Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 1

Menzel G, Heitkam T, Seibt KM, Nouroz F, Müller-Stoerme M, Heslop-Harrison JS, Schmidt T. 2014. The diversification and activity of hAT transposons in Musa genomes. Chromosome Research 22: 559–571. DOI 10.1007/s10577-014-9445-5 andPubmed link ID: 25377178 and author print hATs in Musa _2014_CR_MenzelEtAlAuthorVersion. (see www.molcyt.com)

The diversification and activity of hAT transposons in Musa genomes

Gerhard Menzel, Tony Heitkam, Kathrin M. Seibt, Faisal Nouroz, Manuela Müller-Stoerme, John S. Heslop-Harrison, Thomas Schmidt

G. Menzel, T. Heitkam, K. M. Seibt, M. Müller-Stoermer, T. Schmidt

Institute of Botany, Technische Universität Dresden, D-01062 Dresden, Germany

F. Nouroz, J.S. Heslop-Harrison

Department of Biology, University of Leicester, Leicester LE1 7RH, United Kingdom

T. Schmidt ()

Institute of Botany, Technische Universität Dresden

e-mail: thomas.schmidt@tu-dresden.de

Tel.: (+49) 0351 463 39588

Fax: (+49) 0351 463 39590

Running title hAT transposons in Musa genomes

Electronic supplementary material The online version of this article (doi: XXX) contains supplementary material, which is available to authorized users.

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 2

Abstract

Sequencing of plant genomes often identified the hAT superfamily as largest group of

DNA transposons. Nevertheless, detailed information on the diversity, abundance and

chromosomal localization of plant hAT families are rare. By in silico analyses of the

reference genome assembly and BAC sequences, respectively, we performed the

classification and molecular characterization of hAT transposon families in Musa

acuminata. Musa hAT transposons are organized in three families MuhAT I, MuhAT

II and MuhAT III. In total, 70 complete autonomous elements of the MuhAT I and

MuhAT II families were detected, while no autonomous MuhAT III transposons were

found. Based on the terminal inverted repeat (TIR)-specific sequence information of

the autonomous transposons, 1722 MuhAT I- and MuhAT II-specific miniature

inverted repeat transposable elements (MuhMITEs) were identified. Autonomous

MuhAT I and MuhAT II elements are moderately abundant in the sections of the

genus Musa, while the corresponding MITEs exhibit an amplification in Musa

genomes. By fluorescent in situ hybridization, autonomous MuhAT transposons as

well as MuhMITEs were localized in subtelomeric, most likely gene-rich regions of

M. acuminata chromosomes. A comparison of homoeologous regions of M.

acuminata and Musa balbisiana BACs revealed the species-specific mobility of

MuhMITEs. In particular, the activity of MuhMITEs II showing transduplications of

genomic sequences might indicate the presence of active MuhAT transposons, thus

suggesting a potential role of MuhMITEs as modulators of genome evolution of

Musa.

Keywords Musa acuminata, Musa balbisiana, genome assembly, BAC, hAT transposons, FISH

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 3

Introduction Autonomous transposable elements (TEs) and their non-autonomous derivatives

represent a large fraction of the DNA in most plant genomes (Heslop-Harrison and

Schwarzacher 2011). Their mobility, amplification and decay provide an important

source of genetic variation with respect to modifying gene expression (Capy et al.

1998) or co-amplification along with functional sequences (Alix and Heslop-Harrison

2009). Therefore, TEs can be regarded as important regulators of genome size,

composition and function (Bennetzen 2005).

The ‘cut and paste’ excision and genomic reintegration of DNA-transposons

is enabled by an element-encoded transposase (Finnegan 1989). The superfamilies of

DNA-transposons are mainly categorized by sequence similarities and structural

conservation of domains of the transposase genes, terminal inverted repeats (TIRs)

and target site duplications (TSDs), the latter occurring as a consequence of breakage,

insertion and repair of the DNA at the junctions between the host genomic DNA and

the mobile sequence. In plants, conservation of TIRs, TSDs and catalytic transposase

domains define the six superfamilies hAT, Tc1/mariner, CACTA, PIF/Harbinger,

Mutator and P (Wicker et al. 2007). Non-autonomous derivatives of DNA

transposons, designated as miniature inverted repeat transposable elements (MITEs),

show TIRs and TSDs, but lack transposase activity due to a partial or complete

deletion of the transposase gene. MITE mobilization is mediated by the recognition of

the TIRs by the transposase of related autonomous founder elements (Yang et al.

2006). In contrast to the relatively low copy number of the autonomous transposons,

MITE families like stowaway, tourist or hATpin, are often highly amplified within

plant genomes (Feschotte et al. 2002; Menzel et al. 2012).

Transposons of the hAT superfamily are characterized by an 8 bp TSD, and

short variable TIRs of 5 bp to 27 bp flanking a single gene coding for a transposase

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 4

harbouring the class II-specific DDE amino acid motif at its catalytic core (Zhou et al.

2004). Autonomous transposons and MITEs of the hAT superfamily are widespread

among plants, but their abundance, amplification, and diversification differ

extensively between taxa (Zhang and Wessler 2004; Holligan et al. 2006; Benjak et

al. 2008; Du et al. 2010; Cavallini et al. 2010; Menzel et al. 2011). Therefore, the

comparison of genomic hAT fractions across different species will enable the

identification of principles underlying transposable element activity and

diversification, and their contribution to genome evolution.

The genus Musa, a non-graminaceous monocot in the Musaceae familiy, is now

divided into the sections Musa (x=11) and Callimusa (x=10/9/7) (Häkkinen 2013).

Section Musa includes the important tropical crops banana and plantains with diploid,

triploid and tetraploid hybrids. The majority of the cultivated bananas result from

inter- and intraspecific crosses between the two diploid wild species Musa acuminata

(A genome) and Musa balbisiana (B genome). The genome size (1C) of M.

acuminata was estimated at 523 Mbp (D’Hont et al. 2012), while the M. balbisiana

genome is 550 Mbp in size (Heslop-Harrison and Schwarzacher 2007). The reference

genome sequences of Musa acuminata ssp. malaccensis and M. balbisiana have

recently been completed (D'Hont et al. 2012; Davey et al. 2013), and show the

presence of yet unclassified hAT transposable elements.

In the present work, we analyzed in detail the abundance, diversity and chromosomal

localization of the M. acuminata hAT transposon superfamily based on the genome

assembly (Droc et al. 2013) and of M. acuminata BACs. In particular the comparison

of homoeologous genomic regions from M. acuminata and M. balbisiana for both

Musa species demonstrates a genome-specific mobility of hAT MITEs.

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 5

Materials and methods

Plant material and DNA preparation

Plants of Musa acuminata (2n=2x=22, AA genome constitution) and the triploid

cultivar ‘Giant Cavendish’ (2n=3x=33, AAA), Musa balbisiana (2n=2x=22, BB),

hybrids (eg Musa ‘Obino l‘Ewai’, 2n=3x=33, AAB), Musa textilis, Musa velutina and

Musa ornata as well as Zea mays, Glycine max, Beta vulgaris and Arabidopsis

thaliana were grown under greenhouse conditions. Genomic DNA was isolated from

young leaves using the CTAB (cetyltrimethylammonium bromide) protocol (Saghai-

Maroof et al. 1984). DNA of Pinus elliotii ssp. elliotii was provided by RL Doudrick

(USDA Southern Research Station, Ashville, USA).

