Habenular DBS as a treatment for depression

To understand this paper one needs some context. There was already a wealth of information suggesting the lateral habenula (LHb) had altered activity in depressed patients and that inhibition of the LHb could modulate depressive-like behavior in animals. A few years back, Hauptman et al. (2008) reviewed the neuroanatomical literature and proposed several potential targets, including the LHb, for therapeutic deep brain stimulation (DBS) in patients with treatment resistance depression. Sartorius et al (2010) in a rather fortuitous case study demonstrated therapeutic efficacy of DBS in the LHb. After a few weeks of receiving DBS the patient showed improved symptoms until a bicycle accident sent her back into remission. Unbeknownst to anyone, the accident had dislodged her battery pack and DBS had stopped. Once the pack was repaired the patient soon improved. By shear dumb luck, the researchers had stumbled into a well controlled ABA study design that would otherwise have been unethical. But would this case study prove to be fodder for more research or a fluke?

The former, it seems. Li et al. (2011) provides evidence that DBS of the lateral habenula significantly improves outcomes in two animal models of depression: learned helplessness and forced swim. The researchers first task was to determine whether synaptic transmission onto LHb neurons is affected in a learned helplessness model. LHb neurons exhibiting a retrograde fluorescent marker from VTA neurons were recorded (1,2) by whole cell patch clamp in rats exposed to uncontrollable foot shock (acute learned helplessness; aLH) and rats bred to express learned helplessness (chronic learned helplessness; cLH). A recording of mEPSCs, a measure of neurotransmitter release, showed a higher frequency in aLH and cLH rats relative to control, demonstrating excitatory synaptic potentiation in these animals. But, was this due to learned helplessness or some other factor? Researchers correlated failure to escape shock with mEPSC frequency and found in both wild type and cLH rats there was a positive correlation (3). mEPSC size is unaffected, as are mIPSCs. So, learned helplessness, it seems, leads to potentiation of VTA projecting lateral habenula neurons.

Potentiation implies more input, but synaptic density did not appear to be altered in these rats. That means changes in presynaptic release of neurotransmitter is likely responsible for the potentiation. By evoking release through electrical stimulation, researchers showed that EPSCs amplitude decreased more rapidly in cLH rats with successive stimulation. This implies that more of the neurotransmitter was release during the initial EPSC, meaning the release probability (i.e. the probability that neurotransmitter will be released into the synaptic cleft) is higher in cLH rats. This is confirmed by the lower failure rate (i.e. failure to produce an EPSC) following minimal electrical stimulation and the non-significant difference in amplitude of successful EPSCs.

From this one would predict that repeated stimulation of afferents to the LHb would deplete neurotransmitters and diminish evoked potentials in LHb neurons, which is precisely what was shown in slice by stimulating at 130 Hz, the same frequency used by Sartorius et al. in the above mentioned case study. But does this improve the behavioral outcome? Of course it does or this paper wouldn’t be in Nature. DBS of the LHb led to improved learning of an shock escape task, active shock avoidance task, and elevated mobility during a forced swim task, three separate measures of depressive-like behaviour in animals (4). A small caveat is that DBS was only tried in those animals expressing the highest level of learned helplessness; however, one could argue that is the more representative of treatment resistant depression. What is more interesting, and not really noted in the body of the paper, is that DBS was conducted unilaterally. If that pans out for humans, that means a less invasive surgery would be necessary for probe implantation.

So potentially we have evidence to support the further use of therapeutic DBS in the LHb for treatment resistant depression. One of my main concerns of this paper is the concentration on VTA projecting LHb neurons. While there is some rational for concentrating on these, there is no reason why neurons that project elsewhere aren’t involved. This would require looking at other habenular efferent neurons and devising a way to alter transmission in VTA projecting neurons alone, probably through an optogenetic technique. This of course is beyond the scope of a single study, even for Nature, but the possibility of using such research for the treatment of mood disorders in the future can not be understated.

Li B, Piriz J, Mirrione M, Chung C, Proulx CD, Schulz D, Henn F, & Malinow R (2011). Synaptic potentiation onto habenula neurons in the learned helplessness model of depression. Nature, 470 (7335), 535-9 PMID: 21350486

1) While the VTA is thought of as a dopaminergic center, there are glutamatergic neurons there too. So whether the individual LHb neurons recorded synapse exclusively on one type or the other or both is uncertain. It is also unclear why they targeted the VTA rather than say the dorsal raphe or locus coeruleus. I suspect they started with this experiment and the idea to tack on DBS came later.

2) It would be nice to see where in the LHb neurons were recorded from. Supplementary figure 2 suggest that they are mainly in the medial portion of the LHb, but there are quite a number of subnuclei in this structure.

3) They didn’t test aLH rats. Why? Beats me.

4) No wild type controls for this experiment. I would have liked to see the basal rates of escape, avoidance, and immobility.