(none)
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||
(none)
|
|
|
||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||
The EMBO Journal Vol. 17,pp. 6799-6811, 1998
INRA, Laboratoire de Biologie des Invertébrés, 123 bd F. Meilland, 06600 Antibes, 1 INRA, Laboratoire de Génétique et d'Amélioration des Plantes, Route de Saint Cyr, 78026 Versailles Cedex and 2 INRA, Laboratoire de Biologie Cellulaire, Route de Saint Cyr, 78026 Versailles Cedex, France
 |
Abstract |
|---|
Top
Abstract
Introduction
Results
Discussion
Materials & Methods
References
|
|---|
Sedentary plant-parasitic nematodes are able to induce the redifferentiation of root cells into multinucleate nematode feeding sites (NFSs). We have isolated by promoter trapping an Arabidopsis thaliana gene that is essential for the early steps of NFS formation induced by the root-knot nematode Meloidogyne incognita. Its pattern of expression is similar to that of key regulators of the cell cycle, but it is not observed with the cyst nematode. Later in NFS development, this gene is induced by both root-knot and cyst nematodes. It encodes a protein similar to the D-ribulose-5-phosphate 3-epimerase (RPE) (EC 5.1.3.1), a key enzyme in the reductive Calvin cycle and the oxidative pentose phosphate pathway (OPPP). Quantitative RT-PCR showed the accumulation of RPE transcripts in potato, as in Arabidopsis NFS. Homozygous rpe plants have a germination mutant phenotype that can be rescued in dwarf plants on sucrose-supplemented medium. During root development, this gene is expressed in the meristems and initiation sites of lateral roots. These results suggest that the genetic control of NFSs and the first stages of meristem formation share common steps and confirms the previous cytological observations which indicate that root cells undergo metabolic reprogramming when they turn into NFSs.
Keywords: Arabidopsis thaliana/d-ribulose-5-phosphate 3-epimerase/pentose phosphate pathway/promoter trapping/sedentary endoparasitic nematodes| |
Introduction |
|---|
|
Top
Abstract
Introduction
Results
Discussion
Materials & Methods
References
|
|---|
Among plant pathogens, sedentary endoparasitic nematodes interact with their hosts in a most fascinating way. They are able to induce the redifferentiation of root cells into nematode feeding sites (NFSs). Nematode growth and reproduction depend on the establishment of these NFSs. Nematodes withdraw food from NFSs until the completion of their life cycle, causing a tremendous threat to crop production worldwide (Sasser and Freckman, 1987
). It is not yet understood how these nematodes cause such alterations, but it is suspected that glandular secretions injected into plant cells interact directly or indirectly with the plant nuclear genome (Hussey, 1989). Cytological observations have indicated that NFSs are multinucleated with an enlargement of the nucleus and the nucleolus. Compared with normal cells, NFSs also show an increase in cytoplasmic density, a loss of normal vacuolation and a proliferation of cell organelles. Another characteristic feature of these structures is the development of cell-wall ingrowths, typical of transfer cells (Jones, 1981). These cell-wall ingrowths increase the surface area of the associated membrane and thus facilitate the import of elaborated photosynthates, minerals and other metabolites. Depending on the nematode species, the initial feeding cell develops into either a syncytium (for cyst nematodes such as Heterodera spp. and Globodera spp.) or a system of giant cells (for the root-knot nematodes Meloidogyne spp.) (Jones, 1981). Syncytia result from cell fusions after cell-wall dissolutions between the initial cell on which the nematode starts feeding and an increasing number of neighboring cells. Up to 200 cells can be incorporated in a large syncytium. Conversely, giant cell formation is the result of repeated nuclear divisions of the initial feeding cell without cytokinesis (Huang, 1985). Each root-knot nematode triggers the development of five to seven giant cells, each containing as many as 100 nuclei, which have undergone vast endoreduplication (Wiggers et al., 1990). Because Meloidogyne species can induce similar giant cells in several thousand host species, they probably interact with some fundamental key steps of the plant cell cycle (Niebel et al., 1996). In addition, root-knot nematode development is accompanied by divisions of cortical cells around the NFS, giving rise to a typical root-knot or gall. These complex morphological and physiological changes during the establishment of NFSs are reflected by altered gene expression in affected root cells (Gheysen et al., 1996; Williamson and Hussey, 1996). Approaches based on differential gene expression between healthy and infected roots have allowed the identification of cDNA clones with homology to several known plant defense genes (Niebel et al., 1993, 1995; Lambert, 1995). In parasitized cells, there is upregulation of cDNAs homologous to a key component of the protein ubiquitination pathway (E2 enzyme), a large subunit of RNA polymerase II, a Myb-type transcription factor and a plasmalemma H+-ATPase (Bird and Wilson, 1994). There is also an upregulation of a late embryogenesis-abundant protein (Van der Eycken et al., 1996). Sequencing of differentially expressed genes and computer searching of molecular data banks might indicate a putative function for the products they encode. This approach, however, must be coupled with biochemical and physiological investigations if their actual function is to emerge. More direct evidence for the role of known plant genes in the establishment or maintenance of NFSs came from a variety of promoter-gusA fusion constructs introduced into Arabidopsis and tobacco (Goddijn et al., 1993; Niebel et al., 1996). It has thereby been shown that the root-specific promoter TobRB7 (Conkling et al., 1990), which encodes a presumed water channel expressed in root meristematic and immature vascular cylinder regions, is reactivated in tobacco giant cells induced by Meloidogyne incognita (Yamamoto et al., 1991; Opperman et al., 1994a). Similarly, transcriptional activation of cell cycle markers such as the cyclin-dependent kinase CDC2a and the mitotic cyclin CYC1At is observed during the early stages of NFS formation (Niebel et al., 1996). Moreover, many other genes are downregulated in NFSs. For example, promoters of the bacterial nopaline synthase and the plant phenylalanine ammonia-lyase I gene, which are highly active in non-infected roots, are silenced within a few days of nematode infection (Goddijn et al., 1993).
