Bats with hATs: Evidence for Recent DNA Transposon Activity in Genus Myotis
David A. Ray,* Heidi J. T. Pagan,* Michelle L. Thompson,* and Richard D. Stevens *Department of Biology, West Virginia University; and Department of Biological Sciences, Louisiana State University
Transposable elements make up a significant fraction of many eukaryotic genomes. Although both classes of transposable
elements, the DNA transposons and the retrotransposons, show substantial expansion in plants and invertebrates, the DNA
transposons are thought to have become inactive in mammalian genomes long ago. Here, we report the first evidence for
recent activity of DNA transposons in a mammalian lineage, the bat genus Myotis. Six recently active families of nonautonomous hobo/Activator/TAM transposons were identified in the Myotis lucifugus genome using computational tools.
Low sequence divergence among the individual sequences and between individual sequences and their respective consensus sequences suggest their recent expansion in the M. lucifugus genome. Furthermore, amplification and sequencing of
polymorphic insertion loci in a related taxon, M. austroriparius, confirms their recent activity. Myotis is one of the largest
mammalian genera with 103 species. The discovery of DNA transposon activity in this genus may therefore influence our
understanding of genome evolution and diversification in bats and in mammals in general. Furthermore, the identification
of a likely autonomous element may lead to new approaches for mammalian genetic manipulation.
Introduction
With 103 species and a nearly worldwide distribution,
genus Myotis is a very successful group of mammals whose
diversification is notably extensive. The genus appears to
have been radiating for at least the last 30 Myr (Quinet
1965) with a majority of the diversification occurring since
the late Miocene/early Pliocene, 5–6 MYA (Savage and
Russel 1983; Ariagno 1984). The most recent radiation
has given Myotis a nearly global distribution (being absent
only from polar regions and some oceanic islands) and has
included considerable ecomorphological diversification
(Nowak 1991). For example, New World Myotis appear
to have been radiating for approximately the last 4.6 Myr
(2001) and accumulation of diversity within the New
World clade has not been random. Rather, a number of
New World forms have converged to assume very similar
morphologies as forms found in the Old World (Ruedi
and Mayer 2001). Thus, it appears that Myotis represents
a relatively plastic group of species that can readily adapt
to a variety of habitats. Such morphological plasticity
has made phylogenetic inferences on this genus difficult
(Findley 1972; Ruedi and Mayer 2001; Kawai et al. 2002;
Hoofer and Van Den Bussche 2003). The inclusion of
information on mobile element activity and diversity in
this taxon may contribute greatly to our understanding of
evolutionary relationships and processes. For example, genomic instability due to the presence of mobile element
activity may represent one way to induce the genomic variation needed for the type of rapid diversification observed
in this genus of bats (Kidwell and Lisch 1997, 2001).
Transposable elements, comprising both Class I retrotransposons and Class II DNA transposons, represent a substantial portion of many eukaryotic genomes. For example,
;50% of the human genome is thought to be derived from
transposable element sequences (Lander et al. 2001). Other
genomes, especially in plants, may consist of substantially
higher proportions of transposable element–derived DNA
(SanMiguel and Bennetzen 1998; Bennetzen 2000). In
Key words: mobile element, hAT, transposon, Chiroptera, Myotis.
E-mail: [email protected].
Mol. Biol. Evol. 24(3):632–639. 2007
doi:10.1093/molbev/msl192
Advance Access publication December 5, 2006
Ó The Author 2006. Published by Oxford University Press on behalf of
the Society for Molecular Biology and Evolution. All rights reserved.
For permissions, please e-mail: [email protected]
addition to these observations on genomic content, the activity of transposons and retrotransposons can substantially
alter genome structure, produce novel exon combinations,
and affect gene expression (reviewed in Kidwell and Lisch
2001). It is clear therefore that the study of these elements is
important to our understanding of genomic structure and
function. The focus of this study is the Class II transposable
elements, the DNA transposons, and their derivatives. Although they are common and active in bacteria and many
eukaryotes (Kempken and Windhofer 2001), the analysis of
sequenced mammalian genomes suggests that Class II elements effectively became extinct long ago in mammalian
lineages (Lander et al. 2001; Waterston et al. 2002).
Here, we report the discovery of 6 families of nonautonomous DNA transposons in the Myotis lucifugus genome. Remarkably, these families appear to have been
active in the genome in the recent past and are likely still
active today. This is illustrated by minimal observed differences among individual insertions and the consensus
sequence and by the existence of polymorphic insertion sites
in a closely related taxon, M. austroriparius. The discovery
of an active transposon family in a mammalian lineage represents an important advance on several levels. First, the instability introduced into the Myotis genome by an active
group of transposons has implications for our understanding
of the evolutionary processes underlying the rapid diversification of this group. Furthermore, transposons are efficient
vectors for the introduction of foreign DNA into cells. Low
DNA transposon activity in mammalian cells, however, has
been a stumbling block to their utilization in mammalian systems. The discovery of an active nonautonomous DNA
transposon family in a mammalian lineage suggests the presence of an active autonomous partner. Identification of that
partner could be an important step forward in increasing our
ability to experimentally manipulate mammalian genomes.
