Here, we describe a method to functionally assess the endothelial barrier of cardiac microvessels after ischemia/reperfusion injury via the measurement of the mean fluorescence intensity of extravasated 70,000 Da FITC-dextran in comparison with Evans Blue.
Revascularization therapies for culprit arteries, regardless of percutaneous coronary intervention and coronary artery bypass grafting, are considered the best strategy for improving the clinical prognosis of patients with acute coronary syndrome (ACS). Nonetheless, myocardial reperfusion following effective revascularization can trigger significant cardiomyocyte death and coronary endothelial collapse, known as myocardial ischemia/reperfusion injury (MIRI). Usually, endothelial cells and their intercellular tight junctions cooperatively maintain the microvascular endothelial barrier and its relatively low permeability but fail in reperfusion areas.
Microvascular endothelial hyperpermeability induced by ischemia/reperfusion (IR) contributes to myocardial edema, increased infiltration of pro-inflammatory cells, and aggravated intramyocardial hemorrhage, which may worsen the prognosis of ACS. The tracer used in this study-70,000 Da FITC-dextran, a branched glucose molecule labeled by fluorescein isothiocyanate (FITC)-appears too large to infiltrate the cardiac microvascular endothelium in normal conditions. However, it is capable of infiltrating a broken barrier after MIRI. Thus, the higher the endothelial permeability is, the more FITC-dextran accumulates in the extravascular intercellular space. Thus, the intensity of fluorescence from FITC can indicate the permeability of the microvascular endothelial barrier. This protocol takes advantage of FITC-dextran to evaluate the cardiac microvascular endothelial barrier functionally, which is detected by an automated quantitative pathology imaging system.
The fact that an increasing number of patients with coronary heart diseases have benefited from revascularization therapies for decades does not negate the fatal consequences of MIRI. As described by Yellon and Hausenloy, approximately 30% of the final infarct size can be attributed to MIRI1. Cardiac microvascular endothelial dysfunction is the core mechanism underlying MIRI and the “no-reflow” phenomenon and is the target of many cardioprotective agents and classical cardioprotective pathways2. Moreover, endothelial hyperpermeability of microvessels can trigger myocardial edema3 and the over-infiltration of pro-inflammatory cells, including neutrophils4, macrophages, and erthrocytes5.
MIRI-induced oxidative stress and overexpressed pro-inflammatory cytokines, such as the members of the tumor necrosis factor family6 and the interleukin family7, activate signaling pathways of cell apoptosis to accelerate the death of ischemic myocardium. Erythrocytes penetrate the vulnerable endothelial barrier and accumulate in the extracellular matrix (ECM) to form intramyocardial hematoma8. Macrophage- and neutrophil-derived matrix metalloproteinases disassemble collagen in the ECM, which increases the risk of mechanical complication and post-infarction ventricular aneurysm9. Thus, the permeability of the endothelial barrier is a crucial sign for the prognostic evaluation of acute myocardial infarction and marks the severity of MIRI to some degree.
FITC-dextran is an array of polysaccharides comprised of different lengths of branched glucose molecules of molecular weight ranging from 4,000 Da to 70,000 Da. It is widely used as the standard tracer to demonstrate the permeability of cell membranes, functional physiological barriers such as the blood-brain barrier10, and mucosal membranes11 under specific treatments and pathological conditions. The permeability of one specific membrane varies for different substances according to their molecular weights. Usually, albumin is restricted to cardiac microvessels; however, it can penetrate the endothelium in the early stage of IR, causing myocardial edema. As the molecular weight of albumin is approximately 68,000 Da, 70,000 Da FITC-dextran would be a suitable agent to demonstrate endothelial hyperpermeability after IR. Moreover, dextran has been widely adopted to maintain the blood volume and colloid osmotic pressure of shock patients, and FITC-labeled tracers are used in numerous in vivo or in vitro studies with few adverse effects.
