine, the plant Galega officinalis, and is the most widely prescribed drug for type 2 DM therapy. Despite being used clinically for 60 years, its exact mechanism of action remains incompletely understood.36 Metformin acts primarily through the improvement of blood glucose levels by suppressing hepatic gluconeogenesis.30 However, despite being the first-line agent for the treatment of type 2 DM, metformin is associated with various side effects, with up to 25% of patients experiencing gastrointestinal symptoms such as abdominal pain, diarrhea, nausea and vomiting, and approximately 5% are unable to tolerate metformin.37 It is essential to find novel targets for decreasing hepatic glucose production or to promote liver glucose storage with minimal side effects. Other liver-targeted agents have been investigated in the clinic for the treatment of type 2 DM, including PTP1B inhibitors, glucagon receptor antagonists, GK activators, GP inhibitors, and FBPase inhibitors.3
In the cells, protein phosphorylation occurs mainly on tyrosine (Tyr), serine (Ser) and threonine (Thr) residues, with protein kinases and protein phosphatases being the two super-enzyme families responsible for phosphorylation and dephosphorylation, respectively.40 Each step in the insulin signaling pathway involves a reversible enzymatic reaction, where the activated kinases of the pathway can be dephosphorylated by a phosphatase to stop their action. The discovery of the tyrosine kinase activity of the IR and IRS prompted a search for tyrosine phosphatases that could terminate their activation. PTP1B, a member of the protein tyrosine phosphatases super-family, is a negative regulator of the insulin signaling pathway through the dephosphorylation of the tyrosine residues of IR and IRS (Fig. 3).41 In fact, since PTP1B is a negative regulator of the insulin signaling pathway, increasing evidence demonstrates that inhibitors of this enzyme might increase the levels of phosphorylation of IR and its substrates, improving insulin sensitivity. Therefore, PTP1B is considered a potential target for the management of type 2 DM.42 PTP1B is an intracellular PTP comprising 431 amino acids with a molecular size of 50 kDa. It is ubiquitously expressed, being localized at the cytoplasmic face of the endoplasmic reticulum (Fig. 3). The catalytic domain of PTP1B possesses near 40% sequence homology with the other PTPs.43 In addition to insulin signaling, PTP1B also displays actions in the leptin signaling pathway through the dephosphorylation of Janus kinase 2 (JAK2), leading to its deactivation. Therefore, PTP1B inhibition could decrease leptin resistance, being a possible therapeutic strategy for weight loss.40,44 Obesity is one of the epidemics of the 21st century and is associated with the development of type 2 DM and MAFLD, among other adverse pathological conditions.45 It is considered one of the major risk factors for developing type 2 DM, where around 90% of type 2 DM patients are overweight or obese.46,47 Indeed, as PTP1B can play several roles in different signaling pathways, increasing attention has been dedicated to this enzyme as an interesting therapeutic target.40,44
Up to now, only a few PTP1B inhibitors have reached clinical trials, including trodusquemine, ertiprotafib, and JTT-551. However, these inhibitors were discontinued from clinical trials due to reduced efficacy, lack of specificity, and adverse side effects.40,43 Taking this into account, it is crucial to find novel effective, selective, and safe PTP1B inhibitors.
In the hepatocyte, glucagon binds to its seven-transmembrane receptor, a G protein-coupled receptor that can be also found in pancreatic β cells, activating adenylate cyclase to increase intracellular levels of cyclic adenosine monophosphate (cAMP), which in turn stimulates protein kinase A (PKA) signaling. PKA activation subsequently induces cAMP-response element-binding protein (CREB) and its coactivators CREB-regulated transcriptional coactivator 2 (CRTC2) and CREB-binding protein (CBP), thereby activating gluconeogenic enzymes. The binding of glucagon also activates the phospholipase C (PLC)/inositol triphosphate (IP3)/Ca2+ pathway, responsible for increasing the intracellular Ca2+ levels. Both pathways mediate the hyperglycemic effects of glucagon (Fig. 4).30,51,52
Targeting the glucagon receptor represents a promising approach under investigation to reduce hepatic glucose production. Research findings have demonstrated that mice lacking glucagon receptors in the liver exhibit improved glucose tolerance, elevated plasma glucagon levels and higher insulin sensitivity. Some evidence demonstrates that the hepatic glucagon receptor is linked to the development of pancreatic α cells, where α cell hyperplasia was observed in mice lacking the glucagon receptor. In addition, besides the glucose regulatory effect, pharmacological inhibition of the glucagon receptor also plays a role in lipid metabolism. However, inconsistent results have been observed, with both elevations and reductions in blood cholesterol levels.53 It remains unclear if this effect is translated to humans, highlighting the need for additional studies on this subject.30
Bay 27-9955, MK-0893 and LY2409021 are three glucagon receptor antagonists already studied in human clinical trials.30 Bay 27-9955 decreased fasting glucose levels in humans, but the trials were discontinued. MK-0893 is a selective and reversible competitive glucagon receptor antagonist compound that showed improvements in hyperglycemia. However, it has been observed to cause increases in plasma cholesterol levels. LY2409021 showed a promising reduction in glucose levels in phase I and phase II clinical trials, and clinical development is ongoing.3,30 Recent studies with LY2409021 showed that this glucagon receptor antagonist was able to decrease fasting plasma glucose concentrations, with no improvements in postprandial glucose tolerance in type 2 DM patients.54 While results with glucagon receptor antagonists suggest that this target could be a promising strategy in DM, further research is needed.
