Phenotypic Mutation 'pam_gray' (pdf version)
Mutation Type nonsense
Coordinate20,017,173 bp (GRCm38)
Base Change A ⇒ T (forward strand)
Gene Hps3
Gene Name Hermansky-Pudlak syndrome 3 homolog (human)
Synonym(s) coa, cocoa
Chromosomal Location 19,995,945-20,035,315 bp (-)
MGI Phenotype Homozygotes for spontaneous null mutations exhibit hypopigmentation and prolonged bleeding associated with a platelet defect.
Accession Number
NCBI RefSeq: NM_080634; MGI: 2153839
Mapped Yes 
Amino Acid Change Tyrosine changed to Stop codon
Institutional SourceBeutler Lab
Ref Sequences
Y600* in Ensembl: ENSMUSP00000012580 (fasta)
Gene Model not available
SMART Domains

low complexity region 537 551 N/A INTRINSIC
Phenotypic Category Autosomal Recessive
Penetrance 100% 
Alleles Listed at MGI

All alleles(7) : Spontaneous(6) Chemically induced(1)

Lab Alleles
AlleleSourceChrCoordTypePredicted EffectPPH Score
IGL00545:Hps3 APN 3 20019807 missense possibly damaging 0.94
IGL00846:Hps3 APN 3 20025792 missense probably benign 0.00
IGL01320:Hps3 APN 3 20030469 missense probably benign 0.12
IGL01364:Hps3 APN 3 20003305 missense possibly damaging 0.58
IGL01751:Hps3 APN 3 20010966 missense probably damaging 1.00
IGL01843:Hps3 APN 3 20029001 missense probably benign 0.05
IGL02294:Hps3 APN 3 20014048 missense probably damaging 1.00
IGL02581:Hps3 APN 3 20003221 unclassified noncoding transcript
blue UTSW 3 20030796 missense probably damaging 1.00
earl_grey UTSW 3 20017173 nonsense
gandalf UTSW 3 20012796 nonsense probably null
R0107:Hps3 UTSW 3 20030796 missense probably damaging 1.00
R0245:Hps3 UTSW 3 20012796 nonsense probably null
R0421:Hps3 UTSW 3 20029316 missense probably benign 0.00
R0524:Hps3 UTSW 3 20012776 missense probably damaging 1.00
R0763:Hps3 UTSW 3 20003279 missense probably damaging 1.00
R1795:Hps3 UTSW 3 20012695 critical splice donor site probably null
R1864:Hps3 UTSW 3 20019959 critical splice acceptor site probably null
R2029:Hps3 UTSW 3 20030527 missense probably benign 0.01
R2101:Hps3 UTSW 3 20012783 missense possibly damaging 0.95
R2221:Hps3 UTSW 3 20002363 missense probably benign
R2268:Hps3 UTSW 3 20012935 splice site noncoding transcript
R2520:Hps3 UTSW 3 20029030 missense probably damaging 1.00
R3809:Hps3 UTSW 3 20018812 missense probably damaging 1.00
R3888:Hps3 UTSW 3 20003223 critical splice donor site probably null
R3942:Hps3 UTSW 3 19996939 missense probably damaging 1.00
R4022:Hps3 UTSW 3 20035261 missense possibly damaging 0.69
R4156:Hps3 UTSW 3 20029229 missense probably damaging 1.00
R4739:Hps3 UTSW 3 20030410 missense probably null
R4823:Hps3 UTSW 3 20012726 missense probably benign 0.03
R4912:Hps3 UTSW 3 20014173 missense probably damaging 1.00
R5307:Hps3 UTSW 3 20012701 missense possibly damaging 0.89
R5859:Hps3 UTSW 3 20008870 missense probably benign 0.02
R6140:Hps3 UTSW 3 19996987 missense probably damaging 1.00
R6183:Hps3 UTSW 3 20008868 missense probably benign 0.04
X0021:Hps3 UTSW 3 20030749 missense probably benign 0.14
X0066:Hps3 UTSW 3 20015988 missense probably damaging 1.00
Mode of Inheritance Autosomal Recessive
Local Stock Sperm, gDNA


Last Updated 2016-10-11 6:15 PM by Bruce Beutler
Record Created unknown
Record Posted 2008-07-23
Phenotypic Description

Pam gray was initially identified among N-ethyl-N-nitrosourea (ENU)-induced G3 animals as a hypopigmentation mutant (Figure 1). Pam gray homozygous animals have a gray coat, display normal MCMV resistance (see MCMV Susceptibility and Resistance Screen), and do not have any Toll-like receptor (TLR) signaling defects (see TLR Signaling Screen). Pam grey is allelic to cocoa (1). A second mutant with gray coat color, designated earl grey, was isolated independently as a descendant of the same ENU-mutagenized G0 male that gave rise to pam gray.  However, earl grey arose from a different G1 male founder than pam gray.  The Hps3 mutation found in pam gray was ultimately identified in earl grey.

