Open Access
Add to BibSonomyAbstract
The ability to image neurons anywhere in the mammalian brain is a major goal of optical microscopy. Here we describe a minimally invasive microendoscopy system for studying the morphology and function of neurons at depth. Utilizing a guide cannula with an ultrathin wall, we demonstrated in vivo two-photon fluorescence imaging of deeply buried nuclei such as the striatum (2.5 mm depth), substantia nigra (4.4 mm depth) and lateral hypothalamus (5.0 mm depth) in mouse brain. We reported, for the first time, the observation of neuronal activity with subcellular resolution in the lateral hypothalamus and substantia nigra of head-fixed awake mice.
© 2015 Optical Society of America
1. Introduction
The ability of fluorescent microscopy to monitor the morphology and activity of neurons at subcellular resolution has made it a powerful tool in neuroscience research. Except in the few model organisms of high transparency (e.g., zebrafish and C. elegans), optical microscopy is limited to superficial depths by light scattering in tissue [1]. With current fluorescent reporters, conventional two-photon fluorescence microscopy cannot image neurons much below the neocortex [2]. To reach structures beyond the first 1 mm from the brain surface, one approach is to remove the tissue overlaying the structure of interest [3–5]. This approach, however, is still limited to the more superficial subcortical structures, because the volume of tissue that needs to be extracted is proportional to image depth to the third power. For deep structures such as the lateral hypothalamus, which is 5 mm from the top of the mouse brain, so much tissue would have to be removed (e.g., using a 0.8 numeric aperture microscope objective, the light cone covers 111 mm3 or ~20% of the entire brain [6]) that the remaining tissue can no longer be observed under physiological conditions.
An alternative approach is microendoscopy incorporating gradient refractive index (GRIN) lenses. GRIN lenses are miniature rod-like lenses with a radial refractive index profile of near parabolic shape [7]. Optically, a GRIN lens behaves similarly to a conventional lens. But with its miniature profile of 1 mm diameter or less, it can be embedded inside the brain to relay the excitation and emission light between a conventional microscope objective and the structure of interest [8–10]. Because the tissue volume that a GRIN lens displaces is linear to the image depth, a GRIN lens is uniquely suitable for imaging neurons in deeply buried nuclei.
GRIN lenses can be directly implanted into the mouse brain; however, lenses are costly and surgical procedures have substantial failure rates. Therefore, a more cost- and time-efficient method is to instead implant a guide cannula, which holds the GRIN lens during imaging sessions. Traditionally, these guide cannulas are made of thin-walled glass tubes sealed at one end by a glass cover slip to allow optical access to the tissue [11, 12]. However, the wall thickness of these glass tubes adds significantly to the overall volume. For example, a guide cannula for a 0.5-mm-diameter GRIN lens can have an outer diameter of 0.84 mm [11]. As a result, the overall tissue to be displaced has a volume 2.8 × as large as that occupied by the GRIN lens alone. Since we would like to remove as little overlaying tissue as possible for deeply buried structures, a more optimized guide cannula design is desired.
In this paper, we describe a minimally invasive microendoscopy system utilizing a 0.5-numeric-aperture (NA), 0.5-mm-diameter GRIN lens. We designed a polyimide guide cannula with a 0.61 mm outer diameter, which introduces a mere 50% increase over the GRIN lens volume (e.g., to reach 5 mm, the displaced brain tissue volume is 1.5 mm3, ~0.3% of the overall mouse brain volume). We used this microendoscopy system in combination with genetically encoded calcium indicators for chronic imaging of neurons and neuronal activity in deep structures such as the lateral hypothalamus, striatum, and substantia nigra of head-fixed awake mice. Given the ease in the manufacturing, assembly, and operation of our microendoscopy system, we expect it to find widespread use in the morphological and functional dissection of neural circuits at depth.
2. Materials and methods
2.1 Mice
All experimental protocols were conducted according to United States National Institutes of Health guidelines for animal research and were approved by the Institutional Animal Care and Use Committee at Janelia Research Campus, Howard Hughes Medical Institute. Two- to six-month-old male mice were used for experiments: C57BL/6J (for lateral hypothalamus imaging), Drd1a-cre (for striatum imaging) and Gad2-cre (for substantia nigra imaging).
