Abstract
Purpose
The main therapeutic strategy for Hemophilia B patients involves the administration of recombinant coagulation factors IX (rFIX). Although there are various approaches to increasing the activity of rFIX, targeted protein engineering of specific residues could result in increased rFIX activity through enhanced γ-carboxylation. Specific amino acids in the propeptide sequence of vitamin K-dependent proteins are known to play a role in the interaction with the enzyme γ-carboxylase. The net hydrophobicity and charge of the γ-carboxylic recognition site (γ-CRS) region in the propeptide are important determinants of γ-carboxylase binding. So the contribution of individual γ-CRS residues to the expression of fully γ-carboxylated and active FIX was studied.
Methods
Propeptide residues at positions −14, −13, or − 12 were substituted for equivalent prothrombin amino acids by SEOing PCR. The recombinant FIX variants were transfected and stably expressed in Drosophila S2 cells, and the expression of both total FIX protein and active FIX was assessed.
Results
While overall the substitutions resulted in an increase of both total FIX protein expression as well as an increase in the portion of active FIX, the highest increase in FIX protein expression, FIX activity, and specific FIX activity was observed following the simultaneous substitution of residues at positions −12, −13, and − 14. The enhanced rFIX activity was further confirmed by enrichment for functional, fully γ-carboxylated rFIX species via barium citrate adsorption.
Conclusion
Our findings indicate that by increasing both the net charge and the net hydrophobicity of the FIX γ-CRS region, the expression of fully γ-carboxylated and as such active FIX is enhanced.
Graphical abstract.
.
Keywords: Blood coagulation, Vitamin K-dependent proteins, γ-Carboxylation, Factor IX, Propeptide engineering, Hydrophobicity
Introduction
Vitamin K-dependent (VKD) proteins such as prothrombin (PT), factors VII (FVII), IX (FIX), and X (FX), and proteins C (PC), S (PS), and Z (PZ) play an essential role during the initiation, propagation, and regulation of the blood coagulation cascade [1]. The VKD proteins are derived from a prepro-polypeptide precursor composed of a prepeptide (signal peptide) and a propeptide that are N-terminally connected to the mature protein. The propeptide mediates the recognition and interaction of the VKD precursor with the γ-carboxylase enzyme [2], the latter converting multiple glutamic acid residues (Glu) to γ-carboxylated glutamic acid (Gla) residues in the so called Gla domains of VKD proteins. The Gla residues mediate interaction with anionic membranes of activated platelets or endothelial cells, thereby restricting coagulation to the site of vascular damage. Given the significance of this interaction, γ-carboxylation is critical for the biological activity of VKD blood coagulation proteins. The relative binding affinities characterizing the interaction of the propeptide with γ-carboxylase, which differ over two orders of magnitude [3], are inversely correlated to the turnover rate of γ-carboxylation. This is considered to be due to the substrate release by γ-carboxylase, with a quick release resulting in an efficient γ-carboxylation and as such an efficient expression of fully functional VKD proteins [4]. The γ-carboxylase enzyme interacts with the γ-carboxylase recognition site (γ-CRS) that is located at the N-terminus of the propetide, as disruption of the γ-CRS yields a mature protein that either lacks or is deficient in γ-carboxyl glutamic acids [5, 6]. While specific conserved residues in the γ-CRS such as Phe-16 and Ala-10 have been proven critical for γ-carboxylation [6, 7], non-conserved residues may also mediate the interaction with γ-carboxylase [3, 8, 9]. Among these are hydrophobic amino acids, as their absence leads to failure of γ-carboxylation [10, 11]. Furthermore, the net charge of the γ-CRS region has also been implied to affect γ-carboxylation [11]. Although improving γ-carboxylation through either propeptide replacement or targeted subsitutions within this region has been studied by us and others [4, 9, 12–15], the effect of charge and hydrophobicity of the propeptide on this process is less well-defined. Previously, replacement of the human FIX propeptide for that of the more hydrophobic propeptide of prothrombin suggested an increase in expression of functional FIX [12]. Also, to evaluate the effect of charge on this region, the FIX pro-peptide was replaced with PC that led to much higher fully γ-carboxylated material and activity (1.5 and 2 fold respectivly) [13]. Here we studied both the charge and hydrophobicity of the γ-CRS propeptide region of FIX by targeted substitution of propeptide amino acids at positions −12, −13, and − 14 and replacement by equivalent prothrombin residues.
