INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

DURING THE PAST FEW YEARS, the effect of mechanical force on the regulation of cell functions has been extensively studied (8). Various stresses or strains, such as hydrostatic or hydrodynamic pressure, tensile or biaxial stretching, fluid shear stress and hypergravity, have been applied to various cells in vitro to examine their response to the mechanical stimuli. The most commonly used cell types were bone (31), cartilage (19), smooth muscle (38), endothelial cells (7), and cardiomyocytes (47) that are usually subjected to high fluid shear stresses or pressure loads. The applied forces caused a variety of physiological responses, such as increased bone resorption by osteoclasts (23), changes in matrix protein synthesis by chondrocytes and cardiac fibroblasts (19, 27), pulmonary epithelial cell differentiation (36), changes in smooth muscle contractility (38), and increased migration and tube formation by coronary endothelial cells (50). Most, if not all, of these experiments were performed in vitro and were associated with pathophysiological states such as hemodynamic overload of heart and blood vessels, osteoarthritis, osteoporosis, angiogenesis due to cancer, etc. The mechanism probably involved multiple signal transduction pathways (1).

Recently, we demonstrated that the entry of an egg into the eggshell gland (ESG) of the laying hen imposes a mechanical strain that regulates the expression of a set of genes associated with the eggshell formation (24, 25). Thus the avian ESG is a unique in vivo biological system in which the mechanical forces are coupled to a physiological readout and are imposed in a circadian fashion during the daily egg cycle. Using this system we previously demonstrated a strain-dependent upregulation of genes encoding for the ₁-subunit of the Na-K-ATPase (25) and osteopontin (OPN) (24, 33), which are involved in ion transport and calcification processes in eggshell and bone, respectively.

The eggshell is formed during the passage of the egg through the oviduct, where the various layers of the eggshell are assembled sequentially. After fertilization of the ovum, the egg spends 2-3 h in the magnum where the proteins for embryo consumption are secreted and 1-2 h in the isthmus where the two egg membranes are built. Then the egg enters the ESG, where the calcium deposition occurs, for an additional 18-20 h. During the transport of calcium for shell formation from the plasma to the lumen of the ESG, ~10% of the total body calcium is secreted within 18-29 h. This makes eggshell formation one of the most rapid biomineralization processes known (10).

In the present study we applied the RNA fingerprinting technique in an effort to identify additional genes that might be involved in shell formation and regulated by mechanical strain. The unique circadian pattern of eggshell calcification allows us to compare gene expression in two different physiological states: 1) no calcification and no mechanical strain; 2) peak of calcification and maximal mechanical strain. By this technique, we identified a 382-bp cDNA that was upregulated at the time of maximal mechanical strain and that exhibited 80% homology with the mouse heparan-sulfate proteoglycan, glypican 4 (GPC-4).

Proteoglycans are proteins containing glycosaminoglycan side chains that exist in the extracellular matrix and on the surface of many types of cell. These molecules are thought to play an important role in cell growth and morphogenesis and in cancer development (18, 35). The most abundant proteoglycans are those that bear glycosaminoglycan chains consisting of heparan sulfate. One family of the heparan-sulfate proteoglycans comprises the cell-surface glypicans and includes six members to date (14). All known members of this family have similar core protein sizes and spacing, and a glycosaminoglycans attachment consensus sequence close to the COOH terminus of the proteins; they all exhibit conservation of cystein residues, and all are linked to the cell surface by a glycosyl phosphatidyl inositol (GPI) anchor (12). Members of this family have been associated with the Simpson-Golabi-Behmel syndrome, which is characterized by pre- and postnatal overgrowth (32) and with embryonic medullary renal dysplasia (5), and they modulate the action of stimulatory and inhibitory growth factors during morphogenesis (16). All structural features of the vertebrate glypicans are also represented in the product of dally, a Drosophila melanogaster locus regulating cell division patterning in the developing central nervous system (21). Taken together, these observations implicate glypicans in control of cell division and patterning during development. The gene for the human GPC-4 has been localized to Xq26 in close proximity to GPC-3 (43), and its mRNA is ubiquitously expressed in many tissues such as bone marrow stromal cells and hematopoietic progenitor cells (37), kidney (16), and embryonic brain (45). To date, no specific function has been associated with the GPC-4 gene product beyond its ability to serve as coreceptor to the fibroblast growth factor receptors.

