2.1.

Cell Culture

Primary cultures of rat aortic smooth muscle cells were prepared from the thoracic and abdominal aortas of $5\text{-week}$ -old male Sprague-Dawley rats $\left(150\text{to}200\mathrm{g}\right)$ , as described previously.¹³ The cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum and were used between passages 5 and 10. Purity $(>95\mathrm{\%})$ was confirmed by the staining of $\alpha$ -smooth muscle actin.

2.2.

Murine Thinned Skull Model

All experiments used $7\text{-}\text{to}10\text{-week}$ -old male ICR mice (Charles River Japan, Yokohama, Japan). The procedures for the thinned skull preparation were as described previously.¹⁴ After anesthesia with an intraperitoneal injection of ketamine-xylazine, the scalp is removed and the skull is glued to a custom-designed metal plate (Namil Optical Components, Incheon, Korea) using dental cement. Using a high-speed microdrill, a circular area of skull, $2\text{to}4\mathrm{mm}$ in diameter, over the region of interest is thinned under a dissection microscope until a thin, smooth preparation is achieved. The animal care and experimental procedures were performed under the approval of the Animal Care Committee of KAIST (Daejeon, Korea).

2.3.

Two-Photon Fluorescence Imaging

We used two-photon laser scanning microscopy (LSM510; Carl Zeiss, Oberkochen, Germany ) with a femtosecond pulsed laser (Chameleon; Coherent, Santa Clara, California ) tuned to $800\mathrm{nm}$ and a water immersion objective lens ( $20\times$ , 1.0 numerical aperture). For cellular calcium imaging, cultured cells were loaded with $10\mu \mathrm{mol}∕\mathrm{L}$ of the acetoxymethyl form of Fluo-4 (Fluo-4AM; Molecular Probes, Eugene, Oregon ) at $37°\mathrm{C}$ for $40\min$ in DMEM and then washed with phosphate-buffered saline (PBS). For membrane staining, the cells were loaded with $5\mu \mathrm{g}∕\mathrm{ml}$ Alexa Fluor 594-conjugated wheat germ agglutinin (WGA-Alexa Fluor 594; Invitrogen, Carlsbad, California ) at $37°\mathrm{C}$ for $5\min$ in Hank’s balanced salt solution and then washed twice with PBS. For animal imaging, after the skull thinning surgery, we intravenously injected 2-MDa FITC-dextran ( $100\mu \mathrm{l}$ of 2% w/v in PBS; FD-2000S; Sigma), and the mice were placed on a customized frame (Namil Optical Components, Incheon, Korea) designed for minimizing movement artifacts (Fig. 1 ). We imaged the baseline state of a region of interest for $10\mathrm{s}$ and subsequently imaged the region for up to $1\text{to}2\min$ after laser irradiation. The average laser power used for imaging was less than $50\mathrm{mW}$ .

Fig. 1

In vivo experimental setup. (a, b) Schematics of the animal imaging setup. aCSF, artificial cerebrospinal fluid. (c) A photograph of the animal imaging setup.

2.4.

Laser Irradiation

The optical path and laser source for optical intervention were the same as those used in the two-photon fluorescence imaging. For the in vitro study, we selected a target region of $<0.09\mu {\mathrm{m}}^2$ in the cytosol of a cultured VSMC. The irradiation duration was fixed at $1.6\mathrm{ms}$ , and the average laser power was set to $300\text{to}450\mathrm{mW}$ . We started with low laser power and increased it until a calcium response was observed. For the animal experiments, we selected target arteries that coursed parallel to the cortical surface and that had lumen diameters of $15\text{to}30\mu \mathrm{m}$ and subsurface depths of up to $100\mu \mathrm{m}$ in a thinned skull window model. The arterial vasculature was confirmed by measuring the direction of blood flow in the branching vessels. To target the media layer of the arterial wall, a region located $3\text{to}5\mu \mathrm{m}$ from the luminal surface was selected for irradiation. The irradiation duration was fixed at $1.6\mathrm{ms}$ . The average laser power was set initially at $300\mathrm{mW}$ and was increased to $900\mathrm{mW}$ , until vasoconstriction was observed.

2.5.

Data Analysis

We used ImageJ or MATLAB for image processing and data quantification. Data are expressed as the $\text{means}\pm \mathrm{S}.\mathrm{E}.\mathrm{M}$ .

3.1.

Laser-Induced Calcium Signals in VSMCs

To test whether femtosecond pulsed laser stimulation can trigger calcium waves in VSMCs, we irradiated rat aortic VSMCs stained with the calcium indicator Fluo-4AM. A target cell was selected using the basal calcium fluorescence, and the cytosol region was irradiated as described in Sec. 2. Before starting the laser stimulation experiments, we confirmed that repetitive imaging with an average power of less than $50\mathrm{mW}$ did not cause any notable changes in the calcium signal or cell viability (data not shown). Considering that the laser-induced calcium response in other cell types is known to exhibit a power-dependent, all-or-nothing response, we gradually increased the laser power used for cell stimulation until we observed a change in the calcium signal in the cell.¹⁵ For most cells, we observed an approximately twofold increase in calcium fluorescence signal, which reached a maximum at $\sim 1\mathrm{s}$ and was followed by a subsequent decrease to the basal level. Nonresponding cells underwent necrotic cell death, characterized by loss of the calcium signal and membrane integrity, on repeated irradiation with increasing laser power. In the irradiated cell, increase in a calcium signal was observed in both nuclear and cytosol regions, with a greater signal intensity in the nuclear region, and punctate calcium pattern was observed in the cytosol at the later time points. The calcium wave propagated to the neighboring cells, and the propagation was blocked by treating with a gap junction blocker, 18 $\beta$ -glycyrrhetinic acid, suggesting that the gap junction mediated propagation of the calcium signal (data not shown). To produce a dose-response curve for the laser-induced calcium signals, we irradiated randomly selected cells ( $n=30$ for each laser power) with seven different laser powers (32.9, 99, 164, 230, 296, 370, 444, and $740\mathrm{mW}$ ) and found that $300\mathrm{mW}$ was enough to trigger a calcium signal in VSMCs (Fig. 2 ; also see Video 1 ). We used this laser power for the following in vitro experiments.

