Optogenetics and Its Potential for Use in Research and the Clinic

Brain Stimulation (2011) 4, 1–6

www.brainstimjrnl.com

ORIGINAL ARTICLES A new technique for controlling the brain: optogenetics and its potential for use in research and the clinic

Ryan T. LaLumiere

Department of Psychology, University of Iowa, Iowa City, Iowa

The recent development of optogenetic techniques has generated considerable excitement in neuroscience research. Optogenetics uses light to control the activity of neurons which have been modified to express light-sensitive proteins. Some proteins, such as channelrhodopsin, are cation channels that produce depolarization of neurons when illuminated. In other cases, neuronal activity can be inhibited through illumination of proteins, such as the chloride pump halorhodopsin, that hyperpolarize neurons. Because these proteins can be selectively expressed in specific cell types and/or in specific locations, optogenetics avoids several of the non-specific effects of electrical or pharmacological brain stimulation. This short review will explain the physiology of this technique, describe the basic and technical aspects of the method, and highlight some of the research as well as the clinical potential of optogenetics. Ó 2011 Elsevier Inc. All rights reserved.

Keywords channelrhodopsin; halorhodopsin; ChR2; NpHR

Over the past few years, optogenetics has begun to optogenetics and the potential for its use in research as revolutionize the field of neuroscience as well as opening well as in the clinic. up new possibilities in the field of brain stimulation. The optogenetic technique uses light to control the activity of neurons expressing light-sensitive proteins, thus enabling Physiology of light-sensitive proteins those using the technique to avoid the nonspecific effects of for control of neuronal activity electrical stimulation or pharmacologic manipulation. Moreover, because the expression of the proteins is genet- In the first report on the use of the light-gated cation ically targetable, optogenetics limits the population of channel channelrhodopsin-2 (ChR2; Figure 1 for a depiction neurons that respond to the light, providing an important of ChR2), Boyden et al.1 showed that pulses of blue light at level of resolution. This review will provide a brief approximately 470 nm wavelengths induce spikes of action explanation of the physiology and technical aspects potentials in cells expressing the ChR2 protein. Hippo- involved in this technique and address recent uses of campal neurons expressing these channels can follow trains of light pulses up to 50 Hz with 95% fidelity. 2 Importantly, long-term ChR2 expression does not appear to alter the Correspondence: Ryan T. LaLumiere, Department of Psychology, 11 basal physiology of transduced neurons or show evidence Seashore Hall E, University of Iowa, Iowa City, IA 52242. of toxicity,1 suggesting that neuronal expression of ChR2 E-mail address: [email protected] Submitted June 17, 2010; revised September 3, 2010. Accepted for produces optically controllable, physiologically relevant publication September 24, 2010. systems. Studies in mice with retinal photoreceptor

