Optogenetic approach for functional assays of the cardiovascular system by light activation of the vascular smooth muscle

Abstract

Cardiovascular diseases are the major challenge to modern medicine. Intervention to cardiovascular cells is crucial for treatment of the diseases. Here we report a novel intervention to vascular smooth muscle (VSM) cells by optogenetics. Channelrhodopsin in a tandem with YFP was selectively expressed in smooth muscle of transgenic mice in which YFP fluorescence was found in arterial walls of various tissues. In dissociated VSM cells from the mice blue light evoked inward currents, leading to depolarization and contraction. In isolated mesenteric arterial rings, optostimulation produced vasoconstriction that was reproducible, sustained, light intensity-dependent and comparable to popular vasoconstrictors. Blue light raised robustly coronary resistance without significant effects on heart rate and pulse pressure. Optostimulation produced renal vasoconstriction as well. The optical vasoconstriction had temporal resolutions less than 40s in these organs. These results indicate that optical vasoconstriction can be effectively produced in various organs with channelrhodopsin expression in VSM cells.

Graphical Abstract

1. INTRODUCTION

Cardiovascular diseases are the leading causes of deaths worldwide. According to the National Vital Statistics Reports by the Center for Disease Control in 2010, 29.4% of total causes of death in the USA involve the cardiovascular and cerebrovascular systems (1). One limitation for therapeutic intervention is the accessibility to the cardiovascular system, especially to the vasculature. At present, treatments of the cardiovascular diseases heavily rely on pharmacological agents. Thus, more accessibility to the cardiovascular system is needed for development of effective therapies, especially those with better temporal and spatial resolutions.

Recent development in optogenetics suggests a new way to access the cardiovascular system. The optogenetics is a research approach based on transgenic expression of photosensitive opsins to one group or a few groups of cells, and then these cells are available for optical activation and inhibition (2, 3). Such an optical control of cellular activity has been demonstrated to be highly effective in the manipulation of excitability of individual neurons and recruitment of cells in selective neuronal networks (4–9). Indeed, the optogenetics has been rapidly applied to studies of a number of types of neurons and glial cells (2–11). However, research in the cardiovascular field has barely taken the advantage of the newly developed cell-selective optical intervention. To our knowledge, there is no report on the application of optogenetics to the blood vessels, although optogenetic control of myocardium has been recently demonstrated (12–14).

Therefore, we generated a novel mouse strain with expression of a variant of channelrhodopsin2 (ChR2 H134R) in vascular smooth muscle (VSM) cells. Using tissues from the mice, we tested the expression, membrane excitability, contraction of VSM cells, and vascular tone regulation in several organs. Our results indicate that optical activation of VSM cells can produce potent vasoconstriction that is fast, reproducible, light intensity-dependent and comparable in strength to popular vasoconstrictors.

2. MATERIALS AND METHODS

2.1 Generation of Tagln-ChR Mice

All animal experimental procedures comply with the NIH Guide for the Care and Use of Laboratory Animals, and were approved by the Institutional Animal Care and Use Committee (IACUC) at Georgia State University. Two stains of mice were used to generate mice with ChR2 expression in VSM cells: 1. Tagln-cre mice (Jackson Laboratories, West Grove, PA; 004746, STOCK Tg(Tagln-cre)1Her/J) expressing Cre recombinase under the control of Tagln promoter, and 2. ChR-LoxP mice (Jackson Laboratories; 012569, B6; 129S-Gt(ROSA)26Sor^{tm32(CAG-COP4*H134R/EYFP)Hze}/J) expressing the channelrhodopsin2 (H134R)/eYFP fusion protein following a LoxP-flanked STOP cassette. The male homozygous Tagln-cre mice were crossed with the female heterozygous ChR-loxP mice. Their offspring were found to be ChR2^+/−-Tagln-cre^+/−(Tagln-ChR) or ChR2^−/−-Tagln-cre^+/−(Tagln-cre). Genotypes were determined by using the primers provided by the Jackson Laboratories. Tagln-ChR and Tagln-cre mice were used in the present study as the test and control groups respectively. Genotyping of these mice was shown (Fig. 1). Crossing heterozygous ChR-loxP mice with the homozygous Tagln-cre mice led to the removal of the STOP signal before ChR2-eYFP (Fig. 1A). The offspring Tagln-ChR contained both ChR-loxP and Tagln-cre genes, while ChR-loxP was missing in Tagln-cre mice (Fig. 1B).