In the human body, GK is mainly expressed in the pancreatic β cells and liver cells. This enzyme is responsible for the first step of glucose metabolism, catalyzing the phosphorylation of D-glucose to glucose 6-phosphate.56 In pancreatic β cells, GK acts as a glucose sensor, controlling the threshold for glucose-stimulated insulin secretion. In type 2 DM individuals, insulin secretion is impaired in part due to the reduced activity of GK. Therefore, increased levels of glucose are required to initiate insulin secretion. In hepatocytes, this enzyme plays a key role in glucose uptake and hepatic glucose regulation,57 and its activity is regulated through an interaction with the GK regulatory protein (GKRP), which binds and inactivates GK in the nucleus. In the postprandial state, GK is released to the cytoplasm following GKRP–GK complex disruption, stimulating glycolysis, glycogen production and storage and de novo lipogenesis (Fig. 5). In the fasting state, GKRP inactivates the enzyme, and glycogenolysis and gluconeogenesis pathways are activated to produce glucose.58 In type 2 DM individuals, it is described an impairment of near 50% of the GK activity during the progression of the disease, contributing to the reduction of glycogen storage and improper glucose regulation.57
![]() | ||
Fig. 5 Molecular regulation of GK in hepatic cells. GK: glucokinase; GKRP: GK regulatory protein; GLUT2: glucose transporter 2. |
GK activation is a potential intervention for glucose homeostasis. GK activation targeting the liver stimulates hepatic glucose uptake and glycogen production and storage, inhibiting glycogenolysis. The activation of GK in pancreatic β cells stimulates the secretion of insulin. The net effect of GK actions in the liver and β cells, in addition to its reduced activity in type 2 DM individuals, has led to the development of dual activators targeting both the liver and pancreas, or only selective activators targeting only one tissue.35,59 However, GK activators have not reached clinical use due to adverse side effects, including hypoglycemia, toxicity, and increased plasma concentration of triglycerides. Still, these effects might be preventable with patient selection and liver-selective GK activators.59 The liver-selective GK activator PF-04991532 has shown favorable effects on glucose regulation in diabetic rats. However, additional studies need to be conducted due to the occurrence of oxidative metabolites of the compound found in the human plasma. Additional clinical studies are underway with another liver-selective GK activator, TTP399.3 Recent clinical studies with TTP399 have demonstrated promising effects.60,61 These clinical investigations aim to assess whether glycemic control is improved in patients with type 1 DM, without increasing hypoglycemia or ketoacidosis. The findings indicate enhanced glycemic control without elevating the incidence of hypoglycemia, and reduced occurrence of diabetic ketoacidosis.60,61
A different approach for GK activators is to inhibit the binding to GKRP. This approach could be beneficial since liver-selective GKRP inhibition decreases glucose levels without the risk of hypoglycemia. The compound AMG-3969, which can dissociate GK and its receptor, showed promising results in a diabetic mice model, decreasing blood glucose levels, with no effect in normal mice.3 These effects with liver-selective GK activators or liver-selective GKRP inhibitors have been demonstrating encouraging results as a novel alternative to be used as antidiabetic agents.
GP exists in two interconvertible forms, the active phosphorylated GPa form, and the inactive dephosphorylated GPb form.63 GP activation requires a cascade mechanism in the hepatocyte, which starts with the binding of glucagon to its receptor, activating the cAMP/PKA signaling pathway, already mentioned in section 3.2. PKA phosphorylates phosphorylase kinase (PhK), which in turn activates GP by serine-14 phosphorylation (Fig. 6).29,62
Three different isoforms of GP have been identified, involving different tissue locations and physiological functions, namely in the brain (bGP), liver (lGP), and muscle (mGP). bGP provides an energy supply of glucose for periods of hypoglycemia, mGP delivers energy during muscle contraction, and lGP ensures glucose release from hepatic glycogen to other parts of the body.64,65 Therefore, as a key isoform involved in glucose regulation, lGP is a target for new possible pharmaceutical interventions. The inhibition of lGP may reduce plasma glucose levels by increasing glucose storage in the form of glycogen, reducing hyperglycemia.62 In addition, as glucose is a physiological regulator of lGP, inhibitors of lGP offer an appealing advantage. They demonstrated the ability to inhibit the enzyme at elevated blood glucose levels, but their potency diminishes when blood glucose levels decrease, thereby reducing the risk of hypoglycemia.3 However, there is a sequence homology among brain, muscle and liver isoforms of around 80%, and around 100% homology in the catalytic site of the enzyme.66 This high homology among isoforms is posing challenges in the search for a selective inhibitor of lGP. mGP inhibition is associated with serious adverse side effects in muscle due to the accumulation of glycogen in the tissues. Hence, it is essential to find selective and safe GP inhibitors.62
CP-368962 is a GP inhibitor that has been studied in clinical trials, showing hypoglycemic effects in type 2 DM individuals. However, after 4 weeks of treatment with CP-368962, the hypoglycemic effect was lost, indicating reduced durability of the effect.3 Therefore, additional studies are needed to identify novel GP inhibitors.
![]() | ||
Fig. 7 Representation of the hepatic gluconeogenesis process. FBPase: fructose 1,6-bisphosphatase; GLUT2: glucose transporter 2; PEP: phosphoenolpyruvate; TCA: tricarboxylic acid. |
MB07803 is a FBPase inhibitor already studied in clinical trials in type 2 DM patients. This inhibitor leads to the reduction of blood glucose levels. However, some patients experienced nausea and vomiting at higher doses, highlighting the need for additional studies with this compound. Furthermore, exploring the potential of novel FBPase inhibitors is essential.3
Polyphenols are naturally occurring compounds found in plants, comprising more than 8000 structures discovered to date in several plant species, including vegetables, fruits, grains, and beverages.2,72 Polyphenols comprise an essential part of the human diet, being responsible for some sensory and nutritional aspects of plant foods, including astringency, color and odor. As natural compounds synthesized by plants, these secondary metabolites are mainly involved in the attraction of pollinators, defense against ultraviolet radiation and protection against pathogens.73
Polyphenols consist of one or more phenolic groups in a molecule, varying from a single phenolic structure to a complex polymer with a high molecular mass.2,72 Polyphenols can be categorized in diverse forms due to the high diversity and distribution. The most common polyphenols can be classified according to the number of phenol rings and the structural elements that bind these rings into phenolic acids [benzoic acid derivatives (C1–C6) and cinnamic acid derivatives (C3–C6)], stilbenes (C6–C2–C6), flavonoids (C6–C3–C6, flavanones, flavones, flavanols, flavonols, isoflavones and anthocyanidins) and lignans (C6–C4–C6) (Fig. 8). Flavonoids are the most common and abundant group among polyphenols, with phenolic acids being the second most abundant group.2,74,75
![]() | ||
Fig. 8 Schematic representation of the general backbone structures illustrating phenolic acids, stilbenes, flavonoids, and lignans. R![]() |
Polyphenols have been recognized for numerous pharmacological activities, including antioxidant, cardioprotection, anticancer, anti-inflammatory, antimicrobial, anti-ageing, anti-obesity, and antidiabetic activities, among others, as reviewed by various authors.5–9 Increasing attention has been dedicated to the study of the antidiabetic activity of polyphenols, which can act on several different targets, including α-amylase and α-glucosidase,76 dipeptidyl peptidase-4,77 sodium-glucose cotransporter 2,78 and proliferator-activated receptor γ,79 among others. However, the studies with isolated polyphenols in clinical trials are still limited and often controversial, and therefore further studies are required to better understand the potential of these compounds as antidiabetic agents in humans.80 The limited bioavailability and extensive metabolism of these compounds have constrained their potential in vivo effects, making this a key topic of growing scientific interest that has been extensively addressed in several reviews.2,81–84 In their natural form, dietary polyphenols are often present as glycosides, esters, or complex polymers, which are poorly absorbed in the human gastrointestinal tract.84 Upon ingestion, these compounds undergo extensive transformation by intestinal enzymes and/or by colonic microflora before absorption, resulting in metabolites with substantial structural modifications. Following absorption, these metabolites reach the liver via the portal circulation, where they are further processed. The resulting hepatic metabolites then enter systemic circulation and are distributed to the peripheral tissues, where they may exert biological effects.2,82 The chemical structure of each polyphenol largely determines its absorption, distribution, metabolism, and excretion (ADME).84 These metabolic complexities contribute to the challenges of translating promising in vitro findings into consistent clinical outcomes. Despite the growing body of research, clear conclusions on polyphenol bioavailability remain limited. Reported plasma concentrations following normal dietary intake rarely exceed nanomolar to low micromolar levels,85,86 which contrasts with the higher concentrations commonly used in in vitro studies. Therefore, in clinical settings, it is essential to consider the most effective forms of polyphenol administration to enhance their therapeutic potential. Numerous strategies are currently being explored to improve their bioavailability, with nanotechnology showing promising results in increasing their efficacy and stability in vivo.2
Finding new alternatives to the current antidiabetic medications is required to manage type 2 DM with reduced side effects, and several polyphenols have been demonstrated to be promising compounds. As such, the main purpose of the present review is to highlight the effect of polyphenols on liver targets currently investigated in the clinic for the treatment of type 2 DM, including PTP1B, glucagon receptor, GK, GP, and FBPase.
