experimental reconstruction of double‐strand break repair ......non-mendelian inheritance of the...
TRANSCRIPT
Acc
epte
d A
rtic
le
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process, which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1111/tpj.13769
This article is protected by copyright. All rights reserved.
DR DONG WANG (Orcid ID : 0000-0003-4278-2728)
Article type : Original Article
Experimental reconstruction of double-strand break repair-mediated
plastid DNA insertion into the tobacco nucleus
Dong Wang1,2*
, Jinbao Gu2, Rakesh David
3, Zhen Wang
2, Songtao Yang
4, Iain R. Searle
3,
Jian-Kang Zhu2,5
and Jeremy N. Timmis3
1. Key Laboratory of Molecular Biology and Gene Engineering in Jiangxi Province, College
of Life Science, Nanchang University, Jiangxi, 330031, China
2. Shanghai Center for Plant Stress Biology, Shanghai Institutes for Biological Sciences,
Chinese Academy of Sciences, Shanghai 201602, China
3. Department of Genetics and Evolution, School of Biological Sciences, The University of
Adelaide, South Australia, Australia 5005.
4. Crop Research Institute, Sichuan Academy of Agricultural Sciences, Chengdu, 610066,
China
5. Department of Horticulture and Landscape Architecture, Purdue University, West
Lafayette, IN 47907, USA
Dong Wang: [email protected]
Jinbao Gu: [email protected]
Rakesh David: [email protected]
Iain R. Searle: [email protected]
Zhen Wang: [email protected]
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Songtao Yang: [email protected]
Jian-Kang Zhu: [email protected]
Jeremy N. Timmis: [email protected]
* Corresponding author: Dong Wang ([email protected]) telephone number: (86) 791
83969537
Summary
The mitochondria and plastids of eukaryotic cells evolved from endosymbiotic prokaryotes.
DNA from the endosymbionts has bombarded nuclei since the ancestral prokaryotes were
engulfed by a precursor of the nucleated eukaryotic host. An experimental confirmation
regarding the molecular mechanisms responsible for organelle DNA incorporation into nuclei
has not been performed until the present analysis. Here we introduced double-strand DNA
breaks into the nuclear genome of tobacco through inducible expression of I-SceI, and
showed experimentally that tobacco chloroplast DNAs insert into nuclear genomes through
double-strand DNA break repair. Microhomology-mediated linking of disparate segments of
chloroplast DNA occurs frequently during healing of induced nuclear DSBs but the resulting
nuclear integrants are often immediately unstable. Non-Mendelian inheritance of a selectable
marker (neo), used to identify plastid DNA transfer, was observed in the progeny of about
50% of lines emerging from the screen. The instability of these de novo nuclear insertions of
plastid DNA (nupts) was shown to be associated with deletion not only of the nupt itself but
also of flanking nuclear DNA within one generation of transfer. This deletion of pre-existing
nuclear DNA suggests that the genetic impact of organellar DNA transfer to the nucleus is
potentially far greater than previously thought.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Introduction
A pivotal feature of the evolution of eukaryotes is the formation of mitochondria and plastids,
the cytoplasmic organelles which originated from endosymbiotic prokaryotes. DNA from the
endosymbionts has bombarded nuclei since the ancestral prokaryotes were engulfed by a
precursor of the eukaryotic host (Bock and Timmis, 2008; Timmis et al., 2004). This resulted
in the relocation of many organellar genes to the nucleus, and numerous nuclear sequences
were created by insertion of organelle DNA (Martin et al., 2002; Noutsos et al., 2007). While
functional gene transfers from organelle to nucleus are relatively rare, non-functional
sequence transfer happens very often and most nuclear genomes are extensively punctuated
with organelle genome sequences (Matsuo et al., 2005). Real time DNA migration from
organelle to nucleus has been demonstrated in yeast (Thorsness and Fox, 1990) and tobacco
(Huang et al., 2003; Sheppard et al., 2008; Stegemann et al., 2003), and the frequency of this
movement is unexpectedly high. Organellar DNA fragments incorporate into nuclear DNA
through double-strand break (DSB) repair in yeast (Ricchetti et al., 1999) and tobacco (Wang
et al., 2012) but the molecular mechanisms responsible for insertion remain unclear.
