www.elsevier.com/locate/ygeno

Genomics 85 (20

High-resolution RH map of horse chromosome 22 reveals a putative

ancestral vertebrate chromosome

Ashley Gustafson-Seaburya,1, Terje Raudseppa,1, Glenda Goha, Srinivas R. Katab,

Michelle L. Wagnerc, Teruaki Tozakid, James R. Mickelsonc, James E. Womackb,

Loren C. Skowa, Bhanu P. Chowdharya,e,*

aDepartment of Veterinary Integrative Biosciences, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University,

College Station, TX 77843, USAbDepartment of Veterinary Pathobiology, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University,

College Station, TX 77843, USAcDepartment of Veterinary Biosciences, University of Minnesota, 295f AS/VM, St. Paul, MN 55108, USA

dLaboratory of Racing Chemistry, 1731-2 Tsurutamachi Utsunomiya Tochigi, Utsunomiya, JapaneDepartment of Animal Science, College of Agriculture and Life Science, Texas A&M University, College Station, TX 77843, USA

Received 17 September 2004; accepted 22 October 2004

Available online 9 December 2004

Abstract

High-resolution gene maps of individual equine chromosomes are essential to identify genes governing traits of economic importance in

the horse. In pursuit of this goal we herein report the generation of a dense map of horse chromosome 22 (ECA22) comprising 83 markers, of

which 52 represent specific genes and 31 are microsatellites. The map spans 831 cR over an estimated 64 Mb of physical length of the

chromosome, thus providing markers at ~770 kb or 10 cR intervals. Overall, the resolution of the map is to date the densest in the horse and is

the highest for any of the domesticated animal species for which annotated sequence data are not yet available. Comparative analysis showed

that ECA22 shares remarkable conservation of gene order along the entire length of dog chromosome 24, something not yet found for an

autosome in evolutionarily diverged species. Comparison with human, mouse, and rat homologues shows that ECA22 can be traced as two

conserved linkage blocks, each related to individual arms of the human homologue—HSA20. Extending the comparison to the chicken

genome showed that one of the ECA22 blocks that corresponds to HSA20q shares synteny conservation with chicken chromosome 20,

suggesting the segment to be ancestral in mammals and birds.

D 2004 Elsevier Inc. All rights reserved.

Keywords: Horse; ECA22; Gene mapping; RH map; FISH; Comparative map; Human; Mouse

The first-generation horse whole genome radiation

hybrid (RH) and comparative map represents the most

comprehensive framework of mapped loci in the equine

genome [1]. It contains both gene-specific and polymorphic

markers and integrates the syntenic, cytogenetic, and genetic

linkage maps into a single consensus map. Despite serving

0888-7543/$ - see front matter D 2004 Elsevier Inc. All rights reserved.

doi:10.1016/j.ygeno.2004.10.012

* Corresponding author. Department of Veterinary Integrative Bio-

sciences, College of Veterinary Medicine and Biomedical Sciences, Texas

A&M University, College Station, TX 77843–4458.

E-mail address: bchowdhary@cvm.tamu.edu (B.P. Chowdhary).1 These authors have contributed equally to this work.

as an important foundation for understanding the organ-

ization of the equine genome, the resolution of the current

map is inadequate for rapid discovery of markers closely

linked to traits of interest in the horse. Additionally, it is also

insufficient for use in a candidate gene approach to identify

genes responsible for traits significant to the equine

industry. Among the most apparent drawbacks associated

with the map are: (i) presence of regions on a number of

equine chromosomes that lack adequate numbers of mapped

gene-specific and polymorphic markers and (ii) absence of

uniformity in the distribution of markers over individual

chromosomes.

05) 188–200

A. Gustafson-Seabury et al. / Genomics 85 (2005) 188–200 189

To date very few genes controlling traits of economic

significance have been identified in the horse. Among the

genes identified include those governing monogenic traits,

e.g., coat color and some inherited diseases [2]. The

discoveries of most of these genes were facilitated by

knowledge of homologous genes for similar phenotypes in

humans or other mammalian species. Thus an a priori

knowledge of the causative gene and its mutation led to the

identification of the equine ortholog and its causative

mutation. However, most of the future research aimed at

identifying genes governing economically important traits in

the horse will require a set of genome-wide polymorphic

markers (genome scan panel) appropriate to the population

to be analyzed and high-resolution physically ordered

comparative maps to align the horse map accurately with

its mammalian counterparts. The latter will be critical for the

candidate gene approach used for isolating genes potentially

implicated in the hereditary condition analyzed. The recent

development of the high-resolution RH and comparative

map for horse chromosome 17 [3] and the X chromosome [4]

represents the first steps toward targeted expansion of gene

maps in the horse to obtain dense maps. Though the

approach used for these two chromosomes has provided

resolutions ranging between 1.4 to 1 Mb per marker, it

involved alignment of sequence from multiple species,

identification of conserved regions, and tedious development

of primer pairs to obtain horse-specific amplification [3,4].

Overgos (overlapping oligonucleotide primer pairs) are

probes that hybridize to conserved genomic regions (primar-

ily exons) and can be used for improved multiprobe high-

density filter hybridization to enable rapid identification of

BAC clones containing specific markers [5,6]. To improve

upon the development of expanded gene maps in the horse,

we used this approach to generate a high-resolution RH and

comparative map of horse chromosome 22 (ECA22).

Comparative analysis shows that ECA22 and its homologues

in various mammalian species represent one of the most

conserved genomic segments across a range of evolutionarily

diverged species [7]. The chromosome corresponds to

human chromosome 20 (HSA20) [8], which contains genes

implicated in, for example, spongiform encephalopathies

[9,10], Kindler syndrome [11], and congenital hereditary

endothelial dystrophy of the cornea [12,13].

The current map of ECA22 contains a total of 23 markers

(7 genes/ESTs and 16 microsatellites), assigned by RH and

linkage analysis [1,14,15]. Of the three linkage groups, only

two are physically aligned to the chromosome. The

proximal one-third of the chromosome has very few mapped

markers. To remedy these deficiencies, we undertook

development of a high-resolution map of the chromosome

with the goal of defining gene-specific markers at every

megabase interval. This would finely align ECA22 to the

human and mouse chromosomes, facilitate identification of

candidate loci for various equine conditions associated with

genes located on this chromosome, and help to understand

the relative organization and hence evolution of the

homologues of this chromosome among different mam-

mals/vertebrates.

Results

Overgo hybridization and isolation of BAC clones

containing targeted genes

A total of 45 overgo probes were designed within

conserved regions of genes selected at ~1 Mb intervals

from the terminal part of the short arm to the terminal part of

the long arm of HSA20. The first pooled hybridization with

18 overgo probes for genes located on HSA20p to the high-

density filters of the CHORI-241 horse genomic BAC

library yielded 63 BACs. The second pooled hybridization

with 27 overgo probes representing genes from HSA20q

yielded 103 BACs. The rearray and hybridization of the

positive BACs with individual overgo probes indicated that

4 of the probes from the short arm, and 8 from the long arm,

failed to identify any BACs. Thus, of the 45 overgo probes,

BAC clones were obtained for 33 sets, averaging 5.2 BACs/

overgo pair (Table 1).

