GENETICS | INVESTIGATION

Genetic and Epigenetic Changes in Somatic HybridIntrogression Lines Between Wheat and

Tall WheatgrassShuwei Liu,*,1 Fei Li,*,1,2 Lina Kong,* Yang Sun,* Lumin Qin,* Suiyun Chen,*,3 Haifeng Cui,*

Yinghua Huang,† and Guangmin Xia*,4

*The Key Laboratory of Plant Cell Engineering and Germplasm Innovation, Ministry of Education, School of Life Sciences,Shandong University, Jinan 250100, Peoples’ Republic of China, and †U.S. Department of Agriculture/Agricultural Research Service

Plant Science Research Laboratory, Stillwater, Oklahoma 74075

ABSTRACT Broad phenotypic variations were induced in derivatives of an asymmetric somatic hybridization of bread wheat (Triticumaestivum) and tall wheatgrass (Thinopyrum ponticum Podp); however, how these variations occurred was unknown. We explored thenature of these variations by cytogenetic assays and DNA profiling techniques to characterize six genetically stable somatic introgressionlines. Karyotyping results show the six lines similar to their wheat parent, but GISH analysis identified the presence of a number of shortintrogressed tall wheatgrass chromatin segments. DNA profiling revealed many genetic and epigenetic differences, including sequencesdeletions, altered regulation of gene expression, changed patterns of cytosine methylation, and the reactivation of retrotransposons.Phenotypic variations appear to result from altered repetitive sequences combined with the epigenetic regulation of gene expressionand/or retrotransposon transposition. The extent of genetic and epigenetic variation due to the maintenance of parent wheat cells intissue culture was assessed and shown to be considerably lower than had been induced in the introgression lines. Asymmetric somatichybridization provides appropriate material to explore the nature of the genetic and epigenetic variations induced by genomic shock.

KEYWORDS bread wheat; asymmetric somatic hybridization; introgression line; genomic shock, genetic and epigenetic alteration

WITH the world’s population continuing to increase,achieving a sustainable mode of food production rep-

resents an ever-growing challenge. Plant breeding has nar-rowed the genetic base of many crop species, but as yet hashad little impact on the genetic diversity present in their wildrelatives. In principle, this diversity can be introgressed intocrops via sexual hybridization and subsequent backcrossing.However, in practice, wild-crop manipulation has been se-verely restricted by difficulties in creating the initial sexualhybrid and by sterility issues in the early backcross generations

(Xia 2009). Asymmetric somatic hybridization is a viable alter-native to introgression, especially where wide crosses are notfeasible. It has been successfully exploited in bread wheat totransfer chromosomal segments from a number of related spe-cies (Xia et al. 2003; Xiang et al. 2003, 2004; Cheng et al. 2004;Zhou and Xia 2005; Xia 2009). More importantly, asymmetricsomatic hybridization offers smaller alien chromatin introgres-sion, thereby overcoming a significant problem in wheat sexualhybrids where the Ph1 gene prevents homeologous recombina-tion (Griffiths et al. 2006).

Newly synthesized allopolyploids have provided an oppor-tunity to explore the nature of the genetic and epigeneticchanges triggered by polyploidization (Song et al. 1995;Comai et al. 2000; Ozkan et al. 2001; Shaked et al. 2001;Madlung et al. 2002; Han et al. 2003; Ma et al. 2004; Wanget al. 2004a; Salmon et al. 2005; Tate et al. 2006; Basseneet al. 2010; Xu et al. 2014), although a few allopolyploidswere not accompanied with such changes (Liu et al. 2001).Experimental results indicate that the majority of events arehighly reproducible (Bento et al. 2010). In particular, sequencedeletion is common (Feldman et al. 1997; Liu et al. 1998a,b;

Copyright © 2015 by the Genetics Society of Americadoi: 10.1534/genetics.114.174094Manuscript received December 28, 2014; accepted for publication February 5, 2015;published Early Online February 9, 2015.Supporting information is available online at http://www.genetics.org/lookup/suppl/doi:10.1534/genetics.114.174094/-/DC1.1These authors contributed equally to this work.2Present Address: Cotton Research Center, Shandong Academy of AgriculturalSciences, Jinan 250100, P. R. China.

3Present Address: Plant Science Institute, School of Life Sciences, Yunnan University,Kunming, 650091, P. R. China.

4Corresponding author: School of Life Sciences, Shandong University, Jinan 250100,P. R. China. E-mail: Xiagm@sdu.edu.cn

Genetics, Vol. 199, 1035–1045 April 2015 1035

Ozkan et al. 2001; Shaked et al. 2001; Kashkush et al. 2002;Ma et al. 2004; Ma and Gustafson 2006). Epigenetic modifi-cations, such as changes in the pattern of cytosine methyla-tion, have also been shown to induce changes in geneexpression and activate transposon transcription (Comaiet al. 2000; Shaked et al. 2001; Kashkush et al. 2002,2003). However, these changes induced by “genomic shock”during polyploid synthesis do not represent the changes insomatic introgressions. Allopolyploids represent a combinationof nuclear genomes in a fixed cytoplasmic context, while so-matic hybrids combine both the nuclear and cytoplasmicgenomes within a single cell. The introgression of chromatinsegments by asymmetric somatic hybridization likely occurs vianonhomologous end-joining of fragmented genome piecesrather than by homologous recombination, which would showspecific genetic and epigenetic changes in these materials.Moreover, the epigenetic state of somatic cells tend to be dis-tinct from gametal cells given that the mutagenesis in gametalcells is more tightly controlled to ensure genetic fidelity (Bird1997, 2002). Thus, the variations induced by “somatic geno-mic shock” likely have unique characteristics compared withallopolyploids.

A number of hybrid progenies regenerated from asymmet-ric somatic hybrids [bread wheat cultivar Jinan 177 (JN177)and tall wheatgrass (Thinopyrum ponticum)] have provenphenotypically stable for a number of generations (Xia et al.2003; Chen et al. 2004a; Wang et al. 2004b; Liu and Xia2014). The heterocytoplasmic nature of these hybrid lineswere confirmed, with the chloroplast genomic componentsdominated by wheat, while a few sequences of the chloroplastgenome of tall wheatgrass were also detected in these lines(Chen et al. 2004b). DNA comparison of a well-characterizedset of glutenin proteins among parents and derivatives showsthat all novel glutenin genes in hybrid progenies originatedfrom alien genes of tall wheatgrass and allelic variation ofparent wheat genes (Liu et al. 2007, 2009). Such de novoalleles do not arise simply as a result of UV-induced mutagen-esis, as high frequency of the glutenin alleles were also foundin symmetric somatic hybrids without UV pretreatment (Gaoet al. 2010). Moreover, somaclonal variation of parent wheat istoo rare to account for the observed high frequency of novelglutenin alleles in the somatic hybrids (Feng et al. 2004). It ismore likely that they derive from genomic shock triggered byintrogression of alien chromosome fragments during somatichybridization. Therefore, the suggestion is that somatic hy-bridization provides a means of introgression distinct fromsexual wide crossing and such introgression induces genomicvariations. However, the details of the genomic variations andmechanism of transfer are unknown. Grass genomes are com-posed primarily of repetitive sequences, especially transpos-able elements (Feschotte et al. 2002). Whether somatichybridization induces broad variations in repetitive sequencesdeserves further investigation. Moreover, increasing evidenceshows that epigenetic modifications, such as DNA methyla-tion, play important roles in a wide range of biological pro-cesses, including transposon inactivation and regulation of

gene expression (Bird 2002; Zhang et al. 2006). Whethersomatic hybridization induces epigenetic variations that affectgene expression and/or transposon activation needs furtherexamination. Here, we used DNA profiling techniques to char-acterize genetic and epigenetic alterations from somatic geno-mic shock in six derivatives of bread wheat/tall wheatgrasssomatic hybrids with different phenotypes.

Materials and Methods

Plant materials

Shanrong no. 1 (SR1), Shanrong no. 2 (SR2), Shanrong no. 3(SR3), Shanrong no. 6 (SR6), Shanrong no. 10 (SR10), andShanrong no. 12 (SR12) are six representative introgressionlines derived from asymmetric somatic hybridization betweenbread wheat and tall wheatgrass (Xia et al. 2003; Chen et al.2004a,b). By analogy with terminology applied to generationspostsexual crossing, mutagenesis, and transformation, theregenerated plant postfusion is referred to as R1; segregatingprogeny obtained by self-fertilization in successive generationsare R2, R3, R4, etc. (Figure 1). Only one to two seeds couldproduce from R1 and form a few R2 seedlings, but adequate R3

lines can be used for phenotypic and genomic investigation.We chose the R3 and R10 generations of the six introgressionlines SR1, SR2, SR3, SR6, SR10, and SR12 for analysis. Fourregenerated plants (R1–R4) were derived from protoplastsisolated from cultured embryonic calli of JN177 (R177) bythe method previously described by Feng et al. (2004).

