Identification and detection of the periostin gene in cardiac development
Article first published online: 5 NOV 2004
Copyright © 2004 Wiley-Liss, Inc.
The Anatomical Record Part A: Discoveries in Molecular, Cellular, and Evolutionary Biology
Volume 281A, Issue 2, pages 1227–1233, December 2004
How to Cite
Norris, R. A., Kern, C. B., Wessels, A., Moralez, E. I., Markwald, R. R. and Mjaatvedt, C. H. (2004), Identification and detection of the periostin gene in cardiac development. Anat. Rec., 281A: 1227–1233. doi: 10.1002/ar.a.20135
- Issue published online: 22 NOV 2004
- Article first published online: 5 NOV 2004
- Manuscript Accepted: 6 AUG 2004
- Manuscript Received: 6 JUL 2004
- National Heart, Lung, and Blood Institute/National Institutes of Health. Grant Numbers: RO1-HL66231, RO1-HL33756, Po1-HL52813
- endocardial cushion;
- epithelial-mesenchymal transformation;
- cell adhesion
Periostin, a member of the fasciclin gene family, acts as a cell adhesion molecule through binding to cell surface integrins. Periostin expression has previously been shown to increase substantially following transforming growth factor β (TGF-β) and bone morphogenetic protein stimulation. As these molecules are indispensable for cardiac development, we sought to clone the chicken ortholog of periostin and evaluate its spatiotemporal expression pattern during heart morphogenesis. We show by Northern analysis, whole mount and section in situ hybridization experiments that periostin is predominantly expressed in the developing endothelium of the ventricular trabeculae as well as in the endothelium and mesenchyme of the outflow tract and atrioventricular endocardial cushions. Cardiac expression continues into fetal development where periostin is seen predominantly in the valve leaflets and supporting chordae tendinae. © 2004 Wiley-Liss, Inc.
Periostin is a disulfide-linked 90 kDa cell adhesion protein originally cloned from a mouse osteoblast cell line (Takeshita et al., 1993). To date, the mouse, human, zebrafish, and rat periostin orthologs have been reported (Takeshita et al., 1993; Sasaki et al., 2001; Ito et al., 2002; Kudo et al., 2004). Periostin contains four repeated domains related to those found in the ancestral insect protein, fasciclin I. Functionally, periostin interacts with integrins to support cell adhesion and spreading of chondrocytes, fibroblasts, and a number of cancer cell lines (Gillan et al., 2002; Nakamoto et al., 2002; Bao et al., 2004; Shao et al., 2004). This interaction is accomplished through highly conserved stretches in the second and fourth fasciclin-like repeats, not through typical integrin-RGD binding motifs. In turn, through interactions with heparin as well as yet unidentified extracellular matrix proteins (Sugiura et al., 1995), periostin is also able to associate with the extracellular milieu.
Members of this gene family, including periostin and βig-H3 (TGF-β-induced gene clone 3), are coexpressed in a variety of specialized connective tissues during development. Most notably, murine periostin and βig-H3 are preferentially expressed in the periodontal ligament (PDL), the periosteum, as well as the endocardial cushions and valves of the developing heart (Horiuchi et al., 1999; Kruzynska-Frejtag et al., 2001; Ohno et al., 2002a, b; Doi et al., 2003; Ferguson et al., 2003; Wilde et al., 2003). The regulation of periostin expression in these cell types is not yet well understood. However, there is increasing data demonstrating that stress responses, as well as bone morphogenetic protein (BMP) and transforming growth factor β (TGF-β), stimulate periostin transcription (Horiuchi et al., 1999; Ji et al., 2000; Stanton et al., 2000; Wang et al., 2003; Wilde et al., 2003). For example, mechanical stress forces in the PDL, periosteum, and heart stimulate periostin expression. Similarly, during bone development, periostin is upregulated in part by BMP2/4. However, during osteoblast differentiation, expression decreases, suggesting that this protein may function as an antiosteogenic molecule. In the heart, TGF-β and BMP growth factors secreted by the myocardium stimulate endocardial cushion growth and development. These endocardial cushions play an essential role in valvuloseptal morphogenesis, a developmental process indispensable to the formation of a normal four-chambered heart (Markwald et al., 1977; Eisenberg and Markwald, 1995). Therefore, to investigate the role that periostin plays in valvuloseptal development, we report the cloning of the chicken periostin cDNA and an analysis of its embryonic and fetal cardiac expression.
