Skip Navigation

Cardiovascular Research 2003 58(1):20-31; doi:10.1016/S0008-6363(02)00852-0
© 2003 by European Society of Cardiology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by van den Driesche, S.
Right arrow Articles by Westermann, C. J.J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by van den Driesche, S.
Right arrow Articles by Westermann, C. J.J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 2003, European Society of Cardiology

Hereditary hemorrhagic telangiectasia: an update on transforming growth factor β signaling in vasculogenesis and angiogenesis

Sander van den Drieschea, Christine L. Mummerya,b,* and Cornelius J.J. Westermannc

aHubrecht Laboratory, Netherlands Institute for Developmental Biology, Uppsalalaan 8, 3584 CT Utrecht, The Netherlands
bThe Interuniversity Cardiology Institute of the Netherlands, Utrecht, The Netherlands
cSt. Antonius Hospital, Department of Pulmonary Disease, Nieuwegein, The Netherlands

* Corresponding author. Tel.: +31-30-212-1800; fax: +31-30-251-6464. christin{at}niob.knaw.nl

Received 5 November 2002; accepted 12 December 2002


   Abstract
Top
 Abstract
1 Introduction
 2 TGFβ signaling
 3 Endoglin and HHT...
 4 ALK-1 and HHT...
 5 Other HHT loci
 6 HHT models in...
 7 Vascular phenotypes in...
 8 Primary pulmonary hypertension...
 9 Other (cardio-)vascular...
 10 Conclusions
 References
 
Hereditary hemorrhagic telangiectasia (HHT) is a vascular disorder in humans which has been mapped to two genes, endoglin and activin receptor-like kinase-1 (ALK-1) both of which mediate signaling by transforming growth factor β ligands in vascular endothelial cells. Animal models have shown that these receptors are not only important for maintaining vascular integrity but also for angiogenesis both during embryonic development and during tumor growth. Here, we review the current status of reported mutations in the context of the clinical manifestations and the effects on the vessel wall both in patients and in animal models of the disease.

KEYWORDS Angiogenesis; Growth factors; Receptors; Sequence (DNA/RNA/prot)


   1 Introduction
Top
 Abstract
 1 Introduction
2 TGFβ signaling
 3 Endoglin and HHT...
 4 ALK-1 and HHT...
 5 Other HHT loci
 6 HHT models in...
 7 Vascular phenotypes in...
 8 Primary pulmonary hypertension...
 9 Other (cardio-)vascular...
 10 Conclusions
 References
 
Hereditary hemorrhagic telangiectasia in humans, also known as HHT or Rendu–Osler–Weber syndrome, is an autosomal dominant vascular disorder, which affects ~1:10,000 individuals. HHT patients develop frequent bleedings with increasing age, in particular in nasal, gastrointestinal and cerebral vascular beds. The vascular arteriovenous malformations (AVM) vary in size from 1 mm to several centimetres. Pulmonary AVMs are particularly life threatening because they may hemorrhage or may lead to paradoxical embolism causing brain infarction or brain abscess. HHT is a heterogeneous disease, both between families and among members of a single family. Genetic linkage studies have mapped HHT to chromosome 9q33–34 in some families (HHT type 1) and 12q13 in others (HHT type 2). The mutated genes at these loci have been identified as endoglin and activin receptor-like kinase-1 (ALK-1), respectively. Expression of these genes is mainly restricted to vascular endothelial cells. They mediate binding and signaling of transforming growth factor β (TGFβ) and co-exist in these cells with the more ubiquitously expressed TGFβ receptors for this ligand family. Despite increasing understanding of the molecular regulation of vasculogenesis and angiogenesis, and the role taken by TGFβ in its control, it is still not clear why clinical manifestations of the disease are so variable, why the mutations lead to weak vascular walls in some but not all tissues and why some mutations are not pathogenic at all. There have been some recent suggestions, however, to explain why other TGFβ receptors expressed in endothelial cells do not rescue the phenotype and these are pertinent to our understanding of the disease. Here, we will review recent data on TGFβ signaling in endothelial cells, which are largely based on experiments in mice bearing mutations in TGFβ receptors. Further, we will compare HHT with other vascular disorders in humans and discuss genetic mutations that cause vascular phenotypes in both mice and zebrafish. We will thus provide the clinical geneticist and general physician with a basis for understanding present and future research in this area that may have important implications for patients with HHT.


   2 TGFβ signaling
Top
 Abstract
 1 Introduction
 2 TGFβ signaling
3 Endoglin and HHT...
 4 ALK-1 and HHT...
 5 Other HHT loci
 6 HHT models in...
 7 Vascular phenotypes in...
 8 Primary pulmonary hypertension...
 9 Other (cardio-)vascular...
 10 Conclusions
 References
 
It is clear that HHT is caused by defective TGFβ signaling in vascular endothelial cells. An understanding of how this signaling takes place is thus essential if the implications for the disease are to be fully appreciated. TGFβ family members include TGFβs, three of which are known in mammals, activins and bone morphogenetic proteins (BMPs). They are structurally related, secreted cytokines that regulate a range of cellular functions from proliferation, apoptosis, migration, extracellular matrix production and deposition and differentiation. Cellular responses are elicited by the formation of heteromeric complexes of type I and II serine/threonine kinase receptors, initiated by binding of ligand to type II receptors. Five type II receptors are known (TβRII, BMPR-II, ActR-IIA, ActR-IIB and AMHRII), each generally with different ligand specificities, and seven type I receptors, also known as activin receptor-like kinases (ALKs1–7). Upon ligand binding, the constitutively active type II receptor phosphorylates particular serine and threonine residues in the type I receptor which becomes activated and transduces signals downstream [1]. Type I receptors are fairly promiscuous and associate with different type II receptors depending on the ligand bound (Fig. 1). Downstream targets of the ALKs that are pivotal in mediating signaling to the nucleus are the Smad proteins; these can be divided into three distinct subfamilies: the receptor-regulated Smads (or R-Smads), the common partner Smads (or Co-Smads) and the inhibitory Smads (or I-Smads) [2]. Activated ALKs specifically and transiently interact with and phosphorylate particular R-Smads. The R-Smads can be further divided into two groups: the BMP-Smads (Smads 1, 5 and 8), which are phosphorylated by the BMP type I receptors (ALK-2, ALK-3 and ALK-6) and the TGFβ/activin Smads (Smads 2 and 3), activated by the TGFβ and Activin type I receptors (ALK-5 and ALK-4, respectively). After phosphorylation by the appropriate type I receptors, activated R-Smads form a complex with Co-Smads and subsequently translocate and accumulate in the nucleus. Exceptionally, phosphorylation of tyrosine residues in the linker region of R-Smads by activated tyrosine-kinase receptors (such as the fibroblast growth factor receptor), or activated ras inhibit this translocation to the nucleus [3]. Nuclear Smad complexes can bind to DNA directly or indirectly through other DNA-binding proteins, and regulate transcription of target genes [4,5]. An additional (type III or β-glycan) receptor may be present on a number of cell types. This receptor type has a very short cytoplasmic tail without a signaling domain [6]. The function of this receptor is to facilitate binding of the prototype TGFβs (TGFβ1, -2 and -3) to TβRII; in fact TGFβ2 will not bind to TβRII at all in the absence of TβRIII.


