Geoffrey Burnstock, our friend and magister: the diadenosine polyphosphate connection

Abstract

Development of science needs the cooperation of many creative brains. Sometimes, ideas on a specific area get suddenly exhausted and then it is the time for a privileged mind to think in a different way and reach the turning point to introduce a new paradigm. This happened to Geoffrey Burnstock, a heterodox thinker and nonconformist scientist that has been the paladin of purinergic signalling since 1972, opening neuroscience to the understanding of organs and tissues functioning and development of a new pharmacology. This review summarizes the contribution of our group to the understanding of the role of the diadenosine polyphosphates, Ap_nA, as signalling molecules, describing their tissue and organ distribution, their transport and storage in secretory vesicles and their release and interaction with purinergic receptors. We also have to acknowledge the friendly and kindly support of Professor Burnstock that showed a great interest in the field from our initial findings and actively stimulated our efforts to establish the extracellular roles and biological significance of these dinucleotides.

Introduction

By 1988, our research area focused on chromaffin cell secretion and their exocytotic control. We were lucky because, in addition to catecholamines, ATP and ADP, we found the presence in chromaffin cells of other natural compounds, previously described in platelets [1], such as adenine dinucleotides, now known as diadenosine polyphosphates, diadenosine tetraphosphate (Ap₄A) and diadenosine pentaphosphate (Ap₅A) being the most prominent [2]. Our group was curious about the functional role of these compounds, and I decided to attend the meeting on “Purine Nucleosides and Nucleotides in Cell Signalling,” which took place in Maryland (USA), on September 1989. There, Professor Alexander Ribeiro introduced me to Professor Geoffrey Burnstock. We had a very fruitful talk and I was convinced that it was worth trying to find an answer concerning the physiological role and molecular biology events of Ap_nA.

Once back to Madrid, the effect of diadenosine polyphosphates on catecholamine secretion from isolated chromaffin cells was studied. These compounds significantly increased basal secretion of catecholamines and had an absolute requirement for extracellular Ca²⁺, as it happens with classic neurotransmitters. Once organized the results, the manuscript was sent to British Journal of Pharmacology. It was accepted without any problem, except the classical English style, and was published in 1990 [3]. But the most surprising was the kind letter of Professor Bunstock, asking us for other manuscripts related with this topic. Used to the difficulties to publish the manuscripts, we were astonished, but Prof. Burnstock was always very generous to us and who we frequently turn to in search of an opinion or advice.

The network of ATP signalling was getting more complex since 1985, after the seminal article of Burnstock and Kennedy, where they demonstrated that almost two subtypes of P2 receptors were necessary to explain the diversity of tissue and cell responses, one P2X and the other P2Y [4]. The availability of new agonists and antagonists in the 1990s allowed for better understanding and classification of P2 receptors, defined with great precision later on at the Abbracchio and Burnstock publication in 1994. Then, the subtypes reached the category of families, P2X being ionotropic and P2Y metabotropic [5].

But, what happened with diadenosine polyphosphates during that time? Many questions are required to be studied; the first and more difficult one was the characterization of their receptors and signalling pathways, not forgetting the molecular and cellular biology to explain their physiological role. At this time 1990, new scientists joined the purinergic group of Madrid, among them a talented young student, Jesus Pintor (Suso), who started his PhD Thesis specifically on the topic of “Diadenosine polyphosphates as new neurotransmitters.”

The financial support was another difficult chapter; fortunately, we got to survive thanks to the National and European Union projects, especially a BIOMED project entitled “Nucleotides a novel class of extracellular signalling substances in the nervous system.” This European project allowed the interaction between relevant groups on diverse areas of purinergic field: Prof. Zimmermann, Prof. Burnstock, Prof. Abbracchio, Prof. Ribeiro, Prof. Heilbrom and Prof. Miras-Portugal. The intense scientific interaction, through periodic reunions, and also the special meetings sponsored by the Areces Foundation were at the origin of a fruitful scientific production and deep personal friendship.

The broad tissue and organ distribution of the diadenosine polyphosphates and their release

The exocytotic release of Ap_nA, induced by secretagogues, from perfused bovine adrenal medulla and isolated chromaffin cells, was the starting point to study their presence in other cellular models [6].

The presence of diadenosine polyphosphates in rat brain synaptic terminals and their Ca²⁺-dependent release evoked by 4-aminopyridine and veratridine offered a new perspective on their role at the central nervous system [7]. The presence of Ap_nA at the cholinergic synaptic vesicles of Torpedo electric organ, carried out together with Prof. Zimmermann, was one additional proof of the widespread presence of these compounds in neurosecretory granules [8].

