Molecular cloning of a novel calcium-binding protein in the secreted saliva of the green rice leafhopper Nephotettix cincticeps

Insect Biochemistry and Molecular Biology 42 (2012) 1e9

Contents lists available at SciVerse ScienceDirect

Insect Biochemistry and Molecular Biology journal homepage: www.elsevier.com/locate/ibmb

Molecular cloning of a novel calcium-binding protein in the secreted saliva of the green rice leafhopper Nephotettix cincticeps Makoto Hattori a, *, Masatoshi Nakamura a, Setsuko Komatsu b, Kazuko Tsuchihara a,1, Yasumori Tamura a, Tsuyoshi Hasegawa a a b

National Institute of Agrobiological Sciences, 1-2 Ohwashi, Tsukuba, Ibaraki 305-8634, Japan National Institute of Crop Science, 2-1-18 Kannondai, Tsukuba, Ibaraki 305-8518, Japan

a r t i c l e i n f o

a b s t r a c t

Article history: Received 24 August 2011 Received in revised form 26 September 2011 Accepted 7 October 2011

Green rice leafhoppers (Nephotettix cincticeps) secrete watery and coagulable saliva in the feeding process. In our study, the watery salivary secretion was concentrated by ultrafiltration from “fed diet” and subjected to SDS-PAGE. The N-terminal amino acid sequence of the most predominant band at 84 kDa (designated NcSP84) was analyzed by Edman degradation. This sequence was completely consistent with the most abundant protein in the salivary gland extracts, which was separated by two-dimensional gel electrophoresis. Based on the N-terminal amino acid sequence, the complete cDNA of this protein was cloned by 50 - and 30 -RACE using degenerate primers. The deduced NcSP84 contained an open reading frame of 2061 bp encoding a putative 687 amino acids with a putative signal sequence composed of 19 amino acids. The nucleotide and amino acid sequences of NcSP84 did not share statistically significant homology with any sequences in public databases. Motif search predicted that this protein had EF-hands, the most common motif found in Ca2þ -binding proteins. As predicted, NcSP84 exhibited Ca2þ-binding activity. The SDS-PAGE mobility of purified NcSP84 bound to Ca2þ tended to decline discretely, depending on the concentration of CaCl2 with which it was mixed for 1 h before adding SDS buffer. In situ hybridization and immunohistochemistry showed that the NcSP84 gene and gene product were expressed and stored in type III cells, which are the largest lobes in the primary salivary glands. The NcSP84 protein was detected in the phloem sap of rice exposed to leafhoppers, verifying that the NcSP84 protein was injected into the sieve tubes. These results suggest that NcSP84 could be secreted into the sieve tubes during feeding, which might bind Ca2þ ions that flow into sieve tubes in response to stylet puncturing. This might suppress sieveelement clogging and facilitate continuous ingestion from sieve tubes. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: Watery saliva Salivary glands EF-hand motif Calcium-binding protein Nephotettix cincticeps

1. Introduction The green rice leafhopper Nephotettix cincticeps is a major insect pest of rice in Japan. It uses stylets to intracellularly penetrate plant tissues, enabling the ingestion of nutrients mainly from the phloem and xylem (Oya, 1980; Kawabe, 1985). The leafhopper discharges watery and coagulable saliva in the feeding process (Sogawa, 1967, 1973). The saliva is considered to play important roles in inactivating toxic substances and ensuring continuous ingestion of phloem sap. Laccase has been identified as a salivary component in N. cincticeps, and its cDNA sequence was molecularly cloned from

* Corresponding author. Tel.: þ81 298 38 6085; fax: þ81 298 38 6028. E-mail address: [email protected] (M. Hattori). 1 Present address: Neurosensing and Bionavigation Research Center, Doshisha University, Kyotanabe 610-0321, Kyoto, Japan. 0965-1748/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.ibmb.2011.10.001

the salivary glands (Hattori et al., 2005, 2010). Two biological functions were proposed for this enzyme, i.e., contributing to the rapid polymerization of monolignols, which are potentially toxic substances, and salivary sheath coagulation. In contrast, little is known of the salivary components involved in continuous feeding, particularly from the phloem. The phloem plays a fundamental integrative role in the nutrition, development, and defense of higher plants (Turgeon and Wolf, 2009). The phloem transports hormones, mRNA, and proteins, which may mediate developmental and stress responses, along with photoassimilates (Thompson and Schultz, 1999; van Bel, 2003). The turgor pressure in sieve tubes can reach 30 bar (Geiger et al., 1973). Thus, sieve tubes must possess mechanisms for rapidly sealing or plugging sieve pores in response to injury (van Bel, 2006). Sieveelement sealing is mediated by a variety of substances including callose, forisomes, and plastid inclusions (King and Zeevaart, 1974; Sjolund et al., 1983; Knoblauch and van Bel, 1998; Walsh and