Mitotic and meiotic chromosome analyses

Root-tip meristems were squashed in 45% glacial acetic acidfor GISH karyotyping according to previously describedmethods by Xiang et al. (2003). Tall wheatgrass genomicDNA was labeled as probes and genomic DNA from the breadwheat cv. Chinese Spring was used for blocking. A probe:block ratio of 1:150 was used. For meiotic chromosome anal-ysis, anthers were fixed, squashed, and stained according topreviously described methods by Wang et al. (2004b).

Amplified fragment length polymorphism andmethylation-sensitive amplification polymorphism(MSAP) fingerprinting

Amplified fragment length polymorphism (AFLP) andmethylation-sensitive amplification polymorphism (MSAP) wereperformed according to Shaked et al. (2001), with minormodifications. A set of 44 AFLP and 13 MSAP primer combi-nations were used (Supporting Information, Table S1). Threeindependent technical replicates were performed for each ofthe three biological replicates for AFLP and MSAP reaction,and only intensely stained fragments .100 bp were scored.

RT–PCR analysis

RT–PCR was performed with cDNA templates prepared from2 mg RNA extracted from seedling leaves and roots of cv. JN177and the R3 of six introgression lines and four R177 plantstreated with DNase. The first cDNA strand was synthesized

1036 S. Liu et al.

with M-MLV reverse transcriptase (Invitrogen), following themanufacturer’s instructions. The PCR amplicons were sepa-rated using SSCP (single-strand conformation polymorphism)electrophoresis through 6% nondenaturing polyacrylamidegels at 4� and visualized by silver staining. The set of 80 primerpairs (Table S2) was developed by Bottley et al. (2006). Threeindependent technical replicates were performed for each ofthe three biological replicates for SSCP analysis.

Interretrotransposon amplified polymorphism andretrotransposon microsatellite amplifiedpolymorphism genotyping

Interretrotransposon amplified polymorphism (IRAP) andretrotransposon microsatellite amplified polymorphism(REMAP) PCR were performed in 20-ml reactions, as describedby Kalendar et al. (1999). The 10 IRAP and 12 REMAP primerpairs were developed by Bento et al. (2008) (Table S3). Theamplicons were electrophoresed through 2% agarose gelsand visualized by ethidium bromide staining. Three inde-pendent technical replicates were performed for each primercombination.

Cloning and sequencing of gel fragments

Selected gel fragments were excised from gels and ream-plified using the primers from which they were generated.The amplicons were separated by electrophoresis through

1% agarose gels, gel purified, and ligated into the pEASY-T1plasmid (Transgene), which was transformed into Escheri-chea coli strain DH105 for sequencing based on universalM13 primers. The sequences were BLAST against the NCBIand/or the TREP (the Triticeae Repeat Sequence Database)database.

Southern blot

Fragments with polymorphisms from parent wheat cv.JN177 and the introgression lines in AFLP, SSCP, REMAP,and IRAP analysis were reamplified from the correspondingreconstituted plasmid and labeled with DIG High PrimeDNA Labeling and Detection Starter Kit I (Roche) and usedas hybridization probes for DNA gel-blot analysis. GenomicDNA from the introgression lines, cv. JN177 and R177plants, was digested with restriction enzyme EcoRI and blot-ted onto nylon membranes through vacuum transfer. Prehy-bridization, hybridization, and color detection were performedaccording to protocols provided by the DIG High Prime DNAlabeling and detection starter kit I (Roche).

Bisulfite genomic sequencing

For bisulfite genomic sequencing, genomic DNA was pro-cessed with an EpiTect Bisulfite kit (Qiagen). The amplifi-cation primer pairs for each of MSAP-isolated fragments(MIFs) were designed using MethPrimer software (Li andDahiya 2002). Primer sequences are given in Table S4. ThePCR products were ligated with the pEASY-T Vector (TransGen)and at least 30 clones per insert processed for sequencing. Theratio of C methylation at each cytosine site was calculated andtransformed into a percentage using CyMATE software (http://www.gmi.oeaw.ac.at/research-groups/cymate/cymate/).

Results

Generation of the introgression lines

Our asymmetric somatic hybridization between bread wheatcv. JN177 and tall wheatgrass produced hundreds ofderivatives (Figure 1) classified into five groups accordingto their phenotypes (Xia et al. 2003; Zhang et al. 2005; Xia2009). We chose the six introgression lines SR1, SR2, SR3,SR6, SR10, and SR12 to represent the five phenotypicgroups (SR2 and SR3 belong to the same group), whichhas proven to be phenotypically stable over many genera-tions. The first group showed high grain yield representedby SR1, and the grain yield of SR1 was 35% higher com-pared with JN177 under nonnutrient limiting conditions.These data correlate with the fact that SR1 is registered inChina as a high yielding cultivar. The second group showedhigher abiotic stress tolerance represented by SR2 and SR3;SR2 showed high levels of drought tolerance, while SR3 wasnoticeably more salinity tolerant than cv. JN177. In addition,SR3 showed higher grain yield compared with JN177 undernonnutrient limiting conditions; it is designated as a salinitytolerant cultivar in Shandong Province. The third groupderivatives showed higher disease resistance represented

Figure 1 Generation of somatic hybridization introgression lines. Circlesrepresent nuclei; lines are chromosomes and chromosome segments.

Changes in Somatic Hybrids 1037

by SR6; this derivative was highly resistant to powdery mil-dew and stripe rust. The fourth group derivatives were alldwarfs represented by SR10, and the stems of SR10 were48.20 6 1.31 cm, but they possessed strong tillering abilityand end-use quality. The last group of derivatives showedlarge ears and grains. The 1000 grain weight of one mem-ber, SR12, was nearly 60 g. All measured agricultural traitsof the six introgression lines are listed in Table 1.

Chromosomal constitution of the six introgression lines

We investigated the introgression of Th. ponticum chromo-somal segments into wheat chromosomes in the six lines.Most of the root tip mitotic cells in the R3 generation con-tained 42 chromosomes with a stable somatic chromosomenumber from generation R3 to R10 (data not shown). Karyo-types were nearly identical from cv. JN177, with some minorvariations in arm length ratio and chromosome length (datanot shown). According to genomic in situ hybridization (GISH)analysis, we found several small tall wheatgrass chromosomalsegments present in each line (Figure 2). We also found.90%ring bivalent and .95% total bivalent frequencies in the sixintrogression lines at meiosis (Table S5). This suggests thatintrogressed tall wheatgrass chromosomal segments had littleeffect on chromosome pairing behavior, likely due to the smallsegment size. Given that the karyotypes and DNA fingerprint-ing profiles were stable for each introgression line between theR3 and R10 generations (data not shown), all of the followinganalysis was restricted to the DNA of R3 plants.

Frequency of sequence variation as estimated FromAFLP profiles

In this study, only a few tall wheatgrass segments wereintrogressed into the wheat genome. Our expectation was thatthe AFLP profiles of the introgression lines would be similar tothat of cv. JN177, with the addition of some tall wheatgrass-derived fragments. We focused on fragments present in JN177but lacking in one, or more, of the introgression lines, and thepresence of fragments in the introgression line(s) but not inJN177. Only fragments .100 bp were scored.

AFLP profiling revealed 904 cv. JN177 and 777 tall wheat-grass fragments; 388 were shared by both templates (Figure 3,A and B). Given that AFLP assay markers are dominant, thelevel (percentage) of cv. JN177 fragment loss could be relatedonly to the 516 (904–388) JN177 fragments that did notcomigrate with a tall wheatgrass fragment. The number of

fragments lost from the profile of the six introgression linesranged from 22 to 28, with an overall frequency of 4.7%(Table 2). In addition, the six introgression lines displayed12 to 18 new fragments not present in the cv. JN177 profile,giving an overall frequency of 2.9% (Table 2). AFLP assayresults for cv. JN177 and four R177 lines revealed an aver-age of only 3.5 deleted and 2.0 new bands compared withJN177 (Table 2).

The AFLP-observed fragment changes may have resultedfrom several mechanisms including sequence change atrestriction sites, methylation alteration, or sequence elimi-nation induced by somatic hybridization. To distinguishbetween these possibilities, we isolated four bands deletedin some of the introgression lines from acrylamide AFLP gelsand used them as probes for Southern blot analysis. Theresults confirm deletion of one band, with a single-copy fragment(Figure 3C). The remaining three fragments, which had low-copy DNA, showed no obvious differences between JN177and the introgression lines (Figure 3, D–F). Thus, variationsfound in AFLP analysis are likely due mainly to sequencechanges at restriction sites or from methylation alteration,rather than sequence elimination.

