Interbreed variation of DLA-DRB1, DQA1 alleles and haplotypes in the dog

Interbreed variation of DLA-DRB1, DQA1

alleles and haplotypes in the dog

Lorna J. Kennedya,*, Stuart D. Carterb, Annette Barnesb,Susan Bellb, David Bennettc, Bill Olliera, Wendy Thomsona

aSchool of Epidemiology and Health Sciences, University of Manchester,

Stopford Building, Oxford Road, Manchester, M13 9PT, UKbVeterinary Clinical Science, University of Liverpool, Liverpool, UK

cVeterinary Clinical Studies, University of Glasgow, Glasgow, UK

Received 18 November 1998

Abstract

Although 36 DLA-DRB1 and 10 DLA-DQA1 allele sequences have been published to date, no

data on individual allele frequencies exists, either for specific breeds or cross breeds, and the full

extent of the polymorphism at each of these loci is still not known. We have used sequence-specific

oligonucleotide probing (SSOP) to characterise a series of 367 dogs for their DRB1 and DQA1

alleles. These included individual animals from over 60 different breeds, with numbers per breed

ranging from 1 to 39. DLA types were generated from 218 dogs for DRB1 and from 330 dogs for

DQA1, while 181 dogs were characterised for both these loci. The frequency of individual DRB1

and DQA1 alleles showed considerable interbreed variation, e.g. 83% of West Highland White

Terriers were DRB1�01 as opposed to 9% of Collies. No breed had >9 of the 22 DRB1 types

defined in this study; several breeds had only two DRB1 types. DLA-DQA1 showed less variation

in allele numbers per breed, but also showed considerable interbreed frequency variation.

Haplotype analysis revealed over 44 different DRB1/DQA1 combinations. Of these, 25 were in a

number of animals, and also in an animal that was homozygous for one or both of these loci. Some

DRB1 alleles could be found in combination with several different DQA1 alleles, while others were

only present in one haplotypic combination. DLA allele frequency data in normal dogs will be

critical for disease association studies. It may also be possible to use haplotype data to establish the

genetic relationships between different dog breeds. # 1999 Elsevier Science B.V. All rights

reserved.

Keywords: Dog; DLA-DRB1; DLA-DQA1; Genetics

Veterinary Immunology and Immunopathology

69 (1999) 101±111

* Corresponding author. Tel.: +44-161-275-7349; fax: +44-161-275-5043E-mail address: [email protected], website: http://www.arc.man.ac.uk/dogpage.html (L.J. Kennedy)

0165-2427/99/$ ± see front matter # 1999 Elsevier Science B.V. All rights reserved.

PII: S 0 1 6 5 - 2 4 2 7 ( 9 9 ) 0 0 0 4 6 - X

1. Introduction

Preliminary characterisation of the canine Major Histocompatibility Complex (MHC)

has previously come from three international histocompatibility workshops (Bull et al.,

1987; Deeg et al., 1986; Vriesendorp et al., 1976, 1973). These established that the canine

MHC is similar to that in other species in that it contains class I, II and III genes, although

they did not serologically identify a highly polymorphic system analogous to HLA-DRB1

and DQB1. The Dog Leucocyte Antigen (DLA) region is known to contain class I and II

genes, although the precise number of genes and their relative positions remain unknown.

Since the original canine MHC workshops, there has been only a limited number of

molecular-based studies relating to the DLA region.

In studies of the class II region in the dog, characterisation of the DLA-D region by

restriction fragment length polymorphism (RFLP) using human cDNA probes suggested

canine homologues for class II genes (Sarmiento and Storb, 1988; Williamson et al.,

1989). These studies indicated the presence of at least five DLA class II alpha genes

(DRA, two DQA, DPA and DNA) and seven beta genes (two DRB, two DQB, two DPB

and DOB). Nucleotide sequencing of a canine DRB cDNA clone showed that it was

highly homologous to the human DRB1 gene (Sarmiento and Storb, 1990). More

recently, molecular analysis of the DLA-DR region has confirmed the existence of one

DLA-DRA gene and two DRB genes, one of which is intact (DRB1) and a second gene

which is a pseudogene (DRB2) and is not present in all dogs (Wagner et al., 1996b).

