Lateral, axional and temporal resolutions for imaging in vivo

Lateral, axional and temporal resolutions for imaging in vivo

We are searching data for your request:

Forums and discussions:
Manuals and reference books:
Data from registers:
Wait the end of the search in all databases.
Upon completion, a link will appear to access the found materials.

What are the spatial (lateral and axional), and temporal resolutions that one is expected to get with the following optical imaging techniques?

  • Optical Bioluminiscence Imaging (BLI)
  • Optical Fluorescence Imaging (FLI)
  • Multiphoton Microscopy (e.g. 2-photon microscopy)

Are they any better than the others in providing higher resolutions?

The lateral and axial resolution depends on the numerical aperture of your objective (lateral = lambda/2NA; axial = 2lambda/NA^2). You can improvee the axial resolution with FLI by using a confocal approach or with 2-photon microscopy (2PM). In 2PM you need absorption of two infrared photons to get emission. The probability of 2 infrared absorption depends quadratically on the excitation intensity. Therefore, fluorescence is restricted where the illumination beam is focused which give a better sectioning. An advantage of 2PM is the higher penetration depth and reduce phototoxicity. Indeed you can use 2 infrared photons to excite fluorophore which required visible excitation.

Temporal resolution depends on the area you want to measure and the integration time.

Resolution is limited by the diffraction. In the best case with the 3 imaging techniques you mentioned you can get ~200-300 nm in lateral resolution and 500 - 800 nm in axial resolution Higher resolution can be achieved by using super resolution techniques. They are separated in two groups. The first one exploid the photo-physical properties of the fluorophore to excite and de-excite it. This allows to resolve sub-diffraction areas. This is the case STED (Stimulated emission depletion) microscopy or SSIM (saturated structured illimination microscopy). The second group use the fluctuation of the fluorophore emission to localize it better. It includes techniques such as STORM (STochastic Optical Reconstruction Microscopy, PALM (Photo-Activated Localization Microscopy) and SOFI (super-resolution optical fluctuation imaging).

Lateral, axional and temporal resolutions for imaging in vivo - Biology

In vivo imaging of axonal transport in the sciatic nerve of live anesthetised mice.

Signalling endosomes are monitored using fluorescent probes and confocal microscopy.

Axonal transport in motor and sensory neurons can be differentiated.

This method can be easily adapted to study the axonal transport of other cargoes.

Potential use of this in vivo imaging approach in drug screening.


In this protocol, we describe an imaging technique to study motor axons over time in an acute nerve–muscle explant of the triangularis sterni muscle (the ‘triangularis sterni explant’). The triangularis sterni muscle originates from the inside of the sternum and inserts on the internal rib cage. It is innervated by branches of the upper intercostal nerves, which extend superficially through the muscle and form an endplate band that is typically localized along the craniocaudal axis parallel to the border of the sternum ( Fig. 1 ) 1 . The triangularis sterni explant thus contains distal motor axons, their intramuscular arborizations, neuromuscular synapses and the innervated muscle fibers. A variant of the triangularis sterni explant was first described by McArdle et al. 1 and has been used previously to study neuromuscular physiology and pathophysiology 2 – 5 . We describe a simplified preparation here, on the basis of methods published by our lab 6 , 7 . This muscle provides a number of advantages for imaging experiments. First, it is a particularly thin muscle that consists only of a few (υ) layers of muscle fibers. As a consequence, the innervating motor axons and their branches run superficially and can be followed along their entire course up to their presynaptic terminals on neuromuscular endplates. Second, the triangularis sterni has basically no sensory innervation, so all neurites branching into the muscle are motor axons 8 . Third, as the muscle originates and inserts on bony components of the rib cage, it can be easily dissected without damaging the muscle. This allows the muscle to be isolated at very young ages (as early as embryonic day 13, when neuromuscular junctions first form). Fourth, the triangularis sterni explant contains the entire arborizations and all endplates of the intercostal motor axons entering the muscle. This allows the reconstruction of motor units, e.g., by using transgenic mouse lines, which express fluorescent proteins in a subset of motor axons 9 or which show combinatorial expression of spectrally different fluorescent proteins in individual neurons 10 . Finally, the muscle is multisegmental and, as a consequence, it is innervated by axons derived from different intercostal nerves, which anastomose only once they reach the endplate band. This anatomical pattern allows selective stimulation or partial denervation experiments, in which some inputs are stimulated or transected and the compensatory response of the remaining axons is analyzed.

Innervation pattern of the triangularis sterni muscle. (a𠄼) Triangularis sterni muscle and its innervation in a thy1–YFP16 mouse with YFP-labeled axons 18 . (a) Combined labeling of axons (green), synapses (red) and muscle (gray scale). (b) Axonal labeling. (c) Muscle stain. The violet line in b and c outlines the triangularis sterni muscle between the inner surface of the sternum to the proximal rib cage. (d) Confocal reconstruction of the area boxed in a, showing the multisegmental innervation of the triangularis sterni muscle. (e) High-power confocal reconstruction of a part of the endplate band (boxed area in d). (f) Confocal image showing two individual neuromuscular junctions boxed in e. Accumulations of mitochondria can be found in presynaptic terminals. YFP (axonal cytoplasm), green BTX (acetylcholine receptors), red phalloidin (muscle), gray scale. Scale bars, 500 μm in d and 10 μm in f.

Clearly, there are also certain limitations of this explant system, which need to be considered (for details, see TROUBLESHOOTING table). One obvious limitation is that, during preparation of the explant, the intercostal nerves, which innervate the triangularis sterni muscle, are separated from their neuronal cell bodies. As a result, the axotomized axons start to undergo Wallerian degeneration between 12� h after preparation. This limits the lifespan of the explant and thus the time period over which axonal changes can be imaged. Our studies indicate, however, that axonal morphology and even intracellular processes like the rate of transported mitochondria (i.e., the number of mitochondria that cross a fiduciary mark) remain unchanged compared with the in vivo situation for at least the first 2 h after preparation 6 , 7 . Further studies using muscles derived from mutant mice (WLD S mice), in which Wallerian degeneration is delayed, indicate a way by which the lifetime of the preparation can be extended in some experimental paradigms that are not impacted by the WLD S mutation 11 . Obviously, changes related to the absence of neural activity (such as the internalization of acetylcholine receptors see Fig. 2f ) can occur, but introducing external stimulation is easy. Although the triangularis system can provide valuable insight into acute axonal physiology or pathology, it cannot replace in vivo approaches for the study of chronic axonal changes that require repetitive monitoring over several days, such as in vivo imaging of the sternomastoid muscle 12 – 14 . For in vivo imaging, the anatomical location of the triangularis sterni muscle on the inside of the thorax is prohibitive, preventing a direct in vivoex vivo comparison using this muscle.

High-resolution images from living triangularis sterni explants prepared as described in this protocol. (a) Developing neuromuscular junction with two axonal inputs (arrows) during synapse elimination at postnatal day (P) 9. Fluorescently tagged BTX (red) highlights postsynaptic acteylcholine receptors, axons are transgenically labeled with YFP (green thy1-YFP-16 mouse), and muscle is revealed by oblique illumination (gray scale). #, myonucleus *, motor axon that exits the focal plane to innervate a deeper synapse on the left. (b) Superficial axons in the intercostal nerve. Myelin sheaths are visualized by oblique illumination (gray scale), whereas axonal mitochondria are transgenically labeled with CFP (blue) in a thy1-mitoCFP-S mouse 7 . Arrows, Schmidt–Lantermann incisures *, node of Ranvier. (c) Developing neuromuscular junction (c1 merge c3, BTX, red) innervated by three axons (arrows) (c2, CFP, green) sheathed by Schwann cells (c4, GFP, cyan) in a P10 thy1-CFP5 × s100-GFP Kosmos double transgenic mouse 21 . *, Schwann cell nucleus. The boxed area in c3 is shown at higher magnification in f. (d𠄿) Time-lapse series of cellular and subcellular axonal dynamics in explants. (d) Transgenically labeled axonal filopodia (YFP, green) that extend (blue arrowheads) and retract (orange asterisks) from a developing (P11) neuromuscular junction imaged for approximately 20 min. (e) Transgenically labeled mitochondrion (CFP, blue) that is transported inside an axon (frame rate, 1𠄲 Hz). (f) Postsynaptic acetylcholine receptors (BTX, red) that are endocytosed by the muscle fiber over a period of 𢏁 h (note that this fast turnover likely results from the absence of neurotransmission in this unstimulated explant). Scale bars, 10 μm in a and b 5 μm in c 2.5 μm in d 0.5 μm in e.

The main application of the triangularis sterni explant is thus short-term, high-resolution imaging of peripheral motor axons, their synapses and their surrounding glial cells. Motor axons are ideally suited for imaging studies because their large diameter allows the visualization of intra-axonal processes. Furthermore, neuromuscular junctions between motor axons and muscle fibers are among the biggest and most readily accessible synapses in the mammalian nervous system 15 . Finally, excellent fluorescent probes and transgenic animals that highlight components of the neuromuscular system are available 16 , which can easily be used in combination with this explant system (see Experimental design). Many aspects of motor axon physiology and pathology that have previously been addressed in fixed tissue can now be analyzed in a time-resolved fashion. One example is provided by our recent study of axonal transport, which made use of new transgenic mice with labeled neuronal mitochondria 7 . In this study, we have shown that axonal transport rates in the acute explants were in a similar range as those measured in the sciatic nerves of living mice. However, 2 h after excision of muscle and nerves, axonal transport rates began to decline, possibly because no new mitochondria were provided by the disconnected cell bodies.

Another application of the triangularis sterni explant is the study of neuromuscular development. The neuromuscular junction has long been used as a model to study synapse elimination in the developing nervous system 17 . Key aspects of synapse elimination can be followed ex vivo in triangularis sterni explants derived from early postnatal mice. In one study, we used a combination of in vivo microscopy, ex vivo imaging and serial electron microscopy to understand how axons are lost during synapse elimination 6 . Time-lapse imaging revealed that axons retreat from synapses by shedding part of their material as small 𠆊xosomes’. The dynamic characteristics of axosome shedding became apparent only because we were able to image retreating axons with high temporal and spatial resolution in the explant.

Finally, imaging studies using the triangularis sterni explant can help to explore the pathogenesis of important neurological diseases. Motor axons are affected in a number of human diseases, which include degenerative conditions, such as amyotrophic lateral sclerosis or hereditary polyneuropathies, but also traumatic lesions and inflammatory diseases of the peripheral nerve (such as the Guillian-Barré syndrome). In many cases, isolated motor neurons in cell culture provide an unsatisfying model, as interactions of motor axons with surrounding glial cells and invading immune cells are crucial aspects of the disease process. Such interactions can, however, be followed in acute nerve–muscle explants, which preserve the local milieu of nerve and muscle. In addition to motor axons, neuromuscular junctions as well as the postsynaptic muscle fibers are targeted by diseases, such as myasthenia gravis or the various inflammatory, metabolic and degenerative myopathies, which can also be studied using the triangularis sterni explant depending on the availability of mouse models.

In summary, we provide a protocol that allows imaging of motor axons in a living explant of the triangularis sterni muscle. Owing to the particular advantages of this preparation, a number of aspects of axonal morphology and physiology can be studied with high temporal and spatial resolution. Initial studies already illustrate how this approach can provide novel insights into the dynamic behavior of motor axons in development, health and disease 6 , 7 .

Experimental design

Imaging system

To image motor axons and their synapses in the triangularis sterni explant, we use a wide-field epifluorescence microscopy system. The triangularis sterni muscle is a flat and superficial muscle with most axons running parallel to the surface and many synapses located directly below the parietal pleura that covers the preparation ( Fig. 1 ). These anatomical characteristics make it possible to use wide-field epifluorescence microscopy ( Fig. 3 ) without the need to resort to more expensive techniques that allow deep-tissue penetration, such as two-photon microscopy. However, if the experimental design requires more sophisticated imaging, either a confocal or two-photon microscopy system can be used.

The preparation of the triangularis sterni explant. (a) Left two images show the intact thorax that is dissected after killing the mouse (first image from left, frontal view second image, lateral view from right side after 90° rotation). Third image shows an inside view of the thorax wall after removal of the thoracic viscera (small blue arrows, remnant of diaphragm *, xiphoid cartilage). Dashed line in second image shows paravertebral cut (a symmetrical cut is done on the left side). In the fourth image, the thorax wall is shown folded open and pinned down into a dish. Suitable positions for minuten pins are indicated by blue circles. Blue arrow points to bony part of the sternum arrowheads indicate right internal mammary vessels. (b) Epifluorescence microscopy setup used for imaging triangularis sterni explants. (c) Close-up view of microscope stage with imaging chamber (boxed in b).

Transgenic mouse lines

Imaging of motor axons and their synapses is based on the use of transgenic mice, in which all or a subset of the motor neurons are labeled with fluorescent proteins 18 . Many such mouse lines are now available that differ in labeling densities (i.e., the percentage of axons labeled), spectral characteristics of the expressed fluorescent proteins or the locations to which these proteins are targeted ( Fig. 2 ) 7 , 10 , 19 . An overview of transgenic mouse lines that express fluorescent proteins in different neural cell types can be found in our recent review article 20 . Transgenic labeling of other peripheral nervous system (PNS) structures, such as Schwann cells ( Fig. 2 ) 10 , 21 , muscle cells 22 or macrophages 23 , has also been reported and can readily be combined with vital dye labeling (such as fluorescently tagged α-bungarotoxin (BTX), Invitrogen Fig. 2 ).

Image analysis

The analyzed parameters depend on the experiment. In many of our experiments, the primary set of data is a movie of a recorded series of images. Two conditions are critical for obtaining good movies: stability of focus (absence of z-displacement) and absence of drift (x–y displacement). Whereas the former can only be ensured by careful setting up of the superfusion and suction lines and meticulous focusing during the experiment, the latter can to some degree be corrected by postprocessing using x–y alignment software (e.g., Autoquant, Molecular Devices or the open-source software ImageJ with the ‘stackreg’ plugin, The resulting x–y aligned image series can then be cropped and stored as a stacked tiff format, or alternatively visualized using a standard movie format such as .avi or .mov (Metamorph software offers these capabilities, but increasingly similar features can be found with open-source ImageJ, especially when adding some of the many plugins


In Vivo MR Imaging of Intact Spinal Cord

Gradient-echo MR imaging with ECG triggering and respiratory gating yielded high-resolution images of intact thoracic rat spinal cord in vivo without significant motion artifacts (Fig 1). The typical butterfly-shaped gray matter could be clearly distinguished from the less-intense surrounding white matter axonal tracts. The CSF surrounding the spinal cord appeared as a small hyperintense rim adjacent to the spinal cord due to its longer T2 and T2* relaxation times. With the applied sequence vertebral bodies appeared completely dark, whereas vertebral disks were bright.

In vivo MR imaging of intact spinal cord at 17.6T. Panels A and B show sagittal scans through the thoracic spinal cord (section thickness, 239 μm FOV, 40 × 30 mm in-plane resolution, 156 × 117 μm TR, ∼200 milliseconds, depending on heart rate TE, 4.4 milliseconds). Panels C and D display axial scans through the thoracic spinal cord (section thickness, 500 μm FOV, 17.7 × 35.5 mm in-plane resolution, 69 × 69 μm TR and TE, as above). Scale bar, 2 mm. A, A more lateral sagittal scan depicts primarily white matter (lower signal intensity) with some longitudinally oriented more hyperintense structure, reflecting the gray matter of the lateral ventral horn. CSF appears hyperintense, vertebral bodies are hypointense. B, Most of the spinal cord parenchyma displayed here represents gray matter (hyperintense) surrounded by white matter tracts (hypointense) in a paramedian sagittal scan through the spinal cord. C, An axial scan through the thoracic spinal cord allows the clear distinction between the typical butterfly appearance of the spinal cord gray matter and the surrounding hypointense white matter. Also of note, spinal roots can be clearly identified at this level. D, A subsequent scan more caudally shows the spinal cord in cross-section away from the spinal root entry zone.

Ex Vivo MR Imaging of Contused Spinal Cord

Spin-echo images of excised spinal cord 4 weeks after a thoracic contusion injury revealed a signal intensity pattern away from the injury epicenter in axial sections (Fig 2), which was almost identical to in vivo gradient-echo images in unlesioned animals (Fig 1C, -D). Gray matter was hyperintense compared with white matter. The central canal could be clearly identified. The gray/white matter differentiation completely vanished at the contusion center (Fig 2G, -H). The homogenous hyperintense signal intensity, which was pronounced in the spinal cord center and diminished toward the surface of the cord in all directions, was only interrupted by areas of signal intensity loss in the center of the cord, which were confined mostly to the gray matter (Fig 2G). Trauma-induced hemorrhage and consecutive hemosiderin deposits are the likely pathologic correlates for this signal intensity–free area, which have been described to occur frequently within the spinal cord gray matter after injury. 17 Hypointensities within the lesion center (in particular in sagittal MR images)—not as pronounced as the described hemosiderin-associated changes—corresponded to immunoreactivity for collagen type III, but not GFAP, in sagittal histologic sections (Fig 3B, -J, -N), thus representing components of the fibrous rather than the gliotic scar. The overall diameter of the cord was reduced. In particular the dorsolateral columns were symmetrically diminished in volume (Fig 2G, -H), reflecting substantial irreversible atrophy as soon as 4 weeks after injury. In some Nissl-stained sections (Fig 3G) all types of organized tissue were absent, which suggests a developing fluid-filled cavity however, corresponding MR imaging scans did not display a homogenous hyperintensity (Fig 3C), which would be the MR equivalent of a cystic lesion defect. Instead, only a small hyperintense band—the correlate of the true cystic lesion defect—surrounded a hypointense core. Thus, it is likely that this hypointense area contained nonorganized material (inflammatory cells and cell debris mixed with hemosiderin deposits), which is regularly lost during the process of histologic analysis.

Axial ex vivo MR imaging scans of contused rat spinal cord. Axial sections of ex vivo MR imaging show microscopy grade visualization of morphologic changes in the injured rat spinal cord 4 weeks after contusion injury at midthoracic level (2D multisection spin-echo section thickness, 300 μm FOV, 6 × 6 mm in-plane resolution, 23 × 23 μm TE, 7.5 milliseconds TR, 2 seconds). Scale bar, 1 mm. Panels A–L show consecutive sections in the rostral-caudal direction. In rats the dorsal columns contain not only ascending proprioceptive projections, which are located in the dorsal half of the dorsal column. The crossed corticospinal tract projects in the ventral half of the dorsal columns, unlike humans, where most corticospinal axons are located in the lateral columns. AF, In sections rostral to the contusion site, the spinal cord morphology is still maintained with a clear differentiation of white and gray matter. Note the hypointensity in the dorsal part of the dorsal columns (arrowheads) identical to ascending proprioceptive projections. GI, At the lesion center, the spinal cord diameter is reduced and gray and white matter can no longer be separated. Hypointensities located in the center (G) reflect hemosiderin deposits. JL, Caudal to the lesion, the gray-white matter contrast is preserved. Both in rostral and caudal scans (BD and JL) hyperintense signals are found in the dorsal columns consistent with cystic defects (see also Fig 3). Hypointensities in axial scans rostral to the lesion correspond to ascending sensory projections, whereas in caudal scans they represent the area of the corticospinal tract (JL, arrowheads).

Sagittal ex vivo MR imaging scans of contused rat spinal cord and corresponding histology. Midthoracic contusion injury 4 weeks postlesioning same specimen as Fig 2 (2D multisection spin-echo section thickness, 208 μm FOV, 20 × 6 mm in-plane resolution, 23 × 78 μm TE, 7.5 milliseconds TR, 2 seconds). Scale bar, 2 mm. A–D, Ex vivo MR imaging scans from lateral to medial. E–H, Corresponding Nissl-stained sections. I–L, Corresponding GFAP immunostained sections. M–P, Corresponding collagen type III immunostained sections. Homogenous hyperintensities at the injury center (A) correspond to cystic lesion defects in histologic sections (E, I, and M). In other sections. mixed hypo-/hyperintensities in the lesion center (B and C, arrows) are associated either with cystic lesion defects, hemosiderin deposits, or fibrotic scar formation (F, N, G, and O, arrows). A hypointensity following the path of the dorsal corticospinal tract caudal to the lesion—corresponding to hypointensities in the dorsal columns in axial MR images (Fig 2)—is highlighted by arrowheads (B).

Remote from the lesion center, some neuroanatomically restricted signal intensity changes could be observed (Fig 2A, -JL Fig 3B). Hypointensities strictly confined to either the former main dorsal corticospinal tract caudal to the lesion or the ascending proprioceptive pathways (gracile fascicle) rostral to the lesion were identified. Hyperintense regions within the dorsal columns rostrally and caudally (Fig 2BD, -JL Fig 3A, -B) can be attributed to cystic necrotic zones, which were identified in the corresponding sagittal histologic sections (Fig 3F). There was no difference in overall signal intensity changes compared with the spinal cord sample, which was taken 2 weeks after the contusion injury (data not shown).

In Vivo MR Imaging of Contused Spinal Cord

Five animals with spinal cord contusions at thoracic level applied by the IH Impactor and one intact animal were analyzed with MR imaging 2, 6, and 8 weeks postoperatively. Only 2 of 5 injured animals survived the MR imaging procedure and were thus available for histologic evaluation. Most signal intensity changes observed in ex vivo spin-echo images in the contused spinal cord were almost identical to in vivo gradient-echo images of the contused spinal cord. The cord diameter was reduced at the lesion center in particular, the dorsolateral aspects were atrophic (Figs 4AF and 5E, -F), which was confirmed by identical findings in corresponding Nissl-stained histologic sections. Toward the lesion epicenter, the gray/white matter differentiation completely vanished (Fig 5DG). The signal intensity pattern of the gray matter in the intact spinal cord (Fig 1C, -D) was replaced by hypointense areas in the center of the cord surrounded by a hyperintense rim dorsally and ventrally (Fig 5DG). As described in ex vivo MR imaging (Fig 2F, -G), the hypointensities at the lesion center represent hemosiderin deposits, which are remnants of the hemorrhage caused by the trauma corresponding to dark areas in Nissl- and Prussian-blue–stained sections (Fig 4K, -N). In addition, hypointensities were found confined to the dorsal columns rostral and caudal to the lesion, both in sagittal and axial scans (Figs 4B, -E, and 5H), which are also paralleled by ex vivo MR imaging findings. These hypointensities could be correlated with hemosiderin deposits as depicted by Prussian Blue staining of corresponding spinal cord sections (Fig 4K). There was no correlation with areas of macrophage/monocyte infiltration (Fig 4LO), which are—in addition to being located adjacent to the injury center—commonly identified in white matter tracts remote from the lesion site highlighting areas of ongoing Wallerian degeneration (N. Weidner, unpublished observation).

