DEVELOPMENTAL DYNAMICS 211:215–227 (1998) VEGF Enhances Pulmonary Vasculogenesis and Disrupts Lung Morphogenesis In Vivo XIN ZENG,1 SUSAN E. WERT,1 ROBERT FEDERICI, 2 KEVIN G. PETERS,2 AND JEFFREY A. WHITSETT1* 1Children’s Hospital Medical Center, Division of Pulmonary Biology, Cincinnati, Ohio 2Duke University Medical Center, Department of Internal Medicine, Cardiology Division, Durham, North Carolina ABSTRACT Vascular endothelial growth factor (VEGF) was expressed in developing respiratory epithelial cells under control of the promoter from the human surfactant protein C (SP-C) gene. SP-C-VEGF transgenic mice did not survive after birth. When obtained by hysterectomy on embryonic day 15 (E15) or 17 (E17), abnormalities in the transgenic mice were confined to the lung and were correlated with the expression of transgene mRNA as revealed by in situ hybridization. On E15 and E17, marked abnormalities in lung morphogenesis were observed in transgenic mice. Lungs consisted of large dilated tubules with increased peritubular vascularity. The mRNA levels of the VEGF receptor, Flk-1, and the endothelial cell specific receptor tyrosine kinase, Tie-1, were increased in lung mesenchyme of the transgenic mice. The numbers of acinar tubules and the abundance of mesenchyme were decreased. Endogenous VEGF mRNA was expressed in the respiratory epithelial cells of the developing lungs, and the levels of VEGF mRNA were increased in the SP-C-VEGF transgenic mice. Although the normal pattern of immunostaining for SP-C and Clara cell secretory protein (CCSP) indicated that epithelial cell differentiation was relatively unaltered by the transgene, electron microscopic analysis revealed a lack of alveolar Type I cell differentiation at E18. Expression of VEGF in the developing respiratory epithelium of transgenic mice increased growth of the pulmonary blood vessels, disrupted branching morphogenesis of the lung and inhibited Type I cell differentiation. Dev. Dyn. 1998;211:215–227. r 1998 Wiley-Liss, Inc. Key words: transgenic mice; lung development; alveolarization; vasculogenesis bules have a relatively small lumen and the mesenchyme remains prominent. However, from E15 to birth (canalicular and saccular periods), the terminal buds dilate and the mesenchyme thins. The developing pulmonary vasculature comes into increasing proximity to the epithelial cells in the periphery of the lung with advancing gestation. The close apposition of the pulmonary vasculature to the respiratory epithelium is temporally associated with differentiation of the respiratory epithelium during the canalicular and saccular stages of lung development, ultimately forming the functional gas exchange regions in the alveolus. In the mature alveolus, thin sheets of capillary endothelial cells form a network surrounding the alveolar saccules which are lined primarily by Type I and Type II epithelial cells (Burri, 1984; Adamson, 1991). Respiratory epithelial cells undergo extensive cytodifferentiation during the perinatal and postnatal period of development (Ten Have-Opbroek, 1991; Hilfer, 1983, Weaver and Whitsett, 1991). Inductive interactions between the pulmonary epithelium and mesenchyme are required for the process of branching morphogenesis, sacculation, and epithelial cell differentiation. Autocrine/paracrine signaling, cell-cell and cell-extracellular matrix interactions have been implicated in the process of lung morphogenesis and pulmonary epithelial cell differentiation (Guzowski et al., 1990; Minoo and King, 1994; Peters et al., 1994; Hilfer, 1996). The early stages of lung vascular development involve both angiogenesis and vasculogenesis (deMello et al., 1997, Pardanaud et al., 1989). The signaling molecules that control this process are not known. Vascular endothelial growth factor (VEGF) is a selective mitogen for endothelial cells, influencing angiogenesis and vasculogenesis during normal development (Dvorak et al., 1995). In the mouse, VEGF exists in at least three distinct isoforms that are generated by alternative splicing from a single gene (Breier et al., 1992). VEGF164 INTRODUCTION Lung development in the mouse begins on embryonic day 9.5 (E9.5) as the lung buds, derived from the primitive foregut endoderm, invade the splanchnic mesenchyme. These primitive respiratory tubules, lined by undifferentiated columnar epithelial cells, undergo dichotomous branching to form the conducting airways. During this early developmental stage, the lung tu- r 1998 WILEY-LISS, INC. Grant sponsor: RDP Center for the Cystic Fibrosis Foundation (X.Z., S.E.W., J.A.W.); Grant sponsor: Center for Gene Therapy for CF and Other Lung Diseases (X.Z., S.E.W., J.A.W.); Grant number: HL51832; Grant sponsor: James S. McDonnell Foundation (R.F., K.G.P.). *Correspondence to: Jeffrey A. Whitsett, M.D., Children’s Hospital Medical Center, Divisions of Neonatology and Pulmonary Biology, 3333 Burnet Avenue, Cincinnati, OH 45229-3039. E-mail: WHITJO@CHMCC.ORG Received 26 June 1997; Accepted 19 November 1997 216 ZENG ET AL. is a secreted form of the protein that is expressed in many tissues (Ferrara et al., 1996). VEGF is required for early embryogenesis because genetic inactivation of a single allele of VEGF in the mouse caused abnormal vasculature resulting in early embryonic lethality (Carmeliet et al., 1996; Ferrara et al., 1996). VEGF is also expressed at high levels in well-vascularized organs including the glomeruli in adult kidney, choroid plexus, lung, liver, and corpus luteum (Monacci et al., 1993, Berse