CLINICAL AND EXPERIMENTAL ASPECTS OF NOTCH RECEPTOR SIGNALING: HAJDU-CHENEY SYNDROME AND RELATED DISORDERS

Abstract

There are four Notch transmembrane receptors that determine the fate and function of cells. Notch is activated following its interactions with ligands of the Jagged and Delta-like families that lead to the cleavage and release of the Notch intracellular domain (NICD); this translocates to the nucleus to induce the transcription of Notch target genes. Genetic disorders of loss- and gain-of-NOTCH function present with severe clinical manifestations. Hajdu Cheney Syndrome (HCS) is a rare genetic disorder characterized by acroosteolysis, fractures, short stature, neurological manifestations, craniofacial developmental abnormalities, cardiovascular defects and polycystic kidneys. HCS is associated with NOTCH2 gain-of-function mutations. An experimental mouse model of the disease revealed that the bone loss is secondary to increased osteoclastogenesis and bone resorption due to enhanced expression of receptor activator of nuclear factor kappa B ligand (Rankl). This would suggest that inhibitors of bone resorption might prove to be beneficial in the treatment of the bone loss associated with HCS. Notch2 is a determinant of B-cell allocation in the marginal zone of the spleen and “somatic” mutations analogous to those found in HCS are associated with B-cell lymphomas of the marginal zone, but there are no reports of lymphomas associated with HCS. In conclusion, HCS is a serious genetic disorder associated with NOTCH2 mutations. New experimental models have offered insight on mechanisms responsible for the manifestations of HCS.

1. Notch Receptors

Notch has emerged as a novel signal that plays a key role in cell fate decisions and in the differentiation and function of cells of multiple lineages. There are four Notch receptors (Notch1 to 4) and five classic Notch ligands termed Jagged (Jag)1 and Jag2, and Delta-like (Dll)1, Dll3 and Dll4 [1]. Notch and its ligands are transmembrane proteins that retain structural similarity and mediate communication between neighboring cells. The extracellular domain of Notch is the site of interaction with its ligands, and consists of multiple epidermal growth factor (EGF) repeats. At the junction of the extracellular and the transmembrane domain of Notch rests the negative regulatory region (NRR). This is the site of cleavage required for the initial activation of Notch following its interactions with Jag or Dll ligands [2,3]. The intracellular domain of Notch (NICD) consists of an Rbpjκ-association module (RAM) domain, ankyrin repeats and nuclear localization sequences. These domains are required to induce the transcription of Notch target genes [4,5]. The C-terminus of Notch contains a proline (P)-, glutamic acid (E)-, serine (S)-, threonine (T) (PEST) domain, which is targeted by ubiquitin ligases. This domain is required for the proteasomal degradation of Notch (Figure 1). Notch ligand interactions lead to its proteolytic cleavage and the release of the NICD, and its translocation to the nucleus, where it forms a complex with recombination signal-binding protein for Ig of κ region (Rbpjκ) and mastermind-like (Maml) to regulate transcription [6–8]. The human ortholog of Rbpjκ is termed CBF1, Suppressor of Hairless, Lag 1 (CSL) [9]. The interaction of the NICD, Rbpjκ and Maml leads to the displacement of transcriptional repressors and recruitment of activators of transcription by the NICD to induce the expression of Notch target genes (Figure 2). This has been termed the Notch canonical signaling pathway, and results in the induction of hairy and enhancer of split (Hes)1, Hes5, Hes 6 and Hes7, and HES-related with YRPW motif (Hey)1, Hey2 and HeyL [10–12]. Importantly Rbpjκ, and not the NICD, binds to DNA. Although Rbpjκ acts as a transcriptional activator in Notch signaling, in the absence of Notch-ligand interactions Rbpjκ acts as a transcriptional repressor due to its ability to recruit transcriptional co-repressors and histone deacetylases. This complex is disrupted by the NICD, leading to the recruitment of co-activators of transcription, histone acetylation and enhanced transcription of target genes [13,14]. Cyclin-dependent kinase 8 phosphorylates the PEST domain of the NICD, causing the disassembly of the ternary NICD, Rbpjκ, Maml complex followed by the ubiquitination of the NICD by E3 ubiquitin-ligases and the degradation of Notch [7]. This is necessary to avoid the persistence of Notch signaling which would lead to a gain-of-Notch function.

Figure 1. Domains of the four Notch receptors.

