Read the full article “Expression of human equilibrative nucleoside transporter 1 in mouse neurons regulates adenosine levels in physiological and hypoxic-ischemic conditions” on page 4.

The manuscript by Zhang et al. (2011), from the group of Fiona Parkinson, provides unexpected but convincing evidence that nucleoside transporters into neurons contribute to the removal of adenosine from the synaptic space even during a hypoxic or ischemic insult. Purines, namely ATP and adenosine, are released by most cells under physiologic conditions, their extracellular levels being exacerbated under stressful conditions, where adenosine may act as a protective agent, an action particularly evident in neuronal cells (e.g. Rudolphi et al. 1992). Upon hypoxia or ischemia, extracellular adenosine accumulates and an immediate consequence is the depression of evoked synaptic potentials, as initially shown by Fowler (1989). It is clearly established that this depression is neuroprotective, facilitating recovery of synaptic function after the hypoxic insult (Sebastião et al. 2001). Indeed, when adenosine is not allowed to act on inhibitory A1 receptors, protection can be achieved by preventing the occurrence of evoked synaptic events during the hypoxic period (Sebastião et al. 2001). In spite of our considerable knowledge about neuroprotection by adenosine, little is known about the sources of adenosine during hypoxic insults. There are clearly two key aspects that still deserve further attention: (i) the relative contribution of different cell types, and (ii) whether adenosine is released as such or formed extracellularly from released adenine nucleotides. Once in the extracellular space adenosine is taken up by the cells through the same molecules that promote adenosine release, the equilibrative nucleoside transporters (ENTs), which transport adenosine in or out of the cells according to its concentration gradient. So, the direction and activity of ENTs determine the transient levels of adenosine at synapses. ENT activity can be regulated by adenosine receptors (Pinto-Duarte et al. 2005), in other words, the extracellular levels of adenosine feedback upon the rate of its clearance from the synapses, highlighting the relevance of the control of ENTs for proper adjustment of adenosine-induced synapse fine-tuning (Sebastião and Ribeiro 2009). Control of extracellular levels of adenosine at synapses, can differ from other neuronal regions. Furthermore, different cell types at the tripartite synapse can contribute in a different way for the control of synaptic adenosine levels.

The novelty in the approach by Zhang et al. (2011) is the use of a transgenic mice model that selectively over-expresses an equilibrative nucleoside transporter subtype (ENT1) in neurons (Parkinson et al. 2009), therefore not in astrocytes, together with a functional readout of the synaptic adenosine levels, its ability to depress synaptic transmission. At the synapses, the extracellular levels of adenosine are often higher than the intracellular ones because there is a considerable pool of extracellular adenosine that is formed from ATP, released from nerve endings and from glial cells, namely astrocytes and oligodendrocytes (Fields and Burnstock 2006). ATP acts itself as glia-to-neuron signalling molecule at the tripartite synapses, but is also quickly degraded by ectonucleotidase, contributing in this way to the extracellular pool of adenosine. During hypoxia, it was initially believed that most of the accumulating extracellular adenosine resulted from its increased intracellular accumulation and subsequent release through ENTs. This concept started to be challenged by work from Bruno Frenguelli’s group (Frengelli et al., 2007) showing that the extracellular levels of adenosine during hypoxia, assessed with biosensors, were increased rather than decreased upon inhibition of ENTs. The work by Zhang et al. (2011) further reinforced this surprising conclusion, and firmly established that this does occur at the synaptic level, with a relevant involvement of neuronal ENTs. To evaluate synaptic adenosine levels, Zhang et al. (2011) used a bioassay (A1 receptor-dependent depression of synaptic transmission) to evaluate synaptic adenosine levels during hypoxia or ischemia ex vivo, and compared the synaptic depression in slices from mice over-expressing neuronal ENT1 with slices from wild type littermates. They clearly found that hypoxia or ischemia-induced synaptic depression was smaller in animals over-expressing neuronal ENT1, clearly establishing that neuronal ENTs contribute to the decrease rather than replenishment of the pool of synaptic adenosine even under hypoxic/ischaemic conditions (Fig. 1). Accordingly, synaptic depression was exacerbated by blockade of ENT1, a condition that also abolished the differences between wild type and transgenic mice. Adenosine release precedes ATP release during ischaemia (Frenguelli et al., 2007). So, if ATP breakdown does not significantly contribute to the production of extracellular adenosine during ischaemia, if ENT1 in neurons take up rather than release adenosine during hypoxia/ischaemia, the source of adenosine under such conditions remains to be established. A possibility is the astrocytic pool, since astrocytes contribute to levels of extracellular adenosine during hypoxia (Martín et al. 2007). Mice over-expressing ENT1 in astrocytes and not in neurons, when available, will allow to figure out whether ENT1 in astrocytes contributes to adenosine release (as initially thought) or uptake (as shown in neurons by Zhang et al. 2011) during hypoxia. The understanding of the relative contribution of the different ENT, namely of ENT2, which also has high affinity for adenosine and is expressed in most cells, is another necessary step for a more complete understanding of the influence of equilibrative adenosine transporters in the control of extracellular adenosine in neuronal tissue under pathologic conditions.


Figure 1.  Schematic diagram on the release and uptake of adenosine by equilibrative nucleotide transporters (ENTs) during normoxia and hypoxia. As intracellular concentrations of adenosine are usually lower the extracellular ones, ENTs usually take up adenosine. Under hypoxic, ischemic or other cell stress conditions, the intracellular levels of adenosine increase, forcing the reversal of ENTs, which release adenosine into the extracellular space. Surprisingly, and as shown by Zhang et al. (2011), neuronal ENT1 at synapses keeps taking up adenosine even under hypoxic/ischemic conditions, as highlighted by the red arrows. The relative role of other ENTs, namely ENT2, also abundant in neurons and glia, and of astrocytes for the control of extracellular adenosine under hypoxia/ischemia, awaits further and surely exciting research.

Download figure to PowerPoint

In spite of the questions still waiting to be answered, it is clear from the work by Zhang et al. (2011) that neuronal ENT1 contributes to adenosine removal rather than adenosine release during hypoxia. Transgenic mice models with cell specific expression of high affinity adenosine transporters are indeed a novel and valuable tool to undertake studies on the role of these transporters in the control of extracellular adenosine levels and hence on neuroprotection.


  1. Top of page
  2. References