The modified Landis technique enables paired measurement of the hydraulic conductivity of individual microvessels in the mesentery of normal and genetically modified rats under control and test conditions using microperfusion techniques. It provides a convenient method to evaluate mechanisms that regulate microvessel permeability and transvascular exchange under physiological conditions.
Experiments to measure the permeability properties of individually perfused microvessels provide a bridge between investigation of molecular and cellular mechanisms regulating vascular permeability in cultured endothelial cell monolayers and the functional exchange properties of whole microvascular beds. A method to cannulate and perfuse venular microvessels of rat mesentery and measure the hydraulic conductivity of the microvessel wall is described. The main equipment needed includes an intravital microscope with a large modified stage that supports micromanipulators to position three different microtools: (1) a beveled glass micropipette to cannulate and perfuse the microvessel; (2) a glass micro-occluder to transiently block perfusion and enable measurement of transvascular water flow movement at a measured hydrostatic pressure, and (3) a blunt glass rod to stabilize the mesenteric tissue at the site of cannulation. The modified Landis micro-occlusion technique uses red cells suspended in the artificial perfusate as markers of transvascular fluid movement, and also enables repeated measurements of these flows as experimental conditions are changed and hydrostatic and colloid osmotic pressure difference across the microvessels are carefully controlled. Measurements of hydraulic conductivity first using a control perfusate, then after re-cannulation of the same microvessel with the test perfusates enable paired comparisons of the microvessel response under these well-controlled conditions. Attempts to extend the method to microvessels in the mesentery of mice with genetic modifications expected to modify vascular permeability were severely limited because of the absence of long straight and unbranched microvessels in the mouse mesentery, but the recent availability of the rats with similar genetic modifications using the CRISPR/Cas9 technology is expected to open new areas of investigation where the methods described herein can be applied.
Microperfusion in the vasculature entails establishing controlled flow of an artificial perfusate of known composition via a micropipette in a blood vessel usually less than 40 µm in diameter. The perfused vessel remains within its normal tissue environment and is perfused with the animal’s blood up to the time of cannulation. When used in conjunction with a range of video imaging or fluorometric techniques, in situ microperfusion enables measurement of water and solute flows across the walls of microvessels under conditions where the driving forces for these flows are known and the permeability properties of the vascular wall can be directly evaluated. Further, by controlling the composition of the fluid surrounding the microvessel in the tissue (perfusate and superfusate), the regulation of microvessel permeability and exchange can be investigated by enabling the endothelial cells forming the microvessel wall to be exposed to a variety of experimental conditions (agonists, modified perfusion conditions, fluorescent indicators to measure intracellular composition and signaling) for precisely measured periods of time (sec to hr). In addition, ultrastructural or cytochemical evaluations of key cellular molecular structures regulating the barrier can be investigated in the same microvessels in which permeability is directly measured. The approach thereby forms a bridge between investigation of the cellular and molecular mechanisms to modify endothelial barrier function in cultured endothelial cell monolayers and investigation in intact microvessels. See the following reviews for further evaluation1-6.
A limitation of microperfusion is that it can be used only in microvascular beds that are thin, transparent and have sufficient structural integrity to enable cannulation with a glass micropipette. While early investigations used frog microvessels in mesentery and thin cutaneous pectoris muscle7,8, by far the most commonly used preparation in mammalian models is the rat mesentery9-15. Most investigations have focused on acute changes in vascular permeability studied over periods of 1-4 hr, but more recent investigations have been extended to measurements on individual vessels 24-72 hr after an initial perfusion12,16. The recently developed CRISPR technology, which promises to make more genetically modified rat models available for studying vascular permeability regulation17 should enable the methods described in this communication to be applied in venular microvessels of the mesentery in these important new rat models.