To identify new genes and obtain a more comprehensive view of the molecular mechanisms underlying the induction and maintenance of NFSs, a promoter-trapping strategy was developed, with a promoterless 
-glucuronidase (GUS) construct being introduced randomly into the Arabidopsis genome via Agrobacterium
T-DNA transformation (Kertbundit et al., 1991; Topping et al., 1991; Goddijn et al., 1993). This `tagging' approach has been used in several laboratories, and tagged lines have recently been identified. However, despite interesting expression patterns in NFSs, none of these T-DNA flanking regions have presented homology with known genes (Barthels et al., 1997).
Here we report the molecular characterization of a strongly upregulated nematode-responsive gene in the early steps of giant cell formation induced by M.incognita in Arabidopsis. This gene encodes for the D-ribulose-5-phosphate 3-epimerase (RPE), an enzyme involved in the pentose phosphate pathway. This upregulation is also observed in syncytia induced by Heterodera schachtii, but at a lower level and later after nematode infection. Moreover, we demonstrated that this gene is similarly regulated in potato after infection by both cyst and root-knot nematodes. Finally, the implication of this enzyme in root apex and lateral root initiation sites indicates that the genetic control of NFSs and root formation share common steps and suggests that root cells undergo metabolic reprogramming when turning into NFSs.
| |
Results |
|---|
|
Top
Abstract
Introduction
Results
Discussion
Materials & Methods
References
|
|---|
Characterization of the RPE-tagged line
To isolate molecularly tagged lines showing GUS induction or repression in galls, we screened a collection of T-DNA-tagged Arabidopsis lines (ecotype WS) obtained by in planta transformation (Bechtold et al., 1993; Bouchez et al., 1993). These lines were screened in a GUS assay 7 days after infection with M.incognita. Of 3000 T-DNA tagged lines tested, 25 transgenic lines showed increased GUS expression, and three transgenic lines showed a repressed GUS expression in the galls. In other tissues, specific GUS activity was detected in 7% of the transformants. None of the tagged lines showed GUS expression that was exclusively restricted to the galls. Expression was observed in other parts of the plants and was always found in the root apex. One of these lines, named RPE, showed early and strong GUS expression in the galls. It presented an abnormal segregation (2:1 instead of 3:1) of the kanamycin marker carried by the T-DNA, suggesting mutant-impaired seed germination (embryo-lethal, seed viability or germination mutant). Of the 435 T2 seeds tested in bulk, 287 grew on kanamycin medium and displayed a normal phenotype compared with the wild type. This frequency fitted well with the hypothesis of the ratio of kanamycin-resistant (Kr
) to sensitive (Ks) being 2:1 (
2 = 0.09; P >0.05), suggesting also that the T-DNA was inserted in one nuclear locus.
To analyze precisely at which stage of the interaction the GUS gene is activated, we infected the RPE line in vitro. Time-course experiments pointed out the presence of an early induced plant promoter that activates the transgene in young galls less than 3 days post-infection (d.p.i.), that is to say, 24-48 h after giant cell initiation (Jones, 1981; Wyss et al., 1992). No GUS expression was detected at the penetration site (in the elongation zone) nor during nematode migration. The RPE line showed a strong GUS expression at 7 d.p.i. in galls induced by M.incognita (Figure 1A). This expression was maintained until sexual maturity of the females. To determine the localization of GUS expression, we made thin cryosections of the galls. Cross sections of 10-day-old galls clearly showed GUS staining in the giant cells of the gall (Figure 1B). No significant GUS activity was seen in cortical cells surrounding the feeding cells. Similar results were obtained with other Meloidogyne species such as M.javanica and M.hapla, and with the beet cyst nematode H.schachtii Schmidt (data not shown). Nevertheless, the latter species induced GUS expression later in the feeding cell (15 d.p.i.). During plant development, the GUS gene was expressed in the root meristem and in part of the elongation zone, in which cells divide and expand (Figure 2A). GUS activity was also visible early in the lateral root primordia, before visible evidence of root formation (Figure 2B). High levels of expression were also observed in aerial parts of the plant such as flowers (Figure 2C). In mature embryos from dry seeds, GUS activity was observed through more-intense staining in the zone corresponding to the root apical meristem and in cotyledons (Figure 2D), young tissues with active DNA synthesis for cell polyploidization (Brown et al., 1991).
|
|
To ascertain whether the T-DNA insertion was closely linked to the rpe mutation, we analyzed cosegregation of the mutant phenotype with the T-DNA insertion. Fifty T4 progenies resulting from each of the 40 selfed Kr T3 plants segregated with the expected ratios of 1/4 non-germinating seeds (mutant homozygous phenotype) and 3/4 germinating seeds of which 2/3 were Kr and 1/3 Ks (data not shown; 2 = 0.10; P >0.05). Thus there was no recombination between the rpe mutation and the
T-DNA cassette. These results suggested that rpe mutation was recessive; it prevented germination when homozygous and was actually linked to the T-DNA.
