Aims—The human ccn1 (hccn; hcyr61) gene has been identified previously at the mRNA and protein level as a 1,25-dihydroxyvitamin D3 and growth factor regulated gene in human osteoblasts. This study aimed to analyse genomic clones containing the human ccn1 (cyr61) gene and to provide the 5′ flanking region.
Methods—Genomic clones were isolated by screening a λ library and by array filter hybridisations of a genomic library. Sequencing was performed using the dye terminator method. Promoter activity was measured after transient transfection using a β galactosidase assay. CA repeat motifs were studied by a combined PCR/fragment analysis protocol.
Results—The human 5′ flanking region of 870 nucleotides contains several stretches with high homology to the mouse promoter as well as CA repeat motifs. This first report on the human 5′ flanking sequence of the hccn1 (hcyr61) gene provides important insights into regulation pathways for the expression of this 1,25-dihydroxyvitamin D3 and growth factor responsive early gene. A genomic clone containing the hccn1 (hcyr61) gene region also yielded a CA sequence located 3′ of the ccn1 (cyr61) gene. This CA repeat and one of the CA repeat motifs in the promoter were studied in detail and found to be polymorphic.
Conclusions—The 5′ flanking sequence of the hccn1 (hcyr61) gene provides insights into the mechanisms of regulation of this immediate early gene product. The CA repeat polymorphisms within the gene region will be useful in the genetic study of disorders affecting bone metabolism.
Statistics from Altmetric.com
The human ccn1 (hccn1; hcyr61) gene is a member of an emerging immediate early gene family.1, 2 This CCN family of proteins is named according to three prototypic members (CCN2 (CTGF), CCN1 (CYR61), and CCN3 (NOV)). Additional members are CCN4 (ELM-1/WISP-1), CCN5 (COP-1/WISP-2), and CCN6 (WISP-3). The proteins contain a highly conserved cysteine rich structure, which is organised into discrete domains with partial identities to insulin-like growth factor binding protein (IGFBP), von Willebrand factor, thrombospondin, and a cysteine knot with partial identity to some growth factors.2, 3 Proteins of the CCN family play important roles in cellular processes such as growth regulation and differentiation, adhesion, migration, and angiogenesis.
The hccn1 (hcyr61) gene has been cloned from a human embryo library.4 Originally, ccn1 (cyr61) was identified as a growth factor inducible immediate early gene in mouse fibroblasts.5 The human connective tissue growth factor (CCN2; CTGF) is secreted by umbilical vein endothelial cells.6 A recently identified member (ccn6; wisp-3) is upregulated in colon cancer,7 and mutations of the ccn6 (wisp-3) gene are associated with progressive pseudorheumatoid dysplasia.8 Whereas these three members of the CCN family are believed to be growth stimulators, the other members (ccn4 (elm-1); ccn5 (cop-1)) detected in the rat and mouse are tumour suppressor genes.9, 10 Furthermore, the chicken CCN3 (NOV) protein also displays antiproliferative activity.11 The human analogue of CCN5 (COP-1) was identified in human osteoblasts and promotes cell adhesion.12
ccn1 (cyr61) is an immediate early gene responsive to fetal calf serum (FCS) and serum growth factors in mouse fibroblasts. The secreted protein is associated with the cell surface and the extracellular matrix (ECM). Mouse CCN1 (mCCN1; mCYR61) binds in vitro to integrins; it probably acts as an ECM signalling molecule and is involved in cell adhesion.2, 13 Furthermore, mCCN1 (mCYR61) induces angiogenesis in vitro.14 The expression of ccn1 (cyr61) correlates with the differentiation of mesenchymal cells into chondrocytes.15 Recently, it was demonstrated that the rat ccn1 (cyr61) gene is expressed in fracture callus in a time dependent manner.16 The hccn1 (hcyr61) gene was localised to chromosome 1p22,4, 17 a locus associated with breast cancer,18 neuroblastoma,19 pheochromocytoma,20 and primary hepatoma.21
Previously, we have shown that hccn1 (hcyr61) is an immediate early responsive gene in human osteoblasts and is highly, but transiently, upregulated at the mRNA level by 1,25-dihydroxyvitamin D3 (1,25-(OH)2D3) as well as serum and several growth factors.22 Work from Lau's group has shown that hCCN1 (hCYR61) appears to be a novel ligand for the integrin αvβ3 in endothelial cells.13 In bone, the αvβ3 integrin is expressed predominantly in cells of the osteoclast lineage, suggesting a role for this gene product as a potential 1,25-(OH)2D3 dependent coupling factor within the bone network. This interpretation is strengthened by our observation that in osteoblasts hCCN1 (hCYR61) localises to the Golgi apparatus (a process mediated by an N-terminal sequence motif) and is then secreted.23 Identification of the gene promoter is a prerequisite to the study of gene responses in vitro, which will provide a better understanding of the regulation of hccn1 (hcyr61) expression and help obtain insights into mechanisms of regulation of transcription of this immediate early response gene. Here, we report the isolation of genomic clones containing hccn1 (hcyr61) sequences as well as the 5′ flanking sequence. In addition we report on polymorphic CA repeat motifs present in the promoter and in the hccn1 (hcyr61) gene region.
