Traffic 2004; 5: 69–78 Blackwell Munksgaard
Copyright
#
Blackwell Munksgaard 2004
doi: 10.1046/j.1600-0854.2003.00157.x
Review
How to Roll an Endothelial Cigar: The Biogenesis of Weibel-Palade Bodies Gre´goire Michaux and Daniel F. Cutler* MRC Laboratory for Molecular Cell Biology and Cell Biology Unit and Department of Biochemistry, University College London, London WC1E 6BT, UK * Corresponding author: Daniel F. Cutler,
[email protected] Weibel-Palade bodies (WPB) are the regulated secretory organelles of endothelial cells. These cigar-shaped membrane-bound structures function in both hemostasis and inflammation but their biogenesis is poorly understood. Here, we review what is currently known about their formation. The content of WPBs is dominated by the hemostatic factor von Willebrand factor (VWF), whose complex biogenesis ends in the formation of high molecular weight multimers. VWF is also organized into proteinaceous tubules which underlie the striated interior of WPBs as seen in the EM. VWF expression is necessary for formation of WPBs, and its heterologous expression can even lead to the specific recruitment of WPB membrane proteins, including the leukocyte receptor P-selectin, the tetraspanin CD63, and Rab27a. Unusually, the VWF propeptide is implicated in the biogenesis of WPBs, being essential for formation of the storage compartment. The elongation of the cigars and the formation of the tubules are determined by non-covalent interactions between pro- and mature VWF proteins. Surprisingly, high molecular weight multimers seem neither necessary nor sufficient to trigger formation of a storage compartment, and do not seem to have any role in WPB biogenesis. Von Willebrand’s disease, usually caused by mutations within VWF, has provided many of the insights into the way in which VWF drives the formation of these organelles. Key words: endothelial cells, hemostasis, inflammation, organelle biogenesis, Weibel-Palade bodies Received 24 October 2003, revised and accepted for publication 6 November 2003
First described by Weibel and Palade in 1964 (1), the endothelial-specific regulated secretory organelles known as Weibel-Palade bodies (WPB) are still largely uncharacterized despite playing a key role in both hemostasis and inflammation. It wasn’t until 1982 that von Willebrand factor (VWF) was identified as the protein which forms the proteinaceous tubules that fill the lumen of these cigar-shaped membrane-bound structures (2), and the best characterized membrane protein of this organelle, P-selectin, was first shown to be a component of WPBs only in 1989 (3,4).
The importance of WPBs can be illustrated by the function of these two components. Mutations within VWF are the usual cause of the most common inherited bleeding disorder, von Willebrand disease (VWD; see Table 1). VWD has an estimated prevalence in some human populations of up to 1%, and is most often characterized by prolonged and variable mucocutaneous bleeding. Type III von Willebrand Disease (Table 1) is a severe bleeding disorder, not unlike severe hemophilia type A or B. VWF acts in primary hemostasis to recruit platelets at a site of injury, and is also important in secondary hemostasis, acting as a chaperone for coagulation factor VIII (FVIII). The second of the two components, the only membrane protein whose endothelial-specific function is well understood, is the leukocyte receptor P-selectin. Its storage in WPB ensures its regulated appearance at the apical plasma membrane of endothelial cells to initiate leukocyte recruitment in response to vascular injury. The WPB thus, at a minimum, regulates both the release of a primary hemostatic factor and the appearance of a protein that is central to the inflammatory response.
Expression and Storage of VWF Von Willebrand factor expression is necessary but not sufficient to trigger WPB formation. In porcine aortic endothelial cells, VWF is only constitutively secreted (5), and in megakaryocytes VWF is stored in alpha-granules, which are round organelles (6) instead of having an elongated shape like WPB. In contrast, several non-endothelial cell lines which are not usually VWF-positive can make WPB-like organelles upon heterologous VWF expression [reviewed in (7)], even in cell lines where there are already endogenous secretory granules present (8,9). These de novo formed ‘WPB’ can recruit appropriate membrane proteins, underlining the primary importance of VWF in WPB formation. This review will therefore focus on the biosynthesis of VWF and its role in the formation of WPB. For a review on WPB exocytosis, see (10).
Biosynthesis of VWF Pre-pro-VWF is synthesized at the endoplasmic reticulum (ER) where the signal peptide is cleaved to give a precursor, pro-VWF, composed of a propeptide of 741 amino acids (aa) and a mature protein of 2050 aa (Figures 1 and 2). VWF is 69
Michaux and Cutler Table 1: Classification of von Willebrand’s Disease (VWD) Type 1: A
About 70% of VWD
B
Quantitative partial deficiency: reduced von Willebrand factor (VWF) level in plasma
C
The molecular basis is not well known; most of the identified mutations are substitutions
D
Reduced VWF secretion
E
Normally autosomal dominant but recessives can occur (74–76) Effects due to ER retention (76,77) and degradation by proteasome (78).
F
Modifier of VWF: dominant effect of a glycosyl transferase defect leading to rapid degradation of VWF in plasma in mouse (79)
Type 2: A
About 30% of VWD
B
Qualitative deficiency; smaller multimers; reduced binding to coagulation factor VIII (2N) or platelet binding reduced (2A, 2M) or increased (2B)
C
Almost exclusively substitutions
D
No effect on secretion except for some mutations with a combined Type 1 phenotype (see below)
E
High molecular weight multimers (HMW) are absent in 2A, 2B; superHMW in Vicenza variant; HMW present in 2M and some 2B
Type 3: A
1-2% of VWD
B
Quantitative total deficiency: no VWF in plasma
C
Most mutations are nonsense or deletions/insertions; of 14 identified substitutions, 8 implicate Cys ! probable misfolding
D
No VWF secretion
E
ER retention and degradation (80)
Combined type 1 and 2 have been reported (73,74).
