The protocol describes the use of wire myography to evaluate the transmural isometric tension of mesenteric arteries isolated from mice, with special consideration of the modulation by factors released from endothelial cells and perivascular adipose tissues.
Altered vascular tone responsiveness to pathophysiological stimuli contributes to the development of a wide range of cardiovascular and metabolic diseases. Endothelial dysfunction represents a major culprit for the reduced vasodilatation and enhanced vasoconstriction of arteries. Adipose (fat) tissues surrounding the arteries play important roles in the regulation of endothelium-dependent relaxation and/or contraction of the vascular smooth muscle cells. The cross-talks between the endothelium and perivascular adipose tissues can be assessed ex vivo using mounted blood vessels by a wire myography system. However, optimal settings should be established for arteries derived from animals of different species, ages, genetic backgrounds and/or pathophysiological conditions.
Dilatations and constrictions of arteries are achieved by relaxations and contractions, respectively, of their vascular smooth muscle cells. Changes in vascular responsiveness of small arteries contribute to the homeostatic regulation of arterial blood pressure by autonomic nerves and hormones present in the blood (e.g., catecholamines, angiotensin II, serotonin, vasopressin). At the local level, the vascular responses of smooth muscle cells are modulated by signals from both the endothelial cells of the intima and the adipose tissue surrounding the arteries (Figure 1).
The endothelium is not only a passive barrier, but also serves as a surface to exchange signals between the blood and the underlying vascular smooth muscle cells. By releasing various vasoactive substances, the endothelium plays a critical role in the local control of vascular tone responses1. For example, in response to acetylcholine, endothelial nitric oxide synthase (eNOS) is activated in the endothelium to produce nitric oxide (NO), which induces relaxation of the underlying vascular smooth muscle by activating soluble guanylyl cyclase (sGC)2. Other vasoactive substances include the products of cyclooxygenases (e.g., prostacyclin and thromboxane A2), lipoxygenase (e.g., 12-hydroxyeicosatetraenoic acids, 12-HETE), and cytochrome P450 monooxygenases (HETEs and epoxyeicosatrienoic acids, EETs), reactive oxygen species (ROS), and vasoactive peptides (e.g., endothelin-1 and angiotensin II), and endothelium-derived hyperpolarizing factors (EDHF)3. A delicate balance between endothelium-derived vasodilators and vasoconstrictors maintain the local vasomotor tone4,5.
Endothelial dysfunction is characterized by the impairment in endothelium-dependent vasodilatation6, a hallmark of vascular aging7. With age, the ability of endothelium to promote vasodilatation is progressively reduced, due largely to a decreased NO bioavailability, as well as the abnormal expression and function of eNOS in the endothelium and sGC in the vascular smooth muscle cells8,9,10. Reduced NO bioavailability potentiates the production of endothelium-dependent vasoconstrictors11,12. In aged arteries, endothelial dysfunction causes hyperplasia in the media, as reflected by the marked increases in wall thickness, number of medial nuclei, which are reminiscent of the arterial thickening in hypertension and atherosclerosis observed in human patients13,14. In addition, pathophysiological conditions such as obesity, diabetes or hypertension accelerate the development of endothelial dysfunction15,16.
Perivascular adipose tissue (PVAT) releases numerous adipokines to regulate vascular structure and function17. The anti-contractile effect of PVAT is mediated by relaxing factors, such as adiponectin, NO, hydrogen peroxide and hydrogen sulphide18,19,20. However, depending on the location and pathophysiological condition, PVAT also can enhance contractile responses in various arteries21. The pro-contractile substances produced by PVAT include angiotensin-II, leptin, resistin, and ROS22,23. In most of the studies on isolated blood vessels, PVAT has been considered as a simple structural support for the vasculature and thus removed during the preparation of blood vessel ring segments. Since adipose dysfunction represents an independent risk factor for hypertension and associated cardiovascular complications24, the PVAT surrounding the blood vessels should be considered when investigating the vascular responsiveness of different arteries.
The multi wire myograph systems have been widely used to investigate the vasomotor functions of a variety of blood vessels, including the aorta, mesenteric, renal, femoral, cerebral and coronary arteries25,26. The protocols described herein will use wire myography to evaluate vascular responsiveness in mesenteric arteries isolated from genetically modified mouse models, with a special focus on the modulation by PVAT.
All animals used for the following study were provided by the Laboratory Animal Unit of the Faculty of Medicine, The University of Hong Kong. Ethical approval was obtained from the departmental Committee on Use of Laboratory Animals for Teaching and Research (CULATR, no.: 4085-16).