2

M. Hattori et al. / Insect Biochemistry and Molecular Biology 42 (2012) 1e9

Melarangno, 1981). Forisomes found in Vicia faba are crystalline protein bodies that plug sieve tubes by reacting with the Ca2þ ion released by wounded cells, which leads to dispersed swelling (Knoblauch et al., 2001). However, the vetch aphid Megoura viciae ingests phloem sap from V. faba by ejecting saliva containing a Ca2þbinding protein that suppresses this local defense response in V. faba (Will et al., 2007; Will and van Bel, 2008). Unlike aphids (Will and van Bel, 2006), little is known about the interaction between saliva and sieve tube occlusion in other phloem feeders such as leafhoppers. Our preliminary proteomic analysis of N. cincticeps saliva and salivary glands detected a protein with an 84-kDa that constituted a significant fraction of the total soluble salivary proteins. This salivary protein was predicted to contain EF-hand motifs, which are the most frequent motifs found in Ca2þ-binding proteins (Nakayama and Kretsinger, 1994). Thus, we cloned and sequenced the 84-kDa protein of N. cincticeps, which we designated NcSP84 (N. cincticeps Salivary Protein 84 kDa). We then characterized the biochemical properties of the protein based on the results of sequence analysis and elucidated its function in rice phloem feeding. 2. Material and methods 2.1. Insects and salivary glands N. cincticeps was successively maintained on rice seedlings in the laboratory. Salivary glands were dissected from adult females 3e7 days after emergence, as described previously (Hattori et al., 2005). 2.2. Collection of watery saliva from fed diet One milliliter of 4% sucrose in Milli-Q water was sandwiched between two sheets of parafilm membrane and attached to a 50mm petri dish. Approximately 15 adult leafhoppers were released into each of 72 dishes for 10 h at 25  C. Approximately 55 ml fed diet containing watery saliva was concentrated using an AmiconÒUltra-4 (10000 MWCO, Millipore, MA, USA), after filtering through a Minisart (1.2 mm Sartorius Stedim, Goettingen, Germany). 2.3. N-terminal amino acid sequence analysis of NcSP84 Concentrated watery saliva from the fed diet was separated by SDS-PAGE using 12.5% pre-cast e-PAGEL (ATTO Co., Tokyo, Japan) under reducing and heating conditions. The gel was blotted on a polyvinylidene difluoride (PVDF) membrane by tank blotting, using a Mini Trans-Blot Cell (Bio-Rad) at 300 mA for 1 h. For two-dimensional PAGE (2D-PAGE), 30 pairs of salivary glands were homogenized in lysis buffer containing 8 M urea, 2% NP-40, 2% ampholine (pH 3.5e10) (GE Healthcare, Piscataway, NJ, USA), 5% 2-mercaptoethanol and a protease inhibitor (Complete Mini EDTA-free tablets; Roche, Mannheim, Germany). After centrifugation at 14,000  g for 15 min at 4  C, proteins in the supernatant were separated using a tube gel containing 8 M urea, 3.5% acrylamide, and 2% ampholine for isoelectric focusing in the first dimension (Hirano, 1982). The second dimension was separated using 11% SDS-PAGE, and proteins on the gel were electroblotted onto PVDF membranes (ATTO, Tokyo, Japan) using a semidry blotter (Bio-Rad Trans-Blot SD). PVDF membranes were stained with Coomassie brilliant blue (CBB) R250 or G250. The 84kDa protein band from the fed diet or a large spot of the 80- to 90kDa protein from the salivary glands was excised from the membrane and subjected to Edman degradation in a gas-phase sequencer (HP 241; Hewlett-Packard, Palo Alto, CA, USA).

2.4. Cloning of the cDNA of the 84-kDa salivary protein The salivary glands (50 pairs) of adult females were ground in 200 ml of TRIzol reagent (Invitrogen, Carlsbad, CA). Total RNA was precipitated with isopropanol and reverse-transcribed with an anchor-tagged switching template (SMART II oligonucleotide; Clontech, Mountain View, CA) and an oligo-dT primer for 50 -rapid amplification of the cDNA end (50 -RACE) or with an anchor-tagged oligo-dT primer (for 30 -RACE) using PowerScript reverse transcriptase to prepare the first cDNA strand mixture. For 30 -RACE, a degenerate forward primer was designed for NcSP84-F (Table 1) based on the partial amino acid sequences (DTVPAEV) of NcSP84. For 50 -RACE, a gene-specific primer was designed for NcSP84-R based on the cDNA fragment obtained by 30 -RACE. RACE amplifications were performed with an Advantage PCR System (Clontech), following the manufacturer’s instructions. Bands of amplified cDNA fragments were excised and extracted using a QIAquick Gel Extraction Kit (Qiagen, Valencia, CA). Fragments were ligated into a pGEM-T Easy vector (Promega, Madison, WI) and transformed into DH5a cells. Sequences were determined using a BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA) and DNA analyzer (model 3700; PE Applied Biosystems). Full-length cDNA for NcSP84 was amplified using the primers in 50 UTR and 30 -UTR for the corresponding genes (Table 1). 2.5. Sequence analysis of the cDNA of the 84 kDa salivary protein The completed cDNA and deduced amino acid sequence of the 84-kDa salivary protein were analyzed using the BLAST algorithm. Predictions of signal peptide cleavage sites and glycosylation site usage were carried out with the programs SignalP, NetNGlyc, and NetOGlyc on the CBS Prediction Server (http://www.cbs.dtu.dk/ services/). Putative transmembrane domains (hydrophobicity sites) were predicted using SOSUI (http://bp.nuap.nagoya-u.ac.jp/ sosui/) (Hirokawa et al., 1998) and HMMTOP 2.0 (www.enzim.hu/ hmmtop/index.html) (Tusnády and Simon, 2001). Further biological information relating to protein, sequence motifs and structural domains was acquired using the PROSITE database (http://www. expasy.org/prosite/), which contains a large collection of biologically meaningful signatures described as patterns or profiles (Sigrist et al., 2010) and the Pfam profile HMM (Hidden Markov Model) search engine (http://pfam.sanger.ac.uk/) (Finn et al., 2010). 2.6. Purification of NcSP84 from the salivary glands for the calciumbinding experiment Heads containing salivary glands from approximately 1800 female leafhoppers were homogenized on ice in 30 ml of 50 mM Table 1 Primers used in this study. Orientation

Sequences (50 to 30 )

3 - and 5 -RACE NcSP84-F NcSP84-R

Forward Reverse

GAYACIGTICCIGCIGARGT ACCTTGAACGCGAGCTTCTGCAAGT

Full-length cDNA f-NcSP84F f-NcSP84R

Forward Reverse

CACAGTCCACAGCGTTTAAATG GACGTGCACTTCTATGATAACAA

RT-PCR RT-NcSP84F RT-NcSP84R RT-RpL19F RT-RpL19R

Forward Reverse Forward Reverse

AGCAGAAGAGTTTTCCAATGACTT CCTCAAATTCTTCAGGTGTAACAGT GCTTTGGTAAGAGGAAGGGTACTGC GCCTCATCTTCCTTCTGGTACGATT

Name 0

0

Abbreviations for positions with degenerate codons: I, Y, R, and H indicate inosine, C/T, A/G, and A/T/C, respectively.