Bioinformatic analysis suggested that nonhomologous end joining (NHEJ) is involved in
chloroplast DNA incorporation during DSBs repair (Wang and Timmis, 2013), but direct
experimental evidence was lacking.
De novo experimental nuclear DNA integrants originating from plastid genomes (nupts) are
often unstable. Non-Mendelian inheritance of the selectable marker (neo) used to identify
ptDNA transfers was observed in progeny of about half of the lines emerging from screens
(Huang et al., 2003; Wang et al., 2012). A detailed analysis demonstrated the physical loss of
the neo gene as the cause of instability of kanamycin resistance (Sheppard and Timmis,
2009). No explanations of instability other than deletion have been forthcoming.
In this study, we constructed a reporter system to examine whether DSB repair was involved
in chloroplast DNA insertion into the nuclear genome of tobacco. Three independent
chloroplast integrants were identified at repaired nuclear DSBs. DNA found at these loci
contained multiple disparate DNA sequences from the chloroplast genome linked together by
a process involving microhomology. Furthermore, the instability observed at these de novo
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
nupts was shown to be associated with deletion, within one generation of chloroplast DNA
transfer. Deletions occurred not only in the insert itself but also included flanking native
nuclear DNA.
Results
A genetic screen for nuclear DSB repair events that incorporate chloroplast DNA
In a pilot experiment to investigate the effect of induced DSBs on the frequency of plastid
DNA migration, the transplastomic tobacco (Nicotiana tabacum) line (tpGUS) containing a
gus gene driven by the 35S promoter (Sheppard et al., 2008) was irradiated with UV-C. In
tpGUS, the reporter gus gene is expressed only when it transfers to the nucleus and this event
can be monitored readily by histochemical staining to detect the gus product. The number of
blue sectors resulting from nuclear gus expression observed in seedlings treated with UV-C
was greater than that of control seedlings (Table S1). This tentatively suggested that
increasing DSB frequency increased plastid DNA insertion into the nucleus but the lack of
significance necessitated further experimentation. Therefore, to detect DSB repairs that
mediate chloroplast DNA insertion, an experimental system was developed in tobacco
(Figure 1A). The transplastomic line tp-neo (Huang et al., 2003) contains, in its plastid
genome, a neo gene driven by a eukaryotic promoter. It was therefore poorly expressed in
chloroplasts but strongly activated after transfer to the nuclear genome, conferring kanamycin
resistance. To the nucleus of this line was added an inducible DSB system (Lloyd et al.,
2012; Wang et al., 2012). In this added cassette, the gene encoding the rare cutting
endonuclease I-SceI was controlled by an alcA:35S promoter expressed after ethanol
treatment to cut two adjacent engineered target sites (Figure 1B). Consequently, the plant line
containing both nuclear and plastid transgenes was named pISceI-neo. Two lines of
pISceI-neo: pISceI-neo-9 and pISceI-neo-10, containing low copy numbers of T-DNA
monitored by bar gene quantification (see Figure S1), were used in a screen to monitor
insertion of neo from the transplastome at cut and repaired I-SceI sites (Figure 1). In
pISceI-neo-10, it was also determined that there was only a single nuclear locus containing
the two I-SceI sites (Table S2).
