A question that comes to mind to most when reading an article such as the one Yizhar et al. published online July 27 in Nature is: is this a big step forward or a flashy way to show what we already know? The answer is: both. It has to be.

Optogenetic tools to address neurobiological questions in a well-controlled manner, with selective activation or inactivation of specific brain areas and even cell types, pioneered by Karl Deisseroth’s group, have been around for some time. Deisseroth deserves recognition for developing such a clever tool, but even stronger recognition for proactively sharing these tools with anyone who requests them.

So far, optogenetic tools have been mostly used to address questions for which there were extensive previous studies, albeit with less conclusive techniques. Naturally, some would see that as just a replication of existing knowledge. However, an important first step for a novel tool like this one is to establish credibility, and what better approach than tackling questions for which we know the answer? For example, a couple of papers using optogenetics in 2009 “demonstrated” that parvalbumin (PV) fast-spiking interneurons in cortical circuits are critical for high-frequency oscillations (Cardin et al., 2009; Sohal et al., 2009). Extensive studies had suggested that cortical GABA interneurons are critical for high-frequency oscillations (Atallah and Scanziani, 2009; Puig et al., 2008; Tort et al., 2007; Fries et al., 2007). But the ability to selectively activate or inactivate this neuronal population was not available until the arrival of optogenetics. Addressing an issue with strong data already available was a necessary step, and now we can examine the role of inhibitory cortical neurons in physiology and behavior with well-controlled and reliable tools. One could also place quotation marks around the “we know the answer” in the previous question, since a conclusive causal link between interneuron activity and oscillations, assumed by most, was actually missing. Thus, optogenetic tools may represent flashy confirmations, but one can say the information they have provided so far has nailed old questions in a mechanistic way.

There have been other approaches to establish this type of mechanistic link, including genetic manipulations. For example, knocking out NMDA receptors in PV interneurons yields a number of alterations in adult animals (Belforte et al., 2010), showing that a number of behavioral and electrophysiological properties do depend on interneuron function. These approaches are not mutually exclusive and are quite complementary. Optogenetics allows manipulation of these systems while the animals engage in behaviors, permitting a within-subject comparison, whereas transgenic animals allow the exploration of developmental aspects of interneuron function. Both approaches are here to stay, but are not likely to displace more common and established tools.

In this week’s article, Yizhar et al. provide more than a dramatic confirmation using an array of novel opsins to enhance or suppress activity in pyramidal excitatory neurons or local inhibitory interneurons. Step-function opsins allow lasting depolarizations in either pyramidal neurons or parvalbumin (PV)-positive interneurons in the rat medial prefrontal cortex, enhancing their excitability as a means to alter excitation-inhibition balance within this cortical region while assessing behavioral performance. Shifting excitation-inhibition balance towards increased excitation, but not the other way around, impaired social behaviors and enhanced high-frequency oscillations. There may be some caveats in how physiological a step-function opsin-induced increase in firing may be. A constant depolarization may not be the best way to reproduce firing patterns in fast-spiking interneurons, which rely on a fast and sharp K+ channel-dependent hyperpolarization to rapidly reactivate Na+ channels and allow fast frequencies of firing. Despite this limitation, the authors showed a decrease in pyramidal cell firing following activation of PV neurons, suggesting inhibitory processes were enhanced.

As with other highly visible optogenetic manuscripts, the issue at hand (i.e., whether excitation-inhibition balance is critical for PFC behavior) had been extensively studied. Altered excitation-inhibition balance is a key new element in current views of the pathophysiology of psychiatric disorders, and understanding its link to behavior is certainly critical. We recently showed cognitive deficits in rats with a neonatal ventral hippocampal lesion, a widely used rodent model of major psychiatric disorder that is characterized by loss of activation in prefrontal cortical interneurons (Gruber et al., 2010). However, in ours and all previous studies, the link between interneuron activity or excitation-inhibition balance and behavioral performance was correlational in nature. Yizhar et al. altered behaviors by selectively targeting pyramidal neurons or interneurons, and a causal link could be established. This is both a flashy confirmation and a leap forward. To be able to establish causality where extensive studies have shown strong correlations hinting at causality is a welcome step. Perhaps optogenetics has come of age now, and it is ready to tackle novel questions, but work like what Yizhar et al. produced is indeed an illuminating (a literal blue one) and big step forward.

References:

Atallah BV, Scanziani M. Instantaneous modulation of gamma oscillation frequency by balancing excitation with inhibition. Neuron . 2009 May 28 ; 62(4):566-77. Abstract

Belforte JE, Zsiros V, Sklar ER, Jiang Z, Yu G, Li Y, Quinlan EM, Nakazawa K. Postnatal NMDA receptor ablation in corticolimbic interneurons confers schizophrenia-like phenotypes. Nat Neurosci . 2010 Jan 1 ; 13(1):76-83. Abstract

Cardin JA, Carlén M, Meletis K, Knoblich U, Zhang F, Deisseroth K, Tsai LH, Moore CI. Driving fast-spiking cells induces gamma rhythm and controls sensory responses. Nature . 2009 Jun 4 ; 459(7247):663-7. Abstract

Fries P, Nikolic D, Singer W. The gamma cycle. Trends Neurosci . 2007 Jul 1 ; 30(7):309-16. Abstract

