| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Animal Model |
From the Division of Vascular Biology,* La Jolla Institute for Molecular Medicine, San Diego; and the Department of Medicine, University of California San Diego, La Jolla, California
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
induced leukocyte rolling and adhesion in both arterioles and venules. Treatment with function-blocking anti-selectin mAb revealed that P- and L-selectin are the predominant rolling receptors in the lung microvessels, with E-selectin strengthening P-selectin-dependent interactions. Intravital microscopic studies also demonstrated that during their transit in capillaries, some leukocytes undergo shape change and continue to roll as elongated cells in postcapillary venules. Furthermore, the revascularized microvessels demonstrated the ability to undergo vasoconstriction in response to superfusion with endothelin-1. Overall, these studies demonstrate that the revascularized lung allograft is responsive to various external stimuli such as cytokines and vaso-active mediators and serves as a model to evaluate the interaction of leukocytes with the vascular endothelium in the lung microcirculation under acute as well as chronic experimental conditions.
In previous studies, neonatal hamster pulmonary allografts transplanted into the hamster cheek pouch demonstrated complete revascularization and establishment of blood flow.3-5 Histology studies have shown that the transplanted hamster lung tissue in the hamster cheek pouch retains all of the cell types present in normal lung tissue.4 Extravasation of macromolecules in response to nicotine has been observed in transplanted hamster pulmonary allografts.6 These studies suggest that the LMV of transplanted hamster lung allografts exhibit many of the morphological and physiological characteristics noted in blood vessels of lungs situated intrathoracically in normal animals. However, there are no studies examining leukocyte-endothelial interactions in LM of transplanted lung allografts in this or other animal models. As reagents to study leukocyte adhesion in the hamster are not as readily available as in the mouse, we have used IVM-based techniques to examine the dynamics of leukocyte-endothelial cell interactions in the murine as opposed to the hamster lung vascular bed. We have developed a murine model to continuously visualize the dynamics of leukocyte-endothelial interactions under conditions of flow in LMV by transplanting lung tissue into the dorsal skin-fold window chamber of nude mice.
In this study, we demonstrate that transplanted murine lung allografts undergo revascularization and establish blood flow. Leukocyte rolling and adhesion in response to stimulation with TNF- occurs in both arterioles and postcapillary venules of the murine lung microcirculation, as opposed to murine systemic circulation where rolling is observed predominantly in postcapillary venules and rarely in arterioles.7
P-selectin and L-selectin appear to be the major rolling receptors in inflamed LMV. E-selectin appears to participate by strengthening P-selectin-dependent interactions, which in turn contributes to sequential leukocyte rolling and adhesion in these microvessels. In addition, shape changes in leukocytes during their transit in lung capillaries were noted, with deformed leukocytes continuing to roll as deformed cells in postcapillary venules.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Preparation of Recipient Mouse Dorsal Skin-Fold Chambers
Dorsal skin-fold chambers in nude mice (Jackson Laboratory, Bar Harbor, ME) were prepared as previously described.8,9 In brief, 8- to 10-week old male nude mice (25 to 30 grams body weight) were anesthetized with a subcutaneous injection of a saline solution containing a cocktail of ketamine hydrochloride and xylazine (Phoenix Pharmaceutical Inc., St. Joseph, MO; 7.5 and 2.5 mg, respectively, per 100 mg body weight) and placed on a heating pad. One pair of identical titanium frames was implanted into a dorsal skin-fold (parallel to the animal’s dorsum) so as to sandwich the stretched double layer of skin. One layer of the dorsal skin was completely removed in a circular area of 15-mm diameter. The underlying thin layer of striated skin muscle (M. cutaneous max.), subcutaneous tissue, and epidermis was covered with a coverslip enclosed in one of the frames and the mice were allowed to recover from anesthesia. After a recovery period of 2 to 3 days, the coverslip was removed and lung allografts (up to 3) were carefully placed on the upper tissue layer in the chamber as described below.
Transplantation of Donor Mouse Lung Allograft
Neonatal mice (BALB/c, 5 to 24 hours old) from Harlan (Indianapolis, IN) were used as donors of the lung tissue to be transplanted into the skin-fold window chambers of recipient nude mice. The transplantation of the lung allograft was performed aseptically in a laminar flow hood. Longitudinal lung slices were obtained from the periphery of the lungs and placed into sterile Hank’s buffer containing 5-(and 6)-(((4-chloromethyl)benzoyl)amino) tetramethylrhodamine (CMTMR, Molecular Probes, Eugene, OR) at a concentration of 1 mg/ml for a period of 30 minutes to fluorescently label the lung allograft. The fluorescently labeled lung tissue was briefly washed in Hank’s buffer (containing CMTMR) and used for implantation. The CMTMR-labeled tissue was found to remain fluorescent for 2 to 3 days. The recipient nude mouse was anesthetized as described above and 1 to 3 small slices of donor lung tissue (
1 to 5 mm2) were carefully transplanted into the skin-fold chamber by placing the lung allograft on the subcutaneous tissue of the skin-fold chamber. The chamber containing the lung allograft was then superfused with 25 to 50 µl of sterile saline and covered with a sterile siliconized coverslip kept in position by a "C"-ring.
Observation of the Establishment of Blood Flow in the Lung Allograft by IVM
To observe the transplanted lung microvasculature, the unanesthetized mouse recipient of the lung allograft was placed in a crouched position in a plexiglas tube which was closed at one end and had a longitudinal slit to accommodate the skin-fold chamber as previously described.8-10 The tube had holes to allow the mouse to breathe. The skin-fold chamber containing the lung allograft was then immobilized on a platform and placed on the stage of a Leitz Biomed intravital microscope for observation of the LM within the skin-fold chamber. On the day of implantation (day 0), an overview video picture of the CMTMR-labeled lung allograft was taken using a Leitz PL 1.6X (NA = 0.05) objective. The establishment of vasculature and blood flow was observed either by intravenous (i.v) administration of 50 to 100 µl of 2.5% fluoresceinisothiocyanate (FITC)-dextran 500,000 (Sigma Chemical Co., St. Louis, MO) to obtain vessel contrast and plasma enhancement or by transillumination using a mercury or halogen lamp (VEDCO, St. Joseph, MO). Epi-illumination of FITC-labeled vessels was obtained using a silicone-intensified target camera (SIT68; DAGE MTI, Michigan City, IN) attached to the microscope and connected to a monitor (Panasonic). All images were recorded on a S-VHS videocassette recorder (HC-6600; JVC, Tokyo, Japan) for play back off-line analysis. Observations were made periodically over a 2-week period.
Labeling of Erythrocytes and Microhemodynamics in Lung Allograft Microvessels
Microhemodynamic parameters were measured in the revascularized lung allograft by administering mouse erythrocytes labeled with FITC. Briefly, freshly drawn heparinized blood from 5 to 6 adult mice was centrifuged at 1000 rpm for 5 minutes and the ensuing buffy coat discarded. The erythrocytes obtained by this process were washed with phosphate-buffered saline (PBS) to reduce the final leukocyte contamination below 0.1%. The packed erythrocyte preparation was diluted with PBS to adjust the hematocrit (10%) and labeled with FITC at a final concentration of 0.1 mg/ml. The erythrocyte/FITC solution was incubated for 30 minutes at 37°C in the dark with gentle, intermittent shaking. The FITC-labeled erythrocytes were subsequently washed with PBS three times. The labeled erythrocyte fraction (2.5 x 106) was injected into the tail vein of nude mice with completely revascularized lung allografts (after day 9). The circulation of the injected erythrocytes was visualized by stroboscopic epi-illumination and images recorded using an S-VHS videocassette recorder. The velocity of the erythrocytes in arterioles, capillaries, and postcapillary venules was determined by off-line analysis of recorded video images. In addition to analysis of vessel diameter, shear rates in various LMV were determined as previously described.11
Histology
Tissue sections containing the transplanted lung allograft (day 14) as well as sections of the mouse dorsal skin were excised from the skin-fold chamber and fixed in 10% buffered-formalin overnight. Representative full thickness blocks taken at the site of the lung implant were paraffin-embedded and sectioned (3 to 5 µm). Hematoxylin and eosin (H&E) staining of tissue sections was performed to identify the different structures and confirm the presence of lung cellular architecture.
Effect of Endothelin-1 (ET-1) on Vasoconstriction of the Lung Microvasculature
Mice (n = 22) with transplanted lung allografts (day 9 to 14 post-transplantation) were initially administered with FITC-dextran 500,000 to ensure complete revascularization and establishment of blood flow determined by IVM as described earlier. Video images of the LMV were recorded. Subsequently, the mice were removed from the microscope and the coverslip of the skin-fold chamber carefully removed. The lung allografts in the skin-fold chamber were superfused (50 µl) topically with ET-1 (Sigma, 1.0 µmol/L in PBS containing 0.1% acetic acid) or buffer alone (control). After 1 minute of application, the superfused buffer was replaced with saline solution and the coverslip replaced. Each mouse was immediately transferred to the microscope stage and the effect of ET-1 application on vasoconstriction of LMV was determined over a 60-minute time period. All observations were continuously video recorded and changes in the diameters of individual LMV (n = 88) were measured (µm) from recorded video images before and after superfusion with ET-1.
Effect of TNF- on Leukocyte-Endothelial Interactions in the Lung Allograft Microvessels
Completely revascularized lung allografts in recipient nude mice (day 9 to 14) with well-established blood flow were selected for these studies. The coverslip from the skin-fold chamber was removed and the lung allograft was superfused with TNF- (50 µl at 1 µg/ml) or PBS as a control. The ability of TNF- to induce rolling or adhesion of circulating leukocytes labeled in vivo with acridine orange (i.v. administration at 2 mg/kg body weight) in the LMV was determined by IVM 4 to 6 hours after cytokine stimulation. The interaction of labeled leukocytes in the LMV of the skin-fold chamber (ie, rolling and adhesion) was analyzed by off-line analysis of recorded video images as previously described.8,9,12
Leukocytes visibly interacting with the lung microvascular endothelium and passing at a slower rate than the main blood stream were considered as rolling cells and were quantitated by manually counting the total number of rolling cells passing through a reference point in a vessel segment. The number of rolling cells was expressed as rollers/minute, which was a percentage of the total number of cells passing through the same reference point. Adherent cells were defined as those cells remaining stationary for at least 1 minute and expressed as the number of adherent cells/250-µm length of blood vessel. In certain experiments, the effect of mAbs against P-selectin (mAb 5H1), E-selectin (mAb 9A9), and L-selectin (mAb MEL-14) (all obtained from Dr. Barry Wolitzky, MitoKor, San Diego, CA), as well as 4 (mAb PS/2) (kindly provided by Dr. Eugene Butcher, Stanford University, Palo Alto, CA) on leukocyte rolling and adhesion in LMV was investigated by administering 60 µg of the mAb i.v. immediately before microscopic observations (4 to 6 hours after TNF- stimulation) as previously described in this laboratory.8,9
A species- and isotype-matched mAb was used as the control.
In addition, leukocyte interaction in the capillaries of the LMV (segments and junctions), and the effect of transition through these vessels on shape and form of leukocytes was also examined during off-line analysis of recorded video images. Interacting leukocytes in capillaries were expressed as a percentage of the total cells trafficking per vessel.
Statistics
Comparison of vascular density and leukocyte interactions in the LMV of allografts before and after treatment was analyzed by Student’s t-test using a statistical software package (SigmaStat; Jandel Scientific, San Rafael, CA). Results are expressed as mean ± SEM or SD as indicated, and P values of <0.05 were considered statistically significant.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
CMTMR-labeled lung tissue sections (1 to 5 mm2) excised from the lungs of neonatal mice were carefully implanted adjacent to each other within the dorsal skin-fold window chamber of an anesthetized nude mouse (Figure 1)
. The mice were then allowed to recover and all subsequent observations of the lung allograft were made in unanesthetized mice. The process of revascularization of the transplanted lung allograft visualized by fluorescence IVM in recipient nude mice is shown in Figure 2
. On day 0, the implanted lung allograft initially appeared as a smooth tissue comprised of a cluster of refractile cells. At this stage, no vascular networks were evident in the pulmonary tissue except for regions of occasional redness (Figure 2A)
. The ability of transplanted lung allografts to establish blood flow was visualized by administration of FITC-dextran 500,000. Continued and progressive changes in establishment of blood flow were observed in the transplanted lung allograft over a 9-day period. FITC-dextran failed to circulate through the lung allograft between days 1 to 3, but fluorescence was evident in the surrounding host systemic vessels within the vasculature of the skin-fold chamber indicating normal blood flow (Figure 2B)
. From day 4 to 6 post-implantation, the lung allograft (
50% revascularized) demonstrated increased blood flow (Figure 2C)
. By day 9, administration of FITC-dextran revealed that all vessels in the lung allograft had revascularized and established blood flow with no evidence of visual thrombosis in any of the LMV (Figure 2D)
. Based on the direction of blood flow, the revascularized LMV could be clearly recognized as being comprised of distinct arterioles, capillaries, as well as postcapillary venules. The lung allograft establishes blood flow by initiating contact with the host vessels (Figure 3)
. By day 3, the blood vessels in the periphery of the lung allograft, visualized by plasma enhancement with FITC-dextran, appear to make contact by extending into the host tissue. The white background in Figure 3
(panels A and B) represents blood flow in the recipient cutaneous vessels, but not in the donor lung allograft vessels, clearly demonstrating that the process of revascularization is initiated by the extension of blood vessels from the lung allograft into the host tissue. Complete revascularization of the lung allograft was achieved by day 9 by the establishment of connections between the host blood vessels and vessels of the lung allograft (Figure 4, A and B)
.
|
|
|
|
Microscopic examination of H&E-stained cross-sections of the transplanted mouse lung allograft after 14 days of revascularization within the dorsal skin-fold chamber revealed structures characteristic of normal lung tissue (Figure 5)
. In comparison to a section of subcutaneous tissue (without the implanted lung tissue) obtained from the skin-fold chamber (Figure 5A)
, the section of the revascularized lung tissue demonstrated the presence of a pulmonary artery filled with red blood cells close to the bronchus (Figure 5, B and C)
. The presence of bronchi and bronchoepithelial cells lining the bronchus with protruding cilia is clearly evident and is characteristic of lung tissue (Figure 5D)
. While the alveolar tissue and lymphatics appear collapsed and the lung parenchyma compressed, the overall lung allograft was devoid of any non-specific inflammation (no neutrophil influx in the tissue or pulmonary vessels) and retained the vascular and cellular integrity as observed for normal lung tissue.
|
Evaluation of the circulatory network of the revascularized murine lung allograft revealed a clear pattern of arteriole, capillary and venous blood supply. The average diameter of the arterioles was 22.7 ± 7.8 µm, while that of the venules was 26.4 ± 13.1 µm. Based on the venular diameter, several different orders of venules were evident in the lung microcirculation of the allograft: 12.5 ± 1.4 µm, 29.7 ± 0.8 µm, and 58.1 ± 4.9 µm. Microhemodynamic parameters such as blood flow, erythrocyte velocities as well as shear rates were measured in the revascularized LMV (Table 1)
. FITC-labeled murine erythrocytes were observed to traverse through the entire vascular network of the lung allograft without any sign of retention. The velocity of erythrocytes in arterioles and venules was 609 ± 354 and 977 ± 921 µm/sec, respectively. The wall shear rates in unstimulated lung allograft microvessels ranged from 147 ± 95 to 202 ± 163 seconds-1 and the velocity of labeled leukocytes was 605 ± 334 and 789 ± 473 µm/sec in arterioles and venules, respectively. The values of the microhemodynamic parameters observed in LMV of transplanted mouse lung allografts (Table 1)
were very similar to those previously described for rabbit pulmonary microcirculation using the thoracic window model.1
|
The ability of ET-1 to induce vasoconstriction of revascularized LMV within the allograft was determined to test whether the lung allograft was responsive to external stimuli. Topical superfusion with ET-1 at a concentration of 1.0 µmol/L resulted in an immediate (<5 minutes) vasoconstriction of LMV at multiple sites (Figure 6)
. A concentration of 1 µmol/L was determined to be optimal and increasing the concentration of ET-1 to 10 µmol/L did not result in a further increase in vasoconstriction (data not shown). Superfusion with ET-1 resulted in significant constriction of blood vessels by 66.7 ± 11.5% of the original diameter. There was an overall constriction of the entire blood vessel although the extent of constriction varied at different regions of the blood vessel. Values reported correspond to measurements made at the region where the blood vessel was maximally constricted. In contrast, superfusion with buffer alone (control) had no effect (Figure 6)
. This effect lasted for a period of 45 minutes, after which most of the constricted vessels dilated to their original diameters.
|
Stimulates Leukocyte-Endothelial Interactions in Arterioles and Postcapillary Venules of the Lung Allograft
We examined the ability of circulating leukocytes labeled in vivo with acridine orange to interact with normal and inflamed endothelial cells lining the revascularized arterioles and venules of the lung allograft as well as their transit through the capillary bed. In the absence of cytokine stimulation, few leukocytes were observed to roll or adhere in arterioles or in venules. The velocity of rolling leukocytes in non-inflamed arterioles (36.7 ± 33.2 µm/sec) was appreciably higher than that observed in the venules (16.5 ± 3.8 µm/sec) (Table 1)
. TNF- stimulation of the lung allograft resulted in a significant increase in leukocyte rolling and adhesion in the venules of the lung allograft compared to the control diluent (18.3 ± 2.9 vs. 6.2 ± 3.9 rolling leukocytes/minute, P < 0.05; 5.9 ± 2.1 vs. 1.2 ± 0.5 adherent leukocytes/250 µm length of blood vessel, P < 0.01). TNF- also induced a significant increase in leukocyte rolling and adhesion in arterioles compared to that observed with the control (17.7 ± 6.7 vs. 2.6 ± 1.1 rolling leukocytes/minute, P < 0.01; 3.2 ± 0.6 vs. 0.6 ± 0.2 adherent leukocytes/250 µm length of blood vessel, P < 0.05) (Figure 7, A and B)
. The velocity of rolling leukocytes in TNF--stimulated arterioles (25.5 ± 20.7 µm/sec) was greater than the velocity of rolling leukocytes in venules (9.5 ± 5.5 µm/sec) (Table 1)
. In the absence of a chemoattractant, the adherent leukocytes failed to transmigrate across the TNF--stimulated LMV.
|
Discernible deformation of leukocytes was evident in the capillaries of the lung allograft. Deformed leukocytes were not evident in the arterioles of the lung allograft suggesting that leukocyte deformation occurs after the leukocytes had traversed the arterioles to enter the capillary microcirculation. The fraction of leukocytes that squeezed out from the capillaries were found to flow into postcapillary venules of the lung allograft where they rolled as either spherical or elongated cells (Figure 8)
. However, several leukocytes failed to dislodge from the lung allograft capillaries and remained trapped as spherical or elongated cells for extended periods of time at the junctions (32.1 ± 6.8% of total leukocytes) or within capillary segments (34.5 ± 7.8% of total leukocytes). However, there was no statistically significant difference in the number of deformed (or spherical) leukocytes retained in TNF- stimulated capillaries versus unstimulated capillaries of lung allografts (38.8 ± 14.4 vs. 51.6 ± 18% of total leukocytes, P > 0.05).