Materials and Methods
An initial survey for recently active mobile element
families was performed by obtaining genomic sequence
data for 4 chiropteran taxa from the National Institutes of
Health (NIH) Comparative Vertebrate Sequencing database
(http://www.nisc.nih.gov/). Data were obtained for Target 1
(human homolog—chr7:115404472–117281897) from the
following taxa: Artibeus jamaicensis (DP000001; 584,809
Bats with hATs 633
bp), Carollia perspicillata (DP000018; 1,501,503 bp), Rhinolophus ferrumequinum (DP000028; 1,706,389 bp), and M.
lucifugus (AC174829, AC174830, AC174831, AC174832,
AC181995,
AC181998,
AC182002,
AC182003,
AC182004, AC182774, AC183324, and AC184715;
2,041,025 bp). For all taxa except M. lucifugus, the data were
available as complete genomic scaffolds. For M. lucifugus,
data were obtained from the individual bacterial artificial
chromosome (BAC) clones.
For each taxon, an initial scan for recently active mobile element families was performed using PILER (Edgar
and Myers 2005). Minimum length for discovered repetitive families was set to 100 bp and percentage identity
was set to 90. The output from PILER was organized into
families (all sequences with 90% and higher similarity) and
superfamilies (sequences from 2 or more families that
exhibited sequence similarity). Each superfamily and family alignment was given a numerical designation (table 1).
Superfamily and/or family consensus sequences were subjected to CENSOR (Jurka et al. 2005) searches to determine
similarity to known repetitive elements.
Based on the results of this analysis, a more extensive
analysis of M. lucifugus sequences comprising ;7.2 Mb of
BAC-derivedgenomicsequenceswasconducted(AC174829,
AC174830, AC174831, AC174832, AC181995, AC181998,
AC182002, AC182003, AC182004, AC182774, AC183324,
AC184715, AC175782, AC175783, AC181992, AC181996,
AC182772, AC183326, AC175607, AC175608, AC181991,
AC181994, AC181997, AC182001, AC183841, AC175784,
AC175785, AC181993, AC181999, AC183325, AC186243,
AC182000, AC182325, AC182330, AC186242, AC182326,
AC182327, AC182328, AC182329, AC182510, and
AC182773). For these data, parameters for the PILER analysis
were refined to identify elements with sequence identity of
95% or higher.
Consensus sequences for each of the recovered
transposon-like families (now referred to as Myotis-nhAT)
were used to create a custom library for a local installation
of RepeatMasker (Smit AFA, Hubley R, Green P, RepeatMasker at http://repeatmasker.org). Target 1 scaffolds
from Artibeus, Carollia, and Rhinolophus were subjected to
RepeatMasker analyses using this library to determine if
members of the DNA transposon-like families were present.
Also, as an independent estimate of the distribution and number of Myotis-nhAT elements in the genome, we subjected
;274.9 Mb of sequence (100,000 contigs with an average
length of 2,749 bp; AAPE01000001–AAPE01100000)
from the M. lucifugus–sequencing project to the same RepeatMasker analysis.
For analysis of population level variation at individual
Myotis-nhAT loci, 10 individuals of M. austroriparius were
examined. All animals were collected by mist netting over
water sources in and around the Tunica Hills Wildlife Management Area located in West Feliciana Parish, Louisiana.
Bats were collected between 11 March 2006 and 24 May
2006. Shortly after capture, bats were euthanized and frozen
( 80 °C). Specimens were thawed and tissues (muscle,
heart, liver, and kidney) were removed. Voucher specimens
and tissues can be obtained by contacting the authors. DNA
was isolated from organ tissue using the Wizard SV Genomic DNA Purification System (Promega, Madison, WI).
Table 1
Results of Our Initial PILER Analysis Using NIH Target 1
Sequence Data
Taxon
Number of
Superfamily
Sequences
Designation Families per Family
Artibeus jamaicensis
Total number
of sequences
Carollia perspicillata
0
2
Total number
of sequences
Myotis lucifugus
0
2
Total number
of sequences
Rhinolophus
ferrumequinum
Total number
of sequences
a
CENSOR
Results
0.0
1.1
2.2
3
3
3
9
Ves-SINE
Ves-SINE
NRa
0.4
0.7
1.6
2.3
2.5
3.2
4.1
5.0
3
3
3
3
3
3
4
3
25
L1
L1
NR
L1, ERV11
L1, ERV11
L1
Ves-SINE
Ves-SINE
0.2
0.4
0.8
14
9
3
1.7
2.0
2.6
3.5
4.3
5.1
6.9
7.1
3
16
3
10
6
14
3
3
84
URR1
URR1
URR1B,
MarinerA6A
NRa
Ves-SINE
Ves-SINE
hAT-4_XT
hAT2_MD
hAT1_MD
NRa
Ves-SINE
0.0
3
1.1
2.2
4
4
11
NRa
NRa
hAT2_MD
CENSOR did not find a match with transposable elements in the database.