Figure 1
View larger version (51K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 TGFβ signal transduction. TGFβ family members signal through distinct sets of type I and type II receptors to phosphorylate receptor regulated Smads (R-Smads). Activated R-Smads bind to common Smads (Co-Smads), one of which (Smad4) has been identified in mammals. This complex translocates to the nucleus to regulate transcription. Smads 6 and 7 are inhibitory (or I-) Smads and have a negative effect on the pathway. Type III receptors form complexes with type II receptors and may affect ligand binding affinity. TGFβs can also activate the MAP kinase/Erk pathway via ALK-5. There is no evidence that ALK-1/endoglin activates the MAP kinase/Erk pathway in endothelial cells. Its possible role in HHT is thus presently beyond the scope of the present review. *Receptors capable of binding endoglin [42]. TGFβ, transforming growth factor β; BMP, bone morphogenetic protein; ActR, activin receptor; TβRII, TGFβ receptor II; BMPR, BMP receptor; ALK, activin receptor-like kinase; R-Smads, receptor-regulated Smads; Co-Smads, common partner Smads; I-Smads, inhibitory Smads.

 
Apart from these widely expressed receptors for TGFβ (ALK-5, TβRII and TβRIII) endothelial cells exceptionally also express two other TGFβ receptors, endoglin (a type III receptor resembling β-glycan) and ALK-1. It is these receptors that are mutated in HHT. It is probably because of their highly restricted tissue expression that patients with mutations in these genes survive to birth and adulthood. Instead of phosphorylating the conventional TGFβ receptor Smads 2 and 3, ALK-1 appears to phosphorylate the typically BMP receptor activated Smads 1, 5 and 8 [7,8]. Thus in endothelial cells, in contrast to most if not all other cells, TGFβ activates the BMP-Smads, besides the conventional TGFβ receptor Smads 2 and 3, and therefore activates transcription of their specific target genes. Although there is no evidence so far that the MAP kinase/Erk pathway activated via ALK-5 in many cell types [79] is also activated via ALK-1/endoglin in endothelial cells, the involvement of pathways other than those mediated by Smads in causing the aberrant behavior of endothelial cells, cannot yet be excluded.


   3 Endoglin and HHT type I
Top
 Abstract
 1 Introduction
 2 TGFβ signaling
 3 Endoglin and HHT...
4 ALK-1 and HHT...
 5 Other HHT loci
 6 HHT models in...
 7 Vascular phenotypes in...
 8 Primary pulmonary hypertension...
 9 Other (cardio-)vascular...
 10 Conclusions
 References
 
The first genetic linkage studies on DNA from HHT families mapped the mutations to markers on chromosome 9q33–34 [9,10]. In the same year, McAllister et al. identified the gene responsible as the endoglin gene. Since then, many more mutations have been discovered, most of which appeared to be family specific, although exceptions have been reported [11–22,55]. Table 1 provides a list of all mutations available in the literature. These mutations have all been located in the part of the gene that encodes the extracellular domain of the endoglin protein. The mutations include deletions and insertions, missense mutations (single amino acid changes), splice site changes and nonsense mutations (point mutations that lead to premature stop codons) [23].


View this table:
[in this window]
[in a new window]

  		Table 1 Summary of known endoglin mutations

 
Endoglin is expressed as a disulfide-linked dimer on endothelial cells [24,25]. Missense mutations predicted to have an effect on the extracellular domain were thought to associate negatively with the normal endoglin protein on the cell surface. A dominant-negative model was proposed, whereby overexpression of a mutant polypeptide ‘squelches’ activity of the endogenous gene. In the case of endoglin, a truncated protein would interfere with TGFβ signaling either by binding to normal endoglin or by being secreted and binding to and sequestering extracellular TGFβ [12]. However, overexpression of mutated endoglin proteins with or without co-transfection of normal endoglin in COS-1 cells revealed intracellular expression, and an inability to be secreted or form heterodimers at the cell surface [13,18,21]. Thus mutated forms of endoglin are transiently expressed intracellularly and are not likely to act as dominant negative proteins. This suggested that the majority of mutations represent null alleles where unstable message or protein production from the mutant allele results in a 50% reduction in the levels of protein. This is termed haploinsufficiency, and would lead to levels of endoglin protein being reduced at the surface of vascular endothelial cells. This has been proposed as the predominant mechanism underlying the HHT1 phenotype [13,14,18,21].

Another argument for this haploinsufficiency model comes from observations that several different mutations lead to the same HHT1 phenotype. If a dominant-negative mechanism accounted for HHT, different mutations might be expected to be associated with different clinical phenotypes, which is not the case [14].

In another study the expression of six different missense mutations and two truncation mutations were investigated [26]. Expression of the missense mutants alone revealed that they are misfolded and that most show no cell surface expression. When the missense mutants were co-expressed with wild-type endoglin, the missense mutants were able to dimerize with the normal endoglin protein and the mutants were trafficked to the cell surface. In case of the two truncation mutants studied, one acts through haploinsufficiency, but the other ({Delta}GC in exon 11) at least acts in a dominant negative way. Therefore, either dominant-negative protein interactions or haploinsufficiency can cause HHT1, and it may be that the individual nature of the mutation determines the pathogenesis.

Bourdeau et al. [19] showed in two case studies of HHT1 patients that relative endoglin levels were reduced by half, indicating that only the normal endoglin allele was expressed on endothelium in situ. Analyses of normal, non-affected vessels (without morphological abnormalities) in the two patients showed that the endoglin/PECAM-1 ratio was reduced by up to 50%, when compared with age-matched controls. A similar ratio was observed in the cerebral AVM and pulmonary AVM, suggesting that all blood vessels of HHT1 patients express reduced endoglin in situ and that AVMs are not due to a focal loss of endoglin in the lesions. Furthermore, these data also exclude a second hit hypothesis, in which the general idea is that the second, normal allele of endoglin has to be inactivated in order to have a lesion as seen in HHT1 patients. Thus, reduced levels of endoglin, revealed by in vitro analyses in patients, again support a model of haploinsufficiency for HHT1, in which the endoglin mutations operate as null alleles [19].