Sagittal in vivo MR imaging scans of contused rat spinal cord and corresponding histology. In vivo MR imaging in adult rats at 2 and 8 weeks after thoracic contusion injury displays signal intensity changes, which parallel the ex vivo MR imaging data (2D multisection gradient-echo A–C, section thickness, 311 μm FOV, 30 × 30 mm in-plane resolution, 117 × 117 μm TE, 4.4 milliseconds TR, ∼200 milliseconds, depending on heart rate D–F, section thickness, 300 μm FOV, 40 × 25 mm in-plane resolution, 156 × 98 μm TE, 3.7 milliseconds TR, 200 milliseconds, depending on heart rate). Scale bar AF, 5 mm G–O, 1 mm. Consecutive sagittal MR images are shown at 2 weeks (AC) and 8 weeks (DF) postinjury with corresponding histologic Nissl (GI) and Prussian-blue (JL) –stained sections, and sections processed for ED1 immunohistochemistry (macrophages, monocytes MO), all from the same animal. Arrows in A and B highlight the site of the impact. Hypointensities along ascending and descending axon projections in the dorsal columns (B and E) correlate with hemosiderin deposits (K) rather than macrophage/monocyte infiltration (O respective areas are indicated by arrowheads). These changes increase from 2 weeks (C) until 8 weeks (G) postinjury. The clear reduction in cord diameter over time (B vs E and C vs F) corresponds to the atrophy seen in histologic sections (GO).

Axial in vivo MR imaging scans of contused rat spinal cord. In vivo MR axial scans in adult rats 6 weeks postinjury (2D multisection gradient-echo section thickness, 370 μm FOV, 20 × 20 mm in-plane resolution, 78 × 78 μm TE, 4.2 milliseconds TR, ∼200 milliseconds, depending on heart rate). Scale bar, 2 mm. Scans rostral to the contusion (AC), at the lesion center (DG), and caudal to the lesion (H). The clear differentiation between white and gray matter disappears over subsequent sections. At the lesion center, hypointensities are surrounded by hyperintensities, which are less pronounced toward the cord surface (E and F). The dorsal aspect of the spinal cord at the lesion site appears more homogenously hyperintensive, most likely representing cystic changes (D and E, arrowheads). Signs of atrophy are present in the dorsolateral spinal cord (E and F).

In Nissl-stained histologic sections (Fig 4H), small cigar-shaped cystic areas decreasing in size rostrally were identified, which represent areas of beginning post-traumatic syrinx formation. In vivo MR imaging did not produce correspondingly identifiable signal intensities (Fig 4B,-E). Most likely, the chosen sagittal imaging planes did not include the location of the central canal.


The structural diversity of lateralis afferent peripheral arbors

To unravel the mechanisms responsible for the selection of plane-polarized hair cells by lateralis afferent neurons, we performed a detailed reconstruction of optically isolated axonal arbors in toto at high spatial resolution. This analysis is straightforward because each lateral line organ in the zebrafish larva is innervated by only two or three afferent neurons (Fig. 1A) (Faucherre et al., 2009). Initially, we sought to investigate the elaboration of axonal arbors using UV-mediated photoconversion of single lateralis afferents in Tg(HuC:Kaede) (HuC is also known as elavl3 – Zebrafish Information Network) stable transgenic fish, which has been reported to highlight the afferent neurons of the lateral line (Caron et al., 2008). However, we found that in larvae resulting from a cross between the efferent marker Tg(islet1:gfp) and Tg(HuC:Kaede), efferent neurons of the lateral line were also labeled by Kaede (Fig. 1B). Thus, we could not unequivocally assign an afferent identity to neurons of the lateral line in Tg(HuC:Kaede).

Central and peripheral projections of lateralis afferent neurons. (A) GFP-positive peripheral projections in the HGn39D transgenic zebrafish line. Three afferent neurons projecting to a neuromast are depicted. (B) Neuromast innervation in a HuC:Kaede and islet1:gfp double-transgenic larva after Kaede photoconversion. Green neurons are GFP positive, red neurons are Kaede positive and yellow neurons are GFP/Kaede positive. Neuromasts are innervated by both red and yellow neurons. Arrowheads indicate the neuromasts. Dashed lines delimit the spinal cord. (C) Soma of a single TdTomato-positive afferent neuron in a GFP-positive posterior ganglion of an HGn39D transgenic fish. The soma expressing TdTomato is localized in the posterior lateral line ganglion. (D) Central projection of a single TdTomato-labeled afferent neuron. Anterior is to the left and posterior to the right. (E,F) Peripheral projections of a single TdTomato-labeled afferent neuron at two different time frames (10-minute interval). The arrow indicates a bulged extremity, whereas the arrowheads indicate thin and mobile neurites. (G,H) Anti-Ribeye B immunostaining. The formation of ribbon synapses between the afferent neuron (red) and the hair cells (green) is visualized by the presence of the Ribeye B protein (cyan). Arrowheads indicate (G) a thin neurite that does not contact a hair cell, and (H) bulged neurites containing the Ribeye B protein. (I) The average number of neurites present at each time point for wild-type (wt) and tmie mutant fish. Error bars represent the s.e.m. Wild type, n=6 tmie, n=6. Scale bars: 10 μm in A,G,H 20 μm in B-D 5 μm in E,F.

Central and peripheral projections of lateralis afferent neurons. (A) GFP-positive peripheral projections in the HGn39D transgenic zebrafish line. Three afferent neurons projecting to a neuromast are depicted. (B) Neuromast innervation in a HuC:Kaede and islet1:gfp double-transgenic larva after Kaede photoconversion. Green neurons are GFP positive, red neurons are Kaede positive and yellow neurons are GFP/Kaede positive. Neuromasts are innervated by both red and yellow neurons. Arrowheads indicate the neuromasts. Dashed lines delimit the spinal cord. (C) Soma of a single TdTomato-positive afferent neuron in a GFP-positive posterior ganglion of an HGn39D transgenic fish. The soma expressing TdTomato is localized in the posterior lateral line ganglion. (D) Central projection of a single TdTomato-labeled afferent neuron. Anterior is to the left and posterior to the right. (E,F) Peripheral projections of a single TdTomato-labeled afferent neuron at two different time frames (10-minute interval). The arrow indicates a bulged extremity, whereas the arrowheads indicate thin and mobile neurites. (G,H) Anti-Ribeye B immunostaining. The formation of ribbon synapses between the afferent neuron (red) and the hair cells (green) is visualized by the presence of the Ribeye B protein (cyan). Arrowheads indicate (G) a thin neurite that does not contact a hair cell, and (H) bulged neurites containing the Ribeye B protein. (I) The average number of neurites present at each time point for wild-type (wt) and tmie mutant fish. Error bars represent the s.e.m. Wild type, n=6 tmie, n=6. Scale bars: 10 μm in A,G,H 20 μm in B-D 5 μm in E,F.

Instead, we determined afferent neuronal morph by injecting a plasmid encoding HuC:mem-TdTomato, as previously reported (Faucherre et al., 2009). We selected for analysis only those specimens with single red-fluorescent neurons, the soma of which were localized within the lateralis afferent ganglion, which can be marked using the HGn39D transgenic line (Fig. 1C) (Faucherre et al., 2009). Injections were performed in stable Tg(brn3c:gfp) (brn3c is also known as pou4f3 – Zebrafish Information Network) eggs to visualize all hair cells with EGFP (Xiao et al., 2005). To reconstruct axonal arbors, we devised a simple method using inexpensive and widely accessible commercial tools.

Our 3D reconstruction of optically isolated neurons showed that lateralis afferents project stereotypic central axons that enter the hindbrain and bifurcate, forming an anterior branch and a posterior branch (Fig. 1D) (Metcalfe et al., 1985). By contrast, peripheral axonal arbors were structurally diverse (Fig. 1E,F), typically forming different types of neurites. Bulged neurites were stably associated with hair cells in what are likely to be synaptic connections, as the bulges adjoined synaptic ribbons as revealed by immunolabeling with an antibody to Ribeye B (Ctbp2l – Zebrafish Information Network) (Fig. 1G,H) (Sidi et al., 2004). Thin neurites were longer and did not always contact hair cells (Fig. 1E-G). One explanation for the diversity of peripheral axon structure is that they are dynamic.

To examine their behavior, we reconstructed peripheral terminals by live imaging at high temporal and spatial resolution. We acquired z-stacks of the entire axonal arbor every 10 minutes over a period of 1 hour and subsequently rendered them 3D over time (n=6). Arbors were rebuilt by semi-manual tracing of all the filaments at every time point, which allowed us to quantify the behavior of the different neurites (see Table S1 in the supplementary material). This analysis showed that neurites constantly change length and position. On average, neurites grew or retracted at a rate of 100 nm per minute and were capable of extending as much as 5 μm in 10 minutes (Fig. 1E,F). Despite these rapid changes in shape, arbor complexity did not change because the average number of neurites was maintained during the recorded period (Fig. 1I). This is consistent with previous live imaging observations that the rate of neurite sprouting is equivalent to that of pruning (Faucherre et al., 2009). We also found that neurite sprouting occurred by interstitial branching and that pruning occurred through retraction, as we never observed neurite shedding or degeneration. We conclude that the structural diversity of afferent peripheral arbors results from a very active process of neurite production, extension and retraction.

Peripheral arborization is modulated by hair cell activity

The observation that neuron sprouting activity is lower in quiescent neuromasts than in those undergoing regeneration suggests that peripheral arbors restructure in response to cues from the sensory epithelium (Faucherre et al., 2009). We asked whether these cues originate from hair-cell activity by taking advantage of a zebrafish strain with a loss-of-function mutation in tmie, the ortholog of the mammalian deafness gene transmembrane inner ear. Homozygous tmie mutant zebrafish are viable and fertile, but profoundly deaf due to defects in hair cell mechanoreception (Gleason et al., 2009).

We scatter-labeled lateralis afferents by injecting HuC:mem-TdTomato DNA into wild-type and tmie mutant eggs that also carried the Tg(brn3c:gfp) transgene. Again, axons were identified as afferent when they could be traced to a soma located in the postotic ganglion. We acquired 4D images and found that in every case the peripheral arbors were significantly more complex in tmiefish (n=6) than in wild-type controls (n=6) at 7 dpf. We quantified arbor complexity by adopting different strategies. We assigned numerical complexity by counting the total number of neurite termini per arbor. For dynamic complexity, we generated a visual representation of the arbors by placing a colored dot on each neurite terminus. Subsequently, the position of each terminus was tracked in 3D space every 10 minutes for 1 hour. Termini were assigned a different color at each of the six time points that divided the whole recording period (Figs 2 and 3 see Movies 1-4 in the supplementary material). Comparison of the numerical and dynamic complexity between wild-type and tmie animals indicated that the group of mutants had, on average, a 2-fold increase in the number of neurites at each time point (Fig. 1I and see Fig. S1A in the supplementary material). For example, the tmie mutant in Fig. 2 had 16 neurites at the first time frame, whereas wild-type fish displayed only six branches at the same time point.

Four-dimensional reconstruction of a single labeled afferent neuron. (A-G) Four-dimensional representations of an afferent neuron in wild-type (A-C) and tmie mutant (E-G) fish. (A,E) mem-TdTomato-labeled neuron contacting the GFP-expressing hair cells. The terminal point of each neuron has been plotted with a different color. For each terminal point (i.e. for each color) positions have been plotted over time, every 10 minutes for 1 hour. (B,F) Same as A,E but without the hair cells. (C,G) Same as B,F but with only the dots representing the terminal point. The unique cyan dot indicates the beginning point of the neuronal arbor. The white frame corresponds to the neuronal arbor reconstruction of the first time frame. (D,H) Maximal projections and neuronal arbor reconstruction of the same neuron over time. Images T1-T6 were taken every 10 minutes for 1 hour. Beginning point is in cyan, terminal points are in green, branching points are in red, and special branching points (multiple bifurcations) are in blue. Arrowheads indicate examples of transient neurites. Scale bars: 5 μm.

Four-dimensional reconstruction of a single labeled afferent neuron. (A-G) Four-dimensional representations of an afferent neuron in wild-type (A-C) and tmie mutant (E-G) fish. (A,E) mem-TdTomato-labeled neuron contacting the GFP-expressing hair cells. The terminal point of each neuron has been plotted with a different color. For each terminal point (i.e. for each color) positions have been plotted over time, every 10 minutes for 1 hour. (B,F) Same as A,E but without the hair cells. (C,G) Same as B,F but with only the dots representing the terminal point. The unique cyan dot indicates the beginning point of the neuronal arbor. The white frame corresponds to the neuronal arbor reconstruction of the first time frame. (D,H) Maximal projections and neuronal arbor reconstruction of the same neuron over time. Images T1-T6 were taken every 10 minutes for 1 hour. Beginning point is in cyan, terminal points are in green, branching points are in red, and special branching points (multiple bifurcations) are in blue. Arrowheads indicate examples of transient neurites. Scale bars: 5 μm.

Four-dimensional representations of a single labeled afferent neuron. Four-dimensional representations of single labeled afferent neuron in six wild-type (above) and tmie mutant (below) fish. The terminal point of each neuron has been plotted with a different color. For each terminal point (i.e. for each color) positions have been plotted over time, every 10 minutes for 1 hour.

Four-dimensional representations of a single labeled afferent neuron. Four-dimensional representations of single labeled afferent neuron in six wild-type (above) and tmie mutant (below) fish. The terminal point of each neuron has been plotted with a different color. For each terminal point (i.e. for each color) positions have been plotted over time, every 10 minutes for 1 hour.

Next, we compared arbor stability by counting the number of neurites present throughout the six time points imaged (stable neurites) and also those that were observed in only one to five of the time frames (transient neurites). In tmie mutants 40% of the neurites were transient, significantly higher than the 12% found in wild-type specimens (P<0.01) (Fig. 4A and see Fig. S1B in the supplementary material). We next counted the number of neurites that were visible during one to five of the time frames and defined ‘persistence’ as the ability of a neurite to remain present during a defined period. This analysis revealed that the majority (88%) of wild-type neurites was present throughout the whole recorded period, whereas only 59% of the neurites in tmie fish were present during all six time frames. We found that 14% of tmie neurites were present for only one time frame, compared with 7% in the wild type (Fig. 4B). Therefore, neurites of tmie mutants were less persistent.

We then probed wild-type and tmie neurites for their ‘consistency’. We defined consistency as the persistence of their behavior (Fig. 4C). For example, neurites that have neither grown nor retracted for the whole recorded period were classified as highly consistent, whereas those that performed a sequence of counter behaviors (e.g. growth then retraction) defined a lower degree of consistency. In tmie mutants, only 11% of neurites were absolutely consistent, whereas 42% grew or retracted, and 36%, 9% and 2% performed a sequence of two, three or four counter events, respectively. By comparison, in wild-type fish the respective values were 40%, 37%, 23%, 0% and 0%. Thus, neurites in tmie mutants are less consistent.

Quantitative analysis of neurite dynamics. (A) Average of stable (present throughout the six time frames) versus transient (present in fewer than six time frames) neurites in wild-type and tmie mutant fish. Error bars represent the s.e.m. *, P<0.01. (B) Neurite persistence. Persistence of neurites was calculated from the percentage of neurites present for only one, two, three, four, five or all six time frames. (C) Neurite consistency. ‘Consistent’ refers to a neurite that has neither grown nor retracted over the six time frames. ‘One event’ refers to a neurite that has only grown or retracted over the six time frames. ‘Two events’ refers to a neurite that completed a sequence of two counter movements, i.e. growth followed by retraction or vice versa. Similarly, ‘three events’ refers to a sequence of three counter movements. (D-F) Dynamics of neurites tips. The value of the terminal point position at the first time point was normalized to zero and the same value was subtracted from the subsequent time points in order to establish the distance traveled by the neurite tip. On the right of each graph is shown the average distance (μm) traveled by each type of neurite during one time frame. Stable neurites lack a bulged extremity but are present throughout the six time frames. (D) Dynamics of bulged neurites. (E) Dynamics of stable non-bulged neurites. (F) Dynamics of transient neurites. Red lines, wild type dashed green lines, tmie. Error bars represent s.e.m. *, ***, ****, P<0.01 **, P<0.1 for pairs indicated by number of asterisks wild type, n=6 tmie, n=6.

Quantitative analysis of neurite dynamics. (A) Average of stable (present throughout the six time frames) versus transient (present in fewer than six time frames) neurites in wild-type and tmie mutant fish. Error bars represent the s.e.m. *, P<0.01. (B) Neurite persistence. Persistence of neurites was calculated from the percentage of neurites present for only one, two, three, four, five or all six time frames. (C) Neurite consistency. ‘Consistent’ refers to a neurite that has neither grown nor retracted over the six time frames. ‘One event’ refers to a neurite that has only grown or retracted over the six time frames. ‘Two events’ refers to a neurite that completed a sequence of two counter movements, i.e. growth followed by retraction or vice versa. Similarly, ‘three events’ refers to a sequence of three counter movements. (D-F) Dynamics of neurites tips. The value of the terminal point position at the first time point was normalized to zero and the same value was subtracted from the subsequent time points in order to establish the distance traveled by the neurite tip. On the right of each graph is shown the average distance (μm) traveled by each type of neurite during one time frame. Stable neurites lack a bulged extremity but are present throughout the six time frames. (D) Dynamics of bulged neurites. (E) Dynamics of stable non-bulged neurites. (F) Dynamics of transient neurites. Red lines, wild type dashed green lines, tmie. Error bars represent s.e.m. *, ***, ****, P<0.01 **, P<0.1 for pairs indicated by number of asterisks wild type, n=6 tmie, n=6.

Finally, we measured the distance traveled by each neurite from the first time point. For a more accurate analysis, we subdivided them into three categories: bulged neurites that were present throughout the six time frames (bulged) neurites that were present throughout the six time frames but without bulging (stable) and neurites of low persistence (transient). Wild-type and mutant transient neurites traversed significantly longer distances during one time point than stable and bulged neurites (Fig. 4D-F).

To confirm that the observed effects were caused by the loss of hair cell activity, we applied the same strategy to animals injected with a morpholino (MO) against cadherin 23 (cdh23). Cdh23 morphants phenocopy the sputnik zebrafish strain, which bear a mutation in cdh23 that also causes deafness and lack of mechanoreception due to absent tip links (Sollner et al., 2004). The study was conducted at 4 dpf for maximal MO efficiency. Similar to tmie mutants, Cdh23 morphants showed more complex peripheral arbors than control fish. On average, we observed a 1.7-fold increase in the number of neurites, with 46% of neurites being transient in morphants versus 17% in controls. We also found that neurites were less persistent and less consistent in Cdh23 morphants (controls, n=3 cdh23 MO, n=3) (see Fig. S2 and Table S2 in the supplementary material). From these results, we conclude that the complexity of afferent neuron peripheral arbors is influenced by the sensory epithelium, and that arbor stability depends on evoked hair cell activity.

Central axons appear normal in profoundly deaf zebrafish

We next asked whether profound deafness would also influence neuron central axons. We assessed lateralis afferent central somatotopy and contacts with the Mauthner cell, a characteristic second-order target in the posterior hindbrain (Kimmel et al., 1982 Metcalfe et al., 1985). The lateral dendrite of the Mauthner cell has a dorsal projection at its lateral tip that is the target of the lateralis afferent projections (Fig. 5C). To reveal the Mauthner cell, we injected green fluorescent dextran into the spinal cord of tmie mutant larva. This dye is incorporated into the axon of the Mauthner cell and retrogradely spreads through the entire cell, including its two dendrites. We also injected magenta (in a terminal neuromast of the posterior lateral line) and red (in a supra-orbital neuromast of the anterior lateral line) fluorescent dextran to retrogradely label the afferent neurons that innervate these neuromasts (Sapède et al., 2005).

Central projections of first-order lateral line sensory neurons in tmie mutants. Central somatotopy and targeting of the Mauthner cell are conserved in tmie mutants. (Aa-d) Fluorescent dextran injections in tmie −/− brn3c:gfp transgenic fish showing somatotopy of first-order lateral line sensory neuron projections coming from posterior (b) and anterior (c) neuromasts. Projections coming from P9 (PLL) neuromasts (b Alexa 647-dextran) and SO2 (ALL) neuromasts (c Rhodamine-dextran) project in a dorsoventral manner (a) in the hindbrain. (d) Projections coming from the ear (Brn3c:GFP). (Ba-c) Fluorescent dextran injections in tmie −/− brn3c:gfp transgenic fish showing central targeting of the Mauthner lateral dendrite by posterior and anterior lateral line projections (a). Projections coming from the P9 (PLL) neuromast (b Alexa 647-dextran) connect (arrowheads) to the lateral-most portion of the Mauthner cell lateral dendrite (c fluorescein-dextran), which extends ventrodorsally. (C) Scheme (not to scale) depicting the central somatotopy of the first-order lateral line sensory neuron projections, and their targeting of the Mauthner cell lateral dendrite. In tmie fish, the anteroposterior axis of the lateral line neuromast projects onto the ventrodorsal axis of the Mauthner lateral dendrite dorsal projection, as in wild-type fish. ALL, anterior lateral line PLL, posterior lateral line.

