Skip to main page content
U.S. flag
See More

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2009;4(8):1128-44.
doi: 10.1038/nprot.2009.89. Epub 2009 Jul 16.

Long-term, high-resolution imaging in the mouse neocortex through a chronic cranial window

Anthony Holtmaat  1 , Tobias Bonhoeffer, David K Chow, Jyoti Chuckowree, Vincenzo De Paola, Sonja B Hofer, Mark Hübener, Tara Keck, Graham Knott, Wei-Chung A Lee, Ricardo Mostany, Tom D Mrsic-Flogel, Elly Nedivi, Carlos Portera-Cailliau, Karel Svoboda, Joshua T Trachtenberg, Linda Wilbrecht
Affiliations

Long-term, high-resolution imaging in the mouse neocortex through a chronic cranial window

Anthony Holtmaat et al. Nat Protoc. 2009.

Abstract

To understand the cellular and circuit mechanisms of experience-dependent plasticity, neurons and their synapses need to be studied in the intact brain over extended periods of time. Two-photon excitation laser scanning microscopy (2PLSM), together with expression of fluorescent proteins, enables high-resolution imaging of neuronal structure in vivo. In this protocol we describe a chronic cranial window to obtain optical access to the mouse cerebral cortex for long-term imaging. A small bone flap is replaced with a coverglass, which is permanently sealed in place with dental acrylic, providing a clear imaging window with a large field of view (approximately 0.8-12 mm(2)). The surgical procedure can be completed within approximately 1 h. The preparation allows imaging over time periods of months with arbitrary imaging intervals. The large size of the imaging window facilitates imaging of ongoing structural plasticity of small neuronal structures in mice, with low densities of labeled neurons. The entire dendritic and axonal arbor of individual neurons can be reconstructed.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Schematic view of experimental preparations for in vivo imaging. In the chronic cranial window preparation (left), a piece of the cranial bone (dark gray) is removed (craniotomy) and replaced by a thin cover glass (blue). The dura (grey, arrowhead) remains intact. The exterior is sealed off from the brain surface by dental cement (pink). The craniotomy covers a large surface area (∼3 mm2 in this example; area between dashed lines), which allows imaging of single cells with sparse labeling in the cerebral cortex (GFP-M line in this example, apical dendrite of layer 5 B (L5B) pyramidal cell indicated by arrow). For a thinned skull preparation (right) the cranial bone is usually thinned over a smaller area. The area that becomes transparent for imaging is in the order of 0.15 mm2 (area between dashed lines). This technique is often used for mice with a higher labeling density (YFP-H line in this example). It should be noted that in both preparations the vascular plexus in the trabecular section of the cranial bone (asterisk) is damaged. Imaging is typically performed using ×20, ×40 or ×60 ‘long-working-distance’ water immersion ‘dipping’ objectives with numerical apertures varying from 0.8–0.95. All experiments using animals were carried out under institutional and national guidelines.
Figure 2
Figure 2
In vivo images of pyramidal neurons in the somatosensory cortex of YFP-H and GFP-M transgenic mice. (a) Top view of L2/3, layer 5 B (L5B) and L6 pyramidal neurons in a YFP-H mouse (maximum intensity projection of 160 sections, 5 μm apart, spanning cortical layers 1–5). The inset shows the side view of the same image stack, and the orange box is seen in more detail in c. Individual neurons or dendrites cannot be distinguished. (b) Top view of two L5B pyramidal neurons in a GFP-M mouse (maximum intensity projection of 159 sections, 5 μm apart). The inset shows the side view of the same image stack, and the orange box is seen in more detail in d. Individual neurons and dendrites can clearly be distinguished. See also Supplementary Movie 1. (c) Image of the apical dendritic tufts from the orange box in a (maximum value projection of 12 sections, 1.5 μm apart). Only a few short dendritic regions, likely from different pyramidal cells, are isolated and suitable for analysis of dendritic spines (in between asterisks). (d) Image of an apical dendrite from the orange box in b (maximum intensity projection of 12 sections, 1.5 μm apart). Small synaptic structures, such as axonal boutons (blue arrows) and dendritic spines (orange arrows) can be clearly distinguished. (e–g) Images of dendritic branches in the GFP-M line at different depths in the cortex (best projections of stacks of 4–7 sections, 1.5 μm apart). All images were acquired through a chronic cranial window, with equal excitation intensities (adjusted for increasing depth). Scale bars in a and b represent 10 μm (100 μm for insets). Images were obtained by RM and CP-C. All experiments using animals were carried out under institutional and national guidelines.