   4 ALK-1 and HHT type 2
Top
 Abstract
 1 Introduction
 2 TGFβ signaling
 3 Endoglin and HHT...
 4 ALK-1 and HHT...
5 Other HHT loci
 6 HHT models in...
 7 Vascular phenotypes in...
 8 Primary pulmonary hypertension...
 9 Other (cardio-)vascular...
 10 Conclusions
 References
 
Through the linkage analysis to map the HHT locus, it became clear that some families were excluded from chromosome 9q33–34, and that there was therefore locus heterogeneity [10,27–29]. These 9q33–34 unlinked (or HHT2) families showed a lower incidence of PAVMs [14,27–31], suggesting that the HHT1 gene (endoglin) had an additional role in the pulmonary vasculature not shared by HHT2. However, McDonald et al. [32] reported PAVMs in 6% of affected HHT2 kindred members, suggesting that this pulmonary lesion is just less common in HHT2 patients. Later studies of Kjeldsen et al. [33] and Trembath et al. [34] documented HHT2 families with a higher prevalence of PAVM. On the other hand, Westermann et al. (unpublished) have studied the incidence of PAVMs in the Dutch patient population, and found a significantly higher incidence of PAVMs in HHT1 patients compared to HHT2 patients. It would therefore still be advisable to screen for endoglin mutations in HHT patients of unknown genotype with PAVMs.

This second locus for HHT was subsequently identified on chromosome 12q13 [30,35]. A potential candidate gene was activin receptor-like kinase-1 (ALK-1) and Johnson et al. identified the first mutations in this gene [36]. Since then, more mutations in ALK-1 have been found (Table 2), and include small deletions, insertions and nonsense mutations leading to truncated proteins, as well as missense mutations [33,34,36–41,45,77].


View this table:
[in this window]
[in a new window]

  		Table 2 Summary of known ALK-1 mutations

 
Expression analysis suggested that some of the mutations were null alleles because of apparent instability of the mutant transcript, again supporting a haploinsufficiency model [37,39]. However, no single mutation was a clear example of a true null allele [38], in contrast to endoglin where a mutation destroying the start codon had been found [16]. As for endoglin, the heterogeneity of ALK-1 mutations and their distribution throughout the gene as well as missense mutations in amino acids that are conserved in type I receptors of the TGFβ superfamily, all argue in favor of haploinsufficiency as a model for HHT2 [41].

Abdalla et al. [39] described a novel polyclonal antibody (pAb) against ALK-1. Using this pAb {alpha}-ALK-1 they showed reduced ALK-1 protein levels in HUVEC samples of HHT2 families with ALK-1 missense mutations. This lack of surface expression of the mutant proteins makes it likely that these mutations lead to structural alterations resulting in protein misfolding and intracellular degradation. The inheritance of a single mutant copy would then predispose the individual to develop the vascular lesions observed in HHT. Lesion formation itself could be due to local environmental or mechanical effects at the lumen of the vessel or to additional genetic alterations.

McDonald et al. [32] studied 38 HHT2 patients in single kindred, all with the same identified disease-causing mutation (G998T). They showed variability in clinical symptoms which could not be attributed solely to either allelic or locus heterogeneity. The formation of telangiectases and AVMs clearly depends on factors additional to the dominantly inherited mutation [32]. Others report variability of symptoms even within the same family, also suggesting that factors other than specific mutations influence the clinical manifestations [28,33]. The phenotype of the HHT mice, described below, further supports this.

Diagnosis of HHT2 patients has often been difficult, because of lower penetrance or expression levels, a milder phenotype compared with HHT1 [35,36], and a tendency towards later onset [27,28,31]. A possible explanation might be that endoglin acts more upstream in the TGFβ pathway in vascular endothelial cells than ALK-1 [8]. Endoglin may also be involved in more pathways, since it has been shown to bind TGFβ1, TGFβ3, Activin A, and BMP-2 and -7 via association with their respective ligand-binding receptors [42]. By contrast, McDonald et al. [32] reported that the age-of-onset and severity of epistaxis (nosebleeds) in affected HHT2 kindred members do not suggest a milder phenotype for HHT2 than HHT1 with respect to this most common symptom.


   5 Other HHT loci
Top
 Abstract
 1 Introduction
 2 TGFβ signaling
 3 Endoglin and HHT...
 4 ALK-1 and HHT...
 5 Other HHT loci
6 HHT models in...
 7 Vascular phenotypes in...
 8 Primary pulmonary hypertension...
 9 Other (cardio-)vascular...
 10 Conclusions
 References
 
A third rare variant of HHT has been reported in one large family with hepatic involvement as the major manifestation, with exclusion of linkage to both chromosome 9 and chromosome 12 [43,44]. A recent paper of Olivieri et al. [45], however, reported a new ALK-1 mutation (C199T) in the same family. A different missense mutation (G200A), at the same amino acid (Arg67), was reported by Berg et al. [37] and Trembath et al. [34], confirming that the substitution of this arginine causes HHT2. Therefore, it is very likely that the family studied by Piantanida et al. [43], Buscarini et al. [44], and Olivieri et al. [45] has HHT2, and not an unknown third HHT variant.

A more recent report also describes another HHT family apparently not linked to endoglin and ALK-1 [46]. In contrast to the family described above, however, the incidence of PAVMs in this family is as in HHT1 families, with no obvious higher incidence of hepatic involvement. To date, the exact locus and mutated gene have not yet been identified.


   6 HHT models in mice and zebrafish
Top
 Abstract
 1 Introduction
 2 TGFβ signaling
 3 Endoglin and HHT...
 4 ALK-1 and HHT...
 5 Other HHT loci
 6 HHT models in...
7 Vascular phenotypes in...
 8 Primary pulmonary hypertension...
 9 Other (cardio-)vascular...
 10 Conclusions
 References
 
Once endoglin and ALK-1 had been identified as causally involved in HHT, various research groups started to create mouse models deficient in these genes. Mice lacking either endoglin or ALK-1 die during embryonic development (null mice) or have vascular phenotypes as heterozygotes, at least on some genetic backgrounds. The results of the studies in these mice suggested the involvement of endoglin and ALK-1 in vascular development as well as in maintaining vascular integrity in adults.

Endoglin-null mice (Eng–/–) die with cardiovascular defects early in gestation with most obvious defects in the yolk sac vasculature [47–49]. In contrast to mice lacking TGFβ1 [50] and TGFβ receptors I and II [51,52] where vasculature of the yolk sac is also highly abnormal, vasculogenesis in the embryo proper was apparently unaffected by the absence of endoglin: immunostaining for platelet-endothelial cell adhesion marker (PECAM) revealed the persistence of an immature perineural vascular plexus indicating a failure of endothelial cell remodeling rather than effects on earlier events [47]. Bourdeau et al. [48] confirmed that in situ differentiation of endothelial cells from mesodermally derived precursors, their assembly into the primary capillary network of yolk sac, placenta, and embryo, and the generation of primitive vessels connected to the primordia of the heart take place in the absence of endoglin. These are early steps of vasculogenesis [53] which all appear to be normal, suggesting that endoglin in fact plays a role in angiogenesis, the sprouting of new vessels from existing vasculature [47,48] rather than vasculogenesis, the de novo formation of vessels.

