The AMOR Arabinogalactan Sugar Chain Induces Pollen-Tube Competency to Respond to Ovular Guidance

Report

The AMOR Arabinogalactan Sugar Chain Induces Pollen-Tube Competency to Respond to Ovular Guidance Graphical Abstract

Authors Akane G. Mizukami, Rie Inatsugi, Jiao Jiao, ..., Kenichiro Itami, Narie Sasaki, Tetsuya Higashiyama

Correspondence [email protected]

In Brief Mizukami et al. report the first identification of the female molecule that induces competency of the pollen tube to respond to the ovular attraction signal. This molecule, named AMOR, is an ovular arabinogalactan sugar chain, of which a terminal disaccharide, methylglucuronosyl galactose, is necessary and sufficient for activity.

Highlights d

AMOR makes pollen tubes competent to respond to LURE attractant peptides in Torenia

d

Arabinogalactan sugars, which are abundant in the ovary, are responsible for AMOR

d

A terminal disaccharide structure is necessary and sufficient for AMOR activity

d

The synthesized disaccharide, methyl-glucuronosyl galactose, shows AMOR activity

Mizukami et al., 2016, Current Biology 26, 1–7 April 25, 2016 ª2016 Elsevier Ltd All rights reserved http://dx.doi.org/10.1016/j.cub.2016.02.040

Please cite this article in press as: Mizukami et al., The AMOR Arabinogalactan Sugar Chain Induces Pollen-Tube Competency to Respond to Ovular Guidance, Current Biology (2016), http://dx.doi.org/10.1016/j.cub.2016.02.040

Current Biology

Report The AMOR Arabinogalactan Sugar Chain Induces Pollen-Tube Competency to Respond to Ovular Guidance Akane G. Mizukami,1,2,9 Rie Inatsugi,3,9 Jiao Jiao,4 Toshihisa Kotake,5 Keiko Kuwata,4 Kento Ootani,2 Satohiro Okuda,4 Subramanian Sankaranarayanan,4 Yoshikatsu Sato,4 Daisuke Maruyama,4 Hiroaki Iwai,6 Estelle Gare´naux,7 Chihiro Sato,7 Ken Kitajima,7 Yoichi Tsumuraya,5 Hitoshi Mori,8 Junichiro Yamaguchi,2 Kenichiro Itami,2,4 Narie Sasaki,2 and Tetsuya Higashiyama1,2,4,* 1JST

ERATO, Higashiyama Live-Holonics Project, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Aichi 464-8602, Japan School of Science, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Aichi 464-8602, Japan 3Department of Evolutionary Biology and Environmental Studies, University of Zurich, Winterthurerstrasse 190, 8057 Zurich, Switzerland 4Institute of Transformative Bio-Molecules (WPI-ITbM), Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Aichi 464-8602, Japan 5Graduate School of Science and Engineering, Saitama University, 255 Shimo-Okubo, Sakura-ku, Saitama City, Saitama 338-8570, Japan 6Faculty of Life and Environmental Sciences, University of Tsukuba, Tsukuba, Ibaraki 305-8572, Japan 7Bioscience and Biotechnology Center, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Aichi 464-8601, Japan 8Graduate School of Bioagricultural Sciences, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Aichi 464-8601, Japan 9Co-first author *Correspondence: [email protected] http://dx.doi.org/10.1016/j.cub.2016.02.040 2Graduate

SUMMARY

Precise directional control of pollen-tube growth by pistil tissue is critical for successful fertilization of flowering plants [1–3]. Ovular attractant peptides, which are secreted from two synergid cells on the side of the egg cell, have been identified [4–6]. Emerging evidence suggests that the ovular directional cue is not sufficient for successful guidance but that competency control by the pistil is critical for the response of pollen tubes to the attraction signal [1, 3, 7]. However, the female molecule for this competency induction has not been reported. Here we report that ovular methyl-glucuronosyl arabinogalactan (AMOR) induces competency of the pollen tube to respond to ovular attractant LURE peptides in Torenia fournieri. We developed a method for assaying the response capability of a pollen tube by micromanipulating an ovule. Using this method, we showed that pollen tubes growing through a cut style acquired a response capability in the medium by receiving a sufficient amount of a factor derived from mature ovules of Torenia. This factor, named AMOR, was identified as an arabinogalactan polysaccharide, the terminal 4-O-methylglucuronosyl residue of which was necessary for its activity. Moreover, a chemically synthesized disaccharide, the b isomer of methyl-glucuronosyl galactose (4-Me-GlcA-b-(1/6)-Gal), showed AMOR activity. No specific sugar-chain structure of plant extracellular matrix has been identified as a bioactive molecule involved in intercellular communication. We suggest that the AMOR sugar chain in the ovary

