DEVELOPMENTAL DYNAMICS 207:439-449 (1996) Transcripts for Two Members of the Transforming Growth Factor-p Superfamily BMP-3 and BMP-7 are Expressed in Developing Rat Embryos HIROSHI TAKAHASHI AND TOHRU IKEDA Department of Pathology, Tokyo Metropolitan Institute of Gerontology, Itabashi-ku, Tokyo 173 (H.T.), and Department Oral Pathology, School of Dentistry, Showa University, Shinagawa-ku, Tokyo 142 ( T J J , Japan ABSTRACT Bone morphogenetic proteind (BMP-3) and BMP-7 are members of the transforming growth factor p superfamily that have been implicated in the formation of cartilage and bone. Using in situ hybridization, we localized mRNAs for BMP-3 and BMP-7 during organogenesis in rats. Both mRNAs were expressed in a variety of cells, in particular, in the developing hair follicle, tooth, kidney, and lung tissues, in which reciprocal epithelial-mesenchymal interactions are essential. In some tissues, the distribution of BMP-3 and BMP-7 mRNAs overlapped. In other tissues, the patterns of expression were quite different. Moreover, the site of expression of the transcripts changed from one cell type to another during organogenesis. These results suggest that BMP-3 and BMP-7 play important roles in organogenesis and that the differential patterns of their expression might reflect their distinct roles in embryogenesis. o 1996 Wiley-Liss, Inc. Key words: Bone morphogenetic protein-3, Bone morphogenetic protein-7, Epithelialmesenchymal interaction, In situ hybridization INTRODUCTION The members of the transforming growth factor beta (TGF-P) superfamily are polypeptide signaling molecules that play important roles in growth and development (for review, see Massague, 1990). Based on sequence homology, the members of the TGF-P superfamily have been subdivided into several subfamilies: the TGF-p subfamily; the activin subfamily; the DVR group and others. The DVR group has been further subdivided into the dpp subfamily (bone morphogenetic protein-2 (BMP8), BMP-4, dpp), the 6 0 8 subfamily (BMP-5, BMP-6IVgr-1, BMP-7/0P-1, BMP-8/ OP-2, and 60A) and other molecules (BMPS, Vg-1, GDF-1, Vgr-2IGDF-3, GDF-9, nodal, ~ O F S U (for ~ ) review, see Kingsley, 1994). BMPs were originally identified in extracts of bone on the basis of their ability to induce ectopic bone formation (Wozney et al., 1988). In addition to their presence in developing bone and cartilage, BMP-2, BMP-4 and BMP-6 are expressed in a variety of cells in various 0 1996 WILEY-LISS, INC. of tissues, in particular, in the developing organs in which reciprocal epithelial-mesenchymal interactions are essential (Lyons et al., 1989a,b, 1990; Jones et al., 1991). Recent studies indicate that BMPs play important roles in various stages of embryogenesis (for review, see Lyons et al., 1991). BMP-4 is involved in the mesodermal induction as a posterior ventralizing signal (Jones et al., 19921, and it also mediates epithelialmesenchymal interactions during early tooth development (Vainio et al., 1993). BMP-2 may be an epithelial signal in early limb development (Niswander and Martin, 1993). Mutant short-eared mice have a broad range of skeletal defects which include alterations in the size, shape, and number of many different skeletal elements, as well as marked reduction in the size of the external ear. A genetic study shows that the mouse short ear locus is associated with defects in BMP-5 (Kingsley et al., 1992). Therefore, each BMP molecule seems to have distinct roles in development. Previous studies have shown that mRNAs for BMP-2, BMP-4, and BMP-6 are expressed during embryogenesis in spatially and temporally specialized patterns, suggesting that coordinated expression of BMPs is needed for the regulation of organogenesis (Lyons et al., 1989a,b, 1990; Jones et al., 1991). However, little is known about the role of BMP-3 and BMP-7 in embryonic development. In order better to understand the precise role of each members of the TGFP superfamily in cell-cell interactions and embryonic organogenesis in vivo, it is necessary to define the tissues and cell types that express the different proteins. In the present study, we have examined the patterns of expression of BMP-3 and BMP-7 mRNA in the developing rat tissues by in situ hybridization. We discuss these patterns of expression in a comparison with those of BMP-2, -4, and -6 mRNAs. Received August 22, 1995; accepted July 22, 1996. Address reprint requestdcorrespondence to Hiroshi Takahashi’s current address: Laboratory of Histochemistry, Mitsubishi Kasei Institute of Life Sciences, 11Minamiooya, Machida-shi, Tokyo-194, Japan. 