- Split View
-
Views
-
Cite
Cite
J. David McGee, Judith L. Roe, Teresa A. Sweat, Xuemin Wang, James A. Guikema, Jan E. Leach, Rice Phospholipase D Isoforms Show Differential Cellular Location and Gene Induction, Plant and Cell Physiology, Volume 44, Issue 10, 15 October 2003, Pages 1013–1026, https://doi.org/10.1093/pcp/pcg125
- Share Icon Share
Abstract
Phospholipase D (PLD) has emerged as an important enzyme involved in signal transduction, stress responses, protein trafficking, and membrane metabolism. This report describes the cloning and characterization of three novel PLD genes from rice, designated RPLD3, RPLD4 and RPLD5. The rice PLDs, including the previously isolated RPLD1 and RPLD2, are similar to PLD subfamilies of Arabidopsis. Based on sequence homology and domain conservation, RPLD1 is most similar to the PLDα subfamily of PLDs while RPLD5 most closely resembles the PLDδ type. RPLD2, 3 and 4 represent a unique subfamily, although they are most similar to PLDα. RPLD1 is located on chromosome 1, RPLD5 on chromosome 3, and RPLD2, RPLD3, and RPLD4 are tandemly arrayed on chromosome 5. Transcriptional analysis reveals that RPLD1, present in healthy rice vegetative tissues, is induced rapidly but transiently in wounded leaf tissues. RPLD2, also induced by wounding, is present at lower levels but for a more prolonged duration than RPLD1. Immunolocalization with peptide specific antibodies to each of the five PLDs was used to demonstrate that the isoforms have overlapping but distinct patterns of distribution in healthy rice cells. RPLD1 was detected in mesophyll cell wall, membranes, and chloroplasts, whereas RPLD3 and RPLD4 were located predominantly in the chloroplasts. Labeling of RPLD2 and RPLD5 was sparse, and was most concentrated in the secondary walls of xylem (RPLD2) and guard cells (RPLD2 and RPLD5). This combined information on structural features, expression profiles, and cellular localization will assist the basis for dissection of PLD isoform function in rice.
(Received March 14, 2003; Accepted July 17, 2003)
Introduction
Phospholipase D (PLD) catalyzes the hydrolysis of phospholipids, producing phosphatidic acid (PA) and a hydrophylic head group. This activity, first identified over 50 years ago in carrot as a lecithinase capable of removing choline from plant phospholipids (Hanahan and Chaikoff 1947), has since been observed in plants, animals, fungi and bacteria. Historically, PLD activity in plants was thought to involve primarily phospholipid catabolism, initiating lipolytic cascades that lead to membrane degradation during senescence and necrosis. Recent progress following the cloning of PLDs has greatly expanded this limited view. Models now propose that phospholipid degradation products (PA, diacylglycerol, lyso-PA and free fatty acids) resulting from direct or indirect PLD activity (a) act as second messengers for transmembrane and intercellular signaling, and (b) alter the local structural characteristics of target membranes allowing such mechanisms as localized membrane budding and vesicle fusion (for reviews see Chapman 2000, Munnik 2001, Jones et al. 1999, Liscovitch et al. 2000, Wang 2001, Wang 2002).
Since the cloning of the first eukaryotic PLD gene in castor bean (Wang et al. 1994), PLD cDNAs have been cloned from Arabidopsis, rice, maize, Craterostigma plantagineum, Pimpinella brachycarpa, tomato, cabbage, cowpea, and tobacco (Dyer et al. 1995, Qin et al. 1997, Ueki et al. 1995, Morioka et al. 1997, Frank et al. 2000, Cha et al. 1998, Kim et al. 1999). The identification, cloning and expression of these genes established that plant PLDs are a family of heterologous enzymes that differ in catalytic and regulatory properties. In recent studies, PLD activity has been implicated in a variety of cellular regulation, and intracellular and transmembrane signaling events. These include roles in regulating stress responses (nutritional starvation, freezing, irradiation, wounding), senescence, phytohormone signaling, seed germination, protein secretion, and cytoskeletal rearrangements (for review see Chapman 2000, Munnik 2001, Wang 2001, Wang 2002). Work in our laboratory suggests that at least one isoform of PLD is involved in plant defense (Young et al. 1996). Our overall objectives are to dissect the roles of various PLDs in rice development and responses to pathogen and environmental stresses. Here, we describe the isolation and characterization of three novel PLD genes from rice and discuss their relationship to two previously characterized rice PLDs and to other known plant PLDs. We present transcriptional analysis of the respective PLDs in response to wounding and to infection by the rice bacterial blight pathogen, Xanthomonas orzyae pv. orzyae. Finally, using immunogold labeling with peptide-specific antibodies to each of the five rice PLDs, we demonstrate the cellular locations for each rice PLD.
Results
Gene identification and sequence comparisons
Rice genes encoding sequences homologous to PLD were isolated by screening both rice cDNA and genomic libraries. For cDNA library screening, Arabidopsis cDNAs encoding PLDα, PLDβ, and PLDγ were used as simultaneous probes. All 189 positive plaques obtained after three rounds of screening contained either whole or partial sequence inserts of the previously isolated rice PLD1 (accession number AB001920, Ueki et al. 1995).
Subsequent screening of the rice genomic library using the full-length PLD1 cDNA as probe identified five complete PLD genes. Two of these correspond to previously reported PLD genes, RPLD1 and RPLD2 (Ueki et al. 1995, Morioka et al. 1997, accession numbers AB001920 and AB001919, respectively). The remaining three, identified here, are unique and are referred to as RicePLD3, Rice PLD4, and Rice PLD5 (RPLD3-5). Their accession numbers are AF271356, AF271357, and AF271358, respectively.
Fig. 1a shows a schematic representation of the predicted gene structure of RPLD3-5 as well as those of the previously isolated rice PLD genes and four PLDs from Arabidopsis. Gene structure was confirmed by sequencing of cDNAs (partial or complete) and by comparison to Arabidopsis sequences. RPLD3 has a predicted coding region separated by two introns (positions 1,728–1,825 and 3,768–5,736). The predicted protein has 832 amino acids with a molecular mass of 93 kDa and a pI of 5.96. RPLD4 has a predicted coding region separated by two introns (positions 1,674–1,781 and 3,757–3,845). The predicted protein has 842 amino acids with a molecular mass of 93 kDa and a pI of 5.92. RPLD5 has a predicted coding region containing nine introns (2,316–2,490, 2,608–2,984, 3,182–3,602, 3,864–3,947, 4,017–4,112, 4,265–4,566, 4,716–4,788, 4,960–5,075, and 5,208–5,478). The predicted protein has 845 amino acids with a molecular weight of 95 kDa and a pI of 7.81.
