reSeArcH Article
genomic DNA and the 2-kb promoter region were amplified by PCR from cDNA
and genomic DNA, respectively. The DNA fragment and the promoter region
were cloned into the pENTR vector (Invitrogen). Coding sequences were trans-
ferred from the entry clones to gateway-compatible destination vectors. Transgenic
Arabidopsis lines were generated by Agrobacterium-mediated transformation^54 ,
and homozygous transgenic T3 lines carrying a single insertion were used. The
moca1-2 (SALK_131321) and moca1-3 (GK-856G03) lines were obtained from
the Arabidopsis Biological Resource Center (ABRC). Both insertions are located
within the gene. moca1-2 is found in the fourth exon, and moca1-3 is found in the
eighth exon. We were unable to obtain homozygotes for either mutant allele. Note
that moca1 is moca1-1 in this study.
MOCA1–GFP subcellular localization analysis. For analysis of MOCA1–
GFP in Arabidopsis seedlings, MOCA1 promoter-driven wild-type MOCA1
(pMOCA1::MOCA1–GFP; full-length genomic DNA) and mutant MOCA1
(pMOCA1::mMOCA1–GFP, full-length genomic DNA) and 35S promoter-driven
MOCA1 (p35S::MOCA1–GFP, full-length complementary DNA) transgenic plants
were generated as described previously^17. Seedlings grown in ½ MS medium in
Petri dishes for seven days were subjected to GFP confocal imaging with the Zeiss
LSM 510 microscope or whole-seedling imaging with a Zeiss SteREO Discovery
V20/V16 microscope. Data represent more than ten independent lines examined,
which displayed similar GFP subcellular localization. We analysed the fluorescence
in the pMOCA1::MOCA1–GFP and the Golgi maker mCherry–Golgi co-transgene
lines^68 , and observed that MOCA1–GFP displayed punctate patterns in the cytosol,
and was co-localized with the mCherry–Golgi marker. The Golgi localization is
consistent with the prediction by SUBA4 (http://suba.live/)^69 , and well supported
by several previous studies on Golgi proteomes^70 ,^71.
Histochemical GUS activity analysis. Histochemical staining for GUS activity
using the MOCA1 promoter-driven GUS (pMOCA1::GUS) transgenic lines was
performed as described previously^17. Seedlings grown in ½ MS medium or soil
were used for histochemical staining^72. Data represent six independent lines exam-
ined, which displayed similar staining patterns.
PCR primers and vectors. Genotyping primers: MOCA1-LP, 5′-CATTC
TTGTCCTTAT AGTTGCTGGT; MOCA1-RP, 5′-CTTTCAAGAACCATCTCA
CCGC. Cloning primers: MOCA1-cDNA_Fw, 5′-CACCATGGTG AGACTC
AAGACGAGT; MOCA1-cDNA_Rev, 5′-(TCA)ACAGAGGAAACATA GGGAAT;
Primers for MOCA1 promoter: MOCA1P_Fw, 5′-CACCGGATAATGTT
GAGTAATGG; MOCA1P_Rev, 5′-ACCTTTGTTCCTTGTCACCG. Vectors:
p35S::MOCA1:pGWB502Ω, pMOCA1::GUS:pGWB533, p35S::MOCA1-
GFP:pGWB405, pMOCA1::MOCA1-GFP:pGWB405, pMOCA1::MOCA1-GFP:
pGWB404, and pMOCA1::mMOCA1-GFP:pGWB404.
Sphingolipid analysis. Arabidopsis GIPCs were extracted and purified from 7-day-old
seedlings as described previously with modifications^23 ,^50 ,^73. In brief, seedlings
(∼20 g fresh weight) were blended with 300 ml cold 0.1 N aqueous acetic acid.