Molecular cloning of the RPE gene
DNA hybridization analysis of RPE plants through internal fragments of the T-DNA, GUS (Figure 3A) and left border (data not shown), confirmed that the RPE line carries a single intact T-DNA insert. A 515 bp genomic DNA fragment adjacent to the right border of the T-DNA was isolated by kanamycin plasmid rescue with PstI restriction sites (Bouchez et al., 1996) (Figure 3B). Using an inverse PCR strategy (IPCR; Earp et al., 1990) we cloned 657 bp of DNA flanking the left border (LB) of the T-DNA (BglII site from the IPCR primers T4 and T5) (Figure 3B). These two flanking sequences were used as probes on a Southern blot and indicated that the cloned fragments corresponded to a unique genomic fragment (data not shown). To analyze the insertion point of the T-DNA, we designed the oligonucleotides LA2 and RA2 (Figure 3B). Comparison of the nucleotide sequences of the transformed line with wild-type sequence revealed that a deletion of 77 bp and an insertion of a 5 bp filler sequence at the end of the right border resulted from the insertion of the
T-DNA. In addition, typical deletions in the 24 bp repeat of the left and right borders were observed; they resulted in the presence of only one nucleotide from the left border at the insertion junction (Gheysen et al., 1987; Mayerhofer et al., 1991).
|
Sequence analysis of the RPE cDNA
The RPE cDNA was cloned by rapid amplification of cDNA ends (5' and 3' RACE; Frohman et al., 1988) by using poly(A)+ RNAs from the wild-type Arabidopsis thaliana ecotype WS. Only one class of transcripts was detected. The RPE cDNA is 1259 nucleotides long with an open reading frame (ORF) of 281 amino acids, and with 5'- and 3'-untranslated regions (UTR) of 106 and 311 (268 + 43) nucleotides, respectively (Figure 4A). Translation was assumed to begin at nucleotide 107, the first ATG codon of the open reading frame. The context of this ATG does not match the plant consensus sequence (TAAACAATGGCTA; Joshi, 1987). Nevertheless, the nucleotide 3 bp upstream from the initiation codon is a purine, which is highly conserved in eukaryotic genes (Kozak, 1986). The 3' UTR contains two putative polyadenylation signal sequences, AATAAA and a G/T cluster (GTGTTTTT) at 22 bp and 37 bp, respectively, upstream from the polyadenylation site (Dean et al., 1986). Blast search (Altschul et al., 1990) on the Stanford A.thaliana database revealed a strong sequence identity (98.6% at nucleotide level), with eight expressed sequence tags (EST) from the A.thaliana ecotype Columbia (Figure 4A). cDNA and genomic sequence alignments demonstrated that the RPE gene is interrupted by nine introns, including one in the 3' UTR. The T-DNA was inserted into the third intron, and its resulting integration (deletion of
77 bp and insertion of 5 bp filler sequence) placed the ATG of the uidA gene in-frame with the rpe gene and allowed a functional translational -glucuronidase gene fusion. Both introns and exons were characterized; they varied in size from 71 to 286 and from 60 to 268 nucleotides, respectively. The consensus sequences for the intron/exon splice junctions correspond to those reported for plants (White et al., 1992). A Southern blot analysis of the WS genomic DNA was performed using the cloned cDNA as a probe. The gene appears as a single copy per haploid genome of A.thaliana (Figure 4B), even at low stringency hybridization and wash conditions (data not shown).
|
The predicted amino acid sequence of the RPE protein was compared to the NCBI database by the Blast network service. The protein is closely related to RPEs (EC 5.1.3.1) across species, with two conserved regions (Figure 5). RPE catalyzes the reversible interconversion of ribulose-5-phosphate and xylulose-5-phosphate in the pentose phosphate pathway. In particular 85, 81 and 80% of the resulting amino acids were identical to those of chloroplastic RPE from spinach (Nowitzki et al., 1995), potato (Teige et al., 1995) and rice (protein sequence deduced from rice RPE cDNA obtained by RACE PCR with primers localized on rice ESTs D39200 and D48105), respectively. This high degree of conservation between monocotyledonous and dicotyledonous plant sequences is typical for enzymes of sugar phosphate metabolism (Nowitzki et al., 1995). The region where the encoded Arabidopsis protein is similar to other bacterial or eukaryotic cytosoles (Figure 5) is preceded by a 46-amino-acid sequence with typical properties of chloroplast transit peptides (Von Heijne et al., 1989; Gavel and Von Heijne, 1990). This region is rich in the hydroxylated amino acids serine (19.6%) and threonine (8.7%). We thus predict that RPE is a plastid-localized protein.
|
Homologies were
found also with the RPE from bacteria and yeast. Amino acid identity was
65% with its homolog from the cyanobacteria Synechocystis (Kaneko et al., 1996), and 50 and 48%, respectively, with the corresponding enzymes of Serratia marcescens and Haemophilus influenzae (Fleischmann et al., 1995). In addition, 42% amino acid identity was found with RPE_YEAST (Miosga and Zimmermann, 1996) and RPE_MYCOB (Figure 5). To determine the relationship between these enzymes, we carried out a phylogenetic analysis with a heuristic tree-building program. Plant RPE sequences clustered and formed a lineage distinct from that of yeast RPE. They are more similar to the cyanobacterial and eubacterial homologs than to the other eukaryotic sequence, RPE_YEAST (Figure 6).
|
Finally, the RPE gene was mapped on the A.thaliana CIC genomic library (Creusot et al., 1995). Four positive YAC clones carrying the RPE gene were identified by hybridization. These four clones (CIC3A3, CIC4D8, CIC9G7 and CIC10A10) have previously been anchored to chromosome 5 (contig 29) near the SEP5A and the LFY3 RFLP markers (Schmidt et al., 1997).
