Thinkstock32 OPTICS & PHOTONICS NEWS NOVEMBER 2013 Brain Imaging with Multiphoton Microscopy Ke Wang, Nicholas G. Horton and Chris Xu As scientists seek to unravel the mysteries of the brain, they will need to delve deeper than ever before in order to image individual neurons and their processes. Multiphoton microscopy is a promising new technology for getting there. NOVEMBER 2013 OPTICS & PHOTONICS NEWS 33 he brain initiates everything we do and pro- Optical imaging is noninvasive and pro- cesses all that we experience. Yet the mystery vides the high spatial resolution necessary of the brain rivals that of the universe. In fact, to resolve individual neurons and neuronal we probably know more about the universe processes. However, acquiring images through than we do about our own brains, which significant depths of the brain is no easy task Tconsist of approximately 100 billion neurons since brain tissue is extremely heterogeneous, and 1 trillion support cells. resulting in strong scattering by the various Elucidating how the brain works is a grand tissue components. Optical imaging technol- challenge within science. It will undoubtedly ogy will be essential in addressing these help us to better understand neurological challenges, and it will feature prominently diseases such as Alzheimer’s and Parkinson’s. in U.S. President Obama’s recently announced Research in the last several decades has pro- BRAIN initiative (see sidebar on facing page). Multiphoton microscopy (MPM), which was demonstrated more than two decades Femtosecond lasers are usually ago, has dramatically extended the depth required for multi-photon microscopy penetration of high-resolution optical imaging. It has been a game-changer for neuroscience. to enhance the nonlinear excitation When combined with genetically engineered probability and hence the signal level. fluorescent probes, MPM enables the non- invasive, dynamic measurement of neurons in their native environment. (See, for example, the vided scientists with a good understanding of review paper by J.N.D. Kerr and W. Denk.) It is how neurons work, particularly at the levels of well positioned to play a major role in expand- a single neuron or a small number of neurons. ing our understanding of brain function. With the help of magnetic resonance imaging (MRI), we also have some knowledge MPM for deep tissue imaging of brain activity over a large volume at the time In MPM, image contrast is generated from scale of seconds to minutes. What we now need nonlinearly excited fluorescence or harmonic is technology to bridge the large gap in our generation. For fluorescence, a fluorophore understanding between microscopic interac- is excited to an upper-energy level by tions at the neuronal level and the macroscopic the simultaneous absorption of multiple structures that perform complex computations. photons, and it subsequently emits a single Highly excited energy level Virtual energy level Simultaneous Subsequent emission of third emission of single harmonic fluorescence photon Excited fluorescence Harmonic generation Three-photon microscopy In three-photon microscopy, three photons are converted to a fluorescence photon or third harmonic photon. Compared with single photon excited fluorescence, multiphoton excited fluorescence is confined to the focus only due to the nonlinear excitation and enables 3-D sectioning in imaging. 34 OPTICS & PHOTONICS NEWS NOVEMBER 2013 1047-6938/13/11/46/8-$15.00 ©OSA fluorescence photon. The fluores- increased imaging depth. Absorption cence lifetime is typically on the is easy to determine, as it follows order of nanoseconds. Beer’s law, which relates it to the For harmonic generation, on properties of the material through the other hand, only virtual energy which the light travels. levels are involved, and emission is However, scattering is the The BRAIN practically instantaneous. Femto- dominant factor in reducing the second lasers are usually required excitation intensity at the focus. It Initiative for MPM to enhance the nonlinear causes photons to deviate from their The work summarized in this article excitation probability and hence the ballistic path; deep within tissue, highlights the crucial role that optics signal level. Compared to one-photon these scattered photons have essen- and photonics technologies play in neurological research—an important confocal microscopy, the deep tissue tially zero probability of reaching current policy priority. imaging capability of MPM mainly the focal volume. Thus, only the The Brain Research through Ad- derives from nonlinear and longer- so-called ballistic photons can reach vancing Innovative Neurotechnologies wavelength excitation. the geometrical focus. (BRAIN) Initiative is part of U.S. Presi- First, nonlinear excitation If the power deposited on the dent Obama’s Fiscal Year 2014 budget confines the signal to the focal region sample surface is below the tissue proposal. It is aimed at helping scientists find