sequencer (Le Signor et al. 2009 ). Potential
mutations are visualised as anomalous‘empty’
peaks on overlaid chromatograms using the
GeneMapper®software. This identifies a pool
containing DNA from a number of plants. The
plant carrying the mutation is pinpointed after
conventional PCR and sequencing of the indi-
vidual DNA samples that made up the relevant
pool followed by analysis using the software
MutationSurveyor®which can analyse multiple
sequence tracefiles simultaneously. Seed of the
plant is then sown, and progeny genotyped and
phenotyped to check or select as necessary
homozygous mutants.
21.5 Global Reach and Impact
While it was under development, Lotus TILL-
ING was available to relatively few researchers
within a collaborative programme of research on
the species. When this programme ended how-
ever, and the service was open to all, it rapidly
became globalised. To date, 32 Lotus groups
from 11 countries have used the service
(Table21.2). The depth and spread of TILLING
across the globe demonstrates that it was uni-
versally adopted as the method of choice for
reverse genetics inL. japonicus. To date, 163
genes have been targeted, some in both ecotypes.
These are listed in Table21.3together with the
types of mutation discovered. Due to the random
nature of the chemical mutagenesis, TILLING
can be used to target any sequence with the
expectation that mutations will be identified and
the table shows that this has been the case to
date. TILLING has thus had a significant impact
on numerous aspects of legume research, but
especially the legume–rhizobium symbiosis
(Perry et al. 2009 ); it was used very early on to
establish such genes asCASTORandPOLLUX
(Imaizumi-Anraku et al. 2005 ) as part of the
symbiosis signalling pathway and to obtain
numerous alleles at symbiotic loci (Perry et al.
2009 ). The random nature of the mutagen
ensures that mutations can be found in most
genes whether in nodule signal transduction, as
above, nodule metabolism (e.g. Horst et al.
2007 ), plant development (Welham et al. 2009 )
or indeed many processes for which it would be
very difficult to develop forward screens.
Although knockouts, through the generation of
premature stop codons or incorrect splicing, are
rare—about 5 % as calculated by McCallum
et al. ( 2000 ) and as found in practice—poten-
tially useful mutations are found at a much
higher frequency in most cases. TILLING has
been used, therefore, not only to identify gene
function, but also to confirm a particular pheno-
type (e.g. Yano et al. 2008 ; Krusell et al. 2011 )
where a knockout may not be required.
The impact of the mutations discovered from
our populations has not been confined to TILL-
ING, however. Researchers have used the mu-
tagenized seed for a number of forward screens,
from simple visual screens (e.g.flower develop-
ment; Dong et al. 2005 ), through microscopic
screens (e.g. Groth et al. 2013 ), to chemical
screens for secondary metabolism (e.g. Takos
et al. 2010 ). Most notable of these, however, was
the heroic effort made to isolate spontaneously
nodulating plants by Tirichine and co-workers
(Tirichine et al. 2007 ) using material from the
bulked populations of Gifu (Fig.21.1a).Table 21.2Research groups by country that have
accessed TILLING inLotus japonicus
No. of Research
groupsNo. of Genes
TILLed
Canada 1 5
Denmark 2 40
France 2 3
Germany 4 43
Greece 1 2
Italy 1 4
Japan 9 27
Spain 1 1
Sweden 1 3
UK 8 34
USA 2 2
Total 32 164 a
aA total of 163 genes has been TILLed, one twice21 TILLING inLotus japonicus 233