M. Hattori et al. / Insect Biochemistry and Molecular Biology 42 (2012) 1e9

TriseHCl (pH 6.8) with 15% glycerol, 0.125% ascorbic acid, and protease inhibitors (Complete Mini tablets, Roche). The homogenate was mixed with 1.6 g of Chelex-100 resin (Bio-Rad), stirred for 3 h at 4  C to remove free Ca2þ, and then centrifuged at 15,000  g for 15 min at 4  C. The homogenate was precipitated with 90 ml acetone at 20  C for 16 h and centrifuged at 15,000  g for 15 min. The acetone precipitate was further dissolved in 70% ammonium sulfate solution for 2 h in an ice box. The resulting supernatant was applied to a Resource PHE (1 ml) column in an FPLC system (Pharmacia-LKB Biotechnology, Sweden) and eluted with a 3.0e0.0 M ammonium sulfate gradient in 50 mM TriseHCl (pH 7.8) at a flow rate of 0.5 ml/min. Fractions containing the 84kDa protein (confirmed by SDS-PAGE) were further applied to a HiLoad 26/60 Superdex 200 pg column. The proteins were eluted with 50 mM TriseHCl (pH 7.8) and 150 mM KCl at a flow rate of 0.5 ml/min. Fractions containing purified NcSP84 were pooled and concentrated using AmiconÒUltra-4. An additional 2000 heads containing salivary glands were separated by 2D-PAGE as described above, and gels were stained with EzStain Reverse (ATTO Corporation Tokyo, Japan). NcSP84 (approximately 1 mg) was excised from gel pieces in the 84-kDa spots and used to obtain a rabbit polyclonal antibody against NcSP84, after emulsification in complete Freund’s adjuvant. 2.7. Ca2þ-dependent mobility of purified NcSP84 under SDS-PAGE conditions The calcium-binding property of NcSP84 was confirmed by mixing purified NcSP84 (approximately 1 mg/4 ml) with equal volumes of the following solutions: 10 mM (final concentration) EDTA; deionized water; 0.0015, 0.05, 0.15, 0.5, 1.5, and 5 mM CaCl2; 0.5 mM CaCl2 (non-reducing/non-heating conditions); 0.5 mM CaCl2 (reducing/heating conditions); 5 mM MgCl2; or 5 mM ZnCl2. Each mixture was incubated for 30 min at 25  C, subsequently added to 8 ml of Laemmli sample buffer (Laemmli, 1970), and then subjected to SDS-PAGE under reducing/non-heating conditions, except for two lanes. 2.8. RT-PCR analysis Total RNA was prepared from salivary glands, stomach, Malpighian tubules, and cuticle using RNeasy Mini Kits (Qiagen) according to the manufacturer’s instructions. RNA was reversetranscribed using oligo-dT Primer. Each cDNA sample was amplified using Quick TaqÔ HS DyeMix (TOYOBO) with specific primers. The sense and antisense primers were RT-NcSP84F and RTNcSP84R for NcSP84, and RT-RpL19F and RT-RpL19R for the ribosomal protein L19 (Table 1). PCR was performed for 30 cycles as follows: 30 s at 94  C, 30 s at 55  C and 30 s at 68  C. The amplified DNAs were analyzed on a 1.8% agarose gel. 2.9. Immunohistochemistry Salivary glands were fixed in 4% paraformaldehyde (PFA)/PBS (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, and 1.8 mM KH2PO4) and shaken at 4  C for 2 h. Salivary glands were extensively washed with PBS and incubated with 6% H2O2/methanol for 30 min and stored in methanol at 20  C. Samples were subsequently treated with DMSO/methanol and 2% Triton X-100 and then washed with TST (100 mM TriseHCl, pH 7.8,150 mM NaCl, and 0.5% Triton X-100). Samples were then blocked with 5% dry non-fat milk in TST (TSTM) for 6 h and incubated overnight at 4  C with NcSP84 antibody at a 1:200 dilution in TSTM. Pre-immune serum was used as a negative control. After washing with TST, the samples were treated with horseradish peroxidase (HRP)-labeled goat anti-rabbit IgG (KPL) at

3

a 1:500 dilution in TST and incubated overnight at 4  C. Finally, samples were pre-incubated with TST containing DAB (125 mg/ml) for 1 h and reacted with the same concentration of DAB/TST containing hydrogen peroxide (0.05%) at 0  C for 15 min. The reaction was stopped by rinsing the samples with TST, before fixing in 4% PFA/ PBS and washing with PBS. 2.10. In situ hybridization In situ hybridization was accomplished by the procedure described by Hattori et al. (2010). Salivary gland specimens were cross-sectioned at 6 mm. Hybridization was performed with 300 ng of a DIG-labeled sense and anti-sense RNA probe corresponding to NcSP84 nucleotides 936e1418. Specific hybridization was detected using the anti-DIG alkaline phosphatase system (Roche Applied Science). Coloring reactions were carried out with NBT/BCIP solution (Sigma). 2.11. Daily variation in the NcSP84 protein and mRNA expression in the salivary glands Newly emerged female adults were collected from cages containing rice seedlings between 0e4 h before the light was turned off. Adults were sampled from the cage everyday for a period of 6 days. Each of five pairs of salivary glands from the insects was homogenized in 10 ml of 50 mM Tris buffer (pH 6.8) and 5 mM EDTA, and mixed with 10 ml Laemmli sample buffer (Laemmli, 1970). After centrifugation at 15,000  g for 10 min, each aliquot of the supernatant was subjected to 12.5% SDS-PAGE. After CBB staining, NcSP84 was quantified by scanning densitometry using a Bio-Rad densitometer. The relative amounts of protein in each lane were determined by comparison with CBB-stained gels of BSA samples. 2.12. Western blot analysis Six or no adult N. cincticeps females were introduced into a plastic tube (4  1.8 cm diameter) attached to a 25- to 30-day-old rice plant. After 16 h, the N. cincticeps females were replaced by three Nilaparvata lugens adult females. The stylets of N. lugens were then severed using the beam of a YAG laser (NEC Co., Japan) when the insects began to excrete honeydew (Kawabe et al., 1980; Hattori, 1997). The phloem sap exuded from the severed stylets was collected in a microcapillary tube. A total of 10 ml of sap from 6e8 plants was subjected to SDS-PAGE on a 12.5% gel before electrophoretic transfer onto a PVDF membrane, which was blocked with TBST and 5% non-fat, dry milk. The sample was probed with anti-NcSP84 rabbit anti-serum at a dilution of 1:2000 followed by HRP-labeled goat anti-rabbit Ig (KPL) at a dilution of 1:5000. Immunoblotted proteins were detected using EzWestBlue (ATTO, Japan). 3. Results 3.1. N-terminal amino acid analysis of NcSP84 from saliva and salivary glands SDS-PAGE analysis of saliva secreted in fed diet detected at least three proteins in watery saliva (Fig. 1A). Edman degradation determined the N-terminal amino acid sequence of the major band NcSP84 as SSDTVPAEVQ. This N-terminal sequence was consistent with the 84-kDa spot (SSDTVPAEVQTIVKTPGH) that was the most abundant protein in the salivary gland extracts separated by 2DPAGE (Fig. 1B).

