Essential role of a carboxyl-terminal α-helix motif in the secretion of coagulation factor XI

2.1 Patient and DNA sample

A 75-year-old Japanese male patient was diagnosed with FXI deficiency (FXI: C < 3 IU/dl). Results of the coagulation laboratory test are shown in Supplementary Table S1. Neither the patient nor his family members reported a tendency for excessive bleeding. Blood samples were collected after the patient provided written informed consent. Genomic DNA (gDNA) was extracted from peripheral blood leukocytes, as previously described.²⁰ This study was approved by the Ethics Committee of the Nagoya University Graduate School of Medicine (identification no. 2015-0391).

2.2 DNA sequencing and genetic copy number analysis

All 15 exons, including exon–intron boundaries of F11, were amplified by polymerase chain reaction (PCR) using KOD FX DNA polymerase (Toyobo, Osaka, Japan) and gene-specific primers (Supplementary Table S2). PCR products were separated via electrophoresis on an ethidium bromide stained 1.5% agarose gel. The identified mutation was described according to the nomenclature of the Human Genome Variation Society (http://www.hgvs.org/mutnomen/).

For copy number analysis, quantitative PCR (qPCR) was performed using SYBR Premix Ex Taq II in Thermal Cycler Dice Real-time System II (Takara Bio Inc., Shiga, Japan). The PCR conditions used were chosen according to the manufacturer's instructions. We designed a quantitative primer set for a neighboring region of the mutation in F11 exon 15. ZBTB37, located on chromosome 1, was used as the genomic reference gene in this qPCR. Each chosen quantitative primer set is listed in Supplementary Table S2.

2.3 Mismatch PCR-restriction fragment length polymorphism (PCR-RFLP) analysis

Mutant-specific mismatch PCR-restriction fragment length polymorphism analysis of F11 exon 15 was performed using a forward primer, 5′-AAGGAGATTGACTGGATGAACG-3′, and a mismatched reverse primer, 5′-CTCCCTTTGAGCACAGCCCGCG-3′ (mismatch bases are underlined), to introduce a KspI site in the amplicon derived from the mutant allele (Supplementary Figure S1A). The amplicons were digested using KspI, and the digested fragments were analyzed by electrophoresis on 3% agarose gel.

2.4 Thrombin generation assay

Thrombin generation assays were performed in accordance with our previous report.²¹ In brief, we monitored the reactions for 2 h using Fluoroskan Ascent FL (Thermo Labsystems, Philadelphia, PA, USA) calibrated automated thrombography (Thrombinoscope BV, Maastricht, the Netherlands) and Thrombinoscope software program (Thrombinoscope BV). The excitation and emission wavelengths used were 390 nm and 460 nm, respectively. Serial dilutions of normal pooled plasma in FXI-depleted plasma (Sysmex, Kobe, Japan) were used as references.

2.5 Construction of recombinant FXI expression vector

A complementary DNA (cDNA) fragment of human FXI (wild-type [WT] FXI) was obtained from a human liver cDNA library (Takara Bio). The FXI coding sequence was inserted into a pcDNA3.1 (+) vector (Invitrogen, Carlsbad, CA, USA). Each mutant FXI expression vector was generated using site-directed mutagenesis with specific primer sets (Supplementary Table S3). The sequences of all FXI expression vectors were confirmed by direct sequencing.

2.6 Transient expression of recombinant FXI

HEK293 cells were grown at 37°C under 5% CO₂ in Dulbecco's modified Eagle's medium (DMEM) (Wako, Osaka, Japan) supplemented with 10% fetal bovine serum (Sigma-Aldrich, St. Louis, MO, USA) and penicillin/streptomycin/amphotericin B (Wako). A WT or individual mutant FXI expression vectors was transiently transfected into the cells using Lipofectamine 3000 reagent (Invitrogen) according to the manufacturer's instructions. Five hours after transfection, the medium was changed to DMEM containing 10% fetal bovine serum, and the cells were incubated for additional 19 h. Subsequently, the culture medium was changed to serum-free DMEM, and the cells were further incubated for 24 h.