In the present study we evaluated the expression of the GPC-4 gene in relation to avian eggshell formation and demonstrated for the first time that GPC-4 gene expression is regulated by mechanical strain.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals and mechanical strain induction. Female Loman chickens, 4-8 mo old, were used for all experiments. Samples of magnum, isthmus, and ESG at various stages of the egg cycle and other tissues were collected. Samples were either frozen in liquid nitrogen for RNA extraction or fixed overnight at 4°C in 4% paraformaldehyde for in situ hybridization. Mechanical strain was induced by insertion into the ESG of endotracheal tubes that after inflation resembled the shape of an egg, with volumes of 60 or 20 cm³, as previously described (24, 25).

Preparation of RNA and RNA fingerprinting. Total RNA was extracted with TRI reagent (MRC) according to the company's recommended protocol. The mRNA was prepared from total RNA with the mRNA Isolation Kit (Boehringer Mannheim, Ottweiler, Germany) according to the manufacturer's instructions. Total RNA was fingerprinted with the Delta RNA fingerprinting kit (Clontech, EMC) using dATP from the Radiochemical Centre (Amersham). In brief: first-strand cDNA was synthesized by using 2 µg of total RNA as a template, oligo(dT) as a primer and Moloney murine leukemia virus reverse transcriptase (MoMLV-RT) (Life Technologies). Two dilutions of each cDNA template (corresponding to 5 and 20 ng of reverse transcribed RNA) were used for the PCR reaction. In addition to the template, each PCR reaction contained 50 µM dNTPs, 1 µM primers, 50 nM [³³P]dATP (1,000-3,000 Ci/mmol), and 1 µl Taq DNA polymerase (Perkin-Elmer). The PCR primers used were a pairwise combination of arbitrary "P" and oligo(dT) "T" primers (see Perkin-Elmer's reference manual for oligonucleotide sequences). Thermal cycling was performed according to the following program: one cycle of 94, 40, and 68°C, each for 5 min; two cycles of 94°C for 2 min, 40°C for 5 min, and 68°C for 5 min; 22 cycles of 94°C for 1 min, 60°C for 1 min, and 68°C for 2 min. PCR products were electrophoresed on an 8% acrylamide-8 M urea gel and run in 0.1 M Tris borate-2 mM EDTA (TBE) buffer (pH 8.3). The gels were dried under vacuum and exposed to X-ray films. In a typical RNA fingerprint, ~80-100 bands were evident in each amplification. Differentially expressed cDNAs were eluted from the gel, reamplified, subcloned into the pGEM-T Easy cloning vector, and sequenced. Nucleotide sequences were subjected to FASTA searches for sequence homologies.

RT-PCR and PCR cloning. RNA from ESG was reverse transcribed to single-stranded cDNAs with the aid of oligo(dT) primers and MoMLV-RT. A PCR master mix was added to yield the following concentrations for the cDNA reaction: 1 µM of specific primers, 200 µM dNTPs, 2.5 U of Taq DNA polymerase (Perkin-Elmer), and Taq buffer containing 1.5 mM MgCl₂. PCR was performed through 30 cycles (94°C for 15 s, 60°C for 30 s, and 72°C for 30 s). Amplification products arising from RT-PCR were electrophoresed on a 2% agarose gel and visualized by ethidium bromide staining. A range of forward and reverse primers based on the mouse K-glypican cDNA sequence (accession number X83577) were used for PCR cloning of the chicken GPC-4. Forward 5'-GAAAAGTTGCTCGGAAGTGC-3' and reverse 5'-AAAAGGCTTCAGCTGCTCTG-3' primers produced a new 483-bp cDNA fragment. The forward (F) primer pair based on the sequence received by PCR (5'-GGTACTACGTGGGTGGCAAT-3') and a reverse (R) primer pair based on the sequence received by RNA fingerprinting (5'-GCTGTTCAAGGACCTCTTCG-3') produced a 319-bp fragment that connected two previous cDNA fragments to one 969-bp fragment.