Figure 3
Figure 3
Chronic cranial window surgical procedure. (a) The skin and periosteum have been removed. The right temporalis muscle (arrows) has been separated from the bone. The cranial bone has been covered with a thin layer of cyanoacrylate (PROCEDURE Steps 5–7). Dashed lines indicate cranial sutures, bregma (b) and lambda (l). The skull is exposed from the cerebellum (c, caudal) to the olfactory bulb (r, rostral). (b) A thin layer of dental acrylic (pseudo colored in red) has been applied. The area of interest remains uncovered (PROCEDURE Step 7). (c) A circular groove has been drilled (dashed orange line), and the island of cranial bone has been removed. The brain remains covered with buffer. The superficial blood vessels become clearly visible (PROCEDURE Steps 8–10). (d) A circular cover glass (#1 thickness, blue dashed circle) is covering the brain and part of the skull (PROCEDURE Step 11). (e) Dental cement has been applied on top of the skull and part of the cover glass, sealing off the exterior (PROCEDURE Step 12). (f) All of the exposed skull, wound edges and the temporalis muscle (arrow) have been covered with dental cement, and a titanium bar (asterisk) with threaded holes has been attached. (g) Bright field view of the superficial blood vessels through the glass window immediately after the surgery (day 0). (h) The same preparation 14 d later. Note that the surface vasculature patterns remain unchanged, and the center of the window remains transparent. (The window can remain transparent for several months, see Supplementary Fig. 1.) Surgery in this example was performed by VDP. All experiments using animals were carried out under institutional and national guidelines.
Figure 4
Figure 4
Long-term imaging of GFP-expressing pyramidal neurons in the adult neocortex. (a) Bright field image of the surface vasculature, as seen through a chronic cranial window. (b) Higher magnification of the imaged region. Superposed is the fluorescence image of the dendritic tuft of a layer 5 B (L5B) pyramidal cell. (c) The vasculature (displayed in pink) is used to register regions of interest of the dendrite (gray boxes). (d) Time lapse images of a dendritic (upper series) and axonal (lower series) segment, taken from a region of interest in c (orange box). Imaging was started 15 d after the surgery. Time stamps represent days after the first imaging session. Examples of appearing (arrowheads) and disappearing (arrows) spines and axonal boutons are indicated. (e) The entire neuron (imaged in c) has been reconstructed from in vivo images and the perfusion fixed brain. Top view of the apical tuft (above) and side view (below). Images obtained by AH, LW and GK. All experiments using animals were carried out under institutional and national guidelines.
Figure 5
Figure 5
Examples of spine and bouton scoring criteria. (a) Examples of spines that are typically included in the scoring. The threshold is set at 0.4 μm (typically 5 pixels), which corresponds to lengths that span more than 1 resolution unit. (b) Protrusions that do not reach the threshold are ignored, even if they stand out clear from the haze of the dendrite. (c–e) Only laterally emanating protrusions are scored. The dendritic thickening in c can be resolved as a spine in the z axis of the image stack (d and e). However, it should not be included in the analysis, as this can only be done for very big spines (see also Fig. 6d). (f–i) Time lapse images (time = t1–t3). The spine in g is smaller and longer than the spine in f, but the position of its neck (x) and head (x1) are similar to those of the spine in f. It is therefore to be scored continuously as spine 1. The spine in h is similarly shaped as the spine in f and g; however, the position of its neck (x') differs more than 0.5 μm from the original