In Vivo Clearance of Human Protein S in a Mouse Model
http://www.100md.com
动脉硬化血栓血管生物学 2005年第10期
From INSERM U.143 (C.V.D.), Kremlin-Bicêtre, France; the Laboratory for Thrombosis and Haemostasis (S.J.R., P.J.L.), Department of Haematology, University Medical Center Utrecht, Utrecht, the Netherlands; and the Department of Biochemistry (T.M.H.), University of Maastricht, Maastricht, the Netherlands.
Correspondence to Dr P.J. Lenting, Laboratory for Thrombosis and Haemostasis, Department of Haematology, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX Utrecht, the Netherlands. E-mail p.j.lenting@lab.azu.nl
Abstract
Objective— To explore the effect of the Heerlen polymorphism and C4b-binding protein (C4BP) on protein S catabolism in vitro and in vivo.
Methods and Results— Radiolabeled protein S was efficiently bound and intracellularly degraded by THP-1 macrophages, and both processes were strongly reduced in the presence of the protein S-carrier protein C4BP. To test whether C4BP displays a similar protective effect in vivo, survival experiments were performed in mice. In the absence of C4BP, radiolabeled human protein S disappeared in a biphasic manner (mean residence time [MRT] 2 hours). However, the presence of C4BP resulted in a 4-fold prolonged survival of protein S (MRT 8 hours; P<0.0001). We also applied this experimental model to recombinant protein S-Heerlen, a naturally occurring variant that contains a Ser460Pro substitution. These clearance experiments revealed a strongly decreased survival of recombinant protein S-S460P (MRT 0.6 hours; P=0.021), which could be compensated partially by C4BP (MRT 1.4 hours; P=0.012 compared with protein S-S460P).
Conclusion— Protein S-S460P has a reduced survival in vivo, which may explain the low levels of free protein S in individuals carrying this polymorphism. Furthermore, C4BP prevents premature clearance of protein S and uses this ability to compensate the increased clearance of protein S-S460P.
We established that binding and degradation of protein S by macrophages was partially inhibited by C4BP. C4BP mimicked this effect in vivo by prolonging the half-life of protein S from 2 to 4 hours in a mouse model. The presence of the Heerlen polymorphism reduced survival of protein S 3-fold to 0.6 hours.
Key Words: protein S ? C4b-binding ? clearance
Introduction
Protein S circulates in plasma at a concentration of 0.3 μmol/L, although this value may vary between individuals.1 Only 30% to 40% of protein S circulates in a free form, whereas the remaining part is in complex with C4b-binding protein (C4BP).2 Various isoforms of C4BP exist, and only those that contain the ?-chain (ie, C4BP?) bind protein S.3 Because the C4BP? plasma concentration is lower than that of protein S, the molar excess of protein S over C4BP? determines the amount of free protein S.
The function of protein S relates to various physiological processes, including the clearance of apoptotic cells and neuronal protection during ischemic/hypoxic injury.4–6 However, protein S is originally known from its role in the anticoagulant process,7,8 in which it functions as a cofactor for activated protein C (APC).9,10 In addition, protein S may downregulate the coagulation pathway independently of APC.11,12 The notion that deficiencies in protein S are associated with an increased thrombotic tendency13,14 demonstrates that appropriate plasma levels of protein S are needed to maintain the hemostatic balance.
Two categories of protein S deficiency can be distinguished:15 (1) qualitative defects (type II deficiency), and (2) quantitative defects (type I or III deficiency). Type I deficiency refers to a parallel reduction of total and free protein S. Type III deficiency is characterized by low levels of free protein S but normal levels of total protein S. However, both types of quantitative deficiencies often occur within the same family,1 and more detailed analysis revealed that increased levels of total protein S as seen in type III coincide with an age-dependent increase in protein S levels.16
Quantitative deficiencies of protein S may obviously originate from null alleles, but also, missense mutations can lead to reduced protein S levels.15 Some of these missense mutations result in impaired secretion of the protein.17–19 However, not all amino acid replacements associated with low levels of protein S associate with abnormal secretion. An example hereof is protein S-Heerlen, which contains a Ser to Pro replacement at position 460.20 This substitution results in loss of N-linked glycosylation at position Asn458. Protein S-Heerlen is found in 0.5% of the population20 and is associated with reduced levels of free protein S in heterozygous or homozygous individuals.21,22 Expression studies demonstrated that recombinant protein S-S460P is normally secreted,21 indicating that other mechanisms should be involved that explain the quantitative deficiency of protein S-Heerlen. Possible mechanisms that have been proposed include increased affinity for C4BP? or multivalency of C4BP? for this variant.23 However, protein S-S460P and protein S were similar in their interaction with C4BP? in various other reports.20,21,24,25
The notion that protein S-S460P is similar to wild-type (wt)–protein S in terms of secretion efficiency and binding to C4BP? prompted us to investigate the previously proposed hypothesis that protein S-Heerlen is cleared from the circulation more rapidly than normal protein S (Dr R.M. Bertina, XVI ISTH Congress, Florence, Italy, 1997). Therefore, a mouse model was used to investigate the in vivo survival of wt-protein S and recombinant protein S-S460P in the presence and absence of C4BP?.
Methods
Mice
The mice used in the present study were on a C57BL/6J background and were used between 8 and 12 weeks of age. Housing and experiments were done as recommended by French regulations and the experimental guidelines of the European Community.
Reagents and Proteins
Chloroquine was from Serva. Methyl-?-cyclodextrin (M?CD) was from Sigma-Aldrich. C4BP? was prepared from plasma via barium–citrate absorption and immunoaffinity chromatography to homogeneity (see supplemental data, available online at http://atvb.ahajournals.org). No protein S antigen could be detected in these C4BP? preparations in an immunosorbent assay, whereas the ability to bind protein S was fully retained. Purification of recombinant C4BP, recombinant wt-protein S, and recombinant protein S-S460P has been described previously24,26,27 (see also online supplemental data).
Radiolabeling of Protein S
Purified wt-protein S or protein S-S460P was labeled with Na125I (Amersham Biosciences Inc) using the IODO-GEN method28 (online supplemental data). Final preparations had specific activities varying between 1.4x106 and 2.2x106 cpm/μg protein (n=2 to 10), whereas free iodine was routinely <6%. This method to radiolabel protein S does not affect C4BP? binding capacity (data not shown) and APC cofactor activity.28
Cellular Binding, Internalization, and Degradation Experiments
Cellular binding and degradation experiments were performed using human THP-1 cells (ATCC TIB-202). These cells have been used previously to study the role of protein S in the engulfment of apoptotic cells by macrophages.29 Binding, internalization, and degradation of protein S by THP-1 macrophages were investigated using standard cell biological methods, which are described in detail in the online supplemental data.
Clearance and Biodistribution of Protein S in Mice
Normal C57BL/6J mice were injected intravenously with radiolabeled protein S or protein S-S460P (5 μg per mouse, corresponding to 9x106 cpm per mouse) in the presence or absence of a 2-fold molar excess C4BP?, diluted in PBS containing 3% bovine albumin. To facilitate complex formation, protein S (0.3 μmol/L) and C4BP? (0.6 μmol/L) were incubated for 45 minutes before injection. Because the concentration of both proteins is >300-fold above the KD value (<1 nmol/L),24 these conditions result in >95% of protein S being in complex with C4BP?. Furthermore, concentrations of C4BP? were chosen to be 4-fold below those that saturate the clearance system. At different time points after injection (3, 15, and 30 minutes, and 1, 2, and 4 hours), mice were anesthetized with tribromoethanol (0.15 mL/10 g of body weight), and blood was collected by eye bleed. Subsequently, plasma was prepared as described30 and analyzed for the presence of radioactivity. After bleeding, mice were euthanized, and various organs (heart, lungs, liver, spleen, stomach, kidneys, and left upper leg) were collected, which were analyzed for the amount of radioactivity. Three mice were used for each time point.
Data Analysis and Statistics
Data of in vivo survival experiments were fitted to a biexponential equation as described previously31 to allow calculation of the parameters mean residence time (MRT), t1/2, and t1/2?. Statistical analyses were performed by Student’s unpaired t test using the GraphPad InStat program (GraphPad Instat version 3.0 for Windows; GraphPad Software).
Results
Binding of Protein S to THP-1 Macrophages
Recently, protein S has been reported to mediate the engulfment of apoptotic cells by macrophages.4 Here, we tested whether human THP-1 macrophages are able to bind and deliver protein S to the intracellular degradation pathway. Binding was assessed by incubating radiolabeled protein S (0 to 0.52 μmol/L) with THP-1 macrophages for 2 hours at 4°C. Protein S bound to THP-1 macrophages in a saturable and dose-dependent manner, and half-maximal binding was at a concentration of 0.11±0.04 μmol/L (Figure 1A). The number of binding sites was found to be 4.3±0.5x105 per cell. Specificity of binding was addressed in 2 types of control experiments. First, labeled and unlabeled protein S were applied to the cells at different ratios while maintaining total protein S concentration constant. A linear decrease in binding of radiolabeled protein S was observed (slope –1.0±0.1; Y-intercept 105±3), indicating that labeled and unlabeled protein S bind THP-1 macrophages similarly. Second, binding of radiolabeled protein S was reduced >90% by a 100-fold excess of unlabeled protein S (Figure 1A, inset).