Na+ Cl-

Channelrhodopsin Halorhodopsin Figure 1 Channelrhodopsin (left, in blue) is a light-sensitive, membrane-bound cation channel that opens and closes with the onset and offset of w470 nm (blue) light, respectively. When activated, channelrhodopsin produces depolarization of the expressing neuron, leading to temporally precise control of action potential firing. Halorhodopsin (right, in yellow) is a light-sensitive, membrane-bound chloride pump that, like channelrhodopsin, works with excellent temporal fidelity to the onset and offset of w580 nm (yellow) light. When activated, halorhodopsin hyperpolarizes the neuron, preventing action potentials. degeneration have demonstrated that virally delivered inactivation that can occur with NpHR.8 As new opsins ChR2 in retinal cells restores the ability of the neurons to are developed, they will continue to expand the kinds of transmit information to the visual cortex in response to light questions that are tractable with optogenetics. in a safe, stable long-term manner.3 Since the development of ChR2, other opsins have been identified or engineered and used to control neural activity. Technical aspects in the use of optogenetics Halorhodopsin (NpHR) is a light-sensitive chloride pump (see Figure 1 for a depiction of halorhodopsin) that, when Opsin expression activated, provides significant hyperpolarization of the transduced neuron.4 Although early work with NpHR Transgenic animals can be created with opsin expression demonstrated problems with efficacy and intracellular under the control of general or specific promoters,9,10 aggregates with high levels of NpHR expression, alterations though, with cell class-specific promoters, the protein to the NpHR protein have produced second generation expression is often weak. To achieve the high opsin expres- (eNpHR) and third generation (eNpHR3.0) NpHRs that sion levels necessary for optogenetic control, a mouse with do not produce aggregates and have improved hyperpola- the transgene Cre-recombinase under a cell-specific rizing functioning.5 NpHR has the advantage of being promoter can be used. This system then uses a viral vector sensitive to a wavelength spectrum with a peak at approx- that transduces the neuron with the opsin gene, requiring imately 580 nm, producing minimal overlap with the the action of Cre for opsin expression.11 This system optimal wavelength spectrum for stimulation of ChR2 enables the opsin itself to be under the control of a strong (w470 nm). As a result, it is possible to transduce neurons but general promoter, as the neuronal specificity is with both opsins and stimulate or inhibit neuronal activity conferred by the promoter for the Cre-recombinase and with pulses of light of different wavelengths.4 opsin expression cannot occur without the action of Cre. Considerable advances continue to be made in the This has permitted experiments in which the opsin expres- development of opsins for neurobiologic use. For example, sion is limited to dopaminergic neurons of the ventral optoXRs are optically controllable G-protein-coupled tegmental area.11 The major disadvantage of this method receptors that trigger traditional intracellular signaling is that it requires the use of transgenic animals, thus pathways.6 Recent engineering has produced a ChR2 limiting it to certain species and nonclinical use. mutant labeled ChETA that reduces double-spiking in For optogenetics in other species that are less tractable response to light, a problem with ChR2, and can follow for transgenic models, such as rats, or for potential human trains up to 200 Hz, far beyond the 40-50 Hz ceiling for therapeutic use, viral transduction of cells provides the best many cells expressing ChR2.7 Recent screening of micro- route for opsin expression. In particular, the specificity and bial opsins has led to the discovery and characterization localization of virally mediated expression would be highly of archaerhodopsin-3 (Arch) and the Leptosphaeria macu- desirable in clinical use. Adeno-associated virus (AAV) has lans opsin (Mac), two light-activated proton pumps that already been used to deliver genes into neurons of humans provide powerful inhibition of neurons without the as part of gene therapy,12-14 making AAV a highly feasible Optogenetics and neuroscience 3