To confirm the identity of the BACs, a single represen-

tative BAC clone for each overgo probe was used for direct

sequencing with one of the overgo primers as sequencing

primer. All representative clones provided on average 450

bp sequence with N80% identity with the respective human

ortholog. The BAC for TCF15 showed no amplification.

BAC end sequencing, sequence-tagged site (STS)

development, and optimization of PCR

Analysis of sequence data from one end of the 33

representative clones resulted in the generation of STSs from

only 12 of the ends. BACs that did not provide end sequence

from one end were sequenced from the other end. This

resulted in an additional 20 STS markers. BLAST analysis

[16] of the STS sequences showed that none of the end

sequences corresponded to coding regions of specific genes.

Most of the end sequences that could not be converted to

STSs comprised repetitive elements. Following optimization,

horse-specific amplification was obtained for 29 of the STSs.

Similarly, optimization was also carried out with primer pairs

from eight equine ESTs, eight equine orthologs for human

genes, and 19 microsatellite markers on ECA22 (Table 1).

Fluorescence in situ hybridization (FISH) mapping

All representative BAC clones isolated for the 33 overgo

probes localized to ECA22. One of the clones contained two

genes (BMP2 and CHGB). Further, BACs obtained for

AHT030, UMNe276, and UMNe344 mapped to ECA13q, in

contrast to previous erroneous FISH assignment of AHT030

[17] and subsequent mapping of loci in this RH group to

ECA22 [1]. The remaining five BACs representing ASIP,

Table 1

Composite information about all new genes (ordered according to human sequence map) and microsatellite markers mapped to ECA22

Symbol Name Overgo primers

5V–3V

BAC

CHORI-241

FISH location

on ECA22

PCR primers

5V–3V

Produc

size (bp)

Ta (8C) Reference

Accession No.

ATP5Ea ATP synthase,

H+ transporting,

mitochondrial F1

complex, q subunit

F: ACATCCGATACTCCCAGATCTGTG 161 L16 q19 F: GGAAGTCATTAGCTCCTGTCA 250 62 —

R: TCTCTCACTGCTTTTGCACAGATC R: CATGAAGAGCCTGGAACAAGT

ATP9Aa ATPase, class II, type 9A F: ACTGGCTCATGACAGTGGCGGAGC 355 M06 q18 F: GTCTTGGCTCTTAACGCTCAC 162 58 —

R: CAGGCCAGGCTGAGCAGCTCCGCC R: TCGTGGCTTGTTTGTTGTTAG

BMP2a Bone morphogenetic

protein 2

F: GCTGTCTTCTAGCGTTGCTGCTTC 174 N22 q14dist–q15prox F: AACTCCTCTGCATGGTGTGT 204 58 —

R: CCCAGGAGGACCTGGGGAAGCAGC R: TGTGGGGCCTAAGTTCTCTA

C20orf24 Chromosome 20 ORF 24 — — — F: AGCTGCATGGAACCTGGA 250 58 Equine EST

HEST20-5 R: CGACAGCTCCAATCAGCA BI961817

CDH4a Cadherin 4 F: GTTCATCAACCAGGTCTACAACGG 012 C18 q19 F: GGCCAGGAAATGAGTGTCTA 172 58 —

R: GCCCTCGTCCACGGAGCCGTTGTA R: AATCAGCGTCAATGCTTTTC

CENPBa Centromere protein B

(80 kDa)

F: GAGGCCATGGCTTACTTTGCCATG 185 P08 q14dist–q15prox F: CAGGAAGGTCTCGGCATC 285 60 —

R: TCAGGTACCTCTTGACCATGGCAA R: GTATGAAGGAGGGCACATCA

CHGBa Chromogranin B F: TTCTTCCCAGAATACAACTATGAC 174 N22 q14dist–q15prox Same as BMP2 — — —

R: GCTTTTTCTCCCACCAGTCATAGT

CPXMa Carboxypeptidase X

(M14 family)

F: TGCTTTGAGGTCACTGTGGAGCTG 157 M03 q14dist–q15prox F: ACGTGGCAACACATCTCCTA 167 58 —

R: GGAACTTGTCACAGGACAGCTCCA R: TGGTGATCTTCATTCGTTCTC

CRNKL1 Crooked neck-like 1

(Drosophila)

F: CAGCTTCGAGAATTTGACAGATGC 155 P23 q12–q13 No end sequence from this BAC clone — — —

R: TTTCATAAAGCTTCCGGCATCTGT

DAP4 Discs, large (Drosophila)

homolog-associated protein 4

— — — F: GGTTCGCCTCACAACTAAAC 288 TD60 Equine testis

HEST20-3 R: ACATGGCAGCTTCAAGGTTA cDNA

DOK5a Docking protein 5 F: TGATGCCATCTCCTAACTTAGATG 374 L16 q18 F: GTTTGCTCTCTTCCTTCATGC 182 58 —

R: GCACATTCGCCATGTACATCTAAG R: TGAGAAGCTACAACCAACCAA

DSTNa Destrin F: GCACCTCTGAAAAGTAAAATGATC 054 B22 q12–q13 F: CAAGCACTAAGCTGGGAAGTA 194 58 —

R: CCTTGGAGCTTGCATAGATCATTT R: CGACGGAGTAGGTCACAAAG

EYA2a Eyes absent homolog 2

(Drosophila)

F: ACTGGATGAGGAAGCTGGCCTTCC 027 F10 q18 F: CGGGTCTTACTCAGCTTCAG 191 58 —

R: TTCACCCGCCGGTAGCGGAAGGCC R: GCCTTATACGTGTCTGTCTGC

FKBP1A FK506 binding protein

1A, 12 kDa

— — — F: CAGCCATCATGACAAATCC 166 63 Equine EST

R: TCACCACATCACTACACAGG BI961944

HAO1a Hydroxyacid oxidase 1 F: GGTGCCAGAATGTGAAAGTCATCG 087 N18 q14 F: CCATTGTCATTTGGACCTGT 189 58 —

R: CTCACCAATGTCTTGTCGATGACT R: TGCTATTGGGTAGTTTATGCAC

ITCHb Itchy homolog E3

ubiquitin protein ligase (mouse)