4

M. Hattori et al. / Insect Biochemistry and Molecular Biology 42 (2012) 1e9

Fig. 1. A: Proteins in watery saliva recovered from 5% sucrose diet ingested for 10 h by N.cincticeps. Salivary proteins concentrated by ultrafiltration were subjected to SDS-PAGE (12.5% polyacrylamide gel). B: 2D-PAGE pattern for N. cincticeps salivary gland proteins. Proteins were extracted from 50 pairs of salivary glands from adult females. These proteins were separated by 2D-PAGE involving IEF performed using a tube gel containing 8 M urea, 3.5% acrylamide, and 2% ampholine in the first dimension and SDS-PAGE performed using 11% gel for analysis in the second dimension. The resolved proteins on the gel were electroblotted onto PVDF membranes, and then detected by CBB G250 or R250 staining. The triangular arrows denote NcSP84, N-terminal sequences, which is “SSDTV..”.

3.2. Cloning and analysis of NcSP84 cDNA The nucleotide sequence of the cDNA and deduced amino acid sequence of NcSP84 are shown in Supplement 1. The nucleotide sequence contained an open reading frame of 2061-bp encoding a putative 687 amino acids, with a putative signal peptide

composed of 19 aa. The residues at the beginning of the predicted mature proteins were identical to the N-terminal sequences (SSDTVPAEVQ) that were determined in the 84-kDa protein band on the SDS-PAGE gel and a spot on the 2D-PAGE gel. The calculated molecular weight and pI of the mature protein were 77,557 Da and pH 4.67, respectively. The protein was predicted to have one N-

Fig. 2. Consensus sequence of the canonical EF-hand predicted for NcSP84. The positions of the E and F helices and the intervening loop are shown in the consensus amino acid sequence. The PROSITE EF-hand signature corresponds to the 12 amino acids of the loop. Domain analysis using PROSITE showed that the protein contains five profile PS50222 (EF_HAND_2) EF-hand calcium-binding domain profiles at the indicated positions. Motif analysis using Pfam indicated a cysteine-rich, acidic secretory protein with a calciumbinding region.

M. Hattori et al. / Insect Biochemistry and Molecular Biology 42 (2012) 1e9

linked glycosylation site within the signal sequence but no O-linked glycosylation site. SOSUI and HMMTOP methods predicted a soluble protein with no transmembrane helices in the sequence. BLASTP comparison of the amino acid sequence in Supplement 1 with sequences in the GenBank data base identified two sequences with a low degree of homology. The novel protein was similar to solute carrier family 25 (mitochondrial carrier; phosphate carrier), member 25 (slc25a25) (Danio rerio) with 47% positives (E-value ¼ 0.030), and the predicted calcium-binding p22-like protein (Amphimedon queenslandica) with 59% positives (E-value ¼ 0.062). The PROSITE protein motif search indicated five EFhand calcium-binding domain profiles (EF_HAND_2, PS50222) in NcSP84, despite the low score (Fig. 2). The canonical EF-hand contains a characteristic helix-loop-helix binding motif; the Ca2þbinding loop contains 12 amino acids, beginning with an aspartate and ending with a glutamate (Grabarek, 2006). The amino acid sequence of the loop segment matched the canonical EF-hand (Fig. 2). However, the central glycine residue found in EF-hand motifs was replaced by asparagine, serine, or histidine. The domain search against the Pfam profile HMM predicted a cysteinerich acidic secretory protein with a calcium-binding region domain (SPARC_Ca_bdg) (379e431, E-value ¼ 0.19) with two EF-hands (Fig. 2). The protein was also predicted to have an EF-hand domain (481e499, E-value ¼ 0.2) and a domain of unknown function (DUF3384, not shown) (536e664, E-value ¼ 0.079). 3.3. Purification of NcSP84 from the salivary glands NcSP84 was precipitated from the heads of adult females containing the salivary glands using cold acetone. The acetone precipitate was dissolved with 70% ammonium sulfate solution and kept for 2 h in the ice box. To the supernatant, solid ammonium sulfate was added to a final concentration of 3 M, and the supernatant then was applied to hydrophobic chromatography (Supplement 2). Fractions containing NcSP84 (3 ml) were further purified by gel filtration chromatography (Supplement 3), and those containing NcSP84 (4 ml) were concentrated. Four hundred and eighty microgram of nearly pure NcSP84 was obtained from about 1800 female heads (2.8 mg protein). 3.4. Ca2þ-dependent mobility of purified NcSP84 under SDS-PAGE conditions

5

Fig. 3. Ca2þ-dependent mobility of purified NcSP84 under SDS-PAGE conditions. The NcSP84 protein (approximately 1 mg/lane) was incubated for 30 min at 25  C with each of the following solutions: Lane 1, 5 mM EDTA; Lane 2, deionized water; Lane 3, 0.015 mM CaCl2; Lane 4, 0.05 mM CaCl2; Lane 5, 0.15 mM CaCl2; Lane 6, 0.5 mM CaCl2; Lane 7, 1.5 mM CaCl2; Lane 8, 5 mM CaCl2; Lane 9, 0.5 mM CaCl2 (non-reducing); Lane 10, 0.5 mM CaCl2 (heating); Lane 11, 5 mM MgCl2; and Lane 12, 5 mM ZnCl2. Each mixture was loaded onto the gel after adding SDS buffer under non-heating and reducing conditions, except for Lanes 9 and 10.

the salivary glands immediately after adult emergence, before reaching 0.9 mg (BSA equivalent) during the period of 1e7 days after emergence (Fig. 4).

3.6. Expression of NcSP84 mRNA in different organs The organ-specific expression of NcSP84 in vivo and the time course of NcSP84 expression in the salivary glands were examined using RT-PCR. NcSP84 mRNA was strongly expressed in the salivary glands but not in the stomach (midgut), Malpighian tubules, or the cuticles (epidermis) (Fig. 5A). A slight increase in NcSP84 mRNA levels was observed in the salivary glands during day 1 (compared with day 0), which remained constant until 6 days after adult emergence, as shown in Fig. 5B.