440 TAKAHASHI AND IKEDA RESULTS Expression of mRNAs for BMP-3 and BMP-7 in the Embryos Figure 1 showed that two BMP-3 transcripts of approximately 6.4 kb and 2.9 kb, and two BMP-7 transcripts of approximately 4.6 kb and 2.7 kb, were detected in Northern blot analysis using embryonic day (El15rat embryos, The existence of two BMP-7 transcripts was consistent with the mouse data reported previously (Ozkaynak et al., 1991). These data also showed that each probe for BMPs specifically recognized each molecule. To examine the localization of BMP-3 and BMP-7 mRNA in detail, in situ hybridization was performed. Before presenting the pattern of expression of these molecules in each organ, representative pictures indicating differences of expression pattern of these BMPs were shown in Figure 2. In E l 8 embryos, strong expression of BMP-3 mRNA was detected in bone and cartilage, developing tooth, nasal epithelium, lung, kidney and small intestine (Fig. 2A). In contrast, the strong signals of BMP-7 transcripts were observed in the skin, nasal epithelium, hair follicle, developing tooth, heart, adrenal gland, kidney, esophagus, meningeal tissue, inner ear, and weaker in the developing intestine, lung and bone and cartilage (Fig. 2C).Although the distribution of these two BMPs overlapped in some tissues such as kidney, lung, and developing tooth, cell types expressing each BMPs were different. For instance, expression of BMP-3 showed tubular pattern as BMP-3 was expressed in the developing tubules in the kidney (Fig. 2A, arrowhead). On the other hand, expression of BMP-7 showed spotty pattern as BMP-7 was expressed in the developing glomeruli (Fig. 2C, arrowhead). No hybridization signal was seen in the sections hybridized with sense probes for BMP-3 or BMP-7 (Fig. 2B and D). Expression of mRNAs for BMP-3 and BMP-7 in Skin and Hair Follicles In rats from embryonic day 11( E l l ) to postnatal day 5 (P5),BMP-3 mRNA was not detected in developing skin. During development of hair follicle, low-level of expression of the mRNA was detected in the mesenchyma1 condensation adjacent to the epidermis and in the outer and inner root sheath at E l 5 (data not shown). At E18, BMP-3 mRNA was localized in the dermal sheath (Fig. 3A). This pattern of expression of BMP-3 mRNA continued a t least until P5, although the intensity of the signal declined (Fig. 3B). Expression of BMP-3 mRNA was also detected in the interdigital mesoderm in early limb buds (data not shown). At E l l , BMP-7 mRNA was detected in the somites and epithelium adjacent to the somites. At E13, it was weakly expressed in the oral epithelium (data not shown). At E l 5 and El$,it was strongly expressed in basal cell layers (Fig. 3C). No or little signal was detected in the suprabasal cells in the epidermis. Expression of BMP-7 mRNA declined postnatally (Fig. 3D). Fig. 1. Northern blot analysis of rat BMP-3 and BMP-7 RNA blots were prepared using 1 pg of poly(A)+RNA isolated from E l 5 rat embryos. Two transcripts of 6.4 kb and 2.9 kb in rat BMP-3 and two transcripts of 4.6 k and 2.7 kb in rat BMP-7 are observed. Smaller transcripts in each lane are more abundant than larger transcripts. During development of hair follicle, BMP-7 mRNA was not detected prior to E15. At El$, higher-level expression of BMP-7 mRNA was detected in the inner and outer root sheaths and lower level of expression was detected in the dermal sheath (Fig. 3C). Later in hair follicle development, BMP-7 mRNA was detected in the dermal papilla and in the outer root sheath, with the highest levels being found in the lower half of each follicle (Fig. 3D). Expression of BMP-3 and BMP-7 mRNAs in the Developing Tooth No expression of BMP-3 or BMP-7 mRNA was detected at the tooth bud stage. At the cap stage of tooth development, high level of expression of BMP-3 mRNA was detected in the dental follicle, in particular, adjacent to the outer enamel epithelium (Fig. 3E). This pattern of expression of BMP-3 mRNA persisted through the early and late bell stage of tooth development (Fig. 3F). Strong signals due to BMP-7 mRNAs were seen in the inner enamel epithelium with relatively weak signals in the dental papilla below the enamel epithelium during the cap stage and the bell stage of tooth development (Fig. 3G). At the bell stage, expression of mRNA was detected in the inner enamel epithelium, stratum intermedium, and odontoblasts (Fig. 3H). Expression of BMP-3 and BMP-7 mRNAs in the Nasal Epithelium At E18, expression of BMP-3 mRNA was detected in the mesenchyme adjacent to the folded nasal epithe- BMP-3 AND BMP-7 IN RAT EMBRYOS Fig. 2. Expression of BMP-3 and BMP-7 mRNAs in €18 rat embryos. Parasagittal embryonic sections hybridized with BMP-3 antisense probe (A) or sense probe (B). BMP-3 transcripts are localized in the developing bone and cartilage, tooth, nasal epithelium, skin and hair follicle, lung, kidney, and small intestine (A). Parasagittal embryonic sections hybridized with BMP-7 antisense probe (C) or sense probe (D). Strong signals 441 of BMP-7 mRNA can be detected in the skin, nasal epithelium, hair fol- licle, developing tooth, heart, adrenal gland, kidney, esophagus, meningeal tissue, inner ear, and weaker ones in the developing intestine, lung, and bone and cartilage (C). No obvious signal is recognized in B and D. Expressions of BMP-3 and BMP-7 in the kidney are indicated (A$, arrowheads). 