Encyclopedia of Signaling Molecules

2018 Edition
| Editors: Sangdun Choi

E-NTPDase Family

Reference work entry
DOI: https://doi.org/10.1007/978-3-319-67199-4_197


Historical Background

Nucleoside triphosphate diphosphohydrolases (NTPDases), originally ATP diphosphohydrolases (ATPDases) with the common name apyrase, are by definition enzymes which split the γ- and β-phosphate residues of triphospho- and diphosphonucleosides such as ATP and ADP, respectively. Before the cloning of the gene, the common name “apyrase” was generally used for enzymes that exhibit this activity in plants and invertebrates while the terms ecto-ATPase and ATP diphosphohydrolase (ATPDase) were rather used in vertebrates (Beaudoin et al. 1996). Shortly after cloning the first gene encoding such an enzyme, the nomenclature was unified in mammals to NTPDase, which better reflects the ability of these enzymes to convert not only ATP and ADP but also other triphospho- and diphosphonucleosides (Zimmermann et al. 1999).

A first apyrase was discovered in potato tubers in 1944 and was extensively characterized in the following years, although its function was unknown. Apyrases were later reported in other plant tissues such as in various peas and in cabbage leaves. Beginning in 1979, and mostly during the 1980s, apyrases were found in the salivary glands and saliva of bloodsucking animals such as ticks, mosquitoes, fleas, sand flies, and in the medicinal leech. The function of apyrase in these hematophagous animals was suggested to relate to feeding while preventing coagulation (Beaudoin et al. 1996). Indeed, ADP is a major platelet aggregation agent, and the hydrolysis of this nucleotide by apyrase therefore has an anticoagulating effect.

Ectonucleotidases have long been known in vertebrate tissues, but the true identification of an ATPDase first happened in 1980 by A.R. Beaudoin’s group (Sherbrooke, QC, Canada). They demonstrated the presence of this enzyme in the pig pancreas from which they purified the enzyme to a high degree and studied its kinetics in detail. Cytochemical and biochemical observations showed that this enzyme is associated with the zymogen granule and plasma membrane of acinar cells (Beaudoin et al. 1996). In certain conditions, one could find the enzyme associated with microvesicles secreted from the gland. Until the cloning of ATPDase genes, several papers reported the presence of enzymes with ATPDase activity in virtually all vertebrate tissues. During that period, whether one or two enzymes were responsible for the hydrolysis of ATP and ADP was intensely debated. The definitive identification of these proteins came with the cloning.

Handa and Guidotti in 1996 (Boston, MA, USA) cloned a soluble ATP diphosphohydrolase (apyrase) from potato tubers and noted that similar genes exist in some protozoans, in yeast, and in mammals which were related to a human gene called CD39 cloned 2 years earlier. Following this work, Wang and Guidotti again in 1996 reported that CD39 encodes a protein with ectoapyrase activity (Wang and Guidotti 1996). Independently and at the same time, several peptide sequences of purified ATPDases were obtained which also showed homology to CD39: an ATPDase from human umbilical vessels (S. Christoforidis and coll., Ioannina, Greece), and ATPDases from porcine pancreas and bovine aorta (A.R. Beaudoin’s group). Then the groups of Drs. Beaudoin and Robson (Boston) joined their efforts to clone and characterize the gene responsible for encoding this enzyme with ATPDase activity. They came up with the conclusion that CD39 and ATPDase were identical (Kaczmarek et al. 1996). The enzyme is now known as NTPDase1. Shortly after this discovery, H. Zimmermann’s group (Frankfurt am Main, Germany) and, independently, T.L. Kirley’s group (Cincinnati, OH, USA) cloned another related gene (Kegel et al. 1997; Kirley 1997), now known as NTPDase2. Then, by analyzing expressed sequence tags (ESTs), Chadwick and Frischauf showed three more related genes in humans. We now know that this family comprises eight members in mammals, which are referred to as NTPDase1–8 since 2000 (Zimmermann et al. 1999; Bigonnesse et al. 2004; Robson et al. 2006).

Owing to the cloning of the NTPDase genes, homologous genes that encode enzymes with similar biochemical properties were found throughout the evolutionary tree. Apyrases have been found so far in vertebrates, plants, insects, protozoa, yeasts as well as in bacteria. A more detailed history of apyrase and NTPDases from different species can be consulted in excellent reviews (Plesner 1995; Beaudoin et al. 1996; Robson et al. 2006; Knowles 2011; Zimmermann et al. 2012; Yegutkin 2014). This entry will henceforth focus exclusively on the mammalian enzymes, using the accepted nomenclature as NTPDases. The genes corresponding to human NTPDase1–8 are ENTPD1–8.