Expression of RPE genes in plants
To investigate the expression of the RPE gene, we isolated total RNAs and poly(A)+ RNAs from galls induced by M.incognita, and uninfected root tissues (root meristems, lateral root initiation sites and non-meristematic root fragments) of in vitro-grown wild-type Arabidopsis plants. RNA blots were not sufficiently sensitive to measure RPE gene expression because it was difficult to isolate enough tissue and/or because of the low level of RPE mRNA expression, as observed in roots (Teige et al., 1995). Since no gene was available as a control in the feeding cell, we used competitive PCR to amplify and quantify cDNA copies from low-abundance RPE mRNAs (Gilliland et al., 1990; Kaneko et al., 1992). The target RPE cDNA was coamplified with primers RA2 and LA2 in the presence of a dilution series of a competitor DNA of known concentration, which differs from the RPE cDNA by a small deletion of 35 bp (Figure 7A). The quantification of RPE expression was performed on root tissues from 3 to 15 d.p.i. This time-course experiment showed that the quantity of RPE mRNA in galls is constant (data not shown). The results obtained at 15 d.p.i. showed that no RPE mRNAs were detected in the non-meristematic root fragments, and very few mRNAs were detected in the lateral root initiation sites. The highest quantity of the RPE mRNA was observed in the root meristems and giant cells of NFSs induced by M.incognita (Figure 7B and C).
|
To test the RPE upregulation in plant NFSs, we infected potato plants with Meloidogyne chitwoodi and the cyst nematode Globodera rostochiensis, two very important pathogens on potato crops. Two-week-old galls and syncytia and root tissues from uninfected plants were analyzed for RPE mRNA expression. The competitive PCR was carried out with potato-RPE-specific primers and a competitor issued from potato cDNA. There was as much potato RPE mRNA in the galls as in the root meristems, whereas there was 30% less in the syncytia (Figure 7C). Therefore, the regulation of RPE in potato is similar to that found in Arabidopsis.
Rescue of rpe mutant and complementation
Because the homozygous rpe mutant is impaired in sugar phosphate and lethal, we investigated whether exogenous carbohydrate addition would lead to the rescue of the rpe mutant. When grown in vitro on medium that provides 2% sucrose, kanamycin-resistant dwarf plants appeared with a 1:2 ratio. These plants had light-green leaves and reduced root systems with few lateral roots (Figure 8A). These seedlings developed slowly and died when transferred to the soil. We used PCR assay to test for cosegregation of the rescued phenotypes with the rpe mutation. PCR experiments were carried out with two RPE primers (LA2 and RA2), which span the rpe T-DNA insertion site, and a third primer (U1) specific for the sequence of T-DNA right border (Figure 3C). When genomic DNA from RPE plants was used as a template, both a 437 bp and a 914 bp band were amplified, indicating the presence of both the mutant and wild-type alleles. In contrast, when DNA from dwarf rescue plants was used as a template, only the 437 bp product was obtained from amplifications with all three primers (Figure 8B). This analysis confirmed that all 200 dwarf plants were homozygous for the rpe allele.
|
In an attempt to show genetic complementation of rpe, RPE/rpe heterozygous plants were transformed by vacuum infiltration with the 35S-CMV::RPE cDNA sense construct. As expected, all of the 50 independent primary transformants selected on hygromycin B and sucrose 2% showed a wild-type phenotype. Of these T1 plants, 11 segregated 100% kanamycin-resistant T2 plants, i.e. homozygous for the rpe allele. Transfer of antibiotic-resistant T2 dwarf mutants to agar plates containing hygromycin revealed that all were hygromycin sensitive. Thus these plants did not contain the 35S-CMV::RPE cDNA sense construct. These results confirmed by PCR analysis (data not shown) are in agreement with the complementation of the rpe/rpe homozygous mutant by the cloned RPE cDNA and provides additional evidence that the rpe phenotype is due to the insertion of the T-DNA into the RPE gene.
Nematode sensitivity of the rescued mutants
To test the implication of RPE in NFS formation, we have infected the dwarf mutants with M.incognita and H.schachtii. In vitro H.schachtii infection of these plants showed that all the plants were attacked with a 2-fold reduction of the number of cysts compared with heterozygous and WS plants tested under the same conditions (Table I). This difference could be explained by the lower number of roots produced by the homozygous plants. However, the number of eggs per cyst and the infectivity of the larvae produced by rpe mutants were comparable to those produced by heterozygous and wild-type ones. Conversely, under the same conditions, in vitro inoculation of the homozygous plants with M.incognita showed that the nematodes are not able to produce galls on most plants (95%). On those rare attacked plants, the number of galls was limited to only one, whereas on heterozygous or control plants, 10-20 galls were generally observed. Moreover, the development of the gall was extremely reduced and later aborted. These results indicate that the morphology of the root dwarf mutant does not affect the behavior of nematodes, and show that, in contrast to syncytia formation, RPE is necessary for giant cell formation.
