A long first intron of PROS1 modulates protein S expression across tissues in mice and humans

Keiko Maruyama; Sheng Ye; Yuka Eura; Akihiro Tsuji; Yoshitaka Fujihara; Koichi Kokame

doi:10.1182/bloodadvances.2026019759

. 2026 May 12;10(15):5215–5226. doi: 10.1182/bloodadvances.2026019759

A long first intron of PROS1 modulates protein S expression across tissues in mice and humans

Keiko Maruyama ¹, Sheng Ye ¹, Yuka Eura ¹, Akihiro Tsuji ², Yoshitaka Fujihara ³, Koichi Kokame ^1,^∗

PMCID: PMC13446313 PMID: 42102289

Key Points

•
Elements within PROS1 intron 1 modulate PS expression across tissues.
•
Rare intron 1 variants reduce reporter activity and may contribute to PS deficiency.

Visual Abstract

graphic file with name BLOODA_ADV-2026-019759-ga1.jpg

Abstract

Introns can modulate gene expression through enhancer- or silencer-like elements as well as intron-mediated enhancement. Protein S (PS), encoded by PROS1, is an anticoagulant cofactor for activated protein C and tissue factor pathway inhibitor. PROS1 spans ∼100 kb and harbors an unusually long first intron (∼46 kb). We asked whether intron 1 contributes to PS regulation. Reporter assays in HepG2 cells identified discrete intron 1 regions that enhanced or repressed PROS1 promoter activity. CRISPR/Cas9-mediated deletion of most of intron 1 in HepG2 cells reduced endogenous PS mRNA and secreted PS. Mice carrying large intron 1 deletions were viable but showed markedly reduced PS antigen in plasma and platelets compared with wild-type mice. Tissue profiling revealed variable decreases in PS mRNA across organs, indicating tissue-dependent regulation. In a patient with congenital PS deficiency who was initially negative by conventional coding-region sequencing, long-read sequencing uncovered rare intron 1 variants that reduced reporter activity. These data show that regulatory sequences within PROS1 intron 1 are necessary to maintain PS expression in vivo and suggest that noncoding intronic variants may contribute to congenital PS deficiency. Our findings support the incorporation of PROS1 intron 1 and other noncoding PROS1 regions into genetic testing for PS deficiency.

Introduction

Introns are widespread in eukaryotic genes. In addition to their roles in splicing,¹ some introns harbor regulatory elements that influence transcription, RNA processing, nuclear export, transcript stability, or translation.2, 3, 4, 5, 6, 7 Disease-associated variants are also enriched in noncoding regions, where they can alter regulatory function and contribute to pathogenesis.8, 9, 10, 11, 12

Protein S (PS) is a vitamin K–dependent glycoprotein encoded by the PROS1 gene that acts as a cofactor for activated protein C in the regulation of activated coagulation factors V and VIII and for tissue factor pathway inhibitor in the regulation of activated coagulation factor X.13, 14, 15 In addition, PS functions as a ligand for Tyro3/Axl/Mertk receptors and regulates phagocytosis of apoptotic cells, immune and inflammatory responses, tumor cell metastasis, and angiogenesis.16, 17, 18 Hepatocytes are the major sources of circulating plasma PS,¹⁹ and PS is also expressed in several other cell types (eg, megakaryocytes and vascular endothelial cells).20, 21, 22 Platelet PS limits thrombin generation in venous but not arterial thrombi.²³ PS deficiency increases the risk of venous thromboembolism (VTE), and heterozygous PROS1 pathogenic variants are associated with an increased VTE risk.²⁴^,²⁵ Homozygous or compound heterozygous PS deficiency is a rare disorder that causes neonatal purpura fulminans. In mice, complete deficiency is embryonically lethal, whereas partial deficiency predisposes to thrombosis and vascular abnormalities.26, 27, 28

Despite these well-established clinical consequences, conventional genetic testing targeting PROS1 coding exons and exon-intron boundaries fails to identify causal variants in a substantial proportion of individuals with clinically defined PS deficiency. Large cohort studies have reported that approximately half of affected families lack detectable coding-region variants with standard testing strategies.²⁹^,³⁰ This diagnostic gap suggests that pathogenic variation in noncoding regulatory regions of PROS1 may be underappreciated and motivates a systematic analysis of the noncoding architecture that controls PS expression.

Human PROS1 is located at 3q11.2, spans ∼100 kb, and contains an exceptionally long first intron (46 266 bp). Using the Human Genome Variation Society nomenclature, c. notation is referenced to NM_000313.4 (c.1 = A of the initiator ATG; intron 1 = c.76+1-c.76+46266), and genomic coordinates follow GRCh38.p14 (NC_000003.12). A highly homologous pseudogene, PROS2P, lies nearby and shares 96% sequence identity.31, 32, 33 Notably, PROS2P lacks the promoter and first exon.

The unusual length of intron 1, together with its conserved first-intron architecture across mammals despite limited primary-sequence conservation and its enrichment for regulatory annotations, suggests that this region may harbor key cis-acting elements that tune PROS1 transcription. Therefore, in this study, we focused on a systematic functional dissection of intron 1 as a tractable first step toward defining the noncoding regulatory architecture of PROS1 while recognizing that additional regulatory elements in other noncoding regions (eg, upstream enhancers or more distal introns) are also likely to contribute to PS regulation.

Here, we investigated whether PROS1 intron 1 sequences contribute to PS regulation. We mapped intron 1 regulatory segments with luciferase reporters, deleted most of intron 1 in HepG2 cells and mice, and examined a patient with PS deficiency who was negative by conventional coding-region sequencing. These approaches indicate that intron 1 modulates PS expression in vitro and in vivo and that rare intronic variation can alter PS expression.

Methods

Additional experimental details are provided in supplemental Methods.

Cloning of PROS1 promoter and intron 1 reporter constructs

The PROS1 promoter fragment (c.−2000_−2, relative to the translation start site) was cloned into pGL3-Basic to generate a promoter-only control vector.³⁴ Intron 1 inserts were designed in 2 stages: an initial set of ∼2-kb tiles with ∼500-bp overlaps spanning the full intron (c.76+1_76+46266), followed by finer subtiles (∼500 bp) and shorter segments within regions of interest. Inserts were cloned downstream of the firefly luciferase open reading frame (ORF). Transversion mutations were introduced by site-directed mutagenesis, and all constructs were sequence verified. Oligonucleotides for plasmid construction are listed in supplemental Table 1.

Luciferase reporter assay in HepG2 cells

Human hepatoma HepG2 cells were transiently transfected with PROS1 firefly luciferase reporter plasmids together with a Renilla control plasmid. Luciferase activity was measured 48 hours later with a dual-luciferase assay, and firefly luminescence was normalized to Renilla. Relative luciferase activity was expressed relative to the promoter-only control (set to 1).

CRISPR/Cas9-mediated deletion of regulatory regions within PROS1 intron 1 in HepG2 cells

Guide RNAs flanking the target regions in PROS1 intron 1 were designed and cloned into a Cas9 expression vector. For engineered deletions, HepG2 cells were cotransfected with guide RNA/Cas9 plasmids and the corresponding homology-directed repair donor constructs, selected with puromycin, and screened by polymerase chain reaction (PCR). The target region of established cells was verified by Sanger sequencing. Three independent clones for each genotype were isolated, expanded, and analyzed. Oligonucleotides for plasmid construction are listed in supplemental Table 2.

Generation of PROS1 intron 1–deleted mice

All animal experiments were approved by the animal care and use committees of the National Cerebral and Cardiovascular Center (approval number 25001). Pros1 intron 1–deleted mice were generated by CRISPR/Cas9-mediated genome editing of fertilized BDF1 oocytes as described previously.³⁵ Founder and offspring genotypes were confirmed by PCR.

