Techniques for the determination of dipeptidyl peptidase IV and screening of its inhibitors

Abstract

Dipeptidyl peptidase IV (DPP-IV), an important member of the serine hydrolase family, plays a core regulatory role in various physiological and pathological processes because of its unique broad-spectrum substrate specificity. DPP-IV-mediated proteolytic processing of endogenous peptide substrates serves as a fundamental regulatory mechanism governing physiological processes. An in-depth exploration of the enzymatic characteristics and mechanisms regulating DPP-IV activity will not only provide a theoretical basis for revealing the molecular pathogenesis of related diseases but also contribute to the construction of a complete disease regulatory network map. Furthermore, the development of specific DPP-IV inhibitors constitutes a key strategy for innovative therapy, holding significant promise for clinical translation. This review provides a comprehensive comparison of the applications, merits, and limitations of DPP-IV detection at the messenger RNA (mRNA), protein, and functional levels, along with an in-depth analysis of how DPP-IV enzymatic activity assessment profoundly impacts disease pathogenesis, progression, and therapeutic interventions. Particular emphasis is placed on the superior performance of probes based on drugs and optical substrates (e.g., for fluorescence and bioluminescence) in the functional determination of DPP-IV. Through metabolic phenotyping, we further summarize drug screening systems that target DPP-IV across multiple biological levels, including recombinant proteins (e.g., proteins derived from humans, animals, and gut bacteria), tissues, cells, human organ-on-a-chip models, and whole-animal studies. Furthermore, this review comprehensively evaluates both synthetic and natural DPP-IV inhibitors. Additionally, systematic investigations of their structure-activity relationships (SARs) provide a rational foundation for developing novel inhibitors with enhanced efficacy and safety.

Graphical abstract

Highlights

• •l Systematic comparison of DPP-IV detection methods (mRNA/protein/functional levels); drug/optical probes outperform in functional assays.
• •l Multidimensional analysis of DPP-IV inhibitors: structure, mechanism, source, and target specificity.
• •Novel gut microbiome-based DPP-IV inhibitor screening strategy: microbial DPP-IV affects host metabolism; high-throughput yields selective natural alkali inhibitors.
• •Comprehensive SAR analysis of synthetic/natural DPP-IV inhibitors: active compounds from traditional medicine guide novel inhibitor development.

1. Introduction

Dipeptidyl peptidase-IV (DPP-IV, also called CD26), a key serine hydrolase, plays a crucial role in the metabolism of numerous endogenous substrates as well as exogenous peptides and pharmaceutical compounds [1]. By metabolizing endogenous hormones, chemokines, and growth factors, DPP-IV influences the inflammatory response, oxidative stress, vascular function, and insulin resistance, thereby affecting the onset and progression of various diseases [2]. DPP-IV serves as a key negative regulator of blood glucose homeostasis by rapidly degrading glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP). The inhibition of DPP-IV activity can enhance the physiological effect of incretin, making it an important molecular target for the treatment of type 2 diabetes mellitus (T2DM) [3]. Furthermore, studies have shown that DPP-IV participates in regulating the occurrence and development of cardiovascular diseases, such as atherosclerosis and myocardial infarction, through changes in its enzymatic activity [4]. In cancer biology, DPP-IV exhibits a marked "double-edged sword" effect, and its functional outcomes are highly context dependent and influenced by the tumor type, microenvironment, and molecular isoform (i.e., soluble vs. membrane-bound) [5]. Within the immune system, DPP-IV contributes to immune and inflammatory response regulation by modulating T-cell proliferation, differentiation, and cytokine production [6]. Growing evidence indicates that abnormalities in the plasma level of DPP-IV may serve as potential biomarkers for the early diagnosis and prognostic evaluation of many diseases, including cholestasis, nonalcoholic fatty liver disease and lung diseases [7]. Consequently, the quantification of DPP-IV activity not only will provide critical biomarkers for disease diagnosis, therapeutic monitoring, and prognostic evaluation but also holds dual scientific significance for elucidating pathogenic mechanisms and developing novel targeted therapeutic strategies.

Since the initial Food and Drug Administration approval of sitagliptin, the first DPP-IV inhibitor, in 2006, multiple DPP-IV-targeting agents have been successfully marketed for the clinical management of T2DM. Emerging research has redefined the pivotal role of microbiota-derived DPP-IV within the therapeutic target network for T2DM [8].

Given the pleiotropic biological functions of DPP-IV, its inhibitors have demonstrated significant clinical benefits in the treatment of various systemic disorders. Nevertheless, the use of DPP-IV inhibitors raises unresolved safety and efficacy concerns in both diabetes management and off-label applications. Although currently available synthetic DPP-IV inhibitors demonstrate significant hypoglycemic effects, they still present nonnegligible clinical safety concerns, including potential induction of acute pancreatitis and specific hepatocyte toxicity. Moreover, DPP-IV inhibitors exhibit limited clinical responsiveness, with real-world data indicating that sitagliptin fails to achieve glycemic control in approximately 40% of diabetic patients. Consequently, there is an urgent demand for next-generation DPP-IV inhibitors with enhanced safety profiles and therapeutic efficacy. Based on the aforementioned research background, this article systematically reviews the methodologies for evaluating DPP-IV enzyme activity, critically analyzes the strengths and limitations of current drug screening systems from multiple perspectives, and further proposes optimization strategies for next-generation DPP-IV inhibitor screening platforms. Furthermore, this review systematically compiles the structure-activity relationships (SARs) of existing synthetic and naturally derived DPP-IV inhibitors, providing a theoretical framework for the rational design of novel DPP-IV-targeted therapeutics with optimal efficacy and safety profiles.

1.1. Biological function, catalytic mechanism, and endogenous and exogenous substrates of DPP-IV

DPP-IV has been a focus of interest in the pharmaceutical field because it regulates the levels of endogenous GLP-1, a gastrointestinal hormone that is highly elevated in plasma after food ingestion. Circulating GLP-1 exists in two distinct bioactive isoforms within the circulatory system, GLP-1 (7–37) and GLP-1 (7–36), and is synthesized in enteroendocrine L cells of the terminal ileum and colon. GLP-1 acts on the GLP-1 receptor (GLP-1R) to promote insulin secretion, insulin sensitivity and β-cell proliferation, but its duration of action is short because of its rapid degradation and inactivation in a DPP-IV-mediated process. Plasma from DPP-IV^(−/−) mice showed a significant decrease in hydrolytic activity compared with the synthetic substrates Gly-Pro-p-nitroaniline (GP-PNA) and GLP-1 and a consequent 2-fold increase in plasma insulin and GLP-1 levels, and these biochemical changes led to increased oral glucose tolerance [9]. The DPP-IV gene is a strong determinant of post-oral glucose tolerance test (post-OGTT) levels via glucose-sensing mechanisms that are abrogated in individuals with prediabetes. DPP-IV may be part of the molecular mechanism that controls the impairment of the insulin response after an OGTT as an early metabolic disorder associated with T2DM [10]. Clinically, DPP-IV inhibitors, including the reversible, fast-binding sitagliptin and the cyano-based, covalent, reversible inhibitors vildagliptin and saxagliptin, exhibit high selectivity for DPP-IV over related peptidases. When administered orally, these agents increase circulating GLP-1 levels, reduce fasting glucose, and enhance β-cell function in patients [11]. Since the first DPP-IV inhibitor was approved in 2006, this target has become one of the predominant strategies for developing drugs to treat T2DM. However, with the advent of more potent GLP-1 analogs such as semaglutide, the prominence of DPP-IV inhibitors has somewhat declined [12]. This shift may reflect limitations in their therapeutic profile, including potential off-target effects beyond GLP-1 hydrolysis inhibition that could influence broader physiological processes.

In addition to its role in GLP-1 metabolism, DPP-IV significantly influences the physiological functions of other endogenous substrates. Like GLP-1, GIP is involved mainly in the regulation of postprandial blood glucose levels. DPP-IV can also reduce the hydrolytic activity of GIP (Fig. 1). As established DPP-IV substrates, the neuropeptide peptide YY (PYY) and neuropeptide Y (NPY) play crucial roles in appetite regulation through hypothalamic satiety centers, with their biological activities being modulated by DPP-IV-mediated cleavage [13]. Stromal-derived factor-1 alpha (SDF-1α) is involved in the regulation of cell migration, the homing of stem cells to injury sites, the mobilization and recruitment of lymphocytes to inflammatory sites, and the activation of endothelial cells to initiate cell attachment and infiltration [2,14]. DPP-IV is one of the key metabolic enzymes of SDF-1α, and its specific site of hydrolysis highly coincides with the bioactive region of chemokines. In addition to the aforementioned endogenous substances, the enzymatic hydrolysis of various hormones and peptide molecules plays an important role in maintaining physiological homeostasis and pathological repair in the body. For example, after brain natriuretic peptide is metabolized by DPP-IV, its biological activity can be significantly increased, thereby affecting its ability to repair myocardial function and regulate diuretic effects. Therefore, the physiological regulatory function of DPP-IV and the mechanisms of its influence on multiple systems in the body still need to be further clarified [15]. At present, research has focused mainly on exploring SAR changes in endogenous peptides after specific enzymatic digestion modifications, among which metabolism-induced functional inactivation is the most common regulatory mode. However, NPY activity is significantly greater after metabolism than before metabolism. The partial or complete loss of the biological activity of the substrate caused by DPP-IV-mediated proteolysis remains the mainstream view widely accepted in the current research field. However, an increasing number of studies have shown that the metabolites of DPP-IV still possess selective biological activities, receptor antagonistic effects, and even novel signal transduction characteristics [16]. Therefore, before the physiological regulatory mechanisms of DPP-IV and its endogenous metabolites can be systematically clarified, it is of great scientific necessity to conduct analytical research on the SARs of DPP-IV.

Fig. 1.

Elevated dipeptidyl peptidase-IV (DPP-IV) activity disrupts systemic glucose homeostasis. Elevated DPP-IV activity catalyzes the proteolytic cleavage of the incretin hormones glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP), resulting in inactivation. This consequently impairs glucose-dependent insulin secretion from pancreatic β-cells while diminishing the suppressive effect on glucagon release from α-cells, ultimately leading to elevated blood glucose levels.

1.2. Significance of detecting DPP-IV activity for the study of its biological mechanism

Just as the function and role of endogenous peptide substrates in vivo have been studied above, the metabolic function of DPP-IV plays a very important role. Studies have shown that the activity of DPP-IV is correlated with the function of endogenous substrates [17]. Therefore, detecting DPP-IV activity can predict in advance the function and duration of action of these substrates in the body. In many cases, DPP-IV-truncated peptides exhibit complex new functions that are not fully understood. Understanding this unexpected and mysterious biological activity is an important new area of research because it is underappreciated and represents a major gap in the literature. This ligand activity has important clinical implications and requires new interpretations of some previous findings. Thus, the understanding of enzyme function or the universality of enzyme activity detection still needs more support [15], and the mechanisms and functions of enzyme metabolism and interactions between large and small molecules need to be further clarified.