Figure 1. Binding, internalization, and degradation of wt-protein S by THP-1 macrophages. A, THP-1 macrophages were incubated for 2 hours at 4°C with radiolabeled wt-protein S (0 to 0.5 μmol/L). Bound wt-protein S was determined as described in Methods and corrected for nonspecific binding in the absence of cells (routinely <15%). A, Inset, Binding of radiolabeled wt-protein S (0.13 μmol/L) was assessed in the absence or presence of a 100-fold molar excess of unlabeled wt-protein S. B, THP-1 macrophages were incubated with radiolabeled wt-protein S for 1 hour at 4°C. After washing, cells were placed at 37°C to initiate internalization. At indicated time intervals, the amount of surface-bound (?) and internalized () protein S was determined as described in Methods. Plotted is the amount of radioactivity relative to the amount of surface-bound radioactivity at t=0 (100%). Data are corrected for the amount of radioactivity present in the absence of cells (<15%). C, THP-1 macrophages were incubated with radiolabeled wt-Protein S (0.13 μmol/L) either in the absence (?) or presence of 10 mmol/L M?CD () or 0.1 mmol/L chloroquine () for 10 minutes at 37°C. After washing, incubation was continued at 37°C for indicated time intervals. The amount of degraded wt-protein S was determined as described in Methods and corrected for nonspecific degradation in the absence of cells (routinely <10%). Data represent mean±SEM of 2 to 10 experiments performed in duplicate.
Internalization and Degradation of Protein S
Subsequently, we tested whether binding was followed by transport of protein S to the intracellular degradation pathway. Radiolabeled protein S (0.13 μmol/L) was incubated with macrophages (1 hour at 4°C) to allow surface binding. After washing, cells were put at 37°C to initiate internalization. Cell surface–bound protein S had disappeared by >90% within 40 minutes (Figure 1B). Simultaneously, the amount of internalized protein S had increased and reached a maximum after 40 minutes, which was followed by a gradual decline. Degradation was examined by monitoring the formation of trichloroacetic acid–soluble fragments on incubation with radiolabeled protein S (0.13 μmol/L). In time, increased levels of degradation products were observed (Figure 1C). Degradation was markedly reduced in the presence of M?CD (interferes with caveolae- and clathrin-dependent endocytosis) or the lysosomal inhibitor chloroquine (Figure 1C). Together, these data demonstrate that human macrophages can bind and transport protein S to their intracellular degradation pathway.
C4BP? Partially Inhibits Macrophage-Dependent Protein S Degradation
Because part of protein S circulates in complex with C4BP?, cellular uptake of protein S was tested in the presence of its carrier protein. When assessed for binding to THP-1 macrophages, a 5-fold excess of C4BP? reduced binding by 60%, whereas the nonbinding counterpart C4BP was unable to reduce binding (Figure 2A). A similar effect was observed when protein S degradation was monitored. Whereas C4BP did not inhibit degradation, C4BP? reduced degradation 4- to 5-fold (Figure 2B). These levels were reduced to a similar extent as observed for the degradation inhibitors M?CD and chloroquine. Apparently, complex formation with C4BP? reduces the cellular uptake of protein S.
Figure 2. Effect of C4BP? on binding and degradation of protein S by THP-1 macrophages. A, Binding of radiolabeled wt-protein S was determined in the absence or presence of a 5-fold molar excess of C4BP or C4BP? as described for Figure 1. B, Degradation of radiolabeled wt-protein S (0.13 μmol/L) was monitored in the absence (?) or presence of a 5-fold molar excess of C4BP () or C4BP? () as described for Figure 1. Plotted is the percentage binding or degradation in the presence of C4BP or C4BP? relative to their absence. Data represent mean±SEM of 3 experiments performed in duplicate.
Biphasic Disappearance of Protein S From Plasma
In view of the in vitro effect of C4BP? on the cellular uptake of protein S, we considered the possibility that C4BP? exerts a similar effect in vivo. To test this possibility, we addressed the survival of protein S in a mouse-model. Murine C4BP lacks the ?-chain, rendering this protein unable to bind protein S. This allowed us to use this model to directly examine the effect of human C4BP? on the survival of protein S. First, we evaluated the disappearance of protein S from plasma by administering radiolabeled recombinant wt-protein S (5 μg per mouse) to normal C57BL/6J mice by tail vein injection. Mice were then bled at indicated time points, and plasma was analyzed for residual activity. The recovery (3 minutes after injection) was 61.1±0.5% (n=3 mice; Table). Graphic representation showed that the protein is cleared in a biphasic manner, with a rapid initial phase and a slow secondary phase (Figure 3A). Data analysis allowed calculation of the apparent half-lives, which were 3.6±2.9 minutes and 1.4±0.5 hours for the rapid and slow phase, respectively. The MRT was 1.9±0.6 hours (Table).
Pharmacokinetic Parameters Describing the Clearance of Recombinant wt-Protein S and Protein S-S460P in the Absence or Presence of C4BP?
Figure 3. Disappearance of wt-protein S or protein S-S460P from plasma in the presence or absence of C4BP?. C57BL/6J mice were injected with radiolabeled wt-protein S or protein S-S460P (5 μg per mouse; 1.8x106 cpm/μg protein) in the absence (A and C) or presence (B and D) of a 2-fold molar excess of C4BP? (preincubated 45 minutes before injection). Blood samples were taken at indicated time intervals, and the amount of residual radioactivity in plasma was measured. Plotted is the amount of residual radioactivity relative to the amount injected vs time after injection. The solid line is that obtained after fitting the experimental data to a biexponential equation,31 the parameters of which are summarized in the Table. The dotted lines in B through D represent the disappearance of wt-protein S in the absence of C4BP? and are indicated for comparison. B, Inset, The plasma survival of protein S in the presence of C4BP? (?) compared with the survival of C4BP? itself (). Data represent mean±SD of 3 mice for each time point for all data sets.
Protein S Is Primarily Targeted to the Liver
In the same experiments, we also tested the biodistribution of protein S by collecting the various organs and measuring their content of radioactivity. Minor amounts of radioactivity (<3% of the amount injected) were found in heart, lungs, spleen, kidneys, and left upper leg during the full time course of the experiment (data not shown). In the stomach, similar low amounts were found shortly after injection, whereas some radioactivity (up to 7%) started to accumulate in this organ 1 hour after injection. With regard to the liver, 17.9±0.4% of the amount injected could be detected within 3 minutes after injection. This amount remained stable until 15 minutes (19±3%), followed by a sharp decline between 0.5 and 4 hours. Thus, intravenously administered protein S is mainly directed to the liver.
C4BP? Prolongs the In Vivo Survival of Protein S
To test the effect of C4BP? on the survival of protein S, both proteins were first preincubated in a 2:1 ratio before injection (see Methods). On intravenous application, the initial yield of protein S was significantly higher in the presence of C4BP? than in its absence (72% versus 61%; P=0.048; Table). Furthermore, protein S disappeared slower from plasma in the presence of C4BP?, as illustrated by increased half-lives of both phases of the clearance process (Figure 3B; Table). Indeed, the MRT was increased 4-fold from 1.9 to 7.8 hours (P=0.0001). Corresponding with the slower clearance, lower amounts of radioactivity were found in the liver at 3 and 15 minutes after injection (14.4±0.9 versus 17.9±0.4% [P=0.0035] and 11.7±3.9 versus 19.3±2.7% [P=0.049] for the presence and absence of C4BP?). In addition, a more gradual decline in the amount of radioactivity in the liver was observed in the presence of C4BP?. Thus, C4BP? delays the liver-mediated clearance of protein S in our experimental model.
Increased Clearance of Protein S-S460P
To investigate whether clearance of protein S can be influenced by changes in amino acid sequence, we studied the in vivo survival of protein S-S460P. This mutant comprises a Ser460 to Pro replacement and is associated with reduced antigen levels. Purified recombinant protein S-S460P was 125I-labeled and injected intravenously in the tail vein of normal C57BL/6J mice. At indicated time points, the mice were bled, and plasma was analyzed for the presence of residual radioactivity. The recovery after 3 minutes was 59.0±2.3% (n=3 mice; Table). As for wt-protein S, protein S-S460P disappeared in a biphasic manner, although remarkably more rapidly than wt-protein S (Figure 3C). Indeed, t1/2 and t1/2? are reduced 2- to 3-fold for protein S-S460P compared with wt-protein S (Table). Also, the calculated MRT was 3-fold shorter for the mutant compared with wt-protein S (Table). The increased clearance was associated with a faster appearance of the mutant in the liver at 3 minutes and 15 minutes after injection (22.3±1.3 versus 17.9±0.4% [P=0.005] and 25.6±0.2 versus 19.3±2.7% [P=0.016] for protein S-S460P and wt-protein S, respectively).