Furthermore, the expression of endothelial markers such as Flk-1, Flt-1, and Tie2 and hematopoietic markers Gata-1 and Il-3r was unaffected in Eng–/– mice [47]. Thus, in contrast to TGFβ1 and TGFβ receptors I and II, there is no evidence that endoglin is required for endothelial cell differentiation or primitive hematopoiesis [47,50–52].

Immunohistochemical staining for {alpha}-smooth muscle cell actin, a marker for vascular smooth muscle cells (vsmc), however, show significant differences in vsmc development of Eng+/+ and Eng–/– embryos, which was apparent by embryonic day (E) 9.5. This preceded the differences in endothelial organization observed between E9.5 and E10.5 [47]. At E9.5 the vasculature of Eng–/– yolk sacs failed to organize, as described above; this was even more evident at E10.5. A vsmc defect in the yolk sac of Eng–/– mice was first evident by E8.5 and, as in the embryonic tissue, preceded the defect in endothelial remodeling [47]. In situ hybridization of yolk sacs and embryos at E9.5 using an RNA probe for the vsmc marker SM22{alpha} showed an absence of SM22{alpha} expression in Eng–/– embryos and yolk sacs [47]. Therefore, confirming that endoglin is required for the differential growth and sprouting of endothelial tubes but also showing that it is necessary for the recruitment and differentiation of mesenchymal cells into vsmc and pericytes [47], both processes involved in angiogenesis [53].

There are, however, differences between the Eng–/– mice of Bourdeau et al. [48] and Arthur et al. [49] on the one hand, and Li et al. [47] on the other. Li et al. [47] observed that Eng–/– embryos died in utero at E11.5 and that no large vessels were observed in the yolk sac. In contrast, Bourdeau et al. [48] and Arthur et al. [49] observed that both yolk sac and heart defects were observed simultaneously at E9.0–9.5, and all mice died by E10.0–10.5. In all cases the mice were crossed into the C57BL/6 strain, but the embryonic stem (ES) cells were of 129/SVJ origin in the study of Li et al. [47] and of 129/Ola origin in the other studies. Thus, the relative contributions for different genetic backgrounds might contribute to the differences in phenotype observed in the three studies.

Although on a C57BL/6 genetic background, heterozygous endoglin mice exhibit phenotypes only rarely, on 129/Ola or mixed C57BL/6–129/Ola genetic backgrounds [48,49] Eng+/– mice do develop signs of HHT. Bourdeau et al. [54] reported a disease prevalence of 72% in 129/Ola strain (high), 36% in Ola/129–C57BL/6 intercrosses (intermediate), and 7% in C57BL/6 backcrosses (low). Disease manifestations in these HHT mice include telangiectases on the ears, neck, tail, and genitals, recurrent nosebleeds, extensive dilated and weak-walled vessels, and gastrointestinal, pulmonary, hepatic, and cerebral bleedings [48,49,54,55]. Disease severity in HHT mice generally increased with age. Several older mice (>8 months) developed severe bleeding, marked weight loss and often reduced mobility [55]. Thus, disease manifestations were observed in some mice expressing a single copy of endoglin, confirming that HHT1 is associated with a loss of function of the mutated allele and favoring the haploinsufficiency model, as described above.

Since the development of HHT-like symptoms in the endoglin heterozygotes is dependent on the genetic background, modifier genes may contribute to the development of the vascular abnormalities both in mice and in HHT patients. Furthermore, not all mice of the susceptible strain developed the disease, suggesting that epigenetic factors such as shear stress, environmental conditions, blood pressure, oxygenation, and hormonal levels are also involved in the generation of clinical manifestations [48,49,54,55]. Of interest is the observation of telangiectases in HHT patients on areas of skin exposed to sunlight/DNA damage. This is not the case in African people with pigmented skin. Thus, because a large proportion of the heterozygous endoglin mice do not display clinical signs even after 1 year [54], a single endoglin mutant allele is apparently necessary but not sufficient to cause disease.

It is of note that the Ola/129 mouse strain has significantly lower levels of circulating latent TGFβ1 than the C57BL/6 strain. Furthermore, intercrosses of 129/Ola–C57BL/6 mice and 129/Ola mice heterozygous for endoglin showed a further reduction of plasma TGFβ1 levels compared with wild-type littermate controls. This suggests that modifier genes involved in the regulation of TGFβ1 expression may also act in combination with a single functional copy of endoglin in the development of HHT [54].

In HHT patients, the variability in the expression of the disease, with a wide variance of clinical features even among members of the same family, indicates that the inherited mutation alone does not determine the individual phenotype but shows that other factors are involved [26,54]. The HHT mouse can thus be used as a model to look for these additional factors contributing to the disease. The identification of modifier genes in mice for example should lead to the recognition of their counterparts in humans. HHT mice might also be useful for testing novel therapeutic modalities with important implications for HHT patients.

ALK-1 null embryos have superficially a similar phenotype as Eng–/– embryos [8,56] although by contrast, the vessels are not dilated. This suggests that endoglin and ALK-1 are indeed critical for vascular development and that their function is likely related to an altered response to TGFβ1. Endothelial cells, however, also express ALK-5. The observed lack of functional redundancy of ALK-1 and ALK-5 in endothelial cells of knockout mice was originally explained by the suggestion that they were simply acting as low and high affinity receptors, respectively, and each activated a different Smad pathway. However, this view has recently been challenged. The migratory response of endothelial cells to TGFβ is actually biphasic, with ALK-1 mediating a response to low doses of TGFβ, stimulating migration, and ALK-5 mediating a response to high doses, inhibiting migration [57]. The TGFβ/ALK-1 pathway thus primarily mediates (Smad-dependent) positive effects on migration via Id1, while plasminogen activator induction by ALK-5 inhibits migration and proliferation and stimulates extracellular matrix deposition. ALK-5 thus contributes to TGFβ-induced maturation of blood vessels and it seems likely that the fine balance between ALK-5 and ALK-1 signaling regulates the activation state of the endothelium.