Nature of Mutation
The pam gray mutation was mapped to Chromosome 3, and corresponds to a T to A transversion at position 1815 of the Hps3 transcript, in exon 10 of 17 total exons.
595  -S--E--S--Q--K--Y--E--R--G--L--V-
The mutated nucleotide is indicated in red lettering, and creates a premature stop codon at position 600 of the reading frame (normally a tyrosine) deleting 403 amino acids from the C-terminus of the protein.
Protein Prediction
Mouse Hps3 encodes a 113.1 kDa protein of 1002 amino acids (1).  The translated protein products of mouse Hps3 and its human homolog are 80% identical. The human protein is 113.7 kDa in molecular mass and 1004 amino acids long (2).  HPS3 forms part of the biogenesis of lysosome-related organelle complex 2 (BLOC-2; see Figure 2 and Background).
Figure 1. A, Predicted domains of HPS3. HPS3 contains a putative clathrin-binding (CB) motif, two consensus dileucine-based sorting motifs and an endoplasmic reticulum (ER) membrane retention signal. Amino acid locations are based on the human protein. The location of the pam gray mutation is indicated in red. Image is interactive: click to see other mutations in Hps3 (red indicates phenotypic mutations; green are incidental mutations). B, Components of the biogenesis of lysosomal-related organelle complex 2 (BLOC-2). All three proteins have been shown to co-immunoprecipitate, but only HSP5 and HSP6 bind together in two-hybrid studies suggesting the presence of unknown components of the complex (?).


The HPS3 protein is unrelated to any other known proteins, and contains a possible tyrosine phosphorylation site at codon 295 (1).  The human HPS3 protein is predicted to be 43% α-helix, 19% extended strand, 30% random coil, and 7% β-turn.  Human HPS3 contains a clathrin-binding motif at residues 172 to 176 (LLDFE), 2 consensus dileucine signals (GEKAELL, residues 542-548 and LEPRLLI, residues 711-717), and 12 tyrosine-based sorting signals for targeting to vesicles of lysosomal lineage (2-5).  Clathrin is the main component of protein coats that assist in the formation of vesicles budding from the trans-Golgi network (TGN), plasma membrane, and endosomes, and form clathrin triskelions composed of three heavy chains and three light chains (6).  The clathrin-binding motif is also found in the β subunits of the adaptor protein complexes (see record for bullet gray), and has been shown to be sufficient for binding to the amino terminal domain of the clathrin heavy chain (3).  Human HPS3 also contains a potential endoplasmic reticulum membrane retention signal (KKPL, residues 1000-1003) similar to dilysine motifs present in the carboxy-terminal regions of type Ia membrane proteins that traffic between the endoplasmic reticulum and the Golgi apparatus (7).  Although not always identical, most of these motifs have been conserved in the mouse protein (1;2;8).  A peroxisomal matrix targeting signal (RLDSQHSHL) is found at residues 565-574 of the human protein, but this motif is not conserved in the mouse and may not be significant due to its location in the middle of the protein. This motif is known to function in the amino-terminus of peroxisomal matrix proteins (9). 
The pam gray mutation results in protein truncation after amino acid 599.  The truncated protein would retain the clathrin-binding motif and most of the tyrosine-based sorting signals, but it is unknown whether this protein is expressed and localizes appropriately.
Northern blot analysis in humans demonstrated a 4.4 kb message in all tissues tested, including heart, brain, placenta, lung, liver, skeletal muscle, kidney, and pancreas.  Expression was greatest in kidney, with strong bands in liver and placenta (2).
In mice, Northern blot analysis indicates a 4.0 kb transcript is expressed in all mouse tissues and cell types tested, including heart, brain, spleen, liver, lung, kidney and testis.  Expression is very low in skeletal muscle (1).
