Demonstrated here are protocols for (1) freshly isolating intact cerebral endothelial “tubes” and (2) simultaneous measurements of endothelial calcium and membrane potential during endothelium-derived hyperpolarization. Further, these methods allow for pharmacological tuning of endothelial cell calcium and electrical signaling as individual or interactive experimental variables.
Cerebral arteries and their respective microcirculation deliver oxygen and nutrients to the brain via blood flow regulation. Endothelial cells line the lumen of blood vessels and command changes in vascular diameter as needed to meet the metabolic demand of neurons. Primary endothelial-dependent signaling pathways of hyperpolarization of membrane potential (Vm) and nitric oxide typically operate in parallel to mediate vasodilation and thereby increase blood flow. Although integral to coordinating vasodilation over several millimeters of vascular length, components of endothelium-derived hyperpolarization (EDH) have been historically difficult to measure. These components of EDH entail intracellular Ca2+ [Ca2+]i increases and subsequent activation of small- and intermediate conductance Ca2+-activated K+ (SKCa/IKCa) channels.
Here, we present a simplified illustration of the isolation of fresh endothelium from mouse cerebral arteries; simultaneous measurements of endothelial [Ca2+]i and Vm using Fura-2 photometry and intracellular sharp electrodes, respectively; and a continuous superfusion of salt solutions and pharmacological agents under physiological conditions (pH 7.4, 37 °C). Posterior cerebral arteries from the Circle of Willis are removed free of the posterior communicating and the basilar arteries. Enzymatic digestion of cleaned posterior cerebral arterial segments and subsequent trituration facilitates removal of adventitia, perivascular nerves, and smooth muscle cells. Resulting posterior cerebral arterial endothelial "tubes" are then secured under a microscope and examined using a camera, photomultiplier tube, and one to two electrometers while under continuous superfusion. Collectively, this method can simultaneously measure changes in endothelial [Ca2+]i and Vm in discrete cellular locations, in addition to the spreading of EDH through gap junctions up to millimeter distances along the intact endothelium. This method is expected to yield a high-throughput analysis of the cerebral endothelial functions underlying mechanisms of blood flow regulation in the normal and diseased brain.
Blood flow throughout the brain is regulated by the coordination of vasodilation among cerebral arteries and arterioles in vascular networks1. Endothelial cells lining cerebral resistance arteries command changes in vascular diameter as needed to meet the metabolic demand of neurons1,2,3. In particular, during endothelium-derived hyperpolarization (commonly known as EDH), intracellular Ca2+ ([Ca2+]i) and electrical signaling in endothelial cells coordinate vasodilation among endothelial cells and their surrounding smooth muscle cells through gap junctions for arterial relaxation4. Physiological initiation of EDH sequentially entails stimulation of Gq-coupled receptors (GPCRs), an increase in [Ca2+]i, and activation of endothelial small- and intermediate-Ca2+-activated K+ (SKCa/IKCa) channels to hyperpolarize cerebral endothelial membrane potential (Vm)5,6,7. Thus, the intimate relationship of endothelial [Ca2+]i and Vm is integral to blood flow regulation and indispensable to cardio- and cerebrovascular function6,8. Throughout the broader literature, numerous studies have reported the association of vascular endothelial dysfunction with the development of chronic diseases (e.g., hypertension, diabetes, heart failure, coronary artery disease, chronic renal failure, peripheral artery disease)9,10, indicating the significance of studying endothelial function in physiological as well as pathological conditions.