In addition to FITC-dextran, endothelial permeability has been measured with Evans Blue (EB) staining. After injection into the blood circulation, EB binds to serum albumin tightly and forms a complex12. Thus, this EB-albumin complex, not EB alone, can be used for the measurement of permeability, which means that the accuracy of such an EB-based method can be influenced by different concentrations of serum albumin in experimental subjects. Additionally, EB has been reported to poison the organs of experimental animals, especially the heart, liver, and lungs, via cellular degeneration and increased vascular permeability13. Endothelium undergoing ischemia/reperfusion would be more sensitive to such detrimental effects than normal endothelium, which can confound the significance of experimental results.
Compared to EB dye, FITC-dextran is exogenous, and the binding affinity to albumin is negligible. Hence, FITC-dextran remains in the free state and is not bound by macromolecules, such as albumin and tissue proteins. Thus, if administered according to body weight, FITC-dextran can work at equal blood concentration and is not affected by individual differences in serum albumin concentration and blood volume. Tongxinluo (TXL), a traditional Chinese medicine, was used to test the diagnostic sensitivity of this method. TXL reduces MIRI by maintaining the endothelial cellular ultrastructure and endothelial barrier integrity14. TXL is also used to cure angina pectoris and improve myocardial no-reflow after infarction clinically15. It contains Radix ginseng, Buthus martensi, Hirudo, Eupolyphaga seu steleophaga, Scolopendra subspinipes, Periostracum cicadae, Radix paeoniae rubra, Semen ziziphi spinosae, Lignum dalbergiae odoriferae, Lignum santali albi, and Borneolum syntheticum. TXL has been shown to protect the microvascular endothelium via the MEK/ERK pathway16 and endothelial nitric oxide synthase-mediated mechanisms5,17 and alleviate ischemia/reperfusion injury (IRI) of organs18, especially the heart, in a dose-dependent manner. Here, rats were fed with 0.5 g/kg TXL powder (an equivalent dose to that used clinically in humans) as described19. This paper provides a method to functionally assess the endothelial barrier of cardiac microvessels after IR via the measurement of the total fluorescence intensity of extravasated 70,000 Da FITC-dextran compared to EB.
Animal experiments were performed in accordance with the "Guide for the Care and Use of Laboratory Animals" issued by the US National Institutes of Health (Bethesda, MD, USA, NIH Publication No. 85-23, revised 1996) and the "Regulation to the Care and Use of Experimental Animals" of the Beijing Council on Animal Care (1996). The study protocol was approved by the Care of Experimental Animals Committee of Fuwai Hospital. Male SPF-class Sprague-Dawley rats (SD rats) weighing 150-180 g were used in this study.
1. Preparation of reagents, materials, and equipment
2. Preparing the rats for the surgical procedures
3. Establish a rat model of cardiac ischemia/reperfusion
4. Preparation of frozen tissue slices for fluorescence measurement
5. Fluorescence measurement and data analysis
Before performing the FITC-dextran assay to evaluate the changes in cardiac endothelial permeability after IR, we strongly recommend that experiments be conducted by the same researcher specializing in the establishment of myocardial IR model in rats to confirm that the coronary arteries are occluded in the same position successfully, and the ratios of ischemic area/total area are constant among all rats. As shown in Figure 1A,B, the heart (indicated by yellow arrow) was exposed, and a loop (indicated by green arrow) was formed in the myocardium located in the anterior wall of the left ventricle, approximately 2 mm under the edge of the left atrial appendage with a 0.5 mm (inner diameter) capillary silicone tube and 6-0 suture. As seen in Figure 1C, when the loop was fastened, the ischemic myocardium turned from pink to greyish pink (indicated by the white arrow) immediately.