Because HPS6, another member of the BLOC-2 complex, can be immunoprecipitated from both membrane-associated and soluble fractions of cell lysates, it has been inferred that all components of the BLOC-2 complex, including HPS3, may be both cytosolic and membrane-associated (10).  A FLAG-tagged HPS3 protein transfected into NIH3 and a melanocyte cell line was found in a cytoplasmic distribution with a somewhat reticular appearance and slight perinuclear concentration.  The tagged protein did not colocalize with markers of the endoplasmic reticulum, vesicles and organelles, or the TGN and endosomes (1).  Another study using transfected GFP-tagged HPS3 found that the tagged protein partially co-localized with clathrin predominantly on small vesicles in the perinuclear region following a low-temperature block (8).
Hermansky-Pudlak syndrome (HPS; OMIM #203300) is a disorder with an array of clinical symptoms that are caused by alterations at numerous independent loci.  HPS is commonly characterized by oculocutaneous albinism (OCA), prolonged bleeding, and lysosomal hyposecretion.  In addition, subsets of patients exhibit pulmonary fibrosis, immunodeficiency, and inner-ear abnormalities.  Most of these conditions are known to arise from defects in the biogenesis and/or function of lysosome-related organelles, such as melanosomes and platelet dense granules (11).  These organelles are cell-type specific modifications of the post-Golgi endomembrane system, and may share various characteristics with lysosomes, such as integral membrane proteins and an acidic intralumenal pH (12).  At least eight types of human HPS have been described, and mutations affecting at least 15 loci in mice create HPS-like disease.  The human HPS loci and their mouse equivalents are as follows: HPS1/pale ear (13); HPS2/pearl (14); HPS3/cocoa (2); HPS4/light ear (15); HPS5/ruby-eye 2 (16); HPS6/ruby-eye (16); HPS7/sandy (17); and HPS8/reduced pigmentation (18).  
Most of the genes associated with HPS encode subunits of protein complexes involved in intracellular trafficking, and are important for the trafficking of lysosomes, lysosomal-related organelles or components of these organelles.  Based on biochemical studies, it has been proposed that HPS proteins assemble into four stable complexes, BLOCs (biogenesis of lysosome-related organelle complex) 1, 2, and 3, and the adaptor protein 3 (AP-3) complex (Figure 2).  Mutation of the β3A subunit of the AP-3 complex causes HPS2 (19-22) (mutated in bullet gray).  The following components make up the 200-230 kD BLOC-1 complex: pallidin, dysbindin (HPS7; mutated in salt and pepper), BLOC subunit 1 (BLOS1), BLOS2, BLOS3 (HPS8), cappuccino, muted (mutated in minnie), and snapin.  The proteins HPS3, HPS5 (mutated in toffee and dorian gray), and HPS6 (mutated in stamper-coat) comprise BLOC-2 (350 kD) (10;16;23). BLOC-3 (175 kD) is composed of HPS1 and HPS4 (9;24).  
All of the BLOC-2 components cause HPS in humans and HPS-like disease in mice. While all HPS patients suffer from OCA and prolonged bleeding, different subtypes of HPS are now recognized by their distinct clinical features.  HPS3 patients exhibit a mild form of HPS, with mild oculocutaneous albinism and prolonged bleeding, but no other symptoms (2;25).  Mild forms of HPS also occur in HPS5 and HPS6 patients (2;16;26).  Similarly, ruby-eye (ru), ruby-eye 2 (ru2), and cocoa (coa) mice, with mutations in Hps6, Hps5, and Hps3 respectively, all have very similar to identical phenotypes (16;23;27;28).  Coa mice are characterized by a diluted coat color that darkens with age, light eyes that darken with age to a ruby or maroon color, and platelet storage pool deficiency that causes prolonged bleeding time.  Although platelet numbers are normal, platelets are unable to accumulate dense granule contents resulting in decreased serotonin (28).  In a melanocyte line derived from coa mice, little pigment is observed and melanosomes are reduced in number, small, abnormally shaped, and immature (1).  Similar findings are observed in vivo in melanosomes of the retinal pigment epithelia (RPE) and choroids of the eye (1;23).  In cultured melanocytes derived from human HPS3 patients, melanosomes were immature and unpigmented, but otherwise morphologically normal (29).  Unlike ru and ru2 mice, coa mice have normal secretion of lysosomal enzymes from kidney proximal tubule cells and platelets, suggesting that the HPS3 protein is necessary for melanosomes and platelet dense granules but not lysosomes (28;30).  This result suggests that HPS5 and HPS6 may function independently of HPS3 in some contexts.
