Hypopigmentation in Hermansky–Pudlak syndrome


  • Ai-Hua Wei,

    1. State Key Laboratory of Molecular Developmental Biology, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing, China
    2. Department of Dermatology, Beijing Tongren Affiliated Hospital of Capital Medical University, Beijing, China
    Search for more papers by this author
  • Xin He,

    1. State Key Laboratory of Molecular Developmental Biology, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing, China
    Search for more papers by this author
  • Wei Li

    Corresponding author
    • State Key Laboratory of Molecular Developmental Biology, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing, China
    Search for more papers by this author

Correspondence: Wei Li, Ph.D., Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, 1 West Beichen Road, Chaoyang District, Beijing 100101, China. Email: wli@genetics.ac.cn


Hermansky–Pudlak syndrome (HPS) is characterized by oculocutaneous albinism, bleeding tendency, and ceroid deposition which often leads to death in midlife. Currently, nine genes have been identified as causative for HPS in humans. Hypopigmentation is the prominent feature of HPS, attributable to the disrupted biogenesis of melanosome, a member of the lysosome-related organelle (LRO) family. Current understanding of the cargo transporting mechanisms into the melanosomes expands our knowledge of the pathogenesis of hypopigmentation in HPS patients.


Hermansky–Pudlak syndrome (HPS, Online Mendelian Inheritance in Man [OMIM] no. 203300)[1] is a genetically heterogeneous, autosomal recessive inherited disorder characterized by a triad: oculocutaneous albinism, bleeding tendency and ceroid deposition. The highest prevalence region of HPS is in Puerto Rico with founder effects. The incidence rate is estimated as 1:1800 with a carrier rate of 1:21 in Puerto Rico.[2] HPS-1 and HPS-3 are the two common subtypes of HPS in this region.[3] Non-Puerto Rican HPS patients are scattered in many populations as listed in the HPS Database (HPSD; http://liweilab.genetics.ac.cn/HPSD/).[4] HPS-1 is the relatively common subtype in Japanese[5] and Chinese oculocutaneous albinism (OCA) patients.[6, 7]

Nine genes (HPS1, AP3B1 and HPS39) have been identified as causative genes for HPS in humans (Table 1).[2, 3, 8-13] The application of exome sequencing on undiagnosed HPS samples provides a reasonable method of diagnosis and uncovering novel HPS genes. This has been successful in the identification of a HPS-9 patient[14] and a HPS-2-like patient.[15] The HPS genes revealed in mouse and other species suggest additional causative genes for HPS.[16, 17] The key pathological feature of HPS is the disrupted biogenesis and/or function of specialized lysosomes which are termed lysosome-related organelles (LRO), including melanosomes, platelet dense granules and conventional lysosomes.[16, 18-21]

Table 1. The identified nine human Hermansky–Pudlak syndrome (HPS) genes
Human locus (reference sequence)Chromosomal locationOnline Mendelian Inheritance in Man no.HPS subtypeMouse mutantProtein function
HPS1 (NM_000195)10q24.2604982HPS-1Pale-ear (ep)BLOC-3 subunit
HPS2/AP3B1 (NM_003664)5q14.1608233HPS-2Pearl (pe)AP-3 subunit
HPS3 (NM_032383)3q24606118HPS-3Cocoa (coa)BLOC-2 subunit
HPS4 (NM_022081)22q12.1606682HPS-4Light-ear (le)BLOC-3 subunit
HPS5 (NM_181507)11p15.1607521HPS-5Ruby-eye 2 (ru2)BLOC-2 subunit
HPS6 (NM_024747)10q24.32607522HPS-6Ruby-eye (ru)BLOC-2 subunit
HPS7/DTNBP1 (NM_032122)6p22.3607145HPS-7Sandy (sdy)BLOC-1 subunit
HPS8/BLOC1S3 (NM_212550)19q13.32609762HPS-8Reduced pigmentation (rp)BLOC-1 subunit
HPS9/PLDN (NM_012388)15q21.1604310HPS-9Pallid (pa)BLOC-1 subunit

The most common symptoms of HPS are hypopigmentation, loss of visual acuity, prolonged bleeding, colitis and, in some cases, fatal lung disease.[18] Hemophagocytic lymphohistiocytosis[22] and immune deficiency[23] have been reported in HPS-2 patients. Neuronal symptoms are described in AP-3-deficient mice and dysbindin-null sdy mice,[24, 25] although no apparent neurological symptoms have been documented in HPS-2 or HPS-7 patients. Other tissue-specific phenotypes have been described in HPS mouse mutants.[21] These phenotypes are caused by the disruption of tissue-specific LRO.[16]

The gold standard in diagnosing HPS is the absence of platelet dense granules under electron microscopic (EM) examination. Symptoms of hypopigmentation and bleeding tendency or bruises support the diagnosis. Molecular diagnoses are now available with the identification of nine HPS genes in humans (Table 1). Chediak–Higashi syndrome (CHS, OMIM no. 214500) exhibits similar defects on the biogenesis of LRO due to the mutation on the CHS1/LYST gene.[26] Griscelli syndrome (GS, OMIM no. 214450) also presents with hypopigmentation but lacks defects in platelet dense granules. GS is caused by mutation of GS1/MYO5A,[27] GS2/RAB27A[28] or GS3/MLPH.[29] Molecular diagnosis and EM examination of platelet dense granules in CHS or GS patients allow accurate differentiation from HPS.