renders the pollen tube competent to the chemotropic response prior to final guidance by LURE peptides. RESULTS AND DISCUSSION During growth in the pistil, pollen tubes of flowering plants receive various female molecules including ions, lipids, small bioactive molecules (such as phytohormones), peptides, and glycoproteins until discharge of non-motile sperm cells into the embryo sac (female gametophyte) of the target ovule [1–3]. The gene expression profile of the pollen tubes changes gradually [8, 9], which appears to facilitate gametophytic interactions including pollen-tube attraction and reception by the embryo sac [7]. Torenia is a unique plant species with a protruding embryo sac [10]. In the semi-in vitro Torenia system, pollen tubes growing through a cut style are attracted to the embryo sac by the LURE peptides secreted by two synergid cells on the side of the egg cell [4, 10]. In semi-in vitro culture with ovules, pollen tubes change their physiological properties—including their affinity to LURE peptides—and finally become competent to be attracted by LURE peptides [9]. Discovery of the Ovular Factor AMOR, which Induces Pollen-Tube Competency To determine the timing of pollen-tube acquisition of competency to ovular attraction in the semi-in vitro Torenia system, we performed an ovule-manipulating assay using a glass needle (Figure 1A). A hand-pollinated pistil was immediately excised (15 mm) and cultivated on a thin layer of agar medium containing hundreds of ovules detached from the placenta. Pollen tubes began to emerge from the cut end of the pistil at 6 hr after pollination (hap). For the ovule-manipulating assay, a single ovule excised from a fresh ovary was picked up and positioned in front of pollen tubes using a glass needle. Competent pollen tubes

Current Biology 26, 1–7, April 25, 2016 ª2016 Elsevier Ltd All rights reserved 1

Please cite this article in press as: Mizukami et al., The AMOR Arabinogalactan Sugar Chain Induces Pollen-Tube Competency to Respond to Ovular Guidance, Current Biology (2016), http://dx.doi.org/10.1016/j.cub.2016.02.040

A

C

Figure 1. The Ovular Factor AMOR Is Required for Competency of Pollen Tubes to Respond to Ovular Attraction

B

D

E

F

(A) Schematic of the semi-in vitro pollen-tube guidance assay for quantification of pollen-tube competency (responsiveness) to ovular attraction. The schematic on the left shows a cross-sectional view of a pistil in T. fournieri. As shown in the middle, pollen tubes growing through a cut style were co-cultivated with ovules to examine their response capability by placing a freshly prepared ovule near the tip of an elongating pollen tube using a micromanipulator (inset). EA, egg apparatus; GN, glass needle; PT, pollen tube. (B) Attraction of a competent pollen tube to a manipulated ovule (OV). Time after placing an ovule is indicated as min:s. Arrow heads indicate an attracted (responsive) pollen tube. See also Movie S1. (C and D) Percentages of pollen tubes attracted to a manipulated ovule, in the presence or absence ( ) of co-cultured ovules (C) and in the presence of various types of ovules (D). In (C), pollen tubes emerged from the cut end of the pistil at 6 hr after pollination (hap) and were examined at various time points (in D and others, at 12 hap). (E) Schematic of an ovule-manipulating assay (AMOR assay) for a pre-culture medium of ovules with the placenta. (F) Percentages of pollen tubes attracted to a manipulated ovule in the absence (w/o) or presence of co-cultured ovule tissues, in the absence of ovule tissues but with the pre-culture medium of ovules (pre-culture medium), or with the preculture medium only transiently for 10 min at 10 hap (washout). (G) Dose dependency of AMOR activity in a dilution series of the pre-culture medium. A relative concentration of ‘‘1’’ indicates the normal condition in which ovules from two flowers are cultivated in 200 ml medium. Data are the means and SDs of three independent experiments (n = 30–45 pollen tubes under each condition).

G

were always attracted when the micropylar end of the embryo sac was placed within 30 mm of the front of the tubes (Figure 1B; Movie S1). By conducting the semi-in vitro pollen-tube guidance assay with various cultivation periods, we found that pollen tubes, just after passing through the cut style, were not competent to respond to the attraction signal (Figure 1C, 6–8 hap) and gradually 2 Current Biology 26, 1–7, April 25, 2016

became competent to do so over the course of 4 hr in the medium (Figure 1C, 8–12 hap). To distinguish whether pollen tubes acquired their competency autonomously after passing through the style or whether some factor(s) released from co-cultivated ovules was involved in competency acquisition, we cultured pollen tubes growing through the cut style for 12–14 hap in the absence of co-cultivated ovules. These pollen tubes could not become competent and did not respond to a manipulated ovule placed suddenly in front of the tubes (Figure 1C, 12–14 hap, OV-). This implies that pollen tubes growing through the cut style are rendered fully competent by an unknown factor released from co-cultivated ovules into the medium. We investigated the properties of this putative ovular factor (Figure 1D). When semi-in vitro pollen tubes were cultivated

