Introduction

Abstract Introduction Materials and Methods Results Discussion References

Although formation of the pulmonary vasculature is a critical component of lung development, little is known of the mechanisms that regulate its development (1). Recent information suggests that early vascular development in the lung occurs through both angiogenic and vasculogenic mechanisms (2). Recruitment of mesenchymal cells committed to a smooth-muscle lineage to the nascent endothelial tubes is the next step in vascular development. Although recent studies have evaluated this process in the systemic circulation (3), no such information is available for the pulmonary circulation. As development of the vessel wall proceeds, both proliferation and production of matrix proteins by the committed smooth-muscle cell (SMC) is crucial for normal lung vascular function. When lung development is completed, SMC growth ceases, and under normal conditions the cells remain in a quiescent state. Previous studies in our laboratory have documented dramatic changes in the proliferative rates of aortic SMCs over the time course of vascular development (4). However, an analysis of smooth-muscle proliferation over the entire course of lung vascular development (i.e., from the embryonic period of life through adulthood) has never been done. Significant differences between the two circulations in the control of proliferation may be expected because of the different time courses of their development and their uniquely different functions, especially during fetal life.

Production of extracellular matrix (ECM) proteins by SMCs is essential for vascular development. The diverse proteins made by SMCs and deposited in the basement membrane and vascular interstitium function both to provide physical support for the cellular milieu and to exert a multitude of regulatory functions on the growth and differentiation state of the cells with which they interact (5- 9). ECM proteins reported to be important in both normal vascular development and pathologic conditions of the vasculature include structural matrix proteins (such as type I collagen and elastin) and basement membrane (BM) proteins (such as perlecan, laminin, and type IV collagen) (1, 10). The growth-regulatory properties of specific BM proteins have been particularly well demonstrated in multiple cell types, including SMCs. For example, it has been shown that SMCs cultured on reconstituted BM show markedly reduced replication rates and appear to exhibit a more differentiated phenotype (18). More specifically, perlecan, the predominant BM heparan sulfate proteoglycan, identified in the BM of virtually every cell type (21), has been shown to play a direct role in this growth suppression (18, 25, 26). We showed that perlecan can decrease SMC proliferation by repressing the expression of essential growth-promoting transcription factors (18). Furthermore, heparan sulfate-degrading enzymes enhance the modulation of SMCs from a fully differentiated "contractile" phenotype to a less mature "synthetic" phenotype (27). On the other hand, the interstitial matrix proteins, type I collagen and elastin, are crucial for structural integrity, but in their intact state are not known to affect directly the proliferative phenotype of vascular SMCs (28). The expression of tropoelastin (TE) is known to be developmentally regulated and limited to a brief period in fetal and early neonatal life (29), but this protein has been shown to be reexpressed by SMCs in response to vascular injury (35). Expression of elastin often accompanies abnormal cell replication, but control of its expression has been speculated to be independent of cell replication. The relationship of these important matrix proteins, with significantly different functions, to the changes that occur in SMC proliferation over the full course of pulmonary vascular development has not been evaluated.

We therefore sought first to establish the time course and magnitude of changes in vascular SMC proliferation occurring over the course of lung development, and then to evaluate temporal and spatial changes in the production of BM and interstitial matrix proteins and to relate these changes to changes in SMC proliferation. We chose to evaluate perlecan, a ubiquitous BM protein with purported significant effects on the proliferative capacity of vascular SMCs, and TE, an essential structural ECM protein known to be tightly developmentally regulated. We hypothesized that expression of perlecan would occur concomitantly with decreases in SMC proliferation, but that changes in TE expression would occur independently of changes in SMC proliferation. We used immunohistochemistry, in situ hybridization, and combined immunohistochemistry-in situ hybridization techniques to evaluate both global and cell-specific relationships between matrix protein messenger RNA (mRNA) expression and cell proliferation. We found dramatic changes in proliferation occurring at specific stages of development. Although peak expression of both proteins appeared to follow significant decreases in SMC proliferation, TE transcripts were first detected at embryonic Day 14 (e14), whereas perlecan mRNA was undetectable until fetal Day 19 (f19). Moreover, we found that when we evaluated them on an individual cell basis, perlecan was predominantly expressed in nonreplicating cells, whereas TE mRNA was expressed equally in replicating and nonreplicating cells.