442 TAKAHASHI AND IKEDA Fig. 3. Expression of BMP-3 and BMP-7 rnRNAs in the developing skin and hair follicle, and tooth. Skin and hair follicle hybridized with BMP-3 probe at E l 8 (A) and P5 (B). A: BMP-3 transcripts are localized in the dermal sheath (arrow) and in the mesenchyme adjacent to the epithelium (arrowhead). No hybridization signal can be seen in the epidermis. B: BMP-3 mRNA is located in the dermal sheath (arrow). Skin and hair follicle hybridized with BMP-7 probe at E l 8 (C) and P5 (0). C: BMP-7 transcripts are detected in the innerjouter root sheath and the dermal sheath at E18. Strong signal can be seen in the basal cell layer of the epidermis. D: BMP-7 RNAs are seen in the outer root sheath, espe- cially in the lower half of the follicle (arrow). Developing tooth hybridized with BMP-3 probe at E l 8 (E) and PO (F). Strong signal of BMP-3 is seen in the dental follicle adjacent to the outer enamel epithelium at the cap stage (E) and bell stage (F). No signal can be detected in the enamel epithelium nor dental papilla. Developing tooth hybridized with BMP-7 probe at E l 8 (0)and P5 (H).BMP-7 RNA is expressed in the inner enamel epithelium (arrow) and dental papilla (arrowheads) at cap stage (G). BMP-7 expression is located in the inner enamel epithelium (arrow), and odontoblasts (arrowheads) at bell stage (H). Bars = 100 pm (AE,G,H); 500 pm (F). BMP-3 AND BMP-7 IN RAT EMBRYOS lium (Fig. 4A). BMP-7 mRNA was detected in some parts of the mesenchyme below the nasal epithelium a t E l 8 (Fig. 4B). Expression of BMP-7 mRNA declined postnatally. Expression of BMP-3 and BMP-7 mRNAs in the Developing Lung BMP-3 mRNA was first detected in the epithelium and mesenchyme a t E l 5 (data not shown). Subsequently, it was detected in the bronchial epithelium (Fig. 4C), with little or none in the alveolar epithelium (EM). BMP-7 mRNA was not expressed in the lung prior to E15. At E15, low level of expression of BMP-7 mRNA was detected in the bronchial epithelium and the mesenchyme around the bronchus (data not shown). During development, expression of BMP-7 mRNA declined gradually. Expression of BMP-3 and BMP-7 mRNAs in the Digestive Organs At E15, low-level expression of BMP-3 mRNA was found in the mesenchymal tissues around the epithelium (data not shown). At E18, BMP-3 mRNA was strongly expressed in the mesenchyme at the top of the villi of the small intestine (Fig. 4D). Weaker expression of BMP-3 mRNA was found in the surrounding mesenchyme (Fig. 4D). At E18, BMP-3 mRNA was expressed in the mesenchyme adjacent to the epithelium of the esophagus (data not shown). At E13, low level of expSession of BMP-7 mRNA was first detected in the epithelium of the intestine (data not shown), and this pattern of expression continued until E15. At E18, BMP-7 mRNA was located in the mesenchyme, with especially strong signals at the base of the villi (Fig. 4F). A strong signal was detected in the epithelium (in particular, in the basal cell layer) of the esophagus (Fig. 4E). Expression of BMP-3 and BMP-7 mRNAs in the Kidney and Adrenal Gland Prior to E15, expression of BMP-3 mRNA was first detected in the mesonephric tubules (data not shown). At E18, BMP-3 mRNA accumulated specifically in some parts of the developing tubules (Fig. 4G). At E13, BMP-7 mRNA was first detected in the tubular portion (data not shown). Strong signals of BMP-7 were detected in the cap portion (glomerular epithelium), which surrounds the developing glomerulus, and weaker signals were found in the tubular portion of the developing kidney (Fig. 4H,I). Expression of BMP-7 mRNA was also detected in the developing adrenal gland (Fig. 4H). Expression of BMP-3 and BMP-7 mRNAs in Developing Bone and Cartilage BMP-3 mRNA was expressed in the perichondrium (Fig. 5A,B). It was expressed in the osteoblasts and osteocytes of developing endochondral and intramembranous bones (Fig. 5E,F and I,J), but it was absent 443 from the hypertrophic chondrocytes (Fig. 5A,B and E,F). By contrast, BMP-7 mRNA was found in resting chondrocytes, in proliferating chondrocytes and hypertrophic chondrocytes in cartilage and