Sequence analyses

Musa hAT transposons were identified in silico in the M. acuminata ‘DH Pahang’

reference genome assembly comprising of 473 Mbp. Additionally analyzed were 5.17

Mbp of 38 M. acuminata BAC sequences and 12 BACs of 1.79 Mbp from the M.

balbisiana B genome provided by the Global Musa Genomics Consortium

(http://www.musagenomics.org/), and 13442 ‘DH Pahang’-specific MITE entries of

the P-MITE database (Chen et al. 2014).

For the identification of Musa hAT transposases in the reference assembly and in

BAC sequences, we applied a Hidden Markov Model (HMM)-based approach using

the HMMER3 packages hmmbuild and hmmsearch (Eddy 1998,) to build HMMs and

to query local databases. A hAT-specific HMM was built from representative plant

hAT transposases like Ac (P08770), Mx (AAV32822) from Zea mays, Thelma13

(AAP59878) from Silene latifolia, Tam3 (BAA28817) from Antirrhinum majus, Tag1

(AAC25101) from Arabidopsis thaliana and Tip100 (BAA36225) from Ipomoea

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 6

purpurea. Prior to the HMM search, the reference genome assembly has been

partitioned in 100 kb fragments with 2 kb overlap and translated to all six reading

frames. This thereby created database was queried with the hAT HMM to retrieve

amino acid sequences hAT transposase regions. In order to identify the complete

transposon with TIRs and TSDs, the corresponding nucleotide sequence along with an

upstream and downstream region of 5000 nt, respectively, has been extracted and

subjected to TIRfinder (Gambin et al. 2013) with the following parameters: TSD

mask for all searches: 'NNNNNNNN'; TIR mask for MuhAT I:

'CANNNNTTNNNNNAG'; TIR mask for MuhAT II: 'CANNGTNNNNNNTNNCG'

and 'CAAGGT'.

Sequences of autonomous and non-autonomous transposons as well as homoeologous

genomic BAC regions were aligned with the MAFFT and MUSCLE options of

Geneious 6.1 (Available from http://www.geneious.com), respectively. Open reading

frames of autonomous hAT transposons were identified using the GeneWise algorithm

(Birney et al. 2004). Neighbour-joining consensus trees, including 1,000 bootstrap

replicates, revealing the divergence of hAT transposases and the diversity of Musa

hAT transposons were generated by Geneious 6.1. Sequence logos of hAT-specific

sequence motifs were created by WebLogo v2.8.2 (Crooks et al. 2004). Sequence

information on the hAT transposons identified in the M. acuminata genome assembly

and on Musa BACs is given by the Supplementary file 1 and a Supplementary table 1,

respectively.

Gene association of autonomous hAT transposons and corresponding MITE families

was analyzed with a Python script comparing the genomic coordinates of transposons

and genes based on the GFF annotation file for Musa acuminata DH-Pahang (v1)

(http://banana-genome.cirad.fr). Only transposon copies on pseudochromosomes 1 to

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 7

11 were considered. For each transposon family, the portion of intronic and exonic

copies was determined as well as the distances of the intergenic copies to the nearest

gene.

Southern hybridization and probe amplification

For Southern hybridizations, 10 µg genomic DNA was restricted with different

enzymes, separated on 1.2% agarose gels, and transferred onto nylon membranes (Pall

Biodyne B, Pall, Germany) using alkaline transfer. Southern hybridizations using

radioactively labelled probes were performed using standard protocols (Sambrook et

al., 1989). Filters were hybridized at 60°C and washed at 60°C in 2 x SSC/0.1 % SDS

and 1 x SSC/0.1 % SDS for 10 min each. Signals were detected by autoradiography.

Hybridization probes were generated with primers designed from internal and

flanking regions of MuhAT transposons localized on Musa acuminata BACs, and

used for the amplification of probes with 50-75 ng DNA of BAC DNA. PCR was

performed in 50 µl reaction volume with 2.0 mM MgCl2, 200 mM dNTP, 1 U Taq

polymerase, 10x reaction buffer (Kapa Biosystems), and 10 pmoles of forward and

reverse primers. A MuhAT I-specific transposase fragment of 200 bp was amplified

with the primer pair 5’-ATTTTATAGTGAATGGTGG-3’ and 5’-

TAAAAAAGATCGTGCAATCG-3’. A MuhMITE I probe was generated with the

primers 5’-CAGTGATTGAAAAAGGTGCTC-3’ and 5’-

AGTGATTTAAAAAGCGCTAGG-3’. PCR conditions were 5 min denaturation at

94 ˚C, 34 cycles of 1 min denaturation at 94 ˚C, 1 min at the primer-specific annealing

temperature, 1 min elongation at 72 ˚C, finalized by 5 minutes extension at 72 ˚C.

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 8

Fluorescent in situ hybridization

The meristem of young roots of Musa plants was used for the preparation of mitotic

chromosomes. After fixation in ethanol:acetic acid (3:1), the material was macerated

in an enzyme mixture consisting of 2% (w/v) cytohelicase from Helix pomatia

(Sigma), 4% (w/v) cellulase from Aspergillus niger (Sigma), 2% (w/v) cellulase

Onozuka-R10 from Aspergillus niger (Serva), 0.2% (w/v) pectolyase from

Aspergillus japonicus (Fluka) and 20% (v/v) pectinase from A. niger. After washing

in fixative, the chromosomes were spread on slides as described by Schwarzacher and

Heslop-Harrison (2000) with modifications (Desel 2002). Probes were labelled with

biotin-11-dUTP or digoxigenin-16-dUTP by PCR before hybridization to

chromosomes (Schwarzacher & Heslop-Harrison 2000). Preparations were

counterstained with DAPI (4´, 6´-diamidino-2-phenylindole) and mounted in antifade

solution (CitiFluor).

Slides were photographed with Zeiss Axioplan epi-fluorescence microscopes with

appropriate filters. Micrographs were acquired with the Applied Spectral Imaging

camera ASI BV300-20A or a Jenoptic ProgRes C12 camera and processed in Adobe

Photoshop CS5 using only contrast optimization, Gaussian and channel overlay

functions affecting the whole image equally.

Results

Identification of Musa acuminata hAT transposases

For the identification of hAT transposases, the 473 Mbp genome assembly of Musa

acuminata ‘DH Pahang’ and 38 M. acuminata BAC sequences, comprising of 5.17

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 9

Mbp, were analyzed by a Hidden Markov Model (HMM) built from plant hAT

transposase families from Zea mays, Silene latifolia, Antirrhinum majus, Arabidopsis

thaliana, Vitis vinifera and Ipomoea purpurea. The HMM analysis identified 274

transposase sequences, 114 of which span the five conserved hAT-specific amino acid

blocks B-F (Rubin et al. 2001). The relation of these 114 Musa-specific hAT

(MuhAT) transposases to representative plant hAT transposases is shown in a

neighbour joining (NJ) tree (Fig. 1). 61 MuhAT I transposases and 47 MuhAT II

transposases form two well-supported MuhAT families, which are clearly separated

from the third family MuhAT III consisting only of six members. The MuhAT III

family can be regarded as Ac/Tam3-specific due to the grouping to corresponding

members from different monocotyledoneous and dicotyledoneous plant species.