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
To evaluate the impact of DSB repair on the frequency of plastid-to-nucleus DNA transfer,
pISceI-neo-13 harbouring more than one locus was also used in the screen (Table S2). Leaf
explants of wild type (WT) tobacco were used as negative controls. These were all bleached
within two months when grown on regeneration medium supplemented with 400 mg/L
kanamycin (Figure 1C). Two week old pISceI-neo seedlings were sprayed with 0.7 M ethanol
and 4 days allowed for I-SceI expression, to generate DSBs and DSB repair (Lloyd et al.,
2012). After that, 340 and 400 leaf explants (~25 mm2) from pISceI-neo-9 and pISceI-neo-10
respectively were placed on regeneration medium containing 400 mg/L kanamycin
(Stegemann et al., 2003). In contrast to the control plants, kanamycin resistant shoots
appeared in explants of pISceI-neo leaves (Figure 1C). Six and eight resistant shoots were
generated from pISceI-neo-9 and pISceI-neo-10 lines respectively (Table 1) and these shoots
were designated the kr7 series. Based on cell number estimations (Lloyd and Timmis, 2011),
the occurrence of kanamycin resistant shoots in the screen indicated 1 neo transfer to the
nucleus in approximately 2.1 million and 1.9 million cells in ISceI-neo-9 and pISceI-neo-10
lines respectively, consistent with the plastid-to-nucleus DNA transfer frequency observed in
somatic cells from previous studies (Wang et al., 2012). For pISceI-neo-13, eleven resistant
shoots were generated in 440 leaf explants, indicating a neo transfer to the nucleus of one in
1.5 million cells (Table 1). This frequency is slightly higher than that in pISceI-neo-10,
suggesting that the significant increase results from the higher incidence of DSBs in the
pISceI-neo-13 nuclear genome.
Chloroplast DNA incorporation at nuclear DSB sites
Considering that earlier experimental chloroplast DNA integrants were characteristically
large (tens of kilobases or more in size) and consequently too large to be amplified by the
conventional PCR (Huang et al., 2004; Lloyd and Timmis, 2011), genome walking based on
the nuclear sequences near I-SceI site (Figure 1B) was carried out to characterise DSB
repairs. As expected, associated with the acquisition of kanamycin resistance, transplastomic
DNA was present after I-SceI cleavage and DSB repair in three independent plants obtained
in the screen (Figures 2A, 3 and 4A).
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Sequences adjacent to the repaired I-SceI junctions in kr7.1, kr7.2 and kr7.3 lines (1092, 588
and 1209 nucleotides respectively) revealed complex mosaics of diverse chloroplast
fragments with perfect sequence conservation compared with the plastome or transplastome
(Figure 2A). This sequence fidelity is consistent with the hypothesis that escaped chloroplast
DNAs from pISceI-neo integrated into the nuclear genome at cleaved I-SceI sites during DSB
repair. Besides chloroplast DNA fragments, some unknown sequences presumed to originate
from nuclear DNA were also observed at two of the three repaired sites (Figure 2A). An
intact 35S promoter and part of the neo gene was present in kr7.3 that clearly originated from
the pISceI-neo transplastome (Figure 2A). To exclude the possibility that the other two
chloroplastic integrants from kr7.1 and kr7.2 originated from pre-existing nupts, PCR
amplification was carried out with the primer pairs described in Figure 2A. Corresponding
amplicons were generated only from DNA templates from seedlings containing the
chloroplast integrants and not from WT plants (Figure 2B), confirming that fragments of
chloroplast DNAs had inserted de novo into the nuclear genome at experimentally induced
DSBs.
Upon close examination, the fusion points between chloroplast DNA sequences inside the de
novo nupts suggested that short microhomology was involved in their formation (Figure 3).
Microhomolgy was observed at both chloroplast/chloroplast and chloroplast/nuclear junction
sites, consistent with bioinformatic analyses (Wang and Timmis, 2013).
Instability of chloroplast integrants
In previous experiments, genetic instability was observed within one generation of insertion
in approximately half the kanamycin-resistant lines in a similar screen (Sheppard and
Timmis, 2009). Therefore, the stability of kanamycin resistance was tested in the three kr7
lines reported here by backcrossing them to WT females to replace their transplastomes with
WT and growing the progeny of individual capsules on medium containing kanamycin. Only
one of the three lines (kr7.3) exhibited, in all 8 capsules tested, the 1:1 ratio of
resistant:sensitive expected for a nuclear hemizygote (Table 2). All capsules of kr7.1 and four
of six capsules of kr7.2 presented a very substantial paucity of kanamycin resistant seedlings
(Table 2). No capsules showed an excess of resistant progeny.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
In a previous study, deletion of neo rather than epigenetic silencing was found to be
responsible for the loss of resistance (Sheppard and Timmis, 2009). Therefore, the integrity
of neo in the kr7 backcross progenies was first checked by employing a PCR primer pair
targeting the neo gene. As a positive control, the offspring of kr7.3, the plant that showed
consistent Mendelian segregation, was analyzed first. Because half of its backcross progeny
were expected to be kanamycin sensitive, twelve seeds from a capsule characterised
genetically were grown on medium without selection and their DNAs assayed individually by
PCR. Eight of the 12 samples yielded an amplicon (Figure 4B), consistent with a 1:1 ratio (p
= 0.24813). A similar result was observed (p = 0.24813, Figure 4C top panel), for seedlings
from the fourth capsule of kr7.2 (Table 2). In contrast, DNA from seedlings from a capsule of
kr7.1, which showed high instability of kanamycin resistance, yielded no amplicons of neo
(Figure 4D). These results are consistent with the previous study that determined deletion of
all or part of neo as the cause of loss of kanamycin resistance (Sheppard and Timmis, 2009).