Vascular endothelium is integral to the production of hyperpolarization, vasodilation, and tissue perfusion and thus, examination of its native cellular properties is crucial. As a general study model, preparation of the mouse arterial endothelial tube model has been published before for skeletal muscle11,12, gut13, lung14, and recently for the brain6. Studies of simultaneous [Ca2+]i and Vm measurements in particular have been published for skeletal muscle arterial endothelium15,16 as well as lymphatic vessel endothelium17. In addition to primary studies utilizing the endothelial tube approach, a comprehensive review of its advantages and disadvantages8 can be consulted to determine if this experimental tool is appropriate for a specific study. In brief, an advantage is that the key physiological components of endothelial cell function are retained (e.g., Ca2+ influx and intracellular release, hyperpolarization of Vm up to the Nernst potential for K+ via SKCa/IKCa activation, and endothelial intercellular coupling via gap junctions) without confounding factors such as perivascular nerve input, smooth muscle voltage-gated channel function and contractility, circulation of blood, and hormonal influences8. In contrast, commonly used cell culture approaches introduce significant alterations in morphology18 and ion channel expression19 in a manner that can greatly obfuscate comparisons to physiological observations determined ex vivo or in vivo. Limitations include a lack of integration with other essential components for regulating blood flow, such as smooth muscle and restricted flexibility in an experimental schedule, as this model is optimally tested within 4 h of intact vascular segment isolation from the animal.
Building from a previous video protocol authored by Socha and Segal12 and recent experimental developments in the interim6,15,16, we hereby demonstrate the isolation of fresh endothelium from posterior cerebral arteries and simultaneous measurements of endothelial [Ca2+]i and Vm using Fura-2 photometry and intracellular sharp electrodes, respectively. Additionally, this experiment entails continuous superfusion of salt solutions and pharmacological agents during physiological conditions (pH 7.4, 37 °C). We chose the posterior cerebral artery, as it yields isolated endothelium with the structural integrity (cells coupled through gap junctions) and sufficient dimensions (width ≥50 µm, length ≥300 µm) amenable for intra- and intercellular signaling along and among endothelial cells. In addition, studies of the rodent posterior cerebral artery are substantially represented in the literature and encompass examination of fundamental endothelial signaling mechanisms, vascular development/aging, and pathology20,21,22. This experimental application is expected to yield a high-throughput analysis of cerebral endothelial function (and dysfunction) and will thereby allow for significant advancements in the understanding of blood flow regulation throughout aging and the development of neurodegenerative disease.
Before conducting the following experiments, ensure that all animal care use and protocols are approved by the Institutional Animal Care and Use Committee (IACUC) and performed in accord with the National Research Council's "Guide for the Care and Use of Laboratory Animals" (8th Edition, 2011) and the ARRIVE guidelines. The IACUC of Loma Linda University has approved all protocols used for this manuscript for male and female C57BL/6 mice (age range: 3 to 30 mo).
1. Equipment and Materials
NOTE: Details of materials required for the protocol can be found in the Table of Materials, Reagents, and manuals or websites associated with the respective vendors.
2. Preparation of Solutions and Drugs
3. Dissection and Isolation of Cerebral Artery
NOTE: All dissection procedures require specimen magnification (up to 50x) via stereomicroscopes and illumination provided by fiber optic light sources. To perform dissection procedures for isolation of the brain and arteries, use sharpened dissection instruments. Microdissection tools to isolate and clean arteries include sharpened fine-tipped forceps and Vannas style dissection scissors (3 to 9.5 mm blades).
4. Preparation of Endothelial Tube and Superfusion
NOTE: The endothelial tube is prepared as described previously12, with modifications for the cerebral artery6.
5. Dye Load, Wash-Out, and Temperature Settings
6. Simultaneous Measurement of [Ca2+]i and Vm
7. Visualization of Cell-to-Cell Coupling
The schematic demonstration of the protocol described above is shown in the attached figures. A brain isolated from a young adult male C57BL/6N mouse (5 months) is shown in Figure 1A. Posterior cerebral arteries are carefully isolated from the Circle of Willis, removed without connective tissue, and cut into segments (Figure 1B-D). From partially digested arterial segments, the intact endothelial tube is produced and secured on the glass cover slip using pinning pipettes. Gentle mechanical stretch is applied using two pinning pipettes to approximate linear in vivo length without vessel tortuosity. This has been further clarified in the manuscript (Figure 2A-D). Endothelial tubes isolated from posterior cerebral arteries are ~90–100 µm in width and ≥300 µm in length. The tube is loaded with Fura-2 AM followed by wash-out with PSS. Simultaneous measurement of [Ca2+]i and Vm in the endothelial tube is performed by Ca2+ photometry as well as sharp electrode electrophysiology (experimental rig shown in Figure 3 and Figure 4).
