Biochemical properties

The biochemical characterization of ENTPDases is crucial in developing inhibitors that can be used in pathophysiological conditions. One of the main challenges in the characterization of ENTPDases is the loss of enzymatic activity in the presence of detergents, destabilizing oligomeric complexes, leading to the formation of monomers with significantly reduced catalytic activity. To address this issue, researchers performed transient transfection of COS-7 cells, allowing the analysis of ENTPDases in their native, membrane-bound forms [2–4, 7, 8]. Here, we will focus on the biochemical characterization data of human ENTPDases.

ENTPDases 1, 2, 3, and 8 differ in substrate specificity, pH sensitivity, and preference for Ca²⁺ or Mg²⁺ as cofactors. They all hydrolyze nucleoside triphosphates such as ATP and UTP, but their hydrolysis rates for nucleoside diphosphates vary significantly among subtypes [3, 7]. NTPDase1 hydrolyzes ATP and ADP equally well, converting ATP directly to adenosine monophosphate (AMP) with minimal ADP accumulation [2]. In contrast, NTPDase2 strongly prefers ATP, leading to sustained ADP accumulation before its conversion to AMP. NTPDases 3 and 8 favor ATP over ADP, allowing transient ADP accumulation. These enzymes follow Michaelis-Menten kinetics, with Km values for ATP of 17, 70, 75, and 81 μM for NTPDases 1, 2, 3, and 8, respectively, indicating that NTPDase1 has the highest ATP affinity [2, 7, 8].

Their cofactor preferences also vary. While NTPDases 1 and 2 show no clear preference between Ca²⁺ and Mg²⁺, NTPDases 3 and 8 demonstrate a preference for Ca²⁺ [7, 8]. All enzymes exhibit reduced or abolished activity in the presence of ethylenediaminetetraacetic acid (EDTA) and EGTA and hydrolyze uracil nucleotides more efficiently in the presence of Ca²⁺. They are also highly active at physiological pH, though with some differences: NTPDase1 functions within a more restricted range, predominantly active under neutral to slightly alkaline conditions [7].

Structurally, ENTPDases share about 40% sequence identity among family members and consist of approximately 500 amino acid (aa) residues, with the molecular mass of glycosylated monomers ranging from 70 kDa to 80 kDa [2]. All ENTPDase family members undergo N-glycosylation, essential for proper folding, stability, membrane targeting, and enzymatic activity [4]. However, the specific glycan structures of these enzymes remain unknown, and research in this area is still limited. The predicted numbers of N-glycosylation sites are 6, 6, 7, and 8 for NTPDases 1, 2, 3, and 8, respectively [2].

There are some splicing variants for human NTPDases 2 and 3. NTPDase2 has three variants: NTPDase2α (495 aa), NTPDase2β (472 aa), and NTPDase2γ (450 aa) [9]. Only NTPDase2α is active, as the sequence missing in NTPDase2β and NTPDase2γ variants contains a conserved cysteine residue (C399), essential for disulfide bond formation. Although NTPDase2 is still called CD39L1, the original sequence of human CD39L1 corresponds to the inactive NTPDase2β. For NTPDase3, there are two variants: NTPDase3α (529 aa) and NTPDase3β (459 aa) [10]. The shorter variant, NTPDase3β, lacks the C-terminal end containing the ACR5 domain and does not exhibit measurable activity. However, it may still play an important modulatory role in the activity of NTPDase3α.

Understanding the molecular mechanisms involved in nucleotide hydrolysis and the development of inhibitors has advanced significantly following the resolution of the crystal structure of the extracellular domain of Rattus norvegicus NTPDase2 in 2008 [11]. The crystal structures with substrate analogs reveal two structural domains, each with an extended RNase H fold, a pattern also found in other members of the actin structural superfamily. The two domains are characterized by a central mixed β-sheet and a peripheral layer of predominantly alpha helices [1]. The binding site for the divalent metal ion and di- and triphosphate nucleosides is located in the cleft between the two domains, allowing precise interactions between the substrates and residues of the ACR regions. However, to date, no high-resolution structures, including the transmembrane domains, are available, and they can play a crucial role in modulating the rotational movements of the catalytic domains.