The slurry was filtered through eight layers of miracloth, and the residue was then
re-extracted with hot (70 °C) 70% ethanol containing 0.1 N HCl. The filtrates were
left immediately at − 20 °C overnight. The precipitate was pelleted by centrifuga-
tion at 2,000g at 4 °C. The GIPC-containing pellet was washed with cold acetone,
and subsequently with cold diethyl ether to yield a whitish precipitate. The GIPC
crude extracts were dissolved in tetrahydrofuran (THF)/methanol/water (4:4:1,
v/v/v) containing 0.1% formic acid at 60 °C, further dried, and submitted to a
butan-1-ol/water (1:1, v/v) phase partition. The upper butanolic phase was dried
and the residue was dissolved in THF/methanol/water (4:4:1, v/v/v) containing
0.1% formic acid. The GIPC solutions were analysed using a saturated solution
of 2,6-dihydroxy-acetophenone (DHA) matrix that was prepared in 50:50 (v/v)
ethanol/water containing 3 mM ammonium sulfate. The concentration of GIPC
samples was ∼0.5 mg/ml in THF/methanol/water (4:4:1, v/v/v) containing 0.1%
formic acid. Samples were mixed with matrix solution (0.5 μl) before being loaded
on the MALDI plate. Spectra were acquired in negative ion mode on a MALDI
Q-TOF mass spectrometer (AB SCIEX TOF/TOF 5800). The laser was set to an
energy level of 250 on the instrument scale. The mass range was from m/z 800 to
3,000. The diagram of plant GIPC structure shown in Extended Data Fig. 8h was
adapted from previous studies^23 ,^27 ,^30.
Protoplast isolation and ζ potential measurement. The ζ potential measurements
of Arabidopsis mesophyll protoplasts were carried out using similar methods to
those described previously^74 –^76. Upper epidermis samples from four-week-old
Arabidopsis leaves were vacuum infiltrated for 10 min and then incubated for 90
min in enzyme solutions containing 1% (w/v) cellulase (Onozuka R-10), 0.4%
(w/v) macerozyme (Onozuka R-10), 0.5 M mannitol, and 5 mM MES-KOH, pH
5.7. Released protoplasts were filtered through a 100-μm nylon mesh and washed
twice in solution without enzymes^77 ,^78. For the ζ potential measurements, proto-
plasts in 2 ml solution containing several concentrations of NaCl were loaded into a
one-well chambered cover glass (Nunc) with installed integral platinum electrodes
at the end of the chamber. The migration of the protoplasts under a potential
gradient of about 10 V/cm was analysed^74 using a fluorescence microscope
(Axiovert 200/Axio Observer 3) equipped with filter wheels and a cooled CCD
camera^17. More than five measurements for each sample, each consisting of 30 runs
with duration of ∼30 s, were performed at room temperature. For each concentra-
tion of NaCl, at least three independent preparations were analysed. An average
of the electrophoretic velocity value was analysed and calculated using Image J
(https://imagej.nih.gov/ij/). The values of ζ potentials were further calculated using
the Helmholtz–Smoluchowski equation^74 –^76. For survival rate, protoplasts were
treated with NaCl, and white light images were taken every 30 s for 20 min to
monitor the integrity of protoplasts (Fig. 5f).
Isothermal titration calorimetry. ITC^79 –^82 was used to analyse binding of Na+
to GIPCs with modifications^83 ,^84. The purified lipids were titrations in running
buffer containing 20 mM MES-Tris (pH 5.8) and 240 mM mannitol. For analysis
of binding of Na+ to GIPCs, small unilamellar vesicles of GIPCs from the wild-type
and moca1 seedlings were produced by sonication in running buffer (30 min on
ice, 20 s pulse on, 20 s pulse off, amplitude 25%)^50. Lipid vesicles were produced
from the MS-analysed wild-type GIPC–IPC mixture (calculated as 5.792:0.608
wt:wt) or moca1 GIPC–IPC mixture (calculated as 0.416:5.984 wt:wt) at a final lipid
concentration of 6.40 mg/ml, and the concentrations of the wild-type and moca1
GIPC–IPC mixtures were 5.21 mM and 6.77 mM, respectively. ITC measurements
were performed at 25 °C using MicroCal iTC200 (Malvern Panalytical)^79 –^82. NaCl
at a concentration of 50 mM was injected into a 200 μl sample cell until saturation
was reached. The volume of each injection was 2 μl with a total of 19 injections
and consecutive injections were separated by 2 min to allow the peak to return to
baseline. Similarly, binding of K+ or Li+ to GIPCs was analysed with the exper-
imental procedures developed for Na+. ITC data were analysed with a one-site
fitting model using Origin software (Malvern Panalytical and OriginLab). Error
was calculated from the standard deviation of six titrations.