Moreover, the myocardium reperfused and recovered its pink color immediately after the loop was loosened, as illustrated in Figure 1D. After loosening the loop, an incision was made parallel to the long axis of the femur in the inner thigh. The femoral vein was exposed and punctured with an insulin syringe, as shown in Figure 2A,B. Following a 3 h reperfusion, a "V" incision was made in the lower abdomen, and the aorta abdominalis and vena cava were dissected from the peritoneum (Figure 3A). Cold saline was injected into the vena cava gently, and the aorta abdominalis was cut to prevent the ventricular wall from being stretched excessively due to the intraventricular pressure that would affect its histopathological structure. Bleached lungs (red arrow in Figure 3B) and liver in Figure 3C indicate that the tissue was well-reperfused with saline.
Consequently, there was no significant difference in the cross-sectional area among the three groups: 1.00 ± 0.08 in the Sham group vs. 0.97 ± 0.08 in the IR group vs. 1.02 ± 0.04 in the IR+TXL group (Figure 4A, n = 6 in each group, p > 0.05 between any two groups). As shown in Figure 4B,C, the mean intensity of green fluorescence in the whole sections from Sham rats was extremely low. In contrast, fluorescence emitted by extravasated FITC-dextran increased markedly in the IR group, normalized to the Sham group (5.52 ± 0.85 in the IR group vs. 1.00 ± 0.19 in the Sham group, n = 6, p < 0.05). Furthermore, TXL superfine powder, known to protect the cardiac microvascular endothelium14,16, remarkably reversed the higher fluorescence intensity induced by IR (2.87 ± 0.52 in IR+TXL group vs. 5.52 ± 0.85 in IR group, n = 6, p < 0.05). The images of the regional myocardium indicated that FITC-dextran overwhelmingly accumulated in the extracellular matrix but not in the microvessels, and that TXL improved endothelial barrier function (Figure 5).
Compared with FITC-dextran, EB may be a suboptimal alternative because EB is inferior to 70,000 Da FITC-dextran in both image quality and sensitivity. Similarly, there was no significant difference in the cross-sectional area among the three groups: 1.00 ± 0.08 in the Sham group vs. 0.97 ± 0.08 in the IR group vs. 0.98 ± 0.05 in the IR+TXL group (Supplemental Figure S1A, n = 6 in each group, p > 0.05 between any two groups). As shown in Supplemental Figure S1B,C, the mean intensity of red fluorescence in whole cross-sections from Sham rats was relatively higher, especially in the endocardium and epicardium. Although red fluorescence emitted by EB increased sharply in the IR group (4.41 ± 0.66 in the IR group vs. 1.00 ± 0.22 in the Sham group, n = 6, p < 0.05), the difference between the IR+TXL and IR groups was not significant (3.73 ± 0.47 in the IR+TXL group vs. 4.41 ± 0.66 in the IR group, n = 6, p = 0.0956).
Furthermore, this method was tested in a myocardial infarction (MI) model (Supplemental Figure S2). In the Myocardial Infarction Injection first (MII) group, FITC-dextran solution (at the same concentration mentioned earlier) was injected into the circulation 5 min before LAD ligation. This was in contrast to the FITC-dextran solution injected immediately after ligation in the MIL rats (Myocardial Infarction Ligation first) described earlier in this protocol. Interestingly, only low amounts of FITC-dextran extravasated into the ECM of the injured myocardium in MI rats of both groups, indicated by weak green fluorescence intensity in Supplemental Figure S2.
Figure 1: Visual field of thoracotomy and establishment of ischemia/reperfusion in a rat model. (A) Exposing the rat's heart (indicated by yellow arrow); (B) forming a loop to encircle the coronary artery (indicated by green arrow); (C) fastening the loop and inducing myocardial ischemia for 45 min (indicated by white arrow); (D) loosening the loop and allowing myocardial reperfusion for 3 h (indicated by the blue arrow). Please click here to view a larger version of this figure.
Figure 2: Dissection of the femoral vein and intravenous injection. (A) Exposing the femoral vein (the femoral nerve is indicated by the arrow on the left; the femoral artery is indicated by the arrow in the middle; the femoral vein is indicated by the arrow on the right). (B) Injecting FITC-dextran solution into the femoral vein with an insulin syringe. Abbreviation: FITC = fluorescein isothiocyanate. Please click here to view a larger version of this figure.