— — — F: GACGTTTCTGTTGGGAACTG 248 58 —

R: AAAACAAATGGCTAAACCAAGG

ITGB4BPa Integrin h4binding protein

F: CCCAAGACTTCAATTGAAGACCAG 149 K17 q15–q16prox F: CAGCAAGGTATCCAGAGCAT 284 60 —

R: GAGAAGACAGCTCATCCTGGTCTT R: AGGGAATATGGGTGACGAGT

JAG1a Jagged 1 F: GATTTGGTTAATGGTTATCGCTGT 060 E04 q14 F: TCCCTCTTACTCAAATCAGCC 128 56 —

R: AGCCAGGTGGACAGATACAGCGAT R: GAAAGGTCTAAAAGTGCTGCC

KIAA1404 KIAA1404 protein — — — F: GGGCGTAAATCATACTCTGG 112 58 BI961572

HEST20-4 R: GATCTCCTCAAAGTTCATCAGG

KIAA1847 KIAA1847 protein — — — F: AGGAGAAGATCGAGCAGACC 416 63 BI961508

HEST20-8 R: GTCATCTTCTTGTGGGTGTCTG

MYL9b Myosin, light polypeptide 9, — — — F: GCCAAGGACAAAGACGACTA 230 58 —

regulatory R: CCTGTGTTACATTGGCTGTG

NAPBa N-ethylmaleimide-sensitive F: CTAGAAGCTCATGAAGAACAGAAC 363 K04 q12–q13 F: GATGGATGTGTAAAGGAGTGG 199 58 —

factor attachment protein, h R: CAGTGTAAGCTTCACTGTTCTGTT R: CTTTCTTTGTGGTTGATTGG

NTSR1 Neurotensin receptor 1 F: CGCGGAAGAAGTCGCTGCAGAGCC 396 F21 q19 Repetitive sequence; unsuitable — — —

R: TGCACCGTGCTCTGCAGGCTCTGC for PCR primer design

PAK7

HEST20-2

p21 (CDKN1A)-activated

kinase 7

— — — F: AGCACAACTCTCCTATGTCGG

R: AAGGCCATGAGAATTCCAGTC

279 58 Equine

monocyte

cDNA

HEST20-2

PDYNa Prodynorphin F: ATTCGTCCCAAGCTCAAGTGGGAC 145 E02 q15–q16prox F: CCCAGGTTAGGACAGACTCA 227 62 —

R: CATAGCGCTTCTGGTTGTCCCACT R: CCCCGATAGACACATTTTCTT

PLCB1b Phospholipase C, h1 — — — F: CAACCATACAAAGCCCCATC 227 58 —

R: GCTCAGTTTTCACCAAATGCC

PLCB4 Phospholipase C, h4 — — — F: CGCAGACCAGCAACAGTAG 108 58 AY011799

R: TCCATGGAATTTGTGTCCAG

A.Gusta

fson-Seabury

etal./Genomics

85(2005)188–200

190

(continued on next page)

PPGBa Protective protein for

h-galactosidaseF: TCCCTGGTCTACTTTGCCTACTA

R: TCCCCAGAAGGCCATGGTAGTAGG

370 C22 q17 F: TAGTTTGCCATGCCTATTCA

R: AGTATTTTAGGAGGCTTAATGTGA

151 58 —

PRNPa Prion protein F: TGACTATGAGGACCGTTACTATC 071 K03 q14dist–q15prox F: CTTTTTGGTGGAA TTTAG 493 58 —

R: ACGGTGCATGTTTTCACGATAGTA R: ACTCAACAGCAC TGAA

PSMA7a Proteasome (prosome,

macropain) subunit, a type, 7

F: CCATCACCGTCTTCTCGCCCGAC

R: ACTTGGAAGAGGTGGCCGTCGGGC

012 E24 q19 F: GAATCGGTCAGGA GAGA

R: GAACGTGCAGGG TTTG

193 58 —

PTPRTa Protein tyrosine

phosphatase, receptor

type, T

F: GATCTACATCCAGTGGAAACCTC

R: CCCATTGGTCTCATTGGGAGGTTT

401 L11 q16–q17prox F: GGATGCTTCTAAC CCTT

R: GTTGTGGTCTGGG CAAG

171 60 —

RALYa RNA binding protein

(autoantigenic,

hnRNP-associated

with lethal yellow)

F: CAATGAGCGCCATGCCCGGGCAG

R: ATTCTCTCCCAGCACAGCTGCCCG

352 I04 q15–q16prox F: CCTTCTGCTCACT TTTTA

R: CACTACGCACTGG AAGG

230 58 —

RBL1 Retinoblastoma-like

1 (p107)

F: ACTTTGCTGTAAACAGACTAAAG

R: TACAAAATTTCTGCCAGCTTTAGT

372 P16 q17 PCR primers amplify mster. H analysis. — — —

RPS21a Ribosomal protein S21 F: CAGAACGACGCCGGCGAGTTCGT 056 B24 q19 F: GAATAGCCAGCTT TCTGA 179 58 —

R: GCGGCACGTACAGGTCCACGAACT R: GACACACCAAAG ACAA a

SEC23Ba Transport protein Sec23

isoform B

F: CCTCTCGGGAACTGAAGATTGCA

R: CATGGACCAATGGCTCCTGCAATC

338 P22 q12–q13 F: GGAATAAGTGGAA CCAA

R: CTTACTATGGAGG AGTG TG

208 60 —

SGK2a Serum/glucocorticoid

regulated kinase 2

F: TTGACCCAGAGTTCACCCAGGAA

R: ATGGACTTGGACACAGCTTCCTGG

008 E15 q15–q16prox F: GAAGAGAAGGGG ACAGA

R: CCTCAGAACCACC GGAA

206 60 —

SNAI1a Snail homolog 1

(Drosophila)

F: TTCCAGCAGCCCTACGACCAGGC

R: TGGCTGCCAGCAGGTGGGCCTGGT

104 I23 q18 F: GCTAGGAAGTGGT AAGC

R: TGTGGAGTAAAA CTGG

321 58 —

SNRPB1b Small nuclear

ribonucleoprotein

polypeptides B and B1

— — — F: AAATGTTCTCTCT CAGC

R: AATGCCCAATTCC CCTT

110 58 —

SNRPB2a Small nuclear

ribonucleoprotein

polypeptide B

F: ATTGTGGCTTTAAAGACCATGAA

R: AGGCCTGCCCCCTCATCTTCATGG

397 G22 q12–q13 F: TGACCTTGGGGTC TTTG

R: GGCGTATCGGAGA CATAA

230 58 —

SNTA1a Syntrophin, a1

(acidic component)

F: TGGTGCACTCAGGCCCCTCCAAG

R: TCGTAGGGCACTGAGCCCTTGGAG

189 H02 q15–q16prox F: CCTTTCTGCGCTT TTTC

R: GGGTAAATCCGAA TTGA

383 60 —

SRCb v-src sarcoma

(Schmidt–Ruppin A-2)

viral oncogene

homolog (avian)

— — — F: TCACCTTGGGAAA CCAC

R: AGTGGATGGGGA ACAA

158 58 —

TAF4Aa TAF4 RNA

polymerase II,

TATA box binding

protein-associated

factor

F: ACAAAGAATCACGCGGGTCAACC

R: AAATATGAGGTCCCTGAGGTTGA

095 K05 q19 F: TGGGAGGTAGGCA CAAG

R: TAACGGTCAGGG GGAA

298 58 —

TCF15a??? Transcription

factor 15

F: AGGAGAACCGCAGCGAGAGCGAC

R: AACGACTGGTCCGACGCGTCGCTC

447 D16 q13dist–q14prox F: CTGCAAGAGAGAT AGGA

R: GCCGGAACGTAA GTGA

216 58 —

TCFL5b Transcription

factor-like 5

(basic helix-loop-helix)

