Abstract

Coat color dilution in dogs is a specific pigmentation phenotype caused by a defective transport of melanosomes leading to large clumps of pigment. It is inherited as a Mendelian autosomal recessive trait and may be accompanied by hair loss, the so-called color dilution alopecia (CDA), or black hair follicular dysplasia (BHFD). We previously identified the noncoding c.-22G>A transition in the melanophilin gene ( MLPH ) as a candidate causative mutation for the dilute phenotype. We have now extended our study and genotyped 935 dogs from 20 breeds segregating for dilute coat color. The dilute-associated A allele segregates in many different breeds suggesting an old mutation event. We also investigated skin biopsies of dogs suspected of having either CDA or BHFD, and our data clearly indicate that the dilute mutation is required but not sufficient to develop clinical signs of the disease. The risk to develop CDA/BHFD seems to be breed specific. Interestingly, 22 out of 29 dogs with clinical signs of CDA/BHFD have clumped melanin in the epidermis, the follicular epithelium, and the hair shafts, whereas in dilute dogs without clinical disease, clumped melanin is only found in the follicular epithelium and the hair shafts but not in the epidermis.

Dogs with dilute coat color are known in many breeds. In dilute dogs, the eumelanin- or phaeomelanin-pigmented skin appears paler and is denoted breed specific, foe example, blue, gray, Isabella, fawn, silver, or pale brown ( Schmutz and Berryere 2007 ). As the change in phaeomelanin is not as dramatic as the eumelanin dilution, red-colored dogs are sometimes difficult to detect as dilute. Coat color dilution ( d ) is inherited as a Mendelian autosomal recessive trait in various dog breeds ( Schmutz et al. 1998 ). Coat color dilution is characterized by a defective transport of melanosomes within follicular melanocytes, which is mainly regulated by 3 interacting proteins (MLPH, MYO5A, and RAB27A) ( Barral and Seabra 2004 ; Hume et al. 2006 ; Hume et al. 2007 ). The dilution phenotype occurs in different mammalian species, and causative mutations within the melanophilin gene ( MLPH ) have been identified in, for example, human, mouse, and cat ( Matesic et al. 2001 ; Ménasché et al. 2003 ; Ishida et al. 2006 ). In a previous study, we applied a candidate gene approach and showed that dilute dogs from different breeds share a common approximately 10-kb haplotype block at the 5’ end of the MLPH gene. Within this shared haplotype block, a noncoding single nucleotide polymorphism (SNP) at the splice donor of exon 1 (c.-22G>A) represents a candidate causal mutation for coat color dilution in 7 dog breeds. The MLPH mRNA expression in skin biopsies of dilute beagles carrying the mutant A allele was lower than in beagles carrying the wild-type G allele ( Philipp, Hamann, et al. 2005 ; Philipp, Quignon, et al. 2005 ; Drögemüller et al. 2007 ).

Coat color dilution has been described as a predisposing risk factor for certain forms of hair loss in dogs ( Mecklenburg 2006 ). Both, color dilution alopecia (CDA; also known as color mutant alopecia) and black hair follicular dysplasia (BHFD), also known as dark hair follicular dysplasia ( Selmanowitz et al. 1972 ), are primarily noninflammatory forms of hair loss that occur in various dog breeds. CDA is associated with a dilute coat color ( Laukner 1998 ), and hair loss is usually most severe on the dorsal trunk. BHFD occurs in the pigmented coat areas of spotted dogs, for example, in the Large Munsterlander ( Schmutz et al. 1998 ). Some authors consider both diseases to be etiologically identical ( Carlotti 1990 ). BHFD is usually clinically noted within the first weeks of age and comprises fracture of hair shafts in dark coated regions, resulting in partial alopecia and scaling. First clinical signs of CDA are usually noticed between 3 and 12 months of age, rarely later in life, and lesions are usually slowly progressive with age. Affected dogs are prone to secondary pyoderma. With regard to histopathology, CDA is identical to BHFD ( Gross et al. 2005 , Mecklenburg 2006 ). Affected skin reveals large clumps of melanin within melanocytes in the hair matrix, the outer root sheath of the hair follicle, and within the hair shaft. The affected hair shafts frequently break within the hair canal resulting in a more or less distorted follicular infundibulum, which is often plugged with keratin, fragmented hairs, and large irregular clumps of melanin. The amount of clumped melanin is variable among breeds and individuals and so is the expressivity of clinical disease ( Hargis et al. 1991 ; Mecklenburg 2006 ). Some dogs with exactly the same histological findings in the hair follicles have no signs of alopecia, whereas others may have complete hair loss. About 25% of the gray/blue dogs of both sexes show clinical symptoms, whereas the others do not develop alopecia. In addition, not all dogs that have symptoms develop them at a similar age of onset or with similar severity ( Schmutz and Berryere 2007 ). Obviously, some dog breeds, for example, the Large Munsterlander ( von Bomhard et al. 2006 ), develop more easily clinical symptoms than other breeds, such as the Weimaraner, where if symptoms occur at all, they are less pronounced ( Laffort-Dassot et al. 2002 ; Schmutz and Berryere 2007 ). Unfortunately, no comprehensive study on the breed distribution of CDA or BHFD has been published. A possible influence of other genes besides MLPH influencing the expressivity of clinical disease is under debate, and recently the canine RAB27A gene was analyzed as possible candidate gene. However, no indication for associated nucleotide polymorphisms in the coding region of RAB27A was found ( Schmutz and Berryere 2007 ).