A general view of ENTPDases in pathological conditions

Ecto-NTPDases play a key role in purinergic signaling due to their ability to hydrolyze extracellular di- and triphosphate nucleotides, which act as ligands for P2 receptors involved in pro-inflammatory responses, promoting the release of cytokines and the activation of immune cells [12]. By tightly regulating the availability of nucleotides, ENTPDases act as key modulators of immune responses, preventing excessive inflammation and tissue damage. The activity of ENTPDases is complemented by the cooperative action of 5’-nucleotidases, which hydrolyze AMP to produce ADO, the primary ligand for P1 receptors responsible for anti-inflammatory effects by limiting excessive immune responses and promoting tissue repair. The precise balance between P2 and P1 receptor activation highlights the essential role of ENTPDases in immune homeostasis and inflammatory and neuroinflammatory diseases.

Dysregulation of ENTPDase activity is associated with various pathologies, particularly those involving vascular and immune dysfunction. In endothelial cells, NTPDase1 plays a critical role in modulating vascular inflammation and thrombosis by degrading ADP, the major recruiter for forming occlusive thrombi by activating P2Y₁ and P2Y₁₂ receptors on platelets [13]. This function is essential for maintaining vascular integrity, as an imbalance in NTPDase1 activity can lead to excessive thrombus formation and haemorrhagic tendencies. Additionally, this isoform controls vasorelaxation and is the primary enzyme responsible for regulating nucleotide metabolism on the surface of vascular smooth muscle cells, playing an essential role in the local modulation of vascular tone [2].

Like NTPDase1, NTPDase2 is also expressed in blood vessels, predominantly in vascular adventitial cells, microvascular pericytes, and the subendocardial space [14]. However, it acts predominantly as an ATPase, modulating nucleotide-mediated signaling in vascular and neural tissues. By converting ATP to ADP, NTPDase2 facilitates platelet aggregation and thrombus formation, playing an opposite role to NTPDase1 in regulating hemostasis. Thus, NTPDases 1 and 2 coordinate the regulation of pro- and anti-thrombotic responses, with NTPDase1 inhibiting platelet aggregation and recruitment in intact vessels, whereas expression of NTPDase2 promotes microthrombus formation in areas of extravasation after vascular injury.

In addition, studies show that NTPDase1 plays a critical role in creating an immunosuppressive environment in the tumor microenvironment that promotes cancer growth and progression. The conversion of ATP to ADO through the combined action of CD39 and CD73 favors the suppression of effector T cell activation, the promotion of regulatory T cells (Tregs), and the polarization of macrophages toward the M2 tumor phenotype and angiogenesis. High expression levels of these enzymes have been found in various tumors such as colorectal cancer, pancreatic cancer, and chronic lymphocytic leukemia. CD39 also exists in tumor exosomes and enhances the immunosuppressive effect by activating A2A and A2B receptors [15].

NTPDase2 also regulates ATP levels in the taste buds, preventing desensitization of P2X2 and P2X3 receptors and ensuring proper taste transmission [16]. Its deletion impairs this signaling. In the brain, it is present in germinal zones, suggesting a role in cell proliferation and neuronal differentiation [17]. In liver physiology, its expression by portal fibroblasts can inhibit the activation of P2Y receptor, limiting biliary epithelial proliferation, a process relevant to preventing cholangiopathies [18].

NTPDase3 is prominent in energy metabolism. Its global deletion in mice is associated with resistance to diet-induced obesity and obesity-associated glucose intolerance. This result was not caused by the expression of NTPDase3 in pancreatic beta cells, but probably by metabolic changes in adipocytes [19]. This alteration results from increased expression of uncoupling protein 1 (UCP-1) in brown adipose tissue and increased “browning” in inguinal white adipose tissue with positive regulation of UCP-1 and genes related to thermogenesis. In addition, NTPDase3 may be involved in the regulation of homeostatic functions, since its expression in the hypothalamus coincides with the onset of eating and sleeping rhythms in young rats [20], and there is evidence for its involvement in nociceptive circuits, modulating somatosensory purinergic transmission via P2X and P2Y receptors [21].