	Materials and Methods

Abstract Introduction Materials and Methods Results Discussion References

Pulmonary Developmental Time Course and Tissue Preparation

For lung-developmental studies, female Sprague-Dawley rats of timed pregnancy were obtained from Harlan Sprague Dawley (Indianapolis, IN). At gestational Days 12 to 20, pregnant females were injected intraperitoneally with bromodeoxyuridine (BrdU, 100 mg/kg body weight [BW]; Sigma Chemical Co., St. Louis, MO) at 17 h, 9 h, and 1 h before death in order to label all SMCs traversing the S phase of the cell cycle in a 24-h period (36). Whole embryos (gestational Days 13 to 17 [e13 to e17]) or fetuses (gestational Days 18 to 21 [f18 to f21]) were removed and placed in 4% phosphate-buffered formaldehyde for 16 h at 4°C, and were embedded in paraffin wax for sectioning. Lung-tissue sections from postnatal animals (postpartum days [ppd] 1, 2, 4, 7, 14, 30, 60, and 90) were obtained through similar protocols. At least three animals were used for each time point; tissue sections from the same animals were used for BrdU immunohistochemistry, TE and perlecan in situ hybridization, and combined in situ hybridization-immunohistochemistry protocols.

BrdU Immunohistochemistry and Analysis of DNA Replication by Pulmonary Artery SMCs, Airway SMCs, and Airway Epithelial Cells

Immunoperoxidase staining for BrdU was used to visualize and quantitate replicating SMCs and epithelial cells in the lung, as previously described in the aorta by Cook and associates (4). Briefly, paraffin-embedded sections were deparaffinized, treated with proteinase K, denatured with 2 N HCl, blocked by incubation with normal horse serum, and exposed to a monoclonal antibody directed against BrdU (Becton-Dickinson, San Jose, CA). An avidin-biotin- immunoperoxidase system (Pierce, Rockford, IL) and 3,3'-diaminobenzidine (DAB) substrate kit (Vector Laboratories, Burlingame, CA) were used to detect antigen- antibody complexes. The sections were subsequently counterstained with hematoxylin and eosin (H&E). To distinguish SMCs from surrounding adventitial cells in the pulmonary artery (PA) and airways, selected embryonic sections were stained immunohistochemically with an antibody specific for smooth-muscle -actin. Replication indices were established by determining the percentage of BrdU-positive nuclei in SMCs from the lobar PA, in SMCs from first-branch intrapulmonary airways, and in epithelial cells. At least 600 cells of each cell type, from a minimum of three different animals, were counted for each time point; data are presented as means ± SEM.

In Situ Hybridization

In situ hybridizations for perlecan and TE were performed as previously described (35, 37). Briefly, sections of the tissues described earlier were deparaffinized, rehydrated, then serially incubated in 0.2% Triton X-100/phosphate-buffered saline (PBS), 1 mg/ml proteinase K, and acetic anhydride/0.1 M triethanolamine buffer. The slides were sequentially dehydrated in a graded ethanol series, air-dried, prehybridized for 2 h, and then hybridized overnight to a ³⁵S-labeled riboprobe for TE or perlecan (10⁶ cpm per section; the complementary DNA [cDNA] for rat TE was kindly provided by Dr. W. Parks, Washington University, St. Louis, MO; the cDNA for murine perlecan was kindly provided by Dr. Steve Ledbetter, Upjohn, Inc., Kalamazoo, MI). After hybridization, tissue sections were washed in 2× saline sodium citrate (SSC), incubated with ribonuclease A (RNAase A) (Sigma), and washed several times with 2× SSC at both room temperature and 55°C, and with 0.1× SSC at 55°C. The sections were dehydrated through a graded ethanol series, air-dried, and dipped in nuclear tracking bulk emulsion-2 (Eastman Kodak Co., Rochester, NY). The slides were developed after 5 to 7 d for TE, and at 21 d for perlecan; all slides were counterstained with H&E. Tissue sections from at least three animals were used for each time point. Duplicate sections were used for both sense- and antisense-probe hybridization (Figure 1).

View larger version (209K): [in this window] [in a new window]   Figure 1.   In situ hybridizations demonstrating positive signals with antisense riboprobes for TE mRNA in ppd 1 lung (a) and for perlecan mRNA in Day 19 fetus (c). (b) and (d) Signal was absent with the corresponding sense riboprobes (TE [b]; perlecan [d]).