1.3. Application and clinical limitations of DPP-IV inhibitors in the treatment of multisystem diseases

Although many studies have shown that inhibitors of DPP-IV have no effect on diseases other than diabetes, the function and role of endogenous DPP-IV substances in the body are not known [18]. DPP-IV remains a potential therapeutic target for many diseases. Many studies have shown that DPP-IV inhibitors may be potential therapeutic agents for diseases such as kidney disease, cancer, and myocardial ischemia [19]. The most unclear effect of DPP-IV inhibitors is that it is not possible to monitor endogenous substrates and products in real time at this stage. The main problem is that there are few tools for distinguishing between “intact” ligands and those cleaved by DPP-IV. Most ligand-targeting antibodies also detect intact and N-terminal proteolytic cleavage subtypes very well [20]. Second, the half-life of most DPP-IV ligand substrates is only a few minutes, and it is difficult to simulate the corresponding parameters of metabolic dynamics during the metabolism of small amounts of endogenous substrates, which is often caused by the saturated state of enzymes. Even in vivo and in vitro samples are often mixed with other metabolic enzymes, which makes it impossible to identify major functional proteins that degrade these endogenous peptides [16,21]. Even further processing leads to increased variability in sample ligand measurements, as well as in the identification and quantification of different bioactive lytic ligand subtypes [22]. Therefore, the development of a probe that can determine the rate of endogenous substance metabolism is particularly important, and such a probe could replace the detection of endogenous substances. Unfortunately, none of the assays developed thus far are able to evaluate the metabolism of endogenous substrates. Herein, systems for evaluating the enzymatic activity of DPP-IV, along with their advantages and disadvantages, are systematically summarized and compared, as well as the role of DPP-IV enzyme activity regulation in the occurrence, development and treatment of diseases. DPP-IV produced by gut bacteria has been found to hydrolyze endogenous products. Intestinal inhibition of DPP-IV can also play a role in increasing the levels of endogenous active substrates. However, commercially available DPP-IV inhibitors cannot inhibit DPP-IV produced by gut bacteria [8]. This further increases the scope and need for DPP-IV activity detection. Thus, research on DPP-IV enzyme activity detection systems and the development of new drugs are also summarized. Overall, the findings further emphasize the need to study the functional role of DPP-IV enzymes in the complex metabolic networks of complex biological systems.

2. Methods for DPP-IV detection

Traditional methods to detect DPP-IV, whether at the gene or messenger RNA (mRNA) level or by enzyme-linked immunosorbent assay (ELISA) or mass spectrometry (MS), are common and commercially available. In the past decade, with the discovery that DPP-IV can be used as a diagnostic marker and therapeutic target for diseases, an increasing number of probes have been designed for the detection of DPP-IV [23]. Among them, fluorescent probes, which are fast, convenient, sensitive, and specific, are the most common. Below, we describe assays for detecting function from the gene to protein level, with a focus on active probe assays for DPP-IV.

2.1. DPP-IV gene and protein detection methods

DPP-IV is a type II transmembrane protein that forms a homodimer in most cells and tissues. Transmembrane DPP-IV activity is regulated at multiple levels, including by the control of gene and protein expression, regulation of DPP-IV transfer to the cell surface, interaction with binding partners, and regulation of enzymatic activity [24].

2.1.1. Gene detection methods

The DPP-IV gene phenotype has been widely used in the prediction and treatment of a variety of diseases, including cancer, diabetes, cardiovascular disease, immune system disorders, and coronavirus, among others [25]. The rs3788979 DPP-IV polymorphism may increase the risk of myocardial infarction in patients with coronary artery disease and is associated with reduced plasma DPP-IV levels in patients with myocardial infarction [26]. Single nucleotide polymorphisms (SNPs) of DPP-IV may be correlated with the development of oral cancer in those who smoke cigarettes and consume alcohol. In addition, the DPP-IV SNP rs2268889 could be related to a worse clinical course of oral cancer in nonsmokers [27]. On the one hand, gene mutations may lead to changes in protein structure and its ability to bind with other proteins. On the other hand, structural changes may lead to changes in catalytic function, which would affect not only the metabolism of endogenous substances but also the efficacy of inhibitor treatment [28].

2.1.2. Protein detection methods

The transition of DPP-IV to the cell surface is highly regulated and can be initiated rapidly, such as by interleukin, hypoxia-inducing factor 1, interferon, retinoic acid, hepatic nuclear factor, granulocyte colony-stimulating factor, angiotensinogen II, and classical activation by leukocytes (e.g., T cells and B cells). Additionally, the activity of DPP-IV can be modified through posttranslational glycosylation and subunit dimerization [16,24]. DPP-IV is a relatively large protein composed of 766 amino acids, most of which are located outside the cell, with a hydrophobic transmembrane anchor of 7–28 amino acids and a very short intracellular sequence composed of 6 amino acids [17]. Soluble peptides (CD26/s-DPP-IV) composed of 39–766 extracellular amino acids are released into the extracellular environment by many matrix metalloproteinases [29]. Unlike the highly regulated expression/activity of the parent transmembrane form within cells and tissues, soluble DPP-IV (s-DPP-IV) is widely present in plasma and other tissue compartments and has complete enzymatic activity [30].

In general, DPP-IV antibody can be used in western blotting, immunohistochemistry or other coimmunoprecipitation methods to detect the DPP-IV protein concentration [31,32]. In addition, MS detection is a good method. In human immunodeficiency virus/hepatitis B virus patients with hepatocellular carcinoma (HCC), several differential proteins, including DPP-IV, can be detected in plasma by MS and can be used as screening markers for the early diagnosis of HCC and risk prediction. Monitoring protease expression differences can help in the diagnosis of HCC and determining the prognosis of patients with HCC [33]. Serum DPP-IV can be detected by liquid chromatography and tandem MS and may be a potential biomarker for evaluating the activity and response to treatment of rheumatoid arthritis [7]. The MS detection method can be used for the detection of DPP-IV and has been broadly applied. However, MS is expensive and does not measure a single protein, often resulting in large datasets. When we detected DPP-IV in different individuals, we found that the kinetic parameters of the enzyme varied among different individuals. Therefore, the efficient and rapid characterization of DPP-IV activity may still require substrate simulation.

According to the literature, apart from antibodies, there is currently only one method for the detection of substrates bound to DPP-IV, as shown in Table 1 [[34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55]]. Xing et al. [56] developed a new fluorescent probe by connecting fluorophores to the two inhibitors alogliptin and linagliptin, which bind closely to DPP-IV. This new probe can be used for specific DPP-IV recognition and imaging in living cells (Fig. 2). Importantly, this study provides a general strategy for the construction of new nonenzymatic fluorescent probes for many proteases.

Table 1.

Optical substrates detected based on enzyme functional activity.

No.	Optical substrate	λex/λem (nm)	Catalytic kinetics (Km, μM)	LOD (U/L)	Linearity (U/L)	Refs.
OS-1		405 (LC-UV)	101.2	1.56	1–90	–
OS-2		380/460	34–237.8 (vmax = 0.0052–5.4 μM/min or 9.01 μmol/min/mg)	0.18	1–90	–
OS-3		340/430	60.1	0.10	1–90	–
OS-4		485/535	15.8(vmax = 0.00044 μM/min)	–	–	–
OS-5		630/670	56 (vmax = 0.63 μM/min)	–	–	–
OS-6		330/546	33.6 (vmax = 5.9 μM/min)	–	–	–
OS-7		585/625 (Φ = 0.51)	112 (vmax = 7.8 μM/min)	0.35 ng/mL	2–60 ng/mL	[34]
OS-8		488/500–580	29.6	–	–	[35]
OS-9		320/450	–	–	0.1–0.5 mU/mL (r2 = 0.9608)	[36]
OS-10		300/460	–	0.07 μg/mL	0.07–7.00 μg/mL	[37]
OS-11		430/535 810/535	42.93 (vmax = 15.49 μmol/min/mg)	0.78 ng/mL	0.78–80 ng/mL	[38]
OS-12		458/658	150.1 (vmax = 15.03 μmol/min/mg)	–	0–1.5 μg/mL	[39]
OS-13		595/612	9.2	–	–	[40]
OS-14		640/690	8.76 (70.88 μM/min)	0.19 ng/mL	0.62–10 ng/mL	[41]
OS-15		luciferase	–	0.78 mU/mL (1.15 ng/mL)	2.0–40.0 mU/mL	[42,43]
OS-16		480/666	87.80 (6.90 μmol/min/μg)	5.4 ng/mL	5.4–2000 ng/mL	[[44], [45], [46]]
OS-17		565/618	–	3 μg/mL	–	[47]
OS-18		637/660	29.1 (7.1 × 10−9 M/s)	–	–	[48]
OS-19		594/600–650	24.6	–	–	[49]
OS-20		600/660	–	–	–	[50]
OS-21		665/715	13.38 (0.309 μM/min)	0.29 ng/mL	0.005−0.10 μg/mL	[51]
OS-22		520/673	–	0.84 ng/mL	–	[52]
OS-23		580/775	12.37	2.5 ng/mL	0–2.0 μg/mL	[53]
OS-24		485/535	–	–	–	[54]
OS-25		485/520	–	0.18 mU/mL	1–5 mU/mL	[55]

λ_ex/λ_em: excitation wavelength/emission wavelength; K_m: Michaelis constant; LOD: limit of detection; v_max: maximum reaction velocity; LC-UV: liquid chromatography-ultraviolet detection; −: no data.

Fig. 2.

Strategy for substrate probe design based on dipeptidyl peptidase-IV (DPP-IV) inhibitors. A new fluorescent probe was obtained by connecting fluorescent groups to two inhibitors, alogliptin and linagliptin, which tightly bind to DPP-IV.

2.2. Functional level systems for detecting DPP-IV

2.2.1. Endogenous substrate detection

The detection of endogenous substrates by MS can avoid the problem caused the by inability of antibodies to discriminate. MS characterizes and quantifies different peptides according to their mass, making diamino acid differences easier to detect. However, this approach faces two major challenges: (1) the inherent complexity of biological samples containing numerous peptides with similar molecular weights and (2) the physiological concentrations of many peptides falling below MS detection limits. This often necessitates the addition of supraphysiological ligand concentrations to generate sufficient quantities of both intact and cleaved variants for standard MS analysis [57]. These limitations have historically restricted MS methods to qualitative analysis, primarily for identifying and characterizing specific peptide cleavage variants under constrained experimental conditions [21]. To improve selectivity and achieve lower detection physiological limits, Chappell et al. [58] used immunoaffinity enrichment MS combined with liquid chromatography sample purification to quantify GLP-1(7–36) and GLP-1(9–36) in human plasma. Villaño et al. [59] quantitatively measured GLP-1(7–36), GLP-1(9–36), GIP(1–32) and GIP(3–32) directly in plasma using MS and confirmed that GLP-1(7–36) is indeed elevated after meals but that its metabolic rate is also very fast. In another seminal paper by Wang et al. [8], which followed the same general protocol, circulating intact and DPP-IV-cleaved SDF-1α levels in mice and rhesus monkeys were accurately quantified. Wang et al. [8] found that when measured by other commercially available quantitative methods, while DPP-IV inhibitors reduced the total level of SDF-1α, they significantly increased the level of intact SDF-1α. This was a major surprise and demonstrated the importance of measuring intact and lysed peptides when interpreting experimental data. The authors speculated that this effect represents an overall decrease in the amount of proteolytic isomers while the amount of intact SDF-1α is retained or slightly increased. The inhibition of SDF-1α conversion leads to negative feedback regulation of SDF-1α synthesis and release by C-X-C motif chemokine receptor 4 (CXCR4)-positive cells, which normally increases SDF-1α synthesis when intact SDF-1α/CXCR4 is reduced [20].