2.2 Design of a thin-walled chronically implantable guide cannula
To reduce the size and, consequently, invasiveness of the overall microendoscopy system, we chose thin-walled polyimide tubing (wall thickness: 25 µm) for the guide cannula assembly (Fig. 1(a)). In addition to good thermal stability, mechanical strength, and chemical resistance, polyimide has been shown to be especially compatible with brain tissue [13, 14]. In order to hold 0.5-mm-diameter GRIN lenses, we selected machine-cut polyimide tubes of internal diameter 0.56 ± 0.006 mm, outer diameter 0.61 ± 0.012 mm, and length of 8.4 ± 0.005 mm (MicroLumen) for the body of the cannula. Glass cover slips (No. 1; 0.13 – 0.16 mm thickness) were laser-cut into 0.6 ± 0.025 mm diameter discs (Laser Micromachining Limited). The laser cutting process caused the cross section of the glass disc to be slightly trapezoidal with a taper angle of ~5 – 6 degrees, which facilitated its bonding with one end of the polyimide tube via ultraviolet-cured epoxy (Norland Optical Adhesive 68). The other end of the tube was left open for the insertion and removal of GRIN lenses during experiments and was surrounded by an end piece that provided a large surface area for secure attachment onto the skull with dental cement. Guide cannula assembly was carried out in-house or by Doric Lenses Incorporated.
Fig. 1 Schematics of the minimally invasive microendoscopy system. (a) Design of the guide cannula. Dimensions are in millimeters. (b) A mouse with guide cannula and head-bar implantation. (c) A 0.5-mm-diameter GRIN lens relays the focus of a 0.2-NA objective to a 0.5-NA focus inside a deeply buried nucleus (e.g., lateral hypothalamus). d indicates the distance between the objective focus and the top of the GRIN lens. The brain, guide cannula, and GRIN lens are drawn to scale. Scale bar: 1 mm.
2.3 Characterization of the 0.5-NA, 0.5-mm-diameter GRIN lens
We chose a doublet GRIN lens (NEM-050-25-10-860-DM, GRINTECH GmbH) for deep tissue imaging in mice. With a diameter of 0.5 mm, this lens can easily fit in the guide cannula of inner diameter of 0.56 mm. It has designed working distances of 100 µm on the image side and 250 µm on the sample side. With an overall length of 9.86 mm, the lens is longer than the guide cannula, thus allowing its easy insertion into and removal from the cannula. It is made of two GRIN singlets with NA = 0.5 on the object (brain) side and NA = 0.19 on the microscope side. The internal foci formed by the excitation laser on its way to the brain tissue are located inside the NA = 0.19 part of the lens, thus avoiding the possible damage and autofluorescence background that would occur if a 0.5-NA singlet lens of similar length were used instead [15]. The image NA of 0.5 provides sufficient 3D resolution for functional imaging of neuronal cell bodies.
The GRIN lens was incorporated into a two-photon fluorescence microscope equipped with a 4 × air objective of 0.2 NA (CFI Plan Apochromat Lambda, Nikon Instruments Inc.), which generated the initial focus of the excitation light (of 0.2 NA, to match the 0.19 NA of the GRIN lens on the image side) to be relayed by the GRIN lens to the sample side (Fig. 1(c)). The two-photon fluorescence signal was collected and transported back to the microscope by the GRIN lens and detected with a photomultiplier tube (H7422-40, Hamamatsu Corp.). A Ti:Sapphire femtosecond oscillator (Chameleon Ultra II, Coherent Inc.) tuned to 900 nm was used as the excitation light source for all experiments.