composed of several domains, to some of which functions have been assigned (Figure 1). It is glycosylated at 12 N-linked sites (11) and dimerizes through the formation of C-terminal disulfide bonds. A deletion of the C-terminal region involved in dimerization leads to ER degradation, whereas the peptide resulting from expression of this region on its own is able to dimerize and to leave the ER (12). When HUVECs are treated with tunicamycin to block N-linked glycosylation (13), dimerization is also blocked; the VWF is then trapped in the ER. One major consequence of clinically significant mutations in VWF is to produce misfolding of the protein leading to ER retention (Table 1). ER retention of misfolded protein will induce the unfolded protein response which can in turn activate a caspasedependent cell death by apoptosis (14). The effects of VWD on endothelial cell death have not yet been investigated. VWF that can exit the ER is then transported to the Golgi apparatus, where O-linked sugars are added and the N-linked oligosaccharides are further modified. Blocking such modifications with swainsonine does not affect subsequent traffic to the trans-Golgi network (TGN), where sulfation of both the propeptide and mature VWF (15) likely occurs (16). At this time pro-VWF starts to multimerize through the formation of N-terminal disulfide bonds, a 70
modification that is independent of the ER located C-terminal dimerization (17). Pro-VWF is also cleaved and this cleavage leads to the propeptide and mature VWF still being linked by non-covalent bonds in a 1 : 1 stoichiometric ratio (18,19). It has been found that a non-covalent association, dependent on acidic pH and high calcium concentration, links the propeptide and mature VWF (20) and it is likely that the propeptide interacts with the N-terminal part of mature VWF before cleavage. This continued close association of the propeptide with mature VWF is in contrast with the situation in, e.g. insulin-containing granules, exemplified by dissociation and secretion by a different route of the proinsulin-derived C-peptide from insulin itself in pancreatic beta cells (21). A likely candidate endopeptidase for VWF cleavage is furin (22), an enzyme localized in the TGN (23) which has the appropriate sequence preference [RSKR (24)]. The Golgi-associated multimerization of VWF to produce the high molecularweight multimers, which can reach 20 million kDa, is essential for its function, and is often affected in VWD. Mutations which lead to an excess of low or indeed of high molecular weight multimers have been described (Table 1). The former lead directly to hemostatic problems (see below); the latter can lead to rapid clearance of the secreted variant. Traffic 2004; 5: 69–78
Biogenesis of Weibel-Palade Bodies
Figure 1: Schematic view of pre-pro-von Willebrand factor. The signal peptide (SP) is 22 aa long. The numbers in parentheses indicate the length of the propeptide (red) and mature VWF (green). Domains necessary for multimerization, dimerization and storage are indicated. Note that the C-terminal boundary of the domain required for storage is not clearly identified (see text). Domains implicated in binding to other proteins are represented by black bars.
Figure 2: Trafficking of von Willebrand factor. The left column indicates the location and nature of the main VWF modifications (in black), as well as the stage at which P-selectin and Rab27a are recruited (blue and green, respectively). The right column (in red) summarizes major links between VWF trafficking and von Willebrand’s disease. Note that, so far, no mutation is known to directly prevent WPB formation without affecting the constitutive secretory pathway as well. LMW: low molecular weight: HMW: high molecular weight.
Traffic 2004; 5: 69–78
71
Michaux and Cutler
The intra-Golgi environment is important for modification and processing of VWF. Low concentrations of monensin, which perturbs intra-Golgi pH, reduce cleavage of the propeptide, whereas higher concentrations affect VWF multimerization and glycosylation in a dose-dependent manner (25). Interestingly, when the pH is modulated by ammonium chloride or chloroquine, glycosylation and sulfation are not affected, but multimerization is reduced (not completely prevented) in a dose-dependent manner, confirming that an acidic pH is necessary for late stages of this process to occur, an observation confirmed by in vitro VWF multimerization which occurs only in acidic buffers (26). These data imply also that monensin does not act on pH alone to disrupt glycosylation and cleavage. The latter might reflect an effect of monensin on the intracellular localization of furin (27). When sulfation is inhibited with sodium chlorate, VWF is still able to multimerize normally, showing that sulfation is not involved in this process (15).
to be in recruitment of platelets from the plasma, thereby requiring VWF to be released from the apical plasma membrane, this is a surprising result. VWD is often treated with 1-deamino-8-D-arginine vasopressin (DDAVP) as secretagogue, which rapidly leads to a substantial increase in plasma VWF. Were VWF to be secreted mainly from the basolateral surface, then the DDAVP response would primarily lead to a build-up in subendothelial VWF. This might be consistent with the fact that after vascular injury, which is known to trigger VWF regulated exocytosis, platelets are directly recruited at the newly exposed subendothelial matrix, partly through VWF acting as a bridge between platelets and collagen (note the collagen binding domain on VWF in Figure 1). Alternatively, DDAVPreleased VWF could be transported to the apical side of the endothelial barrier after secretion (33). However, another study of HUVECs treated with PMA concluded that constitutive secretion was mainly basolateral but regulated secretion predominantly apical (34).
When leaving the TGN, VWF is either constitutively secreted, or is stored in newly formed WPB (Figure 2). The regulated and constitutive pathways are independent in as much as thrombin stimulation of regulated exocytosis does not lead to a subsequent reduction in constitutive secretion (28). The degree of multimerization of the VWF released by the two pathways is different. The highest molecular weight multimers are preferentially incorporated in WPB and thus released at sites of injury in a regulated fashion. This is important since the affinity of high molecular weight VWF multimers for their platelet receptor is about 100 times greater than that of low molecular weight forms (29). It has been proposed that up to 95% of VWF is constitutively secreted as a population of partially cleaved dimeric or low molecular weight multimers (30). However, other studies suggest that in HUVECs, constitutive secretion is not significant, VWF being mainly secreted from the storage pool (31).
An indirect analysis of the polarity of VWF secretion has been performed in MDCK-II cells, a well-known epithelial cell line, following VWF expression (35). The authors of this study concluded that the constitutive secretion of VWF was directed slightly more to the apical surface, but that after stimulation by PMA, more than 80% of VWF was targeted to the apical medium, in apparent agreement with the findings of van Buul-Wortelboer and coworkers (34). However, the relative polarity of endothelial and epithelial cells has been separately addressed by heterologous expression of the coagulation cascade component tissue factor. This molecule was secreted apically from endothelial cells but from the basolateral surface of MDCK cells, i.e. was consistently released at the blood side of these two cell types (36). At present, therefore, the findings on both constitutive vs. regulated and apical vs. basolateral targeting are not fully consistent.