General Characteristics

Members of the E-NTPDase family, also called the CD39 family, and sometimes the GDA1 family, are major enzymes that dephosphorylate nucleotides such as ATP and ADP. The “E-” stands for “ecto,” which may be a little confusing since of the eight members of the E-NTPDase family, four, namely, NTPDase1–3 and 8, are specifically expressed at the cell surface and hydrolyze extracellular nucleotides. These enzymes are therefore called ectonucleotidases. On the other hand, NTPDase4–7 are attached to intracellular organelles, via one or two transmembrane domains, with their active site facing the intraorganellar lumen. Therefore, these enzymes, if not ecto, are at least “exo” for exocytoplasmic (or extracytoplasmic), as the catalytic site of NTPDases has never been found to face the cytosol, that is, they always face the extracellular medium or a compartment topologically identical to the extracellular medium. Although heterologous expression showed that NTPDase5 and 6 undergo secretion via proteolytic cleavage, they have a weak activity and a low affinity toward nucleotides compared to the plasma membrane-bound NTPDases (i.e. members 1–3 and 8), which raises the question as to whether these enzymes significantly contribute to the hydrolysis of extracellular nucleotides in vivo.

Other ectonucleotidases also involved in the conversion of nucleotides in the extracellular milieu include nucleotide pyrophosphatases/phosphodiesterases, acid and alkaline phosphatases as well as ecto-5′-nucleotidase/CD73 (Yegutkin 2008; Kukulski et al. 2011). These enzymes differ by their biochemical properties and by their cellular localization. This entry will focus on the four E-NTPDase members that have been clearly demonstrated to contribute to the hydrolysis of extracellular nucleotides and which appear as the dominant ectonucleotidases in physiological conditions (i.e., at pH ∼7.4).

NTPDase1–3 and 8 (EC are type II transmembrane proteins of about 500 amino acid residues. These NTPDases are firmly anchored to the plasma membrane via two transmembrane domains, one near each end of the protein, which delineate a large extracellular loop containing the active site and 10 conserved Cys residues that are obviously important for the tertiary and quaternary structure of these proteins. These four enzymes also encompass about 6–10 N-linked glycosylation sites, depending on the protein and species. Members of this enzyme family share five apyrase conserved regions (ACRs). NTPDase1–3 and 8 readily form homo-oligomeric assemblies. They have indeed been observed as homodimers to tetramers, whereas hetero-oligomeric complexes between NTPDases have not been reported thus far. Homo-oligomeric assemblies appear to be important for the biochemical activity of the enzymes (Robson et al. 2006).

NTPDases catalyze the hydrolysis of the γ- and β-phosphate residues of triphosphonucleosides (e.g., ATP, UTP) and diphosphonucleosides (e.g., ADP, UDP). Optimal NTPDase activity requires low millimolar concentrations of divalent cations (Ca2+ and/or Mg2+). NTPDase1–3 and 8 are all active at physiological pH, and a few of them are active in a wider set of pH values (see below) (Kukulski et al. 2005). The most important difference between these four NTPDases is their ability to hydrolyze diphosphonucleosides, NTPDase1 and NTPDase2 being the most and least efficient forms, respectively. NTPDases are distinct from other ATPases and are insensitive to inhibitors of F-, P-, and V-type ATPases, and of alkaline phosphatases.

The presumptive major role of plasma membrane-bound NTPDases, and more specifically of NTPDase1–3 and 8, is the regulation of the concentration of extracellular nucleotides that are released by cells in a regulated manner as well as during cell death and cell lysis (Yegutkin 2008; Kukulski et al. 2011). Through such an action, NTPDases modulate the biological effects triggered by these nucleotides via the activation of several nucleotide receptors, namely P2Y1,2,4,6,11–14, P2X1–7, and three other receptors that are also activated by nucleotides, namely cysteinyl leukotriene receptor-1 and 2 (CysLT1 and CysLT2), and GRP17. Indeed, micromolar concentrations of UDP have been shown to activate the last three receptors.

Like other ectonucleotidases, NTPDases are involved in multiple aspects of nucleotide (P2) and adenosine (P1) receptor signaling which include the termination of P2 receptor activation, protection of susceptible P2 receptors such as P2Y1 and P2X1 from desensitization, and promotion of some P2 receptors activation as well as of P1 receptors. More specifically, by hydrolyzing ATP, NTPDases terminate P2X1–7 activation as well as P2Y2 and P2Y11 activation. By hydrolyzing UTP, NTPDases end P2Y2 and P2Y4 receptor activation. These enzymes, and more specifically NTPDase2, can also generate the agonist of some receptors from the hydrolysis of ATP and UTP. Indeed, ADP is the ligand of P2Y1, P2Y12, and P2Y13, while UDP is the agonist of P2Y6. In addition, the final product of NTPDase activity on ATP and ADP, AMP, is converted to adenosine by ecto-5′-nucleotidase and/or alkaline phosphatases. The latter process is of high biological significance, since adenosine is the ligand of the four P1 receptors (A1, A2A, A2B, A3). By contributing to the generation of adenosine, NTPDases not only participate in the regulation of P1 receptor activation but also contribute to replenish ATP stores. Indeed, once released from the cells, nucleotides cannot be reaccumulated as long as they bear a negative charge. In other words, all phosphate groups must be removed before the nucleoside core of the molecule can reintegrate into the cell. Then, nucleosides such as adenosine can reenter the cell by active or passive transport across the membrane via specific permeases, namely, concentrative nucleoside transporters (CNTs), and equilibrative nucleoside transporters (ENTs), respectively. These nucleotide salvage pathways are especially important for tissues and cells such as brain, bone marrow, erythrocytes, and intestinal mucosa, which cannot synthesize these valuable molecules de novo.

