Authors: Renate Paddenberg, Petra Mermer, Anna Goldenberg, Wolfgang Kummer
Categories: Medicine, Hypoxic pulmonary vasoconstriction, Issue 83, intra-pulmonary, murine lungs, pre- and intra-acinar arteries, precision cut lung slices, videomorphometry
Source: Journal of Visualized Experiments : JoVE
Doi: 10.3791/50970
Acute alveolar hypoxia causes pulmonary vasoconstriction (HPV) - also known as von Euler-Liljestrand mechanism - which serves to match lung perfusion to ventilation. Up to now, the underlying mechanisms are not fully understood. The major vascular segment contributing to HPV is the intra-acinar artery. This vessel section is responsible for the blood supply of an individual acinus, which is defined as the portion of lung distal to a terminal bronchiole. Intra-acinar arteries are mostly located in that part of the lung that cannot be selectively reached by a number of commonly used techniques such as measurement of the pulmonary artery pressure in isolated perfused lungs or force recordings from dissected proximal pulmonary artery segments^1,2^. The analysis of subpleural vessels by real-time confocal laser scanning luminescence microscopy is limited to vessels with up to 50 µm in diameter^3^.
We provide a technique to study HPV of murine intra-pulmonary arteries in the range of 20-100 µm inner diameters. It is based on the videomorphometric analysis of cross-sectioned arteries in precision cut lung slices (PCLS). This method allows the quantitative measurement of vasoreactivity of small intra-acinar arteries with inner diameter between 20-40 µm which are located at gussets of alveolar septa next to alveolar ducts and of larger pre-acinar arteries with inner diameters between 40-100 µm which run adjacent to bronchi and bronchioles. In contrast to real-time imaging of subpleural vessels in anesthetized and ventilated mice, videomorphometric analysis of PCLS occurs under conditions free of shear stress. In our experimental model both arterial segments exhibit a monophasic HPV when exposed to medium gassed with 1% O2 and the response fades after 30-40 min at hypoxia.
Keywords: Medicine, Issue 83, Hypoxic pulmonary vasoconstriction, murine lungs, precision cut lung slices, intra-pulmonary, pre- and intra-acinar arteries, videomorphometry
In most systemic vascular beds hypoxia induces vasodilatation, in comparison to the vasoconstriction caused by hypoxia in the pulmonary vasculature. This lung-specific response to lowered oxygen tension is called hypoxic pulmonary vasoconstriction (HPV), onsets within seconds and reverses quickly after switchback to normoxic ventilation. Although HPV is known for more than 60 years, the cellular oxygen sensor(s) and the signaling cascade(s) resulting in vasoconstriction are still under debate. There is a relative broad consensus that hypoxia-evoked redox and ROS changes are essential for HPV and the development of pulmonary hypertension (reviewed in Sylvester et al.^4 ^andSchumacker et al.^5^). Our own data support a central role of complex II of the mitochondrial respiratory chain in HPV^6,7^. Recently, Wang et al. presented a completely new concept for oxygen sensing and HPV: Based on their data they propose that alveolar hypoxia is sensed by the adjacent capillaries causing membrane depolarization of the endothelial cells. The response is propagated via connexion 40 gap junctions of the endothelial cells leading to constriction of smooth muscle cells of upstream arterioles^8^.
The arteries of the lung run alongside the airways, branch with them, continuously decrease in diameter, and finally supply blood to the capillary system located in the alveolar walls. This arterial circulation is composed of anatomically and functionally distinct segments. The proximal conduit arteries, characterized by an abundance of elastic fibers in the walls, are followed by fully muscularized intra-pulmonary arteries which largely control the pulmonary vascular resistance. Step-by-step, these arteries transit into segments where the muscle layer becomes incomplete, and finally the vessels are free of smooth muscle actin-immunoreactive cells. The intra-acinar artery feeding an individual pulmonary acinus with blood represents a partially muscular segment^6^. Likewise, the pulmonary arterial system does not represent a uniform structure regarding the hypoxic response but exhibits marked regional diversity^9,10^. For instance, in proximal pulmonary arteries isolated from rat lungs hypoxia induces a biphasic response, displaying an initial rapid contraction of short duration which - after incomplete relaxation - is followed by a second slow but sustained contraction^11^. In resistance arteries isolated from rat lung parenchyma as fourth- and fifth-division of pulmonary arteries (external diameter <300 μm), hypoxia causes monophasic constriction^9^. Already in 1971 Glazier and Murray concluded from measurements of changes in the capillary red blood cell concentration in lungs of dogs ventilated with hypoxic gas mixtures that the hypoxia-induced increase in vascular resistance mainly occurred upstream of the capillaries^12^. Nowadays, intravital microscopy of intact lungs of anesthetized and mechanically ventilated mice represents a powerful tool for the analysis of the pulmonary microvasculature^13,14^. Excision of a circular window in the thoracic wall gives microscopic access to the surface of the lung and allows the analysis of subpleural pulmonary vessels with up to 50 µm in diameter. By combining this technique with the infusion of FITC-dextran, Tabuchi* et al*. demonstrated that only medium sized arterioles with diameters of 30-50 µm exhibit a marked response to hypoxia which sustained over a period of 60 min with a minor attenuation after 30 min. In contrast, small arterioles with diameters of 20-30 µm showed only a minor response to hypoxia^3^. However, this technique does not allow the analysis of arteries with diameter larger that 50 µm since these vessels are located too deep in the lung tissue.
In order to bridge the gap in the analysis of large and very small pulmonary arteries (such as the subpleural vessels) of murine lungs, we adopted a method that was described by Martin et al. for analysis of the reactivity of airways^15^. Based on an agarose gel instilling technique, it facilitates the preparation of precision cut lung slices (PCLS) from this relatively soft and elastic organ. Within the PCLS vasoreactivity of cross-sectioned arteries with inner diameter between 20-100 µm can be directly observed by videomicroscopy. Application of drugs during the hypoxic incubation of the PCLS allows the analysis of their effects on HPV. It is of particular importance that this technique can also be applied to genetically engineered mouse. Based on their location within the lung, we classify the arteries as pre- and intra-acinar vessels, with inner diameters of 20-40 µm and 40-100 µm, respectively. Under a functional view the intra-acinar artery supplies an individual pulmonary acinus with blood and the pre-acinar artery is the preceding vessel sections. Recording of pictures on a digital camera allows the subsequent quantification of the vasoreaction. An obvious attribute of this PCLS model is the lack of shear-stress acting upon the endothelium. In contrast, in perfused vessels acute HPV leads to a rise in shear-stress thereby inducing secondary mechanisms such as NO release^16^. Additionally, the use of PCLS allows measurements of HPV without extrapulmonary neural or hormonal influences. In contrast to cell culture systems, for instance prepared from canine pulmonary arterial smooth muscle cells^17^, the histological architecture of the vessel wall is almost completely preserved.
In summary, this protocol provides a helpful method for analysis of potential molecular oxygen sensors and/or cellular pathways responsible for HPV of intra-pulmonary arteries with inner diameters between 20-100 µm under conditions free of shear stress.
This section describes the necessary equipment and set-up for the protocol. Additional details and manufacturer information can be found in the accompanying table.
Use mice (for instance of the strain C57Bl6) of both genders at an age of 10-25 weeks. HPV can also be analyzed in knockout strains and the corresponding wild type strains.
All experiments were performed according to the NIH guidelines for the care and use of experimental animals, and were approved by the local institutional boards.
