17 Multiscale correlative imaging of the brain

,


Introduction
Modern neurobiology is striving to draw a molecular picture of the processes that determine neuronal plasticity.Since the works of Golgi and Ramon y Cajal, neuroscientists took advantage of light microscopy to observe the brain at many levels.The technologies employed in biophotonics have been constantly evolving over the past fifteen years offering always-new experimental capabilities to inspect the nervous system across multiple scales, from whole brain anatomical reconstructions to ultrastructural imaging of synaptic components, and allowing a constant improvement of sensitivity and resolution.
In this chapter, we will show different light microscopy techniques tackling different aspects of brain anatomy and functionality.A "connectomic" perspective is first described, where neuronal projections are visualized through the entire brain with light sheet microscopy.Advancements in nonlinear imaging led to time-lapse microscopy studies in different brain areas of resting and behaving mice, addressing both structural and functional rewiring.This chapter surveys recent innovations in optical imaging techniques of particular interest to neuroscientists.

Brain anatomy
Mapping neuronal connections across the whole brain (the so-called 'connectome' [1]) is a mandatory step to afford a clearer view of the brain itself.However, the structure of the brain challenges imaging techniques to reconstruct macroscopic volumes with microscopic resolution.Indeed, very thin neuronal processes (dendrites and axons, about 1 µm or less in diameter) usually extend for much larger distances, even the whole encephalon in the case of axons [2].
In the last years, a number of optical methods have been implemented to reconstruct large specimens with micron-scale resolution.Combined with different (sparse) labeling methods, such as Golgi staining [3], transgenic expression of fluorescent proteins [4], dye injection [5], viral transfection [6], and whole-mount immunohistochemistry [7], high-throughput optical microscopy has been exploited to reconstruct fine brain anatomy from the mesoscale to the microscale (Fig. 17.1).In the following, we will briefly review three approaches: serial two-photon sectioning (STP), micro-optical sectioning tomography (MOST), and light sheet microscopy (LSM).

Serial two-photon sectioning
STP systems are standard laser-scanning two-photon fluorescence microscopes coupled with automated translation stages and a vibrating blade microtome, which cuts away thin layers of tissue once they have been imaged (Fig. 17.1 (a)) [8].Two-photon excitation allows imaging below the surface, thus minimizing cutting artefacts.The penetration of two-photon imaging inside aldehyde-fixed mouse brain tissue is lim-ited in practice to about 100 µm; as a consequence, only a single optical section is acquired by STP between consecutive slicing steps (typically at 50 µm depth [8]).This results in poor axial sampling (about 1 µm every 50, discarding about 95 % of the volume), since physical slices cannot be thinner than several tens of microns.Although such sparseness prevents full volumetric imaging, on the plus side it allows imaging an entire mouse brain in a day (a quite short time for a point-scanning microscope) and keeps the size of acquired datasets within reasonable limits (around 100 gigabytes per mouse brain [9]).This method is also very reproducible and scalable, and to date is the only one which has been used for large-scale studies like the mesoscale connectivity atlas realized by the Allen Institute for Brain Sciences, where 469 mouse brains have been scanned [10].

Micro-optical sectioning tomography
MOST exploits a serial sectioning approach similar to STP, but with significant differences in sample preparation.Indeed, the tissue is embedded in a hard resin, allowing cutting very thin slices (about 1 µm) without significant slicing artefacts [11].
The sparse sampling of STP can therefore be overcome in this case.In typical applications, a single line on the sliced section is imaged continuously, directly on the knife edge (Fig. 17.1 (b)) -that's why this technique is also known as knife-edge scanning microscopy (KESM [12]).Image can be acquired using a linear charge-coupled device (CCD) camera [11] or a photomultiplier tube (PMT) in combination with fast acoustooptic scanners [13,14].MOST can produce high-resolution and high-contrast images of the whole mouse brain, on which it is possible to trace single axons throughout the entire encephalon (Fig. 17.2 (a)).The main current limitation of this method is the low rates of image acquisition, as about 10 days are required to image a whole mouse brain.