Somatic hybridization induced changes inmethylation pattern

The isoschizomers MspI and HpaII share the same recogni-tion sequence, but differ in their sensitivity to C methylation.HpaII activity is suppressed when the internal cytosine at theCCGG recognition site is methylated. Therefore, C methyla-tion differences can be detected by MSAP analysis, compar-ing profiles generated by EcoRI–MspI and EcoRI–HpaIIdigestion (Shaked et al. 2001). The standards applied toscoring MSAP profiles and AFLP analyses are similar. The13 MSAP primer combinations amplified 347 fragments, ofwhich 210 were present in cv. JN177 but not in the tallwheatgrass parent (Figure 4). Of the 210, the number ofhypermethylated loci in different introgression lines rangedfrom 16 to 27, giving an overall frequency of 10.5% (Table3). Among the loci methylated in cv. JN177, 35 were hypo-methylated in SR1, 33 in SR3, 29 in SR6, 22 in SR10, and 23in both SR2 and SR12, giving an overall frequency of 13.1%(Table 3). The C methylation in cv. JN177 and four R177lines were also detected by MSAP analysis; the results in-dicated that only 5.5 loci hypermethylated in the regener-ated plant and 8.0 loci hypomethylated (Table 3). The

Table 1 Phenotypic traits associated with the six introgression lines and JN177

SampleGrowth habitof seedling

Plantheight (cm)

Spikelength (cm)

No. ofspikelet/spike

No. ofgrain/spike

Effectivetillers

1000 grainweight (g)

JN177 Arrest 66.90 6 3.07 9.30 6 0.63 19.7 6 0.95 42.8 4.7 6 0.95 37.88SR1 Creep 75.60 6 2.27 12.95 6 0.69 22.9 6 0.74 77.1 8.4 6 3.72 46.96SR2 Creep 74.30 6 3.74 11.40 6 0.94 21.8 6 1.14 71.0 6.4 6 1.71 43.17SR3 Creep 75.70 6 3.05 14.00 6 0.62 24.3 6 1.33 66.7 5.5 6 1.18 41.99SR6 Arrest 75.20 6 2.90 8.00 6 0.33 18.5 6 0.97 50.8 6.8 6 1.62 33.03SR10 Creep 48.20 6 1.31 9.30 6 0.48 16.6 6 0.97 41.5 13.5 6 3.54 38.65SR12 Creep 74.30 6 3.27 12.90 6 0.39 19.8 6 0.79 54.8 4.9 6 1.29 57.77

1038 S. Liu et al.

sequencing of five of the fragments hypomethylated in theintrogression lines showed that three were high-copy-numberretrotransposons (TREP accession nos. TREP255, TREP1418,TREP3251), while the other two showed no significant

similarity to any known sequence. None of the five se-quenced fragments that were hypermethylated in the intro-gression lines showed any significant similarity to DNAsequences of publicly available databases.

Figure 2 Cytogenetic analysis of the six introgression lines at R3. GISH karyotypes of (A) tall wheatgrass and (B) bread wheat cv. JN177, with tallwheatgrass genomic DNA as probe. SR1: (C) GISH karyotype, (D) chromosome configuration at meiotic metaphase I. SR2: (E) GISH karyotype, (F)chromosome configuration at meiotic metaphase I. SR3: (G) GISH karyotype, (H) chromosome configuration at meiotic metaphase I. SR6: (I) GISHkaryotype, (J) chromosome configuration at meiotic metaphase I. SR10: (K) GISH karyotype, (L) chromosome configuration at meiotic metaphase I. SR12:(M) GISH karyotype, (N) chromosome configuration at meiotic metaphase I.

Figure 3 AFLP profile and Southernblot analysis of deleted fragments. (A)AFLP profile based on primer combina-tion EAGC + MCTT, (B) primer combina-tion EAAT + MCTT. Fragments absentfrom the introgression lines or regener-ated plant from JN177 calli (R) areindicated by arrowheads and novel frag-ments indicated by arrows. Southernblots verified the deletion of genomicDNA in SR6 with probes derived fromprimer combination EATC + MCTT (C),but not with probes derived from primercombination EAGG + MCTT (D), EAGC +MCTT (E), and EAAA + MCTT (F). 1, SR1;2, SR2; 3, SR3; 6, SR6; 10, SR10; 12,SR12; 177, parent wheat cv. JN177.R, regenerated JN177 plant R1; Th,tall wheatgrass Th. ponticum; L, the100-bp DNA ladder marker.

Changes in Somatic Hybrids 1039

To further investigate the C-methylation changes inducedby somatic hybridization in the introgression lines, C-methylationstatus in several variant MSAP bands was identified bybisulfite sequencing. We found the methylation rates at CG,CHG, and CHH sites of MSAP-isolated fragment 1 (MIF1)were all elevated in SR2 and SR12 (Figure 5A). The level of Cmethylation at CG sites of MIF2 was elevated in SR1, SR2,SR3, and SR10 but reduced in SR6 and SR12, although theCHG and CHH sites were also hypermethylated in SR3(Figure 5B). Moreover, the level of C methylation at CG andCHG sites of MIF3 was elevated in SR1, and those at CG, CHG,and CHH sites were all elevated in R177 lines (Figure 5C). Thelevel of C methylation at CG sites of MIF4 was elevated in SR1,SR2, SR6, and SR10 (Figure 5D).

Gene expression alterations as estimated by SSCP

To survey the homeologous gene silencing or activationinduced by somatic hybridization, we used SSCP to comparethe amplification profiles of the six introgression lines withthose of seedling shoot and root RNA extracted from cv.JN177 and R177. Expression variation was noted for 11 of the80 targeted sequences. Eight of these involved the loss ofa specific homeologous product in at least one introgressionline (Figure 6, A–D). In two cases, a specific mRNA was pres-ent in some introgression lines but not in cv. JN177 (Figure6C). In three cases, the introgression line profiles includeda variant of one of the cv. JN177 fragments (Figure 6D). Totest whether genetic variations were responsible for thechanges of the expression, genomic DNA was also includedas a template. In four of the eight genes involved in homeol-ogous gene silencing, the corresponding fragment was alsoabsent from the genomic DNA profile (Figure 6, A and D).Southern blot analysis based on genomic DNA showed thatone of the homeologues of BE490384, BE494911, andBE606965 had been deleted in the introgression lines (Figure6, E–G). But, this was not the case for BE399113 (Figure 6H).No variation was found between cv. JN177 and the four R177lines in all 80 targeted sequences.

Genomic rearrangements inretrotransposon-adjacent regions

IRAP and REMAP assays were developed to detect variationin sequence-flanking retrotransposons (Kalendar et al.1999). We used these two techniques to identify genomerearrangements from retrotransposition or inter- and intra-element recombination. Here, 219 wheat specific loci weredetected. Five were lost in SR1, SR2, and SR3; 8 were lost in

SR6 and SR12; 11 were lost in SR10, with an overall meanfrequency of lost fragments across the six introgression linesat 3.2% (Figure 7, A–D and Table 4). An average of 1.5fragments was lost in R177 plants when compared withcv. JN177 (Table 4). Seven novel fragments were found inSR1, SR2, and SR12 that were not amplified in both cv.JN177 and tall wheatgrass. Additionally, 6, 5, and 12 novel

Table 2 AFLP profiles of cv. JN177, the six introgression lines, and regenerated JN177 plants (R177)

Introgression lines SR1 SR2 SR3 SR6 SR10 SR12 Average Frequency (%)

Deletions in introgression lines 23 22 26 25 28 22 24.3 4.7New bands in introgression lines 16 15 18 16 12 13 15.0 2.9Regenerated plants R1 R2 R3 R4Deletions in R177 5 3 3 3 3.5 0.7New bands in R177 1 1 2 4 2.0 0.4

Figure 4 MSAP profiling based on primer combinations EATC + HMTCAA.M, DNA restricted by EcoRI + MspI; H, EcoRI + HpaII. Arrows indicatehypermethylated fragments and arrowheads indicate hypomethylatedfragments. 1, SR1; 2, SR2; 3, SR3; 6, SR6; 10, SR10; 12, SR12; 177,parent wheat cv. JN177. R, regenerated JN177 plant R1; Th, tall wheat-grass Th. ponticum; L, the 100-bp DNA ladder marker.

1040 S. Liu et al.

fragments also appeared in SR3, SR6, and SR10, respec-tively. These values give an overall new-band frequency of3.3% (Figure 7, A-D, Table 4). Only 1.0 novel fragmentappeared in R177 plants when compared with cv. JN177(Table 4).

Southern blot analysis based on genomic DNA usingpolymorphic fragments isolated from IRAP and REMAP gelsas probes shows that the lost fragments were deleted insome of the introgression lines (Figure 7, E and F). Wesequenced a sample of these fragments to identify the genesassociated with these genomic rearrangements. A fragment pres-ent in cv. JN177, but absent in some of the introgression lines,resembled a Ty3-gypsy subclass retrotransposon (ABA97230.1).A second polymorphic fragment (present in some of theintrogression lines, but not in cv. JN177) appeared to behighly homologous to a wheat pore-forming toxin-like proteinHfr-2 (AAW48295.1) involved in plant defense (Puthoff et al.2005).