Chromosomal locations of the five RPLDs were determined by alignment with the currently available rice BAC genome sequence and the map positions of the corresponding BACs were obtained from the TIGR Rice Genome Database (http://www.tigr.org). RPLD1 was detected on chromosome 1 at 24.0 cM and is flanked by two markers, RG472 and RG246. RPLD5 was detected on chromosome 3 at 158.2 cM, distal to the marker RG418X. RPLD2, RPLD3, and RPLD4 are tandemly arrayed at 87.5 cM on chromosome 5, with RPLD3 followed by RPLD2 and RPLD4 in the opposite orientation, and are flanked by the markers RG648 (84.3 cM), RG424 (84.7 cM) and RG172 (102.9 cM).
Domains and motifs
The amino acid sequence alignment of the five rice PLD genes (Fig. 2) shows structural organization and domain features common to plant PLDs. Each rice PLD gene contains two defining fully conserved HXKXXXXD motifs, which are thought to converge to form the active catalytic site for phosphoester bond hydrolysis (Ponting and Kerr 1996, Sung et al. 1997). Each rice PLD also contains a single Ca2+/phosholipid-binding (C2) domain consisting of approximately 125 amino acids near the N-terminus (Fig. 2). In many enzymes, these C2 domains mediate Ca2+ dependent binding of proteins to phospholipids. In addition, a phosphatidylinositol 4,5-bisphosphate (PIP2)-binding motif is found in the rice sequences approximately 50 amino acid residues downstream from the first HXKXXXXD motif (Fig. 2). This motif of 45 amino acids in the Arabidopsis PLDβ was shown recently to bind PIP2, and two Lys residues are critical for the binding (Zheng et al. 2002). The two Lys are conserved in RPLD5 whereas the other rice PLDs, like Arabidopsis PLDα, do not contain the second Lys. Also, RPLD5 contains a putative myristoylation site (MGKSS) at its amino terminus (Towler et al. 1987).
Two previously unidentified motifs that are common to plant PLDs were identified. One motif is directly downstream of the C2 domain and consists of approximately 40 amino acids (Fig. 2). This region shares homology to the B-domain of insect (35% amino acid identity, 51% similarity; RPLD1 vs. Drosophila melanogaster α-amylase accession number AAF57894) and cereal α-amylases (21% amino acid identity, 46% similarity, RPLD1 vs. Oryza sativa α-amylase, accession number P17654). The B-domain, although variable among the different α-amylases, is often involved in enzyme regulation and activity (Janecek et al. 1997, MacGregor 1988).
A second motif, found in all rice PLDs and revealed by the “PRINTS” algorithm/data-base (Attwood et al. 1999), shares homology to a fingerprint of the rhodopsin-like G-protein coupled receptor (GPCR) superfamily (Attwood and Findlay 1994, Birnbaumer and Birnbaumer 1996) (Fig. 2). This fingerprint corresponds to the DRY motif found in greater than 200 GPCRs and is the site of association with heterotrimeric G-proteins (Birnbaumer and Birnbaumer 1996, Gether and Kobilka 1998, Alewijnse et al. 2000). The DRY motif contains an R residue believed to be essential for G-protein binding. Four of the five rice PLDs contain an R residue at or near the equivalent site found in the DRY motif of GCPR.
RPLD transcript levels
RNA blot analysis was used to monitor the steady-state levels of RPLD1 and RPLD2 transcripts in rice leaves at various times following infiltration with either virulent or avirulent strains of the bacterial blight pathogen, Xanthomonas oryzae pv. oryzae or with water (wounding). By comparison, RPLD1 transcription levels were more prevalent relative to RPLD2 based upon ease of RNA blot detection and because only RPLD1 cDNAs were obtained during extensive library screenings. RPLD1 transcription was wound inducible while RPLD2 transcript levels increased after both wounding and inoculation with the bacterial pathogen (Fig. 3). RPLD1 transcripts, present in healthy leaves, increased within 30 min following infiltration in all treatments. This early increase in expression was transient and transcripts returned to levels equivalent to untreated plants within 3 h after treatment. Through four repeated experiments, RPLD2 transcripts were also wound inducible but were present in lower levels than RPLD1 transcripts and were barely detectable in untreated tissues. Transcript levels of RPLD2 steadily increased through 48 h (time points 36 h and 48 h not shown in Fig. 3) following infiltration in all treatments. This increase in RPLD2 transcript was most pronounced in rice plants where X. o. pv. oryzae was present and was apparent after 3 h of pathogen treatment in both resistant and susceptible interactions. By comparison, resistant and susceptible interactions showed similar RPLD2 transcription profiles at each time point (Fig. 3). However, water-infiltrated plants exhibited a smaller and slower induction of RPLD2 transcription compared to pathogen-inoculated plants.
In addition to being detected in leaves, RPLD1 but not RPLD2 transcripts were also detected by RNA blot analysis at similar levels in untreated root and meristem tissues (data not shown). Transcripts of RPLD3, RPLD4, and RPLD5 were not detected by RNA blot analysis of leaf, root and meristem tissues. We obtained partial cDNAs for RPLD3 and RPLD4 by reverse-transcription-PCR methodologies. The sequences of these cDNAs, although truncated, were identical to genomic clones and therefore indicate that the respective genes are transcribed. Although we were not successful in amplifying RPLD5 by Rt-PCR, BLAST analysis of the National Center for Biological Information dbest data base revealed two RPLD5 ESTs from clones isolated from a rice cDNA library constructed from immature seeds 5 d after pollination (GenBank accession numbers AA751500 and AA750856). This indicates that RPLD5 is also transcriptionally functional.
RPLD localization
Peptide-specific antibodies were generated using sequence information deduced from the extreme carboxy terminus of each RPLD (Fig. 2). These antibodies, in addition to a polyclonal anti-castor bean PLD antibody (Wang et al. 1993), were used to detect rice PLDs in centrifugally fractionated and concentrated rice leaf proteins and to determine the RPLDs subcellular location. Each peptide-specific antibody detected proteins of the expected size range for the various PLD proteins. ELISA assays confirmed that the various antibodies were distinct in their ability to detect the respective peptides from which they were induced and that there was little or no cross reactivity among the antibodies (data not shown). In Western blots, only proteins of the expected size for RPLD were detected, suggesting the antibodies did not cross react with other proteins (data not shown). Antibodies against RPLD1, RPLD2, and castor bean PLD detect two closely linked bands (91 and 92 kd) in the cytosolic (soluble) protein fraction of rice leaf tissues, but not in the microsomal fraction (Fig. 4). RPLD3 antibodies detect one band (96 kd) in both the cytosolic and microsomal fractions. RPLD4 and RPLD5 antibodies each detect one broad band at 93 kd in the cytosolic but not microsomal fraction (Fig. 4).