Molecular analysis of the DLA-DQ region shows a similar situation with one DLA-DQA

gene (Sarmiento et al., 1992; Wagner et al., 1996a) and two DQB genes (DQB1 and

DQB2), only one of which is functional (DQB1) (Sarmiento et al., 1993; Wagner et al.,

1998a). The presence of other DLA-DRB and DQB genes cannot as yet be excluded.

The extent of the polymorphism within the DLA class II is only now being determined

at the DNA sequence level and, to date, 36 DRB1 (Francino et al., 1997; Kennedy et al.,

1998; Sarmiento et al., 1990; Wagner et al., 1996c; Kennedy et al., 1999a) and 10 DQA1

(Polvi et al., 1997; Sarmiento et al., 1992; Wagner et al., 1996a) alleles have been

identified. It is likely that this list will expand considerably. The acceptance and naming

of new DLA alleles is now controlled by a nomenclature committee working under the

auspices of the International Society for Animal Genetics (Kennedy et al., 1999a).

As yet two issues have not been addressed in the study of the DLA system. Firstly,

there is no information regarding the distribution of DLA types within different dog

breeds. This is likely to be an important consideration, as breeds have been highly

selected and have a more restricted and static gene pool. In humans, anthropological

studies have demonstrated dramatic differences in HLA allele frequencies between

diverse and inbred ethnic populations. (Charron, 1997).

Secondly, linkage disequilibrium between alleles of different loci is a significant

feature of the MHC region and the haplotypic combination of certain alleles has not been

established in the dog nor in different breeds.

MHC molecules are responsible for the presentation of self and non-self antigens to the

immune system and determine the level of the immune response generated. In humans,

many HLA types are strongly associated with susceptibility to a range of diseases where

there is an underlying immune component (Tiwari and Terasaki, 1985). Furthermore,

102 L.J. Kennedy et al. / Veterinary Immunology and Immunopathology 69 (1999) 101±111

HLA can determine immune response to both vaccination (Hayney et al., 1996) and

natural infection (Hill et al., 1991). Various dog breeds can be highly susceptible to

certain immune mediated conditions and a detailed study of their DLA polymorphisms is

likely to be informative.

As an initial study, we have determined the frequency of DLA-DRB1 and DQA1

alleles in the dog, and investigated whether there is preliminary evidence of restricted

allele distribution in different breeds. Furthermore, we have examined the haplotypic

combinations of DRB1 and DQA1 alleles in a broad population of dogs.

2. Material and methods

Three hundred and sixty seven dogs were included in this study; all were consecutive

consultations at the University of Liverpool Small Animal Hospital. Excess blood taken

from dogs required for clinical diagnostic purposes was collected into EDTA and frozen

until required. Clinical data, age, sex, and breed were recorded for each animal. The panel

of dogs included over 60 different breeds, with numbers per breed ranging from 1 to 39.

DNA was extracted using the DNA-direct system II (Dynal). Of the total number of dogs,

218 were characterised for DLA-DRB1, while 330 were characterised for DLA-DQA1,

by SSOP (Kennedy et al., 1999b; Kennedy et al., submitted). A group of 181 dogs were

typed for both loci. DNA samples were also available from dogs (n � 14) that had been

serologically typed for DLA-B (Professor Grosse-Wilde, Essen, Germany) and from Irish

setters (n � 4) which had been sequenced for DLA-DQA1 and DQB1 alleles (Dr. A.

Polvi, Helsinki, Finland, and Dr. O. Garden, University of London, UK). These were

DNA-sequenced and included as controls in initial experiments. Thereafter, a set of DNA

samples were chosen for each locus so that at least one sample was positive and another

was negative with each probe tested.

DLA-DRB1 intron primers (Wagner et al., 1996c) were used: forward,

50CCGTCCCCACAGCACATTTC; and reverse, 50TGTGTCACACACCTCAGCACCA.