The method requires an inverted microscope equipped with a custom built microscope stage large enough to hold both the animal preparation and at least three micromanipulators used to position microtools close to the perfused vessel and to align a perfusion micropipette with the vessel lumen. For example a custom platform for an x-y microscope stage (about 90 × 60 cm) can be fabricated from a 1 cm thick steel plate with a rust-proof coating. The stage is attached to an engineering index table or two dove-tail slides mounted at right angles and supported on Teflon pillars or ball transfers for movement in the horizontal plane. A typical rig (see Figure 2) has much in common with the microscope and micropositioning equipment used for a range of intravital microcirculation experiments such as those to measure single vessel blood flow and hematocrit, local oxygen delivery by blood perfused microvessels, regulation of vascular smooth muscle tone, and the local microvascular accumulation of fluorescent tracers injected into the whole circulation.18-26
The fundamental aspect of the technique is the measurement of volume flow (Jv) across a defined surface area (S) of the microvessel wall. To accomplish this via the modified Landis technique described herein a simple inverted microscope is adequate. A small video camera is mounted on the image port and the video signal, with an added time base, is displayed on a video monitor and recorded either in digital form on a computer or as a digital or analog signal on a video recorder. Once the microvessel is cannulated the portion of the microvessel visible to the camera can be changed by moving the stage and manipulators as a unit without disrupting the cannulation.
Measurement of transvascular flows may also be combined with more detailed investigations using a sophisticated fluorescence microscope with appropriate filters such as rigs used for measurements of solute permeability, fluorescent ratio monitoring of cytoplasmic calcium or other cellular mechanisms, and confocal imaging 6,12,13,27. A key advantage of all microperfusion approaches is the ability to make repeated measures, on the same vessel, under controlled change of driving force such as hydrostatic and oncotic pressures, or induced change in vessel responses to inflammatory conditions. The most common design is a paired comparison of measured hydraulic conductivity (Lp) on the same vessel with the vessel first perfused via a micropipette filled with a control perfusate and the red cell suspension to establish a baseline permeability state, then with a second pipette with the test agent added to the perfusate. Multiple cannulations are possible with the cycle repeated after reperfusion with the control pipette.
The present protocol demonstrates the cannulation and microperfusion of a venular vessel in rat mesentery to record water fluxes across the microvessel wall and measure the Lp of the vessel wall, a useful index of the permeability of the common pathway for water and solutes across the intact endothelial barrier. The procedure is called the modified Landis technique because the original Landis principle of using the relative movement of red cells as a measure of transvascular fluid exchange after perfusion is blocked is preserved28, but the range of experimental conditions (e.g., the hydrostatic and albumin oncotic pressure differences across the microvessel wall) available after microperfusion is far greater than in uncannulated blood perfused microvessels8,29.
Ethics Statement: All procedures were reviewed and approved by the Institutional Animal Care and Use Committee.
1. Preliminary Fabrication of Micropipettes, Restrainers, and Blockers
2. Animal Preparation and Surgery
3. Prepare Solutions and Erythrocytes for use as Flow Markers
4. Arrange the Tissue on the Animal Tray
5. Fill Micropipette and Mount in Holder
6. Microcannulation and Microvascular Pressure Measurement
7. Microocclusion and Collection of Data
8. Analysis of Data and Measurement of Lp (Water Permeability)
Figure 4 shows the results from measuring the time course of changes in Lp in a rat venular microvessel cannulated successively with four perfusates.33 The magnitude of Lp calculated at a constant pressure was used as a measure of changes in microvessel wall permeability, first in the control state with a perfusate containing 1% bovine serum albumin then when the vessel was exposed to the inflammatory agent bradykinin (Bk) using a second micropipette containing 10 nM Bk. Bradykinin caused a transient increase in Lp that returned towards control value over the subsequent 10-15 min. The vessel was then re-perfused with control perfusate for 30 min to re-establish a stable baseline. When the microvessel was perfused using a fourth micropipette with the same concentration of Bk as well as 1 µM sphingosine-1-phospate (S1P) in the perfusate, the response to Bk was significantly attenuated. The degree of attenuation when Bk and S1P were perfused at the same time was found to be similar to that measured in other experiments where S1P was added to the perfusate up to 20 min before exposure to Bk. These results demonstrate the power of being able to make multiple measurements of Lp on a single vessel, and show that the action of an agent such as S1P to stabilize the vessel wall in the presence of an inflammatory agent can be resolved to time periods on the order of a few seconds. In separate vessels control studies were carried out with a similar protocol in the absence of S1P to show that the attenuation of the Bk response was not due to anaphylaxis.