Peptide-purified peptide-specific antibodies were used in an IgG-gold labeling method to identify intracellular locations of the various RPLD proteins in healthy rice leaf tissues. Each highly purified antibody exhibited a distinct and consistent labeling pattern with minimal background throughout multiple studies (Table 1, Fig. 5). RPLD1 was primarily located on the cell wall, plasma membrane, and chloroplast of leaf mesophyll tissues (Fig. 5D). Additionally, the secondary cell walls of xylem tissues and guard cells were labeled by RPLD1 antibodies (Fig. 5E, F). RPLD2 exhibited sparse labeling relative to most other RPLD antibodies, with little or no labeling present in healthy mesophyll cells (Fig. 5G). However, RPLD2 consistently labeled secondary cell walls of xylem tissues and guard cells (Fig. 5H, I). The locations of RPLD3 and RPLD4 were similar to one another, with the majority of their label being present in the chloroplasts (Fig. 5J, M). Only minor RPLD3 and RPLD4 label (approximate to background) was detected in mesophyll cell walls and plasma membranes and in secondary cell walls of xylem and guard cells (Fig. 5J–O). RPLD5 labeling also was sparse in healthy mesophyll cells and in secondary xylem cell walls (Fig. 5P, Q). However, RPLD5 label was concentrated in the secondary cell walls of guard cells (Fig. 5R).
To obtain a more general view of PLD localization in rice, a polyclonal anti-castor bean PLD antibody (Wang et al. 1993) also was used in immunolocalization studies of leaf tissues. Labeling by this presumably broad-spectrum anti-PLD antibody is consistent with that observed by the peptide-specific antibodies of rice PLDs. That is, like the rice peptide-specific antibodies, the castor bean antibody showed considerable localization to the cell wall, plasma membrane, and chloroplast of mesophyll cells (Table 1, Fig. 5A). Additionally, the castor bean PLD antibody exhibited a strong and distinct labeling of secondary cell walls of xylem tissues and guard cells (Fig. 5B, C). Interestingly, the castor bean PLD antibody also labeled the mesophyll nucleus, with an interesting clustering of the nucleolus (Table 1, Fig. 5A).
Discussion
Similarity to PLD subfamilies
Arabidopsis PLDs, the best characterized of all plant PLDs, are grouped into subfamilies (PLDα, β, γ, δ, and ζ) based upon enzymatic and biochemical differences. This grouping, which is strongly correlated with similarities in sequence identity, gene structure and domain conservation (Qin and Wang 2002), serves as a reference in early classification of newly discovered plant PLDs. RPLD1 is most related to the Arabidopsis PLDα subfamily. RPLD1 shares 71% amino acid identity to PLDα, and like PLDα contains three introns with one present in the 5′ UTR. RPLD2, RPLD3 and RPLD4 are tandemly arrayed on chromosome 5, as reported by Elias et al. (2002). These authors presented an in silico analysis of database sequences and detected 16 complete or partial rice PLD genes, which included our database entries of RPLD3, RPLD4, and RPLD5. Phylogenetic analysis showed that RPLD2, RPLD3, and RPLD4 cluster as a unique subfamily that is most closely related to Arabidopsis PLDα (although RPLD1 is more closely related to PLDα), and that RPLD5 is most closely related to Arabidopsis PLDδ (Qin and Wang 2002, Elias et al. 2002).
Motifs and predicted activity
Alignments of rice PLD sequences reveal previously described motifs that are common to plant PLDs. Comparisons of these domain structures between rice and Arabidopsis PLDs are consistent with subfamily grouping and allow for predictions of catalytic activity. Each rice PLD contains a single putative C2 domain that is predicted to mediate the Ca2+ dependent phospholipid binding involved in associating enzymes with substrates and/or membranes (Essen et al. 1996, Perisic et al. 1998). In plants, Ca2+ is considered an important regulator of PLD activity, and it is proposed that the C2 domain plays a critical role in this regulation (Wang 1999, Pappan and Wang 1999, Liscovitch et al. 2000). Three-dimensional structures determined for C2 domains in various enzymes indicate that at least four conserved acidic amino acids are instrumental in Ca2+ binding (Ponting and Parker 1996). The importance of these conserved residues is supported by studies in Arabidopsis where expressed C2 domains from PLDα and PLDβ show a direct correlation with the preservation of the acidic amino acids and the domains’ ability to bind Ca2+ (Zheng et al. 2000). The Arabidopsis PLDα C2 domain, which contains only two of the amino acids, showed a lower affinity for Ca2+ than did the expressed PLDβ C2 domain which conserves all four. These Ca2+ requirements for the expressed PLDα and PLDβ C2 domains were also reflective of the expressed proteins, which show similar Ca2+ requirements for activity. The C2 domains of the rice PLDs, RPLD1-4, contain few, if any, of the required acidic amino acids, while RPLD5 conserves a majority (Fig. 2). Thus, we predict that RPLD1-4 would require relatively high Ca2+ levels for activity, such as the millimolar amounts required by PLDα, whereas RPLD5 would require micromolar Ca2+ levels similar to PLDβ.
Predictions of PIP2-binding and enzyme activity might also be inferred by domain structure conservation. Of the basic amino acids thought to be involved in PIP2 activation in the first conserved PIP2-binding site (Zheng et al. 2002), Arabidopsis PLDα contains three of four, as do RPLD1, RPLD2, RPLD3, and RPLD4 (Fig. 2). RPLD5, similar to PLDδ, has conserved all the basic residues and thus might also act in a PIP2-dependent manner.
Sequence alignments show several conserved regions among plant PLDs that are not ascribed to known PLD functions and/or activities, but that may be of importance. Two such regions share identity with sequences found in other enzymes. The first, located just downstream of the C2 domain, shares identity with the B-domain of α-amylases in insect and cereals (Fig. 2; Boer and Hickey 1986, Knox et al. 1987, Huang et al. 1990). In cereals, B-domains are essential for enzyme regulation and activity. For example, B-domains of barley α-amylases (AMY1 and AMY2) are instrumental in binding structural Ca2+, conferring stability at low pH, specifying substrate binding, and in binding of the barley α-amylase subtilisin inhibitor, BASI (Rodenburg et al. 1994, Juge et al. 1995). One interesting possibility is that this region is the site for binding β-trefoil-type inhibitors (e.g. BASI in barley and RASI in rice; Yamagata et al. 1998, Murzin et al. 1992) that suppress α-amylase activity in dormant seeds. Alternatively, since the PLD suspected B-domain is in close proximity to the C2 domain (Fig. 2), it is possible that this region may act in a cooperative manner to confer stability to the PLD enzyme in response to Ca2+ binding or to transmit a conformational signal brought about by Ca2+ binding by the C2 domain. This possibility is made even more plausible by recent studies in Arabidopsis where C2 domains were shown to undergo unprecedented conformational changes upon Ca2+ binding (Zheng et al. 2000). Whether any of these traits also are conferred by similar PLD sequences is unknown.