The presence of the engineered I-SceI sites and their known flanking sequences allowed us to
examine the cause of instability of these de novo nupts in greater detail than has been
achieved in previous studies. Primers bar FP and bar RP; ST FP and ST RP (Figure 4A),
designed to amplify neighbouring regions at chloroplast DNA integration sites in nuclear
DNA, were used to assess the extent of sequence deletion when neo was lost. Seedlings from
capsules of kr7.3 and kr7.2 presenting normal segregation of kanamycin resistance gave the
expected products from PCR targeted to sections adjacent to the insertion site (Figure 4A, B
and C top panel), confirming the stability of sequences of both the chloroplast DNA integrant
and its nuclear flanking region. In contrast, seedlings from the fourth capsule of kr7.1 (Table
2), which showed a complete lack of kanamycin resistant progeny, yielded no PCR products
using these same primers (Figure 4A and D). Surprisingly, the sequence missing revealed in
the eighth seedling from the fifth capsule of kr7.2 (Table 2) was found in regions neighboring
the chloroplast DNA insertion site rather than in the neo gene itself (Figure 4A and C bottom
panel). To explore the loss of nucleotides in the nupt lacking neo from kr7.2, one primer
located inside the chloroplast integrant combined with a primer positioned in its nuclear
flanking region were employed (Figure 4A middle panel, P3R for kr7.3 and P2R for kr7.2),
and no products were forthcoming (Figure 4C bottom panel). In contrast, amplicons
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
generated by this strategy showed the same pattern as neo and other adjacent sequences
present in capsules exhibiting stable kanamycin resistance (Figure 4A, B and C top panel).
These results indicate that deletion occurs not only inside the chloroplast DNA integrant, but
also in the nuclear regions nearby. Finally, one more primer pair (DSBFP and DSBRP)
encompassing the positions of both I-SceI sites was used to further explore the nuclear
instability near the nupt (Figure 4A). In this strategy, amplicons were expected only in
pISceI-neo lines, not in plants of the kr7 lines showing kanamycin resistance, emphasizing
again that the sizes of these de novo nupts are extremely large. As for seedlings showing no
amplicons of neo or its nearby region, no PCR products were observed with the primer pair
DSBFP and DSBRP (Figure 4C bottom panel and D) suggesting, in common with all other
PCR results, that large deletions are possible in the nuclear regions flanking some nupts.
Discussion
In this study, we demonstrate that chloroplast DNA inserts into the nuclear genome during
DSB repair in tobacco and these inserts are comprised predominantly of complex mixtures of
disparate regions from the chloroplast genome. The integrants may include other, presumed
nuclear DNA sequences and all are joined together through microhomology (Figures 2A and
3). These findings support and extend the previous suggestion that end-joining of linear DNA
fragments creates mosaic integrants prior to, or during, insertion into the nuclear genome
(Noutsos et al., 2005). These novel chloroplast integrants are large such that they cannot be
amplified in their entirety by standard PCR methods (Figure 4B-D). In addition, three
chloroplast integrants inserted into an I-SceI site, and an intact I-SceI site remained
unchanged (Figure 4A). This observation is different from DSB repair-meditated
mitochondrial DNA insertion events (Wang et al., 2012), in that HincII was not used to
remove the stuffer that contains three HincII sites here as it was in the previous study (Lloyd
et al., 2012). After insertion, sequence loss may occur within the de novo nupts and also in
flanking nuclear DNA segments within one generation of their formation, explaining the
genetic instability that characterises about half of the de novo nupts.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
It appears that primary insertions of organelle DNA are initially large but, in many cases,
they are reduced to shorter stretches over very short time periods. However, it remains a
possibility that the large nupts are a feature of the experimental selection for the entire neo
gene and may not represent the full spectrum of events.