Materials and methods
Selection and substitution of hydrophobic residues
Hydrophobic amino acids in the propeptide region of human FIX were selected using the EXPASy-PROtscale server [16]. The net hydrophobicity of amino acids in γ-CRS was calculated based on the kyte-Doolittle hydrophobicity scale [17]. Amino acids were substituted employing the splicing by overlap extension (SOEing) PCR method. Based on the position of exchange, different primers were designed (Table 1) and the mutated FIX encoding sequence was amplified from the pMT-hFIX template [18] in overlapping A and B fragments. Fragment A contained the mutated pre-pro-sequence of FIX (133 bp), while fragment B consisted of the remaining FIX sequence (1292 bp). These were amplified using KF1/R1-M and F2/XR2 primers, respectively. The modified pre–pro-peptide DNA sequence was ligated to the DNA sequence of FIX via SOEing PCR technique, where primers R1-M and F2 fused the fragments and KF1 and XR2 primers were the outside primers. Following PCR purification, the KpnI-XhoI fragments were inserted into the pMT-V5-HisA vector downstream of the Drosophila Mtn promoter. The resulting plasmid was designated as pMT-hFIX/M and the substitution of nucleotides was also confirmed using nucleotide sequence analysis.
Table 1.
Different primers that designed based on the position of exchange
| Primer name | Primer Sequence |
|---|---|
| KF1 | 5 GGGGTACCCAAAATGCAGCGCGTGAACATG 3 |
| XR2 | 5 CCGCTCGAGAATCCATCTTTCATTAAGTGAGC 3 |
| F2 | 5 ACGCCAACAAAATTCTGAATC 3 |
| R1-M12 | 5 TTTGTTGGCGTTCTGATGATC 3 |
| R1-M13 | 5 TTGTTGGCGTTTTCAGGATC 3 |
| R1-M14 | 5 TGTTGGCGTTTTCATGAGC 3 |
| R1-M12/M13 | 5 TTGTTGGCGTTCTGAGGATC 3 |
| R1-M12/M14 | 5 TGTTGGCGTTCTGATGAGC 3 |
| R1-M13/M14 | 5 TGTTGGCGTTTTCAGGAGC 3 |
| R1-M12/ M13/M14 | 5 TGTTGGCGTTCTGAGGAGC 3 |
K KpnI, X XhoI
Selection of stabe FIX-expressing clones
All constructs along with pCoHygro plasmid were transfected into Drosophila S2 cells (Invitrogen, Lot no; 769,915) employing the calcium phosphate co-precipitation method with minor modifications [19]. After 48 h, the cells were cultured in 300 μg of hygromycin B/ml. To select the clones stably expressing FIX, the single cell strategy was employed [20]. In brief, approximately 150 × 106 parental S2 cells were treated for 4 h with mitomycin C to induce cell cycle arrest, upon which these feeder cells were plated in 24-well plates at 3 × 106 cells/ml [21]. Subsequently, the FIX-transfected cells were added at 1 cell/well. After approximately two weeks, the single cell-derived clones were screened for FIX expression.
Expression and activity analysis
To assess the functional rFIX expression, stable S2 cells were induced with 0.5 mM CuSO4 in the presence of 6 μg/ml of vitamin K1. Total FIX expression was quantified in conditioned media employing a FIX-specific antigen assay (ELISA) following the manufacturer’s instructions (Asserachrom, hFIX: Ag). The FIX clotting activity was determined employing a modified one-stage aPTT assay specific for FIX as described previously [18]. For both analyses, FIX levels were corrected for measurements obtained using conditioned media from non-transfected cells. Reference curves of normal pooled plasma (provided by Hashemi Nezhad Hospital, Iran) were used to calculate the equivalent FIX concentration and clotting activity, respectively. By definition, one unit of FIX activity corresponds to the amount of FIX in 1 ml of normal plasma (~ 5 μg/ml).
rFIX transcript levels analysis
The total RNA was prepared from the induced cells using RNX-Plus Kit (SinaClon, Iran), based on the manufacturer’s instructions. Reverse transcription was performed using primers with random base sequences (usually [d(N)6]) and revertAid M-MuLV. The obtained cDNAs were used as a template for subsequent PCR experiments with two hFIX-specific oligonucleotides including hFIX-F (5’GAATGTTGGTGTCCCTTTGG3’) and hFIX-R (5’AATGGCACTGCTGGTTCAC3’) as forward and reverse primers, respectively. Using a total of 100 ng of RNA from each cell line, real-time PCR was performed using the SYBR green method and premix Amplicon kit on an ABI 7500 real-time PCR system. The ribosomal protein L32 (RPL32) for S2 cells was considered as the internal control. A 165 bp fragment of RPL32 was amplified using oligonucleotides RPL-F (5’ ATCGGTTACGGATCGAACAA 3′)/RPL-R (5’ GACAATCTCCTTGCGCTTCT3’). The real-time PCR conditions consisted of a pre-denaturation step at 95 °C for 5 min, followed by 40 cycles at 95 °C for 15 s and 58 °C for 20s.