As almost every cell expresses more than a single type of P1 and P2 receptors, nucleotides and adenosine have functions in all organs. Therefore, the regulation of the identity and concentration of the receptor ligands by enzymes which metabolize these agonists is critical and should be appropriate for each tissue or cell (Yegutkin 2008; Deaglio and Robson 2011; Kukulski et al. 2011).

Specific Properties

NTPDases differ from each other by their subtle biochemical properties and their cellular localization. Although PCR experiments, which detect low levels of mRNAs, often display the expression of multiple NTPDases by the same tissue/cell, immunohistochemistry with specific antibodies (http://ectonucleotidases-ab.com) most often show that NTPDase1–3 and 8 are expressed by different cells, with the expression of two of these NTPDases by the same cell being the exception.

A great deal of effort is currently dedicated to the development of specific NTPDase inhibitors. Although some progress has been made, much work remains to be done. The inhibitors that have been developed and utilized thus far have been summarized in reviews (Kukulski et al. 2011; al-Rashida and Iqbal 2014). It is also noteworthy that monoclonal antibodies have been developed and two of them specifically inhibit human NTPDase3 and human NTPDase2, respectively (http://ectonucleotidases-ab.com).

Key information toward the design of novel inhibitors is expected from the current work of Dr. Sträter’s team (Universität Leipzig, Germany). They crystallized the extracellular and soluble portion of NTPDase2 (Zebisch and Strater 2008; Zimmermann et al. 2012) and many others more recently. This already confirmed the importance and localization of the 10 conserved Cys residues in the extracellular domain of plasma membrane-bound NTPDases (members 1–3 and 8) (Ivanenkov et al. 2005) and the contribution of all five ACRs to the catalytic activity of the enzyme. These data represent only a prelude to the actual crystallization of NTPDases as this original work was performed on an extracellular domain produced by E. coli which does not glycosylate proteins, and because the transmembrane domains also appear to be important for the activity of NTPDase1–3 and −8 (Grinthal and Guidotti 2006; Knowles 2011).

This section presents the distinct biochemical properties of the plasma membrane-bound form of the E-NTPDase family members, their general cellular localization, and their demonstrated and/or presumed functions. NTPDase functions involve the regulation of the concentration of P2 receptor ligands. In agreement, the functions so far reported for these enzymes have been directly, or at least indirectly, linked to the regulation of nucleotide and nucleoside receptor activation. Therefore, the functions described below will be in line with the assumption that they implicate P2 receptors. This by no means excludes the possibility that NTPDases might have functions unrelated to ATP receptors, but due to the lack of clear evidence at the moment such hypothetical roles in this entry will not be dealt.

Therefore, the identity and localization of the NTPDase(s), together with the identity of the neighboring P2 and P1 receptors, the signaling pathway coupled with the activation of this (these) receptor(s) in a given cell type, as well as the identity and concentration of the nucleotide released will dictate a precise response and function in a given tissue. The current state of knowledge on the various cellular sources of nucleotides has been very well covered by Yegutkin (2008). The signaling and various functions played by P1 and P2 receptors have also been extensively covered in a number of reviews.



NTPDase1 (aka CD39, ATPDase, ecto-apyrase, ecto-ATPase, or ecto-ADPase) is the best characterized NTPDase thus far, and its involvement in the regulation of several distinct biological processes has been clearly demonstrated (see below). The cDNA encoding NTPDase1 was originally cloned from human tissue. In 1994, Maliszewski et al. cloned, what was known at the time as CD39, a lymphocyte activation antigen with unknown function. Partial protein sequences obtained at about the same time by different groups from apyrases/ATPDases purified from different tissues finally allowed two groups to independently and simultaneously clone and express CD39 cDNA as a functional ATPDase (Kaczmarek et al. 1996; Wang and Guidotti 1996). This protein was renamed NTPDase1 in the following years (Zimmermann et al. 1999). This was the first identification of a gene encoding a protein with true ecto-ATPase activity. The gene encoding human NTPDase1 is now named ENTPD1 (or CD39) and maps to chromosome 10q24 (GeneBank access. no. U87967) (Robson et al. 2006).

Two variants, most probably originating from alternative splicing, have been observed in human placenta. One of them bearing a small modification at the N-terminus was shown to be active. The second isoform is probably inactive. Although NTPDase1 expression often appears to be constitutive, it was shown to be modulated by inflammatory cytokines, oxidative stress, hypoxia, cAMP response elements, glucocorticoids as well as by nucleotides (via P2Y1 receptor activation).