MuhAT I members show the closest relationship to the Vitis vinifera element Hatvine-

1 (Benjak et al. 2008), while members of the family MuhAT II are more closely

related to the hAT transposon Bg from maize (Hartings et al. 1991).

Classification of autonomous and non-autonomous MuhAT transposons

To identify complete MuhAT transposons, genomic sequences flanking the coding

regions of the MuhAT transposases in the reference assembly and on BACs of M.

acuminata were in silico scanned for terminal inverted repeat (TIR)- and target site

duplication (TSD)-motifs.

For the MuhAT I family, 49 complete transposons including TIRs and hAT-specific 8

bp TSDs were localized in the genome assembly, while two transposons are located

on M. acuminata BACs. Autonomous MuhAT I transposons range in size from 2206

bp to 6814 bp and are terminated by TIRs of 15 bp. Moreover, two MuhAT I elements

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 10

were shown to contain an intact transposase open reading frame (ORF) with three

short introns at conserved positions, as exampled for MuhAT I-14 (Fig. 2). In silico

analyses of genomic regions adjacent to MuhAT II transposase ORFs revealed 19

autonomous elements, one of which located on a BAC, ranging in size from 4203 bp

to 5644 bp with diverging TIR lengths from 6 bp to 12 bp and flanking 8 bp TSDs.

Two MuhAT II transposons harbour an intact transposase ORF with two conserved

intron positions (MuhAT II-6 in Fig. 2). In contrast to MuhAT I and MuhAT II, no

TIRs and flanking TSD motifs were identified for the six members forming the

MuhAT III family. Nevertheless, translation of the putative ORF regions of MuhAT

III-3 resulted in an intact and continuous transposase ORF of 2148 bp, including start

and stop codons (Fig. 2).

Musa-specific hAT MITEs (MuhMITEs) were identified by queries of 300 bp,

designed from 150 bp of both the 5’-and 3’-untranslated regions including TIR motifs

from autonomous MuhAT I and MuhAT II transposons. These queries were used for

searches within the M. acuminata genome assembly, 38 M. acuminata BACs and

13442 ‘DH Pahang’-specific MITE entries of the P-MITE database (Chen et al.

2014).

Using the MuhAT I-specific query, 116 MITEs ranging in size from 646 bp to 1072

bp were located in the reference assembly (104 MuhMITEs I, Supplementary File 1)

and on M. acuminata BACs (11 MuhMITEs I, Supplementary Table 1). Nucleotide

alignments revealed that MuhMITEs I are exclusively derived from the non-coding

regions of the autonomous MuhAT I transposons (Fig. 3a). For a detailed analysis of

terminal structures specific for the MuhAT I family, the lengths and the relative

nucleic acid residue frequencies of 5’-TIR and 5’-TSD were depicted by sequence

logos. The TIR logo generated from 159 5’-TIR motifs with at most one mismatch

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 11

(Fig. 3b) shows a uniform length of 15 bp starting with the hAT-specific consensus

(T/C)A(A/G)NG (Rubin et al. 2001). The sequence logo of 140 TSDs, allowing only

one mismatch between 5’- and 3’-TSD, indicates a slight preference for an 8 bp target

site containing a 6 bp palindromic motif consisting of T and A residues.

Mining for MuhAT II-specific non-autonomous elements revealed 1701 MuhMITEs

II in the ‘DH Pahang’ reference genome assembly (Supplementary File 1) and 24

MuhMITEs II on M. acuminata BACs (Supplementary Table 1), with a size

distribution from 163 bp to 561 bp. The 5’- and 3’ terminal regions of the 1722

MuhMITEs II, in different extent, are derived from autonomous MuhAT II

transposons (Fig. 4a). In contrast to non-autonomous MuhAT I elements, MuhMITEs

II contain internal sequences from 27 bp to 518 bp in size with no homology to the

transposase ORF of autonomous MuhAT II transposons. A diversity analysis revealed

13 classes of internal regions, as shown by a cladogram (Fig. 4b), which were

regarded as transduplicated sequences acquired from unknown genomic regions of the

Musa genome, as no significant sequence homologies were detected in BLASTN

searches.

The TIRs of the MuhAT II family differ in size from 6 bp to 12 bp, with a strong

conservation of the first five nucleotides containing the hAT-specific TIR consensus

(T/C)A(A/G)NG (Rubin et al. 2001), as shown by a sequence logo derived from 1703

5’-TIR motifs (Fig. 4c). Notably, the TIR lengths differ even between MITEs

containing similar transduplicated genomic sequences.

In contrast to the integration preference of MuhAT I elements to target sites with a

slight conservation at positions 2 and 7, the sequence logo of 902 MuhAT II-specific

5’-TSD motifs shows a significant preference for 8 bp target sites with a strong

conservation of T residues at the nucleotide positions 2 and 3, and A residues at

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 12

positions 6 and 7 (Fig. 4d). Overall, the sequence logo suggests a palindromic target

site specific for the MuhAT II family.

Although the identification of MuhAT III-specific MITEs was hampered by the lack

of TIRs flanking the corresponding six transposase ORFs, genomic regions terminated

by putative hAT-specific TIR-motifs were fused with their reverse complement and

used as queries for BLASTN searches. However, no MuhAT III-derived MITEs were

present in the M. acuminata data set.

In total, 9.31 Mbp of complete autonomous MuhAT transposons and MuhMITEs as

well as fragmented MuhAT-specific transposase ORFs and MITEs sequences,

respectively, were identified. Thus, MuhAT transposons form approximately 1.97%

of the M. acuminata ‘DH Pahang’ genome.

BAC-specific organization and mobility of MuhAT transposons

The genomic organization of MuhMITEs was examined on less complex physical

entities of M. acuminata and M. balbisiana BACs, rather than in the reference

genome assembly which might be misassembled in sequence regions containing

repeated fractions of homologous MuhMITE sequences. From 38 M. acuminata

BACs analyzed, nine contain MuhMITEs I, while MuhMITEs II were localized on 12

BACS. Out of those, collocalization of MuhMITEs I and MuhMITEs II occurs on six

BAC sequences. We also noted a serial localization of seven MuhMITEs II on the M.

acuminata BAC 2G17 (AC226036), and a nested organization of three MuhMITEs

was detected on the M. acuminata BAC MA4_88K20 (AC226051) (Supplementary

Fig. 1a and b).