γ-Carboxylation analysis
To quantify γ-carboxylated rFIX, the barium citrate method was used for adsorbing γ-carboxylated glutamic acid residues [20, 22]. During this process, the proteins that are poorly or not γ-carboxylated at all can not be trapped by barium caption and remain in the soluble fraction. The barium-bound FIX was subsequently eluted from the precipitate through buffer modification. FIX recovery was expressed as the percentage of FIX eluted over the FIX prior to barium citrate adsorption as determined by FIX-specific antigen analysis.
Statistical analysis
All analysis experiments were carried out in duplicates or triplets, and the generated data were presented as the mean ± SD. The ANOVA (analysis of variance) program for the analysis of variance followed by a Tukey post-hoc test were performed for evaluation; P < 0.05 was considered statistically significant. All statistical analyses were carried out with SPSS 16 (SPSS Inc., Chicago, IL, USA).
Results
Selection of FIX γ-CRS residues for targeted substitution
Since amino acids are either hydrophilic or hydrophobic depending on the side chains (R), the hydropathy index of the γ-CRS in various VKD coagulation proteins (defined as amino acid at positions −18 through −9) was calculated emplyong the Kyte-Doolittle scale [17] (Table 2). The latter is based on the interaction of the R group with water, and the free energy transfer (ΔGtrans°) of the solute amino acid between water and the condensed vapor phase is determined. A negative ΔGtrans° indicates a strong affinity of the R group for water (hydrophilic), while a positive value suggests the opposite (hydrophobic) [23]. The γ-CRS net charge was also based on the amino acid side chains, with five residues having a side chain which has the potential to carry an electric charge [24] (Table 2). At pH 7, two are negatively charged: aspartic acid (Asp, D) and glutamic acid (Glu, E) (acidic side chains; charge −1), while three are positively charged: lysine (Lys, K), arginine (Arg, R), and histidine (His, H) (basic side chains; charge 1). Based on these values, the sum of electric charge of amino acids (net electric charge) in the γ-CRS region was calculated (Table 2). The suggest that a lower relative binding affinity (Ki) for γ-carboxylase is correlated with a higher net charge, as observed for PC and prothrombin, and with a trend towards a more hydrophobic γ-CRS region. Since the prothrombin propeptide has the lowest relative binding affinity for γ-carboxylase, the FIX γ-CRS region was modified to resemble that of prothrombin, with the aim of reducing the affinity of the propeptide for γ-carboxylase, thereby increasing γ-carboxylase-mediated turnover. To do so, we specifically targeted those residues with a charged side chain and displaying the lowest hydrophobicity in the FIX γ-CRS and substituted these for the homologous prothrombin residues (Table 3).
Table 2.
Charge and hydrophobicity of the γ-CRS region in blood coagulation VKD proteins
| Name | Ki (nM) [4] | Net charge (−18 to −9) | Net hydrophobicity (−18 to −9) | Residue number | -18 | −17 | −16 | −15 | −14 | −13 | −12 | −11 | −10 | −9 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| FX | 2.6 | 1 | −7.4 | Residue | S | L | F | I | R | R | E | Q | A | N |
| Charge | 1 | 1 | −1 | |||||||||||
| Hydrophobicity | −0.8 | 3.8 | 2.8 | 4.5 | −4.5 | −4.5 | −3.5 | −3.5 | 1.8 | −3.5 | ||||
| FVII | 11.1 | 0 | −5.9 | Residue | R | V | F | V | T | Q | E | E | A | H |
| Charge | 1 | −1 | −1 | 1 | ||||||||||
| Hydrophobicity | −4.5 | 4.2 | 2.8 | 4.2 | −0.7 | −3.5 | −3.5 | −3.5 | 1.8 | −3.2 | ||||
| PS | 12.2 | 1 | −5.8 | Residue | A | N | F | L | S | K | Q | Q | A | S |
| Charge | 1 | |||||||||||||
| Hydrophobicity | 1.8 | −3.5 | 2.8 | 3.8 | −0.8 | −3.9 | −3.5 | −3.5 | 1.8 | −0.8 | ||||
| FIX | 33.6 | −1 | −5.3 | Residue | T | V | F | L | D | H | E | N | A | N |
| Charge | −1 | 1 | −1 | |||||||||||
| Hydrophobicity | −0.7 | 4.2 | 2.8 | 3.8 | −3.5 | −3.2 | −3.5 | −3.5 | 1.8 | −3.5 | ||||
| PC | 230 | 1 | −5.6 | Residue | S | V | F | S | S | S | E | R | A | H |
| Charge | −1 | 1 | 1 | |||||||||||
| Hydrophobicity | −0.8 | 4.2 | 2.8 | −0.8 | −0.8 | −0.8 | −3.5 | −4.5 | 1.8 | −3.2 | ||||
| PT | 277 | 2 | −1.9 | Residue | H | V | F | L | A | P | Q | Q | A | R |
| Charge | 1 | 1 | ||||||||||||
| Hydrophobicity | −3.2 | 4.2 | 2.8 | 3.8 | 1.8 | −1.6 | −3.5 | −3.5 | 1.8 | −4.5 |
Table 3.
