Abstract
Background Preterm infants exposed to O2 with mechanical ventilation often develop bronchopulmonary dysplasia (BPD), a form of chronic lung disease (CLD). The pathogenesis of BPD/CLD involves dysmorphic microvasculature and disrupted alveolarization. This may be due to impaired vascular endothelial growth factor (VEGF) and VEGF receptor expression.
Methods To examine the ontogeny of VEGF and VEGF receptors in baboon lungs from 125 to 185 (term) days gestation and to determine whether exposure to O2 and mechanical ventilation alter these ontogenic profiles, we examined lung specimens from three O2-exposed groups: (1) animals delivered at 125 days gestation and exposed to O2 for 14 days as needed; (2) animals delivered at 140 days gestation and exposed to O2 for 10 days as needed; and (3) animals delivered at 140 days gestation and exposed to 100% O2 for 10 days. Lungs from gestational age-matched controls were also examined at 125, 140, 160, 175, and 185 (term) days.
Results VEGF189 was the most abundant splice variant in the lungs at all stages of development. Extremely premature baboons developing BPD/CLD had higher lung VEGF121 messenger ribonucleic acid (mRNA) expression. However, transcripts for VEGF189, VEGF165, and VEGF receptors (Fms-like tyrosine kinase-1 [Flt-1], kinase-insert domain receptor [KDR]/fetal liver kinase-1 [Flk-1], and neuropilin 1) were suppressed in the BPD models.
Conclusions We conclude that impaired VEGF and VEGF receptor mRNA expression in lungs from extremely premature baboons developing BPD/CLD may contribute to dysmorphic microvasculature and disrupted alveolarization.
Approximately 12% of all infants born in the United States are premature,1most of whom will require O2 therapy. Extremely premature, low birth weight infants exposed to O2 and mechanical ventilation often develop lung inflammation, leading to bronchopulmonary dysplasia (BPD), a form of chronic lung disease (CLD). Despite increased survival of extremely premature infants owing to advancement in respiratory care and management, BPD/CLD remains a cause for concern. The pathogenesis of BPD/CLD is unclear, but elegant studies by Coalson and colleagues have demonstrated that normal processes involved in postnatal lung development, such as microvascular maturation and alveolarization, are disrupted, leading to poor lung development.2-7Specifically, dysmorphic microvasculature may be due to alterations in vascular endothelial growth factor (VEGF), an endothelial cell mitogen that is important for normal vascular development.8,9
Premature infants born at 24 to 26 weeks gestation are in the late canalicular stage, in which the capillaries are beginning to align.10Exposure to high levels of O2 disrupts this process. At autopsy, lung histology reveals arrested lung development, impaired alveolar and vascular growth resulting in alveolar simplification and dysmorphic capillaries, and thickened alveolar septae in infants who died from BPD.11These abnormalities appear to be associated with decreased VEGF signaling.8-15VEGF has also been shown to have a stimulatory effect on human fetal lung epithelial cells, suggesting a role in epithelial cell growth and differentiation.16-18Therefore, abnormal VEGF expression may adversely affect lung epithelial cell proliferation and angiogenesis in the developing lung.
Alternative splicing of a single VEGF gene produces five isoforms consisting of 206, 189, 165, 145, and 121 amino acids, which are differentially expressed in lung development and injury.19VEGF189 is tightly bound to the extracellular matrix (ECM) and is the predominant splice variant in the developing lung.20VEGF165 is partially diffusible because it has one heparin-binding region and mediates the most potent angiogenesis.21VEGF121 does not bind heparin, is highly diffusible, and may not play an important role in angiogenesis.22
The cellular effects of VEGF are mediated via specific receptors (VEGFR-1 or Flt-1 and VEGFR-2 or KDR/Flk-1), which are expressed by endothelial cells.23Knockout studies in mice have shown that the Flt-1 receptor mediates organization of the vasculature,24whereas KDR/Flk-1 mediates endothelial cell differentiation and proliferation.25Soluble forms of Flt-1 (sFlt-1) and KDR/Flk-1 (sKDR) exist and differ significantly from one another. sKDR does not readily bind VEGF and only partially inhibits cell migration.26In contrast, sFlt-1 has a negative regulatory role by binding VEGF with high affinity, thus making it less available to its receptors.27A third receptor, neuropilin 1 (NP-1), is variant specific for VEGF165 and acts to increase VEGF165 potency by enhancing its binding affinity to KDR/Flk-1.22The coordinate expression of VEGF, KDR/Flk-1, and NP-1 results in high vascular density and potent angiogenesis.22
Previous studies have examined the importance of VEGF in maintaining normal pulmonary vascular development10,13,17; however, there is a lack of information regarding the involvement of soluble VEGF receptors and the interaction between VEGF165 and its ligand-specific receptor, NP-1, in the setting of BPD/CLD. We therefore examined the ontogenic profiles of VEGF and VEGF receptors in baboon lungs from 125 days gestation (67% of term; equivalent to 24 to 26 weeks gestation in humans) to 185 days (term gestation in baboons). Further, we determined the effects of O2 and mechanical ventilation on VEGF and VEGF receptor levels in extremely premature baboon lungs.
MATERIALS AND METHODS
Experimental Design
All experiments were carried out according to the guidelines outlined by the American Association for Accreditation of Laboratory Animal Care. Frozen baboon lung tissue and paraffin-embedded sections were obtained from the BPD Resource Center, Southwest Foundation for Biomedical Research, San Antonio, Texas. Lung tissue samples were taken from five gestational control (GC) baboons at 125, 140, 165, 175, and 185 (term) days gestation (n = 3/group) and from three BPD baboon models (n = 3/group) delivered by hysterotomy (timed gestations were confirmed by ultrasonography). All preterm GC animals were sacrificed immediately after delivery, and the GC term animals were sacrificed 1 to 3 days after delivery. All BPD baboon models were intubated, ventilated, and cared for according to standard neonatal intensive care unit protocols, the details of which have been described.3,7,8Briefly, animals were delivered at 125 days gestation (corresponding to 24 to 26 weeks of human gestation), intubated, administered surfactant (Survanta, 100 mg/kg, Ross Abbot, Columbus, OH), and ventilated with oxygen as needed (PRN) for 14 days to maintain partial pressure of arterial oxygen (PaO2) at 55 to 70 mm Hg (125 + 14 days PRN O2 group). Total parenteral nutrition was provided at 24 hours of life, hematocrit was maintained at 45% (using adult donor baboon blood), and broad-spectrum antibiotics were administered for the first 3 days of life.
These animals were exposed to antenatal steroids (betamethasone, 6 mg) at 48 and 24 hours before delivery. The 125-day PRN O2 model closely replicates lung disease that is now seen in 24- to 26-week gestation human infants who develop BPD/CLD.7At the time of sacrifice and tissue sampling, these animals were 140 days postconceptional age and therefore were compared with the 140-day GCs.
Animals assigned to a second BPD group were delivered at 140 days gestation (corresponding to 30 to 32 weeks of human gestation), intubated, and ventilated with O2 PRN for 10 days to maintain PaO2 at 60 to 80 mm Hg (140 + 10 days PRN O2 group). Animals assigned to a third BPD group were also delivered at 140 days gestation, intubated, and ventilated with 100% oxygen for 10 days (140 + 10 days 100% O2 group). The 140-day models were not exposed to antenatal steroids or postnatal surfactant.
This model more closely replicates the “old BPD” before the use of surfactant and prenatal steroids.3At the time of sacrifice and tissue sampling, these animals were 150 days postconceptional age and were compared with the 160-day GCs.