Please cite this article in press as: Mizukami et al., The AMOR Arabinogalactan Sugar Chain Induces Pollen-Tube Competency to Respond to Ovular Guidance, Current Biology (2016), http://dx.doi.org/10.1016/j.cub.2016.02.040

with immature ovules with a non-protruding embryo sac (approximately four-nucleate stage [11]) the tubes were not competent, suggesting that mature ovules render pollen tubes competent. Laser ablation of co-cultivated ovules showed that the embryo sac, including synergid cells, was not required. Ovules of the hybrid species Torenia hybrida (Summer Wave; Suntory Flowers), which does not produce an embryo sac, consistently made pollen tubes competent, suggesting that ovular sporophytic tissue is sufficient as the origin of the factor. No species specificity was found when we used mature ovules of related species of Lindernia, although attractant LURE peptides show species specificity even in more closely related species [12, 13]. To confirm that a diffusible factor derived from the ovule is critical for competency acquisition by the pollen tube, we prepared pre-culture (conditioned) medium by cultivating ovules on the placenta overnight (Figure 1E). We then cultivated pollen tubes growing through a cut style in the pre-culture medium instead of co-cultivating it with living ovules (Figure 1E). After 12–16 hr of cultivation, the competency of the pollen tubes was then examined using a manipulated ovule, and they were found to be competent to respond to the attraction signal (Figure 1F, pre-culture medium). Moreover, pollen tubes became competent when dipped transiently into the pre-culture medium for 10 min followed by washing with fresh medium (Figure 1F, washout), excluding the possibility that an ovular factor is necessary for establishing the attraction signal in the medium (e.g., as a carrier of attractants). Moreover, the effect of the pre-culture medium on pollen-tube competency was found to be concentration dependent (Figure 1G). These results suggest that an ovular factor induces competency of pollen tubes to respond to the attraction signal from the synergid cell. We named this ovular factor AMOR, for activation molecule for response capability. Purification and Characterization of AMOR We next attempted to purify AMOR from ovule pre-culture medium. Pre-culture medium of ovules (ovaries) showed higher AMOR activity compared to those of leaves, sepals, and styles, suggestive of a tissue preference (Figure S1A). We used ovary pre-culture medium for AMOR purification (Figure S1B). Hereafter, the bioassay for fractionated samples using a single manipulated ovule is termed the AMOR assay (illustrated in Figure 1E). AMOR activity is shown as the percentage of responsive pollen tubes cultured in each fraction. We first used ion-exchange columns to purify AMOR. AMOR bound to an anion-exchange column, suggesting the existence of an acidic moiety (Figure S1C). The highest activity was detected in a single-peak fraction, which was subjected to further purification. AMOR bound to various lectin columns, including wheat germ agglutinin (WGA), suggesting the existence of polysaccharides (glycans; Figure S1D). The resultant single-peak AMOR fraction was further purified by size-exclusion column chromatography. The majority of AMOR activity was found in a single peak with a relative molecular mass of 15–25 kDa, as estimated by comparison with protein markers (Figure S1E). Quantification of purified AMOR using a dilution series estimated that 62.5% of AMOR was collected as a single-peak fraction. No other fraction showed major activity, implying AMOR to be a single factor.