NTPDase1 hydrolyzes all nucleotides with a similar efficacy. These include the P2 receptor ligands, namely ATP, ADP, UTP, and UDP. NTPDase1 converts ATP to ADP, and then to AMP. In contrast to the other NTPDases, there is only a minimal transient release of ADP, as the processing step to AMP is favored (Beaudoin et al. 1996). Importantly, this property of NTPDase1 makes it the most efficient to convert ATP to adenosine as it not only rapidly generates AMP, the substrate for ecto-5′-nucleotidase, but also allows the concentration of two inhibitors of the latter enzyme, namely ATP and ADP, to rapidly decrease. It is noteworthy that, in contrast to adenine nucleotides, NTPDase1 hydrolyzes UTP to UMP with a transient accumulation of UDP, which may favor the transient activation of the UDP receptor P2Y6. Finally, NTPDase1 has the narrowest range of pH preference for activity, being active between pH 7 and 10. (Kukulski et al. 2005).

The N-terminal intracytoplasmic domain of NTPDase1 is subject to palmitoylation (Kittel et al. 1999; Koziak et al. 2000), which targets a subpopulation of enzyme molecules to cholesterol-rich lipid rafts/caveolae. The presence of NTPDase1 in these structures might be of physiological importance due to their association with G protein-coupled receptor signaling and to their colocalization with some P2 receptors and ecto-5′-nucleotidase. Although its association with caveolae does not appear to be essential for enzyme activity, cholesterol depletion results in a strong inhibition of NTPDase1. Oxidative stress has also been reported to inhibit NTPDase1 activity in vivo, which affects its normal antiplatelet function such as that observed in transplantation models (Deaglio and Robson 2011).

The N-terminus of human NTPDase1 has been shown to interact with truncated Ran-binding protein M (RanBPM = RanBP9). RanBPM contains conserved SPRY (repeats in splA and RyR) domains which appear to be crucial for its interaction with NTPDase1. RanBPM is known to interact with Sos and to regulate ERK/Ras signaling. Such functions might therefore be altered or regulated by NTPDase1, which may therefore represent a novel, albeit hypothetical function of the enzyme (Robson et al. 2006).

Localization and Function

NTPDase1 is the predominant ectonucleotidase responsible for lowering nucleotide content in the blood. It is expressed in endothelial cells, smooth muscle cells as well as in leukocytes, including neutrophils, monocytes, and lymphocytes (e.g., B lymphocytes, Tregs, memory lymphocytes, natural killer T cells, and natural killer cells) (Kukulski et al. 2011). NTPDase1 has also been immunodetected in macrophages and resident macrophages such as microglia and Kupffer cells, in Langerhans and dendritic cells (Mizumoto et al. 2002). It is also expressed in nonvascular smooth muscle cells as well as in some epithelial cells, for example, in the zymogen granule membrane of pancreatic acini where it was originally demonstrated in mammals (Beaudoin et al. 1996). This protein is also secreted in an active form into the pancreatic juice together with a particulate fraction (shed from the membrane) (Beaudoin et al. 1996; Robson et al. 2006).

As a result of its nucleotide hydrolysis activity, NTPDase1 has been associated with several functions through its regulation of P2 receptor activation and desensitization. For example, NTPDase1 protects P2X1 and P2Y1 receptors from desensitization (Enjyoji et al. 1999; Kauffenstein et al. 2014). Entpd1−/− mice, which are deficient in NTPDase1 expression, display abnormal platelet activation due to P2Y1 receptor desensitization in platelets (Enjyoji et al. 1999). However, it must be noted that this phenotype was observed in NTPDase1-deficient animals which subsisted despite this anomaly. Such protection against P2X1 and P2Y1 receptor desensitization contrasts with another important function of NTPDase1, namely the prevention of platelet aggregation by removal of ADP, the agonist for P2Y1 and P2Y12 receptors in platelets (Robson et al. 2006; Deaglio and Robson 2011). NTPDase1 probably also contributes to this effect by hydrolyzing ATP, the ligand of the P2X1 receptor, which is also expressed in platelets and is also important for their activation in vivo. Therefore, the regulation of the level and duration of NTPDase1 activity should play a crucial role in the overall regulation of P2 receptor activity (activation vs desensitization).

The function of the endothelial NTPDase1 is not limited to platelet aggregation. It also influences inflammatory processes (see below) as well as vascular tone. Although Entpd1−/− mice do not show a global defect in vasomotion, in vitro experiments with arteries from these mice showed that the effects of both exogenous and endogenous nucleotides on vasodilation were dramatically potentiated when NTPDase1 activity is lacking. This regulation was shown to involve the regulation of endothelial P2Y1, P2Y2, and P2Y6 receptors by NTPDase1. Interestingly, the absence of NTPDase1 in vascular smooth muscle cells made the mice more susceptible to vasoconstriction via P2Y6 receptors expressed on smooth muscle cells (Kukulski et al. 2011). These data clearly show the importance of the compartmentalization of nucleotide release as well as the fine regulation of their levels by NTPDase1, which dictates their final output on the biological system, namely the vascular tone in the latter example.