in endochondral bone (Fig. 5C,D and G,H). In the intramembranous bone, BMP-7 was detected in osteoblasts and osteocytes (Fig. 5K,L). Expression of BMP-3 and BMP-7 mRNAs in Other Organs BMP-7 mRNA was expressed in the ducts of the sublingual gland and the ducts of the minor salivary gland in the tongue, but not in acini (Fig. 6A,B). No BMP-3 mRNA was detected in the heart. From Ell, BMP-7 mRNA was expressed in both the atria and the ventricles of the heart (Fig. 6C). This pattern continued postnatally. During the early postnatal period, BMP-3 mRNA was specifically expressed in the postomitotic neurons in the entorhinal cortex (Fig. 6D). BMP-7 mRNA, but not BMP-3 mRNA, was expressed in the meningeal tissues (Fig. 6E). This pattern of expression continued from E l 5 to adulthood. In the brain on E13,expression of BMP-7 mRNA was detected in some neuronal stem cells in the diencephalon (data not shown). DISCUSSION It is important to compare the patterns of expression of BMPs because these patterns should reflect the distinct roles of various BMPs in embryogenesis. Previous studies have shown that BMP-2, -4, and -6 mRNAs are expressed during embryogenesis in spatially and temporally restricted patterns (Lyons et al., 1989a,b, 1990; Jones et al., 1991). However, little information has been available on the localization of BMP-3 and BMP-7 mRNAs in the developing organs. In this study, we used in situ hybridization to localize BMP-3 and BMP-7 mRNAs in developing rat embryos. Each BMP was expressed in a variety of cells in various tissues, in particular, in the organs in which reciprocal mesenchymal-epithelial interactions are essential. Our results suggest that BMP-3 and BMP-7 play important roles in organogenesis. Distribution of human OP-l(BMP-7) protein was examined with an antibody against this protein (Vukicevic et al., 1994). The organs in which BMP-7 was detected are almost identical to the organs in which BMP-7 mRNA was detected in our study, although precise localizations in the various organs differ between the protein and the mRNA. The OP-1 (BMP-7) protein was localized in the basement membrane (rich in type IV collagen) of bronchi in the lung, and in the basement membrane between the dermis and epidermis of the skin. By contrast, BMP-7 mRNA was detected in the bronchial epithelium and the surrounding mesenchyme, as well as in the basal cells of the epidermis in our data. It is possible that the distribution of the protein is different from that of the mRNA. OP-l(BMP-7) has high affinity for type IV col- 444 TAKAHASHI AND IKEDA Fig. 4. Expression of BMP-3 and BMP-7 mRNAs in the nasal epithelium, lung, esophagus, small intestine, kidney and adrenal gland. Nasal epithelium hybridized with BMP-3 (A) and BMP-7 (B) probes at Ei8. Both BMP-3 and BMP-7 transcripts are located in the mesenchyme under nasal epithelium. Signal of BMP-3 mRNA is strong in the folded portion. Developing lung hybridized with BMP-3 probe at El8 (C). BMP-3 signal can be seen in the bronchial epithelium (arrow), not in the alveolar epithelium. D: Small intestine hybridized with BMP-3 probe at Ei8. Strong signal is detected at the top of the villi (arrows) and weak signal in the surrounding mesenchyme. E: Esophagus hybridized with BMP-7 probe at El 8. Strong signal is detected in the basal cell layer of the esophageal epithelium (arrow). F: Small intestine hybridized with BMP-7 probe at E18. BMP-7 transcripts are located in the mesenchyme at the base of the villi (arrows). 0: Developing kidney hybridized with BMP-3 probe. BMP-3 is expressed in some portion of the developing tubules (arrows). H,I: Developing kidney and adrenal gland hybridized with BMP-7 probe. Strong signal can be seen in the cap portion (glomerular epithelium; arrows in I) and weak one in the tubular portion (arrowheads in I).BMP-7 is also expressed in the adrenal gland (arrow in H). Bars = 100 pm (A-F,H,I); 50 pm (G). BMP-3 AND BMP-7 IN RAT EMBRYOS Fig. 5. Expression of BMP-3 and BMP-7 mRNAs in developing bone and cartilage A, C, E,G, I, K are brightfield figures. 6, D, F, H,J, L are darkfield figures corresponding to A, C, E, G, I, K, respectively. A and B: Developing cartilage hybridized with BMP-3 probe. Strong signal of BMP-3 transcripts is seen in the perichondrium (arrow). C and D: Developing cartilage hybridized with BMP-7 probe. BMP-7 transcripts are seen in the perichondrium and hypertrophic chondrocytes. E and F: Developing endochondral bone hybridized with BMP-3 probe. Signal of EMP-3 mRNA is detected in perichondrium (arrowheads) and osteoblasts (arrow). Note that no signal can be detected in hypertrophic chondrocyle. 