Despite constitutively secreted VWF being of low molecular weight, plasma VWF contains a high proportion of high molecular weight multimers. Thus if 90% of VWF is constitutively secreted, extracellular multimerization would need to take place to account for all the high molecular weight multimers found in plasma. Alternatively, a substantial percentage of WPB might be released without secretagogue stimulation, a hypothesis supported by the relatively rapid 24-h turnover of WPB even in the absence of stimulation (32), compared to the granules of PC12 and chromaffin cells in culture which can last for longer than a week. In addition to targeting between constitutive and regulated secretion of VWF, there is an apical/basolateral polarity to VWF release. Wagner and colleagues, using HUVECs, found that while constitutive secretion was not polarized, 90% of PMA-induced secretion was basolateral (33). As the main role of VWF after regulated secretion is thought 72
Storage of VWF in WPBs There are two main hypotheses with which to explain the formation of secretory granules. The first centers on selective coaggregation of stored components in the TGN, leading to their selective retention within the forming granule. The second evokes a central role for sorting receptor(s). While there is considerable experimental support for the former, the latter has always been controversial. In the case of WPBs, there are two findings arguing for selective aggregation being central to the storage of VWF. Firstly, the central role taken by VWF itself, including its ability to drive the formation of independent granules in the presence of endogenous storage organelles; secondly the formation of pseudo-WPBs in non-endothelial cells that do not usually make secretory granules. Together, these strongly suggest the self-organization that is the hallmark of selective aggregation. Traffic 2004; 5: 69–78
Biogenesis of Weibel-Palade Bodies
Detailed information is emerging as to the mechanism of storage of VWF in terms of the processing and modification of this molecule, and in particular the role of the propeptide in relation to mature VWF. The first strong hint as to the importance of the propeptide came from a study in CV–1 cells showing that the propeptide coexpressed in trans (i.e. from a separate plasmid) with mature VWF can induce VWF granular storage*, whereas mature (i.e. with the propeptide deleted) VWF expressed alone was not stored (37). This suggested that the propeptide might be necessary for storage. More recently, several studies from Haberichter, Montgomery and coworkers using the AtT-20 cell line have detailed the key role of the propeptide, using chimeric constructs of canine and human VWF. Having established that normal multimerization and storage occurred when expressing full-length canine or human VWF, they showed that the propeptide is necessary for both multimerization of VWF and for its storage (9). Expression of the propeptide and mature VWF in trans gave the same result, confirming the Voorberg data (36). However, when expressing human/canine chimeric proteins in trans, multimerization occurs normally but such chimeric proteins are stored only if the propeptide is human; with a canine propeptide, a human mature VWF can multimerize but is not stored along with the canine propeptide. These results confirm the earlier finding of Mayadas and Wagner (38) that multimerization is not needed for storage and that these two processes are independent. Further, when expressing the propeptide or mature VWF alone, it was found that only the propeptide is stored, in contrast to mature VWF. This was confirmed by showing that in AtT-20 and in bovine aortic endothelial cells, the propeptide can induce the storage of the precursor of C3, a component of the complement complex, when fused with it (39). When the two peptides are separated by a furin cleavage site, the C3 peptide was not stored, demonstrating that cleavage occurs before storage, and incidentally that cleavage is not necessary for storage, at least in that context. Propeptide cleavage is not necessary for VWF storage in CV–1 cells (37), and when expressed in AtT-20 or RIN 5F cells, non-cleavable VWF may be stored (40) or not stored (41), confirming that cellular context and the precise nature of the cleavage defect may play an important role in storage. The propeptide thus seems to play a crucial role for the formation of the independent storage organelle formed in AtT-20 cells.
*The ‘storage organelles’ found following expression of VWF in heterologous systems have not always been tested for regulated release, they are not always the elongated shape that is the hallmark of true WPBs, nor are they all known to contain the internal striations that represent the proteinaceous tubules that are formed by VWF. For the sake of brevity in this review, we have described all VWF-positive cytoplasmic accumulations as storage organelles. For further discussion, see Hannah et al. (7)
Traffic 2004; 5: 69–78
However, similar experiments in CV–1 and AtT-20 cells have given opposing results. In CV–1 cells, the lack of an antibody specifically recognizing the propeptide alone precluded its localization, and therefore any analysis of its possible independent storage. But when a construct containing the propeptide and the D0 –D3 domains of mature VWF was expressed in CV–1 cells, no storage was found (37), although these peptides contain the domains defined as necessary by Haberichter et al. for interaction and multimerization (see below). Indeed, in AtT-20 cells, a similar construct led to storage of such a peptide (39). The explanation for this discrepancy is not yet clear, but could come from the existence of an endogenous regulated storage organelle in AtT-20, or from differences between the different constructs used. It has recently been shown that the granular storage compartment where the propeptide is found when expressed alone in AtT-20 cells is the ACTH-containing endogenous granules (81). The presence of mature VWF is necessary in these cells to induce an independent storage organelle similar to WPBs. Thus it is not simply the case that the propeptide triggers the formation of an independent storage compartment, and recruits the mature VWF into that organelle by direct interaction between the two. The interaction between the propeptide and mature VWF has been analyzed further by Haberichter et al., who have identified the amino acids mediating the non-covalent interaction between the two peptides necessary to induce the storage of mature VWF (42). They have identified R416 in the propeptide and T869 in mature VWF as being needed for storage, and their data imply that a direct interaction between the propeptide and the mature protein is important to this function. An important feature of WPB is the special arrangement of VWF inside this elongated organelle (Figure 3A). Electron microscopy studies show that the lumen of WPB is filled from one end to the other with striations (18,37,43). Transverse sections demonstrate that the striations are due to the organization of VWF into tubules running along the main axis of the WPB. This particular feature is found not only in endothelial cells but also in transformed cell lines, including CV–1, T24 and HEK293 cells (7,37,44,45). Wagner and coworkers showed that in RIN 5F cells, expression of a VWF protein with the CK domain (Figure 1) deleted is stored in elongated organelles with internal striations formed by tubules. Such a protein can only form dimers through the N-terminal part of mature VWF (D0 -D3 domains – Figure 1). However, a longer C-terminal deletion abolished the formation of elongated structures (40). By expressing human VWF variants in HEK293 cells, it has been shown that two different mutations preventing formation of multimers beyond 6–8-mers had a completely different effect on elongation of WPBs: an R273W mutation almost abolished elongation, whereas C788R has no significant effect on the elongation of WPB (44). Together, 73
Michaux and Cutler
these studies demonstrate that not only is multimerization unnecessary for storage, but also for elongation. The data also strongly suggest that non-covalent interactions are crucial for the formation of the VWF tubules. One further striking observation was that a single tubule is enough to trigger at least partial elongation of the organelle (Figure 3B) (44), whereas it is possible to find bent and flexing WPB [Figure 3C; see the videos in Data supplement to (46)], clearly indicating that, while the tubules are rigid enough to trigger elongation (the simple hypothesis explaining the elongated shape of WPBs), this does not prevent flexibility. Although a number of issues are still controversial, the formation of WPB has been partly elucidated. It is clear that VWF expression is necessary to trigger their formation; that the propeptide is directly implicated in this process (although it is likely that some, probably N-terminal, domains within mature VWF must also be intact and present); and that elongation and tubule formation is determined by non-covalent interactions between VWF proteins. Thus, multimerization, which is neither necessary nor sufficient to trigger storage, does not seem to have any role in WPB biogenesis. If multimerization as measured for VWF is equivalent to the aggregation of content found to be important in forming other secretory granules (47–49), then WPB biogenesis may be a very different process.