Moreover, in addition to the more abundant diadenosine polyphosphates, Ap₄A and Ap₅A, other such as diadenosine triphosphate (Ap₃A) and diadenosine hexaphosphate (Ap₆A) were also detected in secretory granules and at the extracellular level after Ca²⁺-dependent exocytosis [9]. The new techniques of brain perfusion in vivo, known as push-pull, gave us the possibility to have a view on the functioning brain in freely moving rats, in this case, concerning the presence of purinergic compounds at the extracellular space. Dinucleotides have a longer half-life than ATP or ADP at the extracellular level, which was important for their detection at the early times of this technique. One of the most relevant results was the release of these dinucleotides from the basal ganglia of conscious rat after amphetamine administration, which was inhibited by dopamine receptor blockade [10, 11]. In a model of mild hyperammonemia, the presence of diadenosine polyphosphates was found in microdialysis samples from rat cerebellum in vivo, Ap₃A being the prominent one. In this context, the expression of two enzymes belonging to the family of ecto-nucleotide pyrophosphatase/phosphosdiesterase, NPP1 and NPP3, responsible for Ap_nA degradation, were also under control of the hyperammonemia situation [12]. These data give an idea about the complexity of understanding the extracellular levels of Ap_nA and their regulation. Later on, the discovery of diadenosine polyphosphates in human tears and aqueous humour of rabbits opened the field to study their effect on the visual system and the pathologies of dry eye and glaucoma [13, 14].

Dinucleotide transport and storage in secretory vesicles

The presence of Ap_nA in secretory vesicles and their release to the extracellular media opened the question about the storage mechanisms. In 1995, a young scientist, Javier Gualix, started working in this difficult area to carry out his PhD thesis, entitled “Transport of nucleotides and dinucleotides to secretory vesicles.” By this time, the vesicular nucleotide transporter (VNUT) was still elusive to be cloned and had to wait for the seminal discovery made in 2008 by the group of Moriyama in Japan [15]. Thus, VNUT had to be previously characterized by means of pharmacological and biochemical techniques.

The chromaffin granules from adrenal medulla were a largely employed model to study vesicular transporters. The availability of fluorescent ATP and diadenosine polyphosphates analogues (1,N⁶-ethenoadenosine derivatives) allowed a fine identification and quantification of nucleotide uptake into chromaffin granules by HPLC chromatography. Both nucleotides and dinucleotides were transported by the same vesicular transporter, which exhibited a complex allosteric cooperativity. The kinetic behaviour was explained on the basis of a mnemonic mechanism that means a slow transition of the transporter protein at the vesicular membrane [16, 17]. Similar kinetic parameters were observed for the nucleotide transport into rat brain synaptic vesicles [18]. In addition, fluorescent nucleotide derivatives allowed us, for the first time, to analyse chromaffin granule functioning by flow cytometry [19].

The properties of VNUT explain the storage and co-release of a large variety of nucleotides and dinucleotides necessary for their physiological action on receptors and justify the existence of a complex network of specific ecto-nucleotidases for their extracellular metabolism. The VNUT pharmacology is now a fertile area in neuropathic pain and for the understanding of nervous system development (for review, see [20]).

Diadenosine polyphosphate receptors: the long way to reach out and merge with P2X and P2Y receptors

The effects of diadenosine polyphosphates, Ap₄A and Ap₅A, on catecholamine secretion and calcium store mobilization from chromaffin cells were early reported after the discovery of these compounds. Besides, with the scarcity of tools available at that time, it was suggested that these responses were essentially mediated through putative P2Y purinoceptors [21, 22]. In addition, binding studies with Ap₄A and its displacement by the P2Y agonist, adenosine-5′-0-(2-thiodiphosphate) (ADPβS), in rat brain and Torpedo synaptosomes agreed with the presence of a P2Y receptor [23–25].

Besides, in rat midbrain synaptosomes, fluorimetric techniques demonstrated that stimulation with diadenosine polyphosphates was coupled to calcium increase, coming from the extracellular space. Thus, their action through ionotropic receptors was another possibility, increasing the complexity of dinucleotide signalling [26]. On the other hand, the modulatory effects of Ap_nA on different types of Ca²⁺ channels, mainly the N-type, in rat central neurons gave relevance to their effects at the central nervous system [27, 28].

A powerful advance in the understanding of P2 receptors, and therefore of the diadenosine polyphosphate receptors, was the development of molecular biology techniques. Two groups were at the head of the purinergic receptor identification, one from Eric Barnard in UK and the other by Gary Buell in Switzerland. Tania Webb, at the Eric Barnard group, got the cloning and functional expression of the first P2Y receptor [29]. Some diadenosine polyphosphates are able to induce responses from members of the P2Y family [30, 31].

The cloning of P2X receptors was carried out from rat vas deferens by the group of Gary Buell and once expressed in Xenopus oocytes; ATP was the selective nucleotide to activate the cation-selective ion channel. The sequence analysis demonstrated that it was the first subunit member of a new family of ligand-gated ion channels, P2X1–P2X7, associated by homomeric or heteromeric trimers [32].

To study the effects of dinucleotides on the recently cloned P2Y and P2X receptors, our group established a close collaboration with the group of Professor Burnstock, which with his great generosity accepted Dr. Pintor as a post-doc and invited scientist and later on Dr. Gualix, which had synthesized a new diadenosine polyphosphate derivatives, the diinosine polyphosphates.