2.7 Western blotting analysis

Twenty-four hours after the change to serum-free medium as described earlier, the transfected cells and culture media were harvested. Then, cells were lysed in sodium dodecyl sulfate (SDS) sample buffer (50 mM Tris-HCl, 2% SDS, and 10% glycerol; pH 6.8) to prepare the cell lysate. The culture media were centrifuged at 3000g for 10 min to remove debris, and the protein was concentrated by acetone precipitation.²² Thereafter, the precipitated pellet was dissolved in SDS sample buffer. The total protein concentration in each sample was measured using DC Protein Assay Kit (Bio-Rad Laboratories, Hercules, CA, USA) according to the manufacturer's instructions. Equal amounts of total protein in each sample were boiled for 5 min with or without 2-mercaptoethanol, then subjected to SDS–polyacrylamide gel electrophoresis (PAGE) (6% or 7.5%), and transferred onto Immobilon-P Transfer Membrane (Merck, Darmstadt, Germany). After blocking with 2.5% skim milk for 2 h at room temperature, the membranes were incubated overnight with monoclonal mouse anti-human FXI antibody (1:1,000 dilution, clone: G-2; Santa Cruz Biotechnology, Dallas, TX, USA) or monoclonal mouse anti-β-actin antibody (1:2,500 dilution, clone: BA3R; BioVision Inc., Milpitas, CA, USA) at room temperature. The membranes were subsequently incubated with horseradish peroxidase–conjugated goat anti-mouse immunoglobulin G (IgG) (1:5,000 dilution; Cell Signaling Technology, Danvers, MA, USA) for 90 min at room temperature. Next, the membranes were incubated with ECL Select Western Blotting Detection Reagent (GE Healthcare UK Ltd., Little Chalfont, UK) and the signals on them were detected using the Light Capture II system (Atto Corporation, Tokyo, Japan).

2.8 Measurement of FXI protein level

FXI protein levels in the plasma and culture media were measured by enzyme-linked immunosorbent assay (ELISA) using Human Factor XI ELISA Kit (product no. ab168547; Abcam, Cambridge, UK) according to the manufacturer's instructions. The antibodies used for capture and detection in this ELISA kit were anti-FXI polyclonal antibodies. The standard curve was created with a plasma-derived FXI, which was included in the ELISA kit (Abcam).

2.9 Coagulant activity of recombinant FXI

The coagulant activity of recombinant FXI (rFXI) mutants was measured by an activated partial thromboplastin time (APTT)-based one-stage clotting assay using C.K. Prest 2 (FUJIREBIO Inc., Tokyo, Japan). FXI-depleted plasma (George King Bio-Medical, Inc., Overland Park, KS, USA) was supplemented with rFXI-containing culture medium and subjected to the APTT-based assay. Coagulation times from serial dilutions of rFXI-WT were used to create a standard curve. The specific coagulant activity was calculated as the ratio between the coagulant activity (% of rFXI-WT) and the secreted protein level (% of rFXI-WT).

2.10 Immunofluorescence microscopy

COS-7 cells were grown on a cover glass measuring 15 mm in diameter in 12-well plates and transiently transfected with WT or individual mutant FXI expression vectors. Twenty-four hours after transfection, the cells were washed twice with phosphate buffered saline (PBS), fixed with 4% paraformaldehyde in PBS for 15 min, and permeabilized with 0.5% Triton X-100 in PBS for 10 min at room temperature. After washing again with PBS, the cells were incubated with an endoplasmic reticulum (ER) marker, Alexa Fluor 594-conjugated concanavalin A (25 µg/ml; Invitrogen), or a Golgi apparatus marker, Alexa Fluor 488-conjugated lectin Helix pomatia agglutinin (10 µg/ml; Invitrogen) for 30 min at room temperature. Subsequently, the cells were blocked with PBS containing 3% bovine serum albumin and 2% goat serum for 1 h at room temperature, and then incubated overnight with a monoclonal mouse anti-human FXI antibody (1:200 dilution, clone: G-2; Santa Cruz Biotechnology) at 4°C. The secondary antibody, Alexa Fluor 488-conjugated goat anti-mouse IgG or Alexa Fluor 555-conjugated goat anti-mouse IgG (Invitrogen), was 1:1000 diluted. Finally, the cover glass was mounted on a glass slide with VECTASHIELD Antifade Mounting Medium containing 4′, 6-diamidino-2-phenylindole (Vector Laboratories, Burlingame, CA, USA) and observed using an inverted confocal laser microscope (A1Rsi; Nikon). The fluorescence intensity was analyzed using the ImageJ plot profile function (National Institutes of Health). FXI colocalizations with the ER or the Golgi apparatus were statistically evaluated using Pearson's correlation coefficient (PCC), relying on Fiji's plugin, Coloc2 (https://imagej.net/Coloc_2), for the ImageJ software program (National Institutes of Health). For quantitative analysis of FXI colocalization with subcellular organelles, we referred the method of Pignani et al.²³ and Cohen et al.²⁴

2.11 Statistics

All quantitative data are presented as mean ± standard deviation (SD) or standard error of the mean values. Comparisons involving multiple groups were performed using Tukey's one-way analysis of variance. Statistical analyses were carried out using GraphPad Prism 5 (GraphPad Software, San Diego, CA, USA).