Functional assessment of the endothelial tube is performed by applying a classical endothelial GPCR agonist such as ATP into the flow chamber under physiological conditions. As shown in Figure 5, [Ca2+]i and Vm are recorded simultaneously in response to ATP (100 µM). A significant hyperpolarization of Vm (≥5 mV) occurs concomitantly with an intracellular [Ca2+]i increase, indicating that a rise in [Ca2+]i activates SKCa/IKCa channels and thereby allows K+ efflux across the plasma membrane to produce EDH. Evidence of intercellular coupling using current injection (± 0.5 to 3 nA, ~20 s pulses) can be found in our recently published work (reference6, Figure 1). Note that propidium iodide has a mass of ~668 Da in reference to second messengers known for passing through gap junctions such as inositol trisphosphate (IP3, ~420 Da), cyclic adenosine monophosphate (cAMP, ~329 Da).
Figure 1: Isolation of cerebral artery from brain. (A) Brain isolated with intact arteries (white arrow indicates the posterior cerebral artery) in dessection solution. (B) Magnified view of posterior cerebral artery (white arrow; ~3x vs. Panel A). (C) Isolated posterior cerebral arteries secured with stainless steel pins in specimen dish. (D) Posterior cerebral arteries following removal of connective tissues. Inset shows cut segments of arteries; Scale bar = 500 µm. Please click here to view a larger version of this figure.
Figure 2: Preparation of cerebral endothelial tube. (A) Equipment used for triturating the partially digested arterial segments; a = micromanipulator, b = chamber for preparing tube, c = aluminum stage, d = microsyringe, e = microscope, f = microsyringe pump controller. (B) Arterial segments (white arrows) in solution for enzymatic digestion. (C) Endothelial tubes prepared by trituration; a = trituration pipette, b = intact endothelial tube. Inset shows the trituration process to remove adventitia and smooth muscle cells; Scale bar = 100 µm. (D) An intact endothelial tube secured on the glass cover slip with pinning pipette; a = pinning pipette, b = intact endothelial tube with diameter of ~100 µm. Please click here to view a larger version of this figure.
Figure 3: Equipment for superfusion and simultaneous [Ca2+]i and Vm measurements. (A) Equipment for superfusion of the endothelial tube and temperature control, a = high intensity ARC lamp power supply, b = fluorescence system interface, c = temperature controller, d = six-reservoirs of superfusion system, e = fiber optic light sources, f = valve controller, g = hyperswitch of [Ca2+]i photometry system. (B) Superfusion chamber apparatus; a and b = headstages, c = ground electrode, d = inline heater, e = superfusion tubing, f = vacuum suction, g = superfusion chamber, h = aluminum stage holding the chamber. (C) Experimental platform on a vibration isolation table; a = cell framing adaptor with camera, b = microscope light, c = microscope, d = aluminum stage, e = micromanipulator for headstage control, f = vibration isolation table. Please click here to view a larger version of this figure.
Figure 4: Equipment for operating electrophysiology and data recording. a = oscilloscope, b = function generator, c = audible baseline monitor, d = 50/60 Hz noise filter, e and f = electrometers, g = data acquisition system. Please click here to view a larger version of this figure.
