Science
www.sciencemag.org/cgi/content/full/ll67324/DCl
Hams I
Supporting Online Material for
Auxin-Dependent Patterning and Gamete Specification in the Arabidopsis Female Gametophyte
Gabriela C. Pagnussat, Monica Alandete-Saez, John L. Bowman, Venkatesan
Sundaresan*
*To whom correspondence should be addressed. E-mail: sundar@ucdavis.edu
Published 4 June 2009 on Science Express DOI: 10.1126/science.1167324
This PDF file includes:
Materials and Methods Figs. S1 to S7 Tables S1 to S5
References
Supporting Online Material
Materials and methods Plant growth conditions
Seeds were sterilized in 20% (v/v) sodium-hypochlorite, washed with sterile water and plated on MS medium (Sigma-Aldrich). Seedlings were then transferred onto soil and grown in a growth chamber with a 16-hlight/ 8-h-dark cycle at 22°C with 60% relative humidity. For crosses, flowers of the female parent were manually emasculated 2 days before anthesis and cross-pollinated 2 days later.
Histology and microscopy
To prepare cleared whole-mount preparations, pistils containing at least 20 ovules were dissected and cleared overnight in Hoyers solution (S1). The dissected pistils were observed on a Zeiss Axioplan imaging 2 microscope (Zeiss, Oberkochem, Germany) under DIC optics. Images were captured on an Axiocam HRC CCD camera (Zeiss) with the Axiovision program (version 3.1, Zeiss). For GUS staining, developing carpels and siliques were dissected and incubated in GUS staining buffer, as described (S2). Individual ovules were dissected from the pistils/siliques and cleared overnight with Hoyers solution (S1). The ovules were observed under DIC optics. For pollen tube growth pattern analysis, aniline blue staining was performed, as described (S3).
Constructs and Plant Transformation
The synthetic amiR-ARFa gene was designed with miR164b of Arabidopsis as a backbone (S3), targeting a sequence largely conserved between ARF1-8 and ARF19 (ELWHACA =
GAGCTATGGCACGCTTGTGCA, Fig. S6). The synthetic miRNA was constructed by
GenScript (Piscataway, New Jersey, USA). amiR-ARFa was excised from pUC57, cloned into 6OpTATA BJ36 (S2) and subsequently into the binary vector pMLBART (Op::amiR-ARFa BART). Op:amiR-ARFa BART was initially transformed into pAP3:LhG4 driver lines (S4) and individuals lacking petals and stamen development were selected for crossing with pES1:LhG4. 10op:: YUC1 was constructed by inserting YUC1 cDNA (provided by Dr. Y. Zhao) behind an OP array (10OP-TATA-BJ36) (S2) and subsequently subcloned into the binary vector pCAMBIA 1300 (CAMBIA, Canberra, Australia). The plasmid was introduced into Agrobacterium tumefaciens strain GV3101 by electroporation and Ler wild-type plants were transformed with the floral dip method (S5).
Fig. SI: Expression of the synthetic reporter DR5::GUS during female gametophyte development (A) Auxin signaling output could be detected during megasporogenesis at the distal tip of the nucellus. (B) At FG1 the signal is strongly detected in the nucellus, outside the embryo sac. (C) At FG3 stage the signal is detectable inside the embryo sac, at the micropylar pole. Arrowheads indicate the position of nuclei at FG3. Fm, functional megaspore; Fg, female gametophyte.
Fig. S2: Auxin signaling output is downregulated in pESl»amiR-ARFa embryo sacs. (A) A pistils of apESl::LhG4/+; Op::amiR-ARFa/+; DRF::GFP/+ plant showing segregation for GFP expression in embryo sacs. Asterisk indicates embryo sac exhibiting the characteristic expression of the synthetic reporter DR5::GFP. (B) GFP expression is detected in one of the embryo sacs from apES1::LhG4/+; Op::amiR-ARFa/+; DRF::GFP/+ pistil. (C) An embryo sac showing no expression of the auxin reporter DR5::GFP. Embryo sacs are outlined by a dotted line. Es, embryo sac; mi, micropyle. Detailed results and quantification are shown in Table S1.
Fig. S3: Cell identities and pollen tube attraction in embryo sacs downregulated for ARF expression with amiR-ARFa driven by pES1 promoter. (A) two micropylar cells were GUS positive when the expression of a synergid-specific marker was tested in WT embryo sacs. (B) expression of the synergid-specific marker was never detected in ARF-downregulated embryo sacs. (C) Expression of a central cell marker in ARF-downregulated embryo sac was similar to WT. (D) Expression of an antipodal marker was similar to WT (S6). (E) Ovules showing segregation for WT embryo sac attracting a pollen tube on the left, and an ARF-downregulated embryo sac with loss of pollen tube attraction on the right. Ant, antipodal cells; Cc, central cell; Es, embryo sac; F, funiculus; Pt, pollen tube Syn, synergids.