The upper panel shows the domain and motif organization of a generic human/murine Notch receptor before cleavage at the S1 site by furin-like convertases in the Golgi compartment. The extracellular domain contains a leader peptide (LP) and multiple epidermal growth factor (EGF)-like tandem repeats followed by Lin12-Notch repeats (LNR) and the heterodimerization domain (HD). The transmembrane domain (TMD) is located between the extracellular and intracellular domains. The Notch intracellular domain (NICD) contains an Rbpjκ-association module (RAM), a nuclear localization sequence (NLS), ankyrin (ANK) repeats and tandem NLS, which are followed by a proline (P)-, glutamic acid (E)-, serine (S)- and threonine (T)-rich (PEST) domain. The lower panel shows the domains and motifs of heterodimeric individual receptors, the negative regulatory region (NRR) is formed by the LNR and HD following cleavage at the S1 site. Notch1 and Notch2 have 36 EGF-like repeats; in green are those required for binding of Notch1 and Notch2 to cognate Delta/Serrate/Lag2 ligands. Notch1 and Notch2 have a similar NICD, and Notch3 has 34 EGF-like repeats and a shorter NICD than Notch1 and Notch2. Notch4 has 29 EGF-like repeats and an NICD that is shorter than that of other receptors and lacks the tandem NLS located between the ANK repeats and the PEST domain. Reproduced with permission from Zanotti and Canalis, Endocrine Reviews 37:223, 2016.

Figure 2. Canonical Notch Signaling.

Under resting conditions, Notch receptors are intact and Rbpjκ binds histone deacetylases (HDAC) and co-repressors. Following Delta-like or Jagged interactions with the extracellular domain of Notch, the Notch intracellular domain (NICD) is released and translocates to the nucleus to form a complex with mastermind-like (Maml) and Rbpjκ. As a consequence, co-repressors and HDAC is displaced, CBP/p300 is recruited and histone acetylation ensues leading to the induction of Hes and Hey.

Although the four Notch receptors retain structural similarities, their function is not the same. Functional differences between the four Notch receptors are related to individual structural properties, to their temporal and cellular expression, to variations in the affinity of the Notch extracellular domain for its ligands and to differential interactions of the NICD with Rbpjκ [15,16]. For example, Notch1, 2 and 3 are expressed in bone, whereas Notch4 is not detectable in skeletal cells [17]. Notch3 is predominantly expressed in vascular cells, and as a consequence plays a key role in vascular development, and Notch4 is expressed in endothelial cells and plays a role in the development of vertebrate endothelium [18]. Notch4 mRNA is detected in the endothelium of highly vascularized adult tissues, such as lungs, heart and kidney. Notch needs to interact with ligands present in neighboring cells of the same or different lineages for its activation, and Jagged1 is the prevalent ligand expressed by skeletal cells [19]. Notch1 and Notch2 are structurally similar, but either the Notch1 or Notch2 inactivation results in embryonic lethality, indicating that their functions are not redundant [20–22]. The structure of Notch3 diverges, and the amino acid identity of the Notch3 NICD is significantly different from that of Notch1 and 2 [23]. This, as well as a distinct cellular pattern of expression, confers to Notch3 a unique physiological role. The actions of Notch are cell-context dependent and cellular responses are linked to the stage of cell maturation at the time of Notch activation. For example, when Notch1 is activated in osteoblast precursors it inhibits osteoblast differentiation, whereas when activated in mature osteoblasts and osteocytes Notch1 inhibits bone resorption and causes an osteopetrotic phenotype [24,25]. Depending on the cell environment, Notch 2 can lead to progenitor cell depletion or to mature cell reallocation [26,27]

2. Hajdu Cheney Syndrome – Clinical Aspects

In 1948, Hajdu and Kauntze reported a young accountant suffering from severe osteoporosis, acroosteolysis and neurological complications that years later took his life [28]. The clinical condition was reported as a syndrome by Cheney in 1965 [29]. Hajdu Cheney Syndrome (HCS) (OMIN 102500) is a rare genetic disorder, and under 100 cases have been reported although its prevalence is probably higher. HCS is characterized by craniofacial developmental abnormalities that appear at a young age and evolve as the child matures (Table 1). There is high clinical variability and a phenotypical evolution of the disease. Subjects with HCS present with facial dysmorphism, synophrys and epicanthal folds. The distance between the eyes is short, and thick eyebrows extend toward the midline; there is malar hypoplasia, a long and smooth philtrum and micrognathia [30]. The nasal bridge is flattened and becomes broad, and facial features are coarse and the neck is short.