Illustrations of the role(s) of NTPDase1 in immune responses have grown exponentially in the last decade since the enzyme has been shown to be the major ectonucleotidase in leukocytes, including lymphocytes such as Tregs and natural killer cells, as well as macrophages, dendritic cells, and Langerhans cells (Kukulski et al. 2011; Takenaka et al. 2016). In at least a subset of these cell types, NTPDase1 expression protects from ATP-induced cell death and affects angiogenesis, leukocyte trafficking as well as the expression and release of several cytokines (Kukulski et al. 2011). NTPDase1-null mutant mice have revealed various functions of the enzyme in immune responses. Entpd1−/− animals show amplified inflammatory responses to irritant chemicals due to the lack of NTPDase1 suppressive properties and present defective stimulation of hapten-reactive T cells (Mizumoto et al. 2002). Also in mouse, NTPDase1 is a Treg cell surface marker which appears to affect cellular immunoregulation (Takenaka et al. 2016). Modulated nucleotide signaling was also shown to impact NKT-mediated mechanisms that result in liver immune injury. The few observations presented above on NTPDase1-deficient animals clearly suggest that the role of the enzyme in immune reactions is complex. In addition, NTPDase1 has been shown to affect cancer progression, an effect which could be attributed, at least in some models, to NTPDase1 activity in Tregs. Note that NTPDase1 activity detected in some cancers may also be the result of NTPDase1 expression by cancer cells. Altogether, the inhibition of NTPDase1 enzymatic activity has been proposed as an adjunct therapy for primary and secondary malignancies. However, the underlying mechanism is likely very complex, as spontaneous tumors develop in old Entpd1−/− mice, in contrast to the data obtained above with models of more malignant cancers.

Not only NTPDase1 deficiency is linked to distinct anomalies, but the full range of expression and activity for the enzyme also appears to be important for the normal regulation of biological processes, as suggested above. An interesting example of this is an ENTPD1 polymorphism that was reported in human which is associated with reduced NTPDase1 mRNA expression and which correlates with increased susceptibility to Crohn’s disease. In agreement with the latter, severe colitis has been observed in NTPDase1-deficient mice in an experimental model of inflammatory bowel disease (Friedman et al. 2009). The functions of NTPDase1 have been reviewed in recent papers (Deaglio and Robson 2011; Kukulski et al. 2011; Takenaka et al. 2016).



NTPDase2 (= CD39L1 or ecto-ATPase) was originally cloned from chicken and rat (Kegel et al. 1997; Kirley 1997). The ENTPD2 human gene is located on chromosome 9q34 (AF144748). Two inactive splice variants have been observed so far in human and one such form in rat which has the main characteristics of the original NTPDase2 transcript species, with some differences. Fluctuations in NTPDase2 expression have often been noted but these may be related, as for NTPDase1, to the cell type involved. For example, in mouse hepatoma cells, transcription of NTPDase2 can be induced by 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), an environmental contaminant. Interestingly, the NTPDase2 core promoter reveals constitutive transcriptional activity that is independent of TCDD. Moreover, TCDD failed to induce NTPDase2 in a variety of other cell lines derived from various species. In rat Sertoli cells, NTPDase2 is upregulated by follicle-stimulating hormone and cAMP. In the liver, NTPDase2 is selectively downregulated in biliary cirrhosis. In portal fibroblasts, the expression of NTPDase2 is downregulated by IL-6 at the transcriptional level (Robson et al. 2006; Kukulski et al. 2011).

NTPDase2 is a preferential nucleoside triphosphatase. It converts very efficiently ATP and UTP to the respective diphosphate derivative. Although NTPDase2 can further dephosphorylate these diphosphates to their monophosphate derivatives, the latter’s activity is relatively inefficient. Therefore, NTPDase2 facilitates the termination of ATP- and UTP-specific receptor activation in favor of ADP- and UDP-specific receptor activation, respectively. This enzyme is mostly active in the physiological range (as for other NTPDases) as well as in more acidic conditions, but is inactive at pH > 9 (Kukulski et al. 2005).

Localization and Function

The major distribution of NTPDase2 is on the adventitial surface of blood vessels and different types of glial cells in both central and peripheral nervous systems (e.g. astroglia, Schwann cells). NTPDase2 has also been detected in the Bowman’s capsule and in some pericytes as in the heart. Interestingly, NTPDase2 is expressed by “glial-like” type I cells of taste buds, and antibodies to this protein have already become a most efficient way to identify these cells (http://ectonucleotidases-ab.com) (Kukulski et al. 2011).

NTPDase2 expression on the outer surface of blood vessels has been suggested to preserve vascular integrity by favoring the activation of ADP receptors expressed in platelets, namely P2Y1 and P2Y12 (Sévigny et al. 2002). Indeed, upon vascular tissue disruption, ATP is released from damaged cells, and platelets, which are normally found inside vessels, then come in contact with the released nucleotides as well as with NTPDase2, which does not normally face the luminal side where platelets are found under physiological conditions (Deaglio and Robson 2011).