Nucleotide substitutions were not detected as all the nupt sequences remain identical to the
corresponding regions of the plastome. However, some nupts clearly survive for long periods
of evolutionary time and substitutions are seen at these older loci (Ayliffe and Timmis, 1992;
Richly and Leister, 2004; Michalovova et al., 2013; Yoshida et al., 2014; Chen et al., 2015).
Our results suggest that scrambled nuclear organellar DNAs are generated through deletion
during the early stages of post-integration rather than by longer term sequence decay. Of
course, several to many somatic cell divisions must have occurred before our attempts to
characterise nupt sequences and organization. Thus we may be reporting their organization
after an initial period of great flux.
The novel demonstration that incorporation of chloroplast DNA can be accompanied by
deletion of adjacent non nupt DNA, indicates that the process must be an important player in
dynamic nuclear genome evolution. Because analyses of instability within nupts in a previous
study relied on the kanamycin resistance provided by neo (Sheppard and Timmis, 2009),
deletion of non nupt DNA also indicates that the frequency of deletion occurring in nupts was
underestimated. Considering the high frequency of DNA movement from plastid to nucleus
(one event in between 11000 or 16000 pollen grains (Huang et al., 2003; Sheppard et al.,
2008)), high incidence of sequence deletion within nupts not only counterbalances the
resulting problem of increasing genome size, but also provides more chance for prokaryotic
plastid genes to gain the functional elements required for nuclear expression. Moreover, the
intraspecific variation of nupts observed in the progeny (Figure 4C bottom panel) suggests
that the majority of sequence deletion occurs during mitosis rather than meiosis. It is formally
possible that these deletions may be different in each seed of a capsule exhibiting kanamycin
sensitivity (Figure 4D), but it is more likely that local sequence contexts determine and
orchestrate a restricted range of deletion outcomes.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Although the extent of deletion in nupts identified here was characterized, the underlying
mechanisms are still obscure. Direct repeats are thought to be associated with deletions
occurring in nupt DNA (Noutsos et al., 2005), and candidate short repeats were found in
kr7.3 (Figure 2A). Third generation sequencing technology with long read lengths is required
for a thorough understanding of the instability of these nascent nuclear organelle DNAs.
Experimental procedures
Plant growth conditions
Nicotiana tabacum plants were grown with 16 hour light/8 hour dark cycle at 25 °C in jars or
11 cm petri dishes containing half MS medium. Soil grown plants were grown in a growth
room with a 14 hour light/10 hour dark cycle and 25 °C day/18 °C night regime. UV-C
treatment was performed as previously described (Rosa et al., 2013).
Generation of pISceI-neo
pISceI-neo contains a previously-described chloroplast genome insert that is expressed as
kanamycin resistance only after transfer to the nucleus (Huang et al., 2003) and an inducible
DSB system in the nucleus (Figure 1A). The nuclear inducible DSB system was as described
(Lloyd et al., 2012) with the neo gene replaced by a selectable marker bar and a stuffer
fragment of known sequence (Figure 1B).
Experimental induction of DSBs and kanamycin selection
Induction of I-SceI was described previously (Lloyd et al., 2012). Briefly, 2-wk-old seedlings
grown in jars were sprayed with 0.7 M ethanol and four days allowed for producing DSBs.
Leaf explants (~25 mm2) were then placed on 11 cm petri dishes containing regeneration
medium containing 400 mg/L kanamycin. Resistant shoots were moved to 1/2 MS medium to
root and later transferred to soil.
Seedling selection was performed by growing surface-sterilized seeds on half MS medium
supplemented with 150 mg/L kanamycin (Wang et al., 2012), and plates were placed at 25 °C
with a 16 hour light/8 hour dark cycle. The significance of deviation from the expected
Mendelian ratio was calculated by a χ2 test.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
DNA preparation, PCR assays and genome walking
DNA was prepared employing a DNeasy Plant Mini Kit (Qiagen), and PCR amplification
was performed using Taq DNA polymerase (Takara) according to standard protocols.