Recruitment of Membrane Proteins
Figure 3: Weibel-Palade body ultrastructure and Rab27a recruitment. A–C: Ultrastructure of WPB after transient expression of recombinant VWF in HEK293 cells. cDNA encoding either WT-VWF (A), or human variants C1225G-VWF (B) and C788R-VWF (C) was introduced into HEK293 cells which were left for 48 h and then processed for EM as described (44). Note that one tubule (panel B) or only a few tubules (panel C), but certainly less than is needed to fill the organelle as is seen with WT VWF (panel A), may be sufficient to give an elongated shape to the WPB (arrowhead). D: Rab27a-GFP (in green) colocalizes with VWF (in red) in WPB after expression in HUVECs. Note that in these three cells, perinuclear ‘immature’ WPB are GFP negative (arrows), in contrast with peripheral WPB (arrowheads). Scale bar is 200 nm (A–C) and 10 mM (D). Panels A, B and C are reproduced from Gre´goire Michaux, Lindsay J. Hewlett, Anne C. Goodeve, Ian R. Peake, Martina E. Daly and Daniel F. Cutler. Analysis of storage and regulated secretion of three human variants of von Willebrand Factor provides new insights into von Willebrand’s Disease. Blood 2003; 102 : 2452–2458. Copyright American Society of Hematology, used with permission.
74
There are only four proteins known to be recruited to the membrane of WPB. Three of them are integral membrane proteins: the leukocyte adhesion receptor P-selectin (3,4), the tetraspanin CD63 (50) and fucosyl transferase VI (51). P-selectin is only expressed in platelets and in endothelial cells where VWF is also present and is stored in secretory granules where VWF is present. In the absence of VWF to provide a storage compartment, P-selectin is targeted to endosomes and lysosomes where it is degraded (7,52). However, in platelets and in AtT-20 cells, in the absence of VWF, P-selectin is still sorted to the alpha-granules (52) and to ACTH endogenous granules (53), respectively. Several studies have identified a number of targeting signals in the cytoplasmic tail of P-selectin which suffice to recapitulate the trafficking of this protein [for review (7)]. Even in the absence of its lumenal domain, the fusion protein P-selectin-HRP is recruited by VWF (8,44) in heterologous expression experiments. The precise mechanism of P-selectin recruitment is not known. As soon as storage granules are formed by VWF, even in heterologous systems, P-selectin is recruited, indicating that VWF contains some sort of signal that, either directly or indirectly, triggers P-selectin sorting to WPB. This is selective: VWF expression in T24 cells induces WPB formation and recruitment of the endogenous P-selectin (45), but expression of a mutated form of VWF, which does not multimerize but Traffic 2004; 5: 69–78
Biogenesis of Weibel-Palade Bodies
accumulates in the ER, does not recruit P-selectin. However, a VWF human variant which can dimerize but make only very low molecular weight multimers can form WPB and recruit P-selectin and CD63, showing that a high degree of multimerization is not necessary for membrane protein recruitment (44). These data imply that some other event driven by the presence of VWF in a post-Golgi location is needed for the recruitment of P-selectin. This recruitment operates across the bilayer; lumenal VWF drives the recruitment of P-selectin via a tyrosine motif in the latter’s cytoplasmic tail, at least in a heterologous context (7). The recruitment of the universally expressed CD63 to WPBs is similar to that of P-selectin in as much as creation of a storage compartment by expression of VWF leads to its acquisition, but less is known about its targeting signals with respect to secretory organelles. The itinerary of this protein may be more complex than that of P-selectin, since it has a dual intracellular location; CD63 is found at steady state on WPB in endothelial cells but universally (including endothelial cells) in endosomal/lysosomal compartments [reviewed in (7)]. Interestingly, it does not colocalize with P-selectin in the membrane of alpha-granules of platelets, suggesting that VWF in a storage compartment is not sufficient for its recruitment. The third membrane-associated protein known to be present on WPB is the small GTPase Rab27a (54,55). Small GTPases of the Rab family have been described as ‘master regulators’ of membrane traffic, and are thus thought to make a critical contribution to organelle identity (56,57). Rab27a was first found on melanosomes within melanocytes and on lytic granules within T-lymphocytes, and its loss is responsible for the hypo-pigmentation and immunological disorders of Griscelli’s syndrome type 2 (58). Interestingly, Rab27a is not present on the membrane of all WPB in HUVECs (54). Rab27a-negative WPB can be seen in a perinuclear location next to the TGN where WPB are thought to form (Figure 3A). Indeed, when a VWF-GFP fusion protein is expressed in HUVECs, the first GFPpositive cigar-shaped structures which appeared next to the nucleus 4 h after transfection are Rab27a-negative. After 24 h, most of the WPB are Rab27a-positive. This delay in recruitment of Rab27a is reminiscent of the maturation steps which occur during the formation of other classical secretory granules or lysosome-related organelles. This is the first clear molecular distinction between immature and mature WPBs. In addition, Rab27a is recruited by heterologous expression of VWF. The functional consequences of Rab27 recruitment by WPBs are as yet unclear, but this case provides an example of a cargo-driven maturation-dependent recruitment of a Rab protein. As with P-selectin, this recruitment must be driven by a trans-bilayer mechanism, suggesting an important role for the lipid environment created by VWF as it matures. Finally, it has also been proposed that WPB also contain a1.3-Fucosyltransferase VI (FucT-VI) in HUVECs (51). The Traffic 2004; 5: 69–78
surprising presence of the Golgi enzyme FucT-VI in WPB suggests a non-enzymatic role for this protein, including the possibility that it could act as a lectin to help sort fucosylated secretory proteins to the WPB, i.e. as a sorting receptor. As yet, this hypothesis has not been directly tested. Since targeting of glycosyl-transferases has focused on their retention within the Golgi stack via the length of their transmembrane domain, nothing is known as to how such a protein might be targeted to WPBs.