FIGURE 1. Design and genotyping of transgenic mice.

A. Constructs of parental ChR-LoxP and Tagln-cre mice as well as the expected offspring that express both ChR-loxP and Tagln-cre by removal of the stop codon (asterisk) flanked by loxP (solid triangle). This led to expression of ChR2-eYFP in a tandem driven by the CAG promoter in Tagln-ChR mice. B. Mice were genotyped with primers provided by Jac son Laboratory which are targeted at the 3″ UTR with a 212 bp expected product. The presence of the 212 bp band indicated ChR2-eYFP mRNA expression in ChR-loxP and Tagln-ChR mice. Note that a 300 bp non-specific band was found in all mice. Another pair of primers yielding a 100bp PCR product in the cre sequence, which was found in Tagln-cre and Tagln-ChR mice. Abbreviations: CAG, cytomegalovirus-immediate-early (CMV-IE enhancer/chic en β-actin/rabbit β-globin hybrid promoter; PolyA, bovine growth hormone polyadenylation signal and flippase recognition target flanked phosphoglycerate kinase-Neo-polyA cassette; GAPDH, glyceraldehyde 3-phosphate dehydrogenase.

2.2 Histology

Mice at the age 2 to 3 months were euthanized by deep anesthesia followed by thoracotomy. The heart, kidney, mesentery, small intestine, uterus and skeletal muscle tissues were collected from the mice. After four hours fixation in 1% Paraformaldehyde (PFA) at room temperature, the tissues were treated overnight with 30% sucrose in Phosphate Buffered Saline (PBS) buffer at 4°C. The tissues were embedded in the Tissue-Tek^TM CRYO-OCT Compound (Andwin Scientific, Torrance, CA, 4583), and cut into 8 μm slices by using the Microm HM 550 Cryostats system (Thermo Scientific, PA, 22-050-337). YFP fluorescence was detected with 514/527 nm (excitation/emission wavelength) filters under the microscope (Carl Zeiss, Gottingen, Germany, Axiovert 200). Images were analyzed with the ImageJ 1.48 (NIH, Bethesda, MD).

2.3 Acute dissociation of VSM cells

Thoracic aorta was dissected free in the dissection solution containing (mM): 140.0 NaCl, 5.4 KCl, 1.0 MgCl₂, 0.1 CaCl₂, 10.0 glucose, 10.0 HEPES (pH was adjusted to 7.4), cut into 1–2 mm segments, and incubated in 0.1 ml dissection solution containing 20 unit/ml papain (Worthington Lakewood, NJ, LK003176) and 1.25 mg/ml TT for 15 min at 37 °C. The tissue segments were further digested with 440 unit/ml collagenase type IA (Sigma, St. Louis, MO, C0130) for 2–3 min at 37 °C, followed by three washes and gentle trituration with a fire polished glass Pasteur pipettes. The solution containing individual VSM cells was dropped in a Petri dish coated with poly-L-lysine (Sigma, P8920) and prepared for patch clamp and contractile studies The cell suspension was stored at 4 C and used within 4 hrs.

2.4 Patch clamp study

The whole-cell voltage clamp and current patch clamp were used to demonstrate optostimulation of isolated aortic VSM cells. Recorded signals were amplified with an Axopatch 200B amplifier (Molecular Devices, Union City, CA), digitized at 10 kHz, filtered at 2 kHz, and collected with the Clampex 10 data acquisition software (Molecular Devices, Union City, CA). The patch pipettes with resistance of 4–6 MΩ were made with 1.2 mm borosilicate glass capillaries. All recordings were performed at room temperature. The optostimulation was performed by using a xenon arc lamp with high speed switcher Lambda DG-4 system (Sutter Instruments, Novato, CA). The light source was connected to the incident-light illuminator port of the microscope, and passed through a 470 nm bandpass filter (~20 mW/mm²). Light pulse trains were generated with the Digitimer D4030 pulse generator (Digitimer Ltd, Letchworth Garden City, UK). The solution applied to the bath contained (in mM) 130.0 NaCl, 10.0 KCl, 1.0 MgCl₂, 1.5 CaCl₂, 10.0 glucose, 10.0 HEPES and 3.0 NaOH (pH 7.4). The internal (pipette) solution contained (in mM) 10.0 KCl, 133.0 K⁺ gluconate, 5.0 EGTA, 5.0 glucose, 1.0 K₂-ATP, 0.5 Na-ADP, and 10.0 HEPES (pH 7.4), and the final Mg²⁺ concentration was adjusted to 1 mM using a [Ca²⁺ ]/[Mg²⁺ ] calculation software.