Fig. 2

Laser-induced calcium signal in a VSMC. (a) Representative temporal dynamics of the laser-induced calcium wave in a cultured VSMC. The red dot indicates the irradiated region. The numbers above each image indicate the time after laser stimulation at $0\mathrm{s}$ . Scale bar, $20\mu \mathrm{m}$ . (b) Quantification of laser-induced calcium dynamics in the irradiated cell and the neighboring cells. The numbers in the baseline image in (a) indicate the corresponding cells. (c) Quantification of laser-induced calcium dynamics. The gray lines indicate error bars $(n=15)$ . (d) Laser power dependency of calcium wave generation in VSMCs. $n=30$ for each point.

Representative temporal dynamics of the laser-induced calcium wave in a cultured VSMC. The red dot indicates the irradiated region. Scale bar, $20\mu \mathrm{m}$ (MPEG, $1.4\mathrm{MB}$ ). .

3.2.

Laser-Induced Contraction of VSMCs

Elevation of intracellular calcium ion concentration is a well-known signaling pathway responsible for VSMC contraction.¹ We investigated whether the laser-induced calcium increase leads to VSMC contraction. To visualize cell contraction, we used a plasma membrane-staining dye, WGA-Alexa Fluor 594, together with a calcium indicator. Repetitive imaging did not show any noticeable changes in plasma membrane morphology, except minimal photobleaching acceptable for structural imaging (data not shown). Subthreshold laser stimulation did not induce changes in cell morphology or the calcium signal. Following suprathreshold laser stimulation, the plasma membrane showed gradual shrinkage, indicating active cell contraction ( $2.6\pm 0.2\mathrm{\%}$ cell area change per min), along with rapid calcium responses consistent with the previous experiments. After $100\mathrm{s}$ , the cell area had changed to $97\pm 0.3\mathrm{\%}$ $(n=15)$ of the baseline area (Fig. 3 ; also see Video 2 ).

Fig. 3

Laser-induced VSMC contraction. (a) Representative temporal dynamics of laser-induced VSMC contraction. The calcium signal and membrane morphology were imaged simultaneously. The white dashed lines indicate the cell margin at the baseline, and the yellow dashed lines indicate the cell margin at $100\mathrm{s}$ . The numbers indicate the time after laser stimulation at $0\mathrm{s}$ . The red dot indicates the irradiated region. Scale bar, $20\mu \mathrm{m}$ . (b) Quantification of laser-induced calcium dynamics and cell area change $(n=15)$ .

Representative temporal dynamics of the laser-induced VSMC contraction. The red dot indicates the irradiated region. Scale bar, $20\mu \mathrm{m}$ (MPEG, $1.4\mathrm{MB}$ ). .

3.3.

Laser-Induced Arterial Contraction

Next, we examined whether this laser-induced VSMC contraction occurred in a live animal (Fig. 4 ). We used a murine thinned skull model to optically access the cortical vasculature in the brain, in a minimally invasive manner. The vascular lumen was visualized by an intravenous injection of 2-MDa FITC-dextran, which cannot pass through the vessel wall. Arteries were selected by measuring the blood flow directions in the branching vessels, using a line-scanning method. To target the arterial media layer containing smooth muscle, a region located $3\text{to}5\mu \mathrm{m}$ from the luminal surface was irradiated.

Fig. 4

Laser-induced cortical arterial constriction. (a) Vascular structure imaged through a thinned skull window after an intravenous injection of fluorescein conjugated to 2-MDa dextran. The dashed square indicates the region of interest for images in (b). Scale bar, $100\mu \mathrm{m}$ . (b) Representative time points after irradiation. The red dot and dashed line indicate the irradiated region and baseline vessel wall, respectively. Scale bar, $20\mu \mathrm{m}$ . A line scan image reconstructed using time series images is shown on the right. (c) Time course tracing of lumen diameter $(n=6)$ . (Color online only.)

Laser irradiation focused in the arterial vessel wall caused localized vascular contraction, followed by recovery. The vascular lumen contracted to $74.3\pm 3.9\mathrm{\%}$ $(n=5)$ of the baseline diameter, and contraction peaked at $\sim 10\mathrm{s}$ after irradiation. Recovery was completed within $1\min$ , with almost complete reversal of the vascular diameter. The arteriolar contraction dynamics with laser stimulation were comparable to those with other stimuli.¹ Irradiation of a spot in the vessel wall resulted in the rapid lateral spread of contraction within a few seconds.