0 2 10 16 204 6 8 12 14 18

Retention time (min)

Rel

ativ

e in

ten

sity

LeafESI (+)ESI (−)

Supplementary Figure 1 Total ion current chromatogram of lipids from rice leaves.

Supplementary Figure 1

PC

PI

PE

DGDGPG

SQDG

GlcCer

MGDG

SG

Supplementary Figure 2. Structural characterization of triterpenoid saponins. (A) Positive MS/MS of tomatine, and (B) avenancin A-1 are displayed. The m/z values of precursor ions are shown in parentheses. Gal, galactose; Glc, glucose; Xyl, xylose; Ara, arabinose; N-MeAnth, N-methylanthranilic acid.

A

Rel

ativ

e I

nte

nsi

ty (

%)

B

Rel

ativ

e I

nte

nsi

ty (

%)

300 500 600 800 900 1000 1100700400

m/z

m/z

100

75

50

25

0

100

75

50

25

0

m/z 638.404

m/z 932.496m/z 469.330m/z 770.439m/z 451.319

m/z 423.319

Glc(⊿162)

Glc(⊿162)

Ara(⊿132)

N-MeAnth(⊿151)

(m/z 1094.549)

(m/z 1034.535)

300 500 600 800 900 1000 1100700400

Glc(⊿162)

Xyl(⊿132)

Glc(⊿162)

Gal(⊿162)

m/z 578.403

m/z 416.353

m/z 902.515m/z 740.458

Supplementary Figure 2

Supplementary Figure 3

Supplementary Figure 3 TGDG in tgd3 mutant.(A) Positive MS of TGDG in tgd3 mutant (tR 12.5 min). (B) Analysis of 34:6 TGDG in the wild type and tgd3 mutant. The intensity of extracted ion chromatogram of [M+NH4]+ of 34:6 TGDG in the tgd3 is set as 100%. The baseline is shifted for convenience.(C) Heat map expression of profiles of molecular-related ions of TGDG in wild type and tgd3 mutant. Each colored bar within a column represents an ion detected by positive LC-MS in the crude lipid extracts of the wild type and tgd3 mutants (biological replicates = 4). Peak picking and alignments were performed using Shimadzu Profiling Solution software (tolerance for m/z variation, 20 mDa; noise level, 2e5; tolerance for tRvariation, 0.1 min). Deisotope and deconvolution procedures were omitte. Peak intensities were normalized based on [M+H]+ of internal standard. The color of each bar represents the z-scored intensity of the ions of the corresponding lipid species.

m/z 1089.678, [M+NH4+1]+ of 34:6 TGDGm/z 1093.634, [M+Na]+ of 34:6 TGDGm/z 1088.671, [M+NH4]+ of 34:6 TGDG

Col-0 tgd3

-2.9 1.0 2.9

0 2 10 16 204 6 8 12 14 18

Retention time (min)

Rel

ativ

e in

ten

sity

tgd3

WT

B

34:6 TGDG

A

Rel

ativ

e I

nte

nsi

ty (

%)

m/z

100

75

50

25

0

m/z 1088.677, [M+NH4]+ of 34:6 TGDG

m/z 1109.600,[M+K]+ of 34:6 TGDG

m/z 1093.628, [M+Na]+ of 34:6 TGDG

C

900 950 1000 1050 1100 1150 1200 1250

not detected

Supplementary Figure 4. Structural characterization of two types of 34:6 DGDG (A) Negative MS/MS of 34:6 Gal(1→6)GalDG and (B) 34:6 Gal(1→6)GalDG are displayed. The m/z values of precursor ions are shown in parentheses

m/z

A

Rel

ativ

e I

nte

nsi

ty (

%)

100

75

50

25

0300 400 600 700 1000500

2×Gal–2×H2O

(⊿324)

800 900

m/z 907.520, [M–H]–m/z 629.312

m/z

B

Rel

ativ

e I

nte

nsi

ty (

%)

100

75

50

25

0300 400 600 700 1000500

2×Gal–2×H2O

(⊿324)

800 900

m/z 907.521, [M–H]–

m/z 629.314

(m/z 953.5, [M+HCO2]– )

(m/z 953.5, [M+HCO2]– )

Supplementary Figure 4

Supplementary Figure 5. OPLS-DA of lipidome data of ats1-1 and wild type. (A) Score scatter plot (R2Y=0.996, Q2=0.990, R2X=0.698). The confidence intervals correspond to the 2 or 3 sigma limits, i.e., 2 or 3 standard deviations of the vectors is displayed. (B) S-plot. The variables that changed most and have a strong contribution for class separation are circled. Details of variables circled in upper right and lower left are shown in Supplementary Tables 2 and 3.

B

-40

-30

-20

-10

0

10

20

30

40

1 2 3 4 5 6 7 8

Num

2 SD

2 SD

3 SD

3 SD

fad6-1

Col-0Pre

d.

Com

p. 1

A

Supplementary Figure 5

0.1 0.20-0.3 -0.1-0.2

0.0

0.2

0.4

0.6

0.8

1.0

-0.2

-0.4

-0.6

-0.8

-1.0

p(c

orr)

w[1]P

Supplementary Table 2

Supplementary Table 3

Supplementary Figure 6

Supplementary Figure 6. OPLS-DA of lipidome data of ats1-1 and wild type. (A) Score scatter plot (R2Y=0.990, Q2=0.985, R2X=0.736). The confidence intervals correspond to the 2 or 3 sigma limits, i.e., 2 or 3 standard deviations of the vectors is displayed. (B) S-plot. The variables that changed most and have a strong contribution for class separation are circled. Details of variables circled in upper right and lower left are shown in Supplementary Tables 4 and 5.

10 200-0.3 -0.1-0.2

B

-40

-30

-20

-10

0

10

20

30

40

1 2 3 4 5 6 7 8

Num

2 SD

2 SD

3 SD

3 SD

fad7-2

Col-0Pre

d.

Com

p. 1

A

0.0

0.2

0.4

0.6

0.8

1.0

-0.2

-0.4

-0.6

-0.8

-1.0

p(c

orr)

w[1]P

Supplementary Table 4

Supplementary Table 5