Preparation of mouse plasma and platelets

Whole blood was collected from anesthetized mice into 3.2% citrate. Platelet-rich plasma was prepared by low-speed centrifugation, platelets were pelleted and washed, and platelet-free plasma was obtained by additional centrifugation. Platelets were lysed in sodium dodecyl sulfate–polyacrylamide gel electrophoresis sample buffer for immunoblot analysis.

Isolation of mouse hepatocytes, cardiac fibroblasts, and lung vascular endothelial cells

Hepatocytes, lung vascular endothelial cells, and cardiac fibroblasts were isolated from mice by enzymatic digestion and enrichment procedures as described previously.36, 37, 38 Briefly, hepatocytes were obtained by liver perfusion and Percoll enrichment, lung endothelial cells by collagenase digestion followed by CD31 bead enrichment and CD102-based sorting, and cardiac fibroblasts by enzymatic dissociation and adherent culture.

Digital PCR for quantification of PS mRNA

Total RNA was extracted from mouse tissues or isolated cells, reverse transcribed to complementary DNA, and analyzed by digital PCR using TaqMan assays for PS and TATA-box binding protein. Relative PS mRNA was calculated as PS/TATA-box binding protein mRNA and reflects total PS transcripts detected by the assay, without distinguishing splice isoforms.

Quantification of secreted PS in HepG2 cell culture supernatants

HepG2 cells were cultured in serum-containing medium supplemented with vitamin K1, and conditioned media were collected. Secreted human PS was quantified by enzyme-linked immunosorbent assay (ELISA) and normalized to total cellular protein from the corresponding wells.

Quantification of mouse plasma PS antigen by ELISA

Mouse plasma PS antigen was quantified by ELISA using a monoclonal capture antibody and pooled wild-type (WT) plasma as the calibration standard.

Western blot analysis

Platelet PS was assessed by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and western blotting with an anti–mouse PS antibody. Signal intensities were quantified, and equal loading was confirmed by total protein staining.

Genetic analysis of PROS1 in PS deficiency

PROS1 was analyzed by long-range PCR followed by Oxford Nanopore long-read sequencing, as described previously.³⁹ Amplicons covered the full PROS1 genomic region, including all exons and introns (supplemental Figure 1; supplemental Table 3). Reads were analyzed using software tools, including the newly developed TaKoHigh⁴⁰ and Sniffles,⁴¹ and candidate variants were validated by PCR-direct Sanger sequencing. This study was approved by the research ethics committee of the National Cerebral and Cardiovascular Center (approval number M29-108-6) and was conducted in accordance with the Declaration of Helsinki.

Statistical analyses

Statistical analyses were performed with GraphPad Prism 10. Tests are specified in the figure legends, and P < .05 was considered statistically significant.

Results

Effects of PROS1 intron 1 sequences on promoter-driven luciferase activity

To examine whether intron 1 sequences influence transcription from the PROS1 promoter, we constructed a reporter in which the PROS1 promoter (c.−2000_−2) was placed upstream of the firefly luciferase ORF in pGL3-Basic, and intron 1 tiles (∼2 kb) were inserted downstream of the ORF (Figure 1A). Intron 1 was first tiled into 28 ∼2-kb units with ∼500-bp overlaps. Relative luciferase activity, defined as firefly/Renilla normalized to the promoter-only control (pGL3-PROS1[c.−2000_−2; no intron insert]), was reduced for most inserts. In contrast, 5 tiles were not reduced relative to the promoter-only control (Figure 1B, red bars): 2 in the 5′ intron 1 interval (c.76+1_76+2000 and c.76+1501_76+3500) and 3 in a distal interval centered on the 36-kb region of intron 1 (c.76+34499_76+36500, c.76+36001_76+38000, and c.76+42001_76+44000), hereafter referred to as the distal 36-kb intron 1 interval. Therefore, we subdivided these regions into ∼500-bp subtiles and retested them. Four subtiles enhanced luciferase activity relative to the promoter-only control (supplemental Figure 2, black bars): 2 in the 5′ intron 1 interval (c.76+1_76+480 and c.76+381_76+860) and 2 in the distal 36-kb intron 1 interval (c.76+35282_76+35765 and c.76+35661_76+36163). One tile, c.76+34499_76+36500, showed greater variability in this coarse screen (Figure 1); however, we did not rely on this single construct for functional assignment but instead interpreted this region in the context of the overlapping subtiles c.76+35282_76+35765 and c.76+35661_76+36163, which reproducibly showed regulatory activity in the fine-tiling assay (supplemental Figure 2).

Figure 1. — **Reporter design and coarse tiling of *PROS1* intron 1 reveal modulation of promoter-driven luciferase activity.** (A) Schematic of the pGL3-PROS1 reporter. The *PROS1* promoter fragment (c.−2000_−2) was placed upstream of the luciferase ORF in pGL3-Basic, and intron 1 tiles (∼2 kb; ∼500 bp overlaps) were cloned downstream of the ORF at the SalI site. (B) Luciferase activity in HepG2 cells transiently transfected with the promoter-only control (pGL3-PROS1[c.−2000_−2; no intron insert]) or pGL3-PROS1 constructs carrying individual intron 1 inserts (∼2 kb; ∼500-bp overlaps) inserted downstream of the ORF. Firefly luminescence was normalized to Renilla (pRL-SV40), and relative luciferase activity was defined as firefly/Renilla normalized to the promoter-only control across clones 1 and 2 (set to 1). For each construct, 2 independent clones (1 and 2) were analyzed, each measured in triplicate (n = 3 per clone). Bars show mean ± standard deviation (SD). Schematic of *PROS1* exons 1, 2, and intron 1 showing the positions of the coarse tiling fragments (top); luciferase activities for each tile (bottom).

To refine candidate activator elements, we introduced block-transversion substitutions into the positive subtiles. The c.76+381_76+490 subtile enhanced activity (Figure 2A, red bars), and this effect was reduced by block-transversion substitutions spanning c.76+431_76+437 and c.76+438_76+450 (Figure 2A, green bars). Likewise, c.76+760_76+860 enhanced activity, and the effect was counteracted by a block transversion spanning c.76+850_76+860 (Figure 2B). The c.76+35976_76+36076 subtile enhanced activity, which was reduced by block-transversions at both c.76+35985_76+36004 and c.76+36020_76+36030 (Figure 2C). These results delineate putative enhancer-like segments within the 5′ intron 1 interval (c.76+431_76+450 and c.76+850_76+860) and the distal 36-kb intron 1 interval (c.76+35976_76+36076).

We also obtained evidence for silencer-like segments. The c.76+1_76+572 subtile did not affect activity (Figure 2D, red bars), but introducing a block transversion spanning c.76+491_76+500 increased activity (Figure 2D, green bars). Although c.76+760_76+860 enhanced activity (Figure 2E, red bars), extending this subtile to c.76+760_76+950 abolished enhancement (Figure 2E, green bars), whereas block transversions spanning c.76+861_76+905 and c.76+906_76+950 did not alter activity (supplemental Figure 3A). In addition, c.76+35875_76+36076 enhanced activity (Figure 2F, red bars), and this effect was reduced by extending to c.76+35875_76+36163 (Figure 2F, green bars); a block transversion spanning c.76+36107_76+36163 slightly enhanced activity (supplemental Figure 3B). Together, these data suggest silencer-like segments within the 5′ intron 1 interval (c.76+491_76+500 and c.76+861_76+950) and the distal 36-kb intron 1 interval (c.76+36077_76+36163). Thus, intron 1 sequences both enhance and repress transcription from the PROS1 promoter in this reporter context. In the distal 36-kb region, the reporter assay data are most consistent with a composite regulatory interval rather than a single uniformly activating or repressive element.