Figure 3: Tissue perfusion and symptoms of efficient perfusion. (A) Dissecting the abdominal aorta and the vena cava (the abdominal aorta is indicated by the white arrow; the vena cava is indicated by the black arrow). (B) Bleached lungs are a symptom of adequate perfusion; (C) bleached mucosa, bleached kidney, and sweating liver are all symptoms of adequate perfusion (red arrow: mucosa; white arrow: liver; black arrow: kidney). Please click here to view a larger version of this figure.
Figure 4: Representative images and analysis of mean fluorescence intensity by FITC-dextran staining. (A) Relative cross-sectional area of each group normalized to that of the Sham group (p > 0.05); (B) relative fluorescence intensity in the FITC channel of each group normalized to that of the Sham group; (C) representative images at 9.5x magnification in each group. Scale bars = 800 µm. One-way ANOVA with Tukey's post-hoc multiple-group comparisons. * vs. Sham group, p < 0.05; # vs. IR group, p < 0.05. Bar graphs show group mean ± SD. Abbreviations: IR = ischemia/reperfusion; TXL = Tongxinluo; FITC = fluorescein isothiocyanate; DAPI = 4',6-diamidino-2-phenylindole; ANOVA = analysis of variance; SD = standard deviation. Please click here to view a larger version of this figure.
Figure 5: Representative images of areas of interest at 200x magnification of each group after FITC-dextran staining. Scale bars = 50 µm. Abbreviations: FITC = fluorescein isothiocyanate; DAPI = 4',6-diamidino-2-phenylindole. Please click here to view a larger version of this figure.
Supplemental Figure S1: Representative images and analysis of mean fluorescence intensity by Evans Blue staining. (A) Relative cross-sectional area of each group normalized to that of the Sham group (p > 0.05); (B) relative fluorescence intensity in the Texas red channel of each group normalized to that of the Sham group; (C) representative images in each group. One-way ANOVA with Tukey's post-hoc multiple-group comparisons. Scale bars = 800 µm. * vs. Sham group, p < 0.05. Bar graphs show group mean ± SD. Abbreviations: IR = ischemia/reperfusion; TXL = Tongxinluo; DAPI = 4',6-diamidino-2-phenylindole; ANOVA = analysis of variance; SD = standard deviation. Please click here to download this File.
Supplemental Figure S2: Representative images at 9.5x magnification of FITC-dextran staining in a myocardial infarction model. Scale bars = 800 µm. Abbreviations: MII = myocardial infarction injection first (inject FITC-dextran before myocardial infarction); MIL = myocardial infarction ligation first (inject FITC-dextran immediately after myocardial infarction); FITC = fluorescein isothiocyanate; DAPI = 4',6-diamidino-2-phenylindole. Please click here to download this File.
This paper presents a protocol to assess the dysfunction of the cardiac endothelial barrier after being exposed to MIRI. It should be noted that there are several crucial steps in this protocol that determine the reliability and accuracy of the results. All steps involving FITC-dextran are to be performed in a dark room. The FITC-dextran/saline solutions must be stored at 4 °C and wrapped with tin foil. Additionally, coronary arteries must be ligated, as illustrated in Figure 1, so that the myocardium can be reperfused immediately after the loop is loosened. Researchers can choose 5-0 or 6-0 polyethylene sutures because thinner strings under 7-0 are more likely to cut myocardial fibers and vessels. Importantly, the tracer solution should be injected into the femoral veins and not the tail veins. Each heart should be well-perfused and well-rinsed with ice-cold saline before and after being sacrificed, respectively.