2.5 Contractile assessment of individual VSM cells

Zeiss Axiovert 200 system was used to observe cell shape changes with optostimulation. The T-Cube LED Driver (Thorlabs, Newton, NJ, LEDD1B) and Blue (470 nm) Fiber-Coupled High-Power LED (Thorlabs, Newton, NJ, M470F1) were used to generate optostimulation (~24 mW/mm²) with optical fiber (Thorlabs, Newton, NJ, FT400EMT). Cell images were recorded continuously with a 20 s interval. Additional cell contraction was achieved with a 5 min exposure to 20 mM KCl. Data were accepted for further analysis if the features of VSM cells were seen with YFP signal and KCl response. Cell shortening was measured along its longitudinal axis by the ImageJ 1.48 software.

2.6 Arterial rings

The vascular rings were prepared by the technique stated in our previous studies (15). Briefly, mouse mesenteric arteries were dissected free, placed in the Krebs solution containing (in mM/L) 118.0 NaCl, 3.6 KCl, 1.2 MgSO₄, 2.5 CaCl₂, 1.2 KH₂PO₄, 11.0 glucose, 25.0 NaHCO₃ (pH 7.4). The arteries were cut into 2 mm ring segments in which a stainless steel triangle wire was inserted The prepared rings were placed in the 37 C tissue bath filled with Krebs solution bubbled with 95% O₂ and 5% CO₂. Changes of the contractile force were measured with a force-electricity transducer. Signals were amplified, collected with the Clampex 9 software (Molecular Devices), and stored in a PC computer. Blue light was generated by T-Cube LED Driver and Blue Fiber-Coupled High-Power LED (Thorlabs). Intensity of six levels with equal interval (from 10 to 24 mW/mm²) was given to the rings in light-intensity test. Other optical contractile force was generated with ~24 mW/mm² light intensity.

2.7 Langendorff heart

After the mouse was euthanized, the heart was rapidly removed and placed in the ice-cold Krebs-Henseleit (KH) solution composed of (in mM): 119 NaCl, 4.7 KCl, 2.5 MgSO₄, 2.5 CaCl₂, 10 glucose, 0.5 disodium EDTA, 25 NaHCO₃ bubbled with 95% O₂ and 5% CO₂ (pH 7.4). The left ventricle was intubated, the aorta was tightened, and vena cava was cut open. The isolated heart was then placed in a moisture air chamber with the temperature maintained at 35 °C, and perfused with a syringe pump (Syringe Pump, Farmingdale, NY, NE-300, NE4000) at constant flow speed to produce 80 cm H₂O ventricular pressure at baseline. The force-electricity transducer (Model eTH-400, CB Sciences) was used to monitor the perfusion resistance and heart rate. Only hearts that had heart beat >200 times per min with a clear vasodilation response to 10 μM isoproterenol were accepted for further studies. The Lambda DG-4 Plus system (Sutter Instruments, Novato, CA) with a 470 nm filter (fiber diameter 3 mm) was used to generate the blue light. The data was recorded with Clampex 8 (Molecular Devices), and analyzed with Clampfit 9 (Molecular Devices). Recording started after 30 min equilibration when the perfusion resistance and heart rate were stable.

2.8 Isolated kidney preparation

Kidneys from both Tagln-ChR and control mice were prepared in ice-cold KH solution. The kidney was intubated through one end of the abdominal aorta with the other end tightened. The rest of experimental procedure was similar to that for the isolated and perfused heart.

2.9 Data analysis

Data are presented as mean ± s.e.m. (standard error). Comparisons of data were accomplished by one-way ANOVA followed by student’s t-test for parametric data. Kruskal-Wallis and Mann-Whitney (U) tests were used for non-parametric data. The differences between means were considered significantly different when P 0.05.

3. RESULTS

3.1 VSM expression of ChR2

We developed a new strain of transgenic mice, in which ChR2 was selectively expressed in smooth muscle cells. Genotyping shows that the offspring of ChR-loxP and Tagln-cre mice are supposed to express ChR2 with the present of Cre recombinase under the control of Tagln promoter (Fig. 1).In various tissues, we have systematically examined the expression of YFP engineered in frame with ChR2. All arteries in the heart, skeletal muscle, kidney and intestine showed YFP fluorescence in the Tagln-ChR mice but none in the control mice (Fig. 2). Similar expression of YFP was found in dissociated individual VSM cells from Tagln-ChR mice (Fig. 4A). In the Tagln-ChR mice YFP fluorescence was also observed in smooth muscle in the intestine and uterus (Fig. 2K–R).