Several lines of evidence suggest that HPS3, HPS5 and HPS6 are components of the BLOC-2 complex.  All three proteins coimmunoprecipitate, while in yeast two-hybrid experiments HPS6 was found to bind HPS5 (10;16;23).  Similar interactions were not found for HPS3 in these studies, suggesting the presence of additional, unknown components of the complex (16;23).  Double homozygous combinations of Hps3, Hps5, and Hps6 animals exhibited identical phenotypes to each other and to single homozygotes.  Furthermore, HPS6 and HPS5 proteins are destabilized in protein extracts derived from mice mutated in Hps3, Hps5, and Hps6, but are not destabilized in protein extracts derived from other HPS mouse models (23).
Putative Mechanism
The precise molecular function of the BLOC-2 complex remains unknown.  In addition to a yet undefined role in regulating lysosome-related organelle secretion, recent studies suggest that both HPS3 and HPS5 may also regulate protein trafficking during the maturation of melanosomes (29;31;32).  Both tyrosinase (mutated in ghost) and Tyrp1 (tyrosinase-related protein 1) are reduced in HPS5 mutant melanocyte dendrites as measured by immunofluorescence and immunoelectron microscopy (32).  In HPS3 mutant melanocytes, tyrosinase, Tyrp1, and Tyrp2 (tyrosinase-related protein 2) were mislocalized, as were the lysosome associated membrane proteins, LAMP1 and LAMP3 (29;31).  In addition, HPS3, HPS5, and HPS6 mutant melanosomes are predominantly in the early stages of maturation, suggesting that improper trafficking of melanosome proteins impairs the normal maturation of this organelle (1;16;29;32).  In agreement with this hypothesis, BLOC-1 has been shown to physically interact with BLOC-2 to facilitate Tyrp1 trafficking (33).  Another study demonstrates that BLOC-1 regulates Tyrp1 exit from early endosomes toward melanosomes, but interestingly, BLOC-2 affects Tyrp1 trafficking to melanosomes from an endosomal compartment distinct and downstream from that regulated by BLOC-1 (34). Furthermore, HPS3 is able to bind to clathrin through its clathrin-binding domain (8).  Thus, accumulating evidence suggests a role for BLOC-2 in regulating selective cargo trafficking from endosomes to melanosomes.  The step(s) at which BLOC-2 functions and the compartments between which cargo is transferred remain to be more precisely defined. 
Mutations in some HPS proteins cause immunodeficiency.  For example, in HPS2, mutation of the β3A subunit of the AP-3 complex results in immunodeficiency because of defects in natural killer (NK) cells, cytotoxic T lymphocytes (CTLs) and neutrophils (see bullet gray) (19-22). However, BLOC-1 complex mutant mice pallid, muted and sandy have normal CTL function, as measured by ability to kill targets (35), and patients with HPS7 and HPS8 do not display any immunological defects (17;18).  These results contrast with the MCMV susceptibility and impaired NK cell cytotoxicity of the salt and pepper mutation, which is allelic to sandy.  A similar NK cell defect has been reported for pallid mice (36).  These results are similar to mixed immunological phenotypes seen in mice and patients mutant in BLOC-2 components.  HPS3/5/6 patients do not exhibit immunological defects and ru mice display normal CTL activity (2;16;25;26;35), consistent with the normal MCMV resistance seen in pam gray and stamper-coat mice.  However, toffee mice defective in HPS5 are susceptible to infection by MCMV or Listeria monocytogenes, and ru mice are reported to have an increase in transient fusion events at the plasma membrane of mast cells, resulting in a depletion of secretory products (37).  It is likely this defect also causes the decreased quantities of intragranular components of platelet dense granules seen in Hps3, Hps5, and Hps6 mutants (28;38).
Mouse mutants of BLOC-2 components typically have milder pigmentation defects than other HPS mouse models, consistent with the milder phenotypes seen in patients with HPS3, HPS5 and HPS6 (2;16;23;25;26).  It is possible the truncated HPS3 protein in pam gray mice may retain some function as the clathrin-binding site is retained.  However, the coa5J and the coa6J mutation both result in alterations of the HPS3 gene in exon 10, similar to pam gray.  The coa5J mutation is a nonsense mutation at codon 627, while coa6J is an antisense insertion of an intracisternal A particle (IAP) transposable element into exon 10.  These mutants display identical phenotypes to the original coa mutation, which abolishes the 3’ splice site of intron 14 and uses a cryptic splice site in exon 15.  The aberrant coa mRNA produced by this splicing was nearly undetectable in all tissues, suggesting that the coa mutant does not express any HPS3 protein (1). 