Figure 5: Simultaneous [Ca2+]i and Vm recordings. (A) Differential Interference Contrast image (400x) of a mouse cerebral arterial endothelial tube adjusted for experiment in photometric window using a 40x objective. A sharp electrode is placed into a cell as shown at the top of the image (white arrow). (B) Raw traces for simultaneous measurement of F340/F380 ratio (top) and Vm (bottom) in response to ATP (100 µM). Please click here to view a larger version of this figure.
In light of recent developments6,15,16,17, we now demonstrate the method to isolate mouse cerebral arterial endothelium in preparation for simultaneous measurement of [Ca2+]i and Vm underlying EDH consistently for ~2 h at 37 °C. Although technically difficult, we can measure cell-to-cell coupling as well (see reference6, Figure 1). In this manner, we can measure discrete and conducted signaling events simultaneously. For an account of the general challenges faced while isolating mouse arterial endothelium consult references11,12. We emphasize critical steps for isolating mouse cerebral arteries, our particular measurements of endothelial [Ca2+]i and Vm, suggestions for troubleshooting, and considerations for alternative methods in light of experimental strengths/weaknesses below.
The posterior cerebral artery requires careful isolation from the Circle of Willis and connective tissue without damage to smooth muscle and endothelial cells. Surgical tools (forceps and scissors) should be suitable for mouse microcirculation dissection procedures while maintained with clean, sharp edges. Relative to skeletal muscle12, we reduced the time period for enzymatic incubation by two-thirds and the concentrations of papain, collagenase, and dithioerythritol by half while adding use of an empirically optimized concentration of elastase. For consistent success in isolating cerebral arterial endothelial tubes, it is important to order enzymes in batches (2 to 4 vials) from a highly reputable vendor with consistent manufacturing practices in accord with frequency of use and shelf-life. Even with the potential for variation in enzyme activity from one lot of enzyme to the next, it is suggested that enzyme concentrations be maintained while optimizing the time of digestion within 10 to 12 min. It is worth noting that neither the gender nor age of the animal (3 to 30 months) has been a significant barrier to the preparation of isolated endothelium from the cerebral arteries in our experience. Thus, it is recommended that the composition of the enzyme cocktail and time of enzyme digestion remain the same in studies on age and gender.
As emphasized, caution is required during gentle trituration and isolation of the endothelium from adventitia and spindle-shaped smooth muscle cells in order to avoid damage to endothelial cells12. For the trituration step, the viscous mineral oil serves as a hydraulic medium between the piston of the micro-syringe and the aqueous PSS in order to withdraw or eject vessel segments bathed in PSS. It should be noted that if the vessel segment contacts the mineral oil, this step must be repeated using a clean trituration pipette with sequential backfilling of mineral oil and PSS. Before an experiment, it is recommended that an approximately linear in vivo length be restored to the endothelial tube to the extent possible, whereby individual cells assume a "flat", longitudinal morphology (e.g., diameter ~5 µm; length ~100 µm; thickness ~0.5 µm)8. However, axial tension should not be applied to the extent where cells at the edges of the specimen are pulling away from underneath respective pinning pipettes and thereby compromising mechanical stability for reliable cellular measurements during the experiment.
The photometric [Ca2+]i measuring system used in this technique is suitable for continuous measurements of endothelial [Ca2+]i using the ratiometric Fura-2 dye. Individual protocols can typically last ≥10 min without significant photobleaching. Further, we commonly use this approach for measuring [Ca2+]i because it is amenable to timely pauses in data acquisition and frequent movement of microscope objectives as needed to secure sharp electrodes for Vm measurements. Note that this is a photometric approach that only captures global, averaged [Ca2+]i among multiple cells. In general, for [Ca2+]i measurements, we recommend photometry for the initial characterization of endothelial cell responses to pharmacological agents (e.g., time course of peak and plateau phases) while securing plans to proceed to quantitate microdomain signaling events using standard confocal or multiphoton microscopy12,23.
