An Sp1 binding site mutation of the PROS1 promoter in a patient with protein S deficiency

Protein S (PS) is a vitamin K-dependent plasma glycoprotein which functions as an important natural anticoagulant, and its deficiency is known to be one of the major risk factors for thrombosis. PS serves as a cofactor in the protein C (PC) pathway for activated PC, which regulates blood coagulation by inactivating factors Va and VIIIa (Dahlbäck, 1991) and also directly inhibits factors Va (Heeb et al, 1993) and Xa (Heeb et al, 1994) independently of PC. Although many PS (alpha) gene (PROS1) abnormalities associated with PS deficiency have been identified, no mutation has been reported in the PROS1 promoter region to date. This paper describes the first report of a PROS1 promoter mutation responsible for PS deficiency. The mutation, a C-to-T substitution, was within the core binding site of transcription factor Sp1 and disrupted binding of Sp1 to the PROS1 promoter, thereby causing the reduced PROS1 expression in reporter gene assays. The patient was a 32-year-old Japanese male with portal vein thrombosis who had started on anticoagulation therapy with heparin and warfarin. PC decreases more rapidly by warfarisation than PS; however, this patients' plasma levels of free and total PS antigens were notably low (31% and 68% respectively), while that of PC antigen was normal (128%), on the seventh day of the initial low-dose warfarisation. Therefore, he was diagnosed with congenital PS deficiency. Ethical approval for the study was obtained from the Ethics Committee of the Nagoya University School of Medicine. DNA samples were obtained with informed consent in accordance with the Declaration of Helsinki. Polymerase chain reaction (PCR)-based sequence analysis of PROS1 was performed as described previously (Okada et al, 2004). Sequencing of the 5′-flanking region of the patient's PROS1 demonstrated a C-to-T substitution at position 168-bp upstream from the translational start site (-168 C→T). PmlI PCR-restriction fragment length polymorphism analysis using a partially mismatched sense primer (5′-CCCTCTCGGGCTGGGCGCTGGGAGCacG-3′) revealed that the -168 C→T was not a polymorphism (data not shown). To investigate, whether -168 C→T influenced PROS1 promoter function, we analysed the wild-type and mutant promoter activity by luciferase reporter assay in the transiently transfected HepG2 cells. The immediate 5′-flanking region of the wild-type or mutant PROS1 promoter was amplified by PCR and cloned into the pGL3-basic-derived vector (Kokame et al, 2001). The 5-μg luciferase vector and 1-μg pSV-β-galactosidase plasmid (Promega Corp, Madison, WI, USA) were cotransfected to HepG2 cells by the calcium precipitation method. Luciferase and β-galactosidase activities were measured at 24 h after glycerol shock, and the relative promoter activity was represented as the luciferase/β-galactosidase activity. The mean value of the relative luciferase activity with the mutant plasmid was 23·3 ± 12·9% (n = 9) of that with the wild-type. Thus, a fragment of the mutant promoter exhibited impaired activity compared with the corresponding wild-type promoter. The -168 C→T mutation occurs within a putative Sp1 binding site. Sp1 is a versatile protein regulating the expression of many different genes. Binding sites for Sp1 are widely distributed in TATA-less GC-rich promoters (Dynan, 1986) and have been proposed to functionally compensate for the absence of canonical TATA box (Pugh & Tjian, 1991). Therefore, we evaluated the behaviour of a 20-bp PROS1 promoter fragment (from -176 to -157) containing this mutation in interaction with nuclear extracts from HepG2 cells by electrophoretic mobility shift assay (EMSA) (Fig 1). EMSA was performed with 5-μg nuclear extract from HepG2 cells and biotin-labeled double-strand DNA probes (600 fmol) using a LightShiftTM Chemiluminescent EMSA kit (Pierce, Rockford, IL, USA). In the supershift experiment, nuclear extract was incubated with biotin-labeled probes and anti-Sp1 antibody (sc59X) (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Biotin-labeled DNA probes were transferred to Hybond-N membrane (Amersham Bioscience, Piscataway, NJ, USA) and analysed by the Light Capture gel/documentation system (Atto, Tokyo, Japan). The wild-type oligonucleotide was found to bind a protein present in HepG2 nuclear extract, but