Taken together, we refined 6 short intron 1 segments that modulate promoter activity: 4 in the 5′ intron 1 interval (c.76+431_76+450 and c.76+850_76+860, enhancer-like; c.76+491_76+500 and c.76+861_76+950, silencer-like) and 2 in the distal 36-kb intron 1 interval (c.76+35976_76+36076, enhancer-like; c.76+36077_76+36163, silencer-like; Figure 2). The nucleotide sequences are provided in supplemental Table 4.

CRISPR/Cas9-mediated intron 1 deletions alter endogenous PS mRNA levels in HepG2 cells

To test whether the regulatory regions identified in reporter assays affect endogenous expression, we generated HepG2 clones lacking either candidate regulatory elements or most of PROS1 intron 1. Sanger sequencing of the targeted PROS1 region in all edited clones showed both WT and deleted alleles, consistent with monoallelic editing of the PROS1 locus (Figure 3A). Deletion of most of intron 1 (c.76+310_76+45878del) significantly decreased PS mRNA levels (Figure 3B). Among enhancer-element deletions, the 2 edits in the 5′ intron 1 interval (c.76+399_76+451del and c.76+678_76+857del) had different effects on PS mRNA, with only c.76+399_76+451del reducing expression. Among silencer-element deletions, 2 edits in the 5′ intron 1 interval (c.76+491_76+540del and c.76+862_76+964del) did not alter PS mRNA, whereas deletion in the distal 36-kb intron 1 interval (c.76+36096_76+36149del) increased PS mRNA. We were unable to generate clones harboring a deletion spanning c.76+35976_76+36076. Together, these results indicate that sequences within intron 1 contribute to the regulation of endogenous PS mRNA levels in HepG2 cells.

Figure 3. — **CRISPR/Cas9-mediated deletion within *PROS1* intron 1 alters endogenous PS expression in HepG2 cells.** (A) Schematic of *PROS1* intron 1 showing the positions and extents of the CRISPR/Cas9-induced deletions analyzed in HepG2 cells on the same intron 1 schematic and coordinate scale used in Figure 1B and supplemental Figure 2. (B) PS and TATA-box binding protein (TBP) mRNA levels were quantified by digital PCR in edited HepG2 clones and parental cells. Relative PS mRNA was defined as PS/TBP. (C) Secreted PS in culture supernatants from edited HepG2 clones and parental cells, measured by ELISA. PS antigen values were normalized to total cellular protein and expressed as a percentage of WT (set to 100%). Deletions tested included the large intron 1 deletion c.76+310_76+45878del and smaller edits (c.76+399_76+451del, c.76+678_76+857del, c.76+491_76+540del, c.76+862_76+964del, c.76+36096_76+36149del); genotypes are indicated below the bars. Bars show mean ± SD (n = 3). Statistics: 1-way analysis of variance vs WT. ∗P < .05; ∗∗P < .01; ∗∗∗P < .001.

We next examined whether intron 1 deletions also affected PS production at the protein level. Secreted PS was measurable in HepG2 culture supernatants under our assay conditions. Clones carrying the large intron 1 deletion (c.76+310_76+45878del) showed a marked reduction in secreted PS compared with WT clones (Figure 3C). Both enhancer-element deletions, c.76+399_76+451del and c.76+678_76+857del, also reduced PS secretion, even though only c.76+399_76+451del significantly lowered PS mRNA. Among silencer-element deletions, c.76+491_76+540del showed PS secretion comparable to WT, whereas c.76+862_76+964del and c.76+36096_76+36149del caused only minor changes. These data indicate that intron 1 sequences influence PS expression at both mRNA and protein levels in HepG2 cells.

Intron 1 deletion reduces plasma and platelet PS antigen in mice

To examine the effect of intron 1 on the regulation of PS expression in vivo, we generated intron 1–deleted mice. Intron 1 spans 46 266 bp in humans and 30 896 bp in mice. Two independent deletion strains were established (c.76+134_76+30767del and c.76+135_76+30769del; supplemental Figure 4). Homozygous intron 1–deleted mice were born at frequencies close to the Mendelian ratio and exhibited no gross anatomical abnormalities. Plasma PS antigen levels in homozygotes were significantly reduced to ∼20% to 30% of WT levels (Figure 4A). PS antigen in platelets was also reduced to ∼9% of WT levels, as assessed by quantitative analysis of western blot band intensities (Figure 4B; supplemental Figure 5). Thus, intron 1 is required to maintain PS expression in vivo. The magnitude of PS reduction is broadly consistent with levels observed in heterozygous PS deficiency in humans and Pros1-deficient mice, conditions that are associated with a marked increase in VTE risk in experimental models and in patients.24, 25, 26, 27, 28

Figure 4. — **Intron 1 deletion reduces PS antigen in mouse plasma and platelets.** (A) Plasma PS antigen in WT1 (n = 3), c.76+134_76+30767del (Δ1, n = 3), WT2 (n = 5), and c.76+135_76+30769del (Δ2, n = 4) mice; WT1 served as the matched WT control for Δ1; WT2 for Δ2. Values are expressed as a percentage of the pooled WT plasma (set to 100%). Bars show mean ± SD. Statistics: unpaired 2-tailed t tests (Δ1 vs WT1; Δ2 vs WT2). (B) PS antigen in platelet lysates assessed by western blot under reducing conditions after sodium dodecyl sulfate–polyacrylamide gel electrophoresis using a monoclonal anti–mouse PS antibody. A representative immunoblot is shown with genotypes indicated (WT1, n = 3; Δ1, n = 3; WT2, n = 5; Δ2, n = 4). ∗∗∗P < .001.

Tissue- and cell type–dependent effects of intron 1 deletion on PS mRNA in mice

To examine whether intron 1 regulates PS expression across tissues, we quantified PS mRNA in the liver, lung, heart, and kidney in mice. PS mRNA levels were reduced to ∼10% of WT levels in the liver to ∼50% of WT levels in the lung and kidney and only slightly reduced in the heart (Figure 5A), suggesting tissue-dependent effects of intron 1 deletion.

Figure 5. — **Tissue- and cell type–dependent effects of intron 1 deletion on PS mRNA levels in liver, lung, heart, and kidney in mice.** (A) Relative PS mRNA levels in liver, lung, heart, and kidney from homozygous intron 1–deleted mice (Δ1 and Δ2) vs WT (WT1, WT2), quantified by digital PCR. WT1 served as the matched WT control for Δ1; WT2 for Δ2. (B) Relative PS mRNA levels in hepatocytes, lung vascular endothelial cells, and cardiac fibroblasts from homozygous intron 1–deleted (Δ1) and WT1 mice, quantified by digital PCR. For all panels, expression was normalized to TBP and reported as PS/TBP. Bars show mean ± SD. unpaired 2-tailed t tests. ∗P < .05; ∗∗P < .01; ∗∗∗P < .001.

We next measured PS mRNA in cell types that express PS: hepatocytes from the liver, vascular endothelial cells from the lung, and cardiac fibroblasts from the heart. PS mRNA in hepatocytes was reduced to ∼10% of WT levels, whereas lung vascular endothelial cells showed a more modest decrease to ∼70% of WT levels (a ∼30% reduction). No significant reduction was observed in cardiac fibroblasts (Figure 5B). Thus, the impact of intron 1 deletion on PS mRNA differs by tissue and cell type.

Identification of PROS1 intron 1 deletions in a PS-deficient patient

The patient was a Japanese man aged 62 years who, at the age of 60 years, developed high-risk pulmonary embolism and right lower-extremity proximal deep-vein thrombosis. The acute event occurred after a right leg fracture with immobilization in a cast, representing a strong transient provoking factor. On presentation, laboratory testing revealed PS activity of 44% (normal, 65%-105%) and free and total PS antigens of 30% (normal, 60%-128%) and 59% (normal, 70%-140%), respectively, consistent with type 1 PS deficiency. He was treated with anticoagulation and has remained free of recurrent thrombosis during follow-up while receiving rivaroxaban. His medical history included atrial fibrillation treated with catheter ablation at the age of 49 years, after which anticoagulation had been discontinued.