Purified AMOR was further characterized and found to be heat stable (100 C for 10 min) and resistant to proteinase K (Figure 2A). These results suggested that a proteinaceous factor is not required for the activity of AMOR. Therefore, we examined the possibility that polysaccharides contained in the AMOR fraction might be responsible for its activity. When we applied Yariv reagent (b-GlcY), which binds specifically to the main chain of b-(1/3)-galactan of arabinogalactan (AG; type II groups [14]), purified AMOR bound to Yariv reagent, but not to a-galactosyl Yariv (a-GalY) reagent (Figure 2B), an inactive analog of b-GlcY. This suggests that purified AMOR contains AG sugar chains with a b-(1/3)-galactan main chain (see Figure S2 for the structure of a typical AG). Frozen sections of Torenia ovaries stained with Yariv consistently showed that AG sugar chains are abundant in ovules and the surface cell layer of the placenta (Figures 2C–2F). Pollen tubes grow on the surface of the placenta to reach the protruding embryo sac of ovules in Torenia. Similarly, AG sugar chains in the ovule and the transmitting tract where pollen tubes grow have been reported in Arabidopsis [15]. In plants, type II groups of AG are generally synthesized as sugar moieties of arabinogalactan proteins (AGPs), which are heavily glycosylated plant-cell-surface proteins involved in plant growth, development, and reproduction [16–18]. AG sugar chains can be readily released from the plasma membrane as soluble factors with or without a peptide backbone [19, 20]. The importance of AG sugar chains in the function of AGP has long been suggested [21]; however, the function of a specific AG sugar-chain structure has not been reported to date. The Terminal Disaccharide Structure of Arabinogalactan Is Responsible for the Activation of the LURE Signaling Cascade To determine whether AG in the AMOR fraction is necessary for AMOR activity, we treated the purified AMOR fraction obtained in Figure S1E with enzymes specific to AG structures (Figure 3A). AG type II groups consist of a b-(1/3)-galactan backbone with b-(1/6)-galactan side chains [16, 22] (Figure S2). b-(1/6)galactan side chains are partly substituted with other sugars, including single a-(1/3)-L-arabinofuranosyl residues and b-glucuronosyl (with or without 4-O-methylation) residues (Figure S2). Interestingly, hydrolytic degradation of b-glucuronosyl residues attached to b-(1/6)-galactan side chains by b-glucuronidase decreased AMOR activity drastically (5.0% ± 5.0%, n = 3), whereas treatment with a-L-arabinofuranosidase and endob-(1/6)-galactanase, which act on a-L-arabinofuranosyl residues and b-(1/6)-galactan side chains, respectively, did not decrease AMOR activity significantly (66.7% ± 10.2%, n = 3) compared to treatment with heat-inactivated enzymes (53.3% ± 6.9%, n = 3; Figure 3A). These results indicated that AG chains with b-glucuronosyl residues are responsible for AMOR activity, and that terminal modification with b-glucuronosyl residues is critical. Based on the results of enzyme treatment (Figure 3A), we next examined whether (methyl-)glucuronosylated galactose, which is a terminal structure of some AG chains [16, 22] (Figure S2), is sufficient for AMOR activity. We digested and purified side chains of AG from radish root AGPs, which are useful for biochemical analysis of AG structures [16]. To our surprise, even a fraction of the disaccharide, 4-O-methyl-glucuronosyl Current Biology 26, 1–7, April 25, 2016 3

Please cite this article in press as: Mizukami et al., The AMOR Arabinogalactan Sugar Chain Induces Pollen-Tube Competency to Respond to Ovular Guidance, Current Biology (2016), http://dx.doi.org/10.1016/j.cub.2016.02.040

Figure 2. AMOR Contains Arabinogalactan Sugar Chains

B

A

C

D

E

F

galactose (4-Me-GlcA-b-(1/6)-Gal), showed AMOR activity (Figure 3B). Not only the fraction of the disaccharide but also fractions of oligosaccharides including 4-Me-GlcA-b-(1/6)Gal-b-(1/6)-Gal (two galactose residues) and 4-Me-GlcAb-(1/6)-Gal-b-(1/6)-Gal-b-(1/6)-Gal (three galactose residues) showed similar AMOR activity at the same concentration (1 mg/ml; Figure 3B). The other disaccharide fraction, Galb-(1/6)-Gal, did not show any AMOR activity (Figure 3B). This is consistent with the finding that the terminal disaccharide structure is sufficient for AMOR activity. To further examine the specificity of AMOR activity for 4-MeGlcA-b-(1/6)-Gal, we chemically synthesized several variants (Figure 3C). First, stereochemical validation of the anomeric carbon of the glucuronosyl residue in the active AMOR was conducted. Both a and b epimers of 4-Me-GlcA-Gal were synthesized by the sequence shown in Figure S3. We found that the chemically synthesized, pure b isomer of 4-Me-GlcA-Gal showed high AMOR activity (76.7% ± 15.3% at 0.135 mM, n = 3), whereas the a isomer showed low AMOR activity (33.3% ± 5.8% at 1.35 mM, n = 3; Figure 3D). It should be noted that only b isomers are usually found in native AGs of plants [16, 22]. We next examined synthetic GlcA-b-(1/6)-Gal lacking a methyl residue, because AG purified from radish roots was a mixture of GlcA-b-(1/6)-Gal with and without the 4-O-methyl residue. Synthetic GlcA-b-(1/6)-Gal without a 4-O-methyl residue showed significantly lower AMOR activity (3.3% ± 5.8% at 1.35 mM, n = 3), indicating that methyl modification of GlcA was also critical for AMOR activity (Figure 3D). Neither mono4 Current Biology 26, 1–7, April 25, 2016