C4BP? Partially Compensates the Increased Clearance Rate of Protein S-S460P
Because C4BP? reduces the clearance rate of wt-protein S, it was of further interest to investigate the effect of C4BP? on the in vivo survival of protein S-S460P. In preliminary experiments, the interaction between C4BP? and this particular mutant was examined using surface plasmon resonance analysis. As expected, these experiments revealed that the mutant was similar to wt-protein S with regard to C4BP binding in terms of association and dissociation rate constants.24 Protein S-S460P was radiolabeled and subsequently incubated with a 2-fold molar excess of human C4BP? before intravenous injection. The initial recovery 3 minutes after injection was 61.2±6.4%, which is similar to the recovery of protein S-S460P in the absence of C4BP (Table). However, the protein S-S460P/C4BP? complex disappeared less rapidly from the circulation compared with free protein S-S460P (Figure 3D). For instance, at the 15-minute time point, 44.5±1.4% protein S-S460P/C4BP? complex was present compared with 34.9±3.8% of protein S administered in the absence of C4BP? (P=0.015). Calculation of the kinetic parameters revealed that t1/2, t1/2?, and MRT were all increased in the presence of C4BP? (Table). Thus, the increased clearance rate of protein S-S460P is delayed by the presence of C4BP?.
Discussion
Distortion of the balance between procoagulant and anticoagulant pathways can lead to bleeding or thrombotic complications. Such misbalance is often the result of concentrations of components that are too high or too low. For instance, low levels of protein S are associated with an increased thrombotic risk13,14 Low protein S levels may originate from impaired synthesis/secretion, increased clearance, or combinations thereof. Examples of mutations leading to impaired secretion have been described.17–19 However, little is known about the protein S clearance mechanism and how this process is affected by variations in the protein S amino acid sequence or its carrier protein C4BP?.
The possibility that C4BP? affects protein S clearance became apparent in our in vitro experiments regarding the cellular uptake of protein S. We observed that human THP-1 macrophages are able to bind and transport protein S to its intracellular degradation pathways (Figure 1). THP-1 macrophages comprise 4.3x105 binding sites per cell, a number that is in between the amount of binding sites reported previously for bovine aortic endothelial cells (8.5x104 sites per cell32) and human umbilical vein endothelial cells (8.0x105 sites/cell28). Binding and intracellular delivery was markedly reduced in the presence of C4BP? (Figure 2). This effect is in line with the observation by Kask et al that C4BP? interferes with the protein S–dependent uptake of apoptotic cells by macrophages29 and suggests that the C4BP?-binding region (ie, the carboxyl-terminal sex hormone binding globulin-like domain) contributes to the interaction with macrophages. Indeed, preliminary experiments revealed that a recombinant fragment of this region indeed binds to macrophages, albeit with a 5-fold lower affinity (data not shown). However, the notion that C4BP? inhibits cellular uptake only partially points to the possibility that other parts also not covered by C4BP?, such as the amino terminal Gla domain, can mediate cellular binding. Alternatively, the residual uptake may also be facilitated by low-density lipoprotein receptor–related protein, which is present in THP-1 macrophages. We reported previously that this receptor contributes to C4BP? catabolism in vitro and in vivo.33
As to the behavior of protein S in vivo, we found that protein S disappeared in a biphasic manner, with an overall MRT of 1.9 hours (Figure 3; Table). The relatively rapid disappearance from plasma went together with a rapid appearance of radioactivity in the liver (data not shown), suggesting that protein S is mainly directed to this organ. Little contribution of other organs to the uptake of protein S was observed, although in time, accumulation of radioactivity in the stomach was observed. Similar delayed accumulation of radioactivity in the stomach has been described for other proteins, such as von Willebrand factor (vWF) and insulin growth factor binding protein-3.31,34 In case of the latter, this accumulated radioactivity represents small, degraded peptides that are excreted from the liver and reabsorbed by the stomach. Given the time course by which radioactivity appears in liver and stomach, respectively, it seems conceivable that a similar mechanism applies to protein S.
The murine model was particularly useful to study the effect of C4BP? on protein S clearance for 2 reasons: (1) murine C4BP lacks the ?-chain, which renders it unable to bind protein S;35 and (2) murine protein S displays low affinity for human C4BP?, which is mainly because of the presence of a Glu instead of a Gln at position 42736,37. Thus, the complex between human protein S and human C4BP? is affected by their murine counterparts only to a minor extent, if any. The presence of C4BP? resulted in a 4-fold prolonged survival of protein S in the circulation and a delayed uptake in the liver (Figure 3; Table). This demonstrates that C4BP? displays a similar protective effect in the in vivo mouse model compared with the in vitro system using human cells. Of interest, the disappearance of the C4BP?/protein S complex mimicked that of C4BP? alone (Figure 3B, inset), suggesting that clearance of the complex is merely mediated by C4BP?. It should be noted that the half-life of the protein S/C4BP? complex in the mouse is shorter than the half-life of protein S in humans (8 hours compared with 40 hours based on the disappearance of the protein on warfarin treatment). Similar species-specific differences have been reported for numerous other proteins, including factor IX and vWF. Whereas the half-life of vWF in humans is 14 hours, its survival in mice is reduced 4- to 5-fold.31 These differences in clearance rate are probably a reflection of the basal metabolic rate, which is increased in mice.
The protective effect of C4BP? seems to resemble the effect of vWF on the in vivo survival of coagulation factor VIII because the absence of vWF results in increased clearance of factor VIII.38 It is tempting to extrapolate our data obtained with in vitro experiments using human THP-1 cells, and the murine in vivo experiments to the human situation in that C4BP? interferes with protein S clearance. Unfortunately, no direct evidence is available to support this view. It is remarkable, though, that in studies related to the expression of naturally occurring protein S mutants, the levels of total protein S in the corresponding patients were higher than what would be expected from the in vitro expression studies.17,18 More specifically, levels of free protein S were severely reduced, whereas levels of C4BP?-bound protein S were only mildly reduced in these patients.17,18 As such, these data are in support of the possibility that C4BP?-bound protein S displays a longer survival in the circulation than free protein S.
One protein S variant that is consistently associated with reduced antigen levels is protein S-Heerlen, which occurs in 0.5% of the population.20 Although, in particular, the levels of free protein S-Heerlen are decreased,18,22 total protein S levels are also slightly but significantly reduced.18 The underlying mechanism of this phenomenon has remained unclear because expression studies using recombinant variants demonstrated that protein S-S460P is synthesized and secreted as efficiently as wt-protein S.21 Thus, other mechanisms should be responsible for the low levels of protein S-Heerlen. We therefore applied our experimental mouse model to investigate the in vivo survival of free protein S-S460P. Our experiments using purified recombinant protein S-S460P clearly reveal that this mutant disappears from the circulation more rapidly than wt-protein S (Figure 3), resulting in a 3-fold shorter MRT (Table). This demonstrates that the Ser460 to Pro substitution, per se, is associated with increased clearance and explains why levels of free protein S are reduced in individuals carrying the Heerlen polymorphism. Hence, increased clearance is a mechanism that contributes to the pathology of protein S deficiency.
The reason why protein S-S460P is cleared more rapidly remains to be determined, but it may be associated with mutation-induced conformational changes. Alternatively, it may be related to the absence of the carbohydrate chain at position Asn458. Irrespective of the exact mechanism, it should be noted that the increased clearance rate could be compensated partially in the presence of C4BP? (Figure 3). C4BP? interacts with several regions in the sex-hormone–binding-globulin-like domain, including the region 453 to 460, which encompasses the Asn458 glycosylation site and the Ser460Pro replacement.39 This provides C4BP? the potential to mask mutation-induced alterations, allowing C4BP? to counteract the increased clearance. This would explain the near-normal levels of C4BP?/protein S-Heerlen and offers a rationale for the observed Type III deficiency phenomenon.
Acknowledgments
This study was supported by grants from the Dutch Thrombosis Foundation (grant 2002.2 to P.J.L.), the Netherlands Organization for Scientific Research (NWO-VIDI 917-36-372 to T.M.H.), an INSERM-ZonMW/NWO exchange grant (910-48-603 to C.V.D. and P.J.L.) and an INSERM AVENIR program grant (to C.V.D.).
References
Zoller B, Garcia de Frutos P, Dahlback B. Evaluation of the relationship between protein S and C4b-binding protein isoforms in hereditary protein S deficiency demonstrating type I and type III deficiencies to be phenotypic variants of the same genetic disease. Blood. 1995; 85: 3524–3531.
