This protocol describes the induction of pulmonary hypertension (PH) in mice based on the exposure to hypoxia and the injection of a VEGF receptor antagonist. The animals develop PH and right ventricular (RV) hypertrophy 3 weeks after the initiation of the protocol. The functional and morphometrical characterization of the model is also presented.
Pulmonary Hypertension (PH) is a pathophysiological condition, defined by a mean pulmonary arterial pressure exceeding 25 mm Hg at rest, as assessed by right heart catheterization. A broad spectrum of diseases can lead to PH, differing in their etiology, histopathology, clinical presentation, prognosis, and response to treatment. Despite significant progress in the last years, PH remains an uncured disease. Understanding the underlying mechanisms can pave the way for the development of new therapies. Animal models are important research tools to achieve this goal. Currently, there are several models available for recapitulating PH. This protocol describes a two-hit mouse PH model. The stimuli for PH development are hypoxia and the injection of SU5416, a vascular endothelial growth factor (VEGF) receptor antagonist. Three weeks after initiation of Hypoxia/SU5416, animals develop pulmonary vascular remodeling imitating the histopathological changes observed in human PH (predominantly Group 1). Vascular remodeling in the pulmonary circulation results in the remodeling of the right ventricle (RV). The procedures for measuring RV pressures (using the open chest method), the morphometrical analyses of the RV (by dissecting and weighing both cardiac ventricles) and the histological assessments of the remodeling (both pulmonary by assessing vascular remodeling and cardiac by assessing RV cardiomyocyte hypertrophy and fibrosis) are described in detail. The advantages of this protocol are the possibility of the application both in wild type and in genetically modified mice, the relatively easy and low-cost implementation, and the quick development of the disease of interest (3 weeks). Limitations of this method are that mice do not develop a severe phenotype and PH is reversible upon return to normoxia. Prevention, as well as therapy studies, can easily be implemented in this model, without the necessity of advanced skills (as opposed to surgical rodent models).
Pulmonary Hypertension (PH) is a pathophysiological condition, defined by a mean pulmonary arterial (PA) pressure exceeding 25 mm Hg at rest, as assessed by right heart catheterization1,2. There is a variety of diseases that can lead to PH. In an attempt to organize the PH-associated conditions, several classification systems have been developed. The current clinical classification categorizes the multiple PH-associated diseases in 5 different groups1. This distinction is of importance since various groups of patients have diseases that differ in their clinical presentation, pathology, prognosis, and response to treatment2. Table 1 summarizes the current classification, complemented with the basic histopathological characteristics of each disease.
Table 1: Overview of the clinical classification of PH, along with the main histopathological features within the groups. Suitability of the Hypoxia/SU5416 protocol for modeling PH. This table has been modified from19. PH: Pulmonary Hypertension, PAH: Pulmonary Arterial Hypertension
Despite significant advances in the treatment of PH-associated diseases, PH still remains without a cure, with a 3-year mortality rate ranging between 20% and 80%3. This indicates the imperative need for understanding the underlying mechanisms of PH and, thereafter, the development of novel therapies to prevent, slow down the progression, and cure the disease. Animal models are of crucial importance to this scope. Currently, various models exist to study PH. The interested reader is referred to the excellent reviews on this topic2,3,4. Bearing in mind the variety of diseases leading to PH, it is obvious that the diverse conditions of human PH cannot be perfectly recapitulated in one animal model. The animal models available can be categorized in i) single-hit, ii) two-hit, iii) knockout, and iv) overexpression models3. In the single-hit models, PH is induced by a single pathological stimulus, whereas two-hit models combine two pathological stimuli with the goal of inducing more severe PH and thus more closely imitate the complex human disease. Besides the etiological differences, the several stimuli result in PH modeling differences that depend also on the species and the genetic background of the animals4.
One of the most commonly used classic PH rodent models is the chronic hypoxia model2. Hypoxia is known to induce PH in humans as well as in several animal species. Hypoxia has the advantage of being a physiologic stimulus for PH (Table 1). However, while the degree of hypoxia used for inducing PH in rodents is much more severe than in humans, the single insult (hypoxia) leads only to a mild form of vascular remodeling. This does not imitate the severity of the human disease. The addition of a second-hit, an extra stimulus for inducing PH, showed promising results: injection of the compound SU5416 to rodents combined with the hypoxic stimulus induces a more severe PH phenotype2,5,6. SU5416 is an inhibitor of vascular endothelial growth factor (VEGF) receptor-2. It blocks the VEGF receptors and leads to endothelial cell apoptosis. Under hypoxic conditions, this stimulates the proliferation of a subset of apoptosis-resistant endothelial cells. Furthermore, SU5416 leads to smooth muscle cell proliferation. The combination of these effects results in pathologic vascular remodeling of the pulmonary circulation and leads to elevated PA pressure and right ventricular remodeling2,5,7. The model was first described in rats6 and later on applied to mice4,5,7. The mouse model exhibits less severe vascular remodeling compared to rats. Furthermore, when returned to normoxia, PH continues to progress in rats, while in mice it is partially reversible.