Lumenal Proteins Several lumenal proteins are found in WPB, including the chemotactic cytokine interleukin-8 (59,60), which is stored only upon VWF expression (61), in line with a selective coaggregation model. Tissue-type plasminogen activator (t-PA), an essential enzyme in fibrinolysis and thrombolysis, has also been detected in WPB in HUVECs (62–65), although other groups have found it associated with other intracellular vesicles (66,67). Endothelin-1, a peptide mediating vasoconstriction, has also been partially colocalized with VWF in WPB (68). Finally, and most intriguingly, it has been reported that coagulation FVIII can be produced in sinusoidal endothelial cells in the liver (69). While most FVIII is thought to be produced from non-endothelial cells that do not synthesize VWF, thereby implying that the two proteins can associate only after exocytosis – constitutive or regulated – of VWF, these reports (69,70) raise the possibility of co-expression and therefore co-secretion of these two factors. Whether under these circumstances FVIII would be diverted to WPBs is unclear. Certainly, expression of FVIII in VWF-producing endothelial cells and megakaryocytes leads to co-storage in WPBs (40,71) and granules (72). Given these data and the observation that the secretagogue DDAVP causes a rise not only in plasma VWF but also in FVIII (73), there is a strong presumption for at least some co-storage in vivo.
Mutations Within VWF and Their Effect on WPB Formation The presence of membrane and content proteins in WPB whose presence is dependent on VWF expression raises the question of the consequences of mutations affecting VWF, which may be present in as much as 1% of some human populations. Total loss of VWF expression has been analyzed in the mouse (52), where absence of VWF not only leads to a murine version of severe VWD, but also affects the process of inflammation and the ability of P-selectin to be translocated to the plasma membrane when there is no WPB. The same is presumably true also for other proteins stored in WPB which would be constitutively secreted or degraded instead of being stored and released under the appropriate physiological circumstances. 75
Michaux and Cutler
Little is known about how VWF mutations affect the formation of WPB. For most human variants, analysis has understandably been directed towards understanding how VWF function itself is affected, concentrating on functional assays for binding to factor VIII and platelets. Usually, cellular analyses have involved the expression of recombinant cDNA containing the mutations in COS cells where only constitutive secretion and multimerization can be analyzed. Very few cell lines are suitable for analysis of VWF storage and of regulated secretion, and until recently the ability to recruit membrane proteins had not been tested at all (7). However, it has recently been shown that VWF expression in HEK293 cells can recapitulate VWF storage in elongated organelles, recruitment of specific membrane proteins and regulated secretion upon stimulation. Using this expression system, the effects of three human VWF variants affecting multimerization and inducing mild to severe VWD (44) were analyzed. Not only are the variants stored in WPB, but it was also found that these WPB can still recruit P-selectin, CD63 and Rab27a very efficiently (44,54 and our unpublished data). They can also be stimulated to secrete by a secretagogue known to trigger WPB secretion in endothelial cells, showing that HEK293 provide a convenient system to analyze all the aspects of fully functional WPB, from formation to recruitment of membrane proteins and regulated secretion. WPBs thus provide an organelle at the interface of hemostasis and inflammation, whose biogenesis can be recapitulated in model systems, where the mechanisms involved in secretory granule biogenesis may be uniquely addressed and where human genetic disease provides many of the insights and materials for such investigations: the surprise is that this organelle has been neglected.
Acknowledgments GM is supported by a Marie Curie fellowship from the EU. Work in DC’s lab is funded by the MRC. We thank Martina Daly, Sandra Haberichter, Robert Montgomery, Ian Peake, Jan Voorberg and the Cutler Lab for helpful discussion and suggestions.
References 1. Weibel ER, Palade GE. New cytoplasmic components in arterial endothelia. J Cell Biol 1964;23:101–112. 2. Wagner DD, Olmsted JB, Marder VJ. Immunolocalization of von Willebrand protein in Weibel-Palade bodies of human endothelial cells. J Cell Biol 1982;95:355–360. 3. McEver RP, Beckstead JH, Moore KL, Marshall-Carlson L, Bainton DF. GMP-140, a platelet alpha-granule membrane protein, is also synthesized by vascular endothelial cells and is localized in Weibel-Palade bodies. J Clin Invest 1989;84:92–99. 4. Bonfanti R, Furie BC, Furie B, Wagner DD. PADGEM (GMP140) is a component of Weibel-Palade bodies of human endothelial cells. Blood 1989;73:1109–1112.