FIGURE 2. Comparison of YFP fluorescence in various tissues between Tagln-ChR and Tagln-cre mice.

In Tagln-ChR mice (A, C, E, G, I), strong YFP fluorescence was found on arterial walls of the aorta, coronary arteries, skeletal muscle, kidney and gastrointestinal arteries, respectively. In Tagln-cre mice (B, D, F, H, J), arteries in these tissues showed no or rather weak background fluorescence. YFP fluorescence was found in the muscularis externa of small intestine in Tagln-ChR mice (K, M) but not in Tagln-cre mice (L, N) (The lower panel was obtained in the same field of the upper one). Blood vessels in the mucosa were also fluorescent in the Tagln-ChR mice. Similar YFP fluorescence expression was seen in the uterus of the Tagln-ChR mice (O, Q) but not of the Tagln-cre mice P R Scale bars are 200 μm

FIGURE 4. Contraction of the dissociated VSM cell by optostimulation.

A. Fluorescence expression of dissociated individual VSM cells was compared between Tagln-ChR and control groups. B. Longitudinal length of a VSM cell for a Tagln-ChR mouse was measured with and without light stimulation (~24 mW/mm²). Complete recovery was seen 2 min after light stimulation. Final length was measured 5 min after addition of 20 mM KCl. C. Statistically, the blue light produced significant shortening in the longitudinal length (**, P<0.01, n=8). Scale bar in A is 10 μm

3.2 Optical activation of acutely dissociated VSM cells

Whole-cell currents were studied in VSM cells acutely dissociated from aorta. In voltage clamp, the aortic VSM cells exhibited typical voltage-dependent outward currents with the membrane potential held at −50 mV and stepped from −100 to 40 mV (Fig. 3A). Under this condition, exposure of the VSM cells obtained from Tagln-ChR mice to 10 ms pulses of blue light (470 nm) evoked large inward currents (Fig. 3B). The photo currents had a fast onset (0.7 ± 0.04 ms by 10%–90% peak, n = 6 cells) and slow decay with the time constant (30.8 ± 1.7 ms by 90%–10% peak, n = 6 cells). The slow decay was apparently produced by passive membrane properties, as the photo currents showed very little inactivation. With prolonged optostimulations in 200 ms and 300 ms, blue light evoked persistent inward photo currents showing very little adaptation or inactivation over the time periods (Fig. 3C). In the current-voltage (I-V) relationship, the photo currents had the reversal potential ~0 mV with moderate inward rectification (Fig. 3D). In current-clamp, 300 ms blue light stimulation produced depolarization reliably (Fig. 3E), which significantly changed the membrane potentials of VSM cells from −45.4 ± 1.5 mV to −35.1 ± 2.1 mV (n = 5 cells, P<0.01) (Fig. 3F).

FIGURE 3. Optical excitation of dissociated VSM cells.

A. Whole-cell currents were recorded from a VSM cell dissociated from aorta of a Tagln-ChR mouse in voltage clamp. Steps of voltage commands (from −100 to 40 mV with a 10 mV increment) were applied to the cell at a holding potential of −50 mV. B. Stimulation of the VSM cell with 10 ms pulses of blue light (470 nm) (~24 mW/mm²) evoked inward currents. The photo currents onset rapidly, and decayed slowly with the light on and off, which are better seen in the expanded display on the right panel. C. Responses of another VSM cell to longer durations of blue light stimulations. The inward photo currents induced by 200 ms and 300 ms blue light showed only modest reduction in current amplitudes. D. Current-voltage (I-V) relationship of the photo currents showed a reversal potential at 5 mV with moderate inward rectification. E. In current clamp, blue light stimulation produced large depolarization in a VSM cell. F. Comparison of membrane potentials before and during light exposure ata are presented as means ± s e P<0.01 Student’s t-test; n=5 cells).