Recently, mutant zebrafish named violet beauregarde (vbg) have been described with a mutation in ALK-1 [58]. In homozygous vbg mutants, most blood flow is confined to a small number of dilated cranial vessels that contain more than twice as many endothelial cells as their wild-type counterparts. In the affected vessels, the vascular smooth muscle marker sm22{alpha} is normally expressed 2 days post-fertilization, a stage in which the zebrafish embryo has developed a complete activated circulation system. The ALK-1 mutation-induced increase in endothelial cell number and vessel caliber are thus independent of vascular smooth muscle deficits in zebrafish. The heterozygous vbg fish also do not have telangiectases or hemorrhages, which might be due to the absence of modifying factors as in the endoglin heterozygous mice. Nevertheless, since ALK-1 ligand(s) and downstream targets are presently unknown or poorly defined, the embryological, molecular and molecular genetic techniques afforded by the zebrafish system make the vbg mutant a valuable model system for dissecting the molecular pathway by which ALK-1 directs embryonic vessel formation; this could in turn lead to insight into the molecular mechanisms responsible for HHT2 pathogenesis.


   7 Vascular phenotypes in other mutant mice
Top
 Abstract
 1 Introduction
 2 TGFβ signaling
 3 Endoglin and HHT...
 4 ALK-1 and HHT...
 5 Other HHT loci
 6 HHT models in...
 7 Vascular phenotypes in...
8 Primary pulmonary hypertension...
 9 Other (cardio-)vascular...
 10 Conclusions
 References
 
We have recently reviewed mutant mice exhibiting vascular phenotypes to rank the hierarchy of signals that result in the formation of vasculature during development and disease [53]. Although fibroblast growth factors have been implicated in the very first event, mesoderm induction to angioblasts, these studies were carried out in chick and frog; defective angioblast formation is not a feature of mice lacking FGFs or their receptors. Vascular endothelial growth factors (VEGFs) are probably the most critical drivers of vascular formation both in development and the adult; they regulate in situ differentiation of angioblasts into endothelial cells, their proliferation and fusion into tubes with lumina and organization into a primitive plexus (vasculogenesis) as well as later angiogenic sprouting. Subsequent remodeling and maturation of this primitive plexus by sprouting and non-sprouting angiogenesis depends on the coordinated action of VEGFs, the angiopoietin family member Ang1 and ephrin-B2. Ephrin-B2 is particularly important in distinguishing arterial and venous vessels. Following vessel maturation, Ang1 continues to be important in maintaining the quiescence and stability of the mature vasculature; disruption of this in the adult coincides with reinitiation of vascular remodeling, for example in the adult female reproductive system or in tumors, and involves autocrine induction of Ang2, an antagonist of Ang1, in the endothelium to be remodeled. VEGFs, angiopoietins and ephrin B2 apparently recapitulate their developmental roles during vascular remodeling in the adult but are unable to trigger the whole process. For example, VEGF can initiate vessel formation in adults but the vessels that form are weak. This can be counteracted by application of Ang1. This stabilization and activation, known as the activation phase of angiogenesis, results in increased vascular permeability and basement membrane degradation and is essential for the release of endothelial cells in the extracellular space. Here, they can proliferate and migrate to form new capillary sprouts. ALK-1 is thought to play a role in this process. In the second phase of angiogenesis, known as the resolution phase, endothelial cells cease proliferation and migration, reform the basement membrane and mesenchymal cells are recruited that eventually contribute to the robustness of the vessel wall. It is here that TGFβs are thought to act through TβRII and ALK-5 and a set of specific downstream target genes to inhibit endothelial cell proliferation, stimulating extracellular matrix production and inducing the differentiation of mesenchymal cells to pericytes and smooth muscle cells, as described earlier. Deletion of several transcription factors and secreted molecules in mice, regarded as candidate or proven TGFβ targets, has been shown to result in vascular phenotypes in development highly reminiscent of those induced by TGFβ receptor deletion [53]. For example, the vasculatures of embryos lacking ephrins, fibronectin and Id1/Id3 are very similar to those of embryos lacking components of the TGFβ signaling pathway. Analyses like these in mice are likely to provide important information on why vessels in HHT patients are weak and/or fail to repair. In the long term this may result in novel treatment strategies.

Thus although multiple growth factors play essential roles in the development of the vascular system via vasculogenesis and angiogenesis (see Table 3), only TGFβ (and possibly BMP) signaling appears to be affected in HHT patients, specifically affecting their ability to remodel or repair the vasculature.


View this table:
[in this window]
[in a new window]

  		Table 3 Role of various growth factors in vasculogenesis and angiogenesis. This has been reviewed in detail in Ref. [53]

 

   8 Primary pulmonary hypertension and HHT
Top
 Abstract
 1 Introduction
 2 TGFβ signaling
 3 Endoglin and HHT...
 4 ALK-1 and HHT...
 5 Other HHT loci
 6 HHT models in...
 7 Vascular phenotypes in...
 8 Primary pulmonary hypertension...
9 Other (cardio-)vascular...
 10 Conclusions
 References
 
Primary pulmonary hypertension (PPH) is a rare, fatal disorder that involves uncontrolled remodeling of and plexiform lesions in the pulmonary arteries. Typically, the disorder presents as an insidious shortness of breath in mid-life, caused by marked pulmonary arterial hypertension and hypoxaemia [59,60]. The disease is most commonly sporadic, but may also be associated with an autosomal-dominant mode of inheritance (FPPH for familial primary pulmonary hypertension) in 6% of patients [61]. FPPH has an incidence of 1 to 2 cases per million people per year [62]. Its pattern of inheritance appears to be autosomal dominant, and the gene mutated is bone morphogenetic protein receptor II (BMPR-II, see Fig. 1; [59,63–65]). BMP signaling in endothelial cells, like TGFβ activated ALK-1 signaling, is mediated by phosphorylation of Smads 1, 5 or 8 ([66]; Fig. 1) which may explain why some clinical manifestations of PPH and HHT are similar.

Trembath et al. studied rare families of HHT patients who also developed pulmonary hypertension [34]. The pathological features consisted of dilated blood vessels and AVMs characteristic of HHT, but also the occlusion of small pulmonary arteries characteristic of PPH. Some members of a particular family were identified with mutations in ALK-1, while other members had a mutation in BMPR-II. This study indicates that PPH may occur in patients with HHT2, although the histological and pathophysiological features of HHT and PPH seem to be distinct. PAVMs in HHT lead to decreased pulmonary vascular resistance and increased cardiac output, with normal to low pulmonary arterial pressure [67,68]. In contrast, PPH is characterized by obliteration of small pulmonary arteries, leading to increased pulmonary vascular resistance, marked elevation of pulmonary arterial pressure, and ultimately, a reduction in cardiac output [34,69].

Nevertheless, the observation that mutations in two different, but mechanistically related, TGFβ superfamily receptors (i.e. ALK-1 and BMPR-II) can produce the same clinical phenotype suggests that these two receptors may interact directly to modulate vascular-cell growth [70]. However, to date there is no evidence of a direct interaction between ALK-1 and BMPR-II, except their common ability to activate Smads 1, 5 and 8 downstream.

