Love is in the... striatum!

愛	Romantic love is an emotion that stirs our souls. Just by looking at the beloved person, it causes strong feelings like butterflies in the stomach. If love produces so strong emotions one may wonder what it does to the brain.

Bartels and Zekin (2000) wanted to know which regions of the brain are activated when someone in love observes the face of the beloved person. They recruited 17 normal subjects 'truly, deeply and madly in love,' and examine their brains with functional magnetic resonance imaging (fMRI) as the subjects observed pictures of their loved ones and friends. Then, they compare the activation in brain areas induced by the beloved person in relation to the friends. To the surprise of the authors, very few regions were selectively activated; medial insula, anterior cingulate cortex, and striatum (nucleus caudate and putamen). All areas were activated in both hemispheres (bilateral).

The activation of the insula and cingulate cortex did not surprise Bartels and Zekin. These areas are related to several emotions. The unexpected result was to find activation of the corpus striatum, an area involved in motor planning. Would this mean that just by seeing the beloved person we want to run to her? The authors were also puzzled:

"Collectively, these results call for a reappraisal of the role of the putamen and caudatus in emotional states and as part of the extrapyramidal system."

Sometimes, however, relationships are broken, and there is grief, especially when there is separation from the beloved. Najib et al., (2004) studies the brain of 17 women experiencing grief, and who were turning their situation over and over in their mind (i.e. "ruminating"). They used fMRI to show that acute grief increases the activity in several posterior brain regions, and decreased the activity in the thalamus, striatum, and cingulate cortex. For women that experience a larger grief, the decrease in activity in these regions was larger.

In another study, Aron et al., (2005) studied the activation of the brain regions, similar to the study of Bartels and Zekin (2000), but at early stages of the relationship. This time, the subjects were in love for 1–17 months. They found activation of the ventral tegmental area, right postero-dorsal body, and medial nucleus caudatus. Interestingly, the higher the love (in a passionate love scale) is, the larger the response of the nucleus caudatus.

What could be the meaning of striatal activation? Perhaps it is related to the dopamine-reward system, i.e. the brain knows that she can make me happy. Or, it may be related to the eye movements; when one stare at the beloved person, the eyes make saccades to scan her face.

a to-do-list for the striatum

There is abundant information about striatal circuits described in reviews and textbooks. Some believe that the cellular details of the basal ganglia circuits and the role of interneurones have been already described in textbooks. So, one might get the wrong impression that there is nothing further to investigate.

Although much is known about the basal ganglia, some of the synaptic interactions between components are not known. In the classical scheme of the direct and indirect patways, the thalamo-striatal component is not shown. What is the function of this pathway? does it regulate principal neurones or interneurones? There is input from the cerebral cortex directly to the nucleus subthalamicus, skiping the striatum. What is the significance of this pathway, and what information does it carries? The tail of the nucleus caudatus is continuous with the nucleus accumbens, is there any functional significance?

Recently, Ann Graybiel (2005) facilitated our task in finding things to be investigated. In a review published in the December issue of Currents Opinions in Neurobiology, she identify six challenges to the basal ganglia anatomy and function. These are the following:

• Do the direct and indirect pathway project exclusively to different target nuclei?
• Is the pallido-thalamic pathway only inhibitory?
• Is dopamine the only neurotransmitter substance released by the dopamine-containing neurons of the midbrain?
• Do the direct and indirect pathway receives equivalent information from cortical afferents?
• Do the basal ganglia and cerebellum have fully separated functions and pathways to the neocortex?
• Do striosome code reinforcement-related signals?

This list gives us some idea of where to start asking questions. It can be added to the list whether all GABAergic interactions in the striatum are excitatory. In addition to the mechanisms commented in a previous post, GABAergic cell can respond to hyperpolarisation with a rebound spike. Is this spike physiologically significant? i.e. can it lead to excitation of a postsynaptic cell?

Happily for researchers, there is still work to do.

Excitatory GABA goes to the striatum

In textbooks, GABA (gamma-aminobutyric acid) is an inhibitory neurotransmitter. Some investigators have found, however, that GABA_A can mediate excitatory responses (Owens and Kriegstein, 2002). The reason for this curious phenomenon is that in some immature neurones, the chloride gradients are not yet established, setting the reversal potential for chloride above threshold. Now, excitatory responses, mediated by GABA_A receptors have been found in the medium-spiny neurones of the striatum.

Beside the chloride ionic gradient, another mechanism, by which GABAergic responses can be excitatory, is the co-release of GABA and glutamate. The experiments of Bracci and Panzeri (2006) published in the Journal of Neurophysiology are pristine. They recorded from striatal slices of older animals (18–28 postnatal), where the chloride gradients should have been established. They recorded from striatal medium-spiny neurones (the projection neurones) using perforated patch. In presence of NBQX and DAP-5, they found post-synaptic potentials that reversed at -64 mV, and were sensitive to bicuculline.

It is unlikely that another neurotransmitter mediate the responses, because, they were sensitive to a GABAA receptor antagonist (bicuculine). Other possibility would be that the chloride gradients are different for medium-spiny neurones that have reached a mature state. This is an adventurous hypothesis that would need to be tested, and confirmed. Another alternative would be another, yet to be discovered, GABA receptor with poor chloride selectivity.