|
| |
Discussion |
|---|
|
Top
Abstract
Introduction
Results
Discussion
Materials & Methods
References
|
|---|
Isolation of an upregulated nematode-responsive gene
We have isolated, by promoter trapping, an upregulated nematode-responsive gene, RPE, in Arabidopsis. This gene is likely to be directly involved in the biochemical make-up of giant cells. RPE gene encodes a key enzyme in the pentose phosphate pathway. The RPE gene was cloned after T-DNA tagging by promoterless gusA constructs by using in planta transformation (Bechtold et al., 1993). This is the first report of the isolation of an upregulated gene in NFSs through a tagging strategy. In comparison to classical in vitro transformation (Koncz et al., 1989; Kertbundit et al., 1991; Topping et al., 1991), tagged lines obtained with in planta transformation showed lower frequency of GUS activity (7%). Nevertheless, this frequency seems to be in better accordance with a random event of T-DNA insertion in the Arabidopsis genome in the proximity of a plant promoter. However, this difference could be due to the structure of the vector itself, to specific differences in the transformation and selection process, or in the copy number of T-DNA inserts (Bouchez et al., 1993). The average copy number in the tagged lines obtained is 1.5 copies per transformant. In addition, the occurrence of inverted repeats over the right border (20%) may have reduced the frequency of GUS activation.
Screening for GUS expression after M.incognita infection showed that expression of many genes normally expressed at different developmental times or in different cell types are upregulated in giant cells. These results in Arabidopsis are consistent with previous observations in tomato (Bird and Wilson, 1994) and confirm the complex morphological and physiological changes in cells during their modification into nematode feeding site (Jones, 1981; reviewed in Atkinson et al., 1994; Niebel et al., 1994).
Roles of RPE in plants
RPE
(EC 5.1.3.1) (also known as pentose-5-phosphate 3-epimerase) catalyzes the
reversible interconversion of ribulose-5-phosphate and xylulose-5-phosphate. In plants, RPE is integral to both the oxidative pentose phosphate pathway (OPPP) and the reductive Calvin cycle (Dennis and Turpin, 1990; Schnarrenberger et al., 1995). Arabidopsis appears to possess only a single, nuclear-encoded, plast-located RPE enzyme, as in potato (Teige et al., 1995) and spinach (Nowitzki et al., 1995; Schnarrenberger et al., 1995). In this report, we present data on the first rpe mutant in plants. The homozygous rpe mutant plants fail to germinate without exogenous carbohydrate and can be partially rescued with added sucrose. Another plant mutant, the chlorina42, which encodes a light-regulated chloroplast protein of unknown function, has been shown to be rescued on sucrose-containing medium (Koncz et al., 1990). The RPE expression in the photosynthetic tissues and the phenotype of the light-green mutant rescued in a sucrose-containing medium are consistent with the implication of this enzyme in sugar phosphate recycling of the Calvin cycle. Nervertheless, RPE is not expressed only in green tissues containing chloroplasts, as described previously (Teige et al., 1995), but also in specific root cellular types at sites of phloem unloading where
consumption of carbon skeletons is high. The RPE transcript was detected in the proliferating cells of root tips and in the small subset of cells involved in the initiation of lateral roots. Lateral roots are produced when meristems develop de novo from one or a few cells within the pericycle, a differentiated tissue surrounding the vascular cylinder (Peterson and Peterson, 1984). In contrast, expression is never observed in the root cap and in the root specialization zones, which cells enter as they acquire their differentiated attributes (Schiefelbein and Benfey, 1991). The pattern of RPE expression in roots is very similar, if not identical, to those observed in Arabidopsis for other genes implicated in cell division cycle competence or growth, such as mitotic-like cyclin CYC1At (Ferreira et al., 1994) and cyclin-dependent kinase CDC2a (Hemerly et al., 1993). The RPE expression in cotyledons of mature embryos may be correlated to the DNA endoreduplication during cell polyploidization (Brown et al., 1991). These results confirm that a high flux through the OPPP supports a high rate of biosynthesis in actively growing cells. The OPPP plays a crucial role in cells by producing NADPH that is required in numerous biosynthetic reactions (e.g. fatty acids and isoprenoid compounds such as sterols), and by generating carbohydrate intermediates for the synthesis of nucleotides and cell-wall polymers (Figure 9) (Wood, 1986).
|
The degree of conservation between the A.thaliana and spinach sequences suggests that RPE also encodes a functional enzyme. Plant RPE is more similar to the homologs of cyanobacteria and eubacteria than to the sequence from yeast. The gene phylogeny for these sequences supports the hypothesis that higher plants have obtained their RPE gene from eubacteria (perhaps cyanobacteria) via endosymbiotic (organelle to nucleus) gene transfer (Nowitzki et al., 1995), similar to the scenario observed for chloroplast GADPH (Martin et al., 1993) and chloroplast 3-phosphoglycerate kinase (Brinkmann and Martin, 1996).
RPE roles in feeding cells
We showed that the RPE gene is affected during nematode parasitism and is necessary for the giant cell formation. RPE is upregulated in the early steps of feeding cell formation induced by Meloidogyne spp. but not by cyst nematodes (H.schachtii and G.rostochiensis). This is consistent with the admitted hypothesis that giant cells and syncytia, despite analogous ultrastructure, differ completely in the way they are formed (Jones, 1981). Giant cells develop by repeated mitosis without cytokinesis and subsequent cell expansion. The multinucleate state of syncytia results from cell-wall dissolution and coalescence of adjacent cells. In this case, cell division and metaphase chromosomes have never been reported. However, in A.thaliana, key regulators of the cell cycle, CDC2a and CYC1At, are induced early in the initial steps of both NFS formation (Niebel et al., 1996). The lack of early RPE expression in syncytia agrees well with previous 3H-thymidine incorporation experiments which indicate that syncytia are relatively quiescent in terms of DNA synthesis, in contrast to giant cells, which undergo heavy synthesis in early steps of induction (Gheysen et al., 1997).