3.3 Blue light evoked VSM cell contraction

The light-evoked depolarization was expected to affect the contractility of VSM cells. Therefore, we measured the length of isolated VSM cells with and without 470nm light exposure. In Tagln-ChR mice, 10 s blue light exposure shortened individual aortic VSM cells. Typical VSM contraction was the shortening in the longitudinal axis of the VSM cells. The contraction was quantified by measurement of the size in the longitudinal axis of VSM cells. The blue light produced significant shortening in the VSM cells from the Tagln-ChR mice (P<0.01, n=8) (Fig. 4A, B), which could have been even greater if there had been a way to add a preload to the cells. No change in cell shape from WT mice was observed (Fig. 4C). Cell length was recovered to about original level within 2 min (P<0.01, n=8). The contractility was tested by additional 20 mM KCl with further length shrinkage observed (Fig. 4B, C).

3.4 Optostimulation initiated contractions of isolated and perfused mesenteric arteries

The light-evoked VSM contraction should have impact on vascular tones. To demonstrate this, we studied vascular constriction in isolated and perfused mesenteric arterial rings from control and Tagln-ChR mice. Vasoconstriction was observed with the blue light exposure (Fig. 5A). Using high intensity and long-lasting (up to 1 hr) optostimulation, we studied the kinetics of vasoconstriction. The vasoconstriction started fast with a 100 ms latency, reached the peak level within 10 s, and was maintained at the level throughout the photo stimulation period with only a slight decline in the contractile force (Fig. 5A, F). Clear vasoconstriction occurred with the light intensity ~10 mW/mm², and stronger photo stimulations produced stronger vasoconstrictions (Fig. 5B). The contractile force generated by blue light (~24 mW/mm²) was comparable to that produced by 1 μM phenylephrine (16) a selective α-adrenergic receptor agonist, and even larger than that produced by 60 mM KCl (Fig. 5C, E). To test the contraction reversibility, we repetitively stimulated the rings by 1 min light followed by 2 min time interval. As shown in the figure (Fig. 5C), the repetitive optostimulations triggered repetitive vasoconstrictions with modest augmentation in the contractile force to subsequent stimulations. After each photo constriction, baseline contractile force was slightly elevated, which dropped to original level in 1–2 min (Fig. 5C). In contrast, the rings from control (Tagln-cre) mice showed no response to any optostimulation (Fig. 5D). Contractile force which was produced by 1 μM PE (0.26 ± 0.03 g, n=7), 60 mM KCl (0.19 ± 0.04 g, n=7) and light (0.23 ± 0.02 g, n=7) were compared (Fig. 5E). No significant differences were found among these groups. The vasoconstriction force at the peak level of light exposure (0.24 ± 0.03 g, n=7) was not significantly different from the force in the plateau (0.23 ± 0.02 g, n=7) (Fig. 5F). The optostimulation had no significant residual effect on subsequent PE- and KCl- induced vasoconstrictions (Fig. 5G). Long-lasting optostimulation (1 hr) produced sustained vasoconstriction without any relaxation(Fig. 5H). Indeed, the contractile force increased by 24.7 ± 17 % (n=4) at the end of the 1 hr simulation (Fig. 5I). The sustained vasoconstriction was reversible, and vascular tension returned the baseline level within 10 min after light was turned off (Fig. 5H).

FIGURE 5. Characterization of optical vasoconstriction in isolated mesenteric arterial rings.

A. A mesenteric ring was obtained from a Tagln-ChR mouse and studied in-vitro with a 0.3 g preload. Exposure of the ring to blue light (~24 mW/mm²) produced vasoconstriction that started fast when the light was on, was maintained roughly at the same level during the light exposure, and returned to the baseline level rapidly. B. Graded increases in the light intensity led to graded increases in forces of vasoconstriction (from ~10 mW/mm² to ~24 mW/mm²). C. In another ring from a Tagln-ChR mouse, vasoconstriction was first produced with PE at 0.1, 1 and 10 μM followed by the vasodilator acetylcholine Ach 1 μM and washout WS The lac of Ach effect indicated that endothelium was denuded. Exposures to 60 mM KCl also produced vasoconstriction. Subsequently, the ring was exposed to three pulses of blue light, each of which led to vasoconstriction. The amplitude of the light-evoked vasoconstriction increased slightly in the 2^nd and 3^rd exposures. After 15 min rest, the PE and KCl vasoconstrictions were repeated, and the same results were shown Pinacidil 10 μM a vasodilator relaxed rings D. The same experiments were done in another ring from control mice. The ring showed the same responses to all treatments except light. E. Contractile force was produced by PE, KCl and light. No significant differences were found among these treatments. F. The vasoconstriction force at the beginning of light exposure (peak) was not significantly different from the force in the later steady level. G. Comparison of vasoconstrictions by PE and KCl. The vasoconstriction forces at all tested concentrations of PE and KCl showed no significant difference before and after light exposure. Constant optostimulation was indicated as blue bars. H. Long-term stimulation (~24 mW/mm², 1 hr) produced vasoconstriction that started fast when the light was on and did not decline during the light exposure. The vascular tension returned to the initial baseline level within 10 min after light off. I. With the long-lasting optostimulation, the stable contractile force lasted for 1 hr without decline (n=4).