Another conserved but unknown region in plant PLDs shares homology to the DRY motif (Fig. 2) found in rhodopsin-like GPCR (Birnbaumer and Birnbaumer 1996). This motif, located at the cytosolic juncture of GPCR’s third transmembrane domain, is important for coupling to heterotrimeric G-proteins (Alewijnse et al. 2000, Gether and Kobilka 1998). Of major importance is the conservation of a D/e-R-Y/f/h/c sequence that is thought to undergo a conformational change during agonist binding, resulting in G-protein/receptor association (Ballesteros et al. 1998). Four of the rice PLDs (Fig. 2) and the majority of reported plant PLDs (alignments not shown) exhibit this motif as E-R-Y. Although the majority of GPCRs exhibit the residues D-R-Y, conservative substitutions of the D and Y amino acids are allowable (Okana et al. 1992, Hisatomi et al. 1999), but the preservation of the R residue is critical (Alewijnse et al. 2000, Rasmussen et al. 1999). Also of importance, is the conserved hydrophobic region that is located just downstream of D/e-R-Y/f/h/c that is believed to associate with the cell membrane interface (Birnbaumer and Birnbaumer 1996). This region, typified by the sequence V-Y-V-V-V, is present in exact or conservatively modified form in all rice PLDs (Fig. 2) and in most reported plant PLDs. While the presence of the DRY motif in plant PLDs suggests that G-proteins may be involved in regulating PLD activity (Munnik et al. 1995, Park et al. 1996, Chapman et al. 1998, Frank et al. 2000, Ritchie and Gilroy 2000, Lein and Saalbach 2001), conclusive evidence for this interaction and role remain to be determined.
Transcriptional differences
Based upon the ease of transcript detection in leaf, root, and meristem tissues and the ability to obtain only RPLD1 cDNAs from extensive library screenings, RPLD1 transcripts are much more abundant in vegetative rice tissues than those of other PLD family members. By both sequence homology and gene structure, RPLD1 is most closely related to the PLDα subfamily of plant PLDs. This relative high abundance of PLDα-type transcripts has been observed in castor bean (Ryu and Wang 1996, Xu et al. 1997), Arabidopsis (Wang et al. 2000) and in previous studies of rice (Ueki et al. 1995) and suggests that PLDα may play a maintenance or housekeeping role in the plant cell. The present study reveals that RPLD1 expression is increased by wounding in a rapid and transient manner, implying that RPLD1 acts in early stages of wound response. RPLD2 is also wound inducible. However, by comparison with RPLD1, RPLD2 transcripts are in lower amounts, are induced at a slower rate and are more prolonged. In addition, RPLD2 transcription is also influenced by pathogen presence, showing an earlier, more intense induction during both resistant and susceptible interactions than in response to wounding. This difference in expression suggests distinct roles for RPLD1 and RPLD2, with RPLD2 acting in more prolonged wound and pathogen responses. Transcripts of RPLD3, RPLD4, and RPLD5 were rare and not detected by conventional RNA blot analysis of healthy 12-day-old leaf, root, or meristem tissues or in leaf tissues following wounding or pathogen infiltration (data not shown). That these transcripts are expressed in other plant tissues, during specific developmental stages, or by induction with other pathogens and/or stresses remains a possibility. However, the ability to obtain partial cDNA sequences for RPLD3 and RPLD4, the existence of RPLD5 cDNA in DNA databases, the fact that the deduced coding regions from genomic sequence appear functional, and the ability to detect protein with peptide-specific antibodies to each of these is strong evidence that these PLD family members are transcribed. Although the functions of the individual PLD genes remain to be determined in plants, the abundance of RPLD1 transcripts, the differences in wound and pathogen induction of RPLD1 and RPLD2, and the presence of other PLD family members suggest a combination of specified function and tight regulation of PLD activity. Further studies of PLD expression in a variety of plant tissues, during different developmental stages and while undergoing a variety of biotic and abiotic stresses will aid in deducing the physiological roles of the various PLD family members.
Location of RPLDs
Centrifugal fractionation analysis of young rice leaf proteins indicated that all five rice PLDs are extractable in the soluble cytosolic fraction and that RPLD3 also is present in the microsomal portion. In previous studies, plant PLDs were found in various protein fractions, indicating their locations may be developmentally or physiologically dependent. For example, centrifugal fractionation of castor bean cells revealed that most PLD in young leaves was soluble, whereas in mature leaves a majority of PLD was associated with the microsomal fraction (Xu et al. 1996). PLDγ is present mostly in cytosolic fractions (Fan et al. 1999). The detection of the majority of the individual rice PLDs in the soluble cytosolic fraction is consistent with the findings from castor bean.
In their active state, PLDs are thought to be associated with lipid membranes (Pappan and Wang 1999, Liscovitch et al. 2000). Enzymes such as PLD may be regulated by translocation/membrane targeting through mechanisms such as intracellular Ca2+ signaling via C2 domain regulation (Evans et al. 2001, Cho 2001). In this regard, studies of Streptomyces reported that PLD binds to membrane vesicles in a Ca2+-dependent manner and Ryu and Wang (1996) demonstrated that the amount of PLD associated with castor bean microsomal membranes is increased with elevated physiological concentrations of free Ca2+. This relationship between PLD location, regulation, and translocation remains to be determined.
The distribution patterns of all five rice PLDs are distinct and provide valuable insight into the roles of specific rice PLDs. That plant PLDs are associated with the plasma membrane was previously reported (Xu et al. 1996, Ritchie and Gilroy 1998) and is demonstrated here for RPLD1. The RPLD1 location is consistent with a proposed PLD role in transmembrane signaling (Wang 2002). Location of RPLD1, RPLD2, and RPLD5 in the walls of guard cells is consistent with another PLD proposed role, that is, stomatal closure (Jacob et al. 1999). In this regard, anti-sense PLDα Arabidopsis plants show greatly decreased drought tolerance and histochemical studies of the same plants show a greatly decreased presence of the PLD protein in guard cells as compared to wild-type plants (Sang et al. 2001).
RPLD1, RPLD3, and RPLD4 are located in the chloroplast lumen; as RPLD3 and RPLD4 were not detected in the microsomal fraction, they are likely to be in the soluble fraction of the chloroplasts. Since the chloroplast is a major manufacturing site for plant lipids/membranes, this localization supports a role for PLDs in chloroplast lipid metabolism. Another unique location of rice PLDs is that of the primary mesophyll cell wall and secondary cell wall of xylem vessels. This suggests that PLD plays a role in extracellular signaling. How these PLDs become associated with cell walls remains unknown, as no extracellular signal peptide has been identified for them. However, in the two cases where plant proteins have been extracted from plant tissues and used for N-terminal sequencing, both proteins have been truncated at very similar positions (castor bean PLD, amino acid 31, Wang et al. 1994; rice PLD, amino acid 46, Ueki et al. 1995) and might indicate the presence and processing of an uncharacterized signal sequence. In this regard it is interesting to note that two bands of similar size were detected in denaturing PAGE for the RPLD1, RPLD2, and castor bean antibodies and therefore may indicate the processing of the PLD protein.
