The identification of polymorphicmicrosatellite loci in the horse and their use in thoroughbred parentage testing

Br vet.J. (1995). 151, 9

THE IDENTIFICATION OF POLYMORPHIC MICROSATELLITE LOCI IN THE HORSE AND THEIR USE IN THOROUGHBRED PARENTAGE TESTING

M. M. BINNS*, N. G. HOLMES, A. HOLLIMAN and A. M. SCOTT Animitl Health 7)~tst, Lanwades Park, Kennett, Newmarket, Suffolk CB8 7PN, UK

SUMMARY Six new horse microsatellite loci were identified by sequencing M13 clones containing horse genomic inserts which gave positive signals when probed with a C A / G T repeat probe. Oligonucleotide primer pairs were synthesized for these loci and for two previously described horse microsatellites, HTG4 and HTG6. Polymerase chain reaction assays were then carried out on a panel of 20 different unrelated T h o r o u g h b r e d horse DNAs. DNAs from eight cases of double covering which could not be solved by conventional blood typing were also examined. Several of the loci amplified were f o u n d to be polymorphic and, using a limited subset of primers, a clear exclusion could be established for one of the stallions in five of the cases. DNA typing is therefore a useful adjunct to blood typing in the horse and indeed, in the future will probably replace it. Krvwol~ns: T h o r o u g h b r e d horse; microsatellites; equine; genetics; parentage analysis.

INTRODUCTION T h o r o u g h b r e d horses in the UK are subject to registration by Weatherbys, the General Stud Book, whose traditional m e t h o d s of identifying horses have been reinforced by compulsory parentage testing through blood typing since 1986. Current blood typing uses seven red cell antigen loci and 10 serum and intra-elythrocytic enzyme and other protein loci as ~ poly'morphic markers. It is calculated to be efficient at the. 97% exclusion level for paternity testing. Cases of double covering, where a mare has been mated with two stallions, occasionally present problems for parentage assignment by conventional methods, particularly when the stallions are themselves closely related. *To whom correspondenceshould be addressed. 0007-1935/95/010009-07/$08.00/0

© 1995 Bailli~reTindall

10

BRITISH VETERINARY JOURNAL, 151, 1

Until recently, blood typing was also used in human affiliation cases. With the advent of DNA fingerprinting, based initially on minisatellites, paternity and forensic investigations have increasingly adopted the new technologies, although not always without controversy (Lander, 1989). The use of DNA fingerprinting, in theory, offers the ability to identify individuals at a much higher confidence level using potentially simpler and cheaper techniques. Unfortunately, the use of minisatellites has some technical limitations relying as it does on very precise measurements of DNA fragment mobilities on Southern blots. Microsatellites are simple repeated sequences, most frequently CA/GT repeats which are present at many thousand loci in mammalian genomes (Hamada et al., 1982). At any one locus, the number of copies of the repeat may vary (Love et al., 1990), originally probably due to polymerase slippage during replication. The length of each microsatellite is however inherited in a Mendelian fashion, making them ideal polymorphic markers for genome mapping, parentage and identity testing. The use of microsatellites in identity testing in large part avoids the problems inherent in minisateilite use, as it involves the analysis of polymerase chain reaction (PCR) amplified fragments of defined lengths which vary, in two base pair increments in the case of dinucleotide repeats, in different alleles. The results, generated using a series of primer pairs to amplify different microsatellite loci, are easily digitized into a format which is ideal for the construction and maintenance of population databases. Recently, the use of both mini- and microsatellites in equids has been reported. Minisatellite analysis using the human probes 33.15 and 33.6 (Jeffreys et al., 1985) was used to assign the parentage of an Exmoor pony foal in a case where the two possible sires were three-quarter siblings (Hopkins et al., 1991). In contrast, Ellegren et al. (1992) cloned microsatellites from Swedish trotting horses and demonstrated that the microsatellites were highly polymorphic and likely to be useful for parentage analysis in the horse. In this paper we report the cloning of additional microsatellite loci and their use, together with that of previously published loci, to carry out parentage analysis in Thoroughbreds in cases where conventional blood typing was unable to resolve the issue.