2.2.2. Drug substrate detection

DPP-IV is a nontraditional drug-metabolizing enzyme that was initially thought to metabolize only the amide bonds of endogenous peptides and not to participate in drug metabolism. However, recent studies have shown that DPP-IV metabolizes cyanide (CN), and the conversion of CN to carboxyl groups (-COOH) is the most important metabolic pathway of cyanopyrrolidine-based drugs; thus, this pathway has attracted much attention from pharmacological and clinical pharmacy researchers. The products of cyanopyrrolidine-based drugs exhibit increased water solubility after enzymatic metabolism by DPP-IV or its family members, which enables them to be excreted faster, but this also leads to metabolic inactivation of drugs, thus weakening the duration of action and efficacy of drugs. The formation of amide metabolites is not only the initial step in hydrolytic metabolism, but also the rate-limiting step in the formation of carboxylated end products. It has been reported that the nitrile group of saxagliptin cannot be hydrolyzed. The inability of saxagliptin to be hydrolyzed to carboxylic acid may be due to the inability of its amide to form an amide metabolite in the first step; however, further verification is needed [60], and the mechanism of cyanogen metabolism by DPP-IV remains to be further elucidated.

2.2.3. Optical substrate detection

Exogenous visual substrates can be used to detect the activity of a target effectively and can solve the problems of endogenous substrates being difficult to detect, with poorly characterized functional activity. From ultraviolet-sensitive to fluorescent and luminescent substrates, many probes have been designed and developed, as shown in Table 1. In general, these probes do not have optical properties after the binding of dipeptides and produce metabolites with ultraviolet absorption or fluorescence after hydrolysis by DPP-IV. Among them, fluorescent probes for the detection of DPP-IV have been widely used from the molecular level to the whole-organism level.

The main feature of these fluorescent substrates is the selection of easily detectable nuclei with amino acids that can be specifically recognized by DPP-IV. DPP-IV cleaves a dipeptide containing proline, alanine or one of several other amino acids from the second-to-last position of the amino terminus of the substrate. As shown in Table 1, the first exogenous compound used to detect DPP-IV was GP-PNA (OS-1), whose core has a strong absorption capacity. In subsequent decades, coumarin compounds were found to have the characteristics of a fluorescent core that could be used as a probe for DPP-IV, and a series of modified coumarin compounds were synthesized on this basis. These probe substrates (OS-2 and OS-3) are particularly suitable for analyzing relatively simple biological matrices, such as plasma or urine, which contain fewer interfering substances. With the emergence of the importance of DPP-IV in physiology and pathology, researchers are increasingly demanding probe molecules with improved performance. In recent years, improvements have been made in the range of nuclei used to measure the activity of DPP-IV, including cresol violet, rhodamine, and others with high fluorescence quantum yields. These newly designed fluorescent probes (OS-7, OS-8, OS-10, OS-13, OS-17, OS-18, OS-19, OS-20, OS-21, and OS-22) with modified core structures not only enhance DPP-IV detection in simple biological samples but also enable analysis in complex tissue systems. They have been successfully implemented as detection methods for tumor tissues and other pathological specimens. Moreover, several of these probes (OS-4, OS-5, OS-6, and OS-9) have been specifically developed for high-throughput screening (HTS) and the discovery of novel DPP-IV inhibitors. For the first time, our research group used a new fluorophore group, naphthalenediamide, in the design of a DPP-IV probe (OS-11), which has the characteristics of high sensitivity, high fluorescence quantum yield and good selectivity [61,62]. We subsequently developed a seminitrile probe with near-infrared (NIR) wavelengths for deep tissue imaging (OS-16). Consequently, numerous probes (OS-12, OS-14, OS-16, OS-23, and OS-24) have been developed for the real-time visualization of DPP-IV activity in tissue imaging applications, enabling dynamic monitoring of enzymatic activity in situ. Chen et al. [55] developed a novel, simple, isothermal, nonenzymatic DPP-IV detection method based on DNA‒peptide conjugates and dual signal amplification (OS-25). The development of fluorescent probes remains an active research area, and with the clarification of the role of DPP-IV in various diseases, the application of probes is gradually increasing.

Bioluminescence imaging is a widely used optical imaging technique based on the luciferase-catalyzed oxidation of a small molecule called D-luciferin, which results in the production of visible light. D-cysteine, a component of the “split luciferin” system, is “caged” with a DPP-IV-specific amino acid sequence (Gly-Pro), resulting in the tripeptide Gly-Pro-D-Cys (GPc), i.e., OS-15. The proteolytic activity of DPP-IV on this peptide enables the cleavage of D-Cys and the subsequent in situ formation of D-luciferin via reaction with 2-cyano-benzothiazole (CBT), another component of the “split luciferin” system; this process is referred to as click chemistry. The resulting D-luciferin molecule is then oxidized by firefly luciferase, resulting in the production of one photon of light, which can be quantified by a sensitive charge-coupled device (CCD) camera or plate reader. Because free D-cysteine is required for the reaction to proceed, no light is observed in the absence of DPP-IV. Therefore, the probe (OS-15) named DPP-IV-activated luciferin (DAL, OS-X) enables selective real-time imaging and quantification of DPP-IV activity by measuring the photon flux from the luciferase-catalyzed oxidation of D-luciferin [42,43]. This method has already been successfully applied in canine and human cadavers, paving the way for translating functional bioluminescence imaging into noninvasive quantification of biological processes in large animals [63].

3. Classification of DPP-IV inhibitors

The multidimensional classification of disease target inhibitors is of great significance for both basic research and clinical applications. The scope of clinical applications of DPP-IV inhibitors has expanded from the treatment of diabetes to include multiple other diseases, such as cancer and cardiovascular diseases [64]. Inhibitors of different structural types may have different therapeutic effects on specific disease subtypes or patient populations. A scientific classification system can provide a theoretical basis for precision medicine. Furthermore, there are significant differences in the safety profiles of various inhibitors. Systematic classification is helpful for identifying potential adverse reactions, thereby optimizing clinical medication regimens, guiding regulatory decisions, and promoting individualized treatment. At the level of drug research and development, classification based on pharmacological parameters such as affinity, selectivity, pharmacokinetic characteristics and SAR of DPP-IV inhibitors can significantly improve the efficiency of drug design and reduce the costs of trial and error in the research and development process.

DPP-IV inhibitors can be classified according to different criteria, such as peptide inhibitors and nonpeptide inhibitors, according to their structure. Peptide inhibitors are usually based on natural or synthetic linear or cyclic structures composed of amino acid residues that mimic the natural substrate or enzyme inhibition sites of DPP-IV. For example, proteins in food can produce a variety of bioactive peptides after enzymatic hydrolysis, which can directly bind to the active site of DPP-IV, thereby inhibiting the activity of DPP-IV and reducing blood glucose levels [65]. In recent years, a variety of DPP-IV inhibitory peptides have been found in the hydrolysates of marine organisms. For example, peptide extracts from Musculus senhousei contain many fragments with significant DPP-IV inhibitory potential [66]. Nonpeptide DPP-IV inhibitors are structurally independent of amino acid residues and usually bind to the enzyme active site through electron configurations that mimic peptide bonds (e.g., pyrimidine diketone and fluorophenyl). Compared with peptide inhibitors, this type of inhibitor has significant pharmaceutical advantages: its chemical structure is not easily hydrolyzed by enzymes and is more stable. Moreover, it has better bioavailability and pharmacokinetic characteristics. In addition, it carries a much lower risk of immunogenicity risk [67]. Consequently, the structural superiority of nonpeptide inhibitors renders them more suitable as lead compounds for drug development.

Based on their mechanism of action, DPP-IV inhibitors can be classified as competitive inhibitors, noncompetitive inhibitors and irreversible inhibitors. Competitive inhibitors directly compete with the active site of DPP-IV and prevent the substrate from binding to enzymes; for example, vildagliptin and sitagliptin inhibit recombinant human DPP-IV (rhDPP-IV) [68]. Noncompetitive inhibitors act on the inactive sites of DPP-IV, and their effects are independent of the presence of a substrate; such inhibitors include aloe emodin and some compounds containing a phenylethyl phenyl phthalimide skeleton [69]. Irreversible inhibitors form covalent bonds with DPP-IV, resulting in permanent inactivation of the enzyme; for example, X-ray crystallography revealed that saxagliptin covalently binds to the active site of DPP-IV [70]. The processes of covalent bond formation and dissociation of this type of drug are relatively slow, resulting in the duration of the hypoglycemic effect exceeding the half-life of the drug. Therefore, frequent administration is not necessary. In addition, DPP-IV inhibitors can be classified by source, including natural and synthetic sources; examples include aloe vera emodin and sitagliptin, respectively.

In addition to their specific action targets, DPP-IV inhibitors can be further classified as selective inhibitors and nonselective inhibitors. DPP-IV and fibroblast activation protein (FAP) belong to the serine protease family and share a similar catalytic domain. Additionally, these two enzymes are highly homologous prolyl-cleaving dipeptidases, and the substrate spectra of FAP and DPP, including DPP-IV, dipeptidyl peptidase VIII (DPP-VIII), and dipeptidyl peptidase IX (DPP-IX), overlap greatly [17]. Therefore, the development of highly specific substrates is crucial for screening selective inhibitors of DPP-IV and studying their physiological functions [38]. Prusoretine (DBPR108) is a novel highly selective DPP-IV inhibitor developed by CSPC Zhongqi Pharmaceutical Technology (Shijiazhuang) Co., Ltd., with a half maximal inhibitory concentration (IC₅₀) of only 15 nM, and it shows excellent selectivity for DPP-II, DPP-VIII, DPP-IX, and FAP (IC₅₀ > 50 μM) [71]. Both phase II and phase III studies have shown that prusoretine has reliable hypoglycemic efficacy and good safety characteristics [71,72]. DSP-7238, a noncyanide pyrrolidine compound, is a highly selective competitive DPP-IV inhibitor. Long-term administration can improve glycemic control in mouse models with impaired insulin sensitivity and secretion and ameliorate β-cell damage [68]. The low selectivity of drugs for specific enzymes or receptors is a primary cause of adverse effects. For example, vildagliptin, a nonselective DPP-IV inhibitor, concurrently inhibits DPP-VIII and DPP-IX, which are involved in energy metabolism and immune regulation, thereby exerting multiple pharmacological actions beyond their hypoglycemic effects.

4. Screening of DPP-IV-targeting drugs with different levels of action

Screening drugs in different systems is a key step in the drug discovery and development process that helps researchers evaluate the activity, toxicity, selectivity, bioavailability and other properties of drug candidates. As illustrated in Fig. 3, we systematically characterized multiple dimensions of drug screening. Cell models can simulate the action of drugs in the body, providing preliminary pharmacodynamic and toxicity data that can help narrow the field of drug candidates. Animal models can more closely simulate the human physiological environment, and assessments of these models help predict the effects of drugs in humans and inform clinical trials. In vitro enzyme levels or receptor levels can be used to evaluate the inhibition or activation of a drug candidate on a specific enzyme or receptor. In vitro methods are simple, controllable, and extremely fast, thus allowing rapid screening of a large number of compounds and their affinity for a specific target and guiding drug design. In addition, drug screening via microbial models helps to understand the effects of drugs on the gut microecology, which is crucial for assessing the long-term safety and possible side effects of drugs. In recent years, tissue or organ chips have also been used to assess the effects of drug candidates on specific tissues or organs; of course, the ultimate goal of all drug screening is to identify compound suitable for preclinical and clinical trials and then perform final confirmation of their efficacy and safety in human volunteers or patients.

Fig. 3.

Multilevel and multidimensional screening model for dipeptidyl peptidase-IV (DPP-IV) inhibitors. rhDPP-IV: recombinant human DPP-IV.