position (x). It can be scored as a new spine (depending on a preset threshold for x – x'). The neck of the spine in i has the same position as the spine in f and g. However, the position of its head (x1';) differs more than 0.5 μm from the original position (x1), and can thus be scored as a new spine (depending on preset threshold for x1–x1';). (j) Terminaux boutons (TBs) scoring thresholds. TBs are scored when their lengths span more than 1 μm and less than 5 μm. In order for a TB to be scored as a loss in time series, its length has to fall below a threshold of 0.4 μm (not shown). (k) Boutons that are located on stalks longer than 5 μm are scored as part of terminal branches. (l) En passant boutons (EPBs) are analyzed based on their intensity relative to the axon shaft. An axonal varicosity has to be three times brighter than the shaft to be scored as an EPB (see ref. 16 for more details). All experiments using animals were carried out under institutional and national guidelines.
Figure 6
Figure 6
Examples of dendritic spines that have been inconsistently scored by different observers. (a) Example of a spine that has disappeared on day 4 (arrow) and re-appears on day 8 (arrowhead) at the same location. This can be scored as a loss and subsequent gain of a spine, but could also be interpreted as a failure to detect a persistent spine on day 4. (b) Examples of confusion between thin spines and axons. Arrow and arrowhead point to spines that are potentially lost and gained, respectively. However, the fluorescence could also represent boutons that are associated with the axon that crosses the dendrite and possibly has extended between day 0 and 8 (see schematic). (c) Spines of various sizes. The smallest spine (#1) is barely brighter than the background (×1.7) and has a signal-to-noise ratio (SNR) of ∼10. Spines 2–6 are increasingly brighter and will be detected with less ambiguity (e.g., spine 6: ×26 brighter than the background with a SNR of ∼160). (d) Example of a stubby spine (arrow) that emanates from the dendrite, both laterally and perpendicular to the image plane. Its lateral projection is less pronounced on day 20, but its head is still visible in a section, 4 μm above the center of the dendritic shaft. This spine therefore still emanates perpendicular to the plane of imaging on day 20. (e) A new protrusion (arrowhead) extends from a stubby spine or thick nodal part of the dendritic shaft (asterisk). A first criterion should determine whether the thickening represents a spine or not, and a second criterion is needed to decide whether the extending protrusion represents a new spine. Time stamps represent days after the first imaging session in all examples. Images obtained by A.H. and L.W. All experiments using animals were carried out under institutional and national guidelines.
Figure 7
Figure 7
Quantification of layer 5 B (L5B) pyramidal cell dendritic spine density and turnover. (a) Example of spines in somatosensory cortex, imaged through a chronic cranial window 1 d after surgery and 20 d later. Examples of some spines that have disappeared (arrows) and appeared (arrowheads) over this period are indicated. (b) Spine densities vary from cell to cell under the cranial window (black (data from A.H., V.D.P., K.S., L.W.), red (data from R.M., C.P.-C.) and blue (data from T.B., S.B.H., M.H., T.K., T.D.M.-F.)) as well as in the naive cortex (green (data from A.H., J.C., G.K.)), and remain relatively stable on an average over several weeks of imaging. (c) Example of spines that have disappeared (arrows) and appeared (arrowheads) in visual cortex, imaged through a chronic cranial window 2 d after surgery and 12 d later. (d) Average survival fractions of spines. Open markers represent survival fractions in experiments where imaging was started immediately after the surgery, solid markers represent experiments that included a waiting period of 2 weeks after the surgery (black points represent data from earlier studies). (e) Turnover ratio (TOR) over 4 d imaging intervals. Imaging was started immediately after the surgery (red, same mice as in b) or after a 2-week waiting period (black and blue, data from different studies). TORs are, on an average, constant over extended periods of time (> 90 d). All experiments using animals were carried out under institutional and national guidelines.