For some experimental designs it is important to estimate both the solute reflection coefficient for a macromolecule (σ) and the Lp. This is accomplished best by measuring Jv at multiple pressures on each individual microvessel. Figure 5 shows an experiment in which both Lp and the effective oncotic pressure of 5% albumin in the perfusate (πalbumin = 27 cmH2O at 37 °C) were estimated under conditions when there was no albumin in the tissue surrounding the rat microvessel. In these experiments Jv/S was measured at three different pressures. The Lp is the slope of the relation between Jv/S and pressure, and the intercept on the pressure axis is the effective oncotic pressure difference across the vessel wall (σΔπ).
Figure 1. Microtools for perfusion of individual microvessels. (A) A micropipette for cannulation of rat microvessels with a beveled tip. (B) A restrainer for holding mesentery tissue stable near the site of cannulation. (C) An occluding rod, or blocker, used to press down on the mesentery and cannulated microvessel to occlude the flow in the vessel. Please click here to view a larger version of this figure.
Figure 2. Intravital rig for microperfusion. (A) Overview of animal tray with micromanipulators and microtools oriented over glass pillar. (B) Large metal stage mounted on engineering table and supporting bearings. (C) Anaesthetized rat with microtools in position for microperfusion of a microvessel. (D) Close view of mesentery during an experiment. Please click here to view a larger version of this figure.
Figure 3. Measurement of filtration rate per unit area using modified Landis technique. This schematic shows a single microvessel cannulated with a micropipette and an occluding rod positioned over the vessel. Immediately after occluding the microvessel, the distance (ℓ0) between the marker RBC travelling on the centerline of the occluded microvessel and the occlusion site is recorded so that the cell trajectory can be followed for 5-10 sec after occlusion. The initial position (ℓ0) and initial velocity (dℓ/dt) of the cell are used to calculate the average filtration rate per unit area of microvessel wall between the marker cell and the site of occlusion. The microoccluder is then raised, and free perfusion established before additional measurements are made. Please click here to view a larger version of this figure.
Figure 4. Representative data show that sphingosine-1-phosphate (S1P) inhibition acts very fast. Bradykinin (Bk; 10 nM) caused a transient increase in microvessel Lp. S1P applied as concurrent with second bradykinin (Bk) test inhibited the acute Bk response relative to first Bk. These data demonstrate that the time of exposure to a test agent can be modified to evaluate the kinetics of activation of an inhibitory response. Specifically, S1P applied concurrent only with Bk strongly inhibited the Bk response. S1P was 1 µM and Bk was 10 nM. Please click here to view a larger version of this figure.
Figure 5. Measurement of Jv/S at multiple pressures enables both Lp and the effective oncotic pressure to be measured. The filtration flux, Jv/S, was measured in this vessel at various microvessel hydrostatic pressures during perfusion with BSA (50 mg ml−1; π = 27 cmH2O) while the mesentery was bathed with protein-free Ringer's solution. The slope of the relation between Jv/S and pressure is the Lp and the intercept on the pressure axis indicates the effective oncotic pressure difference across the vessel wall. In this experiment the Lp was 1.2 × 10-7 (cm sec-1 cmH2O-1) and the σΔπ was 24 cmH2O. Please click here to view a larger version of this figure.
Details of Lp calculations. Although transvascular fluid movement occurs while the vessel is freely perfused, such exchange is far too small to be measured during free perfusion because it is typically less than 0.01% of the vessel perfusion rate. However, when perfusion is transiently stopped by occluding the microvessel, transvascular flow (i.e., filtration) is measured from the movement of marker red cells in the lumen as the column of fluid between a marker red cell and the site of occlusion shortens as shown in Figure 3. Depending on the complexity and length of the experimental protocol an unbranched vessel 600 to 1,000 µm in length is needed.