MATERIALS AND METHODS
Cloning of Chicken Periostin and βig-H3
Stage 25 chicken heart RNA was isolated using Rneasy Columns (Qiagen) and reverse-transcribed into cDNA (Stratagene). Sequence data from a partial cDNA clone (GB AF31262) was used to design a 5′ oligo (5′-GCGACAGAGGACTAGAGATG-3′). A full-length periostin clone was obtained by PCR using the 5′ oligo and a 3′ linker primer incorporated into the cDNAs. Sequencing of both DNA strands of periostin was performed on an ABI sequence analyzer. An EST for βig-H3 corresponding to bases 450–3000 of GB AB005553 was obtained from University of Delaware, Delaware Biotechnology Institute.
Chicken RNA was isolated from a variety of stages (according to Hamburger and Hamilton) and tissue types (Qiagen). RNA from the craniofacial region was extracted primarily from the mandibular and maxillary processes. The RNA from the heart region was extracted from the entire heart tissue, including the outflow tract. Total RNA was electrophoresed on a 2% agarose/formaldehyde/1Xmops gel. The gel was blotted onto a nitrocellulose membrane, prehybridized and hybridized with radiolabeled probes according to standard protocol (NEN Life Science Products). The periostin probe used corresponds to the first 240 bp of the cDNA isolated. A 780 bp piece of the human GAPDH probe was used for normalizations. Hybridized membranes were washed and subjected to autoradiography for 4 hr. Densitometric analysis was performed using NIH Image.
Whole Mount and Section In Situ Experiments
Whole mount in situ was performed by generating sense and antisense periostin (first 240 bases) and βig-H3 (bases 1599–2069) RNA probes (riboprobes). Based on database sequence homology searches, the sequences for periostin and βig-H3 used in the in situ experiments are not anticipated to cross-react with other family members and are highly specific for either periostin or βig-H3, respectively. Riboprobes were generated by a DIG RNA labeling (SP6/T7) kit (Roche). Riboprobes were purified (Rneasy-Qiagen) and quantified by UV spectroscopy. Whole mount in situ hybridization and detection procedures were followed as previously described with slight modifications (Irving et al., 1996). In situ experiments on tissue sections were performed according to Moorman et al. (2001) using the same riboprobes described above. In situ experiments were carried out on embryos/embryonic hearts from stage 16 to day 12. Embryos were staged according to Hamburger and Hamilton (1951). Specific signal was only observed when sections or whole hearts were hybridized with the antisense riboprobe. No signal was detected when using the control sense riboprobe.
RESULTS AND DISCUSSION
Chicken Periostin Cloning and Expression
cDNA clones for periostin were obtained using RT-PCR from a stage 25 chick heart cDNA library. The longest clone obtained was 2,655 bases in length (GB AY657005), which encodes a putative protein of 841 amino acids (Fig. 1) with a predicted molecular weight of approximately 90 kDa. This sequence contains an additional 30 amino acids in the carboxyl tail as compared to previously described rat, mouse, and human periostin proteins, suggesting a novel isoform in chickens. Overall, chick periostin shows high sequence homology at the amino acid level with other members of the periostin family (65%, 70%, and 73% identical to mouse, human, and rat, respectively). The chicken periostin protein consists of four fasciclin I-like repeats, each of which contain two highly conserved sequences (H1 and H2). Even though these sequences lack a typical RGD-integrin binding motif, they are still capable of interacting with a host of different integrins and thereby affecting cell adhesion properties (Gillan et al., 2002; Bao et al., 2004; Shao et al., 2004).
Northern analysis of periostin was performed on chicken RNA isolated from various stages beginning at stage 16 and ending at stage 29 (Fig. 2). In all samples examined, a single band representing periostin transcript(s) is evident at approximately 4.5 Kb. Initially, faint periostin expression is seen at stage 16 in whole embryo RNA preparations. Due to high levels of periostin expression reported in the murine heart and teeth (Kruzynska-Frejtag et al., 2001, 2004), we focused primarily on cardiac and craniofacial development. Commencing at stage 23, moderate levels of periostin are seen in the heart. Expression increases over time with peak levels around stage 29. A similar trend, albeit at significantly lower levels, was observed during craniofacial development. In order to elucidate a potential role for periostin during heart development, the precise spatiotemporal expression pattern was analyzed through whole mount and section in situ hybridization experiments.