A most fascinating observation was made by Massé et al. who showed the short but key participation of NTPDase2 in eye development in Xenopus lævis (Massé et al. 2007). Hydrolysis of ATP to ADP resulting from the overexpression of NTPDase2 caused the development of ectopic eye-like structures and accordingly increased the expression of eye field transcription factors. Neither NTPDase1 nor NTPDase3 could replace NTPDase2 for normal eye formation. Nevertheless, this does not occur in mammals as the group of Dr. Zimmermann showed that deficient mice in NTPDase2 had normal eye formation (Gampe et al. 2015).

Zimmermann’s group reported the expression of NTPDase2 in the subventricular zone and rostral migratory stream of adult rat brain, and suggested that together with an alkaline phosphatase, the enzyme played a role in neural development and differentiation. In the liver, NTPDase2 has also been associated with the regulation of cell proliferation, albeit at a different level. Thus, bile duct cell proliferation was prevented by NTPDase2 expressed in portal fibroblasts in coculture and, conversely, increased when NTPDase2 expression was downregulated. As mentioned above, NTPDase2 is selectively downregulated in biliary cirrhosis and, accordingly, IL-6 downregulates NTPDase2 transcription. Interestingly, the latter phenomenon might represent a mechanism accounting for the aberrant proliferation of bile duct cells observed in biliary cirrhosis, where IL-6 is indeed markedly upregulated. Additionally, the reconstitution of NTPDase2 expression in a glioma cell line dramatically increased tumor growth in vivo (Robson et al. 2006; Kukulski et al. 2011).



NTPDase3, previously named in different studies as CD39L3 and HB6, was originally cloned from human brain (Smith and Kirley 1998). The ENTPD3 gene is located on chromosome 3p21.3 (AF034840). An inactive splice variant has been identified in a human lung cDNA library. When coexpressed by heterologous transfection with the normal form in the same cells, the inactive splice variant could reduce the biochemical activity of the original form (Robson et al. 2006; Knowles 2011).

NTPDase3 converts both ATP and UTP to the monophosphate derivative with a transient accumulation of the diphospho derivative, leading to the transient activation of ADP- and UDP-specific receptors. NTPDase3 is the E-NTPDase family member which exhibits the broadest spectrum of pH for activity (from 5 to 11). (Kukulski et al. 2005).

Although much less information is available on NTPDase3 than on the two previous isoforms, Kirley’s group (Cincinnati, OH) generated a large number of enzyme mutant forms in order to characterize the biochemistry of the enzyme. These studies have also been helpful to understand the general structure of the plasma membrane-bound NTPDases, which share a common general tridimensional conformation (Knowles 2011).

Localization and Function

Kirley’s group reported the expression of NTPDase3 in brain neurons, in which they suggested that the enzyme might be involved in the modulation of feeding and sleep-wake behaviors. NTPDase3 was also reported in peripheral neurons such as along rodents’ bowels. In addition, it was reported in rodents in certain epithelial cells of the digestive, reproductive, renal, and respiratory systems. In the kidney, NTPDase3 is expressed in thick ascending limb, distal tubules, and in cortical and outer medullary collecting ducts. In the pancreas, NTPDase3 is expressed in all Langerhans islet cells. The pharmacological inhibition of NTPDase3 was shown to increase insulin secretion in low glycemia in the rat β-INS-1 (832/13) insulinoma cell line. NTPDase3 has also been immunolocalized in some enteroendocrine cells of the gastric antrum (Kukulski et al. 2011).



NTPDase8, (= liver canalicular ecto-ATPase or hepatic (h) ATPDase), was originally cloned from mouse liver (Bigonnesse et al. 2004). The cDNA clone obtained had a homology with a gene previously cloned from chicken oviduct (Knowles 2011). The ENTPD8 gene is located in the proximity of ENTPD2, mapping also to chromosome 9q34 (AY430414) (Robson et al. 2006).

Like NTPDase3, NTPDase8 converts both ATP and UTP to the corresponding monophosphate, with transient accumulation of the intermediary, diphospho derivative. An interesting characteristic of NTPDase8 is its ability to hydrolyze nucleotides both at physiological and acidic pH (Kukulski et al. 2005).

Localization and Function

Mammalian NTPDase8 is expressed in liver canaliculi, and in the apical membrane of some epithelial cells such as in the brush border membranes (presumably on proximal tubules) of the kidney (Kukulski et al. 2011). The presence of NTPDase8 mRNA was also detected in the organ of Corti of the inner ear. Although no specific function has yet been demonstrated for this last discovered E-NTPDase family member, its localization in the apical surface of the above structures suggests that NTPDase8 might be important for nucleotide salvaging. Indeed, the liver is likely the main source of purines for tissues incapable of de novo synthesis as mentioned in the section “General Characteristics.” The presence of NTPDase8 in parallel with ecto-5′-nucleotidase and nucleoside permeases in the canalicular domain of hepatocytes is in support of a role for NTPDase8 in nucleotide salvaging in the liver.