We used Primer3 (Rozen and Skaletsky 1998) to design oligonucleotide primers for 27 of the suspected recent
integrations (supplementary table 1, Supplementary Material online) and used them to amplify the loci in a panel of
10 DNA templates from M. austroriparius. Twenty-five–
microliter polymerase chain reaction amplifications were
performed under the following conditions: 10–50 ng template DNA, 7 pM of each oligonucleotide primer, 200 mM
dNTPs, in 50 mM KCl, 10 mM Tris–HCl (pH 8.4), 2.0 mM
MgCl2, and Taq DNA polymerase (1.25 units). An initial
denaturation at 94 °C for 2 min was followed by 30–32
cycles of 94 °C for 15 s, the appropriate annealing temperature for 15 s, and 72 °C for 1 min and 10 s. A final incubation at 72 °C for 5 min prepared the fragments for
potential cloning.
Six loci exhibited patterns that suggested polymorphism with regard to Myotis-nhAT presence/absence. Successful amplicons of representative filled (containing the
transposon integration) and empty (without the transposon
integration) loci were cloned using the TOPO-TA cloning
kit (Invitrogen, Carlsbad, CA), and inserts were sequenced
using chain termination sequencing on an ABI 3130xl
Genetic Analyzer. Sequences were aligned with the matching computationally derived locus from M. lucifugus. All
634 Ray et al.
alignments are available as supplementary data and at
http://www.as.wvu.edu/;dray/pubs.html. All sequences
generated for this manuscript have been deposited in GenBank under accession numbers DQ906144–DQ906155.
We also attempted to identify any candidates for the
autonomous partner of these elements using the available
shotgun sequences from M. lucifugus by creating a custom
Blast database consisting of all available sequences from
that project (GenBank accession numbers AAPE01
000001–AAPE01640786, ;1.76 3 109 bases). The consensus sequences from each suggested autonomous transposonmatch(hobo/Activator/TAM [hAT]-4_XT, hAT1_MD,
and hAT2_MD) from our original search (table 1) were used
to query that database.
Results
Results of our initial analyses of 4 chiropteran taxa (A.
jamaicensis, C. perspicillata, R. ferrumequinum, and M. lucifugus) using Target 1 data from the NIH Comparative
Vertebrate Sequencing database are presented in table 1.
It was clear from the data that M. lucifugus presented by
far the largest proportion of potential repetitive families,
89 sequences in 11 families. However, because of the potential for overlapping BAC clones and thus multiple examples of the same insertion being recovered, we examined
both of the flanking sequences of each element for similarity with the flanking sequences of all other elements. Pairs
with one or both flanking sequences exhibiting 95% identity or higher were assumed to be multiple instances of the
same element, and one from each pair was removed from
further analysis. Characterization of the remaining 84 elements using CENSOR led to the observation that within the
;2 Mb of sequence representing Target 1, there were 56
instances of elements with similarity to known DNA transposons spanning a range of families. Analyses of the same
target in the 3 other chiropteran taxa revealed no comparable numbers of DNA transposon-like sequences.
During the subsequent analysis of 7.2 Mb of BACderived sequence for M. lucifugus, we identified a total of
111 elements with similarity to known DNA transposons.
Multiple instances of the same element were removed using
the methods described above. The remaining elements were
subdivided among 6 groups based on sequence similarity.
These groups have been designated Myotis-nhAT1–6.
Two groups, Myotis-nhAT4 and 5, consisted of 2 subfamilies
each based on sequence differences and indels. As indicated
by the names, details suggest that all of the elements described are nonautonomous members of the hAT superfamily of transposons. This is supported by the observation that
all of the families described share 2 distinguishing features of
the superfamily. First, hAT transposons usually exhibit 8 bp
target site duplications (TSD) and short (5–27 bp) terminal
inverted repeats (TIR) (Kempken and Windhofer 2001).
Each of the 6 families of Myotis-nhAT elements exhibits
TIRs of 16 bp (table 2). These TIRs match well with the consensus TIRs for hAT elements described by Rubin et al.
(2001). Second, TSDs were identifiable for nearly all of
the Myotis-nhAT elements, and consensus TSD sequences
are available in table 2. The TSDs are typically 8 bp long
and almost invariably contain a central TA dinucleotide
(fig. 1 and table 2). Furthermore, as described in table 3,
all of the Myotis-nhAT consensus sequences described here
exhibit some degree of sequence similarity with known
DNA transposons.