Table 1.

Clinical manifestations of Hajdu Cheney Syndrome

Craniofacial Developmental Abnormalities Platybasia, open sutures, wormian bones Characteristic and evolving facial features Maxillofacial and dental abnormalities

Skeletal Features Osteoporosis with fractures Scoliosis, kyphosis Osteolysis of distal phalanges Short stature Joint hypermobility

Renal Polycystic kidneys

Cardiac Septal and valvular defects

Spleen Splenomegaly

The craniofacial developmental defects include wormian bones, open sutures, platybasia and basilar invagination. These are serious manifestations of the disease since they can cause severe neurological complications, including central respiratory arrest and sudden death. Abnormal dental eruptions, and tooth decay with premature loss of teeth are common [31]. Short stature, generalized and local joint hypermobility are reported frequently. Vertebral abnormalities include fractures, kyphosis and scoliosis and long bone deformities. A limited number of subjects undergoing bone biopsies have been reported, but the results have not been conclusive and findings included normal, increased and decreased bone remodeling [32–34]. In selected individuals, an increased number of osteoclasts was found suggesting that enhanced bone resorption is responsible for the bone loss [34].

A characteristic feature of HCS is acroosteolysis of the distal phalanges of fingers and toes. This is associated with an inflammatory reaction, pain and swelling, and the process results in the loss of the distal phalanges and shortening of hands and feet. Biopsies of the lesions have revealed an inflammatory process, neovascularization and fibrosis, but the mechanism responsible for the lesions is not entirely clear. It might be related to effects on Notch2 on B-cell allocation, and subjects with HCS can display splenomegaly [35]. The mechanisms of the acroosteolysis are not considered to be the same as those responsible for the generalized bone loss and fractures observed in HCS. Patients with HCS can present with cardiovascular defects, including patent ductus arteriosus, septal defects, and mitral and aortic valve abnormalities associated with valvular insufficiency or stenosis [36,37]. About 10% of subjects with HCS have polycystic kidneys, and serpentine fibula-polycystic kidney syndrome is most likely the same disorder as HCS [38–40].

3. Hajdu Cheney Syndrome – Genetic Aspects

Mutations in Notch receptors, their ligands, intracellular partners and factors that modify Notch signaling have been associated with genetic disorders, many of them affecting the skeleton. The mutations result either in a gain- or loss-of-Notch function and include genes encoding enzymes that modify the Notch receptor as well as its target genes (Table 2). Genes encoding glycosylating enzymes include EOGT (EGF domain specific O-linked N acetylglucosamine transferase) and LFNG (Lunatic fringe); CHSYS1 encodes chondroitin sulfate synthase, a glucuronyltransferase that downregulates Notch signaling. MESP (mesoderm posterior bHLH transcription factor) and HES7 are Notch target genes, and HES7 regulates LFNG expression.

Table 2.

Genetic disorders associated with mutations in the major components of the Notch signaling pathway.

Gene	Clinical Disorder	Presumed Change in Function
Notch Receptors
NOTCH1	Adams Oliver Syndrome	Loss
NOTCH2	Alagille Syndrome	Loss
	Hajdu Cheney Syndrome	Gain
NOTCH3	CADASIL*	Uncertain
	Lateral Meningocele Syndrome	Gain
NOTCH4	Not reported
Notch Ligands
JAG1	Alagille Syndrome	Loss
JAG2	Not reported
DLL1	Not reported
DLL3	Spondylocostal Dysostosis	Loss
DLL4	Adams Oliver Syndrome	Loss
Intracellular Signal
CSL (Rbpjκ)	Adams Oliver Syndrome	Loss
Modifiers of Notch Signaling and Notch Target Genes
EOGT	Adams Oliver Syndrome	Loss
LFNG	Spondylocostal Dysostosis	Loss
CHSY1	Brachydactyly	Gain
MESP	Spondylocostal Dysostosis	Loss
HES7	Spondylocostal Dysostosis	Loss