Initial PCR-direct sequencing (Sanger method) of all PROS1 coding exons and exon-intron boundaries failed to identify pathogenic variants. Long-read sequencing uncovered 2 rare heterozygous intron 1 deletions in cis: NC_000003.12:g.[93938221_93938304del; 93938219del] NM_000313.4:c.[76+35370_76+35453del; 76+35455del] (Figure 6A-C; supplemental Figure 1C). Genome coordinates follow GRCh38.p14. For brevity, this can also be expressed as c.76+35370_76+35455delinsG. These variants are registered in Single Nucleotide Polymorphism Database (dbSNP) as rs1709219428 (c.76+35370_76+35453del) and rs1367473752 (c.76+35455del), with Genome Aggregation Database (gnomAD) global allele frequencies of 0.0000788 and 0.0000723, respectively. These variants are also registered in the Japanese genetic variation database, ToMMo 61KJPN on jMorp, with allele frequencies of 0.00927 and 0.00928, respectively. Their very similar allele frequencies in both databases, together with the long-read phasing result in our patient, suggest that these 2 deletions may commonly occur together as a linked intron 1 deletion haplotype, although this was not directly assessed at the population level in this study. This deletion haplotype lies within the functionally active distal 36-kb intron 1 interval identified by the overlapping tiles (Figure 1B; c.76+34499_76+36500) and subtiles (supplemental Figure 2; c.76+35282_76+35765).

Figure 6. — **Patient-identified *PROS1* intron 1 rare deletions and their functional impact in luciferase reporter assays.** (A) Long-range PCR followed by long-read sequencing across *PROS1* identified 2 heterozygous intron 1 deletions in cis: NC_000003.12:g.[93938221_93938304del; 93938219del] NM_000313.4:c.[76+35370_76+35453del; 76+35455del]. A representative Nanopore view is shown in genomic orientation on chr. 3 (GRCh38.p14; NC_000003.12), where *PROS1* resides on the minus strand; genomic coordinates increase left to right, whereas transcription runs right to left. (B) Schematic in transcript orientation, with exon 1 at the left. *PROS1* locus and deleted nucleotides (magenta and cyan) identified in the patient with PS deficiency are shown. Coordinates are shown relative to c.76 (the last coding base of exon 1); c.1 denotes the A of the initiating ATG in NM_000313.4. (C) Representative Sanger chromatograms confirming heterozygous c.76+35370_76+35453del and c.76+35455del; arrowheads indicate break points. (D) Luciferase activity in HepG2 cells transiently transfected with pGL3-PROS1 reporters containing intron 1 segments c.76+34499_76+36500 (N1) and c.76+35282_76+35765 (N2), and their double-deletion derivatives (N1+Del, N2+Del), each harboring both c.76+35370_76+35453del and c.76+35455del. Promoter-only control (pGL3-PROS1[c.−2000_−2; no intron insert]) was included. Firefly luminescence was normalized to Renilla (pRL-SV40), and relative luciferase activity was defined as firefly/Renilla normalized to the promoter-only control (set to 1). Bars show mean ± SD (n = 3). Statistics: unpaired 2-tailed t tests. ∗P < .05.

Functional assessment of intron 1 variants by luciferase reporter assay

We measured luciferase activity in HepG2 cells transiently transfected with PROS1 promoter constructs harboring an intron 1 fragment from the distal 36-kb intron 1 interval corresponding to the region that contains the patient’s deletions (Figure 6A). For the WT construct, this fragment was cloned with the reference intron 1 sequence; for the patient construct, we introduced the 2 intron 1 deletions c.76+35370_76+35453del and c.76+35455del in cis to mimic the patient allele. In this reporter context, the patient-derived intron 1 allele significantly reduced PROS1 promoter-driven luciferase activity compared with the WT intron 1 fragment (Figure 6D). These findings indicate that intron 1 variants alter the regulatory output of this interval in the reporter context and support the possibility that they contribute to the patient’s PS deficiency.

Discussion

Figure 7 provides an overview schematic aligning the reporter-defined regulatory segments, the HepG2 and mouse deletions, and the deletions identified in the patient, thereby integrating the functional data sets across experimental systems.

PS deficiency is a risk factor for VTE, and PROS1 contains an exceptionally long intron 1 (46 266 bp). This study provides, to our knowledge, the first evidence that PROS1 intron 1 sequences regulate PS expression in vitro and in vivo. Reporter assays identified intron 1 segments with enhancer- or silencer-like activity. The analysis of HepG2 clones lacking candidate regulatory segments or most of intron 1 revealed that deletion of most of intron 1 significantly decreased PS mRNA. These observations support a model in which first-intron sequences tune PROS1 transcription through multiple cis-regulatory mechanisms, potentially including effects on transcription, chromatin context, and RNA processing.

Transcription is the first step of gene expression, followed by RNA processing to yield mature mRNA. In vertebrates, long first introns are enriched for diverse functional elements, including cytosine-phosphate-guanine (CpG) islands, chromatin marks, and transcription factor (TF)-binding sites, supporting important regulatory roles.⁴²^,⁴³ In addition, intron-intron interactions have been proposed to influence multiple aspects of gene expression and regulation.⁴⁴ In our study, most regulatory segments were initially defined by reporter activity upon transversion mutations, whereas mRNA expression was assessed in deletion clones; differences between mutation-based mapping and deletion-based perturbation may underlie the partial discrepancies observed between reporter readouts and endogenous mRNA levels. In the plasmid-based assay, short intron 1 fragments are assayed outside of their native chromatin environment and at a fixed distance from the PROS1 promoter, whereas the endogenous locus integrates multiple positive and negative elements within a 3-dimensional chromatin context. Therefore, we interpret the tiling data primarily as a high-resolution screen that nominates candidate regulatory segments, and the CRISPR deletions as integrative perturbations that reveal the net contribution of larger intron 1 intervals in situ. The observation that 1 small deletion reduced secreted PS without a statistically significant change in steady-state PS mRNA further suggests that some intron 1 perturbations may influence PROS1 output through mechanisms not fully captured by total mRNA quantification alone, including modest transcriptional effects below our detection threshold or posttranscriptional processes affecting protein production or secretion.

Our tiling reporter assays were designed to identify intron 1 segments that can modulate PROS1 promoter activity, but they do not resolve the precise mechanism by which these elements act. Intronic enhancers and silencers frequently regulate transcription by looping to the promoter in the context of chromatin, and the geometry of a circular plasmid differs from that of the endogenous PROS1 locus. Therefore, it is possible that some elements behave differently in the reporter assay than at the native chromosomal site. Nevertheless, several observations support the relevance of the mapped regions in vivo: (1) deletion of most of intron 1 in HepG2 cells and in mice reduces PS mRNA and antigen levels, (2) smaller CRISPR deletions within intron 1 in HepG2 cells reproduce the direction of effect for a subset of the elements identified in the tiling screen, and (3) patient-specific intron 1 deletions reside within the functionally active distal 36-kb intron 1 interval identified by reporter assays and alter promoter activity when modeled in the same reporter context. These convergent data suggest that tiled intron 1 fragments indeed contain cis-acting regulatory sequences, even though the balance between promoter looping, local chromatin effects, and potential coupling to splicing remains to be fully elucidated.