TaqMan real-time PCR was performed as described previously (Bubner and Baldwin, 2004).
Genome walking utilized a GenomeWalker Universal Kit (Clontech) in accordance with the
manufacturer’s instructions. Details of the primers used are listed in Table S3. Sequence data
for this article can be found in the GenBank data libraries under accession numbers from
KU240015 to KU240017.
Acknowledgments
We thank Mathieu Rousseau-Gueutin and Yuan Li for helping with plant transformation and
TaqMan real-time PCR, respectively. This work was funded by the National Science
Foundation of China (31401077), Natural Science Foundation of Jiangxi Province
(20171ACB20001) and a Nanchang University Seed Grant for Biomedicine
(9202-0210210807). The authors declare no conflict of interest.
Author Contributions
D.W. and J.N.T conceived the study, supervised the experiments, and wrote the article; D.W.,
J.B.G., Z.W., S.T.Y. and R.D. performed the experiments; I.R.S., D.W., J.K.Z and J.N.T.
contributed new reagents and analytic tools; D.W., J.B.G. and J.N.T analyzed data.
List of abbreviations
Double-strand break- DSB
nuclear insertions of plastid DNA- nupts
Nonhomologous end joining- NHEJ
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Short Supporting Information Legends
Figure S1. The comparative copy number of bar in pISceI-neo lines as determined by
real-time quantitative PCR. The least value of copy number is appointed as one.
Table S1. Plastid-to-nucleus DNA transfer in tpGUS seedlings.
Table S2. Segregation of herbicide Basta resistance of self crossed pISceI-neo-9,
pISceI-neo-10 and pISceI-neo-13 lines.
Table S3. Primers used in this study.
References
Ayliffe, M.A. and Timmis, J.N. (1992) Tobacco nuclear DNA contains long tracts of
homology to chloroplast DNA. Theor Appl Genet, 85, 229-238.
Bock, R. and Timmis, J.N. (2008) Reconstructing evolution: gene transfer from plastids to
the nucleus. Bioessays, 30, 556-566.
Bubner, B. and Baldwin, I.T. (2004) Use of real-time PCR for determining copy number
and zygosity in transgenic plants. Plant Cell Rep, 23, 263-271.
Chen, H., Yu, Y., Chen, X., Zhang, Z., Gong, C., Li, J. and Wang, A. (2015) Plastid DNA
insertions in plant nuclear genomes: the sites, abundance and ages, and a predicted
promoter analysis. Funct Integr Genomics, 15, 131-139.
Huang, C.Y., Ayliffe, M.A. and Timmis, J.N. (2003) Direct measurement of the transfer
rate of chloroplast DNA into the nucleus. Nature, 422, 72-76.
Huang, C.Y., Ayliffe, M.A. and Timmis, J.N. (2004) Simple and complex nuclear loci
created by newly transferred chloroplast DNA in tobacco. Proc Natl Acad Sci U S A,
101, 9710-9715.
Lloyd, A.H. and Timmis, J.N. (2011) The Origin and Characterization of New Nuclear
Genes Originating from a Cytoplasmic Organellar Genome. Molecular Biology and
Evolution, 28, 2019-2028.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Lloyd, A.H., Wang, D. and Timmis, J.N. (2012) Single molecule PCR reveals similar
patterns of non-homologous DSB repair in tobacco and Arabidopsis. PLoS One, 7,
e32255.
Martin, W., Rujan, T., Richly, E., Hansen, A., Cornelsen, S., Lins, T., Leister, D.,
Stoebe, B., Hasegawa, M. and Penny, D. (2002) Evolutionary analysis of
Arabidopsis, cyanobacterial, and chloroplast genomes reveals plastid phylogeny and
thousands of cyanobacterial genes in the nucleus. Proc Natl Acad Sci U S A, 99,
12246-12251.