Light sheet microscopy
The fastest optical imaging method for whole brain imaging is probably light sheet microscopy, which provides micron-scale reconstruction of the entire murine encephalon in hours to few (< 3) days [15,16].Indeed, planar illumination (the 'light sheet') confines excitation to the focal plane of detection optics, allowing optical sectioning in a wide-field detection scheme (Fig. 17.1 (c)) [17].This technique is also called selective (or single) plane illumination microscopy (SPIM) [18], or ultramicroscopy [15].To be applied in macroscopic specimens -such as whole mouse brains -LSM needs to be coupled to chemical clearing of the tissue [19].Classic methods to clear the tissue are based on the substitution of water with high-refractive index (≈ 1.56) or- ganic solvents and render the brain very transparent, but no objective lens suitable for immersion in these clearing agents is currently available.Air objectives are commonly used in this case, but the refractive index mismatch introduces strong spherical aberrations [20].
Objective lenses are commercially available for immersion in moderate refractive indices (≈ 1.45), which is the one obtained in CLARITY [7], a clearing protocol which results in highly transparent brains and good fluorescence preservation.In this method, the sample is hybridized into a hydrogel, allowing lipids removal while leaving proteins and nucleic acids in place.This approach greatly increases sample porosity, allowing exogenous labeling over macroscopic portions of tissue, paving the way to anatomical mapping of human tissue.However good the transparency of the tissue, some residual scattering is always present, leading to out-of-focus contributions and thus image blur.To recover image quality, different strategies can be used, like structured illumination [21,22], and confocal line detection [16,23,24] (Fig. 17

Structural plasticity of cortical neurons: from an historical perspective to recent advances in fluorescence imaging in vivo
Large-scale neuroanatomical data are limited to one time point, and thus need to be complemented with studies addressing the temporal evolution of plastic events in live brain.Fluorescence microscopy brought to light neuronal plasticity and its dynamic timeline.The term "structural plasticity" refers to alteration in anatomical connectivity through modifications of synaptic patterns or densities and reshaping of axons and dendrites [25].Structural plasticity plays a major role during development, when new connections are continuously established and refined [26][27][28][29][30], and occurs consistently during adulthood, e.g. as a result of experience and learning [31][32][33].Two-photon fluorescence microscopy (TPFM), by tracking complex dynamics inside live brain, allows the investigation of structural plasticity at the cellular level in vivo [33][34][35][36][37][38].In combination with transgenic mice expressing fluorescent proteins in selected neuronal populations, in vivo TPFM showed events of structural rewiring at synaptic level in the adult brain.Long-term in vivo optical imaging revealed temporal details of dynamic remodeling in the adult brain cortex [26,31,39], olfactory bulbs [40][41][42], and cerebellum [43,44].Time-lapse TPFM of cortical dendrites proved the direct correlation between spine turnover and experience-dependent plasticity [45] in different experimental paradigms, like enriched environments [46,47], long-term stimulation [48,49] and deprivation [30].These studies showed how spine plasticity contributes to experience-dependent rewiring of cortical circuits.Fig. 17  New imaging configurations, by exploiting implanted endoscopes [50] or microprisms [51], three-photon excitation [52] or by means of regenerative amplifiers [53], increased the depth of the imaging area to deep circuits like the hippocampus.Miniaturized portable microscopes [54], fiber optics [50,55] or head-restrained configurations [56] are currently used for imaging in awake and behaving animals.
TPFM recently provided useful insights into the morphological rearrangements of neurons and glial cells involved in recovery from neuronal injury [57][58][59].Technical issues so far limited the investigation of the biological mechanisms of the reactive plasticity after injury in the adult CNS [60].The main obstacle concerns the targeted manipulation of CNS neurons.The spatial localization of multiphoton absorption scales down lesions dimensions to the sub-micrometer scale.Indeed, the intrinsic confinement of nonlinear excitation to the focal volume can be exploited to perform localized targeted photochemistry and photodamage.Laser nanosurgery has been applied in various experimental paradigms, from the investigation of the biological function of subcellular compartments in living cells [61][62][63], to the study of neuronal plasticity of adult mice in vivo [64][65][66][67].A recent TPFM study addressed the re-growing potential of adult cerebellar axons [66]; they showed that the laser-induced ablation of a single axonal branch may elicit axonal sprouting and synaptic remodeling in the surviving portion of the axon (Fig. 17.4 (a)).By combining light and electron microscopy, this study disclosed the formation of new synaptic contacts on the sprouted branch.An analogous study showed the regenerative capabilities of laser axotomized pyramidal neurons in the somatosensory cortex [64] (Fig. 17.4 (b)).Laser-induced lesions have been used to study blood brain barrier disruption [68] and microglial migration towards the lesion site, thus confirming their role in the recovery process [69,70].Pulsed laser irradiation on single blood vessels [71] provides a method to create targeted highly confined strokes mimicking cerebral microvascular diseases [72].The experimental paradigm: immediately after a focal lesion in the retina, the LPZ is unresponsive to visual stimulation, as measured by intrinsic signal optical imaging (bottom, day 0).The size of the unresponsive region decreases with time.Scale bar, 700 µm.(h) Spine density in the visual cortex remains unchanged after a small retinal lesion.(i) Dendritic spines near or in the LPZ display high turnover rates; in the center of the LPZ almost the entire spine population turns over.(j) In the LPZs of small lesions, the fraction of new persistent spines is increased several fold.Reproduced with permission from [25].
TPF imaging and laser nanodissection demonstrated to be a fine tool to dissect at high resolution the dynamics of many neurodegenerative pathologies.