We now conducted an extensive screening experiment to survey in which breeds the MLPH c.-22G>A mutation occurs and for which of the various coat colors in different breeds this mutation might be causative. We also investigated the MLPH genotypes and the histopathological findings in skin biopsies of 45 dogs suspected of having either CDA or BHFD in order to develop an improved phenotypic classification of CDA and a better understanding for additional disease-promoting factors apart from coat color dilution.

Materials and Methods

MLPH c.-22G>A SNP Genotyping

DNA prepared from ethylenediaminetetraacetic acid (EDTA)–stabilized blood samples of 935 purebred individuals was available for genotyping. These dogs were from 20 breeds, in which the dilute phenotype occurs ( Table 1 ). The coat color phenotype of the dogs was recorded based on photographs and/or pedigree certificates as described ( Philipp, Hamann, et al. 2005 ). Additionally, paraffin-embedded skin biopsies of 45 dogs with melanin aggregates within the hair follicles were available. The skin biopsies were from 40 purebred dogs and from 5 mongrels ( Table 2 ). DNA was isolated from the skin biopsies using the DNeasy Blood and Tissue Kit following the manufacturer's instructions (Qiagen, Hombrechtikon, Switzerland).

Table 1

Genotype frequencies of the MLPH c.-22G>A SNP in 935 dogs with Coat color records

    Exon 1 SNP (c.-22G>A)
 
Breed Coat color No. of animals A/A A/G G/G 
German pinscher a Dilute 28 28   
 Wild type 363  131 232 
Doberman pinscher a Dilute 26 26   
 Wild type 208  77 131 
      
Rhodesian ridgeback a Dilute   
 Wild type 112  35 77 
Whippet Dilute 11 11   
 Wild type 64  38 26 
Australian shepherd a Dilute   
 Wild type 21  13 
Briard a Dilute   
 Wild type 12  
Bolonka Zwetna a Dilute   
 Wild type 11  
French bulldog Dilute   
 Wild type 11  
Great Dane Dilute   
 Wild type  
Chihuahua Dilute   
 Wild type   
Beagle a Dilute   
 Wild type  
Large Munsterlander a Dilute   
 Wild type   
Newfoundland Dilute   
 Wild type  
Hovawart Dilute   
 Wild type   
Miniature pinscher Dilute   
Border collie Dilute   
Slovakian rough haired pointer Dilute   
American Staffordshire terrier Dilute   
Italian greyhound Dilute   
Jack Russel terrier Dilute   
Total  935 112 326 497 
    Exon 1 SNP (c.-22G>A)
 
Breed Coat color No. of animals A/A A/G G/G 
German pinscher a Dilute 28 28   
 Wild type 363  131 232 
Doberman pinscher a Dilute 26 26   
 Wild type 208  77 131 
      