Dahlback B, Stenflo J. High molecular weight complex in human plasma between vitamin K-dependent protein S and complement component C4b-binding protein. Proc Natl Acad Sci U S A. 1981; 78: 2512–2516.
Hillarp A, Dahlback B. Novel subunit in C4b-binding protein required for protein S binding. J Biol Chem. 1988; 263: 12759–12764.
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: 87–91.
Webb JH, Blom AM, Dahlback B. Vitamin K-dependent protein S localizing complement regulator C4b-binding protein to the surface of apoptotic cells. J Immunol. 2002; 169: 2580–2586.
Liu D, Guo H, Griffin JH, Fernandez JA, Zlokovic BV. Protein S confers neuronal protection during ischemic/hypoxic injury in mice. Circulation. 2003; 107: 1791–1796.
Dahlback B. Protein S and C4b-binding protein: components involved in the regulation of the protein C anticoagulant system. Thromb Haemost. 1991; 66: 49–61.
Esmon CT. The protein C pathway. Chest. 2003; 124: 26S–32S.
Walker FJ. Regulation of activated protein C by protein S. The role of phospholipid in factor Va inactivation. J Biol Chem. 1981; 256: 11128–11131.
Koedam JA, Meijers JC, Sixma JJ, Bouma BN. Inactivation of human factor VIII by activated protein C. Cofactor activity of protein S and protective effect of von Willebrand factor. J Clin Invest. 1988; 82: 1236–1243.
Heeb MJ, Mesters RM, Tans G, Rosing J, Griffin JH. Binding of protein S to factor Va associated with inhibition of prothrombinase that is independent of activated protein C. J Biol Chem. 1993; 268: 2872–2877.
Hackeng TM, van ’t Veer C, Meijers JC, Bouma BN. Human protein S inhibits prothrombinase complex activity on endothelial cells and platelets via direct interactions with factors Va and Xa. J Biol Chem. 1994; 269: 21051–21058.
Comp PC, Nixon RR, Cooper MR, Esmon CT. Familial protein S deficiency is associated with recurrent thrombosis. J Clin Invest. 1984; 74: 2082–2088.
Broekmans AW, Bertina RM, Reinalda-Poot J, Engesser L, Muller HP, Leeuw JA, Michiels JJ, Brommer EJ, Briet E. Hereditary protein S deficiency and venous thrombo-embolism. A study in three Dutch families. Thromb Haemost. 1985; 53: 273–277.
Gandrille S, Borgel D, Sala N, Espinosa-Parrilla Y, Simmonds R, Rezende S, Lind B, Mannhalter C, Pabinger I, Reitsma PH, Formstone C, Cooper DN, Saito H, Suzuki K, Bernardi F, Aiach M. Protein S deficiency: a database of mutations—summary of the first update. Thromb Haemost. 2000; 84: 918.
Simmonds RE, Zoller B, Ireland H, Thompson E, Garcia de Frutos P, Dahlback B, Lane DA. Genetic and phenotypic analysis of a large (122-member) protein S-deficient kindred provides an explanation for the familial coexistence of type I and type III plasma phenotypes. Blood. 1997; 89: 4364–4370.
Rezende SM, Lane DA, Zoller B, Mille-Baker B, Laffan M, Dalhback B, Simmonds RE. Genetic and phenotypic variability between families with hereditary protein S deficiency. Thromb Haemost. 2002; 87: 258–265.
Espinosa-Parrilla Y, Yamazaki T, Sala N, Dahlback B, Garcia de Frutos P. Protein S secretion differences of missense mutants account for phenotypic heterogeneity. Blood. 2000; 95: 173–179.
Tsuda H, Urata M, Tsuda T, Wakiyama M, Iida H, Nakahara M, Kinoshita S, Hamasaki N. Four missense mutations identified in the protein S gene of thrombosis patients with protein S deficiency: effects on secretion and anticoagulant activity of protein S. Thromb Res. 2002; 105: 233–239.
Bertina RM, Ploos van Amstel HK, van Wijngaarden A, Coenen J, Leemhuis MP, Deutz-Terlouw PP, van der Linden IK, Reitsma PH. Heerlen polymorphism of protein S, an immunologic polymorphism due to dimorphism of residue 460. Blood. 1990; 76: 538–548.
Giri TK, Yamazaki T, Sala N, Dahlback B, Garcia de Frutos P. Deficient APC-cofactor activity of protein S Heerlen in degradation of factor Va Leiden: a possible mechanism of synergism between thrombophilic risk factors. Blood. 2000; 96: 523–531.
Schwarz HP, Heeb MJ, Lottenberg R, Roberts H, Griffin JH. Familial protein S deficiency with a variant protein S molecule in plasma and platelets. Blood. 1989; 74: 213–221.
Duchemin J, Gandrille S, Borgel D, Feurgard P, Alhenc-Gelas M, Matheron C, Dreyfus M, Dupuy E, Juhan-Vague I, Aiach M. The Ser 460 to Pro substitution of the protein S alpha PROS1. gene is a frequent mutation associated with free protein S (type IIa) deficiency. Blood. 1995; 86: 3436–3443.
Koenen RR, Gomes L, Tans G, Rosing J, Hackeng TM. The Ser460Pro mutation in recombinant protein S Heerlen does not affect its APC-cofactor and APC-independent anticoagulant activities. Thromb Haemost. 2004; 91: 1105–1114.
Morboeuf O, Borgel D, Aiach M, Kaabache T, Gandrille S, Gaussem P. Expression and characterization of recombinant protein S with the Ser 460 Pro mutation. Thromb Res. 2000; 100: 81–88.
van Wijnen M, Stam JG, Chang GT, Meijers JC, Reitsma PH, Bertina RM, Bouma BN. Characterization of mini-protein S, a recombinant variant of protein S that lacks the sex hormone binding globulin-like domain. Biochem J. 1998; 330: 389–396.
van de Poel RH, Meijers JC, Rosing J, Tans G, Bouma BN. C4b-binding protein protects coagulation factor Va from inactivation by activated protein C. Biochemistry. 2000; 39: 14543–14548.
Hackeng TM, Hessing M, van ’t Veer C, Meijer-Huizinga F, Meijers JC, de Groot PG, van Mourik JA, Bouma BN. Protein S binding to human endothelial cells is required for expression of cofactor activity for activated protein C. J Biol Chem. 1993; 268: 3993–4000.
Kask L, Trouw LA, Dahlback B, Blom AM. The C4b-binding protein-protein S complex inhibits the phagocytosis of apoptotic cells. J Biol Chem. 2004; 279: 23869–23873.
Denis C, Methia N, Frenette PS, Rayburn H, Ullman-Cullere M, Hynes RO, Wagner DD. A mouse model of severe von Willebrand disease: defects in hemostasis and thrombosis. Proc Natl Acad Sci U S A. 1998; 95: 9524–9529.
Lenting PJ, Westein E, Terraube V, Ribba AS, Huizinga EG, Meyer D, de Groot PG, Denis CV. An experimental model to study the in vivo survival of von Willebrand factor. Basic aspects and application to the R1205H mutation. J Biol Chem. 2004; 279: 12102–12109.
Stern DM, Nawroth PP, Harris K, Esmon CT. Cultured bovine aortic endothelial cells promote activated protein C-protein S-mediated inactivation of factor Va. J Biol Chem. 1986; 261: 713–718.
Westein E, Denis CV, Bouma BN, Lenting PJ. The alpha-chains of C4b-binding protein mediate complex formation with low density lipoprotein receptor-related protein. J Biol Chem. 2002; 277: 2511–2516.
Arany E, Zabel P, Hill DJ. Rapid clearance of human insulin-like growth factor binding protein-3 from the rat circulation and cellular localization in liver, kidney and stomach. Growth Regul. 1996; 6: 32–41.
Rodriguez de Cordoba S, Perez-Blas M, Ramos-Ruiz R, Sanchez-Corral P, Pardo-Manuel de Villena F, Rey-Campos J. The gene coding for the beta-chain of C4b-binding protein (C4BPB) has become a pseudogene in the mouse. Genomics. 1994; 21: 501–509.
Chu MD, Sun J, Bird P. Cloning and sequencing of a cDNA encoding the murine vitamin K-dependent protein S. Biochim Biophys Acta. 1994; 1217: 325–328.
Fernandez JA, Griffin JH, Chang GT, Stam J, Reitsma PH, Bertina RM, Bouma BN. Involvement of amino acid residues 423–429 of human protein S in binding to C4b-binding protein. Blood Cells Mol Dis. 1998; 24: 101–112.
Lenting PJ, van Mourik JA, Mertens K. The life cycle of coagulation factor VIII in view of its structure and function. Blood. 1998; 92: 3983–3996.