Chemical structure | Compound source | Model | Effect | Ref. |
---|---|---|---|---|
↑ means “increase”, ↓ means “reduction”. IC50: half-maximal inhibitory concentration, IRS2: insulin receptor substrate 2, PTP1B: protein tyrosine phosphatase 1B, STZ: streptozotocin, DM: diabetes mellitus. | ||||
In vitro studies | ||||
![]() |
Commercial | Non-cellular enzymatic study (human PTP1B) | IC50 = 0.18 ± 0.02 mg mL−1 | 100 |
![]() |
Commercial | Non-cellular enzymatic study | IC50 = 7.62 ± 0.21 μM | 89 |
![]() |
Commercial | Non-cellular enzymatic study (human PTP1B) | IC50 = 37.14 ± 0.07 μM | 88 |
![]() |
Isolated from Anoectochilus chapaensis | Non-cellular enzymatic study (human PTP1B) | Isorhamnetin: IC50 = 1.75 ± 0.02 μM | 102 |
Isorhamnetin-3-O-β-D-glucoside: IC50 = 1.16 ± 0.03 μM | ||||
Isorhamnetin-3-O-β-D-rutinoside: IC50 = 1.20 ± 0.05 μM | ||||
![]() |
Isolated from Epimedium koreanum | Non-cellular enzymatic study (human PTP1B) | IC50 = 9.94 ± 0.15 μM | 103 |
![]() |
Isolated from Xanthium strumarium | Non-cellular enzymatic study | IC50 = 8.9 ± 0.7 μM | 104 |
![]() |
Isolated from Artemisia capillaris | Non-cellular enzymatic study (human PTP1B) | IC50 = 139.75 ± 2.97 μM | 105 |
![]() |
Commercial | Non-cellular enzymatic study (human PTP1B) | IC50 = 7.76 ± 0.21 μM | 106 |
![]() |
Isolated from Smilax china L. | Non-cellular enzymatic study (human PTP1B) | IC50 = 0.92 ± 0.19 μM | 90 |
![]() |
Not mentioned | Non-cellular enzymatic study | IC50 = 17.5 ± 2.6 μM | 91 |
![]() |
Isolated from Acer ginnala Maxim. | Non-cellular enzymatic study (human PTP1B) | IC50 = 3.46 ± 0.05 μM | 107 |
![]() |
Synthesis | Non-cellular enzymatic study (human PTP1B) | 1: IC50 = 16 ± 2 μM | 96 |
2: IC50 = 10 ± 1 μM | ||||
![]() |
Isolated from Quercus liaotungensis | Non-cellular enzymatic study (human PTP1B) | IC50 = 1.03 ± 0.12 μM | 87 |
![]() |
Isolated from Agrimonia pilosa | Non-cellular enzymatic study (human PTP1B) | Ellagic acid: IC50 = 7.14 ± 1.75 μM | 108 |
Apigenin 7-O-β-D-glucuronide: IC50 = 7.73 ± 0.24 μM | ||||
![]() |
Isolated from Angelica keiskei | Non-cellular enzymatic study | IC50 = 0.82 ± 0.09 μg mL−1 | 109 |
![]() |
Synthesis | Non-cellular enzymatic study (human PTP1B) | IC50 = 0.57 ± 0.2 μM | 101 |
![]() |
Synthesis | Non-cellular enzymatic study | 96.31 ± 1.76% at 20 μg mL−1 | 110 |
![]() |
Isolated from Glycyrrhiza glabra | Non-cellular enzymatic study | IC50 = 6.0 ± 1.4 μM | 97 |
![]() |
Isolated from Salvia amarissima Ortega | Non-cellular enzymatic study (human PTP1B) | IC50 = 62.0 ± 4.1 μM | 111 |
![]() |
Isolated from Hypericum scabrum | Non-cellular enzymatic study | IC50 = 2.19 ± 0.2 μM | 112 |
![]() |
Isolated from Silybum marianum | Non-cellular enzymatic study (human PTP1B) | IC50 = 6.79 ± 0.22 μM | 113 |
![]() |
Isolated from Eremophila denticulata | Non-cellular enzymatic study (human PTP1B) | 76.1 ± 12.4% at 100 μM | 114 |
![]() |
Commercial | Non-cellular enzymatic study (human PTP1B) | IC50 = 6.70 ± 0.03 μM | 115 |
![]() |
Isolated from Geranium collinum | Non-cellular enzymatic study | IC50 = 0.23 ± 0.04 μg mL−1 | 116 |
![]() |
Isolated from Coreopsis tinctoria Nutt. | Non-cellular enzymatic study | IC50 = 7.73 ± 0.48 μg mL−1 | 117 |
![]() |
Isolated from Amomum tsao-ko | Non-cellular enzymatic study | IC50 = 56.4 ± 5.0 μM | 118 |
![]() |
Isolated from Eremophila bignoniiflora | Non-cellular enzymatic study (human PTP1B) | IC50 = 41.4 ± 1.4 μM | 119 |
![]() |
Synthesis | Non-cellular enzymatic study | IC50 = 2.37 ± 0.37 μM | 120 |
![]() |
Isolated from Livistona chinensis | Non-cellular enzymatic study (human PTP1B) | IC50 = 9.41 ± 0.08 μM | 121 |
![]() |
Isolated from Helminthostachys zeylanica | Non-cellular enzymatic study (human PTP1B) | IC50 = 0.6 ± 0.2 μM | 122 |
![]() |
Isolated from Euonymus alatus | Non-cellular enzymatic study (human PTP1B) | IC50 = 13.7 ± 0.2 μM | 123 |
![]() |
Isolated from Ficus tikoua | Non-cellular enzymatic study | IC50 = 11.16 ± 1.88 μM | 124 |
![]() |
Isolated from Psoralea corylifolia | Non-cellular enzymatic study | IC50 = 10.3 ± 0.9 μM | 125 |
![]() |
Synthesis | Non-cellular enzymatic study | IC50 = 1.6 ± 0.9 μM | 126 |
![]() |
Isolated from Flemingia philippinensis | Non-cellular enzymatic study (human PTP1B) | IC50 = 2.4 ± 0.3 μM | 127 |
![]() |
Isolated from the roots of Sophora flavescens | Non-cellular enzymatic study | IC50 = 5.26 ± 0.24 μM | 128 |
![]() |
Isolated from Paulownia tomentosa | Non-cellular enzymatic study | IC50 = 1.9 ± 0.1 μM | 98 |
![]() |
Isolated from Macaranga denticulata | Non-cellular enzymatic study | IC50 = 14.0 ± 1.2 μM | 99 |