MATERIALS A N D M E T H O D S

Identification of horse microsatellites Horse DNA was prepared by isolating white blood cells, subjecting tlaem to sodium dodecyl sulphate and proteinase K followed by phenol and chloroform extractions (Kunkel et al., 1982). The DNA was ethanol precipitated and collected by spooling. After brief air drying, the DNA was resuspended in TE (10 mM Tris, 1 mM EDTA pH 7.4) and its concentration estimated by spectroscopy at 260 nm. One microgranl of DNA was digested with either RsaI or HaeIII and ligated into M13mpl0 digested with Sinai and treated with alkaline phosphatase (Amersham). Following transformation (Hanahan, 1983) into Esche~chia coli TG1, plaque lifts were carried out using Hybond-N (Amersham) and the filters were probed with poly(dC-Ad),~.(dG-dT), (Pharmacia) which had been labelled with :v_,p by nick

EQUINE MICROSATELLITES AND PARENTAGE ANALYSIS

11

translation. Single strand DNA was prepared from positive plaques, re-transformed, re-probed and DNA prepared from well isolated positive plaques. This DNA was sequenced using either a Sequenase kit (Cambridge Bioscience) or a cycle sequencing kit (BRL) following the manufacturers' instructions. Ladders of CA or GT repeats were clearly visible on the majority of positive M13 clones.

Oligonucleotide synthesis Oligonucleotides were synthesized on an Applied Biosystems 391 DNA synthesizer. The oligonucleotides were cleaved from the column with concentrated ammonia. DNA was precipitated from 270 tll of the ammonia solution by the addition of 1/10 volume 3 M sodium acetate (pH 5.2) and 2 volumes of ethanol followed by incubation at -70°C for 1 h. Following centrifugation at 12300 g for 10 min at 4°C, the precipitates were washed once in 70% ethanol and recentrifuged for 5 min. Primers were resuspended in 30 lal milliQ water after they had been vacuum dried and the concentrations estimated from their optical density (OD._,~0). They were diluted to 50 lag ml -l for use in the PCR assays.

Amplification of microsatellites by PCR PCR was optimized by mixing 5 lal 10 x PCR buffer (containing either 1, 1.5, 2, 3 or 4 mM MgC1._,), 8 lal 1.5 mM dNTPs, 4 lal each primer (50 lag ml-I), 28 gl MilliQ water, 1 lal DNA (approximately 0.8 lag). This mixture was overlaid with 50 lal mineral oil. DNA w.as denatured at 95°C for 5 min during which time 0.5/11 (2.5 units) Taq polymerase (Perkin Elmer and Advanced Biotechnologies) was added. Amplification was then carried out either by performing 30 cycles of 94°C, 1 min, annealing temperature (see Table I) 1 min, 72°C, 1 min, or using the 'touchdown' PCR method (Don et al., 1991) using the ranges of annealing temperatures given in Table I. Twenty microlitre samples were run on 3% agarose gels to identify the amplification conditions producing optimum amounts of the expected product. For estimation of the degree of polymorphism exhibited by each of the microsatellites, PCR was carried out as above on 26-30 Thoroughbred DNAs, except that one primer was kinase labelled with :~'-'P and the products were resolved on 5% polyacrylamide/urea DNA sequencing gels, using an M13 sequence ladder as size markers. The gels were dried on Whatmann 3MM paper and autoradiographed overnight using Fuji RX film. Polymorphism information content (PIC) was estimated b y t h e method of Botstein et aL (1980).

RESULTS

Identification and characterization of horse microsateUites When the DNA sequences of the M13 clones hybridizing to :~'-'P-labelled GT/CA probe were examined, microsatellite ladders were detected in the majority of clones. The microsatellites identified included two perfect GT repeats (primers HMB1A and B amplified perfect [GT]I~, whilst primers HMB5A and B amplified [GT]I4). The other four primer pairs amplified imperfect repeats (primers HMB2A and B amplified [GT]I:~ GCATGACT[GT].I, primers HMB4A and B amplified [AC]~sAT[AC],~, primers HMB3A and B amplified

12

BRITISH

VETERINARY

JOURNAl,,

151,

1

e.. o

L~

I

I

~o

°I

I

I e-

%

e"

t~

©

E-E

,z.._~

;.= L~

.u .E E ...a ,.= L,

.=

~a

"0

.,a

e"

E~

~. oo

o~ o,=.=

e--,

?,--. e-'

O

I..