^*CADASIL – Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy

The inheritance of HCS is autosomal dominant although sporadic cases have been reported frequently Exome-wide sequencing of families affected by HCS demonstrated that HCS is associated with mutations in exon 34 of NOTCH2 [35,41–45]. The mutations are nonsense mutations or short deletions leading to the formation of a termination codon upstream of the PEST domain. Since the PEST domain is required for the ubiquitination and degradation of NOTCH2, the mutations lead to the translation of a truncated NOTCH2 protein that is presumably stable; and as a consequence, a gain-of-NOTCH2 function occurs. The truncated NOTCH2 intracellular domain retains all sequences necessary to form a complex with CSL and MAML, and regulate transcription. Similar NOTCH2 mutations in exon 34 leading to a truncated protein devoid of the PEST domain have been associated with B-cell lymphoma and were shown to have enhanced NOTCH2 signaling [46,47]. The increased NOTCH2 activity implies the existence of NOTCH ligand interactions under basal conditions with the subsequent activation of Notch signaling which is enhanced by the prolonged life of the mutated NOTCH2 protein. The diagnosis of HCS can be established by documentation of mutations in exon 34 of NOTCH2 that lead to the creation of a termination codon upstream the PEST domain.

4. Hajdu Cheney Syndrome and Related Genetic Disorders

Lateral Meningocele (LMS) or Lehman Syndrome (OMIM 130720), is a rare genetic disorder with some phenotypic overlap with HCS [48]. LMS is characterized by craniofacial anomalies, hypotonia and meningocele with related neurological dysfunction [49] (Table 3). Skeletal manifestations are numerous, and they include craniofacial developmental defects, short stature, scoliosis and osteopenia [48,50]. Although LMS and HCS are distinct disorders, select phenotypic characteristics overlap [51]. Indeed, a case of LMS was reported initially as having HCS [48]. The subject, a young child, presented with severe scoliosis and dural ectasia surrounding the spinal cord as well as features of HCS [52]. Because of the spinal cord meningoceles, the diagnosis was questioned and the subject was subsequently confirmed to have LMS [51,52]. Exome sequencing of subjects afflicted by LMS has demonstrated the presence of point mutations or short deletions in exon 33 of NOTCH3, upstream of the PEST domain. Like in HCS, the mutations lead to the translation of a truncated and stable protein, devoid of the PEST domain, and presumably a gain-of-NOTCH3 function [51–53]. Interestingly, no genetic disorders associated with similar mutations in NOTCH1 or NOTCH4, leading to the translation of a truncated protein devoid of the PEST domain and gain-of-function, have been reported. Somatic mutations in exon 34 of NOTCH1 are associated with T-cell acute lymphoblastic leukemia [54].

Table 3.

Clinical manifestations of Lateral Meningocele Syndrome.

Craniofacial Features Hypertelorism High arched eyebrows Ptosis Low set eyes Cleft and high arched narrow palate* Micrognathia* Thick calvariae* Dense base of skull* Wormian bones*

Skeletal Features Short stature* Scoliosis* Pectus excavatum Osteopenia*

Other Features Developmental and intellectual delay Decreased muscle mass Hypotonia Lateral meningocele Syringomyelia Cardiac valve abnormalities*

^*Clinical features also present in HCS.

5. Experimental Models of Hajdu Cheney Syndrome

To gain an understanding of the mechanisms responsible for the HCS phenotype, our laboratory created a mouse model reproducing the mutation found in a subject affected by the disease and presenting with bone loss and fractures. The subject harbored a 6949C>T mutation, (translation start site = 1) in exon 34 of NOTCH2 upstream of the PEST domain. To create a mouse model of HCS, we introduced the human mutation into the corresponding nucleotide of the mouse genome, namely a 6955C>T mutation, by homologous recombination and verified by DNA sequencing in F1 pups [55]. The mutation created a STOP codon in exon 34 of Notch2, and the translation of a predicted truncated Notch2 Q2319X protein of 2318 amino acids. Homozygous Notch2HCS mutant mice died at birth, and heterozygous mice were smaller and had shorter limbs than controls. This is in accordance with the known inhibitory effects of Notch on chondrogenesis and is reminiscent of the short stature displayed by humans with HCS [56].

Microcomputed tomography (μCT) of the distal femur revealed that Notch2HCS mutant mice of both sexes have a pronounced decrease in trabecular bone volume due to a reduction in the number of trabeculae, and to a lesser extent to a decrease in trabecular thickness. The decreased cancellous bone volume was associated with decreased connectivity. Cortical bone of Notch2HCS mutants was thin and porous, and total area as well as bone area were reduced. Cancellous bone histomorphometry of femurs from Notch2HCS mutant mice demonstrated decreased bone volume/tissue volume, an increase in osteoclast number and eroded surface, and no change in osteoblast number or bone formation rate. The results indicated that enhanced bone resorption without a coupled bone-forming response was responsible for the skeletal phenotype. Cortical bone histomorphometry confirmed an increase in endocortical osteoclast number and eroded surface in Notch2HCS mutant mice.