Functional imaging and stimulation of neural circuits
Advantages in optical microscopy in combination with calcium indicator and voltagesensitive dyes (VSDs) allow registrations of neuron activity across a population of cells, opening promising prospectives in exploring brain functionality [50,73,74].Further, neurotransmitters uncaging and neuroengineering tools based on microbial opsins provide optical control over the activity of specific populations of neurons.If combined, light can be used to probe and manipulate the electrical activity of neurons singularly, with high spatial-temporal resolution.Here, we will report the state of the art in detecting and manipulating the activity of microcircuits of neurons.

Optical recording of neuronal activity
The first challenge to address for imaging neural activity is the loading of calcium indicators or VSDs into neurons.Fig. 17.5 (a) illustrates the two approaches for dye loading of neurons based on organic and genetically encoded functional probes.The upper panels depict the "acute" network loading: many neurons are labeled simultaneously by pressure injection loading (left panel), by loading with dextran-conjugated dye (middle panel), and by bulk electroporation (right panel).Due to their electrostatic charge, calcium indicators are unable to cross lipid membranes and the use of physical or chemical methods is needed [75].For example, the esterification of the carboxylic groups with an acetoxymethyl (AM) group makes the dye neutral, thus capable of crossing the cell membrane.Besides AM calcium dyes, dextran-conjugated calcium indicators can also be employed for network labeling [76].Electroporation can be alternatively used to allow the membrane passage of charged calcium probes [77].Using similar approaches, bulk loading of VSDs can be achieved by incubating the tissue or using pressure injection [78,79].Further, in recent years, genetically encoded calcium probes and voltage sensitive proteins have become a widely used tool in neuroscience [80][81][82][83][84].There are different possibilities of expressing genetically encoded probes in neurons.The lower row of Fig. 17 Viral constructs can be used to label specific brain areas using stereotaxic injection [85].In utero electroporation of DNA plasmids encoding for the genetically encoded indicator produces relatively sparse labeling.Finally, generating transgenic mice expressing genetically encoded fluorescent proteins enables long-lasting monitoring of neuronal activities at the cellular, circuit, and system level to address questions of nervous system development and maintenance, learning, and memory.
These labeling methods can be exploited to probe cell activity across a population of neurons in an intact circuit and even in vivo.For example, by incubating an acute brain slice with a novel fluorinated VSDs, single spontaneous action potentials ▸ Fig. 17.5: Functional imaging of neuronal circuits with cellular resolution.(A) Dye-loading and genetic-encoded indicators approaches.Reproduced with permission from [75].(B) Real-time multicellular action potential recording by random access two-photon microscopy.Two-photon fluorescence image of a parasagittal acute cerebellar slice stained with voltage-sensitive dye (VSD) by pressure injection.The multiunit optical recording was carried out from the lines drawn (red) on the five Purkinje cells (PCs).The electrical activity (cell-attached recording) of PC1 was also monitored.Reproduced with permission from [86].(C) Calcium imaging of place cells in the CA1 hippocampal region of mice which are placed on a spherical treadmill.Calcium traces are shown in black.Red traces indicate significant calcium transients.In parallel, the position of the mouse along the virtual linear track is recorded.Reward times are shown at the bottom.Reproduced with permission from [88].
were recorded from multiple neurons using a random access multiphoton microscope [86,87] (Fig. 17.5 (b)).Advanced optical methods can also be applied for imaging the activity of neurons with cellular resolution in behaving mice [56,88,89] during virtual navigation.Fig. 17.5 (c) shows an example of two-photon images of neuron cell bodies in stratum pyramidale of CA1 labeled with the genetically encoded calcium indicator, performed while the mouse is running along a virtual linear track.The experimental setup the authors used, consisting of a spherical treadmill, a virtual reality apparatus and a custom-made two-photon microscope, is displayed in the left panel.
Besides monitoring neuronal activity in behaving animals, an elegant alternative to dissect neuronal circuitry implies controlling network firing.