445 G and H: Developing endochondral bone hybridized with BMP-7 probe. Signal is detected not only in the proliferating chondrocytes and resting chondrocytes, but also in hypertrophic chondrocytes. BMP7 mRNA transcripts are also seen in the osteoblasts. I and J: Developing intramembranous bone hybridized with BMP9 probe. Strong signal is detected in the osteoblasts and osteocytes (arrows). K and L: Developing intramembranous bone hybridized with BMP-7 probe. Strong signal is detected in the osteoblasts and osteocytes (arrows). Bars = 100 pm (A, C,E, G, I, K). 446 TAKAHASHI AND IKEDA Fig. 6. Expression of BMP-3 and BMP-7 mRNA in various organs. A,B: BMP-7 transcripts are detected in the duct of sublingual gland (arrows in A) and the duct of minor salivary gland (B) in PO. Note that there is no signal in the acinus. C: BMP-7 expression in the heart. BMP-7 lagen (Vukicevic et al., 1994). Thus, BMP-7 might be secreted into the surrounding mesenchyme or the secreted protein might enter into the blood stream and be trapped at sites with high affhity for it. Another possibility is that BMP-7 mRNA is not translated into protein in some tissues. It is known that localization of the DVR-6 (BMP-6)protein is different from that of the DVR-6 (BMP-6) mRNA. Moreover, Wall et al. (1993) failed to detect DVR-6 (BMP-6) protein in a cell line that expressed DVR-6 RNA. It seems likely that posttranscriptional regulation or modification of BMP-6 oc- mRNA is detected in both atrium and ventricle at E15. D: BMP-3 is expressed in the neurons (arrow) of the entorhinalcortex at PO. E: Strong signal of BMP-7 is detected in the meningeal tissues. Bars (A,B,E); 500 pm (C,D). = 100 pm before expression of BMP-2 in the ectodermal placode. Expression of BMP-4 in the mesenchyme is transient; no expression has been detected at the stage of which the dermal papilla is formed (Jones et al., 1991; Ikeda and Takahashi, unpublished data). In the present study, BMP-3 mRNA was also expressed in the mesenchymal condensation below the ectodermal placode. Thus, both BMP-3 and BMP-4 might play an important role in the initial events in hair development. Furthermore, BMPS mRNA was expressed in the dermal sheath in the immature and mature follicle. By contrast, BMP-7 CWS. mRNA was expressed in the inner and outer root From our data and the data reported previously (Ly- sheaths and the dermal sheath of the immature follicle. ons et al., 1989a,b, 1990; Jones et al., 19911, it appears In the mature follicle, expression of BMP-7 mRNA was that BMP-2, -3, -4, -6 and -7 are expressed in organs detected in the dermal papilla and the outer root sheath. that develop through mesenchymal-epithelial interac- No expression of BMP-4 or BMPS mRNA was observed tions (e.g., hair follicles, tooth buds, kidneys and in the mature hair follicle (Jones et al., 1991;Ikeda and lungs). Therefore, we shall discuss the patterns of their Takahashi, unpublished data). The other members of expression and their possible roles in organogenesis. the TGF-f3 superfamily and fibroblast growth factors The development of hair and whisker follicle involves (FGFs) are expressed in immature and mature hair a sequence of reciprocal interactions between the epi- follicle (Pelton et al., 1990; du Cros et al., 1993). It has dermis and underlying dermis (see review, Hardy, been reported that activin, BMP-4 and FGF are re1992). A previous report showed that expression of quired for mesoderm induction (for review, see Sive, BMP-4 is the earliest event in the hair development. 1993). Also, it is known that FGF-4 and BMP-2 have BMP-4 is expressed in the mesenchymal condensation opposite effects on limb growth: FGF-4 stimulates the BMP-3 AND BMP-7 IN RAT EMBRYOS proliferation of mesenchyme in the early development of mouse limbs, while BMP-2 inhibits limb growth (Niswander and Martin, 1993). Not only interactions among members of the TGF-P superfamily but also interactions between the TGF-P superfamily and the FGF family (or other molecule) must be required for the appropriate development of hair. Only BMP-6 is known to be expressed in the thickened epidermis in prenatal skin and expression of BMPS protein continues in the supra-basal cell layers of the skin until birth (Lyons et al., 1989b; Wall et al., 1993). In the present study, BMP-7 mRNA was localized in the basal cell layer of the epidermis in the developing skin, and the level of expression declined postnatally. The presence of BMP-7 mRNA in the basal cells of the epidermis and its absence from the suprabasal layers suggest that BMP-7 might act as an autocrine mitogen for skin keratinocytes. By contrast, BMP-6 is expressed in the differentiated, postmitotic cells. Thus, BMPS might act as a differentiation factor. Tooth development