The mobility of selected MuhMITEs was revealed by the comparison of a

homoeologous region of 101.7 kb with an 83% nucleotide identity between the BAC

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 13

sequences MA4_82I11 (AC186955.1) from M. acuminata and MBP_81C12

(AC186754) from M. balbisiana (Fig. 5). Within this homoeologous region, the M.

acuminata BAC MA4_82I11 harbours the insertions of the 265 bp MuhMITE II-14

and the 866 bp MuhMITE I-5 flanked by the 8 bp TSDs 5’- TCCCTGAG -3’ and 5’-

GTG(C/T)TA(A/G)C-3’, respectively (Fig. 5a and b). On the M. balbisina BAC

sequence, insertions of MuhMITE II-1723 (1284 bp) and MuhMITE II-1724 (516 bp)

with the flanking TSDs 5’-GTTGCAAC-3’ and 5’-TTCAAATG-3’, respectively,

were identified (Fig. 5c and d). The sequence alignments (Fig. 5) show the absence of

these four MuhMITEs in the corresponding homoeologeous BAC region, while the

lack of these MuhMITEs on both homoeologous chromosomes of the M.acuminata

and M. balbisiana genomes, respectively, was proven by PCR with flanking primers

(Supplementary Fig. 1). In particular, the finding of a single 8 bp target site motif in

the sequences lacking MuhMITEs suggests species-specific integrations of the MITEs

rather than excisions, which would be marked by an excision footprint consisting of

two TSD motifs (Fig. 5a-d).

Genomic organization and chromosomal localization of Musa hAT transposons

The genus Musa is divided into the sections Musa (x=11) and Callimusa (x=10/9/7)

(Häkkinen 2013). The genomic organization of autonomous MuhAT transposons and

MuhMITEs in seven Musa genomes from the sections Musa and Callimusa was

compared by Southern hybridization (Fig. 6). Probing of a 200 bp fragment coding for

the conserved hAT transposase-specific dimerization domain indicates the specificity

of autonomous MuhAT transposons for the genus Musa (Fig. 6a, lanes 1-7). Species

and hybrids from the section Musa (Fig. 6a, lanes 1-4, 6 and 7), and M. textilis from

the section Callimusa (Fig. 6a, lane 5) showed patterns of multiple conserved

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 14

fragments from 0.5 kb to 10 kb, which were different between the sections. Notably,

the section Musa species M. velutina revealed a unique strong 3 kb fragment

suggesting a species-specific diversification of MuhAT transposon variants (Fig. 6a,

lane 6). The coding region of MuhAT-specific dimerization domains was not

detectable in representative angiosperm and gymnosperm species (Fig. 6a, lanes 8-

12).

Hybridization with a MuhMITE I probe to genomic Musa DNA showed a strong

smear over the whole range of separation (Fig. 6b, lanes 1-7). A similar distribution

pattern, but with different intensities is present in the angiosperm and gymnosperm

species tested (Fig. 6b, lanes 8-12). The detection of hybridization signals in outgroup

species, in particular in maize (Fig. 6b, lane 8), is most likely the result of cross-

hybridizations to a MuhMITE I-specific (TAACCC)4-6 microsatellite motif, as

confirmed by at least 500 corresponding hits from BLASTN searches in the maize

genome database (http://www.maizegdb.org/), rather than an indication for the

presence of homologous MuhMITEs in this outgroup genome.

The chromosomal distribution of autonomous and non-autonomous MuhAT

transposons was examined by fluorescent in situ hybridization (FISH) to metaphase

chromosomes of a triploid Musa acuminata (Fig. 7). FISH with the 200 bp fragment

coding for the dimerization domain of the MuhAT I transposase shows hybridization

signals of different intensities on both arms of all chromosomes (Fig. 7a, middle and

right). The signals are predominantly localized in intercalary and subterminal

euchromatic regions.

Similar to autonomous MuhAT I transposons, hybridization signals of a non-

autonomous MuhMITEs I probe are detectable predominantly in subterminal,

presumably euchromatic positions of all 33 chromosomes (Fig. 7b, middle and right).

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 15

Furthermore, a representative close-up showed weak hybridization along

chromosomes with pronounced stronger signals in distal chromosomal regions (Fig.

7d middle and right). At higher resolution of interphase hybridization, signals were

detected adjacent to brightly DAPI-stained heterochromatic foci (Fig. 7b). However,

consistent with signal distribution along metaphase chromosomes (Fig. 7b), co-

localization with few heterochromatic sites were observed.

The FISH hybridization patterns of MuhAT I and MuhMITE I in euchromatic

chromosomal regions suggest a close association of hAT transposons with genes. in

the genome. Therefore, the insertion preference of the MuhAT transposons and

MuhMITEs localized on the 11 M. acuminata pseudochromosomes was in silico

detected (Fig. 8). The 33 MuhAT I and 12 MuhAT II transposons are predominantly

integrated in exons, introns, or close to gene ORFs with a distance up to 1500 bp. The

6 MuhAT III copies were exclusively found in exonic regions. In contrast,

MuhMITEs I and MuhMITEs II, in a portion of at least 70 %, show an intronic

integration.

Discussion

The comparison of genome sequences has revealed contrasting portions of hAT

transposons in plant species. While relatively small hAT fractions are present in

Arabidopsis, Brassica, sunflower and sugar beet (Zhang and Wessler 2004; Cavallini

et al. 2010; Menzel et al. 2011), hAT elements form the largest DNA transposon

superfamily within the grapevine genome (Benjak et al. 2008). According to the

initial analysis of the reference genome sequence of the doubled-haploid M.

acuminata genotype ‘DH-Pahang’ (2n=22), approximately 1.24% of hAT elements

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 16

form the largest portion of DNA transposons (D'Hont et al. 2012, supplementary

information). Our in silico analyses, in particular the detection of a substantial

representation of MuhMITEs, with 9.31 Mbp consisting of complete and incomplete

autonomous as well as non-autonomous MuhAT transposons has increased the hAT

portion to 1.97%. Thus, although the ‘DH Pahang’ reference assembly covers only

90% of the genome and a detailed analysis regarding the abundance of other DNA

transposon superfamilies has not been performed yet, our data suggest the hAT

superfamily as most likely the largest DNA transposon superfamily of the M.

acuminata genome.

The hAT transposon superfamily is ancient and conserved throughout the plant

kingdom. Nevertheless, the number and diversity of hAT families is still unclear.

Members of this superfamily are predominantly grouped into the Ac/Tam3 family,

while transposons like Bg and Tag1 from maize and Arabidopsis, respectively, are

hitherto regarded as members of minor hAT families (Rubin et al. 2001; Xu and

Dooner 2005). A comprehensive analysis of hAT transposons in plant genomes has

been performed only in grapevine and sugar beet so far (Benjak et al. 2008; Menzel et

al. 2011). In grapevine, two Hatvine families were classified as Ac/Tam3-like, while

the majority of eight Hatvine families were grouped together with Bg and Tag1. In

contrast, BvhAT transposons of sugar beet exclusively consist of Ac/Tam3-like hAT

members. Within the M. acuminata genome, hAT transposases can be classified in

three well supported MuhAT families. Similarly to grapevine, the Ac/Tam3-like

MuhAT III family with only six copies is underrepresented, while the Bg-like MuhAT

II family and the MuhAT I family, which is closely related to Hatvine-1, form the

dominant DNA transposon fraction of the ‘DH Pahang’ genome. Thus, M. acuminata

is a prime example of a plant genome with a remarkable amount of hAT transposons

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 17

organized in families different from Ac/Tam3 family. On the other hand, the low copy

MuhAT III transposons, which are truncated and also lack MITE derivatives, or the

absence of Bg-like transposons in the sugar beet reference assembly (Menzel et al.