Rhodesian ridgeback a Dilute   
 Wild type 112  35 77 
Whippet Dilute 11 11   
 Wild type 64  38 26 
Australian shepherd a Dilute   
 Wild type 21  13 
Briard a Dilute   
 Wild type 12  
Bolonka Zwetna a Dilute   
 Wild type 11  
French bulldog Dilute   
 Wild type 11  
Great Dane Dilute   
 Wild type  
Chihuahua Dilute   
 Wild type   
Beagle a Dilute   
 Wild type  
Large Munsterlander a Dilute   
 Wild type   
Newfoundland Dilute   
 Wild type  
Hovawart Dilute   
 Wild type   
Miniature pinscher Dilute   
Border collie Dilute   
Slovakian rough haired pointer Dilute   
American Staffordshire terrier Dilute   
Italian greyhound Dilute   
Jack Russel terrier Dilute   
Total  935 112 326 497 
a

Perfect cosegregation of the A allele with the dilute coat color was also observed in informative families of these breeds.

Table 2

Histological phenotypes of 38 biopsies from dogs, which were homozygous A/A at MLPH c.-22G>A and originally suspected of being affected with CDA or BHFD

Breed No. of animals Clinically CDA/BHFD and clumped melanin in epidermis and follicle Clinically CDA/BHFD and clumped melanin only in follicle Clinically other forms of alopecia and clumped melanin only in follicle 
Doberman pinscher  
Yorkshire terrier  
German pinscher   
Chihuahua 
Rhodesian ridgeback   
Staffordshire bull terrier   
Bernese mountain dog   
Giant schnauzer   
Great Dane   
Italian greyhound   
Miniature pinscher   
Poodle   
Prague Ratter   
Tibetan terrier   
Weimaraner   
Mongrels  
Total 38 22 (58%) 7 (18%) 9 (24%) 
Breed No. of animals Clinically CDA/BHFD and clumped melanin in epidermis and follicle Clinically CDA/BHFD and clumped melanin only in follicle Clinically other forms of alopecia and clumped melanin only in follicle 
Doberman pinscher  
Yorkshire terrier  
German pinscher   
Chihuahua 
Rhodesian ridgeback   
Staffordshire bull terrier   
Bernese mountain dog   
Giant schnauzer   
Great Dane   
Italian greyhound   
Miniature pinscher   
Poodle   
Prague Ratter   
Tibetan terrier   
Weimaraner   
Mongrels  
Total 38 22 (58%) 7 (18%) 9 (24%) 

Polymerase chain reaction (PCR) for amplifying a 312-bp fragment containing canine MLPH exon 1 using the primers MLPH_157395_F (5′-CCTTCCTTCCCCTGTAGGAC-3′) and MLPH_157706_R (5′-GCCTAAAATGAGCTCCCTGA-3′) and sequence reactions were carried out as described before ( Drögemüller et al. 2007 ). Sequence data were analyzed with Sequencher 4.8 (GeneCodes, Ann Arbor, MI) to derive the genotypes at the MLPH c.-22G>A polymorphism. Alternatively, the PCR reaction was carried out using the same primers in a LightCycler where a melting point analysis was performed after amplification using the sensor oligonucleotides Dil(AiO)E1Pw (5′-ROX-GAA AGG AGC CGG TGA GTG CA-PHO-3’), Dil(AiO)E1Pm (5′-Cy5.5-GAA AGG AGC CAG TGA GTG CAG-PHO-3′), and the anchor oligonucleotide Dil(AiO)Ex1A (5′-CCA GGG CCT GCC CGC CCC G-fluoresceine-3′). The mutation is characterized by a sensor probe half melting temperature of 60 °C (Dil(AiO)E1Pw) or 67 °C (Dil(AiO)E1Pm) as compared with 66 °C or 61 °C for the wild-type allele, respectively.

Histopathology

One to five routinely processed skin biopsies stained with hematoxylin and eosin from a total of 45 dogs were examined in this study. They were selected from archival material of the Institute for Animal Pathology, Vetsuisse Faculty, University of Berne, Berne, Switzerland (35 cases), and the Institute of Veterinary Pathology, Vetsuisse Faculty, University of Zürich, Zürich, Switzerland (10 cases). In all histological reports of these biopsies, clumped melanin in the hair follicle and hair shafts had been described. For this study, a blinded histological examination of the biopsies was performed by one of the authors.