Because GRIN lenses may have residue aberrations [16, 17], we measured the on-axis intrinsic aberration of this GRIN lens. At its design image working distance (i.e., d, the distance between the objective focus and the top of the GRIN lens in Fig. 1(c)) of 100 µm, for 1-µm-diameter fluorescent beads at the center of the imaging field of view, correcting the intrinsic aberration of the GRIN lens doubled the signal (Fig. 2(a), 2(b)). The full widths at half maximum (FWHMs) of the lateral and axial images of the beads were 1.29 µm and 12.4 µm before AO correction, respectively. With AO correction [17, 18], the lateral and axial FWHMs were 1.25 µm and 9.9 µm, corresponding to lateral and axial resolutions of 0.85 µm and 8.6 µm, respectively. Because we intended to image cell bodies of neurons, which are much larger than the 1-µm-diameter beads and whose two-photon fluorescence signals are not as sensitive to aberration [19], adaptive optics was not used for the following experiments.
Fig. 2 Imaging performance of the 0.5-NA, 0.5-mm-diameter GRIN lens. (a) Focal series images (at 0.5 µm steps) of a 1-µm-diameter fluorescence bead before and after AO correction. (b) Lateral and axial signals of the focal series images in (a) measured with and without AO correction. (c) Fluorescence beads imaged over a large FOV at different focal planes by varying the image-space distance d between the air objective focus and GRIN lens. Excitation power: 15 mW. Scale bars: (a) 2 µm and (c) 20 µm.
To image in 3D in vivo, we changed the distance between the air objective and the GRIN lens while keeping the relative positioning between the GRIN lens and the brain constant. By moving the objective focus from 400 µm above to 200 µm below the top surface of the GRIN lens (d in Fig. 1(c) from 400 µm to −200 µm), we moved the focus on the sample side from 160 µm to 340 µm away from the bottom surface of the GRIN lens. The imaging FOV of our GRIN-lens-based microendoscope is limited by the off-axis aberrations [16, 17]. Using a dense bead sample, we found that the FOV size at different working distances did not vary substantially, with beads imaged within a ~200-µm-diameter FOV (Fig. 2(c)).
2.4 Stereotaxic viral delivery and guide cannula implantation
Mice were anesthetized with isoflurane and were placed into a stereotaxic apparatus (David Kopf Instruments). The skull was exposed via a small incision and a small hole (< 1-mm diameter) was drilled for virus injection and guide cannula placement. A beveled needle with a 500 µm diameter was first inserted into and then retracted from the brain to create a path for guide cannula insertion. A pulled glass pipette with a 20 – 40 µm tip diameter was inserted into the brain and injections of AAV2/1-Syn-GCaMP6s-WPRE-SV40 (for lateral hypothalamus) or AAV2/1-Syn-GCaMP6f-WPRE-SV40 (for striatum and substantia nigra) virus [20] were made at coordinates for the lateral hypothalamus (Bregma, –1.4 mm; midline, +0.85 mm; dorsal surface, –5.3 mm and –5.0 mm; 30 nl injection at each depth), striatum (Bregma, +0.75 mm; midline, +1.5 mm; dorsal surface, −2.5 mm), or substantia nigra (Bregma, −3.3 mm; midline, +1.4 mm; dorsal surface: −4.4 mm). A micromanipulator (Narishige) was used to control the injection speed at 30 nl/min [21]. Following virus injection, a holder (Doric Lens) adapted to fit the stereotaxic apparatus was used to insert the guide cannula through the craniotomy to the appropriate target depth. Grip cement (DENTSPLY) was used to anchor the guide cannula to the skull. A titanium or plastic head-bar was then installed around the cannula for head-restraint of the mouse during imaging (Fig. 1(b), 1(c)) as described previously [22]. To prevent mice from compromising the cannula, the open end of the cannula was either covered by a piece of Parafilm and then sealed with a silicon elastomer cap (Kwik-Sil, World Precision Instruments) or protected with a reusable custom cap that covered the end piece (Doric Lens). Of 40 mice, the immediate post-surgery survival rate is 100%, with one mouse euthanized 23 days after surgery. Mice were individually housed typically for 6-8 weeks during which inflammatory response decreased and viral transduction occurred.
2.5 In vivo imaging and data analysis
During the recovery and virus transduction period, mice were acclimated to head-fixation using graded exposure techniques [23]. Briefly, we paired the gradually presented anxiety-provoking stimuli (i.e., head restriction) with the anxiety-competing response (i.e., sweet food reward). Mice were placed under head restraints for 5 min and offered sweet food rewards immediately upon their removal from the restrainers. Multiple daily sessions were performed and the head-fixation duration was gradually incremented to 30-60 min.