Discussion

Merging divergent genomes into a single nucleus can triggera “highly-programmed sequence of events within the cellthat serves to cushion the effect of [genomic] shock”(McClintock 1984). However, the mechanisms of genomicshock are not well understood. The introgression of alienchromosome fragments can also lead to severe effects inthe accepting genome, which trigger genetic and genomicchanges. Exploring genetic and genomic changes induced bysomatic genomic shock in the somatic introgression lineswill be helpful for understanding the nature of genomicshock. Here, we demonstrated that the somatic hybridiza-tion process induced widespread genetic and epigeneticchanges including sequence absence, modified regulationof gene expression, alteration of cytosine methylation pat-terns, and the activation of quiescent retrotransposons. Sim-ilar genetic and/or epigenetic alterations occurred followingthe formation of de novo polyploids involving Arabidopsisspp. (Comai et al. 2000; Madlung et al. 2002; Wang et al.2004a), Brassica spp. (Song et al. 1995), Spartina spp.(Salmon et al. 2005), Tragopogon spp. (Tate et al. 2006),and Triticeae spp. (Ozkan et al. 2001; Shaked et al. 2001;Han et al. 2003; Ma et al. 2004). However, the somaticgenomic shock induced by somatic hybridization might bestronger than those in allopolyploids given the formerresulted in more genetic and epigenetic instabilities. Insomatic hybrids, the requirement for a period of in vitro

culture adds an additional potential source of variation.However, the contrast made between the DNA extractedfrom cv. JN177 and R177 plants implies that somaclonalvariation was responsible for only a small proportion ofthe overall variation induced in the introgression lines.Thus, genome introgression was likely responsible for mostgenetic and epigenetic variations. Although we cannotexclude the effect of organellar genome introgression,considering the limited amount of alien organellar DNA

Table 3 Variation in C methylation in the introgression lines and regenerated JN177 plants (R177)

Introgression lines SR1 SR2 SR3 SR6 SR10 SR12 Average Frequency (%)

Hypermethylation in introgression lines 27 23 24 23 16 19 22.0 10.5Hypomethylation in introgression lines 35 23 33 29 22 23 27.5 13.1Regenerated plants R1 R2 R3 R4Hypermethylation in R177 5 9 5 3 5.5 2.6Hypomethylation in R177 5 7 9 11 8.0 3.8

Figure 5 C-methylation status in the MSAP-isolated fragments (MIFs) asidentified by bisulfite sequencing. (A) The level of C methylation at CG,CHG, and CHH sites of MIF1 was all elevated in SR2 and SR12. (B) Thelevel of C methylation at CG sites of MIF2 was elevated in SR3 and SR10but reduced in SR6 and SR12 compared to JN177; those at CHG and CHHsites were also elevated in SR3. (C) The level of C methylation at CG andCHG sites of MIF3 was elevated in SR1 and those at CG, CHG, and CHHsites were all elevated in regenerated JN177 plant R1. (D) The level of Cmethylation at CG sites of MIF4 was elevated in SR1, SR2, SR6, and SR10compared to JN177. 177, parent wheat cv. JN177. R, regenerated JN177plant R1.

Changes in Somatic Hybrids 1041

in the hybrids (Chen et al. 2004b), the introgression ofnuclear DNA was likely responsible for most genetic andepigenetic variations.

Sequence absence

One of the most common responses to polyploidization isthe rapid elimination of DNA sequences (Feldman et al.1997; Liu et al. 1998a,b; Ozkan et al. 2001; Shaked et al.2001; Kashkush et al. 2002; Ma et al. 2004; Ma and Gustafson2006). The frequency of elimination is particularly markedin synthetic tetraploid wheat and triticale (Shaked et al.2001; Ma et al. 2004; Ma and Gustafson 2006). In ourstudy, sequence loss was also observed following somatichybridization (Figure 3), but at a rather lower rate. Only4.7% of the cv. JN177 AFLP fragments were not recoveredin the somatic introgression lines, while 0.7% were con-cluded to have been induced by somaclonal variation (Table2). Moreover, Southern blot analysis of the lost AFLP frag-ments indicated that only a few were eliminated from thegenome of the introgression lines. Most of the band absencemay be due to sequence changes of restriction sites or from

methylation alterations. Sequence elimination during allopo-lyploidization provides the physical basis for the diploid-like meiotic behavior of newly formed allopolyploids (Ozkanet al. 2001). Unlike allopolyploidization, where the wholeparental genomes are combined, the size of the individualdonor chromosome fragments introgressed into cv. JN177via asymmetric somatic hybridization is too small to affectmeiotic chromosome pairing (Figure 2). Thus, meiotic pres-sure is likely to be lower compared with a polyploid. This maypartly explain why sequence elimination is not a predominanteffect in somatic hybridization, unlike in polyploidization.

Epigenetic alterations

Patterns of gene expression are affected by allopolyploidiza-tion, particularly in wheat, where 1–5% of the donor geneshave altered expression in a synthetic allotetraploid, largely viaepigenetic changes (Kashkush et al. 2002). Changes in geneexpression are mediated at either the transcriptional and/orthe post-transcriptional levels. Both transcriptional and post-transcriptional gene silencing are associated with alterationsto DNA methylation (Paszkowski and Whitham 2001). In

Figure 6 RT–PCR and genomic DNA characterization ofdifferentially expressed genes. (A) One homoeoallele ofBE494911 is silenced in SR3 and SR10 while anotherhomoeoallele silenced in SR6 and root of SR12; the si-lencing of BE494911 in SR3, SR6, and SR10 is caused bya deletion in the coding region. (B) BE202265 silenced inshoot and root of SR1, SR2, SR3 and shoot of SR12. (C)One homoeoallele of BE606719 is activated in root ofSR2, while another homoeoallele silenced in SR10 root.(D) For the gene, BE606965, one homeologous copydiffers from that of cv. JN177 in SR1, SR2, SR6, andSR10, while another homoeoallele is deleted in the cod-ing region of SR2, SR6, and SR10. Southern blots dem-onstrate the deletion of genomic DNA in (E) BE490384,(F) BE494911, and (G) BE606965, but not in (H)BE399113. Arrows indicate the genomic fragments de-leted in the introgression lines. 1, SR1; 2, SR2; 3, SR3; 6,SR6; 10, SR10; 12, SR12; 177, parent wheat cv. JN177;R, regenerated JN177 plant R1; Th, tall wheatgrassTh. ponticum.

1042 S. Liu et al.

newly synthesized allohexaploid wheat, �13% of the loci ex-perience an alteration in cytosine methylation (Shaked et al.2001). A commonality between the effect of somatic hybrid-ization and allopolyploidization appears to be that both resultin changes to cytosine methylation, which results in the per-turbation of gene expression (Comai et al. 2000; Shaked et al.2001; Kashkush et al. 2002; Han et al. 2003). However, theepigenetic variations induced by somatic hybridization arestronger than those by allopolyploidization. Based on the anal-ysis of our sample of 80 cDNAs, altered expression was in-duced by the somatic hybridization procedure in 10% ofthe introgression line transcripts (Figure 6). The notable

frequency of cytosine methylation changes observed in theasymmetric somatic hybrids (23.6%) is consistent withlarge-scale epigenetic regulation (Table 3). Thus, we con-cluded that somatic hybridization induced a broad spec-trum of C-methylation changes that perturbed geneexpression to a larger extent than allopolyploidization.

Activation and repression of transposon activity

Genomic shock activates quiescent transposons and causesgenome restructuring (McClintock 1984). In fact, transcrip-tional activation of transposable elements has been shown insome newly synthesized allopolyploids (Kashkush et al. 2002,2003). IRAP and REMAP both represent multilocus PCR-basedmethods designed to detect retrotransposon related genomerearrangements and have been used to uncover sequence rear-rangements associated with retrotransposon-rich regions in trit-icale (Bento et al. 2008, 2010). Here, 6.5% of the IRAP andREMAP fragments were polymorphic between cv. JN177 andthe introgression lines (Figure 7 and Table 4). The introgressionevents were associated with novel sequence polymorphisms atthe junctions between the donor and the recipient DNA. How-ever, given the overall number of wheatgrass segments presentin the introgression lines, frequency of polymorphisms revealedby IRAP and REMAP was too high to merely reflect introgres-sion (Figure 2). Instead, the polymorphisms detected by IRAPand REMAP likely arose from de novo transposition events.These have been detected in the vicinity of genes encodingcertain endosperm storage proteins by Liu et al. (2009).