The cellular locations of PLDs determined by the polyclonal castor bean PLD antibody were consistent with the combined locations found for the individual peptide-specific RPLD antibodies. This consistency in labeling among the various antibodies supports the intercellular localization of the specific rice PLDs of this study and also lends itself to a general view of PLD distribution throughout the plant cell.
The recent cloning of PLD genes has allowed for a variety of genetic, biochemical, immunological, and physiological studies that have advanced our understanding of PLD function. Plant PLDs are now known to exist as part of a multi-gene family, suggesting role specificity. Variations in transcription, expression, and biochemical activity show that these enzymes are regulated at many levels and suggests a tight control of their activity. Still, much remains to be learned about the role of PLD in plants. Why are there multiple distinct PLDs? What are the tissue, temporal and spatial distributions of the isoforms? How are these enzymes regulated and how does this regulation affect cell activity and overall plant response? As a prelude to answering these and other questions, we have presented here the identification of three novel PLD genes in rice. By comparing their sequence, structure, and motif similarities to previously characterized PLDs, we demonstrated commonalities among proposed subfamilies. In addition, we identified two new protein motifs that have homology to regulatory domains of other enzymes. The variation in expression of RPLD genes, particularly the induction by wounding and pathogen infection, as well as the distinct but overlapping locations of the RPLD proteins further suggests specific and distinct roles for the various plant PLDs. Immunological and biochemical studies concerning the roles of the various rice PLD family members in stress and developmental responses are currently in progress.
Materials and Methods
Plant materials, bacterial strains, and media
Rice (Oryza sativa) cultivar IRBB10, that contains the Xa10 gene for bacterial blight resistance, was used as the nucleic acid source for RNA and DNA blot analysis and as the source of protein for fractionation and protein blot analyses. Seedlings were grown in growth chambers as described previously (Reimers and Leach 1991). X. o. pv. oryzae strains [PX099A(pBUavrXa10); resistant interaction with IRBB10; R] and [PX099A(pBU); susceptible interaction with IRBB10; S] were maintained at 28°C on WFP medium (Karganilla et al. 1973). For plant inoculations, bacterial inoculum (5×109 colony forming units ml–1) was prepared and infiltrated into multiple sites of the second fully expanded leaf of 12-day-old rice plants as described (Reimers and Leach 1991).
Library screenings and DNA sequence analysis
Rice cDNAs encoding PLD were isolated from a pathogen-induced cDNA library (Hilaire et al. 2001) constructed from total RNA obtained from 10-day-old IRBB10 rice leaves 12 h after infiltration with X. o. pv. oryzae strain, PX099A(pBUavrXa10) (a resistant interaction). Arabidopsis cDNAs encoding PLDα, PLDβ and PLDγ (Qin et al. 1997; accession numbers U36381, U84568 and AF027408, respectively) were used as simultaneous probes. Plaques hybridizing to the Arabidopsis PLD probes were submitted to three rounds of screening. Phagemids were excised from the phage following manufacturer’s instructions (Stratagene, La Jolla, CA, U.S.A.). Genomic PLD clones were isolated from a XhoI lambda GEM-11 genomic library (Promega, Madison, WI, U.S.A.) constructed from 14-day-old IR54 rice plants described in McGee et al. (2001). A full-length PLD1 cDNA was used as probe. Hybridizing plaques were submitted to three rounds of screening. Phage DNA was isolated by the plaque lysate method (Sambrook et al. 1989) and was analyzed by restriction digest and DNA blot analysis. Fragments of the genomic insert hybridizing to the PLD1 cDNA probe were subcloned into pBluescript KS+ (Stratagene). Both strands of insert DNA were sequenced by primer walking at the Kansas State University/USDA Sequencing Facility (Manhattan, KS, U.S.A.). Sequence homology was compared using BLAST search analysis and the databases at the National Center for Biological Information.
Amino acid sequence analysis
Preliminary amino acid sequence analysis was performed using Baylor College of Medicine’s (Houston, TX, U.S.A.) BCM Search Launcher integrated interface to molecular data base search and analysis (Smith et al. 1996). Motif searches of conserved PLD sequences were analyzed utilizing PROSITE (Hofmann et al. 1999) and PRINTS (Attwood et al. 1999) databases. Alignments of suspected PLD regions with the B-domains of α-amylases or G-protein coupled receptors were constructed using CLUSTALW software (Thompson et al. 1994). The GPCR fingerprint was generated by the PRINTS algorithm/database of protein motifs (Attwood et al. 1999).
DNA blot analysis
Agarose gel electrophoresis, DNA restriction digests, and gel blot transfers were performed using standard procedures (Sambrook et al. 1989). Genomic DNA was isolated from 10-day-old IRBB10 seedlings following the method of McCouch et al. (1988). Digested DNAs (15 µg lane–1) were separated by electrophoresis in 1% agarose gels and blotted onto Hybond N+ membrane (Amersham Pharmacia Biotech, Piscataway, NJ, U.S.A.). Radioactive probe of the full-length RPLD1 cDNA was constructed using the Strip-EZ DNA kit (Ambion, Austin, TX, U.S.A.). Hybridizations were performed at 65°C in 5× SSPE, 5× Denhardt’s solution and 0.5% SDS. Membranes were washed twice (20 min each) in 2× SSPE and 0.1% SDS at 68°C, once (10 min) in 0.2% SSPE and 0.1% SDS, and twice (10 min each) in 0.1% SSPE and 0.1% SDS.
RNA blot analysis
Total RNA was extracted from leaves of 12-day-old IRBB10 rice plants that were untreated (U), infiltrated with water (W), or infiltrated with X. o. pv. oryzae strains eliciting either a susceptible (S) or resistant (R) response using the RNeasy Plant Mini Kit (Qiagen, Valencia, CA, U.S.A.). RNA samples (10 µg lane–1) were separated by electrophoresis under denaturing conditions and transferred to Hybond-N+ membranes (Amersham). Blots prehybridized in Ultrahyb solution (Ambion) for 4 h at 42°C and hybridized in the same solution at 47°C overnight. All hybridization experiments were performed at least four times to assure reproducibility. Gene-specific-probes were made from the putative 3′ untranslated regions of each rice PLD (RPLD1-5) using cloned genomic fragments as template for PCR amplification. Primers used in these constructs were as follows: RPLD1 forward, 5′-CCATCCTCACCTCATAGACGA-3′; RPLD1 reverse, 5′-CCAACTTCCCATCTTACAAACG-3′; RPLD2 forward, 5′-TGCCTCCCTTTCTCACCACA-3′; RPLD2 reverse, 5′-ATTCAGGTCAAATCTCCAAC-3′; RPLD3 forward, 5′-GTTTGTGCTGGTGGTGCAT-3′; RPLD3 reverse, 5′-CGTTTCAGAAGGATTGGGAA-3′; RPLD4 forward, 5′-GTGTTGCTCTGAGTGTGTGA-3′; RPLD4 reverse, 5′-CAGCATTGGTTTCTTAGGTA-3′; RPLD5 forward, 5′-CCCAATGCCCTCACCACCTA-3′; RPLD5 reverse, 5′-CGCACATCACACTCATACACTC-3′. PCR products were subcloned into the pCR II-TOPO cloning vector (Invitrogen, Carlsbad, CA, U.S.A.) and the cloned products sequenced to assure their identity. DNA probes were constructed using the Strip-EZ DNA labeling kit (Ambion). To assess equal loading, each RNA blot was stripped and re-probed using the labeled 1212 bp 18S rRNA DECAprobe (Ambion).