Charge and hydrophobicity of the γ-CRS region in FIX variants
| FIX Variants | Net Hydrophobicity | Net Charge |
|---|---|---|
| Wild-type | −5.3 | −1 |
| E-12Q | −5.3 | 0 |
| H-13P | −3.7 | −2 |
| D-14A | 0 | 0 |
| E−12Q / H-13P | −3.7 | -1 |
| E-12Q / D-14A | 0 | +1 |
| H-13P / D-14A | 1.6 | -1 |
| E-12Q / H-13P / D-14A | 1.6 | 0 |
Expression and activity of rFIX
Following transfection and stable expression of the FIX variants in S2 cells, total FIX expression was monitored during 24–72 h. Introduction of individual substitutions at positions −12, −13 and − 14 in the FIX γ-CRS resulted in enhanced rFIX expression, with an up to 1.7-fold increase as demonstrated by D-14A-FIX (Fig. 1). While combining multiple substitutions resulted in a near-identical increase in total rFIX expression, the largest enhancement (1.8-fold) was observed for simultaneous mutations at positions −12, −13, and − 14 (Fig. 1).
Fig. 1.
Concentration of rFIX expressed by S2-FIX or S2-mFIX. Following propagation of stable clones, rFIX expression was induced with 0.5 mM CuSO4 in the presence of 6 μg/mL vitamin K1. At various post induction times (24–72 h), rFIX expression was assessed in conditioned media by ELISA as described in ‘Materials and methods’. The data are the means ± S.D. of three similar experiments and the number on top of each column represents fold change to wt-FIX
We next assessed the rFIX activity in the 24–72 h following CuSO4-mediated induction of rFIX expression. Consistent with ELISA results, comparison of the variant rFIX activity with that of wild-type rFIX revealed elevated levels (2.2-fold) of functional rFIX expression following the D-14A and E-12Q / D-14A substitutions (Fig. 2a). Furthermore, the double H-13P / D-14A and triple E-12Q / H-13P / D-14A rFIX mutants yielded the highest increase in active rFIX expression, by 2.9- and 3.5-fold, respectively. Since we expect mutations to affect the extent of γ-carboxylation and hence the functional activity of FIX, we determined the specific FIX activity. A 1.2-fold increase in specific FIX activity was observed for the E-12Q, H-13P, and D-14A variants, and a 1.6-fold increase for the single D-14A and double E-12Q / D-14A mutants (Fig. 2b). Finally, the H-13P / D-14A and E-12Q / H-13P / D-14A FIX variants displayed a ≥ 2-fold enhanced specific FIX activity. The improved specific FIX activity observed for the FIX variants indicates that the specific substitutions specifically impact the proportion of active FIX that is expressed.
Fig. 2.
Coagulation activity (a) and specific activity (b) of rFIX expressed by S2 cells. a Following propagation of stable clones, rFIX expression was induced with 0.5 mM CuSO4 in the presence of 6 μg/mL vitamin K1. At various post induction times (24–72 h), rFIX expression was assessed in conditioned media employing the FIX-specific clotting assay as described in ‘Materials and methods’. The data are the means ± S.D. of three similar experiments. b The specific FIX activity was determined by dividing the FIX activity over rFIX antigen values. The number on top of each column represents fold change to wt-FIX
Quantification of γ-carboxylated FIX
To check whether the substitutions would enhance rFIX γ-carboxylation, barium citrate precipitation of γ-carboxylated rFIX was performed. The amount of rFIX recovered from barium citrate adsorption revealed an overall enhanced recovery of rFIX variant relative to wild-type rFIX (Fig. 3). Up to 52% of the mutated rFIX and up to 22% of wild-type rFIX were recovered via barium citrate adsorption, indicating that the specific substitution enhances the γ-carboxylation process. Collectively, these findings suggest that substituting the FIX residues at positions −12, −13, and − 14 for those of prothrombin results in the production of higher levels of active rFIX.
Fig. 3.