Two of the Myotis-nhAT families exhibited very high
degrees of sequence similarity to previously described elements. For example, over portions of their length the
Myotis-nhAT 4 and 5 subfamilies are essentially identical
to the URR1, URR1B, and MarinerNA6A transposon families. One might argue that elements exhibiting high sequence similarity to already described elements should
properly be referred to using those names. However, in each
case the elements recovered from Myotis all share unique
indels (see supplementary Myotis nhAT family alignments,
Supplementary Material online) that distinguish them from
the previously described elements. We argue that the taxonomic distribution and the unique sequence characteristics,
whether as indels or nucleotide differences, merit their distinction as separate families.
On a secondary nomenclatural note, the Repbase
entries for both URR1 and MarinerNA6A extend past the
ends of the Myotis-nhAT elements by 5 bases. The final
2 bases on both ends are TA in each case. These were interpreted by their respective discoverers as TA direct repeats, a common feature of DNA transposons. We
suggest instead that these 5-bp overhangs might represent
incomplete and unrecognized 8-bp TSDs. If that is the case,
we suggest their reclassification as hAT elements.
Calculations based on the ;274.9 Mb that were subjected to a custom RepeatMasker analysis suggest an average density of 1 Myotis-nhAT insertion every 14.3 kb. The
size of the M. lucifugus genome is unknown. However, several estimates for other Myotis species are available at http://
www.genomesize.com. Using the average genome size values for Myotis available, we can presume a size of ;2.4 Gb.
Thus, the total number of Myotis-nhAT elements in the M.
lucifugus genome may be in the neighborhood of 168,000,
occupying ;1.3% of its total sequence (table 2). Of course,
all of these calculations are based on the assumption that the
contigs used are representative of the overall composition
of the M. lucifugus genome and that our estimate of the genome’s size is correct. Thus, they should be taken only as
preliminary estimates.
The presence of multiple insertions with only minimal
divergence from each other and from the consensus suggested that genomes of Myotis may have experienced a recent burst of DNA transposon activity. Myotis-nhAT
elements from the 6 families described fit this criterion
based on genetic distance estimates of the members discovered versus their respective consensus sequence (1.4–3.3%;
table 2). Based on the estimated neutral mutation rate for
mammals (2.2 3 10 9 [Kumar and Subramanian 2002]),
the average age for these transposon families is between
6.4 and 15 Myr. This observation supports the idea that
these elements arose in and are unique to genus Myotis.
To further test the assumption of recent activity, we
hypothesized that very recent integrations, those with minimal divergence from the consensus sequence, might still be
polymorphic in populations of M. lucifugus and/or related
taxa. Ten DNA samples from a related taxon, M. austroriparius (divergence ,6 Myr), were tested using primers for
Bats with hATs 635
Table 2
Descriptive Information for Each Myotis-nhAT Family Recovered
Family
K2P
Divergence
Consensus
from
Size (n) Consensusa
Myotis-nhAT1
Myotis-nhAT2
Myotis-nhAT3
Myotis-nhAT4a
Myotis-nhAT4b
Myotis-nhAT5a
Myotis-nhAT5b
Myotis-nhAT6
235
204
209
212
192
311
223
215
(19)
(14)
(9)
(15)
(5)
(14)
(15)
(15)
0.019
0.016
0.026
0.028
0.021
0.027
0.033
0.014
Consensus
TSD (n)
Consensus TIR
NNHTANNN (18)
VBYTANNB (13)
GBSTAGVC (4)
VNCTANNN (13)
—b
BHNTADAV (14)
NHNTADDN (14)
NNCTAGNV (14)
NNNTANNN (90)
CAGTGATGGCGAACCT
CAGGGGTCCTCAAACT
CAGTGGTCGGCAAACT
CAGYGGTTCTCAACCT
CAGYGGTTCTCAACCT
CAGYGGTTCTCAACCT
CAGYGGTTCTCAACCT
CAGSSGTGGGCAAACT
CAGKGGTBSBCAAMCT
c
Total
Total
Total
Number in Basepairs Percentage Total in Basepairs
d
274.9 Mb in 274.9 Mb of 274.9 Mb Genome in Genomed
3,425
1,307
4,206
3,521
703
1,915
1,964
2,073
19,114
715,304
227,524
467,615
681,585
106,758
529,868
376,809
402,988
3,508,451
0.26%
0.08%
0.17%
0.25%
0.04%
0.19%
0.14%
0.15%
1.28%
30,140 7,082,900
11,502 2,346,326
37,013 7,735,675
30,985 6,568,778
6,186 1,187,789
16,852 5,240,972
17,283 3,854,154
18,242 3,922,116
168,203 37,938,710
a
Based on analysis of all corresponding matches from 7.2 Mb of M. lucifugus genomic sequence identified using RepeatMasker.
Too few recognizable TSD sequences were available to generate a consensus.