![]() |
Isolated from Erythrina subumbrans | Non-cellular enzymatic study | IC50 = 3.21 ± 0.24 μM | 129 |
![]() |
Isolated from Dodonaea viscosa | Non-cellular enzymatic study | IC50 = 13.5 ± 0.3 μM | 130 |
![]() |
Isolated from Sophora alopecuroides L. | Non-cellular enzymatic study (human PTP1B) | Inhibition = 95.22% at 0.1 μg mL−1 | 131 |
![]() |
Isolated from Selaginella uncinata | Non-cellular enzymatic study | IC50 = 4.6 ± 0.5 μM | 132 |
![]() |
Isolated from Ficus racemosa | Non-cellular enzymatic study (human PTP1B) | IC50 = 2.5 ± 0.2 μM | 133 |
![]() |
Isolated from Selaginella tamariscina | Non-cellular enzymatic study (human PTP1B) | IC50 = 4.5 ± 0.1 μM | 134 |
![]() |
Isolated from Silybum marianum | Non-cellular enzymatic study (human PTP1B) | Silybin A: IC50 = 1.54 ± 0.22 μM | 135 |
Isosilybin B: IC50 = 1.37 ± 0.22 μM | ||||
![]() |
Isolated from Cudrania tricuspidata | Non-cellular enzymatic study (human PTP1B) | Cudratricusxanthone N: IC50 = 2.0 ± 0.4 μM | 136 |
Cudracuspixanthone A: IC50 = 1.9 ± 0.4 μM | ||||
![]() |
Isolated from Polygonum multiflorum | Non-cellular enzymatic study (human PTP1B) | 1: IC50 = 2.1 ± 0.08 μM | 137 |
2: IC 50 = 1.9 ± 0.04 μM | ||||
![]() |
Isolated from Morus laevigata | Non-cellular enzymatic study | IC50 = 0.87 ± 0.09 μM | 138 |
![]() |
Isolated from Morus nigra | Non-cellular enzymatic study | Morunigrines A: IC50 = 1.8 ± 0.2 μM | 139 |
Morunigrines B: IC50 = 1.3 ± 0.3 μM | ||||
![]() |
Isolated from Morus alba L. | Non-cellular enzymatic study (human PTP1B) | IC50 = 1.90 ± 0.12 μM | 140 |
![]() |
Isolated from Allium cepa | Non-cellular enzymatic study | 1: IC50 = 1.68 ± 0.02 μM | 141 |
2: IC50 = 1.13 ± 0.01 μM | ||||
![]() |
Synthesis | Non-cellular enzymatic study | IC50 = 0.91 ± 0.33 μM | 142 |
![]() |
Isolated from Morus alba | Non-cellular enzymatic study | IC50 = 2.85 ± 0.30 μM | 143 |
![]() |
Isolated from Rheum undulatum | Non-cellular enzymatic study (human PTP1B) | Piceatannol: IC50 = 4.81 ± 0.21 μM | 144 |
δ-Viniferin: IC50 = 4.25 ± 0.02 μM | ||||
![]() |
Isolated from Cajanus cajan | Cellular study with HepG2 cells overexpressing PTP1B | No differences in PTP1B mRNA expression | 145 |
↓ PTP1B protein expression | ||||
![]() |
Isolated from Juniperus chinensis | Non-cellular enzymatic study (human PTP1B) | IC50 = 25.27 ± 0.14 μM | 146 |
Cellular study with insulin-resistant HepG2 cells | ↓ PTP1B protein expression | |||
![]() |
Isolated from Chrysanthemum morifolium | Non-cellular enzymatic study (human PTP1B) | Diosmetin 7-glucoside: 27.74 ± 2.10% at 1 μM | 147 |
Diosmin: 36.26 ± 1.62% at 1 μM | ||||
Diosmetin: 22.81 ± 2.07% at 1 μM | ||||
Luteolin: 19.80 ± 2.18% at 1 μM | ||||
Apigenin: 29.77 ± 1.48% at 1 μM | ||||
Acacetin: 38.17 ± 1.62% at 1 μM | ||||
Cellular study with CHO-K1 cells | ↓ PTP1B protein expression | |||
![]() |
Isolated from mulberry leaves | Non-cellular enzymatic study | 1: IC 50 = 4.53 ± 0.31 μM | 148 |
2: IC50 = 10.53 ± 1.76 μM | ||||
Cellular study with insulin-resistant HepG2 cells | 1: ↓ PTP1B protein expression | |||
![]() |
Isolated from blueberry fruits | Non-cellular enzymatic study (human PTP1B) | Procyanidin B1: IC 50 = 0.60 ± 0.042 μM | 149 |
Procyanidin B2: IC50 = 4.79 ± 0.023 μM | ||||
Cellular study with insulin-resistant HepG2 cells | ↓ PTP1B protein expression | |||
![]() |
Commercial | Cellular study with insulin-resistant HepG2 cells | ↓ PTP1B protein expression | 92 |
![]() |
Commercial | Cellular study with insulin-resistant HepG2 cells | ↓ PTP1B protein expression | 150 |
![]() |
Synthesis | Non-cellular enzymatic study (human PTP1B) | Inhibition = 18.09 ± 3.39% at 1 μM | 151 |
In vitro cellular study with CHO-K1 cells | ↓ PTP1B protein expression | |||
![]() |
Isolated from Prunus davidiana | Non-cellular enzymatic study (human PTP1B) | IC50 = 5.5 ± 0.29 μM | 93 |
Cellular study with insulin-resistant HepG2 cells | ↓ PTP1B protein expression | |||
![]() |
Commercial | Non-cellular enzymatic study (human PTP1B) | IC50 = 1.23 ± 0.11 μM | 94 |
Cellular study with insulin-resistant HepG2 cells | ↓ PTP1B protein expression | |||
![]() |
Isolated from blueberry fruits | Non-cellular enzymatic study (human PTP1B) | Cyanidin-3-arabinoside: IC50 = 8.91 ± 0.63 μM | 95 |
Cellular study with HepG2 cells overexpressing PTP1B | ↓ PTP1B protein expression | |||
In vivo studies | ||||
![]() |
Synthesis | DM high-fat diet/streptozotocin-induced using C57BL/6J mice (50 and 100 mg kg−1 of compound, administration for 6 weeks) | ↓ PTP1B protein expression | 152 |
![]() |