Evidence suggests that C methylation has evolved asa means of repressing transposon activity. Clearly, highlyrepetitive DNA—much of which is contributed by retrotrans-posons in the large genome grass species—in plant genomesis heavily methylated (Bennetzen et al. 1994). The activityof retrotransposons needs to be tightly controlled to ensureviability and survival of the host, and a major control stepinvolves the inhibition of transcription. Reactivation oftransposons has been observed either as a response to stressor to genomic shock (Bennetzen 2000; Liu and Wendel2000; Kashkush et al. 2002, 2003; Shan et al. 2005, 2009;Bento et al. 2008, 2010). Our MSAP analysis of the intro-gression lines indicates that alteration in C methylationinvolves repetitive sequences. In particular, the hypomethy-lation of retrotransposons revealed is suggestive of retro-transposon activation in the introgression lines.

Although considerable transposition occurred during thesomatic hybridization event, the introgression lines were highlystable from R3 to R10 generation. The implied repression of

Figure 7 REMAP and IRAP analysis, based on primer combinations (A)(GA)9C + sukkula, (B) (GA)9C + sabrina, (C) nikita + sukkula, (D) nikita +sabrina. Southern blots demonstrate the retrotransposon-related dele-tions in the introgression lines with probes derived from (CA)9G + sabrina(E) and nikita + sukkula (F). Arrowheads indicate retrotransposon-relateddeletions in the introgression lines induced by asymmetric somatic hybrid-ization, while arrows indicate retrotransposon-related insertions in theintrogression lines induced by asymmetric somatic hybridization. 1, SR1;2, SR2; 3, SR3; 6, SR6; 10, SR10; 12, SR12; 177, parent wheat cv. JN177;R, regenerated JN177 plant R1; Th, tall wheatgrass Th. ponticum.

Table 4 Genomic rearrangements in retrotransposon related regions in the introgression lines and regenerated JN177 plants (R177)

Introgression lines SR1 SR2 SR3 SR6 SR10 SR12 Average Frequency (%)

Bands lost in introgression lines 5 5 5 8 11 8 7.0 3.2New bands in introgression lines 7 7 6 5 12 7 7.3 3.3Regenerated plants R1 R2 R3 R4Bands lost in R177 2 2 1 1 1.5 0.7New bands in R177 1 0 1 2 1.0 0.5

Changes in Somatic Hybrids 1043

retrotransposon activity closely resembles the behavior of riceintrogression lines (Liu and Wendel 2000), although the un-derlying mechanism is unknown. Wolffe and Matzke (1999)have proposed the involvement of homology-dependent genesilencing (HDGS), an epigenetic phenomenon wherebyone gene is suppressed by the action of a homologous gene(Zaratiegui et al. 2007). When transposon activity is acti-vated by hypomethylation, HDGS is triggered and similarsequences become targets for suppression. How importantthe rapid repression of retrotransposon activity is for thestabilization of the introgression lines following somatichybridization clearly needs further investigation.

Genetic alterations driven by repetitive sequences

Repetitive sequences are a major source of genome instability asthey are susceptible to expansion and contraction (Heidenfelderand Topal 2003). As the most abundant repetitive sequences inwheat genome, the retrotransposon-related sequences showedhigh frequency of alteration (6.5%) in the introgression lines(Table 4). Similarly, high-frequency sequence variation throughindels of repetitive motifs (14/37, 37.8%) was also observed inthe glutenin gene family (Feng et al. 2004; Liu et al. 2007,2009; Gao et al. 2010). This supports the notion that repetitivesequences are a driving force behind the de novo genetic vari-ation generated in the introgression lines. Double-strand breaks(DSBs) appear to be common in regions enriched for repetitivesequences, and these “fragile sites” represent hotspots for re-combination (Wicker et al. 2010). The suggestion is thereforethat repetitive sequences drive the formation of DSBs or othergenetic alterations that occur as a result of genomic shock andsubsequent introgression.

Conclusions

We generated a set of introgression lines displaying a varietyof genetic and epigenetic changes relative to their parentsusing somatic hybridization between wheat and tall wheat-grass. The somatic hybridization process mimics many of thegenetic alterations induced by polyploidization or sexual widehybridization, in a remarkably short time frame and withstronger extent. Therefore, we suggest the somatic hybridiza-tion approach not only provides an effective and time-efficientmeans of achieving introgression into a crop species from itswild relatives, but also provides a means to explore the natureof the genetic and epigenetic events induced by somaticgenomic shock. An understanding of the underlying mecha-nisms may shed light on the variation released in sexual wildhybrids, which has been broadly described as being due togenomic shock.

Acknowledgments

We thank Dr. Austin Cape for careful reading and feedback.This work was supported by the funds of the Natural ScienceFoundation of China (no. 30871320; 31000568), MajorProgram of the Natural Science Foundation of China (no.31030053), Shandong Province Program (no. Q2006D02).

Literature Cited

Bassene, J. B., Y. Froelicher, C. Dubois, R. M. Ferrer, L. Navarroet al., 2010 Non-additive gene regulation in a citrus allotetra-ploid somatic hybrid between C. reticulata Blanco and C. limon(L.) Burm. Heredity 105: 299–308.

Bennetzen, J. L., 2000 Transposable element contributionsto plant gene and genome evolution. Plant Mol. Biol. 42: 251–269.

Bennetzen, J. L., K. Schrick, P. S. Springer, W. E. Brown, and P.SanMiguel, 1994 Active maize genes are unmodified andflanked by diverse classes of modified, highly repetitive DNA.Genome 37: 565–576.

Bento, M., H. S. Pereira, M. Rocheta, P. Gustafson, W. Viegas et al.,2008 Polyploidization as a retraction force in plant genomeevolution: sequence rearrangements in triticale. PLoS ONE 3:e1402.

Bento, M., P. Gustafson, W. Viegas, and M. Silva, 2010 Genomemerger: from sequence rearrangements in triticale to their elim-ination in wheat–rye addition lines. Theor. Appl. Genet. 121:489–497.

Bird, A. P., 1997 Does DNA methylation control transposition ofselfish elements in the germline. Trends Genet. 13: 469–470.

Bird, A. P., 2002 DNA methylation patterns and epigenetic mem-ory. Genes Dev. 16: 6–21.

Bottley, A., G. M. Xia, and R. M. D. Koebner, 2006 Homoeologousgene silencing in hexaploid wheat. Plant J. 47: 897–906.

Chen, S. Y., G. M. Xia, T. Y. Quan, F. N. Xiang, Y. Jin et al.,2004a Introgression of salt-tolerance from somatic hybrids be-tween common wheat and Thinopyrum ponticum. Plant Sci.167: 773–779.

Chen, S. Y., S. W. Liu, C. H. Xu, Y. Z. Chen, and G. M. Xia,2004b Heredity of chloroplast and nuclear genomes of asym-metric somatic hybrid lines between wheat and couch grass.Acta Bot. Sin. 46: 110–115.

Cheng, A. X., G. M. Xia, D. Y. Zhi, and H. M. Chen,2004 Intermediate fertile Triticum aestivum (+) Agropyronelongatum somatic hybrids are generated by low doses of UVirradiation. Cell Res. 14: 86–91.

Comai, L., A. P. Tyagi, K. Winter, R. Holmes-Davis, S. H. Reynolds et al.,2000 Phenotypic instability and rapid gene silencing in newlyformed Arabidopsis allotetraploids. Plant Cell 12: 1551–1567.

Feldman, M., B. Liu, G. Segal, S. Abbo, A. A. Levy et al.,1997 Rapid elimination of low-copy DNA sequences in poly-ploid wheat: A possible mechanism for differentiation of homoe-ologous chromosomes. Genetics 147: 1381–1387.

Feng, D. S., G. M. Xia, S. Y. Zhao, and F. G. Chen, 2004 Twoquality-associated HMW glutenin subunits in a somatic hybridline between Triticum aestivum and Agropyron elongatum.Theor. Appl. Genet. 110: 136–144.

Feschotte, C., N. Jiang, and S. R. Wessler, 2002 Plant transposableelements: where genetics meets genomics. Nat. Rev. Genet. 3:329–341.

Gao, X., S. W. Liu, Q. Sun, and G. M. Xia, 2010 High frequency ofHMW-GS sequence variation through somatic hybridization be-tween Agropyron elongatum and common wheat. Planta 231:245–250.

Griffiths, S., R. Sharp, T. N. Foote, I. Bertin, M. Wanous et al.,2006 Molecular characterization of Ph1 as a major chromo-some pairing locus in polyploid wheat. Nature 439: 749–752.

Han, F. P., G. Fedak, T. Ouellet, and B. Liu, 2003 Rapid genomicchanges in interspecific and intergeneric hybrids and allopoly-ploids of Triticeae. Genome 46: 716–723.

Heidenfelder, B. L., and M. D. Topal, 2003 Effects of sequence onrepeat expansion during DNA replication. Nucleic Acids Res. 31:7159–7164.

1044 S. Liu et al.

Kashkush, K., M. Feldman, and A. A. Levy, 2002 Gene loss, silenc-ing and activation in a newly synthesized wheat allotetraploid.Genetics 160: 1651–1659.

Kashkush, K., M. Feldman, and A. A. Levy, 2003 Transcriptionalactivation of retrotransposons alters the expression of adjacentgenes in wheat. Nat. Genet. 33: 102–106.

Kalendar, R., T. Grob, M. Regina, A. Suoniemi, and A. Schulman,1999 IRAP and REMAP: two new retrotransposon-based DNAfingerprinting techniques. Theor. Appl. Genet. 98: 704–711.

Li, L. C., and R. Dahiya, 2002 MethPrimer: designing primers formethylation PCRs. Bioinformatics 18: 1427–1431.

Liu, B., and J. F. Wendel, 2000 Retrotransposon activation fol-lowed by rapid repression in introgressed rice plants. Genome43: 874–880.

Liu, B., M. J. Vega, and M. Feldman, 1998a Rapid genomicchanges in newly synthesized amphiploids of Triticum andAegilops. II. Changes in low-copy coding DNA sequences. Ge-nome 41: 535–542.

Liu, B., J. M. Vega, G. Segal, S. Abbo, M. Rodova et al.,1998b Rapid genomic changes in newly synthesized amphi-ploids of Triticum and Aegilops. I. Changes in low-copy non-coding DNA sequences. Genome 41: 272–277.

Liu, B., C. L. Brubaker, G. Mergeai, R. C. Cronn, and J. F. Wendel,2001 Polyploid formation in cotton is not accompanied byrapid genomic changes. Genome 44: 321–330.

Liu, H., S. W. Liu, and G. M. Xia, 2009 Generation of high fre-quency of novel alleles of the high molecular weight glutenin insomatic hybridization between bread wheat and tall wheat-grass. Theor. Appl. Genet. 118: 1193–1198.

Liu, S. W., and G. M. Xia, 2014 The place of asymmetric somatichybridization in wheat breeding. Plant Cell Rep. 33: 595–603.

Liu, S. W., S. Y. Zhao, F. G. Chen, and G. M. Xia, 2007 Generationof novel high quality HMW-GS genes in two introgression linesof Triticum aestivum/Agropyron elongatum. BMC Evol. Biol. 7:76.

Ma, X. F., and J. P. Gustafson, 2006 Timing and rate of genomevariation in triticale following allopolyploidization. Genome 49:950–958.

Ma, X. F., P. Fang, and J. P. Gustafson, 2004 Polyploidization-induced genome variation in triticale. Genome 47: 839–848.

Madlung, A., R. W. Masuelli, B. Watson, S. H. Reynolds, J. Davisonet al., 2002 Remodeling of DNA methylation and phenotypicand transcriptional changes in synthetic Arabidopsis allotetra-ploids. Plant Physiol. 129: 733–746.

McClintock, B., 1984 The significance of responses of the genometo challenge. Science 226: 792–801.

Ozkan, H., A. A. Levy, and M. Feldman, 2001 Alloploidy-inducedrapid genome evolution in the wheat (Aegilops-Triticum) group.Plant Cell 13: 1735–1747.

Paszkowski, J., and S. A. Whitham, 2001 Gene silencing and DNAmethylation processes. Curr. Opin. Plant Biol. 4: 123–129.

Puthoff, D. P., N. Sardesai, S. Subramanyam, J. A. Nemacheck, andC. E. Williams, 2005 Hfr-2, a wheat cytolytic toxin-like gene, isup-regulated by virulent Hessian fly larval feeding. Mol. PlantPathol. 6: 411–423.

Salmon, A., M. L. Ainouche, and J. F. Wendel, 2005 Genetic andepigenetic consequences of recent hybridization and polyploidyin Spartina (Poaceae). Mol. Ecol. 14: 1163–1175.

Shaked, H., K. Kashkush, H. Ozkan, M. Feldman, and A. A. Levy,2001 Sequence elimination and cytosine methylation are rapidand reproducible responses of the genome to wide hybridizationand allopolyploidy in wheat. Plant Cell 13: 1749–1759.

Shan, X. H., Z. L. Liu, Z. Y. Dong, Y. M. Wang, Y. Chen et al.,2005 Mobilization of the active MITE transposons mPing andPong in rice by introgression from wild rice (Zizania latifoliaGriseb.). Mol. Biol. Evol. 22: 976–990.

Shan, X. H., X. F. Ou, Z. L. Liu, Y. Z. Dong, X. Y. Lin et al.,2009 Transpositional activation of mPing in an asymmetricnuclear somatic cell hybrid of rice and Zizania latifolia wasaccompanied by massive element loss. Theor. Appl. Genet.119: 1325–1333.

Song, K., P. Lu, K. Tang, and T. C. Osborn, 1995 Rapid genomechange in synthetic polyploids of Brassica and its implicationsfor polyploid evolution. Proc. Natl. Acad. Sci. USA 92: 7719–7723.

Tate, J. A., Z. Ni, A.-C. Scheen, J. Koh, C. A. Gilbert et al.,2006 Evolution and expression of homeologous loci in Trago-pogon miscellus (Asteraceae), a recent and reciprocally formedallopolyploid. Genetics 173: 1599–1611.

Wang, J., L. Tian, A. Madlung, H.-S. Lee, M. Chen et al.,2004a Stochastic and epigenetic changes of gene expressionin Arabidopsis polyploids. Genetics 167: 1961–1973.

Wang, J., F. N. Xiang, and G. M. Xia, 2004b Transfer of smallchromosome fragments of Agropyron elongatum to wheat chro-mosome via asymmetric somatic hybridization. Sci. China Ser. C.47: 434–441.

Wicker, T., J. P. Buchmann, and B. Keller, 2010 Patching gaps inplant genomes results in gene movement and erosion of colin-earity. Genome Res. 20: 1229–1237.

Wolffe, A. P., and M. A. Matzke, 1999 Epigenetics: regulationthrough repression. Science 286: 481–486.

Xia, G. M., 2009 Progress of chromosome engineering mediatedby asymmetric somatic hybridization. J. Genet. Genomics 36:547–556.

Xia, G. M., F. N. Xiang, A. F. Zhou, H. Wang, and H. M. Chen,2003 Asymmetric somatic hybridization between wheat (Tri-ticum aestivum L.) and Agropyron elongatum (Host). Nevishi.Theor. Appl. Genet. 107: 299–305.

Xiang, F. N., G. M. Xia, and H. M. Chen, 2003 Effect of UV dosageon somatic hybridization between common wheat (Triticum aes-tivum L.) and Avena sativa L. Plant Sci. 164: 697–707.

Xiang, F. N., G. M. Xia, D. Y. Zhi, J. Wang, H. Nie et al.,2004 Hybrid plant regeneration in relation to the nuclearand cytoplasmic genomes of wheat and Setaria italica. Genome47: 680–688.

Xu, S. X., D. F. Cai, F. Q. Tan, Y. N. Fang, K. D. Xie et al.,2014 Citrus somatic hybrid: an alternative system to studyrapid structural and epigenetic reorganization in allotetraploidgenomes. Plant Cell Tissue Organ Cult. 119: 511–522.

Zaratiegui, M., D. V. Irvine, and R. A. Martienssen,2007 Noncoding RNAs and gene silencing. Cell 128: 763–776.

Zhang, Q. L., G. M. Xia, T. Y. Quan, Y. G. Zhang, F. X. Li et al.,2005 Trait variation of somatic hybrid progenies from Triticumaestivum/Agropyron elongatum Host (Nevski) (in Chinese withEnglish Abstract). J. Triticeae Crop. 25(6): 11–14.

Zhang, X. Y., J. Yazaki, A. Sundaresan, S. Cokus, S. W. Chan et al.,2006 Genome-wide high-resolution mapping and functionalanalysis of DNA methylation in Arabidopsis. Cell 126: 1189–1201.

Zhou, A. F., and G. M. Xia, 2005 Introgression of the Haynaldiavillosa genome into g-ray-induced asymmetric somatic hybridsof wheat. Plant Cell Rep. 24: 289–296.