Reverse transcription-PCR
Poly (A) RNA was obtained with the Micropoly (A) Pure Kit (Ambion) using an equal-part mixture of total RNA (350 µg) taken from 12-day-old rice plants/tissues after various treatments (uninoculated leaf, 50 µg; uninoculated root, 50 µg; uninoculated shoot meristem, 50 µg; susceptible leaf inoculated 6 h; 50 µg; resistant leaf inoculated 6 h; 50 µg; susceptible leaf inoculated 24 h, 50 µg; resistant leaf inoculated 24 h, 50 µg). First and second-strand cDNA synthesis and adaptor ligations were performed using the Marathon cDNA Amplification Kit (Clontech; Palo Alto, CA, U.S.A.). 5′ RACE PCR was performed using the Advantage 2 Polymerase Mix (Clontech) and the following gene-specific primers: RPLD3, 5′-CACATTCTGATGGAAGCAG-3′; RPLD4, 5′-CAGGAATGGCAACTCGATC-3′. Initial PCR reactions included a denaturing step at 94°C for 30 s followed by 30 cycles of 94°C for 5 s, 59.3°C for 30 s, and 68°C for 4 min. The last cycle consisted of an extension reaction at 68°C for 10 min. Secondary nested PCR reactions were performed for RPLD3 and RPLD4 using 5 µl of a 1 : 50 dilution of the initial reaction and the following gene-specific primers: RPLD3, 5′-CCACCAGCACAAACTTTCACCTG-3′; RPLD4, 5′-GAGCAACACTCTACGACGTGAGG-3′. Nested PCR reactions included a denaturing step at 94°C for 30 s followed by 22 cycles of 94°C for 30 s, 64.1°C for 30 s, 68°C for 4 min with a final extension cycle of 68°C for 10 min. PCR products were separated on TAE agarose gels. Fragments were purified using Qiaex II DNA Purification System (Qiagen) and subcloned into the pCRII-TOPO vector (Invitrogen; Carlsbad, CA, U.S.A.). Inserted DNAs were sequenced to confirm identity.
Peptide synthesis and antibody production
Unique amino acid sequences from the carboxy terminus of the five rice PLD genes (RPLD1:GAKSDYMPPIL; RPLD2:QAPVIGTKGNLPPFL; RPLD3:GKKAINPLMTPDI; RPLD4:GTLASSAYMIPYL; RPLD5:GTQTSLPNALTT) were synthesized by a manual solid-phase method as described (Henry et al. 1992) at the Biotechnology Core Facility of Biochemistry, Kansas State University (Manhattan). Peptides were removed from the support with anhydrous hydrogen sulfide and approximately 6 mg of each peptide was conjugated to keyhole limpet hemocyanin. Concentrations of the peptides were estimated at A214.
Polyclonal PLD antibodies were raised in rabbits against purified PLD from 2-d post-germination endosperm of castor bean (Wang et al. 1993). Peptide-specific antibodies for the individual rice PLDs were prepared by Cocalico Biologicals (Reamstown, PA, U.S.A.) using synthetic peptide-hemocyanin conjugates emulsified with Freund’s complete adjuvant (1 : 1) and injected into female white New Zealand rabbits. Boosters were administered at 14, 21, and 41 d. For electron microscopy, antibodies specific to individual rice PLDs were purified by eluting antibodies from nitrocellulose blots bound with synthetic peptides corresponding to the individual PLDs (Smith and Fisher 1984).
Protein isolation and immunodetection
For fractionation of cytosolic and microsomal proteins, 14-d IRBB10 rice leaves were ground in liquid nitrogen with a mortar and pestle (Young et al. 1996). Proteins were extracted in one equal volume of buffer A (50 mM Tris-HCL, pH 8.0, 1 mM EDTA, 10 mM KCl, 2 mM DTT, 0.5 mM phenylmethylsufonyl fluoride, and 0.5 M sucrose (Wang et al. 1993). The homogenate was filtered through nine layers of cheesecloth and centrifuged for 6,000×g for 20 min. The supernatant was centrifuged at 110,000×g for 1 h. Pelleted proteins were dissolved in buffer B (same as buffer A without sucrose). Protein concentrations were determined colormetrically (Bio-Rad, WI, U.S.A.). Cytosolic or microsomal proteins (100 µg lane–1) were separated on 10% SDS-polyacrylamide gels. Proteins were transferred to polyvinylidone difluoride membranes (Immobilon-PSQ; Millipore) by electrophoresis using Towbin buffer and a final methanol concentration of 5%. The membranes were then probed with either polyclonal anti-PLD antibodies raised to purified PLD from castor bean (Wang et al. 1993) or with peptide-specific antibodies to individual rice PLDs. Antigen–antibody complexes were visualized using the Immun-Star Chemiluminescent Protein Detection System (Bio-Rad).
Electron microscopy and immunochemistry
Rice leaf tissue samples were fixed in 1% gluteraldehyde, 2% paraformaldehyde, 75 mM PIPES (pH 7.3), and 5 mM MgCl2 at 4°C overnight. Samples were washed three times in 75 mM PIPES (pH 7.3) and 5 mM MgCl2 at 15 min intervals. Fixed leaf tissues were dehydrated through graded ethanol series and infiltrated in several changes of LR White resin as described by the manufacturer (Polysciences, Inc., PA, U.S.A.). Tissues were embedded in BEEM capsules (Taab Laboatories, Berks, U.K.) by polymerization at 60°C. Ultra-thin sections on parlodion-covered grids were used for immunogold labelling. Grids containing sections were treated for 10 min on metaperiodate (0.4 g/2.5 ml), followed by 0.1 M HCL for 5 min, rinsed in water, and blocked in EMG buffer (3% BSA) for two periods of 10 min each. Sections were treated with primary antibodies-EMG solution at a dilution of 1 : 8 for peptide-purified peptide-specific anti-rice PLD antibodies and at 1 : 100 for castor bean polyclonal antisera for 3 h. Treated grids were rinsed seven times at 2 min intervals and incubated with 10 nM gold conjugated with anti-rabbit IgG (1 : 50 in EMG) for 2 h. Grids were then rinsed and stained by uranyl acetate (1% for 10 min, dark) and Reynolds lead citrate (1 M NaOH for 3–5 min). Labeled grids were rinsed with 3–4 drops of 0.1 M NaOH, rinsed with water, air dried, and stored desiccated. Sections were analyzed using a Hitachi H-300 transmission electron microscope (Hitachi Instruments, Inc.).