When the vessel is not occluded and the manometer pressure is driving flow through the microvessel, the pressure drop across the pipette tip is large so that microvessel pressure during free perfusion is typically only a few cmH20 above the downstream pressure. In contrast, during an occlusion the pressure drop across the pipette is very small due to fluid slowly entering the vessel lumen to replace fluid filtering across the vessel wall. Thus during occlusion the hydrostatic pressure in the vessel lumen (Pc) equals that set by the manometer. The oncotic pressure in the mesenteric tissue is negligible, so immediately after occlusion, transvascular fluid movements from marker red cells are measured where the hydrostatic pressure is known and the oncotic pressure in the vessel lumen is equal to that set in the perfusate. With successive measurements of Lp, vessel length is lost each time the block site is advanced. In addition, successive recannulation reduces available vessel length.
The transvascular fluid flow per unit area of vessel wall (Jv/S) is estimated from the rate that a measured length of the column of water in the microvessel lumen between a marker red cell and the occlusion site either shortens (net filtration) or lengthens (net reabsorption). The assumption is that neutrally buoyant red cells moving along the centerline of the vessel lumen can be used to track the mean velocity of the water column surrounding the cell (the ratio of red cell velocity to mean water velocity is close to 1.8 for a 6 µm red cell in a 30 µm diameter vessel).8,14 Thus if the velocity of a marker cell (in µm/sec) towards the blocker is V µm/sec over the first 2-5 sec after occlusion, and the vessel radius is r, the volume of fluid filtered across the microvessel wall between that red cell and the site of occlusion (Jv) is πr2V µm3/sec assuming the microvessel is cylindrical. Further the area of a cylindrical microvessel (S) between that marker cell and the occlusion site is 2πrℓ, where ℓ is the initial distance between the marker cell and the occlusion site. Thus the average initial transvascular flow per unit area (Jv/S) at the pressure set by the water manometer Vr/(2ℓ).
The simplest estimate of vessel Lp is made if the oncotic pressure of the perfusate has been set low relative to the microvessel pressure (typically 3.6 cmH2O, the oncotic pressure of bovine serum albumin at a concentration of 10 mg/ml and a capillary pressure greater than 30 cmH2O). Lp is calculated as (Jv/S)/(Pc-3) cm/sec/cm H2O because the albumin reflection coefficient (σalbumin) in the control state is close to 0.9. With this approach rapid changes in Lp with a time resolution of the order of 1 min can be measured by establishing free flow between successive measures of Jv/S at constant pressure as shown in the representative results in Figure 4. The above approach ignores small changes in the oncotic pressure of albumin which fall to less than 1 cmH2O when the reflection coefficient falls below 0.5, but the error in the estimate of Lp remains less than 5% when Pc is greater than 30 cmH2O.
When accurate measurement of both Lp and reflection coefficient is required, Jv/S is measured at a series of microvessel pressures below the expected effective oncotic pressure of the test macromolecule. The slope of the relation between Jv/S and pressure measures Lp and the intercept on the pressure axis when net flow is zero measures σπ (Figure 5). The demonstration that Jv is linearly related to pressure is also an important validation of the use of single pressure measurements using the simpler protocol described above (Figure 4) when the oncotic pressure difference is low relative to hydrostatic pressure.
Possible sources of error. There are several problems that compromise accurate measurement. Failure to properly occlude the vessel or damage to the vessel wall at the occlusion site results in fluid leak past the occlusion site, and movement of marker cells towards the occlusion site faster than that due to transvascular exchange. Whereas marker cells normally move slowly toward the occlusion site, but do not reach it, a leak is indicated if marker cells track all the way to the occluder. On the other hand, failure to seal the pipette into the microvessel lumen at the cannulation site results in underestimation of Jv/S. Fluid leak from the pipette to the superfusate or to the vessel lumen upstream of the cannulation site causes a significant pressure drop across the pipette tip. Such a leak is detected when cells in the micropipette tip flow at a higher velocity than those in the downstream vessel lumen of the occluded vessel. Careful repositioning of the micropipette usually seals such a leak. A different issue arises if the Lp of the vessel wall is not uniform, usually having with one or more sites of localized damage or inflammation. Marker cells track to the site if most transvascular flow is concentrated there. Red cell movement still measures transvascular flow, but the measure Jv/S is not representative of the most of the wall area within the test segment.