In mice, intron 1 deletion lowered PS antigen in plasma (to ∼20%-30% of WT levels) and platelets (to ∼9% of WT levels; Figure 4). Consistent with PS production across tissues, PS mRNA was dramatically reduced in the liver (<10% of WT levels), the major source of circulating plasma PS but showed modest reductions in the lung and kidney (∼50% of WT levels), with minimal change in the heart (Figure 5A). The magnitude of PS mRNA reduction also differed by cell type: in hepatocytes, PS mRNA levels fell to ∼10% of WT levels; in lung vascular endothelial cells, they fell to ∼70% of WT levels; in cardiac fibroblasts, no significant change was observed (Figure 5B). These findings indicate that the contribution of intron 1 to PS expression varies by tissue and cell type. Previous studies have shown that partial Pros1 deficiency predisposes mice to thrombosis and vascular abnormalities.26, 27, 28 Although the long first intron is conserved mainly at the architectural rather than the primary-sequence level, our data support the conservation of intron 1 regulatory function across species. Prior work has shown that certain introns regulate gene expression in tissue- or development-specific manners.45, 46, 47 Consistently, TF binding information within intron 1 retrieved from ChIP-Atlas (https://chip-atlas.org/)⁴⁸ suggests that TF occupancy patterns within intron 1 differ across tissues and cell types.

Finally, we identified 2 rare intron 1 deletions in a patient with congenital PS deficiency who lacked coding-region variants by conventional sequencing. These deletions cluster within a functionally active interval of intron 1 identified by our tiling reporter assays, and, when modeled in the same reporter context, the patient-derived allele significantly reduced PROS1 promoter-driven luciferase activity in HepG2 cells (Figure 6D). We attempted to generate a cell line harboring deletions of c.76+35370_76+35453 and c.76+35455 but were unsuccessful. Notably, causal variants remain undetectable by standard genetic testing in approximately half of families with PS deficiency,²⁹^,³⁰ potentially because of variants in noncoding regions and/or regulatory elements. Our findings support the concept that noncoding intronic variants can contribute to congenital PS deficiency and support broader evaluation of PROS1 intron 1 in diagnostic genetic testing. The 2 rare intron 1 deletions identified here, together with their functional effects in reporter assays, provide proof of principle that clinically relevant noncoding variation in PROS1 may contribute to PS deficiency. Data from a single individual do not establish how often congenital PS deficiency is attributable to PROS1 intron 1 variants. Therefore, systematic long-read or targeted sequencing of PROS1 intron 1 in larger PS-deficient cohorts will be essential to define the overall contribution of intronic variants to disease. Because rs1709219428 has a minor allele frequency of ∼0.9% in the Japanese ToMMo 61KJPN reference panel, future population-based studies that link PROS1 intron 1 genotypes with adjudicated VTE outcomes will be needed to determine whether this variant, or related intronic lesions, contributes measurably to thrombosis risk and diagnostic yield.

Limitations

Reporter assays tested intron 1 fragments outside their native chromatin context and at fixed distances from the promoter on a circular plasmid, which may not fully recapitulate the 3-dimensional organization and looping interactions of the endogenous PROS1 locus. Some segments that modulated promoter activity in the tiling screen had modest or no effects when deleted at the chromosomal locus, likely reflecting context dependence, redundancy, and differences in spacing or orientation of regulatory motifs. We did not define the precise mechanisms or TFs responsible for intron 1 regulation. In addition, the edited HepG2 clones carried monoallelic deletions in a hyperdiploid background and represent a single hepatocyte-derived model; therefore, validating intron 1 regulatory logic in additional hepatocyte and nonhepatocyte systems, particularly endothelial and megakaryocytic cells, will be important. We also did not model patient-specific deletions at the endogenous PROS1 locus in HepG2 cells, perform isoform-resolved RNA sequencing, or conduct chromatin-conformation assays.

Interpretation of PROS1 variation is complicated by the highly homologous pseudogene PROS2P. Our long-range PCR and long-read sequencing strategy reduces mismapping and paralogous sequence interference within intron 1, increasing confidence in variant detection and break point assignment. This technical consideration is critical when extending genetic testing to noncoding regions proximal to PROS1.

In summary, we demonstrate that PROS1 intron 1 sequences harbor transcriptional regulatory elements and that a large intron 1 plays an important role in regulating PS expression in vitro and in vivo. These results provide a framework for interpreting noncoding PROS1 variants and for exploring tissue-specific regulatory mechanisms underlying PS deficiency.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Acknowledgments

This study was supported in part by research grants from the Japan Society for the Promotion of Science (grants 23K06900 and 23K27372), the Ministry of Health, Labour and Welfare of Japan (grant JPMH23FC1022), the Takeda Science Foundation, the SENSHIN Medical Research Foundation, and the Japanese Society of Thrombosis and Hemostasis.

Authorship

Contribution: K.M. designed the research, performed the experiments, analyzed the data, and wrote the manuscript; S.Y. performed the gene analysis and wrote the manuscript; Y.E. performed the gene analysis; A.T. recruited the patient and provided clinical support; Y.F. generated the intron 1–deleted mice; K.K. designed the research, analyzed the data, and wrote the manuscript; and all authors approved the final version of the manuscript.

Footnotes

Original data are available from the corresponding author, Koichi Kokame (kame@ncvc.go.jp), on request.

The full-text version of this article contains a data supplement.

