Skip main navigation

Acceleration of the Development of Microcirculation Embolism in the Brain due to Capillary Narrowing

Originally published 2023;54:2135–2144



Cerebral microvascular obstruction is critically involved in recurrent stroke and decreased cerebral blood flow with age. The obstruction must occur in the capillary with a greater resistance to perfusion pressure through the microvascular networks. However, little is known about the relationship between capillary size and embolism formation. This study aimed to determine whether the capillary lumen space contributes to the development of microcirculation embolism.


To spatiotemporally manipulate capillary diameters in vivo, transgenic mice expressing the light-gated cation channel protein ChR2 (channelrhodopsin-2) in mural cells were used. The spatiotemporal changes in the regional cerebral blood flow in response to the photoactivation of ChR2 mural cells were first characterized using laser speckle flowgraphy. Capillary responses to optimized photostimulation were then examined in vivo using 2-photon microscopy. Finally, microcirculation embolism due to intravenously injected fluorescent microbeads was compared under conditions with or without photoactivation of ChR2 mural cells.


Following transcranial photostimulation, the stimulation intensity-dependent decrease in cerebral blood flow centered at the irradiation was observed (14%–49% decreases relative to the baseline). The cerebrovascular response to photostimulation showed significant constriction of the cerebral arteries and capillaries but not of the veins. As a result of vasoconstriction, a temporal stall of red blood cell flow occurred in the capillaries of the venous sides. The 2-photon excitation of a single ChR2 pericyte demonstrated the partial shrinkage of capillaries (7% relative to the baseline) around the stimulated cell. With the intravenous injection of microbeads, the occurrence of microcirculation embolism was significantly enhanced (11% increases compared to the control) with photostimulation.


Capillary narrowing increases the risk of developing microcirculation embolism in the venous sides of the cerebral capillaries.

Preclinical studies have identified cerebral micro vascular obstruction due to leukocyte plugging as a major cause of reperfusion failure following the recanalization of stroke.1–3 A decrease in cerebral blood flow (CBF) because of increased capillary stall was also observed in animal models of Alzheimer disease.4,5 Even in normal brains, microvascular obstruction spontaneously occurs, leading to an age-dependent reduction in the cerebral capillary density.6 Multiple mechanisms might contribute to microvascular obstruction in healthy and diseased brains, including vascular endothelial dysfunction, inflammation, and reduced glycocalyx layers.7

In theory, microvascular obstruction occurs in the capillary with the lowest pressure gradient (ie, the longest pathway) through the microcirculation networks.8 An in vivo experiment, however, showed an increase in capillary flow stall with a partial increase in capillary diameters induced by pericyte deletion.9 Increasing the capillary lumen space must reduce the flow resistance in the capillary, thereby improving the capillary flow. This paradoxical observation could be explained by a 1-sided inflow toward the dilated capillary, resulting in insufficient flow to the other branch-off capillaries.10 Other biochemical factors, such as the deformability of red blood cells (RBCs) and leukocyte-endothelium interactions,9,11 might also contribute to capillary resistance, and those resistances should vary depending on the size of capillaries. Single-cell transcriptomic analysis has revealed phenotype differences along the arteriovenous axis,12 suggesting that cellular interactions between blood cells and vascular endothelium also vary along the capillary networks. Although these studies indicate regional variations in capillary resistance in addition to pressure gradients, little is known about the causality of capillary size and location of embolism formation in 3-dimensional networks of cerebral microcirculation.

The objective of this study was to determine the effects of the capillary lumen space on the development of microcirculation embolism. To spatiotemporally manipulate capillary diameters in vivo, transgenic mice in which mural cells expressed a light-gated cation channel, ChR2 (channelrhodopsin-2) mutant, a step-function opsin (ChR2-C128S),13 were used. Patch-clamp experiments have shown that photoactivation of cells that express ChR2 in the cell membrane leads to cell depolarization.14 The current animal model can thus be used to test cerebrovascular responses to mural cell depolarization. Pathological depolarization of the pericyte has been shown to occur after cerebral ischemia through increased chloride efflux via activation of Ca2+-gated anion channel, leading to opening of voltage-gated Ca2+ channel and further increased cytoplasmic Ca2+ concentration.15 Previous studies have also shown that photoactivation of ChR2-expressing mural cells causes transient vasoconstriction and a reduction in RBC flow.16–19 However, the effects of mural cell activation on the spatiotemporal regulation of CBF are not well characterized. Moreover, inconsistent observations on the contractile responses of capillaries to ChR2 stimulation have been reported.16,17,19

To address these issues, 3 experiments were performed in this study. First, spatial and temporal changes in the regional CBF for focal illumination of various photostimulation powers were determined using laser speckle flowgraphy. Second, capillary diameter responses to ChR2 mural cell depolarization were directly characterized in vivo using 2-photon microscopy with relatively uniform illumination of the optimized power of photostimulation. Finally, microcirculation embolism due to intravenously injected fluorescent microbeads (4 µm in diameter) was compared under conditions with or without photoactivation of ChR2 mural cells.


The data that support the findings of this study are available from the corresponding author upon reasonable request.