2011), indicates genome–specific expansions, reductions or even eliminations of hAT

transposon families in evolutionary time scales.

Non-autonomous miniature inverted-repeat transposable elements (MITEs) derived

from DNA transposon superfamilies are present in a wide variety of plant species

(Feschotte et al. 2003; Zhang et al. 2004; Macas et al. 2005; Moreno-Vázquez et al.

2005; Benjak et al. 2009; Menzel et al. 2011). Compared to the relatively low copy

number of the autonomous founder elements, MITE derivatives of some DNA

transposon superfamilies are highly amplified within plant genomes, as shown for

Mariner-derived stowaway MITEs (Feschotte et al. 2003; Menzel et al. 2006). In

contrast, low numbers of hAT-specific MITEs in plant genomes have been identified

in rice, grapevine and sugar beet (Fujino et al. 2005; Moreno-Vázquez et al. 2005;

Moon et al. 2006; Benjak et al. 2009; Huang et al. 2009; Menzel et al. 2011). In M.

acuminata, only few unclassified MITEs have been detected on BAC end sequences

as well as in the ‘DH Pahang’ reference genome assembly so far (Cheung and Town

2000; D’Hont et al. 2012). Our analysis identified in total 1722 MuhAT I- and

MuhAT II-specific MuhMITEs lacking transposase coding regions. Thus, hAT-

specific MITEs are highly amplified in Musa genomes, as also shown by Southern

hybridization. This amplification of MuhMITEs is most likely mediated by active

autonomous members of the MuhAT I and MuhAT II families detected in silico.

Beside amplification, MITE of diverse DNA transposon superfamilies are able to

incorporate unrelated genomic sequences, as reported for Stowaway MITEs in the

sugar beet genome and for PIF/Harbinger-derived MITEs in grapevine (Menzel et al.

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 18

2006, Benjak et al. 2009). The acquisition of genomic regions, referred to as

transduplication (Juretic et al. 2005), by MuhMITEs II and the subsequent

amplification of transduplicated sequences resulting in the formation of diverse

MuhMITE II subfamilies might indicate a general mechanism of MITE diversification

in plant genomes.

The comparison of the DNA transposon, and in particular hAT transposon content in

the genomes of M. acuminata and M. balbisiana showed an almost similar percentage

for both species (Davey et al. 2013, Novak et al. 2014). Moreover, Southern analyses

presented in this work indicate a similar abundance and similar families of hAT

transposons and MITE derivatives in all Musa species. The hybridization also

supports the taxonomic revision of the genus into two sections (Häkkinen 2013).

Nevertheless, a comparison of homoeologous BAC regions of M. acuminata and M.

balbisiana suggests significant differences in the MuhMITE landscape of both

species. As MITE mobilization depends on the transposase activity of related

autonomous elements (Feschotte et al. 2002), the differential MuhMITE mobilization

of both Musa genomes is most likely the result of a genome-specific activity of

MuhAT I and MuhAT II transposases.

FISH on M. acuminata metaphase chromosomes clearly demonstrates the insertion

preference of the autonomous MuhAT transposons and derived MITEs in gene-rich

euchromatic regions, as also observed for sugar beet hAT transposons (Menzel et al.

2011). The differences in signal intensities might reflect the chromosome-specific

densities and distributions of genes on the 11 M. acuminata chromosomes (D’Hont et

al. 2012), most likely in combination with local clustering and nested organization in

particular of MuhAT MITEs as observed in M. acuminata BACs.

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 19

In plants and animals, the genomic integration of hAT transposons is obviously

influenced by the structure of the 8 bp target site. Ds elements in maize exhibit no

specificity for a certain target site nucleotide composition, but indicate a target site-

specific increase in GC-content inducing a hydrogen bonding signature, which

possibly mediates the preferential recognition by the transposase (Vollbrecht et al.

2010). In contrast, transposons of the MuhAT I and especially the MuhAT II families

in M. acuminata prefer integration sites containing T residues at positions 2 and 3 and

A residues at positions 6 and 7, respectively. A similar nucleotide dominance within

TSD motifs is detectable for the hAT transposon families in sugar beet and Ds

elements in Arabidopsis (Menzel et al. 2011; Kuromori et al. 2004). Moreover, the

target sites flanking hAT elements in 12 Drosophila species exhibit a strong

specificity requiring the nucleotide residues T and A at positions 2 and 7, respectively

(de Freitas Ortiz et al. 2010). Thus, hAT transposon families might prefer integration

sites with nucleotide positions conserved in plant and animal genomes, rather than

integrate in 8 bp target sites with random nucleotide compositions. The specific

integration of MITEs within genes, which was also shown for MuhMITEs, can affect

gene expression by providing new splicing sites, transcription start sites, new exons,

and poly(A) sites (Benjak et al. 2009). In particular, the targeting of MITEs to

promoter regions might provide novel regulatory sequences most likely resulting in a

modulation of gene expression (Vollbrecht et al. 2010). Therefore, it is tempting to

speculate that the differing insertion preferences of MuhMITEs I and MuhMITEs II

might reflect a family-specific MuhMITE targeting to distinct genic and intergenic

regions of the Musa genome and, in association with the mobility of MuhMITEs II

transduplicating and amplifying various genomic sequences, suggests a possible role

of these MuhMITEs as modulators of gene evolution in Musa genomes.

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 20

Acknowledgements We are indebted to Nadin Fliegner and Ines Walter for excellent

technical assistance. We thank the DAAD-British Council ARC Academic Research

Collaboration project for support of collaboration. We thank the Centre de Cooperation

Internationale en Recherche Agronomique pour le Developpement (CIRAD), Montpellier

(France), and the Musa Genome Resource Centre (MGRC) at the Institute of Experimental

Botany (IEB), Olomouc (Czech Republic) for the production, storage and distribution of

Musa DNA samples used here. We thank the TU Dresden Center for Information Services

and High Performance Computing (ZIH) for computer time allocations.

Conflict of interest The authors Gerhard Menzel, Tony Heitkam, Kathrin M. Seibt, Faisal Nouroz, Manuela Müller-Stoermer, John S. Heslop-Harrison and Thomas Schmidt declare that they have no conflict of interest.