1. Preparations
2. Normalization to determine the optimal initial tension
NOTE: The normalization procedure allows the determination of the optimal internal diameter (IC) of arteries at which the blood vessel experiences a suitable resting transmural pressure (100 mmHg or 13.3 kPa for mesenteric arteries) and produces maximal active forces in response to vasoactive agents.
3. Phenylephrine-induced contractions
NOTE: Drugs that can be selected for inducing the vasoconstrictive responses include the unspecific adrenoceptor agonist norepinephrine, the selective α-1 adrenoceptor agonist phenylephrine, the peptide hormone angiotensin II, and the monoamine neurotransmitter 5-hydroxytryptamine. Phenylephrine is used in the present protocol for examination (Table of Materials).
4. Endothelium-dependent relaxations/contractions
Examination of the length/tension relationships to obtain the normalization factor k
The amount of stretch applied to a vessel segment influences the extent of the actin-myosin interaction and hence the maximal active force developed. Thus, for every type of blood vessel, determining the amount of stretch needed for maximal active force is required for proper myography studies. Here, normalization of the length/tension relationship is performed for mesenteric arteries isolated from mouse models (Figure 2). The arterial segments were suspended in a four-chamber wire myograph system (see Table of Materials) on stainless steel pins (40 μm diameter). Isometric tension was recorded using an analog-to-digital converter connected to a computer with a recording program. Chambers contained 5 mL of Krebs buffer, kept at 37 °C and aerated with 95% O2 and 5% CO2 to maintain pH at 7.4 throughout the experiment. A passive length/tension relationship was established by incremental stretching of the artery segments until the internal circumference corresponding to 100 mmHg transmural pressure (IC100) was obtained. After each stretch (blue arrows), 115 mM KCl was applied to stimulate contractions (green arrows). The active length/tension curves (red) were plotted by extracting the active force data (subtracting the passive force at each stretch from the KCl-activated force) on the Y-axis and then a graph was created manually with the IC values calculated from the micrometer data on the X-axis. An IC value lying within the peak plateau is IC1 (dashed red lines). The normalization factor k was calculated as IC1/IC100 ratios, which could then be applied to samples of the same vessel type in the subsequent experiment.
Presentation and calculation of concentration-response curves
Most concentration-response curves are performed in a cumulative manner. A low concentration of the agonist is added to the bath (preferably starting with a concentration below the threshold for response). After allowing enough time for a possible response (3-5 min), the next concentration is added. When a response is observed, it is allowed to reach a plateau before the next maximal response is obtained. The half-log (average 3.16-fold) increments in concentration of phenylephrine are applied here to study agonist-induced contractions (Figure 3 and Figure 4).
In most cases, contraction-response curves are not expressed as the raw values of tension/force, but as a percentage of the reference response to KCl obtained at the optimal point of the length/tension curve of the individual blood vessel segment. This adjusts for variability in the size or smooth muscle content of the blood vessel as well as corrects for remodeling changes due to aging or pathology. Here, the maximal contractions induced by 115 mM KCl are obtained at the beginning of the experiment and used for calculating phenylephrine-stimulated contractions of mesenteric arteries with or without PVAT, in the absence or presence of L-NAME (Figure 3 and Figure 4).
To study agonists producing relaxation, the vessels are usually contracted to a uniform level—around 50-80% of the maximal contractile response of that tissue. Since the responses to phenylephrine-induced contractions are different between the experimental groups, the present protocol uses U46619 to stimulate stable contractions before applying the cumulative concentrations of acetylcholine. The smooth muscle relaxation is expressed as a percentage of the initial contraction induced by U46619 (Figure 5).
The concentration-response curves can be compared as the presence of a leftward or rightward shift (e.g., between a control curve and one obtained in the presence of an antagonist) by determining the concentrations producing equal responses, e.g., 30% or 50% of the maximum. These are termed EC30 and EC50, respectively (Figure 3B). Statistical comparison of the mean EC50 values should be performed on the logarithm of their values. Depression of curves is examined by comparing their respective maximal responses (Emax) (Figure 3B). In the examples shown, the phenylephrine-induced contractions in mesenteric arteries were enhanced by L-NAME and the concentration-response curves showed a leftward shift as well as an elevation in the maximal contractions (Figure 4B). The acetylcholine-induced relaxations in mesenteric arteries were inhibited by L-NAME, and the concentration-response curve showed a rightward shift as well as reduction in the maximal relaxations (Figure 5B).