Gruber AJ, Calhoon GG, Shusterman I, Schoenbaum G, Roesch MR, O'Donnell P. More is less: a disinhibited prefrontal cortex impairs cognitive flexibility. J Neurosci . 2010 Dec 15 ; 30(50):17102-10. Abstract

Puig MV, Ushimaru M, Kawaguchi Y. Two distinct activity patterns of fast-spiking interneurons during neocortical UP states. Proc Natl Acad Sci U S A . 2008 Jun 17 ; 105(24):8428-33. Abstract

Sohal VS, Zhang F, Yizhar O, Deisseroth K. Parvalbumin neurons and gamma rhythms enhance cortical circuit performance. Nature . 2009 Jun 4 ; 459(7247):698-702. Abstract

Tort AB, Rotstein HG, Dugladze T, Gloveli T, Kopell NJ. On the formation of gamma-coherent cell assemblies by oriens lacunosum-moleculare interneurons in the hippocampus. Proc Natl Acad Sci U S A . 2007 Aug 14 ; 104(33):13490-5. Abstract

View all comments by Patricio O'Donnell

Optogenetics Stimulates Our Thinking About Cortical Pathology in Schizophrenia
We recently reviewed this paper in the weekly Journal Club at the Schizophrenia Research Laboratory in Sydney, Australia. Here are some brief thoughts from our discussion:

First, we were very impressed with the development of the novel stable step function opsin (SSFO) and the anatomical and temporal precision with which it can be used in the mouse. Particularly powerful is the ability to induce cortical excitation in a time frame long enough to impact behavior without the confound of developmental compensatory change, as may occur in genetically engineered mice. The paper did raise a few questions in our minds about how to best relate this to findings in schizophrenia, especially in light of the ongoing debates as to: 1) whether the cortex is actually overactive or underactive (hyperactivity versus hypoactivity by fMRI); 2) whether the cortex shows increased γ band synchrony (at baseline) or decreased γ band synchrony (induced); and 3) whether there is more or less GABA present (postmortem decrease in GAD67 with possible increase by MRS) in people with schizophrenia.

This paper helps us to understand how an exogenously stimulated increase in overall cortical excitability could interfere with some basic social behaviors. It certainly did seem like the "overexcited" mouse was disinterested in the novel immature mouse and preferred not to socialize. Even though the reasons for a mouse not to engage socially are not known, we do know that people with schizophrenia can report that they are feeling paranoid or nervous around other people. So, it would be tempting to wonder whether these thoughts or feelings may be generated by increased prefrontal pyramidal neuron activity, but this, of course, will be difficult to prove. Exploration of the relationship between positive symptoms—hallucinations and delusions—and overactivity of cortical brain regions (by fMRI and with possible disruption by rTMS), in particular in the temporal lobe, may help to answer this question. Additionally, investigation of localized versus cortex-wide changes in excitation/inhibition could shed light on whether pathology in a particular cortical region is necessary and sufficient to drive behavioral changes or even pathological changes, such as increases in subcortical dopamine (Lisman et al., 2008).

While we were convinced by the lack of clear impact of prefrontal cortical "over-inhibition" on behavior, except in combination with "overexcitation," we had some questions regarding the choice of parvalbumin positive (PV+) neurons for modulation in this study. Certainly, there is accumulating evidence that they are pathological in schizophrenia, but recent evidence suggests that PV+ interneurons can mediate both excitatory and inhibitory effects of GABA, depending on the location of their synapse onto the target cell (Szabadics et al., 2006). A subset of PV+ neurons innervate the axon initial segment and may actually stimulate the pyramidal neurons through excitatory actions of GABA at that particular location, whereas PV+ basket neurons innervate the soma, and mediate inhibitory effects of GABA. Therefore, would basket cells be the more likely mediators of the hyperpolarization and loss of excitation reported in this study? Future studies exploring the excitatory/inhibitory balance in schizophrenia may benefit from more anatomically based evaluations of specific subtypes of interneurons in the postmortem human brain. Furthermore, since this technology has the potential to offer a greater understanding of the potential primacy of pathophysiological events, we wonder whether a channel rhodopsin that gates chloride ions into the cells (provided it could be stabilized, as the SSFO was here) would be a way to test if less activity in the different populations of inhibitory interneurons (i.e., somatostatin compared to parvalbumin) could lead to different types of hyperexcitable states in cortical pyramidal neurons? Also, understanding how these more specific changes could lead to changes in more refined and context-dependent tests of social interaction or working memory would be informative alongside measures of pathological change at the level of neurotransmitter release or molecular signaling. In sum, we thoroughly enjoyed discussing the paper, and it gave us much to consider. We applauded the group and hope to continue to have more great papers to discuss!

References:

Lisman, J. E., Coyle, J. T., Green, R. W., Javitt, D. C., Benes, F. M., Heckers, S. and Grace, A. A., 2008. Circuit-based framework for understanding neurotransmitter and risk gene interactions in schizophrenia. Trends in Neurosciences 31, 234-242. Abstract

Szabadics, J., Varga, C., Molnar, G., Olah, S., Barzo, P., Tamas, G., 2006. Excitatory effect of GABAergic axo-axonic cells in cortical microcircuits. Science 311, 233-235. Abstract

View all comments by Cynthia Shannon Weickert
View all comments by Duncan Sinclair
View all comments by Vibeke Catts