References

Alix K, Heslop-Harrison JS (2004) The diversity of retroelements in diploid and allotetraploid

Brassica species. Plant Mol Biol 54:895-909

Altinkut A, Kotseruba V, Kirzhner VM, Nevo E, Raskina O, Belyayev A (2006) Ac-like

transposons in populations of wild diploid Triticeae species: comparative analysis of

chromosomal distribution. Chromosome Res 14:307-317

Bartos J, Alkhimova O, Dolezelová M, De Langhe E, Dolezel J (2005) Nuclear genome size

and genomic distribution of ribosomal DNA in Musa and Ensete (Musaceae):

taxonomic implications. Cytogenet Genome Res 109:50-57

Benjak A, Forneck A, Casacuberta JM (2008) Genome-wide analysis of the "cut-and-paste"

transposons of grapevine. PLoS One 3:e3107

Benjak A, Boué S, Forneck A, Casacuberta JM (2009) Recent amplification and impact of

MITEs on the genome of grapevine (Vitis vinifera L.). Genome Biol Evol 20:75-84

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 21

Bennetzen JL (2005) Transposable elements, gene creation and genome rearrangement in

flowering plants. Curr Opin Genet Dev 15:621-627

Birney E, Clamp M, Durbin R (2004) GeneWise and Genomewise. Genome Res 14:988–995

Capy P, Bazin C, Higuet D, Langin T (1998) Dynamics and evolution of transposable

elements. Springer, Austin

Cavallini A, Natali L, Zuccolo A, Giordani T, Jurman I, Ferrillo V, Vitacolonna N, Sarri V,

Cattonaro F, Ceccarelli M, Cionini PG, Morgante M (2010) Analysis of transposons

and repeat composition of the sunflower (Helianthus annuus L.) genome. Theor Appl

Genet 120:491-508

Chen J, Hu Q, Zhang Y, Lu C, Kuang H (2014) P-MITE: a database for plant miniature

inverted-repeat transposable elements. Nucleic Acids Res 42:D1176-D1181

Cheung F, Town CD (2007) A BAC end view of the Musa acuminata genome. BMC Plant

Biol 7:29

Crooks GE, Hon G, Chandonia JM, Brenner SE (2004) WebLogo: A sequence logo

generator. Genome Res 14:1188-1190

Davey MW, Gudimella R, Harikrishna JA, Sin LW, Khalid N, Keulemans W (2013) A draft

Musa balbisiana genome sequence for molecular genetics in polyploid, inter- and intra-

specific Musa hybrids. BMC Genomics 14:683

de Freitas Ortiz M, Lorenzatto KR, Corrêa BR, Loreto EL (2010) hAT transposable elements

and their derivatives: an analysis in the 12 Drosophila genomes. Genetica

Desel C (2002) Chromosomale Lokalisierung von repetitiven und unikalen DNA-Sequenzen

durch Fluoreszenz- in situ- Hybridisierung in der Genomanalyse bei Beta-Arten.

Dissertation, University of Kiel

138:649-655

D’Hont A, Denoeud F, Aury JM, Baurens FC, Carreel F, Garsmeur O, Noel B, Bocs S, Droc

G, Rouard M, Da Silva C, Jabbari K, Cardi C, Poulain J, Souquet M, Labadie K, Jourda

C, Lengellé J, Rodier-Goud M, Alberti A, Bernard M, Correa M, Ayyampalayam S,

Mckain MR, Leebens-Mack J, Burgess D, Freeling M, Mbéguié-A-Mbéguié D,

Chabannes M, Wicker T, Panaud O, Barbosa J, Hribova E, Heslop-Harrison P, Habas

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 22

R, Rivallan R, Francois P, Poiron C, Kilian A, Burthia D, Jenny C, Bakry F, Brown S,

Guignon V, Kema G, Dita M, Waalwijk C, Joseph S, Dievart A, Jaillon O, Leclercq J,

Argout X, Lyons E, Almeida A, Jeridi M, Dolezel J, Roux N, Risterucci AM,

Weissenbach J, Ruiz M, Glaszmann JC, Quétier F, Yahiaoui N, Wincker P (2012) The

banana (Musa acuminata) genome and the evolution of monocotyledonous plants.

Nature 488:213-217

Droc G, Lariviere D, Guignon V, Yahiaoui, N., This, D., Garsmeur, O., Dereeper, A.,

Hamelin, C., Argout, X., Dufayard, J.-F., Lengelle, J., Baurens, F.-C., Cenci, A.,

Pitollat, B., D'Hont, A., Ruiz, M., Rouard, M., Bocs, S (2013) The Banana Genome

Hub. Database doi:10.1093/database/bat035

Du J, Grant D, Tian Z, Nelson RT, Zhu L, Shoemaker RC, Ma J (2010) SoyTEdb: a

comprehensive database of transposable elements in the soybean genome. BMC

Genomics 11:113

Eddy SR (1998) Profile hidden Markov models. Bioinformatics 14:755-763

Feschotte C, Jiang N, Wessler SR (2002) Plant transposable elements: Where genetics meets

genomics. Nat Rev Genet 3:

Feschotte C, Swamy L, Wessler SR (2003)

329–341

Genome-wide analysis of mariner-like

transposable elements in rice reveals complex relationships with stowaway miniature

inverted repeat transposable elements (MITEs). Genetics

Finnegan DJ (1989) Eukaryotic transposable elements and genome evolution. Trends Genet

5:103-107

163:747-758

Fujino K, Sekiguchi H, Kiguchi T (2005) Identification of an active transposon in intact rice

plants. Mol Genet Genomics 273:150–157

Fujino K, Matsuda Y, Sekiguchi H (2009) Transcriptional activity of rice autonomous

transposable element Dart. J Plant Physiol 166:1537-1543

Gambin T, Startek M, Walczak K, Paszek J, Grzebelus D, Gambin A (2013). TIRfinder: A

Web Tool for Mining Class II Transposons Carrying Terminal Inverted Repeats. Evol

Bioinform Online 9:17-27

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 23

Häkkinen M (2013) Reappraisal of sectional taxonomy in Musa (Musaceae). Taxon 62:809-

813

Han Y, Qin S, Wessler SR (2013) Comparison of class 2 transposable elements at

superfamily resolution reveals conserved and distinct features in cereal grass genomes.

BMC Genomics 14:71

Hartings H, Rossi V, Lazzaroni N, Thompson RD, Salamini F, Motto M (1991) Nucleotide

sequence of the Bg transposable element of Zea mays L. Maydica 36:355–359

Heslop-Harrison JS, Schwarzacher T (2007) Domestication, genomics and the future for

Banana. Ann Bot 100:1073–1084

Heslop-Harrison JS, Schwarzacher T (2011) Organisation of the plant genome in

chromosomes. Plant J 66:18-33

Holligan D, Zhang X, Jiang N, Pritham EJ, Wessler SR (2006) The transposable element

landscape of the model legume Lotus japonicus. Genetics 174:2215-2228

Huang J, Zhang K, Shen Y, Huang Z, Li M, Tang D, Gu M, Cheng Z (2009) Identification of

a high frequency transposon induced by tissue culture, nDaiZ, a member of the hAT

family in rice. Genomics 93:274-281

Juretic N, Hoen DR, Huynh ML, Harrison PM, Bureau TE (2005) The evolutionary fate of

MULE-mediated duplications of host gene fragments in rice. Genome Res 15:1292–

1297

Kunze R, Weil CF (2002) The hAT and CACTA superfamily of plant transposons. In Craig