Supplementary Material

Supplemental Methods, References, Tables, and Figures

BLOODA_ADV-2026-019759-mmc1.pdf^{(4.2MB, pdf)}

References

1.Roy B, Haupt LM, Griffiths LR. Review: alternative splicing (AS) of genes as an approach for generating protein complexity. Curr Genomics. 2013;14(3):182–194. doi: 10.2174/1389202911314030004. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Maniatis T, Reed R. An extensive network of coupling among gene expression machines. Nature. 2002;416(6880):499–506. doi: 10.1038/416499a. [DOI] [PubMed] [Google Scholar]
3.Dahan O, Gingold H, Pilpel Y. Regulatory mechanisms and networks couple the different phases of gene expression. Trends Genet. 2011;27(8):316–322. doi: 10.1016/j.tig.2011.05.008. [DOI] [PubMed] [Google Scholar]
4.Shaul O. How introns enhance gene expression. Int J Biochem Cell Biol. 2017;91(Pt B):145–155. doi: 10.1016/j.biocel.2017.06.016. [DOI] [PubMed] [Google Scholar]
5.Akua T, Shaul O. The Arabidopsis thaliana MHX gene includes an intronic element that boosts translation when localized in a 5' UTR intron. J Exp Bot. 2013;64(14):4255–4270. doi: 10.1093/jxb/ert235. [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Mattick JS, Gagen MJ. The evolution of controlled multitasked gene networks: the role of introns and other noncoding RNAs in the development of complex organisms. Mol Biol Evol. 2001;18(9):1611–1630. doi: 10.1093/oxfordjournals.molbev.a003951. [DOI] [PubMed] [Google Scholar]
7.Abou Alezz M, Celli L, Belotti G, Lisa A, Bione S. GC-AG introns features in long non-coding and protein-coding genes suggest their role in gene expression regulation. Front Genet. 2020;11:488. doi: 10.3389/fgene.2020.00488. [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Wellcome Trust Case Control C Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature. 2007;447(7145):661–678. doi: 10.1038/nature05911. [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Farh KK, Marson A, Zhu J, et al. Genetic and epigenetic fine mapping of causal autoimmune disease variants. Nature. 2015;518(7539):337–343. doi: 10.1038/nature13835. [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Zhang F, Lupski JR. Non-coding genetic variants in human disease. Hum Mol Genet. 2015;24(R1):R102–R110. doi: 10.1093/hmg/ddv259. [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Pagni S, Mills JD, Frankish A, Mudge JM, Sisodiya SM. Non-coding regulatory elements: Potential roles in disease and the case of epilepsy. Neuropathol Appl Neurobiol. 2022;48(3) doi: 10.1111/nan.12775. [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Wu K, Bu F, Wu Y, et al. Exploring noncoding variants in genetic diseases: from detection to functional insights. J Genet Genomics. 2024;51(2):111–132. doi: 10.1016/j.jgg.2024.01.001. [DOI] [PubMed] [Google Scholar]
13.Di Scipio RG, Hermodson MA, Yates SG, Davie EW. A comparison of human prothrombin, factor IX (Christmas factor), factor X (Stuart factor), and protein S. Biochemistry. 1977;16(4):698–706. doi: 10.1021/bi00623a022. [DOI] [PubMed] [Google Scholar]
14.Dahlbäck B. Protein S and C4b-binding protein: components involved in the regulation of the protein C anticoagulant system. Thromb Haemost. 1991;66(1):49–61. [PubMed] [Google Scholar]
15.Hackeng TM, Seré KM, Tans G, Rosing J. Protein S stimulates inhibition of the tissue factor pathway by tissue factor pathway inhibitor. Proc Natl Acad Sci U S A. 2006;103(9):3106–3111. doi: 10.1073/pnas.0504240103. [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Anderson HA, Maylock CA, Williams JA, Paweletz CP, Shu H, Shacter E. Serum-derived protein S binds to phosphatidylserine and stimulates the phagocytosis of apoptotic cells. Nat Immunol. 2003;4(1):87–91. doi: 10.1038/ni871. [DOI] [PubMed] [Google Scholar]
17.Lemke G, Rothlin CV. Immunobiology of the TAM receptors. Nat Rev Immunol. 2008;8(5):327–336. doi: 10.1038/nri2303. [DOI] [PMC free article] [PubMed] [Google Scholar]
18.van der Meer JH, van der Poll T, van 't Veer C. TAM receptors, Gas6, and protein S: roles in inflammation and hemostasis. Blood. 2014;123(16):2460–2469. doi: 10.1182/blood-2013-09-528752. [DOI] [PubMed] [Google Scholar]
19.Fair DS, Marlar RA. Biosynthesis and secretion of factor VII, protein C, protein S, and the protein C inhibitor from a human hepatoma cell line. Blood. 1986;67(1):64–70. [PubMed] [Google Scholar]
20.Fair DS, Marlar RA, Levin EG. Human endothelial cells synthesize protein S. Blood. 1986;67(4):1168–1171. [PubMed] [Google Scholar]
21.Ogura M, Tanabe N, Nishioka J, Suzuki K, Saito H. Biosynthesis and secretion of functional protein S by a human megakaryoblastic cell line (MEG-01) Blood. 1987;70(1):301–306. [PubMed] [Google Scholar]
22.Benzakour O, Kanthou C. The anticoagulant factor, protein S, is produced by cultured human vascular smooth muscle cells and its expression is up-regulated by thrombin. Blood. 2000;95(6):2008–2014. [PubMed] [Google Scholar]
23.Calzavarini S, Prince-Eladnani R, Saller F, et al. Platelet protein S limits venous but not arterial thrombosis propensity by controlling coagulation in the thrombus. Blood. 2020;135(22):1969–1982. doi: 10.1182/blood.2019003630. [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Makris M, Leach M, Beauchamp NJ, et al. Genetic analysis, phenotypic diagnosis, and risk of venous thrombosis in families with inherited deficiencies of protein S. Blood. 2000;95(6):1935–1941. [PubMed] [Google Scholar]
25.Castoldi E, Maurissen LF, Tormene D, et al. Similar hypercoagulable state and thrombosis risk in type I and type III protein S-deficient individuals from families with mixed type I/III protein S deficiency. Haematologica. 2010;95(9):1563–1571. doi: 10.3324/haematol.2010.021923. [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Saller F, Brisset AC, Tchaikovski SN, et al. Generation and phenotypic analysis of protein S-deficient mice. Blood. 2009;114(11):2307–2314. doi: 10.1182/blood-2009-03-209031. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Burstyn-Cohen T, Heeb MJ, Lemke G. Lack of protein S in mice causes embryonic lethal coagulopathy and vascular dysgenesis. J Clin Invest. 2009;119(10):2942–2953. doi: 10.1172/JCI39325. [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Banno F, Kita T, Fernandez JA, et al. Exacerbated venous thromboembolism in mice carrying a protein S K196E mutation. Blood. 2015;126(19):2247–2253. doi: 10.1182/blood-2015-06-653162. [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Caspers M, Pavlova A, Driesen J, et al. Deficiencies of antithrombin, protein C and protein S - practical experience in genetic analysis of a large patient cohort. Thromb Haemost. 2012;108(2):247–257. doi: 10.1160/TH11-12-0875. [DOI] [PubMed] [Google Scholar]
30.Larsen OH, Kjaergaard AD, Hvas AM, Nissen PH. Genetic variants in the protein S (PROS1) gene and protein S deficiency in a Danish population. TH Open. 2021;5(4):e479–e488. doi: 10.1055/s-0041-1736636. [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Schmidel DK, Tatro AV, Phelps LG, Tomczak JA, Long GL. Organization of the human protein S genes. Biochemistry. 1990;29(34):7845–7852. doi: 10.1021/bi00486a010. [DOI] [PubMed] [Google Scholar]
32.Ploos van Amstel HK, Reitsma PH, van der Logt CP, Bertina RM. Intron-exon organization of the active human protein S gene PS alpha and its pseudogene PS beta: duplication and silencing during primate evolution. Biochemistry. 1990;29(34):7853–7861. doi: 10.1021/bi00486a011. [DOI] [PubMed] [Google Scholar]
33.Edenbrandt CM, Lundwall A, Wydro R, Stenflo J. Molecular analysis of the gene for vitamin K dependent protein S and its pseudogene. cloning and partial gene organization. Biochemistry. 1990;29(34):7861–7868. doi: 10.1021/bi00486a012. [DOI] [PubMed] [Google Scholar]
34.Xia Y, Li K, Li J, Wang T, Gu L, Xun L. T5 exonuclease-dependent assembly offers a low-cost method for efficient cloning and site-directed mutagenesis. Nucleic Acids Res. 2019;47(3) doi: 10.1093/nar/gky1169. [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Fujihara Y, Noda T, Kobayashi K, et al. Identification of multiple male reproductive tract-specific proteins that regulate sperm migration through the oviduct in mice. Proc Natl Acad Sci U S A. 2019;116(37):18498–18506. doi: 10.1073/pnas.1908736116. [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Charni-Natan M, Goldstein I. Protocol for primary mouse hepatocyte isolation. STAR Protoc. 2020;1(2) doi: 10.1016/j.xpro.2020.100086. [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Almazloum A, Khalil H. Isolation of adult mouse cardiac fibroblasts. Curr Protoc. 2023;3(7) doi: 10.1002/cpz1.840. [DOI] [PubMed] [Google Scholar]
38.Wang J, Niu N, Xu S, Jin ZG. A simple protocol for isolating mouse lung endothelial cells. Sci Rep. 2019;9(1):1458. doi: 10.1038/s41598-018-37130-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Ye S, Eura Y, Matsumoto M, Kokame K. Genetic analysis using long-read sequencing to overcome the difficulties in VWF gene. Res Pract Thromb Haemost. 2025;9(4) doi: 10.1016/j.rpth.2025.102888. [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Eura Y, Takahashi A, Ye S, Sakai K, Matsumoto M, Kokame K. TaKoHigh enables accurate variant calling and phasing of PCR-based long-read sequencing data. Jxiv. Preprint posted online 10 September 2025 doi: 10.51094/jxiv.1517. [DOI] [Google Scholar]
41.Smolka M, Paulin LF, Grochowski CM, et al. Detection of mosaic and population-level structural variants with Sniffles2. Nat Biotechnol. 2024;42(10):1571–1580. doi: 10.1038/s41587-023-02024-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Li H, Chen D, Zhang J. Analysis of intron sequence features associated with transcriptional regulation in human genes. PLoS One. 2012;7(10) doi: 10.1371/journal.pone.0046784. [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Jo SS, Choi SS. Analysis of the functional relevance of epigenetic chromatin marks in the first intron associated with specific gene expression patterns. Genome Biol Evol. 2019;11(3):786–797. doi: 10.1093/gbe/evz033. [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Li R, Song X, Gao S, Peng S. Analysis on the interactions between the first introns and other introns in mitochondrial ribosomal protein genes. Front Microbiol. 2022;13 doi: 10.3389/fmicb.2022.1091698. [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Fu H, Kim SY, Park WD. A potato Sus3 sucrose synthase gene contains a context-dependent 3' element and a leader intron with both positive and negative tissue-specific effects. Plant Cell. 1995;7(9):1395–1403. doi: 10.1105/tpc.7.9.1395. [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Uveges TE, Shan Y, Kramer BE, Wight DC, Parysek LM. Intron 1 is required for cell type-specific, but not injury-responsive, reripherin gene expression. J Neurosci. 2002;22(18):7959–7967. doi: 10.1523/JNEUROSCI.22-18-07959.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Stemmler MP, Hecht A, Kemler R. E-cadherin intron 2 contains cis-regulatory elements essential for gene expression. Development. 2005;132(5):965–976. doi: 10.1242/dev.01662. [DOI] [PubMed] [Google Scholar]
48.Zou Z, Ohta T, Oki S. ChIP-Atlas 3.0: a data-mining suite to explore chromosome architecture together with large-scale regulome data. Nucleic Acids Res. 2024;52(W1):W45–W53. doi: 10.1093/nar/gkae358. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Methods, References, Tables, and Figures