Synthesis | High-fat diet using C57BL/6J mice (10 mg kg−1 day−1 of compound, gastric gavage for 10 weeks, 5 days a week) | ↓ PTP1B protein expression | 153 |
![]() |
Not mentioned | DM induced by STZ using Wistar rats (50 mg kg−1 of compound, administration daily by gavage daily for 8 weeks) | ↓ PTP1B gene expression | 154 |
![]() |
Commercial | High-fat diet using C57BL/6J mice (20 mg kg−1 of compound, administration by dietary supplementation for 15 weeks) | ↓ PTP1B protein expression | 155 |
![]() |
Commercial | IRS2-deficient using C57BL/6 and 129/Sv mice (2.5 mg kg−1 day−1 of compound, orally administered for 8 weeks via drinking water) | ↓ PTP1B gene expression and activity | 156 |
![]() |
Commercial | High-fat and high-fructose diet using C57BL/6 mice (2 g L−1 of compound, treatment through drinking water for 16 weeks) | ↓ PTP1B protein expression | 150 |
![]() |
Synthesis | Goto-Kakizaki rats (1, 5 and 10 mg kg−1 of compound, subcutaneously daily for 14 days) | =PTP1B protein expression on liver | 157 |
Proença et al.96 studied the in vitro inhibitory activity of a panel of 36 structurally related flavonoids. The authors concluded that the presence of both 7- and 8-OBn groups in the A ring, together with the presence of both 3′- and 4′-OMe groups in the B ring and the 3-OH group in the C ring, are essential for the inhibitory activity of PTP1B. The authors highlight that the presence of methoxy groups, together with hydroxy groups, are determinant for the intended activity,96 contrasting with Jing Xu et al.87 Although both studies investigate the human enzyme, using identical methodologies, the discrepant conclusions may stem from variations in enzyme–substrate ratios, as well as the additional factor of NaOH addition to stop the reaction in the study by Jing Xu et al.87
Upon examination of Table 1, it is evident that the presence of functional groups containing hydrocarbon chains, such as the prenyl and geranyl groups, is significant for the inhibition of the enzyme PTP1B. Particularly, Wei Li et al.97 studied 42 polyphenol derivatives, comprising different groups, including chalcones and flavonoids. Among the compounds investigated, flavonoid glycosides exhibited negligible inhibitory activity, diverging from other studies represented in Table 1. Nevertheless, prenylated compounds demonstrated noteworthy effectiveness. The authors emphasized the critical role of prenyl groups and hydroxy moieties in ortho-position for facilitating the activity of these compounds, eliciting inhibition of PTP1B through diverse modes.97 Geranylated flavonoids also demonstrated promising inhibitory activities, indicating that geranyl is an important substitution for the inhibition of PTP1B.98,99
Among the selected studies of Table 1, it is worth highlighting that only two specifically addressed the issue of polyphenol bioavailability [Rampadarath et al.,100, Zhao et al.,101]. Rampadarath et al.100 conducted an in silico pharmacokinetic analysis of orientin, assessing its ADME properties and drug-likeness using the SwissADME platform. The authors report that a potential therapeutic compound should ideally exhibit at least 10% oral bioavailability, a criterion that favors orientin, which showed a predicted value of 17%. While these computational results are promising and suggest orientin as a potentially bioavailable compound, no in vivo validation of these predictions was performed, thus limiting conclusions regarding its actual pharmacokinetic behavior. On the other hand, Zhao et al.101 carried out a more comprehensive pharmacokinetic profiling of the most active compound identified in their study in Sprague-Dawley rats. After intravenous administration at 2 mg kg−1, the polyphenol showed a long half-life (≈21 h). When given orally at a higher dose (20 mg kg−1), the compound reached a moderate level in the bloodstream over time, with a shorter half-life (≈9 h). These findings underscore the importance of pharmacokinetic characterization in evaluating the therapeutic viability of polyphenolic compounds. However, such studies remain scarce, and further investigations are essential.
Despite the substantial number of in vitro studies on polyphenols in the inhibition of PTP1B, relatively few in vivo studies have been published in the past decade. Notably, overall, all compounds exhibit hydroxy groups, belonging to different groups, including flavonoids, stilbenes, and phenolic acids. In the published in vivo studies, all tested polyphenols have shown the capability to decrease the expression of PTP1B.