As explained above, pulmonary involvement is always considered to occur in HHT1 and to a lesser extent in HHT2. One implication of the study by Trembath et al. [34] is that pulmonary as well as systemic vascular malformations should be considered in all patients with HHT, irrespective of genotype. Similarly, the possibility of HHT should be considered in any patient who presents with unexplained pulmonary hypertension.


   9 Other (cardio-)vascular diseases in humans and the role of endothelial progenitor cells in vascular repair
Top
 Abstract
 1 Introduction
 2 TGFβ signaling
 3 Endoglin and HHT...
 4 ALK-1 and HHT...
 5 Other HHT loci
 6 HHT models in...
 7 Vascular phenotypes in...
 8 Primary pulmonary hypertension...
 9 Other (cardio-)vascular...
10 Conclusions
 References
 
Why vessels with mutations in ALK-1 and endoglin fail to repair is yet unclear; the ‘classic’ responses of endothelial cells from HHT patients to TGFβ (growth inhibition, extracellular matrix production) appear essentially intact although altered responses of endothelial progenitor cells (EPCs) in peripheral blood to local cues or defective TGFβ-induced secondary signaling to vascular smooth muscle cells, recruited to provide robust vascular walls, have not been excluded.

Recent studies have indeed provided increasing evidence that postnatal neovascularization may not only involve angiogenesis but also these bone marrow derived EPCs [71]. EPCs are considered to originate from CD34+ hematopoietic stem cells; they have been shown to express VEGF-R2, contribute to adult blood vessel formation [72], accelerate the restoration of blood flow in diabetic mice [73] and improve cardiac function after infarction [74,75]. More recently, it was shown that functional impairment of EPCs was correlated with a number of risk factors associated with coronary artery disease, most notably hypertension, reducing the migratory response of EPCs and smoking, reducing the number of EPCs in circulating blood [76]. Given the important role for EPCs in neovascularization of ischemic tissue, the decrease in EPC numbers and activity may contribute to impaired vascularization in patients not only with coronary heart disease but also with other vascular defects such as those associated with diabetes and possibly HHT. Determination of EPC numbers in HHT patients and their ability to contribute to vessel repair is an area warranting further research.


   10 Conclusions
Top
 Abstract
 1 Introduction
 2 TGFβ signaling
 3 Endoglin and HHT...
 4 ALK-1 and HHT...
 5 Other HHT loci
 6 HHT models in...
 7 Vascular phenotypes in...
 8 Primary pulmonary hypertension...
 9 Other (cardio-)vascular...
 10 Conclusions
References
 
This review summarizes all mutations in HHT1 (endoglin) and HHT2 (ALK-1) patients and updates on a previous review in 2000 [78]. All mutations were rechecked with the original data; in some cases the published interpretation was incorrect or questionable (these mutations are indicated by an asterisk in Tables 1 and 2Go). So far, there is no evidence for mutations in multiple genes in individual patients causing HHT. Furthermore, it is not likely that the pulmonary phenotypes in HHT2 patients are caused by an additional mutation in BMPR-II.

Despite the detailed genetic data, it is still not clear which aspects of endothelial cell behavior are affected by mutant TGFβ receptors in HHT patients, and cause the disease. Endothelial cells from HHT patients still have to be examined for altered migration, proliferation, extracellular matrix deposition and the interaction with pericytes and vascular smooth muscle cells. Similar studies could be usefully carried out in EPCs derived from HHT patients. Together with the HHT mouse and zebrafish models, these endothelial cells and EPCs may be useful in identifying cell-specific target genes of ALK-1 and ALK-5 that may be deregulated in HHT patients. This may result in better insights into the pathology of HHT, and possibly gene therapies based on EPC cells.

Time for primary review 22 days.


   Acknowledgements
 
We thank Dr. Marie-José Goumans and Dr. Hans Kristian Ploos van Amstel for critically reading the manuscript, Dr. Cesare Danesino and Dr. Michelle Letarte for discussions and sharing unpublished results. This work was supported by grant 99-046 from the Netherlands Heart Foundation and grant MW 902-16-295 from NWO.


   References
Top
 Abstract
 1 Introduction
 2 TGFβ signaling
 3 Endoglin and HHT...
 4 ALK-1 and HHT...
 5 Other HHT loci
 6 HHT models in...
 7 Vascular phenotypes in...
 8 Primary pulmonary hypertension...
 9 Other (cardio-)vascular...
 10 Conclusions
 References
 