Before imaging, mice were secured onto a head-bar holder in either an anesthetized or awake state. The silicon elastomer cap (or custom cap designed by Doric Lens) was removed and a GRIN lens was then placed in the guide cannula. Images of structures of interest were then collected with a two-photon fluorescence microscope. At the end of the imaging session, the GRIN lens was removed from the guide cannula and a new silicon elastomer cap (or the custom cap designed by Doric Lens) was placed to prevent debris from falling into the guide cannula.
Analysis was performed using ImageJ (version 1.47q). For calcium imaging stacks, images were registered using the TurboReg plugin and somata of neurons were identified from the average image of the entire stack as the regions of interest (ROIs). The fluorescence intensity of each ROI across the entire imaging stack was calculated, and from that calculation, the relative fluorescence intensity change, ΔF/F, was calculated.
2.6 Histology
Mice were deeply anesthetized with isoflurane and then euthanized by perfusion with saline followed by 4% paraformaldehyde (PFA) in phosphate buffered saline. Brains were removed and fixed overnight in 4% PFA and then transferred to phosphate buffered saline. Brain sections (thickness: 50 µm) were cut with a Leica VT1200S vibratome, mounted on glass slides using VECTASHIELD mounting medium with DAPI (Vector Laboratories), and coverslipped for imaging. Guide cannula track and neuron images were collected by both widefield (Axiozoom.V16, Carl Zeiss) and confocal microscopy (LSM700, Carl Zeiss). To examine the inflammatory response caused by guide cannula implantation, mice were perfused at 2 and 4 weeks post-surgical procedure and their brains sectioned. Immunostaining for glial markers such as ionized calcium-binding adaptor molecule-1 (IBA1, Wako Chemicals) and glial fibrillary acidic protein (GFAP, Sigma Aldrich) was performed as previously described [24].
3. Results
3.1 Immunohistochemical analysis of reactive glia demonstrates the biocompatibility of chronically implanted polyimide guide cannulas
Invasive surgeries such as virus injection or guide cannula implantation cause inflammatory responses in the brain, during which the neurons do not behave normally and image qualities are low. We analyzed the tissue’s reaction to implanted polyimide guide cannulas by immunostaining with glial markers GFAP and IBA1 in brain sections collected at two or four weeks post-implantation. Two weeks post-implantation, reactive glia formed dense sheaths around the cannula (Fig. 3(a)). However, a remarkable reduction in the inflammatory response was observed at 4 weeks post-implantation (Fig. 3(b)). We quantified this reduction with the ratio of numbers of immunochemically reactive pixels in hemispheres with and without guide cannula insertion. For GFAP, the ratio reduced from 220 × at 2 weeks post-implantation to 9 × at 4 weeks post-implantation; whereas for IBA1, the ratio reduced from 50 × at 2 weeks post-implantation to 6 × at 4 weeks post-implantation. These observations suggest that inflammatory responses triggered by guide cannula implantation have largely dissipated in a time course of ~4 weeks.
Fig. 3 Inflammatory reactions triggered by guide cannula implantation dissipate after ~4 weeks. (a) and (b) Top panels: widefield images of brain sections showing the location of an implanted guide cannula in the right hemisphere above the lateral hypothalamus. Bottom panels: confocal images of GFAP+ and IBA1+ glia (a) two and (b) four weeks after guide cannula implantation. Images from left hemisphere serve as controls of glia populations in intact brain tissue. Black scale bars: 1 mm; white scale bars: 0.1 mm.
3.2 Chronic in vivo morphological imaging of neurons in deep brain nuclei of awake head-fixed mice
With the optical sectioning ability of two-photon fluorescence excitation [25], we obtained three-dimensional image stacks of neurons in deeply buried nuclei in vivo (e.g., neurons in the lateral hypothalamus expressing GCaMP6s, see Visualization 1). At 0.5 NA, the endomicroscope provides enough optical power to resolve subcellular structures such as individual neuronal processes (Fig. 4).