Combined Immunohistochemistry-In Situ Hybridization

The immunohistochemistry and in situ hybridization techniques were combined by sequentially performing immunohistochemistry for BrdU followed by in situ hybridization for TE or perlecan, as previously described (35, 37). Sections were deparaffinized, rehydrated in a graded ethanol series, and incubated in 100 µg/ml proteinase K for 10 min. The sections were then incubated in 2 N HCl for 20 min, followed by incubation in horse serum for 20 min. The remaining immunohistochemistry protocol was performed as described earlier. After exposure to DAB, the slides were washed in 0.1× PBS, incubated in acetic anhydride, dehydrated in a graded ethanol series, and air-dried. The slides were prehybridized for 2 h, and the remaining in situ protocol was performed as described previously. Quantitation of BrdU-positive and BrdU-negative SMCs expressing perlecan or TE was done at fetal Day 21, a time at which both messages are expressed in the PA. A minimum of 10 grains within and immediately surrounding the nucleus was used as a criterion to establish a cell as positive for TE or perlecan message (the threshold of 10 grains was used because the presence of 0 to 8 grains within nuclei in sections hybridized with sense probes). A minimum of 300 PA SMCs from three or four rats were counted.

Perlecan Immunohistochemistry

Formalin-fixed, paraffin-embedded lung tissues from prenatal (Days e17 to f21) and postnatal (ppd 1, 2, 4, 7, 14, 30, 60, and 90) animals were deparaffinized and exposed overnight at 4°C to a polyclonal antibody specific for perlecan (kindly provided by Dr. John Hassell) (38). Antigen-antibody complexes were visualized with an avidin-biotin-immunoperoxidase system and DAB as a substrate. Sections were counterstained with hematoxylin. Controls included the use of a polyclonal antiserum to a human rotavirus (Dako, Inc., Carpinteria, CA) as primary antibody.

Statistical Analysis

Data are expressed as means ± SEM. BrdU incorporation data for vascular SMCs, airway SMCs, and airway epithelial cells were analyzed through one-way analysis of variance (ANOVA) with the Student-Newman-Keuls multiple comparisons test. A value of P < 0.05 was considered significant.

	Results

Abstract Introduction Materials and Methods Results Discussion References

PA SMC DNA Replication Is Highest in the Embryonic Period

The time course of replication of lobar PA SMCs in vivo was determined with BrdU immunohistochemistry (Figure 2). In the PA, SMCs consistently replicated at a high rate (84% per 24 h, on average) during the embryonic stage (Days e14 to e17), but underwent a dramatic decline in replication (P < 0.05) at the transition between the embryonic and fetal periods of development (Days e17 to f19). The PA SMC replication rate remained moderately stable from the late fetal period through ppd 7 (32% BrdU labeled per 24 h, on average). By ppd 14, the BrdU-labeling index had decreased to 8% per 24 h, and by 3 mo of age was < 2% per 24 h.

View larger version (28K): [in this window] [in a new window]   Figure 2.   Bar graph showing quantitation of PA SMC replication rates in vivo during pulmonary development. Embryonic (Days e15 to e17), fetal (Days f18 to f21), and postpartum (ppd 1, 2, 4, 7, and 16; 1 and 3 mo) periods are shown. Replicating cells were identified by BrdU immunohistochemistry as described in MATERIALS AND METHODS. Data represent percent of BrdU-labeled nuclei in the 24 h before tissue harvest, and are presented as means ± SEM.

An in vivo BrdU-labeling index of airway SMCs and epithelial cells in the developing lung was also determined. We found that both airway SMCs and epithelial cells exhibited a pattern of DNA replication remarkably similar to that observed in PA SMCs. High replication indices were present in the embryonic period (80% per 24 h for BrdU-labeled airway cells). The replication rate of airway epithelium and SMCs decreased markedly in the early fetal period (P < 0.05) to 25 to 40% per 24 h, on average (Figure 3). Interestingly, however, unlike PA SMCs, both airway cell types were observed to undergo a transient increase in DNA synthesis from 24 to 48 h after birth (ppd 2 values significantly greater than Day f21 values for both airway SMCs and epithelial cells, P < 0.05). Subsequently, replication rates again declined and were at very low levels by 3 mo of age (< 2% per 24 h; comparable to rates observed in PA SMCs).

View larger version (26K): [in this window] [in a new window]   Figure 3.   Bar graphs showing quantitation of bronchial airway SMC and airway epithelial replication rates in vivo during pulmonary development. Data (percent BrdU-positive cells) from embryonic, fetal, and postpartum time points (as described in Figure 2) are represented, and are presented as means ± SEM.