et al., 1992). VEGF activates endothelial cells by binding to high affinity VEGF receptors, flt-1 (VEGFR-1), and Flk-1/KDR (VEGFR-2), both of which are expressed by endothelial cells (DeVries et al., 1992; Quinn et al., 1993; Mustonen and Alitalo, 1995). In adult lung, VEGF is synthesized primarily by alveolar epithelial cells and macrophages (Monacci et al., 1993). VEGF mRNA is enhanced in alveolar epithelial cells during hypoxia (Tuder et al., 1995) and hyperoxia (Maniscalco et al., 1995). In the embryonic lung, VEGF mRNA is detected in the airway epithelium (Breier et al., 1992; Millauer et al., 1993). In contrast, Flk-1 and flt-1 mRNAs are detected in lung mesenchyme (Millauer et al., 1993; Breier et al., 1995), including the mesenchymal progenitors of pulmonary blood vessels, supporting the concept that VEGF presents a paracrine signal from the epithelium activating target receptors on vessel precusor cells in the stroma. To study the function of VEGF in the vasculogenesis during mouse lung development and the influence of the developing vascular system in the lung mesenchyme on the differentiation of respiratory epithelium, we expressed murine VEGF in the respiratory epithelium of transgenic mice using the promoter from the human surfactant protein-C gene (Glasser et al., 1991; Wert et al., 1993). The altered vascularity caused by the VEGF transgene in transgenic mice was associated with marked disruption of acinar structure. RESULTS VEGF Disrupts Pulmonary Morphogenesis in Transgenic Mice The SP-C-VEGF DNA construct was microinjected into oocytes harvested from FVB/N mice and transferred to surrogate dams. Of the 40 potential founder mice born, 10 died at birth. When screened by PCR or Southern blot analysis, two of the pups bore the VEGF transgene. None of the surviving animals bore the transgene. Analysis of tissues from the dead transgenic animals revealed severe disruption of lung morphogenesis associated with the presence of the transgene (data not shown). Thereafter, embryos were obtained by hysterectomy on E15 and E17, representing the late pseudoglandular and late canalicular stages of mouse lung development, respectively. Of 31 embryos obtained, seven bore the SP-C-VEGF transgene. Weight and appearance of SP-C-VEGF mice were similar to their normal littermates. However, histological examination on E15 and E17 revealed a consistent pattern of abnormalities that were confined to the lungs from all transgenic embryos (Fig. 1). During this period of embryonic development, the bronchial tubules of the normal mouse lung undergo extensive branching, forming acinar buds in the periphery of the lung, which begin to form dilated saccules by E17. In contrast, the lungs from transgenic embryos on E15 and E17 consisted of large, abnormally dilated tubules, and the number of acinar buds and saccules was markedly reduced. The lumen of the remaining acinar buds and saccules were dilated, and the amount of mesenchyme was decreased in the transgenic mice. Except for the marked disruption of lung morphogenesis, the morphology of other organs from the SP-C-VEGF transgenic mice was indistinguishable from those of controls. Consistent histological abnormalities were observed in lungs from all nine transgenic founder animals (including the two dead transgenic pups). Expression of SP-C-VEGF Transgene In order to correlate the lung abnormalities from SP-C-VEGF transgenic embryos with the expression of the transgene, in situ hybridization using transgene specific probes generated from the SV40 t-intron-polyA sequence was performed to determine the presence and the expression pattern of SP-C-VEGF transgenic mRNA (Fig. 2A). Transgene mRNA was detected in the lung on cross sections (E17) and sagittal sections (E15) from the transgenic embryos. The hybridization signal was confined to pulmonary epithelial cells. Transgenic mRNA was restricted to epithelial cells of dilated bronchioles, acinar tubules and buds and was detected at low levels in epithelial cells of larger airways, consistent with the expression pattern noted for cDNAs expressed with the human SP-C promoter in embryonic lungs of transgenic mice reported previously (Korfhagen et al., 1990; Glasser et al., 1991; Wert et al., 1993; Simonet et al., 1995; Zhou et al., 1996a). Transgene mRNA was not detected in other tissues or in lungs of nontransgenic embryos. In situ hybridization using riboprobes generated from VEGF cDNA was also performed to compare the level and expression pattern of VEGF mRNA in lungs from both transgenic and nontransgenic control embryos (Fig. 2B). The endogenous VEGF mRNA was detected primarily in distal airway epithelial cells in normal mouse lung at both E15 and E17. High intensity of the hybridization signal, indicating the elevated level of VEGF mRNA, was observed in the epithelial cells lining the dilated tubules and acinar buds in lungs from transgenic embryos. Consistent with results from in situ hybridization, the VEGF protein, as identified by immunohistochemical staining using the VEGF antibody, was also detected in the distal airway epithelial cells in nontransgenic control embryos and in the epithelial cells lining the dilated acinar tubules and buds in lungs from transgenic embryos at E15 and E17 (Fig. 2C). Fig. 1. Morphology of lungs from SP-C-VEGF transgenic embryos and control littermates. Sagittal sections prepared from embryos on E15 (a,b) and cross sections from embryos on E17 (c,d) were stained with hematoxylin and eosin and photographed under bright