Underlined bases are matches to highly conserved bases for hAT transposons described by (Rubin et al. 2001).
d
Extrapolated based on an estimated size of 2.42 Gb for the M. lucifugus genome.
b
c
specific Myotis-nhAT loci. Samples from M. lucifugus
would have been preferred but were unavailable. Of the
primer pairs tested, 6 showed evidence of polymorphism
for presence and/or absence among the 10 M. austroriparius samples (fig. 2). Sequence analysis of the filled
and empty sites clearly demonstrates the absence of the
Myotis-nhAT sequence at each locus along with only one
copy of the presumed TSD (fig. 3; supplementary fig. 1,
Supplementary Material online).
A Blast search of M. lucifugus genomic sequences
yielded 7 hits with high sequence similarity to hAT1_MD,
a putative autonomous DNA transposon from Monodelphis
domestica. No sequences with significant similarity to other
queried DNA transposons were obtained. An alignment of
the sequences recovered by the Blast search, including
a consensus approximated from them and 1 Myotis-nhAT
sequence, is presented as supplementary figure 2 (Supplementary Material online). One of the hits encompassed
what appears to be a full-length element with very similar
TIRs (5#-CAGTGATGGMGAACCT-3#) to all of the
Myotis-nhAT families (table 2) and easily recognizable 8bp TSDs (5#-ACCTAGGG-3#). However, 3 single nucleotide indels (at positions 215, 884, and 1,741) produce
frameshifts that introduce early stop codons (supplementary
fig. 3, Supplementary Material online). The consensus sequence generated from all 7 hits, however, encodes a presumably functional transposase identical in size to
hAT1_MD, 643 amino acids. We have named this potentially active autonomous hAT transposon Myotis-hAT1.
Given the similarity in TSD sequences and TIR composition, it is likely that each Myotis-nhAT family described
above utilizes the Myotis-hAT1–encoded transposase.
Discussion
The observations presented here suggest that much
care should be taken when extrapolating results gathered
from only a few model organisms to an entire taxonomic
group and illustrate the value of analyzing the genomes
of a broad range of mammalian species. Recent DNA transposon activity in a mammalian lineage and the identification of a potentially active DNA transposon sequence in the
genome of M. lucifugus was unexpected given the lack of
observed transposase activity in mammals in general and in
the 2 most well studied mammals, human and mouse
(Lander et al. 2001; Waterston et al. 2002)—an observation
that has disappointed researchers interested in using transposons as tools for genetic manipulation in mammals (Koga
et al. 2003). Recent work with DNA transposons from other
taxa, including Sleeping Beauty, piggyBac, and Tol2, has
given hope for the potential to use transposons as tools
for such manipulation (Geurts et al. 2003; Koga et al.
2003; Bestor 2005; Ding et al. 2005; Dupuy et al. 2005).
Although these are significant advances, the discovery of
an active mammalian transposase could provide an alternative to these systems. We hope that the description here of
the potentially active Myotis-hAT1 element will revitalize
the search for mammalian transposases.
The high degree of sequence similarity among some
Myotis-nhAT elements and elements in very divergent taxa
raises questions regarding their origin. It is unlikely that elements with such highly similar sequences in such widely
divergent taxa arose by chance alone. We believe the most
likely explanation is horizontal transfer. DNA transposons
are thought to be more prone to horizontal transfer than the
retrotransposons (Kidwell 1992). We suggest the relatively
recent (at least 15 MYA, see Results) introduction of an
autonomous element, Myotis-hAT1, to the ancestral Myotis
genome via horizontal transfer from some unknown source.
After the incorporation of Myotis-hAT1 into the genome,
nonautonomous versions arose via mechanisms that remain
unclear (Hua-Van et al. 2005).
The estimated divergence values of the subfamilies described in table 2 are notably uniform. However, given the
presumed mode of expansion for these elements, it is not
unexpected. The relationship between age of a subfamily
and the divergence of its members from the consensus is
essentially linear when dealing with retrotransposons because they follow a modified master gene model of expansion. Following this model, only one or a few copies of
a given subfamily are active at any given time, and the copies are generated from single or very few active master
genes (Cordaux et al. 2004). Subfamilies of retrotransposons also have a wide range of ages depending on the period
636 Ray et al.
FIG. 1.—Multiple sequence alignment of 10 randomly selected Myotis-nhAT1 elements recovered from a search of 7.2 Mb of Myotis lucifugus
genomic sequence data using PILER. Twenty-two bases of flanking sequence are also included. Shaded regions indicate the 8-bp TSDs typical of
hAT transposons. The central TA dinucleotide within the TSD sequences is also indicated (**).
during which the various subfamilies were active. As an
example, RepeatMasking Target 1 from the human genome
yields raw percentage–divergence values ranging from
0.0% to 24.6% for Alu elements (a primate retrotransposon)—presumably spanning the range of potential ages
from de novo insertions to insertions dating back to the origin of Alu elements ;60 MYA.