The sharp electrode electrophysiology approach is amenable for integrated Vm recordings in physiological multicellular preparations. The measurement of intracellular Vm is by far the most technically challenging with numerous critical steps that can either facilitate or obliterate success of measurements. First, the experimenter needs to determine the optimal program conditions on a glass puller for consistently manufacturing sharp electrodes with a tip resistance of ~150 ± 30 MΩ. Using a filament heat ramp test reading as a reference, it is recommended that the experimenter determine parameters empirically, particularly with respect to variation in the "heat" setting while maximizing the time of pull. Next, mechanical stability of the micromanipulators, headstages, pipette holders, and specimen is absolutely critical. Aside from the obvious necessity for a vibration isolation table, the experimenter will need to ensure that all fittings are secure and that the stage area around and underneath the manipulators is clean. Ideally, extraneous noise should be reduced to the extent that the resolution of sharp electrode recordings is within 1 mV. In our experience, the most common source of noise is silver wire secured in the pipette holder that is in need of a sufficient chloride coating or replacement altogether. If common 50/60 Hz noise is present, "hum silencing" technology can be used and comes standard with a modern electrometer. If noise appears completely unmanageable, check to see whether there is a leak in the bath or overflow where PSS will come into contact with resistors for temperature control. If complications remain with acquiring a stable Vm signal, check the integrity of electrical cords and t-connectors.
Altogether, if establishing a sharp electrode recording is unsuccessful, it is commonly due to an inappropriate electrode, mechanical instability, or poor cellular health. Alternatively, the experimenter may also want to consider a configuration of the patch-clamp technique on individual, electrically uncoupled cells where information on whole cell currents and single ion channel recordings can be obtained. Although faced with the general challenges of using a fluorescent dye (e.g., heterogeneous intracellular loading, bleaching, calibration using ionophores, modest detection resolution), voltage-sensitive dyes such as Di-8-ANEPPS are also commercially available.
Our understanding of the endothelium in blood flow regulation to and throughout the brain is crucial with a rapidly aging human population and the rising incidence of chronic cardiovascular and neurodegenerative diseases. Thus, we provide a method for isolating intact endothelium from a vital cerebral artery, which may be adapted to other vascular structures in the brain. In addition, we show a useful approach for the simultaneous measurements of underlying [Ca2+]i and Vm. Finally, using dual intracellular electrodes, cell-to-cell coupling through gap junctions can also be assessed. With carefully planned pharmacological and/or genetic interventions, the current protocol reasonably and quantitatively encompasses study of cerebral endothelial function in its native form. Our expectation is that experimenters will build from and adapt this information to advance vascular biology and neuroscience in general. In particular, it would be satisfactory to see common experimental struggles involved with electrical measurements minimized altogether. Although this protocol is not a stand-alone set of techniques by any means, it can be used as a complementary approach to determine precisely how endothelial biophysical determinants translate into changes in vascular resistance and how they fine-tune regulation of blood flow relative to conditions representative of health vs. disease.
The authors have nothing to disclose.