A B

M P X N Z Y O

Figure 2 A, One method of targeting specific neurons using optogenetics uses cell-specific promoters that control the expression of the opsin. In this case, neurons X and Y are different classes of neurons but are located in the same structure. Thus, a virus containing channelrhodopsin and a promoter specific to neurons like X can be injected into the structure, but only neuron X will express channelrhodopsin, enabling cell-type specific control of activity in the structure. B, In another method for gaining specificity, the structure on the left (in yellow) is transduced with a channelrhodopsin-containing virus but only the axon terminal region for neuron M is illuminated with the appropriate light. This method allows for selective control of a specific pathway without affecting other efferent pathways from the transduced structure (e.g., neuron N) or other afferent pathways entering the illuminated structure (e.g., neuron O). vector for delivery of opsins in future clinical studies. In used to those small enough to be packaged with the opsin optogenetics, viruses such as AAV or lentivirus (LV) can gene in the virus. be packaged with the transgenic elements containing the With optogenetic methods, one can thus stimulate sequence for the opsin as well as a promoter.15 Both viruses neurons of a particular cell type in a specific location produce long-term expression of the opsin protein. Full, (anatomy) with tremendous temporal precision and control. stable in vivo expression of the opsins in rodents is depen- This kind of precision outstrips traditional electrical dent on both vector and promoter choice and typically stimulation methods and has already provided valuable requires 1-3 weeks for the cell body and 6 or more weeks information on thorny neuroscience questions such as the for the distal ends of axons, depending on the distance possible mechanisms of action of deep brain stimulation and virus used.16 If the virus is administered via microin- (DBS) for Parkinson’s Disease (PD) and the origins of the jections into the neural structure of interest, successful blood oxygen level dependent (BOLD) imaging signal transduction will occur only in the population of neurons commonly used in fMRI studies.19,20 These examples will within a geographically circumscribed area around the be discussed below. site of the injection, depending on, among other factors, the viral titer, volume infused and, for AAV, serotype Technical methods for conducting optogenetic used. Thus, spatial specificity of the opsin expression can experiments be controlled by the location of the infusion as well as titer and type of virus used. In combination with the physical When expression of the opsin has occurred, stimulation of limits of illumination, viral transduction of opsins can the opsin typically occurs via light transmitted through produce a powerful method of gaining spatial resolution. a fiber optic cable. Because the opsins are expressed in the As with transgenic animals, opsin production mediated membranes of the neurons, from the cell bodies to the axon by viral transduction is controlled by promoters in the terminals, the location of the fiber optic tip in the brain can transgene cassette. These promoters can confer specificity confer specificity with regard to spatial resolution in the in terms of the types of cells that will express the protein. control of neurons. The cell bodies of neurons can be Figure 2A provides a visual explanation of the benefits of illuminated, as in Figure 2A, or alternatively, the fiber this kind of specificity. For example, use of the CaMKIIa optic tip can be placed in the axon terminal region, as in promoter will permit the protein to be selectively expressed Figure 2B. Illumination of the axon terminals allows for within excitatory glutamatergic neurons in certain control over specific pathways without directly affecting regions.17 Alternatively, a general neuronal or ubiquitous other afferent pathways to the illuminated structure or promoter can be used if such specificity is not necessary. other efferent pathways from the transgene-expressing The major limitation, however, for viruses is that the size structure.16,19 of the construct that can be packaged into the virus is Traditionally, the light source used in most optogenetics limited. For AAVs, the construct must be under w5.2 kilo- experiments has been lasers (e.g., ref 19), high-power light- bases,18 whereas for LVs, it must be under 8 kilobases.15 In emitting diodes, or Xenon lamps with the appropriate practical terms, this has limited the promoters that can be wavelength filters (e.g., ref 21). Opsins have a peak 4 LaLumiere sensitivity to a particular wavelength of light. Thus, the reducing the symptoms of PD,26 it has not been clear choice of a light source with the appropriate power and how DBS produces these effects, and various hypotheses wavelength to produce strong activation is critical. have been suggested for its effects. Because DBS would The light-generating source is then connected to a fiber be expected to affect activity of the STN principal neurons optic cable. The end of the fiber can then be inserted into as well as any afferent axons, it is impossible to know the brain region of interest (i.e., either the cell body region precisely how DBS is reducing PD symptoms. or the axon terminal region). Given the conical shape of Recent findings using optogenetics,19 however, have light produced by the fiber and sufficient light being begun to provide some answers for how DBS might amelio- produced by the fiber, the tip of