Central projections of first-order lateral line sensory neurons in tmie mutants. Central somatotopy and targeting of the Mauthner cell are conserved in tmie mutants. (Aa-d) Fluorescent dextran injections in tmie −/− brn3c:gfp transgenic fish showing somatotopy of first-order lateral line sensory neuron projections coming from posterior (b) and anterior (c) neuromasts. Projections coming from P9 (PLL) neuromasts (b Alexa 647-dextran) and SO2 (ALL) neuromasts (c Rhodamine-dextran) project in a dorsoventral manner (a) in the hindbrain. (d) Projections coming from the ear (Brn3c:GFP). (Ba-c) Fluorescent dextran injections in tmie −/− brn3c:gfp transgenic fish showing central targeting of the Mauthner lateral dendrite by posterior and anterior lateral line projections (a). Projections coming from the P9 (PLL) neuromast (b Alexa 647-dextran) connect (arrowheads) to the lateral-most portion of the Mauthner cell lateral dendrite (c fluorescein-dextran), which extends ventrodorsally. (C) Scheme (not to scale) depicting the central somatotopy of the first-order lateral line sensory neuron projections, and their targeting of the Mauthner cell lateral dendrite. In tmie fish, the anteroposterior axis of the lateral line neuromast projects onto the ventrodorsal axis of the Mauthner lateral dendrite dorsal projection, as in wild-type fish. ALL, anterior lateral line PLL, posterior lateral line.

Tracing afferent neurons to the hindbrain showed that their central organization was unaffected in tmie mutants (n=7) (Fig. 5B and see Movie 5 in the supplementary material). In addition, the central projections from anterior and posterior neuromast afferents extended normally and were correctly placed along the somatotopic axis in the hindbrain (Fig. 5A). We observed that posterior and anterior lateral line central axons made contacts with the lateral dendrite of the Mauthner cell at the level of the dorsal projection (Fig. 5Ba and see Movie 5 in the supplementary material) as in wild-type controls (data not shown). In addition, live imaging at resolutions identical to those applied to peripheral arbors revealed no dynamic changes in the neuron central arbors (data not shown). These results show that the lack of evoked hair cell activity has no major impact on the central axonal organization of lateralis neurons.

Afferent neurons cannot discriminate target orientation in the absence of hair cell mechanoreception

Because deafness dramatically affected peripheral arborization, we asked whether evoked hair cell activity plays any role in the selection of plane-polarized targets by the afferent neurons. We scatter-labeled neurons and revealed hair cell orientation by labeling their apical stereocilia with fluorescent phalloidin (Fig. 6A,B). Homozygous tmie mutant fish develop hair cells with normal planar cell polarity (Fig. 6B) (Gleason et al., 2009). In wild-type fish, 88% of the bulged neurites associated exclusively with hair cells of a single polarity (n=17), significantly higher than the 38% found in tmie mutants (n=21) (Fig. 6C). In tmie mutants, 62% of bulged neurites were associated with hair cells of both polarities, an occurrence seen much less frequently in wild-type animals (12%). In tmie and wild-type specimens, bulged neurites established synapses with hair cells as revealed by the localized concentration of Ribeye B puncta (Fig. 6D-I) (wild type, n=6 tmie, n=7). Furthermore, several thin neurites also appeared to establish synapses with hair cells (Fig. 6J-L).

Mechanotransmission between a hair cell and its afferent neurons is mediated by glutamate release from the hair cell, which activates AMPA receptors in the neurons (Bailey and Sewell, 2000). To further probe hair cell polarity discrimination by the neurons, we attempted a pharmacological approach using glutamate receptor agonists and antagonists. We treated fish with CNQX (a competitive AMPA/kainate glutamate receptor antagonist), domoic acid (a potent agonist of AMPA/kainate glutamate receptors), kanaic acid (an agonist of the kainate/glutamate receptor), maleate (a non-competitive NMDA receptor antagonist) and cyclothiazide (an allosteric inhibitor of AMPA receptor desensitization that alters receptor gating). None of these drugs produced a phenotype in fish that was consistent with synaptic inhibition (data not shown), suggesting that they did not effectively impact their target in our in vivo experimental set-up. We also tried to electrically silence neuronal activity by expressing the inward-rectifier potassium ion channel Kir2.1 in lateralis afferents (Hua et al., 2005). Although Kir2.1 expression in afferent neurons did not affect their development, it had a profound impact on neuritogenesis and also led to late-onset neuronal degeneration (data not shown). Thus, this approach was also unsuccessful. In spite of this, we interpret our results from the profoundly deaf tmie mutants as a clear indication that evoked hair cell activity is necessary for afferent neurons to correctly select identically oriented targets.

Afferent neurons from tmie mutants are not strict polarity selectors. (A-C) Analysis of polarity-specific innervation in HuC:mem-TdTomato-injected brn3c:gfp tmie mutants. (A) Maximal projection and (B) phalloidin staining of a neuromast innervated by a single labeled afferent neuron in a tmie mutant. Arrows in A indicate bulged neurites establishing contacts with the hair cells. Asterisks in B indicate the innervated hair cells. (C) The percentage of contacts established between an afferent neuron and hair cells of identical polarity or hair cells of two opposing polarities. The numbers in brackets indicate the number of neuromasts (wild type, n=17 tmie, n=21). (D-L) Immunodetection of ribbon synapses in wild type (D-F) and tmie (G-L) fish using the Ribeye B antibody (cyan). Bulged contacts of afferent neurons expressing TdTomato (red) with hair cells (green) are associated with the presence of synapses in wild type (D-F) and tmie (G-I) fish (arrowheads indicate an example in each case). Thin neurites can also form synapses with hair cells as shown by the presence of Ribeye B staining in the neuromast of a tmie mutant (J-L). The boxed area in J is magnified in K,L. Scale bars: 10 μm in A,D,G,J 5 μm in B.

Afferent neurons from tmie mutants are not strict polarity selectors. (A-C) Analysis of polarity-specific innervation in HuC:mem-TdTomato-injected brn3c:gfp tmie mutants. (A) Maximal projection and (B) phalloidin staining of a neuromast innervated by a single labeled afferent neuron in a tmie mutant. Arrows in A indicate bulged neurites establishing contacts with the hair cells. Asterisks in B indicate the innervated hair cells. (C) The percentage of contacts established between an afferent neuron and hair cells of identical polarity or hair cells of two opposing polarities. The numbers in brackets indicate the number of neuromasts (wild type, n=17 tmie, n=21). (D-L) Immunodetection of ribbon synapses in wild type (D-F) and tmie (G-L) fish using the Ribeye B antibody (cyan). Bulged contacts of afferent neurons expressing TdTomato (red) with hair cells (green) are associated with the presence of synapses in wild type (D-F) and tmie (G-I) fish (arrowheads indicate an example in each case). Thin neurites can also form synapses with hair cells as shown by the presence of Ribeye B staining in the neuromast of a tmie mutant (J-L). The boxed area in J is magnified in K,L. Scale bars: 10 μm in A,D,G,J 5 μm in B.


Ongoing internal cortical activity plays a major role in perception and behavior both in animals and humans. Previously we have shown that spontaneous patterns resembling orientation-maps appear over large cortical areas in the primary visual-cortex of anesthetized cats. However, it remains unknown 1) whether spontaneous-activity in the primate also displays similar patterns and 2) whether a significant difference exists between cortical ongoing-activity in the anesthetized and awake primate. We explored these questions by combining voltage-sensitive-dye imaging with multiunit and local-field-potential recordings. Spontaneously emerging orientation and ocular-dominance maps, spanning up to 6 x 6 mm(2), were readily observed in anesthetized but not in awake monkeys. Nevertheless, spontaneous correlated-activity involving orientation-domains was observed in awake monkeys. Under both anesthetized and awake conditions, spontaneous correlated-activity coincided with traveling waves. We found that spontaneous activity resembling orientation-maps in awake animals spans smaller cortical areas in each instance, but over time it appears across all of V1. Furthermore, in the awake monkey, our results suggest that the synaptic strength had been completely reorganized including connections between dissimilar elements of the functional architecture. These findings lend support to the notion that ongoing-activity has many more fast switching representations playing an important role in cortical function and behavior.

Critical dynamics are thought to play an important role in neuronal information-processing: near critical networks exhibit neuronal avalanches, cascades of spatiotemporal activity that are scale-free, and are considered to enhance information capacity and transfer. However, the exact relationship between criticality, awareness, and information integration remains unclear. To characterize this relationship, we applied multi-scale avalanche analysis to voltage-sensitive dye imaging data collected from animals of various species under different anesthetics. We found that anesthesia systematically varied the scaling behavior of neural dynamics, a change that was mirrored in reduced neural complexity. These findings were corroborated by applying the same analyses to a biophysically realistic cortical network model, in which multi-scale criticality measures were associated with network properties and the capacity for information integration. Our results imply that multi-scale criticality measures are potential biomarkers for assessing the level of consciousness.

In cat early visual cortex, neural activity patterns resembling evoked orientation maps emerge spontaneously under anesthesia. To test if such patterns are synchronized between hemispheres, we performed bilateral imaging in anesthetized cats using a new improved voltage-sensitive dye. We observed map-like activity patterns spanning early visual cortex in both hemispheres simultaneously. Patterns virtually identical to maps associated with the cardinal and oblique orientations emerged as leading principal components of the spontaneous fluctuations, and the strength of transient orientation states was correlated with their duration, providing evidence that these maps are transiently attracting states. A neural mass model we developed reproduced the dynamics of both smooth and abrupt orientation state transitions observed experimentally. The model suggests that map-like activity arises from slow modulations in spontaneous firing in conjunction with interplay between excitation and inhibition. Our results highlight the efficiency and functional precision of interhemispheric connectivity.

Purpose: To review the clinical applications and diagnostic value of the retinal function imager (RFI), briefly compare RFI to other optical imaging devices, and to describe recent developments.Methods: The search words "Retinal Functional Imager," "optical imaging," "retina angiography," "avascular zone," "foveal avascular zone," and other closely related terms were used in PubMed to review current literature involving the RFI.Results: The functions of the RFI were utilized in over 44 microvascular studies, which reported that the microvasculature may alter in velocity, morphology, and oximetry when affected by a number of ocular, neurological, or systemic diseases. Recently developed automatic algorithms for noninvasive angiography of large retinal regions, segmenting vessels, measuring blood flow, blood velocity, vessel diameter, and oximetry may enhance the clinical applications of the RFI.Conclusion: The RFI has been used to characterize the retinal microvasculature under various conditions of all prevalent retinal diseases in addition to some central nerve system (CNS) and systemic diseases. Applying the RFI in research and clinical settings should help earlier diagnosis, support disease prevention, and improve treatment management.

Spontaneous internal activity plays a major role in higher brain functions. The question of how it modulates sensory evoked activity and behavior has been explored in anesthetized rodents, cats, monkeys and in behaving human subjects. However, the complementary question of how a brief sensory input modulates the internally generated activity in vivo remains unresolved, and high-resolution mapping of these bidirectional interactions was never performed. Integrating complementary methodologies, at population and single cells levels, we explored this question. Voltage-sensitive dye imaging of population activity in anesthetized rats' somatosensory cortex revealed that spontaneous up-states were largely diminished for similar to 2 s, even after a single weak whisker deflection. This effect was maximal at the stimulated barrel but spread across several cortical areas. A higher velocity whisker deflection evoked activity at similar to 15Hz. Two-photon calcium imaging activity and cell-attached recordings confirmed the VSD results and revealed that for several seconds most single cells decreased their firing, but a small number increased firing. Comparing single deflection with long train stimulation, we found a dominant effect of the first population spike. We suggest that, at the onset of a sensory input, some internal messages are silenced to prevent overloading of the processing of relevant incoming sensory information.

Purpose: Circulatory abnormalities in the retina, optic nerve and choroid have been detected by various technologies in glaucoma patients. However, there is no clear understanding of the role of blood flow in glaucoma. The purpose of this study was to compare retinal blood-flow velocities using the retinal function imager (RFI) between glaucoma and healthy subjects.Materials and Methods: Fifty-nine eyes of 46 patients with primary open-angle glaucoma (POAG), 51 eyes of 31 healthy individuals and 28 eyes of 23 patients with glaucomatous optic neuropathy (GON) but normal perimetry were recruited for this study. Three eyes of 2 patients in the glaucoma group and 2 eyes of 1 patient in the GON group had normal pressure at the time of diagnosis. Eighty-three percent of the glaucoma patients and 73% of the patients in the GON group were treated with anti-glaucoma medications. All patients were scanned by the RFI. Differences among groups were assessed by mixed linear models.Results: The average venous velocity in the GON group (3.8 mm/s) was significantly faster than in the glaucoma (3.3 mm/s, p = 0.03) and healthy (3.0 mm/s, p = 0.005) groups. The arterial velocity in the GON group was not different from any of the other study groups (4.7 mm/s). The arterial and venous velocity in the POAG eyes was not different than in the healthy eyes (arterial: 4.3 versus 4.2 mm/s, p = 0.7 venous: 3.3 versus 3.0 mm/s, p = 0.3). A subgroup of 13 glaucoma patients who had perimetric glaucoma in 1 eye and normal visual field (VF) in the fellow eye showed a trend of lower velocity in the glaucoma eyes.Conclusions: Changes in retinal blood-flow velocity were detected only in the pre-perimetric state, but not in perimetric glaucoma. These findings might represent early dysregulation in the retinal vasculature.

Extracting neuronal spiking activity from large-scale two-photon recordings remains challenging, especially in mammals in vivo, where large noises often contaminate the signals. We propose a method, MLspike, which returns the most likely spike train underlying the measured calcium fluorescence. It relies on a physiological model including baseline fluctuations and distinct nonlinearities for synthetic and genetically encoded indicators. Model parameters can be either provided by the user or estimated from the data themselves. MLspike is computationally efficient thanks to its original discretization of probability representations moreover, it can also return spike probabilities or samples. Benchmarked on extensive simulations and real data from seven different preparations, it outperformed state-of-the-art algorithms. Combined with the finding obtained from systematic data investigation (noise level, spiking rate and so on) that photonic noise is not necessarily the main limiting factor, our method allows spike extraction from large-scale recordings, as demonstrated on acousto-optical three-dimensional recordings of over 1,000 neurons in vivo.

Neural computations underlying sensory perception, cognition and motor control are performed by populations of neurons at different anatomical and temporal scales. Few techniques are currently available for exploring dynamics of local and large range populations. Voltage-sensitive dye imaging (VSDI) reveals neural population activity in areas ranging from a few tens of microns to a couple of centimeters, or two areas up to

10 cm apart. VSDI provides a sub-millisecond temporal resolution, and a spatial resolution of about 50 mum. The dye signal emphasizes subthreshold synaptic potentials. VSDI has been applied in the mouse, rat, gerbil, ferret, tree shrew, cat and monkey cortices, in order to explore lateral spread of retinotopic or somatotopic activation, the dynamic spatiotemporal pattern resulting from sensory activation, including the somatosensory, olfactory, auditory, and visual modalities, as well as motor preparation and the properties of spontaneously-occurring population activity. In this chapter we focus on VSDI in-vivo and review results obtained mostly in the visual system in our laboratory.

A central question in neuronal network analysis is how the interaction between individual neurons produces behavior and behavioral modifications. This task depends critically on how exactly signals are integrated by individual nerve cells functioning as complex operational units. Regional electrical properties of branching neuronal processes which determine the input-output function of any neuron are extraordinarily complex, dynamic, and, in the general case, impossible to predict in the absence of detailed measurements. To obtain such a measurement one would, ideally, like to be able to monitor, at multiple sites, subthreshold events as they travel from the sites of origin (synaptic contacts on distal dendrites) and summate at particular locations to influence action potential initiation. It became possible recently to carry out this type of measurement using high-resolution multisite recording of membrane potential changes with intracellular voltage-sensitive dyes. This chapter reviews the development and foundation of the method of voltage-sensitive dye recording from individual neurons. Presently, this approach allows monitoring membrane potential transients from all parts of the dendritic tree as well as from axon collaterals and individual dendritic spines.

The development of functional imaging techniques applicable to neuroscience and covering a wide range of spatial and temporal scales has greatly facilitated the exploration of the relationships between cognition, behaviour and electrical brain activity. For mammals, the neocortex plays a particularly profound role in generating sensory perception, controlling voluntary movement, higher cognitive functions and planning goal-directed behaviours. Since these remarkable functions of the neocortex cannot be explored in simple model preparations or in anesthetised animals, the neural basis of behaviour must be explored in awake behaving subjects. Because neocortical function is highly distributed across many rapidly interacting regions, it is essential to measure spatiotemporal dynamics of cortical activity in real-time. Extensive work in anesthetised mammals has shown that in vivo Voltage-Sensitive Dye Imaging (VSDI) reveals the neocortical population membrane potential dynamics at millisecond temporal resolution and subcolumnar spatial resolution. Here, we describe recent advances indicating that VSDI is also already well-developed for exploring cortical function in behaving monkeys and mice. The first animal model, the non-human primate, is well-suited for fundamental exploration of higher-level cognitive function and behavior. The second animal model, the mouse, benefits from a rich arsenal of molecular and genetic technologies. In the monkey, imaging from the same patch of cortex, repeatedly, is feasible for a long period of time, up to a year. In the rodent, VSDI is applicable to freely moving and awake head-restrained mice. Interactions between different cortical areas and different cortical columns can therefore now be dynamically mapped through VSDI and related to the corresponding behaviour. Thus by applying VSDI to mice and monkeys one can begin to explore how behaviour emerges from neuronal activity in neuronal networks residing in different cortical areas.

Purpose/Aim of the study: To study changes in retinal blood flow velocity in patients with early and neovascular age-related macular degeneration (AMD). We used the Retinal Function Imager (RFI, Optical Imaging Ltd., Rehovot, Israel), a noninvasive diagnostic approach for measuring blood flow velocity. Materials and Methods: Sixty eyes of 43 AMD patients and 53 eyes of 35 healthy individuals over the age of 50 were recruited for this study. All patients were scanned by the RFI with analysis of blood flow velocity of secondary and tertiary branches of arteries and veins. Differences among groups were assessed by mixed linear models. Results: The average velocity in AMD patients was significantly lower compared to controls in arteries (3.6 +/- 1.4 versus 4.3 +/- 1.0 mm/sec, p=0.009) but not in veins (2.6 +/- 0.9 versus 3.1 +/- 0.6 mm/sec, p = 0.08). When comparing the velocity between low-and high-grade AMD eyes, venous velocity was slower in the high grade AMD eyes only in the "narrow'' group of vessels. Conclusions: Decreased blood flow velocity in retinal arteries in patients with AMD was found. Despite the fact that AMD is essentially a choroidal disease, retinal vessels show a functional abnormality, which may suggest that the vascular abnormality in this disease is more generalized.

Measuring microvascular characteristics in cortical tissue and individual microvessels has important applications for functional imaging, biomedical research, and clinical diagnostics. Multiphoton fluorescence microscopy approaches are most effective and allow to reliably record red blood cell (RBC) velocity in individual vessels, but require injecting fluorescent tracers. Moreover, only one or few vessels in a small area can be imaged at a time. Wide-field CCD/CMOS-based optical imaging of intrinsic absorption or reflection changes in macroscopic vascular networks allows to overcome these shortcomings, by recording RBCs’ trajectories over several mm2 of cortical surface. The RBC velocity can then be extracted from these wide-field data using specialized algorithms. Here, we describe two of those, which provide robust RBC velocity estimations that are independent and can thus be used as a control one for another. Although this approach can be used in any part of the body with optically accessible blood vessels, here we show its application in two cases: first the cerebral cortex and then the eye. In this latter application, we go into some more detail in describing the retinal function imager (RFI): a unique, noninvasive multiparameter functional imaging instrument that directly measures hemodynamic parameters such as retinal RBC velocity, oximetric state, and metabolic responses to photic activation. In addition, it allows capillary perfusion mapping without any contrast agent. These parameters of retinal function are degraded by retinal abnormalities. Here, we thus focus on the characterization of microvessels properties. Indeed, clinical studies suggest that knowing these properties should yield multiple clinical applications for early diagnosis of retinal diseases, possible critical guidance of their treatment, as well as implications for vascular diseases of cortex and eye.

Fundamental understanding of higher cognitive functions can greatly benefit from imaging of cortical activity with high spatiotemporal resolution in the behaving non-human primate. To achieve rapid imaging of high-resolution dynamics of cortical representations of spontaneous and evoked activity, we designed a novel data acquisition protocol for sensory stimulation by rapidly interleaving multiple stimuli in continuous sessions of optical imaging with voltage-sensitive dyes. We also tested a new algorithm for the "temporally structured component analysis" (TSCA) of a multidimensional time series that was developed for our new data acquisition protocol, but was tested only on simulated data (Blumenfeld, 2010). In addition to the raw data, the algorithm incorporates prior knowledge about the temporal structure of the data as well as input from other information. Here we showed that TSCA can successfully separate functional signal components from other signals referred to as noise. Imaging of responses to multiple visual stimuli, utilizing voltage-sensitive dyes, was performed on the visual cortex of awake monkeys. Multiple cortical representations, including orientation and ocular dominance maps as well as the hitherto elusive retinotopic representation of orientation stimuli, were extracted in only 10 s of imaging, approximately two orders of magnitude faster than accomplished by conventional methods. Since the approach is rather general, other imaging techniques may also benefit from the same stimulation protocol. This methodology can thus facilitate rapid optical imaging explorations in monkeys, rodents and other species with a versatility and speed that were not feasible before. (C) 2013 Published by Elsevier Inc.

BACKGROUND AND OBJECTIVE: The Retinal Function Imager (RFI) (Optical Imaging Ltd., Rehovot, Israel) measures retinal blood flow velocity non-invasively. The authors studied the reproducibility of these measurements and assessed the effect of physiological components on them. PATIENTS AND METHODS: Sixty-seven individuals with no retinal pathology were recruited. Velocity reproducibility was verified by comparing repeated RFI measurements. The correlation of the velocity with physiological parameters was assessed by mixed linear and Gaussian models. RESULTS: The average velocity was 4.2 +/- 0.9 mm/sec arterial and 3.3 +/- 0.8 mm/sec venous. Variability was 7.5% +/- 3.7% and interclass correlation coefficient was r = 0.744. Venous velocity decreased after 40 years of age (0.32 mm/sec per decade, P <.01). Arterial velocity increased as mean arterial pressure increased (0.25 mm/sec per 10 mm Hg, P <.01). There was also a positive association between velocities and heart rate (arteries: 0.21 mm/sec per 10 bpm, P <.05 veins: 0.22 mm/sec per 10 bpm, P <.01). CONCLUSION: The RFI provides a reproducible, non-invasive technique to assess retinal velocities.

