News & Views | Published:


Bundling up excitement

Nature volume 407, pages 141143 (14 September 2000) | Download Citation


Glutamate is a ubiquitous amino acid; every cell in our body is packed full of it. But in the brain it functions in a unique way — as a neurotransmitter. In nerve cells, glutamate is bundled into tiny, membrane-encased spheres called synaptic vesicles. When a neuron is stimulated, these vesicles rush to its outer membrane, releasing glutamate outside the cell. Glutamate then rapidly excites other neurons. Neurons are usually defined by the neurotransmitter they release; most release glutamate (they are ‘glutamatergic’). As such, glutamate forms the basis for most of the brain's activity. It is responsible for your ability to see this journal, and to read and understand the words on this page. Over the past decade we have learned much about how glutamate excites neurons, but we know little about what gives a neuron the ability to release glutamate. On page 189 of this issue1 Takamori and colleagues provide the answer, in the shape of the protein responsible for packing glutamate into synaptic vesicles.

A neurotransmitter itself is defined in part by the molecular mechanisms that make it, release it and mop it up from the extracellular environment after it has done its job. Over the past 30 years we have come a long way towards understanding how glutamate is made and cleared up. Now it is the turn of the release mechanisms to face the glare of the researchers' spotlight.

When packets of glutamate are released from a stimulated neuron (Fig. 1), the neurotransmitter diffuses across the ‘synaptic cleft’ to the receiving neuron. There, it activates a series of receptors, leading to stimulation of that neuron, too. The glutamate then needs to be cleared up, and transmembrane glutamate transporters found on the surface of nearby astroglial (supporting) cells are the major synaptic vacuum cleaner. Finally, within the signalling (presynaptic) neuron, more glutamate has to be bundled into synaptic vesicles, ready for another round of stimulation.

Figure 1: What gives a glutamate-releasing neuron its identity? Most of the neurons in the mammalian brain release glutamate as a neurotransmitter.
Figure 1

When the presynaptic neuron is stimulated, synaptic vesicles containing glutamate merge with the neuron's plasma membrane and release their contents outside the cell. Glutamate diffuses to the postsynaptic neuron and binds to receptors there, activating the postsynaptic cell. The released glutamate then needs to be cleared away. This is achieved mainly by the glutamate transporters EAAT1 and EAAT2 (found on astroglial — support — cells), and to a lesser extent by transporters on the postsynaptic neuron (EAAT3 and EAAT4). In astroglial cells, glutamate is enzymatically degraded to glutamine. Glutamine may then be supplied to neurons, through the neutral-amino-acid transporters ASCT1 and ASCT2. Once in the neuron, glutamine is converted to glutamate. Takamori et al.1 and Bellocchio et al.4show that glutamate is then repackaged into vesicles by a protein called BNPI. Expression of BNPI in neurons that do not normally release glutamate is sufficient to make them do so1.

It turns out that most of the new glutamate is synthesized from another amino acid, glutamine, which is released from astroglial cells and taken up by the presynaptic neuron. An enzyme called phosphate-activated glutaminase converts the glutamine to glutamate, which is then bundled into synaptic vesicles. Biochemical studies indicated that a unique transporter was involved in the repackaging, but, until now, the molecular nature of this transporter has eluded researchers. In fact, the repackaging is one of the critical steps in the overall process of glutamate release, so the expression of the molecule required for this process might be what makes a neuron able to release glutamate.

Two years ago, it was discovered that the brain-specific, sodium-dependent phosphate transporter (BNPI) is found mainly in the presynaptic terminals of glutamatergic neurons2. Moreover, electron microscopy showed that BNPI localizes specifically to the synaptic vesicles in those neurons2. But it was not clear whether BNPI has a function in neurotransmitter physiology. Intriguingly, though, a protein called EAT-4 — a relative of BNPI in the nematode worm Caenorhabditis elegans — has an essential role in glutamate transmission3. So it was suspected that BNPI might have a similar function in mammals.

In an elegant series of studies, Takamori et al.1, along with Bellocchio et al.4, now show that the expression of BNPI in cell lines leads to specific, ATP-dependent uptake of glutamate into vesicles. Does this mean that this protein is what makes a neuron glutamatergic? Takamori et al.1 go on to show that the expression of BNPI in a cell line that can be stimulated by the excitatory neurotransmitter acetylcholine results in the release of packets of glutamate from the cells. Finally, they convincingly show that neurons that normally release the neurotransmitter GABA can be persuaded to release glutamate instead by the artificial expression of BNPI.

So it seems that the vesicular glutamate transporter BNPI may be the best marker by which to define a glutamatergic neuron. But in some ways, all that these groups1,4 have done is identify another vesicle-bound neurotransmitter transporter. What makes the work much more exciting is that we may now have a way of specifically controlling neurotransmission within glutamatergic neuronal circuits. A wide range of neurological diseases are characterized by the aberrant regulation of glutamate. These include acute stroke (brain injury resulting from a sudden loss of blood or oxygen to the brain), as well as more chronic neurodegenerative disorders such as Huntington's disease and amyotrophic lateral sclerosis. Both vesicular5 and transmembrane6 glutamate transporters have been implicated in these diseases.

The past decade has seen an enormous effort by pharmaceutical firms to develop potent ‘anti-glutamate’ agents as neuroprotective compounds. But many of these have been either ineffective or too toxic. The new results afford us the opportunity to control the release of glutamate by manipulating its loading into vesicles. But some questions remain unanswered. For example, BNPI was identified as a phosphate transporter. Can it function as such in vivo and, if so, which is its main role — to uptake glutamate or to transport phosphate? In addition, BNPI is not found in all of the known neuronal pathways along which glutamate travels, so other types of vesicular glutamate transporter may exist. If so, we may one day be able to develop drugs that are specific to these different types, allowing us precise control of the glutamatergic neuronal circuitry.


  1. 1.

    , , & Nature 407, 189–194 ( 2000).

  2. 2.

    et al. J. Neurosci. 18, 8648– 8659 (1998).

  3. 3.

    , , , & J. Neurosci. 19, 159–167 (1999).

  4. 4.

    , , & Science 289, 957– 960 (2000).

  5. 5.

    , , , & Nature Genet. 25, 385– 389 (2000).

  6. 6.

    et al. Neuron 20, 589–602 (1998).

Download references

Author information

Corresponding author

Correspondence to Jeffrey D. Rothstein.

About this article

Publication history




  1. Search for Jeffrey D. Rothstein in:


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing