| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Medicine, Johns Hopkins University, Baltimore, Maryland
Submitted 9 January 2007 ; accepted in final form 31 May 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
angiogenesis; chemokines; macrophage inflammatory protein-2
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
LPAL. Our protocol was approved by the Johns Hopkins Animal Care and Use Committee. Male mice, aged 5–6 wk, of several strains were studied: C57Bl/6 mice (total studied, n = 38 mice) and BALB/c mice (n = 24 mice) purchased from Charles River (Wilmington, MA) and CXCR2 (–/–)-deficient mice [knockout (KO)] on a BALB/c background (n = 16 mice, Jackson Laboratories, Bar Harbor, ME). Mice were anesthetized (2% isoflurane in oxygen), intubated, and ventilated at 120 breath/min, 0.2 ml/breath with the anesthetic-gas mixture. As previously described, this model included: left lateral thoracotomy, left pulmonary artery ligation (6–0 silk suture), and closure of the thoracotomy with suture while the animal was placed on positive end-expiratory pressure [1 cmH2O; (14, 15, 22, 27)]. Lidocaine (2%) was applied for analgesia and the skin incision was closed using methyl acrylamide adhesive. The animal was removed from the ventilator, extubated, and allowed to recover.
Neovascularization assessed by blood flow. To determine the extent of neovascularization, systemic blood flow to the left lung was measured after LPAL in mice using radiolabeled microspheres. Diethylenetriamine pentaacetic acid (DTPA)-coated polystyrene microspheres (10 µm diameter; Kisker Products, Steinfurt, Germany) were bound to Technetium-99m radioligand (Cardinal Health; Dublin, OH). Based on previous observations, we made blood flow measurements 14 days after LPAL when a systemic neovasculature arising from intercostal arteries was clearly established (27). Mice were anesthetized, ventilated as described above, the common carotid artery was cannulated (PE-10), and the tip was advanced to a point just within the thorax. Micropheres (150,000; stock = 1.5 million spheres/ml) were infused (0.1 ml at 0.04 ml/min: Harvard Apparatus, Holliston, MA). Mice were killed by exsanguination, and the left lung was excised. Gamma emissions from lodged radiolabeled microspheres were immediately counted in the Hidex Triathler (Bioscan, Washington, DC). Left lung activity was normalized to whole body activity counted in a Capintec counter (Capintec Products, Ramsey, NJ), which had been calibrated to the Bioscan instrument. Systemic perfusion of the left lung was expressed as percent of total measured blood flow (carcass + all organs), i.e., percent of cardiac output.
Experimental protocols. A total of 78 mice were studied, applying three experimental protocols: blood flow assessment, lung MIP-2 protein determination, and endothelial cell harvest and culture for measurement of CXCR1/CXCR2 expression.
Blood flow groups. Neovascularization was assessed by blood flow determination in eight experimental groups of mice (see Table 1). Systemic blood flow to the left lung 14 days after LPAL was determined in control group 1: C57Bl/6 mice; group 2: C57Bl/6 mice administered a CXCR2-neutralizing antibody extensively tested and kindly provided by Dr. Robert Strieter, University of Virginia. The treatment protocol for this group required intraperitoneal delivery of the neutralizing antibody (0.5 ml ip goat anti-mouse serum) 24 h prior to LPAL as well as immediately after surgery and every 48 h after LPAL until blood flow determination at 14 days. This treatment protocol was based on previous studies examining angiogenesis in mouse models of lung cancer (11, 30). A serum control group 3: C57Bl/6 mice treated with intraperitoneal injection of goat serum (Sigma-Aldrich, St. Louis, MO) since the neutralizing antibody was raised in goats and using the same volume and time points as group 2. This group served as a control for the CXCR2-neutralizing antibody.
|
Group 8: C57Bl/6 mice were evaluated 2 days after LPAL and served as an experimental control group since we showed previously that a functional neovasculature does not develop until 5 days after LPAL. Consequently, blood flow measured in this group represents an experimental background.
MIP-2 protein evaluation. Changes in lung MIP-2 protein were evaluated in additional mice 4 h after LPAL. Lung samples were collected from naive C57Bl/6 mice (n = 3) and C57Bl/6 mice after LPAL (n = 6), with CXCR2-neutralizing antibody (n = 4), goat serum control (n = 4), and CXCR2 KO (n = 5). Four hours after LPAL, anesthetized mice were killed (cervical dislocation), and the upper third of the left lung and the right lung were dissected and excised. Lung samples were weighed, homogenized (Fastprep Bio101; Thermo Savant, Holbrook, NY), and aliquoted for ELISA and BCA assays. The samples were processed according to the Quantikine MIP-2 ELISA kit (R&D Systems, Minneapolis, MN). Total protein measurements were made according to the BCA protein assay kit (Pierce, Rockford, IL).
Isolation of mouse aortic endothelial cells. The aortas from C57Bl/6, BALB/c, and CXCR2 KO mice (n = 3 mice/strain) were dissected and placed with the intima side down on Matrigel-coated 35 mm tissue culture dishes. After 4–6 days, endothelial cells that had migrated were replated to a T-25 flask coated with gelatin. Cells were cultured in DMEM supplemented with 20% FCS, 15 µg/ml ECGS, 100 µg/ml penicillin-streptomycin, 0.25 µg/ml amphotericin B, and 0.1 mM MEM with nonessential amino acids. Endothelial cell phenotypes were confirmed by immunostaining techniques for platelet endothelial cell adhesion molecule (PECAM), von Willebrand factor (vWF), and uptake of acetylated low-density lipoprotein (Dil-ac-LDL). Only cell populations with positive staining were used for further experiments. All experiments were carried out using endothelial cells between passages 2 and 8.
Endothelial cell mRNA CXCR1/CXCR2 expression by quantitative real-time RT-PCR.
To confirm the presence of CXCR1 and CXCR2 mRNA in aortic endothelial cells, we performed quantitative real-time RT-PCR. Total RNA was extracted from endothelial cells (5 x 106 cells) using RNeasy Mini Kit (Qiagen, Valencia, CA) with standard procedures. Total RNA was also extracted from lung tissue as a positive control. Total RNA (2 µg) from aortic endothelial cells (n = 3 isolations) was used for first-strand cDNA synthesis with random hexamer primers and a first-strand cDNA synthesis kit (Amersham Biosciences, Piscataway, NJ) according to the manufacturer's protocol. The primers used for amplification were as follows, for CXCR1: CCG TCA TGG ATG TCT ACG TG and CAG CAG CAG GAT ACC ACT GA; and for CXCR2: GGT GGG GAG TTC GTG TAG AA and CGA GGT GCT AGG ATT TGA GC. PCR reactions were performed with QuantiTect SYBR Green PCR Master Mix (Qiagen, Valencia, CA) and IQ5 Multicolor real-time PCR Detection System (Bio-Rad Laboratories, CA), using 1 µl of cDNA as the template in each 25-µl reaction mixture. The absolute copy number of mRNA of interest was determined by interpolation of the standard curve with the threshold cycle value of each sample. The melting curve protocol was performed following the RT-PCR and revealed a single clean melting peak for all samples tested. Data were normalized to the quantity of 
-actin mRNA in individual samples.
Endothelial cell CXCR1/CXCR2 expression by flow cytometry. Endothelial cells were washed, detached (2 mM EDTA/PBS), and resuspended (PBS with 0.2% BSA). After washing, cells were incubated with primary human anti-CXCR1 (0.2 µg/ml) and anti-CXCR2 (1.0 µg/ml) monoclonal antibodies (Abcam, Cambridge, MA), washed, incubated with Alexa-488 mouse anti-human antibody (Molecular Probes, Eugene, OR), and analyzed. Mouse isotype IgG served as a negative control. Since antibodies against mouse CXCR1 are not commercially available, we used antibodies raised against human CXCR1 and CXCR2. Monoclonal antibody against human CXCR1 shows no cross-reactivity with human CXCR2, and CXCR2 antibody does not cross-react with human CXCR1. Human CXCR2 shares 71% identical amino acids with mouse CXCR2 and human CXCR1 shares 65% identity with mouse CXCR1. Cells were evaluated for CXCR1/CXCR2 surface expression using the FACS Calibur (Becton Dickinson, Franklin Lakes, NJ).
Statistics.
All data are presented as the means ± SE. Primary outcome variables (blood flow, MIP-2 protein, and CXC receptor intensity) were not normally distributed (Shapiro-Wilk test for normality) and, consequently, the data were log transformed to assume a normal distribution. Changes in blood flow, MIP-2 protein, and CXC receptor intensity were evaluated by one-way ANOVA. Relevant within-group comparisons were made using either Student's t-test for unpaired data (blood flow) or multiple comparisons using Fisher's test for least significant differences (left lung/right lung protein, receptor fluorescence intensity). A P value 
0.05 was accepted as significant.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Results of the effects of CXCR2 inhibition on systemic neovascularization of the lung after LPAL are presented in Fig. 1, A–C. Overall, neovascularization was significantly different among groups (P = 0.019). Left lung blood flow in control group 1 C57Bl/6 mice was comparable to our previously published values and averaged 2.1 ± 0.5% of total blood flow or cardiac output (14). Control serum-treated mice (group 3; 2.2 ± 0. 5%) did not differ from mice without treatment (group 1), 14 days after LPAL. However, as shown in Fig. 1A, mice treated with the CXCR2-neutralizing antibody demonstrated a significant reduction in blood flow to the left lung 14 days after LPAL (0.7 ± 0.3%; P = 0.025). To determine an experimental zero flow in group 8, mice (n = 5) were evaluated 2 days after LPAL when we previously established there was no neovasculature (15, 28). The values obtained for these control experiments (0.20 ± 0.06%) are considerably lower than after antibody treatment. Thus the neutralizing antibody, although effectively decreasing blood flow, did not completely inhibit neovascularization.
|
Figure 1C shows the results of BALB/c mice after LPAL that had been treated with control serum (group 7) or the CXCR2 neutralizing antibody (group 6). Similar to what had been seen in the C57Bl/6 mice, the neutralizing antibody significantly reduced angiogenesis as determined by blood flow to the left lung after LPAL (P = 0.026).
A comparison of the results of Fig. 1A with Fig. 1C suggests an overall strain-dependent difference in response to LPAL. Since within C57Bl/6 mice, control group 1 was not different from serum control group 3 (P = 0.87) and within BALB/c mice, control group 4 was not different from serum control group 7 (P = 0.97), the control groups of each strain were combined to form larger groups. A comparison of the combined group of C57Bl/6 controls (groups 1 and 3) with the combined BALB/c control groups (groups 4 and 7) demonstrates that neovascularization assessed by blood flow was significantly greater in the C57Bl/6 mice compared with the BALB/c mice, thereby suggesting a strain-dependent difference in responsiveness (P = 0.030).
MIP-2 Protein
To confirm that the marked decrease in vascularization was not related to decreased ligand production, we measured MIP-2 levels in lung homogenates 4 h after LPAL. Absolute MIP-2 protein harvested from left and right lungs of naive C57Bl/6 mice was very low (0.2 ± 0.1 pg/mg of total lung protein). As shown previously and confirmed here, MIP-2 protein in left lungs after LPAL (group 1; 156 ± 19 pg/mg of total lung protein) was significantly greater than paired right lungs (47 ± 10 pg/mg of total lung protein; P = 0.003). Left lung MIP-2 in groups 3, 2, and 5 averaged 94 ± 25, 208 ± 143, and 81 ± 33 pg/mg, respectively (P > 0.05). Figure 2 shows the group results of average fold change in MIP-2 protein in the left lung/total protein compared with right lung/total protein in C57Bl/6 mice and the CXCR2 KO mice. An average two- to fourfold increase in MIP-2 protein in the left lung of all experimental groups after LPAL is shown compared with paired right lungs. These changes were statistically indistinguishable across the experimental groups (P = 0.41).
|
We first confirmed the presence of both CXCR1 and CXCR2 mRNA in aortic endothelial cells. Copy numbers of CXCR1 mRNA normalized to -actin averaged 1.08–3 ± 0.25–3. In these aortic endothelial cells from C57Bl/6 mice, CXCR2 mRNA expression was significantly less than CXCR1 expression, averaging 3.3–4 ± 0.075–4 (P < 0.05). Thus CXCR1 showed an average 3.4-fold greater mRNA expression than CXCR2. To explore one potential explanation for the lack of inhibition of blood flow in the CXCR2 KO mice, cell surface receptor expression was evaluated in primary culture of C57Bl/6, BALB/c, and CXCR2 KO mouse aortic endothelial cells. Figure 3 shows the results of CXCR1 and CXCR2 expression. In mouse aortic endothelial cells cultured from each different group, both CXCR1 and CXCR2 expression was evident and showed differences across groups (P < 0.0001). In endothelial cells from C57Bl/6 mice and CXCR2 KO mice, expression of CXCR1 was significantly greater than CXCR2 (P < 0.01). Although a similar average result was observed in cells from BALB/c mice, the difference failed to reach statistical significance (P = 0.058). Average CXCR1 expression was increased in cells from CXCR2 KO mice compared with expression in background control BALB/c cells (P < 0.01). As expected, CXCR2 expression was very low in the CXCR2 KO (P < 0.01 compared with BALB/c).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
We followed an antibody treatment protocol extensively validated and previously shown to be effective in blocking angiogenesis in mouse models of bronchiolitis obliterans and non-small cell lung cancer (1, 2, 3). The positive response observed as an attenuation in blood flow confirms that the neutralizing antibody reached the site of proliferating intercostal blood vessels and was effective. However, the observation that the neutralizing antibody did not completely inhibit neovascularization could be related to inadequate dosing in this in vivo model as well as the likelihood that other growth factors contribute to the complex process of angiogenesis.
An important series of control experiments was to confirm that proangiogenic agonists were not decreased due to treatment protocols. We measured MIP-2 protein levels in the left lung 4 h after LPAL as representative of the CXC chemokines since this protein was previously shown to be the most highly expressed of the ELR+ CXC chemokines after LPAL (22). We again confirmed increased left lung/right protein expression of MIP-2 and the fold change did not vary among treatment groups (Fig. 2). These results suggested that the levels of CXC ligand, as represented by MIP-2, did not contribute to the reduction in blood flow observed after neutralizing antibody treatment. Additionally, the level of MIP-2 was not different in the C57Bl/6 mice compared with the BALB/c CXCR2 KO, suggesting differences in ligand production were not responsible for strain-dependent differences in angiogenesis. Thus we conclude that the neutralizing antibody reached recovering intercostal arteries and prevented neovascularization despite a proliferative/migratory stimulus from lung chemokines.
Given the results using the CXCR2-neutralizing antibody, we were perplexed by the lack of attenuation in angiogenesis in the CXCR2-deficient mice as assessed by blood flow 14 days after LPAL. The overall hypothesis tested in this study was that CXCR2 is essential for maximal angiogenesis. Thus we fully expected to obtain similar results in antibody-treated mice and CXCR2-deficient mice. The CXCR2-deficient mice used in this study are commercially available, homozygous for the Il8rbtmMwm targeted mutation, and are raised on a BALB/c background. Previous work by Strieter and colleagues (1, 11) showed consistent inhibitory results with both CXCR2-neutralizing antibody and CXCR2-deficient mice. We questioned whether potential compensatory mechanisms exist in the CXCR2-deficient mice that might allow angiogenesis to take place during pulmonary ischemia, circumventing the normal process. With the assessment of the wide range of phenotypic alterations reported on these specific mice (http://jaxmice.jax.org/strain/002724.html), several factors might contribute to a compensatory mechanism in this model. Additionally, since CXCR1 has been shown in human endothelium to contribute to the proangiogenic phenotype, we questioned earlier reports that stated that rodents lack CXCR1 (18). Several recent studies show evidence for CXCR1 in mouse tissue (7, 8, 16). Therefore, we harvested endothelium from each of the mouse strains studied and evaluated CXCR1/CXCR2 receptor expression in primary culture mouse arterial endothelial cells. We first confirmed the presence of mRNA for both CXCR1 and