In the past two decades, the research and development of DPP-IV have focused mainly on its hypoglycemic effects. Therefore, the discovery of the first DPP-IV inhibitor, sitagliptin, has also been regarded as an important milestone in the treatment of diabetes. Recently, the ultra-long-acting DPP-IV inhibitor cofrogliptin, which is taken orally once every two weeks, has become the preferred treatment option, with strong inhibition of the DPP-IV enzyme and a long duration; the inhibition rate of DPP-IV remains above 80% within 2 weeks after oral administration [73]. This drug has shown significant hypoglycemic efficacy in both monotherapy and combination therapy but also has excellent safety features. These features help improve the medication compliance of patients with diabetes and comprehensively enhance the management of diabetes. In addition to diabetes, DPP-IV can also play an important role in cancer, neurodegenerative diseases and cardiovascular diseases by regulating the activity of immune cells and affecting cell proliferation and apoptosis, neuronal function and metabolic processes in the heart [74,75]. However, in addition to a variety of adverse reactions [76], existing DPP-IV inhibitors still have limited efficacy. For example, the therapeutic efficacy of sitagliptin varies widely among individuals, and sitagliptin is ineffective in 40% of patients with diabetes [8]. Therefore, the pathological mechanism of DPP-IV in diabetes remains unclear, and the development of DPP-IV inhibitors with increased efficacy and safety is an important direction for research on hypoglycemic drugs.

4.1. Target-based drug screening with rhDPP-IV

The application of human host recombinant subtype enzymes is an irreplaceable tool in drug screening that takes advantage of the progress of modern molecular biology and bioengineering technology and provides a new perspective and method for drug discovery and development. By expressing the human host recombinant subtype enzyme, it is possible to verify its importance in drug action and thus determine whether it can be used as a target for drug design. In HTS, human recombinant subtype enzymes can be used to rapidly evaluate the selectivity of a large number of compounds for a specific enzyme. Moreover, using recombinant subtype enzymes combined with specific fluorescent probe molecules of drug metabolism enzymes and identification of the “reaction phenotype”, the interaction rules of drug components and metabolic enzymes and new metabolic pathways can be revealed, providing a scientific basis for the study of toxicology mediated by drug metabolism [77].

In the development of drugs that target DPP-IV, many studies have used rhDPP-IV as an enzyme source to achieve standardization and controllability. The evaluation of the inhibitory effects of the human recombinant DPP-IV subtype enzyme on four cannabinoids revealed that cannabidiol, cannabigerol, cannabinol and Δ9-tetrahydrocannabinol can inhibit the activity of DPP-IV or induce conformational changes by binding to active or allosteric sites [78]. Studies have confirmed that amino coumarin derivatives have significant inhibitory effects on human recombinant DPP-IV subtype enzymes. Given that DPP-IV can be involved in tumor processes by regulating the dynamic balance of chemokines and cytokines in the tumor microenvironment, researchers have further evaluated the antitumor potential of this type of compound. Among them, derivative 17I demonstrated outstanding antiproliferative activity in the A549 lung cancer cell line, with an IC₅₀ as low as 24 nM [79]. This type of dual inhibitor is essential for blood glucose management in patients with tumor-associated hyperglycemia. Both vildagliptin and sitagliptin competitively inhibit rhDPP-IV, with inhibition constants (K_i) as low as 0.60 and 2.1 nM, whereas saxagliptin does not competitively inhibit rhDPP-IV [68]. Screening systems using rhDPP-IV as an enzyme source can rapidly identify potential therapeutic DPP-IV inhibitors and provide drug candidates for the development of treatments for T2DM, neurodegenerative diseases, and other diseases associated with DPP-IV activity. However, the rhDPP-IV subtype enzyme has problems such as high cost and poor stability, and the activity of the recombinant subtype enzyme may differ from that of the natural enzyme, which may need to be optimized through protein engineering. Moreover, there may be technical challenges in scaling up production from the laboratory scale to the industrial scale, such as maintaining the stability and activity of the protein.

4.2. Target-based drug screening with intestinal bacteria-derived host DPP-IV

The gut microbiome is composed of trillions of microorganisms that coexist with hosts and participate in host energy metabolism, immune function, and lipid and glucose metabolism through multiple mechanisms, thereby affecting the health of the hosts; therefore, the gut microbiome is also considered an important “functional organ” [80]. In the past ten years, scientists have accumulated much convincing evidence concerning the relationship between the intestinal microbiome and T2DM. Changes in the types, functions and interactions of the intestinal microflora and genera can strongly affect the metabolic balance of the host, leading to insulin resistance and an imbalance of host inflammatory cytokines [81]. Therefore, targeted intervention strategies based on the gut microbiome have shown important prospects for application in the individualized prevention and treatment of T2DM and other metabolic diseases.

However, for DPP-IV inhibitors, there is also a problem of low clinical responsiveness. For example, linagliptin can reduce the mean glycosylated hemoglobin (HbA1c) level in adolescents with T2DM in the short term (<4 weeks), but the level gradually rebounds in 26–52 weeks, and the hypoglycemic effect is not significant (the mean 26-week HbA1c in the linagliptin group was 8.33% versus 8.77% in the placebo group) [82]. After oral sitagliptin treatment, the HbA1c level decreased from the mean baseline level of 8.3% to less than 7% (67 mmol/mol) in only 25% of the patients [83]. Sitagliptin was also less effective in controlling blood glucose and weight in patients with T2DM who did not respond well to metformin [84]. Therefore, improving the clinical efficacy of these drugs is extremely important.

Research has shown that the microbiota may contain microbial enzymes that function similar to human enzymes, such as Streptococcus mutans, which produces X-prolyl dipeptidyl peptidase (Sm-XPDAP), an enzyme analogous to DPP-IV [85]. Functional studies have shown that there are certain differences in substrate recognition mechanisms between microbial DPP-IV and homologous mammalian enzymes. Through examination of the crystal structure of microbial DPP-IV and its complexes with two dipeptides and nonpeptide inhibitors, it was found that microbial DPP-IV contains acylase intermediates in its dipeptide complex, whereas the oligopeptide complex of mammalian DPP-IV contains tetrahedral intermediates, indicating that the two enzymes involved in substrate carbonyl recognition have different structural environments for the active-site arginine (Arg) residue [86]. Recently, Wang et al. [8] proposed the concept of microbe-host isoenzymes (MHIs) in which over the course of long-term coevolution with the host, the intestinal microflora can produce a class of enzymes with the same catalytic function as host enzymes to regulate the physiological state of the host. Further research revealed that microbial DPP-IV can enter the extraluminal intestinal tissue and plasma of the host, degrade GLP-1, and thereby reduce the plasma insulin concentration, leading to impaired glucose tolerance. Subsequently, targeted inhibitors for humans were used to test sensitivity, and it was found that all clinical inhibitors had relatively low inhibitory effects on microbial DPP-IV. Among them, the inhibitory effect of sitagliptin was reduced by approximately 4-fold. Further experiments revealed that patients with microbial DPP-IV enrichment had low responsiveness to sitagliptin, and the weak inhibitory effect of sitagliptin on DPP-IV secreted by Bacteroides thetaiotaomicron (btDPP-IV) was the main reason for the low clinical response (Fig. 4). This study provides new insights into the molecular mechanism by which microbial proteins regulate host physiology and provides a deeper understanding of gut microbiota-host crosstalk.

Fig. 4.

The mechanism of low clinical responsiveness to sitagliptin. In the context of type 2 diabetes mellitus (T2DM), intestinal barrier dysfunction can lead to the translocation of microbiota-derived dipeptidyl peptidase-IV (DPP-IV) to the intestinal mucosal tissues and circulatory system of the host, thereby accelerating the degradation of glucagon-like peptide-1 (GLP-1), reducing plasma insulin levels, and exacerbating impaired glucose tolerance. The inhibitory effects of human targeted DPP-IV inhibitors (such as sitagliptin) on microbial DPP-IV are limited, making it difficult to improve such glucose metabolism disorders effectively. The novel selective inhibitor Dau-d4 specifically targets microbial DPP-IV, significantly inhibiting GLP-1 hydrolysis, maintaining insulin secretion homeostasis, and ultimately improving impaired glucose tolerance.

The validation of microbial DPP-IV as a novel therapeutic target for diabetes not only expands our understanding of microbiome-host interactions but also pioneers new avenues for structure-based precision drug design [8]. The DPP-IV-specific inhibitor diprotein A has inhibitory activity against recombinant Sm-XPDAP, and saxagliptin is also effective to some extent (K_i = 129 ± 16 μM) [87]. Microbial DPP-IV is an important factor affecting host glucose metabolism homeostasis [8]. Therefore, it is necessary to simultaneously inhibit microbial DPP-IV to improve glucose homeostasis effectively. Based on the above findings, Jiang's research team successfully constructed an HTS platform for btDPP-IV. Through systematic screening of a library of 107,000 small-molecule compounds, multiple lead compounds with significant inhibitory activities were identified. The active compounds that showed an inhibition rate of more than 90% at the initial screening concentration (10 μM) included Dau, Jervine, ALK-IN-1, GSK-1070916, Cleviprex, adefovir dipivoxil, 54-J09, 114-F05, 114-L05, and 114-B05. Among them, Dau, a natural alkaloid isolated from medicinal plants, showed highly selective inhibitory activity against microbial DPP-IV (IC₅₀ = 0.37 ± 0.01 μM), whereas these compounds had almost no inhibitory effect on hDPP-IV (IC₅₀ > 100 μM). Structural modification of the lead compound Dau revealed that the derivative Dau-d4 is a specific inhibitor of microbial DPP-IV and can effectively maintain intestinal glucose metabolic homeostasis in mice without significantly affecting the structure and functional integrity of the gut microbiota. Notably, combination therapy with Dau-d4 and sitagliptin can significantly improve the clinical response [8]. The discovery of MHIs opens a new door for the precise regulation of the gut microbiota, and the development of more drugs targeting these isoenzymes is important for increasing the understanding of gut microbiome–host interactions and for improving health, particularly by managing blood sugar and preventing diabetes.

4.3. Target-based drug screening with DPP-IV inhibitors in complex biological systems

Complex biological systems involve multiple interacting molecular and cellular processes. By conducting drug screening in complex systems, researchers can identify potential drug molecules that affect these processes, deepening the understanding of how biological systems work. DPP-IV is a transmembrane glycoprotein with a molecular weight of 110 kDa that is expressed in a constitutive nondimeric form (220 kDa) in various cell types. In addition, DPP-IV exists in a soluble form in the circulatory system. Although sDPP-IV in plasma lacks the intracellular tail and transmembrane domains of proteins, it retains a large amount of enzymatic activity [88]. Therefore, accurate quantification of DPP-IV activity in human plasma is important for the diagnosis of T2DM [46]. The IC₅₀ of sitagliptin was determined using human plasma as the enzyme source and was found to be very close to the IC₅₀ of rhDPP-IV (36.22 nM and 39.18 nM, respectively). This suggests that the expensive rhDPP-IV enzyme can be replaced by human plasma when performing HTS [89]. However, clinical data show that sitagliptin treatment for 12 months reduces systemic DPP-IV activity but does not increase the level of sDPP-IV [90]. Therefore, determining whether sDPP-IV degrades GLP-1 is of practical importance for screening DPP-IV inhibitors using plasma as an enzyme source. Furthermore, several studies have quantitatively determined the activity of DPP-IV in patients with diabetes and in normal individuals using human serum or plasma (Figs. 5A, B and C), and the results revealed that the activity of DPP-IV in patients with diabetes and those with a family history of diabetes was significantly greater than that in normal individuals [41,61]. The inhibitory effects of sitagliptin and vildagliptin were detected using human serum and rhDPP-IV, and the inhibition rates in the two enzyme sources were very similar (Figs. 5D and E). After optimizing the screening method using plasma as the enzyme source for the HTS of DPP-IV inhibitors in 69 natural alkaloids, it was found that fangchinoline and sanguinarine have certain inhibitory effects on these compounds [89]. In addition to existing clinical drugs, anthraquinone compounds were screened using human plasma as an enzyme source, and it was found that alizarin (AQ7), aloe emodin (AQ11), and emodin (AQ13) had significant inhibitory effects on DPP-IV, with IC₅₀ values lower than 5 μM. Further fluorescence imaging at the cellular level also confirmed the screening results (Figs. 5F–H) [69]. However, whether plasma or serum DPP-IV can be used as a therapeutic target for T2DM or a screening target for antidiabetic drugs still requires verification through more experiments.