the fiber frequently rate PD symptoms. Using an animal model of PD, based on terminates approximately 0.3-0.5 mm before the center of a unilateral nigrostriatal lesion, researchers used optoge- intended stimulation for optimal activation and minimal netics to isolate specific sets of neurons to investigate this damage to the structure of interest,19 though the area of issue. Across several experiments using either rats with vir- tissue receiving sufficient light for opsin activation ally mediated transduction or transgenic mice, the authors depends, in part, on the amount of light output at fiber tip demonstrated that high-frequency stimulation of ChR2- and numerical aperture of the fiber. In awake behaving expressing STN cells or inhibition of eNpHR-expressing animals, placement of the fiber has typically been accom- STN cells had no effect on PD symptoms. In contrast, plished by insertion of the fiber optic tip through a previ- high-frequency stimulation of the incoming STN afferents, ously implanted intracranial cannula. For behaving as well as of the M1 motor cortical region that projects to animals, it may also be necessary to have an optical the STN, potently reduced the symptoms in this model. commutator between the light source and the fiber optic That finding strongly suggests that DBS-induced alterations that is inserted into the brain to prevent torsion on the fiber in M1-STN pathway activity are responsible for the thera- optic cable, which can damage the fiber. In addition, if peutic effects of DBS in PD patients. In this study, the op- bilateral illumination is desired, a fiber splitter can also togenetic method has provided a clue about the mechanism be placed after the commutator. Our own unpublished find- of action of a different brain stimulation method (i.e., DBS) ings, however, indicate that commutators and splitters and even points to other ways of therapeutic stimulation, produce a significant loss of light, making the power of such as direct cortical stimulation. By resolving the mech- the light source a crucial variable. anisms underlying clinical therapies, such as DBS, optogenetics may provide a method by which such therapies can be improved. Instrumental research findings Recent work has also illustrated how the genetic targeting of optogenetics enables previously impossible The use of optogenetics in research will undoubtedly experiments to be performed. Using a combination of revolutionize many fields. Already, the safe and long-term transgenic mice and AAV, Tsai et al.11 were able to selec- expression of ChR2 in nonhuman primates, as well as tively express ChR2 in dopamine neurons of the ventral optical stimulation of the ChR2-expressing neurons in those tegmental area to investigate the role of these neurons in animals, has been demonstrated.21 This is a finding of appetitive conditioning. Although pharmacologic and elec- significant value as primate experiments typically last trophysiologic techniques have been used to manipulate the much longer than rodent experiments and can model behav- ventral tegmental area to address this issue, the interpreta- iors much closer to humans’ behavior. In the field of tion of results using such methods has always been limited cortical mapping, optogenetics has led to the use of an auto- by the knowledge that the ventral tegmental area contains mated system for generating such maps, permitting consid- a heterogeneity of cell types.27 With selective control of erably faster mapping with fewer of the problems normally the ventral tegmental area dopamine neurons, phasic (50 found with electrical stimulation-based methods.22 Hz), but not tonic (1 Hz), stimulation of these neurons Although it is beyond the scope of this perspective to induced conditioned place preference for the phasic review every study using optogenetics, the following in- stimulation-paired chamber,11 opening up an entire host depth examples demonstrate the breadth of potential uses of questions that can now been directly investigated of the optogenetic technique, the ability of optogenetics through such methods. to provide a new method of interrogation for long- Other experiments have also used optogenetics with standing questions, and the potential clinical impact of functional magnetic resonance imaging (fMRI) to investi- optogenetics. gate the relationship between local neuronal activity and DBS has become increasingly common as a treatment the BOLD signals measured in the fMRI.28 In particular, for a variety of neurologic and psychiatric diseases and its with BOLD signals, it has not been clearly demonstrated use continues to be explored for other diseases.23,24 In whether such signals result from local excitatory activity, particular, DBS has become known as the treatment of inhibitory neurons, modulatory neurons, or even fibers of last resort for cases of PD.25 Although it is apparent that passage. Therefore, the authors used the ability to geneti- DBS of the subthalamic nucleus (STN) is effective at cally isolate neuronal subtypes with optogenetics to Optogenetics and neuroscience 5 selectively stimulate principal neurons of the M1 region Despite these and other unknown potential problems, and produce positive BOLD signals. In contrast, photosti- optogenetics would offer some immediate benefits as mulation of putative GABAergic neurons produced a replacement for treatments such as DBS. As DBS a zone of negative BOLD signals surrounding the local stimulates all cells and fibers within a given area, it is not positive BOLD signals. Moreover, the authors demon- surprising that unwanted side