field microscopy. The panel on the right is a higher magnification of the same field seen on the left. In the normal lung on E15 and E17, extensive branching of the airway epithelium is associated with the formation of numerous terminal acinar buds (B) and acinar saccules (S) with small lumen (a,c). In the transgenic embryos, respiratory tubules (T) and saccules (S) are markedly dilated and the numbers of terminal buds markedly decreased (b,d). Epithelial cells of the dilated tubules at E15 are columnar in transgenic embryos (f, arrow), and are primarily cuboidal in nontransgenic embryos of the same age (e, arrow). On E17, epithelial cells of the dilated airways in transgenic embryos are cuboidal and comparable to those lining the acinar tubules of nontransgenic controls (h,g, arrows). Bars 5 100 µm. 218 ZENG ET AL. Fig. 2. In situ hybridization for transgene specific and VEGF mRNA and immunohistochemical staining for VEGF protein. Cross (E17) and sagittal (E15) sections of embryos were prepared from both transgenic and nontransgenic mice. Tissue sections were hybridized overnight with 35S-labeled anti-sense and sense (latter not shown) riboprobes generated from transgene specific t-intron-polyA sequence (2A) or VEGF cDNA (2B) or stained with VEGF antibody (2C). 2A: Transgene specific mRNA was readily detectable in the epithelial cells lining the dilated tubules on E15 (c) and the malformed saccules on E17 (d) from the transgenic embryos. The hybridization signal was not observed in bronchi or larger bronchioles in the transgenic embryos. Signal was confined to the lung and was not detected in nontransgenic embryos (a,b). Expression of VEGF in Respiratory Epithelial Cells Induced Ingrowth of Blood Vessels were generally found in close contact with the respiratory epithelium, a situation that does not normally occur until the saccular stage (E17-birth) of normal lung development. VEGF receptor Flk-1 and its related endothelial cell specific receptor tyrosine kinase Tie-1 are expressed in the mesenchyme of developing lung (Korhonen et al., 1994; Breier et al., 1995). To determine whether the increased VEGF level affected the VEGF receptor expression, in situ hybridization was performed on tissue sections from transgenic and nontransgenic embryos at E15. The higher intensity of the hybridization signal for both Flk-1 and Tie-1 mRNA indicated increased vascularity and/or enhanced Flk-1 and Tie-1 mRNAs in the lung mesenchyme (Fig. 4), but not in other tissues (data not shown) in the SPC-VEGF transgenic embryos. To detect other alterations in mesenchymal components, lung sections were stained for a-smooth muscle actin (a-sma). Staining for a-sma, normally detected surrounding the bronchiolar stalks and in the clefts at branch points of bronchial tubules, appeared in a Since VEGF receptors are expressed in the precursors of endothelial cells in mesenchyme, the pulmonary blood vessels are most likely to be affected directly by the expression of VEGF in the airway epithelium. To determine the distribution of pulmonary blood vessels in SP-C-VEGF transgenic mice, immunohistochemical staining for PECAM-1 (CD31), a marker for endothelial cells (Xie et al., 1993; Vecchi et al., 1994), was performed on tissue sections from the SP-C-VEGF transgenic embryos and nontransgenic control littermates. On E15 of normal mouse lung development, microvessels were noted within the pulmonary mesenchyme and were distributed around the distal acinar buds at high density (Fig. 3a,c). Vessels within the mesenchyme were not in close contact with the respiratory epithelium. In contrast, vasculogenesis was markedly advanced in lungs from the transgenic embryos on E15 (Fig. 3b,d). Lumenal diameter of the blood vessels was increased in the transgenic mice. Furthermore, vessels VEGF EXPRESSION 219 Fig. 2B. High levels of VEGF mRNA were detected in epithelial cell lining dilated and malformed acinar tubules on E15 (c) and E17 (d) in the developing lungs of transgenic embryos and colocalized with the transgene specific t-intron-polyA mRNA expression in (A). The endogenous VEGF mRNA was detected at low levels in epithelial cells lining the distal acinar tubules on E15 (a) and slightly increased on E17 (b) in the lungs from nontransgenic littermates. diffuse pattern surrounding the dilated tubules in lungs from the transgenic embryos on E15 (Fig. 5). lial cells forming the abnormal vessels seen in SP-CVEGF mice were joined by tight junctions. No fenestrations or gaps were observed along the endothelium. Thus, overexpression of VEGF in pulmonary epithelium disrupted the normal interface between the pulmonary epithelial and capillary endothelial cells, inhibiting cytodifferentiation of Type I alveolar cells and maturation of Type II alveolar cells. The expression of VEGF in the pulmonary epithelium increased the lumenal size of pulmonary blood vessels and caused areas of interstitial edema consistent with the known effects of VEGF on vascular permeability. Electron Microscopy During the saccular stage of normal mouse lung development (E17.5–birth), the epithelial cells lining the saccules differentiate, producing mature Type II and Type I cells that line the distal airspaces (Fig. 6). Vascularity increases and endothelial cells come into close proximity to the developing squamous Type I cells. In contrast, epithelial cells of the lungs from transgenic mice remained primarily columnar and failed to form the squamous epithelium characteristic of the normal lung in late gestation (Fig. 6A). The