Optical stimulation of neurons
Caged compounds are widely used in neuroscience enabling the optical manipulation of neuronal circuits [90,91] (Fig. 17.6 (a)).Glutamate can be uncaged by focusing light at a specific position in the brain slice, while in the meantime the electrical responses from a selected neuron can be monitored.For example, by scanning the uncaging beam in different position, one can map the territories that generate excitatory or inhibitory responses in the recorded cell.
The capability of stimulating neurons with light has been recently improved by new neuroengineering tools, channelrhodopsin-2 (ChR2) and halorhodopsin (NpHR) [92,93].ChR2 is a monovalent cation channel that allows cations to flow into the cell following illumination to blue light, while NpHR is a chloride pump that activates upon exposure with yellow light (Fig. 17.6).The use of ChR2 in neurons has been applied to a number of investigations on circuit mapping [94], memory storage [95], and Parkinson's disease [96].A fiber optic-based system has also been developed for modulating the activity of ChR2-and NpHR-expressing neurons of freely moving animals [97].If combined with nonlinear excitation, optogenetic tools benefit from increased control of neuronal activation within the intact tissue, even at single-cell level.However, standard ChR2 has low channel conductance and displays fast kinetics; there-  fore, two-photon excitation of a neuron requires complex stimulation strategies [98].The optimization of a new red-shifted chimeric opsin that displays slower off-kinetics and larger photocurrents has been performed and applied for single neuron stimulation in in vivo preparation [99].Recently, simultaneous two-photon optogenetic activation and calcium imaging by coexpression of a red-shifted opsin and a genetically encoded calcium indicator have been performed enabling high-throughput and long-term optical interrogation of specific neural circuits with single-cell and singlespike resolution in vivo [100].

Correlative microscopy 17.5.1 Understanding brain machinery requires multilevel investigation
The structure of the brain encompasses several orders of magnitude, ranging from synapses to neurons, from cortical columns to the whole encephalon.Consequently, neuronal dynamics occurs across different timescales, from the milliseconds of action potential propagation to the continuous remodeling of the brain, which lasts the entire lifespan.Structural and functional data can be collected across all these scales by using a large repertoire of experimental tools (Fig. 17  Minimally invasive neuroimaging methods can provide whole-brain data also in living humans, albeit at low resolution (no more than 100 µm).For instance, functional Magnetic Resonance Imaging (fMRI) allows inferring neuronal activity from local blood oxygenation levels, based on the different magnetic responses of oxygenated and deoxygenated hemoglobin [101][102][103].On the other hand, diffusion Magnetic Resonance Imaging (dMRI) can be used to reconstruct axonal fiber tracts throughout the encephalon, exploiting the anisotropic diffusion of water in the presence of lipid-rich myelin sheaths [101,104,105].
Finer structural details on smaller -though still macroscopic -samples can be explored with high-throughput light microscopy (see Section 17.2).However, given the high density of neuronal processes (which can be separated by just few nm), standard optical methods have to be coupled with sparse fluorescent labeling; spectral multiplexing in Brainbow methods might overcome this limitation [106].Dense neuronal circuits can be fully disentangled using electron microscopy (EM) [107].However, the EM staining is usually nonspecific, and data acquisition rates are very slow, preventing (up to now) its use on the whole mouse brain.EM automated volumetric imaging with resolution of tens of nm can be afforded with approaches based on tissue slicing, like Serial Block-face Scanning Electron Microscopy (SBSEM) [108] -where a scanning electron microscope is coupled with automatic slicing and removal of the most superficial layer after imaging -or Automatic Tape-Collecting Lathe UltraMicrotome (ATLUM) [109] -where ultrathin sample slices are automatically collected on a tape for subsequent high-throughput imaging with transmission electron microscopy.Higher resolution (near-isotropic voxels of ≈ 4 nm), albeit at slower speed, can be achieved with Focused-Ion Beam Scanning Electron Microscopy [110] (FIBSEM).The nm realm can also be explored with optical super-resolution techniques like Photo-Activated Localization Microscopy (PALM) [111] and Stimulated Emission Depletion (STED) [112].
Optical methods such as two-photon fluorescence microscopy can also provide valuable functional data with cellular and subcellular resolution over spatial scales of few hundreds of µm (see Section 17.4).The same approaches can also be used to investigate neuronal reshaping in vivo over weeks to months (see Section 17.3).
The vast collection of imaging techniques developed in the last decades allows researchers to investigate many different aspects of brain architecture and activity with unprecedented detail.Nonetheless, we still lack a comprehensive description of brain machinery across all its organizational levels.A promising approach in this sense is to combine multiple imaging techniques, each one providing different information on different scales.This correlative framework can overcome intrinsic limitations of single methods.