Materials and methods
ISOLATION OF GENOMIC CLONES CONTAINING THE HCCN1 (HCYR61) GENE AND PROMOTER
Screening of a λ library
A genomic clone containing the hccn1 (hcyr61) gene was identified from hybridisation of a phage λ library with an hccn1 (hcyr61) cDNA probe (position 330–2025).22 Briefly, 106 λ phages were plated at 50 000 phages/agar plate on an Escherichia coli layer and incubated overnight at 37°C. Nitrocellulose filter lifts were performed in duplicate and prehybridised in 50% formamide, 5× Denhardts, 3× sodium citrate/sodium chloride (SSC), 0.1% sodium dodecyl sulphate (SDS), and 0.1 mg/ml calf thymus DNA for four hours at 42°C. After addition of the hccn1 (hcyr61) probe (labelled by random priming) at 1 × 106 counts/min/ml of fresh hybridisation solution the filters were incubated overnight at 42°C. Filters were then washed with 3× SSC at room temperature for 30 minutes, with 0.5× SSC and 0.1% SDS for 30 minutes, and finally with 0.5× SSC and 0.1% SDS at 62°C. Filters were exposed to Kodak Xomat XAR x ray films. A positive clone was identified, the λ phages from the corresponding area of the respective agar plate eluted, phages replated, and the procedure repeated. After two rounds of replating a single plaque was obtained, and λ DNA was prepared.
Screening of array filters
Hybridisation of array filters consisting of a human PAC library (RPC16 709; Resource Centre, primary database of the German Human Genome Project at the Max-Planck-Institute for Molecular Genetics, Berlin, Germany) was hybridised with the hccn1 (hcyr61) cDNA probe. Hybridisation and washing were performed according to Leenders et al.24 Filters were exposed to Kodak Xomat XAR x ray films and positive clones identified. The positive clones (LLNLP709J23219Q2 and LLNLP709B1490Q2) were grown on agar plates and the presence of hccn1 (hcyr61) DNA was verified by sequencing polymerase chain reaction (PCR) products obtained from clones on agar plates and direct sequencing after preparation of PAC DNA according to standard procedures.
PCR AND SEQUENCING ANALYSIS
PCR was performed using a PTC-200 thermal cycler (MJ Research, Oldendorf, Germany) in a volume of 30 μl containing 50 ng genomic DNA (λ DNA or PAC clones). The PCR buffer consisted of 10 mM Tris (pH 8.4), 50 mM KCl, 1.5 mM MgCl2, 0.25 mM dNTPs, 5 pmol forward and reverse primers, and 0.5 units of Taq polymerase. Cycling was for four minutes at 94°C, then 35–40 cycles of 94°C for 30 seconds, 58–64°C for one minute (depending on the primers), and 72°C for 1–3 minutes, with a final extension at 72°C for five minutes.
PCR primers for the generation of a PCR product containing three introns from λ DNA were (numbered according to our hccn1 (hcyr61) cDNA sequence entry Z98053): forward, 5′-GGACGGCTGCGGCTGCTG -3′ (358–376) and reverse, 5′-TGCCTCT CACAGACACTCAT-3′ (1836–1817). PCR primers for the generation of hccn1 (hcyr61) specific PCR products from genomic PAC clones were: forward, 5′-GGAACTGGTATC TCCACACG-3′ (947–967) and reverse, 5′-CCACGCAGGAACCGCAGTAC-3′ (1173–1153).