Chemical structure | Compound source | Study method | Effect | Ref. |
---|---|---|---|---|
↑ means “increase”, ↓ means “reduction”. GK: glucokinase, IC50: half-maximal inhibitory concentration, STZ: streptozotocin, DM: diabetes mellitus. | ||||
In vitro studies | ||||
![]() |
Commercial | Cellular assay with HepG2 | Resveratrol: ↓ the GK activity | 158 |
Oxyresveratrol: no effect | ||||
Cyanidin-3-glucoside: ↑ GK activity | ||||
Cyanidin-3-rutinoside: ↑ GK activity | ||||
Resveratrol: ↓ GK protein expression | ||||
Oxyresveratrol: ↓ GK protein expression | ||||
Cyanidin-3-glucoside: ↑ GK protein expression | ||||
Cyanidin-3-rutinoside: ↑ GK protein expression | ||||
Cellular assay with MIN6 | Resveratrol: ↓ GK protein expression | 159 | ||
Oxyresveratrol: ↓ GK protein expression | ||||
Cyanidin-3-glucoside: ↑ GK protein expression | ||||
Cyanidin-3-rutinoside: ↑ GK protein expression | ||||
![]() |
Isolated from Selaginella tamariscina | Cellular assay with HepG2 | 1: ↑ GK protein expression | 165 |
2: ↑ GK protein expression | ||||
3: ↑ GK protein expression | ||||
![]() |
Not mentioned | Cellular assay with HepG2 | ↑ GK protein expression | 162 |
![]() |
Commercial | Cellular assay with INS-1E | Caffeic acid: no differences in GK mRNA expression | 160 |
Naringenin: ↑ GK mRNA expression under normoglycaemic and glucotoxic conditions | ||||
Quercetin: ↑ GK mRNA expression under glucotoxic conditions | ||||
![]() |
Commercial | Cellular assay with L6 | ↑ GK activity | 166 |
![]() |
Isolated from Selaginella tamariscina | Cellular assay with HepG2 | ↑ GK activity | 167 |
![]() |
Commercial | Cellular assay with HepG2 | ↑ GK activity | 168 |
![]() |
Commercial | Cellular assay with HepG2 | ↑ GK protein expression | 169 |
Ex vivo studies | ||||
![]() |
Commercial | Study with isolated perfused rat liver using Wistar rats | Hesperetin: ↓ liver GK activity | 170 |
Naringen: ↓ liver GK activity | ||||
Hesperidin: no effect | ||||
In vivo studies | ||||
![]() |
Commercial | DM induced by NA/STZ using Wistar rats (50 mg kg−1 of compound, daily administration by gastric intubation for 4 weeks) | ↑ mRNA expression of liver GK levels on NA/STZ-induced rats | 163 |
Not mentioned | DM induced by high fat and sugar diet, and STZ, using Sprague-Dawley rats (75 and 150 mg kg−1 of compound, administration by gavage, 6 days weeks, for 8 weeks) | ↑ liver GK protein expression on high fat and sugar diet, and STZ-induced rats | 162 | |
![]() |
Not mentioned | DM induced by high-fat diet and fructose using Wistar rats (50 mg kg−1 of compound, administration by intubation for 30 days) | ↑ liver GK levels on STZ-induced rats. No effect on normal rats | 171 |
![]() |
Commercial | STZ-induced Wistar rats (4, 8 and 16 mg kg−1 of compound, administration by intubation for 45 days) | ↑ liver GK levels on STZ-induced rats. No effect on normal rats | 172 |
![]() |
Isolated from Selaginella tamariscina | STZ-induced CD-1 mice (20 and 40 mg kg−1 of compound, for 8 weeks, type of administration not mentioned) | ↑ liver GK levels on STZ-induced rats. | 173 |
![]() |
Commercial | Megalobrama amblycephala fish fed with high-fat diet (0.04%, 0.36%, 1.08% of compound, administration hand-fed, 3 times a day, for 10 weeks) | ↓ mRNA expression of liver GK levels on fish fed with high-fat diet | 161 |
![]() |
Commercial | DM induced by high-fat diet and STZ using C57BL/6J mice (0.005% w/w of compound, administration orally for 5 weeks) | ↑ liver GK activity and mRNA expression levels on high-fat diet- and STZ-induced mice | 174 |
![]() |
Commercial | DM induced by alloxan using Kunming mice (5, 10, 20 mg kg−1 of compound, administration intragastrically, once a day for 28 days) | ↑ liver GK mRNA expression and protein levels on alloxan-induced mice. | 175 |
![]() |
Synthesis | Wistar rats treated with obesogenic diet, high in sucrose and fat (15, 30 mg kg−1 day−1 of compound, administration orally once a day, for 6 weeks) | ↑ liver GK levels on obesogenic rats, with the lower concentration of pterostilbene | 164 |
![]() |
Isolated | DM induced by alloxan and glucose, using Danio rerio zebrafish (2, 4, 10 μg mL−1 of compound, treatment with incubation for 24 h) | ↑ GK protein expression on alloxan- and glucose-induced zebrafish | 176 |
In two different in vitro studies using different cellular models,158,159 the same group studied the effects of two stilbenes, resveratrol and oxyresveratrol, and two flavonoids, cyanidin-3-glucoside and cyanidin-3-rutinoside, against the enzyme GK. The findings indicate that flavonoids exhibit greater promise in activating GK compared with stilbenes. Stilbenes either demonstrated no effect or led to a reduction in GK activity at higher concentrations.158,159 Bhattacharya et al.160 analyzed the effects of caffeic acid, naringenin and quercetin in the expression levels of GK. Remarkably, both flavonoids, naringenin and quercetin, were able to increase the levels of GK. Caffeic acid showed no effects in this study. It is also possible to verify that flavonoids proved to be more favourable, compared with the studied phenolic acid.160 One ex vivo study was performed with hesperidin, hesperetin and naringenin, using isolated perfused rat liver. Contrasting with the other studies, these flavonoids were not able to increase GK activity, showing no effect or even reduction in its levels.
Regarding the in vivo studies, several polyphenols were explored, including polydatin, ferulic acid, galangin, amentoflavone, resveratrol, myricitrin, formononetin, pterostilbene and theaflavin-3,3′-digallate, as represented in Table 2. Almost all compounds display only hydroxy groups in their backbones, except ferulic acid, formononetin and pterostilbene, which hold hydroxy and methoxy groups. In general, all compounds, including those with hydroxy and methoxy groups in their backbone, exhibited an increase in GK levels.
Interestingly, the stilbene resveratrol was also studied in vivo,161 showing a reduction in the mRNA expression of liver GK, consistent with the findings of in vitro cellular studies.158,159 However, other stilbenes, polydatin and pterostilbene, were also studied in in vivo models, and showed a capacity to increase GK levels.162–164 Particularly, the stilbene polydatin was studied in two different in vivo studies, revealing promising activity for the activation of GK.162,163 Additionally, Hao et al.162 investigated the activity of polydatin in an insulin-resistant HepG2 cell model, which also demonstrated an ability to increase the GK protein expression.