Fig. 4 Chronic in vivo images of neurons from deeply buried nuclei of head-fixed awake mice. (a) Two-photon fluorescence endomicroscopy images of neurons in lateral hypothalamus across 16 days. (b) Two-photon fluorescence endomicroscopy images of neurons in striatum across 36 days. The brain, guide cannula, and GRIN lens were drawn to scale. Black scale bar: 1 mm. White scale bar: 20 µm.
The long-term stability and biocompatibility of these guide cannulas allowed us to chronically monitor neurons in deep brain nuclei such as the lateral hypothalamus (Fig. 4(a)) and striatum (Fig. 4(b)) for weeks. In the lateral hypothalamus, we observed the same population of neurons over a span of 16 days; in striatum, the same neurons were observed for 36 days (the image on day 36 was distorted by slight tilting of the GRIN lens under the microscope). This suggests that our microendoscopy system provides a physiologically stable window for in vivo imaging of neurons in deep brain regions.
3.3. In vivo functional imaging of neurons in deep brain nuclei of head-fixed awake mice
Importantly, the minimal invasiveness of our system also enabled us to monitor neuronal activity in deep brain nuclei in vivo in head-fixed awake mice (Fig. 5). In the mouse lateral hypothalamus (Fig. 5(a)), neurons exhibited no activity when the mouse was anesthetized. Once the mouse awoke, highly synchronized activity was observed across neurons within the entire imaging FOV, as indicated by the calcium transients (ΔF/F) calculated from the fluorescence intensity changes of neurons expressing the calcium indicator GCaMP6s (Fig. 5(b), see Visualization 2). In substantia nigra (Fig. 5(c), 5(d)), neurons had more diverse activity patterns without obvious synchrony. Even though these mice were awake, little in-plane or out-of-plane brain motion was found, with signals from some fluorescent features (structures labeled as ROI 0 in Fig. 5) remaining constant, indicating that the observed fluorescence changes were not caused by brain motion but reflected the neuronal activities of these neurons [26].
Fig. 5 In vivo functional imaging of neuronal activity from deeply buried nuclei of head-fixed awake mice. (a) Left panel: two-photon fluorescence endomicroscopy images of neurons in the lateral hypothalamus; Right panel: regions of interest (ROIs) outline individual neurons. (b) Neuronal activity as measured by the calcium transient ΔF/F of neurons in (a) with the mouse first anesthetized and then awake. (c) Left panel: two-photon fluorescence endomicroscopy images of neurons in substantia nigra; Right panel: ROIs outline individual neurons. (d) Neuronal activity as measured by the calcium transient ΔF/F of neurons in (c) from an awake mouse.
4. Conclusions
In summary, we engineered a minimally invasive microendoscopy system and applied it to in vivo imaging of deeply buried nuclei in the mouse brain. We designed a thin-walled polyimide guide cannula that minimizes tissue removal during its implantation in mouse brain, and allows easy insertion and removal of 0.5-mm-diameter GRIN lenses. It is biocompatible and structurally stable for weeks to months post implantation and has enabled us to chronically monitor morphological and functional properties of neurons in vivo without compromising mouse survival and health.
By using two-photon fluorescence excitation through our microendoscopy system, we imaged individual neurons and neuronal processes expressing GCaMP6 in deeply buried nuclei. High quality images with few motion artifacts were acquired in the lateral hypothalamus, striatum, and substantia nigra of head-fixed awake mice. Calcium transients indicative of action potential firing were also recorded from these neurons. To our knowledge, our results show for the first time neuronal activity with subcellular resolution in these deeply buried structures.
The same thin-walled guide cannula design can be applied to larger GRIN lenses and/or animals (e.g., rat, marmoset, and macaque). Even though we only demonstrated the application of our system for two-photon fluorescence imaging, our system is compatible with other imaging modalities such as single-photon widefield fluorescence imaging [27]. Therefore, we expect widespread adoption of our design for experimental interrogations of brains at depth.