3.1 F11 c.1788dupC is a pathogenic mutation representing the FXI null phenotype

We herein analyzed F11 in a patient with FXI deficiency and identified a novel duplication mutation at exon 15, c.1788dupC (Figure 1A). Gene sequencing revealed a single peak of c.1788dupC, indicating a homo- or hemizygous mutation. Moreover, copy number analysis using qPCR showed that the gDNA of the patient retained a gene dosage of F11 exon 15 equal to those of gDNA from pooled normal individuals (Figure 1B). This finding suggests that the patient carried a homozygous F11 c.1788dupC. Next, to investigate whether F11 c.1788dupC was a polymorphism in the Japanese population, we designed a mismatch PCR-RFLP analysis (Supplementary Figure S1) and analyzed 104 alleles in normal Japanese individuals (n = 52 subjects); however, no mutant alleles were detected.

F11 c.1788dupC causes a unique FXI frameshift, p.E597Rfs*65, which is an extra–C-terminal extension (Figure 1C). In the mutant FXI, the glutamic acid residue located at position 597 is replaced by arginine and carries an abnormal 64-amino-acid alignment. Thrombin generation assays showed that the peak score (199.24) of the patient plasma was comparable with those of FXI 1.6% to 0% plasma, reconstituted by serial dilutions of normal pooled plasma with FXI-deficient plasma (Supplementary Figure S2). This result suggests that the F11 c.1788dupC (p.E597Rfs*65) represents the FXI null mutation. An assessment of plasma FXI levels using commercial ELISA kit revealed that the FXI level in the patient was undetectable (Figure 1D). Thus, we characterized F11 c.1788dupC as a pathogenic mutation causing the FXI-null phenotype.

3.2 Abnormal C-terminal structure causes a defect in FXI extracellular secretion

To discern the possible molecular pathogenesis caused by FXI E597Rfs*65, we constructed the following rFXI expression vectors: rFXI-Mut (E597Rfs*65), WT+35, and WTΔ29 (Figure 2A). rFXI-WT+35 is a C-terminal–extended rFXI mutant in which 35 amino acids of E597Rfs*65 are added to the C-terminus of WT FXI. rFXI-WTΔ29 is a C-terminal–deleted rFXI mutant in which 29 amino acids are deleted from the C-terminus of WT FXI. We transiently expressed WT or each of the rFXI mutants in HEK293 cells and analyzed the resultant rFXI production and extracellular secretion. Western blotting (WB) analysis of cell lysates showed no apparent quantitative differences among intracellular rFXIs (Figure 2B, left). In contrast, WB of culture media revealed a remarkable decrease in the secreted levels of rFXI-Mut, WT+35, and WTΔ29 (Figure 2B, right). In particular, the secreted levels of rFXI-Mut and rFXI-WTΔ29 were undetectable. To confirm this observation, we quantitatively assessed the secreted rFXI levels by ELISA (Figure 2C). The levels of secreted rFXI-WT and WT+35 were 697.3 ± 82.0 ng/ml and 77.2 ± 4.9 ng/ml (11.1 ± 0.7% of rFXI-WT), respectively. Consistent with the findings of the aforementioned WB analysis, the secreted levels of rFXI-Mut and rFXI-WTΔ29 were undetectable by ELISA (Figure 2C). In addition, to assess the coagulation function of the C-terminal–extended rFXI mutant, we measured the coagulant activity of rFXI-WT+35 by APTT-based one-stage assay. The coagulant activity and specific activity of rFXI-WT+35 were 2.5 ± 0.3% of rFXI-WT and 0.22 ± 0.02, respectively (Figure 2D,E). These results suggested that the addition of amino acids to the C-terminus severely affected the secretory behavior and functional activity of FXI. Furthermore, because FXI secretion was dependent on its homodimer structure with a disulfide cross-link,^3-5 we analyzed whether these rFXI mutants intracellularly composed dimers using nonreduced WB (Figure 2F). rFXI-Mut, WT+35, and WTΔ29 were detected as dimers in cell lysates, suggesting that the C-terminal abnormality had no effect on FXI dimer formation.