Communicating editor: J. A. Birchler

Changes in Somatic Hybrids 1045

GENETICSSupporting Information

http://www.genetics.org/lookup/suppl/doi:10.1534/genetics.114.174094/-/DC1

Genetic and Epigenetic Changes in Somatic HybridIntrogression Lines Between Wheat and

Tall WheatgrassShuwei Liu, Fei Li, Lina Kong, Yang Sun, Lumin Qin, Suiyun Chen, Haifeng Cui,

Yinghua Huang, and Guangmin Xia

Copyright © 2015 by the Genetics Society of AmericaDOI: 10.1534/genetics.114.174094

2 SI  S. Liu et al. 

Supplemental data Table S1 Adaptors and primers used in AFLP and MSAP analysis. Name Sequence Adaptors Mse I-adaptor-1 5’-GAC GAT GAG TCC TGA G Mse I-adaptor-2 5'-TAC TCA GGA CTC AT EcoR I-adaptor-1 5’-CTC GTA GAC TGC GTA CC EcoR I-adaptor-2 5’-AAT TGG TAC GCA GTC TAC H/M-adaptor-1 5’-GAT CAT GAG TCC TGC T H/M-adaptor-2 5’-CGA GCA GGA CTC ATG A

Pre-selective primers Mse I+C 5’-GAT GAG TCC TGA GTA AC EcoR I+A 5’-GACTGCGTACCAATTCA H/M+T 5'-CAT GAG TCC TGC TCG GT

Selective primers Mse I+3 MCTT 5'-GAT GAG TCC TGA GTA ACT T MCTA 5'-GAT GAG TCC TGA GTA ACT A MCAG 5'-GAT GAG TCC TGA GTA ACA G MCCA 5'-GAT GAG TCC TGA GTA ACC A MCAA 5'-GAT GAG TCC TGA GTA ACA A EcoR I+3 EAGA 5'-GAC TGC GTA CCA ATT CAG A EACC 5'-GAC TGC GTA CCA ATT CAC C EAAC 5'-GAC TGC GTA CCA ATT CAA C EATC 5'-GAC TGC GTA CCA ATT CAT C EACA 5'-GAC TGC GTA CCA ATT CAC A EAAA 5'-GAC TGC GTA CCA ATT CAA A EAAG 5'-GAC TGC GTA CCA ATT CAA G EAAT 5'-GAC TGC GTA CCA ATT CAA T EAGG 5'-GAC TGC GTA CCA ATT CAG G EAGC 5'-GAC TGC GTA CCA ATT CAG C EAGT 5'-GAC TGC GTA CCA ATT CAG T H/M+4 HMTCAA 5'-CAT GAG TCC TGC TCG GTC AA HMTCCA 5'-CAT GAG TCC TGC TCG GTC CA

  S. Liu et al.  3 SI

Table S2 The primers used for SSCP analysis.

Genbank Identifier

Sequence

BE422968 F CACGACGTTGTAAAACGACGCTCTGTGCTGCATTGATCGT

BE422968 R TCGCTGCCACTTCTTTACCT

BE423963 F CACGACGTTGTAAAACGACGAAATCAAAGAACCCGACACG

BE423963 R GGAGCCAATCATGAAGGGTA

BE424277 F CACGACGTTGTAAAACGACGGAACAATCTGCGGGGTAGAA

BE424277 R TCCCAATTCTCGATTGGAAG

BE425898 F CACGACGTTGTAAAACGACGTGAGGCTGATGTTGTCCTTG

BE425898 R AGGCTACACCAGATGCCTTG

BE438469 F CACGACGTTGTAAAACGACGCGCTGCCACTTCTTTACCTC

BE438469 R AGAAGAACGCGACCATCAAG

BE438643 F CACGACGTTGTAAAACGACGCCAATGTAGCCAGCAATCCT

BE438643 R GGGCAAAATTAGCAGCAGAG

BE438771 F CACGACGTTGTAAAACGACGTCATGGACATCCAAGCTTCA

BE438771 R TGTACTGCTGGAGACCCTTG

BE438810 F CACGACGTTGTAAAACGACGTTGCTTTTCTTTTTCCTTTTGG

BE438810 R CCAAGTCTTTTGCAGCCAAT

BE438866 F CACGACGTTGTAAAACGACGACACCCCTCCAACTCAACTG

BE438866 R TATTTTTCTCTCCGGCAAGG

BE438889 F CACGACGTTGTAAAACGACGTCACCGCTGCAACTCATGG

BE438889 R TAATAGTCTCTCCGGCGACG

BE442851 F CACGACGTTGTAAAACGACGGGAAGCTAGCCTCGTGCTAA

BE442851 R AACGCACAGGGAAACAAAAC

BE443835 F CACGACGTTGTAAAACGACGCAGAAGCCCTGCCTACGTTA

BE443835 R GCAGCTGGATGGTCTCTTTC

BE443905 F CACGACGTTGTAAAACGACGCCAGGTGTCCTTCAGGTTGT

BE443905 R CAGGAACTGGATCCCTTCAA

BE445428 F CACGACGTTGTAAAACGACGCTGAACCTCTGGAAGCAAGG

BE445428 R CGTCACATATCCCATGATCG

BE471203 F CACGACGTTGTAAAACGACGACCTCGAGTTCCTCAAGCAG

BE471203 R TCGAGAATGTGCCACAACTG

BE490041 F CACGACGTTGTAAAACGACGTAGAGAGGCCGTGACCAACT

BE490041 R TCTCTGGCTGCTTCCTCATT

BE490430 F CACGACGTTGTAAAACGACGTCACCGTTCCGATCTCTCTC

BE490430 R GCCTCTCCGTGTGCTTCTC

BE494911 F CACGACGTTGTAAAACGACGGTTCACCAACCCAGGAAAGA

BE494911 R AGGAGAGGAGGAAGCAAAGC

BE497107 F CACGACGTTGTAAAACGACGAGTTCAGCCTCGACTCCAAG

BE497107 R CTTATTGCCAACCAGCATGA

BE500570 F CACGACGTTGTAAAACGACGCAGGTGAGGTGTTGGGAGAT

BE500570 R TTAGCGTCTTGTCCCAGCTT

4 SI  S. Liu et al. 

BE590760 F CACGACGTTGTAAAACGACGAGCTCGCGAAGAAGAGAGTG

BE590760 R CACAAAGCAAGAGCAAGCAG

BF428796 F CACGACGTTGTAAAACGACGGCAAACCCAGCTATCACCTC

BF428796 R GAACGGAGGAAGAGCAACAG

BF473313 F CACGACGTTGTAAAACGACGGCACGATCTGGACAACAAGA

BF473313 R CCAGAAGCCGGATACAAAGA

BM134307 F CACGACGTTGTAAAACGACGCGTTCAGAAGAGACCCAACC

BM134307 R CTCCTCCAGACGTTCCAGTC

BM134506 F CACGACGTTGTAAAACGACGAGCCATAATGGAAGGAAGCA

BM134506 R CTCGCTGATCTCCAGCTGAT

BM136937 F CACGACGTTGTAAAACGACGCCTCGTCTGCAAGCAGTG

BM136937 R GTCCCCTCTGGTGAAGAGC

BM138152 F CACGACGTTGTAAAACGACGTTCTACAGGGCGGAGGAGTA

BM138152 R CAATCCTTGCATTTAGGTAACAG

BE490752 F CACGACGTTGTAAAACGACGAGCAGCCAAGGTGGTGAT

BE490752 R CCCATTAAGCACACTGTTCG

BE404374 F CACGACGTTGTAAAACGACGATGCCCTTTGCTCCTATCCT

BE404374 R CCTGGCGATTTGTTAGCATT

BE404461 F CACGACGTTGTAAAACGACGACCGCTTGTGGAGCAATACT

BE404461 R ATGGTGCAGGCTTTCATCTC

BE404656 F CACGACGTTGTAAAACGACGCTGGTCAAGTGGAGCCTCAT

BE404656 R GTTCGGATGAAGGCAATGAT

BE497566 F CACGACGTTGTAAAACGACGCAGAGTTCAGGGAGCTACGG

BE497566 R GAACGGAGGCAGTAGCATGT

BE518279 F CACGACGTTGTAAAACGACGTGGCGTCACAATGAAGAGAA

BE518279 R CTCCATCTCCAGTTCATGTCG

BE489274 F CACGACGTTGTAAAACGACGGTGGAGGTCAAGGGATGAGA

BE489274 R CACACAGCCATGAAACTTGG

BE405509 F CACGACGTTGTAAAACGACGATTACGAGCAGCACCAGACC

BE405509 R AGCAGTGAACCAGCACATTG

BE586090 F CACGACGTTGTAAAACGACGGGGCATTTAGAGGCTGTTCA

BE586090 R AGGGAATCAATGTCCTGCAC

BE423249 F CACGACGTTGTAAAACGACGGCCAAACTCCCAAAAGCATA

BE423249 R GAGGAGGAGATGGTGGTTCA

BE490274 F CACGACGTTGTAAAACGACGTCATTCTATTGGCGCGTACA

BE490274 R GCATCAATGGCTGCTGTAGA

BM138019 F CACGACGTTGTAAAACGACGCCAACTCCTGGGAAGAAAGA

BM138019 R TCAAATGTTCAGTGCATCCAT

BE517931 F CACGACGTTGTAAAACGACGGTACTCGGTGGACGTCTGGT

BE517931 R GAGAAACGCAGCAGGAAGAC

BF200563 F CACGACGTTGTAAAACGACGACTGGATTGTTGACGGGAAG

BF200563 R CCGAATGGCAGGTTTTTCTA

BF473993 F CACGACGTTGTAAAACGACGACATAACCTGATCCGCAAGG

BF473993 R TACAGTGGCCATGCTTGAAC

  S. Liu et al.  5 SI

BE606719 F CACGACGTTGTAAAACGACGTAAGAAGGCGCTTGCAAAGT

BE606719 R GCAAGGTTTACGTCAAACGAA

BE606965 F CACGACGTTGTAAAACGACGATGCTCAAGCCGAGGAAGTA

BE606965 R TTGATGGAGCGACTGAAATG

BF473348 F CACGACGTTGTAAAACGACGTGTCGAGCAAGGACATCATC

BF473348 R TTGGGCATGAGCTCGTAGAT

BG604766 F CACGACGTTGTAAAACGACGTCTAGCTGGTTTCTCGTGGTG

BG604766 R CTCCAAAATCTTTCCCCAAA

BE498761 F CACGACGTTGTAAAACGACGAAGCAATGTCTTTGGCACCT

BE498761 R AGGAGGAGGTGATGGCTGT

BE499982 F CACGACGTTGTAAAACGACGCCCAAGGGTACATCAGCATC

BE499982 R AGAGTTGGCCTTTCGGTTTT

BF202265 F CACGACGTTGTAAAACGACGAAGGGAAGGGAGGACAGAGA

BF202265 R CAGCAAAAGTGTCGGTGAGA

BE398439 F CACGACGTTGTAAAACGACGCCCTTTTCTGAAAGGTGCAA

BE398439 R GGTTCAGACGCTCATTCACA

BE471132 F CACGACGTTGTAAAACGACGTCACGCTTCTTAATATGCTCTTCA

BE471132 R GGTATACTAGAGCCAATAAAAGCTGGT

BE426364 F CACGACGTTGTAAAACGACGGAAGCCAATTATTGTGGGACA

BE426364 R AATCCCACCTCCAATTGTCA

BE490384 F CACGACGTTGTAAAACGACGGTTGAACTTGATGGCGCTCT

BE490384 R CCGACCTCCCCTACGACTAC

BE403516 F CACGACGTTGTAAAACGACGGTCCGGGTCAATCAAGAAGA

BE403516 R AACCTTCTTGGCCTGGATGT

BE404371 F CACGACGTTGTAAAACGACGGTTCATGGGCTGCACAACT

BE404371 R GAGGAGGACATCGCATTCAT

BE497494 F CACGACGTTGTAAAACGACGCACCAACATTTTTAGTTAATGCTGA

BE497494 R GCGCCAGATCCTCAATAGAA

BE518213 F CACGACGTTGTAAAACGACGGGAACAGTGATGGATTCAAGG

BE518213 R TGTGTATTTGGTGCCGACAT

BF473846 F CACGACGTTGTAAAACGACGAGGATTACTGCGACGAGACG

BF473846 R GGTCATCCAGAACCAGAACG

BE606438 F CACGACGTTGTAAAACGACGAGCCAAAAGCACAAAGAGGA

BE606438 R AAGGAAGATGCTGCAAGGAA

BF473744 F CACGACGTTGTAAAACGACGTTATTCCCACAATCCCAAGC

BF473744 R AATTCATTGAAGACATGGCTGTT

BE499671 F CACGACGTTGTAAAACGACGCCTGATGTCATCCTCGTGAA

BE499671 R CTCTCGCTGCTCGATCTCTC

BE498622 F CACGACGTTGTAAAACGACGCAAGAAGGAGCTAAGGACACTCA

BE498622 R TGGTGTGGTCGTAGGGAATTA

BE499478 F CACGACGTTGTAAAACGACGAGGGCTTGCTTTTCACAAGA

BE499478 R ATACTGCCCCGGGAGTTCTA

BE442655 F CACGACGTTGTAAAACGACGCTTTTGCGGAGGTTCAGAAG

BE442655 R GCTCGGGAGAGCATCATAAG

6 SI  S. Liu et al. 

BE443160 F CACGACGTTGTAAAACGACGGGATCGCACCAAGTACACG

BE443160 R GCATGGATCGGTCGAAAT

BE443948 F CACGACGTTGTAAAACGACGTGTTGTCCAGCAACAACCAC

BE443948 R AGATGGGCCCCAAAATAGAT

BE444599 F CACGACGTTGTAAAACGACGGGATGCTGCTAAGCATATGGAA

BE444599 R GCACCCTTTGCAAGAGCA

BE444178 F CACGACGTTGTAAAACGACGGGTCACTTCTGGTGGCAAAT

BE444178 R AGATTCCGCATCATCACCTC

BE494907 F CACGACGTTGTAAAACGACGCCCAAGGAATGCTGATAGGA

BE494907 R GACTGCGGCACAGTATCAGA

BE493826 F CACGACGTTGTAAAACGACGGTCCTCGGGTACCAACCAG

BE493826 R GGCGGCTTCTTTATCAGGAG

BE445693 F CACGACGTTGTAAAACGACGTTTACTGGAGCTTGGCGTTC

BE445693 R CAATGTGCAAGGTCCATGAT

BE591604 F CACGACGTTGTAAAACGACGGGCGGATGTTGCTTATAGGA

BE591604 R TAGCAGGCTGAGGTCTCGTT

BF202540 F CACGACGTTGTAAAACGACGCTCCCTGCAGGCCTCAAC

BF202540 R AAACTGCAGTGGGGGTCTT

BE637228 F CACGACGTTGTAAAACGACGTGAACCTGAGCTCCCAGACT

BE637228 R CAGCATTCTCCTGAGCTGTG

BF145580 F CACGACGTTGTAAAACGACGCTTCCTTCGTGATGTACAGAGG

BF145580 R CAAATGGGCATGTTGACAAT

BF478648 F CACGACGTTGTAAAACGACGTGATCACGTTGGTGAAGAGC

BF478648 R TCTTCTCCCTCATCCTGTCC

BG313656 F CACGACGTTGTAAAACGACGCTTTCATGCTCCTTGGCTGT

BG313656 R CCAGCATGTCGCTCTTGTT

BG313362 F CACGACGTTGTAAAACGACGGCTTCTCCAAAGGCAACAAA

BG313362 R GGAATGTCTATTGACCCAAAGC

BE399113 F CACGACGTTGTAAAACGACGGAGCCTCCTGCTCAGCTC

BE399113 R ATGCCAGTTCAACGCCACT

BE399363 F CACGACGTTGTAAAACGACGGCTCCTCCACGCGATATG

BE399363 R CACTTGAGCAGCACCACCT

  S. Liu et al.  7 SI

Table S3 Primers for IRAP/REMAP analysis.

Primer

Sequence

Barley LTR Retrotransposon

Nikita 5’-CGCTCCAGCGGTACTGCC

Sabrina 5’-GCAAGCTTCCGTTTCCGC

Sukkula 5’-GATAGGGTCGCATCTTGGGCGTGAC

Stowaway 5’-CTTATATTTAGGAACGGAGGGAGT

SSR

(GA)9C 5’-GAGAGAGAGAGAGAGAGAC

(CT)9G 5’-CTCTCTCTCTCTCTCTCTG

(CA)9G 5’-CACACACACACACACACAG

8 SI  S. Liu et al. 

Table S4 Primers for bisulfite sequencing.

Primer Sequence

MIF1-F 5’-TGGYATGATATTTGTTTGATTGTAT

MIF1-R 5’-CATTCATRTTTACAAAACCTRTTTTT

MIF2-F 5’- GAGAAGGAGYTTATGGAGAAGT

MIF2-R 5’- RTCATACATACCTAAATATCACCCT

MIF3-F 5’- GGGAGTATATGGGTTTTATTGGGTTATA

MIF3-R 5’- ATCGAACAATTCCTATTTACGTCA

MIF4-F 5’-TTAGATGCGATATTATTTTTGAAGGT

MIF4-R 5’- TACAATCTCCCTCAAATCTCCATATACTAA

  S. Liu et al.  9 SI

Table S5 Chromosome configuration in pollen mother cells at meiotic metaphase I of

introgression lines

Lines

Cell

number

observed

Average

chromosome

pair

Chromosome configuration and frequency

univalent ring bivalent rod bivalent trivalent quadrivalent

Ave %

Ave %

Ave %

Ave %

Ave %

SR1 163 20.58 0.23 1.13% 20.25 98.40% 0.08 0.39% 0.02 0.10% —

SR2 159 20.73 0.35 1.69% 20.04 96.67% 0.30 1.45% 0.02 0.10% 0.02 0.10%

SR3 127 20.73 0.16 0.77% 20.23 97.59% 0.32 1.54% 0.01 0.05% 0.01 0.05%

SR6 153 20.64 0.32 1.55% 18.87 91.42% 1.24 6.00% 0.09 0.44% 0.12 0.58%

SR10 149 20.72 0.19 0.92% 18.55 89.52% 1.82 8.78% 0.07 0.34% 0.09 0.43%

SR12 172 20.83 0.10 0.48% 19.12 91.79% 1.46 7.01% 0.02 0.10% 0.13 0.62%