(A and B) Characterization of AMOR fractions purified from ovary culture medium. See Figure S1 for details of the purification. In (A), the AMOR fraction obtained from the anion-exchange column and WGA lectin column purification was used as a positive control (AN-WGA). AMOR was resistant to heat treatment (100 C, 10 min) and proteinase treatment (proteinase K). In (B), AMOR in the fraction purified following separation on the anion-exchange column, WGA lectin column, and size-exclusion column (AN-WGA-SE) was precipitated with b-GlcY but not with a-GalY. a-GalY, a-galactosyl Yariv reagent; b-GlcY, b-glucosyl Yariv reagent. Data are the means and SDs of three independent experiments (n = 30 pollen tubes). (C–F) Yariv staining of Torenia ovaries. Thin frozen sections were stained with b-GlcY (C and D) and a-GalY (negative control; E and F). The direction of ovaries toward the style (ST) is indicated in (C) and (E). Higher-magnification views of ovules are shown in (D) and (F). The ovule sporophytic tissues, funiculus (FU), filiform apparatus in protruding embryo sacs (ES), and surface cell layer of the placenta (PL) show strong staining in (C) and (D).

meric (methyl-) glucuronic acid (GlcA) nor Gal showed AMOR activity (Figure S4), and b-glucuronidase treatment of synthetic 4-Me-GlcA-Gal completely inactivated its AMOR activity (Figure 3E), as in the case of the native AMOR fraction (Figure 3A). Taken together, these results suggested the disaccharide structure of 4-Me-GlcA-b-(1/6)-Gal to be responsible for AMOR activity. What is the role of AMOR in the pistil? We quantified sugars of water-soluble polysaccharides of Torenia pistils using liquid chromatography mass spectrometry (LC-MS; Figure 4A). AG sugar chains generally represent a high proportion of the total soluble carbohydrate in pistil extracts [23]. Consistent with the Yariv staining of ovaries (Figures 2C and 2D), the major components of AG sugar chains, Gal and arabinose (Figure S2), were more abundant in the ovule fraction than in the ovary wall fraction (Figure 4A). We found that not only fucose and GlcA but also 4-methyl glucuronic acid (4-Me-GlcA) existed predominantly in the ovule fraction (Figure 4A). Moreover, 4-Me-GlcA-b-(1/6)Gal was detected using LC-MS in the ovule fraction when the hydrolytic degradation step with trifluoroacetic acid treatment was omitted (Figures S4B and S4C). Considering that the ovary culture medium showed higher AMOR activity (Figure S1A), AG sugar chains containing 4-Me-GlcA-b-(1/6)-Gal are likely to be abundant in the ovary. Next, we performed a LURE bead assay using synthetic AMOR. Pollen tubes growing through a cut style were attracted to recombinant LURE1 and LURE2 of Torenia (Figures 4B and 4C). Micropylar pollen-tube attraction was reconstituted by LUREs and AMOR in the absence of other ovular factors. This suggests that AMOR activates the LURE signaling pathway in the pollen tube. The necessary step involved growing through the style, and this could not be replaced with high concentrations

Please cite this article in press as: Mizukami et al., The AMOR Arabinogalactan Sugar Chain Induces Pollen-Tube Competency to Respond to Ovular Guidance, Current Biology (2016), http://dx.doi.org/10.1016/j.cub.2016.02.040

B

A

Arabinofuranosidase

C

D

E

Figure 3. Me-GlcA-b-(1/6)-Gal Is Responsible for AMOR Activity (A) Susceptibility of AMOR to deglycosylation enzymes. Endo-b-(1/6)-galactanase and a-L-arabinofuranosidase co-treatment hydrolyzes the b-(1/6)galactan side chains of AGP sugar chains. b-glucuronidase hydrolyzes b-glucuronosyl and 4-O-methyl-glucuronosyl residues of AGP sugar chains. See Figure S2 for a schematic. Enzymes heat denatured at 100 C for 10 min were used as negative controls. (B) AMOR activity of 4-O-methyl-glucuronosyl galactan (oligosaccharide) fractions prepared from radish root AGPs. (C) Synthetic scheme of Me-GlcA-Gal. See Figure S3 for more details. (D) AMOR assay using synthetic Me-GlcA-Gal. Epimers (a and b) and GlcA-Gal without 4-O-methyl residues were tested. See Figure S4A for an AMOR assay using related monosaccharides. (E) Susceptibility of AMOR activity of synthetic 4-Me-GlcA-b-(1/6)-Gal to b-glucuronidase treatment. A negative control experiment using a denatured enzyme was performed as in (A). Data are the means and SDs of three independent experiments (n = 30 pollen tubes).

of synthetic AMOR (Figures 4B and 4D, without the style), suggesting that another factor in the style contributes to the competency control. Our understanding of the LURE signaling pathway is limited [3, 24]; however, Torenia pollen tubes are likely to be rendered fully competent by AMOR in the ovary immediately prior to final pollen-tube attraction by the ovule. Precise control of the timing of competency acquisition might be mediated by abundant 4-O-methyl GlcA modification of AG carbohydrate moieties in the ovary. The AG glycosylation pattern of the pistil is controlled in a tissue- and cell-type-specific manner, including the glycosylation gradient of tobacco transmitting tissue-specific (TTS) proteins toward the ovary [25] and ovule-specific AG epitopes [15].