Rapid amplification of cDNA ends. Rapid amplification of cDNA ends (RACE), to complete the 3' region of the chicken GPC-4, was performed with the SMART RACE cDNA amplification kit (Clontech) according to the manufacturer's manual. In brief, the first-strand cDNA was synthesized from chicken ESG mRNA by means of the MoMLV-RT with the aid of SMART II oligonucleotide and 3'-RACE cDNA synthesis primers, at 42°C for 60 min. PCR was then carried out with a 5' gene-specific primer, 5'-GCTGGAGGGGCCTTTTAACATTGAGTC-3', synthesized according to the cloned 969-bp fragment and the mix of two universal adapter primers provided with the kit. Cycling conditions were five cycles of 94°C for 15 s and 72°C for 3 min; five cycles of 94°C for 15 s, 70°C for 30 s, and 72°C for 3 min; and 30 cycles of 94°C for 15 s, 68°C for 30 s, and 72°C for 3 min. A nested PCR reaction was carried out under similar conditions, with the kit primer (Nested Universal Primer) and a specific nested GPC-4 5' primer (5'-GAACAGCATGCAAGTGTCTCA-3'). Figure 1 describes the strategy in the assembly of the chicken GPC-4. The obtained PCR product was separated on 1% agarose and purified, cloned into the pGEM-T Easy vector, and sequenced from both directions with T7, SP6, and specific primers. The cDNA and deduced amino acid sequences were compared with the DNA and protein databases at the National Center for Biotechnology Information.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Assembly of the PCR products of the chicken glypican (GPC)-4.

Preparation of riboprobe and in situ hybridization. A fragment of chicken GPC-4 cDNA (646 bp) was synthesized by RT-PCR from the chicken ESG (see RT-PCR and PCR cloning). This fragment shared no homology with any other known chicken cDNA and was subcloned into the pGEM-T Easy cloning vector to produce a template for riboprobe synthesis. The sense and antisense riboprobes were synthesized by in vitro transcription with T7 and SP6 RNA polymerase, respectively, in the presence of linearized plasmid DNA and DIG RNA labeling mix (Boehringer Mannheim). Serial 5-µm sections of the ESG or magnum were hybridized with a digoxigenin-labeled chicken GPC-4 probe as described previously (25). No hybridization was observed with the sense riboprobe, which was used as a negative control. Hybridization with avian -actin probe was performed to rule out the possibility that changes in GPC-4 expression were due to variation in tissue processing and/or RNA preservation.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Identification of the chicken GPC-4 gene in chicken ESG. RNA fingerprinting was used to identify genes involved in the eggshell calcification. Differential screening was performed between cDNA derived from ESG biopsies from two different states of the daily egg cycle: 1) no egg resides in the ESG, no calcification, and no mechanical strain; 2) the egg resides in the ESG at the peaks of eggshell calcification and of mechanical strain. We isolated and sequenced 14 clones that appeared to be upregulated at the time of eggshell calcification. One of these clones, a 382-bp cDNA sequence, showed 80% similarity to the mouse K-glypican (GenBank accession number X83577). This sequence was extended to both the 3' and 5' directions, resulting in a complete 3' region of the gene and a 5' region with a missing part estimated to comprise the first 71 bp of the open reading frame, according to the mammalian homologues (Fig. 2). The amplified PCR fragments were subcloned, and 11 randomly selected subclones were analyzed by sequencing; the obtained sequence was additionally verified by RT-PCR. Analysis of the obtained sequence revealed 77 and 78% similarity to the mouse and human GPC-4 genes, respectively, with two stretches showing 92% homology. Multiple alignment of the predicted protein product of the GPC-4 gene showed 74 and 75% identity to the amino acid sequences of mouse and human GPC-4, respectively. Of most importance is the conservancy within the sequence of motifs that is characteristic of all members of the glypican family; these motifs include the glypican signature region and the GPI binding site (Fig. 2). Of the 13 cystein residues known to be conserved among all mammalian glypicans, 12 were found to be conserved also in the avian GPC-4. One cystein residue at position 224 corresponding to the mouse sequence was replaced by glycine as confirmed by RT-PCR analysis of cDNA obtained from different avian tissues.

View larger version (94K): [in this window] [in a new window]   Fig. 2.   Predicted amino acid sequence of the coding region of chicken GPC-4. Amino acids are denoted by the single letter code. Gray background represents identity in the amino acids between the chicken and the mammalian genes. , Cystein residues that are conserved in all GPC members. NH2-terminal signal and COOH-terminal hydrophobic region are boxed. Glypican signature region is represented by a solid line.

