Chediak–Higashi Syndrome (CHS) is a very rare autosomal recessive disorder. The syndrome is characterized by severe immune deficiency, oculocutaneous albinism, bleeding tendencies, recurrent pyogenic infection, progressive neurologic defects and a lymphoproliferative syndrome (1–3). The classic diagnostic feature of the syndrome is the presence of giant granules within various cells of the body (Fig. 1). These granules include vesicles such as lysosomes, melanosomes, cytolytic granules and platelet dense bodies. The fact that all of these organelles are affected suggests a common origin or regulation mechanism in their biogenesis. There are different organellar diseases in humans that affect lysosomes, melanosomes, cytolytic granules and platelet dense bodies (Table 1 (4–6). Some of the identified mutant genes have functions relating to vesicular trafficking. Other protein functions, such as Hermansky–Pudlak Syndrome (HPS)-1, remain to be elucidated. The gene responsible for CHS was cloned in 1996 (7–11). However, the function of the protein remains unknown. This review will describe some of the clinical features of this disease, the cloning of the CHS gene and homologs in other species, the cellular defects associated with the disease and the possible role CHS plays in vesicle formation and maintenance.
Clinical Features of CHS
The diagnosis of CHS is usually correlated with the presence of hypopigmentation and frequent, sometimes fatal, pyogenic bacterial infections. The skin, hair and eye pigmentation may be altered. Hypopigmentation of the eyes can result in photophobia and decreased visual acuity (12–14). Cells are capable of making pigment and the pigment is present in melanosomes; however, the melanosomes are abnormally large (15, 16). It is thought that the increased size of the melanosomes inhibits their migration to the cell surface and they are therefore unable to release their contents for absorption by the keratinocytes.
CHS patients may show a mild coagulation defect with bruising, mucosal bleeding and petechiae (17, 18). Normal coagulation occurs in waves with the release of storage pool constituents from platelet dense bodies controlling aggregation. Although platelet numbers are normal in CHS, platelets are defective in coagulation (19). Platelets in CHS patients contain abnormal platelet dense bodies (20, 21). These dense bodies normally contain serotonin, calcium, adenosine diphosphate (ADP), adenosine triphosphate (ATP), and pyrophosphate (22) and are responsible for the second wave of platelet aggregation. Release of storage pools requires regulated secretion and CHS platelet dense bodies appear to be delayed in secretion resulting in increased bleeding times (17).
Patients with CHS have recurrent pyogenic infections. The most common infections occur in the skin and respiratory tract (23, 24). Macrophages and particularly polymorphonuclear leukocytes (PMNs) show decreased chemotaxis (25–27). The phagocytic ability of CHS PMNs is normal, but there is a delay in fusion of phagosomes with lysosomes. This delay permits bacterial replication and escape leading to persistent infections. The current treatment for infections is antibiotics. This is, however, only a temporary solution. If patients survive the onslaught of infections they may then exhibit an `accelerated phase' lymphocyte infiltrate into the major organs of the body (23, 28). The accelerated phase is believed to be an uncontrolled T-cell activation/proliferation (23, 29). A correlation between Epstein–Barr virus infection and the onset of the accelerated phase was shown in eight CHS patients (30), suggesting that a viral insult may be responsible for the accelerated phase and corresponding T-cell proliferation. CHS patients have defective cytolytic T-cell function (31–34), impaired natural killer (NK) cell function (35), and slowed neutrophil and monocyte migration (36, 37).
The treatment for CHS is bone marrow transplant, which alleviates the immunologic problems, the most life-threatening defects of CHS (38–42). Patients who live longer then manifest neurologic defects. Typically, the neurologic symptoms include ataxia and gait problems, weakness, sensory deficits, and advancing neurodegeneration (43, 44). Microscopic examination of neurons revealed cytoplasmic inclusions resembling enlarged lysosomes, similar to that observed in all other cell types of CHS patients (45, 46). These enlarged vesicles may eventually impair synaptic transmissions.