Animal use and the experimental protocols were approved by the Laboratory Animal Care and Use Committee of Keio University (No. A2022-284), and all experimental procedures were performed following guidelines established by the Institute for the Humane Care and Use of Laboratory Animals in compliance with ARRIVE (Animal Research: Reporting In Vivo Experiments) guidelines.20 The sample size was determined based on the variability of the hemodynamic responses in control animals,21 and comparisons between ChR2 and control mice were conducted blindly for the image analysis process. Other methods for animal preparations are described in Supplemental Material.


Focal and global illumination protocols are described in Supplemental Material. For single-cell activation, a 2-photon laser equipped in the microscope system (FV1200 MPE, Olympus Corporation, Japan) was used. The excitation laser was a Ti:sapphire pulse laser (80 MHz, Spectra-Physics, Santa Clara, CA) at a wavelength of 860 nm (an instrumental power setup of 30% to 100%, a mean power of 0.9 to 3.2 mW/mm2 under the objective lens). Excitation at 860 nm was selected to minimize absorption by blood plasma labeled with sulforhodamine22 but to achieve sufficient excitation of ChR2.23

Laser Speckle Flowgraphy

Spatiotemporal changes in CBF were monitored with laser speckle flowgraphy (LSFG-micro, Softcare, Fukuoka, Japan) over the cortical surface with an intact skull, as described previously.21 Details on image acquisition protocols and analysis are described in the Supplemental Material.

Two-Photon Microscopy

Diameter changes in the vessels were measured with a 2-photon laser scanning fluorescence microscope (2-photon microscopy; FV1200 MPE, Olympus Corporation, Japan). Details on image acquisition and analysis are described in the Supplemental Material.

Classification of Vessels and Mural Cells

The 0-order branches were defined as penetrating arteriole (PA) and ascending venule (AV) in the parenchyma, where those vessels branched off from the surface artery (SA) and surface vein, respectively (Figure 1A). Then, the parenchymal capillaries were tracked from the PA or AV and labeled as 1A, 2A, and 3A or 1V, 2V, and 3V capillaries according to the first, second, and third number of branches from the PA or the AV, respectively (Figure 1A and 1B). For those microvessels, ChR2-positive cells (according to the YFP [yellow fluorescent protein] label; see Figure 1C) were morphologically classified into vessels covered with smooth muscle cells (SMCs; a narrow-ring shape), ensheathing pericytes (EPs), mesh pericytes (MPs), thin-strand pericytes (TSPs), and venule SMCs (vSMCs) based on the literature.24

Figure 1.

Figure 1. Definition of vascular types and ChR2 (channelrhodopsin-2) mural cells. A, Vascular tree. The branch order of the parenchymal capillaries was manually traced from the penetrating arteriole (PA) or ascending venule (AV), which are located perpendicularly to the cortical surface and merged to the surface artery (SA) or surface vein (SV), respectively. B, Representative image of the microvascular networks (maximum intensity projection). The minimum route from PA to AV consists of 7 capillaries (ie, 1A, 2A, 3A, 4A [4V], 3V, 2V, and 1V) with variable expression of ChR2 (yellow-green). Scale: 50 µm. C, Representative images (yellow-green: ChR2; red: blood plasma) of the 5 types of ChR2 mural cells (see main text). Scale: 10 µm. EP indicates ensheathing pericytes; MP, mesh pericyte; SMC, smooth muscle cell; TSP, thin-strand pericyte; and VSMC, venule smooth muscle cell.

Evaluation of Microcirculation Embolism

Fluorescent polystyrene microbeads (FluoSpheres sulfate-modified microspheres, 4.0 µm in diameter, yellow–green 505 nm/515 nm, Invitrogen) were injected from a femoral vein for evaluation of microcirculation embolism. Details on image acquisition and analysis are described in the Supplemental Material.


Stimulation Power-Dependent Spatiotemporal Changes in Regional CBF

Transcranial focal photostimulation of the ChR2 mural cells resulted in a rapid decrease and gradual recovery of regional CBF. With an increase in the power of the blue light emitting diode stimulation, the changes in CBF extended from the irradiation point to the periphery (Figure 2A and 2B), and the peak reduction in CBF was enhanced (Figure 2C). No detectable changes in the animal physiology were observed for all simulation conditions. The mean CBF reduction measured at the core of photostimulation showed a nonlinear negative correlation with stimulation power: 14±3%, 24±7%, 34±11%, and 49±8% relative to the baseline CBF for stimulation powers of 0.03, 0.08, 0.26, and 2.5 mW/mm2, respectively (N=5 animals; Figure 2D). Based on these results, the sensitivity of photostimulation to the cause of the CBF reduction was estimated to be a power of 3.6±1.5 µW/mm2. As the normal animal (no expression of ChR2) did not show any detectable change in CBF for all stimulation conditions (Figure S2), the observed CBF reduction is considered the result of the photoactivation of the ChR2 mural cells. Based on the spatial extent of the CBF reduction, 0.26 mW/mm2 blue light stimulation was used for the following experiments.

Figure 2.