°,9

,..,

,-

© .< r <

© ["'

•

©~

J=

.<

b.

©©

,,..,

22

,-e"

EQUINE MICROSATELLITES AND PARENTAGE ANALYSIS

13

[GT]:AT[GT]._,ATGTAT[GT] i.~, and primers HMB6A and B [GT]:CkT[GT] ~.t). A correlation between the length of the microsatellite repeat and the degree of polymorphism exhibited by it has previously been noted, although imperfect repeats showed lower PIC values than expected (Weber, 1990). For the microsatellites isolated here the imperfect repeats have the highest PIC value (see Table I). The sequences of six new primer pairs spanning the equine GT/AC microsatellites described above are presented in Table I together with the optimized PCR conditions for each pair, determined as outlined above. When tested on between 26-30 Thoroughbred horse DNAs, the primer pairs exhibited varying degrees of polymorphisms, and for each primer pair, the number of alleles observed, the degree of heterozygosity and PIC are presented in Table I. Use of microsateUites in double covering cases DNA was prepared from the mare, foal, and the two possible stallions for eight cases of double covering which could not be resolved using conventional blood typing. PCR was carried out with several primer pairs (including two of the most polymorphic microsatellites identified by Ellegren et al., 1992) for each of the double covering cases. The results for four cases demonstrating the clear exclusion of one stallion in each case are presented in Fig. 1. Horse 56 is also excluded as the sire using primers HMB2A and B (data not shown). In a fifth case a clear exclusion was also evident using primers HMB1A and B (data not shown). In contrast, for three further cases of double covering it was not possible to exclude one stallion with the current primer sets.

DISCUSSION

We report here the isolation and characterization of six new microsatellite loci in the horse. The PIC values calculated indicate that all six primer pairs should be informative. The microsatellites together with some of those previously described (Ellegren et al., 1992) have been used to examine the parentage of eight cases of double coverings involving Thoroughbred horses, where conventional blood typing did not provide an answer. In five of the eight cases using a limited set of the microsatellites it was possible to exclude one of the stallions as a possible sire (see Fig. 1). All the microsatellites examined are inherited in a Mendelian fashion. It may well be that for registration purposes exclusions should depend Oll differences in two microsatellites in order to increase the level of confidence in the exclusion. The use of microsatellites is therefore currently a useful adjunct to conventional blood typing in the parentage testing of Thoroughbreds. Using additional microsatellites it should become possible to resolve all cases of double covering. The microsatellites described here and previously (Ellegren et al., 1992) are polymorphic enough ill the Thoroughbred population, a highly inbred population, to be useful in parentage analysis. This suggests that microsatellites should also be usefill genetic markers in other highly inbred populations. With the isolation of additional microsatellite markers the efficacy obtainable with microsatellites should be better than that currently achieved using blood typing. There is

14

BRITISH VETERINARY JOURNAL, 151, 1

(a)

(b)

(d~

(c) .!:

e:x: .: :d'i ::":i:~-' .:,~:~:" :: X:~i:..

O'2

.._:,.. ~:.~, ..~:.~:-: ..:~ .~: ';~. ,-'-, ,: • • . . . . .~.f~ :i. ,:. ,~:~,-..~ ." !'4:~::.::: : - / ~ i':'~ : " : , : . ~~ : . " : , ~ .~'~:

., .

. . . -

~::L:~i.ii ~:I.,~I":~.':-,~.::,~!"::~: '. .::, :

...

.i.!-'t::-~ ~:-.: .:k;--'

~";,~

~'.: ~:. I.~ ~- . ~,

;-- v ~:: ~........., ~--

l:~yz/~::~,l>~.'~,:..,j,~,..~..;:

..

. ....

'

.!:i..

" '.

;

.~::~:,::;.. • .