The expression of the Notch target genes Hey1, Hey2 and HeyL and the activity of transiently transfected Notch reporter constructs were increased in cells from Notch2HCS mice confirming enhanced activation of Notch signaling. In accordance with the resorptive phenotype observed, there was an increase in the expression of Tnfsf11, encoding for Rankl, in femurs, osteoblasts and osteocytes from Notch2HCS mutant mice, and no changes in Tnfrsf11b, encoding for the Rankl decoy receptor osteoprotegerin. The increase in Tnfsf11 was dependent on the activation of Notch signaling because it was prevented by the addition of the γ-secretase inhibitor LY450139, known to prevent the cleavage of Notch and release of the NICD [57]. The Notch2HCS mutation resulted in enhanced osteoclastogenesis in vitro, and there was an increase in the number of osteoclasts in bone marrow cell cultures from Notch2HCS mutants compared to controls.

The phenotype of the Notch2HCS mutant mouse recapitulates aspects of HCS in humans. The osteopenic phenotype of the Notch2HCS mutant mouse can be explained by an increase in bone resorption and osteoclastogenesis [17,58]. The increased number of osteoclasts and bone resorption appeaed to be secondary to enhanced Rankl expression in cells of the osteoblast lineage and possibly to direct effects of Notch2 on osteoclastogenesis. Notch2 enhances osteoclastogenesis directly by interacting with nuclear factor κB on Nfatc1 regulatory elements in cells of the osteoclast lineage [58]. Currently, it is not known whether the induction of Rankl by Notch2 is mediated by Notch canonical signaling and Notch2 NICD interactions with Rbpjκ and Maml. In contrast to the resorptive phenotype, neither bone histomorphometry nor in vitro studies demonstrated changes in osteoblast differentiation or in the bone-forming capacity of mice carrying the Notch2HCS mutation. Bone formation was decreased slightly and not increased as it would be expected in a state of high bone remodeling due to increased bone resorption. It is, therefore, possible that Notch2 inhibited bone formation and uncoupled bone resorption from bone formation.

To determine the individual contributions of osteoclasts and osteoblasts to the osteopenia of the Notch2HCS mouse model, we created a conditional by inversion model where Cre recombination generates a Notch2 mutant allele lacking the PEST domain [59]. The model was designed to introduce a premature STOP codon in exon 34 of Notch2 following Cre-mediated recombination. This would lead to the translation of a truncated Notch2 protein, mimicking the genetic defect associated with HCS. To activate the Notch2 mutation in osteoclasts or osteoblasts, conditional mice were bred with mice expressing Cre under the control of the Lyz2 (encoding for Lysozyme 2) or the BGLAP (encoding for osteocalcin) promoter, respectively. The induction of the Notch2 mutation in Lyz2-expressing cells had no skeletal consequences and did not affect the capacity of bone marrow macrophages to form osteoclasts in vitro, suggesting that the gain-of-function mutation does not have direct effects on cells of the myeloid lineage. In contrast, induction of the mutation in osteoblasts led to generalized osteopenia associated with enhanced cancellous bone resorption and increased Tnfsf11 expression in osteoblasts from Notch2 mutant mice. These experiments reveal that introduction of the HCS mutation in osteoblasts is responsible for the osteopenia due to increased Rankl expression and bone resorption. The results are congruent with the observation that the deletion of Notch2 in Lyz2-expressing cells does not cause a skeletal phenotype [60]. Therefore, neither the activation nor the inactivation of Notch2 in cells of the myeloid lineage in vivo has an effect on skeletal homeostasis, and the effect of Notch2 on bone resorption is secondary to its actions on cells of the osteoblast lineage through the increased expression of Rankl. However, the in vivo observations are not fully congruent with in vitro studies demonstrating that Notch2 enhances osteoclastogenesis directly [58]. This would suggest that the overall effect of Notch2 on osteoclastogenesis is complex and may be derived from its actions on cells of various lineages.