respiratory epithelium was lined predominately by low columnar cells containing extensive glycogen pools and, frequently, lamellar bodies. The lumen of pulmonary blood vessels was enlarged, and the organization of the surrounding stroma was loosely arranged, containing areas of extracellular fluid accumulation, perhaps related to the effect of VEGF on vascular permeability (Fig. 6B). Platelets were frequently observed in dilated, tortuous vessels adjacent to the respiratory epithelium. Endothe- Expression of SP-C and CCSP Genes in the SP-C-VEGF Mice In normal mouse lung, immunostaining of the proSP-C protein in distal airway epithelial cells increases markedly during the canalicular and saccular stage of development in association with other aspects of morphologic maturation and Type II cell cytodifferentiation (Zhou et al., 1996b). CCSP, a marker for Clara cells, is first detected in the mouse conducting airway epithelial cells at E16–17 (Wert et al., 1993; Zhou et al., 220 ZENG ET AL. Fig. 2C. Immunoperoxidase staining for VEGF protein was detected in the epithelial cells of the malformed acinar tubules and saccules in the lungs from SP-C-VEGF transgenic embryos on E15 (c) and E17 (d). The endogenous VEGF protein was also expressed in the epithelial cells of the acinar buds and saccules in the lungs from nontransgenic control embryos on E15 (a) and E17 (b). Bars 5 50 µm. 1996b). These markers were therefore chosen to assess proximal and distal airway epithelial cell differentiation in the SP-C-VEGF mice (Fig. 7). ProSP-C was detected in epithelial cells of acinar tubules from nontransgenic embryos on E15 and E17 and also in epithelial cells of the abnormally dilated tubules seen in the lungs from transgenic mice (Fig. 7A), supporting their identity as terminal acinar tubules. CCSP was present in the bronchiolar epithelium of lungs from both transgenic and nontransgenic embryos on E17 (Fig. 7B), but was not detected in the abnormally dilated acinar tubules in the transgenic mice. Thus, lung morphogenesis was markedly altered in SP-C-VEGF mice, while cytodifferentiation of the respiratory epithelial cells, as assessed by the distribution of CCSP and proSP-C, was only modestly altered by the expression of the transgene. transgene mRNA was restricted to the pulmonary epithelium and was associated with consistent histological alterations in lung morphology that resulted in perinatal death in transgenic offspring. Branching morphogenesis in the SP-C-VEGF mice on E15 in the transgenic embryos was markedly disrupted. Acinar tubules were dilated in association with decreased numbers of terminal buds and decreased pulmonary mesenchyme compared to age-matched controls. Blood vessels were precociously developed and grew in close proximity to the respiratory epithelium. While the expression of SP-C and CCSP in pulmonary epithelial cells were relatively unaltered in the SP-C-VEGF transgenic mice, the differentiation of Type II into Type I epithelial cells was inhibited. Endogeneous VEGF mRNA and protein were detected in the developing respiratory epithelium, while the VEGF receptor, Flk-1, was detected in lung mesenchyme. These findings support a role of VEGF in vessel formation in the developing lung that in turn influences acinar development. The human SP-C promoter has been used to express genes or cDNAs in the pulmonary epithelium in a DISCUSSION In the present study, transgenic mice were generated in which the mouse VEGF164 was directed to the lungs of fetal mice using the promoter/enhancer element derived from the human SP-C gene. Expression of the VEGF EXPRESSION 221 Fig. 3. PECAM-1 staining in the lungs from SP-C-VEGF transgenic mice. The pattern of pulmonary blood vessel distribution was assessed in sagittal sections prepared from SP-C-VEGF transgenic embryos (b,d) and nontransgenic embryos (a,c) on E15 using PECAM-1 (CD31) antibody. The panel on the right is a higher magnification of the field in the window on the left. In lungs from nontransgenic embryos, blood vessels were observed surrounding the distal airways and terminal acinar buds, but remain embedded within the mesenchyme (a,c). In the transgenic developing embryos, the blood vessels are seen in close approximation to the respiratory epithelium (b,d). Note the unusually large lumen of the vessels surrounding the distal epithelial tubules in (d, arrows) and the columnar shape of the epithelial cells in close contact with the blood vessels (d, arrowheads). Bars 5 50 µm. Staining is representative of 2 separate embryos from each group. variety of transgenic mouse models. Transgenic mRNA expression begins as early as E10 and was present only in cells of the distal airway epithelium in transgenic embryos (Wert et al., 1993). The cellular sites of expression of endogeneous VEGF mRNA and protein spatially overlaps with that of the VEGF transgene expressed by SP-C promoter. However, the endogeneous VEGF mRNA was only detectable in respiratory epithelial cells after E15, as determined by in situ hybridization. Therefore, in the present study, the VEGF transgene was likely expressed at higher levels and at an earlier developmental stages in distal airway epithelial cells in SP-CVEGF transgenic embryos. This suggests that the extensive vascularization surrounding the dilated airways of the SP-C-VEGF transgenic mice is related to increased activity or dysregulation of VEGF. This hypothesis is supported by previous studies that have also shown alterations in vascularization in response to VEGF. For