There is currently no cure for HPS. Only symptomatic treatment of the disease exists. Variable response to desmopressin (1-deamino-8-D-arginine vasopressin, DDAVP) treatment has been reported in the correction of bleeding time.[30]

Prolonged bleeding often requires multiple platelet transfusions, and the fibrotic lung disease may lead to death in midlife.[31, 32] Patient-derived induced pluripotent stem cell technology and transdifferentiation technology may be applied to the treatment of HPS in recovering vision loss or treating lung fibrosis in the future.


The identification of HPS1 through positional candidate cloning[2] has prompted the identification of another eight human HPS genes thereafter. The HPS proteins have been grouped into several lysosomal trafficking protein complexes such as AP-3, and BLOC-1, -2 and -3 (Table 1).[16, 33] The mutational alleles of these HPS genes have been collected in the HPSD (http://liweilab.genetics.ac.cn/HPSD/). This database is updated in a timely manner. Notably, most of the homozygous or compound heterozygous mutations are frame-shift or nonsense mutations that cause the loss of the protein and lead to the destabilization of other subunits of the complex in which they reside. The genotype–phenotype relationship remains to be investigated.


The characterization of 15 mouse HPS mutants[21] lead to a series of successful identifications of murine and human HPS genes through positional cloning.[16] Except for the nine mouse HPS genes listed in Table 1, an additional six genes (Ap3d, Rabggta, Vps33a, Cno, Muted, Kxd1) have been identified as causing mouse HPS[16, 34] and are listed in the HPSD (http://liweilab.genetics.ac.cn/HPSD/). Currently, the 15 cloned mouse HPS genes are known to reside in several complexes to mediate the biogenesis of very specialized LRO. These complexes include the AP-3 complex, the HOPS complex, and the BLOC-1, -2 and -3 complexes.[16, 33] Each mouse HPS mutant shows hypopigmentation in coat color or eye color with the BLOC-1 mutants more severely affected.[16, 35] One intriguing observation is the selective loss of pigment in the ears and tail of the BLOC-3 mutant (ep or le) while the eye color is black with giant choroidal melanosomes.[35] Coat color changes in the double or multiple mouse mutants with the combinations of different HPS protein complexes reveals epistatic or synergistic effects.[35, 36] Other putative murine HPS mutants would be expected to develop hypopigmentation when the remaining subunits of the above-mentioned five complexes, for example, snapin, BLOS1 and BLOS2 in the BLOC-1 complex, are deficient. Our recent study showed that KXD1 is a BLOC-1 interactor and the Kxd1–KO mice have features of mild form HPS.[34] The mouse HPS mutants are excellent resources to study the underlying mechanism governing the melanosomal biogenesis.


With the discovery of murine and human HPS genes, three new biogeneses of lysosome-related organelle complexes (BLOC-1, -2 and -3) have been defined. These are: pallidin, muted, dysbindin, CNO, BLOS1, BLOS2, BLOS3 and snapin in BLOC-1;[11, 37-40] HPS3, HPS5 and HPS6 in BLOC-2;[10, 41, 42] and HPS1 and HPS4 in BLOC-3.[9, 43-45] The biochemical features and assembling machineries of these complexes remain to be defined, although some pioneer studies have revealed the binding domains in building these BLOC(k)s[46] and the linear assembly of the BLOC-1 complex in vitro.[47] Together with the AP-3, HOPS and ESCRT complexes, the BLOC complexes function in endolysosomal trafficking to mediate the LRO biogenesis. However, how these complexes work together to regulate the transport of melanosomal proteins is still a mystery.[48]

The study of melanosomal biogenesis provides an excellent example of LRO biogenesis.[49, 50] PMEL is thought to be a key molecule in melanosomal assembly by generating internal matrix fibers in early-stage melanosomes.[51] Other melanosomal components involved in melanin synthesis in late-stage melanosomes include tyrosinase (TYR), OCA2, TYRP1, SLC45A2, SLC24A5, DCT and ATP7A.[52, 53] How these melanosomal proteins are transported into melanosomes is beginning to be understood. During the maturation of cutaneous melanosomes, stages I–IV are characterized based on the ultrastructural analyses.[54] The melanosomal proteins are transported into different stages via different transporting machineries regulated by HPS genes.[19]