Correlative imaging overcomes the limitation of single techniques
The challenge of bridging the gap between post mortem microscopic and in vivo macroscopic worlds can be tackled by novel approaches that, by combining the spatio-temporal resolution of complementary techniques, cross several orders of magnitudes.The following section shows examples of correlative measures that linked different temporal/spatial scales by exploiting the resolution of different techniques, from whole brain fMRI to sub-synaptic imaging with FIBSEM.
Different neuroimaging techniques provide functional connectivity data over different spatial and temporal scales.fMRI, by estimating brain activation based on the alterations in blood oxygenation levels (BOLD contrast), enables non-invasive monitoring of activity in healthy and diseased brains with sub-millimeter spatial resolution in humans and animals [101,113,114].The causal link between BOLD signals and the neural activity has not been well described yet [103].Helmchen et al. demonstrated the possibility of combining fluorescence measures of brain activity with fMRI.By using an optical fiber compatible with the fMRI scanner, the authors could perform BOLD fMRI and simultaneous recording of calcium activity (Fig. 17.8 (a)).This correlative approach allowed defining features of the complex fMRI BOLD signals in terms of neuronal and glial activation.This hybrid method could further disclose crucial features of neurovascular coupling.
TPFM longitudinal studies on the structural plasticity of selected neuronal populations in vivo proved otherwise inaccessible data on micro-connectivity reshaping in different conditions [25,32,73,102,104].These dynamic measures on long timescales can only access a limited portion of the brain (hundredths of microns).By combining TPFM with ex vivo imaging techniques like light sheet microscopy that can explore large volumes it is possible to reframe the in vivo visualized neurons within longrange connectivity settings.In detail, neuroanatomical tracing of the neuron previously observed in vivo allows associating the dynamic fingerprint of a neuron to its connectivity partners.Silvestri et al. used the blood vessel pattern as internal reference to retrieve a neuron whose dendritic arbor was previously imaged in a live mouse using TPFM, and traced the entire neuron from confocal light sheet images [115] (Fig. 17.8 (b)).This correlative two-photon and light sheet microscopy approach is a useful tool to contextualize in vivo long-term imaging within a wider neuroanatomical framework.
Linking data on brain functionality with structural connectivity details at high resolution (nm) gives answers to otherwise impenetrable neurobiological questions.Briggman et al. combined in vivo calcium imaging in the intact retina and SBSEM-based reconstruction of the circuitry in the same piece of tissue to explain the behavior of retinal ganglion cells (DSGCs) [116] (Fig. 17.8 (c)).The authors directly correlated neuronal wiring asymmetry with the computation of direction selectivity.High-resolution 3D maps of neuronal connectivity will soon be a "conditio sine qua non" for the correlation of functionality and structure.The correlation of in vivo imaging of neurons and their synaptic connectivity with electron microscopy allows combining dynamic and ultrastructural information [37,45].The appearance of FIBSEM significantly increased the level of automation, reliability and acquisition speed of EM [110].Recent studies provided evidence of the benefits of correlating in vivo TPFM and FIBSEM for targeting synapses at the micro-and ultrastructural level [117][118][119].Recently, Allegra Mascaro et al. [66] and Canty et al. [64] combined two-photon imaging in vivo with FIBSEM to study the possible formation of synaptic contacts by axonal branches regenerated after injury (Fig. 17.8 (d)).EM images of the distribution of synaptic vesicles and post-synaptic densities allowed deducing the structural and functional relationship between the regenerated axon and post-synaptic dendrites.