Periostin Expression in Embryonic Cardiac and Extracardiac Regions
Whole mount in situ experiments at stage 16 showed no expression of periostin in the heart (Fig. 3A). Cardiac expression of periostin was first detectable at stage 17 (Fig. 3B) in the endothelium of the right ventricle. By stage 18, periostin expression becomes more uniform throughout the trabeculated endothelium of both the right and left ventricles (Fig. 3C). At stage 18, periostin mRNA was also weakly expressed in the endocardial cushions of the outflow tract (OFT) but not in the atrioventricular (AV) cushions (Fig. 3C, arrowheads). In this report, the outflow tract is used to describe the vascular conduit between the embryonic right ventricular segment and the aortic arches. By stage 21, cardiac expression of periostin intensifies in the trabecular endothelium as well as in the OFT and AV cushions (Fig. 3D and E). At stage 25, abundant levels of periostin mRNA persist in the outflow tract and AV cushions as well as in the endothelium of the ventricular trabeculae (Fig. 3F). At stage 29, periostin expression is maintained in the endothelial lining of the trabeculae and becomes intensely expressed in the truncus or distal component of the outflow tract (Fig. 3G).
Section in situ data at stage 25 showed abundant periostin mRNA in the endocardial layer of the OFT and AV cushions (Fig. 4A). In both the OFT and AV cushions, a marked gradient of expression from endothelium to myocardium was observed. A subset of mesenchymal cells immediately adjacent/underlying the endothelium of the OFT cushions was positive for periostin. Similarly, the mesenchyme underlying the endothelium of the AV cushions showed the highest levels of periostin expression (Fig. 4A). Weak or no detectable expression of periostin was seen in cells further away from the endothelium. No periostin expression was detectable in the myocardial layer underlying the mesenchymal cushion cells. This myocardium has been shown to secrete growth factors (BMP2/4 and TGF-βs) that induce an endocardial-mesenchymal transformation (Nakajima et al., 1994; Eisenberg and Markwald, 1995; Markwald et al., 1996; Ramsdell and Markwald, 1997). These growth factors have been shown to induce periostin expression in in vitro experiments (Horiuchi et al., 1999; Ji et al., 2000; Stanton et al., 2000; Wang et al., 2003). Therefore, periostin may act as a signal sensor molecule that is associated with endocardial cushion growth and development.
Section in situ data at stage 29 demonstrate that periostin expression is regionalized in the forming arterial tree, being prominent in the mesenchyme and differentiating smooth muscle cells of the aortic arches (Fig. 4B and D). The pulmonary artery, at the level of the forming pulmonary valve, shows weak expression in the endothelium and underlying condensed mesenchyme (Fig. 4B and C).
Based on the similarity in structure and proposed function of periostin and βig-H3, in situ experiments were performed for βig-H3 on stage 29 sections and compared to the data obtained for periostin (Fig. 4E and F). Unlike periostin, βig-H3 expression was not detected in the cells of the aortic arch arteries. βig-H3 mRNA expression was instead highly restricted to the mesenchymal cells of the pulmonary outlet at the level of the forming pulmonary valve, being most abundant immediately adjacent to the myocardium (Fig. 4E and F). Therefore, at stage 29 at the level of the pulmonary valve, periostin expression is restricted primarily to the apical endothelium and adjacent mesenchyme, whereas the highest levels of βig-H3 expression is restricted to the peripheral aspects of the valve, being subadjacent to the myocardium. This unique complementary and nonoverlapping expression pattern within the valves as well as within the arterial tree may help define the boundaries of the aortic and pulmonary outlets versus the more distal arterial branches.