Many researchers specializing in MIRI are inclined to study the disrupted subcellular structure of the endothelium in the earlier stage of myocardial reperfusion-the formation of apoptotic bodies, swelling mitochondria, and abnormal intercellular junctions-to prove the hyperpermeability of the endothelium via electron microscopic examination. Immunoblotting has also been used to measure microvascular permeability because the decrease in levels of tight conjunction proteins is regarded as the core mechanism of endothelial injury and malfunction of endothelial barrier over time. For instance, VE-cadherin20 and β-catenin21 are key elements of endothelial cell-to-cell adhesion junctions, and these connections contribute to the maintenance of vascular integrity22. Tight junctions, such as JAM-A23,24 and occludin25, also cooperatively participate in the construction of the endothelial barrier in the circulatory system.
Currently, an increasing number of studies suggest that mitochondrial function and mitochondrial homeostasis could be key targets of protective pathways against reperfusion injury. As reported in pathophysiological and pharmacological studies, mitophagy can play an important role in the cellular death of the endothelium and myocardium induced by IR26. In addition, mitochondrial fission, mitochondrial proteostasis, and mitochondrial quality control may protect high-risk myocardium in reperfused areas27,28. The above methods focus more on demonstrating different mechanisms or causes of hyperpermeability than on displaying endothelial permeability directly and visually. Moreover, sample preparation for electron microscopic examination is complicated and dangerous. Glutaraldehyde, acetone, and osmic acid solutions used in electron microscopy probably volatilize and erode the mucosa, conjunctiva, and skin, even threatening lives if used in uncontrolled environments.
Compared to these indirect and life-threatening methods, this method, which measures the fluorescence intensity emitted by FITC-dextran extravasation, is ideal for assessing cardiac microvascular permeability. First, 70,000 Da FITC-dextran cannot penetrate through the endothelium under normal physiological conditions but can enter the myocardial interstitial space under IR, which allows a clear differentiation between rats under different treatments. Additionally, the FITC-dextran solution is injected into the blood via the femoral veins, which guarantees that all of the fluid is injected into the rat’s circulation. In contrast, tail vein injection is more difficult in rats because the tail vein is indistinct and covered with a thick stratum. Further, unlike regular paraformaldehyde fixation and paraffin embedding, myocardial tissue treatment is simplified and optimized using the quick-freeze pathological section technique used in this study. Short storage time, light-proof environment, and cryopreservation may together alleviate fluorescence quenching of FITC.
More importantly, this method can minimize selection bias during statistical analysis. We perform sufficient heart perfusion with saline to ensure that nearly all the blood mixed with FITC-dextran is douched out from the myocardium and the heart chambers, which can eliminate the fluorescence emitted by any FITC remaining in the vessels. If this is not the case, fluorescence merged with capillaries should be excluded and trimmed artificially. While collecting tissue from areas of interest, no visual boundary was observed between the ischemic myocardium and the remote myocardium, especially after EB dye staining. This hindered the collection of samples from these representative areas. Using the method described herein, researchers can obtain images of whole sections and analyze the fluorescence intensity using a quantitative slider scanner.
Interestingly, EB dye can be an alternative, albeit suboptimal, to FITC-dextran. Compared with the relatively harsh storage conditions of FITC-dextran, EB powder or its solution can remain stable at room temperature for a long time. Additionally, EB is relatively inexpensive for most institutions. Nonetheless, certain disadvantages could limit its application. First, EB staining shows lower sensitivity than FITC-dextran, which may underestimate the differences among groups and increase the sample size of each group. Besides, EB shows stronger adhesion to proteins. The endocardium and epicardium are more easily stained by EB, which can influence the results of fluorescence density. Further, obvious filling defects in the reperfusion area remarkedly reduce the image quality, which can limit the use of EB.
To extend the application of this method in an MI model, the method was tested in an MI model. First, FITC-dextran was injected into blood vessels after LAD ligation. As expected, a low level of fluorescence was emitted from the slice because FITC-dextran in blood flow cannot pass through the ligated site, and no FITC-dextran entered the ischemic area. FITC-dextran was also applied 5 min before ligation to ensure the passage of FITC-dextran into all organs. Consequently, FITC-dextran failed to penetrate the microvascular endothelium probably because no perfusion pressure in ligated blood vessels can drive FITC-dextran through the endothelium. Generally, this method probably requires significantly more modification to satisfy its application in the MI model.