These primers amplify all of exon 2, and give a product of 350bp. DLA-DQA1 intron

primers as described by Wagner et al. (1996a), were used: 50TAAGGTT-

CTTTTCTCCCTCT, and 30GGACAGATTCAGTGAAGAGA. These primers amplify

all of exon 2, and give a product of 370bp. Oligonucleotide probes were designed to

detect much of the polymorphism indicated in the published DLA-DRB1 (Francino et al.,

1997; Kennedy et al., 1998; Sarmiento et al., 1990; Wagner et al., 1996c; Kennedy et al.,

1999a). For DRB1, it was impractical to make probes for every single polymorphic

change. Twenty nine probes were made to polymorphisms in the regions around codons

10, 30, 70, 74, 78 and 86. This permitted broad DRB1 types to be assigned. Probes were

also designed to detect all the polymorphisms indicated in the published DLA-DQA1

sequence data (Polvi et al., 1997; Sarmiento et al., 1992; Wagner et al., 1996a). The

length of each probe was adjusted so that the polymorphism was in the middle third of the

probe and the predicted melting temperatures were similar. The probes ranged in length

from 15 to 18 nucleotides, depending on the proportions of the different bases. Probes

were made to detect all known sequences at each polymorphic position. All probes were

50 end labelled with biotin. Probe stocks were used at 10 mM, and probe working

L.J. Kennedy et al. / Veterinary Immunology and Immunopathology 69 (1999) 101±111 103

concentrations were optimised based on the reactions of the known controls. For more

details of the SSOP methods, see Kennedy et al. (1999b) and Kennedy et al. (submitted).

PCR reactions were performed with 500 ng DNA in a 100 ml reaction containing 1x

PCR buffer (10x buffer: 500 mM KCl, 100 mM Tris-Cl pH 8.8, 1% Triton X-100) and

final concentrations of 1 mM each primer, 200 mM each dNTP, 1.5 mM magnesium

chloride and two units of Taq polymerase. The PCR programme consisted of an initial

3 min at 958C, 30 cycles of 958C for 40 s followed by 618C for DRB1 or 488C for DQA1

for 1 min and extension at 728C for 1 min and 20 s, plus a final 10 min extension at 728C.

All PCR reactions were performed in 96 well plates, with a heated lid on the thermal

cycler. Amplifications were monitored by running a 5-ml sample on a 2% agarose gel

stained with ethidium bromide.

Thirty duplicate membranes were prepared. Two microlitres of amplified DNA were

dotted onto positively charged nylon membrane (Hybond N�, Amersham), using an

automatic dot blotter (Robbins), and allowed to dry at room temperature. The DNA was

denatured by incubating the membranes in denaturing solution (1.5 M NaCl/0.5 M

NaOH) for 10 min, and neutralised by washing for 1±2 min in a neutralising solution

(3 M NaCl/1M Tris, pH 7.0). The single stranded DNA was immobilised by UV cross-

linking for 5 min.

The membranes were incubated in hybridisation bottles with 10-ml hybridisation

buffer (5x SSC/0.1% sarcosyl (N-Laurolyl sarcosine)/0.5% blocking reagent (Amer-

sham)/0.02% SDS) in a rotating oven at 428C for at least 30 min. Thereafter, 10±20 pmol

biotinylated probe were added, and incubation continued for at least 90 min at 428C.

Following hybridisation, buffer and probe were discarded and membranes washed briefly

in 15 ml of pre-warmed 5x SSC/0.1% SDS to remove excess probe. One 30±40 min

stringent wash with 1x SSC/0.1% SDS (pre-warmed to 428C) was performed.

Immediately following the stringent wash, membranes were briefly washed in 5x SSC.

Membranes were not allowed to dry at this stage. ECL detection was carried out as

described in the Amersham protocol.

3. Results

3.1. DRB1

Table 1 shows the overall phenotype and `allele' frequencies for the 22 DLA-DRB1

types that could be defined by the SSOP method used. Five DRB1 types had a frequency

of >10%, five had a frequency between 5% and 10%, nine had a frequency of <5%, and

four were not found in this study. One DRB1 type, DRB1�15, was very common, with a

phenotype frequency of 38.9%. This type was also found in over 60% of the breeds

tested. However, the frequency of DRB1�15 within breeds varied considerably, e.g. in

this study, around 50% of Labradors and German Shepherd dogs have DRB1�15, whereas

only 13% of Beagles had DRB1�15. The second most common DRB1 type, DRB1�01,

(26.6%), was found in over 40% of breeds tested and also showed considerable interbreed

frequency variation. Thus, 83% of West Highland White Terriers, but only 9% of Collies

had DRB1�01. Other DRB1 types appeared to have a more limited distribution, e.g.

104 L.J. Kennedy et al. / Veterinary Immunology and Immunopathology 69 (1999) 101±111

DRB1�04 was only found in four breeds: Boxer, Pointer, Pug and Jack Russell Terrier.