Using a model of colitis induced by DSS (dextran sulfate sodium), NTPDase8 was essential for nucleotide hydrolysis in the colon, preventing excessive activation of the pro-inflammatory P2Y₆ receptor [22]. Mice deficient in NTPDase8 showed more intense inflammation, more significant histological damage, and higher expression of pro-inflammatory cytokines compared to wild-type mice, while intra-rectal injection of the enzyme conferred complete protection against colitis. Daily administration of a P2Y₆ antagonist also reduced inflammation in a dose-dependent manner, and bone marrow transplantation from normal mice to P2Y₆-deficient mice also demonstrated a protective effect against inflammation.

ENTPDases in neuroinflammation and neuroinflammatory disorders

Inflammation is a complex signaling cascade in which the innate immune system responds to pathogenic or sterile insults. These processes are mediated by pattern recognition receptors (PRRs), which interact with pathogen-associated molecular patterns (PAMPs) and environment-derived danger-associated molecular patterns (DAMPs) from the environment. In the central nervous system (CNS), microglia and astrocytes are the primary mediators of neuroinflammatory responses, expressing PRRs such as Toll-like receptors (TLRs) that sense extracellular danger signals [23]. During inflammatory responses, inflammasomes are assembled, activating the pro-inflammatory caspase-1, promoting the maturation and release of the cytokines interleukin-1β (IL-1β) and IL-18. In addition, caspase-1 activity contributes to the release of extracellular nucleotides such as ATP and ADP, which amplify inflammatory signaling through purinergic receptors [24].

Neuroinflammation is associated with several CNS disorders, particularly those with a neurodegenerative component, such as Alzheimer’s disease (AD), Parkinson’s disease (PD), multiple sclerosis (MS), and Huntington’s disease (HD) [25]. NTPDases play a critical role in modulating neuroinflammatory pathways by regulating the availability of extracellular nucleotides, thereby influencing microglial activation and neuronal survival [26]. NTPDase1 is highly expressed in microglia, where it regulates extracellular nucleotide concentrations, limits excessive activation of pro-inflammatory P2 receptors, and helps reduce neuroinflammation [18]. In contrast, NTPDase2 is predominantly expressed in astrocytes in rats, which may contribute to the modulation of inflammatory responses within the CNS [27].

In the experimental autoimmune encephalomyelitis (EAE) animal model used to study MS, overexpression of NTPDase1 is associated with microglial polarization toward an anti-inflammatory phenotype, suggesting a possible role in promoting neuroprotection [28]. Studies also show that MS is involved in the release of ADP, which binds to the P2Y₁₃ receptor and activates neuronal survival pathways such as the ERK1/2 pathway [29]. During the progression of EAE, the expression of ENTPDase2 is reduced, limiting the activation of these receptors and favoring the inflammatory response mediated by Th1 and Th17 cells, promoting demyelination and axonal damage [30].

In AD, the accumulation of ATP in the extracellular environment can lead to sustained activation of P2X7 receptors on microglia, promoting a chronic inflammatory state and exacerbating neurodegeneration. NTPDase1 can counteract this process by hydrolyzing ATP, potentially reducing P2X7 receptor overstimulation and subsequent neuronal damage. Data also indicate that in AD, there is increased expression of NTPDase2 and decreased expression of NTPDase3 at different stages of the disease, suggesting an imbalance in purinergic metabolism associated with synaptic impairment and cognitive loss [31]. These enzymes have also been implicated in the modulation of synaptic plasticity and memory, functions that are severely impaired in AD [32].