example, increased expression of VEGF in avian wing bud mesenchyme increased vascularization (Flamme et al., 1995), while malformation of the vascular networks and neovascularization of normally avascular areas were induced by VEGF in developing quail embryos (Drake et al., 1995). In the present study, abnormalities of vascularity in pulmonary mesenchyme were observed in the SP-C-VEGF transgenic mice, including the production of ectopic blood vessels and the formation of vessels with increased lumenal diameter. These findings demonstrated the paracrine interactions between VEGF with its target cells, the endothelial cells, during vasculogenesis. These findings also support the hypothesis that precise regulation of VEGF by the respiratory epithelium is critical for pulmonary vasculogenesis. The finding that targeted inactivation of a single allele of the VEGF gene in the mouse led to abnormal blood vessel formation and early embryonic lethality suggests that the precise control of VEGF production is critical for normal embryonic vessel development (Carmeliet et al., 1996; Ferrara et al., 1996). 222 ZENG ET AL. Fig. 4. In situ hybridization for Flk-1 and Tie-1 mRNA. Tissue sections from transgenic (b,d) and nontransgenic control (a,c) embryos at E15 were hybridized with 35S labeled antisense Flk-1 (a,b) and Tie-1 (c,d) riboprobes. The higher intensity of the hybridization signal for Flk-1 and Tie-1 was observed in lung mesenchyme of the SPC-VEGF transgenic embryos. No hybridization signal was observed using the sense probe (data not shown). Bars 5 50 µm. The coordinate expression of VEGF and its high affinity receptors during early embryogenesis (Breier et al., 1992; Jakeman et al., 1993; Peters et al., 1993; Breier et al., 1995) suggests that this signal transduction system may be involved in paracrine interactions that result in increasing vascularization in developing organs. In the present study, increased VEGF mRNA expression in respiratory epithelium lead to the increased intensity of the in situ hybridization signals for Flk-1 mRNA, demonstrating a paracrine induction of the VEGF receptor in vivo. Although the endothelial cell specific receptor tyrosine kinase Tie-1 is not a cognate receptor for VEGF, the increased intensity of in situ hybridization signals for Tie-1 mRNA detected in SPC-VEGF transgenic lung may indicate that VEGF is involved in the regulation of Tie-1. Expression of VEGF in the respiratory epithelium of transgenic mice disrupted branching morphogenesis of the lung. These findings are distinct from previous observations in which embryonic pattern formation was relatively unaltered after exposure of avian wing bud and developing quail embryo to VEGF, even though VEGF caused marked malformation of the vasculature in those experimental systems (Drake et al., 1995; Flamme et al., 1995). The changes in lung morphology seen in the SP-C-VEGF mice were highly distinct from those produced by KGF and TGF-b. In contrast to VEGF, KGF and TGF-b are mesenchymal signals that directly affect epithelial differentiation. When expressed in pulmonary epithelium by SP-C promoter, KGF induced cystadenomatoid malformation of the lung (Simonet et al., 1995), while TGF-b markedly arrested lung sacculation and inhibited branching morphogenesis (Zhou et al., 1996a). The disruption of lung morphogenesis seen in the SP-C-VEGF mice on E15 may be related to the precocious or dysregulated growth of pulmonary blood vessels, perhaps altering cell-cell or cell-extracellular matrix interactions that are involved in critical aspects of acinar maturation and sacculation. Decreased a-smooth muscle actin staining surrounding the bronchioles of the transgenic embryos indicated that smooth muscle and myofibroblast differentiation or organization within the mesenchymal cell compartment has been altered. VEGF EXPRESSION 223 Fig. 5. a-Smooth muscle actin staining. Sections of SP-C-VEGF transgenic and nontransgenic control embryos on E15 were stained with an FITC-conjugated a-sma antibody and photographed under fluorescence microscopy. a: Reactivity was observed surrounding the blood vessels (V) and bronchioles (B) (bronchioles, stalks, clefts) of lungs from nontransgenic embryos on E15. Note the lack of staining at the terminal buds and leading edges of the branching epithelial tubules of the acinar (a, arrow). b: In transgenic lung, staining was diffuse in regions surrounding the acinar tubules (B). Condensed staining in the vessels (V) was maintained in lungs from transgenic embryos. Bar 5 100 µm. Myofibroblasts may stabilize branch points of the developing bronchiolar tree and restrict expansion of the proximal bronchioles (Mitchell et al., 1990), a process critical to normal lung morphogenesis. The loss of the stabilizing influence of the smooth muscle cell/myofibroblast cell may therefore contribute to the early dilation and decreased development of acinar structures seen in the lungs of the SP-C-VEGF mice. The observed changes in lung structure may represent the indirect effect of VEGF mediated changes in vascularity that in turn influence branching morphogenesis. The precise temporal and spatial regulation of VEGF or VEGF receptor signaling may therefore be critical to normal growth and morphogenesis of acinar structures. Fig. 6. Lung tissue from E18 transgenic and nontransgenic litter mates was processed for ultrastructural analysis as described in Materials and Methods. A: Alveolar septae from E18 nontransgenic control mice (a) were composed of fibroblasts