76
5. Royo T, Martinez-Gonzalez J, Vilahur G, Badimon L. Differential intracellular trafficking of von Willebrand factor (vWF) and vWF propeptide in porcine endothelial cells lacking Weibel-Palade bodies and in human endothelial cells. Atherosclerosis 2003;167:55–63. 6. Cramer EM, Meyer D, le Menn R, Breton-Gorius J. Eccentric localization of von Willebrand factor in an internal structure of platelet alpha-granule resembling that of Weibel-Palade bodies. Blood 1985;66: 710–713. 7. Hannah MJ, Williams R, Kaur J, Hewlett LJ, Cutler DF. Biogenesis of Weibel-Palade bodies. Semin Cell Dev Biol 2002;13:313–324. 8. Blagoveshchenskaya AD, Hannah MJ, Allen S, Cutler DF. Selective and signal-dependent recruitment of membrane proteins to secretory granules formed by heterologously expressed von Willebrand factor. Mol Biol Cell 2002;13:1582–1593. 9. Haberichter SL, Fahs SA, Montgomery RR. von Willebrand factor storage and multimerization: 2 independent intracellular processes. Blood 2000;96:1808–1815. 10. van Mourik JA, Romani de Wit T, Voorberg J. Biogenesis and exocytosis of Weibel-Palade bodies. Histochem Cell Biol 2002;117:113–122. 11. Titani K, Kumar S, Takio K, Ericsson LH, Wade RD, Ashida K, Walsh KA, Chopek MW, Sadler JE, Fujikawa K. Amino acid sequence of human von Willebrand factor. Biochemistry 1986;25:3171–3184. 12. Voorberg J, Fontijn R, Calafat J, Janssen H, van Mourik JA, Pannekoek H. Assembly and routing of von Willebrand factor variants: the requirements for disulfide-linked dimerization reside within the carboxy-terminal 151 amino acids. J Cell Biol 1991;113:195–205. 13. Wagner DD, Mayadas T, Marder VJ. Initial glycosylation and acidic pH in the Golgi apparatus are required for multimerization of von Willebrand factor. J Cell Biol 1986;102:1320–1324. 14. Nakagawa T, Zhu H, Morishima N, Li E, Xu J, Yankner BA et al. Caspase-12 mediates endoplasmic-reticulum-specific apoptosis and cytotoxicity by amyloid-beta. Nature 2000;403:98–103. 15. Carew JA, Browning PJ, Lynch DC. Sulfation of von Willebrand factor. Blood 1990;76:2530–2539. 16. Huttner WB. Tyrosine sulfation and the secretory pathway. Annu Rev Physiol 1988;50:363–376. 17. Voorberg J, Fontijn R, van Mourik JA, Pannekoek H. Domains involved in multimer assembly of von Willebrand factor (vWF): multimerization is independent of dimerization. EMBO J 1990;9:797–803. 18. Ewenstein BM, Warhol MJ, Handin RI, Pober JS. Composition of the von Willebrand factor storage organelle (Weibel-Palade body) isolated from cultured human umbilical vein endothelial cells. J Cell Biol 1987;104:1423–1433. 19. Wagner DD, Fay PJ, Sporn LA, Sinha S, Lawrence SO, Marder VJ. Divergent fates of von Willebrand factor and its propolypeptide (von Willebrand antigen II) after secretion from endothelial cells. Proc Natl Acad Sci USA 1987;84:1955–1959. 20. Vischer UM, Wagner DD. von Willebrand factor proteolytic processing and multimerization precede the formation of Weibel-Palade bodies. Blood 1994;83:3536–3544. 21. Arvan P, Castle D. Sorting and storage during secretory granule biogenesis: looking backward and looking forward. Biochem J 1998; 332:593–610. 22. Wise RJ, Barr PJ, Wong PA, Kiefer MC, Brake AJ, Kaufman RJ. Expression of a human proprotein processing enzyme: correct cleavage of the von Willebrand factor precursor at a paired basic amino acid site. Proc Natl Acad Sci USA 1990;87:9378–9382. 23. Misumi Y, Oda K, Fujiwara T, Takami N, Tashiro K, Ikehara Y. Functional expression of furin demonstrating its intracellular localization and endoprotease activity for processing of proalbumin and complement pro-C3. J Biol Chem 1991;266:16954–16959. 24. Hosaka M, Nagahama M, Kim WS, Watanabe T, Hatsuzawa K, Ikemizu J, Murakami K, Nakayama K. Arg-X-Lys/Arg-Arg motif as a signal for precursor cleavage catalyzed by furin within the constitutive secretory pathway. J Biol Chem 1991;266:12127–12130.