3.5 Manipulation of coronary arterial tones with optostimulation

To further test the effects of light-controlled vascular tones on tissue perfusion at the organ-typical level, we studied coronary vascular responses using the Langendorff heart preparation (17). The heart was prepared to enable the coronary circulation at a constant perfusion volume. The basal perfusion volume was firstly determined at the arterial pressure 80 cm H₂O. Then the perfusion volume was maintained constantly at this level, and the pressure of the coronary arteries was measured with a force-electricity transducer. The viability of the preparation was verified in each heart by the stable heart rate >200 bpm (beat/min) and the vasodilation response to isoproterenol a β-adrenergic receptor agonist (17–19).

The heart rate averaged 300 bpm at about 35°C, consistent with previous observations in isolated heart preparations (20, 21). In the Tagln-ChR heart, optostimulation produced a marked increase in coronary resistance which stands for the coronary resistance with the constant pump output (Fig. 6A), which was not seen in the control heart (Fig. 6B). In both hearts, PE (10 μM) gave rise to a small increase of coronary resistance (Fig. 6A, B). The exposure to 10 μM isoproterenol brought about a clear vasodilation of coronary arteries with increases in the heart rate and the pulse pressure (Fig. 6A, B). Optostimulation (3 min) did not have any evident effects on the heart rate and pulse pressure in both groups of hearts (Fig. 6A, B). In comparison, the light exposure raised the coronary pressure by 20.1 ± 1.6 cm H₂O with ~20 mW/mm² intensity (n=11 hearts), while 10 μM PE increased the pressure by 15.1 ± 3.1 cm H₂O (n=11) (Fig. 6C).

FIGURE 6. Constriction of coronary arteries by optostimulation.

A. In an isolated and perfused heart from a Tagln-ChR mouse, a marked increase in coronary perfusion resistance was produced by blue light followed by PE and isoproterenol (Isop) treatments. Note that the optical vasoconstriction with ~20 mW/mm² blue light was much greater than that produced by 10 μM PE. The optical coronary vasoconstriction did not cause evident changes in heart rate and pulse pressure, while both were augmented by Isop. The heart rate and pulse pressure were obtained from areas indicated by arrows. B. In a control heart, light did not produce coronary vasoconstriction. C. In comparison, the vasoconstriction force was greater by light than by 10 μM PE n=11 D. Time responses of the optostimulation. E. long-term stimulation (30 min) raised coronary pressure to a relatively stable level with a slight decline. The coronary perfusion resistance went back to the initial baseline level in a few minutes after optostimulation. F. The coronary resistance was maintained at a relative stable level during a 30 min optostimulation without obvious decline (n=4). G. long-lasting light stimulation induced the cardiac arrhythmia. Severe arrhythmia was observed after 15 min optostimulation, such as heart block and cardiac arrest.

The temporal resolution was studied by measuring the onset and offset times for the coronary vasoconstriction (Fig. 6D). The onset latency was 1.1 ± 0.2 s (n=7), and the time to 90% peak vasoconstriction was 29.5 ± 8.4 s (n=6). The offset latency was 1.1 ± 0.3 s (n=7), and the time of withdrawal to 10% peak vasoconstriction was even shorter as 15.0 ± 3.1 s (n=7). These indicate that the optical vasoconstriction has a high temporal resolution in isolated hearts.

With a prolonged period (30 min) optostimulation, the increase in coronary resistance was well maintained showing only modest or no relaxation (Fig. 6E, F). However, such a long-lasting optostimulation had an effect on the heart rate. Arrhysmia occurred after 10 min, and became very severe at 20 min (Fig. 6G), suggesting that coronary perfussin is indeed restrained.