Later, RPE accumulates in both types of developing NFSs throughout the nematode life cycle. This expression in both NFSs (15 d.p.i.) is in accordance with the establishment of an active metabolic sink that provides food to the adult females, which require a considerable amount of nutrients for egg production.
From a general point of view, it has been shown that the pentose phosphate pathway plays a crucial role in the host-protozoa parasite relationship. It maintains a pool of NADPH, which serves to protect against oxidant stress and generates carbohydrate intermediates used in nucleotide and other biosynthetic pathways (Barrett, 1997). The human parasite Plasmodium falciparum also stimulates host cell activity (Atamna et al., 1994), and deficiency in the first enzyme of the pathway, glucose-6-phosphate dehydrogenase, protects human erythrocytes from infection (Ruwende et al., 1995). In giant cell formation, pronounced activities of glucose-6-phosphate and 6-phospho-gluconate dehydrogenases found in histochemical preparations (Endo and Veech, 1969) together with the activation of RPE seem to be in accordance with the role of the OPPP in host-parasite interaction.
The implication of RPE in root meristems and initiation sites of lateral roots indicates that the genetic control of NFS and the first stages of meristem formation share common steps. This result confirms that root cells undergo metabolic reprogramming when they turn into NFSs. Some animal parasites, such as Trichinella spiralis, initiate similar changes in differentiated mammalian muscle cells (Jasmer, 1995). These results support the hypothesis that `normal' biochemical functions have been recruited to play key roles in halting pathogen growth (Opperman et al., 1994b). Determining the mechanisms underlying RPE expression at the onset of feeding-cell formation should give insight into the molecular basis of this morphogenetic event.
Towards engineering plant nematode resistance
The mechanisms by which nematodes influence plant cell metabolism should share regulatory features in different plant species, since a range of sedentary nematode species are able to develop NFSs in numerous host plants (Goddijn et al., 1993). We demonstrated that the RPE gene in potato is regulated in a manner similar to that in Arabidopsis after infections by sedentary parasitic nematodes. This result confirms that Arabidopsis, besides being ideally suited for genetic dissection of development (Meyerowitz, 1989), is also a good host plant model for studying molecular interactions with parasitic nematodes (Sijmons et al., 1991; Niebel et al., 1994). The identification and cloning of nematode-responsive plant genes and promoters constitute a major challenge for our knowledge and understanding of the plant-nematode interaction and the development of novel approaches to engineering plant resistance (reviewed in Gheysen et al., 1996; Grundler, 1996). Nematode-responsive promoters may be used to express anti-nematode proteins (Atkinson et al., 1994), phytotoxic proteins or essential genes (anti-sense approaches) that interfere with the development of feeding cells (Sijmons, 1993). Furthermore, determining how a nematode selects particular root cells and modifies them to serve as a feeding site will enhance our understanding of normal cell development and may serve in identifying genes that regulate aspects of cell division.
| |
Materials and methods |
|---|
|
Top
Abstract
Introduction
Results
Discussion
Materials & Methods
References
|
|---|
Plant materials and infection with nematodes
A T-DNA mutagenized A.thaliana (L.) Heyn seed collection (3000 lines) was obtained and characterized previously in INRA Versailles (Bechtold et al., 1993). The plant transformation vector pGKB5, designed for promoter trapping and gene tagging (Bouchez et al., 1993), was used to transform wild-type A.thaliana, ecotype Wassilewskija (WS) by vacuum infiltration. Plants were grown in a greenhouse on sand at 20°C under a light/dark regime of long days (16 h light and 8 h darkness). The seedlings were incubated at 4°C for 2 days in the dark for cold treatment. T-DNA tagged T3 seedlings were selected with Basta herbicide. Three-week-old lines were inoculated with 100 M.incognita second-stage juveniles (J2). One and 2 weeks after inoculation, the plants were carefully harvested by rinsing in water and were GUS-stained as described below. For in vitro analyses of seedlings, seeds were surface-sterilized and grown on Gamborg B5 medium (Sigma) containing 1% sucrose, 0.8% agar (plant cell culture tested, Sigma) and 50 µg/ml kanamycin. Kanamycin resistance was scored in 2-week-old seedlings. For in vitro nematode infection, 100 surface-sterilized freshly hatched J2 of M.incognita or H.schachtii were added on each 2-week-old seedling as described previously (Sijmons et al., 1991). The plates were kept at 20°C in a cycle of 16 h light/8 h darkness. GUS induction time course experiments were done daily after nematode infection. Potato plants Solanum tuberosum L. cv Désirée were grown in a greenhouse and inoculated with M.chiitwoodi J2 or G.rostochiensis cysts under the same conditions.
Histochemical localization of GUS activity
-glucuronidase (GUS) activity was assayed histochemically with
X-Glu according to Jefferson et al. (1987), modified by the use of buffer 0.5 mM K3Fe(CN)6 and 0.5 mM K4Fe(CN)6 in 0.1 M NaH2PO4 pH 7, to prevent potential staining artefacts (De Block and Debrouwer, 1992). For microscopic analyses, galls were excised from X-Glu-stained plants, fixed for 30 s in a microwave oven (temperature 50°C fixed) in freshly prepared 2% formaldehyde/4% glutaraldehyde in 50 mM sodium phosphate buffer pH 7.2 (PBS), washed with PBS, embedded in OCT Compound (BDH Laboratory Supplies) and frozen in nitrogen-cooled isopropane. Frozen embedded material was sectioned at 6 µm (at -20°C) using a Leica JUNG CM 1800 cryomicrotome. Whole-mount Arabidopsis and cryosections were observed through a Zeiss Axioplan 2 microscope with dark-field illumination to increase the sensitivity of the detection.
Southern blot and RNA analysis
For Southern blot analysis, genomic DNA was isolated from 4-week-old plants grown in vitro as described by Doyle and Doyle (1990). Southern blots were performed as described in Sambrook et al. (1989), using 5 µg of genomic DNA. For RPE RNA analysis on wild-type Arabidopsis and potato, galls and syncytia were excised after nematode infection. For comparison, root tissues (non-meristematic root fragments, root tips, initiation sites of lateral roots) were obtained from uninfested plants. All tissues were harvested rapidly and with a minimum of manipulation, immediately frozen in liquid nitrogen and stored at -80°C until use. Total RNA was isolated as described by Chomczynski and Sacchi (1987). Poly(A)+ RNA was isolated with Quickprep Micro mRNA Purification Kit (Pharmacia P-L Biochemicals Inc.) according to the manufacturer's instructions. RNA samples were separated in a formaldehyde-containing agarose gel, blotted and hybridized as described in Sambrook et al. (1989).
Isolation of T-DNA flanking sequences
Kanamycin plasmid rescue of T-DNA right-border flanking sequence was permformed as described by Bouchez et al. (1996). The genomic DNA from transgenic plants was digested by PstI (unique site at the beginning of the nptII gene) and ligated into pResc38 vector. Escherichia coli DH12S cells (Gibco-BRL) were transformed with the ligation mixture. Ampicillin and kanamycin selection provided clones containing the complementing region from the T-DNA plus a flanking genomic DNA region. IPCR amplification of the T-DNA left-border flanking sequences was performed as described by Earp et al. (1990). Genomic DNA of transgenic plants was digested by BglII and ligated with T4 DNA ligase. The PCR was done with T-DNA oligonucleotides T4 (5'-GTA-CAT-TGC-CGT-AGA-TGA-AAG-ACT-3') and T5 (5'-CTA-CAA-ATT-GCC-TTT-TCT-TAT-CGA-C-3') as depicted in Figure 3C. The oligonucleotides T4 and T5 hybridize 500 and 300 bp away from the 3' T-DNA left border, respectively. Amplification was performed with a cycle of 1 min denaturation at 94°C, 1 min annealing at 58°C and 2 min polymerase extension at 72°C, repeated 35 times, with a final 10 min 72°C extension in an MJ Research PTC-200 Peltier Thermal Cycler. The amplified fragment was cloned and sequenced with T7 polymerase sequence kit (Pharmacia).
RACE (Rapid Amplification of cDNA Ends)
The 5' end of the RPE cDNA was cloned using the 5' RACE kit (Gibco-BRL) with poly(A)+ RNA from the wild-type A.thaliana ecotype WS (Frohman et al., 1988). First-strand cDNAs were made using the RPE antisense-specific primer LA1 (5'-CCT-TGA-ACG-CAT-TCG-CTG-GAG-TGA-C-3'). After C-tailing, PCR amplification was performed using an anchor primer (5'-CTA-CTA-CTA-CTA-CTA-GGC-CAC-GCG-TCG-ACT-AGT-ACG-GGI-IGG-GII-GG G-IIG-3') and a nested antisense RPE-specific primer LA2 (5'-GAT-ACC-TTC-TCT-GCA-CAC-ATT-TTC-C-3'). The PCR cycle was 94°C for 1 min; 55°C for 2 min; 72°C for 2 min; repeated 35 times, with a final 10 min 72°C extension. The 3' end of the RPE cDNA was cloned using firststrand cDNAs made with a primer oligo(dT) (5'-TGA-AGG-TTC-CATGAA-TCG-ATA-GGA-ATT-CTT-TTT-TTT-TTT-TTT-TTT-TVN-3').Each PCR was done with the primer 3RACE (5'-TGA-AGG-TTC-CAT-GAA-TCG-ATA-G-3') and the specific primer RA2 (5'-CAG-AAG-GAG-AAC-TCA-TGG-AAT-TG-3'). PCR products were cloned in pUAg (Ingenius) and sequenced.
Mapping
The RPE gene was mapped on the A.thaliana (ecotype Columbia) CEPH-INRA-CNRS (CIC) yeast artificial chromosome (YAC) genomic library (Creusot et al., 1995). Membrane hybridization screening was performed using the RPE cDNA as a probe.
Competitive RT-PCR for quantification of mRNA
Poly(A)+ RNAs from galls induced by M.incognita and WS Arabidopsis root tissues were reverse transcribed using the First-Strand cDNA Synthesis Kit (Pharmacia) with the oligo(dT) primer. The resulting first strand cDNAs were used as template for PCR amplification. To quantify accurately the RPE cDNA abundance, a competitor matrix was created by deleting 35 bp (EcoRV-StuI fragment) from the RA2-LA2 PCR-amplified RPE cDNA (573 bp). A precise dilution series was prepared, ranging from 1 to 10-7 ng of competitor template (538 bp). We performed an initial PCR titration against a broad range of dilutions in log increments. We then performed a second finer titration over a narrower range of dilutions for precise quantification. The accuracy of this method is improved through the use of master mixes. We prepared a master mix containing, per reaction, oligonucleotides RA2 and LA2 10 pmol each, dNTP 200 µM each, 1× PCR buffer, taq DNA polymerase 0.5 U (Appligene) and 1 µl of first-strand cDNA. We added 90 µl of this mixture to 10 µl of previously prepared competitive template. PCRs were performed according to the following temperature profile (35 cycles): 94°C, 1 min; 55°C, 1 min; 72°C, 2 min. Since any change in the variables that influence the efficiency of the PCR will affect equally the yield of both competitive and target cDNA PCRs, the relative ratio of the two products remains unchanged. The relative amounts of each PCR product were readily quantified by direct scanning with a densitometer of ethidium-stained 2% agarose gels. The amount of competitor is multiplied by the ratio of target cDNA bp per competitor bp to correct for increased ethidium staining per mole of the larger fragment. The competitor product per target product was plotted as a function of the starting concentration of competitor. At the point where cDNA and competitor products are in equivalence, the starting concentration of RPE cDNA added to the PCR is equal to the known concentration of the competing deleted cDNA.