Fig. 5.

Biological applications of serum or plasma dipeptidyl peptidase-IV (DPP-IV). (A) Human venous blood-derived plasma DPP-IV and serum DPP-IV. (B, C) DPP-IV enzyme activity as a biomarker of diabetes: serum was used as an enzyme source to detect DPP-IV enzyme activity in patients with and without diabetes (B), and serum was used as an enzyme source to detect DPP-IV enzyme activity in patients with and without a family history of diabetes (C). (D, E) Comparison of the potential of saxagliptin and vildagliptin to inhibit DPP-IV in serum and recombinant human DPP-IV (rhDPP-IV): the inhibition of DPP-IV and human recombinant DPP-IV in serum by saxagliptin (D) and the inhibition of DPP-IV and human recombinant DPP-IV in serum by vildagliptin (E). (F) Screening anthraquinone DPP-IV inhibitors using human plasma as the enzyme source. (G) Structure-activity relationships (SARs) of anthraquinone compounds and human plasma DPP-IV. (H) 29 (sitagliptin), AQ13, and AQ7 inhibited DPP-IV activity in human renal proximal tubular epithelial cells (RTPECs). ^∗∗P < 0.01, ^∗∗∗∗P < 0.001. s-DPP-IV: soluble DPP-IV; m-DPP-IV: membrane-bound DPP-IV; BAN: butyl-4-amino-1,8-naphthalimide; GP-BAN: Gly-Pro-N-butyl-4-amino-1,8-naphthalimide.

4.4. Prospects for future DPP-IV-targeting drug screening

DPP-IV-targeting drugs have shown good potential in the treatment of T2DM and other fields. With the advancement of scientific research technology and a deeper understanding of the mechanism of action of DPP-IV inhibitors, there should be more focus in the screening of DPP-IV-targeting drugs in the future. The first step is to develop more selective DPP-IV inhibitors to reduce their impact on other enzymes and improve the safety and tolerability of drugs. With the analysis of the crystal structure of DPP-IV and its complexes with inhibitors, structure-based drug design methods will become more precise, resulting in the design of drug molecules with higher affinity and fewer side effects. By utilizing combinatorial chemistry techniques, DPP-IV inhibitors with potential activity can be quickly screened, thereby accelerating the process of discovering new drugs. In addition, methods based on bioinformatics and computational biology, such as molecular docking and pharmacophore models, can be used to predict and screen new DPP-IV inhibitors, which will greatly improve the efficiency of drug screening. In addition, considering the impact of the gut microbiota on drug efficacy, future research may focus on how to utilize or regulate the gut microbiota to increase the efficacy of DPP-IV inhibitors. As such, with a deeper understanding of the factors affecting the efficacy and safety of DPP-IV inhibitors, personalized treatment plans for specific populations should be developed in the future. Molecular hybridization is an emerging concept in medicinal chemistry, and in the future, multitarget drug design can be carried out to develop drugs that simultaneously target DPP-IV and other related targets to improve therapeutic efficacy. In addition, while new DPP-IV inhibitors could effectively lower blood sugar, more attention should be given to other protective effects, such as protective effects on pancreatic and cardiovascular function.

5. Marketed DPP-IV inhibitors

DPP-IV inhibitors stimulate insulin secretion in a glucose concentration-dependent manner, promoting secretion when hyperglycemia occurs but not when hypoglycemia occurs [91]. These inhibitors can regulate the function of β-cells at the same time, increase glucose to stimulate insulin secretion, and enhance the glucose-meditated inhibition of glucagon secretion. In the past two decades, the research and development of DPP-IV inhibitors have focused predominantly on optimizing three key pharmacological properties: potency, plasma half-life, and enzymatic selectivity. Recently, Li et al. [92] classified DPP-IV inhibitors into five main categories based on their evolution from initial lead compounds to clinical candidate compounds: (1) class I DPP-IV inhibitors are based on pyrrolidine structures; (2) class II DPP-IV inhibitors are based on xanthine, and their bioelectronic structure is excretory; (3) class III DPP-IV inhibitors are based on phenylethylamine; (4) class IV tricyclic DPP-IV inhibitors are fused with a tricyclic skeleton; and (5) class V DPP-IV inhibitors are based on ortho phenyl aryl groups. This is a more detailed classification that comprehensively summarizes the development of DPP-IV inhibitors over the past thirty years based on the structural changes in lead compounds evolving into long-acting DPP-IV inhibitors. In this review, we systematically elucidate the mechanisms of action and structural basis of currently marketed DPP-IV inhibitors, with a particular focus on the first three categories in the classification.

5.1. Marketed inhibitors

To date, the number of innovative DPP-IV inhibitors worldwide has increased to 14; among them, 10 were developed in China. These include sitagliptin (Merck), vildagliptin (Novartis), saxagliptin (AstraZeneca), alogliptin (Takeda), linagliptin (Eli Lilly), trelagliptin (Takeda), tiagliptin (Daiichi Sankyo/Mitsubishi Tanabe Pharma), cofrogliptin (Hesco) [93], retagliptin (Henri) [94], fotagliptin (Salubris) [95], prusogliptin (Stone Pharmaceutical Group) [71] and shengagliptin (Shengshi Taike) [96]. There are numerous differences in the structure, mechanism of action, pharmacokinetic characteristics, and frequency of use of currently available DPP-IV inhibitors. These differences determine their different characteristics and advantages in clinical applications. In practical use, it is necessary to comprehensively consider factors such as the patient's condition, medication compliance, and safety when selecting appropriate DPP-IV inhibitors. According to their structural characteristics, representative DPP-IV inhibitors that have been approved for marketing are listed in Fig. 6.

Fig. 6.

Representative marketed dipeptidyl peptidase-IV (DPP-IV) inhibitors. A comprehensive list of structurally diverse marketed DPP-IV inhibitors was compiled, including relevant pharmacological data such as developer information, IC₅₀ values, and dosing frequencies. IC₅₀: half maximal inhibitory concentration. SALUBRIS: Salubris' Registered Company in China; Takeda: Takeda Pharmaceutical Company Limited; Mitsubishi Tanabe: Mitsubishi Tanabe Pharma; CGene Tech: Shengshi Taikang Biopharmaceutical Technology (Suzhou) Co., Ltd.; Pfizer: Pfizer Inc.; Merck: Merck & Co., Inc.; Hengrui: Jiangsu Hengrui Pharmaceuticals Co., Ltd.; Haisco: Haisco Pharmaceutical Group Co., Ltd.; Dong-A ST: Dong-A ST Co., Ltd.

The first type of DPP-IV inhibitors contain pyrrolidine or similar compounds as P1 fragments linked to α-aminoacyl chains. In this type of inhibitor, the cyanopyrrolidine and thiazole moieties bind to the S1 pocket of DPP-IV, and the nitrile forms a covalent acyl ester with the hydroxyl group of Ser630. The imide nitrogen forms hydrogen bonds with the hydroxyl group of Tyr547. Vildagliptin is an example of this type of inhibitor. The design concept of early cyanopyrrolidine inhibitors was based on the hydrolytic mechanism of substrate peptide bonds by serine proteases [97]. Saxagliptin adds cyclopropane to cyanopyrrolidine, which improves the chemical stability of cyanopyrrolidine. It also interacts with Tyr666 in the S1 domain to increase its inhibitory effect. In addition, fluorinated pyrrolidine can form hydrophobic interactions in the S1 pocket or hydrogen bonds with Ser630 or Tyr631 to produce more potent inhibitory effects. The carbonyl group in the α-aminoacyl linker can form hydrogen bonds with Asn 710, whereas the primary/secondary amino group forms salt bridges with Glu205 and Glu206. In addition, positively charged group inhibitors can also form linker hydrogen bonds with Tyr662. Replacing the primary/secondary amino group with a tertiary amine or carbon weakens the inhibitory effect. The P2 fragment of this type of inhibitor binds to the S2 subunit of DPP-IV, and the hydroxyadamantyl moiety of vildagliptin and saxagliptin can form hydrogen bonds with Ser209 and Tyr547. However, the S2 pocket of DPP-IV is tolerant of many lipophilic rings, so lipophilic substituents reduce its inhibitory potential. The pyrazole pyrimidine structure or pyrimidine at position 4 of piperazine can interact with the benzene ring of Phe357 in a π‒π manner. The benzene ring of piperazine pyrazole interacts with the S2 broad domain of the enzyme and has hydrophobic interactions with Ser209 and Arg358. The π‒π interaction with Phe357 and the negative environment near Arg358 can significantly improve the inhibition efficiency. However, saxagliptin adds cyclopropane to cyanopyrrolidine, which improves the chemical stability of cyanopyrrolidine. It also interacts with Tyr666 in the S1 domain to increase its inhibitory effect. In addition, fluorinated pyrrolidine can form hydrophobic interactions in the S1 pocket or hydrogen bonds with Ser630 or Tyr631 to produce stronger inhibitory effects.

The second type of inhibitor contain P1 fragments consisting of pyridine-2,4-dione or similar compounds, including xanthine; examples include alogliptin, trelagliptin, and linagliptin, which not only bind to the S1 and S2 domains of DPP-IV but also interact with the S1 and/or S2 pockets of the enzyme. Pyrimidine-2,4-dione interacts with Tyr547 in a π‒π manner, causing conformational changes in the S1 pocket. Alogliptin is a powerful and selective DPP-IV inhibitor based on the pyrimidine diketone. Syrrx (Takeda San Diego) found that alogliptin has higher selectivity to DPP-IV in vitro than DPP-VIII and DPP-IX. Additionally, in a variety of diabetes animal models, it can reduce the levels of HbA1c, plasma glucose, glucagon and triglycerides; improve β-cell function; and significantly increase plasma insulin levels [98]. There can also be π‒π interactions between the xanthine scaffold and the phenyl group of Tyr547. In addition, the tricyclic imidazolo[1,2-a] purinone is stacked on Tyr547. The carbonyl oxygen of pyrimidine-2,4-dione forms hydrogen bonds with Tyr631, and the cyanobenzyl group of pyrimidine-2,4-dione effectively fills the S1 domain. The nitrile group forms hydrogen bonds with Arg125, and the amino group of the aminopiperidine moiety can form salt bridges with Glu205 and Glu206. The butyne group on the core structure of xanthine can bind to the S1 subunit, increasing its inhibitory effect. The phenyl group of the N3 quinazoline substituent of pyrimidine-2,4-dione can form π-π interactions with Trp629 in the S2 pocket, and the imidazole on tricyclic imidazo [1,2-a] purine interacts with Trp629 and Lys554. However, the steric hindrance caused by the conformational change of Trp629 reduces the inhibitory activity, whereas the interaction with Lys554 is beneficial for enhancing the inhibitory effect.