and capillaries containing red blood cells. The overlying epithelium was composed of squamous Type I cells (solid arrowheads) and cuboidal Type II cells (open arrowheads) that contained a few lamellar bodies and little glycogen. Secreted pulmonary surfactant, composed of lamellar bodies and tubular myelin, was found in the alveolar lumen (arrows). Dilated pulmonary tubules found in the SP-C-VEGF transgenic mice (b) were lined by immature, low-columnar to cuboidal Type II cells containing abundant glycogen (asterisks) and lamellar bodies (arrows), seen at higher magnification in panel c (arrows) along with several multivesicular bodies (arrowhead). Scale bar for panel a 5 10 µm; panel b 5 5 µm; and panel c 5 2 µm. 224 ZENG ET AL. Fig. 6B. In the SP-C-VEGF transgenic mice (a), loosely organized, edematous interstitial tissue is found adjacent to a large, dilated vessel (asterisk). An irregular, tortuous blood vessel, containing red blood cells and numerous platelets (arrows), is seen adjacent and closely apposed to the overlying cuboidal epithelium in an oblique section through the basal side of the epithelium (b). 1 5 endothelial cells; 2 5 epithelial cells. Scale bar for panels a and b 5 5 µm. Epithelial cell differentiation in the lungs from SP-C-VEGF mice was not markedly altered as assessed by immunostaining for proSP-C and CCSP. The increased growth of blood vessels did not enhance epithelial cytodifferentiation. The acinar tubules of the transgenic mice were lined primarily by cuboidal epithelial cells with features of mature Type II cells such as surfactant protein C and lamellar bodies. However, during normal lung development from the canalicular stage to the postnatal period, the increasingly close apposition of capillaries to the respiratory epithelium is associated with thinning of the respiratory epithelium in the periphery of the lung (Burri, 1984), producing a squamous epithelium lined by Type I cells characteristic of the normal postnatal lung. The distal respiratory epithelium remained primarily cuboidal in the SP-C-VEGF mice in spite of the extensive vascularity of the tissue. While the alterations in the pulmonary vasculature induced by VEGF disrupted acinar development and branching morphogenesis, the changes in vascularity did not enhance the respiratory epithelial cell differentiation of Type II cells to form Type I cells. In summary, the expression of VEGF in the respiratory epithelium caused an abnormal pattern of blood vessel distribution in the lung mesenchyme which, in turn, disrupted lung morphogenesis. Since VEGF is expressed in alveolar epithelial cells in the adult lung and is subject to regulation by external stimuli such as hyperoxia (Tuder et al., 1995) and hypoxemia (Maniscalco et al., 1995), precise regulation of VEGF may also be critical to repair following lung injury associated with respiratory distress syndrome and bronchopulmonary dysplasia. MATERIALS AND METHODS Transgene Preparation A cDNA (600 bp) encoding mouse VEGF164 (164 amino acid peptide) was isolated by RT-PCR from total RNA purified from mouse kidney. The murine VEGF164 cDNA sequence was verified by analyzing two independent clones and by comparison with published sequence (Breier et al., 1992). The full length (0.6 kb) cDNA of mouse VEGF164 was then inserted into the EcoRI site of the human SP-C promoter plasmid SP-C-3.7 SV40 t-intron-polyA (Glasser et al., 1991). The proper orientation of the construct was verified by restriction enzyme digestion. The SP-C-VEGF chimeric gene was excised with Ndel and Notl and purified from 0.8% agarose gel by Qiaex Extraction Kit (Qiagen Inc, Chatsworth, CA). Purified DNA was dialyzed extensively into 5 mM Tris-HCl (pH 7.4), 0.1 mM EDTA, and microinjected into fertilized eggs of FVB/N mice. The embryos from VEGF EXPRESSION 225 Fig. 7. Immunohistochemical staining for proSP-C, CCSP. 7A: ProSP-C staining was detected in epithelial cells of acinar buds on E15 (a, arrows) and differentiated Type II cells on E17 (b, arrow) in control embryos. Comparable staining was also seen in the epithelium of the dilated acinar tubules in age matched transgenic embryos (c,d, arrows). pregnant mothers were obtained by hysterectomy at embryonic day 15 (E15) and 17 (E17) for analysis. Transgenic founders were identified by PCR analysis of tail DNA using oligonucleotide primers that amplified a 800 bp fragment spanning the VEGF cDNA and the SV40 derived t-intron-polyA sequence and further confirmed by Southern blot analysis on tail DNA probed with the SV40 derived t-intron-polyA sequence. Tissue Preparation Hysterectomy was performed after lethal injection of pentobarbital. The embryos were weighed and fixed in neutral buffered 4% paraformaldehyde at 4°C for 16 hr. Tissues were dehydrated through a graded series of ethanol solutions and embedded in paraffin. Tissues were sectioned transversely through the thorax (E17) or sagittally (E15) at 5 µm thickness and stained with hematoxylin and eosin. Antibodies and Immunohistochemistry Rabbit antiserum against rat CCSP was the kind gift of Dr. Gurmukh Singh (Dept. of Veterans Affairs Medi- cal Center, Pittsburgh, PA) and used at a dilution of 1:1,000. Rabbit antiserum against human proSP-C (#68514) was used at a 1:1,000 dilution. Rabbit antiserum against human VEGF (A-20; Santa Cruz Biotechnology, Inc., San Diego, CA) was used at a dilution of 1:600. Immunostaining for CCSP, VEGF, and proSP-C was performed as previously described (Vorbroker et al., 1995; Zhou et al., 1996a,b). FITC-conjugated