is also controlled by reciprocal epithelial-mesenchymal interactions. At the cap stage, BMP-4 mRNA is detectable in mesenchymal cells of the dental papilla below the enamel epithelium. At the late bell stage, BMP-4 mRNA is expressed in both differentiating (mesenchymal) odontoblasts and differentiating (epithelial) ameloblasts. On the other hand, BMP-2 is expressed in the central cells of dental papilla at the cap stage and in ameloblasts and preodontoblasts transiently a t the bell stage (Vainio et al., 1993; Ikeda and Takahashi, unpublished data). Patterns of expression of BMP-3 and BMP-7 are very different from those of BMP-4 and BMP-2. No expression of BMP-3 or BMP-’7 mRNA was detected before the cap stage. Expression of BMP-3 mRNA was detected in the dental follicles, in particular, adjacent to the outer enamel epithelium and this pattern of expression persisted to the bell stage. BMP-7 mRNA was strongly expressed in the inner enamel epithelium and relatively more weakly in the dental papilla below the enamel epithelium during the cap stage and the early bell stage of tooth development. In late bell stage, expression of BMP-7 mRNA was detected in the inner enamel epithelium and odontoblasts. Coordinate expression of BMP-7, -4, -2 might regulate the differentiation of ameloblasts and odontoblasts in an autocrine or paracrine manner. By contrast, BMP-3 might be involved in follicular formation during the tooth development andlor in the differentiation of the outer enamel epithelium. The morphogenesis of the kidney is a prime example of the importance of the mesenchymal-epithelial interactions. During glomerular development, BMP-7 mRNA was strongly expressed in the cap portion; the glomerular epithelium. BMP-4 mRNA was also located in the cap portion, although expression of BMP-4 mRNA was also observed in mesenchymal tissues around the tubules (Ikeda and Takahashi, unpublished 447 data). BMP-3 and BMP-6 are not expressed in the developing glomerulus. BMP-7 and BMP-4 might play important roles in the formation of the glomerulus. Expression of BMP-3 mRNA was localized within a restricted portion of the developing tubule. Weak expression of BMP-7 mRNA was seen in the tubular portion. BMP-2 mRNA was expressed in the mesenchyme of the medulla (Ikeda and Takahashi, unpublished data). Experiments using organotypic culture and antisense oligonucleotides or antibodies should help to clarify the role of each BMP in renal development. Patterns of expression of BMPs were also distinct in the digestive organs. BMP-3 mRNA was localized in the mesenchyme at the top of villi in the small intestine. BMP-7 mRNA was strongly expressed in the mesenchyme at the base of the villi. Expression of BMP-4 mRNA was seen diffusely in the surrounding mesenchyme. No expression of BMP-2 has been detected in developing digestive organs (Ikeda and Takahashi, unpublished data). The mesenchyme of the developing gut is not homogeneous. For example, the level of expression of the extracellular matrix differs between the top of the villi, the base of the villi and the surrounding mesenchyme of the gut. Before formation of villi, laminin, type IV collagen and nidogen are present a t the epithelial/mesenchymal interface, whereas fibronectin and type I11 procollagen are found throughout the mesenchyme. During the formation of the villi, fibronectin and type I11 procollagen disappear transiently from the mesenchyme a t the top of the villi. (Simon-Assmann et al., 1986). Moreover, each BMP has a different ability to induce cell adhesion molecules in mesenchymal and neural cells (Perides et al., 1992, 1993, 1994). This observation suggests that BMPs might differentially affect the extracellular matrix and differential expression of BMPs might contribute to formation of the specific shape of the organ. Previous studies and our data show that patterns of expression of BMPs overlap, but there are some differences (Lyons et al., 198913,1990;Ikeda and Takahashi, unpublished data). BMP-2, BMPS, and BMP-7 mRNAs ‘areexpressed in hypertrophic chondrocytes, but BMP-3 or BMP-4 mRNAs are not. BMP-3, BMP-7 and BMP-2 mRNAs are expressed in osteoblasts, where expression of BMP-4 and BMPS mRNA cannot be detected. Furthermore, human OP-l (BMP-7) protein is distributed in hypertrophic chondrocytes, osteoblasts, and the periosteum (Vukicevic et al., 1994). These distributions suggest different physiological roles for the various BMPs in skeletogenesis. To clarify the physiological functions of BMPs more precisely, their interactions should be studied. Concluding Remarks The data presented in this study provide information relevant to identification of the functions of the TGF-P