One DRB1 type, DRB1�24 was only found in the Japanese Akita. An indication of some

of the interbreed variation found in this study is shown in Table 2, which shows the

DRB1 frequencies in the nine breeds where >7 animals were tested. In this study, the

maximum number of different DRB1 types found in any one breed was nine out of a

possible 22 DRB1 types. Some breeds, such as German Shepherd Dog and Labrador, had

six and eight different DRB1 types, respectively, whereas others, such as Rottweiler had

only two different DRB1 types. The three Dobermans examined were all found to be

homozygous for DRB1�06.

3.2. DQA1

Table 1 also shows the phenotype and allele frequencies for the 10 known DLA-DQA1

alleles. In this study, DQA1�0101 and DQA1�0601 were two very common alleles, both

with phenotype frequencies of over 30%. Three other alleles, DQA1�0901, DQA1�0401

and DQA1�05011, had frequencies of over 15%. Two alleles, DQA1�0701 and

DQA1�0301 were less frequent at just under 10%, while two alleles, DQA1�05012 and

DQA1�0801, were not found in this group of animals. Most of the DQA1 alleles were

found in the 60 different breeds tested. Three alleles had a more restricted distribution,

Table 1DLA-DRB1 and DQA1 phenotype and allele frequencies (%)

DRB1� Phenotype

n � 218

`Allele'

2n � 436

DQA1* Phenotype

n � 330

Allele

2n � 660

01 26.6 17.2 0101 37.3 25.8

02 12.8 8.3 0201 5.0 3.5

03/09 4.1 2.3 0301 7.9 4.7

04 5.5 2.9 0401 22.1 15.1

05 1.4 0.7 05011 16.7 10.8

06 16.1 11.2 05012 0 0

07 0 0 0601 33.3 21.1

08 0.5 0.2 0701 9.7 5.8

11 6.9 5.1 0801 0 0

12/19 15.1 9.6 0901 19.4 12.4

13 5.1 2.9 New 0.3 0.2

14 0 0

15 38.9 26.2

17 0 0

18 7.3 4.8

20 8.7 4.4

21 0 0

22 0.9 0.5

23 0.9 0.5

24 1.4 0.9

25 0.9 0.7

26 0.5 0.2

New 1.4 0.7

L.J. Kennedy et al. / Veterinary Immunology and Immunopathology 69 (1999) 101±111 105

DQA1�0201, DQA1�0701 and DQA1�0301, which were only found in 7, 10 and 11

breeds, respectively.

Table 3 summarises the DQA1 allele frequencies for the ten different dog breeds

in which ten or more animals were tested. Labradors had the highest number of DQA1

alleles, with eight, whereas Poodles had only two different DQA1 alleles. Within

each of these ten breeds it was clear that the alleles did not occur with the same

distribution in the different breeds, and the most frequent allele was not always

DQA1�0101. Data from other breeds that were tested in fewer numbers are interesting;

six Dobermans were all homozygous for DQA1�0401, three Japanese Akitas were all

homozygous for DQA1�0101 and three Shetland Sheepdogs were all homozygous for

DQA1�0901.