Sorting signals of the cargo proteins serve as the recognition tags for the transporting complexes. TYR, OCA2 and TYRP1 all contain acidic dileucine (aLL) based consensus motifs which can be recognized by AP-1, -2 or -3. Different aLL signals in OCA2 determine the binding affinities with AP-1 and -3.[55] Although DCT lacks the cytoplasmic LL motif, it binds to AP complexes to be sorted into mature melanosomes.[56]

PMEL is sorted into early-stage melanosomes dependent on AP-2[57] and CD63,[58] but not dependent on BLOC-2.[59, 60] As summarized in a recent review,[17] during late-stage melanosomal maturation, TYR is sorted into the AP-3 buds of vacuolar endosomes, where TYR is directed into melanosomes through a BLOC-1-independent pathway. BLOC-2 and Rab38 may act downstream of this AP-3-dependent pathway. A portion of TYR that is sorted into AP-1 buds may compensate for its targeting to melanosomes when AP-3 is deficient. In contrast, OCA2 is sorted into AP-1 or -3 buds and thereafter is directed by BLOC-1 into mature melanosomes. BLOC-2 and RAB38 act downstream of this BLOC-1-dependent pathway. However, TYRP1 sorting signals interact with AP-1 but not AP-3. Therefore, TYRP1 is sorted into mature melanosomes in a similar way as OCA2 via the BLOC-1-dependent pathway, but independent of AP-3. In addition, ESCRT-1 is required for TYRP1 targeting to the melanosomal limiting membrane.[61] Similar to TYRP1, the copper transporter, ATP7A, is sorted into melanosomes via the BLOC-1-dependent pathway.[52] Thus, TYRP1, OCA2 and ATP7A are routed to mature melanosomes in the same pathway.

How two other pre-melanosomal proteins (GPR143 and MART-1) and two other transporters (SLC45A2 and SLC24A5) are sorted into melanosomes remains unclear. On the other hand, although BLOC-3 has been described in regulating melanosomal biogenesis,[16, 18] its exact mechanism in melanosomal protein trafficking is still unknown. It has been suggested that BLOC-3 acts as the RAB9 effector to regulate melanosomal biogenesis.[62] The VPS33A in the HOPS complex may function in the docking of AP-3-coated vesicles during melanosomal biogenesis through the interaction between VPS41 and AP3D.[63, 64] The defects of melanosomes in the buff/Vps33a mutant are less severe than those in BLOC-1, BLOC-2 or AP-3 mutants.[16] However, the defects of melanosomes in either ep/Hps1[35] or le/Hps4[16] are quite different by showing enlarged melanosomes in the choroids or macroautophagosomal structures in HPS-1 skin melanocytes.[65] These suggest that BLOC-3 may play a unique role during melanosomal biogenesis which remains to be investigated. The melanosomal biogenesis pathways regulated by HPS genes are depicted in Figure 1, which is modified from the model by Sitaram and Marks.[17]

Figure 1.

Sorting pathways for melanosomal biogenesis. At the sorting station (TGN or sorting endosomes), the melanosomal proteins are sorted into immature melanosomes and mature melanosomes separately. Different melanosomal proteins (cargoes, listed in right box in different shapes) bind different adaptor proteins (left box in light-blue color with different shapes) in melanosomal sorting. The defined cargoes in sorting-in pathways are depicted in red color, while the cargoes with unknown sorting mechanisms are labeled in green color. Arrows with dotted lines indicate the sorting routes. Arrows in black lines denote the directions in transport. Cargo–adaptor complexes are grouped with dotted circles. The circles with question marks represent undefined position or action. The details of these pathways are described in the text.


Current understanding of the mechanisms for melanosomal biogenesis has gained insights into the pathogenesis of HPS. On the other hand, the biogenesis of melanosomes has provided important clues to understand the sorting mechanisms in the biogenesis of non-melanosomal LRO. With the elucidation of the LRO biogenesis, general mechanisms regulated by HPS could be inferred, as it has been shown that BLOC-1 and AP-3 function together in different cell types such as melanocytes,[66] neurons,[67] platelets[68] and HeLa cells.[69] It is challenging to define how different transporting complexes synergistically or sequentially act on the cargoes during their transport into melanosomes. Systematic approaches such as organellar proteomics and expression profiling in melanosome maturation and melanocyte differentiation should be used to better understand the disciplines in melanosome biology.


This work was partially supported by grants from the National Natural Science Foundation of China (nos. 31230046, 81101182, 31071252), from the Chinese Academy of Sciences (no. KSCX2-EW-R-05) and from The State Key Laboratory of Molecular Developmental Biology, China. We apologize to the authors of relevant works that are not mentioned in this review due to limited space.