Vascular responses to various pharmacological agents can be computed as the area-under-the-contraction-curve (AUCC) for contractions and area-above-the-relaxation-curve (AARC) for relaxations, respectively, by using the nonlinear logistic regression analysis for comparison10 (Figure 3C, Figure 4C and Figure 5C). The effect of L-NAME can be compared by the values of the AUCC/AARC to determine the NO bioavailability (Figure 6). The basal and stimulated release of NO in mesenteric arteries with or without PVAT can be expressed as the differences in the concentration-response curves of phenylephrine-stimulated contraction (DAUCC) and acetylcholine-induced relaxation (DAARC), respectively, in the presence or absence of L-NAME (Figure 6A and Figure 6B). In the example shown, the presence of PVAT reduced the NO bioavailability in mesenteric arteries collected from mice fed with high fat diet (Figure 6C).
Figure 1: A schematic diagram of the wall structure of arteries. Endothelial cells in the tunica intima mediate endothelium-dependent relaxation/contraction of the vascular smooth muscle, whereas signals released from the perivascular adipose tissue modulates the cross-talks between different layers of the arterial wall. Please click here to view a larger version of this figure.
Figure 2: Representative traces illustrating an experiment to determine the optimal initial tensions for mouse mesenteric arteries. The blood vessel segments were prepared from mesenteric arteries collected from the 16-week-old mice fed with standard chow (A) or high fat diet (B). After mounting, the passive and active length/tension curves were obtained by step-wise stretching and sequential stimulation with 115 mM KCl (left panels). The active contraction generated with each stimulation should increase as the vessel is progressively stretched, until it reaches a plateau at the optimal length. Further stretch will lead to a decrease in the active contraction. The IC100 and IC1 were determined by plotting the passive and active length/tensions curves, respectively (right panels). Note that the passive length/tension curve was generated by the Normalization Module after manually inputting the micrometer values, whereas the active length/tensions curves plotted manually after calculating the tension and IC values at each step of KCl stimulation. The IC1/IC100 ratios were calculated as normalization k factor (right panels). ). Note also that only the last four points of the active length/tension curve were shown in the figures in panel (B). STC: Standard chow fed mouse artery; HFD: High-fat diet fed mouse artery. Please click here to view a larger version of this figure.
Figure 3: Output recordings of the vasoconstrictor responses to phenylephrine in mesenteric arteries with or without surrounding PVAT. Contractility studies are performed on preparations of mesenteric arteries from 16-week-old Adipo-SIRT1 transgenic mice, in which the human SIRT1 is overexpressed selectively in adipose tissues29. Cumulative concentrations of phenylephrine were applied to stimulate the contractions of mesenteric arteries without (-PVAT) or with (+PVAT), (A). The contractile responses were recorded and calculated as percentage of 115 mM KCl-induced maximal contraction (B). The area-under-the contraction curves (AUCC) were plotted for comparison (C). Note that PVAT from Adipo-SIRT1 mice elicited an anti-contractile effect on the response to phenylephrine. Please click here to view a larger version of this figure.
Figure 4: Output recordings of the vasoconstrictor responses to phenylephrine in mesenteric arteries with or without the surrounding PVAT, and in the absence or presence of L-NAME. Mesenteric arteries were collected from 16-week-old wild type mice fed with high fat diet. Thirty min after adding 10-4 M L-NAME or vehicle control, cumulative concentrations of phenylephrine were applied to stimulate the contractions of mesenteric arteries (A). The contractile responses were recorded and calculated as percentage of 115 mM KCl-induced maximal contraction (B). The area-under-the contraction curves (AUCC) were plotted for comparison (C). Note that PVAT from dietary obese mice did not elicit anti-contractile effects on the response to phenylephrine. –PVAT: arterial rings prepared without PVAT; +PVAT: arterial rings prepared with the surrounding PVAT; -L-NAME: arterial rings not incubated with L-NAME; +L-NAME: arterial rings incubated with L-NAME. Please click here to view a larger version of this figure.
Figure 5: Output recordings of the vasodilator responses to acetylcholine in mesenteric arteries with or without the surrounding PVAT, and in the absence or presence of L-NAME. Mesenteric arteries were collected from 16-week-old mice fed with high fat diet. At 30 min after adding 10-4 M L-NAME or vehicle control, the blood vessel segments with or without PVAT are pre-contracted with U46619 (1-3 x 10-8 M; Table of Materials), a thromboxane A2 receptor agonist, to induce stable and sustained smooth muscle contractions. Cumulative concentrations of acetylcholine were then applied to stimulate the relaxations of mesenteric arteries with or without the surrounding PVAT (A). The relaxation responses were recorded and calculated as percentage of U46619-induced contraction (B). The area-above-the relaxation curves (AARC) were plotted for comparison (C). Note that only the arterial segment with surrounding PVAT is shown in panel A. –PVAT: arterial rings prepared without PVAT; +PVAT: arterial rings prepared with the surrounding PVAT; -L-NAME: arterial rings not incubated with L-NAME; +L-NAME: arterial rings incubated with L-NAME. Please click here to view a larger version of this figure.