Temporal and spatial expression of the chicken GPC-4 gene during eggshell formation. Temporal and spatial localization of the avian GPC-4 in the ESG during the eggshell formation cycle was performed by in situ hybridization (Fig. 3). Very low levels of expression of the GPC-4 gene were observed before the entry of the egg into the ESG while the egg was still located in the magnum or isthmus (Fig. 3A) or 1 h after the egg entered the ESG (Fig. 3B). Two hours after the egg entered the ESG (Fig. 3C), some of the epithelium cells started to express the GPC-4 gene. Gradually, more epithelium cells expressed the GPC-4 gene, and 8 h after the entry of the egg into the ESG (Fig. 3D), at the time when the shell is being rapidly formed, most of the cells expressed the GPC-4 gene. A high level of expression was observed during all the time of massive calcification (Fig. 3E; 12 h after the entry of the egg into the ESG). One hour before oviposition (Fig. 3F), at the phase of eggshell completion, a massive reduction in the gene expression was observed.

View larger version (140K): [in this window] [in a new window]   Fig. 3.   In situ hybridization of eggshell gland (ESG) biopsies taken at various intervals after oviposition with the avian GPC-4 probe. A: egg resides in the isthmus; B: 1 h, C: 2 h, D: 8 h, E: 12 h, and F: 17 h after the entry of the egg into the ESG. Magnification ×20.

View larger version (67K): [in this window] [in a new window]   Fig. 4.   GPC-4 gene expression in the ESG. A: hematoxylin-eosin staining. B: in situ hybridization with chicken GPC-4 specific probe. PE, pseudostratified epithelium; GE, glandular epithelium.

Effect of mechanical stress on the GPC-4 gene expression. Insertion of an egg-shaped endotracheal tube into the ESG 3 h after oviposition, at a time when the GPC-4 gene is not naturally expressed (Fig. 5A), caused gene induction. The expression of the GPC-4 gene was strain dependent: insertion of a small-size tube (with a volume of 20 cm³) caused a lower induction of GPC-4 gene expression (Fig. 5B) than observed with the larger tube (60 cm³; Fig. 5C). Moreover, removing the mechanical strain by expulsion of the egg close to the time of its entry into the ESG and thus causing premature oviposition attenuated GLC-4 gene expression (Fig. 6A), compared with the level of expression in the control laying hen (Fig. 6B). No change in -actin gene expression was observed in ESG sections 3 and 12 h after oviposition or after mechanical strain application at the time of major increase in the CPC-4 expression, ruling out the possibility of variation in tissue processing and RNA preservation (Fig. 7).

View larger version (41K): [in this window] [in a new window]   Fig. 5.   Effect of mechanical strain on GPC-4 gene expression in the ESG. Mechanical strain was applied when there was no egg in the ESG and the GPC-4 gene was not expressed. A: control, no strain; B: 3 h after mechanical strain induction; inflated volume was 20 cm3, C: 3 h after mechanical strain induction; inflated volume was 60 cm3. Magnification ×20.

View larger version (30K): [in this window] [in a new window]   Fig. 6.   Effect of mechanical strain withdrawal on GPC-4 gene expression. Premature forced oviposition was performed to remove the mechanical strain imposed by the resident egg at the time of maximal mechanical strain induction. A: 2 h after expulsion of the egg from the ESG; B: control, the egg is in the ESG, 9 h after oviposition.

View larger version (139K): [in this window] [in a new window]   Fig. 7.   In situ hybridization of ESG biopsies with the avian -actin probe. A: egg resides in the isthmus; B: 3 h, C: 12 h after the entry of the egg into the ESG. D: ESG after mechanical strain induction.

View larger version (68K): [in this window] [in a new window]   Fig. 8.   Effect of mechanical strain on GPC-4 gene expression in the magnum. A: 3 h after oviposition, the egg resided in the magnum; B: 12 h after oviposition, the egg resided in the ESG. Magnification ×50.

Expression GPC-4 gene in various tissues. The expression of the GPC-4 gene in a variety of tissues was examined by RT-PCR (Fig. 9). In addition to the high levels of the GPC-4 gene that were expressed in the ESG, high levels were also observed in the liver, pancreas, and kidneys. The expression of the GPC-4 gene in these tissues was unaffected by the daily egg cycle (data not shown).