The following protocol describes all the steps for modeling PH in mice using the Hypoxia/SU5416 method (planning, timeline, execution). Additionally, the characterization of the model is described in this protocol: functionally (by invasively measuring the right ventricular (RV) pressure using the open chest technique), morphometrically (by dissecting and weighing both the right and left ventricles), as well as histologically (by evaluating pulmonary vascular remodeling, right ventricular cardiomyocyte hypertrophy and fibrosis).
All the steps and methods described in this protocol can be easily implemented by investigators at any experience level. While the functional measurements of the RV using the open chest technique (described here) is not the gold standard method in the field, it has the advantage that it can be quickly learned and accurately reproduced even by a less experienced experimenter.
Prior to any animal experimentation obtain the local institutional animal care committee authorization. The current experiments were performed after approval by the Institutional Animal Care and Use Committee (IACUC) at the Icahn School of Medicine at Mount Sinai.
1. PH induction
2. Functional characterization by invasive RV pressure measurements
3. Morphometric characterization
In this protocol, we describe in detail the creation of the Hypoxia/SU5416 model for inducing PH in mice. Furthermore, we detail all the needed steps for performing pulmonary vascular and cardiac evaluation at the end of the observation period.
An overview of the experimental design for this model is shown in Figure 1A13,14. Mice are subjected to normobaric hypoxia (10% O2) and subcutaneously injected once a week with SU5416 for three consecutive weeks. The stimuli used to induce PH in this protocol are shown in Figure 1B and 1C.
The VEGF receptor antagonist SU5416 acts by causing endothelial cell apoptosis and, therefore, allowing the proliferation of apoptosis resistant-endothelial cells. This leads to vascular remodeling in the pulmonary vasculature and increased vascular resistance5. The elevated pressure in the pulmonary circulation increases the RV afterload and leads progressively to the RV dysfunction and failure9. In the first step, the success of the Hypoxia/SU5416 protocol can be evaluated by functionally assessing the RV function at the end of the observation period. In this protocol, we describe in detail the invasive assessment of the RV systolic pressure using the open chest RV pressure measurement method. Representative pressure curves and quantitative analysis of the right ventricular pressure are displayed in Figure 2.
How can we quantify vascular remodeling, which leads to elevated vascular resistance and consequently PH? Histomorphometry is the gold standard for characterizing the pulmonary vasculature. In this protocol, we describe in detail the Hematoxylin & Eosin Staining (H&E) protocol. After staining and capturing the images, the pulmonary arteries can be distinguished in small (<50 µm) and larger ones (> 50 µm). Bronchial arteries were excluded from our study. For assessing the medial thickness, the external (ED), as well as the internal diameter (ID) of the arteries, is measured. Representative images of remodeled pulmonary arteries after Hypoxia/SU5416 treatment are shown in Figure 3A. The percentage of arteries medial thickness in relation to cross-sectional diameter is shown in Figure 3B. The morphometric analysis of distal pulmonary arteries demonstrates a significant increase in medial thickness in Hypoxia/SU5416-treated mice in comparison with normoxia animals (Figure 3).
The increased afterload leads to RV hypertrophy and as the disease progresses to RV fibrosis9,15. RV hypertrophy can be assessed morphometrically by measuring the Fulton Index (RV/LV+Septum) as well as by measuring cardiomyocyte (CM) hypertrophy. The weight ratio of the right ventricle (RV) to the left ventricle (LV) plus septum [RV/(LV+S)] is calculated as an index of right ventricular hypertrophy. Representative results from the Fulton Index in Hypoxia/SU5416 and normoxia mice are shown in Figure 4B. The method described here for assessing CM hypertrophy is the staining of right ventricular sections with Wheat Germ Agglutinin (WGA). WGA binds to glycoproteins of the cell membrane and can be used for determining the cross-sectional area of myocytes16,17. Representative images of right ventricular sections stained with WGA are shown in Figure 4A. Quantifications of CM area in both diseased and control mice are shown In Figure 4A. Hypoxia/SU5416 exposure results in a marked increase in cardiomyocyte size and right ventricular hypertrophy (Figure 4). We and others have previously shown that, when compared to the single hit (only hypoxia), Hypoxia/SU5416 aggravates the RV phenotype5,18.
Figure 1: Overview of the Hypoxia/SU5416 method. (A) Experimental design for the Hypoxia/SU5416 mouse model. SU5416 is injected subcutaneously once a week for 3 consecutive weeks. (B) Schematic representation of the hypoxia system. The controller senses and regulates oxygen inside the chamber by infusing nitrogen through the gas infusion tube. (C) Chemical structure of SU5416. Please click here to view a larger version of this figure.