PURPOSE. To study the short-term effects of intravitreal bevacizumab (Avastin) on retinal blood flow velocity and compare them to clinical outcomes assessed by optical coherence tomography (OCT) and tests of visual acuity. METHODS. The Retinal Function Imager (RFI) was used noninvasively and quantitatively to measure retinal blood flow velocity. Eight patients receiving intravitreal injection of Avastin for choroidal neovascularization (CNV) were included in this study. All were imaged by the RFI preinjection and 1 and 7 days postinjection. Visual acuity (VA) and OCT were recorded preinjection and 1 month postinjection. Comparisons were performed using paired Student t test and correlation using Spearman rank test. RESULTS. A good correlation was found between the 1-month change in VA and OCT measurements and the short-term change induced in blood flow velocity. Arterial and venous velocity changes 1 day after the injection correlated with the VA change (p

Comprehensive information on the spatio-temporal dynamics of the vascular response is needed to underpin the signals used in hemodynamics-based functional imaging. It has recently been shown that red blood cells (RBCs) velocity and its changes can be extracted from wide-field optical imaging recordings of intrinsic absorption changes in cortex. Here, we describe a complete processing work-flow for reliable RBC velocity estimation in cortical networks. Several pre-processing steps are implemented: image co-registration, necessary to correct for small movements of the vasculature, semi-automatic image segmentation for fast and reproducible vessel selection, reconstruction of RBC trajectories patterns for each micro-vessel, and spatio-temporal filtering to enhance the desired data characteristics. The main analysis step is composed of two robust algorithms for estimating the RBCs' velocity field. Vessel diameter and its changes are also estimated, as well as local changes in backscattered light intensity. This full processing chain is implemented with a software suite that is freely distributed. The software uses efficient data management for handling the very large data sets obtained with in vivo optical imaging. It offers a complete and user-friendly graphical user interface with visualization tools for displaying and exploring data and results. A full data simulation framework is also provided in order to optimize the performances of the algorithm with respect to several characteristics of the data. We illustrate the performance of our method in three different cases of in vivo data. We first document the massive RBC speed response evoked by a spreading depression in anesthetized rat somato-sensory cortex. Second, we show the velocity response elicited by a visual stimulation in anesthetized cat visual cortex. Finally, we report, for the first time, visually-evoked RBC speed responses in an extended vascular network in awake monkey extrastriate cortex. (C) 2011 E

Purpose: To compare retinal blood flow velocity in small vessels of patients with early diabetes mellitus (DM), without any morphologic changes related to diabetic retinopathy, with that in a control group. Methods: The authors used the retinal function imager to measure blood flow velocities, from many small vessels, simultaneously. Twenty-three eyes of 14 patients with early DM and 51 eyes of 31 healthy subjects were enrolled. Differences between the patients and the control group were assessed by mixed linear models. Results: Venous average velocity significantly increased in the DM group (3.8 +/- 1.2 vs. 2.9 +/- 0.5 mm/second, P <0.0001) than in the healthy subjects. Arterial velocity of DM patients was also significantly higher (4.7 +/- 1.7 vs. 4.1 +/- 0.9 mm/second, P = 0.03). There was no statistically significant difference between groups in age, gender, heart rate, and systolic blood pressure. The diastolic blood pressure in the DM patients was lower than that in the healthy group (P = 0.03). Conclusion: There was an increase in arterial and venous retinal blood flow velocities of patients with early DM with no diabetic retinopathy. These findings support the notion that abnormalities in vessel function exist in diabetic eyes before the development of structural changes. This noninvasive approach facilitated the assessment of early hemodynamic abnormalities and may assist in screening and monitoring. RETINA 32:112-119, 2012

Pyramidal cells in layers 2 and 3 of the neocortex of many species collectively form a clustered system of lateral axonal projections (the superficial patch system-Lund JS, Angelucci A, Bressloff PC. 2003. Anatomical substrates for functional columns in macaque monkey primary visual cortex. Cereb Cortex. 13:15-24. or daisy architecture-Douglas RJ, Martin KAC. 2004. Neuronal circuits of the neocortex. Annu Rev Neurosci. 27:419-451.), but the function performed by this general feature of the cortical architecture remains obscure. By comparing the spatial configuration of labeled patches with the configuration of responses to drifting grating stimuli, we found the spatial organizations both of the patch system and of the cortical response to be highly conserved between cat and monkey primary visual cortex. More importantly, the configuration of the superficial patch system is directly reflected in the arrangement of function across monkey primary visual cortex. Our results indicate a close relationship between the structure of the superficial patch system and cortical responses encoding a single value across the surface of visual cortex (self-consistent states). This relationship is consistent with the spontaneous emergence of orientation response-like activity patterns during ongoing cortical activity (Kenet T, Bibitchkov D, Tsodyks M, Grinvald A, Arieli A. 2003. Spontaneously emerging cortical representations of visual attributes. Nature. 425:954-956.). We conclude that the superficial patch system is the physical encoding of self-consistent cortical states, and that a set of concurrently labeled patches participate in a network of mutually consistent representations of cortical input.

A neural field model is presented that captures the essential non-linear characteristics of activity dynamics across several millimeters of visual cortex in response to local flashed and moving stimuli. We account for physiological data obtained by voltage-sensitive dye (VSD) imaging which reports mesoscopic population activity at high spatio-temporal resolution. Stimulation included a single flashed square, a single flashed bar, the line-motion paradigm - for which psychophysical studies showed that flashing a square briefly before a bar produces sensation of illusory motion within the bar - and moving squares controls. We consider a two-layer neural field (NF) model describing an excitatory and an inhibitory layer of neurons as a coupled system of non-linear integro-differential equations. Under the assumption that the aggregated activity of both layers is reflected by VSD imaging, our phenomenological model quantitatively accounts for the observed spatio-temporal activity patterns. Moreover, the model generalizes to novel similar stimuli as it matches activity evoked by moving squares of different speeds. Our results indicate that feedback from higher brain areas is not required to produce motion patterns in the case of the illusory line-motion paradigm. Physiological interpretation of the model suggests that a considerable fraction of the VSD signal may be due to inhibitory activity, supporting the notion that balanced intra-layer cortical interactions between inhibitory and excitatory populations play a major role in shaping dynamic stimulus representations in the early visual cortex.

Purpose: The purpose of this study was to compare the retinal blood flow velocities of patients with diabetes and healthy control subjects. We used a novel device offering a noninvasive diagnostic of retinal function. Methods: Flow velocities in retinal arterioles and venules were quantitatively analyzed by retinal function imager scanning in 58 eyes of 42 patients with nonproliferative diabetic retinopathy and 51 eyes of 32 normal subjects. Group differences were assessed by the mixed-model effect. Results: Average velocity in arterial compartments (in mm/s) was 3.74 +/- 1.09 for the diabetic group and 4.19 +/- 0.99 for the control subjects. The average velocity of all segments, taking associated heart rate and individual segment widths into account, was 17% slower in the diabetic group (P <0.0001). In both groups, average venous compartment velocity was lower than the arterial velocity (2.61 +/- 0.65 for the diabetic group 3.03 +/- 0.59 for the control subjects). Individual vein velocities, taking heart rate and segment widths into account, was 17% slower, on average, in the diabetic group (P <0.0001). Conclusion: Our measurement showed significantly decreased flow velocities in the retinal arterioles and venules of patients with diabetes compared with healthy control subjects, supporting the view of abnormal vessel function in eyes with nonproliferative diabetic retinopathy. RETINA 30:765-773, 2010

In primary visual areas any local input is initially transmitted via horizontal connections giving rise to a transient peak of activity with spreading surround. How does this scenario change when the stimulus starts to move? Psychophysical experiments indicate that localization is different for stationary flashed and moving objects depending on the stimulus history. We here demonstrate how successively presented stimuli alter cortical activation dynamics. By a combination of electrophysiological and optical recordings using voltage-sensitive dye we arrive at the conclusion that sub-threshold propagating activity pre-activates cortical regions far ahead of thalamic input. Such an anticipatory mechanism may contribute in shifts of the perceived position as observed for the flash-lag effect and line-motion illusion in human psychophysics.

Arousal patently transforms the faculties of complex organisms. Although typical changes in cortical activity such as seen in EEG and LFP measurements are associated with change in state of arousal, it remains unclear what in the constitution of such state dependent activity enables this profound enhancement of ability. We put forward the hypothesis that arousal modulates cortical activity by rendering it more fit to represent information. We argue that representational capacity is of a dual nature-it requires not only that cortical tissue generate complex activity (i.e. spatiotemporal neuronal events), but also a complex cortical activity space (which is comprised of such spatiotemporal events). We explain that the topological notion of complexity-homology-is the pertinent measure of the complexity of neuronal activity spaces, as homological structure indicates not only the degree to which underlying activity is inherently clustered but also registers the effective dimensionality of the configurations formed by such clusters. Changes of this sort in the structure of cortical activity spaces can serve as the basis of the enhanced capacity to make perceptual/behavioral distinctions brought about by arousal. To show the feasibility of these ideas, we analyzed voltage sensitive dye imaging (VSDI) data acquired from primate visual cortex in disparate states of arousal. Our results lend some support to the theory: first as arousal increased so did the complexity of activity (that is the complexity of VSDI movies). Moreover, the complexity of structure of activity space (that is VSDI movie space) as measured by persistent homology-a multi scale topological measure of complexity-increased with arousal as well.

The Retinal Function Imager (RFI Optical Imaging, Rehovot, Israel) is a unique, noninvasive multiparameter functional imaging instrument that directly measures hemodynamic parameters such as retinal blood-flow velocity, oximetric state, and metabolic responses to photic activation. In addition, it allows capillary perfusion mapping without any contrast agent. These parameters of retinal function are degraded by retinal abnormalities. This review delineates the development of these parameters and demonstrates their clinical applicability for noninvasive detection of retinal function in several modalities. The results suggest multiple clinical applications for early diagnosis of retinal diseases and possible critical guidance of their treatment.

Functional maps obtained by various technologies, including optical imaging techniques, f-MRI, PET, and others, may be contaminated with biological artifacts such as vascular patterns or large patches of parenchyma. These artifacts originate mostly from changes in the microcirculation that result from either activity-dependent changes in volume or from oximetric changes that do not co-localize with neuronal activity per se. Standard methods do not always suffice to reduce such artifacts, in which case conspicuous spatial artifacts mask details of the underlying activity patterns. Here we propose a simple algorithm that efficiently removes spatial biological artifacts contaminating high-resolution functional maps. We validated this procedure by applying it to cortical maps resulting from optical imaging, based either on voltage-sensitive dye signals or on intrinsic signals. To remove vascular spatial patterns we first constructed a template of typical artifacts (vascular/cardiac pulsation/vasomotion), using principle components derived from baseline information obtained in the absence of stimulation. Next, we modified this template by means of local similarity minimization (LSM), achieved by measuring neighborhood similarity between contaminated data and the artifact template and then abolishing the similarity. LSM thus removed spatial patterns originating from the cortical vasculature components, including large fields of capillary parenchyma, helping to unveil details of neuronal activity patterns that were otherwise masked by these vascular artifacts. Examples obtained from our imaging experiments with anaesthetized cats and behaving monkeys showed that the LSM method is both general and reproducible, and is often superior to other available procedures. (C) 2008 Elsevier B.V. All rights reserved.

In the neocortex, neurons with similar response properties are often clustered together in column-like structures, giving rise to what has become known as functional architecture-the mapping of various stimulus feature dimensions onto the cortical sheet. At least partially, we owe this finding to the availability of several functional brain imaging techniques, both post-mortem and in-vivo, which have become available over the last two generations, revolutionizing neuroscience by yielding information about the spatial organization of active neurons in the brain. Here, we focus on how our understanding of such functional architecture is linked to the development of those functional imaging methodologies, especially to those that image neuronal activity indirectly, through metabolic or haemodynamic signals, rather than directly through measurement of electrical activity. Some of those approaches allow exploring functional architecture at higher spatial resolution than others. In particular, optical imaging of intrinsic signals reaches the striking detail of similar to 50 mu m, and, together with other methodologies, it has allowed characterizing the metabolic and haemodynamic responses induced by sensory-evoked neuronal activity. Here, we review those findings about the spatio-temporal characteristics of neurovascular coupling and discuss their implications for functional brain imaging, including position emission tomography, and non-invasive neuroimaging techniques, such as funtional magnetic resonance imaging, applicable also to the human brain.

Little is known about the "inverse" of the receptive field-the region of cortical space whose spatiotemporal pattern of electrical activity is influenced by a given sensory stimulus. We refer to this activated area as the cortical response field, the properties of which remain unexplored. Here, the dynamics of cortical response fields evoked in visual cortex by small, local drifting-oriented gratings were explored using voltage-sensitive dyes. We found that the cortical response field was often characterized by a plateau of activity, beyond the rim of which activity diminished quickly. Plateau rim location was largely independent of stimulus orientation. However, approximately 20 ms following plateau onset, 1-3 peaks emerged on it and were amplified for 25 ms. Spiking was limited to the peak zones, whose location strongly depended on stimulus orientation. Thus, alongside selective amplification of a spatially restricted suprathreshold response, wider activation to just below threshold encompasses all orientation domains within a well-defined retinotopic vicinity of the current stimulus, priming the cortex for processing of subsequent stimuli.

In the vertebrate brain external stimuli are often represented in distinct functional domains distributed across the cortical surface. Fast imaging techniques used to measure patterns of population activity record movies with many pixels and many frames, i.e., data sets with high dimensionality. Here we demonstrate that principal component analysis (PCA) followed by spatial independent component analysis (sICA), can be exploited to reduce the dimensionality of data sets recorded in the olfactory bulb and the somatosensory cortex of mice as well as the visual cortex of monkeys, without loosing the stimulus-specific responses. Different neuronal populations are separated based on their stimulus-specific spatiotemporal activation. Both, spatial and temporal response characteristics can be objectively obtained, simultaneously. In the olfactory bulb, groups of glomeruli with different response latencies can be identified. This is shown for recordings of olfactory receptor neuron input measured with a calcium-sensitive axon tracer and for network dynamics measured with the voltage-sensitive dye RH 1838. In the somatosensory cortex, barrels responding to the stimulation of single whiskers can be automatically detected. In the visual cortex orientation columns can be extracted. In all cases artifacts due to movement, heartbeat or respiration were separated from the functional signal by sICA and could be removed from the data set. sICA following PCA is therefore a powerful technique for data compression, unbiased analysis and dissection of imaging data of population activity, collected with high spatial and temporal resolution. (c) 2006 Elsevier Inc. All rights reserved.

Optical imaging, positron emission tomography, and functional magnetic resonance imaging ( fMRI) all rely on vascular responses to image neuronal activity. Although these imaging techniques are used successfully for functional brain mapping, the detailed spatiotemporal dynamics of hemodynamic events in the various microvascular compartments have remained unknown. Here we used highresolution optical imaging in area 18 of anesthetized cats to selectively explore sensory- evoked cerebral blood- volume ( CBV) changes in the various cortical microvascular compartments. To avoid the confounding effects of hematocrit and oximetry changes, we developed and used a new fluorescent blood plasma tracer and combined these measurements with optical imaging of intrinsic signals at a near- isosbestic wavelength for hemoglobin ( 565 nm). The vascular response began at the arteriolar level, rapidly spreading toward capillaries and venules. Larger veins lagged behind. Capillaries exhibited clear blood- volume changes. Arterioles and arteries had the largest response, whereas the venous response was smallest. Information about compartment- specific oxygen tension dynamics was obtained in imaging sessions using 605 nm illumination, a wavelength known to reflect primarily oximetric changes, thus being more directly related to electrical activity than CBV changes. Those images were radically different: the response began at the parenchyma level, followed only later by the other microvascular compartments. These results have implications for the modeling of fMRI responses ( e. g., the balloon model). Furthermore, functional maps obtained by imaging the capillary CBV response were similar but not identical to those obtained using the early oximetric signal, suggesting the presence of different regulatory mechanisms underlying these two hemodynamic processes.

Cortical maps and feedback connections are ubiquitous features of the visual cerebral cortex. The role of the feedback connections, however, is unclear. This study was aimed at revealing possible organizational relationships between the feedback projections from area V2 and the functional maps of orientation and retinotopy in area V1. Optical imaging of intrinsic signals was combined with cytochrome oxidase histochemistry and connectional anatomy in owl monkeys. Tracer injections were administered at orientation-selective domains in regions of pale and thick cytochrome oxidase stripes adjacent to the border between these stripes. The feedback projections from V2 were found to be more diffuse than the intrinsic horizontal connections within V1, but they nevertheless demonstrated clustering. The clusters of feedback axons projected preferentially to interblob cytochrome oxidase regions. The distribution of preferred orientations of the recipient domains in V1 was broad but appeared biased toward values similar to the preferred orientation of the projecting cells in V2. The global spatial distribution of the feedback projections in V1 was anisotropic. The major axis of anisotropy was systematically parallel to a retinotopic axis in V1 corresponding to the preferred orientation of the cells of origin in V2. We conclude that the feedback connections from V2 to V1 might play a role in enhancing the response in V1 to collinear contour elements.

Advancement in the treatment of blindness depends on the development of new technologies that enable early detection, follow-up, and treatment of disease. The authors describe direct, noninvasive imaging of four parameters: blood flow, blood oximetry, metabolic state, and hidden vasculature, particularly capillaries. These are functional parameters of the retina known to be degraded by retinal disease. The new Retinal Function Imager (Optical Imaging, Ltd., Rehovot, Israel) can image all four parameters as intrinsic reflectance intensity differences over the retina's surface. During, the past 2 decades, imaging of small optical signals has been a powerful tool for high-resolution functional mapping in the neocortex. In this article : this technology is applied to the retina and demonstrates a general tool for noninvasively probing retinal function in many modalities. Imaging functional changes before anatomic consequences arise holds promise as a powerful tool for early diagnosis and treatment of retinal disease.

This chapter shows that the spontaneous firing of single neurons is tightly linked to the cortical networks in which they are embedded. The idea of a network is a central concept in theoretical brain research and it is now finally possible to directly visualize the cortical networks and their states in action, at high spatiotemporal resolution.

During the last few decades, neuroscientists have benefited from the emergence of many powerful functional imaging techniques that cover broad spatial and temporal scales. We can now image single molecules controlling cell differentiation, growth and death single cells and their neurites processing electrical inputs and sending outputs neuronal circuits performing neural computations in vitro and the intact brain. At present, imaging based on voltage-sensitive dyes (VSDI) offers the highest spatial and temporal resolution for imaging neocortical functions in the living brain, and has paved the way for a new era in the functional imaging of cortical dynamics. It has facilitated the exploration of fundamental mechanisms that underlie neocortical development, function and plasticity at the fundmental level of the cortical column.

The ultimate goal of high-resolution functional brain mapping is single-condition (stimulus versus no-stimulus maps) rather than differential imaging (comparing two "stimulus maps"), because the appropriate ("orthogonal") stimuli are rarely available. This requires some component(s) of activity-dependent hemodynamic signals to closely colocalize with electrical activity, like the early increase in deoxyhemoglobin, shown previously to yield high-quality functional single-condition maps. Conversely, nonlocal vascular responses dominate in cerebral blood volume (CBV)based single-condition maps. Differential CBV maps are largely restricted to the parenchyma, implying that part of the CBV response does colocalize with electrical activity at fine spatial scale. By removing surface vascular activation from optical imaging data, we document the existence of a capillary CBV response component, regulated at fine spatial scale and yielding single-condition maps exhibiting similar to100 mum resolution. Blood volume and -flow based single-condition functional mapping at columnar level should thus be feasible, provided that the capillary response component is selectively imaged.

Exploring visual illusions reveals fundamental principles of cortical processing. Illusory motion perception of non-moving stimuli was described almost a century ago by Gestalt psychologists(1,2). However, the underlying neuronal mechanisms remain unknown. To explore cortical mechanisms underlying the 'linemotion' illusion(3), we used real-time optical imaging(4-6), which is highly sensitive to subthreshold activity. We examined, in the visual cortex of the anaesthetized cat, responses to five stimuli: a stationary small square and a long bar a moving square a drawnout bar and the well-known line-motion illusion(3), a stationary square briefly preceding a long stationary bar presentation. Whereas flashing the bar alone evoked the expected localized, short latency and high amplitude activity patterns(7,8), presenting a square 60 - 100 ms before a bar induced the dynamic activity patterns resembling that of fast movement. The preceding square, even though physically non-moving, created gradually propagating subthreshold cortical activity that must contribute to illusory motion, because it was indistinguishable from cortical representations of real motion in this area. These findings demonstrate the effect of spatio-temporal patterns of subthreshold synaptic potentials on cortical processing and the shaping of perception.

The rodent primary somatosensory cortex is spontaneously active in the form of locally synchronous membrane depolarizations (UP states) separated by quiescent hyperpolarized periods (DOWN states) both under anesthesia and during quiet wakefulness. In vivo whole-cell recordings and tetrode unit recordings were combined with voltage-sensitive dye imaging to analyze the relationship of the activity of individual pyramidal neurons in layer 2/3 to the ensemble spatiotemporal dynamics of the spontaneous depolarizations. These were either brief and localized to an area of a barrel column or occurred as propagating waves dependent on local glutamatergic synaptic transmission in layer 2/3. Spontaneous activity inhibited the sensory responses evoked by whisker deflection, accounting almost entirely for the large trial-to-trial variability of sensory-evoked postsynaptic potentials and action potentials. Subthreshold sensory synaptic responses evoked while a cortical area was spontaneously depolarized were smaller, briefer and spatially more confined. Surprisingly, whisker deflections evoked fewer action potentials during the spontaneous depolarizations despite neurons being closer to threshold. The ongoing spontaneous activity thus regulates the amplitude and the time-dependent spread of the sensory response in layer 2/3 barrel cortex.