Lung VEGF and Soluble VEGF Receptor Levels
Fresh lung tissue samples (200 mg) were placed in 1.0 mL ice-cold phosphate-buffered saline (PBS) and homogenized using a polytron homogenizer (Brinkman Instruments, Inc., Westbury, NJ). The homogenates were centrifuged at 8,000 rpm for 20 minutes at 4°C, and the supernatant was filtered prior to assay of VEGF (a combination of VEGF165 and VEGF121), sFlt-1, and sKDR using a Quantikine immunoassay (R & D Systems, Minneapolis, MN) according to the manufacturer's assay procedure. Briefly, standard and samples were pipetted into a 96-well plate that was precoated with VEGF, sFlt-1, or sKDR antibody and incubated for 2 hours. After removal of the unbound substances, a substrate was added to the wells and color developed in proportion to the amount of bound VEGF. The VEGF and sFlt-1 standard curves ranged from 0 to 2,000 pg/mL, and the sKDR standard curve ranged from 0 to 5,000 pg/mL. The coefficient of variation from inter- and intra-assay precision assessment was less than 10%. All samples were assayed in duplicate.
Protein Assay
Lung tissue VEGF, sFlt-1, and sKDR levels were standardized using total cellular protein levels. Using a portion (10 μL) of supernatant, total cellular protein levels were determined in the supernatant by the Bradford method (Bio-Rad protein assay kit, Bio-Rad Laboratories, Hercules, CA) using bovine serum albumin as a standard.
Isolation of Total Ribonucleic Acid
Frozen lung specimens (100 mg) were placed in 1.0 mL of ice-cold TriZol reagent (Gibco BRL, Eggenstein, Germany) and homogenized and total ribonucleic acid (RNA) was extracted according to the manufacturer's protocol. Briefly, 200 μL chloroform per milliliter of TriZol was added to the homogenates, shaken vigorously for 15 seconds, and centrifuged at 14,000 rpm for 20 minutes at 4°C. The aqueous phase was removed, and the RNA was precipitated with isopropanol and collected by centrifugation at 14,000 rpm for 10 minutes at 4°C. The supernatant was discarded, and the RNA pellet was washed with 75% ethanol, centrifuged at 8,000 rpm at 4°C for 5 minutes, air-dried, and solubilized in diethylpyrocarbonate (DEPC)-treated water. The integrity of the RNA was determined by gel electrophoresis in 1% agarose gel stained with ethidium bromide. The purity of the RNA was assessed by the ratio of absorbance at 260 and 280 nm. The total RNA concentration was estimated by spectrophotometric measurements at 260 nm, assuming that 40 μg of RNA per milliliter equals one absorbance unit. The total RNA yield was diluted in DEPC-treated water to 1 μg/μL total RNA for all samples.
Reverse Transcriptase Polymerase Chain Reaction
Two micrograms total RNA was reversely transcribed to complementary deoxyribonucleic acid (cDNA) using Muloney murine reverse transcriptase (Perkin Elmer, Norwalk, CT). Amplification of cDNA was performed using specific sense and antisense primers for human glyceraldehyde-3-phosphate dehydrogenase (GAPDH), VEGF, and the receptors Flt-1, KDR/Flk-1, and NP-1 with AmpliTaq DNA polymerase (Perkin Elmer, Norwalk, CT). The sense and antisense primers were prepared by Life Technologies (Carlsbad, CA). The sense and antisense primer sequences for GAPDH were 5′-CCA CCC ATG GCA AAT TCC ATG GCA-3′ and 5′-TCT AGA CGG CAG GTC AGG TCC ACC-3′, respectively.28The sense and antisense sequences for VEGF were 5′-CGA AGT GGT GAA GTT CAT GGA TG-3′ and 5′-TTC TGT ATC AGT CTT TCC TGG TGA-3′, respectively.28The polymerase chain reaction (PCR) products from these primers showed amplification of three VEGF splice variants corresponding to VEGF121, VEGF165, and VEGF189. The PCR cycle profile, which was carried out using a model 480 DNA thermal cycler (Perkin Elmer Cetus, Perkin Elmer), was 94°C for 3 minutes for 1 