NL (ed) Mobile DNA. ASM Press, Washington, pp 565–610

Kuromori T, Hirayama T, Kiyosue Y, Takabe H, Mizukado S, Sakurai T, Akiyama K,

Kamiya A, Ito T, Shinozaki K (2004) A collection of 11 800 single-copy Ds transposon

insertion lines in Arabidopsis. Plant J 37:897-905

Macas J, Koblízková A, Neumann P (2005) Characterization of Stowaway MITEs in pea

(Pisum sativum L.) and identification of their potential master elements. Genome

48:831-839

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 24

Menzel G, Dechyeva D, Keller H, Lange C, Himmelbauer H, Schmidt T (2006) Mobilization

and evolutionary history of miniature inverted-repeat transposable elements (MITEs) in

Beta vulgaris L. Chromosome Res 14:831-844

Menzel G, Krebs C, Diez M, Holtgräwe D, Weisshaar B, Minoche AE, Dohm JC,

Himmelbauer H, Schmidt T. (2012) Survey of sugar beet (Beta vulgaris L.) hAT

transposons and MITE-like hATpin derivatives. Plant Mol Biol

Moon S

78:393-405

, Jung KH, Lee DE, Jiang WZ, Koh HJ, Heu MH, Lee DS, Suh HS, An G (2006)

Identification of active transposon dTok, a member of the hAT family, in rice. Plant

Cell Physiol 47:1473-1483

Moreno-Vázquez S, Ning J, Meyers BC (2005) hATpin, a family of MITE-like hAT mobile

elements conserved in diverse plant species that forms highly stable secondary

structures. Plant Mol Biol

Novák P, Hřibová E, Neumann P, Koblížková A, Doležel J, Macas J. (2014)

58:869-886

Genome-wide

analysis of repeat diversity across the family Musaceae. PLoS One

Rubin E, Lithwick G, Levy AA (2001)

9:e98918

Structure and evolution of the hAT transposon

superfamily. Genetics

Saghai-Maroof MA, Soliman KM, Jorgensen RA, Allard RW (1984) Ribosomal DNA spacer-

length polymorphisms in barley: Mendelian inheritance, chromosomal location, and

population dynamics. Proc Natl Acad Sci USA 81:8014-8018

158:949-957

Sambrook J, Fritsch,EF, Maniatis T (1989) Molecular cloning: A laboratory Manual. Cold

Spring Harbor Laboratory Press, New York

Schwarzacher T, Heslop-Harrison JS (2000) Practical In-situ Hybridization. BIOS Scientific

Publishers, Oxford, p 60

Takagi K, Maekawa M, Tsugane K, Iida S (2010) Transposition and target preferences of an

active nonautonomous DNA transposon nDart1 and its relatives belonging to the hAT

superfamily in rice. Mol Genet Genomics 284:343-355

Vicient CM (2010) Transcriptional activity of transposable elements in maize. BMC

Genomics 11:601

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 25

Vollbrecht E, Duvick J, Schares JP, Ahern KR, Deewatthanawong P, Xu L, Conrad LJ,

Kikuchi K, Kubinec TA, Hall BD, Weeks R, Unger-Wallace E, Muszynski M, Brendel

VP, Brutnell TP (2010) Genome-wide distribution of transposed Dissociation elements

in maize. Plant Cell 22:1667-1685

Wicker T, Sabot F, Hua-Van A, Bennetzen JL, Capy P, Chalhoub B, Flavell A, Leroy P,

Morgante M, Panaud O, Paux E, SanMiguel P, Schulman AH (2007) A unified

classification system for eukaryotic transposable elements. Nat Rev Genet 8:973-982

Xu Z, Dooner HK (2005) Mx-rMx, a family of interacting transposons in the growing hAT

superfamily of maize. Plant Cell 17:375-388

Yang G, Weil CF, Wessler SR (2006) A rice Tc1/mariner-like element transposes in yeast.

Plant Cell

Zhang X

18:2469-2478

, Wessler SR (2004) Genome-wide comparative analysis of the transposable

elements in the related species Arabidopsis thaliana and Brassica oleracea. Proc Natl

Acad Sci USA 101:5589-5594

Zhang X, Jiang N, Feschotte C, Wessler SR (2004) PIF- and Pong-like transposable elements:

distribution, evolution and relationship with Tourist-like miniature inverted-repeat

transposable elements. Genetics 166:971-986

Zhou L, Mitra R, Atkinson PW, Hickman AB, Dyda F, Craig NL. (2004) Transposition of

hAT elements links transposable elements and V(D)J recombination. Nature 432:995-

1001

Figure legends

Fig. 1 Neighbour Joining analysis of Musa hAT transposases. The Neighbour Joining

consensus tree, resulting from a MUSCLE alignment of transposase regions spanning

the five conserved hAT-specific amino acid blocks B-F (Rubin et al. 2001), shows the

relationship of 61 MuhAT I, 47 MuhAT II and six MuhAT III transposases (arcs) to

hAT elements from plants such as Arabidopsis thaliana (Tag1 AAC25101, Tag2

AAD24567, Daysleeper AEE77728), Oryza sativa (Tam3 BAA28817, EEC71336),

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 26

Zea mays (Ac P08770, Bg X56877, ACN33643), Vitis vinifera (Hatvine-1, Hatvine-2,

Hatvine-7, Hatvine-9, Hatvine-10; all from Repbase), and Ipomoea purpurea (Tip100

BAA36225). Bootstrap percentages from 1000 repetitions are indicated at the

branches.

Fig. 2 Schematic organization of MuhAT transposons (not to scale). Autonomous

members of the MuhAT I and MuhAT II families with transposase genes are flanked

by terminal inverted repeats (TIR; grey triangle) of 15 bp and 6 bp, respectively. Open

reading frames (dark grey) with start codon (arrow) and stop codon (*) are interrupted

by introns (white bars) of the designated lengths, and code for transposases of 724

amino acids (aa) and 773 aa, respectively. Characteristic conserved motifs including

untranslated regions and TIRs of MuhAT III are not detectable (n.d.).

Fig. 3 Structural features of the MuhAT I transposon family a Scheme of the

relationship revealed for 51 autonomous transposons and 116 non-autonomous

MuhMITEs I. Transposons of the MuhAT I family are composed of TIRs (patterned

arrows) and 5’- and 3’-untranslated regions (white bars) derived from the autonomous

elements harbouring the transposase ORF (black bar). b Nucleotide composition of

the 15 bp 5’-TIR and c Nucleotide composition 8 bp 5’-TSD motif, shown by

sequence logos generated with (n) sequences. Overall heights of nucleotide stacks

indicate the sequence conservation at the given nucleotide positions measured in bits.

Fig. 4 Structural features of the MuhAT II transposon family a Scheme showing the

relationship of 1722 MITEs and 19 autonomous transposons. MuhMITEs II consist of

TIRs (patterned arrows) and 5’- and 3’-untranslated regions (white bars) as well as

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 27

different internal regions (i1-i13) with no relationship to the transposase ORF (black

bar). b Cladogram showing the diversity of the transduplicated MuhMITE II regions

i1-i13. c nucleotide composition of the 15 bp 5’-TIR and d Nucleotide composition 8

bp 5’-TSD motif, shown by sequence logos generated with (n) sequences. Overall

heights of nucleotide stacks indicate the sequence conservation at the given nucleotide

positions measured in bits. The black arrow in c marks the conserved TIR motif of all

TIRs analyzed.