Although there are few studies on ENTPDases in HD, there is evidence that alterations in ADO metabolism may contribute to pathogenesis. CD73 shows reduced expression, with an imbalance in extracellular levels of hypoxanthine and inosine, possibly associated with hyperactivity of adenosine deaminase (ADA) and a reduction in purine nucleoside phosphorylase [33]. The action of ENTPDases in conjunction with CD73, which provides AMP for the production of ADO, may directly influence the inflammatory and neurodegenerative mechanisms, which are still poorly understood in this disease. Given this scenario, modulating the activity of ENTPDases represents a promising strategy to control inflammation in neurodegenerative and autoimmune diseases.

Several reviews have elaborated on the role of these enzymes in different pathological conditions, including cancer, inflammatory bowel disease, cardiovascular, metabolic, autoimmune, and neurodegenerative diseases, highlighting ENTPDases as therapeutic targets [15, 34, 35]. However, there are still gaps in our understanding of the mechanisms that regulate the expression, localization, and activity of ENTPDases in different tissues and disease stages. Identifying selective modulators—inhibitors or activators—is essential to developing more effective and specific therapies. Further functional and pharmacological studies on ENTPDases are critical to translating their biological potential into concrete clinical applications.

Inhibitors of human ENTPDases and therapeutic applications

Inhibition of NTPDase1 has been shown in experimental models to prevent thrombus formation by maintaining elevated ADP levels, acting as a protective mechanism against vascular complications [4]. Moreover, blocking the conversion of ATP to ADO restores the pro-inflammatory environment and increases the efficacy of immunotherapies. Compounds such as POM-1 have been evaluated in tumor models and shown to reduce tumor progression, strengthening their therapeutic potential [91]. However, low selectivity for specific isoforms and effects on other ectoenzymes remain significant barriers to clinical development, highlighting the need for optimizations that maximize efficacy and minimize toxicities. Furthermore, in addition to preventing ATP degradation, POM-1 blocks central synaptic transmission, an effect not directly related to ENTPDase inhibition [77, 79]. Other limitations include low bioavailability and a lack of advanced clinical trials, with most studies limited to in vitro and animal models.

Other compounds, such as Schiff bases [69] and thiadiazole pyrimidones (8b and 8f), have shown good inhibitory activity against NTPDase1, with potential in the treatment of cancer and cardiovascular diseases such as thrombosis [66]. These compounds showed high specificity for this isoform, with Ki values in the nanomolar range (Table 3). Quinoline derivatives and thienotetrahydropyrimidines also showed good selectivity, with the latter exhibiting dual inhibitory activity, including CD73 inhibition [64, 80]. However, further studies using cellular and animal models must confirm their therapeutic applicability in NTPDase1-related diseases.

In type 2 diabetes, NTPDase3 regulates insulin secretion in pancreatic β-cells. Preclinical studies using specific antibodies against this isoform have shown increased insulin secretion and improved glycaemia control in experimental models [90]. Recently, Afzal et al. [70] demonstrated the potential of carboxyamides to inhibit NTPDase3 with IC₅₀ in the submicromolar range. In vitro tests on pancreatic cells showed that NTPDase3 inhibition modulates insulin secretion [70].

NTPDase2 and NTPDase3 have also been explored in neurodegenerative and neuroinflammatory diseases, aiming to modulate inflammation without disrupting ADO homeostasis [4]. Anthraquinone derivatives have shown moderate selectivity for these isoforms, with IC₅₀ in the nanomolar range (see Table 3). These inhibitors may help regulate inflammatory responses in the CNS by promoting extracellular ADP accumulation, indirectly activating P2Y₁, P2Y₁₂, and P2Y₁₃ receptors in immune and inflammatory modulation. However, additional studies in animal models and cellular systems are needed to validate their therapeutic effects [82].

Monoclonal antibodies are also emerging as promising tools due to their high specificity and ability to target defined molecular epitopes. Anti-NTPDase3 antibodies have shown potential in metabolic diseases such as type 2 diabetes by modulating insulin secretion in pancreatic islets [90]. Antibodies against NTPDase2 may contribute to anti-tumor immunotherapy by inhibiting enzymatic activity and decreasing immunosuppressive ADO levels within the tumor microenvironment. They may also regulate intestinal motility, suggesting therapeutic potential for gastrointestinal disorders [88].