3.3 Threonine at position 622 is essential for FXI secretion

Given the observation that the C-terminus of FXI was associated with its extracellular secretion, we next investigated the amino acid(s) responsible for its secretion. To screen for functional alignment within the C-terminus, we designed the following series of C-terminal–truncated FXI mutants: rFXI-WTΔ19, WTΔ9, WTΔ5, WTΔ4, WTΔ3, WTΔ2, and WTΔ1 (Figure 3A). WB analysis detected all of these rFXI mutants in the cell lysate (Figure 3B, left). On the other hand, rFXI-WTΔ3, WTΔ2, and WTΔ1 were detected in the culture media, whereas further truncated mutants, specifically rFXI-WTΔ19, WTΔ9, WTΔ5, and WTΔ4, were undetectable (Figure 3B, right). ELISA of the culture media showed that the secreted levels of rFXI-WTΔ3, WTΔ2, and WTΔ1 were 16.5 ± 1.3%, 67.4 ± 8.7%, and 94.5 ± 12.7% of rFXI-WT, respectively (Figure 3C). The secreted level of rFXI-WTΔ4 was undetectable by ELISA, which was consistent with the WB data. To assess the coagulation function of C-terminal–truncated rFXI mutants, we measured the coagulant activity of the secreted rFXI deletion mutants, rFXI-WTΔ3, WTΔ2, and WTΔ1 (Figure 3D,E). The coagulant activity values of rFXI-WTΔ3, WTΔ2, and WTΔ1 were 20.4 ± 1.6%, 79.9 ± 2.2%, and 76.9 ± 7.1% of rFXI-WT respectively (Figure 3D). Moreover, the specific activity values of these rFXI mutants were 1.24 ± 0.07 (WTΔ3), 1.20 ± 0.16 (WTΔ2), and 0.86 ± 0.05 (WTΔ1) (Figure 3E). Although the specific activity of rFXI-WTΔ1 decreased with a level of statistical significance relative to rFXI-WTΔ2 or WTΔ3, we considered that the coagulant activities of rFXI-WTΔ3, WTΔ2, and WTΔ1 were all retained. The sequential C-terminal deletion of FXI, especially the second or third residue from the C-terminus, causes a decrease in their extracellular secretion level while maintaining relevant activity, which indicates a major contribution of these residues on FXI secretion. Meanwhile, the deletion of the fourth residue from the FXI C-terminus, threonine at position 622 (Thr622), causes a severe FXI secretion defect, which indicates an essential role of Thr622 in FXI secretion.

3.4 FXI C-terminal α-helix is a crucial motif for extracellular secretion

To investigate the characteristics of Thr622 in the context of FXI secretion, we conducted an in silico structure simulation. In silico analysis suggested that Thr622 was a component residue of the C-terminal α-helix (Figure 4A). In the α-helix motif, Thr622 is bound to leucine at position 619 (Leu619) via main-chain conjunction with a hydrogen bond, which is formed between a hydrogen atom of the amino group in Thr622 and an oxygen atom of the carboxy group in Leu619. This simulation suggested that the lack or amino acid substitutions of Thr622 led to the destruction of the FXI C-terminal α-helix motif. To verify the validity of the simulation, we generated rFXIs with alanine substitution of Thr622 (T622A) and/or Leu619 (L619A) (Figure 4B). WB analysis showed that these alanine-substituted rFXI mutants were all detected in the cell lysate and culture media (Figure 4C). ELISA of culture media showed that the secreted levels of rFXI-T622A, L619A, and T622/L619A were 35.2 ± 2.1%, 92.8 ± 4.9%, and 26.8 ± 1.3% of rFXI-WT, respectively (Figure 4D). To further assess the properties of Thr622, we measured the specific activity of rFXI-T622A, which was found to be 0.84 ± 0.18 (Figure 4E). These observations indicated that the alanine substitution at position 622 of FXI reduced its secretory capacity but retained the normal specific coagulant activity. The present results suggest that the α-helix at the C-terminus is an indispensable motif for extracellular secretion of FXI.

3.5 C-terminal disruption in FXI leads to its abnormal ER retention

To investigate the subcellular localization of C-terminal–disrupted rFXIs, we expressed rFXI-Mut (E597Rfs*65) or a C-terminal–truncated rFXI mutant (rFXI-WTΔ4 and rFXI-WTΔ3) in COS-7 cells and observed their colocalization with the ER (Figure 5) or the Golgi apparatus (Figure 6). Colocalization of rFXI with the ER or the Golgi apparatus was quantitatively examined by the calculation of PCC.

In double staining of FXI and the ER, all rFXIs were found to be highly colocalized with the ER (Figure 5A,B). Although the PCC value was >0.8 for all rFXIs, rFXI-Mut showed a statistically significant decrease in PCC when paired with the ER, indicating an intracellular distribution abnormality of rFXI-Mut. Double staining of FXI and the Golgi apparatus showed that secretory rFXIs (rFXI-WT and rFXI-WTΔ3) were colocalized with the Golgi apparatus (Figure 6A). In contrast, nonsecretory rFXIs (rFXI-Mut and rFXI-WTΔ4) were barely detectable in the Golgi apparatus. Fluorescence intensity analysis visualized the distribution difference of the secretory- and nonsecretory-rFXIs in the Golgi apparatus (Figure 6A, right graphs), and the PCC measurements further supported these findings (Figure 6B). The PCC values of the nonsecretory rFXIs (rFXI-Mut and rFXI-WTΔ4) were significantly lower than that of the secretory rFXIs (rFXI-WT and rFXI-WTΔ3). These observations indicated that rFXI mutants with disrupted C-terminus were abnormally retained in the ER and not transported into the Golgi apparatus. In fact, the lack of a C-terminal α-helix motif severely impaired ER-to-Golgi trafficking of FXI.