cycle, followed by 35 cycles of 94°C for 1 minute, 56°C for 1 minute, and 72°C for 1 minute, followed by 1 cycle of 72°C for 10 minutes.28The sense and antisense primers for Flt-1 were 5′-CAG CGG CTT TTG TGG AAG ACT CAC-3′ (sense) and 5′-ACA TCT CGG TGT CAC TTC TTG GAC-3′, respectively (735 bp), and the sense and antisense primers for KDR/Flk-1 receptor were 5′-CAA CAA AGT CGG GAG AGG AG-3′ (sense) and 5′-ATG ACG ATG GAC AAG TAG CC-3′, respectively (819 bp). The PCR cycle profile was 95°C for 30 seconds, 56°C for 5 seconds, and 72°C for 1 minute, for 40 cycles, followed by 72°C for 10 minutes.29The sense and antisense primer sequences for NP-1 were 5′-CAA CGA TAA ATG TGG CGA TAC T-3′ and 3′-TAT ACT GGG AAG AAG CTG TGA T-5′, respectively (820 bp). The PCR cycle profile was 94°C for 1 minute, 63°C for 2 minutes, and 72°C for 3 minutes for 30 cycles, followed by 72°C for 3 minutes.22
Densitometric Scanning
Gel electrophoresis of the PCR products was performed on 1.5% agarose gel stained with ethidium bromide. The intensities of the bands were measured with the use of a Gel Doc 1000 Darkroom Imager and Molecular Analyst software (Bio-Rad Laboratories). The PCR fragments were identified according to their molecular mass using a DNA mass ladder (Perkin Elmer). The amount of DNA in each specimen was semiquantitated by the integrated density of the product bands within a closed rectangle, which was then normalized to the density of the GAPDH bands. The data are expressed as mean VEGF189, VEGF165, VEGF121, Flt-1, KDR/Flk-1, or NP-1/GAPDH ratio ± SEM.
Immunohistochemistry
VEGF immunohistostaining was performed using ABC staining kits (Santa Cruz Biotech, Santa Cruz, CA) according to the manufacturer's protocol, with minor modifications as outlined previously.8Briefly, paraffin-embedded sections were deparaffinized through xylene and ethanol prior to digestion in trypsin (0.1% in PBS) to unmask the antigens. Following incubation in methanol and 30% hydrogen peroxide, the slides were washed and blocked with blocking serum. Immunohistochemistry was performed using 1:100 dilution of rabbit antihuman VEGF polycolonal antibody raised against a peptide mapping at the amino terminus of VEGF of human origin and detects the 189-, 165-, and 121-amino acid splice variants of precursor and mature VEGF (Santa Cruz Biotech). Following incubation overnight at 4°C, the slides were rinsed and incubated with biotinylated secondary antibody, after which they were incubated with streptavidin-horseradish peroxidase. The slides were stained with 3,3′-diaminobenzidine and counterstained with hematoxylin-eosin. We received six slides per animal. Each slide contained three sections. For the 125-day, 140-day, and term GCs, sections were taken from the right middle and lower lobes. For the 160-day and 175-day GCs, sections were taken from the right middle lobe. For the 125-day BPD model, sections were taken from the right lower lobe; for the 140-day PRN O2 BPD model, sections were taken from the right upper and lower lobes; and for the 140-day 100% O2 BPD model, sections were taken from the right middle and lower lobes.
Statistical Analysis
Statistical analysis was accomplished with the use of GraphPad Instat (GraphPad, San Diego, CA). One-way analysis of variance was used to examine differences among the groups for the ontogeny data. Post hoc analysis was performed using Student-Neuman-Keuls multiple comparisons test or Dunnett's test following Levene's test for equality of variances. Unpaired Student's t-test was used to examine differences between the BPD models and their matched GCs for normally distributed variables and Wilcoxon rank sum test for non-normal data. Data are expressed as mean ± SEM. A p value of ≤ .05 was considered significant.
RESULTS
Effect on VEGF
VEGF levels (soluble forms 165 and 121) in the GCs were consistently low in early gestation but increased several-fold at term (data not shown). Data for VEGF in the lung homogenates from the BPD groups proved to be unreliable because one sample in the 140-day + 100% O2 group was hemorrhagic and generated extremely high VEGF levels. The two other samples in that group had mild hemolysis. Because hemolysis interferes with the VEGF assay, resulting in abnormally high VEGF levels, we excluded the VEGF protein data. Figure 1 represents the messenger ribonucleic acid (mRNA) expression of VEGF189 in the baboon lungs. Data are standardized using GAPDH mRNA expression. VEGF189 is highly expressed in early gestation until 175 days gestation, at which time, the expression decreases fourfold, with minor elevations at term (see Figure 1A). Interestingly, despite the drop in VEGF189 mRNA expression at 175 days and term gestation, the levels were higher than that of VEGF165 at the corresponding ages. Exposure to oxygen significantly reduces VEGF189 mRNA expression in all of the BPD models (see Figure 1, B and C) compared with their age-matched controls.
As noted in Figure 2, mRNA expression of VEGF165 produced a response pattern similar to that of VEGF189, despite its lower abundance in the lungs (see Figure 2A). VEGF165 mRNA was lower in the baboon lungs from the 125-day PRN O2 group compared with the 140-day GC (see Figure 2B). In contrast, there were no appreciable changes in VEGF165 in the 140-day baboons developing BPD/CLD (see Figure 2C). VEGF121 mRNA expression was several-fold lower in the baboon lungs compared with VEGF189 and VEGF165 and exhibited a quite different ontogenic pattern (Figure 3). VEGF121 mRNA expression was biphasic, low at 125 and 140 days, increased at 160 days, and gradually decreased at 175 and 185 (term) days gestation (see Figure 3A). In the 125-day BPD model, VEGF121 mRNA expression was increased compared with the 140-day GC (see Figure 3B). However, in both 140-day BPD models, VEGF121 mRNA expression was decreased compared with the 160-day GC (see Figure 3C). VEGF immunostaining (189, 165, and 121 isoforms) is shown in Figure 4. VEGF protein (brown stain) was intense in the septae from 140-day GC (see Figure 4A) and was decreased in the 125-day BPD model, which had significantly less alveolar formation (see Figure 4B). VEGF protein is still seen in the septae from the 160-day GC (see Figure 4C) and 140-day BPD model (see Figure 4D) but was decreased in the 140-day 100% BPD model despite thickened septae (see Figure 4E).
Effect on VEGF Receptors
sFlt-1 levels were approximately 100-fold higher than VEGF levels in the lung homogenates throughout normal lung development (Figure 5A). Soluble Flt-1 levels were decreased in the 125-day BPD model compared with the 140-day GC (Figure 5B). A similar drop is noted in the 140-day BPD models compared with the 160-day GC, although significance was not achieved. This may be due to the small sample size because one sample was excluded owing to pulmonary hemorrhage in the 140 + 10-day 100% O2 group. Although the two other samples in that group had mild hemolysis, this did not interfere with the soluble receptor protein assays. Soluble KDR levels in the lung homogenates were also several-fold higher than that of VEGF, although they were lower than that of sFlt-1 (Figure 6). During normal lung development, sKDR remained relatively constant, with only a minor elevation at 175 days compared with term gestation (see Figure 6A). In the 125-day BPD model, no change in sKDR levels was noted compared with the 140-day GC (see Figure 6B). In contrast, the 140-day BPD models had lower sKDR levels compared with the 160-day GC (see Figure 6C).