Periostin Expression in Valvulogenesis
During embryonic development, endocardial cushions undergo an inductive process resulting in the formation of the valve leaflets and their supporting suspensory apparatus (Nakajima et al., 1997; van den Hoff et al., 1999; Schroeder et al., 2003). Increasing evidence suggests that the valvular structures are derived primarily from the endocardial cells of the developing cushions (Lincoln et al., 2004). Data presented here demonstrate that periostin is intensely expressed in the endocardium of the cushions, suggesting that expression may be maintained in the valve structures. By embryonic day 12, the valve leaflets and supporting structures have matured and been remodeled. At this time point, cardiac expression of periostin mRNA was evaluated by whole mount and section in situ hybridization experiments. Abundant periostin message was detected in the chordae tendinae of the left ventricular AV junction, the semilunar valve leaflets of the aortic and pulmonary outlet segments, and the annulus fibrosae (Fig. 5). Periostin mRNA was detected in the mitral valve but not in the tricuspid valve. Within the mitral valve, expression was further restricted to the ventricular side of the leaflets. The lack of periostin expression in the tricuspid valve is different in chick than has been observed previously in mouse. In mouse, periostin mRNA is abundant in the tricuspid and mitral valves at E17 (Kruzynska-Frejtag et al., 2001). However, the tricuspid valve in chicken development is unique to avians since it is MF20/Nkx2.5-positive and consists almost entirely of muscle (Lincoln et al., 2004).
Additionally, punctate periostin mRNA expression was detected throughout the myocardium of the atria and ventricles (Fig. 5G). The function of periostin in the myocardium is not known, but may play a structural role in stabilizing the heart. This would indeed substantiate a potential role that periostin may play in remodeling the adult ventricle following stress responses, as has been reported in cases of myocardial infarction and hypertrophy (Stanton et al., 2000; Wang et al., 2003).
In conclusion, chicken periostin is expressed in a temporospatial pattern during cardiac development. The highest levels of periostin mRNA are detected in the embryonic trabecular endothelium and endocardial cushions, as well as the valvular structures of the fetal heart. To date, the role that periostin and other fasciclin I domain containing proteins play in valvulogenesis is not known. However, βig-H3 and periostin are able to function as antiosteogenic/chondrogenic molecules during development (Ohno et al., 2002a; Saito et al., 2002). Molecular markers of chondrogenesis and tendon development (type II collagen, scleraxis, tenascin, BMP, and TBX20) are present in the mature valve leaflets and supporting structures in the chick and mouse. Therefore, in order to maintain a normal phenotype, Periostin, βigH3, and possibly other fasciclin family members may define a novel gene family essential in maintaining proper nonosteo/chondrogenic valvular structures. The identification of the cellular origins and molecular pathways involved in valvulogenesis will undoubtedly provide important information on the etiology of congenital valve malformations as well as valvular calcification disease.
Supported by National Heart, Lung, and Blood Institute/National Institutes of Health grants RO1-HL66231 (to C.H.M.) and RO1-HL33756 and Po1-HL52813 (to R.R.M. and A.W.).
- 2004. Periostin potently promotes metastatic growth of colon cancer by augmenting cell survival via the Akt/PKB pathway. Cancer Cell 5: 329–339. , , , , , , , , , .
- 2003. Mechanical stimuli enhances the expression of RGD-CAP/betaig-h3 in the periodontal ligament. Arch Oral Biol 48: 573–579. , , , , , , , , , , .
- 1995. Molecular regulation of atrioventricular valvuloseptal morphogenesis. Circ Res 77: 1–6. , .
- 2003. Developmental expression patterns of βeta-ig (betaIG-H3) and its function as a cell adhesion protein. Mech Dev 120: 851–864. , , , .
- 2002. Periostin secreted by epithelial ovarian carcinoma is a ligand for alpha(V)beta(3) and alpha(V)beta(5) integrins and promotes cell motility. Cancer Res 62: 5358–5364. , , , , , .
- 1951. A series of normal stages in the development of the chick embryo. Dev Dyn 195: 231–272. , .
- 1999. Identification and characterization of a novel protein, periostin, with restricted expression to periosteum and periodontal ligament and increased expression by transforming growth factor beta. J Bone Miner Res 14: 1239–1249. , , , , , , , , .
- 1996. Progressive spatial restriction of Sek-1 and Krox-20 gene expression during hindbrain segmentation. Dev Biol 173: 26–38. , , , , .
- 2002. Tornado extraction: a method to enrich and purify RNA from the nephrogenic zone of the neonatal rat kidney. Kidney Int 62: 763–769. , , , , , , .