Purified NcSP84 protein was subjected to SDS-PAGE after mixing with different concentrations of CaCl2 to determine whether the predicted EF-hand motifs of NcSP84 possessed any calcium-binding capability. The mobility of the NcSP84 band on the SDS-PAGE gel was decreased by the addition of CaCl2 at a concentration of 0.015 to 5 mM. One to three obscure zones with lower mobility were observed in the lanes in which the protein was incubated with higher concentrations of calcium (Fig. 3). Thus, NcSP84 mobility appeared to decline discretely, depending on the concentration of calcium. In contrast, NcSP84 showed no reduction in migration when a sample mixed with CaCl2was heated before subjecting to SDS-PAGE (lane 10). Thus, Ca2þ bound to the protein may have been detached and replaced by SDS under heating conditions. The protein also exhibited selective binding to Ca2þ over Mg2þ and Zn2þ cations, even at a concentration of 5 mM (lane 11, 12). 3.5. Daily variation in the NcSP84 protein in the salivary glands of females after adult emergence The amount of NcSP84 per female was examined during the 7 days after adult emergence. A low level of NcSP84 was produced in

Fig. 4. Daily variation in the NcSP84 protein in the salivary glands of females after adult emergence. Five adult females were tested each day. The amount of NcSP84 is expressed as BSA equivalents. The error bars indicate the standard error.

6

M. Hattori et al. / Insect Biochemistry and Molecular Biology 42 (2012) 1e9

Fig. 5. Expression of the NcSP84 gene in different organs and its daily variation in the salivary glands of females after adult emergence. A: RT-PCR was performed on cDNA prepared from salivary glands, stomach (midgut), Malpighian tubules, and the newly ecdysed cuticle of adult females. NcSP84 was amplified using gene-specific primers. Ribosomal protein L19 primers were used as a control in equal template amounts. B: cDNA was prepared from salivary glands of adult female 0e6 days after adult emergence.

3.7. Localization of the NcSP84 protein and mRNA in the salivary glands Immunohistochemical analysis indicated that the type III cells stained positive for the NcSP84 protein (Fig. 6). Pre-immune serum showed no significant level in the salivary glands (Fig. 6). In situ hybridization showed that the type III cells in the posterior lobe of the main salivary glands were positively stained for the NcSP84 transcript (Fig. 7). The negative control with a sense probe showed no hybridization to RNA in the gland (Fig. 7). Thus, the immunolocalization of the NcSP84 protein in the salivary glands agreed with the region where the NcSP84 transcript was localized. 3.8. Western blot analysis of NcSP84 in the phloem sap of rice ingested by leafhoppers Western blot analysis with NcSP84 antibodies was performed to verify whether NcSP84 was secreted into the sieve tube of rice plants. The NcSP84 protein was detected in the phloem sap of plants exposed to leafhoppers for 16 h (lane 2), whereas it was undetectable in the sap of plants unexposed to leafhoppers (lane1, Fig. 8). NcSP84 was detected in protein extracts from the salivary glands (0.004 female salivary gland equivalents), as shown in lane 3 of the control. These results indicate that NcSP84 was injected as a salivary component into the sieve tube of rice plants during N. cincticeps feeding. 4. Discussion Preliminary proteomic analysis showed that the watery saliva and salivary glands of N.cincticeps contained NcSP84 as the most abundant protein. RT-PCR, in situ hybridization and immunohistochemistry experiments indicated that NcSP84 was specifically expressed and stored in the type III cells (lobes) of the salivary

glands. The type III cells are the largest of the principal salivary glands, which consist of four types of secretory cells (lobes), and their cytoplasm is largely occupied by intracellular canaliculi. These are presumed to be serous cells for storage of aqueous saliva because of their characteristic alveolate-like structure (Sogawa, 1965). Thus, it was considered that NcSP84 is predominantly produced in this particular lobe of the salivary glands and secreted as a major salivary component during feeding. Molecular cloning and sequence analysis using bioinformatics approaches showed that the NcSP84 gene had a signal sequence with low homology to any previously reported sequence. Protein motif search for NcSP84 using PROSITE predicted the existence of five EF-hand motifs despite the low scores. Domain search against Pfam also predicted a SPARC Ca bdg (379e431), which contains two EF-hands (Delostrinos et al., 2006) and another independent EFhand motif (481e499). The EF-hand structure is found in a large set of Ca2þ-binding proteins, such as calbindin D9K, parvalbumin, and calmodulin, which contain two, three, and four EF-hands, respectively (Kawasaki et al., 1998; Lewit-Bentley and Rety, 2000). EF-hand motifs almost always occur in pairs (Nelson and Chazin, 1998), and in parvalbumin, two of the three EF-hands are paired to bind Ca2þ tightly, while the first N-terminal EF-hand does not bind Ca2þ (Swain et al., 1989). Multiple EF-hand motifs were predicted in the NcSP84 sequence, which raised the possibility that this protein has Ca2þ-binding ability. However, the number of NcSP84 EF-hands participating in Ca2þ binding remains unclear. The mobility of NcSP84 on the SDS-PAGE gel was discretely reduced by increasing the concentration of calcium from 0.015 to 5 mM. NcSP84 binding to calcium via the EF-hand motif led to a shift in its mobility on SDS-PAGE gels, with a more slowly migrating form of NcSP84 bound to the calcium cation. This behavior on the gel depended on the calcium concentration, which was consistent with the predicted multiple EF-hand motifs in NcSP84. NcSP84 was secreted as an apoprotein in watery saliva and

Fig. 6. Immunohistochemical localization of the NcSP84 protein in the posterior lobe of salivary glands using whole-mount preparations. Left, antibody-detected; Right, control of pre-immune rabbit serum. The localization of signals was detected in the type III cells of the salivary glands. The tissue marked with an arrow is a type III cell. The bar represents 100 mm.