3.6 Light stimulation increased vessel resistance in perfused kidneys

In isolated and perfused kidneys from Tagln-ChR and Tagln-cre mice, renal vascular tension was measured with a constant perfusion flow that produced the renal pressure 100 cm H₂O at baseline. The viability of kidney was tested by 10 μM PE and 10 μM isoproterenol. In the Tagln-ChR kidney, optostimulation with ~20 mW/mm² intensity raised the renal pressure by 20–30 cm H₂O (Fig. 7A). Such a light response was not seen in control kidney (Fig. 7B), although its PE response was the same to the Tagln-ChR kidney. Vasodilation effects of 10 μM isoproterenol were relatively small in both groups of idneys li ely due to the lac of expression of β2 adrenoceptors (22). In comparison, the Tagln-ChR kidney responded to light with an increase of renal pressure by 27.2 ± 2.5 cm H₂O (n=7), while the pressure reached 126.9 ± 12.0 cm H₂O with 10 μM PE exposure (Fig. 6C). The temporal resolution was 38.3 ± 8.7 s (n=6) for the onset time of 90% peak vasoconstriction, and 30.4 ± 4.43 s (n=6) for the time to return to baseline (10% peak vasoconstriction) (Fig. 7D).

FIGURE 7. Optical vasoconstriction in isolated kidney.

A, B. Optical vasoconstriction was observed in an isolated and perfused kidney from a Tagln-ChR mouse but not that from a control mouse. C. The optical renal vasoconstriction was compared with that produced by 10 μM PE D. Onset/offset time responses to optostimulation. E. Long-lasting optostimulation (30 min) induced a prolonged increase in vessel resistance. F. The renal vascular resistance took about 4 min to reach the peak level and slightly declined (by 19 %, n=4) at the end of 30 min stimulation.

The renal vasoconstriction was well maintained with prolonged optostimulation (up to 30 min). After the peak pressure increase was reached at about 4 min, renal vascular pressure dropped modestly at 30 min followed by a full recovery (Fig. 7E). On average the renal resistance did not significantly decline by the end of 30 min of opteostimulation (Fig. 7F).

4. DISCUSSION

This is the first demonstration of optical vasoconstriction using an optogenetic approach. By taking the advantage of Cre-Lox mice, we have generated a new strain of transgenic mice that express ChR in VSMs. This allows us to stimulate the VSM cells and produce optical vasoconstriction. The optical vasoconstriction is robust, reproducible, light intensity-dependent, and superb in temporal resolution, which we have observed in various tissues.

4.1 Successful generation of VSM optogenetic mice

The Tagln gene encodes the protein transgelin, an actin cross-linking protein found in smooth muscle cells. Tagln mRNA is detected in the aorta, uterus, intestine and lung at high levels, especially in the SMC lines (23, 24). The promoter Tagln has been well characterized, and used previously to deliver genes in the tissue-specific manner (25–27). By cross-breeding mice with Cre driven by the Tagln promoter with LoxP-ChR2 mice, we have developed a new mouse strain, in which ChR is expressed in VSM cells. This is supported by our morphological studies of YFP fluorescence in various tissues. Strong YFP fluorescence is found only in vessel walls, although several tissues in both control and Tagln-ChR mice have weak background fluorescence that can be easily distinguished by the fluorescence intensity. Weak YFP fluorescence has been observed in veins of Tagln-ChR mice, especially large veins that have a thin layer of smooth muscle as well. In addition to VSM, we also examined YFP fluorescence in the skeletal muscle, heart, kidney, intestine and uterus. YFP fluorescence is negative in skeletal muscle, adipocyte, heart, and kidney, and is weak in intestine and uterus. In kidney tissues where pericytes and intraglomerular mesangial cells are located, we did not find any YFP fluorescence, sugging that ChR may not be expressed in these cells.

4.2 Characteristics of the optical activation of VSM cells

Photo currents have been recorded in VSM cells dissociated from aorta of Tagln-ChR mice in voltage clamp. These are inward currents seen more clearly at hyperpolarized membrane potentials. Their reversal potential is close to 0 mV, indicating that the currents are carried by cations non-selectively. These currents show very little adaptation/inactivation with a prolonged light exposure, and show moderate inward rectification in the I-V relationship, which are consistent with our results of optical vasoconstriction showing little/no adaption as well. In current clamp, optical activation of these currents produces depolarization known to be necessary for VSM contraction. The VSM cells in our study seem to have a hight level of ChR2 expression although the critical expression intensity for activation individual VSM cell is unknown. All these indicate that ChR has been successfully expressed in VSM cells of the Tagln-ChR mice (27, 28).