— — — F: CTCTCATTCGACA CATC

R: GCAATCCAATATC GGTG

500 50 —

A.Gusta

fson-Seabury

etal./Genomics

85(2005)188–200

191

TT

AT

A

AC

AG

T

AA

C

ha

C

AC

A

TG

G

T

G

CC

CC

CA

TA

T

AT

C

A

CC

G

AT

G

GT

TC

CT

G

AGG

G

TC

TTC

No R

C

GG

GTT

AG

C

G

T

AA

C

A

CA

A

AG

G

AC

C

TG

Symbol Name Overgo primers

5V–3V

BAC

CHORI-241

FISH location

on ECA22

PCR primers

5V–3V

Produc

size (bp)

Ta (8C) Reference

Accession No.

TDE1a Tumor differentially

expressed 1

F: ATGCTCTGCTTGGCTTCCTTGTA

R: TCAGGGTCATCATGATGTACAAGG

099 O22 q15–q16prox F: GCCCTGGCACACAATAGA

R: GCATTAGTTTTCCCATCTGC

172 60 —

TKY498 Microsatellite — — — F: GGTGGGAGCATTATCTTTG 256 58 AB103716

R: CTGTCTTTGTGCGTTTGGAG

TKY548 Microsatellite — — — F: CAGCAAGACGTCTGTCCAT 173 58 AB103766

R: TACCCTGGGACTCATGCTCT

TKY554 Microsatellite — — — F: GGCCTTCCTGACTCCAATT 240 58 AB103772

R: AGCGTTTGCTGAAACCCATA

TKY560 Microsatellite — — — F: ATCCACCCTCTGCTCCTCT 150 58 AB103778

R: CCCAAGGGAAAAATGGAAAA

TKY785 Microsatellite — — — F: TTAGGTGTCAAGTCTAGGA 192 58 AB104003

R: GTGGGGAAGACGCTAAAGAG

TKY1009 Microsatellite — — — F: ACAATGTCTGCAACTCACT 121 58 AB104227

R: GCACTTTGCCACATTTGTTTC

TKY1106 Microsatellite — — — F: TGCCACAACTAGAAGGATT 280 58 AB104324

R: GCAGAACGTGCGAACTTAAC

TOP1b Topoisomerase (DNA) I — — — F: TCCCATTCACTCTTCTCCC 113 58 —

R: ATCCAGCCAAGAAAAGGCTC

UMNe105 Microsatellite — 078 B07 q15–q16prox F: TGTGTGGCCAGTTTATAACCATA 210 60 AF536249

R: AAACAAGCCAATCCCAAAAG

UMNe137 Microsatellite — — — F: CTGCTTTTTACTGCTTCAGTG 114 58 AY391299

R: GATTTGAGTCGAGGTCTGCC

UMNe183 Microsatellite — — — F: AGAGAACAGGAGAAGACATGC 106 58 AY391316

R: ACCCTGTCTCATGGAAGCTG

UMNe208 Microsatellite — — — F: AGAGCAGAACGCAACTTTC 140 58 AF536290

R: AGGAGATGCGCATATGTGC

UMNe215 Microsatellite — — — F: AACTGCAGCCTGGTGAG 197 58 AF536296

R: GACCTCCTCCTTTCCTTTCTG

UMNe355 Microsatellite — — — F: TCATGCAATTCTGAAAAAGAG 200 58 AY391346

R: CAGAGCAAATCTTCCTCACC

UMNe441 Microsatellite — — — F: GATAAACTGTCCTCCTGCTGA 150 62 —

R: GGCAGAAAGTTCTGTGGAGTG

UMNe466 Microsatellite — — — F: AGCATCCAGTCCACTGCA 378 58 —

R: GGCAGGTAAGTTTGAGAACCC

UMNe490 Microsatellite — — — F: TTGAAAAGGACACGTGCAA 300 55 —

R: TAGGCCACCTTCCAAAATTG

UMNe499 Microsatellite — — — F: CTGCAGCCCGTGTGTATG 100 60 —

R: TTAAAAGGGACTTGAGAGGGC

UMNe537 Microsatellite — — — F: TGCATTACGGGAAAGGAATC 150 58 —

R: CTGAACTGCACACTGAGAAATG

VIAATa Vesicular inhibitory

amino acid transporter

F: TCGACTTTGAGCACCGCCAGGGC

R: AGGATGTCCATCTGCAGGCCCTGG

156 E20 q15–q16prox F: GGCAGCATGGTTGTAATAGT

R: TGTATGTATTCTTCGGGGTCA

178 58 —

XRN2b 5V–3Vexoribonuclease 2 — — — F: TGGACATCAGTACCCTTCA 184 58 —

R: AGCACACAAAACACATGCAG

YWHAB

HEST6-7

Tyrosine

3-mono-oxygenase/tryptophan

5-monooxygenase activation

protein, h polypeptide

— — — F: CCCAATGCTACACAACCA

R: CCGAGAAATTGAGTGCCAG

~1000 62 Equine EST

BI961675

FISH locations shown in bold font are from this study. TD, touchdown PCR.

a Primers for RH designed from BAC end sequences.

b Primers for RH designed through multiple alignment of other mammalian sequences.

Table 1 (continued)A.Gusta

fson-Seabury

etal./Genomics

85(2005)188–200

192

Fig. 1. Retention frequency of 83 gene-specific and STR markers on horse chromosome 22. The x axis shows distribution of markers from the pter to th ter region on the chromosome and the y axis shows the

frequency of retention for individual markers. The centromere is demonstrated by an arrowhead. Marker retention was on average uniform throughou e chromosome (~21.4%).

A.Gusta

fson-Seabury

etal./Genomics

85(2005)188–200

193

e q

t th

A. Gustafson-Seabury et al. / Genomics 85 (2005) 188–200194

AHCY, COR022, UMNe077, and UMNe105 mapped to

ECA22. Thus a total of 38 new FISH assignments provided

markers on each band of the chromosome. A summary of

these localizations is presented in Fig. 2 (markers followed

by a solid circle). The physical order of 19 of the closely

located/overlapping markers was deduced using double-

color FISH on metaphase or interphase chromosomes (Fig.

3; see markers in bold italic font; Fig. 2).

RH typing and map development

The 64 new markers described above were typed on

the 5000-rad horse � hamster RH panel. The data were

added to previously published results on 21 markers (7

genes and 14 microsatellites) [1]. Three of the markers

belonging to a terminal RH group were excluded from the

analysis following discovery that they map to another

A. Gustafson-Seabury et al. / Genomics 85 (2005) 188–200 195

chromosome (described above). Final analysis was thus

carried out with a total of 83 markers, of which 52

represented putative genes and 31 were microsatellites.