Summary and Perspectives

The E-NTPDase members 1–3 and 8 are major enzymes responsible for the hydrolysis of nucleoside triphosphates and diphosphates at the cell surface, which regulate P2 receptors and, by extension, adenosine receptor activation as well as the recycling of purines for tissues and cells deficient in de novo synthesis. The study of the functions of these enzymes will be facilitated with the availability of inhibitors for these enzymes, thanks to the current efforts made in several drug discovery laboratories. Although some progress has been made in the last few years, more potent and specific inhibitors of each NTPDase form are still very much needed. The recent cloning, production of antibodies as well as generation of gene-deficient animals, recently achieved for each of these enzymes should prove very useful in the near future to help defining the exact functions of these enzymes, and especially for NTPDase2, 3, and 8. Indeed, a great deal of information about NTPDase1 function has come from mice deficient in the protein which was reported in 1999 (Enjyoji et al. 1999). In addition, the current intense research activity on nucleotide release modes, on nucleotide and adenosine receptors, and on nucleoside transporters that are all part of the purinome system should be very fruitful for NTPDase research as well.


  1. al-Rashida M, Iqbal J. Therapeutic potentials of ecto-nucleoside triphosphate diphosphohydrolase, ecto-nucleotide pyrophosphatase/phosphodiesterase, ecto-5′-nucleotidase, and alkaline phosphatase inhibitors. Med Res Rev. 2014;34:703–43. doi: 10.1002/med.21302.PubMedCrossRefGoogle Scholar
  2. Beaudoin AR, Sévigny J, Picher M. ATP-diphosphohydrolases, apyrases, and nucleotide phosphohydrolases: biochemical properties and functions. In: Lee AG, editor. ATPases. Greenwich: JAI Press Inc; 1996. p. 369–401.CrossRefGoogle Scholar
  3. Bigonnesse F, Lévesque SA, Kukulski F, Lecka J, Robson SC, Fernandes MJG, et al. Cloning and characterization of mouse nucleoside triphosphate diphosphohydrolase-8. Biochemistry. 2004;43:5511–9.PubMedCrossRefGoogle Scholar
  4. Deaglio S, Robson SC. Ectonucleotidases as regulators of purinergic signaling in thrombosis, inflammation, and immunity. Adv Pharmacol. 2011;61:301–32.PubMedCrossRefGoogle Scholar
  5. Enjyoji K, Sévigny J, Lin Y, Frenette PS, Christie PD, Esch II JSA, et al. Targeted disruption of CD39/ATP diphosphohydrolase results in disordered hemostasis and thromboregulation. Nat Med. 1999;5:1010–7.PubMedCrossRefGoogle Scholar
  6. Friedman DJ, Kunzli BM, AR YI, Sévigny J, PO B, Enjyoji K, et al. From the Cover: CD39 deletion exacerbates experimental murine colitis and human polymorphisms increase susceptibility to inflammatory bowel disease. Proc Natl Acad Sci U S A. 2009;106:16788–93.PubMedPubMedCentralCrossRefGoogle Scholar
  7. Gampe K, Haverkamp S, Robson SC, Gachet C, Huser L, Acker-Palmer A, et al. NTPDase2 and the P2Y1 receptor are not required for mammalian eye formation. Purinergic Signalling. 2015;11:155–60.PubMedCrossRefGoogle Scholar
  8. Grinthal A, Guidotti G. CD39, NTPDase 1, is attached to the plasma membrane by two transmembrane domains. Why? Purinergic Signalling. 2006;2:391–8.PubMedPubMedCentralCrossRefGoogle Scholar
  9. Ivanenkov VV, Meller J, Kirley TL. Characterization of disulfide bonds in human nucleoside triphosphate diphosphohydrolase 3 (NTPDase3): implications for NTPDase structural modeling. Biochemistry. 2005;44:8998–9012.PubMedCrossRefGoogle Scholar
  10. Kaczmarek E, Koziak K, Sévigny J, Siegel JB, Anrather J, Beaudoin AR, et al. Identification and characterization of CD39 vascular ATP diphosphohydrolase. J Biol Chem. 1996;271:33116–22.PubMedCrossRefGoogle Scholar
  11. Kauffenstein G, Pelletier J, Lavoie EG, Kukulski F, Martín-Satué M, Dufresne SS, et al. Nucleoside triphosphate diphosphohydrolase-1 ectonucleotidase is required for normal vas deferens contraction and male fertility through maintaining P2X1 receptor function. J Biol Chem. 2014;289:28629–39.PubMedPubMedCentralCrossRefGoogle Scholar