Matsuo, M., Ito, Y., Yamauchi, R. and Obokata, J. (2005) The rice nuclear genome
continuously integrates, shuffles, and eliminates the chloroplast genome to cause
chloroplast-nuclear DNA flux. Plant Cell, 17, 665-675.
Michalovova, M., Vyskot, B. and Kejnovsky, E. (2013) Analysis of plastid and
mitochondrial DNA insertions in the nucleus (NUPTs and NUMTs) of six plant
species: size, relative age and chromosomal localization. Heredity, 111, 314-320.
Noutsos, C., Kleine, T., Armbruster, U., DalCorso, G. and Leister, D. (2007) Nuclear
insertions of organellar DNA can create novel patches of functional exon sequences.
Trends Genet, 23, 597-601.
Noutsos, C., Richly, E. and Leister, D. (2005) Generation and evolutionary fate of
insertions of organelle DNA in the nuclear genomes of flowering plants. Genome Res,
15, 616-628.
Ricchetti, M., Fairhead, C. and Dujon, B. (1999) Mitochondrial DNA repairs
double-strand breaks in yeast chromosomes. Nature, 402, 96-100.
Richly, E. and Leister, D. (2004) NUPTs in sequenced eukaryotes and their genomic
organization in relation to NUMTs. Molecular Biology and Evolution, 21, 1972-1980.
Rosa, M., Von Harder, M., Cigliano, R.A., Schlogelhofer, P. and Scheid, O.M. (2013)
The Arabidopsis SWR1 Chromatin-Remodeling Complex Is Important for DNA
Repair, Somatic Recombination, and Meiosis. Plant Cell, 25, 1990-2001.
Sheppard, A.E., Ayliffe, M.A., Blatch, L., Day, A., Delaney, S.K., Khairul-Fahmy, N.,
Li, Y., Madesis, P., Pryor, A.J. and Timmis, J.N. (2008) Transfer of plastid DNA
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
to the nucleus is elevated during male gametogenesis in tobacco. Plant Physiol, 148,
328-336.
Sheppard, A.E. and Timmis, J.N. (2009) Instability of plastid DNA in the nuclear genome.
PLoS Genet, 5, e1000323.
Stegemann, S., Hartmann, S., Ruf, S. and Bock, R. (2003) High-frequency gene transfer
from the chloroplast genome to the nucleus. Proc Natl Acad Sci U S A, 100,
8828-8833.
Thorsness, P.E. and Fox, T.D. (1990) Escape of DNA from mitochondria to the nucleus in
Saccharomyces cerevisiae. Nature, 346, 376-379.
Timmis, J.N., Ayliffe, M.A., Huang, C.Y. and Martin, W. (2004) Endosymbiotic gene
transfer: organelle genomes forge eukaryotic chromosomes. Nature Reviews Genetics,
5, 123-135.
Wang, D., Lloyd, A.H. and Timmis, J.N. (2012) Environmental stress increases the entry of
cytoplasmic organellar DNA into the nucleus in plants. Proc Natl Acad Sci U S A,
109, 2444-2448.
Wang, D. and Timmis, J.N. (2013) Cytoplasmic organelle DNA preferentially inserts into
open chromatin. Genome Biol Evol, 5, 1060-1064.
Yoshida, T., Furihata, H.Y. and Kawabe, A. (2014) Patterns of genomic integration of
nuclear chloroplast DNA fragments in plant species. DNA research, 21, 127-140.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Tables
Table 1. Frequency of DNA transfer from plastid to nucleus observed in pISceI-neo-9,
pISceI-neo-10 and pISceI-neo-13 lines.
Leaf explants Resistant shoots Estimated transfer rate
pISceI-neo-9 340 6 ~2.1×106
pISceI-neo-10 400 8 ~1.9×106
pISceI-neo-13 440 11 1.5×106
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Table 2. Segregation of kanamycin resistance of progenies from the three kr7 lines
backcrossed to female WT plants.