Fig. S4: PIN expression in the developing ovule. PIN1 was detected by following the localization of the reporter gene GFP in transgenic plants carrying a PIN1::GFP translational fusion (gift of J. Friml) (A) Epifluorescence image showing PIN1 expression in the nucellus at FG1 stage. (B) Confocal image showing PINl localization in the external layer of nucellar cells. (C) Confocal image showing PIN1 expression in internal cells of the nucellus. (D) PIN1 localization suggests that auxin flux in the nucellus might be involved in establishing an auxin maximum at FG1 stage. Green lines in the diagram indicate PIN 1 localization. (E) Epifluorescence image showing PIN1 expression in the nucellus at sporogenesis, prior to FG1 stage. (F) No PIN1 expression could be detected in the nucellus or embryo sac after FG1 stage. Picture shows an ovule at FG6 stage in which only auto fluorescence is visible. (G) Epifluorescence image from PIN3::GFP transgenic plant showing no PIN3 expression in the nucellus at FG1 stage. Inset, PIN3 expression in the root is shown as a positive control for GFP activity. (H) Epifluorescence image from PIN4::GFP transgenic plant showing no PIN4 expression in the nucellus at FG1 stage. Inset, PIN4 expression in the root is shown as a positive control for GFP activity. Di, developing integuments; Es, embryo sac; Fm, indicates position of the functional megaspore; mi, indicates position of micropyle; nu, nucellus.
Fig. S5: DR5::GFP activity was detected inpESl»YUCl female gametophytes from stages FG1 to FG6. (A) DR5 activity at FG1. (B) DR5 activity at FG3. (C) DR5 activity at FG4. (D) DR5 activity at FG6 (embryo sac after cellularization). Low activity was detected at the nucellus at FG1 (A). As a strong activity was detected inside the developing embryo sac, the exposure used was not enough to reveal the nucellar GFP. Arrowheads indicate positions of nuclei inside the embryo sac. Fm, functional megaspore; Fg, female gametophyte; N, nucellus; Es embryo sac
CTCGAGGAGAATGATGAAGGTGTGTGATGAGCAAGA.rGCflCflAGCGTGCCflTAGCrCTTACTAGCTCATATATACACTCTCACCACAAATG CGTGTATATATGCGGAATTTTGTGATATAGATGTGTGTGTGTGTTGAGTGTGATGATATGGATGAGTTAGTTCGCGCTATGGCAATCTTGA GCATCATGACCACTCCACCTTGGTGACGATGACGACGAGGGTTCAAGTGTTACGCACGTGGGAATATACTTATATCGATAAACACACACGT GCGGGATCC
10 20 30 40 50 60 70
| A        CG UCUU A ---- U -  - A G A A
GAUGC CAAG UGCCAUAGC ACUAGCUCAU UAUA CAC CUCA C CACA AUGCGU UAUAU UGCGG A CUACG GUUC ACGGUAUCG UGAUUGAGUA GUAU GUG GAGU G GUGU UGUGUA AUAUA GUGUU U A A        UA CGCU G        AGUA      U        U U        G G - U
150 140 130 120 110 100 90 80
position 221111111111
in miR       3'    109876543210987654321 5'
rc amiR-ARFa GAGCUAUGGCACGCUUGUGCA
ARF5/MP
ARF7/NPH4
ARF2
ARF19
ARF8
ARF6
ARF3/ETTIN
ARF4
ARF1
GAGCUAUGGCACGCUUGUGCA GAGCUAUGGCACGCUUGUGCU GAGCUAUGGCACGCUUGUGCU CAACUAUGGCACGCUUGUGCA GAGCUAUGGCAUGCUUGUGCU GAGCUCUGGCAUGCUUGUGCU GAGCUGUGGCAUGCUUGUGCU GAGCUUUGGCAUGCUUGUGCU GAGCUCUGGCAUGCCUGUGCU
A
B
C
D
Fig. S6: Design of the synthetic amiR-ARFa gene. (A) miR164b of Arabidopsis was used as a backbone, targeting a sequence largely conserved between ARF1-8 and ARF19 (ELWHACA = GAGCTATGGCACGCTTGTGCA). These ARFs belong to the MP, ETT and ARF1 clades, sometimes also referred to as the Class IIa, IIb and Ia ARFs (S7-S9). (B) predicted stem-loop. italic = amiR-ARFa. (C) Alignment of the reverse complement of the amiR-ARFa sequence with potential ARF targets in Arabidopsis. Mismatches are shown in red. On the basis of the effect of mismatches on cleavage of the PHABULOSA gene (S10), the first 4 genes are very likely to be targeted by the a-miRNA, with the remaining 5 genes in descending order are potential targets for cleavage or translational repression (S8 and S9). (D) Flower from a plant with down-regulated ARF expression in floral organ primordia. When amiR-ARFa is driven in the second and third whorls of the flower with the APETALA3 (AP3) promoter (pAP3»amiR-ARFa), there was a loss of lateral organs in these whorls, a result consistent with amiR-ARFa targeting ARF5/MONOPTEROS (S11). Lines with strong phenotypes lacked both petals and stamens, whereas weaker phenotype lines lacked only petals as shown in this image (two sepals have been removed).