6. Hajdu Cheney Syndrome, B-cells and the Spleen

Notch receptors play an important role in the fate and function of cells of the immune system. Notch1 is expressed preferentially by T-cells, and its inactivation prevents T-cell formation and causes ectopic B-cell development in the thymus, whereas its constitutive activation is associated with T-cell acute lymphoblastic leukemia [61]. Notch1 is an important modulator of T cell-mediated immune responses [62]. Notch2 is expressed preferentially in mature B-cells, and Notch2 signaling is indispensable for marginal zone (MZ) B-cell development in the spleen [63,64]. MZ B-cells reside in the spleen at the junction of the red pulp, constituted by red cells, and white pulp, constituted by follicular cells; MZ B-cells originate from circulating B-cells [65,66]. Notch2 haploinsufficiency and the inactivation of Notch2 in Mx- or CD19-expressing B-cells result in a marked reduction in MZ B-cells of the spleen [67–69]. The inactivation of Rbpjκ in CD19-expressing cells phenocopies the inactivation of Notch2 indicating that canonical signaling and Notch2 NICD interactions with Rbpjκ are responsible for changes in MZ B cell allocation [64]. Moreover, the overexpression of the Notch2 NICD in CD19-expressing cells leads to the allocation of B-cells to the MZ of the spleen at the expense of follicular cells [27]. Individuals with NOTCH2 loss-of-function mutations have decreased MZ B-cells, whereas subjects with gain-of-NOTCH2 function mutations, such as those occurring in HCS, may present with splenomegaly [35,69]. Importantly, somatic mutations in the same region of exon 34 of NOTCH2, as those found in HCS, are associated with diffuse large B-cell lymphomas and lymphomas of the MZ of the spleen [46,47]. Notch2HCS mutant mice have altered B-cell allocation with an increase in MZ B-cells and a proportional reduction in follicular B-cells in the spleen. ADAM10 expression results in activation of Notch signaling and the commitment of T1 B transitional cells to MZ B-cells [70]. Accordingly, the MZ does not develop following the deletion of Adam10 in CD19-expressing cells [71]. Treatment of Notch2HCS mutant and wild type mice with anti-Notch2 antibodies have confirmed the dependency of the MZ B-cell population on Notch2 signal activation [72]. The mechanisms responsible for the effect of Notch2 activation on B-cell allocation in the spleen have not been elucidated, but may involve canonical Notch signaling and the induction of the Notch target genes Hes1 and Hes5, since they are expressed in the spleen and are Notch-dependent.

Although subjects with HCS have not been reported to have alterations in B-cell allocation, a recent report documents splenomegaly in an affected individual [35]. It is of interest that subjects with Alagille Syndrome associated with JAG1 mutations do not have abnormalities in B-cell populations, whereas individuals with Alagille Syndrome associated with NOTCH2 haploinsufficiency display a marked reduction in MZ B-cells arguing for a role of NOTCH2 in B-cell allocation also in humans [69]. However, it is important to note that humans have a less well-developed MZ of the spleen than rodents, and the organization of the marginal and follicular zone in the human spleen differs from that in the mouse [73]. Genome-wide association studies have identified CSL as a risk allele for rheumatoid arthritis, but it is not known whether subjects with HCS are at an increased risk of autoimmune disorders [74].

7. Controlling Notch Signal Activation

The approach to downregulate Notch signaling has been diverse, and includes the use of biochemical inhibitors of Notch activation, antibodies to Notch receptors or to Notch ligands, and the use of small permeable molecules that prevent the formation of an NICD/Rbpjk/Maml ternary complex [75]. γ-secretase inhibitors are frequently used to block the cleavage and prevent activation of the Notch receptor induced by presenilins [76]. The limitation of γ-secretase inhibitors is their lack of specificity since nearly 100 substrates of the γ-secretase complex are known to exist [77]. Thapsigargin is an inhibitor of the sarco/endoplasmic reticulum Ca²⁺-ATPase that precludes the maturation and folding of the Notch receptor and its exit from the endoplasmic reticulum; and as a consequence, it has been used to prevent Notch activation [78,79]. Synthetic small cell permeable molecules that prevent the assembly of an active Notch transcriptional complex can be used for the inhibition of all Notch signaling, but their long-term efficacy is unknown [80]. A limitation of γ-secretase inhibitors, thapsigargin and small permeable molecules that prevent the formation of a Notch transcriptional complex is that they interfere with the indiscriminate activation of all four Notch receptors.

To target specific Notch receptors, antibodies to the NRR of Notch1, Notch2 and Notch3 have been developed [81,82]. The NRR consists of three LIN-12 Notch repeats (LNR) and a heterodimerization domain (HD), and contains the initial cleavage sites (S1 and S2) of Notch which are required for protein maturation and signal activation (Figure 1) [5,83]. The epitope for the anti-Notch NRR bridges the LNR and HD so that the antibody locks the receptor in its quiescent state preventing Notch activation [82,84]. The targeting of this region prevents the initial cleavage and activation of specific Notch isoforms, making the use of anti-Notch NRR antibodies ideal for the neutralization of individual Notch receptors. However, one needs to be cautious with the extrapolation of results from studies in preclinical models to their application to human disease since information on the inhibition of Notch signaling in humans is quite limited. Moreover, Notch neutralization is not without unwanted events, and it may cause vascular tumors and gastrointestinal toxicity [85,86].

Recently, anti-Notch2 NRR antibodies were tested for their effects on the skeletal phenotype of Notch2HCS mutant and control mice. Treatment of 1 month old mice with an anti-Notch2 NRR antibody for a 4 week period reversed the osteopenic phenotype of Notch2HCS mutant mice. Although encouraging, extrapolation of these results to the human condition is not possible and one should exert caution with their interpretation and applicability beyond the species studied.

8. Treatment of Hajdu Cheney Syndrome

The treatment and management of HCS is defined by the organs affected by the disease. Realistically this would entail the repair of cardiac defects and treatment of osteoporosis with the goal of preventing fractures. There are no controlled clinical trials on the management of the disease since the number of subjects affected is limited. There are no known ways to either prevent or treat the acroosteolysis, and no reports on the treatment of the growth failure. There are no trials on the treatment of the bone loss or fractures in HCS. Only anecdotal cases treated with bisphosphonates, teriparatide or both exist although the benefit of these therapies is uncertain [87,88]. There should be caution with the use of teriparatide because experimental models of HCS reveal a resorptive phenotype, which could be augmented. Moreover, there is evidence that Notch signaling is enhanced in human osteosarcoma and the prolonged activation of Notch in osteoblasts results in osteosarcoma in experimental preclinical models [89,90]. Although there are no reports of osteosarcoma in HCS, it would seem prudent not to treat with teriparatide a population with a theoretical risk of osteosarcoma. Since Notch2HCS mutant mice exhibit increased osteoclastogenesis due to enhanced expression of Rankl by osteoblasts, a treatment to consider is denosumab, which by binding RANKL can reduce osteoclast formation. There is a case report demonstrating the beneficial effects of denosumab in HCS [35]. In this woman, there was an increase in femoral and lumbar spine bone mineral density, and fractures did not occur while the subject was receiving denosumab. It is of interest that the individual reported had increased serum levels of RANKL, confirming that increased osteoclastogenesis and bone resorption are mechanistically relevant to the skeletal manifestations of HCS.

9. Conclusions

In conclusion, HCS is a rare genetic disorder associated with gain-of-function mutations in exon 34 of NOTCH2. HCS influences the development of multiple organs and affected subjects present with fractures, acroosteolysis and classic craniofacial features. Neurological complications can be fatal. Preclinical murine models have offered an understanding of the pathogenesis of the syndrome. Treatment is defined by the organs affected.

Abbreviations

ANK

Ankyrin

CADASIL

cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy

CSL, CBF1

Suppressor of Hairless Lag 1

CHYS1

chondroitin sulfate synthase

Dll

Delta-like

EGF

epidermal growth factor

EOGT

EGF domain specific O-linked N acetylglucosamine transferase

HCS

Hajdu Cheney Syndrome

Hes

hairy and enhancer of split

Hey

HES-related with YRPW

HD

heterodimerization domain

Jag

Jagged

LP

leader peptide

LFNG

Lunatic fringe

LMS

Lateral Meningocele Syndrome

LNR

LIN-12 Notch repeats

MZ

marginal zone

Maml

mastermind-like

MESP

mesoderm posterior bHLH transcription factor

μCT

microcomputed tomography

NRR

negative regulatory region

NICD

Notch intracellular domain

NLS

nuclear localization sequence

PEST

proline (P), glutamic acid (E), serine (S) and threonine (T)

RAM

Rbpjκ-association module

Rankl

receptor activator of nuclear factor kappa B ligand

Rbpjκ

recombination signal-binding protein for Ig of κ region

TMD

transmembrane domain