Lines Capsule Resistant Sensitive P
kr7.1 1 2 168 ***
2 3 88 ***
3 7 176 ***
4 1 164 ***
5 5 76 ***
kr7.2 1 21 96 ***
2 50 143 ***
3 29 110 ***
4 85 83 NS
5 9 53 ***
6 22 30 NS
kr7.3 1 58 55 NS
2 21 24 NS
3 47 43 NS
4 50 51 NS
5 49 43 NS
6 59 57 NS
7 44 42 NS
8 56 52 NS
Seeds from individual capsules were grown on half MS medium supplemented with 150 mg/L of
kanamycin. The p values signify deviation from a 1:1 ratio of resistant:sensitive (χ2 test). NS, not
significant; ***, p < 0.001.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Figure legends
Figure 1. A genetic screen for DNA transfer from chloroplast to nucleus through DSB repair.
(A) Structure of tp-neo. The neo and aadA genes are under CaMV 35S and psbA promoter
control, respectively (Huang et al., 2003). The black box represents the STLS2 intron. (B)
Schematic representation of the DSB system in the nucleus. The top panel presents vector
T-DNA harbouring a hygromycin selectable marker gene (hyg), an AlcR gene and the I-SceI
gene driven by an ethanol inducible promoter (Lloyd et al., 2012). The bottom panel shows
vector T-DNA holding a herbicide resistant gene (bar) and a stuffer fragment of known
sequence flanked by two I-SceI target sites. LB, left border; RB, right border of the
Agrobacterim transformation system. (C) Stable plastid-to-nucleus transfer. Leaf explants
(~25 mm2) were grown on regeneration medium supplemented with 400 mg/L kanamycin.
As resistant plants emerged (as presented in the right hand panel), they were transferred to
1/2 MS media to root and subsequently moved in soil. After 2 months, all leaf explants from
Petite Havana (WT) were totally bleached, and the screen ended at this time point. Plates are
11 cm petri dishes.
Figure 2. Chloroplast DNAs inserted at repaired I-SceI sites. (A) Schematic representations
of partial chloroplast integrants at DSB repair junctions. DNA from three seedlings contained
mosaic integrants comprised of different chloroplast and transplastomic DNA fragments
(different shades of green, shades are not scale). Numbers in the green boxes indicate
corresponding nucleotides of the native tobacco chloroplast genome (Shinozaki et al., 1986).
Grey boxes represent presumed nuclear DNA. The T-DNA vector flanking sequence of
I-SceI restriction site is shown in black. The three primer pairs shown (P1F and P1R, P2F and
P2R, P3F and P3R for kr7.1, kr7.2 and kr7.3, respectively) were designed to amplify specific
chloroplast integrants in the nucleus. (B) The de novo experimental nupts are not present as
nupts elsewhere in WT tobacco. One section of the tobacco nuclear genome amplified by
primers from a recent study (Lloyd and Timmis, 2011) was used as a positive control.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Figure 3. Microhomology involved in nupt-mediated DSB repair. Alignments show nupts in
three kr7 lines, their corresponding nuclear sequence in pISceI-neo (pISceI-neo_N), and the
chloroplast DNA sequences (tabacum_pt) or transplastomic DNA sequence (pISceI-neo_tp).
Short microhomology at fusion points is highlighted in red, green is used to mark DNA
originating from chloroplast or transplastome, and the unidentified DNA sequence in nupts is
shown in blue. Numbers inside two slashes indicate nucleotides in nupts that are identical to
chloroplast DNA or unidentified DNA origins.
Figure 4. PCR analysis of kr7 lines showing genetic instability for kanamycin resistance. (A)
Schematic representations of nupt flanking sections. Primers used are labelled and their
direction shown arrowed. Primer pairs bar FP and bar RP, ST FP and ST RP cover most of
the bar gene and the stuffer fragment, respectively. The primer pair DSBFP and DSBRP
amplifies a region that includes the two I-SceI sites. P3R and P2R are primers specific for
chloroplast integrants in kr7.3 and kr7.2, respectively. Pnos and Tnos signify the nos
promoter and nos terminator respectively. (B, C and D) PCR analysis of backcross progeny
from three kr7 lines. DNAs prepared from the corresponding pISceI-neo seedlings were
applied in the reaction. The abbreviations kr7.3_K+, kr7.2_K+ and kr7.3_K+ stand for
kanamycin resistant seedlings from each line. The primer pair used as a control is described
in Figure 2B. Ve-, no template.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.