Figure 2: Right ventricular pressure in mice exposed to chronic hypoxia combined with SU5416 injection. (A) Representative tracings of invasive pressure measurements of the right ventricle (RV). (B) RV systolic pressure in Hypoxia/SU5416 mice and control animals exposed to normoxia. n = 6-8 mice per group. *** p < 0.001. All quantitative data are reported as means ± SEM. Please click here to view a larger version of this figure.
Figure 3: Hypoxia/SU5416 induces pulmonary vascular remodeling. (A) Representative Hematoxylin/Eosin-stained sections of lungs from the indicated groups demonstrate increased media wall thickness in pulmonary arteries of Hypoxia/SU5416 mice. Scale bar: 50 μm. (B) Percentage of arteries medial thickness in relation to cross-sectional diameter. n = 5 mice per group. *** p < 0.001. All quantitative data are reported as means ± SEM. Please click here to view a larger version of this figure.
Figure 4: Right ventricular hypertrophy in mice exposed to chronic hypoxia combined with SU5416 injection. (A) (Left) Representative WGA (Wheat Germ Agglutinin) staining of right ventricular tissue after the indicated treatment. Scale bar: 50 μm. (Right) Quantitative analysis of the data. n = 5 mice per group. (B) RV hypertrophy reflected by the RV weight over LV plus interventricular septum (S) weight ratio (Fulton index= RV/LV+ S) in each group. n = 8 mice per group. *** p < 0.001. All quantitative data are reported as means ± SEM. Please click here to view a larger version of this figure.
This protocol describes how to model PH in mice by combining two pathological stimuli: chronic hypoxia and SU5416 injection (Hypoxia/SU5416)18. In an attempt to correlate this mouse model with the human PH condition, one inevitably must look at the current PH classification, shown in Table 1. PH in almost all forms is characterized by pulmonary vasoconstriction and aberrant proliferation of endothelial and smooth muscle cells. This leads to elevated pressure in the pulmonary arteries and consequently to increased afterload of the right ventricle.
Every attempt to characterize an animal model of PH should include evidence of the histopathological remodeling of the pulmonary vasculature and the right ventricle. The single-hit hypoxia mouse model leads to a mild form of vasculature remodeling2,3. These pathological findings include the muscularization of previously non-muscularized vessels, accompanied by endothelial cell, smooth muscle cell and fibroblast proliferation. These findings are aggravated by the addition of the second hit (SU5416 injection). The effects are reversible in the single-hit (hypoxia) model and only partly reversible in the Hypoxia/SU5416 model.
The main cause of death for PH patients is the right ventricular failure (RVF)4,20. Pulmonary vascular remodeling in animal models is not always accompanied by RVF. In order to characterize an animal model in terms of RVF morphological, functional and molecular data should be analyzed. The latter is beyond the scope of this protocol. RV morphological remodeling includes both macro- and microscopical aspects. At the macroscopical level, the main index for RV hypertrophy is the Fulton index, defined as the weight of RV divided by the left ventricular (LV) and Septum (S) weight (RV/LV+S). At the microscopical level, fibrosis, inflammation, and hypertrophy can be assessed by Sirius red, Hematoxylin/Eosin and WGA staining, respectively.
The mouse Hypoxia/SU5146 model (which is described here) shows an RV dysfunction, as measured by elevated systolic pressures and morphological criteria. Regarding pulmonary vascular remodeling, medial hypertrophy is observed three weeks after the initiation of the protocol. Compared to the Hypoxia/SU5416 model in rats, the mouse model does not cause RV Failure (only moderate dysfunction), does not lead to severe obliterative angiopathy, as observed in severely diseased humans, and the pulmonary pathology ameliorates after return to normoxia. Overall, the mouse Hypoxia/SU5416 model is suitable for imitating vascular injury as encountered in PH, predominantly Group I (partially Group III, see Table 1)1,19. The advantage of this model is the application in wild type (genetically unmodified) mice, the relatively easy and low-cost implementation, the relatively low mortality of the diseased animals, and the quick development of the disease of interest (3 weeks). PH prevention and therapy studies can easily be implemented in this model, without the necessity of advanced skills as opposed to surgical rodent models.