effects are a problem in the strated that the use of optogenetics with fMRI could eluci- use of DBS. Optogenetics would largely avoid this issue date the connections among structures found with BOLD through the selective transduction of only the appropriate signals, in contrast to the difficulties found with traditional neurons as well as the application of the photostimulation electrical stimulation of structures that would be expected to the ideal region, which may not be the same region to activate local neurons, afferent inputs, and fibers of where DBS electrodes are placed. Recent findings also passage and, thereby, confound results. These findings demonstrate the precision with which neurotransmitters can illustrate the clear benefits of optogenetics in terms of be released via optogenetic control of neurons as well as resolving long-standing questions, particularly in regard a lack of change in blood flow and pH resulting from to issues that will have a major therapeutic impact. optogenetic stimulation of neurons,30 suggesting that optogenetically controlled brain stimulation may have other significant advantages over other stimulation techniques. Potential future uses in research and As with the research potential for optogenetics, it is difficult therapeutics to predict the clinical future for this technique. Nonetheless, for any psychiatric disease or neurologic disorder in which The research highlighted above illustrates the types of dysfunction is at least expressed through alterations in clinically relevant questions that optogenetics has already neuronal or network activity, it remains a distinct possibility begun to address. In particular, as noted, such questions that photocontrol over specific neurons or even larger may have previously been intractable but can now be networks will reduce or eliminate the symptoms of the investigated with optogenetic tools. With viral delivery of disorder, even if it does not provide a ‘‘cure’’ for the disease. the opsin genes, control of a genetically targetable cell For example, PD results from the loss of dopamine input to the population occurs through localized transduction of cells as caudate-putamen. Although DBS of the STN provides one well as through the limited expression depending on the potential therapeutic avenue, optogenetics may also provide promoter used. Moreover, as the photostimulation can be a means by which the effects of dopamine itself on basal provided at the cell body region or anywhere along the ganglia activity can be replicated, perhaps through increased axon, including the terminal region, this provides another stimulation of the remaining substantia nigra neurons. In fact, controlled level of spatial resolution. Because these systems recent work has used optogenetic control of rat dopamine work not only in reduced systems but also in awake, neurons in the substrantia nigra to produce dopamine release behaving animals, the questions that can be addressed by in the dorsal striatum in a temporally precise manner,30 this technique are limited by the imagination of the providing the groundwork for potential clinical uses in the researcher. future. Likewise, evidence showing that photostimulation of Can this technique ever be used therapeutically in cortical interneurons can induce gamma oscillations,31,32 human patients? Certainly, there are significant hurdles which are believed to be critical in attention and focusing, that would need to be overcome for this technique’s clinical may provide the foundation for treatments of a variety of disor- use. Recent work has used virally mediated transduction of ders, such as attention deficit hyperactivity disorder, in which ChR2 in nonhuman primate cortex and found safe, stable, attentional problems are a significant symptom. long-term expression of ChR2 with excellent control over Even without direct clinical use of optogenetics, optical neuronal firing, suggesting that, at least between rodents control of neurons in animal models of human diseases and and primates, there are no significant problems in the use disorders will enable a far more precise exploration of the of optogenetics.21 Using viruses to introduce transgenes, parameters and sites of stimulation necessary for symp- although a relatively old concept, has only recently had tomatic relief. The results of optogenetics studies could the appropriate technical advancement to show its promise therefore be applied to less precise techniques such as DBS, for the treatment of disease. In recent trials, AAV-mediated transcranial magnetic stimulation (TMS), or even the more gene therapy has been shown to be at least modestly effec- recent discovery of using ultrasound to stimulate neuronal tive and safe.12,13,28,29 Although AAV is generally consid- activity in deeper brain structures.33 Thus, this technique ered unlikely to produce immune responses, it should be will almost certainly begin having significant clinical noted that it is capable of eliciting an immune response impact through knowledge gained about brain stimulation in the brain and, thus, work must be monitored for such methods already in use. Optogenetics is a technique for safety issues.29 On a smaller scale, the development of controlling neuronal activity that has burst onto the scene very small, implantable yet powerful light sources with within the past 5 years and has already begun to revolu- long-lasting batteries will be a necessity for clinical use tionize our ability to interrogate the nervous system and of optogenetics. potentially to control it for clinical purposes. 6 LaLumiere