Table 2DLA-DRB1 phenotype frequencies (%) in nine different dog breeds

DRB1� GSDb

n � 21

Labrc

n � 21

Retrd

n � 7

Beage

n � 15

WHWTf

n � 12

Collg

n � 11

Setth

n � 11

Boxi

n � 10

Rottj

n � 7

01 29 19 14 40 83 9 73 Ða 57

02 19 10 14 33 Ða 36 Ða 10 Ða

03/09 Ða Ða Ða Ða Ða 9 Ða Ða Ða

04 Ða Ða Ða Ða Ða Ða Ða 70 Ða

05 Ða Ða Ða Ða 17 Ða 18 Ða Ða

06 Ða 10 14 20 Ða Ða 27 20 86

07 Ða Ða Ða Ða Ða Ða Ða Ða Ða

08 Ða 5 Ða Ða Ða Ða Ða Ða Ða

11 43 Ða Ða 7 Ða Ða Ða Ða Ða

12/19 5 48 86 7 Ða 9 18 Ða Ða

13 Ða Ða Ða 7 Ða 18 Ða 40 Ða

14 Ða Ða Ða Ða Ða Ða Ða Ða Ða

15 57 48 29 13 33 45 Ða 30 Ða

17 Ða Ða Ða Ða Ða Ða Ða Ða Ða

18 Ða 5 Ða Ða Ða 27 9 Ða Ða

20 5 24 14 Ða Ða Ða 27 Ða Ða

21 Ða Ða Ða Ða Ða Ða Ða Ða Ða

22 Ða Ða Ða Ða Ða Ða Ða 10 Ða

23 Ða Ða Ða Ða Ða Ða Ða Ða Ða

24 Ða Ða Ða Ða Ða Ða Ða Ða Ða

25 Ða Ða Ða Ða Ða Ða Ða Ða Ða

26 Ða Ða Ða 7 Ða Ða Ða Ða Ða

New Ða Ða Ða 7 Ða Ða Ða Ða Ða

a Allele not found in this breed in this study.b German Shepherd Dog.c Labrador.d Retriever.e Beagle.f West Highland White Terrier.g Collie.h Setter.i Boxer.j Rottweiler.

106 L.J. Kennedy et al. / Veterinary Immunology and Immunopathology 69 (1999) 101±111

3.3. DRB1/DQA1 haplotypes

DLA-DRB1/DQA1 haplotypes were identified by following a sequential analytical

process. Firstly, 64 dogs that were homozygous at both DRB1 and DQA1 were selected,

and from these we found 16 different DRB1-DQA1 combinations which occurred in

varying numbers from 1 to 17. Dogs that were homozygous only at DRB1 (n � 22) or at

DQA1 (n � 36) were then selected. From these, we confirmed many of the previous

haplotypes, and also identified 10 further haplotypes. The remaining dogs were then

examined using the haplotype data already identified, and haplotypes were assigned to

each of these dogs. From these, we identified a further 18 possible haplotypes, making a

total of 44 different haplotypes. However, 19 of these 44 haplotypes were only found in

one heterozygous individual and these were grouped together as `others' until such time

as more animals are found with those haplotypes. This left 25 haplotypes which were

found in at least two different animals, and all of which were identified in at least one dog

that was homozygous for one or both of the loci studied. Table 4 shows the haplotype

frequencies for these DRB1/DQA1 haplotypes. As expected, the most common

haplotypes feature the most frequent DRB1 and DQA1 alleles: DRB1�15/DQA1�0601

(20.9%) and DRB1�01/DQA1�0101 (14.6%). Most DRB1 types are found in haplotype

combination with two or three different DQA1 alleles. However, there are some DRB1

types that were only found in one haplotype combination, e.g. every DRB1�03,

DRB1�09, DRB1�13, DRB1�18 and DRB1�24 was found with DQA1�0101, and every

Table 3DLA-DQA1 phenotype frequencies (%) in ten different dog breeds

DQA1� GSDb

n � 28

Labrc

n � 30

Retrd

n � 18

Beage

n � 12

WHWTf

n � 17

Collg

n � 22

Setth

n � 11

Boxi

n � 16

Rottj

n � 12

Poodk

n � 10

0101 25 30 17 58 71 50 100 44 Ða 50

0201 21 7 6 Ða Ða Ða Ða 25 8 Ða

0301 Ða 10 Ða 8 6 5 18 Ða 58 Ða

0401 25 47 72 8 Ða 14 18 13 Ða Ða

05011 4 20 11 Ða Ða Ða 27 13 92 Ða

05012 Ða Ða Ða Ða Ða Ða Ða Ða Ða Ða

0601 57 40 33 Ða 29 59 9 25 Ða 70

0701 18 7 Ða 8 Ða Ða Ða 50 8 Ða

0801 Ða Ða Ða Ða Ða Ða Ða Ða Ða Ða

0901 18 10 11 42 35 41 Ða Ða Ða Ða

New Ða Ða Ða Ða Ða Ða Ða Ða Ða Ða

a Allele not found in this breed in this study.b German Shepherd Dog.c Labrador.d Retriever.e Beagle.f West Highland White Terrier.g Collie.h Setter.i Boxer.j Rottweiler.k Poodle.

L.J. Kennedy et al. / Veterinary Immunology and Immunopathology 69 (1999) 101±111 107

DRB1�12 and DRB1�19 with DQA1�0401. Conversely, all the DQA1 alleles, except

one, were found in several haplotype combinations. The exception, DQA1�05011, was

only present with DRB1�06, although DRB1�06 was also identified with four other

DQA1 alleles.