CXCR2 using real-time RT-PCR in mouse endothelial cells from C57Bl/6 mice. CXCR1 showed an average 3.4-fold greater mRNA expression than CXCR2. Using human antibodies raised against CXCR1 and CXCR2, we further evaluated differences in cell surface expression of these two receptors. In endothelial cells from both C57Bl/6 and BALB/c mice, measurement of fluorescence intensity representing cell surface receptor levels, showed the presence of both receptors with the signal for CXCR1 predominating in all cells. Interestingly, Salcedo and colleagues (20) showed differential expression of these chemokine receptors in different types of human endothelial cell subtypes, and in human dermal microvascular endothelial cells, approximately equivalent levels of CXCR1 and CXCR2 as we report (average fluorescent intensity = 261 and 156 fluorescent units respectively). In the present study, CXCR1 was increased significantly in endothelial cells from CXCR2 KO mice compared with cells from wild-type BALB/c mice (Fig. 3). Based on these results, while acknowledging the limitations imposed by the use of human antibodies, we suggest that CXCR1 expression may have been increased in the CXCR2 KO mice, which contributed to endothelial cell signaling and sustained levels of angiogenesis. This confirmation of CXCR1 on mouse arterial endothelial cells provides a potential explanation for the lack of complete effectiveness of the CXCR2-neutralizing antibody. Previously, it was shown that this neutralizing antibody did not cross-react with human CXCR1 nor did it prevent binding of IL-8 with CXCR1 in human cells (1). A recent study by Fan and colleagues (7) suggested that another chemokine ligand of CXCR1, CXCL6, plays a critical role in angiogenesis. However, in the present model, gene array profiling did not indicate a change in this chemokine after LPAL (22). Further clarification of the proangiogenic role of both chemokine receptors is needed and definitive answers await the availability of specific antibodies targeted to mouse CXCR1 for in vivo use.
It should be noted that the neovascularization that occurs in this mouse model is different from other mammals where bronchial arterial neovascularization predominates. Because mice lack a developed subcarinal systemic vasculature, after LPAL, intercostal arteries play the major role in perfusing the ischemic lung (15). However, case reports in human subjects with chronic pulmonary embolism demonstrate growth of systemic vessels into ischemic lung from bronchial vessels as well as intercostal, mammary, and phrenic arteries (19). Thus, in mice and humans, systemic vessels in proximity to ischemic lung tissue appear to be primed to respond in situations of pulmonary ischemia. In this mouse model, we believe a chemotactic gradient is provided by growth factors within the ischemic left lung, which is directly apposed to systemic vessels within the chest wall. Although we have not specifically prevented left lung association with the thoracic wall to test this hypothesis, it is the only explanation we can provide for the growth of intercostal vessels into the ischemic lung that lacks pulmonary perfusion. As noted previously, we see no intercostal vessel attachments in sham-operated mice (thoracotomy but no LPAL).
In summary, we showed that treatment of mice with a neutralizing antibody to CXCR2 results in an attenuation of neovascularization after LPAL. We confirmed that ligand production, assessed by MIP-2 levels, was not limiting or altered despite differences in the absolute level of neovascularization. Using commercially available antibodies, we confirm the presence of both CXCR1 and CXCR2 in primary cultured mouse arterial endothelial cells. Mice with the genetic deletion of CXCR2 may have other compensatory mechanisms that allow for the essential function of new vessel growth and proliferation.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Sukkar, J. Jenkins, J. Sanchez, and E. M. Wagner Inhibition of CXCR2 attenuates bronchial angiogenesis in the ischemic rat lung J Appl Physiol, May 1, 2008; 104(5): 1470 - 1475. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. A. Abdel-Malak, C. B. Srikant, A. S. Kristof, S. A. Magder, J. A. Di Battista, and S. N. A. Hussain Angiopoietin-1 promotes endothelial cell proliferation and migration through AP-1-dependent autocrine production of interleukin-8 Blood, April 15, 2008; 111(8): 4145 - 4154. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. M. Wagner, J. Sanchez, J. Y. McClintock, J. Jenkins, and A. Moldobaeva Inflammation and ischemia-induced lung angiogenesis Am J Physiol Lung Cell Mol Physiol, February 1, 2008; 294(2): L351 - L357. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