Fig. S7: Cell identity is altered in tir1 afb1-1 afb2-2 afb3-3 quadruple mutant embryo sacs. (A) Phenotype of an embryo sac from a tir1 afb1-1 afb2-2 afb3-3 quadruple mutant plant showing two egg cell-like cells and only one synergid cell. (B and C), GUS expression patterns detected after pollinating a tir1 afb1-1 afb2-2 afb3-3 quadruple mutant with pollen from transgenic plants carrying the PFAC1IE:GFP-GUS:TFAC1 construct. FAC1 (Embryonic Factor 1) encodes a putative AMP deaminase (EC:3.5.4.6) and is expressed in both the developing embryo and endosperm from the paternal allele (Xu et al., 2005). (B) WT pattern showing that paternal FAC1::GUS expression is detected both in the embryo and in the endosperm after fertilization. (C) Abnormal GUS expression pattern in a tir1 afb1-1 afb2-2 afb3-3 embryo sac showing paternal FAC1::GUS expression in an embryo proper and in a proembryo arrested after the first zygotic division, and no endosperm development. From a total of 399 tir1 afb1-1 afb2-2 afb3-3 embryo sacs examined, 82 showed defects in egg cell and synergid morphology while 26 aborted at FG1 stage. Ccn, central cell nucleus; EcL, egg cell-like; End, endosperm; Ep, embryo proper; PE, proembryo; Syn, synergid. Nuclei in panel (A) are artificially colored in red, and the proembryo in (C) and embryos in (B) and (C) are outlined by dotted lines. Bar indicates 20 (im.
Table SI. Expression of the synthetic reporter DR5::GFP in embryo sacs defective for ARFs expression. Pistils ofpES1::LhG4/+; Op::amiR-ARFa/+; DR5::GFP /+ plants were examined for GFP expression in the embryo sac.
Pistils studied
GFP expression GFP positive     GFP negative
Total
P value *
Op::amiR-ARFa/+; pES1::LhG4; DR5::GFP/+    102 (32.07 %)     216 (67.92 %)      318 (100 %) <0.005
pES1::LhG4/+; DR5::GFP/+
257 (51.19 %)    245 (48.81 %)      502 (100%)
>0.05
*X2-Test from the expectation that 50 % of the gametophytes will be GFP positive as a result of DR5::GFP being expressed in the embryo sacs. A second x2-Test was performed for the expectation that 37.5% of the gametophytes will be GFP positive as a result of amiR-ARFa being expressed in DR5::GFP embryo sacs, if the amiR-ARFa is fully penetrant. The P value obtained was >0.1, meaning that the deviation observed in the Op::amiR-ARFa / +; pES1::LhG4 / +; DR5::GFP / + pistils from the expected 37.5% for GFP positive embryo sacs is not statistically significant.
Table S2. . Embryo sac analysis in pistils from plants carrying the Op::amiR-ARFa construct crossed to plants with LhG4 expression driven by the embryo sac promoter pESl. Ten independent transgenic plants carrying the Op::amiR-ARFa construct were crossed to plants with LhG4 expression driven by the embryo sac promoter pES1 and 30 F1 plants were studied. Eight F1 plants showed a strong phenotype, with a ratio of defective embryo sacs near 25%. In carpels of pES1::LhG4/+; Op::amiR-ARFa/+ plants, 25% of the embryo sacs are predicted to inherit both constructs and result in the down-regulation of ARF expression.
Phenotypes observed Total p value*2
Pistils studied
WT embryo sacs      Defects in micropylar Collapsed embryo sacs cells polarity
Op::amiR-ARFa / + x
pES1:: LhG4 / + 534 (77.84 %) 121 (17.64 %) 31 (4.52 %)*' 686 (100 %) >0.05
*1 The proportion of aborted embryo sacs is comparable to the one described for WT pistils.
*2 x2-Test from the expectation that 25% of the gametophytes will be defective as a result of amiR-ARFa expression in the embryo sacs.
Table S3. Expression of cell specific markers in embryo sacs defective for ARF expression. Pistils ofpES1 ::LhG4/+; Op::amiR-ARFa/+; GUS/+ plants were examined for GUS expression in the embryo sac.