  1. Massague J. TGF-beta signal transduction. Annu Rev Biochem (1998) 67:753–791.[CrossRef][ISI][Medline]
  2. Itoh F., Asao H., Sugamura K., et al. Promoting bone morphogenetic protein signaling through negative regulation of inhibitory Smads. EMBO J (2001) 20:4132–4142.[CrossRef][ISI][Medline]
  3. Kretzschmar M., Doody J., Timokhina I., et al. A mechanism of repression of TGFbeta/Smad signaling by oncogenic Ras. Genes Dev (1999) 13:804–816.[Abstract/Free Full Text]
  4. Derynck R., Zhang Y., Feng X.H. Smads: transcriptional activators of TGF-beta responses. Cell (1998) 95:737–740.[CrossRef][ISI][Medline]
  5. Massague J., Wotton D. Transcriptional control by the TGF-beta/Smad signaling system. EMBO J (2000) 19:1745–1754.[CrossRef][ISI][Medline]
  6. Wang X.F., Lin H.Y., Ng-Eaton E., et al. Expression cloning and characterization of the TGF-beta type III receptor. Cell (1991) 67:797–805.[CrossRef][ISI][Medline]
  7. Chen Y.G., Massague J. Smad1 recognition and activation by the ALK1 group of transforming growth factor-beta family receptors. J Biol Chem (1999) 274:3672–3677.[Abstract/Free Full Text]
  8. Oh S.P., Seki T., Goss K.A., et al. Activin receptor-like kinase 1 modulates transforming growth factor-beta 1 signaling in the regulation of angiogenesis. Proc Natl Acad Sci USA (2000) 97:2626–2631.[Abstract/Free Full Text]
  9. McDonald M.T., Papenberg K.A., Ghosh S., et al. A disease locus for hereditary haemorrhagic telangiectasia maps to chromosome 9q33–34. Nat Genet (1994) 6:197–204.[CrossRef][ISI][Medline]
  10. Shovlin C.L., Hughes J.M., Tuddenham E.G., et al. A gene for hereditary haemorrhagic telangiectasia maps to chromosome 9q3. Nat Genet (1994) 6:205–209.[CrossRef][ISI][Medline]
  11. McAllister K.A., Grogg K.M., Johnson D.W., et al. Endoglin, a TGF-beta binding protein of endothelial cells, is the gene for hereditary haemorrhagic telangiectasia type 1. Nat Genet (1994) 8:345–351.[CrossRef][ISI][Medline]
  12. McAllister K.A., Baldwin M.A., Thukkani A.K., et al. Six novel mutations in the endoglin gene in hereditary hemorrhagic telangiectasia type 1 suggest a dominant-negative effect of receptor function. Hum Mol Genet (1995) 4:1983–1985.[Free Full Text]
  13. Pece N., Vera S., Cymerman U., et al. Mutant endoglin in hereditary hemorrhagic telangiectasia type 1 is transiently expressed intracellularly and is not a dominant negative. J Clin Invest (1997) 100:2568–2579.[ISI][Medline]
  14. Shovlin C.L., Hughes J.M., Scott J., et al. Characterization of endoglin and identification of novel mutations in hereditary hemorrhagic telangiectasia. Am J Hum Genet (1997) 61:68–79.[ISI][Medline]
  15. Yamaguchi H., Azuma H., Shigekiyo T., et al. A novel missense mutation in the endoglin gene in hereditary hemorrhagic telangiectasia. Thromb Haemost (1997) 77:243–247.[ISI][Medline]
  16. Gallione C.J., Klaus D.J., Yeh E.Y., et al. Mutation and expression analysis of the endoglin gene in hereditary hemorrhagic telangiectasia reveals null alleles. Hum Mutat (1998) 11:286–294.[CrossRef][ISI][Medline]
  17. Gallione C.J., Scheessele E.A., Reinhardt D., et al. Two common endoglin mutations in families with hereditary hemorrhagic telangiectasia in the Netherlands Antilles: evidence for a founder effect. Hum Genet (2000) 107:40–44.[CrossRef][ISI][Medline]
  18. Pece-Barbara N., Cymerman U., Vera S., et al. Expression analysis of four endoglin missense mutations suggests that haploinsufficiency is the predominant mechanism for hereditary hemorrhagic telangiectasia type 1. Hum Mol Genet (1999) 8:2171–2181.[Abstract/Free Full Text]
  19. Bourdeau A., Cymerman U., Paquet M.E., et al. Endoglin expression is reduced in normal vessels but still detectable in arteriovenous malformations of patients with hereditary hemorrhagic telangiectasia type 1. Am J Pathol (2000) 156:911–923.[Abstract/Free Full Text]
  20. Cymerman U., Vera S., Pece-Barbara N., et al. Identification of hereditary hemorrhagic telangiectasia type 1 in newborns by protein expression and mutation analysis of endoglin. Pediatr Res (2000) 47:24–35.[ISI][Medline]
  21. Paquet M.E., Pece-Barbara N., Vera S., et al. Analysis of several endoglin mutants reveals no endogenous mature or secreted protein capable of interfering with normal endoglin function. Hum Mol Genet (2001) 10:1347–1357.[Abstract/Free Full Text]
  22. Dakeishi M., Shioya T., Wada Y., et al. Genetic epidemiology of hereditary hemorrhagic telangiectasia in a local community in the northern part of Japan. Hum Mutat (2002) 19:140–148.[CrossRef][ISI][Medline]
  23. Shovlin C.L., Letarte M. Hereditary haemorrhagic telangiectasia and pulmonary arteriovenous malformations: issues in clinical management and review of pathogenic mechanisms. Thorax (1999) 54:714–729.[Free Full Text]
  24. Gougos A., Letarte M. Primary structure of endoglin, an RGD-containing glycoprotein of human endothelial cells. J Biol Chem (1990) 265:8361–8364.[Abstract/Free Full Text]
  25. Gougos A., St Jacques S., Greaves A., et al. Identification of distinct epitopes of endoglin, an RGD-containing glycoprotein of endothelial cells, leukemic cells, and syncytiotrophoblasts. Int Immunol (1992) 4:83–92.[Abstract/Free Full Text]
  26. Lux A., Gallione C.J., Marchuk D.A. Expression analysis of endoglin missense and truncation mutations: insights into protein structure and disease mechanisms. Hum Mol Genet (2000) 9:745–755.[Abstract/Free Full Text]
  27. Heutink P., Haitjema T., Breedveld G.J., et al. Linkage of hereditary haemorrhagic telangiectasia to chromosome 9q34 and evidence for locus heterogeneity. J Med Genet (1994) 31:933–936.[Abstract]
  28. McAllister K.A., Lennon F., Bowles-Biesecker B., et al. Genetic heterogeneity in hereditary haemorrhagic telangiectasia: possible correlation with clinical phenotype. J Med Genet (1994) 31:927–932.[Abstract]
  29. Porteous M.E., Curtis A., Williams O., et al. Genetic heterogeneity in hereditary haemorrhagic telangiectasia. J Med Genet (1994) 31:925–926.[Abstract]
  30. Vincent P., Plauchu H., Hazan J., et al. A third locus for hereditary haemorrhagic telangiectasia maps to chromosome 12q. Hum Mol Genet (1995) 4:945–949.[Abstract/Free Full Text]
  31. Berg J.N., Guttmacher A.E., Marchuk D.A., et al. Clinical heterogeneity in hereditary haemorrhagic telangiectasia: are pulmonary arteriovenous malformations more common in families linked to endoglin? J Med Genet (1996) 33:256–257.[Abstract]
  32. McDonald J.E., Miller F.J., Hallam S.E., et al. Clinical manifestations in a large hereditary hemorrhagic telangiectasia (HHT) type 2 kindred. Am J Med Genet (2000) 93:320–327.[CrossRef][ISI][Medline]