Chemical structure | Compound source | Study method | Effect | Ref. |
---|---|---|---|---|
↑ means “increase”, ↓ means “reduction”. GP: glycogen phosphorylase, IC50: half-maximal inhibitory concentration, STZ: streptozotocin, DM: diabetes mellitus. | ||||
In vitro studies | ||||
![]() |
Commercial and synthesis | Non-cellular enzymatic study (rabbit muscle GPa) | Norwogonin (IC50 = 13.2 ± 1.4 μM) | 62 |
Baicalein (IC50 = 23.5 ± 2.9 μM) | ||||
Baicalin (IC50 = 20.5 ± 2.5 μM) | ||||
Baicalein-7-methylether | ||||
(IC50 = 22.6 ± 0.4 μM) | ||||
![]() |
Commercial | Non-cellular enzymatic study (rabbit muscle GPb) | Chrysin (Ki = 19 μM) | 179 |
Flavopiridol (Ki = 1.24 μM) | ||||
![]() |
Resveratrol glucoside isolated from Paeonia clusii and tachioside isolated from Dorycnium pentaphyllum | Non-cellular enzymatic study (rabbit muscle GPb) | Resveratrol glucoside (IC50 = 258.4 ± 0.6 μM) | 185 |
Tachioside (IC50 = 571.1 ± 45.5 μM) | ||||
![]() |
Isolated from Seriphidium stenocephalum | Non-cellular enzymatic study (rabbit muscle GP) | Stenocephol (IC50 = 8.794 μM) | 186 |
![]() |
Commercial | Non-cellular enzymatic study (rabbit muscle GPa and GPb) | Ellagic acid (Ki = 13.4 ± 1.2 μM for GPb and 7.52 ± 0.36 μM for GPa) | 187 |
Gallic acid (Ki = 1.73 ± 0.16 mM for GPb and 3.86 ± 0.27 mM for GPa) | ||||
![]() |
Synthesis | Non-cellular enzymatic study (rabbit muscle GPa and GPb, and human liver GPa) | Quercetin (Ki = 43.52 ± 1.65 μM for human GP) | 180 |
Chrysin (Ki = 7.28 ± 0.09 μM for human GP) | ||||
1 (Ki = 7.39 ± 0.09 μM for human GP) | ||||
2 (Ki = 3.39 ± 0.22 μM for human GP) | ||||
3 (Ki = 3.89 ± 0.08 μM for human GP) | ||||
4 (Ki = 2.23 ± 0.08 μM for human GP) | ||||
5 (Ki = 21.36 ± 0.98 μM for human GP) | ||||
Cellular study with HepG2 | Flavonoid 4 ↓ glycogenolysis in hepatocytes (IC50 = 70 μM) | |||
In vivo studies | ||||
![]() |
Not mentioned | DM induced by high-fat diet and fructose using Wistar rats (50 mg kg−1 of compound, with administration by intubation for 30 days) | ↓ liver GP levels on STZ-induced rats. No effect on normal rats | 171 |
![]() |
Commercial | DM induced by NA/STZ using Wistar rats (100 mg kg−1 day−1 of compounds, administration by oral gavage for 4 weeks) | ↓ liver GP levels with the same effect in hesperidin- and quercetin-treated rats | 181 |
![]() |
Commercial | DM induced by high-fat diet and STZ using Wistar rats (100 mg kg−1 day−1 of compound, administration orally for 30 days) | ↓ liver GP levels on DM rats. No effect on normal rats | 188 |
![]() |
Commercial | DM induced by STZ using Wistar rats (20, 40, 80 mg kg−1 of compound, administration intragastrically for 30 days) | ↓ liver GP levels on DM rats. No effect on normal rats | 184 |
![]() |
Commercial | STZ–cadmium-induced diabetic nephrotoxic Wistar rats (1 mg kg−1 of compound, administered intraperitoneally once a day for 12 weeks) | ↓ liver GP levels on STZ–cadmium-induced rats. No effect on normal rats | 189 |
![]() |
Commercial | DM induced by nicotineamide/STZ using rats (100 mg kg−1 day−1 of compounds, by oral gavage for 4 weeks) | ↓ liver GP levels on nicotineamide/STZ-induced rats, with the same effect for naringenin and naringin. | 183 |
![]() |
Commercial | STZ-induced Wistar rats (25, 50, 100 mg kg−1 of compound, administered orally using an intragastric tube for 30 days) | ↓ liver GP levels on STZ-induced rats. No effect on normal rats. | 182 |
![]() |
Commercial | STZ induced Wistar rats (100 mg kg−1 of compound, administered orally daily for 30 days) | ↓ liver GP levels on STZ-induced rats. No effect on normal rats. | 190 |
![]() |
Isolated from Semecarpus anacardium | DM induced by high-fat diet and STZ using Wistar rats (20, 40, 80 mg kg−1 of compound, administration intragastrically for 30 days) | ↓ liver GP levels on high-fat diet and STZ-induced rats. No effect on normal rats. | 191 |
![]() |
Commercial | STZ-induced Wistar rats (20 mg kg−1 of compound, administered orally daily for 6 weeks) | ↓ liver GP levels on STZ-induced rats. | 192 |
Rocha et al.62 explored the activity of 17 polyphenols in vitro against the enzyme rabbit muscle GPa. The authors concluded that the hydroxylation of the A ring is determinant for the inhibitory activity, whereas methoxy substituents were disadvantageous for GPa inhibitory activity. The flavonoid norwogonin (Table 3) was the most promising compound among the panel of polyphenols tested, showcasing heightened inhibitory activity in the presence of glucose. This characteristic suggests a potential for reducing the risk of hypoglycemia, particularly relevant in diabetic contexts marked by elevated plasma glucose levels. The authors also studied the flavonoid chrysin (Table 3), one of the most active compounds of the study. This flavonoid was studied by other authors,179,180 being recognized as a potent GP inhibitor.
Chetter et al.180 studied the inhibitory potential of 10 flavonoids against rabbit muscle GPa and GPb, and human liver GPa. The flavonoid, which contains hydroxy groups at the C-5 and C-7 positions and a methyl at the C-3′ position (Table 3), was the most active compound of the group. This compound was also studied using hepatocarcinoma HepG2 cells, where it effectively inhibited endogenous GP activity.180 In this study, the flavonoid quercetin (Table 3) showed the highest IC50 among all the tested compounds. Comparably, Rocha et al.62 also studied quercetin, which showed no inhibitory activity.
Polyphenols were already studied using in vivo models for testing the inhibition of GP (Table 3), including ferulic acid, hesperidin, quercetin, theaflavin, naringin, myricetin, naringenin, naringin, hesperidin, tangeretin, a biflavonoid and hesperetin. In general, these polyphenols are hydroxylated, except ferulic acid, hesperidin, tangeretin and hesperetin. The flavonoid hesperidin was already explored in 2 different studies and demonstrated a significant reduction of the GP levels in both studies.181,182 Quercetin, explored in two in vitro studies,62,180 was also studied in vivo by Ali et al.,181 demonstrating its ability to reduce lGP levels in treated animals, exhibiting a similar effect to hesperidin.
Ahmed et al.183 explored the antidiabetic effects of two flavonoids, naringin and naringenin, using nicotineamide- and streptozotocin-induced diabetic rats. Both flavonoids reduced lGP levels, showing that the presence of the disaccharide at the C-7 of the A ring, when compared with naringin, was not valuable for the reduction of lGP levels.183 Naringin was also investigated by Pari and Chandramohan184 using diabetic rats induced with streptozotocin. The results are in accordance with Ahmed et al.,183 illustrating the ability of naringin to reduce lGP levels.
There are few reports on the literature in the last 10 years exploring the inhibitory potential of polyphenols, as represented in Table 4. Considering the studies using polyphenols as FBPase inhibitors, flavonoids are the most explored compounds. Pterostilbene, a stilbene, is the only compound studied as an FBPase inhibitor that does not belong to the flavonoid family.