Spontaneous cortical activity - ongoing activity in the absence of intentional sensory input - has been studied extensively(1), using methods ranging from EEG (electroencephalography)(2-4), through voltage sensitive dye imaging(5-7), down to recordings from single neurons(8,9). Ongoing cortical activity has been shown to play a critical role in development(10-14), and must also be essential for processing sensory perception, because it modulates stimulus-evoked activity(5,15,16), and is correlated with behaviour(17). Yet its role in the processing of external information and its relationship to internal representations of sensory attributes remains unknown. Using voltage sensitive dye imaging, we previously established a close link between ongoing activity in the visual cortex of anaesthetized cats and the spontaneous firing of a single neuron(6). Here we report that such activity encompasses a set of dynamically switching cortical states, many of which correspond closely to orientation maps. When such an orientation state emerged spontaneously, it spanned several hypercolumns and was often followed by a state corresponding to a proximal orientation. We suggest that dynamically switching cortical states could represent the brain's internal context, and therefore reflect or influence memory, perception and behaviour.

Sensory processing and its perception require that local information would also be available globally. Indeed, in the mammalian neocortex, local excitation spreads over large distances via the long-range horizontal connections in layer 2/3 and may spread over an entire cortical area if excitatory polysynaptic pathways are also activated. Therefore, a balance between local excitation and surround inhibition is required. Here we explore the spatiotemporal aspects of cortical depolarization and hyperpolarization of rats anesthetized with urethane. New voltage-sensitive dyes (VSDs) were used for high-resolution real-time visualization of the cortical responses to whisker deflections and cutaneous stimulations of the whisker pad. These advances facilitated imaging of ongoing activity and evoked responses even without signal averaging. We found that the motion of a single whisker evoked a cortical response exhibiting either one or three phases. During a triphasic response, there was first a cortical depolarization in a small cortical region the size of a single cortical barrel. Subsequently, this depolarization increased and spread laterally in an oval manner, preferentially along rows of the barrel field. During the second phase, the amplitude of the evoked response declined rapidly, presumably because of recurrent inhibition. Subsequently, the third phase exhibiting a depolarization rebound was observed and clear, and similar to16 Hz oscillations were detected. Stimulus conditions revealing a net surround hyperpolarization during the second phase were also found. By using new, improved VSD, the present findings shed new light on the spatial parameters of the intricate spatiotemporal cortical interplay of inhibition and excitation.

Spontaneous cortical activity of single neurons is often either dismissed as noise, or is regarded as carrying no functional significance and hence is ignored. Our findings suggest that such concepts should be revised. We explored the coherent population activity of neuronal assemblies in primary sensory area in the absence of a sensory input. Recent advances in real-time optical imaging based on voltage-sensitive dyes (VSDI) have facilitated exploration of population activity and its intimate relationship to the activity of individual cortical neurons. It has been shown by in vivo intracellular recordings that the dye signal measures the sum of the membrane potential changes in all the neuronal elements in the imaged area, emphasizing subthreshold synaptic potentials and dendritic action potentials in neuronal arborizations originating from neurons in all cortical layers whose dendrites reach the superficial cortical layers. Thus, the VSDI has allowed us to image the rather illusive activity in neuronal dendrites that cannot be readily explored by single unit recordings. Surprisingly, we found that the amplitude of this type of ongoing subthreshold activity is of the same order of magnitude as evoked activity. We also found that this ongoing activity exhibited high synchronization over many millimeters of cortex. We then investigated the influence of ongoing activity on the evoked response, and showed that the two interact strongly. Furthermore, we found that cortical states that were previously associated only with evoked activity can actually be observed also in the absence of stimulation, for example, the cortical representation of a given orientation may appear without any visual input. This demonstration suggests that ongoing activity may also play a major role in other cortical function by providing a neuronal substrate for the dependence of sensory information processing on context, behavior, memory and other aspects of cognitive function.

Optical imaging studies of orientation and direction preference in visual cortex have typically used vector averaging to obtain angle and magnitude maps. This method has shown half-rotation orientation singularities (pinwheels) located within regions of low orientation vector magnitude. Direction preference is generally orthogonal to orientation preference, but often deviates from this, particularly in regions of low direction vector magnitude. Linear regions of rapid change in direction preference terminate in or near orientation singularities. The vector-averaging method is problematic however because it does not clearly disambiguate spatial variation in orientation tuning width from variation in height. It may also wrongly estimate preferred direction in regions where preference is weak. In this paper we analyze optical maps of cat visual cortex by fitting model tuning functions to the responses. This new method reveals features not previously evident. Orientation tuning height and width vary independently across the map: tuning height is always low near singularities, however regions of broad and narrow orientation tuning width can be found in regions of low tuning height, often alternating in a spoke-like fashion around singularities. Orientation and direction preference angles are always closely orthogonal. Reversals in direction preference form lines that originate precisely in orientation singularities.

The spatiotemporal dynamics of the sensory response in layer 2/3 of primary somatosensory cortex evoked by a single brief whisker deflection was investigated by simultaneous voltage-sensitive dye (VSD) imaging and whole-cell (WC) voltage recordings in the anesthetized rat combined with reconstructions of dendritic and axonal arbors of L2/3 pyramids. Single and dual WC recordings from pyramidal cells indicated a strong correlation between the local VSD population response and the simultaneously measured subthreshold postsynaptic potential changes in both amplitude and time course. The earliest VSD response was detected 10-12 msec after whisker deflection centered above the barrel isomorphic to the stimulated principal whisker. It was restricted horizontally to the size of a single barrel-column coextensive with the dendritic arbor of barrel-column-related pyramids in L2/3. The horizontal spread of excitation remained confined to a single barrel-column with weak whisker deflection. With intermediate deflections, excitation spread into adjacent barrel-columns, propagating twofold more rapidly along the rows of the barrel field than across the arcs, consistent with the preferred axonal arborizations in L2/3 of reconstructed pyramidal neurons. Finally, larger whisker deflections evoked excitation spreading over the entire barrel field within similar to50 msec before subsiding over the next similar to250 msec. Thus the subthreshold cortical map representing a whisker deflection is dynamic on the millisecond time scale and strongly depends on stimulus strength. The sequential spatiotemporal activation of the excitatory neuronal network in L2/3 by a simple sensory stimulus can thus be accounted for primarily by the columnar restriction of L4 to L2/3 excitatory connections and the axonal field of barrel-related pyramids.

A novel method of chronic optical imaging based on new voltage-sensitive dyes (VSDs) was developed to facilitate the explorations of the spatial and temporal patterns underlying higher cognitive functions in the neocortex of behaving monkeys. Using this system, we were able to explore cortical dynamics, with high spatial and temporal resolution, over period of less than or equal to1 yr from the same patch of cortex. The visual cortices of trained macaques were stained one to three times a week, and immediately after each staining session, the monkey started to perform the behavioral task, while the primary and secondary visual areas (V1 and V2) were imaged with a fast optical imaging system. Long-term repeated VSD imaging (VSDI) from the same cortical area did not disrupt the normal cortical architecture as confirmed repeatedly by optical imaging based on intrinsic signals. The spatial patterns of functional maps obtained by VSDI were essentially identical to those obtained from the same patch of cortex by imaging based on intrinsic signals. On comparing the relative amplitudes of the evoked signals and differential map obtained using these two different imaging methodologies, we found that VSDI emphasizes subthreshold activity more than imaging based on intrinsic signals, that emphasized more spiking activity. The latency of the VSD-evoked response in V1 ranged from 46 to 68 ms in the different monkeys. The amplitude of the V2 response was only 20-60% of that in V1 As expected from the anatomy, the retinotopic responses to local visual stimuli spread laterally across the cortical surface at a spreading velocity of 0.15-0.19 m/s over a larger area than that expected by the classical magnification factor, reaching its maximal anisotropic spatial extent within similar to40 ms. We correlated the observed dynamics of cortical activation patterns with the monkey's saccadic eye movements and found that due to the slow offset of the cortical response relative to its ons

To facilitate the combination of optical imaging with various electrode-based techniques, we have designed and produced a skull-mounting 'sliding-top cranial window' and a removable 'electrode positioner microdrive'. These new devices were used to study sensory processing in chronic and acute experiments in the cerebral cortices of cats and monkeys. This assembly allows simultaneous optical imaging of intrinsic signals or voltage-sensitive dyes combined with extracellular recording (single and multiple unit recording and local field potential), intracellular recording, microstimulation, or targeted injection of tracers. After the functional architecture is determined by optical imaging, electrodes are targeted into a selected cortical site under full visual control, at a variety of penetration angles (30-90degrees), accessing a large cortical area. The device consists of three parts: (1) a skull-mounting chamber, (2) a sliding cap, and (3) a microdrive. The microdrive can easily be removed and the cranial window is then sealed and covered with a flat protective cover. For chronic experiments, this arrangement allows the animal to be handled over a long period while fitted with a sealed cranial window of minimal volume and weight, and with negligible risk of accidental damage or infection. (C) 2002 Elsevier Science B.V. All rights reserved.

We explored the spatio-temporal dynamics of odor-evoked activity in the rat and mouse main olfactory bulb (MOB) using voltage-sensitive dye imaging (VSDI) with a new probe. The high temporal resolution of VSDI revealed odor-specific sequences of glomerular activation. Increasing odor concentrations reduced response latencies, increased response amplitudes, and recruited new glomerular units. However, the sequence of glomerular activation was maintained. Furthermore, we found distributed MOB activity locked to the nasal respiration cycle. The spatial distribution of its amplitude and phase was heterogeneous and changed by sensory input in an odor-specific manner. Our data show that in the mammalian olfactory bulb, odor identity and concentration are represented by spatio-temporal patterns, rather than spatial patterns alone.

We present a transparent silicone dural substitute, which we have been using for the last 7 years for imaging cortical dynamics in awake behaving monkeys. This substitute enabled us to record optically for more than a year intrinsic or voltage sensitive dye signals. It is thin and elastic enough to allow microelectrode to pass through without any damage, using full visual control to target the electrode to the desirable recording site. This implant has proved crucial for maintaining the cortex in a good physiological condition and for preserving its optical characteristics that are necessary for optical imaging. We describe the details of the surgical implantation of the silicone dural substitute, the maintenance of the exposed cortex over long periods of time, the cortical reaction to this implant and its possible clinical implications in humans, and the rehabilitation procedure in monkeys. (C) 2002 Elsevier Science B.V. All rights reserved.

The frontal eye field and neighboring area 8Ar of the primate cortex are involved in programming and execution of saccades. Electrical microstimulation in these regions elicits short-latency contralateral saccades. To determine how spatiotemporal dynamics of microstimulation-evoked activity are converted into saccade plans, we used a combination of real-time optical imaging and microstimulation in behaving monkeys. Short stimulation trains evoked a rapid and widespread wave of depolarization followed by unexpected large and prolonged hyperpolarization. During this hyperpolarization saccades are almost exclusively ipsilateral, suggesting an important role for hyperpolarization in determining saccade goal.

How does the high selectivity to stimulus orientation emerge in the visual cortex? Thalamic feedforward-dominated models of orientation selectivity predict constant selectivity during the visual response, whereas intracortical recurrent models predict dynamic improvement in selectivity. We imaged the cat visual cortex with voltage-sensitive dyes to measure orientation-tuning dynamics of a large neuronal population. Tuning-curve width did not narrow after response onset, whereas the difference between preferred and orthogonal responses (modulation depth) first increased, then declined. We identified a suppression of the evoked responses, referred to as the evoked deceleration-acceleration (DA) notch, which was larger for the orthogonal response, Furthermore, peak selectivity of the tuning curves was contemporaneous with the evoked DA notch. These findings suggest that in the cat brain, sustained visual cortical processing does not narrow orientation tuning rather, intracortical interactions may amplify modulation depth and suppress the orthogonal response relatively more than the preferred, Thus, feedforward models and recurrent models of orientation selectivity must be combined.

Understanding of the spatio-temporal characteristics of the sensory-evoked cortical blood-volume and oxygenation changes is important from the physiological perspective as well as for the interpretation of results obtained by various neuroirnaging techniques, such as optical imaging, PET and f-MRI, and for their improvement. The detailed picture, however, has remained elusive for more than a century. We investigated the blood-volume and oxygenation changes in anesthetized cats and awake monkeys using intrinsic imaging at isosbestic and other wavelengths, laser Doppler, imaging spectroscopy, phosphorescence quenching and fluorescence imaging of activity-dependent responses of intravenously injected extrinsic probes. We found that the onset of blood-volurne changes was delayed (> 300 ms) with respect to a fast decrease in blood oxygenation. Thus, the blood-volume effects cannot merely explain the "initial dip". 570-nm measurements and high-resolution imaging (80 ms, 7 pm) of a fluorescent tracer injected into the blood circulation facilitated the resolution of the responses of different microvascular compartments. Preliminary results show that the arterioles led the blood volume increase, rapidly spreading towards the other microvascular compartments. Veins lagged behind. Functional maps of stimulus vs. blank (single condition maps) in a conscious macaque and an anesthetized cat were obtained only during the early deoxygenation phase at 605 nm. At later times or at 570 nm, the vessel artifacts dominated. These results indicated a stronger co-localization of oxygen consumption and electrical activity as compared to the subsequent volume and flow increase, which are not well regulated at the cortical column level. (C) 2002 Elsevier Science B.V. All rights reserved.

High-resolution images of the somatotopic hand representation in macaque monkey primary somatosensory cortex (area S-I) were obtained by optical imaging based on intrinsic signals. To visualize somatotopic maps, we imaged optical responses to mild tactile stimulation of each individual fingertip. The activation evoked by stimulation of a single finger was strongest in a narrow transverse band (similar to1 x 4 mm) across the postcentral gyrus. As expected, a sequential organization of these bands was found. However, a significant overlap, especially for the activated areas of fingers 3-5, was found. Surprisingly, in addition to the finger-specific domains, we found that for each of the fingers, weak stimulation activated also a second "common patch" of cortex, located just medially to the representation of the finger. These results were confirmed by targeted multiunit and single-unit recordings guided by the optical maps. The maps remained very stable over many hours of recording. By optimizing the imaging procedures, we were able to obtain the functional maps extremely rapidly (e.g., the map of five fingers in the macaque monkey could be obtained in as little as 5 min). Furthermore, we describe the intraoperative optical imaging of the hand representation in the human brain during neurosurgery and then discuss the implications of the present results for the spatial resolution accomplishable by other neuroimaging techniques, relying on responses of the microcirculation to sensory-evoked electrical activity. This study demonstrates the feasibility of using high-resolution optical imaging to explore reliably short- and long-term plasticity of cortical representations, as well as for applications in the clinical setting.

Explorations of learning and memory, other long-term plastic changes, and additional cognitive functions in the behaving primate brain would greatly benefit from the ability to image the functional architecture within the same patch of cortex, at the columnar level, for a long period of time. We developed methods for long-term optical imaging based on intrinsic signals and repeatedly visualized the same functional domains in behaving macaque cortex for a period extending over 1 year. Using optical imaging and imaging spectroscopy, we first explored the relationship between electrical activity and hemodynamic events in the awake behaving primate and compared it with anesthetized preparations. We found that, whereas the amplitude of the intrinsic signal was much larger in the awake animal, its temporal pattern was similar to that observed in the anesthetized animals. In both groups, deoxyhemoglobin concentration reached a peak 2-3 sec after stimulus onset. Furthermore, the early activity-dependent increase in deoxyhemoglobin concentration (the "initial dip") was far more tightly colocalized with electrical activity than the delayed increase in oxyhemoglobin concentration, known to be associated with an increase in blood flow. The implications of these results for improvement of the spatial resolution of blood oxygenation level-dependent functional magnetic resonance imaging are discussed. After the characterization of the intrinsic signal in the behaving primate, we used this new imaging method to explore the stability of cortical maps in the macaque primary visual cortex. Functional maps of orientation and ocular dominance columns were found to be stable for a period longer than 1 year.

Cat visual cortex contains a topographic map of visual space, plus superimposed, spatially periodic maps of ocular dominance, spatial frequency and orientation. It is hypothesized that the layout of these maps is determined by two constraints: continuity or smooth mapping of stimulus properties across the cortical surface, and coverage uniformity or uniform representation of combinations of map features over visual space. Here we use a quantitative measure of coverage uniformity (c') to test the hypothesis that cortical maps are optimized for coverage. When we perturbed the spatial relationships between ocular dominance, spatial frequency and orientation maps obtained in single regions of cortex, we found that cortical maps are at a local minimum for c'. This suggests that coverage optimization is an important organizing principle governing cortical map development.

Revealing the layout of cortical maps is important both for understanding the processes involved in their development and for uncovering the mechanisms underlying neural computation. The typical organization of orientation maps in the cat visual cortex is radial complete orientation cycles are mapped around orientation singularities. In contrast, long linear zones of orientation representation have been detected in the primary visual cortex of the tree shrew. In this study, we searched for the existence of long linear sequences and wide linear zones within orientation preference maps of the Eat visual cortex. optical imaging based on intrinsic signals was used. Long linear sequences and wide linear zones of preferred orientation were occasionally detected along the border between areas 17 and 18, as well as within area 18. Adjacent zones of distinct radial and linear organizations were observed across area fs of a single hemisphere. However, radial and linear organizations were not necessarily segregated long (7.5 mm) linear sequences of preferred orientation were found embedded within a typical pinwheel-like organization of orientation. We conclude that, although the radial organization is dominant, perfectly linear organization may develop and perform the processing related to orientation in the cat visual cortex.

The relation between the activity of a single neocortical neuron and the dynamics of the network in which it is embedded was explored by single-unit recordings and real-time optical imaging. The firing rate of a spontaneously active single neuron strongly depends on the instantaneous spatial pattern of ongoing population activity in a Large cortical area. Very similar spatial patterns of population activity were observed both when the neuron fired spontaneously and when it was driven by its optimal stimulus. The evoked patterns could be used to reconstruct the spontaneous activity of single neurons.

Conventional imaging techniques have provided high-resolution imaging either in the spatial domain or in the temporal domain. Optical imaging utilizing voltage-sensitive dyes has long had the unrealized potential to achieve high resolution in both domains simultaneously, providing subcolumnar spatial detail with millisecond precision. Here, we present a series of developments in voltage-sensitive dyes and instrumentation that make functional imaging of cortical dynamics practical, in both anesthetized and awake behaving preparations, greatly facilitating exploration of the cortex. We illustrate this advance by analyzing the millisecond-by-millisecond emergence of orientation maps in cat visual cortex.

Modern functional brain mapping relies on interactions of neuronal electrical activity with the cortical microcirculation. The existence of a highly localized, stimulus-evoked initial deoxygenation has remained a controversy. Here, the activity-dependent oxygen tension changes in the microcirculation were measured directly, using oxygen-dependent phosphorescence quenching of an exogenous indicator. The first event after sensory stimulation was an increase in oxygen consumption, followed by an increase in blood flow. Because oxygen consumption and neuronal activity are colocalized but the delayed blood flow is not, functional magnetic resonance imaging focused on this initial phase will yield much higher spatial resolution, ultimately enabling the noninvasive visualization of fundamental processing modules in the human brain.

A number of new imaging techniques are available to scientists to visualize the functioning brain directly, revealing unprecedented details. These imaging techniques have provided a new level of understanding of the principles underlying cortical development, organization and function. In this chapter we will focus on optical imaging in the living mammalian brain, using two complementary imaging techniques. The first technique is based on intrinsic signals. The second technique is based on voltage-sensitive dyes. Currently, these two optical imaging techniques offer the best spatial and temporal resolution, but also have inherent limitations. We shall provide a few examples of new findings obtained mostly in work done in our laboratory. The focus will be upon the understanding of methodological aspects which in turn should contribute to optimal use of these imaging techniques. General reviews describing earlier work done on simpler preparations have been published elsewhere (Cohen, 1973 Tasaki and Warashina, 1976 Waggoner and Grinvald, 1977 Waggoner, 1979 Salzberg, 1983 Grinvald, 1984 Grinvald et al., 1985 De Weer and Salzberg, 1986 Cohen and Lesher, 1986 Salzberg et al., 1986 Loew, 1987 Orbach, 1987 Blasdel, 1988, 1989 Grinvald et al., 1988 Kamino, 1991 Cinelli and Kauer, 1992 Frostig, 1994).

Modern functional neuroimaging methods, such as positron-emission tomography (PET), optical imaging of intrinsic signals, and functional MRI (fMRI) utilize activity dependent hemodynamic changes to obtain indirect maps of the evoked electrical activity in the brain. Whereas PET and flow-sensitive MRI map cerebral blood flow (CBF) changes, optical imaging and blood oxygenation level-dependent MRI map areas with changes in the concentration of deoxygenated hemoglobin (HbR), However, the relationship between CBF and HbR during functional activation has never been tested experimentally, Therefore, we investigated this relationship by using imaging spectroscopy and laser-Doppler flowmetry techniques, simultaneously, in the visual cortex of anesthetized cats during sensory stimulation, We found that the earliest microcirculatory change was indeed an increase in HbR, whereas the CBF increase lagged by more than a second after the increase in HbR, The increased HbR was accompanied by a simultaneous increase in total hemoglobin concentration (Hbt), presumably reflecting an early blood volume increase. We found that the CBF changes lagged after Hbt changes by 1 to 2 sec throughout the response, These results support the notion of active neurovascular regulation of blood volume in the capillary bed and the existence of a delayed, passive process of capillary filling.

In the primary visual cortex, neurons with similar response properties are arranged in columns. As more and more columnar systems are discovered it becomes increasingly important to establish the rules that govern the geometric relationships between different columns. As a first step to examine this issue we investigated the spatial relationships between the orientation, ocular dominance, and spatial frequency domains in cat area 17. Using optical imaging of intrinsic signals we obtained high resolution maps for each of these stimulus features from the same cortical regions. We found clear relationships between orientation and ocular dominance columns: many iso-orientation lines intersected the borders between ocular dominance borders at right angles, and orientation singularities were concentrated in the center regions of the ocular dominance columns. Similar, albeit weaker geometric relationships were observed between the orientation and spatial frequency domains. The ocular dominance and spatial frequency maps were also found to be spatially related: there was a tendency for the low spatial frequency domains to avoid the border regions of the ocular dominance columns. This specific arrangement of the different columnar systems might ensure that all possible combinations of stimulus features are represented at least once in any given region of the visual cortex, thus avoiding the occurrence of functional blind spots for a particular stimulus attribute in the visual field.