No one breed appeared to have all the possible haplotypes represented. Fig. 1 shows

the haplotype frequencies (%) in five different breeds, where at least ten animals were

typed. The breed with the highest number of different haplotypes was the Labrador with

15, however, this was also the breed with the most dogs tested (n � 20). West Highland

White Terriers (n � 11) had only five different haplotypes. Interestingly, while Beagles

and Boxers both had six different haplotypes, none of them were in common. Some

haplotypes were found in many different breeds, e.g. DRB1�15/DQA1�0601 was found

in 28/51 (55%) of breeds tested, whereas others show a more restricted distribution, e.g.

DRB1�15/DQA1�0901 in 10 different breeds, DRB1�0101/DQA1�0301 in three

different breeds, and DRB1�24/DQA1�0101 in only one breed.

4. Discussion

We have typed a cohort of dogs for DLA-DRB1 and DQA1, and established the

haplotypic combinations present between alleles of these two loci. While haplotypes were

not established using family data, the use of DLA homozygous animals allowed us to

assign most haplotypes unequivocally. The data shows that there is considerable

interbreed variation of both allele and haplotype frequencies and that the number of

different alleles and haplotypes may be restricted in many breeds. As yet, many breeds

have only been tested in small numbers, and this data can only be considered as

preliminary. However, there are many indications that this interbreed variation will be

confirmed. In many cases, although there were only two or three unrelated individuals

Table 4DLA-DRB1-DQA1 haplotype frequencies (%)

Haplotype DRB1-DQA1 Frequency 2n � 362 Haplotype DRB1-DQA1 Frequency 2n � 362

01-0101 14.6 11-0201 2.8

01-0301 1.9 11-0701 2.2

01-0901 0.8 12/19-0401a 8.8

02-0901 6.1 13-0101 3.3

04-0201 1.1 15-0401 1.7

04-0701 1.7 15-0601 20.9

05/23-0301a 0.8 15-0901 5.3

06-0101 1.1 18-0101 4.7

06-0201 0.8 20-0401 1.9

06-0401 1.4 20-0601 1.1

06-05011 7.2 20-0701 0.6

06-0701 0.8 24-0101 1.1

03/09-0101a 1.9 Others 5.3

a Some of the early experiments could not distinguish DRB1�03 from DRB1�09, DRB1�05 from DRB1�23, norDRB1�12 from DRB1�19.

108 L.J. Kennedy et al. / Veterinary Immunology and Immunopathology 69 (1999) 101±111

from one breed, they either had identical alleles, or shared more alleles than might be

expected by chance.

The number of dogs homozygous at both DRB1 and DQA1 (35.4%), is high

considering the number and frequency of alleles present. These data do not fit the Hardy±

Weinberg equilibrium (p < 0.0001) and would suggest that some breeds are highly inbred

with respect to their MHC region. This may be relevant to diseases where the immune

system is involved in disease pathology, and justifies further analysis of the dog MHC in

terms of disease susceptibility.

It is not clear why some breeds have a more restricted set of DLA alleles and haploypes

than others, but there are several possible explanations. While recent pedigree dogs

follow carefully controlled breeding programmes, with strict control on which animals

mate, early domestic dog populations will probably have had repeated exposure to new

alleles and gene admixture, as there would have been little to prevent domesticated

females mating with wild males. Thus, we might expect older dog breeds to have a wider

range of alleles at any particular locus and younger breeds to be more restricted in their

range. Many breeds have been created relatively recently, within the last 150 years, often

from a small number of founder animals. Current practice by dog breeders often results in

only a few `Champion' sires being used to father most of the next generation. Some rarer

breeds, which never had a large population base may have been through genetic

bottlenecks, where only a few animals survived through a time when that particular breed

was not popular. Other breeds such as Japanese Akitas, which date from the 1600s, have

Fig. 1. Haplotype frequencies (%) in five different breeds.

L.J. Kennedy et al. / Veterinary Immunology and Immunopathology 69 (1999) 101±111 109

never achieved a large population base, because they were exclusively owned by certain

groups of people for many years. All these factors may combine to reduce the number

and variety of DLA alleles and hapolotypes in any one breed.