Significance of the method with respect to existing methods. The two most commonly used methods to investigate the regulation of endothelial barrier permeability and transvascular exchange are (1) in vitro measurement of exchange across cultured endothelial cell monolayers and (2) measurement of tracer accumulation in a tissue sample after the tracer is injected intravenously into the whole animal. Experiments in cultured endothelial cell monolayers are favored because the cellular functions can be more directly modified and there is sufficient cellular material available for detailed evaluation of molecular events in endothelial cells. However under most culture conditions the endothelial cells do not usually form the barrier representative of vascular permeability in vivo. The modified Landis technique described above has been demonstrated to be a robust and reliable technique to investigate the regulation of endothelial barrier function in intact venular microvessels. In the hands of an experienced investigator paired measurements of permeability under control and test conditions where some of the approaches used in cell culture (e.g., imaging of cellular components, modification of signaling pathways) can be applied. The importance of experiments in intact microvessels has been demonstrated by showing significant differences between the mechanisms that regulate barrier permeability in individually perfused microvessels and those often described in cultured endothelial monolayers. Examples include responses to shear, thrombin exposure, and the contribution of contractile mechanisms to inflammatory gap formation2-6,9,16. Measurements of tracer accumulation in an intact vascular bed preserves more normal physiological function but direct investigation of real changes in vascular permeability are compromised because tracer accumulation may increase as the result of increased perfusion (increased surface area for exchange) with or without an increase in the permeability of the wall. Thus the extent of a permeability increase may be overestimated when vasodilation increases the number of perfused vessels and thereby total solute accumulation is greater than that due to increase in vascular permeability alone34.
Future applications. The range of options to modify cellular structure and function under the same conditions as permeability of individual vessels is directly measured during microperfusion is expected to increase. Furthermore it is expected that the application of the methods in genetically modified rat mesentery may solve a long-term problem that has existed for genetically modified mice. Specifically, when genetically modified mice became available it was expected that microperfusion could be extended to investigate microvessels in their mesenteric vessels. While some microperfusion experiments have been completed in microvessels of mouse mesentery35, mice generally have fewer microvessels in their mesenteric tissue than rats, and these vessels are often short and highly branched, as opposed to more suitable longer (> 500 µm) straight vessels that can be found in rats. Thus experiments to investigate the modulation of transvascular exchange in genetically mice have relied on assays such as the Miles34 or the technically demanding, but more reliable tracer experiments which measure changes in both vascular and extravascular accumulation of fluorescent of radiolabelled test probes36.
Modifications of the method. In the above protocols, changes in perfusate composition required change of micropipettes. An alternate approach is to refill the micropipette in situ using a refilling apparatus. This procedure has been accomplished as described in detail37. The limitation is that the time to change the composition of the perfusate in the pipette is not as fast as achieved by pipette replacement, but the procedure is especially useful if it is necessary to perfuse a microvessel for extended periods of time.
The same microperfusion approached can be extended to measure permeability coefficients to fluorescently labeled solute31,38-40. In this case the single barreled micropipette is replaced by a double barreled (theta) pipette41 and the vessel is perfused alternately with the test fluorescent solute to measure tissue accumulation relative to lumen content, and the same perfusate without the fluorescent solute to clear the tissue and allow repeated measurement of solute accumulation. When measurements of transvascular water flows are to be combined with measurements of solute permeability coefficient using fluorescently labeled test solutes, fluorescent indicators of intracellular composition, or confocal imaging of cellular components, a more sophisticated computer controlled fluorescence microscope is required. Since these investigations usually require lenses with higher magnification and much shorter working distance, the quartz pillar used to hold the mesentery is replaced by a glass cover slip glued to a Plexiglas support within the tissue well on the microscope tray. The increased availability of 3-D printers, will allow high precision manufacture of custom trays, which meet the narrow tolerance required to position larger diameter lenses with shorter working distance relative to selected microvessels.
The availability of usable microvessels in the mesentery varies with age, gender, and strain. Therefore, the choices of strain of rat and the use of male or female will depend on the specific questions being addressed. For many of our recent studies we have used male Sprague-Dawley rats which are acclimated in our animal facility at least one week before use at about age 3-4 months at which time we find they generally have available long, straight mesenteric microvessels. By contrast, our colleagues have successfully used female Sprague-Dawley rats of 2-3 months age13.