Acknowledgments
This project was supported by the Janelia Visitor Program and funded by the Howard Hughes Medical Institute and NIDA Intramural Research Program. M. E. B. and Y. A. were supported by the NIDA Intramural Research Program, W-C. J., C. W., J. T. D., and N. J. were supported by Howard Hughes Medical Institute. We thank J. Rouchard, K. Morris for mouse breeding and procedures; M. Copeland, A. Hu, B. Shields for histology support; S. Sarsfield for help with figures and comments on the manuscript; Branka Prijovic, Sead Doric, and Jean-Luc Neron from Doric Lenses Inc. for assistance with guide cannula assembly.
References and links
1. N. Ji, “The practical and fundamental limits of optical imaging in mammalian brains,” Neuron 83(6), 1242–1245 (2014). [CrossRef] [PubMed]
2. W. Mittmann, D. J. Wallace, U. Czubayko, J. T. Herb, A. T. Schaefer, L. L. Looger, W. Denk, and J. N. Kerr, “Two-photon calcium imaging of evoked activity from L5 somatosensory neurons in vivo,” Nat. Neurosci. 14(8), 1089–1093 (2011). [CrossRef] [PubMed]
3. A. Mizrahi, J. C. Crowley, E. Shtoyerman, and L. C. Katz, “High-resolution in vivo imaging of hippocampal dendrites and spines,” J. Neurosci. 24(13), 3147–3151 (2004). [CrossRef] [PubMed]
4. D. A. Dombeck, C. D. Harvey, L. Tian, L. L. Looger, and D. W. Tank, “Functional imaging of hippocampal place cells at cellular resolution during virtual navigation,” Nat. Neurosci. 13(11), 1433–1440 (2010). [CrossRef] [PubMed]
5. J. H. Marshel, A. P. Kaye, I. Nauhaus, and E. M. Callaway, “Anterior-posterior direction opponency in the superficial mouse lateral geniculate nucleus,” Neuron 76(4), 713–720 (2012). [CrossRef] [PubMed]
6. N. A. Bock, N. Kovacevic, T. V. Lipina, J. C. Roder, S. L. Ackerman, and R. M. Henkelman, “In vivo magnetic resonance imaging and semiautomated image analysis extend the brain phenotype for cdf/cdf mice,” J. Neurosci. 26(17), 4455–4459 (2006). [CrossRef] [PubMed]
7. C. Gomez-Reino, M. V. Perez, and C. Bao, Gradient-Index Optics (Berlin: Springer-Verlag 2002).
8. J. C. Jung and M. J. Schnitzer, “Multiphoton endoscopy,” Opt. Lett. 28(11), 902–904 (2003). [CrossRef] [PubMed]
9. M. J. Levene, D. A. Dombeck, K. A. Kasischke, R. P. Molloy, and W. W. Webb, “In vivo multiphoton microscopy of deep brain tissue,” J. Neurophysiol. 91(4), 1908–1912 (2004). [CrossRef] [PubMed]
10. J. C. Jung, A. D. Mehta, E. Aksay, R. Stepnoski, and M. J. Schnitzer, “In vivo mammalian brain imaging using one- and two-photon fluorescence microendoscopy,” J. Neurophysiol. 92(5), 3121–3133 (2004). [CrossRef] [PubMed]
11. R. P. J. Barretto, T. H. Ko, J. C. Jung, T. J. Wang, G. Capps, A. C. Waters, Y. Ziv, A. Attardo, L. Recht, and M. J. Schnitzer, “Time-lapse imaging of disease progression in deep brain areas using fluorescence microendoscopy,” Nat. Med. 17(2), 223–228 (2011). [CrossRef] [PubMed]
12. R. P. J. Barretto and M. J. Schnitzer, “In vivo optical microendoscopy for imaging cells lying deep within live tissue,” in Imaging: A Laboratory Manual, R. Yuste, ed. (Cold Spring Harbor Laboratory Press, 2011).