superfamily in morphogenesis. BMP-3 and BMP-7 mRNAs were detected in various tissues, in particular, in organs in which reciprocal mesenchymal-epithelial 448 TAKAHASHI AND IKEDA exposed to the imaging Plate for 24 hours and the result was analysed using a Biolmage Analyzer (BAS 2000, Fuji Photo Film Co., Japan). [3SSlUTP-labelledsinglestranded RNA probes for in situ hybridization were synthesized by in vitro transcription. Treatment of the slides and hybridization conditions were the same as described previously (Ikeda et al., 1992)and the protocol was essentially that of Nomura et al. (1988) In brief, sections were pretreated with proteinase K (BoehringEXPERIMENTAL PROCEDURES er-Mannheim, Germany) and acetylated with acetic anPreparation of Tissues hydride/triethanolamine-HC1.Tissue sections were alEmbryos were obtained from female rats of Wistar lowed to hybridize with probes overnight in a moist strain. Embryonic age was determined on the basis of chamber a t 55°C. After hybridization, the sections were detection of the vaginal plug (day 0).Embryonic tissues treated with 12.5 pgh1 of RNAse A in 10 mM Tris-HC1, were fixed with 4% paraformaldehyde in phosphate pH 7.6,l mM EDTA and 500 mM NaCl at 37°C for 30 buffer a t 4°C for 24 hr, dehydrated and embedded in min, and washed with 2 x SSC and 0.2x SSC at 55°C. paraffin. Tissues from postnatal rats were also fixed in Dried slides were dipped in NTB-3 emulsion (Eastman 4% paraformaldehyde solution by cardiac perfusion Kodak: Rochester, NY) that had been diluted (1:l) with and embedded in paraffin. Four-pm-thick serial sec- 2% glycerol, dried and stored for 2 weeks at 4°C in tions of tissue were cut and transferred to siliconized decicated slide boxes. After development and fixation, the sections were stained with hematoxylin and eosin. slides. interactions are essential, and the patterns of expression of these mRNAs are temporally and spatially regulated. Identification and study of receptors for BMPs will help us to elucidate the functions of BMPs. Analysis of gene-targetted mice and studies of organotypic cultures should also help to identify the precise functions of these molecules during vertebrate organogenesis. Preparation of Probes, Northern Blotting and In Situ Hybridization Poly(A)+RNAwas isolated from whole bodies of rats on E l 5 with a Fast Track mRNA isolation kit (InVitrogen, San Diego, CAI. Single-stranded cDNA was synthesized with a cDNA Synthesis Kit plus (Amersham, U.K.). To generate the cDNA sequences for rat BMP-3 and BMP-7, polymerase chain reactions (PCRs) were performed with the single-stranded cDNA from E l 5 rats as the template. The degenerate oligonucleotide pairs that were used to amplify the sequences corresponded to nucleotides 1,136-1,530 of the human gene for BMP-3 (Wozney et al., 1988) and 924-1,349 of the human gene for BMP-7 (Celeste et al., 19901, respectively. Products of PCR were ligated into the pGEM-T vector (Promega, Madison, WI). They were sequenced and identified as fragments of cDNAs for rat BMP-3 or BMP-7. (GenBank accession numbers: D29768, D29769.1 Data base analyses showed that the fragment of rat BMP-3 had 87% identity with human BMP-3 and only 10-15% homology with other members of BMPs in nucleotide level. The fragment of rat BMP-7 had 91% identity with human BMP-7 and 63% identity with mouse BMP-6 (which had the highest homology among BMPs except for human BMP-7). Thus, each BMP cDNA in the plasmid vectors is suitable as the probe for Northern blot analysis and in situ hybridization. Northern blot filters containing 1pg of poly(A)+RNA from whole bodies of rats on E l 5 were prepared. DNA probes for BMP-3 and BMP-7 were isolating from the plasmid vectors described above and were radiolabelled with 32P-dCTP by random-prime labelling kit (Boehringer-Mannheim, Germany). The Northern blot filters were hybridized with each probe at 65°C for 15 hours in hybridization buffers and washed a t room temperature in l x SSC/O.l% SDS for 40 minutes and 65°C in 0.1 x SSC/O.l% SDS for 40 minutes. The filters were ACKNOWLEDGMENTS The authors thank Ms. Eiko Moriizumi for excellent technical assistance, Dr. Shintaro Nomura for expert advice, and Professor Shusaku Yoshiki and Dr. Akira Yamaguchi for support during this study. The authors also thank Dr. Chikako Nor0 and Dr. Nobuhiro Nor0 for helpful comments on the manuscript. REFERENCES Celeste, A.J., Iannazzi, J.A., Taylor, R.C., Hewick, R.M., Rosen, V., Wang, E.A., and Wozney, J.M. (1990) Identification of transforming growth factor p family members present in bone-inductive protein purified from bovine bone. Proc. Natl. Acad. Sci., USA 87:98439847. du Cros, D.L., Tsaacs, K., and Moore, G.P.M. (1993) Distribution of acidic and basic fibroblast growth factors in ovine skin during follicle morphogenesis. J. Cell. Sci. 105:667-674. Hardy, M.H. (1992) The Bemet life of the hair follicle. Trend Genet. 