Giri TK, Linse S, Garcia de Frutos P, Yamazaki T, Villoutreix BO, Dahlback B. Structural requirements of anticoagulant protein S for its binding to the complement regulator C4b-binding protein. J Biol Chem. 2002; 277: 15099–15106.(Cécile V. Denis; Sarah J.)
Correspondence to Dr P.J. Lenting, Laboratory for Thrombosis and Haemostasis, Department of Haematology, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX Utrecht, the Netherlands. E-mail p.j.lenting@lab.azu.nl
Abstract
Objective— To explore the effect of the Heerlen polymorphism and C4b-binding protein (C4BP) on protein S catabolism in vitro and in vivo.
Methods and Results— Radiolabeled protein S was efficiently bound and intracellularly degraded by THP-1 macrophages, and both processes were strongly reduced in the presence of the protein S-carrier protein C4BP. To test whether C4BP displays a similar protective effect in vivo, survival experiments were performed in mice. In the absence of C4BP, radiolabeled human protein S disappeared in a biphasic manner (mean residence time [MRT] 2 hours). However, the presence of C4BP resulted in a 4-fold prolonged survival of protein S (MRT 8 hours; P<0.0001). We also applied this experimental model to recombinant protein S-Heerlen, a naturally occurring variant that contains a Ser460Pro substitution. These clearance experiments revealed a strongly decreased survival of recombinant protein S-S460P (MRT 0.6 hours; P=0.021), which could be compensated partially by C4BP (MRT 1.4 hours; P=0.012 compared with protein S-S460P).
Conclusion— Protein S-S460P has a reduced survival in vivo, which may explain the low levels of free protein S in individuals carrying this polymorphism. Furthermore, C4BP prevents premature clearance of protein S and uses this ability to compensate the increased clearance of protein S-S460P.
We established that binding and degradation of protein S by macrophages was partially inhibited by C4BP. C4BP mimicked this effect in vivo by prolonging the half-life of protein S from 2 to 4 hours in a mouse model. The presence of the Heerlen polymorphism reduced survival of protein S 3-fold to 0.6 hours.
Key Words: protein S ? C4b-binding ? clearance
Introduction
Protein S circulates in plasma at a concentration of 0.3 μmol/L, although this value may vary between individuals.1 Only 30% to 40% of protein S circulates in a free form, whereas the remaining part is in complex with C4b-binding protein (C4BP).2 Various isoforms of C4BP exist, and only those that contain the ?-chain (ie, C4BP?) bind protein S.3 Because the C4BP? plasma concentration is lower than that of protein S, the molar excess of protein S over C4BP? determines the amount of free protein S.
The function of protein S relates to various physiological processes, including the clearance of apoptotic cells and neuronal protection during ischemic/hypoxic injury.4–6 However, protein S is originally known from its role in the anticoagulant process,7,8 in which it functions as a cofactor for activated protein C (APC).9,10 In addition, protein S may downregulate the coagulation pathway independently of APC.11,12 The notion that deficiencies in protein S are associated with an increased thrombotic tendency13,14 demonstrates that appropriate plasma levels of protein S are needed to maintain the hemostatic balance.
Two categories of protein S deficiency can be distinguished:15 (1) qualitative defects (type II deficiency), and (2) quantitative defects (type I or III deficiency). Type I deficiency refers to a parallel reduction of total and free protein S. Type III deficiency is characterized by low levels of free protein S but normal levels of total protein S. However, both types of quantitative deficiencies often occur within the same family,1 and more detailed analysis revealed that increased levels of total protein S as seen in type III coincide with an age-dependent increase in protein S levels.16
Quantitative deficiencies of protein S may obviously originate from null alleles, but also, missense mutations can lead to reduced protein S levels.15 Some of these missense mutations result in impaired secretion of the protein.17–19 However, not all amino acid replacements associated with low levels of protein S associate with abnormal secretion. An example hereof is protein S-Heerlen, which contains a Ser to Pro replacement at position 460.20 This substitution results in loss of N-linked glycosylation at position Asn458. Protein S-Heerlen is found in 0.5% of the population20 and is associated with reduced levels of free protein S in heterozygous or homozygous individuals.21,22 Expression studies demonstrated that recombinant protein S-S460P is normally secreted,21 indicating that other mechanisms should be involved that explain the quantitative deficiency of protein S-Heerlen. Possible mechanisms that have been proposed include increased affinity for C4BP? or multivalency of C4BP? for this variant.23 However, protein S-S460P and protein S were similar in their interaction with C4BP? in various other reports.20,21,24,25
The notion that protein S-S460P is similar to wild-type (wt)–protein S in terms of secretion efficiency and binding to C4BP? prompted us to investigate the previously proposed hypothesis that protein S-Heerlen is cleared from the circulation more rapidly than normal protein S (Dr R.M. Bertina, XVI ISTH Congress, Florence, Italy, 1997). Therefore, a mouse model was used to investigate the in vivo survival of wt-protein S and recombinant protein S-S460P in the presence and absence of C4BP?.
Methods
Mice
The mice used in the present study were on a C57BL/6J background and were used between 8 and 12 weeks of age. Housing and experiments were done as recommended by French regulations and the experimental guidelines of the European Community.
Reagents and Proteins
Chloroquine was from Serva. Methyl-?-cyclodextrin (M?CD) was from Sigma-Aldrich. C4BP? was prepared from plasma via barium–citrate absorption and immunoaffinity chromatography to homogeneity (see supplemental data, available online at http://atvb.ahajournals.org). No protein S antigen could be detected in these C4BP? preparations in an immunosorbent assay, whereas the ability to bind protein S was fully retained. Purification of recombinant C4BP, recombinant wt-protein S, and recombinant protein S-S460P has been described previously24,26,27 (see also online supplemental data).
Radiolabeling of Protein S
Purified wt-protein S or protein S-S460P was labeled with Na125I (Amersham Biosciences Inc) using the IODO-GEN method28 (online supplemental data). Final preparations had specific activities varying between 1.4x106 and 2.2x106 cpm/μg protein (n=2 to 10), whereas free iodine was routinely <6%. This method to radiolabel protein S does not affect C4BP? binding capacity (data not shown) and APC cofactor activity.28
Cellular Binding, Internalization, and Degradation Experiments
Cellular binding and degradation experiments were performed using human THP-1 cells (ATCC TIB-202). These cells have been used previously to study the role of protein S in the engulfment of apoptotic cells by macrophages.29 Binding, internalization, and degradation of protein S by THP-1 macrophages were investigated using standard cell biological methods, which are described in detail in the online supplemental data.
Clearance and Biodistribution of Protein S in Mice
Normal C57BL/6J mice were injected intravenously with radiolabeled protein S or protein S-S460P (5 μg per mouse, corresponding to 9x106 cpm per mouse) in the presence or absence of a 2-fold molar excess C4BP?, diluted in PBS containing 3% bovine albumin. To facilitate complex formation, protein S (0.3 μmol/L) and C4BP? (0.6 μmol/L) were incubated for 45 minutes before injection. Because the concentration of both proteins is >300-fold above the KD value (<1 nmol/L),24 these conditions result in >95% of protein S being in complex with C4BP?. Furthermore, concentrations of C4BP? were chosen to be 4-fold below those that saturate the clearance system. At different time points after injection (3, 15, and 30 minutes, and 1, 2, and 4 hours), mice were anesthetized with tribromoethanol (0.15 mL/10 g of body weight), and blood was collected by eye bleed. Subsequently, plasma was prepared as described30 and analyzed for the presence of radioactivity. After bleeding, mice were euthanized, and various organs (heart, lungs, liver, spleen, stomach, kidneys, and left upper leg) were collected, which were analyzed for the amount of radioactivity. Three mice were used for each time point.
Data Analysis and Statistics
Data of in vivo survival experiments were fitted to a biexponential equation as described previously31 to allow calculation of the parameters mean residence time (MRT), t1/2, and t1/2?. Statistical analyses were performed by Student’s unpaired t test using the GraphPad InStat program (GraphPad Instat version 3.0 for Windows; GraphPad Software).
Results
Binding of Protein S to THP-1 Macrophages
Recently, protein S has been reported to mediate the engulfment of apoptotic cells by macrophages.4 Here, we tested whether human THP-1 macrophages are able to bind and deliver protein S to the intracellular degradation pathway. Binding was assessed by incubating radiolabeled protein S (0 to 0.52 μmol/L) with THP-1 macrophages for 2 hours at 4°C. Protein S bound to THP-1 macrophages in a saturable and dose-dependent manner, and half-maximal binding was at a concentration of 0.11±0.04 μmol/L (Figure 1A). The number of binding sites was found to be 4.3±0.5x105 per cell. Specificity of binding was addressed in 2 types of control experiments. First, labeled and unlabeled protein S were applied to the cells at different ratios while maintaining total protein S concentration constant. A linear decrease in binding of radiolabeled protein S was observed (slope –1.0±0.1; Y-intercept 105±3), indicating that labeled and unlabeled protein S bind THP-1 macrophages similarly. Second, binding of radiolabeled protein S was reduced >90% by a 100-fold excess of unlabeled protein S (Figure 1A, inset).