Barium citrate adsorption of rFIX. Following propagation of stable clones, rFIX expression was induced with 0.5 mM CuSO4 in the presence of 6 μg/mL vitamin K1. At various post induction times (24–72 h), rFIX expression was assessed in conditioned media employing the FIX-specific clotting assay as described in ‘Materials and methods’. rFIX antigen was determined in conditioned media prior to barium citrate adsorption (–Ba2+) or after barium citrate adsorption and elution (+Ba2+), and FIX recovery was expressed as the percentage of FIX eluted over FIX prior to barium citrate adsorption as described in ‘Materials and methods’. The data are the means ± S.D. of three independent experiments and the number on top of each column represnts folds change to wt-FIX
The rFIX translation and secretion efficiency
Given that the expression of functional rFIX variants was enhanced, we assessed rFIX mRNA levels employing quantitative real-time PCR. As expected, no difference was observed in rFIX mRNA level for the S2 cells expressing the triple E-12Q / H-13P / D-14A rFIX mutant relative to those of wild-type rFIX (Fig. 4). However, examination of the conditioned media and cell lysates for rFIX antigen levels indicated that the total expression of wild-type rFIX increased over time up to 607 ng/ml/106 cells with 72% secretion efficiency (Table 4). Interestingly, up to 705 ng/ml/106 cells with 94% secretion efficiency was observed for rFIX variant E-12Q / H-13P / D-14A, demonstrating an increased translation and secretion efficiency for FIX following specific substitutions in the γ-CRS region.
Fig. 4.
Real-time PCR analysis for rFIX expression. The rFIX mRNA levels in S2-FIX/12.13.14 cells 72 h after CuSO4-mediated induction were assessed by real-time RT-PCR as described in ‘Materials and methods’. rFIX mRNA levels in S2-FIX cells 72 h after CuSO4-mediated induction were used as control, and all mRNA levels were expressed as ratio over the control. The data are means ± SEM of three independent experiments
Table 4.
rFIX antigen levels in conditioned media and cell lysates
| FIX variant | FIX expression (hours post induction) | Secreted rFIX (ng/ml/106 cell) | Intracellular rFIX (ng/ml/106 cell) | Total (ng/ml/106 cell) | Secretion efficiency (%) |
|---|---|---|---|---|---|
| E-12Q / H-13P / D-14A | |||||
| 24 | 369 ± 13 | 29 ± 4 | 398 | 93 | |
| 48 | 507 ± 21 | 50 ± 3 | 557 | 91 | |
| 72 | 660 ± 19 | 45 ± 4 | 705 | 94 | |
| wild-type | |||||
| 24 | 204 ± 15 | 99 ± 7 | 303 | 67 | |
| 48 | 367 ± 17 | 134 ± 11 | 501 | 73 | |
| 72 | 438 ± 23 | 169 ± 8 | 607 | 72 | |
Discussion
In the current study we investigated the contribution of specific residues in the γ-CRS propeptide region of blood coagulation FIX to the production and expression of active and as such fully γ-carboxylated recombinant FIX. The 18 amino acid propeptide of VKD precursor proteins is known to mediate the recognition and interaction with the γ-carboxylase enzyme [2]. This is consistent with the observation that precursors lacking a propeptide are a poor substrate for γ-carboxylase, demonstrated by a 10- to 100-fold reduction in substrate affinity [25]. Further studies on the contribution of specific amino acids within the propeptide indicated that residues N-terminal to the consensus sequence motif Lys/Arg-X-X-L-X-X-X-X-Lys/Arg (sequence −8 to −1, with L being a Leu or other hydrophobic residue) mainly facilitate the γ-carboxylase interaction [25]. This is corroborated by studies demonstrating a dramatic impairment of γ-carboxylation following either deletion of the −17 to −13 propeptide sequence or point mutations at positions −18, −17, −16, −15, −10, −9, or − 6 [3, 7, 8, 26], implying that these residues are part of the γ-CRS region. Substitution at positions −8, −6, −4 or − 1 did not affect γ-carboxylation [8], indicating that these are not γ-CRS residues. Collectively, these findings suggest that the γ-CRS in the VKD propeptides spans amino acids −18 to −9.