Conclusions To our surprise, a simple disaccharide structure (4-Me-GlcAb-(1/6)-Gal; Figure 3C, bottom right) contained in AG sugar chains was responsible for cell-cell communication in the pollen-pistil interaction. The mechanism of action of AMOR remains to be elucidated, i.e., whether it is recognized by a cell-surface receptor, as in the case of chitin receptor for defense [26, 27]. Disaccharide structures can be recognized by lectins [28], and many receptor-like kinases contain lectin domains [29]. Many AGPs involved in various intercellular signaling pathways have been identified [17, 18], including TTS, 120 kDa glycoprotein, AGP6, and AGP11 proteins for reproduction, xylogen and AGP1 for development, and AGP17 for defense and symbiosis. Current Biology 26, 1–7, April 25, 2016 5

Please cite this article in press as: Mizukami et al., The AMOR Arabinogalactan Sugar Chain Induces Pollen-Tube Competency to Respond to Ovular Guidance, Current Biology (2016), http://dx.doi.org/10.1016/j.cub.2016.02.040

A

B

C

Figure 4. AMOR Activates Signaling Pathway

the

LURE

(A) Quantification of sugars (galactose, Gal; arabinose, Ara; fucose, Fuc; glucuronic acid, GlcA; 4-methyl glucuronic acid, 4Me-GlcA) in watersoluble polysaccharides of Torenia pistils. The structural formula of each sugar is indicated. Data are shown in relative units (R.U.) compared with the average in stigmas. Note that 4Me-GlcA exists predominantly in ovules with the placenta. See also Figures S4B and S4C. (B–D) Semi-in vitro LURE attraction assay. Arrowheads in (B) and (C) mark the positions of the tips of pollen tubes when gelatin beads (circles) containing recombinant LURE1 (B) and LURE2 (C) were added (0 min). Note that semi-in vitro pollen tubes growing through the style are attracted to LUREs in the presence of synthetic disaccharide AMOR. The asterisks in (D) indicate significant differences based on ANOVA followed by Bonferroni-Dunn test for multiple comparisons (LURE1; p < 0.01) or by Fisher’s exact test (LURE2; p < 0.01). Data are the means and SDs of three independent experiments. The number of pollen tubes examined under each condition is indicated in (D).

D

However, the function of AG in signaling is largely unknown [16–18, 30]. The identification of AMOR suggests the importance of a specific structure of AG for intercellular signaling. Recently, several genes that encode key enzymes for AG biosynthesis in Arabidopsis have been identified, including the potent hydroxyproline O-galactosyltransferases HPGT1, -2, and -3 for initiation of AG glycosylation on the peptide backbone [21] and b-glucuronosyltransferase for addition of GlcA to the galactan [22, 31]. Whether AMOR plays a role in the pollen-tube guidance of Arabidopsis remains unknown. Use of a combination of genetic analysis using mutants defective in AG biosynthesis with a synthetic chemistry approach will address such a general issue and open up a new frontier in plant cell-cell communication mediated by sugar chains. 6 Current Biology 26, 1–7, April 25, 2016

SUPPLEMENTAL INFORMATION Supplemental Information includes Supplemental Experimental Procedures, four figures, and one movie and can be found with this article online at http://dx.doi.org/10.1016/j.cub.2016.02.040. AUTHOR CONTRIBUTIONS T.H. directed the study. A.G.M., T.H., and R.I. designed and conducted a majority of the biological experiments, including characterization and purification, and A.G.M. identified the disaccharide structure for AMOR activity. K.O. and Y.S. performed Yariv staining. K.O., S.O., and S.S. performed semi-in vitro pollen-tube attraction assays using LURE beads. T.K. and Y.T. prepared AG-specific enzymes and oligosaccharides from radish AGP. J.J. and J.Y. synthesized mono- and disaccharides. K. Kuwata, K.O., A.G.M., E.G., C.S., and H.I. contributed to sugar analysis using LC-MS. H.M. assisted with the