Figure 2. Spatiotemporal changes in cerebral blood flow (CBF) induced by focal photostimulation. A, A representative raw image of the mean blur rate (MBR) generated by laser speckle flowgraphy of the cortical surface vasculature. The image was averaged for 20 s measured under prestimulation baseline conditions. The circle indicates the place of blue light stimulation. Scale: 1 mm. B, Stimulation power-dependent decreases in regional CBF (0.03, 0.08, 0.26, and 2.5 mW/mm2; left to right). The image was averaged for 20 s measured after photostimulation. C, Temporal dynamics of the relative changes in CBF (arbitrary unit [a.u.]) at a core of photostimulation. The stimulation (stim.) was 3 s of blue light emitting diode (LED) and subsequent 3 s of orange LED (vertical bars). The color indicates 4 different powers of blue LED from 0.03 to 2.5 mW/mm2 (gray to black). Each of the 3 trials (thin lines) and mean time courses (bold) are represented from the animal shown in A. D, Stimulation power-dependent decreases in CBF at the core of photostimulation. Error bar: SD (N=5 animals).

Vascular Type–Dependent Vasoconstriction Induced by Photoactivation of ChR2 Mural Cells

Comparisons of XYZ images captured before and after global photostimulation showed stimulation-induced vasoconstriction for the arteries but not for the veins (N=25 animals; Figure 3A). Temporal dynamic analysis of the vessel diameter changes further revealed transient vasoconstriction of the SA and PA but no detectable changes in the surface vein and AV (Figure 3B). A larger diameter change was observed for the SA (20%±10%, P<0.05, prestimulation versus poststimulation, n=28 vessels) than for the PA (12%±9%, P<0.05, n=20 vessels), where the baseline diameter was also greater for the SA than for the PA (Figure 3C and 3D). However, no significant change in diameter was observed for the surface vein (0%±5%, n=32 vessels) and AV (1%±5%, n=20 vessels; Figure 3C and 3D).

Figure 3.

Figure 3. Vascular responses to global photostimulation. A, Representative 2-photon microscopy images of the vascular structures (red) at cortical depths of 0, 100, and 200 µm (top to bottom) captured before (pre) and after (post) photostimulation. Localization of the ChR2 (channelrhodopsin-2) mural cells (yellow-green) was merged with the prestimulation vascular image (left). All images were the mean image for a 15-s acquisition. The white rectangle indicates the location of the region of interest placed for diameter measurements of the surface artery (SA) and surface vein (SV; a depth of 0 µm) or the penetrating arteriole (PA) and ascending venule (AV; depths of 100 and 200 µm). Scale: 50 µm. B, Temporal dynamics of the diameters of the SA and SV (a depth of 0 µm) and the PA and AV (depths of 100 and 200 µm; top to bottom). Three seconds of photostimulation (stim) was applied at time 0, while imaging was interrupted. C, Prestimulation baseline diameters. Mean±SD (SA, PA, AV, and SV for n=28, 20, 20, and 32, respectively). D, Diameter responses to photostimulation. *P<0.05 (paired t test, prestimulation vs poststimulation).

In the parenchymal capillaries, all capillaries irrespective of the branch orders showed a statistically significant vasoconstriction response to photostimulation (Figure S3). Vasoconstriction of the capillaries (eg, from 5.7 µm to 5.1 µm) occurred with an occasional stall of blood cell flow where motions of nonfluorescent shadows (ie, unlabeled blood cells) stopped (Figure 4A; Video S1). Once the vasoconstriction was recovered (eg, to 5.3 µm), the stalled flow was recovered (Figure 4B). Comparisons of the pericyte types showed variable baseline diameters: 28±11 µm (n=36), 12±3 µm (n=29), 5.8±1.4 µm (n=27), 5.5±1.7 µm (n=35), and 27±15 µm (n=30) for the SMC, EP, MP, TSP, and vSMC, respectively. Among the ChR2-positive pericytes (Figure 4C), EPs were present at the precapillary arterioles, and the capillaries (less than 8 µm in diameter) had MPs and TSPs. Significant vasoconstriction (P<0.05) was also observed for the activation of all ChR2 pericytes (EP: 13%±11%, MP: 5%±11%, and TSP: 6%±9% relative to the baseline diameters, Figure 4D). Although the baseline diameter and stimulation-induced vasoconstriction were identical for the capillaries irrespective of the branch orders, a flow stall during vasoconstriction (n=9 out of 56 capillaries) was observed for the capillaries of the venous side (within the 3 branches from the AV, Figure 4E).

Figure 4.

Figure 4. Capillary responses to global photostimulation. A, Representative images of capillary (1V) structures (red) at a cortical depth of 25 µm. Localization of ChR2 (channelrhodopsin-2) thin-strand pericyte (TSP; yellow-green) was merged with the prestimulation image (top left, a mean of 15-s baseline images). The raw capillary images (top left to bottom right) showed motion of the nonfluorescence cells (white triangles) between the first 2 frames, whereas flow stall occurred over 15 to 75 s after the photostimulation (magenta triangles). Scale: 20 µm. B, Temporal dynamics of the representative capillary diameters shown in A. Photostimulation (stim.) was given at time 0, while imaging was interrupted. Red blood cell (RBC) flow stopped approximately 30 s after photostimulation and recovered at 60 s, and the corresponding capillary diameters were 5.1 µm and 5.3 µm, respectively. C, Prestimulation baseline diameters. Mean±SD (ensheathing pericyte [EP], mesh pericyte [MP], and thin-strand pericyte [TSP] for n=29, 27, and 35, respectively). D, Diameter responses to photostimulation (arbitrary unit [a.u.]). *P<0.05 (paired t test, prestimulation vs poststimulation). E, Relationships between capillary diameter responses and flow stall. All capillary responses (n=56) were plotted according to their baseline diameters and divided into arterial (left) and venous (right) sides. The photostimulation-evoked capillary flow stall occurred only on the venous side (black square), and no flow stall (white circle) was evident on the arterial side.