the mutant showed a decreased nuclear protein binding. In the bound HepG2 nuclear proteins, two complexes (Fig 1, arrows) were specific, as they could be competed away by incubation with increasing amounts of unlabeled wild-type oligonucleotide, but not by equivalent amounts of unlabeled mutant. Thus, the wild-type oligonucleotide served as the stronger competitor as opposed to the mutant. The presence of the ubiquitous transcription factor Sp1 in the slowest-mobility complex (Fig 1, *) was shown by the supershift experiment with anti-Sp1 antibody, indicating that the wild-type oligonucleotide of the human PROS1 promoter sequence contains an actual Sp1 binding site. Electrophoretic mobility shift assay using wild-type and mutant oligonucleotide sequences. The -168 C-to-T mutation (-168 C→T) diminished binding of Sp1 to the PROS1 promoter sequence. A biotin-labeled wild-type (WT) oligonucleotide encompassing the Sp1 binding site (from -176 to -157 prior to the translation start site) in the human PROS1 promoter region showed binding to nuclear extract proteins (arrows) from HepG2 cells (lanes 1 and 7). A biotin-labeled mutant (MT) oligonucleotide showed decreased signals of protein combinations (lane 2). Reactions were performed with incubation of unlabeled competitor oligonucleotides at 6× concentrations of WT and MT sequence (lanes 3 and 4 respectively), and 25× concentrations of WT and MT sequence (lanes 5 and 6 respectively). Sp1-specific antibody was added to the reaction, resulting in a supershift of the antibody-protein-oligonucleotide probe complex (lane 8, *). Transcription from the PROS1 promoter is directed from multiple start sites (de Wolf et al, 2005), and recently two groups have characterised essential regulatory elements in the 5′-flanking region of the PROS1 (Tatewaki et al, 2003; de Wolf et al, 2006). Various transcription factor binding sites were identified within the first 400-bp proximal to the PROS1 translational start site, including multiple binding sites for ubiquitous transcription factors Sp1 and Sp3. Further confirmation that the Sp1 binding disruption by this mutation leads to reduced promoter activity is shown in Fig 2. One-microgram of pEVR2/Sp1 vector encoding Sp1 (kindly provided by Dr Suske, Phillipps-Universität Marburg, Marburg, Germany), or pEVR2 without Sp1 cDNA as a negative control vector was used for the cotransfection experiment (16 h after glycerol shock). Cotransfection with Sp1 expression plasmid elevated luciferase expression from both wild-type and mutant reporter vectors, although to a significantly lesser extent with the mutant than with the wild-type plasmid. These data suggested that a -168 C→T mutation would reduce PROS1 mRNA transcription, because of an impairment of Sp1 binding to the PROS1 promoter sequence, leading to PS deficiency. Cotransfection of an Sp1 expression vector alters expression of reporter gene from vectors with wild-type and -168 C→T mutant PROS1 promoter sequence. wild-type (WT) or mutant (-168 C→T; MT) reporter vector was transiently transfected into HepG2 cells with or without Sp1 expression vector as described in methods. The results shown (37·0 ± 9·3% for mutant plasmid vs. 100·0 ± 11·3% for wild-type plasmid, in the absence of Sp1 expression vector; 133·4 ± 51·5% for mutant plasmid vs. 328·9 ± 81·1% for wild-type plasmid, in the presence of Sp1 expression vector) were the averages from three experiments. Each of the relative luciferase activities was normalised for transfection efficiency by cotransfection with a β-galactosidase control plasmid and calculated as a percentage of the mean value of wild-type reporter vector without Sp1 expression vector (mean ± SD %). The total number of replicates in each group is shown. In summary, this is the first report of a PROS1 promoter mutation (-168 C→T) in the Sp1 binding site of a PS deficient patient that strongly impaired its promoter function. The authors declare no competing financial interests. We gratefully acknowledge C. Wakamatsu for expert technical assistance, and members of the Kojima's and Murate's laboratories for their helpful discussions. This study was supported in part by Grants-in-Aid from the Japanese Ministry of Education, Culture, Sports, Science and Technology (17590490), and the Japanese Ministry of Health, Labour and Welfare (Research on Measures for Intractable Diseases).