High throughput sequencing of flanking
regions of T-DNA inserts
Isolation of regions flanking the left and right
border of T-DNA inserts was performed using an adapter-anchor
PCR method Figure
8 (Siebert et al 1995 Nucleic Acids Res. 23, 1087-1088
). This method relies on the use of a blunt end restriction enzyme, a specific
adapter and two steps of nested PCR amplification using primers specific
to the adapter and the T-DNA (Sallaud
et al 2003 Theor Appl Genet, 106:1396-1408,
Sallaud et al. Plant J, 39:450-464). The method proved to be very
efficient for large scale isolation of T-DNA flanking sequences in
Arabidopsis (Ortega et al, 2002: C.R. Acad. Sci.
serie III, Paris 325 :773-80 ; Balzergue
et al 2001: Biotechniques 30, 496-8, 502,
504 ).
As illustrated in the flow chart Figure
1, DNA was extracted from tissues of young
shoots of regenerating plantlets before their transfer to test tubes
for further development. Isolation of regions flanking the T-DNA left
border at the integration site was performed separately from DraI
and SspI digests for each DNA sample. Plants
corresponding to tracks exhibiting a unique PCR2 product on agarose gels with either DraI
or SspI were first analyzed since this unique
PCR2 product can be directly sequenced without any additional purification
step. A unique PCR2 product is observed in 35% and 41% of the DNA
samples following digestion with DraI and
SspI respectively, whereas 17% of the plants
exhibited a unique fragment for both enzymes. For the right
border, flanking regions were first amplified using EcoRV. The samples not yielding a PCR product
(ca 50%) were then PCR amplified using Ssp1
(For more information about walk-PCR amplification
of flanking regions of T-DNA inserts download pdf
of the protocol)
Results of sequencing a PCR2 products using
a T-DNA LB and RB specific primers are summarized in Table
3. Frequent occurrence of tandem T-DNA and/or binary vector
sequences appear a major limitation of the system. Therefore,
when possible, sequencing the PCR product before the transfer of the
plants to the greenhouse could further improve the value of a T-DNA
library for reverse genetics.
27,717 PCR products yielded good
sequences larger than 30bp (average length 250 bp). 13% of the LB FST
sequences proved to be redundant due to parallel amplification from
the border of the same T-DNA insert in both DraI and SspI digests.
Six percent of sequences where found redundant. They may result from
the presence of a few clonal lines deriving from the same transformation
event most likely resulting from the rare fragmentation of a hygromycin-resistant
cell line and further subculturing of callus pieces. Contamination during DNA extraction
or PCR experiments could also be another source of redundancy.
96% of the good sequences
were assigned to at least one position in the IRGSP Nipponbare
genome sequence. T-DNA insertions appear to be randomly distributed
over the 12 chromosome pseudomolecules,
followed that of the predicted coding sequences with a lower insertion
density around the centromere region and a higher density in the subtelomeric regions. T-DNA inserts were found to rarely
integrate into repetitive sequences (Figure 9).
A detailed analysis of
Chromosome 1 integrations showed that 47% of the T-DNA inserted within
an interval extending from -250 bp upstream
the ATG to the STOP codon of predicted genes,
thereby generating reliable knock outs. Preferential insertion was
also observed within the first 250 bp upstream
the putative ATG start codon. Using 4 kb of sequences surrounding the insertion
points, 62% of the sequences showed significant similarity to gene
encoding known proteins (E-Value <1.00E-05).
Overall, 30% and 10%
of the T-DNA FSTs had a significant hit on the whole population of full length cDNAs (n= 28,000) (Kikuchi et al 2003) or a sub population
(n= 9,734) successfully classified in Gene Ontology biological functions.
(For more details about recovery of flanking regions download a poster presented at
5th Rice Genetics Symposium, Maniela, Philippines download pdf
or see Sallaud et al, (2004) Plant J.
, 39: 450-464)
High thoughput sequencing of flanking regions of new Tos 17 inserts
In a collaborative project with the french national sequencing center (genoscope)
we have successfully sequenced 15,000 flanking regions of new Tos17 inserts in the
T-DNA insertion library. A novel method of high-throughput selective amplification
of flanking regions of newly transposed Tos17– called TOSTRAP – has been developed
in that aim (Piffanelli et al. 2007, Plant Molecular Biology). This walk-PCR based
protocol takes advantage of the absence of EcoRV restriction site in the vicinity of
the right border of Tos17 copies residing on chromosomes 7 and 10 in cv. Nipponbare, thereby generally
preventing PCR amplification of regions flanking these resident copies (Figure 10).
Using the TOSTRAP protocol we amplified flanking sequences from at least one newly-transposed
copy of Tos17 in 70% of the insertion lines. We typically generated PCR fragments of size ranging
from 300 bp to 2000 bp (average size 700 bp) that contain a TAG sequence that enable the use of one
primer to sequence all the generated PCR products. The TOSTRAP protocol enables to amplify multiple
PCR products from the same DNA sample in 25% of the analysed rice lines. This implies that 75% of our
products are single PCR products and 25% multiple PCR products (in most cases 2 PCR products) (Figure 11).
Moreover, Sequences tagging two independent new inserts can be frequently retrieved from the sequencing
of multiple PCR fragments (Figure 12).