Perlecan Gene Expression in the Lung Is Ubiquitous, But Tightly Regulated Temporally

During development, perlecan message was undetectable in the lung throughout the embryonic period (Days e13 to e17)(Figure 4a). However, in the early fetal period (Day f19), perlecan mRNA was expressed ubiquitously throughout the lung (Figure 4b). With in situ hybridization, perlecan mRNA was detectable in most pulmonary cell types, including vascular SMCs and endothelial cells, parenchymal interstitial cells, and airway and alveolar epithelial cells. Perlecan transcripts remained abundantly expressed throughout the remainder of the fetal period (Figures 4c through 4d) and during early postpartum life (Figure 4e), but decreased by ppd4 to low but detectable levels (Figure 4f). Perlecan mRNA expression in the adult lung was limited to a low level of expression and was observed only in the PA.

View larger version (144K): [in this window] [in a new window]   Figure 4.   Perlecan mRNA expression during pulmonary development. In situ hybridizations were used to detect perlecan mRNA in lung tissue sections from embryonic Day 17 (a); fetal Days 19 (b), 20 (c), and 21 (d); and ppd 1 (e) and 4 (f ) animals. Perlecan mRNA was first significantly expressed at Day f19, when pulmonary-cell replication began to decline, and continued to be detected in pulmonary vascular tissue at basal levels throughout adult life.

Perlecan Immunostaining Closely Follows the Pattern of mRNA Expression in the Developing Lung

As shown in Figure 5, minimal perlecan immunostaining was observed in the embryonic period (Figure 5a). In the early fetal period (Day f19), however, pulmonary vessels and, to a lesser extent, airways and parenchyma, showed distinct perlecan immunostaining (Figure 5b). Perlecan protein expression remained high throughout late fetal (Figure 5c) and early neonatal (Figure 5d) life. A decrease in perlecan immunostaining was observed in all pulmonary structures during the later neonatal period, but remained detectable at all postnatal time points examined (Figure 5e). No staining was observed with the irrelevant antirotavirus antibody (Figure 5f) or when the primary perlecan antibody was omitted (not shown).

View larger version (171K): [in this window] [in a new window]   Figure 5.   Distribution of perlecan protein in the developing lung. A monospecific polyclonal antibody was used to localize perlecan protein in lung tissue sections from embryonic Day 17 (a), fetal Days 19 (b) and 21 (c), and ppd 1 (d) and 90 (e) animals; the reaction product is dark brown. In the tissue shown in f, antirotavirus antiserum was used as the primary antibody; similar negative results were obtained when the primary antibody was omitted. In accord with the in situ hybridization results, perlecan protein was first detected at Day f19, and continued to be readily detectable throughout adult life. (a-f ) Bar = 100 µm.

TE mRNA Expression in the Developing PA Follows an Embryonic Period of Rapid SMC Replication

During lung development, TE message was detected not only in SMCs of pulmonary arteries, veins, and airways, but also in vascular endothelial cells and in the lung parenchyma (Figure 6). At all developmental ages examined, TE gene expression appeared to be more intense in PA SMCs than in SMCs surrounding airways. Expression of TE mRNA by endothelial cells was similar in intensity to that observed in pulmonary vascular SMCs, and followed the same time course of expression as in SMCs. At embryonic Day e14, TE mRNA transcripts were observed in the PA (and aorta), whereas airway SMCs (visualized at higher magnification and confirmed to be SMCs by positive actin staining) showed no TE mRNA expression (Figure 6a). By Day f18 (Figure 6b), not only pulmonary blood vessels but also airway SMCs and parenchymal cells were positive for TE message; this pattern persisted through ppd 7 (Figures 6c through 6e). At ppd 14, TE expression persisted in the pulmonary vessels, but was now only minimally detectable in airway SMCs (data not shown), and by 1 mo postpartum, only pulmonary vessels were TE-positive (Figure 6f). At 3 mo, TE mRNA was undetectable in all pulmonary structures examined (data not shown).