Sequence analysis was performed by dye terminator sequencing of λ DNA or its PCR products as well as PAC DNA and its PCR products using the ABI 310 Sequencer. Analysis of the 5′ flanking sequence for recognition motifs was done using the Matinspector program (available at www:/gsf.de/cgi-bin/matinspector.pl).25 The sequence was submitted to the EMBL data base (accession number AJ249826).
TRANSFECTION AND MEASUREMENT OF PROMOTER ACTIVITY
The 5′ flanking region of hccn1 (hcyr61) was amplified from the PAC clone using the primer 5′-GGAGAAGGCGCGGAGGG-3′ and a primer in the 5′ region of the hccn1 (hcyr61) cDNA sequence 5′-GGAGAAGGCGCGG AGGG-3′, position 47–32, according to our cDNA sequence entry Z98053. The PCR product was cloned into the pBlue-TOPO vector (InVitrogen, Groningen, the Netherlands), which allows measurement of promoter activity using the standard photometric β galactosidase assay. COS-7 cells were transiently transfected using liposome mediated gene transfer with lipofectamine. Twenty four hours after transfection, cells were lysed and cellular extracts measured for β galactosidase activity. Results were normalised to cellular protein concentrations, as measured by means of the BioRad protein assay. Untransfected cells were also measured for β galactosidase enzyme activity, which was subtracted from values obtained with the promoter construct (less than 1% of total value). Results are expressed as nmoles OPNG (o-nitrophenyl βD galactopyranoside) hydrolysed for each minute and mg of cellular protein.
SOURCE AND ISOLATION OF A CA REPEAT SEQUENCE IN THE GENE REGION
A further CA repeat sequence was identified from hybridisation of array filters (human PAC library (RPC16 709; Resource Centre, primary database of the German Human Genome Project at the Max-Planck-Institute for Molecular Genetics, Berlin, Germany) with an hccn1 (hcyr61) cDNA probe22 and two positive clones were identified. The presence of hccn1 (hcyr61) sequences and the CA repeat was verified by PCR and sequencing analysis. The CA repeat was amplified from genomic DNA isolated from the blood of healthy donors using the forward primer and a 6-FAM labelled reverse primer as described below. The CA repeat motif is localised about 4–5 kb 3′ of the hccn1 (hcyr61) gene.
ANALYSIS OF CA REPEAT MOTIFS
PCR amplification of products containing CA repeat elements
PCR was performed in a PTC 200 PCR machine (MJ Research) in a volume of 30 μl containing 50–100 ng genomic DNA. The PCR buffer consisted of 10 mM Tris (pH 8.4), 50 mM KCl, 1.5 mM MgCl2, 0.25 mM dNTPs, 5 pmol forward and 6-FAM labelled reverse primers, and 0.5 units of Taq polymerase. Cycling (for PCR product with CA repeat motif in the 3′ region of hccn1 (hcyr61)) was four minutes at 94°C, then 35 cycles of 94°C for 30 seconds, 64°C for one minute, and 72°C for one minute, with a final extension at 72°C for one hour (to obtain +A products only). The PCR product with the CA repeat motif within the promoter was amplified using a similar protocol except for a different annealing temperature of 55°C.
PCR primers (CA repeat motif 3′ of the hccn1 (hcyr61)): forward, 5′-CAGTTTTC AAACATTGGCGGCAAGC-3′ (6-FAM labelled) and reverse, 5′-CATTGTGGTGG ATGCCTGTGGTC-3′. PCR primers (CA repeat motif within the promoter): forward, 5′-CAACTACCATCACCACCATCACGC-3′ (6-FAM labelled) and reverse, 5′-CTGTC TCCCAGACCCGTCAAG-3′.
Analysis of fluorescent labelled PCR products
The CA repeat in the 3′ region of the gene and one CA repeat motif in the promoter were analysed with the use of an ABI 310 Sequencer and the Gene Scan software package. Briefly, PCR products were purified using G50 AutoSeq columns and a 0.1–1.0 μl aliquot was analysed for the size of the PCR products.