A hydrophobicity analysis indicated that the net hydrophobicity of propeptide sequence −18 to −9 is inversely correlated with the binding affinity of the VKD proteins for γ-carboxylase (Ki, Table 2). Consistent with this, we have earlier shown that replacing the FIX propeptide with a less hydrophobe γ-CRS region (net hydrophobicity −5.3) for that of prothrombin with a more hydrophobe γ-CRS sequence (net hydrophobicity −1.9) results in increased expression of fully γ-carboxylated and functional FIX [12]. Interestingly, although the PC and FIX γ-CRS sequences display a similar net hydrophobicity, the affinity of FIX for γ-carboxylase is almost 7-fold enhanced relative to PC, with the latter displaying a binding affinity similar to prothrombin (Table 2). This suggests that hydrophobicity is not the single determinant of γ-carboxylase binding. Given the sequential increase in net charge of the −18 to −9 sequences of FIX, PC, and prothrombin, respectively (Table 2), the net charge of this region may contribute to γ-carboxylase binding as well. The latter is supported by our previous work, in which the FIX propeptide was exchanged for that of PC, thereby increasing the net charge of the −18 to −9 propeptide sequence while maintaining the net hydrophibicity [13] . Despite the substantial difference in Ki for PC and FIX (230 vs. 33.6 nM), the replacement of the FIX propeptide for the PC propeptide resulted in the expression of more γ-carboxylated and as such functional FIX. The results suggested that in addition to the hydrophobicity, the charge of γ-CRS region is also important in the rate of γ-carboxylation.
To examine the effect of both charge and hydrophobicity of the γ-CRS region on the γ-carboxylation of FIX, amino acids −12, −13, and − 14 were selected and substituted for the equivalent amino acids in prothrombin. While all individual substitutions enhanced both FIX total and active protein expression as well as the specific FIX activity (Figs. 1, 2), simultaneous substitution at all three positions yielded the highest increase in active rFIX expression. Specifically, the FIX E-12Q / H-13P / D-14A variant displayed a ≥ 2-fold enhanced specific FIX activity (Fig. 2), indicating that the substitutions enhance the proportion of active FIX that is expressed. Substitution of amino acids −12, −13, and − 14 in the FIX propeptide by the equivalent prothrombin residues resulted in both an increase in the net hydrophobicity from −5.3 to 1.6 and net charge from −1 to 0 of the γ-CRS region, relative to that of wild-type FIX (Table 3). Given the enhanced production of active FIX following these substitutions, our findings indicate that by increasing both the net charge and the net hydrophobicity of the FIX γ-CRS region, the turnover and therefore activity of γ-carboxylase may be enhanced.
Support for the modulation of the net charge and/or hydrophobicity of the γ-CRS region resulting in a modified turnover by γ-carboxylase and subsequent γ-carboxylation comes from several studies. For instance, replacement of -10A by a Glu residue significantly reduced the γ-carboxylation of FIX [7]. Alanine is a hydrophobic amino acid with a hydrophobicity of 1.8, but glutamate is a negative amino acid with an hydrophobicity of −5.3. Therefore, the replacement of A to E substitution at position −10 reduces the net charge (from −1 to −2) and net hydrophobicity (from −5.3 to −10.6) in the FIX γ-CRS region. Similar findings were obtained for prothrombin, where the Ala at position −10 was substituted for an aspartic acid, thereby reducing the net charge from −1 to −2 and the net hydrophobicity from −5.3 to −10.6 in the γ-CRS region [3], leading to a significant reduction in γ-carboxylation [8]. Conversely, the -9 N to K substitution in FIX, while not affecting the hydrophobicity (from −5.3 to −5.7), did affect the net charge of the γ-CRS region (from −1 to 0). This substitution resulted in an increase in Ki from 33 to 370 nM and enhanced γ-carboxylation of FIX [3]. Moreover, while the net charge remined constant following the the -9 N to L substitution in FIX, the net hydrophobicity will change to (from −5.3 to 2), which coincides with an increase in Ki from 33 to 460 nM [3]. As such, these examples further point to a role for both charge and hydrophobicity of the FIX γ-CRS region in the functional interaction with γ-carboxylase, which is essential for an efficient turnover and γ-carboxylation.
As anticipated, no difference was observed in the rFIX mRNA levels when comparing between wt-FIX and the triple mutant 72 h after induction as assessed by real-time analysis (Fig. 4), which is consistent with previous findings [27, 28]. These studies have demonstrated that the mRNA levels of the transgene are largely unaffected even when a heterologous signal peptide is used. Interestingly, we observed an improved secretion efficiency for the FIX E-12Q / H-13P / D-14A variant (Table 4). Wspeculate e that the enhanced secretion of fully γ-carboxylated rFIX from the ER to the media allows for rapid processing of newly synthesized rFIX through the ER-lumen pathway. Confirmation of process could follow from quantification of the mRNA-ribosome complex to that was halted for translation by the signal recognition particle (SRP) in the cytoplasm through Fluorescence Resonance Energy Transfer (FRET) [29] or fluorescence-based stopped-flow techniqes [30].
In conclusion, our results indicate that an increased functional expression of rFIX is due to the presence of hydrophobic amino acids in the propeptide. Furthermore, increasing the net charge as well as the net hydrophobicity of the FIX γ-CRS region increased the turnover by γ-carboxylase thereby enhancing the production of fully γ-carboxylated and functional FIX.