BLOODA_ADV-2026-019759-mmc1.pdf^{(4.2MB, pdf)}

[bib1] 1.Roy B, Haupt LM, Griffiths LR. Review: alternative splicing (AS) of genes as an approach for generating protein complexity. Curr Genomics. 2013;14(3):182–194. doi: 10.2174/1389202911314030004. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib2] 2.Maniatis T, Reed R. An extensive network of coupling among gene expression machines. Nature. 2002;416(6880):499–506. doi: 10.1038/416499a. [DOI] [PubMed] [Google Scholar]

[bib3] 3.Dahan O, Gingold H, Pilpel Y. Regulatory mechanisms and networks couple the different phases of gene expression. Trends Genet. 2011;27(8):316–322. doi: 10.1016/j.tig.2011.05.008. [DOI] [PubMed] [Google Scholar]

[bib4] 4.Shaul O. How introns enhance gene expression. Int J Biochem Cell Biol. 2017;91(Pt B):145–155. doi: 10.1016/j.biocel.2017.06.016. [DOI] [PubMed] [Google Scholar]

[bib5] 5.Akua T, Shaul O. The Arabidopsis thaliana MHX gene includes an intronic element that boosts translation when localized in a 5' UTR intron. J Exp Bot. 2013;64(14):4255–4270. doi: 10.1093/jxb/ert235. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib6] 6.Mattick JS, Gagen MJ. The evolution of controlled multitasked gene networks: the role of introns and other noncoding RNAs in the development of complex organisms. Mol Biol Evol. 2001;18(9):1611–1630. doi: 10.1093/oxfordjournals.molbev.a003951. [DOI] [PubMed] [Google Scholar]

[bib7] 7.Abou Alezz M, Celli L, Belotti G, Lisa A, Bione S. GC-AG introns features in long non-coding and protein-coding genes suggest their role in gene expression regulation. Front Genet. 2020;11:488. doi: 10.3389/fgene.2020.00488. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib8] 8.Wellcome Trust Case Control C Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature. 2007;447(7145):661–678. doi: 10.1038/nature05911. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib9] 9.Farh KK, Marson A, Zhu J, et al. Genetic and epigenetic fine mapping of causal autoimmune disease variants. Nature. 2015;518(7539):337–343. doi: 10.1038/nature13835. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib10] 10.Zhang F, Lupski JR. Non-coding genetic variants in human disease. Hum Mol Genet. 2015;24(R1):R102–R110. doi: 10.1093/hmg/ddv259. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib11] 11.Pagni S, Mills JD, Frankish A, Mudge JM, Sisodiya SM. Non-coding regulatory elements: Potential roles in disease and the case of epilepsy. Neuropathol Appl Neurobiol. 2022;48(3) doi: 10.1111/nan.12775. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib12] 12.Wu K, Bu F, Wu Y, et al. Exploring noncoding variants in genetic diseases: from detection to functional insights. J Genet Genomics. 2024;51(2):111–132. doi: 10.1016/j.jgg.2024.01.001. [DOI] [PubMed] [Google Scholar]

[bib13] 13.Di Scipio RG, Hermodson MA, Yates SG, Davie EW. A comparison of human prothrombin, factor IX (Christmas factor), factor X (Stuart factor), and protein S. Biochemistry. 1977;16(4):698–706. doi: 10.1021/bi00623a022. [DOI] [PubMed] [Google Scholar]

[bib14] 14.Dahlbäck B. Protein S and C4b-binding protein: components involved in the regulation of the protein C anticoagulant system. Thromb Haemost. 1991;66(1):49–61. [PubMed] [Google Scholar]

[bib15] 15.Hackeng TM, Seré KM, Tans G, Rosing J. Protein S stimulates inhibition of the tissue factor pathway by tissue factor pathway inhibitor. Proc Natl Acad Sci U S A. 2006;103(9):3106–3111. doi: 10.1073/pnas.0504240103. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib16] 16.Anderson HA, Maylock CA, Williams JA, Paweletz CP, Shu H, Shacter E. Serum-derived protein S binds to phosphatidylserine and stimulates the phagocytosis of apoptotic cells. Nat Immunol. 2003;4(1):87–91. doi: 10.1038/ni871. [DOI] [PubMed] [Google Scholar]

[bib17] 17.Lemke G, Rothlin CV. Immunobiology of the TAM receptors. Nat Rev Immunol. 2008;8(5):327–336. doi: 10.1038/nri2303. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib18] 18.van der Meer JH, van der Poll T, van 't Veer C. TAM receptors, Gas6, and protein S: roles in inflammation and hemostasis. Blood. 2014;123(16):2460–2469. doi: 10.1182/blood-2013-09-528752. [DOI] [PubMed] [Google Scholar]

[bib19] 19.Fair DS, Marlar RA. Biosynthesis and secretion of factor VII, protein C, protein S, and the protein C inhibitor from a human hepatoma cell line. Blood. 1986;67(1):64–70. [PubMed] [Google Scholar]

[bib20] 20.Fair DS, Marlar RA, Levin EG. Human endothelial cells synthesize protein S. Blood. 1986;67(4):1168–1171. [PubMed] [Google Scholar]

[bib21] 21.Ogura M, Tanabe N, Nishioka J, Suzuki K, Saito H. Biosynthesis and secretion of functional protein S by a human megakaryoblastic cell line (MEG-01) Blood. 1987;70(1):301–306. [PubMed] [Google Scholar]

[bib22] 22.Benzakour O, Kanthou C. The anticoagulant factor, protein S, is produced by cultured human vascular smooth muscle cells and its expression is up-regulated by thrombin. Blood. 2000;95(6):2008–2014. [PubMed] [Google Scholar]

[bib23] 23.Calzavarini S, Prince-Eladnani R, Saller F, et al. Platelet protein S limits venous but not arterial thrombosis propensity by controlling coagulation in the thrombus. Blood. 2020;135(22):1969–1982. doi: 10.1182/blood.2019003630. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib24] 24.Makris M, Leach M, Beauchamp NJ, et al. Genetic analysis, phenotypic diagnosis, and risk of venous thrombosis in families with inherited deficiencies of protein S. Blood. 2000;95(6):1935–1941. [PubMed] [Google Scholar]