The third type of DPP-IV inhibitor using phenylethylamine or similar compounds as P1 fragments linked to β-aminobutyryl. Halogen substitution of phenyl ethylene diamine-based compounds results in enhanced inhibitory efficacy. These inhibitors include evogliptin, sitagliptin and gemigliptin. Trifluorophenyl or similar structures bind to the S1 pocket of DPP-IV and exhibit more potent inhibitory effects than difluorophenyl derivatives do. Nonfluorine atoms can form hydrogen bonds with Tyr631, whereas some fluorine atoms can form hydrophobic interactions with Tyr666 and Tyr662. The β-amino group of the butyryl linker forms hydrogen bonds with the hydroxyl oxygen of Tyr662 and the carboxyl oxygen of Glu205 and Glu206 in the S2 subsite, whereas the carbonyl group forms hydrogen bonds with Tyr547. Replacing β-aminobutyryl with cyclohexylamine as a linking structure also results in excellent pharmacokinetic properties, where nitrogen can form hydrogen bonds with Glu205, Glu206, and Tyr662. For the P2 fragments of drugs such as sitagliptin, the triazole pyrazine moiety of the trifluoromethyl substituent binds to the S2 subsite of DPP-IV, and the triazole piperazine moiety is located on Phe357. The trifluoromethyl substituent interacts with Arg358 and Ser209, which helps stabilize inhibitor–enzyme binding. The P2 can also be substituted with trifluoromethyl pyrimidine pyridine or piperazine-2-one. The tert butoxymethyl group on piperazine-2-one interacts with Arg125, and the electrostatic intermolecular interactions generated by the negative environment near Arg125 increase its affinity for DPP-IV.

Based on clinical needs, the development of long-acting agents can help simplify treatment and improve patient compliance. In the process of developing DPP-IV inhibitors, reducing the frequency of administration has been the main focus, with reductions from twice a day to once a day, once a week, and biweekly. Additionally, the beneficial effects of long-acting hypoglycemic preparations are progressing. The ultra-long-acting DPP-IV inhibitor coagliptin, a biweekly oral administration recently launched in China, is expected to be the preferred treatment. Pharmacodynamic studies have shown that this biweekly dosing preparation has comparable efficacy to the daily preparation in terms of blood glucose control, demonstrating excellent safety characteristics both as a monotherapy and in combination therapy regimens. Furthermore, this drug does not require dose adjustment in special patient groups (such as those with impaired liver or kidney function), and this characteristic significantly enhances its clinical applicability in the comprehensive management of T2DM [99].

5.2. Adverse reactions of marketed DPP-IV inhibitors

Although the overall safety of DPP-IV inhibitors is good, their potential adverse reactions still cannot be ignored, which has become one of the main reasons for the discontinuation of such drugs in clinical practice [100]. When used as a monotherapy, DPP-IV inhibitors primarily exert their glucose-lowering effects by potentiating the activity of endogenous GLP-1 and GIP, thereby stimulating insulin secretion in a glucose-dependent manner. Owing to the moderate and glucose-sensitive nature of this insulinotropic effect, these agents are generally not associated with significant hypoglycemia. However, when combined with insulin secretagogues such as sulfonylureas, the risk of hypoglycemia may be substantially increased [101]. This synergistic effect may be attributed to dual mechanisms of action: DPP-IV inhibitors prolong GLP-1-mediated insulin secretion by preventing its degradation, whereas sulfonylureas directly stimulate insulin release from pancreatic β-cells through ATP-sensitive potassium channel blockade. The concurrent use of these agents may therefore lead to excessive insulin secretion and consequent hypoglycemia. The hypoglycemic events associated with this combination therapy are significantly correlated with hepatic dysfunction, sex differences, and frequent alcohol consumption (≥3 times/week), all of which may contribute to glucose dysregulation through the impairment of hepatic gluconeogenesis pathways [102]. In the immune system, DPP-IV, also known as CD26, is an important lymphocyte surface protein that performs multiple key functions in the immune system, especially in the process of T-cell activation. Its activity plays an irreplaceable role in maintaining lymphocyte homeostasis, regulating T-cell activation and costimulatory signal transduction, and promoting the generation of memory T cells [16]. Therefore, DPP-IV inhibitors can exert immunomodulatory effects by inhibiting the activity of this enzyme. Clinical research data further show that treatment with DPP-IV inhibitors may increase the risk of infection, with common infections including nasopharyngitis, urinary tract infections and upper respiratory tract infections [73,103]. Pancreatitis is an established but rare side effect of DPP-IV inhibitors and occurs very infrequently [104]. Sitagliptin carries a greater risk of pancreatitis, whereas alogliptin is potentially associated with an increased risk of pancreatic cancer according to the pharmacovigilance data [105]. Animal experiments have confirmed that sitagliptin can increase circulating amylase levels and the proportion of REG3β-positive cells in the pancreas of mice fed normal and high-fat diets. REG3β is one of the most well-known activators of pancreatitis; it can activate the CXCL12/CXCR4 signaling cascade to promote the interaction between epithelial cells and immune cells, evade immune surveillance to promote tumor escape, and promote pancreatic ductal adenocarcinoma metastasis [106]. In addition, DPP-IV inhibition increases GLP-1 circulation time, which in turn induces excessive growth of pancreatic acinar and ductal cells, leading to occlusion and pancreatitis. A cohort study from Taiwan revealed that the use of DPP-IV inhibitors may be associated with liver dysfunction and an increased risk of liver failure in patients with T2DM complicated with cirrhosis [107]. Further analysis showed that compared with GLP-1 receptor agonists, DPP-IV inhibitors statistically significant associations between DPP-4 inhibitors and hepatobiliary adverse events like cholelithiasis and cholecystitis [108]. One possible mechanism is that the inhibition of DPP-IV enzyme activity may lead to an increase in the levels of proinflammatory chemokines (such as eotaxin-1), which has been shown to be a key pathogenic factor in drug-induced liver injury. A postmarketing surveillance study of linagliptin in patients with T2DM in South Korea revealed that 1.8% of these patients might have adverse reactions, among which the most common adverse reaction was gastrointestinal disease (0.7%) [109]. The mechanism of gastrointestinal adverse reactions related to DPP-IV inhibitors may be related to drug selectivity. Studies have shown that in inflammatory bowel disease, DPP-VIII/DPP-VI is highly expressed in colonic tissues [110]. However, owing to its relatively low selectivity, saxagliptin can simultaneously inhibit the activity of DPP-VIII and DPP-VI at extremely low concentrations, so the incidence of gastrointestinal adverse reactions associated with this drug is relatively low. Due to adverse reactions to DPP-IV inhibitors, systematic monitoring and standardized management are key to ensure the safety and effectiveness of treatment, improving the quality of clinical medical services, and promoting evidence-based medical research.

6. DPP-IV inhibitors from TCM screening

Traditional Chinese medicines (TCMs), including medicines derived from plants, animals and minerals, hold an important position in the history of human health. These traditional medicines are rich in bioactive components and show significant therapeutic potential. The diversity of unique chemical structures among these compounds provides a valuable resources for the identification of lead compounds and development of new drugs. By structurally modifying and pharmacologically optimizing these lead compounds, their therapeutic indices can be significantly enhanced, and ultimately, new drugs with greater clinical value can be developed. Natural medicines in the traditional medical system have been verified through long-term clinical practice, and their safety and therapeutic efficacy have been confirmed in a wide range of folk applications. Modern drug research and development can use this empirical evidence and transform traditional drugs into treatment methods that meet modern medical standards through systematic scientific methods. Natural sources of DPP-IV inhibitors refer to bioactive molecules isolated from plants, animals or microorganisms that can specifically inhibit the enzymatic activity of DPP-IV. These compounds show significant potential for the treatment of diabetes and related metabolic diseases. Plant-based DPP-IV inhibitors, including alkaloids, phenolic acids, flavonoids, quercetin and coumarin, have strong potential for treating diabetes [111]. As such, a fluorescent probe system with high specificity and anti-interference ability can be designed and developed for the efficient screening of DPP-IV inhibitors and large-scale system validation of TCM formulas.

The discovery of active ingredients under the guidance of “spectrum effect combination” is a new strategy that combines modern high-resolution chromatographic separation and analysis techniques with enzymatic systems (Fig. 7) to separate and analyze chemical components in natural medicinal resources (such as animals, plants, and microorganisms). In this process, a complex crude extract is separated to obtain corresponding components, and each component is further analyzed to identify the key substances. After the main chemical components are identified, their pharmacological activity is further evaluated, and their enzyme inhibition spectra are plotted. Finally, through multivariate correlation analysis and dose-effect curve fitting, the key components, such as those with significant inhibitory activity against DPP-IV, are identified, and the K_i of the main active components can be detected by enzyme kinetics experiments [112]. Furthermore, directly screening specific inhibitors of target enzymes (such as DPP-IV) from natural product compound libraries and their structural analog libraries is also a highly efficient screening strategy that is widely used in the field of drug discovery.

Fig. 7.

Efficient discovery of inhibitors based on the “spectrum effect combination” strategy and natural product library. (A) Screening of components with inhibitory potential from crude extracts of traditional Chinese medicines (TCMs). (B) Preparation of the liquid phase for component separation and collection. (C) Confimation of the components in distillates. (D) Evaluation of inhibition effect. (E) Research on inhibition mechanism. (F) Verification in vitro molecular docking. DPP-IV: dipeptidyl peptidase-IV.

6.1. TCM extracts

Screening DPP-IV inhibitors from TCM prescriptions or TCMs with hypoglycemic effects is a promising research direction that could provide new drugs for the treatment of diabetes and offer new opportunities for the modernization and internationalization of TCMs. In the theoretical system of TCM, diabetes belongs to the category of “eliminating thirst”, and it is prevented and treated mainly using compound preparations with the effects of nourishing yin and heat, supplementing qi and yin, and promoting blood circulation and removing blood stasis. Accordingly, the inhibitory effects of Liuwei Dihuang Wan and various TCM extracts with hypoglycemic effects on DPP-IV have been studied. Among them, extracts of Huanglian and Mudan bark have shown excellent inhibitory activity [113]. However, for specific chemical components with inhibitory activity, structural identification and functional verification still need to be carried out through systematic separation and purification experiments. In addition, researchers have successfully isolated and identified 43 chemical components from the Xiaokean formula, 13 of which had inhibitory effects on DPP-IV activity and 5 of which had significant inhibitory effects on DPP-IV. Among them, salvianolic acid C, ginsenoside Rg5, and saponin AI from Anemarrhena chinensis inhibited DPP-IV in a dose-dependent manner at concentrations ranging from 5 to 50 μmol/L [114]. Sponge diterpenoids isolated from the leaves of Amomum tsao-ko also inhibit the activity of DPP-IV [115]. Ulva reticulata extracts (petroleum ether, benzene, ethyl acetate, and methanol) can inhibit DPP-IV enzyme activity in a dose-dependent manner [116]. The extract of gardenia has potential inhibitory effects on DPP-IV, and further results revealed that the effective components of iridoid glycosides have significant inhibitory activity (IC₅₀ = 53 ± 0.63 μg/mL); among them, genipin 1-gentioside has been identified as a promising novel DPP-IV inhibitor in Gardenia jasminoides Ellis [117]. As shown in Table 2 [[113], [114], [115], [116], [117], [118], [119], [120], [121]], the three-stage progressive screening mode based on compound-component monomers can significantly improve the efficiency of discovering active ingredients in TCM and effectively reduce the randomness and blindness of traditional screening methods.

Table 2.

Inhibitory effects of Chinese medicines on dipeptidyl peptidase-IV (DPP-IV).