Please cite this article in press as: Mizukami et al., The AMOR Arabinogalactan Sugar Chain Induces Pollen-Tube Competency to Respond to Ovular Guidance, Current Biology (2016), http://dx.doi.org/10.1016/j.cub.2016.02.040

biochemical experiments. T.H., N.S., T.K., H.M., K. Kitajima, and K.I. supervised the project and commented on the manuscript. The manuscript was written by A.G.M., D.M., and T.H. ACKNOWLEDGMENTS We thank Y. Amagai, Y. Yano, and S. Sakamoto, undergraduate students of the University of Tokyo and Nagoya University, for their assistance in preparing medium for large-scale ovule and ovary culture. We also thank M.M. Kanaoka, T. Kuroiwa, T. Ueda, and A. Nakano for helpful comments; H. Takeuchi for critical reading of the manuscript; M. Kuzuya and R. Yui for help with biochemical analysis; K. Ogasawara and A. Terashima for help with frozen sectioning of Torenia pistils; K. Takahashi for help with LC-MS analysis; N. Kamiya for help with LURE bead assays; and D. Kurihara, Y. Hamamura, and S. Nagahara for their assistance with time-lapse imaging. A.G.M. and S.O. were supported by grants 7802 and 30004, respectively, from the Japan Society for the Promotion of Science Fellowships. This work was supported in part by grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan (17027006, 17657022, 18075004, and 19370017 to T.H.), Program for the Promotion of Basic Research Activities for Innovative Biosciences (PROBRAIN to T.H.), Yamada Science Foundation (to T.H.), Mitsubishi Foundation (to T.H.), Japan Advanced Plant Science Network, and Japan Science and Technology Agency (CREST, PRESTO, and ERATO projects to T.H.). Received: November 17, 2015 Revised: January 24, 2016 Accepted: February 12, 2016 Published: April 7, 2016 REFERENCES 1. Palanivelu, R., and Tsukamoto, T. (2012). Pathfinding in angiosperm reproduction: pollen tube guidance by pistils ensures successful double fertilization. Wiley Interdiscip. Rev. Dev. Biol. 1, 96–113. 2. Dresselhaus, T., and Franklin-Tong, N. (2013). Male-female crosstalk during pollen germination, tube growth and guidance, and double fertilization. Mol. Plant 6, 1018–1036. 3. Higashiyama, T., and Takeuchi, H. (2015). The mechanism and key molecules involved in pollen tube guidance. Annu. Rev. Plant Biol. 66, 393–413. 4. Okuda, S., Tsutsui, H., Shiina, K., Sprunck, S., Takeuchi, H., Yui, R., Kasahara, R.D., Hamamura, Y., Mizukami, A., Susaki, D., et al. (2009). Defensin-like polypeptide LUREs are pollen tube attractants secreted from synergid cells. Nature 458, 357–361. 5. Ma´rton, M.L., Fastner, A., Uebler, S., and Dresselhaus, T. (2012). Overcoming hybridization barriers by the secretion of the maize pollen tube attractant ZmEA1 from Arabidopsis ovules. Curr. Biol. 22, 1194– 1198. 6. Takeuchi, H., and Higashiyama, T. (2012). A species-specific cluster of defensin-like genes encodes diffusible pollen tube attractants in Arabidopsis. PLoS Biol. 10, e1001449. 7. Leydon, A.R., Chaibang, A., and Johnson, M.A. (2014). Interactions between pollen tube and pistil control pollen tube identity and sperm release in the Arabidopsis female gametophyte. Biochem. Soc. Trans. 42, 340–345. 8. Qin, Y., Leydon, A.R., Manziello, A., Pandey, R., Mount, D., Denic, S., Vasic, B., Johnson, M.A., and Palanivelu, R. (2009). Penetration of the stigma and style elicits a novel transcriptome in pollen tubes, pointing to genes critical for growth in a pistil. PLoS Genet. 5, e1000621. 9. Okuda, S., Suzuki, T., Kanaoka, M.M., Mori, H., Sasaki, N., and Higashiyama, T. (2013). Acquisition of LURE-binding activity at the pollen tube tip of Torenia fournieri. Mol. Plant 6, 1074–1090. 10. Higashiyama, T., Yabe, S., Sasaki, N., Nishimura, Y., Miyagishima, S.-y., Kuroiwa, H., and Kuroiwa, T. (2001). Pollen tube attraction by the synergid cell. Science 293, 1480–1483. 11. Susaki, D., Takeuchi, H., Tsutsui, H., Kurihara, D., and Higashiyama, T. (2015). Live imaging and laser disruption reveal the dynamics and cell-