[bib25] 25.Castoldi E, Maurissen LF, Tormene D, et al. Similar hypercoagulable state and thrombosis risk in type I and type III protein S-deficient individuals from families with mixed type I/III protein S deficiency. Haematologica. 2010;95(9):1563–1571. doi: 10.3324/haematol.2010.021923. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib26] 26.Saller F, Brisset AC, Tchaikovski SN, et al. Generation and phenotypic analysis of protein S-deficient mice. Blood. 2009;114(11):2307–2314. doi: 10.1182/blood-2009-03-209031. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib27] 27.Burstyn-Cohen T, Heeb MJ, Lemke G. Lack of protein S in mice causes embryonic lethal coagulopathy and vascular dysgenesis. J Clin Invest. 2009;119(10):2942–2953. doi: 10.1172/JCI39325. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib28] 28.Banno F, Kita T, Fernandez JA, et al. Exacerbated venous thromboembolism in mice carrying a protein S K196E mutation. Blood. 2015;126(19):2247–2253. doi: 10.1182/blood-2015-06-653162. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib29] 29.Caspers M, Pavlova A, Driesen J, et al. Deficiencies of antithrombin, protein C and protein S - practical experience in genetic analysis of a large patient cohort. Thromb Haemost. 2012;108(2):247–257. doi: 10.1160/TH11-12-0875. [DOI] [PubMed] [Google Scholar]

[bib30] 30.Larsen OH, Kjaergaard AD, Hvas AM, Nissen PH. Genetic variants in the protein S (PROS1) gene and protein S deficiency in a Danish population. TH Open. 2021;5(4):e479–e488. doi: 10.1055/s-0041-1736636. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib31] 31.Schmidel DK, Tatro AV, Phelps LG, Tomczak JA, Long GL. Organization of the human protein S genes. Biochemistry. 1990;29(34):7845–7852. doi: 10.1021/bi00486a010. [DOI] [PubMed] [Google Scholar]

[bib32] 32.Ploos van Amstel HK, Reitsma PH, van der Logt CP, Bertina RM. Intron-exon organization of the active human protein S gene PS alpha and its pseudogene PS beta: duplication and silencing during primate evolution. Biochemistry. 1990;29(34):7853–7861. doi: 10.1021/bi00486a011. [DOI] [PubMed] [Google Scholar]

[bib33] 33.Edenbrandt CM, Lundwall A, Wydro R, Stenflo J. Molecular analysis of the gene for vitamin K dependent protein S and its pseudogene. cloning and partial gene organization. Biochemistry. 1990;29(34):7861–7868. doi: 10.1021/bi00486a012. [DOI] [PubMed] [Google Scholar]

[bib34] 34.Xia Y, Li K, Li J, Wang T, Gu L, Xun L. T5 exonuclease-dependent assembly offers a low-cost method for efficient cloning and site-directed mutagenesis. Nucleic Acids Res. 2019;47(3) doi: 10.1093/nar/gky1169. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib35] 35.Fujihara Y, Noda T, Kobayashi K, et al. Identification of multiple male reproductive tract-specific proteins that regulate sperm migration through the oviduct in mice. Proc Natl Acad Sci U S A. 2019;116(37):18498–18506. doi: 10.1073/pnas.1908736116. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib36] 36.Charni-Natan M, Goldstein I. Protocol for primary mouse hepatocyte isolation. STAR Protoc. 2020;1(2) doi: 10.1016/j.xpro.2020.100086. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib37] 37.Almazloum A, Khalil H. Isolation of adult mouse cardiac fibroblasts. Curr Protoc. 2023;3(7) doi: 10.1002/cpz1.840. [DOI] [PubMed] [Google Scholar]

[bib38] 38.Wang J, Niu N, Xu S, Jin ZG. A simple protocol for isolating mouse lung endothelial cells. Sci Rep. 2019;9(1):1458. doi: 10.1038/s41598-018-37130-4. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib39] 39.Ye S, Eura Y, Matsumoto M, Kokame K. Genetic analysis using long-read sequencing to overcome the difficulties in VWF gene. Res Pract Thromb Haemost. 2025;9(4) doi: 10.1016/j.rpth.2025.102888. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib40] 40.Eura Y, Takahashi A, Ye S, Sakai K, Matsumoto M, Kokame K. TaKoHigh enables accurate variant calling and phasing of PCR-based long-read sequencing data. Jxiv. Preprint posted online 10 September 2025 doi: 10.51094/jxiv.1517. [DOI] [Google Scholar]

[bib41] 41.Smolka M, Paulin LF, Grochowski CM, et al. Detection of mosaic and population-level structural variants with Sniffles2. Nat Biotechnol. 2024;42(10):1571–1580. doi: 10.1038/s41587-023-02024-y. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib42] 42.Li H, Chen D, Zhang J. Analysis of intron sequence features associated with transcriptional regulation in human genes. PLoS One. 2012;7(10) doi: 10.1371/journal.pone.0046784. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib43] 43.Jo SS, Choi SS. Analysis of the functional relevance of epigenetic chromatin marks in the first intron associated with specific gene expression patterns. Genome Biol Evol. 2019;11(3):786–797. doi: 10.1093/gbe/evz033. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib44] 44.Li R, Song X, Gao S, Peng S. Analysis on the interactions between the first introns and other introns in mitochondrial ribosomal protein genes. Front Microbiol. 2022;13 doi: 10.3389/fmicb.2022.1091698. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib45] 45.Fu H, Kim SY, Park WD. A potato Sus3 sucrose synthase gene contains a context-dependent 3' element and a leader intron with both positive and negative tissue-specific effects. Plant Cell. 1995;7(9):1395–1403. doi: 10.1105/tpc.7.9.1395. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib46] 46.Uveges TE, Shan Y, Kramer BE, Wight DC, Parysek LM. Intron 1 is required for cell type-specific, but not injury-responsive, reripherin gene expression. J Neurosci. 2002;22(18):7959–7967. doi: 10.1523/JNEUROSCI.22-18-07959.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib47] 47.Stemmler MP, Hecht A, Kemler R. E-cadherin intron 2 contains cis-regulatory elements essential for gene expression. Development. 2005;132(5):965–976. doi: 10.1242/dev.01662. [DOI] [PubMed] [Google Scholar]

[bib48] 48.Zou Z, Ohta T, Oki S. ChIP-Atlas 3.0: a data-mining suite to explore chromosome architecture together with large-scale regulome data. Nucleic Acids Res. 2024;52(W1):W45–W53. doi: 10.1093/nar/gkae358. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

A long first intron of PROS1 modulates protein S expression across tissues in mice and humans

Keiko Maruyama

Sheng Ye

Yuka Eura

Akihiro Tsuji

Yoshitaka Fujihara

Koichi Kokame

Key Points

Visual Abstract

Abstract

Introduction

Methods

Cloning of PROS1 promoter and intron 1 reporter constructs

Luciferase reporter assay in HepG2 cells

CRISPR/Cas9-mediated deletion of regulatory regions within PROS1 intron 1 in HepG2 cells

Generation of PROS1 intron 1–deleted mice

Preparation of mouse plasma and platelets

Isolation of mouse hepatocytes, cardiac fibroblasts, and lung vascular endothelial cells

Digital PCR for quantification of PS mRNA

Quantification of secreted PS in HepG2 cell culture supernatants

Quantification of mouse plasma PS antigen by ELISA

Western blot analysis

Genetic analysis of PROS1 in PS deficiency

Statistical analyses

Results

Effects of PROS1 intron 1 sequences on promoter-driven luciferase activity

Figure 1.

Figure 2.

CRISPR/Cas9-mediated intron 1 deletions alter endogenous PS mRNA levels in HepG2 cells

Figure 3.

Intron 1 deletion reduces plasma and platelet PS antigen in mice

Figure 4.

Tissue- and cell type–dependent effects of intron 1 deletion on PS mRNA in mice

Figure 5.

Identification of PROS1 intron 1 deletions in a PS-deficient patient

Figure 6.

Functional assessment of intron 1 variants by luciferase reporter assay

Discussion

Figure 7.

Limitations

Acknowledgments

Authorship

Footnotes

Supplementary Material

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases