Two SINE families associated with equine microsatellite loci
Patrick C. Gallagher, Teri L. Lear, Linda D. Coogle, Ernest Bailey
M.H. Gluck Equine Research Center, Department of Veterinary Science, University of Kentucky, Lexington, Kentucky 40546-0099, USA
Received: 20 July 1998 / Accepted: 29 September 1998
Abstract. BLAST searches of 61 equine microsatellite sequences
revealed two related families of retroposons. The first family in-
cluded seven markers, all of which showed significant homology
to the Equine Repetitive Element-1 (ERE-1) Short Interspersed
Nucleotide Element (SINE) sequence. Length of homology ranged
from 76 to 171 bases with identities to the ERE-1 consensus se-
quence ranging from 71% to 83%. The second family referred to
as Equine Repetitive Element-2 (ERE-2) has a consensus sequence
that showed homology to ERE-1 over approximately 60 bases.
These 60 bases comprised subunit I. Sequence comparisons for the
two retroposons led to the identification of a subunit II, subunit III,
as well as the tRNA
subunit. The subunit structure of ERE-1 was
tRNAser-I-II. By contrast, the subunit structure of ERE-2 was
I-III-III. The nine markers related to ERE-2 showed homology
lengths ranging from 84 to 163 bases with identities ranging from
75% to 99%.
In addition to being present in microsatellites, ERE-2 appeared
in three separate equine genes. It occurred in an intron of DNA-
PK, in an untranslated region as well as in the promoter of PGHS,
and in the coding region of PAM. The amino acids corresponding
to the ERE-2 sequence in PAM were not present in the human or
mouse PAM homologs. These amino acids associated with the
ERE-2 sequence were present on the cytosolic side of the trans-
membrane domain of the PAM enzyme.
Microsatellite markers in the ERE-1 and ERE-2 families were
found throughout the genus equus and also for rhinoceros, indi-
cating that the appearance of both retroposons predates the diver-
gence of equids from the other perissodactyls. The markers did not
amplify in human or bovine DNA. This indicated that ERE-1 and
ERE-2 are, at least, perissodactyl specific.
Short Interspersed Nucleotide Elements (SINEs) are present in all
mammalian genomes (Smit and Riggs 1995). These sequences
originated from RNA molecules that developed the ability to un-
dergo retrotransposition. The human Alu family is an example of
such a molecule (Batzer et al. 1996). It was derived from 7sl RNA.
Most other reported SINE families were derived from tRNA mol-
ecules or snRNAs (Rogers 1985).
An equine SINE family, Equine Repetitive Element-1 (ERE-
1), was reported that is derived in part from tRNA
al. 1994). This equine SINE had an estimated copy number of
between 20,000 and 80,000. It contained the elements character-
istic of a retroposon including an RNA polymerase III promoter
box A and box B, terminal repeats, homology to an RNA mol-
ecule, and a poly-A sequence that corresponded to a poly-A tail.
The internal RNApol III promoter is a common feature of retro-
posons as it allows them to carry their promoter with them as they
move around the genome (Rogers 1985).
An artiodactyl family of SINEs was reported that was found in
association with poly-(AC) microsatellite markers. This family is
known as the C-A family and also contains a tRNA derivative. The
microsatellites are often tails of retroposons (Kaukinen and Varvio
1992). It was shown that the subunits that constitute artiodactyl
SINEs can be organized in several orientations to yield different
families of SINES (Lenstra et al. 1993).
Reported here is a similar association of SINEs with equine
microsatellites. Specifically, a sequence was present at many dif-
ferent loci throughout the genome and often occurred next to a
dinucleotide repeat microsatellite. Also, the subunits that com-
prised SINE families could be arranged in different orientations.
Materials and methods
BLAST searches (Altschul et al. 1990) were
done on 61 microsatellite markers developed in our lab (LEX002-LEX011,
LEX013-LEX063) (Coogle and Bailey 1997; Coogle et al. 1996a, 1996b,
1996c, 1997) as well as 79 microsatellite possessing clones that were not
developed into markers (Coogle and Bailey, unpublished). Sequences of
the published markers have been deposited in Genbank. Markers that re-
vealed similar results from the non-redundant database were aligned with
programs in the GCG suite (Wisconsin Package, Version 9.0). Multiple
alignments were done using pileup. The ERE-2 consensus sequence was
generated by pretty. Percentage identity and length of homology to the
consensus were determined with the pairwise alignment program bestfit.
Genes that were found as a result of the BLAST
searches [equine prostaglandin G/H synthase (PGHS) mRNA (AF027334),
PGHS partial cds and 5Ј flanking region (AF027335), DNA-protein kinase
(DNAPK) partial cds (U97529), peptidylglycine monooxygenase and pep-
tidylamidoglycolate lyase (PAM) (D29625)] were aligned with the Equine
Repetitive Element-2 (ERE-2) consensus sequence using bestfit and gap.
The peptide sequence corresponding to PAM was used to query the non-
redundant database through BLAST to find homologs in other species.
Multiple alignment of the equine, human, and mouse homologs was
done with the Multiple Alignment Construction and Analysis Workbench
DNA from several species was tested for the presence of
the ERE-1 and ERE-2 sequences. DNA tested came from three horses
(Equus caballus), one donkey (Equus asinus), one zebra (Equus burchelli),
one onager (Equus hemionus), one rhinoceros (Diceros bicornis), one cow
(Bos taurus), and one human (Homo sapiens). Primer pairs for microsat-
ellite sequences possessing ERE-1 were redesigned to include the ERE-1
SINE sequence with data for markers ASB28 and VHL81. (ASB28f 5ЈTTC-
CAAATGTCTGTCTGTCTG, ASB28r 5ЈGGTTTTGATGAACAGTGCC,
VHL81f 5ЈCGTCAGGCCACACTTAGG, VHL81r 5ЈGTGGT-
CAGACTCGGAGTAC). Primer pairs for microsatellite sequences pos-
sessing ERE-2 subunits in one of the primers were used for markers
LEX042, LEX056, and LEX057 (Coogle and Bailey 1997; Coogle et al.
1997). Reactions were run as reported previously (Coogle and Bailey 1997;
Coogle et al. 1997), with slight modifications. Specifically, reactions ran
for 30 cycles: 30 at 95°C denaturing, 30 at 50°C annealing, and 30 at 75°C
extension. After 30 cycles, a 10-min extension period at 75°C was in-
cluded. Each reaction included 1× PCR buffer (Sigma; 10 m
Tris-HCl,Correspondence to: P.C. Gallagher
Mammalian Genome 10, 140–144 (1999).
© Springer-Verlag New York Inc. 1999