Figure 1. Binding, internalization, and degradation of wt-protein S by THP-1 macrophages. A, THP-1 macrophages were incubated for 2 hours at 4°C with radiolabeled wt-protein S (0 to 0.5 μmol/L). Bound wt-protein S was determined as described in Methods and corrected for nonspecific binding in the absence of cells (routinely <15%). A, Inset, Binding of radiolabeled wt-protein S (0.13 μmol/L) was assessed in the absence or presence of a 100-fold molar excess of unlabeled wt-protein S. B, THP-1 macrophages were incubated with radiolabeled wt-protein S for 1 hour at 4°C. After washing, cells were placed at 37°C to initiate internalization. At indicated time intervals, the amount of surface-bound (?) and internalized () protein S was determined as described in Methods. Plotted is the amount of radioactivity relative to the amount of surface-bound radioactivity at t=0 (100%). Data are corrected for the amount of radioactivity present in the absence of cells (<15%). C, THP-1 macrophages were incubated with radiolabeled wt-Protein S (0.13 μmol/L) either in the absence (?) or presence of 10 mmol/L M?CD () or 0.1 mmol/L chloroquine () for 10 minutes at 37°C. After washing, incubation was continued at 37°C for indicated time intervals. The amount of degraded wt-protein S was determined as described in Methods and corrected for nonspecific degradation in the absence of cells (routinely <10%). Data represent mean±SEM of 2 to 10 experiments performed in duplicate.
Internalization and Degradation of Protein S
Subsequently, we tested whether binding was followed by transport of protein S to the intracellular degradation pathway. Radiolabeled protein S (0.13 μmol/L) was incubated with macrophages (1 hour at 4°C) to allow surface binding. After washing, cells were put at 37°C to initiate internalization. Cell surface–bound protein S had disappeared by >90% within 40 minutes (Figure 1B). Simultaneously, the amount of internalized protein S had increased and reached a maximum after 40 minutes, which was followed by a gradual decline. Degradation was examined by monitoring the formation of trichloroacetic acid–soluble fragments on incubation with radiolabeled protein S (0.13 μmol/L). In time, increased levels of degradation products were observed (Figure 1C). Degradation was markedly reduced in the presence of M?CD (interferes with caveolae- and clathrin-dependent endocytosis) or the lysosomal inhibitor chloroquine (Figure 1C). Together, these data demonstrate that human macrophages can bind and transport protein S to their intracellular degradation pathway.
C4BP? Partially Inhibits Macrophage-Dependent Protein S Degradation
Because part of protein S circulates in complex with C4BP?, cellular uptake of protein S was tested in the presence of its carrier protein. When assessed for binding to THP-1 macrophages, a 5-fold excess of C4BP? reduced binding by 60%, whereas the nonbinding counterpart C4BP was unable to reduce binding (Figure 2A). A similar effect was observed when protein S degradation was monitored. Whereas C4BP did not inhibit degradation, C4BP? reduced degradation 4- to 5-fold (Figure 2B). These levels were reduced to a similar extent as observed for the degradation inhibitors M?CD and chloroquine. Apparently, complex formation with C4BP? reduces the cellular uptake of protein S.
Figure 2. Effect of C4BP? on binding and degradation of protein S by THP-1 macrophages. A, Binding of radiolabeled wt-protein S was determined in the absence or presence of a 5-fold molar excess of C4BP or C4BP? as described for Figure 1. B, Degradation of radiolabeled wt-protein S (0.13 μmol/L) was monitored in the absence (?) or presence of a 5-fold molar excess of C4BP () or C4BP? () as described for Figure 1. Plotted is the percentage binding or degradation in the presence of C4BP or C4BP? relative to their absence. Data represent mean±SEM of 3 experiments performed in duplicate.
Biphasic Disappearance of Protein S From Plasma
In view of the in vitro effect of C4BP? on the cellular uptake of protein S, we considered the possibility that C4BP? exerts a similar effect in vivo. To test this possibility, we addressed the survival of protein S in a mouse-model. Murine C4BP lacks the ?-chain, rendering this protein unable to bind protein S. This allowed us to use this model to directly examine the effect of human C4BP? on the survival of protein S. First, we evaluated the disappearance of protein S from plasma by administering radiolabeled recombinant wt-protein S (5 μg per mouse) to normal C57BL/6J mice by tail vein injection. Mice were then bled at indicated time points, and plasma was analyzed for residual activity. The recovery (3 minutes after injection) was 61.1±0.5% (n=3 mice; Table). Graphic representation showed that the protein is cleared in a biphasic manner, with a rapid initial phase and a slow secondary phase (Figure 3A). Data analysis allowed calculation of the apparent half-lives, which were 3.6±2.9 minutes and 1.4±0.5 hours for the rapid and slow phase, respectively. The MRT was 1.9±0.6 hours (Table).
Pharmacokinetic Parameters Describing the Clearance of Recombinant wt-Protein S and Protein S-S460P in the Absence or Presence of C4BP?
Figure 3. Disappearance of wt-protein S or protein S-S460P from plasma in the presence or absence of C4BP?. C57BL/6J mice were injected with radiolabeled wt-protein S or protein S-S460P (5 μg per mouse; 1.8x106 cpm/μg protein) in the absence (A and C) or presence (B and D) of a 2-fold molar excess of C4BP? (preincubated 45 minutes before injection). Blood samples were taken at indicated time intervals, and the amount of residual radioactivity in plasma was measured. Plotted is the amount of residual radioactivity relative to the amount injected vs time after injection. The solid line is that obtained after fitting the experimental data to a biexponential equation,31 the parameters of which are summarized in the Table. The dotted lines in B through D represent the disappearance of wt-protein S in the absence of C4BP? and are indicated for comparison. B, Inset, The plasma survival of protein S in the presence of C4BP? (?) compared with the survival of C4BP? itself (). Data represent mean±SD of 3 mice for each time point for all data sets.
Protein S Is Primarily Targeted to the Liver
In the same experiments, we also tested the biodistribution of protein S by collecting the various organs and measuring their content of radioactivity. Minor amounts of radioactivity (<3% of the amount injected) were found in heart, lungs, spleen, kidneys, and left upper leg during the full time course of the experiment (data not shown). In the stomach, similar low amounts were found shortly after injection, whereas some radioactivity (up to 7%) started to accumulate in this organ 1 hour after injection. With regard to the liver, 17.9±0.4% of the amount injected could be detected within 3 minutes after injection. This amount remained stable until 15 minutes (19±3%), followed by a sharp decline between 0.5 and 4 hours. Thus, intravenously administered protein S is mainly directed to the liver.
C4BP? Prolongs the In Vivo Survival of Protein S
To test the effect of C4BP? on the survival of protein S, both proteins were first preincubated in a 2:1 ratio before injection (see Methods). On intravenous application, the initial yield of protein S was significantly higher in the presence of C4BP? than in its absence (72% versus 61%; P=0.048; Table). Furthermore, protein S disappeared slower from plasma in the presence of C4BP?, as illustrated by increased half-lives of both phases of the clearance process (Figure 3B; Table). Indeed, the MRT was increased 4-fold from 1.9 to 7.8 hours (P=0.0001). Corresponding with the slower clearance, lower amounts of radioactivity were found in the liver at 3 and 15 minutes after injection (14.4±0.9 versus 17.9±0.4% [P=0.0035] and 11.7±3.9 versus 19.3±2.7% [P=0.049] for the presence and absence of C4BP?). In addition, a more gradual decline in the amount of radioactivity in the liver was observed in the presence of C4BP?. Thus, C4BP? delays the liver-mediated clearance of protein S in our experimental model.
Increased Clearance of Protein S-S460P
To investigate whether clearance of protein S can be influenced by changes in amino acid sequence, we studied the in vivo survival of protein S-S460P. This mutant comprises a Ser460 to Pro replacement and is associated with reduced antigen levels. Purified recombinant protein S-S460P was 125I-labeled and injected intravenously in the tail vein of normal C57BL/6J mice. At indicated time points, the mice were bled, and plasma was analyzed for the presence of residual radioactivity. The recovery after 3 minutes was 59.0±2.3% (n=3 mice; Table). As for wt-protein S, protein S-S460P disappeared in a biphasic manner, although remarkably more rapidly than wt-protein S (Figure 3C). Indeed, t1/2 and t1/2? are reduced 2- to 3-fold for protein S-S460P compared with wt-protein S (Table). Also, the calculated MRT was 3-fold shorter for the mutant compared with wt-protein S (Table). The increased clearance was associated with a faster appearance of the mutant in the liver at 3 minutes and 15 minutes after injection (22.3±1.3 versus 17.9±0.4% [P=0.005] and 25.6±0.2 versus 19.3±2.7% [P=0.016] for protein S-S460P and wt-protein S, respectively).