Competitive RT-PCRs were performed on poly(A)+ RNA from galls induced by M.chitwoodi, syncytia induced by G.rostochiensis and potato root tissues. The primers RP2 (5'-CTT-TCA-CCA-GGA-GGA-GAA-TTC-A-3') and LP2 (5'-CTC-AAG-TCC-GAG-ATT-TTC-TT-3') were complementary, respectively, to the 5'- and 3'-ends of the S.tuberosum RPE-coding sequence. In this case, the competitive template is a 50-bp-deleted potato RPE cDNA obtained after MseI restriction.
Plant transformation
For the sense expression of the RPE cDNA in transgenic plants of A.thaliana, the complete RPE cDNA was amplified on A.thaliana (WS) cDNA using 5' (5'-GTC-CCG-GGT-CAA-CCT-CCG-CCG-CTT-CCT-TG-3') and 3' primers (5'-GTC-CCG-GGG-ACA-AAT-GAA-ATG-CTT-TAT-TAT-AAG-TC-3'). The SmaI fragment was ligated into the binary vector pCAMBIA 1302 digested by NcoI and BstEII. The proper orientation with regards to the 35S-CMV promoter was verified by sequencing. The sense construct was introduced into A.tumefaciens strain C58C1 and used to transform heterozygous Rpe/rpe A.thaliana plants by vacuum infiltration (Bechtold et al., 1993). Primary transformants were selected on GB5 medium containing 2% (w/v) sucrose and 75 µg/ml hygromycin B and transferred to soil. Plants were grown to maturity and allowed to self.
DNA sequence analysis
The BLAST search program (Altschul et al., 1990) was used for sequence analysis and comparisons in the DDBJ/EMBL/GenBank
and swissProt databases at the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/cgi-bin/BLAST/) and in the Arabidopsis Database (AtDB) Project (http://genome-www.stanford.edu/Arabidopsis/). Multiple sequence alignments and the unrooted neighbor-joining dendrogram were determined using CLUSTAL W (Thompson et al., 1994). The significance of the phylogenetic results was assessed using bootstrap analysis (Felsenstein et al., 1985). PROSITE (http://expasy.hcuge.ch/sprot/scnpsit1.html) allows a protein sequence to be scanned for signature patterns.
Accession numbers
The sequence data of the A.thaliana and rice RPE cDNA have been submitted to the DDBJ/EMBL/GenBank databases under accession Nos AF015274 and AF047444, respectively.
| |
Footnotes |
|---|
| |
Acknowledgements |
|---|
We thank Florence Exbrayat for technical assistance, Didier Mugniery and Laurent Marché (INRA Rennes) for infected potato plants, Nicola Von Mende (IACR Rothamsted, GB) for advice on in vitro nematode infection, CAMBIA for pCAMBIA 1302 vector and Claude Castella (INRA Antibes, F) for advice on cryosectioning. We also thank Thierry Candresse (INRA Bordeaux, F), Philippe Castagnone, David Pauron and Marie Noëlle Rosso (INRA Antibes) for helpful discussions. This work was supported by the European Community's BIOTECH Programme Bio4-CT96-0318 and the GDR Arabidopsis, INRA AIP No. 176.
| |
References |
|---|
|
Top
Abstract
Introduction
Results
Discussion
Materials & Methods
References
|
|---|
-glucuronidase assay. Plant J., 2, 261-266.-glucuronidase as a sensitive and versatile gene fusion marker in higher plants.
EMBO J.,
6,
3901-3907[Abstract].-glucuronidase gene fusions in Arabidopsis thaliana.
Proc. Natl Acad. Sci. USA,
88,
5212-5216[Abstract].Received on January 19, 1998; revised and accepted on September 28, 1998.
The EMBO Journal Vol. 17 No. 23 pp. 6799-6811, 1998, Copyright İ1998 Oxford University PressThis article has been cited by other articles:
|
|
 |
|
  B. Favery, L. A. Chelysheva, M. Lebris, F. Jammes, A. Marmagne, J. de Almeida-Engler, P. Lecomte, C. Vaury, R. A. Arkowitz, and P. Abad Arabidopsis Formin AtFH6 Is a Plasma Membrane-Associated Protein Upregulated in Giant Cells Induced by Parasitic Nematodes PLANT CELL, September 1, 2004; 16(9): 2529 - 2540. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Savenstrand, M. Brosche, and A. Strid Regulation of Gene Expression by Low Levels of Ultraviolet-B Radiation in Pisum sativum: Isolation of Novel Genes by Suppression Subtractive Hybridisation Plant Cell Physiol., April 15, 2002; 43(4): 402 - 410. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. S. Springer Gene Traps: Tools for Plant Development and Genomics PLANT CELL, July 1, 2000; 12(7): 1007 - 1020. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