Identification of the CHS/Beige Gene
The beige mouse has long been thought of as a homolog of CHS and this was formally demonstrated using somatic cell fusion (47). Wild type human fibroblasts were able to complement the enlarged lysosome phenotype of mouse beige fibroblasts, while CHS fibroblasts were not able to complement the beige fibroblasts nor Aleutian mink fibroblasts. These studies demonstrated that the beige/Aleutian mink mutation was in the orthologous gene to the CHS mutation. The Beige gene had been localized to chromosome 13 (48) and the CHS gene to chromosome 1q42 (49, 50). Investigators used traditional positional cloning approaches to narrow the region of the Beige gene. Two different groups simultaneously cloned the Beige gene using either a strict positional cloning strategy (7) or positional cloning combined with YAC complementation (8). The cloning of Beige and the generation of Beige cDNAs allowed for the identification of a candidate gene for CHS (7, 9). CHS1 is among the largest genes identified in the human genome with a messenger RNA (mRNA) of approximately 11.5 kb. The gene encodes a protein of 3801 amino acids with a predicted molecular weight of 430 kDa. The CHS1 and Beige protein sequences are 85% identical. Mutations in CHS1 span the entire gene with frameshift mutations identified at codons 40, 489, 633, 1026 and 3197 and non-sense mutations at codons 50, 566, and 1029 (1) as well as missense mutations (51). Together these data establish that the full-length protein is needed for function.
Several animal models of CHS have been identified, including the beige mouse (52), the beige rat (53), the Aleutian mink (23), CHS in cats (54), a reported case of CHS in killer whales (55), and Chediak in cattle (23). The bovine Chediak gene has been sequenced and a mutation identified in which a single amino acid change from an arginine to a histidine gives rise to the Chediak phenotype (56, 57). This arginine residue is conserved in CHS1 genes in other species. It is unknown if this amino acid change results in an unstable protein thus giving rise to the phenotype. This suggests that either this region of the protein is important for function or the region is required for the stability of the protein.
CHS1/Beige Protein Characterization
The CHS1 protein is predicted to be a cytosolic protein with three to four defined domains including a weak ARM/HEAT repeat domain containing alpha-helices, a possible perilipin domain, a BEACH domain (BEige And CHediak) containing the amino acid sequence WIDL and a series of WD-40 repeats at the carboxyl terminus (Fig. 2). ARM motifs are suggested to mediate membrane associations (58) and HEAT repeats are thought to play a role in vesicle transport (59, 60). The perilipin domain is suggested to be a lipid association domain (61) and is found in another vesicle trafficking protein, TIP47 (62, 63). The BEACH domain has no known function but is found in several proteins in eukaryotes (3, 64). The WD-40 repeat domain has been identified as a protein–protein interaction domain (65).
Database searches have identified a family of eukaryotic proteins that contain the BEACH and WD-40 repeat domains. There are homologs in all eukaryotes examined with several family members present in mammals, Drosophila melanogaster, Caenorhabditis elegans, Dictyostelium discoidium and one homolog in Saccharomyces cerevisiae (Fig. 3). Some of the identified CHS1 homolog have been further analyzed. Neurobeachin, a mammalian homologs, has been localized to the trans-Golgi and has been characterized as a protein kinase A anchoring protein (66). FAN, another mammalian family member, is involved in neutral sphingomyelinase activation (67). No function has been described for CDC4L (68). One of the D. melanogaster homologs, dAKAP550, has also been shown to be a protein kinase A anchoring protein (69), similar to Neurobeachin. Studies in D. discoidium have identified six BEACH/WD40 containing proteins, LvsA-F (64). Two of these proteins, LvsA and LvsB, have been further characterized. D. discoideum has a contractile vacuole complex that serves to excrete cytoplasmic water through a vesicular trafficking and insures survival in a hypo-osmotic environment. LvsA protein has been shown to be associated with the contractile vacuole of D. discoidium (70). LvsB, which has the most homology to the CHS1/Beige protein, affects vacuole morphology. The lvsB-null mutants contain enlarged acidic lysosomes, similar to Chediak or beige and the authors suggest that the LvsB protein may regulate vesicle fusion (71). Recently, a D. melanogaster homolog was disrupted and analyzed for developmental defects. A mutation in CG9011/DBEACH1 results in defects in bristle formation (72). This is the first reported information that a CHS homolog may be involved in development.
Somatic cell fusion experiments revealed that CHS and beige fibroblast lysosomes are capable of fusion (47). Mutant cells fused to other mutant cells by UV-Sendai virus showed lysosome mixing (fusion). Mutant cell lysosomes were capable of fusion with wild type or mutant lysosomes. No alterations in the kinetics of lysosome fusion were observed (D.M. Ward and J. Kaplan, unpublished). It is of interest to note that in fusions of CHS/beige cells with wild type cells the mutant lysosomes required fusion with wild type lysosomes in order to regain wild type morphology. These data suggest that the CHS1 protein may either associate with lysosomes at some point or alter some property of lysosomes that is essential to maintain normal lysosome size/distribution. Overexpression of the Beige protein results in smaller than normal lysosomes, suggesting a role for Beige/CHS1 in vesicle fission (73).
It was hypothesized that the cellular defect in CHS/beige might be caused by a microtubule defect (74–78). Vesicles are often transported along microtubules using retrograde and anterograde microtubule-associated motor proteins. Microtubules are normal in CHS/beige cells. Perou and Kaplan demonstrated that microtubule-associated proteins appear to function normally in CHS/beige fibroblasts (79). That is, lysosomes could move to the periphery and back to a perinuclear region in the absence of the CHS/Beige protein. These data rule out a possible defect in microtubules and microtubule-associated proteins. A previous report suggested that the CHS1/Beige protein was associated with microtubules (80). Those results have been proven incorrect (W. Faigle, personal communication), and any association of the CHS1/Beige protein with membranes or cytoskeletal elements has not been demonstrated. To date the function of the CHS1/LYST/Beige protein remains unknown.
Cellular Defects Associated with CHS
Investigators have used cell biology and biochemical techniques to try to determine the function of the CHS1/LYST/Beige protein. Some cell types have unique organelles (i.e. secretory granules, melanosomes) that perform specific functions. Because these organelles are believed to be of the same lineage as lysosomes, investigators have studied cells with these specific organelles. For example, cytotoxic T lymphocytes (CTLs) from CHS patients show abnormal secretory granules with only one to two enlarged secretory granules/lysosomes. Normally CTLs contain several small secretory granules. Baetz et al. (31) and Stinchcombe et al. (34) have shown that the CHS CTLs are unable to secrete the giant granules that contain the lytic proteins. Stinchcombe et al. (81) further demonstrated that early secretory granule biogenesis is normal in CHS. However, granule maturation is altered and CHS granules appear to fuse and generate the abnormally large granules.
Antigen presentation is dependent upon appropriate vesicular trafficking. Lem et al. (82) demonstrated that the class II loading compartment of CHS antigen-presenting cells was enlarged and that antigen processing was delayed. Faigle et al. (80) also showed a delay in antigen presentation in lymphocytes from CHS patients. These data may explain the observed defects in host defense mechanisms but they do not ascribe a function to the CHS1/LYST/Beige protein.
Changes in melanosome content, morphology, or location serve as classic identifiers of pigment disorders (83, 84) and indeed melanosomes are large and clustered near the nucleus of melanocytes in CHS. Normal pigmentation occurs by the biogenesis of melanosomes, the movement of melanosomes to the periphery of the melanocyte and release of the melanosome to the epithelium or keratinocyte. Studies have demonstrated that enlarged melanosomes in CHS/beige melanocytes contain pigments (16). Enlarged melanosomes in CHS appear to be the result of an aberrant fusion similar to observations made in CTLs (34), mast cells (85), and pancreatic acinar cells (86) where abnormal fusion of specific vesicles has been observed.
The underlying biochemical defect in CHS/beige has yet to be identified. Many studies have speculated that alteration in cyclic nucleotides (75), protein kinase C (PKC) levels (87–91), or lipid turnover/composition (91–95) may be the underlying biochemical defect. Previous studies have shown that PKC is abnormally rapidly down-regulated in beige PMNs and NK cells (89). Investigators have also shown that treating cells with inhibitors of PKC proteolysis could eliminate the rapid PKC down-regulation (90, 96). The relationship between the down-regulation of PKC and the lysosomal abnormalities in beige/CHS remains unknown. Tanabe and colleagues reported increases in ceramide production in beige fibroblasts and suggest that this results in the rapid down-regulation of PKC (91). They further showed that treatment of normal fibroblasts with a PKC inhibitor, chelerythrin, resulted in enlarged lysosomes clustered in a perinuclear region. These data suggest that in the absence of the Beige/CHS protein or when PKC levels are misregulated lysosome morphology can be altered. It is clear that the absence of the CHS/Beige protein affects the origin and/or maintenance of a specific subset of vesicles. How the CHS/Beige protein functions in this pathway remains to be determined.