Fusing multiple levels of investigation might boost our understanding of the brain
The correlative methods presented here showed fundamental insight into different spatio-temporal scales of brain functioning.This approach can be promisingly expanded towards a unified method covering most perspectives.In vivo imaging by noninvasive fMRI could be the starting point of a long pipeline that investigates the longterm plasticity of small populations of neurons through in vivo TPF imaging.Once functional and structural data have been obtained in vivo at the small-circuitry level, the same sample shall be processed with LSM for ex vivo long-range anatomical analysis.This contextualization into a wider framework is refined up to the synaptic scale when imaged through super-resolution techniques or electron microscopy.This multilevel approach could be useful in a variety of investigations, including neurological disease studies like stroke.Stroke alters and triggers changes in intra-and inter-hemispheric connectivity; this rewiring aims at compensating for the loss of function [120].fMRI can easily tell the progression of the pathology, showing the plastic remapping over the whole brain [121].Simultaneously, TPF imaging of fluorescent labeled neurons can highlight the structural and functional rewiring underlying fMRI signals in the newly activated cortical area of the same mouse with cellular detail [58].Nonlinear imaging on stroke animal models is capable of providing fundamental insight into axonal rewiring and spine plasticity [58,122,123] while accurately depicting the functional remapping of the damaged cortical area [124,125].Then, alterations in longrange projections underlying inter-hemispheric plasticity can be explored ex vivo on the same cleared brain using LSM.Stroke-induced expression of several molecules and proteins, like growth-associated factors and inflammatory chemokines, can be addressed additionally with multi-round immunohistochemistry.At this point, as discussed above, fine details like the presence of synaptic contacts on regenerated axons are available by electron microscopy on targeted regions of the same sample.This multidimensional hybrid strategy can be extremely useful in the investigation of com-plex brain diseases, and would speed up the translation of neurobiology studies to clinical settings.

Fig. 17 . 1 :
Fig. 17.1: Schematic representation of the three main optical techniques used to reconstruct neuroanatomy on a brain-wide scale: serial two-photon sectioning (a), micro-optical sectioning tomography (b), and light sheet microscopy (c).Reproduced with permission from [9].

Fig. 17 . 2 :
Fig. 17.2: Mapping brain-wide anatomy.(a) Tracing of long-distance projecting neurons in the brain of a Thy1-eYFP-H mouse.Traced neurons (depicted with different colors) are registered on an MRI reference mouse brain.(b)-(e) Full reconstruction of Purkinje cells neuroanatomy in the cerebellum of an L7-GFP mouse.(b) 3D volume rendering of the whole cerebellum.The superimposed planes refer to transverse (red), sagittal (green) and coronal (blue) digital sections shown in (c)-(e) respectively.Each section is a maximum intensity projection of 40 µm thick slabs.Scale bars: 1 mm.Reproduced with permission from [13] (a), and [16] (b)-(e).
.3 shows two examples of experience-dependent spine plasticity in the adult neocortex investigated by two-photon microscopy in vivo.Fig.17.3 (a)-(e) shows how chessboard whisker trimming modifies whisker representational maps in the barrel cortex at the synaptic level.Fig.17.3(a)  shows examples of spine stability (yellow arrowhead), and turnover (white arrowhead).The authors showed that new persistent spines (orange arrowhead) are more likely to grow after whisker trimming, whilst previously persistent spines (green arrowhead) are more likely to disappear.In a different experimental paradigm, i.e., focal lesion in the retina (Fig.17 .3 (f)-(j)), dendritic spines near or in the lesion projection zone (LPZ) in the neocortex display high turnover rates; in the center of the LPZ almost the entire spine population turns over.Although some spines are present over more than two months of imaging (for example, yellow arrowhead), most spines are lost (for example, green arrowhead) and replaced by new persistent spines (orange arrowhead).

▸ Fig. 17 . 3 :
(a) Time-lapse TPF images of dendritic spines in the barrel cortex before and after chessboard whisker trimming.(b) The experimental paradigm: chessboard whisker trimming causes changes in the whisker representational map in the barrel cortex.(c) Spine density in the barrel cortex does not change after whisker trimming.(d) The fraction of surviving spines is slightly decreased after whisker trimming due to an increased loss of persistent spines.(e) The fraction of new persistent spines increases ≈ 2.5 fold after whisker trimming.(f) Time lapse of dendritic spines in the visual cortex after a unilateral focal lesion in the retina.(g)

Fig. 17
Fig. 17.4: Morphological rewiring of axonal connectivity after laser transection in different brain regions.(a) Time course (from day 0 to day 5) of a cerebellar axon before and after laser dissection.The first image (d 0) was acquired one day before laser irradiation.The laser beam was focused on the axon where the red arrow points on d1.The red and green arrowheads at d5 highlight the degeneration of distal portion and the protrusion of new branches, respectively.Scale bar, 15 μm.The histogram compares the sprouting frequency (SF) in control and laser axotomized climbing fibers.Reproduced with permission from [66].(b) Following laser dissection (yellow arrow in the +5 min panel) and subsequent degeneration of the axonal ending (red line), few axons of pyramidal layer 6 (L6) neuron demonstrated sustained regrowth over a 3 month time period (green line).The red dotted line box indicates the area followed over time in the time-stamped panels.Insets at +5 d and +51 d show the growth cone.The graph on the lower left shows that 55 % of L6 axons (red) demonstrated regrowth over several weeks compared with only ≈ 20 % of all other axons (blue).Reproduced with permission from[64].