Traffic 2004; 5: 69–78
Biogenesis of Weibel-Palade Bodies 25. Wagner DD, Mayadas T, Urban-Pickering M, Lewis BH, Marder VJ. Inhibition of disulfide bonding of von Willebrand protein by monensin results in small, functionally defective multimers. J Cell Biol 1985;101:112–120. 26. Mayadas TN, Wagner DD. In vitro multimerization of von Willebrand factor is triggered by low pH. Importance of the propolypeptide and free sulfhydryls. J Biol Chem 1989;264:13497–13503. 27. Chapman RE, Munro S. Retrieval of TGN proteins from the cell surface requires endosomal acidification. EMBO J 1994;13:2305–2312. 28. Mayadas T, Wagner DD, Simpson PJ. von Willebrand factor biosynthesis and partitioning between constitutive and regulated pathways of secretion after thrombin stimulation. Blood 1989;73:706–711. 29. Federici AB, Bader R, Pagani S, Colibretti ML, De Marco L, Mannucci PM. Binding of von Willebrand factor to glycoproteins Ib and IIb/IIIa complex: affinity is related to multimeric size. Br J Haematol 1989;73:93–99. 30. Sporn LA, Marder VJ, Wagner DD. Inducible secretion of large, biologically potent von Willebrand factor multimers. Cell 1986;46: 185–190. 31. Tsai HM, Nagel RL, Hatcher VB, Seaton AC, Sussman II. The high molecular weight form of endothelial cell von Willebrand factor is released by the regulated pathway. Br J Haematol 1991;79:239–245. 32. Kobayashi T, Vischer UM, Rosnoblet C, Lebrand C, Lindsay M, Parton RG, Kruithof EK, Gruenbeng J. The tetraspanin CD63/lamp3 cycles between endocytic and secretory compartments in human endothelial cells. Mol Biol Cell 2000;11:1829–1843. 33. Sporn LA, Marder VJ, Wagner DD. Differing polarity of the constitutive and regulated secretory pathways for von Willebrand factor in endothelial cells. J Cell Biol 1989;108:1283–1289. 34. van Buul-Wortelboer MF, Brinkman HJ, Reinders JH, van Aken WG, van Mourik JA. Polar secretion of von Willebrand factor by endothelial cells. Biochim Biophys Acta 1989;1011:129–133. 35. Hop C, Fontijn R, van Mourik JA, Pannekoek H. Polarity of constitutive and regulated von Willebrand factor secretion by transfected MDCK-II cells. Exp Cell Res 1997;230:352–361. 36. Camerer E, Pringle S, Skartlien AH, Wiiger M, Prydz K, Kolsto AB, Prydz H. Opposite sorting of tissue factor in human umbilical vein endothelial cells and Madin-Darby canine kidney epithelial cells. Blood 1996;88:1339–1349. 37. Voorberg J, Fontijn R, Calafat J, Janssen H, van Mourik JA, Pannekoek H. Biogenesis of von Willebrand factor-containing organelles in heterologous transfected CV-1 cells. EMBO J 1993;12:749–758. 38. Mayadas TN, Wagner DD. Vicinal cysteines in the prosequence play a role in von Willebrand factor multimer assembly. Proc Natl Acad Sci USA 1992;89:3531–3535. 39. Haberichter SL, Jozwiak MA, Rosenberg JB, Christopherson PA, Montgomery RR. The von Willebrand factor propeptide (VWFpp) traffics an unrelated protein to storage. Arterioscler Thromb Vasc Biol 2002;22:921–926. 40. Rosenberg JB, Foster PA, Kaufman RJ, Vokac EA, Moussalli M, Kroner PA, Montgomery RR. Intracellular trafficking of factor VIII to von Willebrand factor storage granules. J Clin Invest 1998;101:613–624. 41. Journet AM, Saffaripour S, Cramer EM, Tenza D, Wagner DD. von Willebrand factor storage requires intact prosequence cleavage site. Eur J Cell Biol 1993;60:31–41. 42. Haberichter SL, Jacobi P, Montgomery RR. Critical independent regions in the VWF propeptide and mature VWF that enable normal VWF storage. Blood 2003;101:1384–1391. 43. Wagner DD, Saffaripour S, Bonfanti R, Sadler JE, Cramer EM, Chapman B, Mayadas TN. Induction of specific storage organelles by von Willebrand factor propolypeptide. Cell 1991;64:403–413. 44. Michaux G, Hewlett LJ, Messenger SL, Goodeve AC, Peake IR, Daly ME, Cutler DF. Analysis of intracellular storage and regulated
Traffic 2004; 5: 69–78
45.
46.
47.
48.
49.
50. 51.
52.
53.
54.
55.
56. 57. 58.
59.
60.
61.
62.
63.
secretion of 3 von Willebrand disease-causing variants of von Willebrand factor. Blood 2003;102:2452–2458. Hop C, Guilliatt A, Daly M, de Leeuw HP, Brinkman HJ, Peake IR, van Mourik JA, Pannekoek H. Assembly of multimeric von Willebrand factor directs sorting of P-selectin. Arterioscler Thromb Vasc Biol 2000;20:1763–1768. Romani De Wit T, Rondaij MG, Hordijk PL, Voorberg J, Van Mourik JA. Real-time imaging of the dynamics and secretory behavior of WeibelPalade bodies. Arterioscler Thromb Vasc Biol 2003;23:755–761. Chanat E, Huttner WB. Milieu-induced, selective aggregation of regulated secretory proteins in the trans-Golgi network. J Cell Biol 1991;115:1505–1519. Colomer V, Kicska GA, Rindler MJ. Secretory granule content proteins and the luminal domains of granule membrane proteins aggregate in vitro at mildly acidic pH. J Biol Chem 1996;271:48–55. Arvan P, Zhang BY, Feng L, Liu M, Kuliawat R. Lumenal protein multimerization in the distal secretory pathway/secretory granules. Curr Opin Cell Biol 2002;14:448–453. Vischer UM, Wagner DD. CD63 is a component of Weibel-Palade bodies of human endothelial cells. Blood 1993;82:1184–1191. Schnyder-Candrian S, Borsig L, Moser R, Berger EG. Localization of alpha 1,3-fucosyltransferase VI in Weibel-Palade bodies of human endothelial cells. Proc Natl Acad Sci USA 2000;97:8369–8374. Denis CV, Andre P, Saffaripour S, Wagner DD. Defect in regulated secretion of P-selectin affects leukocyte recruitment in von Willebrand factor-deficient mice. Proc Natl Acad Sci USA 2001;98:4072–4077. Koedam JA, Cramer EM, Briend E, Furie B, Furie BC, Wagner DD. P-selectin, a granule membrane protein of platelets and endothelial cells, follows the regulated secretory pathway in AtT-20 cells. J Cell Biol 1992;116:617–625. Hannah MJ, Hume AN, Arribas M, Williams R, Hewlett LJ, Seabra MC, Cutler DF. Weibel-Palade bodies recruit Rab27 by a content-driven, maturation-dependent mechanism that is independent of cell type. J Cell Sci 2003;116:3939–3948. Manneville