cell communication during Torenia fournieri female gametophyte development. Plant Cell Physiol. 56, 1031–1041. 12. Higashiyama, T., Inatsugi, R., Sakamoto, S., Sasaki, N., Mori, T., Kuroiwa, H., Nakada, T., Nozaki, H., Kuroiwa, T., and Nakano, A. (2006). Species preferentiality of the pollen tube attractant derived from the synergid cell of Torenia fournieri. Plant Physiol. 142, 481–491. 13. Kanaoka, M.M., Kawano, N., Matsubara, Y., Susaki, D., Okuda, S., Sasaki, N., and Higashiyama, T. (2011). Identification and characterization of TcCRP1, a pollen tube attractant from Torenia concolor. Ann. Bot. (Lond.) 108, 739–747. 14. Kitazawa, K., Tryfona, T., Yoshimi, Y., Hayashi, Y., Kawauchi, S., Antonov, L., Tanaka, H., Takahashi, T., Kaneko, S., Dupree, P., et al. (2013). b-galactosyl Yariv reagent binds to the b-1,3-galactan of arabinogalactan proteins. Plant Physiol. 161, 1117–1126. 15. Coimbra, S., Almeida, J., Junqueira, V., Costa, M.L., and Pereira, L.G. (2007). Arabinogalactan proteins as molecular markers in Arabidopsis thaliana sexual reproduction. J. Exp. Bot. 58, 4027–4035. 16. Seifert, G.J., and Roberts, K. (2007). The biology of arabinogalactan proteins. Annu. Rev. Plant Biol. 58, 137–161. 17. Ellis, M., Egelund, J., Schultz, C.J., and Bacic, A. (2010). Arabinogalactanproteins: key regulators at the cell surface? Plant Physiol. 153, 403–419. 18. Pereira, A.M., Pereira, L.G., and Coimbra, S. (2015). Arabinogalactan proteins: rising attention from plant biologists. Plant Reprod. 28, 1–15. 19. Oxley, D., and Bacic, A. (1999). Structure of the glycosylphosphatidylinositol anchor of an arabinogalactan protein from Pyrus communis suspension-cultured cells. Proc. Natl. Acad. Sci. USA 96, 14246–14251. 20. van Hengel, A.J., Tadesse, Z., Immerzeel, P., Schols, H., van Kammen, A., and de Vries, S.C. (2001). N-acetylglucosamine and glucosamine-containing arabinogalactan proteins control somatic embryogenesis. Plant Physiol. 125, 1880–1890. 21. Ogawa-Ohnishi, M., and Matsubayashi, Y. (2015). Identification of three potent hydroxyproline O-galactosyltransferases in Arabidopsis. Plant J. 81, 736–746. 22. Knoch, E., Dilokpimol, A., and Geshi, N. (2014). Arabinogalactan proteins: focus on carbohydrate active enzymes. Front. Plant Sci. 5, 198. 23. Hoggart, R.M., and Clarke, A.E. (1984). Arabinogalactans are common components of angiosperm styles. Phytochemistry 23, 1571–1573. 24. Liu, J., Zhong, S., Guo, X., Hao, L., Wei, X., Huang, Q., Hou, Y., Shi, J., Wang, C., Gu, H., and Qu, L.J. (2013). Membrane-bound RLCKs LIP1 and LIP2 are essential male factors controlling male-female attraction in Arabidopsis. Curr. Biol. 23, 993–998. 25. Wu, H.M., Wang, H., and Cheung, A.Y. (1995). A pollen tube growth stimulatory glycoprotein is deglycosylated by pollen tubes and displays a glycosylation gradient in the flower. Cell 82, 395–403. 26. Kaku, H., Nishizawa, Y., Ishii-Minami, N., Akimoto-Tomiyama, C., Dohmae, N., Takio, K., Minami, E., and Shibuya, N. (2006). Plant cells recognize chitin fragments for defense signaling through a plasma membrane receptor. Proc. Natl. Acad. Sci. USA 103, 11086–11091. 27. Cao, Y., Liang, Y., Tanaka, K., Nguyen, C.T., Jedrzejczak, R.P., Joachimiak, A., and Stacey, G. (2014). The kinase LYK5 is a major chitin receptor in Arabidopsis and forms a chitin-induced complex with related kinase CERK1. eLife 3, e03766. 28. Weis, W.I., and Drickamer, K. (1996). Structural basis of lectin-carbohydrate recognition. Annu. Rev. Biochem. 65, 441–473. 29. Vaid, N., Macovei, A., and Tuteja, N. (2013). Knights in action: lectin receptor-like kinases in plant development and stress responses. Mol. Plant 6, 1405–1418. 30. Majewska-Sawka, A., and Nothnagel, E.A. (2000). The multiple roles of arabinogalactan proteins in plant development. Plant Physiol. 122, 3–10. 31. Geshi, N., Johansen, J.N., Dilokpimol, A., Rolland, A., Belcram, K., Verger, S., Kotake, T., Tsumuraya, Y., Kaneko, S., Tryfona, T., et al. (2013). A galactosyltransferase acting on arabinogalactan protein glycans is essential for embryo development in Arabidopsis. Plant J. 76, 128–137.

Current Biology 26, 1–7, April 25, 2016 7