. 4 :
Fig. 17.4: Morphological rewiring of axonal connectivity after laser transection in different brain regions.(a) Time course (from day 0 to day 5) of a cerebellar axon before and after laser dissection.The first image (d 0) was acquired one day before laser irradiation.The laser beam was focused on the axon where the red arrow points on d1.The red and green arrowheads at d5 highlight the degeneration of distal portion and the protrusion of new branches, respectively.Scale bar, 15 μm.The histogram compares the sprouting frequency (SF) in control and laser axotomized climbing fibers.Reproduced with permission from [66].(b) Following laser dissection (yellow arrow in the +5 min panel) and subsequent degeneration of the axonal ending (red line), few axons of pyramidal layer 6 (L6) neuron demonstrated sustained regrowth over a 3 month time period (green line).The red dotted line box indicates the area followed over time in the time-stamped panels.Insets at +5 d and +51 d show the growth cone.The graph on the lower left shows that 55 % of L6 axons (red) demonstrated regrowth over several weeks compared with only ≈ 20 % of all other axons (blue).Reproduced with permission from[64].
.5 (a) shows the expression of genetically encoded indicators (GE) by viral transduction (left panel), in utero electroporation (middle panel), and generation of transgenic mouse lines (right panel).

Fig. 17
Fig. 17.6: Optical interrogation of neuronal circuit.(a) Neuron excitation by glutamate uncaging.Structure of Ruthenium-bipyridine-trimethylphosphine-Glutamate (RuBi-Glutamate) and glutamate photorelease reaction.Layer 2/3 pyramidal cell loaded with Alexa-594 and position of the multiplexed uncaging laser targets (eight subtargets) on the soma of the cell.Action potentials triggered by uncaging of RuBi-Glutamate in a layer 2/3 pyramidal neuron.Input map for a layer 5 pyramidal neuron and superimposed morphological reconstruction of its dendritic tree.Gray areas outline stimulated neurons.Colored areas are outlines of all positive cells, color-coded according to peak EPSP amplitude.Dotted outline marks the patch pipette location.Scale bar, 100 µm.Voltage recordings from the patched cell during uncaging in locations labeled by arrows.Reproduced with permission from [90, 91].(b) Optogenetic tools.Schematic of channelrhodopsin-2 (ChR2) and the halorhodopsin (NpHR) pump.Action spectra for ChR2 and NpHR.Cell-attached (top) and whole-cell current-clamp (bottom) traces from hippocampal neurons showing all-optical neural activation and inhibition.Blue pulses represent the blue light flashes used to drive ChR2-mediated activation and the yellow bar denotes NpHR-mediated inactivation.A cannula is implanted into the head of the experimental animal to guide an optical fiber to the targeted brain region.The optical fiber is coupled to a strong light source to bring blue or yellow light into the brain.Reproduced with permission from [93].

. 6 :
Fig. 17.6: Optical interrogation of neuronal circuit.(a) Neuron excitation by glutamate uncaging.Structure of Ruthenium-bipyridine-trimethylphosphine-Glutamate (RuBi-Glutamate) and glutamate photorelease reaction.Layer 2/3 pyramidal cell loaded with Alexa-594 and position of the multiplexed uncaging laser targets (eight subtargets) on the soma of the cell.Action potentials triggered by uncaging of RuBi-Glutamate in a layer 2/3 pyramidal neuron.Input map for a layer 5 pyramidal neuron and superimposed morphological reconstruction of its dendritic tree.Gray areas outline stimulated neurons.Colored areas are outlines of all positive cells, color-coded according to peak EPSP amplitude.Dotted outline marks the patch pipette location.Scale bar, 100 µm.Voltage recordings from the patched cell during uncaging in locations labeled by arrows.Reproduced with permission from [90, 91].(b) Optogenetic tools.Schematic of channelrhodopsin-2 (ChR2) and the halorhodopsin (NpHR) pump.Action spectra for ChR2 and NpHR.Cell-attached (top) and whole-cell current-clamp (bottom) traces from hippocampal neurons showing all-optical neural activation and inhibition.Blue pulses represent the blue light flashes used to drive ChR2-mediated activation and the yellow bar denotes NpHR-mediated inactivation.A cannula is implanted into the head of the experimental animal to guide an optical fiber to the targeted brain region.The optical fiber is coupled to a strong light source to bring blue or yellow light into the brain.Reproduced with permission from [93].