In Figure 1 the results of the measurement of HPV of a large pre-acinar artery and in Figure 2 of small intra-acinar arteries are given. In the phase contrast images (Figures 1A and 2A) it becomes clear that it is possible to discriminate these 2 classes of arteries based on their location within the lung Pre-acinar arteries run in close neighborhood to bronchi and bronchioli (Figure 1A), whereas the intra-acinar arteries are located at gussets of alveolar septa and surrounded by alveolae (Figure 2A). With a little practice it is possible to see the changes in the luminal area in response to U46619 on the phase contrast pictures (Figures 1A and 2A). However, hypoxic pulmonary vasoconstriction is often not so pronounced and becomes clear only after complete evaluation of the changes of the luminal areas (**Figures 1B, 1C, 2B, **and, 2****C). For didactic reason we have given an example of a pre-acinar artery which shows an unusually pronounced vasoconstriction. On average, HPV results in a 20-30% reduction of the luminal area.
In Figure 2 the recordings of small intra-acinar arteries incubated with hypoxic-gassed medium with or without 50 µM pinacidil (a nonselective opener of mitochondrial ATP-sensitive potassium channels) are shown and the inhibitory effect of the drug becomes clearly visible. The final part of the experiment demonstrates the selectivity of the action of the drug on HPV: Vasoconstriction induced by the thromboxane analog U46619 is not changed by the addition of pinacidil. Indeed, the curve for the artery exposed the pinacidil runs distinctly lower than the other two, but the extent of reduction of the luminal area by U46619 alone is comparable. In this case it was the incomplete reversion of HPV which causes the difference between the curves. In this graph also an artery exposed to normoxic gassed medium is included as additional control. Under this condition, no changes of the luminal area are detectable.
In Figure 3 the data of the complete series of measurements on the influence of pinacidil on HPV are shown. For comparison of the two groups for statistical differences, the data sets of the time points indicated were analyzed with the Kruskal-Wallis- and the Mann-Whitney-test. HPV was clearly abolished in presence of pinacidil whereas U46619-induced contraction was unchanged.
Alternatively, it is possible to identify differences between groups by comparison of the area under the curve as described in Müller-Redetzky et al.^22^
**Figure 1. Measurement of HPV of a large pre-acinar artery. **(A) Phase contrast images of a cross-sectioned pre-acinar artery (a) which runs in close neighborhood to a cross-sectioned bronchus (B). The pictures are taken at the time points indicated in (B) by at the beginning of the measurement (a), at the end of the treatment with U46619 (b), at the end of exposure to Nipruss (c), after 30 or 40 min in hypoxic gassed medium (d, e), after washing with normoxic gassed medium (f), and after the final application of U46619 (g). In the graph (B) changes of the luminal area are plotted against time whereas the luminal area at the beginning of the experiment is defined as 100% and vasoconstriction/-dilatation are given as relative values. In this case hypoxia induces a 60% reduction of the luminal area. (C) For a clearer presentation of the hypoxic response, the initial phase of the experiment in which the vasoreactivity was tested is not included, but the value obtained immediately before exposure to reduced oxygen is set as 100% (see also Figure 2). Click here to view larger image.
Figure 2. Impact of pinacidil (a nonselective opener of mitochondrial ATP-sensitive potassium channels; mitoKATP) on HPV of small intra-acinar arteries. (A) Intra-acinar arteries (a) are located at gussets of alveolar septa. Alv = alveolus. The sequence of the individual conditions applied in these experiments is given in the headline of the graph (B). Hypoxic exposure is performed in presence or absence of 50 µM pinacidil. Control incubations are done with normoxic gassed medium. Pictures shown in (A) are taken at the beginning of the measurement (a, a', a"), at the end of the treatment with U46619 (b, b', b"), at the end of exposure to Nipruss (c, c', c"), after 30 or 40 min in hypoxic or normoxic gassed medium (d, d', d"; e, e', e"), after washing with normoxic gassed medium (f, f', f"), and after the final application of U46619 (g, g', g"). For a clearer presentation of the response to normoxia/hypoxia/hypoxia+pinacidil, the values obtained immediately before exposure to normoxic-/hypoxic-gassed medium are set as 100% (C). Click here to view larger image.