The role of the red cells. The primary role of red cells as described above was to track transvascular water movement assuming the cell acted as inert, neutrally buoyant indicators of water flows after the microvessels were occluded. In addition to this fundamental role, it is now recognized that the red cells contribute to the stability of the barrier by supplying S1P to the perfusate when albumin in present31,42,43. There are several important implications of these observations. In the absence of red cells or with the use of alternate inert flow markers the baseline permeability is likely to be less stable. It is recommended that measurements of S1P should be made in all perfusates used in microperfusion experiments.
The authors have nothing to disclose.
This work was supported by National Institutes of Health grants HL44485 and HL28607.
MICROSCOPE, TABLE AND STAGE | |||
inverted microscope (metallurgical type) with trinocular head for video: example | Olympus | CK-40 | try to place eyepieces higher relative to stage–you have to look through eyepieces while reaching around to top of stage over intervening micromanipulators |
inverted microscope (metallurgical type) with trinocular head for video: example | Leica | DMIL | try to place eyepieces higher relative to stage–you have to look through eyepieces while reaching around to top of stage over intervening micromanipulators |
narrow diameter, long working distance objective: example | Nikon | Nikon E Plan 10×/0.25 LWD | |
stage platform–1/2 inch or 1 cm sheet steel | welding shop | this should be heavy to reduce vibration | |
Unislide x-y table: dove tail slides | Velmex | AXY4006W1 | |
VIDEO | |||
CCD video camera: example | Pulnix | TM-7CN (no longer available) | no color needed |
video capture system with audio–generic | |||
video playback system (completely still frame, single frame motion) | |||
small microphone | |||
MICROMANIPULATORS, HOLDERS | |||
micromanipulator, XYZ (3) | Prior/Stoelting (no longer available) | look for fine Z, and larger range of travel in coarse drives for ease of positioning | |
hydraulic probe drive, one way | FHC | 50-12-1C | need to buy either manual drive or electronic drive |
manual drum drive | FHC | 50-12-9-02 | |
or hydraulic drive, 3 way | Siskiyou Corporation | MX610 (1-way) or MX630 (3-way) | great for short arms, water filled and must be sent back for refill ~every 2 years |
connectors/rods/holders | Siskiyou Corporation | MXC-2.5, MXB etc. | |
pin vise | Starrett | 162C | to hold restrainer |
pipette holder | World Prescision Instruments | MPH3 | |
water manometer ~120 cm | |||
MICROSCOPE TRAY | |||
clear Plexiglas for microscope tray for animal | |||
3/4 inch polished quartz disc ~1/4 inch tall | Quartz Scientific Inc. | custom | (or polished plexiglass, glass); make sure the height is less than working distance of objective |
Plexiglas glue (Weld-on 4: CAUTION CARCINOGEN) | |||
medical adhesive for tissue well | NuSil | MED-1037 | |
All-purpose silicone rubber heat mat, 5" L x 2" W | Cole Parmer | EW-03125-20 | heater for microscope tray–needs cord and controller–240V version available |
Power Cord Adapter for Kapton Heaters and Kits, 6 ft, 120 VAC | Cole Parmer | EW-03122-75 | |
STACO 3PN1010B Variable-Voltage Controller, 10 A; 120 V In, 0-140 V Out | Cole Parmer | EW-01575-00 | |
PIPET MANUFACTURE | |||
vertical pipette puller | Sutter Instrument Company | P-30 with nichrome filament | |
1.5 mm OD thin wall capillary tubing | Sutter Instrument Company | B150-110-10 | |
pipette grinder air stone and dissection microscope–see reference in text | or purchase a package from Sutter Instruments or World Precision Instruments | ||
RX Honing Machine, System II | RX Honing Machine Corporation | MAC-10700 Rx System II Machine | alternative for air stone, use with a dissecting microscope mounted at an angle |
with ceramic sharpening disc | RX Honing Machine Corporation | use "as is" or attach lapping film | |
lapping film sheets, 0.3 or 0.5 um | 3M | part no. 051144 80827 | 268X Imperial lapping film sheets with adhesive back–can be purchased from Amazon |