When implementing the protocol there are some critical steps, which one should keep in mind. When planning the study, one should keep in mind that in the Hypoxia/SU5416 group the mortality of the animals varies between 0-10% (unpublished observations). Therefore, in order to reach statistical power and avoid underpowered studies, at least 10 mice per group are recommended. The solubility of SU5416 is low. Therefore, DMSO or another solvent (e.g. Carboxymethyl cellulose, CMC) have to be used. DMSO in high doses can be toxic. The LD50 for subcutaneous (s.c.) use in mice has been reported to be 13.9 - 25.6 g/kg21,22. LD50 is defined as the dose required to kill 50% of the members of a tested population after a specified test duration21,22. For a mouse that weighs 25 g, 4.4 g/Kg of DMSO is used (calculations based on DMSO density of 1.1 g/mL and 0.1 mL applied s.c./mouse). Therefore, the subcutaneously given dose is much lower than the LD50 value. In our hands, the application of SU5416 dissolved in DMSO, as described here, can cause skin irritation in some cases, but no other toxic effects are observed. However, several reports recommend the use of CMC as an alternative vehicle to SU541614. When performing the RV functional measurements, close attention has to be paid at the body temperature, bleeding, and depth of anesthesia, as assessed by testing the mouse reflexes. The open chest technique for assessing the RV pressure, as described here, has the advantage of easily being implemented even by an inexperienced user. The closed-chest method (described elsewhere23,24,25) has the advantage of being less invasive and can, therefore, be implemented also in non-terminal experiments. It requires though a high level of expertise.
After the first description of the Hypoxia/SU5416 model in rats, the mouse model has been successfully used in several studies5,9,13. However, there is evidence that the results depend on the genetic background and sex of the mice, the manufacturer of SU5416 and the frequency of SU5416 injection26. While injecting SU5416 over three consecutive weeks leads to PH in mice, a single dose would not induce PH4. Furthermore, other forms of PH, such as those associated with left heart disease or due to chronic thromboembolic disease, require etiology-related models. New therapies should be tested in at least 2 different animal models, before being able to pave the way to translational studies.
The authors have nothing to disclose.
This work was supported by grants from the American Heart Association (AHA- 17SDG33370112 and 18IPA34170258) and from the National Institutes of Health NIH K01 HL135474 to Y.S. O.B was supported by the Deutsche Herzstiftung.
Acetic acid glacial | Roth | 3738.1 | |
Acetone, Histology Grade | The Lab Depot | VT110D | |
ADVantage Pressure-Volume System | Transonic | ADV500 | |
Bouin's solution | Sigma | Ht10132 | |
Cautery System | Fine Science Tools | 18000-00 | |
Connection tubing and valves | |||
Cotton-Tipped Applicators | Covidien | 8884541300 | |
Coverslips, 24 x50 mm | Roth | 1871 | |
Data Acquisition and Analysis | Emka | iox2 | |
Direct Red 80 | Sigma | 365548-5G | |
DMSO (Dimethyl Sulfoxide) | Sigma Aldrich | 276855 | |
Dry ice | |||
Dumont # 5 forceps | Fine Science Tools | 11251-10 | |
Dumont # 7 Fine Forceps | Fine Science Tools | 11274-20 | |
Embedding molds | Sigma Aldrich | E-6032 | |
Eosin Solution Aqueous | Sigma | HT110216 | |
Ethanol, laboratory Grade | Carolina Biological Supply Company | 861285 | |
Fast Green FCF | Sigma | F7252-5G | |
Fine scissors | Fine Science Tools | 14090-09 | |
Goat Serum | invitrogen | 16210-064 | |
Heating pad | Gaymar | T/Pump | |
Hematoxylin 2 | Thermo Scientific | 7231 | |
Hypoxic chamber | Biospherix | A30274P | |
Induction chamber | DRE Veterinary | 12570 | |
Intubation catheter (i.v. catheter SurFlash (20 G x 1") ) | Terumo | SR*FF2025 | |
Iris scissors | Fine Science Tools | 14084-08 | |
Isoflurane | Baxter | NDC-10019-360-40 | |
Isoflurane vaporizer | DRE Veterinary | 12432 | |
Mice (C57BL/6) | Charles River | ||
Needles 25 G x 5/8" | BD | 305122 | |
OCT | Tissue Tek | 4583 | |
PBS (Phosphate Buffered Saline) | Corning | 21-031-CV | |
Piric Acid- Saturated Solution 1.3 % | Sigma | P6744-1GA | |
Pressure volume catheter | Transonic | FTH-1212B-4018 | |
Retractor | Kent Scientific | SURGI-5001 | |
Static oxygen Controller ProOx 360 | Biospherix | P360 | |
SU 5416 | Sigma Aldrich | S8442 | |
Surgical Suture, black braided silk, 5.0 | Surgical Specialties Corp. | SP116 | |
Surgical tape | 3M | 1527-1 | |
Syringe 10 ml | BD | 303134 | |
Syringes with needle 1 ml | BD | 309626 | |
Sytox Green Nuclein Acid Stain | Thermo Scientific | S7020 | |
Tenotomy scissors | Pricon | 60-521 | |
Toluol | Roth | 9558.3 | |
Ventilator | CWE | SAR-830/P | |
WGA Alexa Fluor | Thermo Scientific | W11261 | |
Xylene | Roth |