Chinese patent medicine	Chinese medicine	Components	Enzyme source	Inhibition rate (%)	Residual activity (%)			IC50	Refs.
					10 μg/mL	100 μg/mL	1000 μg/mL
Liuwei Dihuang pill	Aqueous extract	–	Plasma	–	93.47	91.96	70.41	–	[113]
–	Cortex Moutan	–	–	–	80.76	52.46	35.44	–	–
–	Poria Cocos	–	–	–	99.90	100.15	97.84	–	–
–	Wine Cornus	–	–	–	96.85	96.20	81.13	–	–
–	Rhizoma Dioscoreae	–	–	–	88.94	91.14	89.87	–	–
–	Radix Rehmanniae Preparata	–	–	–	93.96	91.46	89.64	–	–
–	Alisma Orientalis	–	Plasma	–	97.93	99.52	95.28	–	–
Jiangtangjia tablet	Aqueous extract	–	–	–	94.93	90.93	70.36	–	–
–	Rhizoma Polygonati	–	–	–	106.12	96.47	95.56	–	–
–	Rehmannia Glutinosa	–	–	–	104.21	102.19	85.92	–	–
–	Radix Pseudostellariae	–	–	–	96.18	94.23	95.80	–	–
–	Trichosanthis Radix	–	–	–	97.92	98.49	95.27	–	–
–	Astragalus Membranaceus	–	–	–	93.04	97.52	95.49	–	–
Ganlu Xiaoke capsule	Aqueous extract	–	Plasma	–	92.84	84.38	71.75	–	–
–	Radix Rehmanniae Preparata	–	–	–	93.96	91.46	89.64	–	–
–	Rehmannia Glutinosa	–	–	–	104.21	102.19	85.92	–	–
–	Lycii Fructus	–	–	–	100.33	97.90	99.22	–	–
–	Cortex Lycii	–	–	–	101.76	98.45	81.56	–	–
–	Fructus Corni	–	–	–	104.40	96.15	66.29	–	–
–	Radix Scrophulariae	–	–	–	104.80	98.08	96.41	–	–
–	Ginseng Radix et Rhizoma	–	–	–	91.40	91.88	91.52	–	–
–	Radix Codonopsis	–	–	–	97.90	88.48	79.44	–	–
–	Astragalus Membranaceus	–	–	–	93.04	97.52	95.49	–	–
–	Semen Cuscutae	–	–	–	89.24	79.92	52.99	–	–
–	Trichosanthis Radix	–	–	–	97.92	98.49	95.27	–	–
–	Radix Angelicae Sinensis	–	–	–	95.08	94.88	89.46	–	–
–	Rhizoma Coptidis	–	–	–	99.54	82.99	27.90	–	–
–	Rhizoma Atractylodis Macrocephalae	–	–	–	97.27	99.98	94.79	–	–
–	Ootheca Mantidis	–	–	–	102.54	97.81	89.97	–	–
–	Radix Asparagi	–	–	–	102.86	98.42	99.39	–	–
–	Ophiopogon Japonicus	–	–	–	99.50	100.87	95.26	–	–
–	Alisma Orientalis	–	–	–	97.93	99.52	95.28	–	–
–	Trichosanthis Radix	–	–	–	97.92	98.49	95.27	–	–
–	Ophiopogon Japonicus	–	–	–	99.50	100.87	95.26	–	–
–	Radix Glycyrrhizae	–	–	–	101.03	97.37	90.46	–	–
–	Fructus Schisandrae Chinensis	–	–	–	100.36	98.97	87.29	–	–
–	Puerarialobata	–	–	–	94.86	93.68	84.58	–	–
–	Rehmannia Glutinosa	–	–	–	104.21	102.19	85.92	–	–
–	Radix Glycyrrhizae	–	–	–	101.03	97.37	90.46	–	–
–	Fructus Corni	–	–	–	104.40	96.15	66.29	–	–
–	Stigma Maydis	–	–	–	84.90	82.80	63.90	–	–
–	Rehmannia Glutinosa	–	–	–	104.21	102.19	85.92	–	–
–	Cortex Lycii	–	–	–	101.76	98.45	81.56	–	–
–	Ophiopogon Japonicus	–	–	–	99.50	100.87	95.26	–	–
–	Poria Cocos	–	–	–	99.90	100.15	97.84	–	–
Yuquan pill	Aqueous extract	–	Plasma	–	92.54	90.61	79.73	–	–
–	Trichosanthis Radix	–	–	–	97.92	98.49	95.27	–	–
–	Ophiopogon Japonicus	–	–	–	99.50	100.87	95.26	–	–
–	Radix Glycyrrhizae	–	–	–	101.03	97.37	90.46	–	–
–	Fructus Schisandrae Chinensis	–	–	–	100.36	98.97	87.29	–	–
–	Puerarialobata	–	–	–	94.86	93.68	84.58	–	–
–	Rehmannia Glutinosa	–	–	–	104.21	102.19	85.92	–	–
Jiangtangning capsule	Aqueous extract	–	Plasma	–	100.15	94.39	81.82	–	–
–	Radix Glycyrrhizae	–	–	–	101.03	97.37	90.46	–	–
–	Fructus Corni	–	–	–	104.40	96.15	66.29	–	–
–	Stigma Maydis	–	–	–	84.90	82.80	63.90	–	–
–	Rehmannia Glutinosa	–	–	–	104.21	102.19	85.92	–	–
–	Cortex Lycii	–	–	–	101.76	98.45	81.56	–	–
–	Ophiopogon Japonicus	–	–	–	99.50	100.87	95.26	–	–
–	Poria Coco	–	–	–	99.90	100.15	97.84	–	–
–	Trichosanthis Radix	–	–	–	97.92	98.49	95.27	–	–
–	Astragalus Membranaceus	–	–	–	93.04	97.52	95.49	–	–
–	Rhizoma Anemarrhenae	–	–	–	94.97	88.57	89.81	–	–
–	Linn Gypsum	–	–	–	93.52	95.79	94.31	–	–
–	Rhizoma Dioscoreae	–	–	–	88.94	91.14	89.87	–	–
–	Ginseng Radix et Rhizoma	–	–	–	91.40	91.88	91.52	–	–
Xiaokeling tablet	Aqueous extract	–	Plasma	–	97.30	96.27	80.58	–	–
–	Ophiopogon Japonicus	–	–	–	99.50	100.87	95.26	–	–
–	Fructus Schisandrae Chinensis	–	–	–	100.36	98.97	87.29	–	–
–	Cortex Moutan	–	–	–	80.76	52.46	35.44	–	–
–	Rhizoma Coptidis	–	–	–	99.54	82.99	27.90	–	–
–	Astragalus Membranaceus	–	–	–	93.04	97.52	95.49	–	–
–	Radix Ginseng Rubr	–	–	–	105.68	100.99	108.88	–	–
–	Poria Cocos	–	–	–	99.90	100.15	97.84	–	–
–	Rehmannia Glutinosa	–	–	–	104.21	102.19	85.92	–	–
–	Linn Gypsum	–	–	–	93.52	95.79	94.31	–	–
–	Lycii Fructus	–	–	–	100.33	97.90	99.22	–	–
–	Trichosanthis Radix	–	–	–	97.92	98.49	95.27	–	–
Xiaokean formula	–	Salvianolic acid C	pDPP-IV	95.57 (50 μM)	–	–	–	–	[114]
–	–	Ginsenoside RG5	–	92.12 (50 μM)	–	–	–	–	–
–	–	Ginsenoside-Ro	–	55.73 (50 μM)	–	–	–	–	–
–	–	Timosaponin AI	–	53.61 (50 μM)	–	–	–	–	–
–	–	Timosaponin AIII	–	45.15 (50 μM)	–	–	–	–	–
–	Leaves of Amomum tsao-ko	–	–	–	–	–	–	311.0 μM	[115]
–	Ulva reticulata	methanolic extract	pDPP-IV	–	–	–	–	42.12 μg/mL	[116]
–	Gardenia jasminoides Ellis	Gardenia jasminoides Ellis extract	hDPP-IV	–	–	–	–	2270 μg/mL	[117]
–	–	Iridoid glycosides	–	–	–	–	–	53 μg/mL	–
–	–	Genipin 1-gentiobioside	–	–	–	–	–	–	–
–	Elatostema tenuicaudatum W.T.Wang	Crude extract	hDPP-IV	–	–	–	–	220.5 μg/mL	[118]
–	–	Compound 5	–	–	–	–	–	141.7 μM	–
–	–	Compound 6	–	–	–	–	–	151.2 μM	–
–	–	Compound 8	–	–	–	–	–	107.9 μM	–
–	–	Compound 14	–	–	–	–	–	71.9 μM	–
–	Senna siamea	Resveratrol	hDPP-IV	–	–	–	–	611.80 μM	[119]
–	Moringa oleifera Lam.	O-Ethyl-4-[(α-l-rhamnosyloxy)-benzyl] carbamate	hDPP-IV	–	–	–	–	0.798 μM	[120]
–	Caryopteris incana (Thunb.) Miq	Caryopincaolide C	hDPP-IV	–	–	–	–	54.2 μM	[121]
–	–	Caryopincaolide D	–	–	–	–	–	222.9 μM	–
–	–	Caryopincaolide L	–	–	–	–	–	168.7 μM	–
–	–	Prineoparaquinone	–	–	–	–	–	115.9 μM	–
–	–	6,11-Dihydroxy-12-methoxy 5,8,11,13-abietatetraen-7-one	–	–	–	–	–	228.9 μM	–
–	–	Salvicanaraldehyde	–	–	–	–	–	178.3 μM	–

pDPP-IV: pig-derived DPP-IV; hDPP-IV: human-derived DPP-IV; IC₅₀: half maximal inhibitory concentration; –: no data.

6.2. Flavonoid inhibitors

As one of the most structurally diverse and pharmacologically active metabolites, natural flavonoids remain an important source for discovering new drugs and active molecular templates. Research shows that high intake of dietary flavonoids is significantly negatively correlated with the risk of diabetes, and high intake can reduce the risk of diabetes by 28% [122]. Flavonoids can significantly reduce blood glucose levels, and their mechanism of action is related to the inhibition of alpha glucosidase activity and pancreatic tissue protection. In addition, flavonoids can exert further synergistic regulatory effects through antioxidant, anti-inflammatory, and glucose tolerance improvement pathways [123]. In vitro studies have shown that various flavonoids can significantly inhibit DPP-IV activity in Caco-2 cells and downregulate its protein expression level [124]. The fermentation of black wolfberry juice with Lactobacillus rhamnosus GG (LGG) can increase the content of total flavonoids, resulting in greater DPP-IV inhibition [125]. The inhibitory effects of 30 dietary flavonoids on DPP-IV were studied, and it was found that isorhamnetin-3-O-glucoside (IC₅₀, 6.53 ± 0.280 μM) had the strongest inhibitory effect, followed by anthocyanin-3-O-glucoside and isorhamnetin-3-O-rutinoside. The SAR revealed that the 3rd position of the C ring is advantageous for large hydrogen bond acceptors as well as hydrophilic and electron-donating substituents. The presence of secondary and electron withdrawing groups at position 4 of ring B and 5 and 7 of ring A can improve the inhibition of DPP-IV, and all three flavonoids can inhibit the expression of DPP-IV in Caco-2 cells [124]. The flavonoid schaftoside was screened from 41 sugarcane leaf extracts and shown to significantly inhibit DPP-IV, and the residues Arg125 and Tyr662 may play key roles in inhibiting DPP-IV activity [126]. Three flavonoid glycosides (1–3) isolated from the seeds of a legume plant (Lens culinaris Medikus) can inhibit DPP-IV activity in a concentration-dependent manner, and their flavonoid core structure shows the best electrostatic attraction interaction with the catalytic triad residues of DPP-IV [127]. Ethanol extracts of Acacia Arabica, which contains mainly flavonoids, tannins and anthraquinone compounds, can inhibit DPP-IV activity in 3T3L1 cells, increase muscle glucose uptake, and inhibit DPP-IV enzyme activity in rats fed a high-fat diet (HFD), improving glucose tolerance, plasma insulin, and GLP-1 levels [128]. The ethyl acetate extract BsE from the stem of Scutellaria baicalensis contains the most phenolic and flavonoid compounds, which strongly inhibited DPP-IV. Among them, quercetin has the strongest ability to inhibit DPP-IV (IC₅₀ = 8.25 μM) [129]. Quercetin can increase glucose tolerance and insulin secretion by pancreatic β-cells while inhibiting α-glucosidase and DPP-IV enzymes, thereby prolonging the half-lives of GLP-1 and GIP [130]. Compared with treatment with sitagliptin alone, combination therapy with quercetin/sitagliptin can improve blood glucose control, metabolic profile, oxidative and inflammatory status, pancreatic islet structure, and β-cell function in patients with diabetes mellitus [131].

