The Lotus japonicus Genome

intensity and analytical threshold, and 342 iSRMs
were successfully optimized, 88 of which were
significantly different between B-129 and MG-20
(Sawada et al. 2012 ).
Each iSRM is linked to MS2T data. Thus, in
principle, all the detectable metabolites can be
annotated on the basis of the putative structures
indicated by the MS/MS spectra. The annotation
rate of metabolites depends on the availability of
MS and MS/MS databases and the similarity of
search algorithm (Smith et al. 2005 ; Fiehn et al.
2005 , 2008 ; Wishart et al. 2007 , 2009 ; Cui et al.
2008 ; Soh et al. 2003 ; Mylonas et al. 2009 ). A
plant-specific MS/MS database and a search
algorithm established in 2012 will accelerate the
annotation rate of plant metabolomics data
(http://spectra.psc.riken.jp/).
The new version of widely targeted analysis
using iSRM was applied to RILs between MG-
20 and B-129 (Hayashi et al. 2001 ; Kawaguchi
et al. 2001 ; Klein and Grusak 2009 ; Gondo et al.
2007 ) to reveal eight mQTLs (LOD score[10).
Annotation of the peaks detected by iSRMs was
successful using ReSpect search and suggested
flavonoid glycosides. These results suggest that
widely targeted analysis with iSRM and MS2T-
based annotation by MS/MS database search
have the potential for effective elucidation of

plant metabolism. Sawada and Hirai ( 2013 )

developed iSRM for soya bean recombinant

inbred lines derived fromG. maxandG. sojaand

found several mQTLs.

16.6 Future Aspects

for Metabolomics

As practical metabolomics platforms, iSRM based

on MS2T and HR/MS of LC-FTICR-MS have

been established for quantitative and qualitative

analysis and for elemental composition analysis,

respectively (Nakabayashi et al.2013b). SRM,

iSRM, MS2T and HR/MS are now integrated for

quantitative and qualitative metabolomics. As an

application of the new metabolomics, mQTL

analysis of L. japonicushas the potential to

improve metabolite annotations. If an mQTL is

assigned to a biosynthetic gene already charac-

terized, the metabolite annotations could be vali-

dated by the gene function. The gene annotation

associated with metabolomics information could

narrow down the candidate structure of the

unknown metabolites, e.g., a cytochrome P450-

associated metabolite can be predicted to have

hydroxyl groups in its chemical structure. As a

source of high-density markers, single nucleotide

Fig. 16.2 Workflow of integrated metabolomics. On the
left side(black arrows), quantitative data can be obtained
by SRM and iSRM analyses. In case of annotation of
metabolite biosynthesis genes, data are used for mQTL
analysis (linkage mapping and GWAS). On theright side

(grey arrows), qualitative data can be obtained by MS2T

and HR/MS analysis. In case of chemical annotation of

detectable metabolites, data are used for searching

external databases. The goal of the integrated metabolo-

mics database is efficient mining of biological discoveries

178 Y. Sawada and T. Aoki