When dealing with DNA transposon families, however, the pattern would necessarily differ because they follow a different model of mobilization. As long as the
transposase is active and as long as there are TIRs for
the transposase to recognize and mobilize, many copies
can be ‘‘active’’ and many loci can act as sources. Thus,
the relationship that exists for retrotransposons between
age and divergence from the consensus sequence likely
does not apply to these elements. Instead, we would be
more likely to find a uniform set of ages for each of these
families given that the source of their transposition activity
is the same, the autonomous transposon responsible for the
mobilization of the entire group. In other words, unlike retrotransposons all of the nonautonomous families will likely
be active or inactive at any given time depending on the
state of their autonomous partner.
The detection of multiple nearly identical instances of
these nonautonomous DNA transposons in M. lucifugus is
strong evidence for their recent activity in that genome. Observing instances of insertion polymorphism adds experimental evidence to further support this hypothesis. Of
course, we would have preferred to test for insertion polymorphism in a panel of taxa from that species, and it is unfortunate that we did not have access to such samples.
However, the detection of polymorphic nonautonomous
DNA transposon insertion sites in a panel of 10 M. austroriparius using loci first identified in M. lucifugus DNA
sequence is interesting at several levels. First, such results
clearly indicate the recent incorporation of Myotis-nhAT
Bats with hATs 637
Table 3
Best CENSOR Matches of Myotis-nhAT Elements to Known Transposon Sequences
Myotis-nhAT1
Myotis-nhAT2
Myotis-nhAT3
Myotis-nhAT4a
Myotis-nhAT4b
Myotis-nhAT5a
Myotis-nhAT5b
Myotis-nhAT6
CENSOR Match
Taxon
hAT1_MD
hATN6_AG
hAT2_MD
Chaplin1A_FR
URR1
MarinerNA6A_MD
URR1
MarinerNA6A_MD
URR1B
MarinerNA6A_MD
URR1B
MarinerNA6A_MD
hAT4_XT
Monodelphis domestica
Anopheles gambiae
M. domestica
Takifugu rubripes
Rodentia
M. domestica
Rodentia
M. domestica
Muridae
M. domestica
Muridae
M. domestica
Xenopus tropicalis
Portion of nhAT
Consensus
Sequence
1–84, 181–235
1–35, 173–204
1–35, 43–204
1–13, 163–209
1–212
1–72, 73–212
1–32, 39–192
1–31, 45–192
1–84, 86–311
1–85, 86–311
1–99, 156–223
1–91, 156–223
1–46, 168–215
transposons at these loci in a common ancestor of M. lucifugus and M. austroriparius. The observation that some loci
are heterozygous also supports the hypothesis of recent activity. No divergence estimates for M. lucifugus and M. austroriparius are available in the literature. However, given
that both are New World taxa and an assumption of monophyly for New World Myotis (Ruedi and Mayer 2001;
Hoofer and Van Den Bussche 2003), it is likely that they
shared a common ancestor no later than the late Miocene
(5–6 MYA) (Ruedi and Mayer 2001). Their polymorphic
status in M. austroriparius indicates that the alleles are undergoing a period of lineage sorting. This is common
among species that are experiencing or have experienced
a rapid burst of speciation. Such insertions have been observed previously when short interspersed element (SINE)
% of nhAT
59.1
32.8
96.5
28.7
100
100
96.9
93.8
100
100
74.9
71.3
43.7
Matching Portion of
Known Element
Consensus Sequence
7–91, 2941–2992
1–35, 284–315
8–42, 3134–3291
1–13, 472–518
6–231
6–78, 555–707
6–37, 78–231
6–36, 679–707
1–84, 171–225
6–90, 482–707
1–97, 158–225
6–96, 640–707
1–46, 3353–3400
p Distance from
CENSOR Match
23.7
8.6
11.4
15.0
4.7
13.3
1.1
9.4
1.8
10.8
3.1
10.7
11.7
loci were examined in cichlids and charr (Hamada et al.
1998; Takahashi et al. 2001). The recent radiation
suggests the possibility for shared polymorphisms among
taxa, and the observation that in as few as 4.6 Myr a large
portion of the ;103 species have emerged clearly makes
Myotis spp. candidates for observing lineage sorting in
action.
Given the transposition mechanism of DNA transposons, it is possible that the observed Myotis-nhAT polymorphisms could have a different source—excision of the
elements via transposase activity. Although this is possible,
we think it is unlikely given the clear absence of any telltale
signatures associated with the removal of a transposon. If
the elements had undergone a precise ‘‘cut and paste’’ mobilization, we would expect to at least see the presence of 2
FIG. 2.—Agarose gel chromatographs illustrating the 6 polymorphic Myotis-nhAT insertion sites in Myotis austroriparius. Numbers across the top are
identifiers for individual specimens. Fragment sizes for filled and empty sites are indicated.