View larger version (33K): [in this window] [in a new window]   Fig. 9.   GPC-4 gene expression in chicken tissues. RNA was isolated from various tissues and GPC-4 expression was evaluated by RT-PCR.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The avian GPC-4 gene exhibits high homology with the mammalian ones and presents highly conserved sequences such as the glypican signature and the GPI-binding site in the same locations as the mammalian homologues. Moreover, the conserved locations of the cystein residues suggest similar three-dimensional structures (Fig. 2). To date, GPC-4 is the second member of the glypican family identified in the chicken [the first being GPC-1 (29)], which suggests that, similarly to mammals, avian species may contain more members of this family. In the present study we demonstrated for the first time that the avian GPC-4 gene is expressed in the ESG in a circadian fashion and is probably regulated by mechanical strain. This hypothesis is supported by the following observations: 1) the gene is expressed in the ESG only when an egg resides in the ESG and imposes a mechanical strain (Fig. 3); 2) removal of the mechanical strain caused reduction in the gene expression (Fig. 6); 3) artificial application of the mechanical strain caused induction of the GPC-4 expression that was related to the level of the strain (Fig. 5); and 4) the GPC-4 gene was induced by mechanical strain in other parts of the oviduct too (Fig. 8). In the ESG, the GPC-4 was expressed only by the GE cells and not by the cells facing the lumen (Fig. 4). Previously, we (24) demonstrated that OPN expressed by the PE cells was regulated by the mechanical strain, whereas calbindin, a calcium-binding protein expressed by the GE cells in a circadian fashion, was not. Moreover, the ₁-subunit of the Na-K-ATPase that was expressed by both cell types was regulated by mechanical strain only in the PE cells, whereas in the GE cells it was regulated by the calcium flux (25). Thus GPC-4 is the first example of a mechanical strain-dependent gene expressed by the GE cells. This suggests that the mechanical strain signal that is first sensed by the cells facing the lumen had to be transduced to the inner cell layer either directly or mediated by factor(s) from the PE. The expression of other proteoglycan genes such as versican, biglycan, perlecan, and decorin, some of which are differentially regulated in various tissues, is also known to be affected by mechanical strain (26). For example, biglycan is activated in smooth muscle cells (26) and attenuated in lung epithelium (46) in response to mechanical strain. Mechanical signals activate many different signal transduction pathways, but currently no single transcriptional regulatory element or combination of elements can account for cell-specific mechanically induced responses. It is still to be determined if the expression of the GPC-4 gene in other tissues such as liver and kidney (Fig. 9) is also regulated by mechanical stimulus, as in other parts of the oviduct (Fig. 8).

Although the precise mechanism by which mechanical strain regulates gene expression has not yet been elucidated, it is interesting to note that OPN (44, 49), the ₁-subunit of Na-K-ATPase (22, 48) and the glypican genes (3, 20), all of which were activated in the ESG by the same mechanical strain, share some potential transcription factor binding motif such as Sp1. These Sp1 binding sites have been found in shear stress response element (SSRE) (34). In our search for mechanical strain-dependent genes in the ESG, we found that Rho A, a small GTP-binding protein known to be involved in mechanical strain signal transduction, was also differentially displayed at the time of eggshell formation (data not shown). Rho A has been reported to play a role in mechanical stress-induced responses in cardiac myocytes (1) and aortic smooth muscle cells (30) and has been implicated as a downstream target of the integrin-dependent signal pathway (9, 15). Rho A has been shown to activate the expression of various transcription factors, such as MyoD (6), c-fos (42), and various other genes, via the serum response factor (4, 41), which requires Sp1 factor binding sites (39). The involvement of Rho A in the strain-dependent transcriptional activation of the GPC-4 in the ESG can only be speculated on, but the correlation between the temporal and spatial activation of these two genes supports this hypothesis.

The avian ESG is a tissue specialized in the massive calcium transport needed for eggshell formation and is the source of the organic matrix of the shell. The role of GPC-4 in the process of eggshell formation is not known. One possibility is that the function of GPC-4 in the ESG is related to its ability to serve as a coreceptor and/or a modulator of the activity of various growth factors (17). Alternatively, similar to other proteoglycans (40) and glypicans (12, 28, 45), GPC-4 may be shed from the cell surface and its role might be to become part of the assembly of proteoglycans that contribute to the biochemical properties of the mature product (13). Avian-specific glypican antibodies would enable its presence in the eggshell to be verified. It is interesting to note that calcium flux due to increase loading such as hypertension was observed in other tissues as well (11).

In a sexually immature prelaying hen, before the onset of reproduction, the GPC-4 gene in the ESG was found to be silent and could not be induced by a mechanical strain. These results suggest a priming mechanism of the gene that occurs during the transition from a nonlaying to a laying state, only after which can the GPC-4 gene be induced by the mechanical strain. Hormones such as estrogen, progesterone, etc., that are involved in maturation of the laying hen may, therefore, be involved in this process.

In summary, in this study we demonstrated that the avian homologue of the mammalian GPC-4 is expressed in the ESG in a circadian fashion and is regulated by mechanical strain.