Results and Discussion

In a previous study, we have identified the c.-22G>A transition at the last nucleotide of the 5′-untranslated first exon of the MLPH gene as a candidate causative mutation for the dilute phenotype ( Drögemüller et al. 2007 ). We have now extended our study and genotyped 935 dogs with recorded coat color from 20 different dog breeds at this SNP marker ( Table 1 ). In 9 of the 20 dog breeds, informative 2-generation families were available, and in these families the A allele showed perfect cosegregation with the dilute phenotype. Furthermore, we observed perfect association of the dilute phenotype with the c.-22G>A polymorphism in all 20 different breeds, and each of the 112 dilute colored dogs in our study was homozygous for the A allele at the c.-22G>A polymorphism ( Table 1 ). We still cannot formally exclude the possibility that c.-22G>A represents just a neutral polymorphism in linkage disequilibrium with the causative mutation until the complete associated haplotype block has been fully characterized in the dilute animals. However, the new genotype data from the extended large cohort are in agreement with our previous findings and support the hypothesis that c.-22G>A is indeed the causative mutation. The wide breed distribution across diverse types of dogs implies an old mutation event predating the creation of modern dog breeds.

Dilute colored dogs are assumed to be predisposed to develop hair loss in the form of CDA or BHFD. In order to get a more accurate phenotypic classification, we histologically examined archived biopsies of 45 unrelated dogs with an original histological finding of clumped melanin within the hair follicle and clinical hair loss. On re-evaluation in 29 of the 45 cases, the clinical findings were compatible with the described findings in CDA or BHFD ( Gross et al. 2005 ). In all 45 cases, clumped melanin of variable size in the hair shaft and the outer root sheath of the follicle was present. Some hair shafts were fractured or attenuated. The sometimes dilated infunbibula contained variable amounts of keratin, hair fragments, and melanin clumps. A rather high percentage of hair follicles was in telogen, and perifollicular pigmentary incontinence of variable degree was present. In some of the biopsies, additional findings compatible with a secondary pyoderma were seen. The 45 biopsies were also genotyped at the MLPH c.-22G>A SNP. The coat color phenotypes of the 45 dogs were not available. A total of 38 out of 45 dogs were homozygous A/A for the dilute-associated allele. Two dogs with aggregated melanin within the hair follicles and hair shafts were heterozygous A/G, and 5 dogs were homozygous G/G. Surprisingly, some clumped melanin was present in the outer root sheath of the follicles in these 7 dogs. However, clumping of melanin is not only specific for CDA and BHFD but also occurs in hair follicle dystrophy as a consequence of preterminal catagen induction ( Hendrix et al. 2005 ). All remaining 38 homozygous A/A cases showed clumped melanin in the hair follicles. In 22 dogs (58%), where the clinical anamnesis was exactly compatible with the criteria of CDA or BHFD, clumped melanin was visible in the epidermis ( Figure 1a ). In 7 dogs (18%), which also showed typical clinical symptoms of CDA or BHFD, the clumped melanin could only be seen in the hair follicles, and we consider these dogs to be most likely affected with CDA or BHFD ( Figure 1b ). In the remaining 9 dogs (24%), where the clumped melanin was also apparently restricted to the hair follicles, the clinical symptoms were not compatible with CDA or BHFD after careful re-evaluation of the available anamneses ( Figure 1c ). In some of these dogs, the re-evaluation resulted in the diagnosis of other well-known hair loss phenotypes, such as, for example, seasonal flank alopecia or endocrine disorders.

Figure 1

Representative histological phenotypes of dogs, which were homozygous A/A at MLPH c.-22G>A and suspected of being affected with CDA or BHFD. Biopsies from dogs with ( a ) confirmed clinical symptoms of CDA/BHFD and clumped melanin in epidermis, hair follicles, and hair shafts; ( b ) clinical symptoms compatible with CDA/BHFD and clumped melanin only in the hair follicle; and ( c ) clinical symptoms different from accepted CDA/BHFD criteria and clumped melanin restricted to the hair follicle. Note that melanin clumping in the hair follicle is similar in all 3 cases, and the definitive diagnosis of CDA/BHFD can only be made based on additional clinical findings. Hematoxylin and eosin ×200. (This figure appears in color in the online version of Journal of Heredity .)