  12. Kegel B, Braun N, Heine P, Maliszewski CR, Zimmermann H. An ecto-ATPase and an ecto-ATP diphosphohydrolase are expressed in rat brain. Neuropharmacol. 1997;36:1189–200.CrossRefGoogle Scholar
  13. Kirley TL. Complementary DNA cloning and sequencing of the chicken muscle ecto-ATPase: homology with the lymphoid cell activation antigen CD39. J Biol Chem. 1997;272:1076–81.PubMedCrossRefGoogle Scholar
  14. Kittel A, Kaczmarek E, Sévigny J, Lengyel K, Csizmadia E, Robson SC. CD39 as a caveolar-associated ectonucleotidase. Biochem Biophys Res Commun. 1999;262:596–9.PubMedCrossRefGoogle Scholar
  15. Knowles AF. The GDA1_CD39 superfamily: NTPDases with diverse functions. Purinergic Signalling. 2011;7:21–45.PubMedPubMedCentralCrossRefGoogle Scholar
  16. Koziak K, Kaczmarek E, Kittel A, Sévigny J, Blusztajn JK, Esch II JSA, et al. Palmitoylation targets CD39/endothelial ATP diphosphohydrolase to caveolae. J Biol Chem. 2000;275:2057–62.PubMedCrossRefGoogle Scholar
  17. Kukulski F, Lévesque SA, Lavoie EG, Lecka J, Bigonnesse F, Knowles AF, et al. Comparative hydrolysis of P2 receptor agonists by NTPDases 1, 2, 3 and 8. Purinergic Signalling. 2005;1:193–204.PubMedPubMedCentralCrossRefGoogle Scholar
  18. Kukulski F, Lévesque SA, Sévigny J. Impact of ectoenzymes on P2 and P1 receptor signaling. Adv Pharmacol. 2011;61:263–99.PubMedCrossRefGoogle Scholar
  19. Massé K, Bhamra S, Eason R, Dale N, Jones EA. Purine-mediated signalling triggers eye development. Nature. 2007;449:1058–62.PubMedCrossRefGoogle Scholar
  20. Mizumoto N, Kumamoto T, Robson SC, Sévigny J, Matsue H, Enjyoji K, et al. CD39 is the dominant Langerhans cell-associated ecto-NTPDase: modulatory roles in inflammation and immune responsiveness. Nat Med. 2002;8:358–65.PubMedCrossRefGoogle Scholar
  21. Plesner L. Ecto-ATPases: identities and functions. Int Rev Cytol. 1995;158:141–214.PubMedCrossRefGoogle Scholar
  22. Robson SC, Sévigny J, Zimmermann H. The E-NTPDase family of ectonucleotidases: structure function relationships and pathophysiological significance. Purinergic Signalling. 2006;2:409–30.PubMedPubMedCentralCrossRefGoogle Scholar
  23. Sévigny J, Sundberg C, Braun N, Guckelberger O, Csizmadia E, Qawi I, et al. Differential catalytic properties and vascular topography of murine nucleoside triphosphate diphosphohydrolase 1 (NTPDase1) and NTPDase2 have implications for thromboregulation. Blood. 2002;99:2801–9.PubMedCrossRefGoogle Scholar
  24. Smith TM, Kirley TL. Cloning, sequencing, and expression of a human brain ecto-apyrase related to both the ecto-ATPases and CD39 ecto-apyrases. BBA-Protein Struct M. 1998;1386:65–78.CrossRefGoogle Scholar
  25. Takenaka MC, Robson S, Quintana FJ. Regulation of the T cell response by CD39. Trends Immunol. 2016;37:427–39.PubMedPubMedCentralCrossRefGoogle Scholar
  26. Wang T-F, Guidotti G. CD39 is an ecto-(Ca2+, Mg2+)-apyrase. J Biol Chem. 1996;271:9898–901.PubMedCrossRefGoogle Scholar
  27. Yegutkin GG. Enzymes involved in metabolism of extracellular nucleotides and nucleosides: functional implications and measurement of activities. Crit Rev Biochem Mol Biol. 2014;49:473–97.PubMedCrossRefGoogle Scholar
  28. Yegutkin GG. Nucleotide- and nucleoside-converting ectoenzymes: important modulators of purinergic signalling cascade. Biochim Biophys Acta. 2008;1783:673–94.PubMedCrossRefGoogle Scholar
  29. Zebisch M, Strater N. Structural insight into signal conversion and inactivation by NTPDase2 in purinergic signaling. Proc Natl Acad Sci U S A. 2008;105:6882–7.PubMedPubMedCentralCrossRefGoogle Scholar
  30. Zimmermann H, Beaudoin AR, Bollen M, Goding JW, Guidotti G, Kirley TL, et al., editors. Proposed nomenclature for two novel nucleotide hydrolyzing enzyme families expressed on the cell surface. Ecto-ATPases and Related Ectonucleotidases; 1999 2000; Diepenbeek: Shaker Publishing BV, Maastricht.Google Scholar
  31. Zimmermann H, Zebisch M, Strater N. Cellular function and molecular structure of ecto-nucleotidases. Purinergic Signalling. 2012;8:437–502.PubMedPubMedCentralCrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  1. 1.Department of Microbiology-Infectiology and Immunology, Faculty of MedicineCentre de Recherche du CHU de Québec - Université LavalQuebec CityCanada