The average retention frequency of the markers was

21.4%, with no significant peak in any region (Fig. 1).

Initial analysis showed that all markers aligned in a single

linkage group at LOD z7. Multipoint analysis of the

markers showed that at odds of 1:1000, 32 of the markers

formed the framework, to which the remaining 51 markers

were added in successive stages. The total length of the

map is 831 cR. The only notable gap between markers is

found in the middle of the map, between markers

COR022 and MYL9. Though the published linkage maps

[14,15] have only 11 markers, the order of these loci in

the new ECA22 RH map is essentially the same except

for minor inversions.

A comparative map of ECA22 in relation to a range of

vertebrate species was developed as described earlier [1],

using the physical order of the 52 putative equine gene loci

obtained by RH mapping. Additionally, 3 loci (CRNKL1,

RBL1, and NTSR1) were bplacedQ in the physical order,

based on their location in the FISH map in relation to

adjacent markers. Megabase positions of orthologs for the

55 putative equine genes were obtained for different species

(see Fig. 2 legend for details). As described earlier [1]

maximally contiguous chromosomal regions with identical

gene content and order—referred to as conserved link-

ages—were identified by clustering the contiguous mega-

base locations of individual loci in human, mouse, rat, dog,

and chicken (see Fig. 2). Two major clusters of conserved

gene order were found between horse and human, mice, or

rat. The dog data, however, showed remarkable conserved

linkage with ECA22. Comparison with the chicken

sequence data showed conserved synteny between the distal

half of ECA22 and the entire chicken chromosome 20

(GGA20). The proximal half of the equine chromosome

corresponded primarily to parts of two chicken chromo-

somes, viz., GGA3 and 14. Segments depicting the

conserved order of loci across all mammalian species are

denoted by shaded horizontal boxes (Fig. 2).

Fig. 2. A high-resolution radiation hybrid and comparative map of the horse c

representation (idiogram) of G-banded ECA22. Next to the idiogram are all the FIS

by .). Most of the FISH loci are connected by diagonal lines to corresponding loc

and is depicted by a vertical color bar calibrated on the left at 50 cR intervals. To

microsatellites in light font. Framework markers (odds 1:1000) are shaded green. bshaded loci were bplacedQ in the map based on their proximity to adjacent markers

(HSA), mouse (MMU), rat (RNO), dog (CFA), and chicken (GGA) orthologs of t

information for all species available at this Web site). Orthologs showing conserve

boxes (conserved linkages) demonstrating gene order conservation in different spec

separating conserved linkages in human, mouse, and rat indicate putative breakag

order of the horse loci corresponds closely to that observed in dog (see text fo

preserved gene order in horse, human, mouse, rat, and dog, signifying core ancestr

(BTA), pig (SSC), and cat (FCA) for orthologs with available mapping informatio

loci in cattle, pig, and cat. Mapping data for HSA20/ECA22 orthologs in zebra

comparative genome organization of these two distantly related species in relati

indicating each locus.

Discussion

ECA22 RH and FISH map

The map of ECA22 reported herein represents a

fourfold increase in the number of mapped markers (83

loci: 52 gene specific and 31 microsatellites) compared to

the previously published RH map that had just 21 loci (7

gene specific and 14 microsatellites) [1]. The number of

FISH-mapped markers has increased from 3 to 41. The

markers are distributed uniformly along the length of the

chromosome. The size of this chromosome, as predicted

from the size of the human counterpart HSA20, is appro-

ximately 64 Mb (http://www.ensembl.org/Homo_sapiens/

mapview?chr = 20). Thus on average, the map presented

herein provides markers every ~770 kb or 10 cR. In terms

of gene-specific markers, the map has on average a marker

every 1.2 Mb. Overall, the resolution of the map is to date

the densest for the horse and among the highest in any of

the domesticated animal species for which annotated

sequence data are not yet available. The only region

where the markers seem to be poorly represented from

HSA20 is the 25–31 Mb region. The majority of this

region represents the human centromere. The remaining

small pericentromeric region of HSA20 comprises ~20

members of CST: cystatin precursors. The sequence

similarity in these genes hampered efforts to design

overgos representing this region. Additionally, the

ECA22 RH map showed relatively higher than average

centiray distances (or gaps) in the 350–450 cR region,

although megabase positions of human loci in this region

do not reflect a noteworthy gap.

The approach used in this study to carry out targeted

development of a high-resolution map has been largely

successful. Though the final success rate for mapping

gene-specific markers using overgo probes was ~60%,

addition of 16 putative genes through equine ESTs and

heterologous primer sets from conserved gene sequences

significantly complemented the overall number and

distribution of markers. No BAC clones were obtained

hromosome 22. To the extreme left of the RH map is the diagrammatic

H localizations (41 loci), of which 38 were conducted in this study (denoted

i in the RH map. A single RH group (RHMAP2PT; LOD 7.0) was obtained

the right of the bar are mapped equine loci: genes are shown in bold and

?Q shows disagreement between FISH and RH assignments. The three blue

. Next to the RH map are the sequence locations in megabases for all human

he mapped horse genes (http://www.genome.ucsc.edu/cgi-bin/hgGateway—

d gene order compared to the derived order of equine genes are grouped in

ies compared to that seen in the horse. Red arrow on ECA22 and the red line

e/fusion point in the evolution of this chromosome. The overall pter Y qte

r other details). Yellow-shaded horizontal regions represent loci that have

al segments in these species. A comparative status is also provided in cattle

n. Solid vertical bars with arrows show known pter Y qter arrangement o

fish and medaka have been added toward the extreme right to show the

on to mammals. Linkage groups are enumerated with color-coded circles

-

r

f

Fig. 3. Double-color FISH shows relative order of loci on metaphase (top) and interphase chromatin (bottom).

A. Gustafson-Seabury et al. / Genomics 85 (2005) 188–200196

for 12 of the overgo probes. This could be attributed to

the difficulty associated with accurate identification of

conserved sequences within targeted regions, especially

where sequence data for a specific gene was available in

only one species. For example, no comparative sequence

data were available for some of the genes located at the

terminal part of HSA20q. Consequently, overgo probe

design was based solely on the human sequence, resulting

in probes that may not necessarily have had sufficient

homology to give strong hybridization signal on the horse

BAC library.