View larger version (136K): [in this window] [in a new window]   Figure 6.   TE mRNA expression during pulmonary development. In situ hybridizations were used to detect TE mRNA in lung sections from embryonic Day 16 (a); fetal Days 18 (b) and 20 (c); and ppd 1 (d), 7 (e), and 30 (f) animals. Short arrows denote PA, long arrows denote airways, and the open arrow (a) denotes the aorta in a to f. TE mRNA was first detected in the aorta and pulmonary vessels during embryonic life, and later in pulmonary airway cells during fetal life. Although TE mRNA remained detectable in vascular tissue at basal levels throughout adult life, its expression was undetectable in airway cells of the adult animal.

On an Individual Cell Basis, Perlecan Transcripts Are Exclusively Expressed by Nonreplicating Cells, Whereas TE Expression Occurs in Both Replicating and Nonreplicating Cells

Because TE expression was observed during the highly replicative period of embryonic development, whereas perlecan transcripts were first detected during the embryonic-to-fetal transition, when replication rates were markedly decreasing, we wanted to determine more precisely the relationship between cell replication and individual matrix-protein expression. Using a combined in situ hybridization-BrdU immunohistochemistry technique, we quantitated the number of cells expressing either or both markers. We found that only nonreplicating cells expressed large amounts of perlecan message in the rat lung. As illustrated for Day f21, replicating SMCs (BrdU-positive nuclei) from the proximal intralobar PA (Figure 7a) expressed minimal perlecan mRNA compared with nonreplicating cells. This same pattern was observed at all levels of the PA and pulmonary vein, from lobar to small peripheral intralobar branches (data not shown), and in the airway epithelium and stroma (much of the stroma is -actin-positive; Figure 7b). In contrast, as seen in the extrapulmonary PA on ppd 1 (Figure 7c), both replicating and non-replicating SMCs appeared to express similar amounts of TE message. This pattern of TE expression was true for all time points examined in the embryonic, fetal, and postnatal periods, and in all sizes of PA examined, including extralobar (Figure 7c), hilar (Figure 7d), and peripheral intralobar PA (data not shown). Interestingly, this pattern contrasted sharply with that seen in cardiac myocytes and airway chondrocytes, in which notable TE message was observed only in nonreplicating cells (data not shown). Quantitation of SMCs expressing perlecan or TE message and showing BrdU labeling confirmed that perlecan transcripts were abundantly expressed by nonreplicating compared with replicating PA SMCs, whereas TE was equally expressed in both replicating and nonreplicating SMCs (Figure 8).

View larger version (148K): [in this window] [in a new window]   Figure 7.   Relationship between cell replication and perlecan or TE mRNA expression. Sections of BrdU-treated Day f19 (perlecan message [a and b]) and ppd 1 (TE message [c and d]) lung tissues were examined with in situ hybridization to detect perlecan or TE mRNA (black dots under brightfield magnification) concurrently with BrdU immunohistochemistry (brown nuclei) to detect DNA synthesis. Arrows indicate BrdU-positive, replicating cells; arrowheads indicate BrdU-negative, nonreplicating cells. Replicating cells were typically devoid of perlecan message, whereas TE mRNA was equally expressed by both replicating and nonreplicating cells. (a-d ) Bars = 50 µm.

View larger version (18K): [in this window] [in a new window]   Figure 8.   Bar graph quantifying the relationship between PA SMC replication and matrix gene expression. For the data presented in Figure 7, individual SMCs were scored with respect to the presence of positive signals for perlecan or TE (> 10 silver grains within the cell nucleus was considered positive) and for BrdU (brown reaction product). Data are presented as means ± SEM.