C4BP? Partially Compensates the Increased Clearance Rate of Protein S-S460P
Because C4BP? reduces the clearance rate of wt-protein S, it was of further interest to investigate the effect of C4BP? on the in vivo survival of protein S-S460P. In preliminary experiments, the interaction between C4BP? and this particular mutant was examined using surface plasmon resonance analysis. As expected, these experiments revealed that the mutant was similar to wt-protein S with regard to C4BP binding in terms of association and dissociation rate constants.24 Protein S-S460P was radiolabeled and subsequently incubated with a 2-fold molar excess of human C4BP? before intravenous injection. The initial recovery 3 minutes after injection was 61.2±6.4%, which is similar to the recovery of protein S-S460P in the absence of C4BP (Table). However, the protein S-S460P/C4BP? complex disappeared less rapidly from the circulation compared with free protein S-S460P (Figure 3D). For instance, at the 15-minute time point, 44.5±1.4% protein S-S460P/C4BP? complex was present compared with 34.9±3.8% of protein S administered in the absence of C4BP? (P=0.015). Calculation of the kinetic parameters revealed that t1/2, t1/2?, and MRT were all increased in the presence of C4BP? (Table). Thus, the increased clearance rate of protein S-S460P is delayed by the presence of C4BP?.
Discussion
Distortion of the balance between procoagulant and anticoagulant pathways can lead to bleeding or thrombotic complications. Such misbalance is often the result of concentrations of components that are too high or too low. For instance, low levels of protein S are associated with an increased thrombotic risk13,14 Low protein S levels may originate from impaired synthesis/secretion, increased clearance, or combinations thereof. Examples of mutations leading to impaired secretion have been described.17–19 However, little is known about the protein S clearance mechanism and how this process is affected by variations in the protein S amino acid sequence or its carrier protein C4BP?.
The possibility that C4BP? affects protein S clearance became apparent in our in vitro experiments regarding the cellular uptake of protein S. We observed that human THP-1 macrophages are able to bind and transport protein S to its intracellular degradation pathways (Figure 1). THP-1 macrophages comprise 4.3x105 binding sites per cell, a number that is in between the amount of binding sites reported previously for bovine aortic endothelial cells (8.5x104 sites per cell32) and human umbilical vein endothelial cells (8.0x105 sites/cell28). Binding and intracellular delivery was markedly reduced in the presence of C4BP? (Figure 2). This effect is in line with the observation by Kask et al that C4BP? interferes with the protein S–dependent uptake of apoptotic cells by macrophages29 and suggests that the C4BP?-binding region (ie, the carboxyl-terminal sex hormone binding globulin-like domain) contributes to the interaction with macrophages. Indeed, preliminary experiments revealed that a recombinant fragment of this region indeed binds to macrophages, albeit with a 5-fold lower affinity (data not shown). However, the notion that C4BP? inhibits cellular uptake only partially points to the possibility that other parts also not covered by C4BP?, such as the amino terminal Gla domain, can mediate cellular binding. Alternatively, the residual uptake may also be facilitated by low-density lipoprotein receptor–related protein, which is present in THP-1 macrophages. We reported previously that this receptor contributes to C4BP? catabolism in vitro and in vivo.33
As to the behavior of protein S in vivo, we found that protein S disappeared in a biphasic manner, with an overall MRT of 1.9 hours (Figure 3; Table). The relatively rapid disappearance from plasma went together with a rapid appearance of radioactivity in the liver (data not shown), suggesting that protein S is mainly directed to this organ. Little contribution of other organs to the uptake of protein S was observed, although in time, accumulation of radioactivity in the stomach was observed. Similar delayed accumulation of radioactivity in the stomach has been described for other proteins, such as von Willebrand factor (vWF) and insulin growth factor binding protein-3.31,34 In case of the latter, this accumulated radioactivity represents small, degraded peptides that are excreted from the liver and reabsorbed by the stomach. Given the time course by which radioactivity appears in liver and stomach, respectively, it seems conceivable that a similar mechanism applies to protein S.
The murine model was particularly useful to study the effect of C4BP? on protein S clearance for 2 reasons: (1) murine C4BP lacks the ?-chain, which renders it unable to bind protein S;35 and (2) murine protein S displays low affinity for human C4BP?, which is mainly because of the presence of a Glu instead of a Gln at position 42736,37. Thus, the complex between human protein S and human C4BP? is affected by their murine counterparts only to a minor extent, if any. The presence of C4BP? resulted in a 4-fold prolonged survival of protein S in the circulation and a delayed uptake in the liver (Figure 3; Table). This demonstrates that C4BP? displays a similar protective effect in the in vivo mouse model compared with the in vitro system using human cells. Of interest, the disappearance of the C4BP?/protein S complex mimicked that of C4BP? alone (Figure 3B, inset), suggesting that clearance of the complex is merely mediated by C4BP?. It should be noted that the half-life of the protein S/C4BP? complex in the mouse is shorter than the half-life of protein S in humans (8 hours compared with 40 hours based on the disappearance of the protein on warfarin treatment). Similar species-specific differences have been reported for numerous other proteins, including factor IX and vWF. Whereas the half-life of vWF in humans is 14 hours, its survival in mice is reduced 4- to 5-fold.31 These differences in clearance rate are probably a reflection of the basal metabolic rate, which is increased in mice.
The protective effect of C4BP? seems to resemble the effect of vWF on the in vivo survival of coagulation factor VIII because the absence of vWF results in increased clearance of factor VIII.38 It is tempting to extrapolate our data obtained with in vitro experiments using human THP-1 cells, and the murine in vivo experiments to the human situation in that C4BP? interferes with protein S clearance. Unfortunately, no direct evidence is available to support this view. It is remarkable, though, that in studies related to the expression of naturally occurring protein S mutants, the levels of total protein S in the corresponding patients were higher than what would be expected from the in vitro expression studies.17,18 More specifically, levels of free protein S were severely reduced, whereas levels of C4BP?-bound protein S were only mildly reduced in these patients.17,18 As such, these data are in support of the possibility that C4BP?-bound protein S displays a longer survival in the circulation than free protein S.
One protein S variant that is consistently associated with reduced antigen levels is protein S-Heerlen, which occurs in 0.5% of the population.20 Although, in particular, the levels of free protein S-Heerlen are decreased,18,22 total protein S levels are also slightly but significantly reduced.18 The underlying mechanism of this phenomenon has remained unclear because expression studies using recombinant variants demonstrated that protein S-S460P is synthesized and secreted as efficiently as wt-protein S.21 Thus, other mechanisms should be responsible for the low levels of protein S-Heerlen. We therefore applied our experimental mouse model to investigate the in vivo survival of free protein S-S460P. Our experiments using purified recombinant protein S-S460P clearly reveal that this mutant disappears from the circulation more rapidly than wt-protein S (Figure 3), resulting in a 3-fold shorter MRT (Table). This demonstrates that the Ser460 to Pro substitution, per se, is associated with increased clearance and explains why levels of free protein S are reduced in individuals carrying the Heerlen polymorphism. Hence, increased clearance is a mechanism that contributes to the pathology of protein S deficiency.
The reason why protein S-S460P is cleared more rapidly remains to be determined, but it may be associated with mutation-induced conformational changes. Alternatively, it may be related to the absence of the carbohydrate chain at position Asn458. Irrespective of the exact mechanism, it should be noted that the increased clearance rate could be compensated partially in the presence of C4BP? (Figure 3). C4BP? interacts with several regions in the sex-hormone–binding-globulin-like domain, including the region 453 to 460, which encompasses the Asn458 glycosylation site and the Ser460Pro replacement.39 This provides C4BP? the potential to mask mutation-induced alterations, allowing C4BP? to counteract the increased clearance. This would explain the near-normal levels of C4BP?/protein S-Heerlen and offers a rationale for the observed Type III deficiency phenomenon.
Acknowledgments
This study was supported by grants from the Dutch Thrombosis Foundation (grant 2002.2 to P.J.L.), the Netherlands Organization for Scientific Research (NWO-VIDI 917-36-372 to T.M.H.), an INSERM-ZonMW/NWO exchange grant (910-48-603 to C.V.D. and P.J.L.) and an INSERM AVENIR program grant (to C.V.D.).
References
Zoller B, Garcia de Frutos P, Dahlback B. Evaluation of the relationship between protein S and C4b-binding protein isoforms in hereditary protein S deficiency demonstrating type I and type III deficiencies to be phenotypic variants of the same genetic disease. Blood. 1995; 85: 3524–3531.
Dahlback B, Stenflo J. High molecular weight complex in human plasma between vitamin K-dependent protein S and complement component C4b-binding protein. Proc Natl Acad Sci U S A. 1981; 78: 2512–2516.