4.3 Effective vasoconstriction by optostimulation

We have studied the effects of optostimulation on vascular tones in several tissues. In isolated VSM cells from the aorta, shortening in cell length is seen. We believe that the optical activation effect is under-presented in the dissociated cell preparation. The VSM cell contraction could be greater if there were a preload added to the cells. It is known that the preload can drastically enhance muscle contraction according the muscle length-tension relationship.

Indeed, the optical vasoconstriction is found quite robust in the isolated and perfused mesenteric rings. The force produced by the optical vasoconstriction is comparable to 1 μM PE or 60 mM KCl, both of which are known to contract VSMs in mesenteric rings potently. The optical vasoconstriction is reproducible, which is seen with repetitive optostimulation, and shows no decline in the contractile force. The optical vasoconstriction is light intensity-dependent, which is seen with blue light as low as ~10 mW/mm².

We have observed optical vasoconstriction in isolated and perfused heart and kidney. Light penetration is limited in these organs, allowing the blood vessels on and close to the surface to be stimulated. Even with the poor transparence, optostimulation produces strong vasoconstriction. In the coronary circulation, the optical vasoconstriction is more potent than the vasoconstrictions produced by 10 μM PE. In addition to the robustness, optical vasoconstriction has an extraordinarily high temporal resolution in isolated organs. The onset and offset times (measured as the 10–90% peak value) are within 40 s, which cannot be achieved by any non-invasive approaches at present. The optical vasoconstriction is weaker in the renal circulation, which is likely to be due to anatomical nature of the kidney. Unlike the heart, the kidney is not a hollow organ in which substantial blood vessels may be not accessible to surface light. In perfused heart, light source was positioned on the surface of the heart, adjacent to aortic root from which left and right coronary arteries branch off. About 30% of the superficial vessels of the heart can be directly illuminated (single beam, 3mm diameter light guide). In contrast, the coverage area of the light accounts for up to 35% of the entire kidney surface. Attenuation coefficients of solid portion of heart and kidney at 470 nm are similar (29). Therefore hollow chamber of the heart helps the heart to gain better light transmittance. In addition, a change in the tissue thickness from 1.0 to 1.1 mm leads to a 10% decrease of light intensity. Thus, only a small proportion of blood vessels near to the surface are activated by optostimulation. As a result, optostimulation may not lead to strong vasoconstriction.

No evident injury or side effect was seen in the vasculature even after repetitive and long-lasting optostimulation (1 hr in rings and 30 min in perfused organs). This is consistent with the fact that VSM cells unlike neurons can survive relatively hight Ca²⁺ load to maintain persistent contractile forces (30).

4.4 Potential usage in biomedical research

The preparation of optical vasoconstriction has several potentials for research applications. Firstly, owing to the nature of optostimulation, blood vessels in a well-defined region can be activated, an approach that is not currently available with systemic administration of vasoconstrictors. Such an approach can be useful for the understanding of cortical functions of the brain. The spatial accuracy of optical vasoconstriction could be used to reveal the mechanism of neuron-astrocyte-vascular coupling in certain brain areas, such nucleus regulating breathing (10, 31). Secondly, the temporal resolution and robustness of optical vasoconstriction may allow a better control of vascular tension where timing is critical. The approach may be useful for the study of microcirculation when precision controls of the arteriole tone are needed. In the presence of a basal level of vasodilator, optical vasoconstriction may allow a fast manipulation of the arteriole tone to desired levels. Thirdly, the optical vasoconstriction can be used to set a basal vascular tone for further vasodilation by potential therapeutic agents. The optogenetic approach does not involve vasoconstrictors, many of which may interfere with the vasodilators to be studied. Finally, the optical vasoconstriction may be useful to manipulate the metabolic rate temporospatially by reducing O₂ and nutrient supplies to certain tissue. This even allow clinical applications when rhodopsins can be delivered to a specific tissue. Our findings thus may open an avenue for future studies to achieve such a therapeutic objective.

In conclusion, this study shows a novel intervention to vascular tones by optical activation of VSM cells that express ChR. In dissociated VSM cells, blue light evokes large inward currents, leading to depolarization and contraction. In isolated mesenteric arterial rings, optostimulation produces vasoconstriction that is reproducible, sustained and light intensity-dependent. In isolated and perfused heart, blue light raises coronary resistance even more potently than that produced by 10 μM PE with fast onset and offset responses. These results indicate that the optogenetics can be effectively applied to the vasculature, and thus opens a brand new avenue for intervention to the cardiovascular system.