	Discussion

Abstract Introduction Materials and Methods Results Discussion References

The acquisition of a medial coat of SMCs around developing vascular channels is required for normal pulmonary vascular development. Knowledge of the factors that control vascular SMC proliferation, as well as those that establish and maintain SMC quiescence once vascular morphogenesis is complete, is therefore necessary for a better understanding of vascular development. Assessment of the changes that occur in the proliferative state of SMCs during different phases of vascular development is a potentially crucial first step in this process and might provide important directions for research involving mechanisms that control pulmonary vascular SMC proliferation in both normal and pathologic situations. In addition to proliferating, SMCs in the developing circulation must also produce and assemble ECM proteins for normal vascular development to proceed. Besides their vital role in providing structural support, ECM molecules play important roles in cell adhesion (5), migration (7), differentiation (8, 9), and replication (6, 7, 9). In the present study, we therefore sought first to establish the changes in SMC proliferation that normally occur during pulmonary vascular development, and then to evaluate the relationship between SMC replication and the gene expression of two matrix proteins with significantly different functions, perlecan and TE, in the PA of the developing lung. We found dramatic differences in the rates of SMC proliferation at specific times during the course of lung develpment. Of particular interest was the high rate of proliferation in embryonic life and the dramatic decline that occurred at the transition to the fetal period of development. We also demonstrated that peak expression of two unrelated matrix proteins occurs after the PA has undergone a dramatic decrease in SMC replication. However, the two matrix-protein genes showed strikingly different patterns of regulation in relation to DNA replication in individual SMCs. Whereas TE message was first apparent during the highly replicative embryonic period and appeared to be expressed equally in both replicating and nonreplicating SMCs, significant perlecan gene expression was first observed following the dramatic decline in replication and was present predominantly in nonreplicating SMCs. Interestingly, this same pattern of an inverse relationship between perlecan expression and cell proliferation was also observed in airway SMCs and epithelial cells. The observation that perlecan expression is limited to nonreplicating cells is consistent with in vitro observations suggesting a potential role for perlecan in the regulation of SMC growth state and/or differentiated functions (18, 25, 27).

The observed pattern of PA SMC replication, consisting of rapid proliferation during the embryonic period followed by a sharp decline in DNA synthesis after the embryonic-to-fetal transition, is strikingly similar to that previously reported in the rat aorta (4), with the exception that PA SMCs demonstrate slightly higher replication rates in the adult animal (PA = 2%; aorta = 0.5%). SMCs isolated from the rat aorta during the embryonic period have been reported to grow autonomously in culture (i.e., in the absence of serum growth factors) and to become serum-dependent following the transition to the fetal stage (4). Because the replication indices of PA SMCs in the embryonic and fetal periods approximate those reported for rat aortic SMCs during development, it is possible that PA SMCs also have the capacity for self-driven replication during the rapid embryonic growth period and undergo a similar transformation to serum dependency at the embryonic-to-fetal transition. Although the molecular mechanisms regulating these growth characteristics remain unknown, it seems plausible that any interruption of these processes during crucial periods of development could contribute to the excessive SMC proliferation observed in perinatal pulmonary diseases such as persistent pulmonary hypertension of the newborn. In support of this, recent data from our group have shown a reexpression or persistence of fetal-like proliferative characteristics in PA SMCs isolated from newborn calves with severe neonatal pulmonary hypertension (39, 40).

Interestingly, we observed a similar developmental pattern of replication for airway SMCs and airway epithelial cells. These observations of similar replication patterns by such diverse pulmonary cell types suggest that proliferation of PA SMCs during the embryonic and early fetal periods of development are controlled by intrinsic, developmentally timed cellular mechanisms, and not by environmental factors such as blood flow or wall stress. However, unlike PA SMCs, airway cells demonstrated a slight increase in replication in the immediate postpartum period. Given the marked changes that occur in the airways during the transition to extrauterine life, it can be speculated that physical and biochemical mechanisms unique to the airways themselves positively influence the growth state of these cells during the early postpartum period. Additionally, although our data appear to represent the first complete analysis of cellular replication during development in large airways, the replication indices reported here for at least one time point in development (ppd 30) are consistent with those observed in an earlier study of airway epithelial cell and SMC replication indices of weanling rats exposed to normoxic or hyperoxic conditions (41).

Because the mature arterial wall consists of quiescent, fully differentiated tissue, one of our research goals is to identify endogenous mechanisms that contribute to the suppression of SMC replication during vascular development. The data from our work suggest that one process contributing to developmental growth inhibition is the gradual deposition of a growth-inhibitory ECM. The timing of perlecan gene expression reported here correlates with the decline in PA SMC growth during lung development, suggesting that perlecan plays a role in the acquisition of SMC quiescence observed in the mature blood vessel. Perlecan, a large proteoglycan containing a unique heparan sulfate-binding domain and domains homologous to the low-density lipoprotein receptor, laminin, and N-CAM (22, 42), is present in the basement membranes of a variety of cell types (14, 43), in all vascular tissues (46), and in the pulmonary parenchyma (44). This proteoglycan has been ascribed multiple properties, including cell adhesion (through ₁ and ₃ integrins) (47, 48) and binding of other matrix molecules (48) and growth factors (43). Heparin and endogenous heparan sulfates, including perlecan, are known potent inhibitors of vascular SMC proliferation (25, 49). Inhibition by these substances may occur through many mechanisms, including inhibition of protein kinase C-mediated responses (50, 51), downregulation of c-myc (52) and JunB (54) expression, inhibition of mitogen-receptor (53) and matrix-cell surface (55) interactions, and inhibition of proteinases involved in SMC migration and proliferation (56). Although perlecan has been shown to be a potent inhibitor of vascular SMC growth, previous studies have shown that it can facilitate mitogenesis in certain cell types, presumably through its interactions with fibroblast growth factor (43), and it may play an integral role in the onset of tumor invasion (59). It is possible that such diverse functions of perlecan derive from the various distinct domains of the core protein. Alternatively, compositional alterations in the ECM, such as would occur during the process of tumor invasion, could account for the growth-promoting activity of perlecan in invasive melanomas. Despite the apparent contradictory effects of perlecan on the growth of tumor cells and SMCs, an overwhelming abundance of data suggest that heparin and heparan sulfate proteoglycans, including perlecan, exert growth-inhibitory effects on vascular SMCs.