Fig. 5 Genome-specific insertions of MuhMITEs on Musa BAC clones. The

alignment of four regions (a, b, c and d) of the Musa acuminata BAC MA4_82I11

(AC186955) and the Musa balbisiana BAC MBP_81C12 (AC186754) revealing the

presence of MuhMITEs and their absence in the homoeologous genomic regions. The

relative positions of the MuhMITE insertions are indicated at the flanking nucleotide

residues (bp). Target site duplications flanking MuhMITE insertions are given in bold

and underlined, while the corresponding single nucleotide motifs are underlined.

Fig. 6 Abundance of MuhAT transposons in Musa species. Genomic DNA of the

Musa species, from the sections Musa (M) and Callimusa (C), M. acuminata (lane 1),

M. balbisiana (lane 2), Giant Cavendish (lane 3), M. obino l‘Ewai (lane 4), M. textilis

(lane 5), M. velutina (lane 6) and M. ornata (lane 7), as well as DNA of Zea mays

(lane 8), Glycine max (lane 9), Beta vulgaris (lane 10), Arabidopsis thaliana (lane 11)

and the gymosperm Pinus elliotii (lane 12) was restricted with the endonuclease DraI.

The Southern filter was subsequently probed with the M. acuminata-specific sequence

region coding for the dimerization domain of the MuhAT I transposase (a) and a full-

length MuhMITE I (b).

Menzel et al. hAT transposons in Musa genomes 10.1007/s10577-014-9445-5 P. 28

Fig. 7 Chromosomal localization of MuhAT transposases and MuhMITEs I.

Fluorescent in situ hybridization of probes derived from an autonomous MuhAT I

transposon (a) and a MuhMITE I to interphase nuclei (c) and metaphase

chromosomes (b and d) from M. acuminata ‘Giant Cavendish’ (2n=3x=33, AAA

genome). The blue fluorescence of DAPI stained DNA (left in a, b and d, above in c)

shows the morphology of the chromosomes, while hybridization signals are visible as

red fluorescent signals (middle and right overlay in a, b and d, overlay below in c).

The scale bars correspond to 5 µm.

Fig. 8 Association of MuhAT transposons and MuhMITEs with M. acuminata genes.

Element positions of MuhAT I, MuhAT II, MuhAT III, MuhMITE I and MuhMITE II

were compared to gene annotations on the 11 M. acuminata ‘DH-Pahang’

pseudochromosomes to identify exonic (yellow), intronic (green), and intergenic

transposon copies (grey), with specific grey shading according to distance to nearest

gene. The number of transposon copies located on the pseudochromosomes is

provided below the x-axis.

Figure 1 Menzel et al. 2014

100

100

100

97.4

98 53.3

58.2 Hatvine-7

Hatvine-10

Hatvine-2 Tag1

Bg

Tip100 Tam3

Hatvine-9 Ac

Tag2

Hatvine-1

MuhAT III

MuhAT II

MuhAT I

Figure 2 Menzel et al. 2014

MuhAT I-1

MuhAT II-1

MuhAT III-1 n.d. n.d.

3626 bp

4465 bp

2148 bp

*

*

*

724 aa

773 aa

716 aa

27 bp 53 bp 53 bp

116 bp 55 bp

15 bp 15 bp

6 bp 6 bp

Figure 3 Menzel et al. 2014

v n=159 n=140

autonomous MuhAT I (2206-6814 bp)

MuhAT I MITEs (646-1072 bp)

(A)

(B) (C)

v

n=51

n=117

430-786 bp 116-286 bp

5‘-TSD 5‘-TIR

v

5‘-TIRs

n=902

autonomous MuhAT II (4203-5644 bp)

MuhAT II MITEs (163-561 bp)

(A)

(C) (D)

v

n=20

n=1824

Figure 4 Menzel et al. 2014

n=1803

(B)

i1

i2

i3 i4 i5 i6 I7

i8 i9

i10

i11

i12

i13

i1-i13

23-150 bp 20-100 bp

5‘-TSD

27-518 bp

MuMITE I-8

81,400 81,800 82,200 82,600 83,000 83,400 (B) AC226051

MuMITE II-24 MuMITE I-9

MuMITE II-8 MuMITE II-7 MuMITE II-9 MuMITE II-13 MuMITE II-12 MuMITE II-11 MuMITE II-10

2000 3000 4000 5000 6000 7000 8000 9000 10,000 11,000 12,000 13,000 14,000 15,000 16,000 17,000 18,000 19,000 (A) AC226036

Figure Deleted in final MS Menzel et al. 2014

Figure 5 Menzel et al. 2014

MA4_82I11

MBP_81C12

GTCCCCATTCATCCAAAAGATCCCTGAG--------------------TCCCTGAGTATATATCTCTTTTCTAAAT GTCCCCATTCATCCAAAAGA----------------------------TCCCTGAGTATATATCTCTTTTCTGAAT

MuMITE II-15

GTCATTGACTTGTTTTGGTAGTGCTAAC--------------------GTGTTAGCCCCCTAAAAGGCATATGATA GTCATTGACTTGTTTTGGTA----------------------------GTGTTAGCCCCCTAAAAGGCATATGATA

AGCTGGGAGCCAAGAGATGT----------------------------GTTGCAACTAAAGCAGAAGATTAAAGAT AGCTGGGAGCCAAAAGATGTGTTGCAAC--------------------GTTGCAACTAAAGCAGAAGATTAAAGAT

CAGATTGCTCTTGAAATGAG----------------------------TTCAAATGTACATAATTAACCATTATAC CGGATTGCTCTTGAAATGAGTTCAAATG--------------------TTCAAATGTACATAATTAACCATTATAC

18.292 18.565

18.339 18.331

MA4_82I11

MBP_81C12

MA4_82I11

MBP_81C12

MA4_82I11

MBP_81C12

MuMITE I-5

MuMITE II-26

MuMITE II-27

45.164 46.039

45.390 45.381

51.711 51.720

53.135 51.842

79.828 79.837

79.544 79.019

(A)

(B)

(C)

(D)

(B) 1 2 3 4 5 6 7 8 9 10 11 12

kb

0.5

1.0

2.0

3.0

10.0

6.0

(A)

kb

0.5

1.0

2.0

3.0

10.0

6.0

kb

0.5

1.0

2.0

3.0

10.0

6.0

1 2 3 4 5 6 7 8 9 10 11 12

Figure 7 Menzel et al. 2014

Figure 7 Menzel et al. 2014

Figure 8 Menzel et al. 2014

0

10

20

30

40

50

60

70

80

90

100

Porti

on o

f cop

ies

[%]

Exon Intron ≤ 500 bp≤ 1000 bp ≤ 1500 bp > 1500 bp

33 12 6 96 1512