Gene flow of unique sequences between Mus musculus domesticus and
Rhonda Greene-Till, Yingping Zhao, Stephen C. Hardies
Department of Biochemistry, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive,
San Antonio, Texas 78284-7760, USA
Received: 14 May 1999 / Accepted: 5 November 1999
Abstract. Allelic diversity has been examined from a variety of
Mus musculus subspecies and Mus spretus strains by sequencing at
a 453-bp unique sequence locus. One M. m. domesticus classic
inbred strain, C57BL/KsJ, contained a sequence identical to that in
the M. spretus wild-derived inbred strain SEG, and other wild M.
spretus isolates. Such a result should have been precluded by the
expected divergence between the species unless there has been
interspecies gene flow. Examination of C57BL/KsJ for M. spretus-
specific repetitive sequences shows that it is neither a mis-
identified spretus strain nor a domesticus/spretus hybrid. Thus, in
addition to the previously reported presence of small amounts of
Mus spretus-specific repetitive DNA in M. m. domesticus, there is
a detectable flow of unique sequence between the two species.
There was also ancestral polymorphism observed among the spre-
tus alleles. The difficulty of distinguishing ancestral polymorphism
from horizontal transfer is discussed.
The Mus musculus complex of house mouse subspecies separated
from its nearest relatives, M. spretus and M. spicilegus, about 1.1
Mya (She et al. 1990). The coalescence of alleles within the M.
musculus complex indicates a population restriction about 0.9 Mya
for the ancestral M. musculus species followed by further restric-
tions during the derivation of several subspecies including M. mus-
culus domesticus and M. musculus musculus (Bourset et al. 1996;
Din et al. 1996). It is unclear whether the current M. musculus
subspecies have retained much polymorphism dating back to the
common ancestor with M. spretus. Currently, M. m. domesticus
and M. spretus share a geographical range. These two species can
interbreed and produce fertile offspring in captivity (Bonhomme et
al. 1978, 1979), although hybrids have not been found in nature.
Previously, we reported two instances of spretus LINE-1 re-
petitive sequences found in the classic inbred strain C57BL/6J
(Rikke et al. 1995; Zhao et al. 1996). The classic inbred strains
were derived from strains developed by mouse fanciers, and hence
their exact history is clouded. However, genetic studies have de-
tected components derived from both M. m. domesticus and M. m.
musculus (Nagamine et al. 1992; Festig 1996). Therefore, although
classified as M. m. domesticus, the classic inbred strains preserve
more of the diversity of the ancestral M. musculus species than do
wild M. m. domesticus.
The two spretus-like LINE-1 sequences in C57BL/6J closely
match their M. spretus counterparts in DNA sequence, implying a
recent interspecies exchange. For example, with the neutral diver-
gence rate of 1%/Myr, which accommodates the rodent speed-up
and is consistent with the speciation times quoted (She et al. 1990),
the LINE-1 sequence at the D4Sh1 locus (L1C105) shared a com-
mon ancestor with M. spretus LINE-1s within the last 88,000
years. This is actually the time at which the last known parental
LINE-1 sequence in the lineage leading to L1C105 came into
existence in M. spretus. So, naturally the actual insertion of
L1C105 and the transfer into M. m. domesticus must have come
some time afterwards. Since M. m. domesticus, but not M. m.
musculus, has geographical contact with M. spretus, it is presumed
that the spretus-like L1C105 insert must have made its way into
C57BL/6J through the M. m. domesticus component of its genome.
The total amount of spretus LINE-1 in C57BL/6J is about
0.65% of that in strains of M. spretus. This implies a level of about
0.65% of the M. m. domesticus genome at large should be derived
from M. spretus in recent times. A caveat to the latter conclusion
is that spretus-derived LINE-1s may have amplified after an origi-
nal introduction into M. m. domesticus and hence may exaggerate
a more miniscule amount of gene flow. On the other hand, if the
LINE-1s have just been passively transferred along with random
components of the spretus genome, then spretus alleles should
exist at ordinary nontransposable loci in M. m. domesticus at a
frequency that could be observed. A nonrepetitive allele that has
transferred between these two species is reported here.
Materials and methods
Genomic mouse DNAs (C57BL/6J, C57BL/KsJ, C58/J,
CBA/J, M. m. castaneus (CAST/Ei), M. spicilegus (PANCEVO, called M.
hortulanus by Jackson Laboratory), M. m. molossinus (MOLD/Rk), SJL/J,
SPRET/Ei, and WSB/Ei) were from the Mouse DNA Resource, Jackson
Genomic DNA samples from the wild-derived M. spretus strains SEG
and STF were gifts from Jean-Louis Gue´net. Five wild spretus genomic
DNAs (animals D3878, D3879, D3889, D3890, D3891) were generously
provided by Michael Potter. D3878 and D3879 were from an outbred
colony started from mice sampled in Morocco, and D3889, D3890, and
D3891 were from an outbred colony started from animals sampled in
Spain. Spret/Ei also originated from the latter colony. M. cervicolor DNA
was prepared as described (Sambrook et al. 1989) from livers generously
provided by Frank Berger. DNAs for the wild-derived inbred strains M. m.
domesticus DMZ and M. spretus SMZ were isolated from spleen samples
provided by Francois Bonhomme, with a variation of the method in Strauss
(1994). The protocol was modified as follows: ammonium acetate was
omitted from the ethanol precipitation, and DNA was harvested by hooking
it with the sealed end of a Pasteur pipet instead of by centrifugation.
PCR amplification and cloning.
PCR amplification prior to cloning the
unique sequence alleles was conducted with the “hot start” protocol
(D’Aquilla et al. 1991). PCR amplification was otherwise under standard
conditions (Innis and Gelfand 1990) for 30 cycles. The forward and reverse
primers, oC105B-4 (5Ј-TGGATATACCTGTTGAAGCAT) and
oC105B-5 (5Ј-CGGTTACAATTTTGTATATCCTCTGT) respectively,
were annealed at 55°C. PRC products were directly cloned into pCRII
(Clark 1988; Mead et al. 1991) with the TA cloning kit from Invitrogen, or
into pGEM-T using a cloning kit from Promega.
Correspondence to: S.C. Hardies, e-mail: email@example.com
Mammalian Genome 11, 225–230 (2000).
© Springer-Verlag New York Inc. 2000