- 2000. Patterns of gene expression associated with BMP-2-induced osteoblast and adipocyte differentiation of mesenchymal progenitor cell 3T3-F442A. J Bone Miner Metab 18: 132–139. , , , , , .
- 2001. Periostin (an osteoblast-specific factor) is expressed within the embryonic mouse heart during valve formation. Mech Dev 103: 183–188. , , , , .
- 2004. Periostin is expressed within the developing teeth at the sites of epithelial-mesenchymal interaction. Dev Dyn 229: 857–868. , , , , , , , .
- 2004. Zebrafish periostin is required for the adhesion of muscle fiber bundles to the myoseptum and for the differentiation of muscle fibers. Dev Biol 267: 473–487. , , , , .
- 2004. Development of heart valve leaflets and supporting apparatus in chicken and mouse embryos. Dev Dyn 230: 239–250. , , .
- 1977. Structure and development of endocardial cushions. Am J Anat 148: 85–120. , , .
- 1996. Epithelial-mesenchymal transformations in early avian heart development. Acta Anatomica 156: 173–186. , , , , .
- 2001. Sensitive nonradioactive detection of mRNA in tissue sections: novel application of the whole-mount in situ hybridization protocol. J Histochem Cytochem 49: 1–8. , , , .
- 1994. Myocardial regulation of transforming growth factor-beta expression by outflow tract endothelium in the early embryonic chick heart. Dev Biol 165: 615–626. , , .
- 1997. Expression of smooth muscle alpha-actin in mesenchymal cells during formation of avian endocardial cushion tissue: a role for transforming growth factor beta3. Dev Dyn 209: 296–309. , , , , .
- 2002. Analysis of gene expression profile in p130(Cas)-deficient fibroblasts. Biochem Biophys Res Commun 294: 635–641. , , , , , , , .
- 2002a. RGD-CAP (betaig-h3) exerts a negative regulatory function on mineralization in the human periodontal ligament. J Dent Res 81: 822–825. , , , , , , , , , .
- 2002b. RGD-CAP ((beta)ig-h3) is expressed in precartilage condensation and in prehypertrophic chondrocytes during cartilage development. Biochim Biophys Acta 1572: 114–122. , , , , , , .
- 1997. Induction of endocardial cushion tissue in the avian heart is regulated, in part, by TGFbeta-3-mediated autocrine signaling. Dev Biol (Orlando) 188: 64–74. , .
- 2002. A cell line with characteristics of the periodontal ligament fibroblasts is negatively regulated for mineralization and Runx2/Cbfa1/Osf2 activity, part of which can be overcome by bone morphogenetic protein-2. J Cell Sci 115(Pt 21): 4191–4200. , , , , , , , , , .
- 2001. Expression of Periostin, homologous with an insect cell adhesion molecule, as a prognostic marker in non-small cell lung cancers. Jpn J Cancer Res 92: 869–873. , , , , , , , , , .
- 2003. Form and function of developing heart valves: coordination by extracellular matrix and growth factor signaling. J Mol Med 81: 392–403. , , , .
- 2004. Acquired expression of periostin by human breast cancers promotes tumor angiogenesis through up-regulation of vascular endothelial growth factor receptor 2 expression. Mol Cell Biol 24: 3992–4003. , , , , , , , , .
- 2000. Altered patterns of gene expression in response to myocardial infarction. Circ Res 86: 939–945. , , , , , , , , , .
- 1995. Expression and characterization of murine osteoblast-specific factor 2 (OSF-2) in a baculovirus expression system. Protein Expr Purif 6: 305–311. , , , .
- 1993. Osteoblast-specific factor 2: cloning of a putative bone adhesion protein with homology with the insect protein fasciclin I. Biochem J 294(Pt 1): 271–278. , , , .
- 1999. Myocardialization of the cardiac outflow tract. Dev Biol 212: 477–490. , , , , , , .
- 2003. Effects of pressure overload on extracellular matrix expression in the heart of the atrial natriuretic peptide-null mouse. Hypertension 42: 88–95. , , , , , , , , .
- 2003. The divergent expression of periostin mRNA in the periodontal ligament during experimental tooth movement. Cell Tissue Res 312: 345–351. , , , , .