The identity of the PCR products and the determination of the number of CA repeats were verified by dye terminator sequencing analysis using the ABI 310 Sequencer.
ISOLATION OF HCCN1 (HCYR61) GENOMIC CLONES AND EXON–INTRON MAPPING
A single genomic clone was obtained from a phage λ human genomic library screened with an hccn1 (hcyr61) cDNA probe.22 Sequencing resulted in the detection of three introns (introns 1–4; table 1). For each exon–intron boundary three independently generated PCR products were sequenced on both strands. However, this genomic clone did not contain the most 5′ hccn1 (hcyr61) cDNA sequences.
Two genomic PAC clones were obtained from a human genomic library by array filter hybridisation with the hccn1 (hcyr61) cDNA probe. Direct sequencing as well as sequencing of the PCR products identified another intron (intron 1; table 1).
SEQUENCE ANALYSIS OF THE 5′ UPSTREAM REGION
Sequencing analysis of the two genomic PAC clones using primers complementary to the 5′ region of the hccn1 (hcyr61) cDNA revealed that one PAC clone (LLNLP709B1490Q2) contained all exons of hccn1 (hcyr61) but was lacking 5′ flanking sequences. The other clone (LLNLP709J23219Q2) contained sequences 5′ upstream of the first intron. From analysis of the 934 bp of hccn1 (hcyr61) 5′ flanking region (fig 1), several regions with high similarity to the mouse promoter were apparent. In addition, two CA repeat motifs were identified from direct sequencing of the PAC clones. To exclude the possibility that these motifs are accidentally present in the human genomic clone (that is, not representative of genomic DNA at this locus), the region containing these CA repeat motifs was amplified from genomic DNA isolated from the blood of two healthy human donors. Sequencing analysis of the resulting PCR products confirmed the presence of the two CA repeat motifs.
As is shown in fig 1, the 5′ flanking region of hccn1 (hcyr61) contains a TATA region; however, it lacks a CAAT box. Several motifs for transcription factors such as stimulating protein 1 (SP1), CCAAT/enhancer binding protein β (CEBPβ), cAMP responsive element binding protein (CREB), forkhead transcription factor 1 (FREAK-1), and nuclear factor κB ( NF-κB) are present.
MEASUREMENT OF PROMOTER ACTIVITY
To prove functional activity of the hccn1 (hcyr61) 5′ flanking sequences a standard photometric reporter gene assay (β galactosidase) was applied. The COS-7 cell line was used as a test system because it can easily be transfected at high efficiency using liposome mediated gene transfer. For controls, the vector only, and the promoter subcloned in the wrong orientation, were similarly transfected and cellular extracts measured for enzyme activity. The construct containing the promoter resulted in easily measurable colour development with a mean (SD) of 220 (26) nmoles OPNG hydrolysed/min/mg protein (n = 3), whereas the control constructs were negative in this assay.
ANALYSIS OF CA REPEAT POLYMORPHISMS
3′ CA repeat polymorphism and allele frequency
We identified 11 different alleles from 107 unrelated healthy donors. Figure 2 shows representative examples of individual PCR products containing the CA repeat. The typical pattern of a dinucleotide repeat sequence was obtained with groups of four peaks for each allele separated by a 2 bp size difference. The observed heterozygosity was 0.888. Table 2 shows the size and frequency of the 11 alleles and table 3 shows the size distribution of the homozygous individuals.
CA repeat polymorphisms and allele frequency of the element within the promoter
We identified five different alleles from 36 unrelated healthy donors. Again, the typical pattern of a dinucleotide repeat sequence was obtained with groups of four peaks in each allele separated by a 2 bp size difference (data not shown). The observed heterozygosity was 0.75. Table 4 shows the size and frequency of the five alleles and table 5 shows the size distribution of the homozygous individuals.