M. Hattori et al. / Insect Biochemistry and Molecular Biology 42 (2012) 1e9

7

Fig. 7. Localization of NcSP84 mRNA in the posterior lobe of the salivary gland tissue sections by in situ hybridization. Left, anti-sense; Right, sense probe. The tissue marked with an arrow is a type III cell. Localization of signals was detected in the type III cells of the salivary glands. The bar represents 100 mm.

it may have bound to multiple calcium ions with each molecule in plant tissues after being injected. The protein exhibited selectivity for Ca2þ over the chemically similar Mg2þ and Zn2þ cations. The migratory difference between NcSP84 in its apo-form and its calcium-binding form under SDS-PAGE does not necessarily imply a conformational change. Under this condition, the native structure of the protein could be lost by addition of SDS, which provides a negative charge. Instead, NcSP84 would have a higher positive electrical charge depending on the number of Ca2þ ions bound, which is a divalent cation. Thus, NcSP84 bound to Ca2þ ions might counteract the negative charges contributed by SDS, resulting in the restricted movement to the anode side. Functional EF-hand proteins can be divided into two general classes, i.e., Ca2þ signaling proteins and the Ca2þ buffering/transport proteins (Bhattacharya et al., 2004; Gifford et al., 2007). The first group is the largest and includes family members such as calmodulin and troponin C. These proteins typically undergo a Ca2þ-dependent conformational change which opens a target binding site. The second group is represented by parvalbumin and calbindin D9k and they do not change their conformation on Ca2þ

Fig. 8. Detection of the NcSP84 protein in the phloem sap of rice plants using purified NcSP84 antibodies. Lane 1: Phloem sap of rice plant collected without leafhopper feeding, which served as a negative control. Lane 2: Phloem sap of rice plant collected after leafhopper feeding. Lane 3: Protein extract of salivary glands (0.004 female). Lane 4: Pre-stained marker protein.

binding, which appears to preclude their Ca2þ-dependent regulatory function (Nelson and Chazin, 1998). NcSP84 may perform a function as a Ca2þ buffer rather than having a direct regulatory role in sieve elements because the protein is secreted with saliva and does not originally exist in the sieve elements. A Ca2þ-binding protein in insect saliva was first found in the fed diet secreted by the vetch aphid, M. viciae (Will et al., 2007), although its molecular structure remains obscure. Carolan et al. (2009) reported that saliva of the pea aphid, Acyrthosiphon pisum contains a protein which is a member of the SMP-30 family including regucalcin. However, no obvious Ca2þ-binding domain has been confirmed. Two proteins that were predicted to have an EF-hand motif were found in the salivary glands of N. lugens (Konishi et al., 2009), although whether these proteins exhibit Ca2þ affinity in secreted saliva is unknown. Thus, Ca2þ-binding proteins may be a universal component in the saliva of phloem feeders. Blood-feeding arthropods including the mosquito have mechanisms to effectively block the hemostasis system with their saliva, which contains a mixture of secreted proteins. The saliva inhibits platelet aggregation and blood coagulation and causes vasodilation resulting in successful feeding (Ribeiro and Francischetti, 2003). On the other hand, aphids, phytophagous Hemiptera have been suggested to feed from sieve elements by circumventing or suppressing the sieve plate sealing response (Douglas, 2003; Tjallingii, 2006). Sieve tubes possess various mechanisms for sealing or plugging sieve pores in response to injury, which prevents the loss of phloem sap. Slow sealing is mediated by callose, which is b-1,3 glucan polymer produced around sieve pores, and plasmodesmata (King and Zeevaart, 1974; Zabotin et al., 2002). Rapid plugging is mediated by phloem-specific proteins in dicotyledons, such as Pprotein crystalloids in Fabaceae (Knoblauch et al., 2001) and phloem proteins 1 (PP1) and 2 (PP2) in Cucurbitaceae (Read and Northcote, 1983). Ca2þ is an important mediator of sieve tube plugging in various plants, including rice, where Ca2þ chelating agents such as EDTA have been demonstrated to prevent sieve tube occlusion (King and Zeevaart, 1974; Fellows et al. 1978; Grayer et al., 1994; Knoblauch

8

M. Hattori et al. / Insect Biochemistry and Molecular Biology 42 (2012) 1e9

et al., 2001). Increased cytoplasmic Ca2þ can induce the formation of callose, which is known to be involved in the wound response in sieve tubes (Kauss, 1987; Evert, 1982). Ca2þ is required for callose synthase activity in some plant species (Kauss et al., 1983; Delmer et al., 1984; Morrow and Lucas, 1986; Hayashi et al., 1987). An influx of Ca2þ ions into the sieve elements of Fabaceae stimulates the reversible dispersal of crystalloid P-protein, which occludes sieve plate pores (Knoblauch et al., 2001). The forisomes of V. faba are crystalline protein bodies that are assumed to adopt a crystalline condensed conformation in low Ca2þ solutions while swelling to a dispersed conformation in high Ca2þ solutions (Knoblauch et al., 2003). Aphids feeding on Fabaceae plants are predicted to have salivary Ca2þ chelating activity, which prevents Ca2þ-dependent plugging of the plant sieve elements (Will et al., 2007). Will and van Bel (2008) reported that Ca2þ-binding proteins in the watery saliva of M. viciae counteract forisomes, which induce Ca2þ-dependent sieve plate occlusion in Vicia faba. Interactions between aphid saliva and sieve-element proteins might be a universal mechanism for preventing sieve plate occlusion (Will et al., 2009). The monocotyledons, which lack structural proteins such as forisomes, release the proteinaceous inclusions from sieve-element plastids to seal sieve pores (Walsh and Melarangno, 1981; Eleftheriou, 1990; Paivaa and Machado, 2008). The detailed molecular mechanism governing sieve tube plugging or sealing in rice plants remains incompletely understood, although callose deposition was observed in the sieve tubes of rice around N. lugens puncture sites (Hao et al., 2008). However, Ca2þ-dependent protein kinases have been identified in the phloem sap of rice, suggesting a role for Ca2þ in signaling within sieve tubes (Nakamura et al., 1995). Our in vitro experiments showed that NcSP84 can bind Ca2þ ion and that this protein is injected into sieve tubes during feeding. We have no direct evidence of NcSP84 suppressing the occlusion of sieve tubes in rice, but the findings in this study may support the following hypotheses. NcSP84 secreted in its apo-form rapidly binds Ca2þ ions that flow into sieve tubes in response to stylet puncturing, thereby preventing an increase in the Ca2þ concentration in the sieve tube. This may inhibit activation of callose synthase or Ca2þ-dependent protein kinases, which initiate the signal transduction pathway that induces sieve tube plugging. Phloem sap exudation of rice plant was maintained for more than 3 h after the stylets of N. lugens were severed by laser beams during feeding from the sieve elements (Kawabe et al., 1980). This indicates that continuous injection of saliva into the sieve tubes throughout feeding is not necessary for suppressing sieve tube occlusion in rice. Thus, injection of a large amount of NcSP84 might be required to adequately trap Ca2þ ions immediately after the stylets penetrate the sieve tubes. In summary, we demonstrated that the green rice leafhopper secreted a novel Ca2þ-binding protein with multiple EF-hand motifs in its saliva, which might facilitate continuous feeding by inhibiting sieve-element plugging. Acknowledgments We thank M. Watanabe for her assistance with the experiments, and H. Niki and K. Hashino for their help in rearing insects. This work was partly supported by a Grant-in-Aid for Scientific Research, KAKENHI (22580063) to M.H. Appendix. Supplementary data Supplementary data associated with this article can be found in the online version, at doi:10.1016/j.ibmb.2011.10.001.