Acknowledgments
This work was supported by a USDA-NRI grant (#97-3503-4550) and NASA-NAG (#100–0142) to Leach, Guikema and Wang, and the Kansas State University Agricultural Research Station (01-95-J). We thank J. Zhou and R. Welti for critically reviewing the manuscript.
Corresponding author: E-mail, jeleach@ksu.edu; Fax, +1-785-532-5692.
Mesophyll tissue (gold particle/unit measure) b | Number of gold particles associated with purified peptide-specific antibody to PLD a | |||||
CBPLD | RPLD1 | RPLD2 | RPLD3 | RPLD4 | RPLD5 | |
Cell wall (gp cm–1) | 3.0±0.4 | 0.9±0.1 | 0.2±0 | 0.1±0 | 0.2±0 | 0.2±0 |
Membrane (gp cm–1) | 1.5±0.2 | 0.4±0.1 | 0.1±0 | 0 | 0.1±0 | 0 |
Chloroplast (gp cm–2) | 10.3±0.9 | 0.8±0.1 | 0.2±0.1 | 1.8±0.2 | 1.3±0.1 | 0.3±0.1 |
Vacuole (gp cm–2) | 0.4±0.1 | 0.3±0.1 | 0.2±0.1 | 0.2±0.1 | 0.7±0.1 | 0.2±0.1 |
Cytoplasm (gp cm–2) | 0.6±0.2 | 0.3±0.1 | 0.2±0.1 | 0.5±0.1 | 0.3±0 | 0.2±0.1 |
Nucleus (gp cm–2) | 1.6±0.1 | 0.3±0.1 | 0.2±0.1 | 0.3±0.1 | 0.5±0.1 | 0.1±0.1 |
Ex cell (gp cm–2) | 0.2±0.1 | 0.2±0.1 | 0.2±0.1 | 0.2±0.1 | 0.5±0.1 | 0.3±0.1 |
Mesophyll tissue (gold particle/unit measure) b | Number of gold particles associated with purified peptide-specific antibody to PLD a | |||||
CBPLD | RPLD1 | RPLD2 | RPLD3 | RPLD4 | RPLD5 | |
Cell wall (gp cm–1) | 3.0±0.4 | 0.9±0.1 | 0.2±0 | 0.1±0 | 0.2±0 | 0.2±0 |
Membrane (gp cm–1) | 1.5±0.2 | 0.4±0.1 | 0.1±0 | 0 | 0.1±0 | 0 |
Chloroplast (gp cm–2) | 10.3±0.9 | 0.8±0.1 | 0.2±0.1 | 1.8±0.2 | 1.3±0.1 | 0.3±0.1 |
Vacuole (gp cm–2) | 0.4±0.1 | 0.3±0.1 | 0.2±0.1 | 0.2±0.1 | 0.7±0.1 | 0.2±0.1 |
Cytoplasm (gp cm–2) | 0.6±0.2 | 0.3±0.1 | 0.2±0.1 | 0.5±0.1 | 0.3±0 | 0.2±0.1 |
Nucleus (gp cm–2) | 1.6±0.1 | 0.3±0.1 | 0.2±0.1 | 0.3±0.1 | 0.5±0.1 | 0.1±0.1 |
Ex cell (gp cm–2) | 0.2±0.1 | 0.2±0.1 | 0.2±0.1 | 0.2±0.1 | 0.5±0.1 | 0.3±0.1 |
a A polyclonal antibody to castor bean PLD (CBPLD) and peptide-specific antibodies to RPLD1, RPLD2, RPLD3, RPLD4, and RPLD5 were used to label thin sections of rice leaf tissues. The bound antibodies were detected by treatment with anti-rabbit IgG conjugated to 10 nM gold.
b The mean numbers of gold particles on the wall and membrane of mesophyll cells were determined per cm of tissue, whereas the numbers of particles in the cytoplasm, chloroplasts, vacuoles, or extracellular spaces (Ex Cell) were determined per cm2.
Mesophyll tissue (gold particle/unit measure) b | Number of gold particles associated with purified peptide-specific antibody to PLD a | |||||
CBPLD | RPLD1 | RPLD2 | RPLD3 | RPLD4 | RPLD5 | |
Cell wall (gp cm–1) | 3.0±0.4 | 0.9±0.1 | 0.2±0 | 0.1±0 | 0.2±0 | 0.2±0 |
Membrane (gp cm–1) | 1.5±0.2 | 0.4±0.1 | 0.1±0 | 0 | 0.1±0 | 0 |
Chloroplast (gp cm–2) | 10.3±0.9 | 0.8±0.1 | 0.2±0.1 | 1.8±0.2 | 1.3±0.1 | 0.3±0.1 |
Vacuole (gp cm–2) | 0.4±0.1 | 0.3±0.1 | 0.2±0.1 | 0.2±0.1 | 0.7±0.1 | 0.2±0.1 |
Cytoplasm (gp cm–2) | 0.6±0.2 | 0.3±0.1 | 0.2±0.1 | 0.5±0.1 | 0.3±0 | 0.2±0.1 |
Nucleus (gp cm–2) | 1.6±0.1 | 0.3±0.1 | 0.2±0.1 | 0.3±0.1 | 0.5±0.1 | 0.1±0.1 |
Ex cell (gp cm–2) | 0.2±0.1 | 0.2±0.1 | 0.2±0.1 | 0.2±0.1 | 0.5±0.1 | 0.3±0.1 |
Mesophyll tissue (gold particle/unit measure) b | Number of gold particles associated with purified peptide-specific antibody to PLD a | |||||
CBPLD | RPLD1 | RPLD2 | RPLD3 | RPLD4 | RPLD5 | |
Cell wall (gp cm–1) | 3.0±0.4 | 0.9±0.1 | 0.2±0 | 0.1±0 | 0.2±0 | 0.2±0 |
Membrane (gp cm–1) | 1.5±0.2 | 0.4±0.1 | 0.1±0 | 0 | 0.1±0 | 0 |
Chloroplast (gp cm–2) | 10.3±0.9 | 0.8±0.1 | 0.2±0.1 | 1.8±0.2 | 1.3±0.1 | 0.3±0.1 |
Vacuole (gp cm–2) | 0.4±0.1 | 0.3±0.1 | 0.2±0.1 | 0.2±0.1 | 0.7±0.1 | 0.2±0.1 |
Cytoplasm (gp cm–2) | 0.6±0.2 | 0.3±0.1 | 0.2±0.1 | 0.5±0.1 | 0.3±0 | 0.2±0.1 |
Nucleus (gp cm–2) | 1.6±0.1 | 0.3±0.1 | 0.2±0.1 | 0.3±0.1 | 0.5±0.1 | 0.1±0.1 |
Ex cell (gp cm–2) | 0.2±0.1 | 0.2±0.1 | 0.2±0.1 | 0.2±0.1 | 0.5±0.1 | 0.3±0.1 |
a A polyclonal antibody to castor bean PLD (CBPLD) and peptide-specific antibodies to RPLD1, RPLD2, RPLD3, RPLD4, and RPLD5 were used to label thin sections of rice leaf tissues. The bound antibodies were detected by treatment with anti-rabbit IgG conjugated to 10 nM gold.
b The mean numbers of gold particles on the wall and membrane of mesophyll cells were determined per cm of tissue, whereas the numbers of particles in the cytoplasm, chloroplasts, vacuoles, or extracellular spaces (Ex Cell) were determined per cm2.