Despite the above advantages, there are limitations to this method. First, this protocol was not tested in IR models of other animals, which may limit its application in preclinical and clinical studies. Second, FITC-labeled dyes are more expensive than other common dyes such as EB. However, unlike EB, as 10% w/v FITC-dextran solution, diluted with serum, rarely changes the color of the myocardium, FITC-dextran staining and EB/2,3,5-triphenyltetrazole chloride (TTC) dual staining may be conducted in the same heart, which can trim the budget with respect to the number of animals and the dyes used. Last, the slide scanner used in this study-the Quantitative Slide Scanner and similar instruments-may not be available in some laboratories. In summary, this protocol presents a simplified, reliable, and visual method for the detection of cardiac microvascular endothelial permeability in an IR model in rats using a fluorescence assay based on an automated quantitative pathology imaging system. This protocol demonstrates that 70,000 Da FITC-dextran performs better than EB in the functional assessment of the endothelial barrier of cardiac microvessels after IR injury.
The authors have nothing to disclose.
This study was supported by grants from The National Key Research and Development Program of China (No. 2017YFC1700503) and the National Natural Science Foundation of China (No. 81874461 and No. 81774292).
16 G Angiocather | Becton Dickinson, USA | 2089312 | |
3-0 polypropylene suture, taper needle | Ethicon, USA | KGP693 | |
4’,6-diamidino-2-phenylindole (DAPI) | Invitrogen, USA | D3571 | Alternative to Mounting medium with DAPI |
6-0 polypropylene suture, round needle | Ethicon, USA | KGJ656 | |
Adhesive Microscope Slides | CITOTEST, China | 188105 | |
ALC-V8S Ventilator | Shanghai Alcott Biotech, China | ALC-V8S | |
Constant Temperature Warming System | Shanghai Alcott Biotech, China | ALC-HTP203 | |
Dissecting Forceps-Large Cvd | RWD life Science, China | F12007-10 | |
Dissecting Forceps-Str | RWD life Science, China | F12005-10 | |
FITC-dextran, 70,000Da | Sigma-Aldrich, USA | 53471 | |
Hair Removal Cream | Nair, USA | N/A | |
Insulin Syringe,1ml | Becton Dickinson, USA | 8155672 | |
Microneedle holders | RWD life Science, China | F31047-12 | |
Mosquito Forceps-Cvd | RWD life Science, China | F22007-12 | |
Mounting medium with DAPI | ZSGB-BIO, China | ZLI-9557 | |
Multi-position Fiber Optic Illuminator System | RWD life Science, China | 76301 | |
Operating Scissors | RWD life Science, China | S14014-12 | |
Pannoramic SCAN | 3DHISTECH, Budapest, Hungary | SC150-211305 | Alternative slide scan to Vectra Polaris |
Phenochart 1.0.8 | PerkinElmer, USA | N/A | Slide viewer software |
Phosphate buffered saline | HyClone,USA | SH30256.01 | |
Retractors,30mm Spread,5cm | RWD life Science, China | F22005-45 | |
SP8 Confocal microscope | Leica, German | TCS-SP8 SR | Alternative to Vectra Polaris |
Surgical tape | 3M, USA | 1527C-0 | |
Tissue-Tek O.C.T. Compound | SAKURA,Japan | 0822-00 | |
Tongxinluo superfine powder | Yiling, China | Z19980015 | |
Ultraviolet Freezing Slicer | Leica, German | CM1950 | |
Vectra Polaris 1.0.7 software | PerkinElmer, USA | N/A | Slide scan software |
Vectra Polaris Quantitative Slide Scanner | PerkinElmer, USA | VP1814N1068 |
.