References Bhattacharya, S., Bunick, C.G., Chazin, W.J., 2004. Target selectivity in EF-hand calcium binding proteins. Biochim. Biophys. Acta-Mol. Cell Res. 1742, 69e79. Carolan, J.C., Fitzroy, C.F., Douglas, A.E., Wilkinson, T.L., 2009. The proteome of pea aphid saliva characterized by LC/MS-MS. Proteomics 9, 2457e2467. Delmer, D.P., Thelen, M., Marsden, M.P.F., 1984. Regulatory mechanisms for the synthesis of -glucans in plants. In: Dugger, W.M., Bartnicki, S. (Eds.), Structure, Function, and Biosynthesis of Plant Cell Walls. American Society of Plant Physiologists, Rockville, MD, pp. 133e149. Delostrinos, C.F., Hudson, A.E., Feng, W.C., Kosman, J., Bassuk, J.A., 2006. The Cterminal Ca2þ-binding domain of SPARC confers anti-spreading activity to human urothelial cells. J. Cell. Physiol. 206, 211e220. Douglas, A.E., 2003. The nutritional physiology of aphids. Adv. Insect. Physiol. 31, 73e140. Eleftheriou, E.P., 1990. Monocotyledons. In: Behnke, H.D., Sjolund, R.D. (Eds.), Sieve Elements. Springer-Verlag, Berlin, pp. 103e137. Evert, R.F., 1982. Sieve-tube structure in relation to function. Bio. Sci. 32, 789e795. Fellows, R.J., Egli, D.E., Leggett, J.E., 1978. A pod leakage technique for phloem translocation studies in soybean (Glycine max [L.] Merr.). Plant Physiol. 62, 812e814. Finn, R.D., Mistry, J., Tate, J., Coggill, P., Heger, A., Pollington, J.E., Gavin, O.L., Gunesekaran, P., Ceric, G., Forslund, K., Holm, L., Sonnhammer, E.L., Eddy, S.R., Bateman, A., 2010. The Pfam protein families database: nucleic acids research. Database Issue 38, D211eD222. Geiger, D.R., Giaquinta, R.T., Sovonick, S.A., Fellows, R.J., 1973. Solute distribution in sugar beet leaves in relation to phloem loading and translocation. Plant Physiol. 52, 585e589. Gifford, J.L., Walsh, M.J., Vogel, P.H., 2007. Structures and metal-ion-binding properties of the Ca2þ-binding helixeloopehelix EF-hand motifs. Biochem. J. 405, 199e221. Grabarek, Z., 2006. Structural basis for diversity of the EF-hand calcium-binding proteins. J. Mol. Biol. 359, 509e525. Grayer, R.J., Harborne, J.B., Kimmins, F.M., Stevenson, P.C., Wijayagunasekera, H.N.P., 1994. Phenolics in rice phloem sap as sucking deterrents to the brown planthopper, Nilaparvata lugens. Acta Horticult. 381, 691e694. Hao, P., Liu, C., Wang, Y., Chen, R., Tang, M., Du, B., Zhu, L., He, G., 2008. Herbivoreinduced callose deposition on the sieve plates of rice: an important mechanism for host resistance. Plant Physiol. 146, 1810e1820. Hattori, M., 1997. Feeding behavior of the green rice leafhopper, Nephotettix cincticeps (Homoptera: Cicadellidae) to pure phloem sap collected from the resistant and susceptible rice varieties. Appl. Entomol. Zool. 32, 409e412. Hattori, M., Konishi, H., Tamura, Y., Konno, K., Sogawa, K., 2005. Laccase-type phenoloxidase in salivary glands and watery saliva of the green rice leafhopper, Nephotettix cincticeps. J. Insect Physiol. 51, 1359e1365. Hattori, M., Tsuchihara, K., Noda, H., Konishi, H., Tamura, Y., Shinoda, T., Nakamura, M., Hasegawa, T., 2010. Molecular characterization and expression of laccase genes in the salivary glands of the green rice leafhopper, Nephotettix cincticeps (Hemiptera: Cicadellidae). Insect Biochem. Mol. Biol. 40, 331e338. Hayashi, T., Read, S.M., Bussell, J., Thelen, M., Lin, F.C., Brown, R.M., Delmer, D.P., 1987. UDP-Glucose: (1 / 3)-beta-glucan synthases from mung bean and cotton: differential effects of ca and mg on enzyme properties and on macromolecular structure of the glucan product. Plant Physiol. 83, 1054e1062. Hirokawa, T., Boon-Chieng, S., Mitaku, S., 1998. SOSUI: classification and secondary structure prediction system for membrane proteins. Bioinformatics 14 (4), 378e379. http://bp.nuap.nagoya-u.ac.jp/sosui/. Hirano, H., 1982. Varietal differences of leaf protein profiles in mulberry. Phytochemistry 21, 1513e1518. Kauss, H., Köhle, H., Jeblick, W., 1983. Proteolytic activation and stimulation by Ca2þ of glucan synthase from soybean cells. FEBS Lett. 158, 84e88. Kauss, H., 1987. Some aspects of calcium-dependent regulation in plant metabolism. Ann. Rev. Plant Physiol. 38, 47e72. Kawabe, S., Fukumorita, T., Chino, M., 1980. Collection of rice phloem sap from stylets of homopterous insects severed by YAG laser. Plant Cell Physiol. 21, 1319e1327. Kawabe, S., 1985. Mechanism of varietal resistance to the rice green leafhopper, Nephotettix cincticeps Uhler. Jpn. Agric. Res. Quater. 19, 115e124. Kawasaki, H., Nakayama, S., Kretsinger, R.H., 1998. Classification and evolution of EF-hand proteins. Biometals 11, 277e295. King, R.W., Zeevaart, J.A., 1974. Enhancement of phloem exudation from cut petioles by chelating-agents. Plant Physiol. 53, 96e103. Knoblauch, M., van Bel, A.J.E., 1998. Sieve tubes in action. Plant Cell 10, 35e50. Knoblauch, M., Peters, W.S., Ehlers, K., van Bel, A.J.E., 2001. Reversible calciumregulated stopcocks in legume sieve tubes. Plant Cell 13, 1221e1230. Knoblauch, M., Noll, G.A., Müller, T., Prüfer, D., Schneider-Hüther, I., Scharner, D., van Bell, A.J.E., Peters, W.S., 2003. ATP-independent contractile proteins from plants. Nat. Mater. 2, 600e603. Konishi, H., Noda, H., Tamura, Y., Hattori, M., 2009. Proteomic analysis of the salivary glands of the rice brown planthopper, Nilaparvata lugens (Stål) (Homoptera: Delphacidae). Appl. Entomol. Zool. 44, 525e534. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227, 680e685. Lewit-Bentley, A., Rety, S., 2000. EF-hand calcium-binding proteins. Curr. Opin. Str. Biol. 10, 637e643.