Figure 1

Representative histological phenotypes of dogs, which were homozygous A/A at MLPH c.-22G>A and suspected of being affected with CDA or BHFD. Biopsies from dogs with ( a ) confirmed clinical symptoms of CDA/BHFD and clumped melanin in epidermis, hair follicles, and hair shafts; ( b ) clinical symptoms compatible with CDA/BHFD and clumped melanin only in the hair follicle; and ( c ) clinical symptoms different from accepted CDA/BHFD criteria and clumped melanin restricted to the hair follicle. Note that melanin clumping in the hair follicle is similar in all 3 cases, and the definitive diagnosis of CDA/BHFD can only be made based on additional clinical findings. Hematoxylin and eosin ×200. (This figure appears in color in the online version of Journal of Heredity .)

From the detailed histological analysis of the 38 biopsies from homozygous A/A dogs, we suspect that the current histopathological diagnostic criterium of clumped melanin in the hair follicle does not allow an unambiguous phenotype classification of clinically suspected CDA or BHFD cases. The histhopathological findings have to be evaluated in light of the clinical symptoms to allow a precise and definitive diagnosis of CDA or BHFD. In about a quarter of the dogs that had originally been diagnosed with CDA or BHFD, the hair loss is apparently caused by a different pathological mechanism, and therefore, these dogs should not be considered to be genuinely affected by CDA or BHFD. According to our findings, clumped melanin in the epidermis is a very strong indicator of true CDA or BHFD.

In conclusion, we found perfect association of the MLPH c.-22G>A SNP with dilute coat color in more than 900 dogs supporting the hypothesis that this polymorphism is indeed the causative mutation. The wide breed distribution of the mutant MLPH c.-22G>A allele suggests an old mutation event. Although our data clearly indicate that the MLPH mutation increases the risk for CDA/BHFD, there seem to be additional modifying factors. A characteristic feature of most CDA/BHFD–affected dogs is the presence of clumped melanin in the epidermis. In some breeds, such as the pinscher breeds and the Rhodesian ridgebacks, the reported coat quality of dilute dogs ranges from normal to severely CDA affected. Therefore, these breeds offer the chance to search for modifier genes, which influence the risk of developing CDA/BHFD in dogs with dilute coat color.

Funding

Albert-Heim-Foundation (grant no. 85).

The authors would like to thank the numerous owners, breeders, and veterinarians who contributed samples to this study. The help of Brigitta Colomb, Manuela Bozzo, Erika Garchi, and Evelyne Rohrer for expert technical assistance is acknowledged.

References

Barral
DC
Seabra
MC
The melanosome as a model to study organelle motility in mammals
Pigment Cell Res
 , 
2004
, vol. 
17
 (pg. 
111
-
118
)
Carlotti
DN
Von Tscharner
C
Halliwell
REW
Canine hereditary black hair follicular dysplasia and colour mutant alopecia. Clinical and histopathological aspects
Advances in veterinary dermatology
 , 
1990
, vol. 
vol. 1
 