Other breeds such as the German Shepherd Dog and Labrador are, however, known to

have existed as breeds for over 1000 years (Wayne, 1993). Both of these breeds are also

very common, and so have a large population base, which has probably never dropped

very low in numbers. German Shepherd Dogs are also phenotypically closer and, thus,

probably genetically nearer to the domestic dogs ancestors, the grey wolf, than many

other breeds. These dogs may well have had repeated exposure to the wild-type gene

pool, especially in the early centuries of domestication. All these factors would help

maintain a higher number of DLA alleles and haplotypes than other rarer breeds.

In a recent report from the American Kennel Club (AKC) on the major health issues in

dogs, four out of the top eleven conditions had an immune component. These were

hypothyroidism, autoimmune diseases (e.g. canine rheumatoid arthritis, SLE, diabetes),

cancer and allergic dermatitis. These diseases represent important veterinary problems

and also serve as interesting naturally occurring models for human disease. One of the

features common to immune diseases is that they often have a genetic component and this

may be particularly important in pure bred dogs which have a restricted gene pool. This

restriction is important because: (a) most infectious diseases have an immune component;

and (b) immune diseases can be triggered by exogenous influences such as drugs or

vaccines. As in man, variation in the immune response to infection and vaccination is

expected to be dependent on genetic background.

If we are to address these major canine health issues and the role of the immune

response, it is important to fully understand the genetic influence on the immune response

in dogs. Central to this is the characterisation of the MHC and the delineation of how

polymorphic the genes in this region are. In dogs, these regions are poorly characterised

despite the fact that a greater understanding of these genes and how they operate is

important for dissecting disease immunopathology. Characterisation of the genes within

the canine MHC will assist in the study of canine autoimmune diseases and variations in

immune response to infection and vaccination. This could ultimately lead to a better

understanding of the causes of these conditions and, hence, their treatment or even

prevention via selective breeding programmes.

References

Bull, R.W., Vriesendorp, H.M., Cech, R., Grosse-Wilde, H., Bijma, A.M., Ladiges, W.L., Krumbacher, K.,

Doxiadis, I., Ejima, H., Templeton, J., Albert, E.D., Storb, R., Deeg, H.J., 1987. Joint report of the Third

International Workshop of Canine Immunogenetics. II. Analysis of the serological typing of cells.

Transplantation 43, 154±161.

Charron, D., 1997. HLA, Genetic Diversity of HLA. EDK, Paris.

Deeg, H.J., Raff, R.F., Grosse, W.H., Bijma, A.M., Buurman, W.A., Doxiadis, I., Kolb, H.J., Krumbacher, K.,

Ladiges, W., Losslein, K.L., Schoch, G., Westbroek, D.L., Bull, R.W., Storb, R., 1986. Joint report of the

Third International Workshop on Canine Immunogenetics. I. Analysis of homozygous typing cells

Transplantation 41, 111±117 [published erratum appears in Transplantation 41 (1986) 667].

Francino, O., Amills, M., Sanchez, A., 1997. Canine MHC DRB1 genotyping by PCR-RFLP analysis. Animal

Genetics 28, 41±45.

110 L.J. Kennedy et al. / Veterinary Immunology and Immunopathology 69 (1999) 101±111

Hayney, M.S., Poland, G.A., Jacobson, R.M., Schnaid, D.J., Lipsky, J.J., 1996. The influence of the HLA-

DRB1�13 allele on measles vaccine response. J. Investig Med. 44, 261±263.

Hill, A.V., Allsopp, C.E., Kwiatkowski, D., Anstey, N.M., Twumasi, P., Rowe, P.A., Bennett, S., Brewster, D.,

McMichael, A.J., Greenwood, B.M., 1991. Common West African HLA antigens are associated with

protection from severe malaria. Nature 352, 595±600.

Kennedy, L.J., Carter, S.D., Barnes, A., Bell, S., Bennett, D., Ollier, W.E.R., Thomson, W., 1998. Nine new dog

DLA-DRB1 alleles identified by sequence based typing, Immunogenetics. 48, 296±301.

Kennedy, L.J., Alter, L., Angles, J.M., Barnes, A., Carter, S.D., Francino, O., Gerlach, J.A., Happ, G.M., Ouier,

W.E.R., Polvi, A., Thomson, W., Wagner, J.L., 1999a. Nomenclature for the factors of the Dog Major

Histocompatibility system (DLA), 1998: First report of the ISAG DLA Nomenclature Committee. Animal

Genetics, in press.