855-61. Ikeda, T., Nomura, S., Yamaguchi, A., Suda, T., and Yoshiki, S. (1992) In situ hybridization of bone matrix proteins in undecalcified adult rat bone sections. J. Histochem. Cytochem. 40:1079-1088. Jones, C.M., Lyons, K.M., and Hogan, B.L.M. (1991) Involvement of Bone morphogenetic protein-4 (BMP-4) and Vgr-1 in morphogenesis and neurogenesis in the mouse. Development 111:531-542. Jones, C.M., Lyons, K.M., Lapan, P.M.,Wright, C.V.E., and Hogan, B.L.M. (1992) DVR-4 (bone morphogenetic protein-4) as a posteriorventralizing factor in Xenopus mesoderm induction. Development 115639-647. Kingsley, D.M. (1994) The TGF-p superfamily: New members, new receptors, and new genetic tests of function in different organisms. Genes Dev. 8133-146. Kingsley, D.M., Bland, A.E., Grubber, J.M., Marker, P.C., Ruseell, L.B., Copeland, N.G., and Jenkins, N.A. (1992) The mouse short ear skeletal morphogenesis locus is associated with defecte in a bone morphogenetic member of the TGFp superfamily. Cell 71:399-410. Lyons, K., Graycar, J.L., Lee, A., Hashmi, S., Lindquist, P.B., Chen, E.Y., Hogan, B.L.M., and Derynck, R. (1989a) Vgr-1, a mammalian gene related to Xenopus Vg-1, is a member of the transforming growth factor p gene family. Proc. Natl. Acad. Sci. USA 8645544568. Lyons, K.M., Pelton, R.W., and Hogan, B.L.M. (1989b) Patterns of expression of murine Vgr-1 and BMF'-2a RNA suggest that trans- BMP-3 AND BMP-7 IN RAT EMBRYOS forming growth fador-&like genes coordinately regulate aspects of embryonic development. Genes Dev. 3:1657-1668. Lyons, K.M., Pelton, R.W., and Hogan, B.L.M. (1990) Organogenesis and pattern formation in the mouse: RNA distribution patterns suggest a role for bone morphogenetic protein-ZA (BMP-ZA). Development 109:833-844. Lyons, K.M., Jones, C.M., and Hogan, B.L.M. (1991) The DVR gene family in embryonic development. Trend Genet. 7:408-412. Massague, J. (1990)The transforming growth factor-p family. Annu. Rev. Cell Biol. 6:597-641. Niswander, L., and Martin, G.R. (1993) FGF-4 and BMP-2 have opposite effects on limb growth. Nature 361:68-71. Nomura, S., Willis, A.J., Edwards, D.R., Heath, J.K., and Hogan, B.L.M. (1988) Developmental expression of Zar (osteopontin) and SPARK (osteonectin) RNA as revealed by in situ hybridization. J. Cell Biol. 106441-450. Ozkaynak, E., Schnegelsberg, P.N., and Oppermann, H. (1991) Murine ostwgenic protein (OP-1):High levels of mRNA in kidney Biochem. Biophys. Res. Commun. 179:116-123. Pelton, R.W., Dickinson, M.E., Moses, H.L., and Hogan, B.L.M. (1990) In situ hybridization analysis of TGFP3 RNA expression during mouse development: Comparative studies with TGFpl and pZ. Development 110:609-620. Perides, G., Safran, R.M., Rueger, D.C., and Charness, M.E.(1992) Induction of the neural cell adhesion molecule and neuronal aggregation by osteogenic protein 1. Proc. Natl. Acad. Sci., USA 89: 10326-10330. Perides, G., Hu, G., Rueger, D.C., and Charness, M.E. (1993) Osteo- 449 genic protein-1 regulates L1 and neural cell adhesion molecule gene expression in neural cells. J . Biol. Chem. 26825197-25205, Perides, G., Safran, R.M., Downing, L.A., and Charness, M.E. (1994) Regulation of neural cell adhesion molecule and L1 by the transforming growth factor-p superfamily. J. Biol. Chem. 269:765-770. Simon-Assmann, P., Kedinger, M., and Haffen, K. (1986) Immunocytochemical localization of extracellular-matrix proteins in relation to rat intestinal morphogenesis. Differentiation 32:59-66. Sive, H.L. (1993) The frog prince-ss: A molecular formula for dorsoventral patterning in Xenopus. Genes Dev. 7:l-12. Vainio, S.,Karavanova, I., Jowett, A., and Thesleff, I. (1993) Identification of BMP-4 as a signal mediating secondary induction between epithelial and mesenchymal tissues during early tooth development. Cell 75:45-58. Vukicevic, S., Latin, V., Chen, P., Batorsky, R., Reddi, A.H., and Sampath, T.K. (1994)Localization of osteogenicprotein-1 (bone morphogenetic protein-7) during human embryonic development: High affinity binding to basement membranes. Biochem. Biophys. Res. Commun. 198:693-700. Wall, N.A., Blessing, M., Wright, C.V.E., and Hogan, B.L.M. (1993) Biosynthesis and in vivo localization of the Decapentaplegic-Vgrelated protein, DVR-6 (Bone Morphogenetic Protein-6). J. Cell Biol. 120:493-502. Wozney, J.M., Rozen, V., Celeste, A.J., Mitsock, L.M., Whitters, M.J., Kriz, R. W., Hewick, R.M., and Wang, E.A. (1988) Novel regulators of bone formation: Molecular clones and activities. Science 242: 1528-1534.