The 38 new FISH localizations in this study are largely

consistent with the physical order obtained by RH analysis

and were vital in providing/confirming relative arrange-

ment of some of the markers with the same centiray

location. The most significant discrepancy corrected by

FISH mapping of UMNe344, UMNe276, and AHT030 was

the relocation of an RH group comprising six markers,

viz., TKY997, AHT030, UMNe344, UMNe276, TKY421,

and AHT095 from ECA22 [1] to ECA13q. This typically

demonstrates the significance of FISH assignments in

accurate placement/orientation of RH groups to specific

chromosomes or chromosomal regions, especially when

evidence based on linkage to adjacent markers is inad-

equate. The findings are in agreement with recent linkage

analysis results that indicate the location of AHT30 on

ECA13 [15] instead of ECA22. The FISH placement of

RBL1 close to PPGB contradicts the location of the marker

proximally, close to RPN2 and DAP4, based on the

comparative map. Availability of horse-specific primers for

RBL1, and subsequent RH mapping, will help to resolve

the issue. Finally, the minor disagreement between the

physical order of VIAAT deduced by FISH and RH

mapping in the q16 region could be due to typing

difficulties associated with the marker.

Comparative map

A greater than seven-fold increase in the number of

gene-specific markers on ECA22 attained in this study

provides 52 anchor points that facilitate comparison of the

organization of this chromosome in relation to its counter-

parts in human, mouse, rat, dog, and other mammalian/

vertebrate species. ECA22 is an acrocentric chromosome.

The human homologue, HSA20, however, is metacentric.

As far as is known, the homologues of this chromosome in

other species (Fig. 4) either are acrocentric, e.g., in dog

(CFA24) [18], Asian/African elephant (EMA23/LAF23)

[19], and some bats [20], or comprise a single complete

arm of a chromosome, e.g., cat (A3p) [21] and aardvark

(OAF2p) [19] (see Fig. 4). In a group of other species, the

chromosome corresponds to parts of one chromosome.

Last, in species with low chromosome numbers, e.g., In-

dian muntjac, common shrew, the chromosome corre-

sponds to part of a large chromosome. In essence, the

chromosome demonstrates a remarkable degree of con-

served synteny (predominantly a single block) in a wide

range of species.

Comparison of the gene map between horses and

humans shows that the proximal half of ECA22 corre-

sponds to the short arm of HSA20, while the distal half

corresponds to the long arm of HSA20. A putative

breakage/fusion point compared to the human homologue

is located on ECA22q15, between FKBP1A and AHCY

(Fig. 2; see red arrow). It is notable that the order of loci

from this point onward is essentially conserved between

the two species, though the order of genes from the short

arm of HSA20 is inverted on ECA22. Interestingly, a

similar situation is also observed in mouse (MMU2; distal

one-third) and rat (RNO3; distal one-third), in which

HSA20 short arm and long arm orthologs are distinctly

Fig. 4. Structural organization of horse chromosome 22 (ECA22) homologues in different species. The high degree of conserved synteny observed for the

homologues of this chromosome in a range of evolutionarily distantly related species reiterates the ancestral nature of this chromosome. The green star symbol

indicates the most common configuration for the ECA22 homologues.

A. Gustafson-Seabury et al. / Genomics 85 (2005) 188–200 197

grouped together and demonstrate a conserved gene order.

Like in humans, the order of the genes corresponding to

HSA20p in mouse and rat is also inverted in relation to

the horse. One noteworthy comparative observation,

however, is that FKBP1A is telomeric on 20p in humans,

while it is on the perimeter of HSA20q orthologs in

mouse and rat. This locus, together with AHCY and ASIP,

can be useful in determining the likely breakage/fusion

point of the two conserved linkages (Fig. 2).

One of the most notable observations of the comparative

data was the order of the 52 loci mapped on ECA22 and its

dog homolog CFA24. Except for minor discrepancies, the

gene order was essentially identical in the two species. The

discrepancies could be attributed either to statistical analysis

of the horse data or to the draft compilation of the dog

genome sequence (http://www.genome.ucsc.edu/cgi-bin/

hgGateway). Irrespective of this, small intrasegmental

rearrangements during karyotype evolution in either species

cannot be completely ruled out. Next, comparison of the

ECA22 map with the recently published map of the bovine

homolog BTA13 (based on 37 loci) [22] reveals more

rearrangements than seen between horse and human, mouse,

or rat. Segments corresponding to HSA20p seem to be

present in three blocks, while those corresponding to

HSA20q as two distinct blocks with inversed gene order

(suggesting inversion). In cats, in which relatively few

comparative loci have as yet been mapped on ECA22/

HSA20 homolog FCAA3p, rearrangements in relation to

horse and human are evident [23,24]. Last, in pigs, most

comparative loci (~10) [25,26] are mapped by somatic cell

hybrid analysis, due to which valid comparisons regarding

gene order conservation and/or rearrangement cannot be

made.

Of considerable interest is the comparison of the ECA22

map data with the recently available draft sequence of the

chicken genome (http://www.genome.wustl.edu/projects/

chicken/). GGA20 shows complete synteny conservation

with the distal half of the horse chromosome and hence also

HSA20q. Intrachromosomal rearrangements in relation to

the horse and human homologs are, however, abundant on

the chicken chromosome. Compared to this, the proximal

half of ECA22 (or HSA20p) does not show en bloc synteny

conservation and shares synteny primarily with parts of two

chicken chromosomes, GGA3 and GGA14. Additionally,

correspondence limited to one or two loci is found with

GGA2, 6, 7, and 22.

Putative mammalian/vertebrate ancestral configuration

The two outstanding features emerging from the com-

parison of ECA22 and its homologues across various

mammalian species analyzed to date are:

(i) the chromosome is present as a single syntenic block in

most species (either as a whole chromosome or arm or

as a single part of another chromosome) and

(ii) rearrangements detected now predominantly involve

segments corresponding to the entire HSA20p or

HSA20q (i.e., nomajor intrasegmental rearrangements).

Assuming that ancestral mammalian chromosomes were

primarily acrocentric [27], it is tempting to hypothesize

A. Gustafson-Seabury et al. / Genomics 85 (2005) 188–200198

that the near-identical horse and dog chromosomes

(ECA22 and CFA24) are ancestral among placental

mammals. However, a contrary possibility cannot be ruled

out when the human, mouse, and rat homologues of

ECA22 (HSA20 and the terminal one-third part of

MMU3p and RNO2p) are analyzed. It is noteworthy that

the three homologues have an almost identical gene order.

If the primate/rodent gene order is more primitive, the

human homologue most likely resulted from a simple

reinsertion of the centromere from the end to the middle of

the acrocentric ancestral chromosome. In this scenario, the

horse and dog counterparts should have undergone a

paracentric inversion, with breakages adjacent to the

centromere and the middle of the chromosome (Fig. 2,

red arrow). Emerging sequence data in a range of

evolutionarily diverged mammals (including marsupials

and monotremes) will help to validate either ancestral

configuration. Finally, primates and rodents are proposed

to be phylogenetically more closely related to each other

than with either horse or dog, which are anticipated to

have radiated separately from a common ancestor [28,29].

Considering this, the exceptional conserved linkage seen in

humans, mice, and rats may not be surprising. However,

maintenance of this linkage despite extensive rearrange-

ments in the rodent karyotype in comparison to humans is

indeed remarkable.