Figure 8
Figure 8
Quantification of L6 terminaux bouton density and turnover. (a) Examples of appearing (arrowheads) and disappearing (arrows) terminaux boutons in the somatosensory cortex, imaged through a chronic cranial window on the day of the surgery, and 14 and 21 d later. (b) Bouton density under the cranial window remains constant over time and is comparable with bouton densities in naive fixed brains (fixed control). (c) Survival fractions of boutons imaged immediately after the surgery are similar to those imaged after a recovery period of 10–14 d. Open markers represent survival fractions in experiments, where imaging was started immediately after the surgery (same mice as in b), solid markers represent experiments that included a waiting period of 2 weeks after the surgery (results from an earlier study). (d) Turnover ratio (TOR) over 4-day imaging intervals. Imaging was started ∼10 d after the surgery (data from different studies). TORs are, on average, constant over long periods of imaging (> 45 d). Data by V.D.P. All experiments using animals were carried out under institutional and national guidelines.
Figure 9
Figure 9
Qualitative assessment of the ultrastructure under a chronic cranial window. Electron micrographs showing the upper region of L1 in the primary somatosensory cortex in a control (a) and an operated (b) mouse 15 d after surgery. The image in b is taken at the position below the center of the cranial window. The basal lamina is indicated (arrowheads) in each image and below this are astrocytic profiles (stars). Both images show the tightly packed arrangement of axons and dendrites in this upper region of cortical layer with no clear differences in the amount of glia or neurons present. Boxed areas (c and d) show typical astrocytic profiles that can be found in the vicinity of synapses (stars) between axonal boutons (b) and dendritic spines (sp). Data obtained by G.K. and J.C. All experiments using animals were carried out under institutional and national guidelines.
Figure 10
Figure 10
Immunohistochemical staining of glial markers under a chronic cranial window. (a) Immunostaining for the astrocyte marker GFAP, at several time points after surgery (time stamp in left lower corners). Operated hemisphere is displayed on the right, control hemisphere on the left. A transient and mild enhanced immunostaining is observed at 2 d and 10 d (GFAP) after surgery. The enhanced immunoreactivity is partly caused by an increase in the number of moderately GFAP-immunopositive astrocytes in the L1 and 2/3, as well as by an increase of GFAP levels in some cells (e.g., arrows). (b) Alternating sections of the same mice as in a, stained for Iba1, a microglia marker. Some microglial profiles are slightly thicker on day 2 after the surgery (arrowheads), but remain essentially unchanged over the course of the experiment. (c,d) Quantification of GFAP (c) and Iba1 (d) immunopositive cells in L1 (black) and L2/3 (red) from a different experiment in a different lab. The number of GFAP immunopositive cells is significantly (*P < 0.05, ANOVA) increased during the first 2 weeks after surgery, but return to control levels afterwards. Iba1 immunoreactivity remains unchanged. Data in a and b were obtained by R.M. and C.P.-C. Data in c and d were obtained by W.-C.A.L. and E.N. Qualitatively all labs observed phenomena that were comparable to the examples presented in this figure. All experiments using animals were carried out under institutional and national guidelines.
See this image and copyright information in PMC

References

    1. Denk W, Strickler JH, Webb WW. Two-photon laser scanning microscopy. Science. 1990;248:73–76. - PubMed
    1. Svoboda K, Yasuda R. Principles of two-photon excitation microscopy and its applications to neuroscience. Neuron. 2006;50:823–839. - PubMed
    1. Lendvai B, Stern E, Chen B, Svoboda K. Experience-dependent plasticity of dendritic spines in the developing rat barrel cortex in vivo. Nature. 2000;404:876–881. - PubMed
    1. Gray NW, Weimer RM, Bureau I, Svoboda K. Rapid redistribution of synaptic PSD-95 in the neocortex in vivo. PloS Biol. 2006;4:2065–2075. - PMC - PubMed
    1. Feng G, et al. Imaging neuronal subsets in transgenic mice expressing multiple spectral variants of GFP. Neuron. 2000;28:41–51. - PubMed

Publication types

Substances