13. P. J. Rousche, D. S. Pellinen, D. P. Pivin Jr, J. C. Williams, R. J. Vetter, and D. R. Kipke, “Flexible polyimide-based intracortical electrode arrays with bioactive capability,” IEEE Trans. Biomed. Eng. 48(3), 361–371 (2001). [CrossRef] [PubMed]
14. Y.-Y. Chen, H.-Y. Lai, S.-H. Lin, C.-W. Cho, W.-H. Chao, C.-H. Liao, S. Tsang, Y.-F. Chen, and S.-Y. Lin, “Design and fabrication of a polyimide-based microelectrode array: Application in neural recording and repeatable electrolytic lesion in rat brain,” J. Neurosci. Methods 182(1), 6–16 (2009). [CrossRef] [PubMed]
15. K. König, A. Ehlers, I. Riemann, S. Schenkl, R. Bückle, and M. Kaatz, “Clinical two-photon microendoscopy,” Microsc. Res. Tech. 70(5), 398–402 (2007). [CrossRef] [PubMed]
16. C. Wang and N. Ji, “Pupil-segmentation-based adaptive optical correction of a high-numerical-aperture gradient refractive index lens for two-photon fluorescence endoscopy,” Opt. Lett. 37(11), 2001–2003 (2012). [CrossRef] [PubMed]
17. C. Wang and N. Ji, “Characterization and improvement of three-dimensional imaging performance of GRIN-lens-based two-photon fluorescence endomicroscopes with adaptive optics,” Opt. Express 21(22), 27142–27154 (2013). [CrossRef] [PubMed]
18. N. Ji, D. E. Milkie, and E. Betzig, “Adaptive optics via pupil segmentation for high-resolution imaging in biological tissues,” Nat. Methods 7(2), 141–147 (2010). [CrossRef] [PubMed]
19. N. Ji, T. R. Sato, and E. Betzig, “Characterization and adaptive optical correction of aberrations during in vivo imaging in the mouse cortex,” Proc. Natl. Acad. Sci. U.S.A. 109(1), 22–27 (2012). [CrossRef] [PubMed]
20. T.-W. Chen, T. J. Wardill, Y. Sun, S. R. Pulver, S. L. Renninger, A. Baohan, E. R. Schreiter, R. A. Kerr, M. B. Orger, V. Jayaraman, L. L. Looger, K. Svoboda, and D. S. Kim, “Ultrasensitive fluorescent proteins for imaging neuronal activity,” Nature 499(7458), 295–300 (2013). [CrossRef] [PubMed]
21. Y. Aponte, D. Atasoy, and S. M. Sternson, “AGRP neurons are sufficient to orchestrate feeding behavior rapidly and without training,” Nat. Neurosci. 14(3), 351–355 (2011). [CrossRef] [PubMed]
22. J. E. Osborne and J. T. Dudman, “RIVETS: A mechanical system for in vivo and in vitro electrophysiology and imaging,” PLoS One 9(2), e89007 (2014). [CrossRef] [PubMed]
23. C. Schwarz, H. Hentschke, S. Butovas, F. Haiss, M. C. Stüttgen, T. V. Gerdjikov, C. G. Bergner, and C. Waiblinger, “The head-fixed behaving rat--procedures and pitfalls,” Somatosens. Mot. Res. 27(4), 131–148 (2010). [CrossRef] [PubMed]
24. A. E. Cardona, E. P. Pioro, M. E. Sasse, V. Kostenko, S. M. Cardona, I. M. Dijkstra, D. Huang, G. Kidd, S. Dombrowski, R. Dutta, J.-C. Lee, D. N. Cook, S. Jung, S. A. Lira, D. R. Littman, and R. M. Ransohoff, “Control of microglial neurotoxicity by the fractalkine receptor,” Nat. Neurosci. 9(7), 917–924 (2006). [CrossRef] [PubMed]
25. W. Denk, J. H. Strickler, and W. W. Webb, “Two-photon laser scanning fluorescence microscopy,” Science 248(4951), 73–76 (1990). [CrossRef] [PubMed]
26. D. A. Dombeck, A. N. Khabbaz, F. Collman, T. L. Adelman, and D. W. Tank, “Imaging large-scale neural activity with cellular resolution in awake, mobile mice,” Neuron 56(1), 43–57 (2007). [CrossRef] [PubMed]
27. K. K. Ghosh, L. D. Burns, E. D. Cocker, A. Nimmerjahn, Y. Ziv, A. E. Gamal, and M. J. Schnitzer, “Miniaturized integration of a fluorescence microscope,” Nat. Methods 8(10), 871–878 (2011). [CrossRef] [PubMed]