However, essential gaps must be addressed to expand the possible clinical use of these antibodies. Studies such as those by Munkonda et al. [90] (2009) and Pelletier et al. [88] (2017) have demonstrated the exceptional specificity of antibodies against NTPDases. However, in vivo validations are lacking to assess therapeutic effects in more complex biological systems. In addition, challenges such as low tissue penetration in specific organs and high production costs can limit their clinical use. Advances in molecular engineering and improvements in inhibitory functionality are critical to turning these antibodies into effective therapeutic interventions.

Inhibitors of parasite ENTPDases and their therapeutic potential

The development of specific inhibitors targeting parasitic ENTPDases represents a promising strategy for controlling infections. These enzymes regulate extracellular nucleotide levels, modulating purinergic signaling that influences critical immunological and inflammatory processes. Through this regulation, parasites create an immunosuppressive microenvironment that supports their survival and replication within the host. Inhibiting these enzymes can turn off the immune evasion mechanisms of parasites, restore the host’s immune response, and enhance the efficacy of conventional therapies. Additionally, ENTPDase inhibitors can act as adjuvants, reducing the doses and side effects of traditional medications, which is particularly advantageous in chronic infections where the immune system is compromised [36, 37].

Classic inhibitors such as DIDS, sodium azide, suramin, and ARL67156 have effectively inhibited ENTPDases and are widely used in characterization studies [46]. However, these compounds exhibit low specificity and non-selectivity on mammalian and parasitic ENTPDases, limiting their therapeutic application. In vitro studies have shown that suramin significantly reduces T. cruzi infectivity [47]; however, in murine models, it induced adverse effects, including exacerbated inflammation, worsened myocarditis, and increased mortality. These findings suggest that suramin’s broad interference with host purinergic signaling may lead to undesirable systemic outcomes [92], underscoring the need for more selective inhibitors.

Derivatives of ent-isoquercetin, particularly compound 16, demonstrated 94% inhibition of recombinant TcNTPDase1 with a Ki of 8.39 µM. Molecular docking studies confirmed that the compound binds at the catalytic site and interacts with critical residues such as Gly369, Gly370, and Ser371, reinforcing its potential for treating Chagas disease [93]. Furthermore, this inhibitor may support structural studies to deepen the understanding of enzyme conformation and catalytic mechanisms. Given the structural similarity between Leishmania ENTPDases and TcNTPDase1 [94], these compounds may also be applicable against Leishmania spp. Nonetheless, further studies in cellular and animal models and specificity assessments against human ENTPDases are needed to ensure therapeutic efficacy and safety.

In S. mansoni, investigations involving cardamonin and alkylaminoalkanesulfonic acid derivatives (AAATs) have highlighted the importance of NTPDase inhibition. Cardamonin exhibited an IC₅₀ of 23.54 µM against SmATPDase1 and significantly reduced parasite burden (46.8%) and egg production (54.5%) in murine models. AAATs demonstrated strong inhibitory activity, with efficacy influenced by their lipophilicity and interactions with cysteine residues. Despite these encouraging results, many studies have used indirect enzymatic models, such as potato apyrase, limiting conclusions regarding SmNTPDase isoform specificity. Furthermore, there is a lack of pharmacokinetic and toxicity data in experimental models—critical factors in drug development [95–97].

In the case of T. gondii, resveratrol (RSV) has emerged as a modulator of NTPDases, reducing ATP and ADP hydrolysis by approximately 20% and 10%, respectively, while partially restoring ADA activity. These findings suggest RSV can regulate the inflammatory microenvironment and oxidative stress without significantly decreasing brain cyst load. Thus, RSV’s effects appear to relate more to neuroprotection and immunomodulation than direct parasitic clearance, reaffirming the therapeutic potential of NTPDases in neurological infections caused by T. gondii [98].