|
Since TNF- induced significant leukocyte rolling and adhesion in LMV, we examined if these adhesive interactions in the LMV were dependent on the engagement of one or more of the known leukocyte (L-selectin and 4 integrins) and endothelial rolling receptors (including E- and P-selectins) (Figure 9)
. Antibody blockade studies revealed that treatment of mice with antibodies directed against leukocyte-expressed L-selectin significantly inhibited leukocyte rolling in postcapillary venules (approximately 40% vs. control). Similarly, mAb against P-selectin significantly inhibited leukocyte rolling in venules (approximately 54% vs. control), whereas treatment with anti-E-selectin mAbs alone did not significantly influence leukocyte rolling (approximately 9% vs. control) (Figure 9A)
. However, combination treatment with anti-E-selectin and anti-P-selectin mAb resulted in near complete inhibition of leukocyte rolling (approximately 97% vs. control) (Figure 9B)
, suggesting that E-selectin contributes significantly to leukocyte rolling in combination with P-selectin, conceivably by strengthening P-selectin-initiated leukocyte rolling in LMV. In a limited number of arterioles that were examined, combined treatment with anti-P- and E-selectin resulted in a significant reduction in rolling but anti-E-selectin alone failed to significantly alter the number of rolling cells (data not shown).
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
. By day 3, the blood vessels in the periphery of the lung allograft appear to make contact by extending into the host tissue. These studies suggest that the revascularized blood vessels most likely originate from transplanted allograft. The lung allografts exhibit a well-defined network of LMV within 5 to 9 days, which could be classified as terminal arterioles, capillaries, as well as first, second, and third order venules based on their branching patterns. The LMV also established a directed, continuous blood flow, ie, arterioles to venules via capillaries. Further, the re-established microvessels were found to respond to mediators of pulmonary vasoconstriction, such as ET-1, and to cytokine activation resulting in leukocyte-endothelial adhesive interactions. While it conceivable that the use of nude mice might potentially compromise inducible effects of extravascular stimuli, this possibility appears to be remote based on our observation of ET-1-induced vasoconstriction and TNF--induced leukocyte rolling and adhesion in the revascularized LMV. A fraction of the leukocytes transiting the capillaries were found to undergo a shape change. Using this experimental model, the implanted lung allografts, which have remained viable for several weeks and demonstrate characteristic features observed in normal lung tissue allow visualization of lung blood vessels in acute as well as chronic experimental conditions by IVM. Previous studies with transplanted hamster neonatal pulmonary allografts have revealed that revascularized pulmonary microvessels retain microcirculatory functions of the adult lung in the hamster cheek pouch.3-6,13 Our studies extend these observations by using IVM to demonstrate that microvessels in murine lung allografts, even though transplanted in immune-compromised mice, have the ability to respond to external stimuli such as cytokine stimulation (to induce leukocyte-endothelial cell adhesion) and vaso-active mediators (to induce vasoconstriction). In addition, the shear rates we observed in the lung allografts are characteristic of the shear rates noted in the pulmonary circulation of the lung as demonstrated by other investigators.1,2
Elucidation of mechanisms by which circulating leukocytes interact with LMV and are subsequently recruited in the lung during episodes of lung inflammation requires the visualization of such interactions in vivo. Using IVM to visualize leukocytes in the systemic circulation has allowed the demonstration of the importance of adhesion molecules in mediating an initial, but reversible, adhesion (rolling) of circulating leukocytes (including eosinophils) to vascular endothelium.14-16
In response to activation-dependent stimulation, rolling leukocytes are known to subsequently adhere firmly and emigrate into extravascular sites of inflammation.14,15