mouse monoclonal antibody against a-smooth muscle actin (Sigma Chemicals, St. Louis, MO) was used at a dilution of 1:50 as previously described (Zhou et al., 1996a). Immunostaining with monoclonal rat-antimouse PECAM-1 (CD31) antibody (Pharmingen, San Diego, CA) was performed based on the procedure for proSP-C immunostaining with minor modifications. Briefly, 5 µm paraffin sections were deparaffinized and rehydrated. After 15 min incubation at room temperature (RT) with 3% H2O2 to quench endogenous peroxidase, a brief incubation with 0.1% trypsin (Difco Laboratories, Detroit, MI), pH 7.6 at 37°C was performed to unmask the antigen. Sections were blocked with 2% normal rabbit serum in phosphate buffered saline with 226 ZENG ET AL. plasmid with mouse VEGF cDNA in pCRScript SK (1) was linearized and used to generate radiolabeled sense and antisense transcripts by in vitro transcription in the presence of 35S-labeled UTP (1310 Ci/mmole, Dupont NEN, Boston, MA) using T3 or T7 polymerase and reagents contained in the Maxscript in vitro transcription kit (Ambion, Inc., Austin, TX). The mouse Flk-1 and Tie-1 partial cDNA clone containing the kinase domain were generated by PCR from the full length Flk-1 and Tie-1 cDNA respectively and used to generate the radiolabeled antisense transcript. In situ hybridization was performed as previously described (Wert et al., 1993) except that the overnight hybridization was carried out at 55°C. Electron Microscopy Fetal lung tissue was minced into 1–2 mm3 pieces, fixed overnight in the cold with 2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M sodium cacodylate buffer (SCB) at pH 7.3, and post-fixed in 1% osmium tetraoxide in 0.1 M SCB for 2 hr at room temperature. The tissue was then dehydrated through a graded series of alcohol solutions, washed in propylene oxide, and infiltrated and embedded in Embed-812 (Electron Microscopy sciences, Ft. Washington, PA). Semithin plastic sections were stained with toluidine blue and photographed with a Nikon Microphot FXA photomicroscope. Ultrathin sections were stained with lead citrate and uranyl acetate and examined with JEOL 100CX electron microscope. Fig. 7B. CCSP was detected in bronchial epithelial cells in lungs from nontransgenic (a) and transgenic embryos on E17 (b). 2% Triton X-100 (PBST) for 2 hr at RT. After overnight incubation with primary antibody (diluted to 1 µg/ml in blocking serum) at 4°C, sections were washed in PBST and incubated with biotinylated rabbit anti-rat antibody (Novocastra Laboratories Ltd, UK), diluted 1:200 in blocking solution, for 30 min at RT. Sections were then washed, incubated with avidin-biotin-peroxidase complex from a Vectastain Elite ABC standard kit (Novocastra Laboratories, UK) diluted in PBST. Reaction products were visualized by incubation with nickelcontaining 3,38-diaminobenzidine, followed by a brief incubation with Tris cobalt to achieve a black precipitate. Sections were then counter-stained with nuclear fast red. In Situ Hybridization In situ hybridization using transgene specific riboprobes generated from SV40 derived t-intron-polyA sequences was performed on tissue sections from transgenic and nontransgenic embryos as previously described (Zhou et al., 1996a). In situ hybridization using VEGF, Flk-1, and Tie-1 riboprobes were performed on tissue sections from transgenic and normal control mouse embryos. The ACKNOWLEDGMENTS The authors thank Dr. Lan Zhou for helpful discussions and Dr. Karen Yaeger, Sherri Proffit, Kevin Kirwin, and Chitta R. Dey for technical assistance during the course of these studies. REFERENCES Adamson IYR. Development of lung structure. In: Crystal RG, West JB, eds. The Lung: Scientific Foundations. New York: Raven Press, 1991;663–670. Berse B, Brown LF, Van De Water L, Dvorak HF, Senger DR. Vascular permeability factor (vascular endothelial growth factor) gene is expressed differentially in normal tissues, macrophages, and tumors. Mol. Biol. Cell 1992;3:211–220. Breier GM, Clauss M, Risau W. Coordinate expression of vascular endothelial growth factor receptor-1 (flt-1) and its ligand suggest a paracrine regulation of murine vascular development. Dev. Dyn. 1995;204:228–239. Breier G, Albrecht U, Sterrer S, Risau W. Expression of vascular endothelial growth factor during embryonic angiogenesis and endothelial cell differentiation. Development 1992;114:521–532. Burri PH. Fetal and postnatal development of the lung. Ann. Rev. Physiol. 1984;46:617–628. Carmeliet P, Ferreira V, Breier G, Pollefeyt S, Kieckens L, Gertsenstein M, Fahrig M, Vandemhoeck A, Harpal K, Eberhardt C, Declercq C, Pawling J, Moons L, Collen D, Risau W, Nagy A. Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele. Nature 1996;380:435–439. deMello DE, Sawyer D, Galvin N, Reid LM. Early fetal development of lung vasculature. Am. J. Respir. Cell. Mol. Biol. 1997;16:568–581. VEGF EXPRESSION DeVries C, Escobedo JA, Ueno H, Houck K, Ferrara N, Williams LT. The fms-like tyrosine kinase, a receptor for vascular endothelial growth factor. Science 1992;255:989–991. Drake CJ, Little CD. Exogenous vascular endothelial growth factor induces malformed and hyperfused vessels during embryonic neovascularization. Proc. Natl. Acad. Sci. USA 1995;92:7657–7661. Dvorak HF, Brown LF, Detmar M, Dvorak AM. Vascular permeability factor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis. Am. J. Pathol. 1995;146:1029–1039. Ferrara N, Carver-Moore K, Chen H, Dowd M, Lu L, Sue O’Shea K, Powell-Braxton L, Hillan KJ, Moore MW. Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature 1996;380:439–442. Ferrara N, Houch KA, Jakeman LB, Winer J, Leung DW. The vascular endothelial growth factor family of polypeptides. J. Cell. Biochem 1991;47:211–218. Flamme I, von Reutern M, Drexler HCA, Syed-Ali S, Risau W. Overexpression of vascular endothelial growth factor in the avian embryo induces hypervascularization and increased vascular permeability without alterations of embryonic pattern formation. Dev. Biol. 1995;171:399–414. Glasser SW, Korfhagen TS, Wert SE, Bruno MD, McWilliams KM, Vorbroker DK, Whitsett JA. Genetic element from human surfactant SP-C gene confers bronchiolar-alveolar cell specificity in transgenic mice. Am. J. Physiol. 1991;261:L349–356. Guzowski DE, Blau H, Bienkowski RS. Extracellular matrix in developing lung. In: Scarpelli, EM ed. Pulmonary Physiology: Fetus, Newborn, Child, and Adolescent. Philadelphia, Lea & Febiger 1990;83–105. Hilfer SR. Development of terminal buds in the fetal mouse lung. Scanning Electron Microscopy 1983;III:1387–1401. Hilfer SR. Morphogenesis of the lung: control of embryonic and fetal branching. Annu. Rev. Physiol. 1996;58:93–113. Jakeman LB, Armanini M, Phillips HS, Ferrara N. Developmental expression of binding sites and messenger ribonucleic acid for vascular endothelial growth factor suggests a role for this protein in vasculogenesis and angiogenesis. Endocrinology 1993;133:848–859. Korfhagen TR, Glasser SW, Wert SE, Bruno MD, Daugherty CC, Mcneish JD, Stock JL, Potter SS, Whitsett JA. Cis-acting sequences from a human surfactant protein gene confer pulmonary-specific gene expression in transgenic mice. Proc. Natl. Acad. Sci. USA 1990;87:6122–6126. Korhonen J, Polvi A, Partanen J, Alitalo K. The mouse tie receptor tyrosine kinase gene: expression during embryonic angiogenesis. Oncogene 1994;9:395–430. Maniscalco WM, Watkins RH, Finkelstein J, Campbell MH. Vascular endothelial growth factor mRNA increases in alveolar epithelial cells during recovery from oxygen injury. Am. J. Respir. Cell Mol. Biol. 1995;13:377–386. Millauer B, Wizigmann-Voos S, Schnürch H, Martinez R, Møller NPH, Risau W, Ullrich A. High affinity VEGF binding and developmental expression suggest Flk-1 as a major regulator of vasculogenesis and angiogenesis. Cell 1993;72:835–846. Minoo P, King RJ. Epithelial-mesenchymal interactions in lung development. Annu. Rev. Physiol. 1994;56:13–15. 227 Mitchell JJ, Peynolds SE, Leslie KO, Low RB, Woodcock-Mitchell J. Smooth muscle cell markers in developing rat lung. Am. J. Respir. Cell Mol. Biol. 1990;3:515–523. Monacci WT, Merrill MJ, Oldfield EH. Expression of vascular permeability factor/vascular endothelial growth factor in normal rat tissues. Am. J. Physiol. 1993;264:c995–1002. Mustonen T, Alitalo K. Endothelial receptor tyrosine kinases involved in angiogenesis. J. Cell Biol. 1995;129:895–898. Pardanaud L, Yassine F, Dieterlen-lievre F. Relationship between vasculogenesis, angiogenesis and haemapoiesis during avian ontogeny. Development 1989;105:473–485. Peters K, Werner S, Liao X, Wert S, Whitsett J, Williams L. Targeted expression of a dominant negative FGF receptor blocks branching morphogenesis and epithelial differentiation of the mouse lung. EMBO J 1994;13:3296–3301. Peters KG, DeVries C, Williams LT. Vascular endothelial growth factor receptor expression during embryogenesis and tissue repair suggests a role in endothelial differentiation and blood vessel growth. Proc. Natl. Acad. Sci. USA 1993;90:8915–8919. Quinn TP, Peter KG, DeVries C, Ferrara N, Williams LT. Fetal liver kinase 1 is a receptor for vascular endothelial growth factor and is selectively expressed in vascular endothelium. Proc. Natl. Acad. Sci. USA 1993;90:7533–7537. Simonet WS, DeRose ML, Bucay N, Nguyen HQ, Wert SE, Zhou L, Ulich TR, Thomason A, Danilenko DM, Whitsett JA. Pulmonary malformation in transgenic mice expressing human keratinocyte growth factor in the lung. Proc. Natl. Acad. Sci. USA 1995;92:12461– 12465. Ten Have-Opbroek AAW. Lung development in the mouse embryo. Exp. Lung Res. 1991;17:111–130. Tuder RM, Flook BE, Voelkel NF. Increased gene expression for VEGF and the VEGF receptors KDR/FLK-1 in lungs exposed to acute or chronic hypoxia. J. Clin. Invest 1995;95:1798–1807. Vecchi A, Garlanda C, Lampugnani MG. Monoclonal antibodies specific for endothelial cells of mouse blood vessels. Their application in the identification of adult and embryonic endothelium. Eur. J. Cell Biol. 1994;63:247–254. Vorbroker DK, Profitt SA, Nogee LM, Whitsett JA. Aberrant processing of surfactant protein C in hereditary SP-B deficiency. Am. J. Physiol. 1995;268;(4 pt 1):L647–656. Weaver TE, Whitsett JA. Function and regulation of expression of pulmonary surfactant-associated proteins. Biochem. J. 1991;273: 249–264. Wert SE, Glasser SW, Korfhagen TR, Whitsett JA. Transcriptional elements from the human SP-C gene direct expression in the primordial respiratory epithelium of transgenic mice. Dev. Biol. 1993;156:426–443. Xie Y, Muller WA. Molecular cloning and adhesive properties of murine platelet/endothelial cell adhesion molecule 1. Proc. Natl. Acad. Sci. USA 1993;90:5569–5573. Zhou L, Dey CR, Wert SE, Whitsett JA. Arrested lung morphogenesis in transgenic mice bearing an SP-C-TGF-b1 chimeric gene. Dev. Biol. 1996a;175:227–238. Zhou L, Lim L, Costa RH, Whitsett JA. Thyroid transcription factor-1, hepatocyte nuclear factor-3b, surfactant protein B, C, and Clara cell secretory protein in developing mouse lung. J. Histochem. Cytochem 1996b;44:1183–1193.