Pistils studied
pES1>>amiR-ARFa x Egg cell marker line
pES1>>amiR-ARFa
x Synergid cell marker line
GUS positive WT
expression pattern
189 (38.72 %)
GUS negative
289 (59.22 %)
97 (23.26 %) 320 (76.74 %)
GUS positive Abnormal expression pattern
10 (2.05 %)
Total
489 (100 %)
417 (100 %)
P values *
6.376 e
< 2.2 e "
pES1>>amiR-ARFa
x Central cell marker     185 (46.48
line %)
213 (53.52 %)
398 (100 %)
0.1605
pES1>>amiR-ARFa
x Antipodal cell        220 (49.33 marker line %)
226 (50.67 %)
446 (100 %)
0.7763
0
0
0
* x2-Test from the expectation that 50% of the gametophytes will be GUS positive, WT expression pattern (i.e. express the cell-specific marker in the correct cells), due to segregation of the GUS reporter. Of the total embryo sacs, 25% are predicted to be defective for ARF expression, and 12.5% are predicted to be both defective for ARF expression and carrying the GUS reporter.
Table S4. Embryo sac analysis in pistils from plants carrying the Op::YUCl construct that were crossed to plants with LhG4 expression driven by the embryo sac promoterpESl. Ten independent transgenic plants carrying the Op::YUCl construct were crossed to plants with LhG4 expression driven by the embryo sac promoter pESl and 30 F1 plants were studied. Six F1 plants showed a strong phenotype, presenting a ratio of defective embryo sacs near 25%. In carpels of pESl::LhG4/+; Op::YUCl plants, 25% of the embryo sacs are predicted to inherit both constructs resulting in over-expression of YUCl.
Phenotypes observed
Total
P value *
Pistils studied
Defects in
Unfused polar
Collapsed
WT
micropylar cells
nuclei
embryo sacsa
polarity
64
5
59
504
>0.05
(12.70 %)
(1.00 %)
(11.70 %)
(100 %)
aIn WT pistils, the frequency of collapsed embryo sacs is 4-5% (see Table S1).
* x2-Test from the expectation that 25% of the gametophytes will be defective as a result of YUCl overexpression in the embryo sacs.
Table S5. Expression of cell specific markers in embryo sacs overexpressing YUC1 of pESr..LhG4/+; Op::YUC1/+; GUS/+ plants were examined for GUS expression in the embryo sac.
GUS positive Pistils studied     WT expression pattern
pES1>>YUC1 x Egg cell marker line
111 (21.94 %)
GUS negative
362 (71.54 %)
GUS positive Abnormal expression pattern
Whole E.       Central Antipodal
Sac            cell cells
2 31
0
(0.40 %) (6.13 %)
Total P values * (100 %)
506
(100 %)
< 2.2 e 46
pES1>>YUC1
x Synergide cell marker line
154 328 12 5 21 520
(29.61 %) (63.07 %) (2.30 %)      (0.96 %)       (4.04 %)        (100 %)
< 2.2 e 46
pES1>>YUC1
x central cell 260 355
marker line       (42.28 %)        (57.72 %)
615
(100 %)
0.0001277
pES1>>YUC1
x Antipodal cell marker line
118 (27.76 %)
307
(72.24 %)
425 < 2.2 e 46
(100 %)
0
0
0
0
0
0
* x2-Test from the expectation that 50% of the gametophytes will be GUS positive, WT expression pattern (i.e. express the cell-specific marker in the correct cells), due to segregation of the GUS reporter. Of the total embryo sacs, 25% are predicted to overexpress YUC1, and 12.5% are predicted to be both overexpressing YUC1 and carrying the GUS reporter
Supplemental references
51 C. M. Liu, D. W. Meinke, Plant Journal 16, 21 (1998).
52 G. C. Pagnussat, H. J. Yu, V. Sundaresan, Plant Cell 19, 3578 (2007).
53 J. P. Alvarez et al., Plant Cell 18, 1134 (2006).
54 A. Goldshmidt et al., Plant Cell 20:1217-1230 (2008).
55 S. J. Clough, A. F. Bent, Plant Journal 16, 735 (1998).
56 H-J. Yu et al., Plant Physiol. 139, 1853-1869 (2005).
57 Y. Okushima et al., Plant Cell 18, 1134 (2006).
58 D. Wang et al., Gene 394, 13-24 (2007).
59 D. L. Remington et al., Plant Physiol. 135, 1738-1752 (2004).
510 A. C. Mallory, et al. , EMBO J., 23, 3356-3364 (2004).
511 M. Schuetz et al., Plant Physiol. 148, 870-880 (2008).
512 J. Xu, et al., Plant J. 42: 743-756 (2005).