Acknowledgements
This work has been supported by a grant (Project No. 96004545) from the Iran National Science Foundation (INSF) to Jafar Vatandoost.
Funding
This study was funded by the Iran National Science Foundation (Project No. 96004545).
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Roberts HR, Monroe D, Hoffman M. Molecular biology and biochemistry of the coagulation factors and pathways of hemostasis. Williams hematology. 2001;6:1409–1434. [Google Scholar]
- 2.Pan LC, Price PA. The propeptide of rat bone gamma-carboxyglutamic acid protein shares homology with other vitamin K-dependent protein precursors. Proc Natl Acad Sci. 1985;82:6109–6113. doi: 10.1073/pnas.82.18.6109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Stanley TB, Jin D-Y, Lin P-J, Stafford DW. The propeptides of the vitamin K-dependent proteins possess different affinities for the vitamin K-dependent carboxylase. J Biol Chem. 1999;274:16940–16944. doi: 10.1074/jbc.274.24.16940. [DOI] [PubMed] [Google Scholar]
- 4.Camire RM, Larson PJ, Stafford DW, High KA. Enhanced γ-carboxylation of recombinant factor X using a chimeric construct containing the prothrombin propeptide. Biochemistry. 2000;39:14322–14329. doi: 10.1021/bi001074q. [DOI] [PubMed] [Google Scholar]
- 5.Wallin R, Hutson SM. Warfarin and the vitamin K-dependent γ-carboxylation system. Trends Mol Med. 2004;10:299–302. doi: 10.1016/j.molmed.2004.05.003. [DOI] [PubMed] [Google Scholar]
- 6.Chu K, Wu S-M, Stanley T, Stafford DW, High KA. A mutation in the propeptide of factor IX leads to warfarin sensitivity by a novel mechanism. J Clin Invest. 1996;98:1619–1625. doi: 10.1172/JCI118956. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Jorgensen MJ, Cantor AB, Furie BC, Brown CL, Shoemaker CB, Furie B. Recognition site directing vitamin K-dependent γ-carboxylation resides on the propeptide of factor IX. Cell. 1987;48:185–191. doi: 10.1016/0092-8674(87)90422-3. [DOI] [PubMed] [Google Scholar]
- 8.Huber P, Schmitz T, Griffin J, Jacobs M, Walsh C, Furie B, et al. Identification of amino acids in the gamma-carboxylation recognition site on the propeptide of prothrombin. J Biol Chem. 1990;265:12467–12473. [PubMed] [Google Scholar]
- 9.Handford P, Winship P, Brownlee G. Protein engineering of the propeptide of human factor IX. Protein Eng. 1991;4:319–323. doi: 10.1093/protein/4.3.319. [DOI] [PubMed] [Google Scholar]
- 10.Czerwiec Eva, Begley Gail S., Bronstein Mila, Stenflo Johan, Taylor Kevin, Furie Barbara C., Furie Bruce. Expression and characterization of recombinant vitamin K-dependent γ-glutamyl carboxylase from an invertebrate, Conus textile. European Journal of Biochemistry. 2002;269(24):6162–6172. doi: 10.1046/j.1432-1033.2002.03335.x. [DOI] [PubMed] [Google Scholar]
- 11.Vermeer C. Gamma-carboxyglutamate-containing proteins and the vitamin K-dependent carboxylase. Biochem J. 1990;266:625–636. doi: 10.1042/bj2660625. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Khorshidi S, Zomorodipour A, Behmanesh M, Vatandoost J, Bos MH. Functional expression of the human coagulation factor IX using heterologous signal peptide and propeptide sequences in mammalian cell line. Biotechnol Lett. 2015;37:1773–1781. doi: 10.1007/s10529-015-1868-3. [DOI] [PubMed] [Google Scholar]
- 13.Vatandoost J, Hassankhani M. Effect of prepro-peptide replacement on γ-carboxylation and secretion of recombinant factor IX in Drosophila S2 cells. Iran J Pharm Res. 2018; in press.
- 14.Vatandoost J, Zarei Sani O. Effect of propeptide amino acid substitution in γ-carboxylation, activity and expression of recombinant human coagulation factor IX. Biotechnol Prog. 2018;34:515–520. doi: 10.1002/btpr.2582. [DOI] [PubMed] [Google Scholar]
- 15.Rabiet M-J, Jorgensen MJ, Furie B, Furie BC. Effect of propeptide mutations on post-translational processing of factor IX. Evidence that beta-hydroxylation and gamma-carboxylation are independent events. J Biol Chem. 1987;262:14895–14898. [PubMed] [Google Scholar]
- 16.Gasteiger E, Hoogland C, Gattiker A, Wilkins MR, Appel RD, Bairoch A. Protein identification and analysis tools on the ExPASy server. The proteomics protocols handbook. Springer; 2005. p. 571–607.