Phylogenetic analysis. Multiple sequence alignment of the GT8 and UDPGP
regions was performed using MAFFT v7.05 with automatic method^85. The phy-
logeny was constructed using FastTree v2.1.7 with default parameters^86. FastTree
implements an ultrafast and fairly accurate approximate ML method. Phylogenetic
trees were represented and edited using FigTree (http://tree.bio.ed.ac.uk/software/
figtree/). Statistical values are shown beside selected major nodes with black circles.
The scale bar indicates the number of amino acid residue substitutions per site.
The full tree could be divided into three major classes (I to III), consistent with
previous results^87.
Statistical analysis. To minimize the system variations, wild-type and moca1 seed-
lings as well as transgenic lines were always grown side-by-side in agar medium
in Petri dishes or in soil in trays, and the Petri dishes and trays were rotated every
other day in positions in the growth chambers or rooms to have even tempera-
ture and light. Individual seedlings or pools of several seedlings were analysed.
For instance, in Fig. 2d, e, 8–12 seedlings were pooled together as one pool, and
12 of these pools were analysed to give mean ± s.d. Independent experiments were
performed at least three times. Statistical analysis was performed using Excel 2016
software (Microsoft), and P values were calculated via T.TEST (Student’s t-test,
two-sided). Data are presented as mean ± s.d. or s.e.m. To analyse the difference
between genotypes or treatments in line graphs, two-way analysis of variance
(ANOVA) was carried out using SAS 9.3/9.4 software (SAS Institute). Values of
P < 0.05 were considered statistically significant.
No statistical methods were used to predetermine sample size. The experiments
were not randomized and investigators were not blinded to allocation during
experiments and outcome assessment.
Reporting summary. Further information on research design is available in
the Nature Research Reporting Summary linked to this paper.Data availability
The data that support the findings of this study are available from the correspond-
ing author upon request.- Monshausen, G. B., Messerli, M. A. & Gilroy, S. Imaging of the Yellow Cameleon
3.6 indicator reveals that elevations in cytosolic Ca^2 + follow oscillating
increases in growth in root hairs of Arabidopsis. Plant Physiol. 147 , 1690–1698
(2008). - Feldmann, K. A. T-DNA insertion mutagenesis in Arabidopsis - mutational
spectrum. Plant J. 1 , 71–82 (1991). - Weigel, D. et al. Activation tagging in Arabidopsis. Plant Physiol. 122 , 1003–1013
(2000). - Clough, S. J. & Bent, A. F. Floral dip: a simplified method for Agrobacterium-
mediated transformation of Arabidopsis thaliana. Plant J. 16 , 735–743 (1998). - Schubert, D. et al. Silencing in Arabidopsis T-DNA transformants: the
predominant role of a gene-specific RNA sensing mechanism versus position
effects. Plant Cell 16 , 2561–2572 (2004). - Li, J. et al. BAK1, an Arabidopsis LRR receptor-like protein kinase, interacts with
BRI1 and modulates brassinosteroid signaling. Cell 110 , 213–222 (2002). - Pei, Z.-M. et al. Calcium channels activated by hydrogen peroxide mediate
abscisic acid signalling in guard cells. Nature 406 , 731–734 (2000).