M. Hattori et al. / Insect Biochemistry and Molecular Biology 42 (2012) 1e9 Morrow, D.L., Lucas, W.J., 1986. (1 / 3)-b-D-glucan synthase from sugar beet. I. Isolation and solubilization. Plant Physiol. 81, 171e176. Nakamura, S., Hayashi, H., Mori, S., Chino, M., 1995. Detection and characterization of protein kinases in rice phloem sap. Plant Cell Physiol. 36, 17e27. Nakayama, S., Kretsinger, R.H., 1994. Evolution of the EF-hand family of proteins. Annu. Rev. Biophys. Biomol. Struct. 23, 473e507. Nelson, M.R., Chazin, W.J., 1998. Structure s of EF-hand Ca2þ-binding proteins: diversity in the organization, packing and response to Ca2þ binding. Biometals 11, 297e318. Oya, S., 1980. Feeding habits and honeydew components of the green rice leafhopper, Nephotettix cincticeps UHLER (Hemiptera: Deltocephalidae). Appl. Entomol. Zool. 15, 392e399. Paivaa, É.A.S., Machado, S.R., 2008. Can sieve-element plastids in Panicum maximum (Poaceae) leaves act in the blockage of injured sieve-tube elements? Flora Morphol. Distribut. Funct. Ecol. Plants 203, 327e331. Read, S.M., Northcote, D.H., 1983. Chemical and immunological similarities between the phloem proteins of three genera of the Cucurbitaceae. Planta 158, 119e127. Ribeiro, J.M., Francischetti, I.M., 2003. Role of arthropod saliva in blood feeding: sialome and post-sialome perspectives. Annu. Rev. Entomol. 48, 73e88. Sigrist, C.J.A., Cerutti, L., de Castro, E., Langendijk-Genevaux, P.S., Bulliard, V., Bairoch, A., Hulo, N., 2010. PROSITE, a protein domain database for functional characterization and annotation. Nucleic Acids Res. 38, 161e166. Sjolund, R.D., Shiha, C.Y., Jensen, K.G., 1983. Freeze-fracture analysis of phloem structure in plant tissue cultures: III. P-protein, sieve area pores, and wounding. J. Ultrastruct. Res. 82, 198e211. Sogawa, K., 1965. Studies on the salivary glands of rice plant leafhoppers: I. Morphology and histology. Jpn. J. Appl. Entomol. Zool. 19 (4), 275e290. Sogawa, K., 1967. Studies on the salivary glands of rice plant leafhoppers. II. Origins of the structural precursors of the sheath material. Appl. Entomol. Zool. 2, 195e202. Sogawa, K., 1973. Feeding of the rice plant- and leafhoppers. Rev. Plant Protect. Res. 6, 31e43.

9

Swain, A.L., Kretsinger, R.H., Amma, E.L., 1989. Restrained least squares refinement of native (Ca2þ) and Cd- substituted carp parvalbumin using X-ray crystallographic data at 1.6-Å resolution. J. Biol. Chem. 264, 16620e16628. Tjallingii, W.F., 2006. Salivary secretions by aphids interacting with proteins of phloem wound responses. J. Exp. Bot. 57, 739e745. Thompson, G.A., Schultz, A., 1999. Macromolecular trafficking in the phloem. Trends Plant Sci. 4, 354e360. Turgeon, R., Wolf, S., 2009. Phloem transport: cellular pathways and molecular trafficking. Annu. Rev. Plant Biol. 60, 207e221. Tusnády, G.E., Simon, I., 2001. The HMMTOP transmembrane topology prediction server. Bioinformatics 17, 849e850. Walsh, M.A., Melarangno, J.E., 1981. Structural evidence for plastid inclusions as a possible ‘sealing’ mechanism in the phloem of monocotyledons. J. Exp. Bot. 32, 311e320. van Bel, A.J.E., 2003. The phloem, a miracle of ingenuity. Plant. Cell. Environ. 26, 125e149. van Bel, A.J.E., 2006. Sieve-pore plugging mechanisms in cellecell channels. In: Baluska, F., Volkmann, D., Barlow, P.W. (Eds.), pp. 1e6. WWW.Eurekah.com. Will, T., van Bel, A.J.E., 2006. Physical and chemical interactions between aphids and plants. J. Exp. Bot. 57, 729e737. Will, T., Tjallingii, W.F., Thonnessen, A., van Bel, A.J.E., 2007. Molecular sabotage of plant defense by aphid saliva. Proc. Natl. Acad. Sci. U.S.A. 104, 10536e10541. Will, T., van Bel, A.J.E., 2008. Induction as well as suppression: how aphid saliva may exert opposite effects on plant defense. Plant Signal Behav. 3, 427e430. Will, T., Kornemann, S.R., Furch, A.C., Tjallingii, W.F., van Bel, A.J.E., 2009. Aphid watery saliva counteracts sieve-tube occlusion: a universal phenomenon? J. Exp. Biol. 212, 3305e3312. Zabotin, A.I., Barysheva, T.S., Trofimova, O.I., Lozovaya, V.V., Widholm, J., 2002. Regulation of callose metabolism in higher plant cells. Russ. J. Plant Physiol. 49, 792e798.