- 17.Kyte J, Doolittle RF. A simple method for displaying the hydropathic character of a protein. J Mol Biol. 1982;157:105–132. doi: 10.1016/0022-2836(82)90515-0. [DOI] [PubMed] [Google Scholar]
- 18.Vatandoost J, Zomorodipour A, Sadeghizadeh M, Aliyari R, Bos MH, Ataei F. Expression of biologically active human clotting factor IX in Drosophila S2 cells: γ-carboxylation of a human vitamin K-dependent protein by the insect enzyme. Biotechnol Prog. 2012;28:45–51. doi: 10.1002/btpr.723. [DOI] [PubMed] [Google Scholar]
- 19.Cherbas L, Cherbas P. Transformation of Drosophila cell lines. Baculovirus and insect cell expression protocols. Springer; 2007. p. 317–40.
- 20.Vatandoost J, Bos MH. Efficient expression of functional human coagulation factor IX in stably transfected Drosophila melanogaster S2 cells; comparison with the mammalian CHO system. Biotechnol Lett. 2016;38:1691–1698. doi: 10.1007/s10529-016-2156-6. [DOI] [PubMed] [Google Scholar]
- 21.Nilsen SL, Castellino FJ. Expression of human plasminogen inDrosophilaSchneider S2 cells. Protein Expr Purif. 1999;16:136–143. doi: 10.1006/prep.1999.1045. [DOI] [PubMed] [Google Scholar]
- 22.Yao SN, Wilson JM, Nabel EG, Kurachi S, Hachiya HL, Kurachi K. Expression of human factor-ix in rat capillary endothelial-cells - toward somatic gene-therapy for hemophilia-b. Proc Natl Acad Sci U S A. 1991;88:8101–8105. doi: 10.1073/pnas.88.18.8101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Burns A., Olszowy P., Ciborowski P. Proteomic Profiling and Analytical Chemistry. 2016. Biomolecules; pp. 7–24. [Google Scholar]
- 24.Alberts B, Bray D, Hopkin K, Johnson A, Lewis J, Raff M et al. Chemical compound of cell Essential cell biology Garland Science; 2013. p. 55.
- 25.Czerwiec Eva, Kalume Dario E., Roepstorff Peter, Hambe Björn, Furie Bruce, Furie Barbara C., Stenflo Johan. Novel γ-carboxyglutamic acid-containing peptides from the venom ofConus textile. FEBS Journal. 2006;273(12):2779–2788. doi: 10.1111/j.1742-4658.2006.05294.x. [DOI] [PubMed] [Google Scholar]
- 26.Foster Donald C., Rudinski Mark S., Schach Barbara G., Berkner Kathleen L., Kumar A. Ashok, Hagen Frederick S., Sprecher Cindy A., Insley Margaret Y., Davie Earl W. Propeptide of human protein C is necessary for .gamma.-carboxylation. Biochemistry. 1987;26(22):7003–7011. doi: 10.1021/bi00396a022. [DOI] [PubMed] [Google Scholar]
- 27.Tröße C, Ravneberg H, Stern B, Pryme IF. Vectors encoding seven oikosin signal peptides transfected into CHO cells differ greatly in mediating Gaussia luciferase and human endostatin production although mRNA levels are largely unaffected. Gene Regul Syst Bio. 2007;1. [PMC free article] [PubMed]
- 28.Knappskog S, Ravneberg H, Gjerdrum C, Tröβe C, Stern B, Pryme IF. The level of synthesis and secretion of Gaussia princeps luciferase in transfected CHO cells is heavily dependent on the choice of signal peptide. J Biotechnol. 2007;128:705–715. doi: 10.1016/j.jbiotec.2006.11.026. [DOI] [PubMed] [Google Scholar]
- 29.Noriega Thomas R., Tsai Albert, Elvekrog Margaret M., Petrov Alexey, Neher Saskia B., Chen Jin, Bradshaw Niels, Puglisi Joseph D., Walter Peter. Signal Recognition Particle-ribosome Binding Is Sensitive to Nascent Chain Length. Journal of Biological Chemistry. 2014;289(28):19294–19305. doi: 10.1074/jbc.M114.563239. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Mercier E, Holtkamp W, Rodnina MV, Wintermeyer W. Signal recognition particle binds to translating ribosomes before emergence of a signal anchor sequence. Nucleic Acids Res. 2017;45:11858–11866. doi: 10.1093/nar/gkx888. [DOI] [PMC free article] [PubMed] [Google Scholar]