To rule out the possibility that capillary vasoconstriction occurred due to the reduction in blood flow resulting from the vasoconstriction of arteries upstream, photostimulation of individual cells was further performed with single-cell stimulation using the 2-photon laser (N=13 animals). Photostimulation of the ChR2 pericytes resulted in local vasoconstriction of the neighboring capillary (Figure 5A and 5B). In the stimulated location, there was a significant decrease in capillary diameter (7%±7%, P<0.05, n=12 capillaries from N=5 animals). However, in the vicinity of the location within a distance of 5 µm and >5 µm from the stimulated area, both showed no detectable changes in the diameters (Figure 5C). Consistent with the vasoconstriction induced by blue light stimulation, 2-photon laser stimulation evoked significant vasoconstriction in the arteries and capillaries but not the veins (Figure S4). In addition, all ChR2 pericytes (EPs, MPs, and TSPs) consistently evoked significant vasoconstriction but not vSMCs (Figure 5D).

Figure 5.

Figure 5. Capillary responses to 2-photon laser stimulation of ChR2 (channelrhodopsin-2) pericytes. A, Representative images of capillary (2V) structures (red) at a cortical depth of 40 µm captured before (pre) and after (post) 2-photon laser stimulation (white square). Localization of ChR2 thin-strand pericyte (TSP; yellow-green) was merged with the prestimulation image (left, a mean of 15-s baseline images). A 2-photon laser was applied to a portion of the ChR2-expressing pericyte (laser-stim., second from the left). Vasoconstriction of the capillary after stimulation was evident at the capillary (arrowheads) near the stimulated cell. Scale: 10 µm. B, Temporal dynamics of the representative capillary diameters shown in A. Two-photon laser stimulation (stim.) was given at time 0, while imaging was interrupted. Average (black) and individual trials (gray, 3 trials) measured at the stimulated sites showed consistent vasoconstriction after the stimulation. C, Comparisons of capillary diameter changes within stimulated (S), proximal (P; within a distance of 5 µm), and distal (D; >5 µm) locations from the stimulated area (arbitrary unit [a.u.]). Mean±SD (N=5 animals). *P<0.05 (Dunnett test). D, Diameter responses to 2-photon laser stimulation. Mean±SD (ensheathing pericyte [EP], mesh pericyte [MP], thin-strand pericyte [TSP], and venule smooth muscle cell [vSMC] for n=13, 13, 21, and 8, respectively). *P<0.05 (paired t test, prestimulation vs poststimulation).

Enhancement of the Formation of Microcirculation Embolism due to Capillary Narrowing

Finally, the development of microcirculation embolism was compared between the conditions with (on) or without (off) photostimulation in the same locations (Figure 6A). In the control animals (N=6), the ratio between stalled (> at least 5 s at the same place) and circulating microbeads was constant between the light-on and light-off conditions (Figure 6B). In contrast, the ChR2 animals (N=4) showed an 11%±19% increase in stalled microbeads for the light-on conditions compared to the light-off conditions (Figure 6B). On average, 13%±14% and 77%±25% of the microbead stalls were found in the capillaries of arterial and venous proximity (ie, the remainder were not identified), respectively (P<0.05, n=7 locations from 3 animals). No obstruction was found in the capillaries with 3 branches of PA, whereas most of the stalled microbeads were within 3 branches of AV. A significant increase (P<0.05) in the number of stalled microbeads for a short duration (<30 s when the capillaries are constricted) was observed compared with the control conditions (Figure 6C and 6D). A relative increase in stalled microbeads for a longer duration (more than 30 s) was also observed, but this was not statistically significant (P=0.08). The capillary diameters upstream and downstream where the microbeads were stalled were 3.5±0.4 µm and 3.6±0.8 µm, respectively, for the ChR2 animals and 4.5±0.6 µm and 3.6±0.6 µm, respectively, for the control animals (P>0.05; Figure 6E).

Figure 6.

Figure 6. Comparisons of microcirculation embolism development with intravenous infusion of fluorescent microbeads. A, Two-photon microscopy images representing microvessels and ChR2 (channelrhodopsin-2) pericyte. Mean projection images of vasculature (red), ChR2 pericytes (green), and overlays of both images. The microvasculature was extracted to count several microbeads detected in each vessel (bottom right). Scale: 50 µm. B, The location of the microbeads (yellow) observed in the capillary was identified using the vessel image (top). The number distribution of microbeads (black dot) in each image (bottom) was compared for conditions with no photostimulation (left: light off) and photostimulation (right: light on). The number on the y axis indicates an identical vessel segment for the light-off and light-on conditions, and # indicates a microbead found at the off-target location. Scale: 50 µm. C, Comparisons of the rate of microbeads that were moved through the vessels per a total number of microbeads (arbitrary unit [a.u.]) identified in the image in the control (Cont, N=6) and ChR2 mice (ChR2, N=4). *P<0.05. D, Comparison of the number of stacked microbeads between light-on and light-off conditions for short (<30 s) or long (> 30 s) periods in the ChR2 mice (n=8 capillaries from 4 animals, *P<0.05). E, Comparisons of the capillary diameters upstream (Up) and downstream (Dn) of the stacked microbeads in the control and ChR2 mice. Mean±SD.


This study demonstrates the causality of microcirculation embolism due to capillary narrowing. Photoactivation of ChR2 mural cells consistently showed significant vasoconstriction in the arteries, arterioles, and capillaries but not in the venules and veins (Figures 3 through 5). This results in a focal reduction in CBF (Figure 2) and flow stalls in the capillaries of the venous sides (Figure 4). A significant increase in the trapping of intravenously injected microbeads was observed in the capillaries with a smaller diameter than that of the microbeads, which was significantly enhanced by vasoconstriction of the capillaries due to photoactivation of the ChR2 mural cells (Figure 6).

The new findings are that microcirculation embolism during vasoconstriction occurs preferentially in the capillaries on the venous sides but not on the arterial sides. The capillary diameters at rest are rather uniform regardless of arterial or venous proximities despite the variable localization and morphology of the capillary pericytes.25 There was no detectable difference in the morphological reactions in the capillaries for both blue light and 2-photon laser stimuli with respect to the topological orders or morphological diversity of the pericytes (Figure 4; Figure S2). For SMC and EP, photostimulation-induced vasoconstriction occurred in both parts of the vessels covered and uncovered with ChR2 mural cells (Figure S5). It is suggested that these vasoconstrictions are not propagated via gap junction pathways (Figure S6). In contrast, photostimulation-induced vasoconstriction in capillaries was localized to the site of photostimulation (Figure S7). Therefore, the degree of vasoconstriction correlated with a coverage rate with ChR2 pericytes for global photostimulation (Figure S8). These findings indicate that other factors, in addition to the capillary structure, also contribute to the occurrence of capillary flow stalls, such as a difference in the pressure gradient between capillaries, network structures, flow speed, and flow-dependent rheological properties or cellular interactions between blood cells and vascular endothelium. Nevertheless, the physical size of the capillary lumen is critical to trapping a certain size of the intravascular embolus (Figure 6).

The present results indicate that pericyte-dependent capillary contraction may play a role under pathological conditions. Ultrastructure analysis of cortical microvessels in patients with Alzheimer disease revealed narrowing of capillary lumen partly due to swelling of the pericytes.26 Human cortical biopsies showed relatively narrow capillaries around the pericyte soma with amyloid β deposition, indicating the causal relationship between amyloid pathology and microvascular disturbances due to dysfunctional pericytes.27 A significant reduction in capillary diameter during reperfusion in major arteries after cerebral ischemia, compared to normal controls of anesthetized gerbil.28 Although previous studies have not classified capillaries into arterial or venous proximities, all pericyte types would be a promising therapeutic target to prevent microvascular obstruction and improve poststroke recovery.15

Another factor that may affect capillary flow stall is the deformability of RBCs, which is known to decrease with age.29 Interestingly, the reduced deformability of RBCs is also found in patients with Alzheimer disease.30 The normal size of RBCs is 7.3 µm and 5.8 µm in humans and mice, respectively.31 If the capillary becomes narrower than this size, the flow resistance should increase, depending on the deformability of the RBCs. Reduction of the deformability, therefore, increases the probability of capillary stalls. In pathological conditions, the inflammatory response can further exaggerate the capillary stalls. In the Alzheimer disease model, capillary stall containing leukocytes rather uniformly occurred, suggesting overall damage to vascular endothelial cells.4 A significant contribution of leukocyte plugging in the cerebral capillaries for the failure of postischemia reperfusion was also indicated.1 However, other studies have shown mild effects of leukocyte plugging in the capillaries for postischemia no reflow.32,33 The discrepancy in the previous reports could be related to variable inflammation and damage to vascular endothelial cells. The current animal model with optogenetics may therefore help distinguish the dysfunction of vascular mural cells or endothelial cells.

The vasoconstriction of capillaries caused by the photoactivation of ChR2 pericytes with blue light illumination was in good agreement with a previous study.17 In addition, 2-photon laser stimulation of ChR2 pericytes led to significant vasoconstriction of the brain capillaries (Figure 5), which was also consistent with the published work.19 However, an initial attempt by Hill et al16 showed no detectable changes in the capillaries following 2-photon laser stimulation of ChR2 pericytes but significant vasoconstriction in the arteries due to the activation of ChR2 SMCs. As discussed in detail in a later publication, careful optimization of the laser power, wavelength, and scan speed is a prerequisite for the photoactivation of ChR2 with 2-photon laser illumination.18 In addition, the slow dynamics of capillary pericyte reactions relative to SMCs and EPs may contribute to the inconsistency of these observations.34 In this study, a step-function opsin variant, ChR2(C128S), was used for optogenetic manipulation, which differs from the ChR2 variant used in previously published works (ie, H134R variant of ChR2). The benefit of this opsin is that it is light sensitive and stable and permits a prolonged open state35; therefore, it is feasible for optogenetic manipulation of slow dynamic cellular functions. The power of the excitation laser was set to a minimum level below a threshold that causes the pericyte to be damaged (Figure S9).

Cerebral arteries and parenchymal capillaries showed reproducible contractile responses to photoactivation, but veins or venules did not (Figures 3 through 5). The absence of the venous vessel responses was not due to the lack of ChR2 expression, although the level of expression was significantly lower (P<0.05) in the surface vein (9%±4% overlap with the vessel wall area) than in the SA (20%±10%). The results indicate that vSMCs may play a distinct role from other contractile mural cells. This animal model expresses ChR2 in NG2 (nerve/glia antigen 2)-positive cells. Although NG2 is expressed by glial and mural cells,36 an earlier study showed that photoactivation of ChR2-NG2 glia has no effect on CBF.13 The finding indicates that the changes in CBF driven by the photoactivation of ChR2 in NG2 cells are due to the activation of NG2 mural cells. Note that changes in the venous vessel structure were not detected for blue light or 2-photon laser illumination, while the same power of photostimulation produced a 35% reduction in CBF (Figure 2). In turn, a venous vessel maintains a rigid structure insensitive to the upstream flow. This further implies that the stimulation-induced reduction in the cerebral blood volume dominantly occurs in the arteries/arterioles, which is in line with our previous report on blood volume responses to neural activity.37

In conclusion, capillary narrowing in the brain increases the risk of microcirculation embolism, particularly in brain capillaries on the venous sides. Maintaining the normal structure of capillaries with contractile healthy pericytes is key to preventing the age-dependent decline in CBF and dementia. Optogenetic manipulation of the cerebrovasculature and cerebral microcirculation provides a powerful tool to elucidate the causality of the pathogenesis and progression of neurodegenerative diseases and thus to prevent stroke and dementia.



The authors thank Hiroki Suzuki and Takuma Sugashi for their support in analyzing the data.

Supplemental Material

ARRIVE Checklist

Supplemental Methods

Figure S1–S9

Video S1

Reference 38

Nonstandard Abbreviations and Acronyms


ascending venule


cerebral blood flow




ensheathing pericyte


mesh pericyte


nerve/glia antigen 2


penetrating arteriole


red blood cell


surface artery


smooth muscle cell


thin-strand pericyte


venule smooth muscle cell

Disclosures None.


For Sources of Funding and Disclosures, see page 2143.

Supplemental Material is available at

Correspondence to: Kazuto Masamoto, PhD, Graduate School of Informatics and Engineering, Center for Neuroscience and Biomedical Engineering, University of Electro-Communications, 1-5-1 Chofugaoka, Chofu, Tokyo 182-8585, Japan. Email


  • 1. El Amki M, Glück C, Binder N, Middleham W, Wyss MT, Weiss T, Meister H, Luft A, Weller M, Weber B, et al. Neutrophils obstructing brain capillaries are a major cause of no-reflow in ischemic stroke.Cell Rep. 2020; 33:108260. doi: 10.1016/j.celrep.2020.108260CrossrefMedlineGoogle Scholar
  • 2. Erdener SE, Tang J, Kiliç K, Postnov D, Giblin JT, Kura S, Chen IA, Vayisoğlu T, Sakadžić S, Schaffer CB, et al. Dynamic capillary stalls in reperfused ischemic penumbra contribute to injury: a hyperacute role for neutrophils in persistent traffic jams.J Cereb Blood Flow Metab. 2021; 41:236–252. doi: 10.1177/0271678X20914179CrossrefGoogle Scholar
  • 3. Rolfes L, Riek-Burchardt M, Pawlitzki M, Minnerup J, Bock S, Schmidt M, Meuth SG, Gunzer M, Neumann J. Neutrophil granulocytes promote flow stagnation due to dynamic capillary stalls following experimental stroke.Brain Behav Immun. 2021; 93:322–330. doi: 10.1016/j.bbi.2021.01.011CrossrefGoogle Scholar
  • 4. Cruz Hernández JC, Bracko O, Kersbergen CJ, Muse V, Haft-Javaherian M, Berg M, Park L, Vinarcsik LK, Ivasyk I, Rivera DA, et al. Neutrophil adhesion in brain capillaries reduces cortical blood flow and impairs memory function in Alzheimer’s disease mouse models.Nat Neurosci. 2019; 22:413–420. doi: 10.1038/s41593-018-0329-4CrossrefMedlineGoogle Scholar
  • 5. Ali M, Falkenhain K, Njiru BN, Murtaza-Ali M, Ruiz-Uribe NE, Haft-Javaherian M, Catchers S, Nishimura N, Schaffer CB, Bracko O. VEGF signalling causes stalls in brain capillaries and reduces cerebral blood flow in Alzheimer’s mice.Brain. 2022; 145:1449–1463. doi: 10.1093/brain/awab387CrossrefGoogle Scholar
  • 6. Reeson P, Choi K, Brown CE. VEGF signaling regulates the fate of obstructed capillaries in mouse cortex.Elife. 2018; 7:e33670. doi: 10.7554/eLife.33670CrossrefMedlineGoogle Scholar
  • 7. Crumpler R, Roman RJ, Fan F. Capillary stalling: a mechanism of decreased cerebral blood flow in AD/ADRD.J Exp Neurol. 2021; 2:149–153. doi: 10.33696/neurol.2.048CrossrefGoogle Scholar
  • 8. Obrist D, Weber B, Buck A, Jenny P. Red blood cell distribution in simplified capillary networks.Philos Trans A Math Phys Eng Sci. 2010; 368:2897–2918. doi: 10.1098/rsta.2010.0045CrossrefGoogle Scholar
  • 9. Choe YG, Yoon JH, Joo J, Kim B, Hong SP, Koh GY, Lee DS, Oh WY, Jeong Y. Pericyte loss leads to capillary stalling through increased leukocyte-endothelial cell interaction in the brain.Front Cell Neurosci. 2022; 16:848764. doi: 10.3389/fncel.2022.848764CrossrefGoogle Scholar
  • 10. Berthiaume AA, Schmid F, Stamenkovic S, Coelho-Santos V, Nielson CD, Weber B, Majesky MW, Shih AY. Pericyte remodeling is deficient in the aged brain and contributes to impaired capillary flow and structure.Nat Commun. 2022; 13:5912. doi: 10.1038/s41467-022-33464-wCrossrefGoogle Scholar
  • 11. Wei HS, Kang H, Rasheed ID, Zhou S, Lou N, Gershteyn A, McConnell ED, Wang Y, Richardson KE, Palmer AF, et al. Erythrocytes are oxygen-sensing regulators of the cerebral microcirculation.Neuron. 2016; 91:851–862. doi: 10.1016/j.neuron.2016.07.016CrossrefMedlineGoogle Scholar
  • 12. Vanlandewijck M, He L, Mäe MA, Andrae J, Ando K, Del Gaudio F, Nahar K, Lebouvier T, Laviña B, Gouveia L, et al. A molecular atlas of cell types and zonation in the brain vasculature.Nature. 2018; 554:475–480. doi: 10.1038/nature25739CrossrefMedlineGoogle Scholar
  • 13. Oishi M, Passlick S, Yamazaki Y, Unekawa M, Adachi R, Yamada M, Imayoshi I, Abe Y, Steinhäuser C, Tanaka KF. Separate optogenetic manipulation of Nerve/glial antigen 2 (NG2) glia and mural cells using the NG2 promoter.Glia. 2023; 71:317–333. doi: 10.1002/glia.24273CrossrefGoogle Scholar
  • 14. Tanaka KF, Matsui K, Sasaki T, Sano H, Sugio S, Fan K, Hen R, Nakai J, Yanagawa Y, Hasuwa H, et al. Expanding the repertoire of optogenetically targeted cells with an enhanced gene expression system.Cell Rep. 2012; 2:397–406. doi: 10.1016/j.celrep.2012.06.011CrossrefGoogle Scholar
  • 15. Korte N, Ilkan Z, Pearson CL, Pfeiffer T, Singhal P, Rock JR, Sethi H, Gill D, Attwell D, Tammaro P. The Ca2+-gated channel TMEM16A amplifies capillary pericyte contraction and reduces cerebral blood flow after ischemia.J Clin Invest. 2022; 132:e154118. doi: 10.1172/JCI154118CrossrefGoogle Scholar
  • 16. Hill RA, Tong L, Yuan P, Murikinati S, Gupta S, Grutzendler J. Regional blood flow in the normal and ischemic brain is controlled by arteriolar smooth muscle cell contractility and not by capillary pericytes.Neuron. 2015; 87:95–110. doi: 10.1016/j.neuron.2015.06.001CrossrefMedlineGoogle Scholar
  • 17. Nelson AR, Sagare MA, Wang Y, Kisler K, Zhao Z, Zlokovic BV. Channelrhodopsin excitation contracts brain pericytes and reduces blood flow in the aging mouse brain in vivo.Front Aging Neurosci. 2020; 12:108. doi: 10.3389/fnagi.2020.00108CrossrefGoogle Scholar
  • 18. Tong L, Hill RA, Damisah EC, Murray KN, Yuan P, Bordey A, Grutzendler J. Imaging and optogenetic modulation of vascular mural cells in the live brain.Nat Protoc. 2021; 16:472–496. doi: 10.1038/s41596-020-00425-wCrossrefMedlineGoogle Scholar
  • 19. Hartmann DA, Berthiaume AA, Grant RI, Harrill SA, Koski T, Tieu T, McDowell KP, Faino AV, Kelly AL, Shih AY. Brain capillary pericytes exert a substantial but slow influence on blood flow.Nat Neurosci. 2021; 24:633–645. doi: 10.1038/s41593-020-00793-2CrossrefMedlineGoogle Scholar
  • 20. Percie du Sert N, Hurst V, Ahluwalia A, Alam S, Avey MT, Baker M, Browne WJ, Clark A, Cuthill IC, Dirnagl U, et al. The ARRIVE guidelines 2.0: updated guidelines for reporting animal research.J Cereb Blood Flow Metab. 2020; 40:1769–1777. doi: 10.1177/0271678X20943823CrossrefGoogle Scholar
  • 21. Masamoto K, Unekawa M, Watanabe T, Toriumi H, Takuwa H, Kawaguchi H, Kanno I, Matsui K, Tanaka KF, Tomita Y, et al. Unveiling astrocytic control of cerebral blood flow with optogenetics.Sci Rep. 2015; 5:11455. doi: 10.1038/srep11455CrossrefGoogle Scholar
  • 22. Appaix F, Girod S, Boisseau S, Römer J, Vial JC, Albrieux M, Maurin M, Depaulis A, Guillemain I, van der Sanden B. Specific in vivo staining of astrocytes in the whole brain after intravenous injection of sulforhodamine dyes.PLoS One. 2012; 7:e35169. doi: 10.1371/journal.pone.0035169CrossrefGoogle Scholar
  • 23. Mohanty SK, Reinscheid RK, Liu X, Okamura N, Krasieva TB, Berns MW. In-depth activation of channelrhodopsin 2-sensitized excitable cells with high spatial resolution using two-photon excitation with a near-infrared laser microbeam.Biophys J. 2008; 95:3916–3926. doi: 10.1529/biophysj.108.130187CrossrefGoogle Scholar
  • 24. Berthiaume AA, Hartmann DA, Majesky MW, Bhat NR, Shih AY. Pericyte structural remodeling in cerebrovascular health and homeostasis.Front Aging Neurosci. 2018; 10:210. doi: 10.3389/fnagi.2018.00210CrossrefMedlineGoogle Scholar
  • 25. Ross JM, Kim C, Allen D, Crouch EE, Narsinh K, Cooke DL, Abla AA, Nowakowski TJ, Winkler EA. The expanding cell diversity of the brain vasculature.Front Physiol. 2020; 11:600767. doi: 10.3389/fphys.2020.600767CrossrefGoogle Scholar
  • 26. Miyakawa T, Uehara Y, Desaki J, Kimura T, Kuramoto R. Morphological changes of microvessels in the brain with Alzheimer’s disease.Jpn J Psychiatry Neurol. 1988; 42:819–824. doi: 10.1111/j.1440-1819.1988.tb01171.xCrossrefGoogle Scholar
  • 27. Nortley R, Korte N, Izquierdo P, Hirunpattarasilp C, Mishra A, Jaunmuktane Z, Kyrargyri V, Pfeiffer T, Khennouf L, Madry C, et al. Amyloid β oligomers constrict human capillaries in Alzheimer’s disease via signaling to pericytes.Science. 2019; 365:eaav9518. doi: 10.1126/science.aav9518CrossrefMedlineGoogle Scholar
  • 28. Hauck EF, Apostel S, Hoffmann JF, Heimann A, Kempski O. Capillary flow and diameter changes during reperfusion after global cerebral ischemia studied by intravital video microscopy.J Cereb Blood Flow Metab. 2004; 24:383–391. doi: 10.1097/00004647-200404000-00003CrossrefMedlineGoogle Scholar
  • 29. Abe H, Orita M, Arichi S. Erythrocyte deformability in aging.Mech Ageing Dev. 1984; 27:383–390. doi: 10.1016/0047-6374(84)90060-5CrossrefGoogle Scholar
  • 30. Nardini M, Ciasca G, Lauria A, Rossi C, Di Giacinto F, Romanò S, Di Santo R, Papi M, Palmieri V, Perini G, et al. Sensing red blood cell nano-mechanics: toward a novel blood biomarker for Alzheimer’s disease.Front Aging Neurosci. 2022; 14:932354. doi: 10.3389/fnagi.2022.932354CrossrefGoogle Scholar
  • 31. Namdee K, Carrasco-Teja M, Fish MB, Charoenphol P, Eniola-Adefeso O. Effect of variation in hemorheology between human and animal blood on the binding efficacy of vascular-targeted carriers.Sci Rep. 2015; 5:11631. doi: 10.1038/srep11631CrossrefGoogle Scholar
  • 32. Dirnagl U, Niwa K, Sixt G, Villringer A. Cortical hypoperfusion after global forebrain ischemia in rats is not caused by microvascular leukocyte plugging.Stroke. 1994; 25:1028–1038. doi: 10.1161/01.str.25.5.1028LinkGoogle Scholar
  • 33. Ritter LS, Orozco JA, Coull BM, McDonagh PF, Rosenblum WI. Leukocyte accumulation and hemodynamic changes in the cerebral microcirculation during early reperfusion after stroke.Stroke. 2000; 31:1153–1161. doi: 10.1161/01.str.31.5.1153LinkGoogle Scholar
  • 34. Hartmann DA, Coelho-Santos V, Shih AY. Pericyte control of blood flow across microvascular zones in the central nervous system.Annu Rev Physiol. 2022; 84:331–354. doi: 10.1146/annurev-physiol-061121-040127CrossrefMedlineGoogle Scholar
  • 35. Berndt A, Yizhar O, Gunaydin LA, Hegemann P, Deisseroth K. Bi-stable neural state switches.Nat Neurosci. 2009; 12:229–234. doi: 10.1038/nn.2247CrossrefGoogle Scholar
  • 36. Nishiyama A, Komitova M, Suzuki R, Zhu X. Polydendrocytes (NG2 cells): multifunctional cells with lineage plasticity.Nat Rev Neurosci. 2009; 10:9–22. doi: 10.1038/nrn2495CrossrefMedlineGoogle Scholar
  • 37. Kim T, Hendrich KS, Masamoto K, Kim SG. Arterial versus total blood volume changes during neural activity-induced cerebral blood flow change: implication for BOLD fMRI.J Cereb Blood Flow Metab. 2007; 27:1235–1247. doi: 10.1038/sj.jcbfm.9600429CrossrefGoogle Scholar
  • 38. Sekiguchi Y, Masamoto K, Takuwa H, Kawaguchi H, Kanno I, Ito H, Tomita Y, Itoh Y, Suzuki N, Sudo R, et al. Measuring the vascular diameter of brain surface and parenchymal arteries in awake mouse.Adv Exp Med Biol. 2013; 789:419–425. doi: 10.1007/978-1-4614-7411-1_56CrossrefGoogle Scholar


eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.

Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.