Spatial and temporal frequencies are important attributes of the visual scene. It is a long-standing question whether these attributes are represented in a spatially organized way in cat primary visual cortex(1-4). Using optical imaging of intrinsic signals(5-10), we show here that grating stimuli of different spatial frequencies drifting at various speeds produce distinct activity patterns. Rather than observing a map of continuously changing spatial frequency preference across the cortical surface, we found only two distinct sets of domains, one preferring low spatial frequency and high speed, and the other high spatial frequency and low speed. We compared the arrangement of these spatio-temporal frequency domains with the cytochrome oxidase staining pattern, which, based on work in primate striate cortex, is thought to reflect the partition of the visual cortex into different processing streams. We found that the cytochrome oxidase blobs in cat striate cortex coincide with domains engaged in the processing of low spatial and high temporal frequency contents of the visual scene. Together with other recent results(11), our data suggest that spatiotemporal frequency domains are a manifestation of parallel streams in cat visual cortex, with distinct patterns of thalamic inputs and extrastriate projections.


The goal of this study was to explore the functional organization of direction of motion in cat area 18. Optical imaging was used to record the activity of populations of neurons, We found a patchy distribution of cortical regions exhibiting preference for one direction over the opposite direction of motion, The degree of clustering according to preference of direction was two to four limes smaller than that observed for orientation. In general, direction preference changed smoothly along the cortical surface however, discontinuities in the direction maps were observed. These discontinuities formed lines that separated pairs of patches with preference for opposite directions. The functional maps for direction and for orientation preference were closely related typically, an iso-orientation patch was divided into regions that exhibited preference for opposite directions, orthogonal to the orientation. In addition, the lines of discontinuity within the direction map often connected points of singularity in the orientation map, Although the organization of both domains was related, the direction and the orientation selective responses were separable whereas the selective response according to direction of motion was nearly independent of the length of bars used for visual stimulation, the selective response to orientation decreased significantly with decreasing length of the bars. Extensive single and multiunit electrical recordings, targeted to selected domains of the functional maps, confirmed the features revealed by optical imaging, We conclude that significant processing of direction of motion is performed early in the cat visual pathway.

Evoked activity in the mammalian cortex and the resulting behavioral responses exhibit a large variability to repeated presentations of the same stimulus. This study examined whether the variability can be attributed to ongoing activity, Ongoing and evoked spatiotemporal activity patterns in the cat visual cortex were measured with real-time optical imaging local field potentials and discharges of single neurons were recorded simultaneously, by electrophysiological techniques. The evoked activity appeared deterministic, and the variability resulted from the dynamics of ongoing activity, presumably reflecting the instantaneous state of cortical networks, In spite of the large variability, evoked responses in single trials could be predicted by linear summation of the deterministic response and the preceding ongoing activity, Ongoing activity must play an important role in cortical function and cannot be ignored in exploration of cognitive processes.

An image enhancement method, wherein video analog signals are provided, a reference analog image is selected and the signals of the image are digitized, the digitized signals are converted to video analog reference signals, the reference signals are synchronized with video analog signals to be enhanced, the converted analog reference signals are subtracted from the video analog signals, and only then the subtracted analog signals are amplified, so as to obtain a sequence of enhanced images, in real time. There is also provided an image enhancement system.

1. We examined the spatiotemporal organization of ongoing activity in cat visual areas 17 and 18, in relation to the spontaneous activity of individual neurons. To search for coherent activity, voltage-sensitive dye signals were correlated with the activity of single neurons by the use of spike-triggered averaging. In each recording session an area of at least 2 X 2 mm of cortex was imaged, with 124 diodes. In addition, electrical recordings from two isolated units, the local field potential (LFP) from the same microelectrodes, and the surface electroencephalogram (EEG) were recorded simultaneously. 2. The optical signals recorded from the dye were similar to the LFP recorded from the same site. Optical signals recorded from different cortical sites exhibited a different time course. Therefore real-time optical imaging provides information that is equivalent in many ways to multiple-site LFP recordings. 3. The spontaneous firing of single neurons was highly correlated with the optical signals and with the LFP. In 88% of the neurons recorded during spontaneous activity, a significant correlation was found between the occurrence of a spike and the optical signal recorded in a large cortical region surrounding the recording site. This result indicates that spontaneous activity of single neurons is not an independent process but is time locked to the firing or to the synaptic inputs from numerous neurons, all activated in a coherent fashion even without a sensory input. 4. For the cases showing correlation with the optical signal, 27-36% of the optical signal during spike occurrence was directly related to the occurrence of spontaneous spikes in a single neuron, over an area of 2 X 2 mm. In the same cortical area, 43-55% of the activity was directly related to the visual stimulus. 5. Surprisingly, we found that the amplitude of this coherent ongoing activity, recorded optically, was often almost as large as the activity evoked by optimal visual stimulation. The ampli

We have used optical imaging based on intrinsic signals to explore the functional architecture of owl monkey area MT, a cortical region thought to be involved primarily in visual motion processing. As predicted by previous single-unit reports, we found cortical maps specific for the direction of moving visual stimuli. However, these direction maps were not distributed uniformly across all of area MT. Within the direction-specific regions, the activation produced by stimuli moving in opposite directions overlapped significantly. We also found that stimuli of differing shapes, moving in the same direction, activated different cortical regions within area MT, indicating that direction of motion is not the only parameter according to which area MT of owl monkey is organized. Indeed, we found clear evidence for a robust organization for orientation in area MT. Across all of MT, orientation preference changes smoothly, except at isolated line- or point-shaped discontinuities. Generally, paired regions of opposing direction preference were encompassed within a single orientation domain. The degree of segregation in the orientation maps was 3-5 times that found in direction maps. These results suggest that area MT, like V1 and V2, has a rich and multidimensional functional organization, and that orientation, a shape variable, is one of these dimensions.

Processing of retinal images is carried out in the myriad dendritic arborizations of cortical neurons. Such processing involves complex dendritic integration of numerous inputs, and the subsequent output is transmitted to multiple targets by extensive axonal arbors. Thus far, details of this intricate processing remained unexaminable. This report describes the usefulness of real-time optical imaging in the study of population activity and the exploration of cortical dendritic processing. In contrast to single-unit recordings, optical signals primarily measure the changes in transmembrane potential of a population of neuronal elements, including the often elusive subthreshold synaptic potentials that impinge on the extensive arborization of cortical cells. By using small visual stimuli with sharp borders and realtime imaging of cortical responses, we found that shortly after its onset, cortical activity spreads from its retinotopic site of initiation, covering an area at least 10 times larger, in upper cortical layers. The activity spreads at velocities from 100 to 250 mu m/msec. Near the V1/V2 border the direct activation is anisotropic and we detected also anisotropic spread the ''space constant'' for the spread was similar to 2.7 mm parallel to the border and similar to 1.5 mm along the perpendicular axis. In addition, we found cortical interactions between cortical activities evoked by a small ''center stimulus'' and by large ''surround stimuli'' positioned outside the classical receptive field. All of the surround stimuli used suppressed the cortical response to the center stimulus. Under some stimulus conditions iso-orientation suppression was more pronounced than orthogonal-orientation suppression. The orientation dependence of the suppression and its dependency on the size of some specific stimuli indicate that at least part of the center surround inhibitory interaction was of cortical origin. The findings reported here raise the possibility that distribu

Keywords: Anatomy & Morphology Developmental Biology Neurosciences

In primate primary visual cortex, neurons sharing similar response properties are clustered together forming functional domains that appear as a mosaic of patches or bands, often traversing the entire corticaL depth from the pia to the white matter. Similarly, each cortical site connects laterally through an extensive network of intrinsic projections that are organized in multiple clusters (patches) and reach distances of up to a few millimeters. The relationship between the functional domains and these lateraLly connected patches has remained a controversial issue despite intensive research efforts. To investigate this relationship, we obtained high-resolution functional maps of the cortical architecture by in vivo optical imaging. Subsequently, extracellular injections of the sensitive anterograde tracer biocytin were targeted into selected functional domains. Within the ocular dominance system, we found that long-range intrinsic connections tended to link the monocular regions of same-eye ocular dominance columns. Furthermore, we discovered that binocular domains formed a separate set of connections in area V1 binocular regions were selectively connected among themselves but were not connected to strictly monocular regions, suggesting that they constitute a distinct columnar system. In the other sub-system subserving orientation preference, patches of intrinsic connections tended to link domains sharing similar orientation preferences. Analyses of the precision of these connections indicated that in both functional subsystems,

In this study we used optical imaging based on activity-dependent intrinsic signals to determine the distribution of cells responding to gratings of various orientations moving in different directions in area 18 of cat visual cortex. To test directional-selective clustering of neurons, we compared cortical activity maps obtained by stimulation with two gratings of identical orientation but moving in opposite directions. We found those maps to be almost identical, suggesting that neurons are not notably clustered into directionality columns. We also compared activity maps obtained with gratings of different orientations. Each of the orientation maps was similar to the 2-deoxyglucose maps previously reported. Having compiled the information obtained from the different orientations into one ''orientation preference map,'' we found, in contrast to earlier reports, that iso-orientation domains are not elongated parallel bands but are small patches organized in ''pinwheels'' around points that we refer to as ''orientation centers.'' We furthermore show that the only locations at which orientation preference changes rapidly are these orientation centers and not lines or loops. In addition, this report clarifies that our observations on the functional architecture of cat area 18, although at first sight at variance with earlier observations, are actually fully consistent with them. We therefore propose that in cat visual cortex pinwheel-like patterns of orientation preference form an irregular mosaic of modular units with an average density of 1.2 pinwheels per square millimeter.

The relationships between cytochrome oxidase blobs, ocular-dominance columns, and iso-orientation domains, subsystems underlying visual perception, were explored in primary visual cortex of macaque monkey. High-resolution maps of these three subsystems were acquired. Optical imaging based on activity-dependent intrinsic signals revealed that the most prominent organizational feature of orientation preference was a radial arrangement, forming a pinwheel-like structure surrounding a singularity point. More than 80% of these pinwheels were centered along the midline of ocular-dominance columns. The iso-orientation contours of adjacent pinwheels crossed borders of ocular-dominance columns at approximately right angles. Pinwheels with the same or opposite directions of orientation-preference change were smoothly connected with each other. On the average, all orientations were equally represented. In exactly the same cortical area, the cytochrome oxidase blobs, thought to be involved in color processing, were also mapped, using cytochrome oxidase histology. Like the centers of pinwheels, the centers of blobs also lie along the midline of ocular-dominance columns. However, the centers of pinwwheels did not coincide with the centers of blobs these two subsystems are spatially independent. "Hypercolumn" modules, each including two complete pinwheels in two adjacent columns of complementary ocularity, as well as portions of a few blobs, were frequently found but did not seem to be the primary unit of cortical organization. An alternative to hypercolumns is proposed.

Long standing questions related to brain mechanisms underlying perception can finally be resolved by direct visualization of the architecture and function of mammalian cortex. This advance has been accomplished with the aid of two optical imaging techniques with which one can literally see how the brain functions. The upbringing of this technology required a multi-disciplinary approach integrating brain research with organic chemistry, spectroscopy, biophysics, computer sciences, optics and image processing. Beyond the technological ramifications, recent research shed new light on cortical mechanisms underlying sensory perception. Clinical applications of this technology for precise mapping of the cortical surface of patients during neurosurgery have begun. Below is a brief summary of our own research and a description of the technical specifications of the two optical imaging techniques. Like every technique, optical imaging also suffers from severe limitations. Here we mostly emphasize some of its advantages relative to all alternative imaging techniques currently in use. The limitations are critically discussed in our recent reviews. For a series of other reviews, see Cohen (1989).

Optical imaging of the functional architecture of cortex, based on intrinsic signals, is a useful tool for the study of the development, organization, and function of the living mammalian brain. This relatively noninvasive technique is based on small activity-dependent changes of the optical properties of cortex. Thus far, functional imaging has been performed only on anesthetized animals. Here we establish that this technique is also suitable for exploring the brain of awake behaving primates. We designed a chronic sealed chamber and mounted it on the skull of a cynomolgus monkey (Macaca fascicularis) over the primary visual cortex to permit imaging through a transparent glass window. Restriction of head position alone was sufficient to eliminate movement noise in awake monkey imaging experiments. High-resolution imaging of the ocular dominance columns and the cytochrome oxidase blobs was achieved simply by taking pictures of the exposed cortex when the awake monkey was viewing video movies alternatively with each eye. Furthermore, the functional maps could be obtained without synchronization of the data acquisition to the animal's respiration and the electrocardiogram. The wavelength dependency and time course of the intrinsic signal were similar in anesthetized and awake monkeys, indicating that the signal sources were the same. We therefore conclude that optical imaging is well suited for exploring functional organization related to higher cognitive brain functions of the primate as well as providing a diagnostic tool for delineating functional cortical borders and assessing proper functions of human patients during neurosurgery.

THE mammalian cortex is organized in a columnar fashion: neurons lying below each other from the pia to the white matter usually share many functional properties. Across the cortical surface, cells with similar response properties are also clustered together, forming elongated bands or patches. Some response properties, such as orientation preference in the visual cortex, change gradually across the cortical surface forming 'orientation maps'. To determine the precise layout of iso-orientation domains, knowledge of responses not only to one but to many stimulus orientations is essential. Therefore, the exact depiction of orientation maps has been hampered by technical difficulties and remained controversial for almost thirty years. Here we use in vivo optical imaging based on intrinsic signals to gather information on the responses of a piece of cortex to gratings in many different orientations. This complete set of responses then provides detailed information on the structure of the orientation map in a large patch of cortex from area 18 of the cat. We find that cortical regions that respond best to one orientation form highly ordered patches rather than elongated bands. These iso-orientation patches are organized around 'orientation centres', producing pinwheel-like patterns in which the orientation preference of cells is changing continuously across the cortex. We have also analysed our data for fast changes in orientation preference and find that these 'fractures' are limited to the orientation centres. The pinwheels and orientation centres are such a prominent organizational feature that it should be important to understand their development as well as their function in the processing of visual information.

The design of a macroscope constructed with photography lenses is described and several applications are demonstrated. The macroscope incorporates epi-illumination, a 0.4 numerical aperture, and a 40 mm working distance for imaging wide fields in the range of 1.5-20 mm in diameter. At magnifications of 1 x to 2.5 x, fluorescence images acquired with the macroscope were 100-700 times brighter than those obtained with commercial microscope objectives at similar magnifications. In several biological applications, the improved light collection efficiency (20-fold, typical) not only minimized bleaching effects, but, in concert with improved illumination throughput (15-fold, typical), significantly enhanced object visibility as well. Reduced phototoxicity and increased signal-to-noise ratios were observed in the in vivo real-time optical imaging of cortical activity using voltage-sensitive dyes. Furthermore, the macroscope has a depth of field which is 5-10 times thinner than that of a conventional low-power microscope. This shallow depth of field has facilitated the imaging of cortical architecture based on activity-dependent intrinsic cortical signals in the living primate brain. In these reflection measurements large artifacts from the surface blood vessels, which were observed with conventional lenses, were eliminated with the macroscope.

In this book questions regarding neuronal populations were primarily investigated using single- and multi-unit recordings with electrodes. The complexity of the functional organization which is revealed by these studies indicates a need for new methods which could obtain complementary information that is hard or impossible to obtain with microelectrodes. Imaging methods that provide high spatial or temporal resolution maps of the cortical organization of large cortical areas are of particular promise for studies of functional organization.

Optical measurements from rat optic nerve, loaded with the new Ca2+ indicator Fura-2, provide the first evidence for the presence of activity-dependent fast intracellular [Ca2+] transients in mammalian central nervous system (CNS) myelinated axons. The results suggest that voltage-dependent Ca2+ channels are present in some of the myelinated axons. Optical measurements from axons stained with anterogradely transported voltage-sensitive dye suggest the presence of Ca2+-dependent potassium conductances in these axons. This report also demonstrates that Fura-2 can readily detect changes in [Ca2+] inside cells as a result of electrical activity, and establishes its suitability for measurements of intracellular Ca2+ transients in the millisecond time domain.

The availability of suitable voltage-sensitive dyes and arrays of photodetectors has facilitated the optical monitoring of electrical activity simultaneously from hundreds of sites on the processes of single nerve cells, both in culture and in invertebrate ganglia. This method also provides a unique ability to detect activity in many individual neurons in an entire invertebrate ganglion controlling particular behavioral responses. The in-vitro activity of individual populations of neuronal elements (cell bodies, axons, dentrites or nerve terminals) at many neighboring loci in mammalian brain slices or isolated brain structures has been investigated. Recently dynamic patterns of electrical activity evoked in the intact vertebrate or mammalian brain by natural stimuli have also been monitored. By employing computerized optical recording and a display processor, video-displayed images of neuronal elements can be superimposed on the corresponding patterns of the optically detected electrical activity, thus allowing the spatio-temporal patterns of intracellular activity to be visualized in slow motion.

This chapter describes a novel approach to investigate the spatiotemporal distribution of electrical activity in nervous systems. Using voltage-sensitive dyes and an electro-optical measuring system, it has recently become possible to monitor electrical activity simultaneously from multiple sites on the processes of single nerve cells, either in culture or in an intact central nervous system (CNS) in vitro, to detect the activity of many individual neurons controlling a behavioral response in invertebrate ganglia, or to follow the activity of populations of neurons at many neighboring loci in mammalian brain slices or in the intact brain. Employing optical recordings and a display processor, the images of nerve cells light up on a TV monitor when they are electrically active. Thus, the spread of electrical activity can literally be visualized in slow motion. This chapter describes recent progress in the implementation of this new technique.


Basic principles

The advent of fluorescence microscopy has been a major step forward in the study of living cells. Leveraging the characteristic emissions of excited biological fluorophores, such as fluorescent proteins, it is possible to gain insight on cell structure and function (Figure 2). Following traditional fluorescence microscopy has been the development of multi-photon methods, where fluorophores are excited by two or more photons [35]. Multi-photon absorption is achieved with a single pulsed laser focused to a diffraction-limited spot on the specimen. With higher peak power, there is an increase in probability for multi-photon absorption leading to fluorophore excitation. Two-photon fluorescence is depicted in Figure 3a. To meet the excitation energy in this case, two 800 nm photons are used. One 400 nm photon is of equivalent energy, as can be used in single photon excitation, but with multi-photon methods only the area of the laser focus on the specimen is excited. Due to more focused excitation, there is a lower overall phototoxic effect. Also, as scattering of longer wavelength photons is less, multi-photon methods have deeper penetration when compared to single photon excitation.

Fluorescent protein applications. (a) Three Madin-Darby canine kidney epithelial cells with GFP-rac1 and dsRed-E-cadherin. Rac1 is a pleiotropic signaling molecule that is closely associated with cell-cell adhesion and cell motility. E-cadherin is a cell-cell adhesion protein responsible for facilitating communication between two contacting cells. Scale bar: 10 μm. Contributed by Lance Kam (Columbia University, New York). (b) Membranes of human umbilical cord endothelial cells visualized using EYFP. Scale bar: 40 μm. (c) GFP-actin labeled human umbilical cord endothelial cell undergoing mitosis, with actin filaments aligned toward the centrioles. Scale bar: 30 μm. Contributed by Samuel Sia (Columbia University, New York).

Two-photon microscopy of in vivo brain function. (a) Basic mechanism of two-photon fluorescence. (b) Schematic of surgical preparation of exposed cortex, with sealed glass window and microscope objective positioning. Green dot shows location of two-photon fluorescence. (c) Examples of two-photon maps of the vasculature following intravenous injection of dextran-conjugated fluorescein. Black dots and stripes show red blood cell motion. (d) Dual-channel imaging of neuronal (green) and vascular (red) signals: (left) Oregon Green 488 BAPTA-1 AM calcium sensitive dye stained neurons and (right) transgenic mouse expressing green fluorescent protein (GFP) in a subpopulation of neurons (mouse supplied by Jeffrey M. Friedman, Rockefeller University, New York) [101]. Texas dextran red is the intravascular tracer in both cases. (e) Three channel imaging of Tg2576 APP Alzheimer's disease mouse model with amyloid-targeting dye (blue), GFP expressing neurons and dendrites (green) and vasculature (red). Adapted from [52] and contributed by Elizabeth Hillman (Columbia University, New York).

In STED (Stimulated Emission Depletion) microscopy [36, 37], two pulsed lasers are used in tandem to break the diffraction barrier. The first laser pulse has a wavelength that excites fluorophores, and is immediately followed by a second laser pulse that depletes fluorescence. The fluorescence depletion is achieved as the wavelength of the second laser is tuned to be longer than the fluorescence emission. Absorption of a photon from the second laser induces electrons to drop to a lower energy level (stimulated emission) preventing typical fluorescence. The difference in area of the two focused beams leaves only a very small area from where fluorescence is detected. This area is smaller than a diffraction-limited spot. Using STED, images have been captured with a resolution of

30 nm [35]. In another study using rhomadine amide and a photoswitching technique, a resolution of 15 nm was achieved [38]. In [39, 40], an optical trapping system was used to make angstrom resolution measurements of pair based stepping of RNA polymerase, and thus has established an important resolution benchmark in molecular biology.

Electron microscopy has offered a resolution of

5 nm for imaging biological tissue [41]. However, to prepare a sample to be imaged by an electron microscope is a rigorous process that does not allow for imaging of live samples [42]. One common sample preparation technique is cryofixation, which is a high pressure and deep freezing technique that results in contrast in electron microscopy [43]. Even though the samples are no longer viable, electron microscopy has provided invaluable insights on the structural details of organelles and membranes [44].