6.3. Polyphenol inhibitors

Phenolic acids are a class of organic acids containing phenolic rings that generally exist widely in nature as secondary metabolites. Multiple phenolic compounds have also been proven to have strong inhibitory effects on DPP-IV. After high-pressure homogenization, free phenols, esterified phenols, and insoluble bound phenols in "Lijiang Snow" peach juice can effectively inhibit α-glucosidase and DPP-IV via hydrogen bonds and van der Waals forces [132]. Sea fennel extract showed high DPP-IV inhibition ability (73% at 1 mg/mL), and further research revealed that chlorogenic acid seemed to play an important role in its inhibitory activity [133]. Gastrointestinal digestion simulated in vitro revealed that the digestible fraction of adzuki bean paste subjected to high pressure for 10 min had increased DPP-IV inhibitory activity; the digestible fraction contained various phenolic substances, such as coumaric acid, sinapine, and eugenic acid [134]. The halophyte ice plant (Mesembryanthemum crystallinum) showed strong DPP-IV inhibitory activity (with an inhibition rate of 98.6% at a final concentration of 1 mg/mL), and fractionation studies of the extract revealed that polyphenols may be the main reason for the above activity [135]. Phenolic compounds from fermented berry beverages modulated DPP-IV and its substrate GLP-1 to increase insulin secretion and to upregulate the expression of mRNAs encoding insulin receptor-associated genes and proteins in pancreatic β-cells [136]. The ethyl acetate components of Taiwanese black grape stems were separated into (+)-hopeaphenol, (+)-vitin A, and (−)-vitin B, which inhibited pig kidney DPP-IV (IC₅₀ = 401, 90.75, and 15.3 μM), and (+)-vitin A and (−)-vitin B exhibited mixed noncompetitive inhibitory effects [137]. Among natural phenolic compounds, curcumin, eugenic acid and resveratrol have high affinity for DPP-IV, and curcumin and resveratrol have shown potential inhibitory effects on DPP-IV. For example, curcumin can inhibit DPP-IV activity in Caco-2 cells. In addition, animal experiments have shown that curcumin can target DPP-IV to reduce blood glucose [138]. To improve the stability of curcumin and its pharmacokinetic properties, researchers found that curcumin can form complexes with ruthenium (II) coordination and retain excellent DPP-IV inhibitory activity [139].

6.4. Alkaloid inhibitors

Alkaloids typically inhibit the activity of DPP-IV by competitively binding to its active site, thereby preventing the degradation of hormones such as GLP-1 by DPP-IV and helping to maintain blood glucose balance. Berberine is a known natural DPP-IV inhibitor with an IC₅₀ of 16.33 ± 1.34 μM, and studies have shown that inhibiting DPP-IV is one of the mechanisms of the antihyperglycemic effects of berberine [140]. Short-term administration of berberine can reduce the postprandial level of streptozotocin-induced diabetes in rats by the local inhibition of intestinal DPP-IV [141]. In addition, berberine has protective effects on myocardial necrosis induced by isoproterenol in diabetes, which may be attributed to its beneficial effects on diabetes and hyperlipidemia [142]. The DPP-IV inhibitory activity of nine alkaloids in Huanglian was detected; among them, columbamine, demethylenebergerin and coptisine exert significant inhibitory effects, with IC₅₀ values below 10 μM. These alkaloids exhibit rapid binding and dissociation with DPP-IV through electrostatic interactions and van der Waals forces [143]. The improvement in blood glucose control, metabolic parameters and liver function in diabetic rats mediated by trigonelline may be related to its inhibition of DPP-IV in plasma and the small intestine [144]. Alkaloids, as DPP-IV inhibitors, have good pharmaceutical potential, but further research and optimization are still needed to promote their clinical application.

6.5. Protein hydrolysate inhibitors

Food ingredients, including food-derived bioactive peptides, have been recommended as sources of DPP-IV inhibitors with no side effects. Adzuki beans are a type of bean with a low glycemic index, which is traditionally used as a metric to help regulate T2DM. In healthy and insulin-resistant cell models, digested adzuki bean β-vignin significantly increased hepatocyte glucose uptake by inhibiting DPP-IV (>40%) [65]. Both fermented and unfermented lentil flour subjected to in vitro digestion inhibited DPP-IV activity in Caco-2 cells [145]. Yak hemoglobin hydrolysate also contains a variety of DPP-IV inhibitory peptides, and the hydrogen bonds formed by the core residues in the DPP-IV S2 pocket and electrostatic attraction may be the main reasons for their inhibitory properties [146]. In the active peptides of tilapia peel gelatin hydrolysate, peptide chain length plays a dominant role in DPP-IV inhibition, with 4 to 9 peptides showing the strongest inhibitory efficacy, with IC50 values 2.15-10.43 times lower than Xaa Pro type tripeptides, demonstrating stronger inhibitory activity [147]. The ultrafiltration fraction isolated from black tea extract showed DPP-IV inhibitory activity in vitro, with peptide II, with a molecular weight of 976 Da, exhibiting the highest DPP-IV inhibitory activity in vitro. Animal experiments have shown that after the administration of peptide II, GLP-1 and insulin levels in hyperglycemic mice increased, and pancreatic beta-cell function improved [148]. Although significant progress has been made in the research of food-based proteolytic peptides as DPP-IV inhibitors, their effects in cell models and animal models still need to be further studied to verify their bioavailability and efficacy. In addition, more efficient enzymatic hydrolysis processes and screening methods need to be explored to improve the yield and activity of DPP-IV inhibitory peptides. As DPP-IV inhibitors, these peptides have broad application prospects, but more research is needed to support practical application.

6.6. Other types of compounds

The synthesized amino chalcone compounds 3c, 3b, 3d, and 3e and the natural hydroxy chalcone compound 3i isolated from medicinal plants can also significantly inhibit the activity of DPP-IV [149]. A series of liquorice chalcone A analogs also exhibit DPP-IV inhibitory activity, with nitro-substituted analog 27 showing the strongest inhibitory activity (K_i = 0.96 μM) [150]. Deep learning and sequential metabolism research has revealed that the effective components of iridoid glycosides in Gardenia jasminoides have significant DPP-IV inhibitory activity (IC₅₀ 53 ± 0.63 μg/mL), among which genipin 1-gentiopicroside has been identified as a promising novel DPP-IV inhibitor [117]. For these compounds with different structures, in the future, their structures can be further optimized through computer-aided drug design and molecular dynamics simulations to increase their bioavailability and metabolic stability.

7. Conclusion and prospects

Systems for DPP-IV enzyme activity detection are based on specific substrate enzymatic hydrolysis reactions, and the cleavage products can be quantitatively analyzed by spectrophotometry or fluorescence methods. These tests are conducted using standardized systems with strict control of the reaction conditions (including pH and temperature), and the enzymatic activity level is evaluated by measuring the substrate conversion rate, which accurately reflects the functional status of DPP-IV in different biological systems. Current enzymatic activity-based detection methods face several critical challenges, including: 1) the uncertainty of enzyme selectivity, where multiple enzymes may react simultaneously with the substrate, leading to potentially distorted results; 2) the unclear metabolic relevance between artificial substrates and endogenous peptide metabolism in vivo; and 3) the availability of very few substrates that are suitable for direct in vivo detection in clinical practice, despite numerous reported biological applications. To overcome these limitations, we believe that future research should prioritize optimizing enzyme specificity as the most crucial aspect. Additionally, establishing clearer correlations between the metabolism of different peptides of DPP-IV is essential for better biological simulation and inhibitor optimization. Furthermore, the development of NIR or infrared-based probes may represent a promising approach to facilitate the clinical translation of these detection methods.

As DPP-IV inhibitors are important therapeutic agents for T2DM, with well-established clinical efficacy, DPP-IV inhibitors may still cause a series of adverse reactions, including potentially serious adverse events. Given these safety considerations, probe-based detection methods have significant potential for the future development of DPP-IV inhibitors. The core goal of the research and development of DPP-IV inhibitors has shifted to enhancing the efficacy and safety of these drugs. By integrating the structural biology insights obtained from high-dimensional data analysis, researchers can conduct rational drug design for DPP-IV-specific conformations or allosteric binding sites, and this strategy is expected to lead to the development of new inhibitors with several advantages. (1) New inhibitors may show enhanced target specificity for DPP-IV. Significant progress has been made in this respect. Extensive inhibitor screening and protein conformational analyses have elucidated selective binding sites, substantially reducing off-target molecular effects. (2) New inhibitors may show optimized pharmacodynamic profiles. Beyond improved target selectivity, establishing intervention strategies for specific peptide metabolic pathways through elucidating correlations between probe detection and endogenous peptide metabolism can increase disease-specific selectivity. This approach simultaneously mitigates adverse effects caused by reduced hydrolysis of other peptides. A prime example is the development of selective inhibitors that specifically target the metabolism of GLP-1 and GIP by DPP-IV. (3) New inhibitors may have expanded therapeutic indications. Notably, the physiological functions of DPP-IV extend far beyond glycemic control. Substantial evidence has demonstrated the key regulatory roles in pathological processes, including tumorigenesis, autoimmune disorders, and neurodegenerative diseases. Therefore, in-depth mechanistic studies of DPP-IV in multiple diseases not only hold significant theoretical value but could also provide novel targets and strategies for treating various diseases. (4) The development of DPP-IV inhibitors targeting different species represents a crucial direction in drug discovery, particularly given the substantial therapeutic potential arising from the differences between gut microbe- and human-derived enzymes. Concurrently, developing inhibitors with differential metabolic impacts will establish a solid foundation for broadening therapeutic applications. Future research on DPP-IV enzyme inhibitors will include the development and application of new high-dimensional detection systems to elucidate the regulatory mechanism of this enzyme in complex biological systems. Such advanced technology platforms, owing to their outstanding resolution and detection sensitivity, will be capable of accurately analyzing the dynamic interaction networks between DPP-IV and substrates, inhibitors, and other biological macromolecules.

Enhancing medication safety constitutes a critical scientific challenge in DPP-IV inhibitor development. The ongoing advancement of probe technology promises to establish essential foundations for the clinical translation of both diagnostic probes and therapeutic inhibitors. This comprehensive review evaluates the application of various probes and inhibitors in both preclinical and clinical settings, with particular emphasis on the screening of TCMs for DPP-IV inhibitory components. Through the implementation of lifecycle safety optimization strategies in drug development, we aim to facilitate the creation of next-generation DPP-IV inhibitors characterized by enhanced efficacy and reduced toxicity, ultimately delivering superior clinical therapeutics for diverse disease populations.

CRediT authorship contribution statement

Hong-Hong Ma: Writing – review & editing, Writing – original draft, Validation, Investigation, Conceptualization. Xiao-Dong Li: Writing – review & editing, Writing – original draft, Visualization, Investigation. Xing-Kai Qian: Supervision, Project administration, Funding acquisition, Conceptualization. Li-Wei Zou: Supervision, Project administration, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.