Abbreviations
- PA
phosphatidic acid
- PIP2
phosphatidylinositol 4,5-bisphosphate
- PLD
phospholipase D
- RPLD
rice phospholipase D.
The nucleotide sequences reported in this paper are in GenBank under accession numbers AF271356, AF271357, and AF271358.
References
Alewijnse, A.E., Timmerman, H., Jacobs, E.H., Smit, M.J., Roovers, E., Cotecchia, S. and Leurs, R. (
Attwood, T.K. and Findlay, J.B. (
Attwood, T.K., Flower, D.R., Lewis, A.P., Mabey, J.E., Morgan, S.R., Scordis, P., Selley, J.N. and Wright, W. (
Ballesteros, J., Kitanovic, S., Guarnieri, F., Davies, P., Fromme, B.J., Konvicka, K., Chi, L., Millar, R.P., Davidson, J.S., Weinstein, H. and Sealfon, S.C. (
Birnbaumer, M. and Birnbaumer, L. (
Boer, P.O. and Hickey, D.A. (
Cha, Y.R., Lee, K.-W., Moon, Y.-H., Kim, J.C., Han, T.J., Lee, W.S. and Cho, S.H. (
Chapman, K.D. (
Chapman, K.D., Tripathy, S., Venables, B.J. and Desouza, A.D. (
Dyer, J.H., Zheng, S. and Wang, X. (
Elias, M., Potocky, M., Cvckova, F. and Zarsky, V. (
Essen, L., Perisic, O., Cheung, R., Katan, M. and Williams, R.L. (
Evans, J.H., Spencer, D.M., Zweifach, A. and Leslie, C.C. (
Fan, L., Zheng, S., Cui, D. and Wang, X. (
Frank, W., Munnik, T., Kerkmann, K., Salamini, F. and Bartels, D. (
Gether, U. and Kobilka, B.K. (
Hanahan, D.J. and Chaikoff, I.L. (
Henry, R.L., Takemoto, L.J., Murphy, J., Gallegos, G.L. and Guikema, J.A. (
Hilaire, E., Young, S., Willard, L., McGee, J.D., Sweat, T., Chittoor, J., Guikema, J. and Leach, J.E. (
Hisatomi, O., Takahashi, Y., Taniguchi, Y., Tsukahara, Y. and Tokunaga, F. (
Hofmann, K., Bucher, P., Falquet, L. and Bairoch, A. (
Huang, N., Sutliff, T.D., Litts, J.C. and Rodriguez, R.L. (
Jacob, T., Ritchie, S., Assmann, S.M. and Gilroy, S. (
Janecek, S., Svensson, B. and Henrissat, B. (
Jones, D., Morgon, C. and Cockcroft, S. (
Juge, N., Rodenburg, K.W., Guo, X-J., Chaix, J.-C. and Svensson, B. (
Karganilla, A., Paris-Natural, M. and Ou, S.H. (
Kim, D.U., Roh, T.Y., Lee, J., Noh, J.Y., Jang, Y.J., Hoe, K.L., Yoo, H.S. and Choi, M.U. (
Knox, C.A.P., Sonthayanon, B., Chandra, G.R. and Muthukrishnan, S. (
Lein, W. and Saalbach, G. (
Liscovitch, M., Czarny, M., Fiucci, G. and Tang, X. (
McCouch, S.R., Kochert, G., Yu, Z.H., Wang, Z.Y., Khush, G.S., Coffman, W.R. and Tanksley, S.D. (
McGee, J.D., Hamer, J.E. and Hodges, T.K. (
Morioka, S., Ueki, J. and Komari, T. (
Munnik, T. (
Munnik, T., Arisz, S.A., de Vrije, T. and Musgrave, A. (
Murzin, A.G., Lesk, A.M. and Chothia, C. (
Nalefski, E.A. and Falke, J.J. (
Okana, T., Kojima, D., Fukada, Y., Schichida, Y. and Yoshizawa, T. (
Pappan, K. and Wang, X. (
Park, C., Park, M.-H. and Chae, Q. (
Perisic, O., Fong, S., Lynch, D., Bycroft, M. and Williams, R.L. (
Ponting, C.P. and Kerr, I.D. (
Ponting, C.P. and Parker, P.J. (
Qin, W., Pappan, K. and Wang, X. (
Qin, C. and Wang, X. (
Rasmussen, S.G.F., Jensen, A.D., Liapakis, G., Ghanouni, P., Javitch, J.A. and Gether, U. (
Reimers, P.J. and Leach, J.E. (
Ritchie, S.M. and Gilroy, S. (
Ritchie, S.M. and Gilroy, S. (
Rodenburg, K.W., Juge, N., Guo, X.-J., Sogaard, M., Chaix, J.-C. and Svensson, B. (
Ryu, S.B. and Wang, X. (
Sambrook, J., Fritsch, E.F. and Maniatis, T.A. (
Sang, Y., Zheng, S., Li, W., Huang, B. and Wang, X. (
Smith, D.E. and Fisher, P.A. (
Smith, R.F., Wiese, B.A., Wojzynski, M.K., Davidson, D.B. and Worley, K.C. (
Sung, T., Roper, R.L., Zhang, Y., Rudge, S.A., Temel, R., Moss, B., Engebrecht, J. and Frohman, M.A. (
Thompson, J.D., Higgins, D.G. and Gibson, T.J. (
Towler, D.A., Adams, S.P., Eubanks, S.R., Towery, D.S., Jackson-Machelski, E., Glaser, L. and Gordon, J.I. (
Ueki, J., Morioka, S., Komari, T. and Kumashiro, T. (
Wang, C., Zien, C.R., Afitlhile, M., Welti, R., Hildebrand, D.F. and Wang, X. (
Wang, X. (
Wang, X., Dyer, J.H. and Zheng, L. (
Wang, X., Xu, L. and Zheng, L. (
Xu, L., Paulsen, A.Q., Ryu, S.B. and Wang, X. (
Xu, L., Zheng, S., Zheng, L. and Wang, X. (
Yamagata, H., Kunimatsu, K., Kamasaka, H., Kuramoto, T. and Iwasaki, T. (
Young, S.A., Wang, X. and Leach, J.E. (
Zheng, L., Krishnamoorthi, R., Zolkiewski, M. and Wang, X. (