We thank Charles Hewitt for excellent technical assistance while establishing equipment and supplies needed for the current protocols. We thank Drs. Sean M. Wilson and Christopher G. Wilson, from the LLU Center for Perinatal Biology, for providing us with an additional inverted microscope and electrometer, respectively. This research has been supported by National Institutes of Health grant R00-AG047198 (EJB) and Loma Linda University School of Medicine new faculty start-up funds. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Glucose | Sigma-Aldrich (St. Louis, MO, USA) | G7021 | |
NaCl | Sigma | S7653 | |
MgCl2 | Sigma | M2670 | |
CaCl2 | Sigma | 223506 | |
HEPES | Sigma | H4034 | |
KCl | Sigma | P9541 | |
NaOH | Sigma | S8045 | |
ATP | Sigma | A2383 | |
HCl | ThermoFisher Scientific (Pittsburgh, PA, USA) | A466250 | |
Collagenase (Type H Blend) | Sigma | C8051 | |
Dithioerythritol | Sigma | D8255 | |
Papain | Sigma | P4762 | |
Elastase | Sigma | E7885 | |
BSA | Sigma | A7906 | |
Propidium iodide | Sigma | P4170 | |
DMSO | Sigma | D8418 | |
Fura-2 AM dye | Invitrogen, Carlsbad, CA, USA | F14185 | |
Recirculating chiller (Isotemp 500LCU) | ThermoFisher Scientific | 13874647 | |
Plexiglas superfusion chamber | Warner Instruments, Camden, CT, USA | RC-27 | |
Glass coverslip bottom (2.4 × 5.0 cm) | ThermoFisher Scientific | 12-548-5M | |
Anodized aluminum platform (diameter: 7.8 cm) | Warner Instruments | PM6 or PH6 | |
Compact aluminum stage | Siskiyou, Grants Pass, OR, USA | 8090P | |
Micromanipulator | Siskiyou | MX10 | |
Stereomicroscopes | Zeiss, NY, USA | Stemi 2000 & 2000-C | |
Fiber optic light sources | Schott, Mainz, Germany & KL200, Zeiss | Fostec 8375 | |
Nikon inverted microscope | Nikon Instruments Inc, Melville, NY, USA | Ts2 | |
Phase contrast objectives | Nikon Instruments Inc | (Ph1 DL; 10X & 20X) | |
Fluorescent objectives | Nikon Instruments Inc | 20X (S-Fluor), and 40X (Plan Fluor) | |
Nikon inverted microscope | Nikon Instruments Inc | Eclipse TS100 | |
Microsyringe pump controller (Micro4 ) | World Precision Instruments (WPI), Sarasota, FL, USA | SYS-MICRO4 | |
Vibration isolation table | Technical Manufacturing, Peabody, MA, USA | Micro-g | |
Amplifiers | Molecular Devices, Sunnyvale, CA, USA | Axoclamp 2B & Axoclamp 900A | |
Headstages | Molecular Devices | HS-2A & HS-9A | |
Function generator | EZ Digital, Seoul, South Korea | FG-8002 | |
Data Acquision System | Molecular Devices, Sunnyvale, CA, USA | Digidata 1550A | |
Audible Baseline Monitors | Ampol US LLC, Sarasota, FL, USA | BM-A-TM | |
Digital Storage Oscilloscope | Tektronix, Beaverton, Oregon, USA | TDS 2024B | |
Fluorescence System Interface, ARC Lamp + Power Supply, Hyperswitch, PMT | Molecular Devices, Sunnyvale, CA, USA | IonOptix Systems | |
Temperature Controller | Warner Instruments | TC-344B or C | |
Inline Heater | Warner Instruments | SH- 27B | |
Valve Controller | Warner Instruments | VC-6 | |
Inline Flow Control Valve | Warner Instruments | FR-50 | |
Electronic Puller | Sutter Instruments, Novato, CA, USA | P-97 or P-1000 | |
Microforge | Narishige, East Meadow, NY, USA | MF-900 | |
Borosilicate Glass Tubes (Trituration) | World Precision Instruments (WPI), Sarasota, FL, USA | 1B100-4 | |
Borosilicate Glass Tubes (Pinning) | Warner Instruments | G150T-6 | |
Borosilicate Glass Tubes (Sharp Electrodes) | Warner Instruments | GC100F-10 | |
Syringe Filter (0.22 µm) | ThermoFisher Scientific | 722-2520 | |
Glass Petri Dish + Charcoal Sylgard | Living Systems Instrumentation, St. Albans City, VT, USA | DD-90-S-BLK | |
Vannas Style Scissors (3 mm & 9.5 mm) | World Precision Instruments | 555640S, 14364 | |
Scissors 3 & 7 mm blades | Fine Science Tools (or FST), Foster City, CA, USA | Moria MC52 & 15000-00 | |
Sharpened fine-tipped forceps | FST | Dumont #5 & Dumont #55 |