Systematic studies to assess the selectivity and activity of adenine dinucleotides on recombinant P2X and P2Y receptors demonstrated that P2X1, P2X2 and P2X3 were susceptible of activation by Ap₄A and Ap₅A, with different affinities. On the other hand, P2Y₁ also offered a significant response to these natural compounds. These results, although at the beginning of more extended experimental studies, clearly showed that they are also active, once released, at some of the cloned purinergic receptors [33, 34].

A turning point on the understanding of diadenosine polyphosphate receptors, versus the P2X receptors already cloned, was possible by the new compounds synthetized by Dr. Gualix, who by enzymatic methods changed adenosines into inosines, obtaining the series of diinosine polyphosphates, Ip_nI. These compounds were able to blockade calcium responses to Ap_nA in rat brain synaptic terminals [35]. An additional evidence of the antagonistic effects of Ip_nI on purinergic receptors was the discovery that diinosine pentaphosphate (Ip₅I) antagonized contractions, mediated via P2X receptors, in the guinea-pig isolated vas deferens [36]. Further experiments with recombinant rat P2X receptors demonstrated that Ip₅I was a powerful antagonist of the P2X1 and P2X3 receptors. It is to notice that, in the case of P2X1, Ip₅I was able to inhibit ATP responses at the pM concentration [37]. No inhibitory effects were observed for P2X7 receptors expressed in diverse neural cells.

Although the synaptic terminals in central nervous system are a difficult model to study, due to their diversity, new technical approaches provide a way to elucidate the presence of functional receptors for ATP and Ap_nA [38–40]. Studies on intracellular calcium increases in single synaptic terminals by fluorometry and video microscopy showed responses to Ap₄A and Ap₅A in a large variety of terminals. The calcium signalling was coupled to the release of classical neurotransmitters, such as acetylcholine, noradrenaline, GABA or glutamate [41–44]. After the functional studies, the identification of responding terminals was carried out by immunohistochemistry. The presynaptic vesicular marker, synaptophysin, was combined with a second marker, usually a specific vesicular transporter. As an example, glutamatergic terminals from rat midbrain were identified by the presence of vesicular glutamate transporters, either VGLUT1 or VGLUT2, these markers being present in 31% and 16% of the total synaptosomal population, respectively. Not all the glutamatergic terminals responded to ATP or Ap₅A, responses being present in about 40% of them [44]. Similar procedure was employed to identify the GABAergic or cholinergic terminals [42, 45]. In addition to vesicular transporters, the availability of P2X reliable antibodies made possible to associate the nucleotide and dinucleotide responses with specific P2X receptors. The presence of diverse homomeric and heteromeric P2X receptors was a handicap to clearly associate a response with one specific receptor. However, we demonstrated a very abundant presence of P2X3 subunits, assessed by immunohistochemistry, in the synaptic terminals responding to Ap₅A at the central nervous system, the presence of P2X1 or other P2X receptors, being much more scarce [40, 45]. It is relevant to emphasize that this technique also allows to analyse the interaction between different receptors present at the presynaptic structures. One example is the cholinergic synaptosomal population from rat midbrain, where about 63% of them respond to ATP and Ap₅A. These studies also permitted to demonstrate the wide co-expression of functional nicotinic and nucleotidic receptors. The percentage values of the terminals responding to both nucleotidic agonists and nicotine is about 40% of the total population. Immunological studies also confirmed the presence of P2X3 subunits and the α4 and α7 nicotinic receptor subunits [45].

The presence of diverse ionotropic receptors altogether at the presynaptic level suggested the possibility of interaction and regulation among them. In fact, most of the receptors can influence the activity of the nucleotide receptors, and vice versa, in specific synaptic terminals such as cholinergic, GABAergic, etc., organizing complex cross talks [46–48]. Besides, when this cross talk takes place in specific neural cells, such as cerebellar astrocytes in culture, it becomes much more complex and the signalling cascades where nucleotides and dinucleotides are implicated include much more elements, which are involved in neuroprotection and differentiation [49, 50].

Conclusion remarks

Diadenosine polyphosphates are relevant signalling molecules, as demonstrated by their vesicular storage, Ca²⁺-induced release and interaction with a wide range of purinergic receptors. Regarding their inactivation, we have shown that N2a neuroblastoma cells display an ectoenzymatic activity capable of degrading Ap_nA. Activity assays carried out with differentiated N2a cells showed that NPP1 is the main isozyme involved in the extracellular degradation of dinucleotides in these cells, this enzyme reducing its activity and changing its subcellular location following neuronal differentiation [51].

To conclude, we would like to remember and acknowledge again Prof. Burnstock by which we felt a sincere and deep admiration. It was a great honour for us that he accepted to become a Foreign Member of the Spanish Royal Academy of Pharmacy in 2004. Also, a Plenary Lecture in his honour was celebrated at the First European Purine Meeting in Santiago de Compostela in September 2019.

Funding

This work was supported by grants MEC (BFU2014-53654-P) and Fundación Ramón Areces (PR2018/16-02).

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

This article does not contain any studies with human participants or animals performed by any of the authors. Please note that the statements given above will only be JOURNAL ROLLOUT DETAILS 2 S/SD-12/1936 used if none is given in the manuscript.