Atomic force microscopes do not acquire information optically, but rather by recording intermolecular forces between a probe tip and a surface. The primary information acquisition component of an atomic force microscope is a cantilever with a nanometer-scale silicon tip. The tip is brought in close proximity with the sample and the deflection of the cantilever due to Van der Waals forces is recorded to generate a contour map of the sample surface [45, 46]. In comparison with an electron microscope, the sample does not need any special treatment that would actually destroy the sample and prevent its reuse. However, using contact or tapping mode of an atomic force microscope, which impinges on the sample surface to acquire measurements such as strain, can mechanically damage cells and tissue. Other probe microscopy techniques include scanning tunneling microscopy and near-field scanning optical microscopy [47].

Current applications

Fluorescence microscopy is often used in systems biology and there is a strong push for the development of high throughput methods. In the application of genome-wide RNAi screens to document the phenotype for each suppressed gene [48, 49], there can be millions of images from a single screen which can amount to several terabytes of data [50]. It is systems biology modeling that relieves the bottleneck of processing this large amount of RNAi screen image data by providing an efficient means of classification. With high throughput microscopy there is much more data generated than can be annotated or evaluated manually, and so developing a fine-tuned and efficient classification model is paramount for unlocking the potential of high throughput methods.

As seen in Figure 2, fluorescent proteins can be used to visualize many functional and structural aspects of cells. Using multi-photon methods as well can provide insight on cell structural and biochemical changes [51]. Multi-photon methods have a wide array of applications including in vivo brain imaging in animals (Figure 3) [52–54], where cortical micro-architecture has been investigated with single cell resolution [55]. Electron microscopes have been used to elucidate macromolecule structure [41]. As for cancer applications, atomic force microscopes have been used to monitor the super-coiled state of DNA, which is preferential to the binding of the tumor suppressing protein p53 [56]. Also in regard to detecting levels of p53 in cells, fluorescence microscopy has been used to determine the effectiveness of oncolytic adenoviruses. Specially designed oncolytic adenoviruses target cancerous tissues and are programmed to replicate if the cellular p53 level is low. Viral oncolytic therapy is an intense research area for cancer treatment and as microscopy techniques advance so will the ability to assess the effectiveness of viral vectors for tumor ablation [57, 58].

Advantages and limitations

It has been long held that the wave nature of light imposes a seemingly fundamental limit on the resolving power of a microscope. The limitation was approximately half the wavelength of visible light or 200 nm. Recently, there has been over a 10-fold resolution improvement with advances in microscopy [35, 38]. However, optical techniques have limited penetration as light readily scatters in tissue. This can be partially ameliorated by using more powerful lasers, but this in turn can lead to increased photobleaching effects which can limit the amount of time that an experiment can run.

Microscopy, as with other modern imaging techniques, has become ever more dependent on software for image acquisition and analysis. Imaging technology can be enhanced or limited by the software it is coupled with. Table 2 contains an overview of current microscopy image analysis software. With advances in acquisition algorithms and optics holographic microscopy has been achieved, by which full three-dimensional information can be acquired in a single image [59, 60]. As a result, volumetric time series data can be collected without the need of changing focus and scanning multiple z-planes.


Here, we studied the spatiotemporal patterns and mechanisms of WD in the cerebral cortex. We combined in vivo 2-photon time-lapse imaging of a fluorescent membrane tracer through a cranial window and laser-mediated lesions to track the full extent of WD for individual unmyelinated cortical axonal arbours (Fig. 1). We provide new insights into cortical axon degeneration: (1) segments below

600 μm in length consistently undergo a faster form of WD (roWD) (2) synaptic density, but not arbor complexity, contributes to the timing of cortical axon WD (3) AAD is negligible in lesioned cortical axons and (4) similar to WD in other parts of the nervous system and AAD, an NAD + -dependent signalling pathway, independent of transcription, delays cortical axon roWD and WD.

In vivo optical imaging of spinal cord injury has provided unique insights into the mechanisms of axon degeneration and regeneration in recent years [20, 23, 58]. However, we are only now starting to exploit this tool to understand the principles of axon degeneration and regeneration in the mammalian brain [37,38,39, 59,60,61]. Limitations of optical methods through glass windows to induce local damage include the inability to model more complex types of injury conditions such as those found in traumatic brain injury or multiple sclerosis and the difficulty of getting therapeutic compounds into the brain in order to manipulate the local environment in vivo [62]. There are three main advantages of laser-mediated axonal transection compared to previously reported methods. Firstly, laser microsurgery reveals the entire degeneration process as close as 4–10 μm from the injury site along to the terminal ending of the axon [13, 38]. In contrast, following mechanical or needlestick injury, anatomical distortions mean that axonal changes can only be observed at a distance (> 100 μm) from the injury site. Secondly, local axonal injury can be induced without rupturing the dura mater and overlying blood vessels, decreasing contributions from inflammatory mediators, thereby reducing the complexity of cellular interactions and improving experimental reproducibility [38, 63]. Thirdly, several neurons, or even parts of a neuron [25], can be simultaneously targeted with minimal damage to the surrounding tissue [13, 36, 38, 64,65,66,67]. The combination of this lesion method with the selective monitoring of the axolemma, rather than the axoplasm as in previous studies, allowed us to accurately measure the timing of the execution phases of axon degeneration, ruling out instances of cytoplasmic fragmentation not directly accompanied by loss of membrane integrity (as reported previously [4, 15], and as our data suggest it is the case for AAD, Fig. 5). These results indicate that axoplasmic components such as microtubules and neurofilaments are the first to change in WD, while the axonal membrane persists for longer time, in agreement with these previous reports [4, 15]. We expect that this paradigm and its refinements, for example by simultaneously imaging cytoskeletal [68], mitochondria [69] and membrane markers, will be useful to investigate axonal degeneration in other mouse models of disease (e.g. Alzheimer’s disease) or human-mouse chimeric models, which employ patient-specific genetic backgrounds [31, 70].

To our knowledge, ours is the first study to tackle the challenge of monitoring the spatiotemporal dynamics, as well as shedding light on the mechanisms, of WD in the mammalian cerebral cortex at single axon resolution. We used a labelling strategy targeted mainly to excitatory unmyelinated axons originating in cortical layer 2/3/5/6 and the thalamus, because of their vulnerability in several neurodegenerative and acute injury conditions. The reported degeneration kinetics fall within published time frames for WD in the murine spinal cord [20], peripheral nerves [71] and optic nerve [18], but reveal that transected cortical axon segments <

600 μm have a different spatial and temporal degeneration progression pattern compared to both WD and AAD. Even though only a small percentage (< 1%) of the axonal arbor when lesioned undergoes roWD, roWD affects up to 2–3 times more axon length than AAD [20] (see Additional file 5).

In earlier work in zebrafish larvae, the length of the disconnected axonal segment was not a critical parameter, not affecting the duration of the lag phase (which we here call onset) and only slightly influencing the fragmentation rate [13]. The difference in fragmentation rates between WD and roWD in mammalian cortical neurons (Fig. 2d, g) could reflect the natural variability of the degeneration process across species and the nervous system. It is well established that degeneration kinetics are variable between axons in adults [11], and even more so during development. For example, axotomised trigeminal axons in the developing zebrafish show embryonic stage-dependent differences in the time of the degeneration onset and a distinct lack of AAD [13]. Other axon subtypes, such as those of the zebrafish lateral line projections, do undergo a process resembling AAD in that a gap rapidly forms on both sides of the lesion, although no evidence of fragmentation was detected [29].

For cortical axons, the timing of both the initiation (onset, Fig. 2d, e) and execution phase (fragmentation kinetics, Figs. 2d and 3c) differs depending on a cut-off length of approximately 600 μm. Our data are not consistent with a linear length-based mechanism to explain the onset of cortical axon degeneration as suggested by others who find that longer disconnected segments take longer to commence degeneration, for example in transected zebrafish trigeminal axons [13]. A ‘threshold-dependent’ mechanism likely controls cortical roWD and WD as illustrated in an additional file (see Additional file 2). It could be either a ‘protective factor’ or a ‘degeneration trigger factor’ level-based threshold. For a ‘protective factor’ level-based threshold, the amount of a degeneration inhibitor (e.g. NAD + , or neuro-protective NAD synthesising enzyme Nmnat2 [14, 72, 73]) would decrease [74] below a critical threshold level before termination of the broken membrane resealing process (which can occur within 30 min post injury in vitro [75]), triggering rapid onset of fragmentation. In longer segments, the membrane from the cut end would reseal before depletion, and as such, these fragments persist. For a ‘degeneration trigger factor’ level-based threshold, a critical amount of a trigger such as calcium influx (e.g. via downstream effectors such as calpains [76]), mediated by extracellular calcium influx through the cut end [77, 78], would be rapidly reached in shorter segments which would therefore fragment quickly compared to longer segments. The rapid removal of short, damaged axonal segments from the neuropil may have evolved as a mechanism to rapidly restore brain homeostasis and keep the neuropil free from debris.

Our results on the regulation of cortical roWD and WD by NAD + -dependent mechanisms (Fig. 6) are consistent with previous data of delayed dopaminergic and striatal synaptic degeneration in Wld S mice [79, 80], providing additional in vivo evidence that Wld S -based neuroprotective therapeutics may be developed not only for myelinated but also for unmyelinated, grey matter fibre degeneration. These results demonstrate further conservation of the molecular mechanisms controlling WD, despite different spatiotemporal patterns of degeneration in different areas of the mammalian nervous system [26].

Cortical axons have more delayed degeneration onset the higher the density of synaptic connections (Fig. 4). Synapse loss in neurodegeneration can occur independently of axonal and cell body loss, suggesting that different mechanisms regulate the degeneration of these compartments [81,82,83]. We hypothesise that the presence of synapses limits the depletion of axon protective factors or accumulation of degeneration triggers, possibly through either slower diffusion along the axon, or due to being crosslinked with synaptic machinery, prolonging the lag phase before fragmentation.

In the future, the in vivo optical imaging approach used here could lead to a better understanding of the molecular and cellular mechanisms that regulate cortical axon degeneration, a necessary step in the development of effective therapies to counteract or stabilise the progression of axon and synaptic loss in both the diseased and injured brain.

Pushing Biology Forward with Super Resolution

Super resolution microscopy has revolutionized our understanding of biology by shedding light on the molecular world beyond the diffraction limit. Traditional light microscopy is limited to about 200 nm resolution under ideal circumstances, where resolution is roughly one-half the wavelength of emitted light. Everything smaller than this limit appears as a diffraction limited spot.

The diffraction limit (d) is related to the wavelength of emitted light (λ) and the numerical aperture (NA) of the objective.

Super resolution microscopy has revealed previously unknown structures in biology and shown us the dynamic behavior of individual molecules. Super resolution microscopy uses physical tricks to see beyond the diffraction limit, including illuminating samples with a shifting pattern, depleting fluorescence at the edge of the excitation beam, and imaging single molecules. These techniques are constantly improving with advances in dye creation, acquisition speed, and reducing phototoxicity.

Talley J. Lambert (2015). Cow lung artery cell highlighting the cellular components actin (black), mitochondria (red), and DNA (blue). Nikon Small World Competition.

Structured Illumination Visualizes Patterns and is Ideal for Live Imaging

Structured illumination microscopy (SIM) uses a moveable grating to visualize patterns and structures beyond the diffraction limit. When two grids are misaligned, they create interference patterns known as Moiré fringes. If you know the identity of one grid, then you can mathematically deduce and reconstruct the second grid from the Moiré fringes. SIM uses excitation light focused into a striped pattern to make a grid. The sample acts as the unknown grid that is reconstructed.

“When a grid’s misaligned with another behind, that’s a Moiré.”

SIM is ideal for super resolution imaging of live samples due to its fast acquisition and low illumination intensity, although its resolution is more modest than other super resolution techniques. SIM achieves a lateral resolution of 100 nm and optical sectioning with an axial resolution of about 300 nm. Another advantage of SIM is that it does not require specialized dyes, allowing super resolution imaging across the color spectrum using standard dyes. It is therefore trivial for a user to take a sample prepared for confocal or widefield microscopy and perform SIM.

SIM does have some caveats. Because the sample is reconstructed, the final images are not quantitative, therefore colocalization analysis of SIM images should not be done using correlations of pixel intensity, but rather object-based methods, such as measuring the overlap of segmented objects (see Hong et al., (2017)). Careful consideration of the biology and the reconstruction parameters is essential, as out of focus (‘honeycomb’) pseudo-structures are common, especially with non-punctate signal such as cell-fills. Lastly, SIM does not perform well when imaging deep sections of tissue, as scattering distorts the illumination and emission patterns.

Suppressing Fluorescence for a Sharper Image with STED

Stimulated emission depletion (STED) microscopy works similarly to beam scanning confocal but shrinks the size of the fluorescent spot using stimulated emission, producing a lateral resolution of about 50 nm. As with imaging, focusing the excitation beam on a sample is diffraction limited. STED uses a donut shaped depletion beam (also referred to as the STED beam) that encircles the excitation beam to reduce the excited area that fluoresces.

Schematic of STED depletion to reduce the effective fluorescence point spread function (PSF). From Hiersemenzel et al., (2013)

STED is useful for fixed and live imaging, but the imaging area must be reduced to have millisecond scale time resolution, and the high intensity STED beam can contribute to phototoxicity with prolonged imaging. Samples require minimal special preparation aside from the use of STED compatible fluorescent proteins or dyes like CF®405M, CF®488A, and CF®680R that can be depleted by the STED beam. Unlike other super resolution techniques, STED is quantitative because it does not rely on post-processing or reconstruction.

Confocal (top) vs STED (bottom) images of peroxisomes (blue), vimentin (green), giantin (red), and nuclear pores (grey). Adapted from Winter et al., (2017)

Improving STED with Superior Dyes and Lasers

STED is commonly limited to one or two STED channels per experiment because of the impracticality of using multiple STED beams. With advances in dye engineering, hyperspectral STED (“Hyper-STED”) has been developed to simultaneously image up to four dyes including CF®680R using a single STED beam (Winter et al., 2017). Coupled with spectral detection, this technique increases imaging speed and reduces phototoxicity. The white light supercontinuum laser used in Hyper-STED is currently cost-prohibitive, but accessibility should improve in the future with lower hardware costs and the availability of more STED compatible dyes.

3 color Hyper-STED in live cells labeling tubulin (green), endoplasmic reticulum (blue), and endosomes (red). Adapted from Winter et al., (2017)

Imaging Single Molecules Gives Insight into Structure and Molecular Dynamics

The advent of bright and photostable dyes and fluorescent proteins has enabled a new era of single molecule imaging to uncover ultrastructure and the dynamic behavior of individual molecules. By collecting photons from isolated spots, single molecule techniques calculate the molecule’s most likely position within a resolution of 20 nm. Stochastic optical reconstruction microscopy (STORM) relies on random photoswitching of dyes between light and dark states, while its sibling, photoactivation localization microscopy (PALM), uses photoactivation of assemblies of isolated single molecules to construct an image. Single particle tracking (SPT) shows us the dynamic behavior of molecules as they move.

Single molecule localization from a diffraction limited point-spread function (Gustafsson et al., n.d.).

STORM Provides Unparalleled Resolution at a Cost

STORM offers the highest resolution in optical microscopy. By localizing single molecules, STORM constructs a composite image made of these localized spots by multiple cycles of photoactivation and bleaching. STORM has the intricate structure of intracellular assemblies, including the elegant axonal cytoskeleton, in a way that was not possible with conventional techniques. However, STORM suffers from low acquisition speed and phototoxicity due to the sample’s constant exposure to the intense photoactivation laser.

STORM revealed the periodic structure of the axon (Xu et al., 2013).

STORM uses photoswitching, “blinking” dyes, such as several CF® dyes, that have distinct light and dark states. While live imaging with STORM is possible, it is best applied to fixed samples where sufficient imaging cycles can be performed or live samples where object movement is minimal. Buffers, labeling, and mounting conditions need to be carefully optimized to promote dye switching. As with other super resolution techniques, dye innovations are driving improvements in STORM. STORM relies on intense and constant illumination, and multicolor STORM requires dyes that can be spectrally separated. Spectrally distinct, blinking dyes with improved molecular brightness reduce the intensity of light illumination and phototoxicity, and allow multicolor STORM with simple organic dyes.

Single Molecule Tracking Reveals Molecular Behavior

While STORM is used to gain structural insights, single molecule tracking (SMT) studies the dynamic behavior of molecules. Live imaging experiments using STED, SIM, or conventional microscopy image the average behavior of the population of labeled molecules, but SMT experiments allow the study of individual molecules, yielding valuable insights such as the existence of distinct populations of diffusing molecules and molecular partners.

SPT experiments track diffusion rates of transmembrane proteins (purple tracks) that interact with PSD-95 (pseudocolored density map) with (SEP-TM-Bind) or lacking PSD-95 binding domain (SEP-TM-Nonbind) (Li & Blanpied, 2016).

Single molecule tracking relies on photostable fluorophores. Quantum dots have been used due to their high brightness and photostability but are large compared to organic dyes and fluorescent proteins. Development is focused on producing photostable organic dyes and fluorescent proteins with high molecular brightness and low rates of photobleaching. The CF® dyes CF®633 and CF®680R are ideal for single molecule tracking, as their red and far-red excitation reduces phototoxicity.

Imaging at Depth with 2-Photon Microscopy

Super resolution and conventional microscopy suffer when attempting to image deep into tissue or behaving organisms. Visible light is absorbed or scattered by tissue, limiting its penetration, but infrared and near-IR light is less affected. While traditional single-photon microscopy excites a fluorophore with an absorbed wavelength of light, 2-photon microscopy instead uses two longer wavelength photons. 2-photon microscopy is commonly used at the cellular or macroscopic level with long working distance objectives to image organ-level organization with dyes including CF®488A, CF®594, and CF®680R.

2-photon microscopy is also commonly used to image biosensors, such as calcium and recently, ultrafast imaging of dopamine release (Patriarchi et al., 2018). Thus, developments in 2-photon microscopy center on acquisition speed and improving resolution at the cellular scale through dye engineering. While 2-photon microscopy can better penetrate tissue to excite dye molecules, emitted fluorescence is also subject to scattering. To that end, red-shifted dyes like CF®680R have been developed to improve resolution in deep imaging.

Super Resolution Trends Going Forward

Super resolution microscopy continues to improve. Key areas for improvement are in dyes and sample preparation. Recently, SIM has been combined with STORM principles to extend its resolving ability further (D. Li et al., 2015). Super resolution microscopy has been combined with new sample preparation methods like expansion microscopy to look even deeper into molecular organization (Tillberg et al., 2016 Wang et al., 2018 H. Xu et al., 2019).

  1. Dodgson, J., Chessel, A., Cox, S., & Carazo Salas, R. E. (2015). Super-Resolution Microscopy: SIM, STED and Localization Microscopy. In T. E. S. Dahms & K. J. Czymmek (Eds.), Advanced Microscopy in Mycology (pp. 47–60). Springer International Publishing.
  2. Hiersemenzel, K., Brown, E. R., & Duncan, R. R. (2013). Imaging Large Cohorts of Single Ion Channels and Their Activity. Frontiers in Endocrinology, 4.
  3. Hong, S., Wilton, D. K., Stevens, B., & Richardson, D. S. (2017). Structured Illumination Microscopy for the Investigation of Synaptic Structure and Function. Methods in Molecular Biology (Clifton, N.J.), 1538, 155–167.
  4. Gustafsson, M. G. L., Betzig, E., Hess, H. F., Patterson, G. H., Lippincott-Schwartz, J., & Davidson, M. W. (n.d.). ZEISS Microscopy Online Campus | Introduction to Superresolution Microscopy. Zeiss Education in Microscopy and Digital Imaging. Retrieved July 7, 2020, from
  5. Li, D., Shao, L., Chen, B.-C., Zhang, X., Zhang, M., Moses, B., Milkie, D. E., Beach, J. R., Hammer, J. A., Pasham, M., Kirchhausen, T., Baird, M. A., Davidson, M. W., Xu, P., & Betzig, E. (2015). ADVANCED IMAGING. Extended-resolution structured illumination imaging of endocytic and cytoskeletal dynamics. Science (New York, N.Y.), 349(6251), aab3500.
  6. Li, T. P., & Blanpied, T. A. (2016). Control of Transmembrane Protein Diffusion within the Postsynaptic Density Assessed by Simultaneous Single-Molecule Tracking and Localization Microscopy. Frontiers in Synaptic Neuroscience, 19.
  7. Patriarchi, T., Cho, J. R., Merten, K., Howe, M. W., Marley, A., Xiong, W.-H., Folk, R. W., Broussard, G. J., Liang, R., Jang, M. J., Zhong, H., Dombeck, D., Zastrow, M. von, Nimmerjahn, A., Gradinaru, V., Williams, J. T., & Tian, L. (2018). Ultrafast neuronal imaging of dopamine dynamics with designed genetically encoded sensors. Science, 360(6396), eaat4422.
  8. Tillberg, P. W., Chen, F., Piatkevich, K. D., Zhao, Y., Yu, C.-C. (Jay), English, B. P., Gao, L., Martorell, A., Suk, H.-J., Yoshida, F., DeGennaro, E. M., Roossien, D. H., Gong, G., Seneviratne, U., Tannenbaum, S. R., Desimone, R., Cai, D., & Boyden, E. S. (2016). Protein-retention expansion microscopy of cells and tissues labeled using standard fluorescent proteins and antibodies. Nature Biotechnology, 34(9), 987–992.
  9. Wang, Y., Yu, Z., Cahoon, C. K., Parmely, T., Thomas, N., Unruh, J. R., Slaughter, B. D., & Hawley, R. S. (2018). Combined expansion microscopy with structured illumination microscopy for analyzing protein complexes. Nature Protocols.
  10. Winter, F. R., Loidolt, M., Westphal, V., Butkevich, A. N., Gregor, C., Sahl, S. J., & Hell, S. W. (2017). Multicolour nanoscopy of fixed and living cells with a single STED beam and hyperspectral detection. Scientific Reports, 7(1), 46492.
  11. Xu, H., Tong, Z., Ye, Q., Sun, T., Hong, Z., Zhang, L., Bortnick, A., Cho, S., Beuzer, P., Axelrod, J., Hu, Q., Wang, M., Evans, S. M., Murre, C., Lu, L.-F., Sun, S., Corbett, K. D., & Cang, H. (2019). Molecular organization of mammalian meiotic chromosome axis revealed by expansion STORM microscopy. Proceedings of the National Academy of Sciences, 116(37), 18423–18428.
  12. Xu, K., Zhong, G., & Zhuang, X. (2013). Actin, Spectrin, and Associated Proteins Form a Periodic Cytoskeletal Structure in Axons. Science, 339(6118), 452–456.