This paradigm of leukocyte adhesion involving sequential rolling, adhesion, and transmigration has thus far not been examined in the lung microcirculation of mice. Data from the present study suggest that leukocytes can roll and adhere efficiently in TNF--stimulated microvessels of the lung allograft. This increase in leukocyte rolling and adhesion in the lung allograft microvessels was observed during TNF-induced inflammation. Studies with neutralizing antibodies demonstrate that leukocyte rolling in cytokine-stimulated postcapillary venules is mediated by both leukocyte (L-selectin) and endothelial (P-selectin alone and in combination with E-selectin) selectins, with smaller contributions from leukocyte 4 integrins. Our IVM studies of the murine lung allograft support the concept that the multi-step paradigm of leukocyte adhesion, well documented in the systemic circulation, is also evident in the lung microcirculation, although leukocyte rolling and adhesion occurs in both arterioles and postcapillary venules of the lung microcirculation, contrary to the systemic circulation where rolling is observed only in postcapillary venules. Additional support for the multi-step leukocyte-endothelial adhesion cascade occurring in the lung microcirculation is derived from some, but not all, animal models used to visualize the pulmonary microcirculation. For instance, in rabbits, leukocyte rolling and adhesion has been demonstrated in pulmonary venules and arterioles of normal lungs using a thoracic window.1,2
In a canine study, which also used a thoracic window, leukocytes rolled but failed to adhere to pulmonary venules or arterioles with transient leukocyte arrest being observed in pulmonary capillaries,17
while leukocyte sequestration was noted in the pulmonary capillaries but failed to interact in either arterioles or postcapillary venules in a contrasting study.18,19
More recent studies in rats using isolated, reperfused lungs demonstrated that leukocytes roll more appreciably along pulmonary arteriolar walls than in venules.20
In contrast to rolling, leukocytes failed to adhere to the pulmonary venules or arterioles, however transient and sustained leukocyte arrest was observed in the pulmonary capillaries.20
In the currently described mouse lung allograft model, low levels of spontaneous leukocyte rolling and adhesion is evident in both venules and arterioles, which could be up-regulated by stimuli such as cytokines, making the lung allograft an useful model to study the regulation of leukocyte adhesion in the lung.
The mouse lung allograft model can also be used to study whether leukocytes undergo intravascular shape change as they traffic through the capillary bed. Our studies demonstrate that a fraction of the leukocytes do indeed undergo shape change (deform into stretched or elongated forms) as they transit the capillary bed, and roll as either elongated or spherical (cells that have not undergone shape change) cells in the postcapillary venules. At the concentration tested, TNF- failed to augment the number of leukocytes trapped in the capillaries, however, it is likely that a higher concentration of TNF- or the presence of other cytokines could potentially result in increased capillary retention in LMV. Recent IVM studies with both animal and human lungs have shown that fully dilated capillary segments have an average diameter of 7.5 ± 2.3 µm and an average length of 14.4 ± 5.8 µm21
similar to what we have observed in the mouse lung allograft. Leukocytes traversing the lung capillaries cross between 45 to 90 capillary segments,22,23
which would provide them with multiple opportunities for encountering capillaries that are smaller than their own diameter.21
Electron microscopy studies have demonstrated that due to the smaller size of the capillaries in comparison to the traversing leukocyte populations, intravascular leukocytes undergo shape change and deform into elongated cells.21
The ability of intravascular leukocytes to undergo shape change (deform/elongate) may prevent them from being trapped during their transit across the capillary bed.24
Furthermore, IVM studies using a canine thoracic window model revealed that half of the polymorphonuclear leukocytes (PMNs) pass through the capillaries without stopping, whereas the other half stops one or more times.18,19,25,26
Leukocytes, on the other hand, have been observed to deform during transit through the canine capillary bed.17
In addition, leukocytes stop more frequently in canine microvessel segments than in junctions as well as roll along the walls of canine arterioles and venules. Interestingly, most of the leukocytes rolling in canine arterioles were spherical, while leukocytes passing through canine capillaries deformed into elongated shapes and continued to roll in the canine venules as elongated cells, similar to what we have noted with mouse leukocytes in the murine lung allograft microvessels. IVM studies of microvessels in the subpleural surface of the rabbit and canine lungs, as well as our studies in mouse lung tissue, have revealed that leukocytes in all three species deform during their transit through the capillary bed with discernible leukocyte rolling in venules and arterioles.17-19,25,26