Chemical structure | Compound source | Study method | Effect | Ref. |
---|---|---|---|---|
↑ means “increase”, ↓ means “reduction”. FBPase: fructose 1,6-bisphosphatase, IC50: half-maximal inhibitory concentration, STZ: streptozotocin, DM: diabetes mellitus. | ||||
In vitro studies | ||||
![]() |
Commercial | Non-cellular enzymatic study (human FBPase) | Baicalein (IC50 = 29 ± 3 μM) | 68 |
Scutellarein (IC50 = 38.2 ± 0.4 μM) | ||||
Herbacetin (IC50 = 8.7 ± 0.7 μM) | ||||
Gossypetin (IC50 = 24 ± 2 μM) | ||||
![]() |
Isolated from Desmodium caudatum | Non-cellular enzymatic study (human FBPase) | 8-Prenylquercetin (IC50 = 3.62 ± 0.37 μM) | 193 |
In vivo studies | ||||
![]() |
Commercial | STZ-induced Wistar rats (4, 8 and 16 mg kg−1 of compound, with administration by intubation for 45 days) | ↓ liver FBPase levels on STZ-induced rats. No effect on normal rats | 172 |
![]() |
Not mentioned | STZ-induced Swiss Albino mice (5 and 10 mg kg−1 of compound, intraperitoneal administration for 5 weeks) | ↓ liver FBPase levels on STZ-induced mice. No effect on normal rats | 194 |
![]() |
Commercial | DM induced by STZ using Wistar rats (20, 40, 80 mg kg−1 of compound, administered orally by intragastric intubation, daily for a period of 30 days) | ↓ liver FBPase levels on STZ-induced rats. No effect on normal rats | 184 |
![]() |
Commercial | STZ–cadmium-induced diabetic nephrotoxic Wistar rats (1 mg kg−1 of compound, administered intraperitoneally once a day for 12 weeks) | ↓ liver FBPase levels on STZ–cadmium-induced rats. | 189 |
↓ FBPase levels of normal rats | ||||
![]() |
Commercial | STZ-induced Wistar rats (25 and 50 mg kg−1 of compound, administered orally to for 30 days) | ↓ liver FBPase levels on STZ-induced rats. No effect on normal rats | 195 |
![]() |
Commercial | STZ-induced Wistar rats (25, 50, 100 mg kg−1 of compound, administered orally using an intragastric tube for 30 days) | ↓ liver FBPase levels on STZ-induced rats. No effect on normal rats | 182 |
![]() |
Commercial | STZ-induced Wistar rats (100 mg kg−1 of compound, administered orally daily for 30 days) | ↓ liver FBPase levels on STZ-induced rats. No effect on normal rats | 190 |
![]() |
Isolated from Semecarpus anacardium | DM induced by high-fat diet and STZ using Wistar rats (20, 40, 80 mg kg−1 of compound, administration intragastrically for 30 days) | ↓ liver FBPase levels on high-fat diet and STZ-induced rats. No effect on normal rats | 191 |
Proença et al.68 investigated the effect of 55 structurally related flavonoids, using an in vitro non-cellular assay with isolated human FBPase. The results showed that herbacetin (Table 4), was the most potent inhibitor of the study, with similar potency to the positive control AMP. The authors concluded that, in general, the addition of hydroxy substituents increased the inhibitory effect of the flavonoids, and methoxy substituents are disadvantageous for the intended effect. In particular, the authors reported that hydroxy groups at the C-3, C-4′, C-5, C-7, and C-8 positions, as well as the double bond between C-2 and C-3 and the 4-oxo function at the pyrone ring, are determinant for the inhibitory potential of the compounds. For example, the flavonoid quercetin was not able to inhibit the enzyme.68 Similarly, Zhang et al.193 studied the effect of hydroxylated flavonoids in an in vitro non-cellular assay, and verified that the existence of hydroxy groups together with a prenyl group were essential for the inhibitory activity of the flavonoids. In accordance with Proença et al.,68 the flavonoid quercetin was not able to inhibit the enzyme. However, when comparing quercetin with 8-prenylquercetin, the addition of the prenyl group at the C-8 position was essential for the effect.193 Proença et al.68 and Zhang et al.193 also studied in common the flavonoids apigenin and naringenin, showing no effect on both studies.
Some authors have already studied polyphenols in in vivo studies, including galangin, pterostilbene, naringin, myricetin, morin, hesperidin, tangeretin and a biflavonoid, as represented in Table 4. These polyphenols were able to reduce the levels of FBPase in streptozotocin-induced animals. In general, these polyphenols are hydroxylated, except pterostilbene and hesperidin, holding methoxy and hydroxy groups, and tangeretin, bearing only methoxy groups. Proença et al.68 studied the flavonoids galangin, myricetin, and morin, revealing their lack of inhibitory activity against the human enzyme. The different results observed are possibly due to the distinct complexity associated with both in vitro and in vivo experiments.
Polyphenols as PTP1B inhibitors emerge as the most extensively studied category among the reviewed targets. In contrast, research on the other targets, including glucagon receptor antagonists, GK activators, GP inhibitors, and FBPase inhibitors, remains relatively limited. Notably, studies investigating polyphenols as glucagon receptor antagonists are virtually absent, highlighting a critical gap in the literature. Flavonoids have garnered particular interest across various targeted pathways, with the presence of hydroxyl groups playing a key role in their activities. However, despite the significant number of studies, cellular-based investigations remain scarce for all targets, limiting mechanistic insights into their efficacy and specificity. The development of in vitro mechanistic studies, particularly using cellular models, 3D cultures, and co-cultures, would be valuable in addressing this gap, as no such studies were identified within our search criteria over the past decade.
Additionally, it is important to acknowledge that in vitro enzymatic studies are conducted under varying conditions, with differences in enzyme concentrations and buffer compositions, while in vivo studies are performed in distinct models with variations in study duration, dosing regimens, and administration routes. These factors introduce challenges in making direct comparisons between studies. Moreover, polyphenols are subject to extensive metabolism and rapid elimination in the human body, which may limit their therapeutic potential in vivo despite promising in vitro results. Despite the increasing research on polyphenols, pharmacokinetic studies remain scarce, making it difficult to fully understand their ADME profiles. To address these challenges, future research should prioritize pharmacokinetic characterization and explore novel delivery systems aimed at enhancing bioefficacy and stability. In this context, approaches using nano-based delivery systems are gaining attention and may significantly improve the clinical translation of polyphenol-based therapies.
Overall, while considerable progress has been made in the last decade in elucidating the potential of polyphenols as inhibitors across various targets, there exists a pressing need for further investigation, particularly in cellular and in vivo models, to comprehensively understand their mechanisms of action, bioavailability, safety and therapeutic efficacy.
This journal is © The Royal Society of Chemistry 2025 |