'

' , ~,~

~:,.,:.:,~::.-~~. : .!( :, , ,-.., ~.~.~,~...,~-,~:.i~,,~.e,:..,~.:.:~z : ~.,'.~ ~.~ ..-

, :--. ,.~ .,:-~

Fig. 1. (a) Primers HMB2A and HMB2B on horse DNAs 25 ( 2 ) , 26 (ex), 27 (c~l) and 29 (c~ 2). Twenty-seven is clearly excluded as the sire of 26 with these primers. (b) Primers HTG4A and HTG4B on horse DNAs 31 ( ~ ) , 32 ( 6 1 ) , 33 (ex) and 34 ((32). Thirty-four is excluded as the sire of 33. (c) Primers HTG4A and HTG4B o n horse DNAs 53 ( ~ ) , 54 (ex), 55 (c~l) and 56 ( ~ 2). Fifty-six is excluded as the sire of 54. (d) Primers HTG6A and HTG6B on horse DNAs 37 ( ~ 1), 44 (~ 2), 43 (ex) a n d 42 ( 2 ) . Thirty-seven is excluded as the sire of 43.

EQUINE MICROSATELLITESAND PARENTAGE ANALYSIS

15

t h e r e f o r e a case for the r e p l a c e m e n t o f conventional blood typing m e t h o d s by m o l e c u l a r techniques. If microsatellites are to be used it will be advisable to use tri- and tetra nucleotide repeats. T h e s e have several advantages over the dinucleotide repeats ifi that the 'stutter' bands typical o f dinucleotide repeats and visible in Fig. 1 are greatly r e d u c e d . Also, as the alleles are three or f o u r bases apart they can be resolved m o r e easily, often on agarose gels thereby removing the n e e d to use labelling techniques and polyacrylamide gels. T h e high level o f p o l y m o r p h i s m at microsatellite loci makes t h e m valuable markers for the establishment o f a linkage map. Such studies could lead to the identification o f markers for c o m p l e x genetic traits in the horse such as performance, soundness and fertility.

ACKNOWLEDGEMENTS Part o f this work was f u n d e d by a g e n e r o u s grant from Weatherbys. We would like to thank Karen Smith for excellent technical assistance, Jan Butler for photography and Mr Peter Willett for helpful suggestions.

REFERENCES BorsrEIN, D., WHrrz, R. I., SKOLNICr;,M. & DA'~qS,R. W. (1980). Construction of a genetic linkage map in man using resu-iction fi'agment length polymorphisms. AmericanJou~vzal of Human Genetics 32, 314-31. DoN, R. H., Cox, P. T., W,xINWm(.rrr, B.J., BA~, K. & MArricK,J. S. (1991). "Touchdown" PCR to circunavent spurious priming during gene amplification. Nucleic Acids Research 19, 4008. EH.VC;~I':N, H., JOHA~SSOX,M., SAND~ERC;,K. & ANDERSON,L. (1992). Cloning of highly polymorphic microsatellites in the horse. Animal Genetics 23, 133-42. HAMADA,H., P~'TRINO,M. G. & I~U~UN,aGA,T. (1982). A novel repeated element with Z-DNAforming potential is widely found in evolutionary diverse eukaryotic genomes. Proceedings of the National Acadenty of Science USA 79, 6465-9. H:~NAHAN,D. (1983). Studies on transformation of Esche~qchia coli with plasmids. Journal of Molecular Biology 166, 557-80. HOPKINS,B., O'CoNNELL,F. M. 8c HOVKINS,J. (1991). Use of DNA fingerprinting in paternity analysis of.closely-related Exmoor ponies. Equine VeterinmyJourna123, 277-9.. J~:vrRl~S, A. J., WILSON,V. & THElY, S. L. (1985). Hypervariable "minisatellite" regions in hunaan DNA. Nature 314, 67-73. KI_'NKEL,L. M., TANTR.-WAHI,U., EISENI-IARD,M. 8c La'Vl', S. A. (1982). Regional localisation on the httnaan X of DNA segments cloned from flow sorted chromosomes. Nucleic Acids Research 10, 1557-78. LaNDEg, E. S. (1989). DNA fingerprinting on trial. Nature 339, 501-5. Lovl~,J. M., K~IGIIT,A. M., MCALEER,M. A. & TODD,J. A. (1990). Towards construction of a high resolution map of the mouse genome using PCR-analysed microsatellites. Nucleic Acids Research 18, 4123-30. Wt~.m.:R,J. L. (1990). Informativeness of human (dC-dA),,. (dG-dT),, polymorphisnas. Genomits 7, 524-30. (Acceptedfor publication 22 Aplil 1994)