638 Ray et al.
FIG. 3.—Representative sequence alignment for 1 polymorphic Myotis-nhAT insertion site. The top sequence is the original sequence obtained from
BAC-derived Myotis lucifugus genomic sequence. The bottom 2 sequences correspond to filled and empty sites pictured in figure 1. The underlined
segments are TSDs, and the boxed sequences represent TIR. Dots represent identical bases, and dashes indicate indels.
TSD sequences in the empty sites. Furthermore, the typical
footprint in cases of excision is a few basepairs of the transposon. The lack of any of these telltale signs strongly suggests that these empty sites were never occupied by the
transposon and thus the polymorphisms result from insertion rather than excision events. Even if excision were the
case, it would serve as additional evidence of the activity of
the presumed autonomous element. Another possibility that
must be addressed is illegitimate recombination. We do not
believe that this process is playing a major role in our observations given the easily identifiable TSDs for most loci.
It is also possible that the polymorphisms we have observed are actually the result of paralogous insertions. This
scenario would require the presence of duplicated regions
of the M. austroriparius genome, an occurrence that is not
uncommon in some mammals (Cheung et al. 2003; Tuzun
et al. 2004; Cheng et al. 2005). One copy of the duplicated
region would have to have been the site of an insertion,
whereas the other remained clear. In such cases, recent activity of the Myotis-nhAT transposons is still indicated
given that the paralogous insertions are not fixed in the population and are heterozygous in some individuals.
The discovery of this activity in a taxonomic group
that has undergone a recent diversification suggests a possible model system to study the effects of mobile element
activity on genomic diversification and speciation. Future
work will be needed in several areas. First, we will need
to determine the taxonomic limits of the transposon activity. Is the observed activity truly limited to Myotis? Is it
limited to only the North American Myotis? Is it common
for species-rich mammalian genera to have transposon activity? Four hAT2-like elements and 2 unidentified repetitive families were recovered from the R. ferrumequinum
genome during the initial analysis. Rhinolophus is also a relatively species-rich genus with 77 taxa. Although it is possible that this taxon also experienced a burst of transposable
element activity, they have apparently not been as active as
in Myotis given the substantially smaller number of elements recovered. Alternatively, a burst of activity in Rhinolophus may have occurred in the more distant past. If
that were the case, our search for very recent elements using
PILER would likely not have identified them.
In conclusion, we have demonstrated that nonautonomous members of the hAT transposon superfamily have recently been active in the genomes of at least 2 bat species
from genus Myotis. The presumed autonomous partner was
also identified and experiments are currently being designed to determine if mobilization can be observed in
mammalian cells. These observations represent the first
suggestion of an active DNA transposon in a mammalian
lineage, and the future utilization of the full-length autonomous element may be a stepping stone to an increased
ability to manipulate mammalian genomes. The burst of activity may have influenced the ability of Myotis genomes to
adapt to a wide variety of habitats and feeding strategies
through its effects on genome structure and by consequence
phenotypic change.
Supplementary Material
Supplementary table 1 and figures 1–3 and the sequences of the transposon loci, both filled and empty sites that
have been assigned GenBank accession numbers
(DQ906144–DQ906155), are available at Molecular Biology and Evolution online (http://www.mbe.oxfordjournals.
org/).
Bats with hATs 639
Acknowledgments
This work was supported by the Eberly College of Arts
and Sciences at West Virginia University (D.A.R.). R.D.S.
was supported by National Science Foundation (Grant
#DEB 0535939). Drs S. DiFazio and J. Xing and an anonymous reviewer contributed valuable comments to earlier
drafts.
Literature Cited
Ariagno D. 1984. Myotis myotis et Myotis blythi. In: Faryard A,
editor. Atlas des mammiferes sauvages de la France. Paris:
S.F.E.P.M., p. 68–71.
Bennetzen JL. 2000. Transposable element contributions to plant
gene and genome evolution. Plant Mol Biol. 42:251–269.
Bestor TH. 2005. Transposons reanimated in mice. Cell. 122:322–
325.
Cheng Z, Ventura M, She X, Khaitovich P, Graves T, Osoegawa
K, Church D, de Jong P, Wilson RK, Pääbo S, Rocchi M,
Eichler EE. 2005. A genome-wide comparison of recent
chimpanzee and human segmental duplications. Nature. 437:
88–93.
Cheung J, Wilson MD, Zhang J, Khaja R, MacDonald JR, Heng
HH, Koop BF, Scherer SW. 2003. Recent segmental and gene
duplications in the mouse genome. Genome Biol. 4:R47.
Cordaux R, Hedges DJ, Batzer MA. 2004. Retrotransposition of
Alu elements: how many sources? Trends Genet. 20:464–467.
Ding S, Wu X, Li G, Han M, Zhuang Y, Xu T. 2005. Efficient
transposition of the piggyBac (PB) transposon in mammalian
cells and mice. Cell. 122:473–483.
Dupuy AJ, Akagi K, Largaespada DA, Copeland NG, Jenkins NA.