Fig. 17
Fig. 17.7: Multiple spatial scales in the brain.On the top, different structures are depicted in proximity to their typical size, showing how relevant spatial scales in the brain span several orders of magnitude.On the bottom, typical working ranges of state-of-the-art imaging technologies.

. 7 :
Fig. 17.7: Multiple spatial scales in the brain.On the top, different structures are depicted in proximity to their typical size, showing how relevant spatial scales in the brain span several orders of magnitude.On the bottom, typical working ranges of state-of-the-art imaging technologies.

Fig. 17
Fig. 17.8: Correlative microscopy in neuroscience.(a) Experimental approach for examining the relationship between single-cell activity and fMRI signals.Single-cell Ca 2+ measured by two-photon microscopy was correlated with bulk responses in the same region measured by one-photon fiberoptic microscopy.Then the bulk one-photon responses were correlated with the BOLD fMRI signals.Figure modified with permission from [126].(b) On the left, maximum intensity projections of several stacks TPF stitched together in a single image.The red dashed line highlights blood vessel shadows.Red arrows highlight characteristic features of a dendritic arbor, to help finding it back in the CLSM images.On the right, CLSM imaging of the same neuron observed with TPF.Starting from the apical portion of the dendritic tree, the neuron has been segmented and is shown inside a maximum-projection 3D rendering.The scale of the figure can be inferred from the red cube down on the right, which has 100 µm sides.Figure modified with permission from [115].(c) Functional characterization of direction-selective retinal ganglion cells (DSGCs) and their localization within SBSEM volume.On the left, Polar tuning curves for 25 DSGCs sorted and color-coded by preferred direction.The corresponding soma locations superimposed onto a two-photon image from the recorded region of the ganglion cell layer and the acquired SBSEM volume (scale bars: 100 nm).On the right, skeleton reconstructions of DSGCs.DSGCs, color-coded by preferred direction (inset), normal to the plane of the retina (scale bars: 50 µm).Figure modified with permission from [116].(d) Correlative in vivo TPF and focused ion beam scanning electron microscopy of cortical neurons.In vivo TPF imaging of an axon showing two stable boutons (scale bar: 5 μm).Both boutons make multiple synaptic contacts, as visible in a single plane of the correspondent EM images, with multiple dendritic spines (scale bar: 500 nm).3D rendering of the same axon imaged in TFP microscopy.The cytoplasm of the axon is represented in light blue, mitochondria in green, synaptic vesicles in yellow and synapses in red.The postsynaptic spiny neurons are shown in gray.Figure modified with permission from [119].

. 8 :
Fig. 17.8: Correlative microscopy in neuroscience.(a) Experimental approach for examining the relationship between single-cell activity and fMRI signals.Single-cell Ca 2+ measured by two-photon microscopy was correlated with bulk responses in the same region measured by one-photon fiberoptic microscopy.Then the bulk one-photon responses were correlated with the BOLD fMRI signals.Figure modified with permission from [126].(b) On the left, maximum intensity projections of several stacks TPF stitched together in a single image.The red dashed line highlights blood vessel shadows.Red arrows highlight characteristic features of a dendritic arbor, to help finding it back in the CLSM images.On the right, CLSM imaging of the same neuron observed with TPF.Starting from the apical portion of the dendritic tree, the neuron has been segmented and is shown inside a maximum-projection 3D rendering.The scale of the figure can be inferred from the red cube down on the right, which has 100 µm sides.Figure modified with permission from [115].(c) Functional characterization of direction-selective retinal ganglion cells (DSGCs) and their localization within SBSEM volume.On the left, Polar tuning curves for 25 DSGCs sorted and color-coded by preferred direction.The corresponding soma locations superimposed onto a two-photon image from the recorded region of the ganglion cell layer and the acquired SBSEM volume (scale bars: 100 nm).On the right, skeleton reconstructions of DSGCs.DSGCs, color-coded by preferred direction (inset), normal to the plane of the retina (scale bars: 50 µm).Figure modified with permission from [116].(d) Correlative in vivo TPF and focused ion beam scanning electron microscopy of cortical neurons.In vivo TPF imaging of an axon showing two stable boutons (scale bar: 5 μm).Both boutons make multiple synaptic contacts, as visible in a single plane of the correspondent EM images, with multiple dendritic spines (scale bar: 500 nm).3D rendering of the same axon imaged in TFP microscopy.The cytoplasm of the axon is represented in light blue, mitochondria in green, synaptic vesicles in yellow and synapses in red.The postsynaptic spiny neurons are shown in gray.Figure modified with permission from [119].