While extending the ECA22/HSA20 comparison to

other vertebrates, it was observed that the distal half of

ECA22 (HSA20q) shares striking similarity with GGA20.

This extraordinary synteny conservation in species as

diverged as mammals and birds (~310 Myr ago) [30]

leads us to postulate that the segment/arm is ancestral.

Extending the comparisons to the limited mapping data

available for zebrafish and medaka (shared a common

ancestor with mammals ~450 Myr ago) [30] shows that the

distal half of ECA22 (HSA20q) shares synteny mainly

with two to three linkage groups in zebrafish (LG6, 7, and

23) [31,32] as well as medaka (LG5, 7, and 11) [33].

Though a clearer picture of ECA22/HSA20 protovertebrate

chromosomes will emerge only after a finished whole

genome sequence is available for fishes (zebrafish, puffer-

fish, and medaka), the present comparisons add to the

current knowledge regarding the likely ancestral homo-

logue of ECA22/HSA20.

Conclusions

The high-resolution physically ordered map of ECA22

represents another important step toward targeted expansion

of the horse gene map and its alignment with maps of a

range of mammalian/vertebrate species, some of which have

sequence data available for detailed comparison. The

uniform distribution of markers and the accessibility of

comparative data in a range of species will considerably

facilitate the search for bcandidateQ genes on this equine

chromosome. The overall conservation of the chromosome

in a range of evolutionarily distantly related mammalian

species reinforces earlier suggestions that ECA22/HSA20 is

an ancestral configuration. The striking similarity in gene

order seen between ECA22 and CFA24 is unprecedented for

autosomes in species that diverged ~70–80 Myr ago from a

common ancestor.

Methods

Design of overlapping oligonucleotide (overgo) probes

Genes at ~1Mb intervals were identified for overgo primer

design using the human genomic draft sequence for

chromosome 20 (http://www.genome.ucsc.edu, version July

2003). Conserved regions within individual genes were

identified following multiple species alignment of the human

sequence using GenBank pair-wise BLAST [34]. These

regions were utilized to design overgo primer sets with the

Overgo Maker program (http://www.genome.wustl.edu/

tools?overgo.html). Finally, individual overgos were

screened against nr and EST sequences using BLAST

(http://www.ncbi.nlm.nih.gov/BLAST/) [16] to confirm their

specificity and to exclude homology with repetitive elements.

Overgo labeling, BAC library screening, isolation of BAC

clones, and end sequencing

The overgo primer sets were individually labeled radio-

actively with [32P]dATP and [32P]dCTP, pooled, and

hybridized onto high-density filters of an equine genomic

BAC library (CHORI-241; http://www.bacpac.chori.org/

equine241.htm) using a previously described protocol [35].

The filters were washed three times at 428C for 15 min in 2�SSPE (0.2 M NaH2PO4–H2O, 0.02 M EDTA, 3.0 M NaCl)

and exposed to film over intensifying screens for 4 days at

�808C prior to development of the autoradiograms. Analysis

of the autoradiograms led to identification of positive BAC

clones. These clones were isolated and used to generate

secondary filters. The secondary filters were hybridized with

individual radioactively labeled overgo probes and processed

following the protocol described above. Following develop-

ment and analysis of the autoradiograms, BAC clones

specific for individual overgo probes were identified. In

instances in which multiple BAC clones were positive for a

single overgo probe, a representative clone was chosen for

subsequent analysis. Individual BAC clones were cultured,

and DNA was extracted using the Qiagen Plasmid Midi kit

(Valencia, CA, USA). One of the overgo primers was used as

sequencing primer to obtain partial information on the

targeted gene. The identity of the sequence (hence the BAC

clone) was verified by using GenBank BLAST [16]. One

representative BAC clone for each overgo probe was end

sequenced using pTARBAC2.1-derived sequencing primers

T7.29 (5V-GCCGCTAATACGACTCACTATAGGGAGAG)

A. Gustafson-Seabury et al. / Genomics 85 (2005) 188–200 199

and SP6.26 (5V-CCGTCGACATTTAGGTGACACTATAG)[35].

FISH

The representative BAC clones in the final collection

were individually labeled with biotin and/or digoxigenin

using the BioNick Labeling System (Invitrogen) and DIG-

Nick Translation Mix (Roche). The labeled clones were

hybridized to horse metaphase chromosome spreads to

determine their chromosomal origin and physical location

along the length of the chromosome. Differentially labeled

clones were cohybridized to determine their relative order

along ECA22. DNA labeling, in situ hybridization, signal

detection, microscopy, and image analysis were carried out

as described previously [1,36].

Primer design and RH analysis

BAC end sequences were analyzed for repetitive elements

using RepeatMasker (http://www.repeatmasker.genome.

washington.edu/cgi-bin/RepeatMasker) and then exam-

ined for homology with available sequences using BLASTn

(http://www.ncbi.nlm.nih.gov/BLAST/) [16]. PCR primers

were designed from satisfactory end sequences using

Primer3 software (http://www.frodo.wi.mit.edu/cgi-bin/

primer3/primer3_www.cgi). Additionally, eight primer

pairs were generated from EST sequences obtained from

the horse testis cDNA library [37] that provided significant

hits to HSA20 genes. Finally, heterologous primer pairs for

eight additional HSA20 genes were developed following a

previously described approach of multiple alignment of the

human sequences with sequences available from other

species and identification of conserved regions for horse-

specific DNA amplification [3,38]. PCR amplification

conditions for individual primer pairs were optimized

using horse and hamster genomic DNA to ensure horse-

specific amplification. The successful primer pairs were

then used to type in duplicate the 5000-rad horse �hamster RH panel by PCR. Analysis of the typing data and

construction of a physically ordered RH map were carried

out as described previously [1].

Acknowledgments

This project was funded by grants from the Texas Higher

Education Board (ARP 010366-0162-2001, B.P.C.; ATP

000517-0306-2003, B.P.C., J.E.W.), NRICGP/USDA Grant

2003-03687 (B.P.C.), the Texas Equine Research Foundation

(B.P.C., L.C.S.), the Link Endowment (B.P.C., L.C.S.), the

American Quarter Horse Association, and the Dorothy

Russell Havemeyer Foundation. Additional support was

available from the USDA-NRSP-8 Coordinators Fund. We

thank Dr. Pat Venta, Michigan State University, for design

and supply of several primer pairs.

References

[1] B.P. Chowdhary, et al., The first-generation whole-genome radiation

hybrid map in the horse identifies conserved segments in human and

mouse genomes, Genome Res. 13 (2003) 742–751.

[2] B.P. Chowdhary, E. Bailey, Equine genomics: galloping to new

frontiers, Cytogenet. Genome Res. 102 (2003) 184–188.

[3] E.J. Lee, et al., A 1.4-Mb interval RH map of horse chromosome 17

provides detailed comparison with human and mouse homologues,

Genomics 83 (2004) 203–215.