In support of a role for perlecan in the suppression of SMC growth in adult blood vessels are data showing that after experimental balloon-catheter-induced injury, peak neointimal perlecan expression occurs at a time point when SMC proliferation is decreasing (60); data for the pulmonary system documenting that perlecan derived from the PA endothelium is a potent inhibitor of PA SMC replication (25); and data demonstrating that heparin administration limits hemodynamic changes and PA medial thickening in experimental models of pulmonary hypertension (61). Furthermore, our previous data showed that an intact, perlecan-rich matrix inhibits SMC growth and suppresses the expression of specific growth-essential transcription factors (18). Although perlecan's presence in the adult lung and in the peri-implantation embryo has been documented (42, 64), there is no information describing the time course of perlecan expression during lung development. The present study shows that perlecan mRNA expression appears to be very tightly regulated in the perinatal period, with peak expression occurring from Day f19 through ppd 2, a time span in lung development during which PA SMC replication is rapidly decreasing. These data are very similar to those from earlier studies of the expression of perlecan during aortic development (37), suggesting that an inhibitory effect of perlecan on SMC replication during development is a characteristic feature of the developing vascular system.

We compared the expression pattern of perlecan to that of an unrelated but structurally essential matrix protein, TE. TE is an important ECM protein that contributes to the structural support and inherent elastic recoil of the vessel wall (28, 65). TE appears not only to be an important component of the normally developing blood vessel, but also to be involved in numerous pathologic processes of the systemic and pulmonary vasculature, including systemic and pulmonary hypertension (13, 15, 65, 66), atherosclerosis (67), and stenosis of the injured systemic vascular wall (60). Previous work has reported TE gene expression to be maximal in the PA during the late fetal period (30, 31, 68), whereas maximal elastin protein deposition occurs within the first 3 wk postpartum (32, 69). Our results concur with those in these earlier studies. However, we found a wider window of TE gene expression by PA SMC than had previously been reported (from Day e16 to ppd 30) and we found that expression of TE by PA SMCs occurs over a longer period during lung development than does perlecan expression.

Our results show that when compared with PA SMC replication (Figure 2), maximal levels of expression of both perlecan and TE follow peak DNA replication in the developing PA. However, on an individual cell basis, we found perlecan mRNA synthesis to be essentially limited to nonreplicating cells, whereas TE expression was unrelated to the replication state of the individual cell. These data suggest that the developmentally timed expression of perlecan could be associated with the induction of PA SMC quiescence. Alternatively, expression of perlecan by postreplicative SMCs may serve to maintain these cells in a quiescent state. Moreover, it appears likely that the expression pattern of TE is governed by a similar, but separate, developmental timing mechanism that controls cell replication, but that TE does not directly affect DNA replication. These results are not surprising, given perlecan's proposed role in SMC growth inhibition and elastin's primary function of providing structural support and elasticity to the vessel wall.

In summary, DNA replication in all pulmonary cell types examined in the present study was greatest during the embryonic period, and decreased during the early fetal period. Gene expression of both matrix proteins examined was greatest after the decline in DNA replication, with perlecan mRNA expression being highest in the fetal period and TE gene expression being highest in the early postnatal period. Importantly, on an individual cell basis, perlecan message was apparent only in nonreplicating pulmonary cells (which was not true for TE), indicating a role for perlecan in the induction and/or maintenance of quiescence of different cell types in the pulmonary vasculature and parenchyma.