Philadelphia
Baillière Tindall
(pg. 
43
-
46
)
Drögemüller
C
Philipp
U
Haase
B
Günzel-Apel
AR
Leeb
T
A non-coding melanophilin gene ( MLPH ) SNP at the splice donor of exon 1 (c.-22G>A) represents a candidate causal mutation for coat color dilution in dogs
J Hered
 , 
2007
, vol. 
98
 (pg. 
468
-
473
)
Gross
TL
Ihrke
PJ
Walder
EJ
Affolter
VK
Gross
TL
Ihrke
PJ
Walder
EJ
Affolter
VK
Dysplastic diseases of the adnexa: color dilution alopecia and black hair follicular dysplasia
Skin diseases of the dog and cat: clinical and histopathological diagnosis
 , 
2005
2nd ed
Oxford (UK)
Blackwell Science Ltd
(pg. 
518
-
522
)
Hargis
AM
Brignac
MM
Al-Bagdadi
FAK
Muggli
F
Mundell
A
Black hair follicular dysplasia in black and white saluki dogs: differentiation from color mutant alopecia in the doberman pinscher by microscopic examination of hairs
Vet Dermatol
 , 
1991
, vol. 
2
 (pg. 
69
-
83
)
Hendrix
S
Handjiski
B
Peters
EM
Paus
R
A guide to assessing damage response pathways of the hair follicle: lessons from cyclophosphamide-induced alopecia in mice
J Invest Dermatol
 , 
2005
, vol. 
125
 (pg. 
42
-
51
)
Hume
AN
Tarafder
AK
Ramalho
JS
Sviderskaya
EV
Seabra
MC
A coiled-coil domain of melanophilin is essential for Myosin Va recruitment and melanosome transport in melanocytes
Mol Biol Cell
 , 
2006
, vol. 
17
 (pg. 
4720
-
4735
)
Hume
AN
Ushakov
DS
Tarafder
AK
Ferenczi
MA
Seabra
MC
Rab27a and MyoVa are the primary Mlph interactors regulating melanosome transport in melanocytes
J Cell Sci
 , 
2007
, vol. 
120
 (pg. 
3111
-
3122
)
Ishida
Y
David
VA
Eizirik
E
Schäffer
AA
Neelam
BA
Roelke
ME
Hannah
SS
O'Brien
SJ
Menotti-Raymond
M
A homozygous single-base deletion in MLPH causes the dilute coat color phenotype in the domestic cat
Genomics
 , 
2006
, vol. 
88
 (pg. 
698
-
705
)
Laffort-Dassot
C
Beco
L
Carlotti
CN
Follicular dysplasia in five Weimaraners
Vet Dermatol
 , 
2002
, vol. 
13
 (pg. 
253
-
260
)
Laukner
A
Coat color in dogs. 2: Clinical significance
Tierarztl Prax Ausg K Klientiere Heimtiere
 , 
1998
, vol. 
26
 (pg. 
124
-
132
)
Matesic
LE
Yip
R
Reuss
AE
Swing
DA
O'Sullivan
TN
Fletcher
CF
Copeland
NG
Jenkins
NA
Mutations in Mlph, encoding a member of the Rab effector family, cause the melanosome transport defects observed in leaden mice
Proc Natl Acad Sci U S A
 , 
2001
, vol. 
98
 (pg. 
10238
-
10243
)
Mecklenburg
L
An overview on congenital alopecia in domestic animals
Vet Dermatol
 , 
2006
, vol. 
17
 (pg. 
393
-
410
)
Ménasché
G
Ho
CH
Sanal
O
Feldmann
J
Tezcan
I
Ersoy
F
Houdusse
A
Fischer
A
de Saint Basile
G
Griscelli syndrome restricted to hypopigmentation results from a melanophilin defect (GS3) or a MYO5A F-exon deletion (GS1)
J Clin Invest
 , 
2003
, vol. 
112
 (pg. 
450
-
456
)
Philipp
U
Hamann
H
Mecklenburg
L
Nishino
S
Mignot
E
Günzel-Apel
AR
Schmutz
SM
Leeb
T
Polymorphisms within the canine MLPH gene are associated with dilute coat color in dogs
BMC Genet
 , 
2005
, vol. 
6
 pg. 
34
 
Philipp
U
Quignon
P
Scott
A
Andre
C
Breen
M
Leeb
T
Chromosomal assignment of the canine melanophilin gene ( MLPH ): a candidate gene for coat color dilution in Pinschers
J Hered
 , 
2005
, vol. 
96
 (pg. 
774
-
776
)
Schmutz
SM
Berryere
TG
Genes affecting coat colour and pattern in domestic dogs: a review
Anim Genet
 , 
2007
, vol. 
38
 (pg. 
539
-
549
)
Schmutz
SM
Moker
JS
Clark
EG
Shewfelt
R
Black hair follicular dysplasia, an autosomal recessive condition in dogs
Can Vet J
 , 
1998
, vol. 
39
 (pg. 
644
-
646
)
Selmanowitz
VJ
Kramer
KM
Orentreich
N
Canine hereditary black hair follicular dysplasia
J Hered
 , 
1972
, vol. 
63
 (pg. 
43
-
44
)
von Bomhard
W
Mauldin
EA
Schmutz
SM
Leeb
T
Casal
ML
Black hair follicular dysplasia in Large Munsterlander dogs: clinical, histological and ultrastructural features
Vet Dermatol
 , 
2006
, vol. 
17
 (pg. 
182
-
188
)

Author notes

Corresponding Editor: Francis Gailbert