2005. Mammalian mutagenesis using a highly mobile somatic
Sleeping Beauty transposon system. Nature. 436:221–226.
Edgar RC, Myers EW. 2005. PILER: identification and classification of genomic repeats. Bioinformatics. 21 (Suppl 1):i152–
i158.
Findley JS. 1972. Phenetic relationships among bats of the genus
Myotis. Syst Zool. 21:31–52.
Geurts AM, Yang Y, Clark KJ, Liu G, Cui Z, Dupuy AJ, Bell JB,
Largaespada DA, Hackett PB. 2003. Gene transfer into
genomes of human cells by the sleeping beauty transposon system. Mol Ther. 8:108–117.
Hamada M, Takasaki N, Reist JD, DeCicco AL, Goto A, Okada N.
1998. Detection of the ongoing sorting of ancestrally polymorphic SINEs toward fixation or loss in populations of two species of charr during speciation. Genetics. 150:301–311.
Hoofer SR, Van Den Bussche RA. 2003. Molecular phylogenetics
of the chiropteran family Vespertilionidae. Acta Chiropt. 5
(suppl):1–63.
Hua-Van A, Le Rouzic A, Maisonhaute C, Capy P. 2005. Abundance, distribution and dynamics of retrotransposable elements
and transposons: similarities and differences. Cytogenet Genome Res. 110:426–440.
Jurka J, Kapitonov VV, Pavlicek A, Klonowski P, Kohany O,
Walichiewicz J. 2005. Repbase Update, a database of eukaryotic repetitive elements. Cytogenet Genome Res. 110:462–467.
Kawai K, Nikaido M, Harada M, Matsumura S, Lin LK, Wu Y,
Hasegawa M, Okada N. 2002. Intra- and interfamily
relationships of Vespertilionidae inferred by various molecular
markers including SINE insertion data. J Mol Evol. 55:
284–301.
Kempken F, Windhofer F. 2001. The hAT family: a versatile
transposon group common to plants, fungi, animals, and
man. Chromosoma. 110:1–9.
Kidwell MG. 1992. Horizontal transfer. Curr Opin Genet Dev.
2:868–873.
Kidwell MG, Lisch D. 1997. Transposable elements as sources of
variation in animals and plants. Proc Natl Acad Sci USA.
94:7704–7711.
Kidwell MG, Lisch DR. 2001. Perspective: transposable elements,
parasitic DNA, and genome evolution. Evolution Int J Org
Evolution. 55:1–24.
Koga A, Iida A, Kamiya M, Hayashi R, Hori H, Ishikawa Y,
Tachibana A. 2003. The medaka fish Tol2 transposable element can undergo excision in human and mouse cells. J
Hum Genet. 48:231–235.
Kumar S, Subramanian S. 2002. Mutation rates in mammalian
genomes. Proc Natl Acad Sci USA. 99:803–808.
Lander ES, Linton LM, Birren B, et al. 2001. Initial sequencing
and analysis of the human genome. Nature. 409:860–921.
Nowak RM. 1991. Walker’s mammals of the World. Baltimore
(MD): Johns Hopkins University Press.
Quinet GE. 1965. Myotis misonnei. chiroptere de l’Oligocene de
Hoogbutsel. Bull Inst R Sci Nat Belg. 41:1–11.
Rozen S, Skaletsky H. 2000. Primer3 on the WWW for general
users and for biologist programmers. In: Krawetz S, Misener
S, editors. Bioinformatics Methods and Protocols: Methods in
Molecular Biology. Totowa (NJ): Humana Press, pp 365–386.
Rubin E, Lithwick G, Levy AA. 2001. Structure and evolution of
the hAT transposon superfamily. Genetics. 158:949–957.
Ruedi M, Mayer F. 2001. Molecular systematics of bats of the genus Myotis (Vespertilionidae) suggests deterministic ecomorphological convergences. Mol Phylogenet Evol. 21:436–448.
SanMiguel P, Bennetzen JL. 1998. Evidence that a recent increase
in maize genome size was caused by the massive amplification
of intergene retrotransposons. Ann Bot. 82:37–44.
Savage DE, Russel DE. 1983. Mammalian paleofaunas of the
World. Reading (MA): Addison-Wesley.
Takahashi K, Terai Y, Nishida M, Okada N. 2001. Phylogenetic
relationships and ancient incomplete lineage sorting among
cichlid fishes in Lake Tanganyika as revealed by analysis
of the insertion of retroposons. Mol Biol Evol. 18:2057–2066.
Tuzun E, Bailey JA, Eichler EE. 2004. Recent segmental duplications in the working draft assembly of the brown Norway rat.
Genome Res. 14:493–506.
Waterston RH, Lindblad-Toh K, Birney E, et al. 2002. Initial sequencing and comparative analysis of the mouse genome. Nature. 420:520–562.
Adriana Briscoe, Associate Editor
Accepted November 29, 2006