Acknowledgments 16. Zhang F, Gradinaru V, Adamantidis AR, et al. Optogenetic interrogation of neural circuits: technology for probing mammalian brain I thank Drs. David Moorman and Elena Vazey for their structures. Nat Protoc 2010;5:439-456. 17. Aravanis AM, Wang LP, Zhang F, et al. An optical neural interface: in helpful comments on a draft of this manuscript. vivo control of rodent motor cortex with integrated fiberoptic and optogenetic technology. J Neural Eng 2007;4:S143-S156. 18. Wu Z, Yang H, Colosi P. Effect of genome size on AAV vector References packaging. Mol Ther 2010;18:80-86. 19. Gradinaru V, Mogri M, Thompson KR, Henderson JM, Deisseroth K. 1. Boyden ES, Zhang F, Bamberg E, Nagel G, Deisseroth K. Millise- Optical deconstruction of parkinsonian neural circuitry. Science 2009; cond-timescale, genetically targeted optical control of neural activity. 324:354-359. Nat Neurosci 2005;8:1263-1268. 20. Lee JH, Durand R, Gradinaru V, et al. Global and local fMRI signals 2. Zhang F, Wang LP, Boyden ES, Deisseroth K. Channelrhodopsin-2 driven by neurons defined optogenetically by type and wiring. Nature and optical control of excitable cells. Nat Methods 2006;3:785-792. 2010;465:788-792. 3. Bi A, Cui J, Ma YP, et al. Ectopic expression of a microbial-type 21. Han X, Qian X, Bernstein JG, et al. Millisecond-timescale optical rhodopsin restores visual responses in mice with photoreceptor control of neural dynamics in the nonhuman primate brain. Neuron degeneration. Neuron 2006;50:23-33. 2009;62:191-198. 4. Zhang F, Wang LP, Brauner M, et al. Multimodal fast optical interro- 22. Ayling OG, Harrison TC, Boyd JD, Goroshkov A, Murphy TH. gation of neural circuitry. Nature 2007;446:633-639. Automated light-based mapping of motor cortex by photoactivation 5. Gradinaru V, Thompson KR, Deisseroth K. eNpHR: a natronomonas of channelrhodopsin-2 transgenic mice. Nat Methods 2009;6: halorhodopsin enhanced for optogenetic applications. Brain Cell 219-224. Biol 2008;36:129-139. 23. Rezai AR, Machado AG, Deogaonkar M, Azmi H, Kubu C, 6. Airan RD, Thompson KR, Fenno LE, Bernstein H, Deisseroth K. Boulis NM. Surgery for movement disorders. Neurosurgery 2008;62 Temporally precise in vivo control of intracellular signalling. Nature (Suppl 2):809-838; discussion 38-39. 2009;458:1025-1029. 24. Lakhan SE, Callaway E. Deep brain stimulation for obsessive- 7. Gunaydin LA, Yizhar O, Berndt A, Sohal VS, Deisseroth K, compulsive disorder and treatment-resistant depression: systematic Hegemann P. Ultrafast optogenetic control. Nat Neurosci 2010;13: review. BMC Res Notes 2010;3:60. 387-392. 25. Remple MS, Sarpong Y, Neimat JS. Frontiers in the surgical treatment 8. Chow BY, Han X, Dobry AS, et al. High-performance genetically of Parkinson’s disease. Expert Rev Neurother 2008;8:897-906. targetable optical neural silencing by light-driven proton pumps. 26. Poewe W. Treatments for Parkinson diseasedpast achievements and Nature 2010;463:98-102. current clinical needs. Neurology 2009;72(7 Suppl):S65-S73. 9. Yamanaka A, Tsunematsu T. New approaches for the study of orexin 27. Kalivas PW. Neurotransmitter regulation of dopamine neurons in the function. J Neuroendocrinol 2010;22:818-824. ventral tegmental area. Brain Res Brain Res Rev 1993;18:75-113. 10. Arenkiel BR, Peca J, Davison IG, et al. In vivo light-induced activa- 28. Eberling JL, Jagust WJ, Christine CW, et al. Results from a phase I tion of neural circuitry in transgenic mice expressing channelrhodop- safety trial of hAADC gene therapy for Parkinson disease. Neurology sin-2. Neuron 2007;54:205-218. 2008;70:1980-1983. 11. Tsai HC, Zhang F, Adamantidis A, et al. Phasic firing in dopaminergic 29. Mandel RJ, Manfredsson FP, Foust KD, et al. Recombinant neurons is sufficient for behavioral conditioning. Science 2009;324: adeno-associated viral vectors as therapeutic agents to treat neurolog- 1080-1084. ical disorders. Mol Ther 2006;13:463-483. 12. Bainbridge JW, Smith AJ, Barker SS, et al. Effect of gene therapy on 30. Bass CE, Grinevich VP, Vance ZB, Sullivan RP, Bonin KD, visual function in Leber’s congenital amaurosis. N Engl J Med 2008; Budygin EA. Optogenetic control of striatal dopamine release in 358:2231-2239. rats. J Neurochem 2010;114:1344-1352. 13. Maguire AM, Simonelli F, Pierce EA, et al. Safety and efficacy of 31. Sohal VS, Zhang F, Yizhar O, Deisseroth K. Parvalbumin neurons and gene transfer for Leber’s congenital amaurosis. N Engl J Med 2008; gamma rhythms enhance cortical circuit performance. Nature 2009; 358:2240-2248. 459:698-702. 14. Kaplitt MG, Feigin A, Tang C, et al. Safety and tolerability of gene 32. Cardin JA, Carlen M, Meletis K, et al. Driving fast-spiking cells therapy with an adeno-associated virus (AAV) borne GAD gene for Par- induces gamma rhythm and controls sensory responses. Nature kinson’s disease: an open label, phase I trial. Lancet 2007;369:2097-2105. 2009;459:663-667. 15. Heilbronn R, Weger S. Viral vectors for gene transfer: current status of 33. Tufail Y, Matyushov A, Baldwin N, et al. Transcranial pulsed gene therapeutics. Handb Exp Pharmacol 2010;197:143-170. ultrasound stimulates intact brain circuits. Neuron 2010;66:681-694.