Lung mRNA expression of Flt-1 progressively decreased with advancing gestation and reached almost undetectable levels at term (Figure 7A). All BPD models had decreased Flt-1 mRNA expression compared with their age-matched controls (Figure 7, B and C). Lung KDR/Flk-1 mRNA expression progressively increased to peak at 160 days gestation and then suddenly decreased to almost nondetectable levels at 175 days and term gestation (Figure 8A). Similar to the Flt-1 receptor, lung KDR/Flk-1 receptor mRNA expression was decreased in all BPD models compared with their gestational age-matched controls (Figure 8, B and C). Lung NP-1 receptor mRNA expression demonstrated a response pattern similar to that of VEGF165 mRNA expression, high in early gestation with progressive reductions to almost undetectable levels at term (Figure 9A). NP-1 mRNA expression was mildly reduced in the 125-day BPD model (Figure 9B), whereas in the 140-day BPD models, a dramatic reduction in NP-1 mRNA expression was noted (Figure 9C).
DISCUSSION
In this study, the ontogeny of VEGF and VEGF receptors in lungs from normal baboons at different gestational ages was examined. Changes in VEGF and VEGF receptors were also examined in three premature baboon BPD models. We found that a predominant VEGF splice variant, VEGF189, was coexpressed in developing lungs with VEGF165 and, to a lesser degree, with VEGF121. VEGF189 and VEGF165, as well as VEGF receptor transcripts, were decreased in lungs from extremely premature baboons developing BPD/CLD. These findings demonstrate an association between abnormal VEGF and VEGF receptor levels and the development of BPD/CLD.
Previous studies have demonstrated reduced VEGF mRNA and protein expression as well as decreased Flt-1 receptor mRNA expression in lungs from infants with fatal BPD.10Using baboon BPD models, similar to those used in the present study, Maniscalco and colleagues also showed decreased VEGF immunostaining and decreased Flt-1 receptor mRNA in the 125-day PRN O2 model compared with the 140-day GC.8The present study provides further information regarding the responses of the VEGF splice variants and their tyrosine kinase receptors in premature baboons developing BPD/CLD. However, to our knowledge, this is the first study to describe the responses of soluble VEGF receptors and the VEGF ligand-specific receptor NP-1 in the setting of BPD/CLD. Our findings, as well as those of Maniscalco and colleagues,8Bhatt and colleagues,11and Brown and colleagues,16demonstrate that VEGF is essential for normal baboon lung development.
Premature baboon lungs at 125 days gestation are developmentally equivalent to that of a 24- to 26-week human premature infant, whereas premature baboon lungs at 140 days gestation are equivalent to a 30- to 32-week human premature infant.13Therefore, premature baboons delivered at 125 days gestation and managed with ventilation and supplemental O2 PRN develop lung pathophysiology similar to extremely premature human newborns with BPD/CLD.7The saccular (27-37 weeks gestation) and alveolar (postnatal) stages are characterized by increased angiogenesis and a reduction in mesenchymal tissue.30Our findings of high VEGF levels in the lung homogenates from GCs at term compared with early gestation (data not shown) were not surprising and provide further support that VEGF increases during late fetal lung development.13,30
VEGF189 is expressed by epithelial cells and is highly bound to the extracellular matrix.31The expression of VEGF189 isoform during the canalicular stage of lung development, extending to the terminal sac period, suggests that this splice variant mediates epithelial-endothelial cell interaction as the capillary-alveolar network is formed. Localized expression of VEGF189 in the extracellular matrix, where this isoform is tightly bound, implicates an additional role of mesenchymal-endothelial cell interaction in lung organogenesis, as recently described by Greenberg and colleagues.32Hyperoxia damages epithelial cells, which are the source of VEGF35. The damaging effects of hyperoxia on lung epithelial cells were proposed by Das and colleagues in a recent study and demonstrated that apoptosis is the likely mechanism for alteration in lung architecture in a similar baboon BPD model.33Therefore, in the present study, apoptosis may contribute to the decrease in VEGF189 mRNA expression observed in the BPD models. Of the five isoforms, VEGF165 is the most potent.21It interacts specifically with NP-1, and the coordinate expression of VEGF165, NP-1, and KDR/Flk-1 promotes potent angiogenesis.22VEGF165 is a more potent mitogen for endothelial cells than VEGF121. VEGF121 is soluble, lacks a heparin-binding domain, and is constitutively expressed.22Therefore, despite its increase in the 125-day BPD model, it may not be an important molecule for the control of angiogenesis. Instead, these findings suggest conversion of the ECM-bound isoforms to the soluble VEGF121 isoform. Increased VEGF121 with oxygen injury was previously demonstrated by Watkins and colleagues in rabbit lungs.20