[4] T. Raudsepp, et al., Exceptional conservation of horse–human gene

order on X chromosome revealed by high-resolution radiation

hybrid mapping, Proc. Natl. Acad. Sci. USA 101 (2004)

2386–2391.

[5] M.T. Ross, S. La Brie, J. McPherson, V.P. Stanton Jr., Screening large

insert libraries by hybridization, in: A. Boyl (Ed.), Current Protocols

in Human Genetics, Wiley, New York, 1999, pp. 5.6.1–5.6.52.

[6] C.S. Han, et al., Construction of a BAC contig map of chromosome

16q by two-dimensional overgo hybridization, Genome Res. 10

(2000) 714–721.

[7] B.P. Chowdhary, T. Raudsepp, L. Fronicke, H. Scherthan, Emerg-

ing patterns of comparative genome organization in some mam-

malian species as revealed by Zoo-FISH, Genome Res. 8 (1998)

577–589.

[8] T. Raudsepp, L. Fronicke, H. Scherthan, I. Gustavsson, B.P.

Chowdhary, Zoo-FISH delineates conserved chromosomal segments

in horse and man, Chromosome Res. 4 (1996) 218–225.

[9] R.S. Sparkes, et al., Assignment of the human and mouse prion

protein genes to homologous chromosomes, Proc. Natl. Acad. Sci.

USA 83 (1986) 7358–7362.

[10] J. Collinge, Prion diseases of humans and animals: their causes and

molecular basis, Annu. Rev. Neurosci. 24 (2001) 519–550.

[11] F. Jobard, et al., Identification of mutations in a new gene encoding a

FERM family protein with a pleckstrin homology domain in Kindler

syndrome, Hum. Mol. Genet. 12 (2003) 925–935.

[12] M. Callaghan, et al., Homozygosity mapping and linkage analysis

demonstrate that autosomal recessive congenital hereditary endothelial

dystrophy (CHED) and autosomal dominant CHED are genetically

distinct, Br. J. Ophthalmol. 83 (1999) 115–119.

[13] C.K. Hand, et al., Localization of the gene for autosomal recessive

congenital hereditary endothelial dystrophy (CHED2) to chromosome

20 by homozygosity mapping, Genomics 61 (1999) 1–4.

[14] J. Swinburne, et al., First comprehensive low-density horse linkage

map based on two three-generation, full-sibling, cross-bred horse

reference families, Genomics 66 (2000) 123–134.

[15] G. Guerin, et al., The second generation of the International Equine

Gene Mapping Workshop half-sibling linkage map, Anim. Genet. 34

(2003) 161–168.

[16] S.F. Altschul, W. Gish, W. Miller, E.W. Myers, D.J. Lipman, Basic

local alignment search tool, J. Mol. Biol. 215 (1990) 403–410.

[17] J. Swinburne, et al., First comprehensive low-density horse linkage

map based on two three-generation, full-sibling, cross-bred horse

reference families, Genomics 66 (2000) 123–134.

[18] M. Switonski, I. Szczerbal, J. Nowacka, The dog genome map and its

use in mammalian comparative genomics, J. Appl. Genet. 45 (2004)

195–214.

[19] F. Yang, et al., Reciprocal chromosome painting among human,

aardvark, and elephant (superorder Afrotheria) reveals the likely

eutherian ancestral karyotype, Proc. Natl. Acad. Sci. USA 100 (2003)

1062–1066.

[20] M. Volleth, K.G. Heller, R.A. Pfeiffer, H. Hameister, A comparative

ZOO-FISH analysis in bats elucidates the phylogenetic relationships

between Megachiroptera and five microchiropteran families,

Chromosome Res. 10 (2002) 477–497.

[21] W.J. Murphy, et al., A radiation hybrid map of the cat genome:

implications for comparative mapping, Genome Res. 10 (2000)

691–702.

A. Gustafson-Seabury et al. / Genomics 85 (2005) 188–200200

[22] A. Everts-van der Wind, et al., A 1463 gene cattle–human

comparative map with anchor points defined by human genome

sequence coordinates, Genome Res. 14 (2004) 1424–1437.

[23] M. Menotti-Raymond, et al., Second-generation integrated genetic

linkage/radiation hybrid maps of the domestic cat (Felis catus),

J. Hered. 94 (2003) 95–106.

[24] S. Sun, W.J. Murphy, M. Menotti-Raymond, S.J. O’Brien, Integration

of the feline radiation hybrid and linkage maps, Mamm. Genome 12

(2001) 436–441.

[25] H. Hayes, et al., Mapping of 195 genes in cattle and updated

comparative map with man, mouse, rat and pig, Cytogenet. Genome

Res. 102 (2003) 16–24.

[26] S. Cirera, et al., Comparative mapping in the pig: localization of 214

expressed sequence tags, Mamm. Genome 14 (2003) 405–426.

[27] F. Richard, M. Lombard, B. Dutrillaux, Reconstruction of the

ancestral karyotype of eutherian mammals, Chromosome Res. 11

(2003) 605–618.

[28] W.J. Murphy, Resolution of the early placental mammal radiation

using Bayesian phylogenetics, Science 294 (2001) 2348–2351.

[29] M.S. Springer, W.J. Murphy, E. Eizirik, S.J. O’Brien, Placental

mammal diversification and the Cretaceous–Tertiary boundary, Proc.

Natl. Acad. Sci. USA 100 (2003) 1056–1061.

[30] S. Kumar, S.B. Hedges, A molecular timescale for vertebrate

evolution, Nature 392 (1998) 917–920.

[31] J.H. Postlethwait, et al., Zebrafish comparative genomics and the

origins of vertebrate chromosomes, Genome Res. 10 (2000)

1890–1902.

[32] I.G. Woods, et al., A comparative map of the zebrafish genome,

Genome Res. 10 (2000) 1903–1914.

[33] K. Naruse, et al., A medaka gene map: the trace of ancestral vertebrate

proto-chromosomes revealed by comparative gene mapping, Genome

Res. 14 (2004) 820–828.

[34] T.A. Tatusova, T.L. Madden, BLAST 2 sequences, a new tool for

comparing protein and nucleotide sequences, FEMS Microbiol. Lett.

174 (1999) 247–250.

[35] A.L. Gustafson, et al., An ordered BAC contig map of the equine

major histocompatibility complex, Cytogenet. Genome Res. 102

(2003) 189–195.

[36] T. Raudsepp, et al., Comparison of horse chromosome 3 with donkey

and human chromosomes by cross-species painting and heterologous

FISH mapping, Mamm. Genome 10 (1999) 277–282.

[37] B.P. Chowdhary, et al., Construction of a 5000(rad) whole-genome

radiation hybrid panel in the horse and generation of a compre-

hensive and comparative map for ECA11, Mamm. Genome 13

(2002) 89–94.

[38] D.M. Shubitowski, P.J. Venta, C.L. Douglass, R.X. Zhou, S.L. Ewart,

Polymorphism identification within 50 equine gene-specific sequence

tagged sites, Anim. Genet. 32 (2001) 78–88.