The murine transplanted lung allograft model offers several advantages compared to the rabbit or canine thoracic window model or the ex vivo observation of isolated rat perfused lungs. For example, in the mouse lung allograft, direct visualization of leukocytes in the lung circulation can be made continuously and is not limited by the movement or shape of the lungs during respiration or ventilation. As genetically engineered mice deficient in adhesion molecules and reagents for studying leukocyte-endothelial interactions are readily available in the mouse, it is important to develop methods to visualize the lung circulation in mice. While implantation of a thoracic window can be readily accomplished in larger animals such as dogs, rabbits, and rats, such techniques have not as yet been described in mice necessitating new approaches to visualizing the lung vasculature in mice. To this effect, histological examination of the revascularized lung allograft demonstrated cellular and morphological features characteristic of normal murine lung. While the lung allograft model has been developed primarily with a view to study leukocyte-endothelial cell interactions in LMV, certain aspects of the physiological environment are likely to be different between the transplanted lung allograft and intact lungs. For example, the pulmonary artery normally receives deoxygenated blood at low systolic pressure, whereas arterioles in lung grafts appear to receive oxygenated blood at prevalent systemic pressures. Also, the tissue environment with regard to ambient oxygen tension and mechanical stress associated with respiration is likely to be different in the lung allograft and intact lungs. Nonetheless, leukocyte-endothelial cell interactions observed in the LMV of the allograft (such as rolling in arterioles, capillary trapping, etc) are in contrast to those observed in the surrounding host vessels in the skin-fold chambers, suggesting that the lung allograft retains several of the hemodynamic features of the intact lung rather than the surrounding systemic microcirculation. Since macrophages and mast cells play an important role during inflammatory events by releasing various mediators that modify local endothelial properties, their involvement in the transplanted allograft needs further investigation. This model is unique in that it enables the simultaneous quantification of microhemodynamics and leukocyte kinetics in murine lung arterioles, capillaries, and venules using fluorescence IVM and in vivo labeling of leukocytes.
In summary, our studies have revealed that local lung inflammation induced by TNF- results in increased leukocyte rolling and adhesion in both arterioles and venules of lung allografts transplanted into the dorsal skin-fold chamber of nude mice. This lung allograft model using neonatal or adult lungs will facilitate identifying mechanisms of leukocyte-endothelial adhesion in LMV, shape change of leukocytes in capillary beds, as well as provide a tool to study pulmonary vascular responses to vaso-active mediators by direct visual observations using IVM. Using this model it will be possible to determine how individual subsets of leukocytes, including eosinophils, neutrophils, monocytes, and T-lymphocytes, transit through microvessels during episodes of lung inflammation in mice. Furthermore, this mouse model has significant advantages over previously described animal models such as dogs, rabbits, and rats in that the effects of specific gene deletions (knockout mice lacking adhesion molecules, cytokines, chemokines, etc) can readily be examined.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by California Tobacco-Related Disease Research program grants 7RT-0197, 10 RT-0171, and National Institutes of Health grant AI 35796 to P.S.
Accepted for publication February 13, 2003.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  L. P. Veliz, F. G. Gonzalez, B. R. Duling, J. C. Saez, and M. P. Boric Functional role of gap junctions in cytokine-induced leukocyte adhesion to endothelium in vivo Am J Physiol Heart Circ Physiol, September 1, 2008; 295(3): H1056 - H1066. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. L. Patrick, J. Rullo, S. Beaudin, P. Liaw, and A. E. Fox-Robichaud Hepatic leukocyte recruitment in response to time-limited expression of TNF-{alpha} and IL-1beta Am J Physiol Gastrointest Liver Physiol, October 1, 2007; 293(4): G663 - G672. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. M. Kuebler, K. Parthasarathi, J. Lindert, and J. Bhattacharya Real-time lung microscopy J Appl Physiol, March 1, 2007; 102(3): 1255 - 1264. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Reutershan, A. Basit, E. V. Galkina, and K. Ley Sequential recruitment of neutrophils into lung and bronchoalveolar lavage fluid in LPS-induced acute lung injury Am J Physiol Lung Cell Mol Physiol, November 1, 2005; 289(5): L807 - L815. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Sikora, S. P. Rao, and P. Sriramarao Selectin-dependent rolling and adhesion of leukocytes in nicotine-exposed microvessels of lung allografts Am J Physiol Lung Cell Mol Physiol, September 1, 2003; 285(3): L654 - L663. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