Kennedy, L.J., Carter, S.D., Barnes, A., Bell, S., Bennett, D., Ollier, W.E.R., Thomson, W., 1999b. DLA-DRB1

polymorphisms in dogs defined by sequence specific oligonucleotide probes (SSOP). Tissue Antigens 53,

184±189.

Polvi, A., Garden, O.A., Elwood, C.M., Sorensen, S.H., Batt, R.M., Maki, M., Partanen, J., 1997. Canine major

histocompatibility complex genes DQA and DQB in Irish setter dogs. Tissue Antigens 49, 236±243.

Sarmiento, U.M., DeRose, S., Sarmiento, J.I., Storb, R., 1992. Allelic variation in the DQ subregion of the

canine major histocompatibility complex: I. DQA. Immunogenetics 35, 416±420.

Sarmiento, U.M., DeRose, S., Sarmiento, J.I., Storb, R., 1993. Allelic variation in the DQ subregion of the

canine major histocompatibility complex: II. DQB. Immunogenetics 37, 148±152.

Sarmiento, U.M., Sarmiento, J.I., Storb, R., 1990. Allelic variation in the DR subregion of the canine major

histocompatibility complex. Immunogenetics 32, 13±19.

Sarmiento, U.M., Storb, R.F., 1988. Restriction fragment length polymorphism of the major histocompatibility

complex of the dog. Immunogenetics 28, 117±124.

Sarmiento, U.M., Storb, R., 1990. Nucleotide sequence of a dog DRB cDNA clone. Immunogenetics 31, 396±

399.

Tiwari, J.L., Terasaki, P.I., 1985. HLA and Disease Associations. Springer-Verlag, New York.

Vriesendorp, H.M., Albert, E.D., Templeton, J.W., Belotsky, S., Taylor, B., Blumenstock, D.A., Bull, R.W.,

Cannon, F.D., Epstein, R.B., Ferrebee, J.W., Grosse-Wilde, H., Hammer, C., Krumbacher, K., Leon, S.,

Meera, Khan.P., Mickey, M.R., Motola, M., Rapaport, F.T., Saison, R., Schnappauf, H., Scholz, S.,

Schroeder, M.L., Storb, R., Wank, R., Westbroek, D.L., Zweibaum, A., 1976. Joint report of the Second

International Workshop on Canine Immunogenetics. Transplant Proc. 8, 289±314.

Vriesendorp, H.M., Westbroek, D.L., D'Amaro, J., van der Does, J.A., van der Steen, G.J., van Rood, J.J.,

Albert, E., Bernini, L., Bull, R.W., Cabasson, J., Epstein, R.B., Erikson, V., Feltkamp, T.E., Flad, H.D.,

Hammer, C., Lang, R., Largiader, F., Loringhoven, K.v., Los, W., Meera Khan, P., Saison, R., Serrou, B.,

Schnappauf, H., Swisher, S.N., Templeton, J.W., Uhlschmidt, G., Zweibaum, A., 1973. Joint report of 1st

International Workshop on Canine Immunogenetics. Tissue Antigens 3, 145±163.

Wagner, J.L., Burnett, R.C., DeRose, S.A., Storb, R., 1996a. Molecular analysis and polymorphism of the DLA-

DQA gene. Tissue Antigens. 48, 199±204.

Wagner, J.L., Burnett, R.C., Storb, R., 1996b. Molecular analysis of the DLA DR region. Tissue Antigens. 48,

549±553.

Wagner, J.L., Burnett, R.C., Works, J.D., Storb, R., 1996c. Molecular analysis of DLA-DRBB1 polymorphism.

Tissue Antigens. 48, 554±561.

Wagner, J.L., Hayes-Lattin, B., Works, J.D., Storb, R., 1998a. Molecular analysis and polymorphism of the

DLA-DQB genes. Tissue Antigens 52, 242±250.

Wagner, J.L., Works, J.D., Storb, R., 1998b. DLA-DRB1 and DLA-DQB1 histocompatibility typing by PSR-

SSCP and sequencing. Tissue Antigens 52, 397±401.

Wayne, R.K., 1993. Molecular evolution of the dog family. Trends Genetics 9, 218±224.

Williamson, P., Nicholas, F.W., Stewart, G.J., 1989. Restriction fragment length polymorphism analysis of

canine class II major histocompatibility complex genes. Transplant Proc. 21, 3751±3752.

L.J. Kennedy et al. / Veterinary Immunology and Immunopathology 69 (1999) 101±111 111