Christopher Pratt, PhD

Christopher Pratt earned his PhD in cell biology and neuroscience from Carnegie Mellon University in 2016 under the guidance of Marcel Bruchez, PhD. For his thesis, he developed and applied novel fluorescent probes to understand synaptic vesicle and ion channel trafficking in the brain, making him a ardent microscopist. With a passion for communicating science and data, Christopher is a regular contributor to Biotium’s Full Spectrum Blog and is a freelance science and medical writer, analyst, and research consultant based in Chicago, Illinois. Visit his website, twitter, or LinkedIn and get in touch!


Glutathione hydropersulfides (GSSH) are alluded to play crucial roles in signal transduction, redox homeostasis, and metabolic regulation. However, the detailed biological functions of GSSH in these aspects are extremely ambiguous. The key barrier to understand the role of GSSH in biological systems is a lack of detection tools with high spatiotemporal resolution. To address the issues, we are seeking novel chemical tools for GSSH detection. We herein develop the first two-photon ratiometric fluorescent probe (TP-Dise) for GSSH detection with high spatial and temporal resolution in living cells and tissue. On the basis of our probe TP-Dise, we investigate the biosynthesis of GSSH, and the results indicate that GSSH is mainly from two sulfurtransferases, CBS and CSE. Furthermore, we explore the biological function of GSSH in protecting cells from mercury ion-induced cell damage for the first time. The experimental results indicate that mercury ions may induce cell death by causing mitochondrial autophagy. GSSH acts both as antagonist and as antioxidant and can effectively alleviate the damage caused by mercury stress.



All studies and procedures involving animals were in strict accordance with the recommendations of the European Community Directive (2010/63/UE) for the protection of vertebrate animals used for experimental and other scientific purposes. The project was specifically approved by the ethics committees of Institut Curie CEEA-IC #118 and Institut Pasteur (reference #2015-0008) and by the French Ministry of Research (authorization references #0424003, #13310-2018020112578233-v1, #6354-2016080912028839-v4, #11206-2017090816044613-v2). We comply with internationally established principles of replacement, reduction, and refinement in accordance with the Guide for the Care and Use of Laboratory Animals (National Research Council 2011). Husbandry, supply of animals, as well as mouse maintenance and care in the Animal Facility of Institut Curie (facility license #C75-05-18) before and during experiments fully satisfied animal needs and welfare. Suffering of the animals was kept to a minimum.

The following mouse lines were used in this work: R26 mTmG mice (24), NMRI-Foxn1 nu/nu (purchased from Janvier Labs), Pax7 CreERT2 mice (30), and Rag KO/KO -GFP mice [B6(Cg)-Rag2 tm1.1Cgn /JTg(UBC-GFP) (25, 26)]. All experiments were performed in 4- to 8-week-old female mice.


Three-dimensional (3D) conception of the window, mold, and holder was performed using SOLIDWORKS 2018 (Dassault Systèmes). 3D printing of window holder prototypes was performed using a home-made system [Fused deposition modeling (FDM)] and MultiJetFusion [Hewlett-Packard (HP)] using various materials, predominantly generic acrylonitrile butadiene styrene (FDM) and polyamide and polypropylene (PP) (fusion). We recommend PP fusion for higher printing resolution, temperature, and hygroscopic stability.

Window production

Biocompatible transparent silicone (Elkem) components (A and B) were mixed at a 1:1 ratio (A:B) according to manufacturer guidelines for 5 min per 50 g of mix. The mix was then degassed for 5 and 10 min (<10 mbar) with a vacuum release in between to maximize air removal. Air removal was completed by centrifuging the mix at 1000g for 5 min. The degassed silicone mix was kept for a maximum of 24 hours at 4°C to prevent spontaneous polymerization. Windows were produced using a custom three-stage single footprint compression mold made of machined steel, which was maintained at 100°C throughout molding. Approximately 1 ml of silicone was placed on the footprint before closing the mold and applying compression at 5 tons for 10 min. Polymerized windows were then postcured for at least 24 hours at 60°C and subsequently washed in absolute ethanol before autoclaving. While untested to date, implant sizes may be modified (e.g., smaller versions for specific tissue sites) subject to optimizing elastomer stiffness to preserve the mechanical properties of the device required for suture-free maintenance.

Optical characterization of PDMS windows

Optical, chromatic, and morphological aberrations depend on the optical path, which includes the objective and microscope core. Therefore, all assays were performed using a single microscope setup and water as the imaging medium. All measurements were acquired using a 40×/1 numerical aperture (NA) Water DIC PL APO VIS-IR objective on an upright spinning disk microscope (CSU-X1 scan-head from Yokogawa Carl Zeiss, Roper Scientific, France), equipped with a CoolSNAP HQ2 charge-coupled device (CCD) camera (Photometrics) and MetaMorph software.

Optical aberrations (PSF). To measure the PSF representing optical aberrations arising from RI changes, we prepared 2% agarose gels containing 1:200 fluorescein and 1:400 diluted FluoSpheres carboxylate-modified microspheres (0.2-μm diameter). For both glass and PDMS windows, we acquired eight stacks of 100 μm with a step size of 0.2 μm at increasing depths from the window/gel interface (Z0, defined using the background signal in the fluorescein channel). PSF was calculated using the MetroloJ plugin (33) in Fiji (ImageJ v1.53) on at least 20 beads per section.

Chromatic aberrations. To measure the chromatic aberrations representing optical aberrations induced by wavelength-specific RI changes, we prepared 2% agarose gels containing TetraSpeck Microspheres (4-μm diameter) diluted at 1:400. Data were acquired over 200 μm from the first bead detected in Z for both glass and PDMS windows. Laser intensities and exposures were set to obtain similar intensity range (12 bits). R and Rref were calculated using the MetroloJ plugin (33) in Fiji (ImageJ v1.53) on at least 30 beads per condition.

Morphological aberrations. To measure morphological aberrations that may arise from an uneven window surface, we imaged a microgrid (Argolight SLG-075, pattern B) using a laser excitation wavelength of 405 nm through either water only (uncovered), glass, relaxed PDMS, or stretched (to 200%) PDMS (wavy) window in duplicates. Dot grids were segmented using Fiji v1.53 (threshold > analyze particles), and dot coordinates were extracted. Using a central motif (cross 00) and the top left dot (−105 −75), we corrected the rotation of the grid under the different materials and compared the corrected coordinates of each dot to either “medium only” (measured) or the theoretical (15 × 15 μm) pattern B matrix. Graphical deformation maps were generated using the R package ggplot2 (34).

Window implantation

Before surgery, all instruments were sterilized by autoclaving or heat sterilization. PDMS windows were sterilized by autoclaving to avoid darkening. To ensure sterility throughout the procedure, mouse preparation and handling, adjustment of anesthesia flow rates, and similar tasks were performed by an assistant, limiting the surgeon to only contacting sterile surgical tools and the surgical field. The physiological body temperature of the anesthetized mouse was maintained throughout the procedure using a heated induction chamber and heated surgical station.

Preoperative care

Before anesthesia, mice were administered 125 mg/kg of Noroclav (140 mg/ml of amoxicillin 35 mg/ml of clavulanic acid) and an analgesic cocktail [0.1 mg/kg of buprenorphine (Buprecare) and 5 mg/kg of carprofen (RIMADYL)] by subcutaneous injection. Analgesics were administered into the neck skin fold for all experiments apart from implantation in PDX models, where they were administered in the rear leg skin fold.

Surgical field preparation

Mice were anesthetized in an induction chamber using 4% isoflurane (ISOFLURIN). Once anesthetized, mice were transferred onto a heat pad, and anesthesia was maintained using an isoflurane concentration of 1.5 to 2%. Ophthalmic ointment was applied to eyes to prevent corneal drying. At the planned implantation site, mouse hair was trimmed using an electronic pet clipper (Aesculap Exacta). Residual hairs were removed using depilatory cream for a maximum of 2 min to minimize the risk of skin damage and drying. Stray hairs and cream were removed using Betadine Scrub 4% before a soap rinse with sterile water. The shaved skin and surrounding areas were subsequently disinfected using dermal Betadine 10%. Finally, the prepared surgical area was covered by precut sterile film (Tegaderm).

Implantation (performed by the surgeon). We recommend the following surgical tools to limit skin and tissue damage: Graefe extra fine straight 1 × 2 teeth forceps [for skin and membranes Fine Science Tools (FST) #11153-10], flat blunt forceps (for the window and tissue), serrated blunt forceps (for the window), fine and sharp straight scissors (for skin and membranes FST #14568-09), and Strabismus blunt straight scissors (for blunt dissecting the subcutaneous pocket FST #14574-09).

To implant the PDSM window (Ø18mm) in the prepared surgical area, a 10-mm incision was made in the skin overlying the tissue of interest [corresponding to approximately 15 mm when stretched (Fig. 2A)]. Next, using blunt scissors, the skin was carefully detached from underlying tissues approximately 5 to 10 mm around the incision to generate sufficient space for the implanted window. Throughout the procedure, prewarmed saline solution (sterile 0.9% NaCl) was used to prevent tissue dehydration. Before positioning the window, the incision site was filled with prewarmed saline solution to minimize the risk of introducing air underneath the window during implantation. Using thin forceps with blunt and flat edges, the PDMS window was folded in half and placed into the incision pocket where it was unfolded under the skin. Typically, the procedure from the first incision to prepositioning of the window under the skin was performed in less than 3 min (fig. S2, A and B). To fit the window in place, the skin edges were positioned into the window groove using two forceps that alternate between holding the upper frame and sliding the skin around the window (similar to the technique for adjusting a tire on a rim). During this step, the incision size can be adjusted, if required, to tightly seal the skin around the window. Experienced users are capable of performing the entire surgical procedure in

Postoperative care

After surgery, mice were placed in a prewarmed cage (on a heating pad) to recover from the anesthesia. Mice were closely monitored for signs of pain, discomfort, and infection after window implantation. The second and last Noroclav antibiotic dose was provided by subcutaneous injection the day after surgery. The skin surrounding the window was cleaned using dermic Betadine 1% every 2 to 4 days to prevent infection. Ibuprofen was administered in drinking water (0.4 mg/ml) for 7 days after surgery to minimize inflammation and pain. If signs of pain are observed, then postoperative analgesics [0.1 mg/kg buprenorphine (Buprecare) and 5 mg/kg of carprofen (RIMADYL)] can be administered by subcutaneous injection the day after surgery.

Mice bearing windows were group housed where possible after surgery and provided with cage enrichment suitable for postsurgery recovery, including nesting material, housing, and chew sticks. Recovery food (DietGel Recovery, ClearH20) was provided in the cage to limit mechanical stress from reaching the feeder. Singly housed mice without sufficient environmental enrichment may be more likely to damage the device during the first 1 to 2 days after surgery. Stringent aseptic surgical technique, and including preventative antibiotic and ibuprofen treatment, minimizes this risk by reducing inflammation and irritation. Overall, younger animals were more prone to skin damage arising from the procedure, which may be due to the window/animal size ratio and/or increased activity at earlier ages. Cage cleanliness did not appear to affect the risk of implant infection. Typically, mice exhibited minor weight loss the day after surgery (−4.20 ± 4.74%, within ethical limits), which were restored by day 2 (+2.93 ± 3.18%). As expected, PDMS windows retain their dimensional characteristics after implantation, displaying no discernible swelling as a result of marginal exposures to biological fluids during study time frames (maximum 35 days tested).

Handling window-bearing mice

When the implantation is performed correctly with stringent asepsis, no mouse should spontaneously lose their implanted windows. Windows are at a risk of displacement, however, in the first few days after surgery when handling mice using conventional restraining techniques. Thus, we recommend adapting handling techniques when assessing mice in the days immediately after window implantation and consider the use of light gas anesthesia where possible, e.g., when performing procedures such as intraperitoneal injection.

Troubleshooting guide for the implantation procedure

(A) Air is trapped under the window during implantation. Prepare a 1-ml syringe with sterile, prewarmed 0.9% NaCl, and a 30-gauge needle. Position the mouse so that the injection port is facing upward and disinfect it with 70% ethanol. Using fine forceps, hold the injection site and inject 200 to 500 μl of saline solution underneath the window: The air bubble(s) will move close to the injection site. Slowly aspirate using the syringe to remove air and as much of the saline solution as possible (as demonstrated in movie S4). Dry the injection port. Wait for the saline solution to be absorbed prior to imaging.

(B) Air appears under the window after implantation. During surgery, take care to make a straight incision. Irregular and ragged incisions increase the risk of air leakage. Check the window integrity: If the window is perforated, then it must be discarded. Check carefully the skin in the groove for breakages as small cuts can cause air leakiness. If a small break or nick (1 to 2 mm maximum) is observed, then place a small amount of glue on the groove to seal the leakage and proceed to (A). The next day, check carefully for the absence of air bubbles.

(C) Presence of small amount of fluid (exudate) and immune cells between the tissue and the window. This is common for a few days after surgery and should disappear after 2 to 3 days. To minimize this risk, review the surgery procedure to ensure stringent asepsis, check/change the ibuprofen in the drinking water, and ensure that food is accessible to limit mechanical stress. If the exudate precludes imaging, then administer prewarmed saline through the injection port to flush underneath the window [see (A) fig. S3E]. As immune cells may return to the site after flushing in the days following surgery, this step may need to be repeated in a subsequent acquisition session, depending on the desired imaging frequency.

(D) Hair underneath the window. Review the preparation step. Trim a larger area if necessary. A spray plaster can be used before incision to trap residual stray hairs.

(E) Skin becomes thinner or dry around the window. If this issue occurs within the first 7 days from implantation, then reduce depilatory cream incubation and/or change the brand. Use 1 × 2 teeth forceps to reduce skin damage. Limit damage to vasculature when performing blunt dissection to detach the skin. This issue may appear after long-term maintenance (>2 weeks), but, in our experience, it did not affect imaging (no infection, no air bubbles) when postsurgery care was properly performed.

(F) Skin or tissue become necrotic. Immediately terminate the experiment. This should never happen review the surgery procedure (including asepsis, tissue drying, and damage to vasculature).

Intravital imaging through PDMS windows

Imaging on the day of implantation is not recommended to avoid additional stress to the surgery area and overconstraining the window. Residual physiological saline under the window immediately after implantation can also affect imaging depth. Instead, perform imaging the next day (day 1) at the earliest. While the mouse is anesthetized, administer postoperative antibiotics by subcutaneous injection. Carefully inspect the device and clean the surgical area as described above. Mice are gas anesthetized using 4% isoflurane (ISOFLURIN) initially and maintained using 1.5 to 2.5% isoflurane for short-term imaging. When performing time-lapse imaging spanning several hours, reduce the isoflurane concentration to between 0.8 and 1.2%. These anesthesia levels are optimal for long-term maintenance of mice in a nonresponsive state with a slow, constant, and nonforced breathing pattern. Irregular and abnormal breathing patterns are associated with persistent anesthesia greater than 1.5%, which perturbs imaging (increases motion-induced image deformations) and can decrease survival times (35). Fast breathing and reflex responses indicate insufficient anesthesia, which also disturbs imaging and increases the risk of the mouse awakening during imaging.

Before imaging

To maintain animal hydration for short-term imaging (<2 to 3 hours), administer 250 μl/10 g of saline solution by subcutaneous injection after anesthesia induction. In experiments exceeding 3 hours, mouse hydration can be maintained during imaging using a subcutaneous or intraperitoneal infusion of glucose and electrolytes (

50 to 100 μl/hour) via an indwelling line. Our custom-made holder, adapted for upright microscopes (ready to print STL files provided in data file S1), permits regional body immobilization through compression. A rigid conical structure (containing immersion medium, usually water) presses on the edge of the window, allowing the tissue and window to be exposed and stretched beneath the objective as the mouse rests on a foam bed, which preserves breathing while cushioning movements. The mouse is positioned on the imaging holder and gently pushed on the static stage, centering the window in the conic aperture. The mouse should be tightly constrained the foam absorbs excessive compression. The mouse is then observed for 5 min to check breathing and the stability of the anesthesia. Before placing the holder onto the microscope stage, the conic aperture is filled with 2 to 3 ml of pure water. The holder may be adapted for inverted microscope configurations, using gel-based immersion medium instead of water for imaging. Critical: Do not use oil-based imaging medium with PDMS windows, as hydrophobic solvents may result in window swelling, fragility, and toxicity to underlying tissues. Placing a glass coverslip over the PDMS window may enable the use of oil-based objectives for imaging superficial tissue structures, provided that the underlying material remains protected throughout.

During imaging

Find structures of interest using brightfield or fluorescence with the oculars. During scanning, optimize laser power and exposure times to minimize phototoxicity and pixel saturation. PDMS windows are resistant to high-power laser bursts and are therefore compatible with localized photobleaching and photoablation applications.

After imaging

The imaging medium must be removed from the conic aperture using a paper towel. The mouse is released from the holder and placed on a heated pad (set at 38°C) for recovery. During this time, the window is cleaned using a paper towel and 70% ethanol, and the surrounding skin is disinfected with dermic Betadine 1%.

Troubleshooting guide for imaging procedures

(G) The window cannot be correctly positioned on the conic aperture. Use softer foam in the holder that allows fitting to any position.

(H) The imaging medium (water) leaks. The conic aperture is not centered on the window. Reposition it or use a gel-based imaging medium.

(I) The holder disturbs breathing. Reduce the isoflurane dose. Reduce the compression on the animal. Sculpt the foam, ensuring that the mouse’s airway is not restricted. Chisel the foam to modify the mouse position (a horizontal plane may not be optimal for the tissue of interest).

(J) Poor visibility during imaging. Exudate accumulation and/or epithelioid “membrane” formation beneath the window (which can develop with long-term maintenance) can interfere with imaging. This has also been observed with glass/titanium windows (6, 13) and is likely due to inflammatory reactions in response to surgery. To minimize this risk, review the surgical procedure, taking care to limit tissue damage and maintain stringent asepsis throughout. While the epithelioid membrane–like structure affects tissue imaging depths, overall, it has no detrimental impact on image quality. However, when coupled with immune cells and/or fibrosis, it can preclude imaging. To alleviate this issue, prewarmed saline can be administered via the injection port to flush underneath the window [following the protocol in (A) fig. S3E and movie S4). Alternatively, reposition the window on a different area of the tissue of interest or terminate the experiment.

Intravital microscopy

Animals were anesthetized and positioned for imaging using the custom-made holder as described above. Intravital imaging was performed on an upright Nikon A1R MP multiphoton confocal microscope equipped with a pulsed Spectra-Physics Insight DeepSee laser (680 to 1300 nm, 120-fs pulses with auto-alignment), Luigs & Neumann XY motorized stage and four GaAsP nondescanned detectors with SP492, BP 525/50, BP 575/50, and BP 629/56 filter sets. The microscope was surrounded by a heated dark box maintained at 38°C. All images were acquired using 16× NA 0.8 or 25× NA 1.1 Plan Apo LambdaS water objectives. An excitation wavelength of 960 nm was used for GFP and TdTomato in addition to SHG imaging of collagen.

Image processing and visualization

Time-lapse acquisitions were corrected for movements and drifts using the “Correct 3D Drift” plugin (36) in Fiji (ImageJ v1.53) using the SHG channel as a reference.

Generation of tdTomato-expressing PDX tumor cells

Ewing sarcoma PDX tumors (IC-pPDX-87) were surgically removed and enzymatically dissociated to single cell level using the protocol described in Stewart et al. (37). Cells were resuspended (2.25 × 10 7 cells per 1.5 ml) in Dulbecco’s modified Eagle’s medium/F12 supplemented with 1% penicillin-streptomycin and 2% B27 (all from Thermo Fisher Scientific) and transduced overnight with a TdTomato-lentivirus in a 2-ml tube using an Intelli-Mixer RM-2L (Dutcher), program F1, 5 revolutions per minute, in a 37°C incubator with 5% CO2. TdTomato-lentivirus was generated by replacing the GFP sequence with tandem Tomato in the Lenti-sgRNA-GFP plasmid (Addgene plasmid no. 65656 a gift from C. Vakoc). The next day, cells were spun at 500g for 10 min. For one injection, 1 × 10 7 cells were resuspended in 50 μl of media and kept on ice. Fifty microliters of Matrigel (Corning, ref. 354234) was added to the cell suspension before injection into the interscapular fat tissue using an insulin syringe (BD, ref. 324891).

Muscle regeneration studies

Muscle injury was performed in Pax7 CreERT2 R26 mTmG adult mice at the time of PDMS imaging window implantation by intramuscular injection of 50 μl of cardiotoxin (1 μM Latoxan, L8102) diluted in 0.9% NaCl. Cardiotoxin was administered either before window positioning or after through the injection port. Mice received 3 mg of tamoxifen free base (Euromedex) by intraperitoneal injection 2 days before intravital imaging.

IVIS imaging

NIR fluorescence imaging was performed using a highly sensitive CCD camera mounted in a light-tight specimen box (IVIS 50 Living Image version: Animals were anesthetized in a warm induction chamber using 4% isoflurane and subsequently placed onto a warmed stage inside the camera box and supplied with 1.5 to 2% isoflurane to maintain anesthesia during imaging. The emitted fluorescence signal was detected by the IVIS camera system, integrated, digitized, and displayed. Excitation filters for Tomato and GFP fluorescence were 535 (DsRed filter, position 3) and 465 (GFP filter, position 2), respectively. Exposure times and pixel binning were auto-optimized for each fluorescence channel to minimize overexposure and normalized for comparison using the formula normalized_intensity = (measured_intensity/exposure_time)/binning_factor 2 . Raw parameters are provided in data file S2.