Figure 3. Inhibition of HPV by pinacidil. The recordings of small intra-acinar arteries exposed to hypoxic gassed medium with or without 50 µM pinacidil are summarized and presented as means ± SEM. In (A) the complete recordings are given, in (B) the relative data in relation to the value at the beginning of the hypoxic incubation are shown. No vasoreactivity is detectable in PCLS which are exposed to hypoxic-gassed medium containing pinacidil. Vasoconstriction induced by U46619 is not affected by the drug. "n" in brackets refers to the number of arteries/number of animals from which PCLS were made. In other words the first number describes the count of lung sections analyzed and the second number gives the count of mice from which these sections were prepared. At the given time points the differences between both groups are tested for significance. n.s.: not significant, *: p≤0.05, **: p≤0.01, ***: p≤0.001. Click here to view larger image.
**Figure 4. Schematic overview of the method. **Briefly, mice are killed by cervical dislocation. After opening of the chest, the lungs are filled with low melting point agarose and after cooling down cut into 200 µm thick precision cut lung slices (PCLS). After removing the agarose at 37 °C one PCLS is transferred into the flow-through superfusion chamber in which it is exposed to normoxic medium or to medium gassed with 1% O2. Vasoreactivity is recorded as changes in the luminal area. Click here to view larger image.
The isolated ventilated and perfused mouse lung is an excellent model for the analysis of the physiological response of the pulmonary vascular system on changes in oxygen supply and among others permits the continuous measurement of the pulmonary arterial pressure^1^. However, this model does not allow the identification and analysis of those vascular segment(s) showing the strongest response to hypoxia. This is the advantage of our videomorphometric analysis of PCLS which facilitates the measurement of HPV of individual arteries with inner diameters of 20-100 µm. PCLS represent an attractive in vitro model since they closely resemble the organ from which they are prepared. In contrast to cell culture systems, all cell types are present in their original tissue-matrix configuration. Furthermore, one lung is sufficient for the preparation of many PCLS, so that at least partially experiments can be standardized by the use of sections from the same mouse. According to the 3R concept (reduction, refinement and replacement of laboratory animals in the life sciences) of Russell and Burch^23^ this fact also argues for the use of PCLS.
However, one has to keep in mind, that the tissue is damaged by cutting with a vibratome and longitudinal signaling for instance via the endothelial cells as postulated by Kübler et al.^14^ is no longer possible.
Initially, PCLS were mainly applied for biochemical, pharmacological and toxicological studies, but in the meantime they are also used for measurement of bronchial contractility, mucociliary function and vascular responses (for reviews see Sanderson^20^ and Davies^21^). Held et al. have performed a study in which they compared the models of isolated perfused and ventilated mouse lung and of PCLS^24^. They found by analysis of the responses of airways and pulmonary vessels to a variety of endogenous mediators that important characteristics of the whole lung were maintained in PCLS.
In PCLS, hypoxic conditions are not established via the airways as in the intact lung but by incubation of the lung section in hypoxic-gassed medium. We have analyzed the oxygen partial pressure (pO2) of medium pregassed with 1% O2, 5.3% CO2, 93.7% N2 and with 21% O2, 5.3% CO2, 73.7% N2, respectively, using a blood gas analyzer. Immediately before feeding it into the perfusion chamber, the pO2 of the hypoxic gassed MEM was 40 mmHg and that of the normoxic gassed medium 160 mmHg^6^. In the intact lung HPV is induced when alveolar pO2 drops below 50 mmHg^25^, a situation which can be obviously mimicked by application of hypoxic-gassed medium. Our data on the extent of HPV matches well to results obtained with a different experimental approach. Yamaguchi et al. have applied isolated rat lungs to examine microvessels with diameter of 20-30 µm by real-time confocal laser scanning luminescence microscopy coupled to a high-sensitivity camera with an image intensifier^10^. They observed a mean reduction in diameter of 2.7 µm after exposure of the lungs to hypoxia. One can calculate that a 20% reduction of the luminal area as we measure it in our system corresponds to about 15% decrease in diameter.