The hccn1 (hcyr61) gene belongs to an emerging family of early response genes, the CCN family.2, 3 The proteins encoded by these genes are involved in processes such as embryogenesis, differentiation, growth modulation, and angiogenesis and might also play a role in wound healing and tumour formation. hCCN1 (hCYR61) was identified by us as a novel 1,25-(OH)2D3 and growth factor target in osteoblasts,22 by means of differential display PCR analysis of a conditionally immortalised cell line (hFOB cells).26
This first report on the 5′ flanking sequence of hccn1 (hcyr61) could provide general information regarding promoter organisation in this CCN family of immediate early response genes.
We did not succeed in determining the transcriptional start site by primer extension, probably because of the high GC content of the 5′ region of the cDNA. However, from several hccn1 (hcyr61) cDNA sequence entries, as well as the high percentage of homology between hccn1 (hcyr61) cDNA and the immediate 5′ flanking region of the mouse27 and the hccn1 (hcyr61) gene presented here, the promoter region is easily identified. The 5′ flanking region subcloned in a reporter gene vector displayed promoter activity using the β galactosidase assay. Therefore, the identified 5′ flanking region in vitro behaves as a functional promoter.
Some parts of the 5′ flanking region are homologous to the mccn1 (mcyr61) promoter. A first stretch of homology, at a region 5′ of the TATA box, contains a CEBPβ motif, which is known to be a general regulator of mesenchymal cell differentiation and is important for interleukin 6 (IL-6) dependent gene regulation.28 This supports our previous finding of IL-6 dependent modulation of hccn1 (hcyr61) mRNA in osteoblasts.22 Two long CA repeat elements are present in the human 5′ flanking sequence. In contrast, the mccn1 (mcyr61) promoters contain only one element, which is at a different location (−1000).27 Sequence variations of one of the CA repeat motifs in the human DNA were analysed and revealed a polymorphic distribution within the normal human population. Potentially, different alleles of this polymorphism may be linked to different promoter activity.
Further upstream in the 5′ flanking sequence of hccn1 (hcyr61) is a CREB motif, which is important for cAMP dependent gene regulation,29 and might mediate the growth factor dependent gene regulation observed for hccn1 (hcyr61) mRNA.22 The same holds true for a NFκB/c-rel binding motif30 further downstream, because we observed tumour necrosis factor α (TNF-α) dependent upregulation of hccn1 (hcyr61) mRNA in human osteoblasts,22 and NFκB participates in TNF-α dependent signal transduction.31 A motif for FREAK-1 is also present: this transcriptional activator is expressed at epithelial–mesenchymal interfaces.32 In situ hybridisation data indicate high expression of hccn1 (hcyr61) mRNA in mesenchymal cells of the bone marrow (N Schütze et al, 2001, unpublished results). Therefore, this FREAK-1 element could participate in the regulation of hccn1 (hcyr61) mRNA expression throughout cell differentiation.
The coding region of hccn1 (hcyr61) contains four small introns. All intron–exon boundaries followed the GT … AG rule. Exon–intron localisation is homologous to the introns of the orthologue mccn1 (mcyr61) gene, which has also been described previously.33
A highly polymorphic CA repeat motif immediately 3′ of the hccn1 (hcyr61) gene was identified, which, because of its high frequency of heterozygosity, provides a useful marker for this gene region on chromosome 1p22. Because chromosomal alterations have been observed at this locus in different carcinomas, and because hCCN1 (hCYR61) belongs to a family of growth modulating factors, this polymorphism will be useful in the genetic study of disorders affecting bone metabolism and malignancies. Further studies to link the distribution of alleles with regard to metabolic bone diseases would be helpful to identify a potential disease association of hccn1 (hcyr61) alleles. From a cell biology point of view, hccn1 (hcyr61) is an attractive candidate gene because of the regulation of hccn1 (hcyr61) mRNA and protein by 1,25-dihydroxyvitamin D3 and growth factors in osteoblasts.22 Furthermore, Kireeva and colleagues13 have shown that hccn1 (hcyr61) in human endothelial cells can bind to the αvβ3 integrin (vitronectin receptor), a membrane receptor that in bone is expressed predominantly in cells of the osteoclast lineage. Thus, hccn1 (hcyr61) might act as a soluble, extracellular matrix associated factor within the bone microenvironment.
This work was supported by a grant from the Deutsche Forschungsgemeinschaft to F Jakob (Ja 504/4–1/4–2).
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.