new ways to treat, cure and possibly (i.e., the excitation confinement), damage threshold, the signal decay prevent brain diseases. which enables 3-D sectioning with- from absorption and scattering If approved by Congress, BRAIN would out using a confocal pinhole in the can be compensated by raising the include over $100 million in investments, signal path. Thus, with MPM, one input power exponentially as depth with approximately: can still efficiently collect signals increases. However, there is a c $50 million allocated to generated deep within scattering tis- fundamental limit that determines the Defense Advanced Research sue. Second, a longer wavelength is the maximum imaging depth for Projects Agency. used in MPM to excite fluorophores 2PM, which was shown by Theer et c $40 million to the National because the energies of two or three al. in 2003. The exponential decay Institutes of Health. photons are combined to generate of the excitation intensity at the the molecular transition. The longer focus when imaging deep into a c $20 million to the National excitation wavelength significantly scattering tissue causes nonlinear Science Foundation. reduces the effect of tissue scatter- excitation to be no longer confined A recent panel event at OSA’s annual ing. The combined effect of a longer to the focal volume. meeting described some of the exciting excitation wavelength and efficient A quantitative measure of the 3-D work being done in optics and photonics signal collection makes MPM the excitation confinement is the signal- to help the BRAIN initiative to meet its goals to: method of choice for imaging the to-background ratio (SBR), which activity of individual neurons deep is simply the ratio of the amount of c UNDERSTAND how brain activity inside an intact brain. excitation within the focal volume to leads to perception, decision-making Researchers have used two-pho- the amount outside. When this SBR and action. ton microscopy (2PM) extensively for approaches unity, the signal gener- c DEVELOP new imaging technolo- deep imaging in animal studies that ated at the focus is overwhelmed by gies and understand how information delve into the mouse brain. Recently, the background; hence, no useful is stored and processed in neural a record imaging depth of 1.6 mm feature can be resolved at the focus. networks. was attained in the cortical tissue, Indeed, it is the SBR, rather than c PROVIDE the knowledge for ad- or gray matter, of a mouse brain the decreasing signal strength, that dressing debilitating conditions. in vivo (D. Kobat et al., 2011). The ultimately limits the maximum c PRODUCE a sophisticated under- question then arises: What prevents penetration depth in 2PM. For standing of the brain, from genes to deeper imaging? example, this depth limit for 2PM in neuronal circuits to behavior. Tissue absorption and scattering the mouse neocortex at 775-nm and For more information, visit www.white- lead to the exponential decay of 1,280-nm excitation is about 700 and house.gov/infographics/brain-initiative. excitation light at the focus with 1,600 µm, respectively. NOVEMBER 2013 OPTICS & PHOTONICS NEWS 35 1,280 nm Depth [µm] 775 nm 0 decays approximately as 1/z2 and 1/z4 for 2PM and 3PM, respectively. 3PM has a much better signal confinement than 2PM because it suppresses the 200 out-of-focus background. This shows that, for the same imaging depth, 3PM has a higher SBR and 400 can be used to image deeper than 2PM. Indeed, within an imaging depth of 2.5 mm in brain tissue, SBR does not appear to limit 3PM. 600 100 µm Excitation wavelength for deep-tissue imaging 800 As we stated earlier, attenuation of excitation leads to SBR degradation in deep-tissue imaging. The smaller the attenuation, the greater the imaging depth that 1,000 can be obtained before SBR reaches unity. What wavelength attenuates least in brain tissue? Two-photon microscopy depth limit A previous 2PM experiment of in vivo brain The signal-to-background ratio limits the maximum penetration tissue shows that a 1,280-nm excitation wavelength depth in 2PM. In the mouse neocortex, the depth limit at 775-nm can penetrate deeper than a 775-nm excitation and 1,280-nm excitation is about 700 and 1,600 µm, respectively. wavelength (D. Kobat et al., 2011). Excitation attenu- Opt. Express 17, 13354–64 (2009). ation is characterized by the effective attenuation –1 length (le=(1/ la +1/ ls) ), which factors both the tissue 3PM for deeper imaging absorption length (la) and scattering length (ls). At 1,280 nm (285 μm), l is approximately twice what it The depth limit set by the SBR strongly suggests that e is at 775 nm (131 μm) due to the reduced scattering at deeper brain imaging requires tighter excitation confine- the longer wavelength. ment. Three-photon microscopy (3PM) was demonstrated At first glance, it seems that longer excitation in the mid-1990s, mainly to extend the spectral range of wavelengths will always correlate to greater imag- existing lasers.