In the lung homogenates, sFlt-1 is more than 100-fold higher than VEGF and 2 to 4 times higher than sKDR. High sFlt-1 during early gestation may not have a physiologic role because VEGF levels remain low until late gestation, when sFlt-1 starts to decrease. It was interesting to note that exposure to O2 and mechanical ventilation also has a suppressive effect on the soluble VEGF receptors, except for sKDR in the 125-day BPD model. Significant suppression of sFlt-1 in the lung homogenates from the 125-day PRN O2 model may be counterintuitive because this would suggest increased VEGF activity in the lungs. However, considering that VEGF immunostaining was decreased in the lung tissue (see Figure 4) from that same group, we do not believe that VEGF activity was increased. It may be a compensatory response to decreased VEGF levels. It should be noted that because one sample in the 140 + 10-day 100% O2 group was excluded owing to contamination with hemorrhaging, caution should be applied when interpreting these data. When the sample number is so small (n = 3), removal of even one sample owing to contamination would have an impact on the data.
We have shown that the VEGF receptor Flt-1 mRNA remains relatively high in early gestation, whereas KDR/Flk-1 peaks at 160 to 175 days in baboon lung development. The increase in Flt-1 during this period when the lung is highly vascularized is necessary for organization of the microvasculature. On the other hand, peaking of KDR/Flk-1 at the latter part of gestation may represent an intense period of endothelial cell differentiation during the formation of a large capillary surface. Flt-1 is important for vascular organization. Its decrease in the BPD models suggests dysmorphic lung capillaries. In contrast to the findings of Maniscalco and colleagues,8we showed that KDR/Flk-1 was suppressed during hyperoxia and mechanical ventilation. This apparent discrepancy could be explained by differences in sampling sites between the two studies. Nevertheless, the decrease in KDR/Flk-1 observed in our study may represent decreased endothelial cell mass and/or decreased epithelial cells.
Our study suffers from several limitations, not the least of which is small sample size. Despite our small sample size, we were able to demonstrate significant changes in VEGF and VEGF receptor expression. A previous report by Maniscalco and colleagues corroborates some of our findings.8However, inconsistencies in lung sampling sites could also account for the wide variation in some of our data. Further studies with a larger sample size and use of alternative methodologies, such as ribonuclease protection assays, are required to confirm our data. Another important limitation was the lack of a 125-day PRN O2 group that was not exposed to antenatal steroids and postnatal surfactant. This group would have provided important information regarding the effects of antenatal steroids and postnatal surfactant on VEGF and VEGF receptors in BPD/CLD. However, this group would not have survived without antenatal steroids and postnatal surfactant. We also lacked a true 150-day GC for the 140-day BPD models. Comparing lungs from the 150-day BPD models with lungs from a more mature 160-day GC may lead to errors in interpretation.
In summary, we have confirmed that VEGF189 is the predominant isoform in the lungs and that VEGF and VEGF receptor expression is impaired in extremely premature, low birth weight infant baboons developing BPD/CLD. Although we speculate that lung injury leads to conversion of the ECM-bound isoforms, further studies are needed to provide a fuller understanding of the underlying mechanisms associated with increased lung VEGF121 mRNA expression during alternating hyperoxic/normoxic episodes. These data provide an association between decreased lung VEGF and VEGF receptor mRNA expression and BPD/CLD and may suggest a potential therapy to promote normal lung growth and development in infants at risk.
ACKNOWLEDGMENT
We would like to thank Jacqueline J. Coalson, PhD, and the BPD Resource Center, Southwest Foundation for Biomedical Research, San Antonio, Texas, for providing the baboon frozen lung tissue samples and the paraffin-embedded tissue sections.