In our experiments we have classified the arteries as pre- and intra-acinar vessels, respectively, with inner diameters of 40-100 µm and 20-40 µm. In humans the transition from muscular to nonmuscular arteries occurs in the diameter range of 70-100 µm. In mice, smooth muscle cells are present down to an external diameter of 20 µm^26^. For this reason it is not possible to analyze arteries with diameters below 20 µm since they can not reliable be identified based on the phase contrast image. At the other end of the scale, vessels with diameters above 100 µm are hardly to find in PCLS and commonly stripped from the surrounding tissue.
Actually, a number of molecular candidates are discussed as molecular oxygen sensor(s) or as component of the signaling cascade resulting in HPV (for a review see Sylvester et al.^4^). Once appropriate knockout mice are available videomorphometry can be used for analysis of vasoreactivity of pre- and intra-acinar arteries as compared to wild type animals. However, PCLS have also been used for other Faro et al. employed them to characterize the development of the endothelium dependent dilation in lung after birth^29^ and PCLS prepared from guinea pigs exposed to smoke or air daily for 2 weeks were used to demonstrate the impact of cigarette smoke on vasoreactivity via induction of endothelial dysfunction^30^.
In our experiments we classified the arteries as pre-acinar (inner diameters of 40-100 µm) and intra-acinar (inner diameters of 20-40 µm). Especially for the preparation of lung sections which should be used for the analysis of larger vessels it is important to add sodium nitroprusside to the perfusion buffer. This drug prevents the contraction of the vessels during the sample preparation and thereby their rip off from the surrounding tissue leading to incomplete vasodilation. Sodium nitroprusside in the perfusion buffer is not so important for the preparation of lung section which should be used for analysis of small arteries because they are strongly anchored to the alveolar septa.
All experiments should be started with incubations in which the reactivity of the arteries is tested. Rarely, we obtained lung preparations in which no response of vessels to contractors or dilators was detectable. We do not know the reason for May be that the volume of the agarose filled into the lungs was too great or too low so that cutting of the organ into PCLS was not optimal. Alternatively, it is imaginable that the agarose was cooling down too fast during the instillation procedure resulting in damaging shear stress. In case that in an individual PCLS no viable artery is detectable, the section has to be discarded and replaced by another one.
The decision on the viability of an artery was made based on the response to U46619. Application of U46619 at a concentration of 0.1 µM induces a vasoconstriction which - after some exercise - is visible directly in the image sequence at the screen. Since there are some variances in the vasoreactivity we investigate the impact of a drug on HPV by measuring the vasoresponse in lung sections exposed to the drug or to the medium alone in turn.
HPV of an individual artery is often hardly detectable in the microscope, and in average it results in a reduction of the luminal area of about 20-30%. However, small changes in the diameter of an artery have a distinct input on flow resistance. According to the equation "R = 1/r^4^" with R=resistance and r=radius, the flow resistance is inversely proportional to the fourth power of the radius. Let me give an An "ideal artery" exhibiting a circular cross-section with a diameter of 40 µm (r=20 µm) has a luminal area of about 1,260 µm^2^. When the luminal area is reduced by 20%, we can calculate that the diameter of the vessel is reduced by 10.5% to 35.8 µm (r=17.9 µm). According to the equation given above, the flow resistance of this vessel would increase from 6.25 x 10^-6^ to 9.71 x 10^-6^ that is by about 55%. In case of a reduction of the luminal area by 30% the radius would decrease by about 16%, but the flow resistance would increase by about 100%. Although these calculations are an oversimplification in which a laminar blood flow and a vessel form of a rigid pipe are assumed it is suggestive of the impact of already minor changes of the diameter on flow resistance.
Authors have nothing to disclose.
This research is sponsored by the Excellence Cluster Cardio-Pulmonary System.