For T. vaginalis, the alkaloids lycorine and aldimine significantly inhibited extracellular nucleotide hydrolysis. Lycorine exhibited an IC₅₀ of 32 µM, while aldimine inhibited up to 77% of ATP hydrolysis. However, lycorine’s cytotoxicity in human cells indicates the need for chemical modifications to improve its selectivity. Moreover, targeting the purinergic system, including ADA inhibition, has reduced reactive oxygen species (ROS) and inflammatory cytokines in neutrophils exposed to the parasite, indicating that combinatorial strategies may enhance infection control [99–101].

Although compounds like cardamonin and flavonoid derivatives have shown promise, the repertoire of specific inhibitors for parasitic ENTPDases remains extremely limited, and the field lacks sufficient chemical diversity. Notably, there are no dedicated studies on selective ENTPDase inhibitors for Leishmania spp. and P. falciparum, constituting a significant research gap. Since these enzymes play central roles in the purinergic metabolism of these parasites, they represent attractive targets for therapeutic innovation.

Advances in hNTPDases research offer a valuable framework for developing inhibitors targeting parasite isoforms. Mechanistic, structure-activity relationship (SAR), and molecular dynamics data from human enzymes can be adapted to identify functional and structural divergences in parasitic isoforms. Computational tools such as molecular docking and homology modeling—already widely applied in hNTPDase research—will be instrumental in this process [102]. Such an integrated approach may facilitate the development of novel therapeutic strategies against pathogenic parasites.

Challenges and future perspectives in the development of ENTPDase inhibitors

Despite recent progress, the development of ENTPDase inhibitors—especially for infectious and inflammatory disease contexts—still faces significant challenges that hinder clinical translation. A key issue is the low selectivity of existing compounds, directly resulting from the high structural conservation among human and parasitic ENTPDase isoforms. This homology compromises molecular specificity and increases the likelihood of adverse effects, particularly considering the widespread distribution of these enzymes across vital tissues such as the brain, liver, heart, intestine, and immune system [67].

Cross-reactivity with other elements of the purinergic signaling cascade, including P2 receptors and E-5’-NTs, presents another significant limitation. This is especially concerning in physiological contexts where ENTPDases play protective roles, as indiscriminate inhibition may lead to harmful systemic consequences. As mentioned earlier, compounds like suramin exemplify this issue, showing severe adverse effects in animal models due to widespread interference with host purinergic signaling [92]. Furthermore, many candidate compounds remain restricted to in vitro evaluations and lack in vivo validation for efficacy, toxicity, and pharmacokinetics [36, 37]. The lack of high-resolution structural data, such as crystal structures and accurate 3D models, further limits rational drug design strategies, making it harder to progress toward developing high-affinity and highly selective inhibitors [46, 67].

Targeting the CNS poses additional challenges. Most ENTPDase inhibitors have poor blood-brain barrier (BBB) permeability, limiting their use in neuroinflammatory disorders like AD, MS, and parasitic encephalitis. While monoclonal antibodies offer high specificity, they face practical constraints including high production costs, poor tissue penetration, and limited inhibitory function [35].

For parasitic ENTPDases, the scenario is further complicated by external factors such as limited funding, low commercial interest, and pharmaceutical disinvestment in neglected diseases like leishmaniasis and Chagas disease [36].

Despite these obstacles, recent developments provide encouraging perspectives. Allosteric modulators represent a promising alternative to overcome the selectivity limitations of competitive inhibitors. These compounds fine-tune enzymatic activity instead of entirely blocking it, potentially reducing off-target effects and improving tissue-specific activity [46].

Computational approaches have been central to these advances. Robust biochemical assays, including malachite green-based methods, supply reliable potency data [80]. At the same time, molecular docking and dynamics simulations enable precise analysis of ligand-target interactions, guiding the rational design of next-generation inhibitors [102]. X-ray crystallography has also been vital in characterizing active sites, facilitating the creation of stable and selective compounds [103]. More recently, artificial intelligence and machine learning have accelerated high-throughput screening, enhancing affinity, bioavailability, and toxicity predictions while addressing species-related extrapolation issues [87]. Additionally, novel delivery systems—such as nanostructured carriers, transport vectors, and intranasal administration—are being explored to improve drug penetration into hard-to-reach tissues, including the CNS [104].