Characterization of newly developed SSLP markers for the rat
Takeshi K. Watanabe,
Michael R. James,
G. Mark Lathrop,
Otsuka GEN Research Institute, Otsuka Pharmaceutical Co. Ltd. 463-10 Kagasuno, Kawauchi-cho, Tokushima 771-0192, Japan
Laboratory of Genome Database, Human Genome Center, Institute of Medical Science, The University of Tokyo, 4-6-1 Shirokanedai Minato-ku,
Tokyo 108-8639 Japan
Laboratory of Molecular Medicine, Human Genome Center, Institute of Medical Science, The University of Tokyo, 4-6-1 Shirokanedai Minato-ku,
Tokyo 108-8639, Japan
The Wellcome Trust Centre for Human Genetics, Roosevelt Drive, Oxford OX3 7BN, UK
Received: 8 July 1999 / Accepted: 3 December 1999
Abstract. We have isolated more than 12,000 clones containing
microsatellite sequences, mainly consisting of (CA)n dinucleotide
repeats, using genomic DNA from the BN strain of laboratory rat.
Data trimming yielded 9636 non-redundant microsatellite se-
quences, and we designed oligonucleotide primer pairs to amplify
8189 of these. PCR amplification of genomic DNA from five
different rat strains yielded clean amplification products for 7040
of these simple-sequence-length-polymorphism (SSLP) markers;
3019 markers had been mapped previously by radiation hybrid
(RH) mapping methods (Nat Genet 22, 27–36, 1998). Here we
report the characterization of these newly developed microsatellite
markers as well as the release of previously unpublished micro-
satellite marker information. In addition, we have constructed a
genome-wide linkage map of 515 markers, 204 of which are de-
rived from our new collection, by genotyping 48 F
crosses. This map spans 1830.9 cM, with an av-
erage spacing of 3.56 cM. Together with our ongoing project of
preparing a whole-genome radiation hybrid map for the rat, this
dense linkage map should provide a valuable resource for genetic
studies in this model species.
In recent years, high-density linkage and physical maps of the
human and mouse genomes have been constructed at a remarkable
speed (Dib et al. 1996; Dietrich et al. 1996; Gyapay et al. 1996;
Schuler et al. 1996; Stewart et al. 1997; Deloukas et al. 1998). In
comparison, the rat genome project has been progressing slowly,
and its genome resources have been relatively poor (Jacob et al.
1995; Pravenec et al. 1996; Bihoreau et al. 1997; Brown et al.
1998; Wei et al. 1998). Nevertheless, several international collabo-
rative efforts have recently resulted in the development of highly
useful rat genome resources. In particular, SSLP markers for rat
genome research number almost 10,000—to our knowledge, the
largest library in any species. Of these, 4733 have been RH
mapped (Watanabe et al. 1999a), and 4736 are genetically mapped
(Steen et al. 1999).
This report describes the characterization of our microsatellite
markers, including polymerase chain reaction (PCR) amplification
efficiencies and informative rates. Comparison of our markers with
those developed by Steen et al. (1999) reveals that the two sets are
fully complementary. This results in part from the use of different
restriction enzymes to produce a less-redundant genomic micro-
satellite library. In addition, the SSLP marker maps themselves
shared a large proportion of framework SSLP markers, allowing
full integration of genome data. Finally, we show that some mark-
ers are closely linked to functional gene sequences, enabling us to
further refine the whole-genome comparative map (Watanabe et al.
1999a). Using these new markers and integrating different sources,
we have constructed a genome-wide linkage map of 515 markers.
Materials and methods
Isolation of microsatellite-containing genomic clones.
of short-insert genomic libraries and screening procedures to identify mi-
crosatellite-positive clones have been described previously (Watanabe et
al. 1999a). Plasmid DNA was extracted with a PI-100 automatic DNA
extraction system (Kurabo Industries Ltd., Japan) and digested with the
appropriate restriction endonucleases to excise inserts. 100–200 ng of each
purified DNA fragment was sequenced with ABI 377XL sequencers with
a dye-terminator cycle sequencing kit (Applied Biosystems, PE Bio sys-
tems, Foster City, Calif.) and standard “reverse” and “−21” primers.
Establishment of SSLP markers.
Raw sequence data were transported
semi-automatically from sequencing instruments to the database-
management system. To extract non-redundant microsatellite sequences,
we trimmed the raw data in the following ways: (1) Vector-derived se-
quences were removed after comparing raw sequence data with the vector
sequence. (2) Recognition sites of the restriction enzymes used to prepare
the libraries were flagged. To exclude chimeric inserts, only sequences
lying between two recognition sites were considered. (3) Pairwise com-
parisons were performed to remove internally redundant clones, and the
remaining microsatellites were designated as “core sequences.” (4) We
used Repeat Masker 2 software to reduce the risk of problematic amplifi-
cations from SINE- or LINE-like sequences (http://ftp.genome.
washington.edu/cgi-bin/RepeatMasker). (5) We used Primer version 2.3
software to design primers for amplifying the targeted microsatellite re-
gions (Primer version 3.0 is now available at http://www-genome.wi.
mit.edu/cgi-bin/primer/primer3.cgi). Melting temperatures for all primers
were adjusted to 65°C, and SSLP markers were synthesized by the Central
Research Center of Nippon Flour Mills Co., Ltd. (Tokyo, Japan) and Am-
ersham Pharmacia Biotech (Piscataway, NJ, USA).
We used the BLAST sequence similarity search program (Altschul et
al. 1990) to determine the redundancy between our marker sequences and
those established by Steen et al. (1999). Core sequences were compared
with 4362 marker sequences obtained from the databases of the Center for
Genome Research at the Whitehead Institute for Biomedical Research
(WIBR/MIT, http://waldo.wi.mit.edu/rat/public/). These 4362 sequences
represent 92% of the markers published by Steen et al. (1999).
To identify microsatellite sequences closely linked to functional genes,
we compared all 9636 core sequences to unique sequences in the UniGene
databases (http://www3.ncbi.nlm.nih.gov/UniGene/), using the BLAST
program. In total, 62,851 human, 15,275 mouse, and 23,003 rat genes
registered with UniGene were compared. Some microsatellite sequences
Correspondence to: A. Tanigami; e-mail: firstname.lastname@example.org
Mammalian Genome 11, 300–305 (2000).
© Springer-Verlag New York Inc. 2000