Figure 6: Illustration of the procedure to calculate NO bioavailability. The area-under-the-contraction-curves (AUCC) and the area-above-the-relaxation-curves (AARC) were calculated based on the responses to cumulative concentrations of phenylephrine (A) and acetylcholine (B), respectively. The differences between preparations pre-treated without and with L-NAME were defined as ∆AUCC (A) and ∆ AARC (B) to represent basal NO contribution and stimulated NO release, respectively. Accordingly, the ∆AUCC (calculated from Figure 4C), ∆AARC (calculated from Figure 5C), and the sum of both (total NO bioavailability) were presented for comparing the NO bioavailability in mesenteric arteries without and with the surrounding PVAT (C). –PVAT: arterial rings prepared without PVAT; +PVAT: arterial rings prepared with the surrounding PVAT; -L-NAME: arterial rings not incubated with L-NAME; +L-NAME: arterial rings incubated with L-NAME. Please click here to view a larger version of this figure.
Apart from the endothelial cells, signals derived from PVAT play an important role in the regulation of smooth muscle tone reactivity30. Healthy PVAT releases NO and anti-inflammatory adiponectin to exert an anti-contractile effect on arteries, which is lost under pathological conditions such as obesity and metabolic syndrome31,32. In disease states, PVAT contributes to the development of endothelial dysfunction and other cardiovascular abnormalities33,34. Abnormal eNOS expression and function have been reported in PVAT of arteries from obese animals35,36. Since both endothelial and PVAT dysfunctions contribute to the development of cardiovascular and metabolic abnormalities23,37, when performing ex vivo vascular experiment, their role should be considered by including in or removing them from the preparations.
The wire myography system provides a convenient platform to dissect the vasoactive signals released from PVAT using different pharmacological probes10,38. However, the compositions in PVAT of different types of arteries, or the same arteries from animals of different genetic background, are not same39. Therefore, the wire myography results involving PVAT should not be compared across different types of arteries or the same type of arteries from mice of different strains. Age and the underlying disease states also affect the cellular compositions in PVAT. Here, mice from the same genetic background but with different genetic modifications in their adipose tissue were used for comparing the vasomodulating activity of PVAT.
As a main source of resistance to blood flow, mesenteric arteries are chosen for the present study. Resting tension determines the amount of vasomotor responsiveness40. The optimal initial tension of the blood vessel is affected by the type of artery, age, diet, treatment and genetic background of the animals, thus should be determined individually before examining relaxation/contraction-response curves. For the present demonstration, the superior mesenteric arteries were collected from 16-week-old mice fed with standard chow or high fat diet starting from the age of four weeks. The present protocol emphasizes the establishment of optimal settings for maximal active force production of the arterial segments before assessing pharmacological responses. Both passive and active length/tension relationships are studied for mesenteric arteries collected from in-house mouse models. A normalization k factor of 1 has been established for preparations from the 16-week-old animals, which is different from the default value of 0.9 or those used by previous publications41. Caution is needed when comparing the normalization ratios in the literature due to possible differences in the technique, buffer composition and instrument models, etc. In particular, age, diet and other pathophysiological conditions affect the passive and active tension as well as the pharmacodynamics characteristics of arteries42.
The authors have nothing to disclose.
This work was financially support by the grants from Research Grant Council of Hong Kong [17124718 and 17121714], Hong Kong Health and Medical Research Fund [13142651 and 13142641], Collaborative Research Fund of Hong Kong [C7055-14G], and the National Basic Research Program of China [973 Program 2015CB553603].
Acetylcholine | Sigma-Aldrich | A6625 | Stock concentration: 10-1 M Working concentration: 10-10 to 10-5 M |
L-NAME (Nω-nitro-L-arginine methyl ester) | Sigma-Aldrich | N5751 | Stock concentration: 3 x 10-2 M Working concentration: 10-4 M |
Phenylephrine | Sigma-Aldrich | P6126 | Stock concentration: 10-2 M Working concentration: 10-10 to 10-5 M |
U46619 (9,11-dideoxy-9α,11αmethanoepoxy prostaglandin F2α) | Enzo | BML-PG023-0001 | Stock concentration: 10-5 M Working concentration: 1-3 x 10-8 M |
Multiwire myograph | Danish MyoTechnology (DMT) | 620M | |
PowerLab 4/26 | ADInstruments | ML848 | |
Labchart7 | ADInstruments | – | |
Adipo-SIRT1 wild type mice | Laboratory Animal Unit, The University of Hong Kong | CULATR NO.: 4085-16 | |
Silicon-coated Petri dishes | Danish MyoTechnology (DMT) | ||
Tungsten wires | Danish MyoTechnology (DMT) | 300331 | |
Surgical tools |