Primers Primers cannot be located by automatic search.
Pam gray genotyping is performed by amplifying the region containing the mutation using PCR, followed by sequencing of the amplified region to detect the single nucleotide change.  The KAPA2G Fast DNA polymerase was needed using the KAPA2G Fast HotStart kit from KAPABIOSYSTEMS.
Primers for PCR amplification
PCR program
1) 95°C             1:00
2) 95°C             0:10
3) 56°C             0:10
4) 72°C             0:15
5) repeat steps (2-4) 40X
6) 72°C             2:00
7) 4°C               ∞
Primers for sequencing
The following sequence of 1694 nucleotides (from Genbank genomic region NC_000069 for linear DNA sequence of Hps3) is amplified:
19893                                    aggcatat taaggtcaat cgagttttgt
19921 atttctgaga accatttttt ttaaattttg tattgattct ttgtgcattt catgtcacgc
19981 actctaatcc cctcatcgcc ctgtcccttc acacccactc ccccccaaaa gaaacaaaca
20041 aaataaaaca aaaaaaatcc acaaaaataa cacaaaacat ctcttcatgg aagctctagt
20101 gtatcccagt gtgtcacaca gtgtacactt ttgtctacac atctttactt gtaaatgttt
20161 attacagcca ggcatggtga tgcacacctt taatcccagc actcgggggg caggggcagg
20221 ggcaggggca ggggcagggg caggggcagg ggcatttctg agtttgaggc cagcctggtc
20281 tacagagtga gtttcaggac atcccagggc tacacagaga aaccctgtct caaaaaaaaa
20341 aaaaccaaat aatgtttatt gcagtgagta attggcatgt gcgacagtat gaacactgca
20401 tgctcaccgg tctgatctcc tgttgtggcc ctgtgtcaga agatcctgaa gctttggatc
20461 tgctgaacca gccccttcac aaggtccatt acttcataga tgaggtagat gttggagcgg
20521 tccaacatct gtatctgccc tgggtctagt tggcagctgg gttggtcagc ctgccaaagc
20581 tgtcccacac cctcacaacc agagccagca ctgtcctggc tagctcaccc agtgctgcag
20641 caggtaagga cccaggccag ctctcctgct ctcctgccct cagggccagc cacactgtgc
20701 tgcccaggtg aagggcgggg ctggcttctc ctcagtgcta taggtgctga gggccaggga
20761 cagctctcca gttctgagaa tcatgcttat agaatttttg tagacagtgt atcataggct
20821 tacttcacct gtgtaagaga ttgtcgatga ctgagatgag gtagactata aatgccattt
20881 ataatattaa atattaatga atgattgctt ttctaccttt gatatgctca ggaaaaatgc
20941 tactgagagt ggctttataa aagtcaagct gtgaggctta tgcctttgaa aagtgtgatc
21001 tgttttatta cttactgtgt taagtagctg ccacacaaat gcacatttga agccctttaa
21061 catttctctc tctctaggct taccactgag caatcccatc tcgccttgcc gtactataag
21121 atgtctggtt tgtctctggc tgaagtcttg gcccgtgtgg actggacaga agagagtgaa
21181 tcacagaaat atgagagagg actagtattt tatattaatc attctcttta tgaaaacctg
21241 gatgaagaat taagtaaagt gaatataatt ttctataaag ttatagcttc tctttttaac
21301 ttttagtagt atttagaatt taagttccaa aattttcttt tttttttttc ctgaaaaaat
21361 tagcatgccc tttacttctt taatggtctt agaattatag cttggaaaag aatttaggtg
21421 tgactttccc atgtgtcctt agtattcagc atggaggcta ttgtggctga ggggcctgcc
21481 caggcacagt gagatctatt ctcaagtgaa gacagaatgc tggtccctgc cttcctctgc
21541 agtatgtttt ttccccataa ctccatgtgc cctattttca acgtcc
PCR primer binding sites are underlined; sequencing primer binding sites are highlighted in gray; the mutated T is shown in red text.
 27.  Silvers, W. K. (1979) The Coat Colors of Mice.
Science Writers Nora G. Smart
Illustrators Diantha La Vine
AuthorsOwen Siggs, Amanda L. Blasius, Bruce Beutler
Edit History
2011-03-24 4:30 PM (current)
2011-03-23 1:19 PM
2011-01-07 9:28 AM
2010-10-11 2:43 PM
2010-10-11 2:43 PM
2010-02-03 4:29 PM