JB, Etienne-Manneville S, Skehel P, Carter T, Ogden D, Ferenczi M. Interaction of the actin cytoskeleton with microtubules regulates secretory organelle movement near the plasma membrane in human endothelial cells. J Cell Sci 2003;116:3927–3938. Munro S. Organelle identity and the targeting of peripheral membrane proteins. Curr Opin Cell Biol 2002;14:506–514. Pfeffer SR. Rab GTPases. Specifying and deciphering organelle identity and function. Trends Cell Biol 2001;11:487–491. Menasche G, Ho CH, Sanal O, Feldmann J, Tezcan I, Ersoy F, Houdusse A, Fischer A, de Saint Basile G. Griscelli syndrome restricted to hypopigmentation results from a melanophilin defect (GS3) or a MYO5A F-exon deletion (GS1). J Clin Invest 2003;112:450–456. Wolff B, Burns AR, Middleton J, Rot A. Endothelial cell ‘memory’ of inflammatory stimulation: human venular endothelial cells store interleukin 8 in Weibel-Palade bodies. J Exp Med 1998;188:1757– 62. Utgaard JO, Jahnsen FL, Bakka A, Brandtzaeg P, Haraldsen G. Rapid secretion of prestored interleukin 8 from Weibel-Palade bodies of microvascular endothelial cells. J Exp Med 1998;188:1751–1756. Romani de Wit T, de Leeuw HP, Rondaij MG, Laaf RT, Sellink E, Brinkman HJ, Voorberg J, van Mourik JA. Von Willebrand factor targets IL-8 to Weibel-Palade bodies in an endothelial cell line. Exp Cell Res 2003;286:67–74. Rosnoblet C, Vischer UM, Gerard RD, Irminger JC, Halban PA, Kruithof EK. Storage of tissue-type plasminogen activator in WeibelPalade bodies of human endothelial cells. Arterioscler Thromb Vasc Biol 1999;19:1796–1803. Pepper MS, Rosnoblet C, Di Sanza C, Kruithof EK. Synergistic induction of t-PA by vascular endothelial growth factor and basic fibroblast growth factor and localization of t-PA to Weibel-Palade
77
Michaux and Cutler
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
78
bodies in bovine microvascular endothelial cells. Thromb Haemost 2001;86:702–709. Huber D, Cramer EM, Kaufmann JE, Meda P, Masse JM, Kruithof EK, Vischer UM. Tissue-type plasminogen activator (t-PA) is stored in Weibel-Palade bodies in human endothelial cells both in vitro and in vivo. Blood 2002;99:3637–3645. Datta YH, Youssoufian H, Marks PW, Ewenstein BM. Targeting of a heterologous protein to a regulated secretion pathway in cultured endothelial cells. Blood 1999;94:2696–2703. Emeis JJ, van den Eijnden-Schrauwen Y, van den Hoogen CM, Priester W, Westmuckett A, Lupu F. An endothelial storage granule for tissue-type plasminogen activator. J Cell Biol 1997;139:245–256. Schick BP, Gradowski JF, San Antonio JD. Synthesis, secretion, and subcellular localization of serglycin proteoglycan in human endothelial cells. Blood 2001;97:449–458. Russell FD, Skepper JN, Davenport AP. Evidence using immunoelectron microscopy for regulated and constitutive pathways in the transport and release of endothelin. J Cardiovasc Pharmacol 1998;31:424–430. Do H, Healey JF, Waller EK, Lollar P. Expression of factor VIII by murine liver sinusoidal endothelial cells. J Biol Chem 1999;274: 19587–19592. Hollestelle MJ, Thinnes T, Crain K, Stiko A, Kruijt JK, van Berkel TJ, Loskutoff D, van Mourik JA. Tissue distribution of factor VIII gene expression in vivo – a closer look. Thromb Haemost 2001;86:855–861. Rosenberg JB, Greengard JS, Montgomery RR. Genetic induction of a releasable pool of factor VIII in human endothelial cells. Arterioscler Thromb Vasc Biol 2000;20:2689–2695. Shi Q, Wilcox DA, Fahs SA, Kroner PA, Montgomery RR. Expression of human factor VIII under control of the platelet-specific alphaIIb promoter in megakaryocytic cell line as well as storage together with VWF. Mol Genet Metab 2003;79:25–33. Mannucci PM, Pareti FI, Holmberg L, Nilsson IM, Ruggeri ZM. Studies on the prolonged bleeding time in von Willebrand’s disease. J Lab Clin Med 1976;88:662–671.
74. Castaman G, Eikenboom JC, Missiaglia E, Rodeghiero F. Autosomal dominant type 1 von Willebrand disease due to G3639T mutation (C1130F) in exon 26 of von Willebrand factor gene: description of five Italian families and evidence for a founder effect. Br J Haematol 2000;108:876–879. 75. Casana P, Martinez F, Haya S, Espinos C, Aznar JA. Association of the 3467C>T mutation (T1156M) in the von Willebrand’s factor gene with dominant type 1 von Willebrand’s disease. Ann Hematol 2001;80:381–383. 76. Eikenboom JC, Matsushita T, Reitsma PH, Tuley EA, Castaman G, Briet E, Sadler JE. Dominant type 1 von Willebrand disease caused by mutated cysteine residues in the D3 domain of von Willebrand factor. Blood 1996;88:2433–2441. 77. O’Brien LA, James PD, Othman M, Berber E, Cameron C, Notley CR, Hegadorn CA, Sutherland JJ, Hough C, Rivard GE, O’Shaunessy D, Lillicrap D. Association of Hemophilia Clinic Directors of Canada. Founder von Willebrand factor haplotype associated with type 1 von Willebrand disease. Blood 2003;102:549–557. 78. Bodo I, Katsumi A, Tuley EA, Eikenboom JC, Dong Z, Sadler JE. Type 1 von Willebrand disease mutation Cys1149Arg causes intracellular retention and degradation of heterodimers: a possible general mechanism for dominant mutations of oligomeric proteins. Blood 2001;98: 2973–2979. 79. Mohlke KL, Purkayastha AA, Westrick RJ, Smith PL, Petryniak B, Lowe JB, Ginsburg D. Mvwf, a dominant modifier of murine von Willebrand factor, results from altered lineage-specific expression of a glycosyltransferase. Cell 1999;96:111–120. 80. Allen S, Goodeve AC, Peake IR, Daly ME. Endoplasmic reticulum retention and prolonged association of a von Willebrand’s diseasecausing von Willebrand factor variant with ERp57 and calnexin. Biochem Biophys Res Commun 2001;280:448–453. 81. Haberichter SL, Fahs SA, Jozwiak MA, Retzlaff KL, Montgomery RR. The sorting signal for von Willebrand factor granular storage lies within the first 140 amino acids of the D2 domain of the VWF propeptide (Abstract). J Thromb Haemost 2003;1 Supplement 1 July: OC 401.
Traffic 2004; 5: 69–78