Hillarp A, Dahlback B. Novel subunit in C4b-binding protein required for protein S binding. J Biol Chem. 1988; 263: 12759–12764.
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: 87–91.
Webb JH, Blom AM, Dahlback B. Vitamin K-dependent protein S localizing complement regulator C4b-binding protein to the surface of apoptotic cells. J Immunol. 2002; 169: 2580–2586.
Liu D, Guo H, Griffin JH, Fernandez JA, Zlokovic BV. Protein S confers neuronal protection during ischemic/hypoxic injury in mice. Circulation. 2003; 107: 1791–1796.
Dahlback B. Protein S and C4b-binding protein: components involved in the regulation of the protein C anticoagulant system. Thromb Haemost. 1991; 66: 49–61.
Esmon CT. The protein C pathway. Chest. 2003; 124: 26S–32S.
Walker FJ. Regulation of activated protein C by protein S. The role of phospholipid in factor Va inactivation. J Biol Chem. 1981; 256: 11128–11131.
Koedam JA, Meijers JC, Sixma JJ, Bouma BN. Inactivation of human factor VIII by activated protein C. Cofactor activity of protein S and protective effect of von Willebrand factor. J Clin Invest. 1988; 82: 1236–1243.
Heeb MJ, Mesters RM, Tans G, Rosing J, Griffin JH. Binding of protein S to factor Va associated with inhibition of prothrombinase that is independent of activated protein C. J Biol Chem. 1993; 268: 2872–2877.
Hackeng TM, van ’t Veer C, Meijers JC, Bouma BN. Human protein S inhibits prothrombinase complex activity on endothelial cells and platelets via direct interactions with factors Va and Xa. J Biol Chem. 1994; 269: 21051–21058.
Comp PC, Nixon RR, Cooper MR, Esmon CT. Familial protein S deficiency is associated with recurrent thrombosis. J Clin Invest. 1984; 74: 2082–2088.
Broekmans AW, Bertina RM, Reinalda-Poot J, Engesser L, Muller HP, Leeuw JA, Michiels JJ, Brommer EJ, Briet E. Hereditary protein S deficiency and venous thrombo-embolism. A study in three Dutch families. Thromb Haemost. 1985; 53: 273–277.
Gandrille S, Borgel D, Sala N, Espinosa-Parrilla Y, Simmonds R, Rezende S, Lind B, Mannhalter C, Pabinger I, Reitsma PH, Formstone C, Cooper DN, Saito H, Suzuki K, Bernardi F, Aiach M. Protein S deficiency: a database of mutations—summary of the first update. Thromb Haemost. 2000; 84: 918.
Simmonds RE, Zoller B, Ireland H, Thompson E, Garcia de Frutos P, Dahlback B, Lane DA. Genetic and phenotypic analysis of a large (122-member) protein S-deficient kindred provides an explanation for the familial coexistence of type I and type III plasma phenotypes. Blood. 1997; 89: 4364–4370.
Rezende SM, Lane DA, Zoller B, Mille-Baker B, Laffan M, Dalhback B, Simmonds RE. Genetic and phenotypic variability between families with hereditary protein S deficiency. Thromb Haemost. 2002; 87: 258–265.
Espinosa-Parrilla Y, Yamazaki T, Sala N, Dahlback B, Garcia de Frutos P. Protein S secretion differences of missense mutants account for phenotypic heterogeneity. Blood. 2000; 95: 173–179.
Tsuda H, Urata M, Tsuda T, Wakiyama M, Iida H, Nakahara M, Kinoshita S, Hamasaki N. Four missense mutations identified in the protein S gene of thrombosis patients with protein S deficiency: effects on secretion and anticoagulant activity of protein S. Thromb Res. 2002; 105: 233–239.
Bertina RM, Ploos van Amstel HK, van Wijngaarden A, Coenen J, Leemhuis MP, Deutz-Terlouw PP, van der Linden IK, Reitsma PH. Heerlen polymorphism of protein S, an immunologic polymorphism due to dimorphism of residue 460. Blood. 1990; 76: 538–548.
Giri TK, Yamazaki T, Sala N, Dahlback B, Garcia de Frutos P. Deficient APC-cofactor activity of protein S Heerlen in degradation of factor Va Leiden: a possible mechanism of synergism between thrombophilic risk factors. Blood. 2000; 96: 523–531.
Schwarz HP, Heeb MJ, Lottenberg R, Roberts H, Griffin JH. Familial protein S deficiency with a variant protein S molecule in plasma and platelets. Blood. 1989; 74: 213–221.
Duchemin J, Gandrille S, Borgel D, Feurgard P, Alhenc-Gelas M, Matheron C, Dreyfus M, Dupuy E, Juhan-Vague I, Aiach M. The Ser 460 to Pro substitution of the protein S alpha PROS1. gene is a frequent mutation associated with free protein S (type IIa) deficiency. Blood. 1995; 86: 3436–3443.
Koenen RR, Gomes L, Tans G, Rosing J, Hackeng TM. The Ser460Pro mutation in recombinant protein S Heerlen does not affect its APC-cofactor and APC-independent anticoagulant activities. Thromb Haemost. 2004; 91: 1105–1114.
Morboeuf O, Borgel D, Aiach M, Kaabache T, Gandrille S, Gaussem P. Expression and characterization of recombinant protein S with the Ser 460 Pro mutation. Thromb Res. 2000; 100: 81–88.
van Wijnen M, Stam JG, Chang GT, Meijers JC, Reitsma PH, Bertina RM, Bouma BN. Characterization of mini-protein S, a recombinant variant of protein S that lacks the sex hormone binding globulin-like domain. Biochem J. 1998; 330: 389–396.
van de Poel RH, Meijers JC, Rosing J, Tans G, Bouma BN. C4b-binding protein protects coagulation factor Va from inactivation by activated protein C. Biochemistry. 2000; 39: 14543–14548.
Hackeng TM, Hessing M, van ’t Veer C, Meijer-Huizinga F, Meijers JC, de Groot PG, van Mourik JA, Bouma BN. Protein S binding to human endothelial cells is required for expression of cofactor activity for activated protein C. J Biol Chem. 1993; 268: 3993–4000.
Kask L, Trouw LA, Dahlback B, Blom AM. The C4b-binding protein-protein S complex inhibits the phagocytosis of apoptotic cells. J Biol Chem. 2004; 279: 23869–23873.
Denis C, Methia N, Frenette PS, Rayburn H, Ullman-Cullere M, Hynes RO, Wagner DD. A mouse model of severe von Willebrand disease: defects in hemostasis and thrombosis. Proc Natl Acad Sci U S A. 1998; 95: 9524–9529.
Lenting PJ, Westein E, Terraube V, Ribba AS, Huizinga EG, Meyer D, de Groot PG, Denis CV. An experimental model to study the in vivo survival of von Willebrand factor. Basic aspects and application to the R1205H mutation. J Biol Chem. 2004; 279: 12102–12109.
Stern DM, Nawroth PP, Harris K, Esmon CT. Cultured bovine aortic endothelial cells promote activated protein C-protein S-mediated inactivation of factor Va. J Biol Chem. 1986; 261: 713–718.
Westein E, Denis CV, Bouma BN, Lenting PJ. The alpha-chains of C4b-binding protein mediate complex formation with low density lipoprotein receptor-related protein. J Biol Chem. 2002; 277: 2511–2516.
Arany E, Zabel P, Hill DJ. Rapid clearance of human insulin-like growth factor binding protein-3 from the rat circulation and cellular localization in liver, kidney and stomach. Growth Regul. 1996; 6: 32–41.
Rodriguez de Cordoba S, Perez-Blas M, Ramos-Ruiz R, Sanchez-Corral P, Pardo-Manuel de Villena F, Rey-Campos J. The gene coding for the beta-chain of C4b-binding protein (C4BPB) has become a pseudogene in the mouse. Genomics. 1994; 21: 501–509.
Chu MD, Sun J, Bird P. Cloning and sequencing of a cDNA encoding the murine vitamin K-dependent protein S. Biochim Biophys Acta. 1994; 1217: 325–328.
Fernandez JA, Griffin JH, Chang GT, Stam J, Reitsma PH, Bertina RM, Bouma BN. Involvement of amino acid residues 423–429 of human protein S in binding to C4b-binding protein. Blood Cells Mol Dis. 1998; 24: 101–112.
Lenting PJ, van Mourik JA, Mertens K. The life cycle of coagulation factor VIII in view of its structure and function. Blood. 1998; 92: 3983–3996.
Giri TK, Linse S, Garcia de Frutos P, Yamazaki T, Villoutreix BO, Dahlback B. Structural requirements of anticoagulant protein S for its binding to the complement regulator C4b-binding protein. J Biol Chem. 2002; 277: 15099–15106.(Cécile V. Denis; Sarah J.)