  33. Kjeldsen A.D., Brusgaard K., Poulsen L., et al. Mutations in the ALK-1 gene and the phenotype of hereditary hemorrhagic telangiectasia in two large Danish families. Am J Med Genet (2001) 98:298–302.[CrossRef][ISI][Medline]
  34. Trembath R.C., Thomson J.R., Machado R.D., et al. Clinical and molecular genetic features of pulmonary hypertension in patients with hereditary hemorrhagic telangiectasia. N Engl J Med (2001) 345:325–334.[Abstract/Free Full Text]
  35. Johnson D.W., Berg J.N., Gallione C.J., et al. A second locus for hereditary hemorrhagic telangiectasia maps to chromosome 12. Genome Res (1995) 5:21–28.[Abstract/Free Full Text]
  36. Johnson D.W., Berg J.N., Baldwin M.A., et al. Mutations in the activin receptor-like kinase 1 gene in hereditary haemorrhagic telangiectasia type 2. Nat Genet (1996) 13:189–195.[CrossRef][ISI][Medline]
  37. Berg J.N., Gallione C.J., Stenzel T.T., et al. The activin receptor-like kinase 1 gene: genomic structure and mutations in hereditary hemorrhagic telangiectasia type 2. Am J Hum Genet (1997) 61:60–67.[ISI][Medline]
  38. Klaus D.J., Gallione C.J., Anthony K., et al. Novel missense and frameshift mutations in the activin receptor-like kinase-1 gene in hereditary hemorrhagic telangiectasia. Mutations in brief no. 164. Online Hum Mutat (1998) 12:137.
  39. Abdalla S.A., Pece-Barbara N., Vera S., et al. Analysis of ALK-1 and endoglin in newborns from families with hereditary hemorrhagic telangiectasia type 2. Hum Mol Genet (2000) 9:1227–1237.[Abstract/Free Full Text]
  40. Lin W.D., Wu J.Y., Hsu H.B., et al. Mutation analysis of a family with hereditary hemorrhagic telangiectasia associated with hepatic arteriovenous malformation. J Formos Med Assoc (2001) 100:817–819.[ISI][Medline]
  41. Abdalla SA, Cymerman U, Johnson RM et al. Disease-associated mutations in conserved residues of the ALK-1 kinase domain. Eur J Hum Genet, 2002; in press.
  42. Barbara N.P., Wrana J.L., Letarte M. Endoglin is an accessory protein that interacts with the signaling receptor complex of multiple members of the transforming growth factor-beta superfamily. J Biol Chem (1999) 274:584–594.[Abstract/Free Full Text]
  43. Piantanida M., Buscarini E., Dellavecchia C., et al. Hereditary haemorrhagic telangiectasia with extensive liver involvement is not caused by either HHT1 or HHT2. J Med Genet (1996) 33:441–443.[Abstract]
  44. Buscarini E., Buscarini L., Danesino C., et al. Hepatic vascular malformations in hereditary hemorrhagic telangiectasia: Doppler sonographic screening in a large family. J Hepatol (1997) 26:111–118.[CrossRef][ISI][Medline]
  45. Olivieri C., Mira E., Delu G., et al. Identification of 13 new mutations in the ACVRL1 gene in a group of 52 unselected Italian patients affected by hereditary haemorrhagic telangiectasia. J Med Genet (2002) 39:E39.[CrossRef][Medline]
  46. Wallace G.M., Shovlin C.L. A hereditary haemorrhagic telangiectasia family with pulmonary involvement is unlinked to the known HHT genes, endoglin and ALK-1. Thorax (2000) 55:685–690.[Abstract/Free Full Text]
  47. Li D.Y., Sorensen L.K., Brooke B.S., et al. Defective angiogenesis in mice lacking endoglin. Science (1999) 284:1534–1537.[Abstract/Free Full Text]
  48. Bourdeau A., Dumont D.J., Letarte M. A murine model of hereditary hemorrhagic telangiectasia. J Clin Invest (1999) 104:1343–1351.[ISI][Medline]
  49. Arthur H.M., Ure J., Smith A.J., et al. Endoglin, an ancillary TGFbeta receptor, is required for extraembryonic angiogenesis and plays a key role in heart development. Dev Biol (2000) 217:42–53.[CrossRef][ISI][Medline]
  50. Dickson M.C., Martin J.S., Cousins F.M., et al. Defective haematopoiesis and vasculogenesis in transforming growth factor-beta 1 knock out mice. Development (1995) 121:1845–1854.[Abstract]
  51. Oshima M., Oshima H., Taketo M.M. TGF-beta receptor type II deficiency results in defects of yolk sac hematopoiesis and vasculogenesis. Dev Biol (1996) 179:297–302.[CrossRef][ISI][Medline]
  52. Larsson J., Goumans M.J., Sjostrand L.J., et al. Abnormal angiogenesis but intact hematopoietic potential in TGF-beta type I receptor-deficient mice. EMBO J (2001) 20:1663–1673.[CrossRef][ISI][Medline]
  53. Carvalho RLS, Driessens MHE, Mummery C. Vasculogenesis and angiogenesis: lessons from genetics in mice. Commun Theor Biol, 2002; in press.
  54. Bourdeau A., Faughnan M.E., McDonald M.L., et al. Potential role of modifier genes influencing transforming growth factor-beta1 levels in the development of vascular defects in endoglin heterozygous mice with hereditary hemorrhagic telangiectasia. Am J Pathol (2001) 158:2011–2020.[Abstract/Free Full Text]
  55. Bourdeau A., Faughnan M.E., Letarte M. Endoglin-deficient mice, a unique model to study hereditary hemorrhagic telangiectasia. Trends Cardiovasc Med (2000) 10:279–285.[CrossRef][ISI][Medline]
  56. Urness L.D., Sorensen L.K., Li D.Y. Arteriovenous malformations in mice lacking activin receptor-like kinase-1. Nat Genet (2000) 26:328–331.[CrossRef][ISI][Medline]
  57. Goumans M.J., Valdimarsdottir G., Itoh S., et al. Balancing the activation state of the endothelium via two distinct TGF-beta type I receptors. EMBO J (2002) 21:1743–1753.[CrossRef][ISI][Medline]
  58. Roman B.L., Pham V.N., Lawson N.D., et al. Disruption of acvrl1 increases endothelial cell number in zebrafish cranial vessels. Development (2002) 129:3009–3019.[ISI][Medline]
  59. Lane K.B., Machado R.D., Pauciulo M.W., et al. Heterozygous germline mutations in BMPR2, encoding a TGF-beta receptor, cause familial primary pulmonary hypertension. The International PPH Consortium. Nat Genet (2000) 26:81–84.[CrossRef][ISI][Medline]
  60. Scott J. Pulling apart pulmonary hypertension. Nat Genet (2000) 26:3–4.[CrossRef][ISI][Medline]
  61. De Caestecker M., Meyrick B. Bone morphogenetic proteins, genetics and the pathophysiology of primary pulmonary hypertension. Respir Res (2001) 2:193–197.[CrossRef][ISI][Medline]
  62. Rich S., Dantzker D.R., Ayres S.M., et al. Primary pulmonary hypertension. A national prospective study. Ann Intern Med (1987) 107:216–223.[CrossRef][ISI][Medline]
  63. Nichols W.C., Koller D.L., Slovis B., et al. Localization of the gene for familial primary pulmonary hypertension to chromosome 2q31–32. Nat Genet (1997) 15:277–280.[CrossRef][ISI][Medline]
  64. Morse J.H., Jones A.C., Barst R.J., et al. Mapping of familial primary pulmonary hypertension locus (PPH1) to chromosome 2q31–q32. Circulation (1997) 95:2603–2606.[Abstract/Free Full Text]
  65. Machado R.D., Pauciulo M.W., Fretwell N., et al. A physical and transcript map based upon refinement of the critical interval for PPH1, a gene for familial primary pulmonary hypertension. The Internation